There are inherent controversies in both aspirations.

To focus on a single topic in an issue would turn it into a collective monograph, probably adding research value; yet with a population of 1.3 million, Estonia simply lacks sufficient research capacity for such efforts. Abandoning a central axis would certainly expedite the work of an author, but he or she would no longer need to consider the matters that are existential to Estonian research—to avoid ambiguity, to seek to address the most important topics, with little regard to perhaps intriguing but still marginal issues.

Focussing one’s attention on only internationally significant topics would discount the simple fact that the law in the world these days is still largely national law, and it is the development of national law that drives the progress in classical international, supra- and transnational as well as European law. If the reader now takes a look out of the window, he or she will immediately realise that there is no abstract global national environment but it is the tree growing by his or her window that is unique. The constitutional rules must pass an international human rights’ test, while the Constitution still governs the functioning of a particular country and its population. A referendum must be carried out with due regard to democratic principles, but it is the people who vote, not abstract subjects of law. Consumer protection has been thoroughly regulated on the EU level, but it is essentially experienced as the treatment you receive in a nearby supermarket or the amount of trust you have in your publisher’s agent.

The second part of the title, Reciprocal Impact, sets out to demonstrate that it is a two-way street. The above reference to Estonia’s population implies that we do not have the resources available to address all research topics. Yet we do not distinguish strictly between important and unimportant topics. All the topics discussed here are important, and only time will tell in due course which of them will have more influence on us and which will become our contribution to the development of internationally recognised jurisprudence. Perhaps some topics will sink into oblivion, but we are lucky not to know that yet; this is not likely either, given the substance and quality of the articles.

Dear reader, come and learn about Estonian law by reading this issue. And do not be concerned, you will not be leaving the international dimension while you go through the articles, based on their content and level. Estonian law does not detach itself from international jurisprudence, it is the very jurisprudence.

Estonia has used remote Internet-based voting in five elections: twice each in municipal and Riigikogu (parliamentary) elections and once in European Parliament elections. The number of ‘I-voters’ has grown sharply from less than 10,000 in 2005’s municipal elections to over 140,000 in the 2011 parliamentary elections. The latter account for 24.3% of all votes cast and 56.4% of the advance votes. Initially, no individual complaints claiming unconstitutionality of I-voting were filed in court. In 2011, the situation has changed: critical public debate has re-emerged, followed by several complaints.

Only Estonia, Switzerland, Norway and a few other countries allow legally binding remote I‑voting, though some countries are on their way toward its countrywide use. The list of countries that have abandoned the use of e-voting in various forms is much longer, including the US, Germany, Finland, and the Netherlands. ^*1 France, for example, tries to keep alive the tradition of voting only at the polling station, as this ritualises citizenship ^*2 , but has allowed proxy voting and recently remote I-voting from abroad. The reasons for allowing or giving up on I-voting are different, but constitutional questions of whether fair and free voting can be secured in the case of remote I-voting have always been raised.

We are facing the pressure of the information society ^*3 : people require e-services, yet, on the other hand, cyber-threats are more serious than ever before. ^*4 Social changes have already forced countries to allow remote postal or proxy voting. ^*5 We have to admit that holding on to old traditions (one single election day, casting of paper ballots in a controlled environment as the only option, etc.) will not be possible in the future, but free and fair elections, anonymity of the vote, and the principle of uniformity must be guaranteed. The Council of Europe has adopted recommendation ^*6 and guidelines ^*7 for electronically enabled elections, and the OSCE/ODIHR is looking for ways to observe and evaluate various forms of e-voting, including I-voting. Estonia’s I-voting experience is internationally followed with special attention; any failures would have very negative consequences not only for Estonian democracy but for all I-voting projects, worldwide.

The concept of the Estonian I-voting system is described and analysed here in the light of theoretical literature, judgements of the Supreme Court of Estonia, and the empirical data available. In addition to statistics, the results of sociological surveys are used.

2. Description of the concept of Estonian I-voting

Estonia’s I-voting system is based on an electronic roll of voters, a compulsory e-ID, the public/private key infrastructure (‘virtual double-envelope scheme’), and the right to change a vote given online (‘virtual voting booth’). The elements of the system are meant to guarantee the compliance of the I-voting with constitutional principles of elections: only people entitled to vote can vote, access to voting shall be equal, one vote per voter shall be counted, free voting shall be granted, and both counting of the voting results and election results shall be fair and sound. Brief description of the elements of the Estonian I‑voting system is given in this section; the constitutional analysis follows in Section 3.

2.1. Electronic Population Register

The Estonian Population Register is a uniform database of personal data of Estonian citizens and foreigners with Estonian residence permits. The Estonian voter roll is held on the basis of the Population Register, and voters do not have to enrol specially before elections. The Estonian electoral law ^*8 states that electoral rolls are drawn up 30 days before election day but additions to the list can be made until the very end of elections. This gives the list the property of being constantly up to date in practice. During Internet voting, the voting roll is updated daily. ^*9

2.2. ID card and m-ID

The cornerstone of most e-services, public as well as private, is the e-ID. ^*10 Since 2002, an ID card has been the new generation’s mandatory primary identification document. The ID cards are issued by the government and contain certificates for remote authentication and digital signature. Every Estonian citizen or resident alien above age 15 must have an ID card.

Each ID card contains two discrete PKI-based digital certificates—one for authentication and one for digital signing. The certificates contain only the holder’s name and personal code and have two associated private keys on the card, each protected by a unique user PIN. The certificates have no restrictions of use: they are by nature universal and meant to be used in any form of communication, whether between private persons or organisations or within the government. The e-ID card can be used also for encryption of documents so that only the person intended to view the document can decrypt it. This is an efficient means for secure transfer of documents over public networks. In addition, each ID card has all data printed on it also in electronic form, in a special publicly readable data file.

The number of ID cards issued grew in June 2010 to exceed 1.1 million. Over 2/3 of cardholders have used the e-ID card for remote personal identification and more than 1/3 for digital signature. Here it has to be noted that Internet voting has strongly promoted electronic use of ID cards. Another important promoting factor has been the agreement among banks to allow Internet banking only with an ID card or PIN calculator. The old password cards can be used only for very small transactions.

To use the ID card, one needs a smartcard reader and a computer with the relevant software installed (free for download from the Web page https://installer.id.ee/); an Internet connection; and a Windows, Mac, or Linux operating system.

A couple of years ago, a new e-ID solution was brought to the market: the m-ID, where a mobile telephone (via its SIM card) acts as an ID card and a card reader at the same time. In addition to having the functionality of an ordinary SIM, a mobile-ID SIM holds a person’s mobile identity that enables providers of Internet services to identify the person and to issue digital signatures. ^*11 Personal identification and digital signature functionality are secured by up-to-date security technology and corresponding personal identification numbers. Making the solution more convenient, with this, one does not need an ID card reader for the computer any longer; instead, one can perform electronic transactions just as one would with an ID card: it enables logging in to databases, Internet banks, etc. and signing various types of contracts digitally. The m-ID certificate is issued by the state and is thereby an equally e-enabled document to the ID card. The m-ID can be used as a means of authentication and digital signature in elections from 2011.

In practice, an e-ID is used for user authentication with several databases ^*12 ; the above‑mentioned state portal serving as an e-service centre, e-tickets for public transportation, a customer loyalty programme identification tool in several private companies, and even insertion of comments for the online daily newspaper Eesti Päevaleht, which has prohibited anonymous comments in order to prevent libel cases. The use of e-ID is steadily widening, although the initial aim of combining e-ID with all possible other documents, such as driving licences, and replacing all possible password‑based solutions has not been fulfilled yet. ^*13

2.3. System architecture

The Estonian IT security experts in their security analysis ^*14 published in 2003 and revised in 2010 declared that in a practical sense the Estonian I-voting system was secure enough for implementation. In absolutely secure systems, unexpected events are not possible. One may dream about such systems, but they can never be realised in practice. This applies particularly to I-voting systems. Considering the security level of personal computers, it is impossible to design I-voting systems that are absolutely secure for every user. The most important security goal of voting is not to affect the final results and not to abuse the constitutional principles. Single incidents with users are still important, but they do not have an influence on the final result. Moreover, small-scale incidents are acceptable even in traditional voting systems. ^*15

The part of I-voting in the whole process of organising elections is relatively small. The system uses existing information systems—the Population Register for the polling list, election information system of the National Electoral Committee (hereinafter referred to as the NEC) for the collection and publication of information on candidates and voting results, and the infrastructure of Certification Centre Ltd. for checking ID card (or m-ID) certificates.

The main components of the Estonian I-voting systems are the voter application; the Vote Forwarding Server; and the back office, which is divided in two: the Vote Storage Server and the Vote Counting Application. These components support the following processes:

–     The voter application is a Web-based application or an application on voters’ personal computers.

–     The Vote Forwarding Server is responsible for authentication, checking of enfranchisement, sending a list of candidates to voters, and receiving signed and encrypted ballots.

–     The network server immediately transfers the received encrypted ballots on the Vote Storage Server and transfers the acknowledgements of receipt from the Vote Storage Server to the voters. The network server completes the work when the I-voting period finishes.

–     The Vote Storage Server receives encrypted ballots from the network server and stores them until the end of the voting period. The Vote Storage Server is responsible for cancellation and management of votes.

–     The Vote Counting Application is an off-line program that summarises all encrypted ballots. The encrypted ballots are transferred from the Vote Storage Server to the Vote Counting Application via data carriers. The Vote Counting Application does not receive voters’ digital signatures, and it does not know voters’ personal data.

Additionally, the I-voting system delivers independent log files, which consist of tracing data for the received encrypted ballots from the Vote Forwarding Server, all annulled encrypted ballots, all encrypted ballots sent to the Vote Counting Application, and all counted encrypted ballots. The cryptographic protocol used links all records in the log files. The NEC has the right to use the log files to resolve disputes. Hence, there is an independent audit trail to verify the e-voting process and help solve problems should they appear. ^*16 The legality of all elections depends on the presence and proper functioning of these components.

2.4. Measures used to ensure voting secrecy

In order to understand how the I-voting system guarantees secret and equal voting, we should briefly describe the envelope voting method used in Estonia for advance paper voting. The latter gives the voter the possibility to vote outside the polling station for the voter’s residence in any rural municipality or city. A voter presents a document for entry in the list of voters and then receives the ballot and two envelopes. The inner envelope has no information about the identity of the voter, and the ballot paper is put in it. The inner envelope is placed in an outer envelope, on which the voter’s details are written, so that, after the end of the advance poll, the envelope can be delivered to the voter’s polling station of residence. There it is verified whether the voter has the right to vote; then, the inner envelope is taken out and placed unopened into the ballot box. The two-envelope system guarantees that the voter’s choice remains secret. The same system but electronically built is used in Internet voting. ^*17

Asymmetric cryptography is used to guarantee the secrecy of votes. A pair of keys is generated for the system in a special safety module so that its private component never leaves this environment. The public component of the pair of keys is integrated into the voter application and is used to encrypt the votes. The private component of the pair of keys is used in the vote-counting application to open the votes on the evening of election day. The NEC can open the votes—i.e., use the private component—only collegially. After the period for dealing with any complaints has elapsed, the private key is destroyed.

2.5. ‘Virtual voting booth’

In order to guarantee the freedom of voting, I-voters have the right to replace the vote cast on the Internet by means of another I-vote or a paper ballot. However, this can be done only on advance polling days. In the case of several I-votes being cast, only the last one is counted; in the event of contradiction between an I-vote and paper ballot, the paper ballot is deemed definitive. If multiple physical ballots are cast, all votes are declared invalid. ^*18 Thus the ‘one voter—one vote’ principle is guaranteed.

3. Analysis of the constitutionality of Internet voting

According to the Estonian Constitution ^*19 , members of the Riigikogu, as well as local government councils and the European Parliament shall be elected in free, general, equal, and direct elections, and voting shall be secret. There is no special regulation of I-voting in the Constitution. The legal framework for I-voting is laid down in electoral law. The provisions are almost the same in all legal acts regulating voting procedures. In the case of I-voting, almost all principles of democratic elections give rise to several questions in constitutional law and, more broadly, in social sciences.

3.1. A teleological interpretation of the principle of secrecy

The secrecy of voting has traditionally been viewed in Estonia as the right and obligation to cast one’s vote alone in a voting booth. In the case of Internet voting, it is impossible to ensure the privacy aspect of the voting procedure. The voter’s right to anonymity during the counting of the votes can be guaranteed, indeed to the extent to which this can be secured in the case of remote postal voting. Therefore, remote Internet voting requires rethinking of the privacy principle.

The principle of privacy is there to protect a person from any pressure or influence acting counter to his or her free expression of political preference. Such a teleological approach to the Constitution was the basis of the I-voting provisions from the very beginning of the whole project. In short, the provisions enabling Internet voting are based on the premise that the government has to trust the individual and avoid, whenever possible, interference with decision-making at the individual level. ^*20 The individual has to be aware of the risks—e.g., technical risks—and he or she has to have the right to decide whether or not to exercise the Internet voting opportunity. The Supreme Court has agreed with this teleological approach to the principle of secrecy. ^*21

Buchstein, on the other hand, does not agree:

Mandatory secrecy is a principle which goes beyond constitutional law, its fundaments are based on the idea of auto-paternalism and it is understood as a mechanism of self-binding of autonomous citizens in order to avoid situations of external pressure or corruption. In this concept, it is not the individual him- or herself, but a warranted outside agent or authority—normally the state—that is responsible for providing the necessary means to allow for the secret ballot. ^*22

Indeed, postal voting as another form of absentee ballot is widespread and is becoming accepted in Germany. There, the Federal Constitutional Court has twice declared remote postal voting to be constitutional, arguing that facilitation of voter turnout outweighs, in this case, the problems possible in relation to security and public scrutiny of electoral processes. ^*23 In France, by contrast, postal voting was abolished in 1975 because of incidents of fraud. ^*24

3.2. Increase of turnout

One of the declared aims of launching online voting in Estonia was to increase voter turnout, which perhaps could be described more realistically as broadening access possibilities and stopping the decrease in participation. Scholars point out on the positive side of I-voting also that I-voting could and should better accommodate the needs of disabled voters. ^*25

The actual impact of Internet voting on the turnout does not lend itself to objective analysis. One can determine the variations of turnout in different election years (comparing equivalent types of elections) and attempt to clarify the causes underpinning variations with the aid of sociological studies. Perhaps the most important question is what proportion of the electorate would not have participated in the voting had the Internet voting opportunity not been provided. There does not exist a way of obtaining empirical evidence. We must, therefore, come to terms with unverifiable claims made by the voters themselves. The only exception is the case where Internet voting provides the only possibility for the elector to vote and he or she takes advantage of this possibility. For example, the local government council elections in Estonia do not provide the possibility of voting abroad by postal ballot or at a diplomatic representation. Nonetheless, it is possible to vote over the Internet when abroad. ^*26

Table 1. I-voting statistics for 2005–2011 ^*27

Source: National Electoral Committee

I-voting seems to have had, in 2005, a slight effect on the increase in the turnout of voters who sometimes vote and sometimes not. ^*28 In 2007, approximately 10% of those I-voters questioned said that they certainly or probably would not have voted without having had the possibility to vote via the Internet. Moreover, Trechsel and Vassil show that the percentage of the I-voters questioned who certainly or probably would not have voted without having had the possibility to vote via the Internet has risen to 16.3%, which allows the conclusion that the overall turnout might have been as much as 2.6% lower in the absence of such a method of voting. That is already a significant marker when one looks at the impact of Internet voting on the overall turnout. ^*29

3.3. Uniformity

3.3.1. The digital divide and equal opportunities for representation

Trechsel et al. concluded in the report prepared for the Council of Europe following the experience of the Internet voting in 2005 and 2007 that education and income, as well as type of settlement, are insignificant factors in the choice of Internet voting over other voting methods. One of the most important findings of that study was that it is not so much the divide etc. between the Internet access ‘have’s and ‘have-not’s as, clearly, computing skills, frequency of Internet use, and trust in the I-voting procedure that direct voters’ decisions to use or not use I-voting. Age has remained a significant factor for some years. ^*30 Moreover, some interesting conclusions have been drawn in the latest report by Trechsel and Vassil, in 2010, where they state that the ICT variables (computing knowledge and frequency of Internet usage) have disappeared since the 2009 elections as predictors of Internet voting usage. ^*31

In the discussion of equal access to the place of voting, some authors ^*32 ignore the fact that in Estonia there are quite many different voting methods; for example, if a voter is unable to vote at a polling place as a result of his or her state of health or for another good reason, he or she may apply to vote by paper ballot at home on the day of election day (Riigikogu Election Act, §46 (1)).

The Estonian Supreme Court has stated:

The principle of equal treatment in the context of electing representative bodies does not mean that absolutely equal possibilities for performing the voting act in equal manner should be guaranteed to all persons with the right to vote. In fact, those who use the different voting methods provided by law (advance polls, voting outside the polling division of residence, voting in custodial institutions, home voting, voting in a foreign state, etc) are in different situations. For example, the voters who have to use the possibility of advance polls, are in a situation different from that of the voters who can exercise their right to vote on the election day. The guarantee of absolute actual equality of persons upon exercising the right to vote is infeasible in principle and not required by the Constitution. ^*33

In the future, the number of people without Internet access will probably decrease, but the digital divide is going to be even deeper than before. People without Internet access will have significantly less information, no access to voting-advice applications, etc. In this case, it is not the access to I-voting (as long as other methods of voting remain) but access to the candidates’ and parties’ information that might be the constitutional problem.

3.3.2. Impact on the voting results

The most intriguing question for political parties is probably that of the impact of the use of I-voting on results. Impact on the voting results can result from the fact that votes cast by those voters who would not participate if I-voting did not exist may not be distributed proportionally over the political spectrum. However, studies have shown that the left– right auto‑positioning of the voter does not play any important role in the choice of a voting channel. The same applies to the 2009 ^*34 and 2011 elections.

Table 2. Relationship of I-votes to all votes cast for a political party

Data: National Electoral Committee

a) = Percentage of I-votes

b) = Proportion of I-votes to total votes, in per cent

RP = Reform Party

PRU = Pro Patria and Res Publica Union (in 2005 only Res Publica)

PP = Pro Patria Union (merged with Res Publica to form PRU since 2007)

SD = Social Democratic Party

GP = Green Party

CP = Centre Party

In comparison of the overall distribution of votes in Internet voting or e-votes with that for total votes, not only the growing proportion of e-votes could be observed. According to Table 2, the party that is most popular in electronic voting is not always the one that profits the most from e-voting. The PRU (PP) and the GP (instead of the winner, RP) have been the greatest beneficiaries of Internet voting. The small numbers of e-votes on the account of the otherwise popular CP can be explained by that party’s strong opposition to Internet voting from the very beginning ^*35 but probably also by specifics of the electorate.

The hypothesis that I-voting rewards advantages to urban voters found no proof. Gender is also not an important factor when one chooses I-voting from among the possible voting channels. Age, by contrast, is quite an important factor in choosing Internet voting. ^*36 Yet still, as can be seen in Figure 1, no age group is clearly dominant. The 55+ age group, with up to 20% of all Internet voters, is worthy of note here. So, while being younger correlates with use of the Internet as a means of voting, age does not give all the answers.

(Source: National Electoral Committee)

Figure 1. Age of I-voters in 2005 to 2011

It is, nevertheless, very interesting to compare the age groups taking part in Internet voting with the general electorate. For lack of a more comprehensive reference, we examine survey data from an exit poll conducted at the 2007 parliamentary elections by the Tartu University Department of Political Science. ^*37 According to the poll the age groups break down as follows: ages up to 24 accounting for 12.3%, 25–34 for 16.3%, 35– 44 for 19.5%, 45–55 for 16.5%, and over-55s for 35.4%. When comparing these figures to the Internet voting results for 2007, we see a strong over-representation in the under-35 group and under-representation in the over-55 age group. This appears to be consistent with the importance of age in the decision to choose Internet voting as a voting method.

3.3.3. The right to change one’s I-vote

The President refused to promulgate amendments, which allowed I-voting and gave to the I-voter the right to replace I-vote once given with another I-vote or paper-ballot, to the Local Government Council election act ^*38 , arguing that I-voters are in a better position when compared to other voters, who do not have any right to change their vote once cast. ^*39 The initial version of the I‑voting law included the possibility of changing the I-vote with a paper ballot not only during advance voting but also on election day. To solve some of the problems indicated by the President, the Riigikogu restricted the time of I-voting to advance voting days. The chance to change their election preferences on Sunday after receiving additional information about candidates in the second half of the week had really placed I-voters in a better position. After this change, all voters who take advantage of advance poll possibilities were formally acting in the same conditions. The President did not see these changes as sufficient and initiated constitutional review.

The Supreme Court Chamber of Constitutional Review pointed out that, despite repeated electronic voting, there was no possibility of an I-voter affecting the voting results to a greater degree than can those voters who use other voting methods. From the standpoint of the voting results, this vote was deemed in no way more influential than a vote cast by paper ballot.

The most important arguments of the Supreme Court were the following. The principle of freedom of the vote gives rise to the obligation of the state to protect voters from persons attempting to influence their choice.

The aim of increasing voter turnout is without any doubt legitimate. The measures the state takes for ensuring the opportunity to vote for as many voters as possible are justified and advisable. Another aim in allowing I-voting is the modernisation of voting practices that coincides with the aims of I-voting listed in the recommendation Rec(2004)11, on legal, operational, and technical standards for I-voting, of the Council of Europe.

In accordance with the Penal Code, preventing a person from freely exercising his or her right to elect or be elected in an election or to vote in a referendum, if such prevention involves violence, deceit, or threat or takes advantage of a service, economic, or other dependency relationship of that person with the offender, is punishable by a pecuniary punishment or up to one year of imprisonment. The possibility for the voter to change the vote cast by electronic means throughout the advance polling period constitutes an essential supplementary guarantee to the observance of the principle of free elections and secret voting upon voting by electronic means.

A voter who has been illegally influenced or watched in the course of electronic voting can restore his or her freedom of election and the secrecy of voting by voting again, either electronically or via a ballot paper, after having been freed from the illegal influence. In addition to the possibility of subsequently rectifying a vote given under such influence, the possibility of voting again serves an important preventive function. When the law guarantees a voter who is voting electronically the possibility of changing a vote cast by electronic means, the motivation to influence him or her illegally decreases.

There are no measures as effective as the possibility of changing a vote cast by electronic means for guaranteeing the freedom of election and secrecy of voting upon electronic voting by means of an uncontrolled medium. The infringement of the right to equality and of uniformity, which the possibility of I-voters to change their vote an unlimited number of times can be regarded as amounting to, is not sufficiently intensive to outweigh the aim of increasing participation in elections and introducing new technological solutions. ^*40 Norwegian scholars arrived at similar principles independently before obtaining in-depth knowledge of the Estonian Internet voting system. ^*41

In fact, the number of changed and replaced votes has been low in all elections. The maximum number of replaced votes has been 100, and the percentage of repeated votes does not exceed 4% of total e-votes. ^*42 So, any fears of misuse of these opportunities cannot be validated.

In short, the fact that the Internet voter is in a somewhat different position from the traditional voter does not in itself indicate an infringement of the constitutional values. The Supreme Court thus confirmed the constitutionality of one of the main premises of the remote Internet voting project.

3.3.4. Computing skills and security of the voter’s computer

It has been noted that good computing skills have been an important factor in choosing Internet voting as a mode of voting in the 2005 and 2007 elections. Since 2009, the ICT variable has lost its meaning in defining the reasons behind the choice of using e-enabled voting. However, since the absolute number of Internet voters has steadily risen, the question of technical uniformity and usability emerges. I-voting has been offered in a variety of environments and on several platforms claiming to cover the maximum number of possible voters. In addition, comprehensive informational materials and a 24-hour help line are available. ^*43 However, a peculiar issue arose in the 2011 elections. There were a few voters who used a very rare combination of screen resolution, Windows 7, and font sizes on their computer. When these people used the Internet voting application, some of the interface and control buttons were left behind the Windows taskbar. This would not have been a greater problem unless some of the candidates’ names too were covered by the taskbar. One of the candidates brought a complaint to the Supreme Court that stated:

The chamber adds that in organising Internet voting the state has to guarantee the accordance of the application with most common hardware, operating systems, resolutions, and fonts. In some cases, compliance cannot be guaranteed. In the event of such problems, the voter has the option of contacting the technical support staff. If the issues cannot be resolved, the voter can use the traditional means of voting. ^*44

Therefore, ensuring the compatibility of the computer with the Internet voting application is clearly left to the user.

The security analysis of the Internet voting concept ^*45 states clearly that one of the fundamental security problems with electronically enabled voting is the necessity of trusting the voters’ computer. The central system can be, and is, protected by the state. The spread of malware on private computers, on the other hand, cannot easily be limited—either by the state or through private efforts. The analysis even says that the modern personal computer is a ‘black box’ that nobody is able to control. Therefore, the security of the computer on which the voting application is run remains an issue in actuality. The user—the voter—can, of course, take actions to protect the computer, but, nevertheless, this cannot resolve all possible consequences. Accordingly, the security of the voting application is a topic that is being given extra attention.

However, the issue of secrecy became prominent during the 2011 elections when a computer enthusiast hacked his own vote in the voting application on his own computer. He was able to modify the vote and create an illusion of the vote not having been sent to the central system. He was also keen to go public with his discovery (to national media) and later bring the issue up to the Supreme Court. It is important to state that all of the problems and situations discovered were monitored in the central system and that the threats revealed had been discussed already in the 2003 security analysis.

Subsequently, the Supreme Court, in its judgement No. 3-4-1-4-11 ^*46 , stated that knowingly manipulating one’s own vote cannot be seen as grounds for indictment of the overall security of the Internet voting system. In an analogy with traditional voting, a voter could easily go to the polling booth and make the polling paper invalid (by scrapping or doodling on the paper, etc.). That is a conscious decision and is completely legitimate.

However, the debate about secrecy is never resolved. Another issue that was raised by the computer enthusiast described earlier is the traceability of a vote. The reasoning behind this is that the online environment cannot be trusted and additional external proof of compliance has to be generated. A very interesting Internet voting pilot project is to be introduced in late 2011 in Norway. ^*47 In this project, external means of confirming one’s choices are used. Namely, voters receive a special printed polling card (by post) with all candidates who are running for election represented by code names. After voting, the voter can request the code name matching the vote cast, via independent channels. This should, in theory, guarantee that the vote can be traced and that it has been accepted.

However, some additional concerns arise with this. Firstly, new channels of communication have to be built and secured between the state and the voter. Secondly, issues with the principle of anonymity come up where the voter has to understand that under some circumstances the state knows how he or she has voted. Thirdly, how does this traceability affect the possibility of buying or selling one’s vote over the Internet?

4. Certification and auditing

Certification is, in broader term, a process of confirmation that an e-voting system is in compliance with prescribed requirements and standards and that it at least includes provisions for ascertaining that the system is functioning as intended. This can be done through measures ranging from testing and auditing to formal certification. The end result is a report and/or a certificate. An audit is an independent pre- or post-election evaluation of a person, organisation, system, process, entity, project, or product, which includes quantitative and qualitative analysis. ^*48

Currently, there is no domestic or international public body that would be ready to certify and audit all the elements of an entire I-voting system before, during, and after election procedures. In Estonia, hired specialists performed comprehensive tests in order to check the functionality and accuracy of the system both as experienced by testers and in public (in demo voting). A third party audits the source code and the procedures that have been carried out.

The Estonian I-voting system was developed to follow the principle that all components of the system must be transparent for audit purposes. Procedures should be fully documented, with those that are critical being logged, audited, observed, and videotaped as they are conducted. A common requirement is that the source code of the voting application be available for auditing. In Estonia, though the code is not universally available, it could be audited if so agreed by the NEC.

As a rule, the process audit is ordered from external internationally certified IT auditors. The audit reviews and monitors sensitive aspects of the process, such as updating of the list of voters, preparation of hardware and its installation, loading of election data, maintenance and updating of election data, and the process of counting the votes etc. At the counting event on election day, auditors publicly declare their opinion about the soundness of the procedures of the electoral administration to that point. The report of the auditors, released after all procedures are complete (including the destruction of all voting equipment—I-votes along with it), states whether the I-voting procedures followed the rules described in the system’s documentation and whether the integrity and confidentiality of the system was not endangered. To date, all reports have been positive.

The I-voting system produces a wealth of system log information that can be used to monitor the work of the system thoroughly. In its different stages, the I-voting system produces a variety of logs concerning received, cancelled, and counted votes, also invalid and valid votes. The audit application enables determining what happened to an I-vote cast by a specific person without revealing the voter’s choice. These logs provide external auditors as well as observers with information that they can use to ensure that the system is working correctly.

According to the Estonian electoral laws, all activities related to elections are public. Observers have access to the meetings of all election committees and can follow all electoral activities, including the voting process, the counting, and tabulation of results. Internet voting is no different. All significant documents describing the I-voting system are public. In order to enhance the observers’ knowledge of the system, the political parties are invited to take part in a training course before each election, in which I-voting is used. Besides political parties, auditors and other persons interested in the I-voting system take part in the training. In addition, observers are invited to follow the testing of the whole process and take part in other preparatory procedures. However, few political parties have so far exercised their opportunity to observe the I-voting procedures. ^*49 It is important that observers be deployed for an amount of time that suffices to allow meaningful observation. If some important stages influencing the correctness of final results have not been observed, conclusions cannot be made as to the integrity of the system.

The OSCE did audit the 2007 elections, and in its report it states that the “election administration implemented the [I-voting] system in a fully transparent manner, and appeared to take measures to safeguard the conduct of Internet voting to the extent possible”. ^*50 Professional, independent, reliable, and comprehensive IT audit and certification procedures should compensate for the lack of simple public scrutiny.

5. Conclusions

In Estonia, as well as in many other countries that have prepared systems for, and allowed, postal voting, advance voting, and other supplementary voting methods, voting at a polling station has virtually lost its significance as a ritual of transforming people into a nation-state and a carrier of sovereign nationhood.

In discussion surrounding the introduction of I-voting, the classical arguments concerning the conformity of I-voting with the principles of fair elections (including the reliability of the electronic voting systems) have gained renewed force. For example, one of the typical arguments against I-voting is that people who have no commitment to go to the polling station to execute their citizen’s duty should not participate in governing at all, which contradicts the axiom that the higher the turnout the better.

A possible lack of legitimacy of the election results could result from either of the following situations:

–     The privacy of individual I-voting procedure cannot be supervised by authorities or observed in a traditional way. Therefore, massive buying and selling of votes, as well as exercise of other influence or pressure on the voter, is possible.

–     The people themselves cannot verify the I-voting results, and people need to have absolute faith in the accuracy, honesty, and security of the whole electoral system (its people, procedures, software, and hardware) if it is to be legitimate. For people who didn’t take part in developing the system, the computer operations can be verified only by knowing the input and comparing the expected with the actual output. In a secret ballot system, there is no known input, nor is there any expected output with which to compare the electoral results. ^*51

Therefore, the question of whether remote Internet voting with binding results in public political elections complies with the constitutional principles of fair voting cannot be answered simply with a ‘yes’ or ‘no’. Actually, the question and answer should be divided into two parts. The first sub-question should be whether the legal norms in the abstract comply with the constitutional provisions and the second whether the technical solution used to conduct voting procedures in a certain election guarantees constitutionality.

The first sub-question can be answered on the basis of theoretical analysis, but the second should be examined before and after the relevant elections. The fact that it is possible not to fulfil the legal requirements set for an I-voting system is not enough per se for declaring I-voting as a concept unconstitutional. As a matter of fact, this underscores the importance of qualified certification and auditing of the system as well as the need for a new approach in electoral observation. The second sub-question can be answered with a ‘yes’ only if sufficient measures are in place to check whether the IT solutions work properly. This leads to a requirement that auditing, certification, and evaluation as required in the Council of Europe guidelines ^*52 be foreseen by law or NEC regulation.

In the Estonian case, the first sub-question could be answered ‘yes’, as e-ID enables secure remote identification, e-ID has wide penetration, all advance voters are placed in the same conditions, and the ‘virtual voting booth’ (the right to replace an I-vote with another I-vote or a paper ballot) and ‘virtual double-envelope system’ ensure freedom of voting and uniformity of elections. Moreover, the system is justified by the aim to guarantee universal suffrage in an information society where e-services (including Internet voting) are demanded by a significant proportion of the electorate. Whilst formal equality can be provided, the questions of material equality and the issue of the digital divide remain. In addition, complying with the principle of secrecy poses new obstacles for many countries. According to the above teleological interpretation of the principle of secrecy, the voting act is to be seen not as an aim but as a measure to guarantee freedom of voting, and the anonymity aspect of the principle of secrecy can be guaranteed. The analysis of the compliance of the Estonian I-voting system with the United Nations International Covenant on Civil and Political Rights has given positive result as well ^*53 but emphasises the importance of special procedures to facilitate auditing and observation of I-voting. ^*54

 The answer to the second sub-question is more complicated. Internet voting in concrete election is constitutional if the provisions of the law are fulfilled in practice: only people entitled to vote can vote, e-votes cast over the Internet are recorded and counted properly, and only one vote per voter shall be counted. Independent IT auditing that covers all aspects of the system can prove its soundness. The proper performance of the IT system should be certified and audited before, during, and after voting. The personal computer and the Internet remain a weak point of the system. The scholars are probably right in saying that “[a]lthough perfect real-time knowledge of all cyber threats is an impossible goal, it is realistic to do mach better at providing a richer, better integrated picture of our cyber security to the technologists, attorneys, and political leaders who will have to collaborate to avert the next cyber attack”. ^*55 Both new threats and I-voting are part of the information society.

Notes:

^*1 See the database for the Competence Center for Electronic Voting and Participation, at http://db.e-voting.cc/. German constitutional court decision to declare the use of voting machines unconstitutional: BVerfG, 2 BvC 3/07 vom 3.3.2009, Absatz-Nr. (1-163). Available at http://www.bverfg.de/entscheidungen/cs20090303_2bvc000307.html (9.10.2011). The core of the decision in German:

            Der Grundsatz der Öffentlichkeit der Wahl aus Art. 38 in Verbindung mit Art. 20 Abs. 1 und Abs. 2 GG gebietet, dass alle wesentlichen Schritte der Wahl öffentlicher Überprüfbarkeit unterliegen, soweit nicht andere verfassungsrechtliche Belange eine Ausnahme rechtfertigen.

            Beim Einsatz elektronischer Wahlgeräte müssen die wesentlichen Schritte der Wahlhandlung und der Ergebnisermittlung vom Bürger zuverlässig und ohne besondere Sachkenntnis überprüft werden können.

^*2 L. Monnoyer-Smith. How I-voting technology challenges traditional concepts of citizenship: An analysis of French voting rituals. – R. Krimmer (ed.). Electronic Voting 2006: 2nd International Workshop Co‑organised by the Council of Europe, ESF TED, IFIP WG 8.6, and E-Voting.CC. Bonn: Gesellschaft für Informatik 2006, pp. 63–64.

^*3 W. Drechsler. Dispatch from the Future. – The Washington Post, 5.11.2006.

^*4 J. Farwell, R. Rohozinski. Stuxnet and the Future of Cyber War. – Survival 2011 (53) 1, pp. 23–40.

^*5 See, e.g., the thorough overview of remote postal voting in N. Kersting. Briefwahl im Internationalen Vergleich. – Österreichische Zeitschrift für Politikwissenschaft 2004 (33) 3, pp. 325–328.

^*6 Recommendation on legal, operational, and technical standards for e-voting, Rec(2004)11; Recommendation on electronic democracy, Rec(2009)1. Available on the Council of Europe Web site.

^*7 Certification of E-voting Systems, GGIS(2010)3E; Guidelines on transparency of e-enabled elections, GGIS(2010)5E. Available on the Council of Europe Web site.

^*8 Up-to-date translations of election laws are available on the National Electoral Committee Web site: http://www.vvk.ee/?lang=en (9.10.2011).

^*9 For more information, visit the Web site of the Ministry of Internal Affairs, specifically http://www.siseministeerium.ee/35796/ (9.10.2011).

^*10 Detailed information about e-IDs, the areas of their use, etc. can be found at http://www.id.ee/?lang=en (9.10.2011).

^*11 More about the m-ID project can be found at http://id.ee/?id=10995.

^*12 For example, the Estonian Research Portal, at https://www.etis.ee/index.aspx?lang=en, which compiles information on all Estonian researchers and their scientific projects, publications, and activities.

^*13 Comprehensive coverage of the ID card can be found in the work of T. Martens and E. Maaten. E-voting is here to stay. – Baltic IT&T Review 2006 (1).

^*14 Available from the NEC Web site at http://www.vvk.ee/public/dok/E-voting_concept_security_analysis_and_measures_2010.pdf.

^*15 T. Mägi. Practical Security Analysis of I-voting Systems. Master’s Thesis 2007. Available at http://triinu.net/e-voting/master%20thesis%20e-voting%20security.pdf.

^*16 General description of the Estonian Internet voting system, 2010. Available at http://www.vvk.ee/public/dok/General_Description_E-Voting_2010.pdf (9.10.2011).

^*17 Details of the double-envelope scheme and description can be found in the General Overview document (Note 16).

^*18 Riigikogu Election Act (Riigikogu valimise seadus), §40 (6). – RT I 2002, 57, 355; RT I, 10.12.2010, 1 (in Estonian).

^*19 Translations of Estonian legal acts can be found at http://www.just.ee/6906. Up-to-date official versions of all legal acts are available from the State Gazette, at http://www.riigiteataja.ee/ (in Estonian).

^*20 The ideological foundation and parliamentary debates are explored by W. Drechsler, Ü. Madise. E-voting in Estonia. – Trames 2002 (6) 3, pp. 234–244; W. Drechsler, Ü. Madise. Electronic Voting in Estonia. – N. Kersting, H. Baldersheim (eds.). Electronic Voting and Democracy: A Comparative Analysis. Basingstoke: Palgrave Macmillan 2004, pp. 97–108.

^*21 Available at http://www.nc.ee/?id=381 (9.10.2011).

^*22 H. Buchstein. Online Democracy. Is It Viable? Is It Desirable? Internet Voting and Normative Democratic Theory. – N. Kersting, H. Baldersheim (eds.). Electronic Voting and Democracy: A Comparative Analysis, Basingstoke: Palgrave Macmillan, pp. 39–58.

^*23 BVerfGE 21, 200 (15.02.1967); BVerfGE 59, 119 (24.11.1981). Available at http://www.wahlrecht.de/wahlpruefung/index.htm (9.10.2011).

^*24 L. Monnoyer-Smith (Note 2), p. 63.

^*25 See, e.g., M. Loncke, J. Dumortier. Online voting: A legal perspective. – International Review of Law, Computers & Technology 2004 (18) 1, pp. 60–61.

^*26 Ü. Madise, E. Maaten. Internet Voting in Estonia. – D. R. Insua, S. French (eds.). Advances in Group Decisions and Negotiation Vol 5 e-Democracy: A Group Decision and Negotiation Perspective. Dordrecht, Heidelberg, New York, London: Springer 2010, pp. 314–316.

^*27 LE—Local Elections, PE—Parliament Elections, EPE—European Parliament Elections.

^*28 F. Breuer, A. Trechsel. E-voting in the 2005 local elections in Estonia: Report for the Council of Europe 2006, available at the Council of Europe Web site.

^*29 A. Trechsel, K. Vassil. Internet Voting in Estonia: A Comparative Analysis of Four Elections Since 2005. Council of Europe and European University Institute 2010. Available at http://www.vvk.ee/public/dok/Report_-_E-voting_in_Estonia_2005−2009.pdf (9.10.2011).

^*30 A. Trechsel. Internet voting in the March 2007 Parliamentary Elections in Estonia: Report for the Council of Europe, 2007. Available at http://www.vvk.ee/public/dok/CoE_and_NEC_Report_E-Voting_2007.pdf.

^*31 A. Trechsel, K. Vassil (Note 29).

^*32 See, e.g., S. Meagher. When Personal Computers Are Transformed into Ballot Boxes: How Internet Elections in Estonia Comply with the United Nations International Covenant on Civil and Political Rights. – American University International Law Review 2008/23, pp. 374–376.

^*33 CRCSCd, 1.9.2005, 3-4-1-13-05, paragraph 24. Available at http://www.nc.ee/?id=381 (9.10.2011).

^*34 A. Trechsel, K. Vassil (Note 29).

^*35 Ü. Madise, E. Maaten, P. Vinkel. Internet Voting at the Elections of Local Government Councils on [sic] October 2005, Report on Internet voting to the NEC, 2006. Available at http://www.vvk.ee/public/dok/report2006.pdf (9.10.2011).

^*36 A. Trechsel, K. Vassil (Note 29).

^*37 R. Toomla. Results of 2007 Riigikogu elections exit polls. Conducted by the Department of Political Science of Tartu University. Unpublished, available to the authors.

^*38 Draft No. 607 SE in X Riigikogu proceedings. The draft, information regarding parliamentary procedures, and motions to change the draft are available on the Parliament Web site at http://www.riigikogu.ee/?page=eelnou2&op=ems&eid=607&assembly=10&u=20110420131938 (9.10.2011) (in Estonian). The I-voting provisions were first adopted as a law in 2002; see drafts 747 SE, 748 SE, 771 SE, and 906 SE in IX Riigikogu proceedings. Right before the very first use of I-voting in 2005 municipal elections, the Riigikogu decided to change some I-voting provisions and the President used his suspensive veto foreseen in §107 of the Constitution of Estonia.

^*39 Decision No. 873, 22.6.2005. Available at http://vp2001-2006.president.ee/et/ametitegevus/otsused.php?gid=64640 (in Estonian).

^*40 CCRSCd, 1.9.2005, 3-4-1-13-05 (Note 33).

^*41 G. Skagestein, A. V. Haug, E. Nodtvedt, J. Rossebo. How to create trust in electronic voting over an untrusted platform. – R. Krimmer (ed.) (Note 2), p. 108.

^*42 See Table 1 for further data.

^*43 Available at http://www.valimised.ee/internet_eng.html (9.10.2011).

^*44 In 3-4-1-6-11. Available at http://www.riigikohus.ee/?id=11&tekst=RK/3-4-1-6-11 (in Estonian).

^*45 See Note 14.

^*46 Available at http://www.riigikohus.ee/?id=11&tekst=RK/3-4-1-4-11 (9.10.2011) (in Estonian).

^*47 For more information about the Norwegian Internet voting system, see http://www.regjeringen.no/en/dep/krd/prosjekter/e-vote-2011-project.html?id=597658 (9.10.2011).

^*48 Council of Europe Rec(2004)11 and guidelines based on that recommendation (see Note 7).

^*49 E. Maaten, T. Hall. Improving the Transparency of Remote I-voting: The Estonian Experience. – R. Krimmer, R. Grimm (eds.). Electronic Voting 2008, Gesellschaft für Informatik. Bonn 2008, pp. 31–43.

^*50 OSCE/ODIHR 2007. Election Assessment Mission Report, Republic of Estonia, Parliamentary Elections, 4 March 2007. Available at http://vvk.ee/public/dok/OSCE_report_EST_2007.pdf (9.10.2011).

^*51 Ü. Madise, T. Maaten (Note 26).

^*52 See Note 7.

^*53 S. Meagher (Note 32), pp. 349–380.

^*54 Ibid., pp. 384–386.

^*55 Th. C. Wingfield, E. Tikk. Frameworks for International Cyber Security: The Cube, the Pyramid, and the Screen. – International Cyber Security. Legal & Policy Proceedings. Tallinn: CCD COE Publications 2010, p. 21.

This article explores the position of referendum provisions in the Constitution and also, more broadly, the subject of direct democracy in the context of Estonian constitutional law and in international comparison.

1. Referendum typology and application in the world

Direct democracy operates in two ways: either by a referendum or a citizen initiated referendum. In literature, various forms of referendum are distinguished between ^*5 :

1)    mandatory constitutional referendum—set out expressis verbis in the constitution of the country as a mechanism required to make decisions related to the sovereignty of the state or amend the constitution; the outcome of a referendum is binding and mandatory for all of the public bodies; used, e.g., in Australia, Denmark, Ireland and Switzerland;

2)   abrogative referendum—a procedure allowing the people to decide on a law already passed in the parliament; it has also been called facultative or veto referendum. For example, in Sweden and Austria such a referendum may be initiated by at least one third of the members of the parliament; in Italy such a referendum may be initiated either by a part of the parliament, citizens (at least 500,000) or regional councils (at least five). In Italy such a referendum may repeal any law (including those already enforced), elsewhere it is used to reject new laws (referendum is held before enforcement). The outcome is binding;

3)   consultative referendum—also called the plebiscide or ad hoc referendum. A referendum may address any issues and is initiated either by the government or parliament. The outcome is non-binding. Such a regulation may be incorporated in the constitution (e.g., in France), be provided for by a special act (e.g., in Canada) or in the decision on the initiation of a referendum itself (e.g., in Great Britain).

As a rule, a referendum is general and direct; it may be held over a certain issue, including a draft act. A referendum should be distinguished from a public poll and a citizen initiated referendum. ^*6 With the latter, a draft act is initiated by the people and requires collection of a certain amount of signatures from the citizens. The number of signatures required to initiate a referendum varies and is prescribed either in the constitution or in a special act. The outcome of a referendum is usually binding, although its legitimacy may sometimes be assessed by a court or the parliament (e.g., in Switzerland, New-Zealand and some states of the US).

Jurisprudent Markku Suksi believes that the provisions containing direct democracy in some form are present in 85 constitutions of the world. ^*7 The table on the application of direct democracy (see Table 1) shows that direct democracy is more existent in European countries. In Europe, Switzerland has the greatest tradition of applying direct democracy—between 1945 and 2006 it held 396 direct democratic referendums. ^*8 In other European countries, direct democracy was widely applied in the enlargement process of the European Union.

Table 1. Application of direct democracy in the continents of the world in 1951−2010

Note: The table is based on data from the Initiative & Referendum Institute (hereinafter referred to as the IRI) Europe. ^*9

To sum up, alongside representative democracy direct democracy in its various forms is being applied worldwide, including in Europe as an active mechanism of democracy.

2. Referendum and citizen initiated referendum in Estonian law

Next let us have a look at the position of referendum clauses in Estonian legal order.

Two sections of the Constitution regulate referendums: §§105 and 106. Section 105 provides that the Riigikogu has the right to submit a draft act or other national issue to a referendum; the decision of the people shall be made by a majority of the participants in the voting. A law which is passed by a referendum shall promptly be proclaimed by the President of the Republic; the decision of the referendum shall be binding on all state institutions. If a draft act which is submitted to a referendum does not receive a majority of votes in favour, the President of the Republic shall declare extraordinary elections to the Riigikogu. Section 106 of the Constitution provides that issues regarding the budget, taxation, financial obligations of the state, ratification and denunciation of international treaties, the declaration or termination of a state of emergency, or national defence shall not be submitted to a referendum. An act adopted by referendum normally has the same legal power as any act adopted by the parliament. The stance of the Constitutional Review Chamber of the Supreme Court is that the Riigikogu may submit an issue to a referendum only where such an issue is in the remit of the Riigikogu and holding a referendum is not prohibited under §106 of the Constitution; for example, deciding on the extraordinary elections to the Riigikogu is not within the powers of the Riigikogu—pursuant to §78 3) of the Constitution such elections are proclaimed by the President of the Republic and therefore the Riigikogu cannot submit that issue to a referendum. ^*10

Such forms of direct democracy as referendum and citizen initiated referendum were also present in the constitutions of the Republic of Estonia before 1940. Section 29 of the constitution adopted by the Constituent Assembly in 1920 gave extensive powers to the people—besides elections of the Riigikogu and the form of a referendum, it also provided for the institute of a citizen initiated referendum. The people, i.e., the active citizenship, even had the right to step up as a legislator under the referendum regime—they could issue, amend or repeal acts (§31).

What was significant was that the active citizenship had exclusive rights to amend the constitution, whether by a citizen initiated referendum ^*11 or by the Riigikogu (§88). ^*12 Section 29 of the 1933 constitution provided that the people exercise the power of state by a referendum, citizen initiated referendum, election of the Riigikogu and election of the Riigivanem (president).

The 1938 constitution relinquished, inter alia, the referendum and kept only the institute of citizen initiated referendum (§35 4)).

Returning to current law, besides the Constitution the issues related to referendum are regulated also by a specific act. Clause 104 5) of the Constitution provides that the Referendum Act belongs to the constitutional laws, i.e., it may be passed and amended only by a majority of the membership of the Riigikogu. The Referendum Act ^*13 (hereinafter referred to as the RA) was adopted by the Riigikogu on 13 March 2002 and entered into force on 6 April 2002. Section 1 of the act provides for the issues already addressed in the Constitution (repeats the provisions of §§105 and 106 of the Constitution). A referendum is free, general, uniform and direct. Voting is secret, each voter has one vote. An Estonian citizen who has attained eighteen years of age by the date of a referendum may participate in the referendum. A person cannot participate in the voting if he or she has been divested of his or her active legal capacity by a court judgment or has been convicted by a court and is serving a sentence in a custodial institution (RA §2). Section 3 of the RA provides for the time of referendum so that a referendum is held not earlier than three months after the passage of a resolution to this effect by the Riigikogu (subsection 1) whilst a referendum is not held at a time when less than ninety days remain until elections to the Riigikogu. A referendum on a draft act to amend the Constitution or on another national issue may be scheduled for a time after the next elections to the Riigikogu; a referendum on another draft Act must not be scheduled for a time after the next elections to the Riigikogu; a referendum may be scheduled for the same day as Riigikogu elections or local government council elections (subsection 3). What is also important is the principle that referendums are not held on the same day on issues which are mutually exclusive or for the passage of acts which are in conflict with each other (subsection 4). Subsection 30 (1) of the RA provides that if a draft act is submitted to a referendum, the title of the draft act or, pursuant to a resolution of the Riigikogu, the text of the draft act, the question “Kas Teie olete seaduseelnõu seadusena vastuvõtmise poolt?” [Are you in favour of passage of the draft Act?] and spaces marked with the possible answers “jah” [yes] and “ei” [no] are entered on the ballot paper. If another national issue is submitted to a referendum, the wording of the issue and spaces marked with the possible answers “jah” [yes] and “ei” [no] are entered on the ballot paper.

A referendum is initiated pursuant to the provisions of Chapter 14 of the Riigikogu Rules of Procedure and Internal Rules Act ^*14 (hereinafter referred to as the RRPIRA). Pursuant to §128 (1) of the RRPIRA a member of the Riigikogu, a faction and a committee have the right to initiate a referendum. ^*15 Pursuant to §129 (6) of the RRPIRA a decision to hold a referendum in order for an act to be passed is taken by the Riigikogu by a final vote on the draft resolution at the third reading and in order for the draft resolution to be passed, a majority of votes in favour is required. In order to submit to a referendum draft acts of the laws listed in §104 (2) of the Constitution, the draft act must have a majority of the votes in favour of the membership of the Riigikogu. If the draft resolution is not adopted, the draft act is also deemed to have been rejected (RRPIRA §129 (7)). In order for the draft resolution on submitting other national issues to a referendum to be passed, a majority of votes in favour is required (RRPIRA §130 (2)). Pursuant to §6 (2) of the RA, the Riigikogu shall not amend or repeal a resolution to hold a referendum. Pursuant to §7 (1) of the RA, a referendum is postponed if a state of emergency or a state of war is declared. In case of a state of emergency, the Riigikogu may postpone the referendum by a resolution (subsection 2). The Riigikogu determines a new date for the referendum within two weeks after the reasons cease to exist, observing the term prescribed in §3 (1) of the RA which shall be calculated from the date on which the resolution of the Riigikogu determining the new referendum date is passed. Subsection 8 (1) of the RA provides that a referendum is not held if:

1)    the Supreme Court repeals the resolution of the Riigikogu concerning submission of a draft act or other national issue to a referendum;

2)   the Riigikogu has not passed a resolution determining a new date for the referendum within the term prescribed in §7 (3);

3)   the time of the referendum is not in accordance with §3 (3) of the RA due to extraordinary Riigikogu elections being called.

In addition to a referendum, it was possible during a limited time period after the adoption of the Constitution in Estonia to call forth a citizen initiated referendum. Second paragraph of §8 of the Constitution of the Republic of Estonia Implementation Act ^*16 provided that the right to initiate amendment of the Constitution during the three years following the adoption of the Constitution by a referendum also rests, by way of public initiative, with not less than ten thousand citizens with the right to vote. However, no law was passed to specify the procedure of a citizen initiated referendum. In 1994, 10,632 citizens of Estonia with the right to vote initiated a draft act to amend the Constitution. The draft act proposed to amend §28 of the Constitution to incorporate a guaranteed pension to everybody according to their labour input, and §56, supplementing it with the provision that the president is to be elected directly by the people. ^*17 The draft act was rejected without a debate by the Riigikogu in the autumn of the same year (with the votes 32 in favour, three against and four undecided). ^*18

The current law of Estonia provides only for a referendum. Also, in addition to a citizen initiated referendum the current law does not provide for the institute of a public poll. A public poll is favoured by the principle of democracy according to which it is essential to take the opinions of the people into consideration—in such a case, the public authorities would have the right to ask non-binding opinions from the people in order to get information about the feelings and stances of the people. The Supreme Court has noted that the term ‘public poll’ is not used in the Constitution. The outcome of a public poll is, unlike the outcome of a referendum, not binding to a state or local government body and is limited just to the clarification of the opinion of the persons entitled to participate in a public poll. ^*19 The non-binding nature of a public poll, however, would be a problem—in terms of politics a public poll would acquire a binding substance and thus a referendum should be preferred to a public poll. ^*20

3. Application of direct democracy in Estonia

A total of seven referendums and one citizen initiated referendum have been held in Estonia’s history (see Table 2). ^*21

Pre-1940, direct democracy was applied in Estonia in five instances. Only two referendums were held during the prolonged Soviet occupation and the restoration of Estonia’s independence—the referendum on independence (3 March 1991) and the referendum on the constitution (28 June 1992) were the major political events of the transitional period and though close in time they took place under totally different circumstances. ^*22

Just one referendum has been held under the current Constitution: On 14 September 2003, the Constitution of the Republic of Estonia Amendment Act ^*23 was adopted by a referendum. It was not possible to ratify the Treaty on accession to the European Union by a referendum as §106 of the Constitution prohibits to submit the issues related to the ratification and denunciation of international treaties to a referendum and there was no desire to amend §106 of the Constitution. It was important to amend the Constitution and hold a referendum as accession to the European Union was a matter of principle in terms of the nationhood and future of Estonia. A referendum was needed to amend the Constitution as the nature of such sections was altered whose amendment is possible only by a referendum; the decision of the Riigikogu on holding a referendum ^*24 was based on §§105, 162, 163, 164 and 167 of the Constitution and §30 (1) of the RA. The ballot paper presented the following question: “Kas Teie olete Euroopa Liiduga ühinemise ja Eesti Vabariigi põhiseaduse täiendamise seaduse vastuvõtmise poolt?” [Are you in favour of accession to the European Union and the passage of the Constitution of the Republic of Estonia Amendment Act?]. ^*25 However, a member of the Riigikogu Igor Gräzin noted that the legally correct question would have been “Kas olete nõus loobuma Eesti praegusest iseseisvusest Eesti astumise korral Euroopa Liitu?” [Are you in favour of relinquishing the current independence of Estonia upon Estonia’s accession to the European Union?] “as the State of Estonia ceases to exist in the form we know albeit the sovereignty of the nation (the right to an armed resurrection, national self-determination, etc. is inalienable”. ^*26 Jurisprudent Lauri Mälksoo finds that irrespective of the form and title of the amendment, in essence the Constitution was changed. Adoption of the Constitution Amendment Act enabled the Riigikogu to ratify the already signed treaty on Estonia’s accession to the European Union. ^*27 Section 1 of the adopted act provided that Estonia may belong to the European Union in accordance with the fundamental principles of the Constitution of the Republic of Estonia and §2 provided that as of Estonia’s accession to the European Union, the Constitution of the Republic of Estonia applies taking account of the rights and obligations arising from the Accession Treaty. The Constitution Amendment Act can be amended only by a referendum, meaning that should the legal nature of the European Union change drastically in the future and become unacceptable to Estonia and contrary to the provisions of the Constitution Amendment Act, the latter can be amended. The Riigikogu has the exclusive powers to organise a referendum, i.e., the Riigikogu must decide whether, for example, an issue concerning the future structure of the European Union is so fundamental that in order to form a stance a referendum is required to be held in Estonia.

The 2003 referendum was disputed quite actively and some disputes were referred to the Supreme Court. ^*28 However, none of the complaints was successful as the voting procedure had not been violated; the accession to the European Union was also disputed by neither the Chancellor of Justice nor the President of the Republic. ^*29

Just one referendum has been held under the current Constitution. However, the Riigikogu has on several occasions attempted to broaden the forms of direct democracy: The 10th and 11th compositions of the Riigikogu conducted a proceeding on a draft act which sought to amend the Constitution in order to legitimise citizen initiated referendums (210SE) and the 11th composition conducted a proceeding on a draft act (477SE) seeking to introduce the institute of a local referendum. ^*30 The Government of the Republic was unsupportive of both draft acts. As regards the legitimisation of citizen initiated referendums, the Government of the Republic noted that the circle of subjects entitled under §103 of the Constitution (inter alia, every member of the Riigikogu has the right to initiate laws) is sufficient to ensure that all the draft acts enjoying considerable support in society will eventually be proceeded by the Riigikogu. There is no need to broaden the options of a legislative initiative ^*31 . As regards the draft Local Referendum Act, the Government of the Republic noted that the draft act was unnecessary and, moreover, adds excess financial liability upon local self-governments. ^*32

Table 2. Application of direct democracy in Estonia in 1923–2003  ^*33

Also, the Treaty establishing a Constitution for Europe (Treaty of Lisbon) caused the question whether or not a referendum should be held to approve the treaty. For instance, the then Chancellor of Justice noted that a referendum was required to approve the treaty, reasoning that: the question of whether Estonia agrees to share its sovereignty with other Member States to that extent should be posed to the bearer of the supreme power—the people. It would also be helpful in that the nation would subsequently embrace the Treaty establishing a Constitution for Europe. It is not insignificant that one of the key objectives of the treaty is to bring the European Union closer and make it more understandable to people and thus the practical added value of a referendum should not be underestimated. ^*34 The group of experts formed within the constitutional committee of the Riigikogu noted that holding a referendum under §6 was out of question as the issue concerned the ratification of a foreign treaty. The Riigikogu considered holding a referendum unnecessary and approved the treaty (in favour 91, against 1). ^*35

4. Problems related to the referendum provisions of the Constitution

David E. Butler and Austin Ranney, who have analysed issues related to direct democracy, note that citizens themselves consider a referendum to be the most authentic and direct way to express their will. Therefore the decisions adopted by a referendum are more legitimate than the decisions based on representative democracy. However, Butler and Ranney note that this does not mean that the decisions made in direct democracy are necessarily wiser or that all the decisions made within a political system should be directly democratic, but rather the question is primarily about the legitimacy of politics. ^*36 The application of direct democracy does not concern just the question of legitimacy but has many other aspects—both positive and negative (see Table 3). ^*37

Table 3. Positive and negative aspects of a referendum

At the same time, the main objective of applying direct democracy—to ensure that the people as the bearer of the supreme power of state can express their will beyond parliamentary elections—does not mean that the political decision-making process becomes uncontrollable or that the fulfilment of the functions of state becomes paralysed. The drafting of the current constitution was guided by the principle that Estonia is a parliamentary republic and the application of direct democracy depends mainly on parliamentary decisions; thus a provision was made for a referendum but not for a citizen initiated referendum. Some jurisprudents have maintained that direct democracy should have been more broadly provided for in the constitution. The lawyers who prepared the report of the Committee for Expert Analysis of the Constitution pointed out that the current constitution was too Riigikogu-centred and that the people were not sufficiently trusted to exercise the powers of state. The State expects acknowledgement from all individuals through the people, but does not sufficiently take the opinions of the people in direct consideration. ^*38 This standpoint was shared by several jurisprudents already while the Constitution was being drafted. Professor Ilmar Rebane noted in his analysis of the draft constitution (wording of 15 December 1991) tabled in the Constitutional Assembly that the opportunities for the people to exercise the supreme power of state are rather negligible and limited just to the election of the Riigikogu and referendums. Exercise of the power of state via public polls was omitted (a provision to that effect was contained in one of the original versions) and the right of citizen initiated referendums was also not provided for. ^*39 Dr. Heinrich Schneider has claimed that the Estonian experience speaks not only about the need to broaden the functions and powers of the people but also about preventing the restriction of existing functions and ensuring their realisations. This, however, was not considered in drafting the 1992 Constitution, empowering the Riigikogu to decide on the holding of a referendum (Constitution §65 (2)). ^*40

There is some conceptual confusion regarding the referendum clauses of the current Constitution: A referendum is in part mandatory (e.g., to amend certain provisions of the Constitution), however, in part it is optional (as in other issues, holding a referendum depends purely on the relevant decisions and choices of the Riigikogu) but notwithstanding the outcome of a referendum is binding upon all public bodies. It may well be that the fourth paragraph of §105 of the Constitution requiring that if a draft act is rejected at a referendum, the Riigikogu should be dismissed curbs the parliament’s enthusiasm to submit any draft acts to a referendum. ^*41 In essence, the fourth paragraph of §105 makes the whole section redundant as there is no parliament that would risk holding a referendum if the failure of the submitted draft act results in extraordinary parliamentary elections. Jurisprudent G. Carcassonne opined in the expert analysis of the Constitution that it is most likely that no referendums will be held. ^*42 However, it should be pointed out that the negative outcome of a referendum in other national issues does not cause extraordinary elections. ^*43

The substantial problems related to the application of direct democracy do not preclude the need to analyse the referendum clauses of the Constitution and discuss, for example, the need to legitimise citizen initiated referendums. Jurisprudent Rait Maruste has highlighted the latter point in recent debates on the amendment of the Constitution. ^*44 The opinion of the Committee for Expert Analysis of the Constitution also proposed to recreate the institute of a citizen initiated referendum in a restricted form, making its outcome politically but not legally binding. ^*45 It may be added here that citizens’ political initiatives were granted recently at the EU level in order to bridge the alienation between the government level and the citizens (so-called deficit of democracy). The Treaty of Lisbon provided for a citizens’ initiative which in substance means citizen initiated referendums. Namely, the citizens who are nationals of the European Union Member States may, under Article 11 (4) of the Treaty establishing the European Union, take the initiative of inviting the European Commission, within the framework of its powers, to submit any appropriate proposal on matters where citizens consider that a legal act of the Union is required for the purpose of implementing the Treaties. ^*46 One million citizens from at least one quarter of the EU Member States can invite the European Commission to bring forward proposals for legal acts in areas where the Commission has the power to do so. The organisers of a citizens’ initiative, a citizens’ committee composed of at least seven EU citizens who are resident in at least seven different Member States, will have one year to collect the necessary statements of support. The number of statements of support has to be certified by the competent authorities in the Member States. The Commission will then have three months to examine the initiative and decide how to act on it. European citizens’ initiatives can be launched as from 1 April 2012. The procedure and conditions of the European citizens’ initiative have been set out in Regulation No. 211/2011 of the European Parliament and the Council (16 February 2011). ^*47 In addition to the already realised initiative, a proposal has been tabled to provide for a pan-European referendum—important EU agreements would no longer be debated in individual Member States (i.e., the states which allow referendums) but in the European Union as a whole and all citizens of the European Union can participate. ^*48

5. Conclusions

The principle of people’s sovereignty mainly places emphasis on the people as the bearer of the power of state and its role as a source of legitimacy. The people express their will through elections and various forms of direct democracy; direct democracy is being increasingly applied throughout the world—the most recent instances are related to the reform process of the European Union. The provisions of direct democracy are present in the constitutions of many countries. Referendums are addressed in §§105 and 106 of the current Constitution of the Republic of Estonia, and also in the Referendum Act. A referendum, as set out in the Constitution, covers both the referendums to amend the Constitution, referendums on other draft acts and referendums on other national issues. An act adopted by referendum normally has the same legal power as any act adopted by the parliament. Just one referendum has been held in Estonia under the current Constitution, although draft acts have on several occasions been tabled in the Riigikogu seeking to broaden the forms of direct democracy (local referendum, citizen initiated referendum). There have also been developments in European Union legislation (legitimisation of a pan-European citizens’ initiative). Such a development dynamic justifies the need to analyse the justifiability of the application of direct democracy in Estonian legal regime.

Notes:

^*1 For example, the referendums on the Treaty establishing a Constitution for Europe in France and Holland in 2005; the referendums held in 2008 and 2009 in Ireland on the ratification of the Treaty of Lisbon. There were 24 referendums on the European Union treaties and accession to the EU between 1998 and 2008.

^*2 Here the following sources can be cited: S. Binzer Hobold. Europe in Question: Referendums on European Integration. Oxford: Oxford University Press 2009; M. Setälä, T. Schiller. Referendums and Representative Democracy: Responsiveness, Accountability and Deliberation. Routledge/ECPR Studies in European Political Science 2009; Z. T. Pallinger, B. Kaufmann, W. Marxer, T. Schiller. Direct Democracy in Europe. Developments and Prospects. Wiesbaden: VS Verlag für Sozialwissenschaften/GWV Fachverlage GmbH 2007; M. Qvortrup. A Comparative Study of Referendums: Government by the People. Manchester University Press 2005; T. Schiller. Direkte Demokratie. Forschung und Perspektiven. Westdeutscher Verlag 2003. There is also an abundance of other articles. As regards Estonia, there is a significant monograph by Vallo Olle. Kohaliku omavalitsuse teostamine vahetu demokraatia vormis: kohalik rahvaalgatus ja rahvahääletus (Exercising Local Government in the Form of Direct Democracy: Local Citizens’ Initiatives and Referendums). Tartu Ülikooli Kirjastus 2002 (in Estonian).

^*3 M. Suksi. Bringing in the People. A Comparison of Constitutional Forms and Practices of the Referendum. Martinus Nijhoff Publishers, Kluwer Academic Publishers 1993, pp. 16–28.

^*4 Eesti Vabariigi põhiseadus. – RT 1992, 26, 349; RT I 2007, 33, 210 (in Estonian).

^*5 L. LeDuc. The Politics of Direct Democracy: referendums in global perspective. Broadview Press 2003, p. 39.

^*6 Committee for Expert Analysis of the Constitution, final report. Kommentaarid põhiseaduse 3. peatükile „Rahvas“ (Comments on Chapter 3 ‘People’ of the Constitution). Available at http://www.just.ee/10726 (19.3.2011) (in Estonian).

^*7 M. Suksi (Note 3), p. 137.

^*8 M. Gallagher. Elections and Referendums. – D. Caramani (ed.). Comparative Politics. Oxford: Oxford University Press 2008, p. 257.

^*9 The IRI Guidebook to Direct Democracy in Switzerland and beyond 2010. Köniz: Initiative & Referendum Institute Europe, Ast & Jakob, Vetsch AG, 2010, p. 208.

^*10 CRCSCd, 30.10.2009, 3-4-1-20-09.

^*11 In comparison, it should be pointed out that currently, under §161, the right to initiate amendment of the Constitution rests with not less than one-fifth of the membership of the Riigikogu and with the President of the Republic (subsection 1).

^*12 H. Schneider. Riigiteoreetilisi probleeme: miks nad on tekkinud ja kuidas neid lahendada? (Theoretical Constitutional Problems: Why Have They Emerged and How to Solve Them?) – Juridica 1998/7, p. 355 (in Estonian).

^*13 Rahvahääletuse seadus. – RT I 2002, 30, 176; 2009, 29, 175 (in Estonian). The act repealed the Referendum Act which had been in force since 19 June 1994 (RT I 1994, 41, 659; 2000, 95, 610).

^*14 Riigikogu kodu- ja töökorra seadus. – RT I 2003, 24, 148; 2010, 28, 144 (in Estonian).

^*15 Pursuant to §40 (1) of the RRPIRA a faction may be formed by and must comprise not less than five members of the Riigikogu who are elected from a list of candidates of the same political party. Pursuant to §17 of the RRPIRA the Riigikogu has standing committees, select committees, committees of investigation and study committees. Pursuant to §18 (2) of the RRPIRA committees prepare draft legislation for deliberation by the plenary assembly of the Riigikogu, exercise supervision over the exercise of executive power within their particular field and perform other functions assigned to the committees by law or by a resolution of the Riigikogu. Standing committees are formed under law; select committees are formed by a resolution of the Riigikogu which sets out the composition, including an alternate member to substitute for each committee member, functions and procedure for reporting on the activities of the committee (RRPIRA §19 (2)); committees of investigation and study are formed by a resolution of the Riigikogu which sets out the composition, including an alternate member to substitute for each committee member, functions and term of authority of the committee (RRPIRA §20 (2) and §21 (2)).

^*16 Põhiseaduse rakendamise seadus. – RT I 1992, 26, 350 (in Estonian).

^*17 Pursuant to §79 of the Constitution, the President of the Republic is elected by the Riigikogu or an electoral body. The electoral body is comprised of members of the Riigikogu and representatives of the local government councils. Each local government council elects at least one representative to the electoral body, who must be an Estonian citizen. The specific procedure for the election of the President of the Republic is provided by the President of the Republic Election Act (Vabariigi Presidendi valimise seadus. – RT I 1996, 30, 595; RT I, 16.11.2010, 9 (in Estonian)).

^*18 A. Leps. Eesti põhiseaduse muutmisest (On the Amendment of the Estonian Constitution). – Akadeemia 1999/3, p. 461 (in Estonian).

^*19 CRCSCd, No. III-4/1-2/93, 11.8.1993. – RT I 1993, 59, 841 (in Estonian).

^*20 T. Annus. Riigiõigus (Constitutional Law). AS Juura 2006, p. 56 (in Estonian).

^*21 The following sources were used in the table: Eesti Entsüklopeedia (Encyclopaedia of Estonia). Vol. 7. Eesti Entsüklopeediakirjastus 1994, article “Rahvahääletus” (Public Poll), p. 674; T. Raitviir. Eesti üleminekuperioodi valimiste (1989–1993) võrdlev uurimine (Elections in Estonia during the Transition Period: A Comparative Study (1989–1993). Institute of International and Social Studies. Tallinn: Teaduste Akadeemia Kirjastus 1996, pp. 165–181; informative Web site of the Electoral Committee of the Republic, available at http://www.vvk.ee (25.3.2011).

^*22 T. Raitviir (Note 21); J. Ruus. Estonia. – Direct Democracy: The Eastern and Central European Experience. A. Auer, M. Butzer (ed.). Ashgate 2001, pp. 51–55.

^*23 Eesti Vabariigi põhiseaduse täiendamise seadus. – RT I 2003, 64, 429 (in Estonian).

^*24 Decision of the Riigikogu “Rahvahääletuse korraldamine Euroopa Liiduga ühinemise ja Eesti Vabariigi Põhiseaduse täiendamise küsimuses“ (Holding a Referendum on Accession to the European Union and Amendment of the Constitution of the Republic of Estonia). Adopted and entered into force 18.12.2002. – RT I 2003, 60 (in Estonian).

^*25 J. Laffranque. Euroopa Liidu õigussüsteem ja Eesti õiguse koht selles (European Union Legal System and Estonia’s Place in It). AS Juura 2006, pp. 69–71 (in Estonian); A. Albi. EU Enlargement and the Constitutions of Central and Eastern Europe. Cambridge 2005, pp. 138–162.

^*26 I. Gräzin. Eesti põhiseadus ja Euroopa Liidu referendum (Estonian Constitution and European Union referendum). – Riigikogu Toimetised 2002 (5). Available at http://www.riigikogu.ee/rito/index.php?id=11891&highlight=Gräzin&op=archive2 (25.3.2011) (in Estonian).

^*27 L. Mälksoo. Eesti suveräänsus 1988–2008. Iganenud või igavene? – Tekste kaasaegsest suveräänsusest (Estonia’s Sovereignty 1998−2008. Obsolete or Infinite?—Texts on Modern Sovereignty). H. Kalmo, M. Luts-Sootak (ed.). Tartu Ülikooli Kirjastus 2010, p. 141 (in Estonian).

^*28 CRCSCr, 12.9.2003, 3-4-1-10-03; CRCSCr, 24.9.2003, 3-4-1-11-03; CRCSCd, 29.9.2001, 3-4-1-12-03; CRCSCr, 30.9.2003, 3-4-1-15-03; CRCSCd, 2.10.2003, 3-4-1-13-03; CRCSCd, 2.10.2003, 3-4-1-16-03; CRCSCd, 3.10.2003, 3-4-1-17-03; CRCSCr, 10.10.2003, 3-4-1-20-03; CRCSCd, 14.10.2003, 3-4-1-19-03; CRCSCr, 17.10.2003, 3-4-1-21-03; CRCSCr, 26.2.2004, 3-4-1-6-04.

^*29 J. Laffranque (Note 26), pp. 72–73.

^*30 E. Liivik. Legitimacy through Direct Democracy in the EU Member State: Direct Democratic Initiatives in the Estonian Parliament. The Proceedings of the Institute for European Studies. – Journal of Tallinn University of Technology 2010/8, pp. 91–93.

^*31 Opinion of the Republic of the Government on draft act 210SE, 4.2008, No. 10.2-3/3174. Available at http://www.riigikogu.ee/index.php?page=en_vaade&op=ems&eid=254354 (25.3.2011) (in Estonian); draft act No. 210 SE II, initiated 28.2.2008. Available at http://www.riigikogu.ee/index.php?page=en_vaade&op=ems&eid=254354 (25.3.2011) (in Estonian).

^*32 Opinion of the Government of the Republic on draft act No. 477SE, 4.6.2009, No. 10.2-3/5487. Available at http://www.riigikogu.ee/?page=en_vaade&op=ems&eid=611481 (25.3.2011) (in Estonian); draft act No. 477 SE II-1. Available at http://www.riigikogu.ee/?page=en_vaade&op=ems&eid=611481 (25.3.2011) (in Estonian).

^*33 The League of Veterans (in Estonian vabadussõjalased or vapsid)—in actuality the League of Veterans of the Estonian War of Independence was founded in 1929. It was headed by General Andres Larka (1879–1943) and lawyer Artur Sirk (1900–1937). The veterans sought, among other things, to introduce the institute of a president in the constitution. In 1934, the leaders of the veterans were imprisoned upon the initiative of the then Prime Minister Konstantin Päts. See A. Kasekamp. Vaps. – World Fascism. A Historical Encyclopedia. C. P. Blamires, P. Jackson (eds.). ABC-CLIO, Inc California 2006, Vol. 2: L-Z, p. 696.

^*34 A. Jõks. Rahvahääletust ei ole vaja karta (No Need to Fear the Referendum). – Postimees, 28.4.2005. Available at http://www.postimees.ee/280405/esileht/arvamus/164191_1.php (25.3.2011) (in Estonian).

^*35 Lissaboni lepingu, millega muudetakse Euroopa Liidu lepingut ja Euroopa Ühenduse asutamise lepingut, ratifitseerimise seadus (Act Ratifying the Treaty of Lisbon amending the Treaty on European Union and the Treaty establishing the European Community). – RT II 2008, 14, 36 (in Estonian). Information about the proceeding of the draft act available at http://www.riigikogu.ee/?page=en_vaade&op=ems&eid=240477&u=20110408144446 (29.3.2011) (in Estonian).

^*36 D. Butler, A. Ranney. Referendums Around the World: The Growing Use of Direct Democracy. AEI Press 1994, pp. 14–15.

^*37 M. Gallagher (Note 8), p. 256.

^*38 Committee for Expert Analysis of the Constitution (Note 6).

^*39 Põhiseadus ja põhiseaduse assamblee. Koguteos (Constitution and the Constitutional Assembly. The Complete Work). Eesti Vabariigi Justiitsministeerium; Juura, Õigusteabe AS 1997, p. 1290 (in Estonian).

^*40 H. Schneider (Note 12).

^*41 T. Annus (Note 20), pp. 54–55.

^*42 K. Merusk, R. Narits. Eesti konstitutsiooniõigusest (On Estonian Constitutional Law). AS Juura 1998, pp. 39–40 (in Estonian).

^*43 M. Suksi (Note 3), p. 121.

^*44 R. Maruste. Miks põhiseadus vajab korrastamist (Why Constitution Needs Structuring). – Riigikogu Toimetised 2010 (22), p. 33 (in Estonian).

^*45 Committee for Expert Analysis of the Constitution (Note 6).

^*46 European citizens’ initiative. Available at http://ec.europa.eu/dgs/secretariat_general/citizens_initiative/index_-et.htm (25.3.2011).

^*47 OJ L 65, 11.3.2011, p. 1. Available at http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2011:065:0001:0022:ET:PDF (4.7.2011).

^*48 R. Rose. The case for pan-European referendums. Europe’s World, Spring 2011. Available at http://www.europesworld.org/NewEnglish/Home_old/Article/tabid/191-/ArticleType/articleview/ArticleID/21795/language/en-US/Default.aspx (25.6.2011).

This article analyses the compliance of the amendments made in Estonian health insurance and pension insurance with international requirements and the principles set forth in the Constitution.

1. The right to social security as a fundamental right

The right to social security has been recognised as one of the fundamental rights. In Estonia, the rights related to social security are regulated by §28 of the Constitution. ^*4 Pursuant to subsection 2 of that section, a person has the right to state assistance in the event of old age, incapacity for work, loss of someone on whom he depends, and need. With that provision, the Estonian state has assumed the responsibility to ensure the payment of social benefits necessary for an individual in the above-mentioned circumstances.

Pursuant to §28 (1) of the Constitution of the Republic of Estonia, everybody is entitled to health protection. This provision has sparked discussions ^*5 of whether the right to health protection would automatically entail the right to health insurance, or if the right provided by the Constitution means only a person’s right to receive state-warranted health care in cases of emergency (e.g., accident), while routine treatment is not covered by §28 (1) of the Constitution. ^*6

The Constitution presumes that the state has to ensure a decent existence for everyone. Social rights, therefore, consist in a person’s right to receive benefits from the state. At the same time, the state is obliged to take active steps to ensure these benefits. ^*7 It is important to keep in mind that fundamental social rights (incl. the right to social security) are so-called soft rights, meaning that the state can only provide benefits to the extent for which it has the resources. ^*8 As a result, the standards that states follow may indeed be identical while the actual economic value (purchasing power) of the benefits provided through social security varies greatly by country.

In the past year, several fundamental reforms intended to ensure the sustainability of the national social security system have been carried out in Estonia. Important changes have been made in pension insurance. With regard to this field of insurance, the retirement age was raised and state instalments in the second pension pillar were suspended. ^*9 International opinion on the sustainability of the Estonian health insurance system is negative, since the financing founded on the principle of solidarity used thus far is not sustainable and is hence in need of reform. ^*10

2. International legal instruments in the field of social security that are binding on Estonia

Various requirements have been set for social security at the international level. The main organisations enforcing these requirements are the International Labour Organization (hereinafter referred to as the ILO) and the Council of Europe. ^*11 Two documents of the Council of Europe in the area of social security—the European Social Charter ^*12 and the European Code of Social Security ^*13 —are binding on Estonia. These two documents jointly prescribe the development and functioning of health insurance and pension insurance systems to Member States.

Estonia has not ratified ILO conventions for regulation of the field of social security. Nor is there currently any need for Estonia to ratify these conventions, because, bound by the documents approved by the Council of Europe, Estonia has met all of the international requirements set for social security systems.

3. Changes in pension insurance

3.1. Raising the retirement age

Various countries have begun to seek options for reforming the social security systems effective in their jurisdiction. A major part of the social security systems to be reformed is pension insurance. ^*14 One of the main reasons for reforming pension insurance is the ageing of the population.

Both the European Code of Social Security and ILO Convention 128 set the age of retirement at 65 but permit raising the retirement age. ^*15 The current age for old-age pension in Estonia is 63, but the Riigikogu recently decided to raise the retirement age to 65. ^*16 Behind this decision is the demographic situation of Estonia, which clearly points to the inevitability of an increased retirement age. ^*17

The demographic situation leaves the state a choice between, on the one hand, leaving the age limit required for receiving old-age pension at the same level and raising the relevant taxes to the levels needed for financing pension insurance and, on the other hand, raising the retirement age and thereby in the longer term ensuring the functioning of the pension system and a fundamental right—the right to a decent standard of living and state assistance in the event of old age. Although it may seem at first sight that raising or not raising the retirement age is more of a demographic issue, it also has legal consequences. The retirement age is an important condition by which receipt of a pension can be ensured also for future generations. Pursuant to the State Pension Insurance Act ^*18 , the pension is a monthly monetary payment based on the principle of solidarity (§2). According to that principle, the older generation presumes that the younger generations pay for the pensions of the older. Not raising the retirement age creates a situation wherein the younger generations are no longer able to finance pensions and the principle of solidarity provided by the State Pension Insurance Act is violated as well. That, in turn, entails violation of §28 (2) of the Constitution, pursuant to which a person is entitled to state assistance in old age. The non-functioning of the principle of solidarity also follows from the fact that the younger generation supposed to enter the labour market are doing so later and later in life. Today’s pensioners typically started their working life at the age of 14–15, while nowadays professional careers begin at 20 at the earliest, and in some cases even later. ^*19 It can therefore be said that professional life has shortened by 10 years when compared to that of today’s pensioners. Accordingly, the shorter life span that has emerged as a main argument in current discussions ^*20 is not the only serious argument for not raising the retirement age.

Insofar as international standards retain the option of raising the retirement age, each country is entitled to do so. In fact, it can be argued that not raising the retirement age might entail legal problems in the context of international documents, as well as with respect to the right to assistance in old age prescribed by the Estonian Constitution. It follows from the above that raising the retirement age does not violate international standards or the right to state assistance due to old age prescribed by §28 (2) of the Constitution.

3.2. Enforcing additional pension schemes

Another major aspect of reform to pension insurance, in addition to that of the ageing population, is states’ wish to make the population itself take responsibility for its pensions. These ideas have served as the basis for the Estonian pension reform ^*21 , whose components are the mandatory funded pension and voluntary funded pension.

The mandatory funded pension has to provide old-age pensioners with the chance to receive an old-age pension. Pursuant to the Funded Pensions Act ^*22 §40 (1), a person is entitled to receive a mandatory funded pension once he has reached the old-age pension age. In keeping with this requirement, the legislator has provided that if a person receives a state old-age pension, he shall also receive a mandatory funded pension, which jointly with the state old-age pension should guarantee him a pension equal to at least half of his former wages. ^*23

An important aspect of the functioning of a mandatory funded pension system is the financing of that system. The financing of the Estonian pension system is divided between the insured person and the state. The insured person thus pays 2% of his wages into the pension fund and the state adds to that another 4% from the social tax received on the behalf of said person. ^*24 Such a financing scheme means in broad terms that 6% of labourers’ earnings goes toward financing the funded pension.

The institution of the mandatory funded pension as such may raise the question of whether it is possible, constitutionally, to enforce a pension insurance scheme that is based on private insurance ^*25 and for which the state is not directly responsible. ^*26 Pursuant to the Constitution, a person shall have the right to turn to the state for assistance if the need is due to old age (§28 (2)). The Constitution does not forbid the creation of additional mandatory social security schemes in a situation where a state insurance mechanism has been ensured. Given that the State Pension Insurance Act has been adopted and functions in Estonia, the minimum requirement of the creation and existence of a state pension system, as set forth by the Constitution, has been fulfilled.

Insofar as the mandatory funded pension is a supplementary scheme created by the state, that scheme need not be administered by the state even by international standards. According to the requirements of both the European Code of Social Security ^*27 and the corresponding ILO convention, pension systems should be administered by public institutions. However, neither the European Code of Social Security nor ILO Convention 128 prohibits the creation of supplementary private pension schemes. The mandatory funded pension is a pension type that is primarily intended to supplement and support state pension insurance. By means of this system, the state is able to show that it has fulfilled its duty to maintain the pension system at the required level, but, at the same time, no international requirements have been set for mandatory funded pension systems.

The state, therefore, is free to create additional social security mechanisms and delegate the administration of such supplementary social security schemes to private persons.

The supplementary pension scheme is, however, subject to the provision for protection of property that follows from the Constitution. Pursuant to §32 of the Constitution, everybody’s property is inviolable. Property may be expropriated in the public interest for a fair price in cases provided by law. Although this provision is not directly linked to social security benefits, one’s years of pensionable service for social security (with the right of claim to future pension) have, in a sense, been regarded as property of the insured person. ^*28 Accordingly, these years of pensionable service cannot be taken away from the insured person or reduced without fair compensation being offered in return.

In order to reduce the budgetary expenses of the state and taking into account the unstable economic situation, the Estonian state decided in 2009 to no longer contribute to the mandatory funded pension. ^*29 That decision was based on the need to prevent the under-financing of state pensions and to avoid a situation in which the state is unable to fulfil its Constitutional obligation to ensure protection to people in their old age. Such a decision, with which the state unilaterally withdraws from financing the social security scheme, can legally be considered violation of an agreement between the state and an individual. At the same time, one may ask whether this behaviour by the state was justified or not.

Here it must be pointed out that the decision of the Estonian state to stop making payments to the mandatory funded pension fund is of a temporary nature; in other words, the payment of state contributions has not been terminated but was suspended until 1 January 2011. The state has also provided for a certain compensatory mechanism, according to which state contributions will be restored step by step, then, in a later phase, the state will compensate for the uncollected amounts at a higher percentage rate. While the rate normally is 4%, it will become 6% after some time. ^*30 So it seems, at first sight, that no expropriation of property has occurred in the mandatory funded pension system.

However, expropriation of property may be seen as occurring in the mandatory funded pension system if the state were to fail, for whatever reason, to resume its payments into the mandatory funded pension fund after 1 January 2012. In that case, persons who have voluntarily continued to make contributions to the mandatory funded pension fund would be entitled to file a claim against the state, since the state, by promising to resume making contributions to the mandatory funded pension fund at a certain time and then failing to do so, would be breaching its promise and violating the legal expectation of future pensioners to receive a supplementary pension.

4. The right to health protection and health insurance

Health insurance benefits are divided into two kinds. On the one hand, health insurance is used for providing health services; at the same time, the health insurance system is intended to provide benefits in the event of temporary incapacity for work.

4.1. Payment of benefits in cases of temporary incapacity for work

Although the Constitution of the Republic of Estonia does not prescribe that a person must be provided with assistance through the health insurance system in cases of illness, §28 (2) of the Constitution provides that a person is entitled to state assistance in the event of need. Pursuant to the notes added to the Constitution, need primarily signifies a person’s right to social assistance here ^*31 , but at the same time the Constitution leaves open the possibility of regarding as need also other situations wherein the person loses his income and as a result must receive social‑security‑related benefits.

Up to 30 June 2009, benefits responding to temporary incapacity for work were paid through the Estonian Health Insurance Fund. ^*32 From 1 July 2009, the payment of benefits for those with temporary incapacity for work has been distributed between the employer and the Health Insurance Fund: for the first eight days of sickness, liability for maintaining income has been transferred from the Health Insurance Fund to the employer. With this amendment, the labourer is deprived of health insurance benefits for the first three days; for sickness days 4–8 the employer shall pay the labourer a benefit that equals 70% of the labourer’s average wages, and on day 9 the Health Insurance Fund takes over the obligation to pay health insurance benefit. The reason behind this amendment is the determination to reduce the expenses of the Health Insurance Fund related to the amounts of benefits paid for temporary incapacity for work. However, the matter of payment of health insurance benefits for days 4–8 on the employer’s account has, in the main, so far been only partially resolved. ^*33

Pursuant to the Occupational Health and Safety Act ^*34 , the employer is given the same rights as the Health Insurance Fund in some cases (§12²), but the realisation of these rights remains unclear. One problem, for instance, is the right of the Health Insurance Fund to refuse to pay the benefit for those with temporary incapacity for work if the temporary incapacity for work developed while the worker was in an intoxicated state. ^*35 Since the Health Insurance Fund has been given the right not to pay the benefit in question in that case, the same right is provided to the employer. The employer may, therefore, refuse to pay the benefit if the certificate of temporary incapacity for work ^*36 states that the insured person was in a state of intoxication. If the insured person nevertheless wishes to receive money in compensation, he needs to turn to the county court (civil court) to settle the dispute. ^*37

The right of the employer to check the legality of issue of a certificate of temporary incapacity for work might also prove problematic. Under §60 (2) of the Health Insurance Act, the Health Insurance Fund is entitled to delay the payment of benefits related to temporary incapacity for work by up to 30 days if there is suspicion that the certificate of incapacity for work may not have been issued properly. The Health Insurance Fund has been given this right in order to prevent abuse of the Health Insurance Fund’s resources, but it remains unclear to what extent an employer might or could exercise this right and delay the payment of health insurance benefits.

The basis for calculation of the benefit differs as well—the Health Insurance Fund calculates the benefit on the basis of the previous year’s income taxed with social tax, whereas the employer must pay a benefit to the labourer at the rate of 70% of his average wages. ^*38

Obliging the employer to pay benefits to a labourer at a time when the labourer is unable to fulfil his duties is not by nature in contradiction with the Constitution or with social security in general, but the duties and rights entailed with respect to it must be specified. At present, it can be said that the situation analysed above does not provide the labourer with a sense of security about the extent to which the employer can refuse to pay the benefit for sickness days 4–8, and the employer’s situation is not clear either, because the employer’s authority in the payment of this benefit has not been explicitly defined. ^*39

4.2. Financing of health insurance

In order for the right to health protection to be ensured for every person, it is important that the financing of health insurance be ensured as well. In the Estonian health insurance system, those covered by health insurance include persons who pay health insurance contributions or who have been equated with insured persons for social reasons. The Estonian health insurance scheme has been founded on financing that is based on the principle of solidarity. In accordance with that principle, the health insurance scheme is financed only from the sums collected from insured persons. In Estonia, as in other Member States of the European Union, the percentage of elderly people is on the rise. ^*40 Because of that trend, the number of people who are able to finance a social security system based on the principle of solidarity is growing smaller. In the longer term, this will lead to a situation wherein Estonia is unable to provide health insurance because there are not enough people who are able to finance this system. As a result, the Constitutional requirement that each person have access to health protection will not be satisfied. Although the Constitution does not specify that it is the state, in particular, that needs to ensure health protection and in no other form than health insurance, the state’s obligation to guarantee health services remains. A study published in March 2010 clearly indicated that the current Estonian health insurance financing scheme is not sustainable and also endangers the principle mentioned in §28 (1) of the Constitution. ^*41 In order to find a solution to this problem, the ‘future concept’ ^*42 prescribes four distinct approaches:

1)    expanding the revenue base of the public sector and bringing additional revenues to the health system—i.e., state contribution to the providing of health insurance must increase;

2)   changing the scope of insurance cover with respect to the right to health insurance, the range of benefits, and elements of cost-sharing (application of this requires reduction of the number of persons equated with insured persons, as well as an increase in cost‑sharing by the insured person ^*43 );

3)   making the management of budgetary resources more efficient;

4)   strengthening the strategic management of the health sector.

4.3. Ensuring health insurance for everyone

Subsection 28 (1) does not prescribe that health insurance should extend to all persons. According to Health Insurance Fund statistics ^*44 , 1,216,000 residents of Estonia were covered by health insurance as of 31.12.2010. Given that Estonia’s population totals 1.34 million people, this means that about 80,000 people in Estonia currently have no health insurance cover. Accordingly, the question of how and on what conditions these 80,000 people may exercise the right to health protection as well has become the main concern of the Estonian state. In the current discussion ^*45 , there is no clear consensus that these 80,000 people should gain full protection in the health insurance system, but an opportunity to ensure at least first-contact care, or access to family physicians, is being sought. Since the Estonian health insurance system is based on the Bismarck model, the system includes from the start the principle that health insurance is provided only to those who finance that system. As mentioned above, this system also benefits certain persons who do not finance the system themselves. At the same time, Estonia has not made a clear decision that the right to health care should be universal and that all persons in Estonia should be able to realise that right.

5. Conclusions

Estonia has managed to create a social security system that covers most of the social risks included in the European Code of Social Security and ILO Convention 128. As for the only risk still not covered, Estonia has thus far not managed to create a system for occupational accident insurance and occupational disease insurance.

The Estonian Constitution also establishes the state’s duty to ensure state assistance in cases of old age, incapacity for work, or loss of a provider of livelihood. At the same time, the Constitution provides a person’s right to health protection. Even though Estonia is bound by the requirements of the European Code of Social Security and the European Social Charter, there are still situations wherein the state needs to make radical changes in the social security system, both for practical reasons and due to economic situation. Although such changes may not go against international requirements, they might be in violation of the Constitution (e.g., §28 (1) of the Constitution’s right to health protection or the Constitution’s §32’s inviolability of property). As a result, we can say today that the changes made for the purpose of achieving economic welfare may entail a danger of direct violation of the social-security-related requirements guaranteed to members of the public.

Notes:

^*1 See analysis Responding to the challenge of financial sustainability in Estonia’s health system. WHO 2010. Available at http://www.haigekassa.ee/uploads/userfiles/E93542.pdf (9.10.2011).

^*2 Riikliku pensionikindlustuse seaduse muutmise ja sellega seonduvalt teiste seaduste muutmise seadus (Act Amending the State Pension Insurance Act and Other Related Acts), §7 (1). – RT I 2010, 18, 97 (in Estonian).

^*3 See, e.g., ‘Valitsus püüab eripensione kaotada järk järgult’ (Government Tries to Abolish Special Pensions Gradually). – Postimees, 18.5.2010. Available at http://www.postimees.ee (in Estonian); ‘Valitsus tahab eripensionid kaotada’ (Government Set to Abolish Special Pensions). – Postimees.ee, 28.4.2011. Available at www.reporter.ee/2011/04/28 (in Estonian).

^*4 Eesti Vabariigi põhiseadus. – RT 1992, 26, 349; RT I 2007, 33, 210 (in Estonian).

^*5 See M. Malvet. Õigus sotsiaalkindlustusele kui põhiõigus. Magistritöö (Right to Social Security as a Fundamental Right. Master’s Thesis). Tartu 2006, pp. 40–41 (in Estonian).

^*6 T. Annus. A. Nõmper. The Right to Health Protection in the Estonian Constitution. – Juridica International 2002 (VII), pp. 122–35.

^*7 See: Põhiseadus: kommenteeritud väljaanne (Constitution: Commented Edition). 2nd edition. Tallinn 2008, §28 comment 1 (in Estonian). The duties prescribed to the state have been provided by the European Social Charter. Article 12 of the European Social Charter thus provides that member states shall create or maintain a social security system, as well as ensure the conformity of the social security system to the conditions of the European Code of Social Security.

^*8 Põhiseadus: kommenteeritud väljaanne (Note 7), §28 comment 3. See also The International Covenant on Economic, Social and Cultural Rights, Article 2. – RT II 1993, 10/11, 13.

^*9 From this year (1.1.2011), pension contributions to the second pillar have partially been restored. Here it is relevant to give a brief description of the Estonian pension system. According to the proposal made by the Estonian pension reform committee, Estonia’s pension system comprises so-called three pillars: 1st pillar—state pension system, which is backed by the state. The second pension pillar is the mandatory funded pension. Every person of working age in Estonia is required to subscribe to the second pension pillar. The second pension pillar is administered by various private pension funds. Pension funds are a subset of investment funds. The third pillar of the pension reform is the voluntary funded pension. With this pillar, it is up to the person to decide if he wants to insure himself in this pillar or not. See http://www.pensionikeskus.ee/?lang=en.

^*10 Responding to the challenge of financial sustainability in Estonia’s health system. WHO 2010. Available at http://www.haigekassa.ee/uploads/userfiles/E93542.pdf (9.10.2011).

^*11 F. Pennings. An overview of interpretation issues of international social security standards. – F. Pennings. International Social security Standards. Intersentia 2007, pp. 1–30.

^*12 European Social Charter, revised version. – RT II 2000, 15, 93.

^*13 European Code of Social Security. – RT II 2004, 6, 17.

^*14 See, e.g. T. Sakellaropoulus, M. Angelaki. The Politics of Pension Reform in South European Welfare States. – J. van Langendonck. The Right to Social Security. Intersentia 2007, pp. 121–139.

^*15 European Code of Social Security, Article 26; ILO Convention 128, Article 15.

^*16 The retirement age of 65 years will enter into force on 1.1.2017.

^*17 In political discussions, the demographic (rhetorical) objection that if the retirement age is to be raised, the life-span of persons must be increased as well, has been raised at once (See M. Lauristin. Esmalt tuleb tõsta inimeste eluiga (At First, People’s Life-span Must be Increased). – Postimees, 23.3.2010. Available at http://www.postimees.ee (in Estonian). At the same time, it must be kept in mind that when the part of population of working age decreases, the revenue base of the state is inevitably diminished. As a result, the system of social benefits is also no longer sustainable.

^*18 Riikliku pensionikindlustuse seadus. – RT I 2001, 100, 648: RT I 2010, 41, 240 (in Estonian).

^*19 B. Contini. Youth Employment in Europe: Institutions and Social Capital Explain. Better than Mainstream Economics 2010, p. 6, etc. Available at http://ftp.iza.org/dp4718.pdf.

^*20 H. Pevkur: Pensioniea tõstmine kaitseb (Raising the Retirement Age Protects). – Postimees, 25.3.2010. Available at http://www.postimees.ee (in Estonian); S. Oviir. Pensioniea tõstmine karistab (Raising the Retirement Age Penalises). – Postimees, 25.3.2010. Available at http://www.postimees.ee (in Estonian); L. Leppik. Mida teha pensionieaga (What to Do with the Retirement Age). – Eesti Päevaleht, 26.2.2009. Available at http://www.epl.ee (in Estonian).

^*21 The Estonian pension reform was initiated in 1999, when the social tax was transformed into an individually registered tax. The social tax is a tax by which both health insurance and pension insurance are financed. The state pension system became insurance-based as well. The last stage of the pension reform took place in 2002, when regulation on mandatory funded pension as well as voluntary funded pension entered into force. See Monetary Developments & Policy Survey: Estonian Pension Reform and its Implications on Balanced Budget. Bank of Estonia 2003. Available at http://www.eestipank.info/pub/en/dokumendid/publikatsioonid/see- riad/ylevaade/_2003_01/_2003_01/_2mar_taust.pdf?ok=1.

^*22 Kogumispensionide seadus. – RT I 2004, 37, 252; RT I 24.3.2011, 1 (in Estonian).

^*23 For information on mandatory funded pension, see http://www.pensionikeskus.ee.

^*24 The percentage of social tax collected into the budget of state pension insurance is 20% of the wage fund paid by the employer. The rate of social tax as a whole is 33%, of which 13% is collected into the health insurance budget.

^*25 Different pension funds responsible for the mandatory funded pension are essentially private investment funds.

^*26 It should be added that both the European Code of Social Security and ILO Convention 128 include the principle that if the old-age benefit is not ensured by a public institution, representatives of public authority shall participate in the activity of said institutions.

^*27 European Code of Social Security Article 71, ILO Convention 128, Articles 35 and 36.

^*28 M. Cousins. Social Security as a Possession. – The European Convention on Human Rights and Social Security Law. Intersentia 2008, pp. 17–45.

^*29 Kogumispensionide seaduse ja sotsiaalmaksuseaduse muutmise seadus (Act for the Amendment of Funded Pensions Act and Social Tax Act), §1. – RT I 2009, 26, 161 (in Estonian); see also L. Leppik. II pensionisamba külmutamine ei ole ainus võimalus (Freezing the 2nd Pension Pillar is Not the Only Option). – ERR News, 6.4.2010. Available at http://uudised.err.ee/index.php?06160447 (9.10.2011) (in Estonian).

^*30 In 2011 state pays 2%, in 2012–2013 state pays 4%, 2014–2017 state pays 6%. Since 2018 the normal system 2+4 will be restored. See http://www.pensionikeskus.ee/?id=3006.

^*31 Põhiseadus: kommenteeritud väljaanne (Note 7), §28 comment 10.4.4.

^*32 The Estonian Health Insurance Fund is a legal person in public law that is responsible for health services and the payment of benefits for temporary incapacity for work.

^*33 Increasing the liability of the insured person is not an issue. Both the European Code of Social Security and ILO Conventions allow not paying benefits to the insured person during the first three days.

^*34 Töötervishoiu ja tööohutuse seadus. – RT I 1999, 60, 616; RT I, 2010, 31, 158 (in Estonian).

^*35 Ravikindlustuse seadus (Health Insurance Act), §60 (1) 2). – RT I 2002, 62, 377; RT I, 10.06.2011, 7 (in Estonian).

^*36 Certificate of temporary incapacity for work is a document issued by a physician that certifies the insured person’s temporary incapacity for work.

^*37 However, if the Health Insurance Fund refuses to pay benefits, the labourer may turn to administrative court. It is therefore theoretically possible that the insured person must simultaneously turn to two courts to receive benefits, in order to convince both the employer and the Health Insurance Fund that there is no causal relationship between the intoxicated state of the insured person and the temporary incapacity for work.

^*38 In addition to this problem, the nature of the benefit paid by the employer poses a specific problem: should it be regarded as wages or as health insurance benefit? The difference lies in the fact that health insurance benefits are not subject to social security tax, whereas any sums that the employer pays on the basis of the employment contract of the labourer are also to be taxed with social security contributions.

^*39 In his statement to the Estonian Employers’ Confederation, the Chancellor of Justice has also reached the conclusion that the new regulation contradicts §12 of the Constitution, namely that the principle of equality before law has been violated, but also §§13 and 14 of the Constitution, which provide the principle of legal clarity. That means, above all, that if the state has established certain rules, these rules must be clear to the appliers of these rules. See: Õiguskantsleri märgukiri (Memorandum of the Chancellor of Justice) 10.9.2010. Available at http://www.oiguskantsler.ee (in Estonian) (9.10.2011). Although the Chancellor of Justice has in principle expressed the opinion that the procedure for the payment of these benefits is unconstitutional, no amendments have so far been made in Estonian laws and employers are still obligated to pay benefits pursuant to §12 (2) of the Occupational Health and Safety Act.

^*40 A. Maamägi. Eesti võimalik rahvaarv ja vanuskoosseis aastani 2050 (Potential Population and Age Distribution of Estonia up to Year 2050). Available at http://www.stat.ee/rahvastikupyramiid (in Estonian) (9.10.2011).

^*41 Responding to the challenge of financial sustainability in Estonian’s health system (Note 10).

^*42 Ibid.

^*43 Considering that the rate of cost-sharing is already high enough, i.e., in addition to the sums collected into the health insurance budget, persons need to cover an additional ca 26%, the application of this principle is not realistic.

^*44 See www.haigekassa.ee.

^*45 See: G. Härmsalu, K. Palu, P. Talpsepp. Ravikindlustusega hõlmamata iskute olukord Eesti Vabariigi kohalikes omavalitsustes. Õiguskantsleri kantselei 2009 (Situation of Persons without Health Insurance Cover in the Municipalities of the Republic of Estonia. Office of the Chancellor of Justice 2009). Available at http://www.oiguskantsler.ee (9.10.2011) (in Estonian); Ravikindlustuseta saab üksnes vältimatut abi (Only Emergency Care Available without Health Insurance). See Medistsiiniuudised http://www.mu.ee/ (9.10.2011) (in Estonian).

This article concentrates on the ordering by rules of the internal security of a state as an absolutely important area of practical reality. In adopting this emphasis, we are interested less in the content of the respective legal rules than we are in the problems related to the application of the principles of better regulation where the ordering of the area of internal security is concerned. The subject matter is of vital importance for Estonia as the ministries responsible for internal security and the respective legal regulation—the so-called internal security law—are facing certain difficulties in meeting the requirements set forth by the principles of better regulation. ^*1

The authors are of the opinion that better regulation, or, primarily, following of the requirements of better regulation does not replace but rather complements political decision-making. It is precisely because of that supportive function that the link between the main instruments touching upon the area of internal security policy in Estonia and the implementation of the respective policies has to be recognised. As indicated above, this implementation is possible only by and with the aid of legal rules and other utterances of law. Approaching the problem from that perspective, one can establish the necessary connection between that contained in the instruments of internal security policy (i.e., concepts, measures, means of governance, etc.), and the principles of better regulation in the legislative process at both national and EU level. ^*2

In this article, the essence of better regulation, the main features of internal security in the context of the actual state of internal security of Estonia, and the implementation side of better regulation with an emphasis on the internal security policy instruments of Estonia shall be discussed.

1. Better regulation

1.1. The concept of better regulation

In the last few decades, governments in most well-developed Western countries have made increasing efforts to improve the quality of legislation by means of various ‘better regulation’ (hereinafter referred to as the BR) programmes. ^*3 Better regulation as a concept lacks a universal definition and therefore serves as an umbrella term to cover a myriad of initiatives, such as deregulation, reducing the administrative burden, improving the quality of impact assessment, reducing the quantity of legislation, and simplification. ^*4

According to C. Radaelli, better regulation is a process addressing the whole life cycle of the regulations, laying down general rules for determination, assessment, enforcement, implementation, and ex post assessment of legal rules. Consequently, the guidelines for better regulation may embrace a vast array of measures, including simplification of administrative procedures, consolidation of legal acts, alleviation of the administrative burden, use of market-friendly alternatives, risk-based review, funds allocated for rule-making, standards for consultation of interest groups, assessment of the sustainability of the existing as well as of the new regulation, and ex post review of the effects. Of the elements of the better regulation ‘package’, regulatory impact analysis (hereinafter referred to as the RIA) has to be regarded as the most important. ^*5

RIA is a set of procedures to be followed in order to appraise regulation. It can be used both ex ante (i.e., at the stage of policy formulation, to appraise proposals) and ex post. It typically revolves around the steps of problem definition, the identification of a range of options, consultation, the classification of costs and benefits, a plan for monitoring and review, and the choice of an option on the basis of certain decision-making criteria (such as cost-effectiveness, minimisation of the administrative burden, cost–benefit analysis ratios, or thresholds). ^*6

The fact that RIA has been considered the most important element of BR programmes might be explained by the strong instrumental view of legislation, which is the predominant way of understanding the role and functions of regulation for politicians, within governmental bureaucracies, and also for most stakeholders. First and foremost, legal regulations are simply understood as means for those in power to achieve the desired goals. Therefore, the outcome of regulation is currently perceived as the basic issue, linked to the quality of legislation, and not, for example, the question of the legitimacy and justification of the government for intervening in the behaviour of ordinary people or companies. ^*7 But it also has strong potential in terms of evidence- based policy, accountability, and transparency of policy formulation processes. RIA is a general procedure applicable to a large number of initiatives, from proposals for new legislation coming from the government to departmental/ministerial regulation (such as statutory instruments in the UK) and rule-making delegated to independent agencies.

In most countries, the RIA is carried out by the government, not by Parliament, but MPs and also state audit offices, interest groups, etc. outside Parliament are among the users of RIA-related information, since it assists in describing the socio-legal problems, alternatives, and consequences of policy and legislation. It should also be noted that discussions of RIA are only beginning to rise from governmental to parliamentary level as far as parliamentary functions, such as representation, legislation, and supervision of the executive power, are concerned. ^*8

In pursuit of reaching the ultimate goal of uniform quality in legislation at the national, regional, and international level, drafters and commentators in both the civil and common law systems have turned to compilations of principles of drafting that could lead to consistently successful legislation. ^*9 On the international scene of regulation, the Organisation for Economic Co-operation and Development (hereinafter referred to as the OECD) has since the beginning of the 1990s had the leading role in enhancing principles and ideas of BR and quality standards for regulation. The OECD has been able to collect experiences and research data from its member countries, on the basis of which it has formulated programmes, recommendations, and policy guidebooks for the member countries for the successful adoption and implementation of RIA. ^*10

In the European Union, the subject matter of better regulation began being considered more intensely after the analysis carried out by the OECD in the area of regulatory reforms in the Member States ^*11 , and the Lisbon (European Council) Summit where the high-level advisory group chaired by M. Mandelkern (hereinafter referred to as the Mandelkern Group) was formed. The Mandelkern Group Report, serving as the first agreement aimed at better regulation on the European Union level, includes seven recommendations aimed at achieving better regulation: 1) considering the full range of options regarding policy implementation; 2) regulatory impact assessment; 3) consultation with interested parties; 4) simplification; 5) better access to regulation; 6) dealing with the respective supporting structures and co-ordination, and 7) effective implementation of European regulation. ^*12

Newer Member States have been encouraged, and indirectly requested, to pass legislation on drafting laws as a means of responding to the European Commission’s insistence that implementation of the acquis for the purposes of accession means correct implementation of EU law with national transposition measures of good quality. ^*13

In Estonia, the training and other development measures to build the conditions needed for better regulation and regulatory impact assessment were initiated in co-operation with the OECD in 1998. ^*14 Given the experience of OECD and EU Member States in the 1990s, there is no reason to think that good law-making and governance practices will start to function without political commitment in regulatory policy, methodological guidelines, systematic training, and basic surveillance mechanisms. ^*15 To respond to those problems and to harmonise the processes and terminology of impact assessment, the concept of regulatory impact analysis was developed at national level in Estonia by the Ministry of Justice in 2007–2009 and the Development Plan for Legal Policy until 2018 was adopted by the Riigikogu (Parliament) on 23 February 2011. It should be emphasised that now Estonia belongs to the OECD family of better regulation. ^*16

1.2. Problems in the application of better regulation

But why should politicians, civil servants, and different stakeholders be interested in better regulation? One potential function of BR programmes and quality standards for legislation could be that following them increases the legitimacy and acceptance of the proposed rules. ^*17 These, in turn, are preconditions for a state based on the rule of law. The legitimacy of rules is especially crucial in internal security policy because here the rules directly constrain people’s constitutional rights. ^*18 It is worth mentioning here that many of the security-related policy documents described below emphasise the importance of the rule of law in Estonia. Therefore, if a gap exists between the principle of rule of law and actual better regulation measures (impact assessment, civic engagement, simplification, etc.), it indicates serious problems in governance, because without regulatory impact assessment information, it is difficult to talk about knowledge-based and responsible law-making and public administration. ^*19

Jyrki Tala points out another risk is BR—namely, that genuine, formal decision-makers in the law-making process (the politicians) are left almost completely outside any BR activities:

The borderline between law drafters and policy planners on the one hand and the political decision makers, responsible principally to their electors on the other, is sharply maintained. Of course, the courts are also left outside the scope of BR measures, when developing the content of the legal system by means of decisions in single cases. ^*20

Correspondingly, the man on the street, often the main target of the regulation, obviously has very little say in those processes. ^*21 Thus bad implementation and bad judicial application may interfere with the results of BR. The extent of the margin for incorrect implementation and judicial application is directly linked to the quality of the draft legislation, of course, but it is possible that the error in the draft may be attributed to a fault in the content of the policy pursued or in the calculations in the regulatory impact assessment made for the allocation of resources for implementation. ^*22

In most European countries, the analytical information on social, budgetary, economic, environmental, and administrative objectives and impacts of proposed legislation has to be given in an explanatory memorandum (in note or letter form) accompanying the draft law. The explanatory memorandum on the draft law is (and indeed has to be) a normatively structured legal document that includes the results of socio-legal impact assessment and public consultations. ^*23 If the draft law is not accompanied by impact assessment documentation, the legislator cannot vow and declare that the new law is in conformity with the rule of law or the constitutional norms.

Even more, if the target groups and effects on their lives are not specified in the memoranda, the EU better regulation assessment principles cannot be applied. The Mandelkern Group Report’s principles describe a comprehensive overall approach with a set of seven core principles: necessity, proportionality, subsidiarity, transparency, accountability, accessibility, and simplicity. ^*24 In Estonia, those principles are further supplemented by legality, legal certainty, openness, and responsibility. ^*25

The same applies a fortiori to internal security regulations, for the reasons described above.

And, naturally, the task of analysing the impact of legislation would go partly uncompleted if it were not followed by specific activities to enhance the quality of legislation by various methods (such as simplification and/or codification). ^*26 For example, the UK Better Regulation Task Force, which was set up in 1997, recommends that a rolling programme of simplification be developed to identify regulations that can be simplified, repealed, reformed, and/or consolidated. ^*27 Regulators should undertake more frequent and better post-implementation reviews of regulation. Such reviews should assess whether the measure is working as expected, whether the costs and benefits are as predicted, whether there have been unintended consequences, and whether there is scope for simplification. The results of these reviews should feed in to future policymaking and simplification proposals. ^*28 The authors of this article have similar considerations in mind concerning the internal security law in Estonia. That will be discussed in the next two parts of this contribution.

2. Internal security policy

Before any system can be developed, the definitions in it must be agreed upon. Definitions are decisive for any system because only after clarification of the definitions can norms be arranged in a uniform system. ^*29

Defining internal security itself has proved to be a difficult task. It traditionally referred to the territorial state and its geographic borders beyond which ‘inner’ should become ‘outer’ and where security is traditionally one-dimensional, as military security. ^*30 Olivier Brenninkmeijer states that security priorities have now shifted. They encompass the prevention of crime and of illegal transnational trafficking and smuggling, the control of clandestine migration, and the fight against urban juvenile delinquency. ^*31

The European Security Research and Innovation Forum (hereinafter referred to as the ESRIF) stated in its final report that, on one hand, its role was not to define security policy but, on the other, it aims for a common understanding of security, research, and innovation to support a more harmonised approach. ^*32 The report further assures that ‘the ESRIF took a holistic approach to security, taking the widest definition of security and examining how that can be achieved regarding society itself and the freedoms we want to maintain or enhance’. ^*33

The Internal Security Strategy for the European Union gives an overall definition: ‘In this context EU internal security means protecting people and the values of freedom and democracy, so that everyone can enjoy their daily lives without fear.’ ^*34 The strategy also emphasises the importance of a broad-based approach to the concept of security:

The concept of internal security must be understood as a wide and comprehensive concept which straddles multiple sectors in order to address these major threats and others which have a direct impact on the lives, safety, and well-being of citizens, including natural and man-made disasters such as forest fires, earthquakes, floods and storms. ^*35

Defining internal security through threats posed to people and measures taken to avoid these threats is quite common. All of the Estonian policy documents on security that are mentioned below take this approach. Internal security is not defined precisely; rather, a number of threats, activities, or actors are listed and analysed. The Internal Security Strategy also attempts to concretise the concept by describing the measures encompassed in it. A horizontal dimension of security is described: ‘to reach an adequate level of internal security in a complex global environmental requires the involvement of law‑enforcement and border-management authorities, with the support of judicial co‑operation, civil protection agencies and also of the political, economic, financial, social and private sectors, including non-governmental organisations’; also, there is a vertical dimension: ‘international co-operation, EU-level security policies and initiatives, regional co-operation between Member States and Member States’ own national, regional and local policies.’ ^*36

The notion of security, particularly as related to the expression ‘internal security’, has become increasingly diversified also in the sense of both the overall security that the state offers to society and the feeling of personal safety of the citizen. It contrasts with what used to be major security concerns of both state and citizen—namely, external aggression by a foreign power. Rather than that, internal security is now considered to encompass such diverse issues as economic security, the prevention of all forms of crime and violence, and social security. ^*37

The vice-chancellor of the Ministry of the Interior, Erkki Koort, categorises a problem as being part of internal security if a certain act brings with it danger to people’s life and health. ^*38 In today’s Europe, those acts are considered to consist of terrorism, serious and organised crime, drug trafficking, cyber-crime, trafficking in human beings, sexual exploitation of minors and child pornography, economic crime and corruption, trafficking in arms, and cross-border crime. ^*39 As we can see, many of those cannot be considered merely internal threats. This might be one of the reasons it is so hard to separate the ‘internal’ from ‘external security (i.e., security from defence) and give a clear definition to both. As the internal security strategy action plan states, ‘[i]nternal security cannot be achieved in isolation from the rest of the world, and it is therefore important to ensure coherence and complementarity between the internal and external aspects of [EU] security’. ^*40

Three common factors can be found in the attempts made to define internal security: 1) it must be seen as a wide and comprehensive concept; 2) it cannot be looked at separately from the ‘outer’ security, and 3) it involves an increasingly diversified situation. Regardless of such specification, a uniform definition clearly has not formed yet.

3. Better regulation and internal security

In the view of the authors of this contribution, better regulation is especially important in the field of internal security. Rules concerning people’s security, their rights and obligations toward a state, and infringements of their constitutional rights must be very clear and thoroughly analysed. We will now turn to some of the key policy documents of Estonian internal security in order to find out whether or not they embrace the concept of better regulation. Policy documents are analysed because they form the basis for ministerial measures in internal security—including legislation, planning, and allocation of funds. If better regulation is not included on that level, it cannot be transformed into knowledge-based draft legislation procedures, adequate laws, or an effective internal security policy.

We can find some minor pieces of comparative research on implementation of better regulation tools in the field of internal security regulation and regulatory management. Three policy documents are analysed here. The analysis focuses on whether and to what extent better regulation guidelines are related to the concept of internal security policy and to its objective and methods. ^*41

The National Security Concept of the Republic of Estonia (2004) specifies that:

Estonia’s Internal Security Policy encompasses the functions of the state’s internal security agencies and the general structure of the system as well as participation in activities ensuring international security. The main functions of the Internal Security Policy, for achieving the goal of the National Security Policy, are the ensuring of domestic stability and the protecting and saving of human lives.

The importance of government based on the rule of law is also emphasised, but the policy process and better regulation concept as a precondition for a sustainable policy process is not mentioned in the document. ^*42

The Main Guidelines of Estonia’s Security Policy Until 2015 develops this policy further and ‘specifies the standard principles, vision, directions and long-term effect-based objectives of the security policy—principles which must be adhered to, and objectives which must be facilitated by the public sector, non-profit sector and the private sector’. Three policy planning phases are described in the definition of security policy―development, improvement, and implementation of legal acts; development plans and activity plans with the aim of preventing threats to public order; and, in cases of a suspected threat, ascertaining and eliminating them. ^*43

This is amended by the definition of security:

[A] social state of affairs which is created with the help of many, which allows individuals to feel protected, and which ensures a truly safe living environment by reducing the probability of hazardous situations as well as enhancing the ability to react to threats and alleviate the damage caused by realisation of the threat.

The generally accepted principles of involvement of stakeholders and public consultations are also stressed as a method for preventing deviant behaviour, which is a positive step toward better regulation. ^*44 Some guidelines for impact assessment, involvement, and better regulation are specified in the implementation-related parts of the policy document, but their emphasis remains on the implementation of said policy itself, not its quality as a whole.

The Development Plan for the Ministry of the Interior for 2011–2014 states that the field of internal security encompasses the creation of an internal security policy that is composed of crisis management, rescue, migration, border guard, law enforcement, and criminal justice policies and also internal security education.

Better regulation concepts and activities are not included in the body of this document, in spite of the fact that more ‘effective and transparent processes’ are foreseen as one of the objectives of it. However some of the concepts—e.g., consultations, risk analysis, and administrative burden—are mentioned in the annexes (‘Overview of the current situation’) to the document. ^*45

The analysis confirms that better regulation guidelines are not systematically integrated with the concept or development measures for internal security policy. They are, however, occasionally mentioned in annexes or background information. Unlike equivalent work of many other ministries in Estonia, the Development Plan of the Ministry of Interior does not include a special portion on organisational development measures, where better regulation guidelines usually belong.

Therefore, it is too early at this stage to speak about systematic implementation of better regulation guidelines in the context of internal security. Some significant improvements have been made in recent years, but the importance of the quality and sustainability of the whole policy process is still not emphasised enough.

The question of institutional analysis is this: At what level do the factors precluding the implementation of better regulation principles in internal security policy exist? In connection to Estonia, three observations can be made: 1) There are no international obstacles, and the better regulation programmes apply to Estonia; ^*46 2) Estonia took a step closer to the countries leading the OECD by approving the Development Plan for Legal Policy on 23.2.2011; ^*47 and 3) the application of better regulation principles depends greatly on the choices made by vice-chancellors and departments, and on their values, work routines, and understandings of the better regulation policy. ^*48

It seems therefore that political commitment in relation to the principles of good legislation is one of the most important conditions for introducing the methods of RIA. To be useful, impact assessment should be institutionally linked to decision-making and the creation of laws. The White Paper on European Governance (2001) also raises the important question of political will: ‘Carrying these actions forward does not necessarily require new Treaties. It is first and foremost a question of political will.’ ^*49

Another problem worth mentioning is the absence of a consistent definition of internal security policy in Estonia, which probably precludes effective communication between the political-administrative and operational management and also affects consultation with target groups. This could lead to much worse results when action is needed urgently.

Correspondingly, the actual laws of internal security cannot be of the best possible quality. Mapping of Estonian regulations in the field of internal security reveals that there are roughly 150 relevant laws in force today. ^*50 This is an overwhelming quantity in light of the fact that almost all of these laws have ‘lower’ (implementing) acts as well. This mass of laws and regulations taken as a whole is deemed the law of internal security today, but such situations are not new or recently developed. As French scholar E. Catta states, many deficiencies exist in the laws of virtually every state today. Chief among these are overabundance (usually it is unclear how many laws there are in any given state), pileup of laws (usually the legislator does not summarise former laws or abolish the contradicting, excessive, or expired and therefore useless text), and instability (many laws or even paragraphs are changed several times in a year). ^*51 All of these deficiencies seem to be found in Estonian internal security law.

Problems such as these can be solved through the better regulation instruments enumerated above. Along with these instruments, Catta suggests a few practical steps: 1) compilation: the grouping of texts by subject area, or in chronological order; 2) consolidation: amendments being inserted in the initial law to achieve a uniform and up‑to-date work; and 3) codification: use of the previous two solutions to classify norms and integrate them by areas of law. ^*52

The Development Plan of the Ministry of the Interior for 2011–2014 and the Main Guidelines for Estonia’s security policy until 2015 foresee the codification in the area of crisis management. ^*53 But it is of vital importance to map out the whole internal security area before work is started on codifying a specific part of it. If an overall analysis is not conducted, the codification will probably have gaps and contradictions in it.

In light of the imperfections described above that internal security law faces, the authors suggest systematising this field of law. Systematising objective law is not merely a technical task. Systematisation of legal provisions creates and develops the system of concepts that frames all legal thinking—including clarification of the content of legal provisions. In its final stage, systematisation of objective law is an essential tool in implementing the rule of law as an idea in applied form. It is, therefore, not correct to reduce codification as a traditional element of systematising legal provisions to mere compilation of a code from different parts of a legal order or set of laws. ^*54 The aim in codification is, above all, to create legal certainty and clarity by making it easier for those applying the law to find the necessary regulation and providing a more general view of the applicable law. ^*55

The Development Plan of the Ministry of Justice of the Republic of Estonia that applied until 2005 stipulated as one of the main functions of the ministry the correspondence of laws to society’s expectations. To that end, the development plan provided for preparations for so-called codification plans. The document included observations of the idea of, need for, and methods of codification, along with recommendations on how to codify Estonian law. ^*56 Unfortunately, the development plan for the next period did not include such codification plans.

The French doctrine of codification prescribes four types of codification: reformatory codification, codification of constant law, consolidation, and compilation. In the context of this article, the second type of codification proves interesting and may even be feasible in connection with internal security law. Codification of constant law is aimed at gathering and structuring objective law as it is, not as it should be. An overview is made of the current state of affairs that paves the way to necessary reforms. ^*57 This is systematic codification that is thematic; i.e., systems of law as well as laws and regulations that belong to the same subject field are organised in a code. A homogenous and well-integrated collection must be achieved without gaps and superficiality. ^*58 Constant law is, above all, valid, applicable law. If deficiencies exist in the law (overabundance, pileup, instability, etc.), it could be very hard to find out what constant law is. ^*59

If the systematisation of internal security law reveals that such a uniform field of law indeed exists, common denominators in each law should be pointed out and a ‘General Part’ formed for internal security law.

4. Conclusions

A subtle connection exists between what is contained in the instruments of internal security policy (i.e., concepts, measures, means of governance, etc.) and the principles of better regulation in the legislative process. But analysis of the quality of draft regulations and the excessive number of laws in the field of internal security gives evidence of a lack of both ex ante and ex post systematic regulatory impact assessment.

Owing to the fact that implementation of the common principles of the Main Guidelines of the Security Policy as well as attainment of the objectives is supervised by the Ministry of the Interior, the authors offer some suggestions that could be included in the development plans in the future:

1) Regulatory impact assessment guidelines should be drafted for the development plans and laws in the field of internal security, to increase the analytical and administrative capacities of the ministries concerned.

2) In connection with the excessive quantity of laws in force in the field of internal security, an integral analysis should be conducted before any new laws are drawn up. Comprehensive systematisation and/or partial codification of law in force should be considered.

3) Training for civil servants should be conducted before drafting and adoption of any new laws on internal security.

In this contribution, the authors have been able to take only a glimpse at the given problem but, nonetheless, hope they have demonstrated that the connection between internal security policy and better regulation deserves further research.

Notes:

^*1 A. Kasemets. Lõhe õigusloome normide ja faktide vahel mõjude hindamise ja huvirühmade kaasamise teabe esitamisel (The Gap between Law-making Norms and Facts (1998–2009) in the Field of Impact Assessment and Civic Engagement). – Riigikogu Toimetised 2009 (19), pp. 104–115 (in Estonian).

^*2 In order to refer to problematic areas as regards rule-making in the field of internal security, the authors use the concept of ‘selective compliance with the rules of rule-making’ throughout the text. The term in itself implies certain problems in this area.

^*3 J. Tala. Better regulation through programs and quality standards—are new perspectives needed? – Legisprudence 2010, p. 193.

^*4 L. Clapinska. Post-Legislative Scrutiny of Legislation Derived from the European Union. – European Journal of Law Reform 2007, p. 343.

^*5 C. M. Radaelli. Regulating Rule-Making via Impact Assessment. – Governance 2010/1, p. 90.

^*6 OECD. Policy Recommendations on Regulatory Reform. Regulatory Impact Analysis: Best Practices in OECD Countries. OECD/PUMA, Paris 1997; OECD Sigma. Regulatory Management Capacities of New EU Member States, Paris 2004.

^*7 J. Tala (Note 3), pp. 203–204.

^*8 A. Kasemets. Impact Assessment of Legislation for Parliament and Civil Society: A Comparative Study. (22 countries). Proceedings of ECPRD seminar. Riigikogu Kantselei. Tallinn 2001.

^*9 H. Xanthaki. Drafting Manuals and Quality in Legislation: Positive contribution towards certainty in the law or impediment to the necessity for dynamism of rules? – Legisprudence 2010, p. 112.

^*10 J. Tala (Note 3), p. 197; OECD. Building an Institutional Framework for Regulatory Impact Analysis (RIA): Guidance for Policy Makers. Paris 2008.

^*11 OECD. Reducing the risk of policy failure: Challenges for regulatory compliance. Paris 2000. Available at http://www.oecd.org/dataoecd/23/37/46466287.pdf (5.4.2011).

^*12 Mandelkern Group on Better Regulation. Final Report. 13 November 2001. Available at http://ec.europa.eu/governance/better_regulation/documents/mandel-kern_report.pdf (5.4.2011).

^*13 H. Xanthaki (Note 9), p. 121.

^*14 See for example M. Ben-Gera. Impact Assessment: Role, Procedures, Methods and Good Practices in OECD and CEE Countries. – Society, Parliament and Legislation. A. Kasemets et al. (eds.). Tallinn: Riigikogu Kantselei 1999, pp. 27–35.

^*15 A. Kasemets. Teadmistepõhise ning säästva poliitika institutsionaalsed eeldused Eestis (The Institutional Preconditions for Knowledge-based and Sustainable Politics in Estonia). – Riigikogu Toimetised 2006 (14), pp. 152–160 (in Estonian).

^*16 Justiitsministeerium. Õigusaktide mõjude analüüsi süsteemi üldkirjeldus. Õiguspoliitika osakond (Ministry of Justice. General Description of impact assessment system. Legislative Policy Department), 2008. Available at http://www.just.ee/41314 (5.4.2011) (in Estonian); Õiguspoliitika arengusuunad aastani 2018 (Development Plan for Legal Policy until 2018). – RT III, 7.3.2011, 1 (in Estonian).

^*17 J. Tala (Note 3), p. 207.

^*18 D. Bigo, S. Carrera, E. Guild, R. Walker. The changing landscape of European liberty and security. – International Social Science Journal 2008/59, pp. 283–308.

^*19 Rule of law, i.e., legal certainty and predictability of administrative actions and decisions, which refers to the principle of legality as opposed to arbitrariness in public decision-making and to the need for respect of legitimate expectations of individuals. Openness and transparency, accountability and efficiency are also specified as common standards for action within public administration. OECD Sigma. Preparing Public Administration for the European Administrative Space. Sigma No. 23. Paris 1998.

^*20 J. Tala (Note 3), p. 203.

^*21 Ibid., p. 205.

^*22 H. Xanthaki (Note 9), pp. 114–115.

^*23 See for example M. Kiviorg. Õigusakti mõju ja rakendamise analüüs (Analysis of the Effect and Application of a Legal Act); – K. Merusk et al. Õigusriigi printsiip ja normitehnika (Principle and Normative Techniques of a State Based of the Rule of Law). Tartu: Sihtasutus Eesti Õiguskeskus 1999, pp. 19–39 (in Estonian); R. Narits. Good Law Making Practice and Legislative Drafting Conforming to It in the Republic of Estonia. – Juridica International 2004 (IX), pp. 4–13.

^*24 Mandelkern (Note 12).

^*25 Ministry of Justice (Note 16).

^*26 R. Narits. Systematisation of Objective Law. – Juridica International 2005 (X), p, 164.

^*27 L. Clapinska (Note 4), p. 339.

^*28 Better Regulation Task Force, Regulation—Less is More. Reducing Burdens, Improving Outcomes. A BRTF report to the Prime Minister. United Kingdom 2005, p. 33.

^*29 R. Narits. Objektiivse õiguse korrastamisest Eestis: kodifitseerimisest õiguse ümberkujundamiseni (Systematisation of Objective Law in Estonia: From Codification to Reformation of Law). – Riigikogu Toimetised 2005 (12), pp. 71–79 (in Estonian).

^*30 O. Brenninkmeijer. Internal Security Beyond Borders: Public Insecurity In Europe And The New Challenges To State And Society. Peter Lang Pub Inc. 2001, pp. 30–31.

^*31 Ibid.

^*32 European Security Research and Innovation Forum. ESRIF Final Report. December 2009, p. 12. Available at http://www.esrif.eu/documents/esrif_final_report.pdf (5.4.2011).

^*33 Ibid., p. 11.

^*34 Council of the European Union. Internal Security Strategy for the European Union. Brussels. 25.2.2010. 6870/10 (Presse 44).

^*35 Ibid.

^*36 Ibid.

^*37 O. Brenninkmeijer (Note 30), p. 42.

^*38 E. Koort. Siseturvalisusus või sisejulgeolek (Internal Safety or Internal Security). – L. Tabur, E. Koort (eds.). Turvalisuspoliitika 2010 (Safety Policies 2010). Tallinn 2010, pp. 39–43 (in Estonian).

^*39 Internal Security Strategy for the European Union (Note 34).

^*40 European Commission. The EU Internal Security Strategy in Action: Five steps towards a more secure Europe. Brussels. 22.11.2010. COM(2010) 673 final. Parenthesis added.

^*41 Mandelkern Group on Better Regulation Final Report (2001). Available at http://ec.europa.eu/governance/better_regulation/documents/mandel-kern_report.pdf (5.4.2011).

^*42 Eesti Vabariigi julgeolekupoliitika alused (The National Security Concept of the Republic of Estonia). – RT I 2004, 49, 344 (in Estonian). Available at http://www.vm.ee/?q=node/4107 (5.4.2011).

^*43 Eesti turvalisuspoliitika põhisuunad aastani 2015 (Main Guidelines of Estonia’s Security Policy until 2015). – RT I 2008, 25, 165 (in Estonian). English text available at http://www.siseministeerium.ee/29744/ (5.4.2011).

^*44 Kaasamise hea tava (The Good Practice of Public Participation). Available at http://www.valitsus.ee/et/riigikantselei/ valitsuskommunikatsioon/kaasamine/kaasamise-hea-tava (5.4.2011) (in Estonian).

^*45 Siseministeeriumi valitsemisala arengukava 2011−2014 (The Development Plan of the Ministry of Interior 2011−2014). Available at http://www.siseministeerium.ee/17410/ (5.4.2011) (in Estonian).

^*46 Mandelkern (Note 12), OECD (Note 10).

^*47 See Note 16.

^*48 A. Kasemets. Teadmistepõhise ning säästva poliitika institutsionaalsed eeldused Eestis (Institutional Prerequisites of Knowledge-based and Sustainable Policy in Estonia). – Riigikogu Toimetised 2006 (14), p. 154 (in Estonian).

^*49 European Commission. European Governance. A White Paper. Brussels, 25.7.01, COM(2001) 428. Available at http://europa.eu.int/comm/governance/white_paper/en.pdf (5.4.2011).

^*50 See Note 45.

^*51 E. Catta. Kodifitseerimise ja süstematiseerimise tähtsus õige seaduse leidmise lihtsustamisel (Importance of Codification and Systematisation in Simplifying the Finding of a Correct Act). – Juridica 2002/9, pp. 588–589 (in Estonian).

^*52 Ibid., p. 589.

^*53 Note 45. As stated in the latter document: A general legal act will be established in the field of crisis management, organising the entire legislative basis for crisis management.

^*54 R. Narits (Note 26), pp. 161–162.

^*55 Ibid., p. 163.

^*56 Ibid.

^*57 E. Catta (Note 51), p. 589.

^*58 Ibid., p. 590.

^*59 Ibid., p. 591.

In recent years, as the EU consumer acquis is being systematically revised and the fundamental principles of European contract law are being drafted, the academic discussion in international legal literature has mainly been focussed on the aims and principles of consumer contract regulation. ^*1 On the other hand, the issue of determination of the circle of persons entitled to extended protection as consumers has been relatively less touched upon in the discussion hitherto. In this article the authors attempt to bridge this gap and address in greater detail the bases of the concept of consumer in the light of the changing EU consumer acquis. The article analyses the concept of consumer in current EU consumer acquis and from the perspective of Member States, in particular from that of Estonian national law. Then the impact of the recent EU consumer law initiatives on the concept of consumer is explored, relying on the new consumer rights directive adopted by the European Parliament. ^*2 For that purpose, the main bases of changes are highlighted while addressing the potential risks and challenges of implementing the law in the future.

1. Notion of consumer de lege lata

1.1. The notion of consumer in current EU consumer acquis

In current EU consumer acquis, the notion of consumer has been specified separately in each adopted instrument. Thus the notion of consumer has been defined in several directives in the area of contract law ^*3 ; in the area of non-contractual obligations, a ‘consumer’ has been defined in the EU producer liability directive ^*4 (85/374/EC) and, in the area of procedural law, in the regulations Brussels I ^*5 and Rome I. ^*6

The definitions of consumer, as provided for in various EL instruments, do not entirely coincide. A majority of current EU directives defines the consumer as a natural person who, in transactions covered by the directive, is acting for purposes which are outside his trade, business or profession. ^*7

The notion of consumer as defined in the price indication directive and the original doorstep selling directive is essentially similar, treating the consumer as any natural person who, in transactions covered by the directive, is acting for purposes that do not fall within the sphere of his commercial or professional activity. Albeit differently worded, the scope of the notion of consumer is, in essence, the same in the Brussels I and Rome I regulations, pursuant to which a consumer is a natural person who has concluded a contract for a purpose outside his commerce or profession. Article 2 (a) of the unfair commercial practices directive goes a little further and excludes from among the persons who can be regarded as having the characteristics of a consumer such persons who operate in crafts (‘is acting for purposes which are outside his trade, business, craft or profession’).

Thus, as evidenced above, the definitions of consumer as provided in EU legislation do not completely overlap as far as their content is concerned. Nevertheless, having analysed the notion of consumer in the current EU consumer acquis it can be said that most instruments describe a consumer as sharing two central characteristics:

(a) a consumer is a natural person, and

(b) in concluding a contract he is acting for purposes which are outside his commercial or professional activities. ^*8

As already said, this applies to most of the legislative acts. The package travel directive ^*9 is exceptional among the other directives in that in Article 2 (4) it defines consumer as a person ‘who takes or agrees to take the package (‘the principal contractor’), or any person on whose behalf the principal contractor agrees to purchase the package (‘the other beneficiaries’) or any person to whom the principal contractor or any of the other beneficiaries transfers the package (‘the transferee’)’. Though said directive includes a definition of consumer, as do other directives regulating the area of contract law, it is clearly different from the other directives in that the notion of consumer covers not just natural persons but all other buyers of package travel, including persons who conclude a package travel contract inside their commercial or professional activities. ^*10

1.2. The notion of consumer in Member States’ law

While exploring the notion of consumer as defined in EU Member States’ national law, a rather mottled picture unravels influenced by the Member States’ legal traditions and facilitated by the principle of minimal harmonisation used in EU directives to date. Several Member States, such as Estonia, Germany, Belgium, Poland and others, provide for a negative definition of consumer. For example, pursuant to §2 (1) of the Estonian Consumer Protection Act ^*11 (hereinafter referred to as the CPA), consumer means a natural person to whom goods or services are offered or who acquires or uses goods or services for purposes not related to his or her business or professional activities. Several other Member States (e.g., Finland, Sweden) use the so-called positive method to define consumer by listing the characteristics of a consumer. ^*12

In transposing the directives, some Member States (e.g., Hungary, United Kingdom, Cyprus, Ireland, and Luxembourg) have followed the example of the directives and incorporated the definition of consumer into each instrument transposing an EU legislative act. However, several Member States have adopted an approach to develop a general definition of consumer applicable to all of the transposed directives. ^*13

Estonia is among those Member States which have several general definitions of consumer. ^*14 Besides the definition provided in the CPA, consumer has also been defined in the Estonian Law of Obligations Act ^*15 (hereinafter referred to as the LOA). Section 34 of the LOA sets out that for the purposes of the LOA, a consumer is a natural person who performs a transaction not related to an independent economic or professional activity. This definition does not fully overlap with that provided for in §2 (1) of the CPA. Firstly, as formulated in the LOA, a consumer is a person who performs a transaction and not a person to whom goods are offered, i.e., a person who might not enter into the transaction. However, the criterion of transaction is not necessarily the decisive constitutive element in clarifying the position of the consumer, as the LOA also sometimes uses the notion of consumer in a context where there is no transaction per se (e.g., LOA §99: Provision of goods or services not ordered). Another difference between §34 of the LOA and §2 (1) of the CPA is more fundamental by its nature: Namely, the definition provided in the LOA makes no reference to the purpose for which the consumer is purchasing or using the goods or services, but limits itself to transactions not related to an independent economic or professional activity. Therefore, it can be argued that LOA’s definition is broader than that of the CPA and allows a much wider circle of contracts to be treated as consumer contracts. This conclusion is corroborated by earlier Estonian case-law regarding the contracts of suretyship entered into by the members of the management board of a legal person to guarantee the fulfilment of the obligations of the person. Combining §34 of the LOA and §2 (1) of the CPA, the Supreme Court has taken the view that a person who enters into a contract of suretyship due to having interest in the economic activities of the company cannot be deemed to be a consumer. ^*16 In other words, the court has found that the definition of consumer contained in §34 of the LOA should be delimited and used the definition provided in the CPA for that purpose. It is, however, true that as of 5 April 2011 this practice of the Supreme Court no longer applies to the suretyship issued by a member of the management board because, in order to improve the situation of all sureties who are natural persons ^*17 , §143 (1) of the LOA was amended so that a contract of consumer surety is a contract of suretyship where the surety is a natural person.

Teleological interpretation of the definition of consumer as specified in §2 (1) of the CPA leads to the conclusion that the law delimits the notion of consumer only to such a person who purchases or uses goods or services solely outside his economic or professional activity. For example, the Austrian and Belgian regulations follow similar principles, while several Member States (Finland, Sweden, Denmark, etc.) proceed from the principal purpose of use, i.e., the predominant purpose of use. ^*18 The original wording of the CPA (1.4.2003) too defined a consumer as a natural person who intends to purchase or use goods or services for a purpose which is not directly related to his economic or professional activity. ^*19 However, in the course of processing the draft, the definition of consumer was changed and protection was precluded for persons who use, even if just in part, the goods or services in their economic or professional activities.

As evidenced above, similarly to several other Member States and the EU consumer acquis, the general notion of consumer as provided for in current Estonian law, as a rule ^*20 , proceeds from a narrow definition of consumer, delimiting the concept to a natural person who uses the commodity solely outside his economic or professional activity. There are, however, Member States which, under certain circumstances, also extend consumer protection provisions to legal persons. As far as the purpose of consumer transactions is concerned, based on the national law of most Member States it cannot be said whether treating a contract as a consumer contract depends on the principal purpose of the contract or if just any connection with the person's professional activity precludes the application of consumer provisions. ^*21

2. Main characteristics of the concept of consumer in the context of the changing EU consumer acquis

2.1. Revision of the EU consumer acquis

The revision of the EU consumer acquis is a part of the measures implemented to apply Community’s contract law uniformly and improve the functioning of the internal market.The review was necessitated by the need to tackle the fragmentation of consumer protection legislation and to modernise and simplify the system of rules which had become out-dated due to rapid development of technical capabilities. The procedure was initiated back in 2004 by the Commission, with its communication ^*22 , and it covers eight consumer directives. ^*23 The Green Paper on the review of Community’s consumer acquis, published in February 2007, notes that the overarching aim of the Review is to ‘achieve a real consumer internal market striking the right balance between a high level of consumer protection and the competitiveness of enterprises, while ensuring the strict respect of the principle of subsidiarity’ and at the end of the exercise it should, ideally, be possible to say to EU consumers ‘wherever you are in the EU or wherever you buy from it makes no difference: your essential rights are the same’. ^*24 The Green Paper also notes that there are certain issues which are common to all directives of the consumer acquis which could be extracted from the existing directives and regulated in a horizontal instrument. The notion of consumer is one of such issues: The Commission stresses that a consistent definition of the notions of consumer and professional is important since it permits to delimit the scope of the acquis more accurately and that during the review the widening of the definitions to cover transactions for mixed purposes should be considered. ^*25

The new consumer credit directive ^*26 and timeshare contract directive ^*27 , replacing previous directives, were the first to be adopted during the revision of the EU consumer acquis. The new directives approach the notion of consumer the same way as the original directives and, as a result, each directive defines the notion of consumer separately but not in the same wording. Thus, it can be said that unlike the emphasis of the Green Paper, it was not deemed necessary to use a uniform and consistent approach to the notion of consumer in the first revised directives.

In October 2008, in the framework of revising the consumer acquis the Commission tabled a proposal for a directive on consumer rights. ^*28 The European Parliament adopted the new directive on 23 June 2011. The new directive, in whose drafting the standpoints that had evolved during the drafting of the Draft Common Frame of Reference ^*29 (hereinafter referred to as the DCFR) were taken into account, combines into one whole two previous EU directives. ^*30 The goal is to enhance consumers’ trust and curb the bureaucracy which has hindered companies’ activities in other EU Member States, thus stripping the consumers of the freedom of choice and competing offers. In legal literature the new directive has already been seen as a suitable foundation for a future European code of consumer rights (contracts). ^*31 The consumer rights directive uses for the first time in the EU consumer acquis a uniform general notion of a consumer applicable to all of the consumer contracts covered by the framework directive.

2.2. Consumer as a natural person

As noted above, in the EU consumer acquis the notion of consumer covers as a rule just natural persons. This has been on several occasions emphasised in the case law of European Court of Justice. In joined cases Idealservice the European Court of Justice took a firm stance in interpreting the unfair terms directive ^*32 that the notion of consumer covers only natural persons. ^*33

In the Di Pinto case the European Court of Justice interpreted, for the purposes of the doorstep selling directive ^*34 , the notion of consumer in a narrower sense, noting that a trader canvassed with a view to the sale of his business is not to be regarded as a consumer protected by the directive and explained that the directive does not afford protection to legal persons even if they are in a position similar to that of a consumer. ^*35

In a number of Member States, including in Estonia’s current law and case-law, the notion of consumer is solely delimited to natural persons ^*36 and this position is also supported in legal literature. ^*37

In the discussions held in the Commission on the consumer regulation of the Draft Common Frame of Reference, the experts agreed that the notion of consumer should cover solely natural persons. ^*38 This stance is also expressed in the DCFR: Pursuant to Article I.-1:105 of the DCFR, a consumer is a natural person. Legal persons performing transactions outside their normal economic or professional activities are not regarded as consumers.

Pursuant to Article 2 (1) of the proposal for a consumer rights directive ^*39 the notion of consumer initially covered just natural persons. The opinion published on 16 July 2009 by the European Economic and Social Committee ^*40 , however, noted that the framework directive should adopt a clear position, inter alia, regarding whether the notion of consumer could be extended to certain legal persons, as a number of Member States have done. Based on the opinion of the European Economic and Social Committee, the proposal for a directive was supplemented by a clause to the effect that the Member States may maintain or extend the rules of the directives to natural or legal persons who are not consumers within the meaning of the directive. ^*41

Recital 13 of the adopted directive includes the right of the Member States to maintain or introduce national legislation corresponding to the provisions of the directive or certain of its provisions in relation to transactions that fall outside the scope of the directive. However, the same recital sets out the right of the Member States to decide to extend the application of the rules of the directive to legal persons or to natural persons who are not ‘consumers’ within the meaning of the directive, such as non-governmental organisations, start-ups or small and medium-sized enterprises.

This clause may be seen as being different from the original proposal for a directive, as proposed by the Commission, whose aim was maximum harmonisation, but rather as the retention of the principle of minimum harmonisation and concession to those Member States which have provided in their current laws for an option to regard certain legal persons as consumers (e.g., Spain, Belgium, Slovakia, Denmark, Greece, and Austria). ^*42

For those countries, the requirement of maximum harmonisation would have meant significant restriction of the scope of the consumer acquis. It has been likewise noted in legal literature that in such a situation maximum harmonisation does not widen the consumer’s rights but instead restricts them. ^*43

On 9 October 2010, the European Commission published an unofficial note to the proposal for a consumer rights directive ^*44 , which is intended to serve as a guide to the implementation of the proposal for a directive as regards certain types of contracts. The note provides clarification as to who might be these legal or natural persons which are not ‘consumers’ in the meaning of Article 2 (1) regarding whom the Member States may maintain or extend the application of the rules of the proposal. According to the note, a Member State may, e.g., give NGOs or small businessmen status equal to that of consumers. However, it is pointed out that such persons that have equal rights with the consumers should not be referred to as ‘consumers’ as that would be incompatible with the definitions in the proposal for a directive. ^*45

Given that the initiatives of the past few years do not directly regard legal persons as consumers and continue to stress the need to delimit the notion of consumer to just natural persons, one may conclude that there have not been significant changes in the paradigm of EU law regarding this issue.

2.3. Consumer acting outside his economic or professional capacity

As a rule, the wording of the consumer directives does not provide a clear answer to the question whether or not consumer protection is also extended to a natural person in cases where his transaction is at least slightly connected with his economic or professional activity. ^*46

The European Court of Justice has in its earlier case-law supported the approach that even the minor connection with professional activity is sufficient to preclude the application of consumer provisions. In the Dietzinger case, the European Court of Justice adopted in 1998 the position that a contract of guarantee concluded by a private person for the purpose of guaranteeing the obligations of a building firm cannot be treated as a consumer contract in the meaning of the doorstep selling directive 85/577. The court noted that as regards a guarantee, the provisions applicable to a consumer can be implemented only if the commitment is entered into for a purpose which is unconnected with the guarantor’s trade or profession. ^*47 The European Court of Justice has also found that a person who had entered into a contract for the purpose of engaging in business in the future cannot be treated as a consumer. ^*48

In the 2005 Gruber case ^*49 , the European Court of Justice demonstrated a somewhat more accommodating approach. In that case, a decision was needed on whether a contract entered into by a farmer for the purpose of buying roof tiles was a consumer contract in view of the fact that the area of the building used for personal purposes was slightly more than 60% of the total floor area. The European Court of Justice found that a person may be treated as a consumer where the link between the purpose of a transaction and the trade or profession is so slight as to be marginal. Hence, the European Court of Justice acknowledged that a slight connection to a professional activity does not preclude the application of the consumer provisions. As mentioned above, it cannot be unambiguously said on the basis of the national laws of the majority of Member States whether or not a contract should be regarded as a consumer contract on the basis of the primary purpose of use. ^*50 A number of Member States have extended the application of consumer provisions to transactions primarily entered into outside professional either expressis verbis (Denmark, Finland, and Sweden) or address the issue in their case-law (Germany). It is precisely from the German case-law that the often cited in legal literature example of a lawyer who buys a car, which he then uses both for business and private purposes, comes. ^*51 Likewise it has been concluded in Estonian legal literature that where a natural person enters into a contract which is partly linked to his professional activities, the contract is qualified depending on the predominant purpose of the transaction. ^*52

The Commission’s Green Book notes that during the review of consumer acquis the widening of the definitions is needed to cover transactions for mixed purposes, i.e., transactions whose object is used in part to satisfy personal needs, in part for business or professional activity. ^*53 This would be a major conceptual change in the definition of a consumer compared to the current approach employed in EU consumer acquis which does not recognise transactions for mixed purposes as consumer transactions. The DCFR too mirrors the changes in the perception of the extent of the concept of a consumer: pursuant to Article I.-1:105 (1), a consumer is any natural person who is acting primarily for purposes which are not related to his or her trade, business or profession. However, the definition of consumer as worded in the DCFR allows partial use of the transaction object in trade or profession (so-called dual use). The DCFR also regards as consumer contracts such contracts whose purpose is to generate one-time profit—for example, resale of a purchased item, unless such activity is regular. The frequency and volume of transactions determine whether or not such resale activity qualifies as a consumer contract. ^*54 Article I.-1:105 (3) of the DCFR deals with contracts that have different purposes of use. If a person can be treated both as a consumer and a trader, he is deemed to be a consumer in the context of provisions providing protection to consumers. This clause is intended to solve those situations where a transaction carries a personal as well as trade and profession related purpose. Thus, a person who buys a computer which he primarily uses for personal purposes and, to a slight extent, also for the purposes of trade, is a consumer. ^*55

Article 2 (1) of the original proposal tabled by the Commission regarding the consumer rights directive defined a consumer as any natural person who, in contracts covered by the directive, is acting for purposes which are outside his trade, business, craft or profession.This approach matched the more narrow approach that had become rooted in EU consumer acquis but differed significantly from the approach to consumers employed in the DCFR, and, as such, the proposal was criticised in the legal literature exploring the proposed directive. ^*56 Also, the opinion of 16 July 2009 of the European Economic and Social Committee, mentioned above, referred to a contradiction between the proposed directive and the approach of several Member States regarding the transactions for mixed purposes.

Pursuant to the note of the European Commission dated 9 October 2010, the scope of the directive includes contracts concluded both for professional and private purposes when the private purpose is clearly predominant. The note recommends that the predominant purpose be assessed by national courts on a case by case basis. ^*57 The legal definition of a consumer was changed in the proposal for a directive submitted on 24 March 2011 to the European Parliament for the first reading ^*58 , defining it as follows: ‘consumer’ means any natural person who, in contracts covered by this directive, is acting primarily for purposes which are outside his trade, business, craft or profession.

Article 2 (1) of the consumer rights directive adopted by the European Parliament defines a consumer as any natural person who, in contracts covered by the directive, is acting for purposes which are outside his trade, business, craft or profession. This regulation should be viewed in conjunction with Recital 17 of the directive, pursuant to which if the contract is concluded for purposes partly within and partly outside the person’s trade (dual purpose contracts) and the trade purpose is so limited as not to be predominant in the overall context of the supply, that person should also be considered as a consumer.

The authors hold that considering a person as a consumer based on the main goal of the transaction represents, at the EU level, a significant paradigmatic change in the concept of a consumer, resulting in a major expansion of the scope of consumer provisions.

Therefore it is all the more surprising how the definition of a consumer has been structured in the two already revised directives compared with the DCFR and the new consumer rights directive. Namely, Article 3a of the new consumer credit directive and Article 2 (1) of the timeshare contract directive are based on the early dogmatic approach that a natural person can be regarded as a consumer only if the purpose of his transactions is in no way related to his economic or professional activity. At that, Article 22 (1) of the consumer credit directive sets out that Member States may not maintain or introduce in their national law provisions diverging from those laid down in this directive. For the Estonian legislator, e.g., it means that a contradiction between the directive and §403 (2) of the LOA needed to be removed as it extends the definition of consumer credit to cover also the transactions entered into to start a business. As the Estonian legislator did not want to relinquish the already existing protection of the recipient of loan or credit ^*59 , the following solution was found: a natural person who concludes a contract to take out credit in order to start independent economic or professional activity is not considered as a consumer, however, consumer credit provisions apply to such a contract.

Upon enforcement of the consumer rights directive, there is also a question on how to address the situation where under national law the same general definition of a consumer applies both to consumer credit and consumer sale (as it is under Estonian law), but EU directives foresee concepts of a consumer with varying scope for those contracts. The authors believe that in such a situation it would be reasonable to proceed from the wider general definition of a consumer and provide for necessary restrictions for specific types of contracts. One must, however, acknowledge that such a regulative method is justified only as long as the wider general definition covers more types of contracts than contracts to which the more narrow definition applies. Several Member States have attempted to introduce a general definition of a consumer in their national law on the basis of EU consumer acquis; however, as things stand currently, the situation is much better in those Member States which have defined the notion of a consumer in each law separately. This is a paradox given that the main goal of revising the consumer acquis is to achieve greater coherence and that the Commission has in the Green Book emphasised the need to harmonise basic terms.

2.4. Recognition of a person as a consumer— a subjective or objective approach?

Having concluded that the provisions protecting consumers also apply to natural persons (Section 2.2) who primarily enter into transactions outside their economic or professional activity (Section 2.3), a question emerges whether the interpretation of his conduct by the other party is relevant in recognising a person as a consumer. In other words, if a natural person enters into a transaction in his economic or professional activity but that purpose is not recognisable to his contractual partner, should the consumer provisions be applied to such a transaction? Or should the consumer provisions also be applied in a case where a person enters into the contract for personal purposes but his conduct creates an impression that he is a person engaged in economic or professional activity?

An example from German Supreme Court case-law illustrates such a situation. ^*60 A car dealer wished to sell a used car only to such a person who is not a consumer because he wanted to use the possibility to exclude the seller’s guarantee. The buyer, who wanted the car for personal use, was aware of the fact; however, as the case was, the car would not have been sold to him and so he maintained that he was buying it for his economic and professional activity. Later, the buyer wished to withdraw from the contract due to the seller’s breach and invoked that the contract was a consumer contract. The court found that a person who deceives his contractual partner in order to avoid the contract being qualified as a consumer contract cannot, based on the principle of good faith, subsequently enjoy legal remedies designed for consumers.

Under Estonian positive law and the case-law to date, it is also possible to restrict the rights of a contractual party who, in concluding the contract, poses as an undertaking but wishes to invoke consumer protection provisions in submitting claims arising out of the contract. The authors find that such conduct has elements of contradictory conduct (venire contra factum proprium). Pursuant to §138 (1) of the General Part of the Civil Code Act ^*61 (hereinafter referred to as the GPCCA), rights shall be exercised and obligations shall be performed in good faith. Under §6 (1) and (2) of the LOA, obligees and obligors shall act in good faith in their relations with one another and nothing arising from law, a usage or a transaction shall be applied to an obligation if its contrary to the principle of good faith. The Supreme Court’s case-law has several examples where the principle of good faith has been used to restrict the claims of a person due to his contradictory conduct. ^*62

The same question has been raised in the case-law of European Court of Justice. In the above cited Gruber case the court found that the consumer contract provisions of the Brussels convention do not apply where the supposed consumer had in fact, by his own conduct with respect to the future contractual party, given the latter the impression that he was acting for business or professional purposes (e.g., the person uses corporate letterheads, asks the goods to be delivered to a business address, etc.). According to the approach of the European Court of Justice, such a conduct should be treated as waiver of the protection provided for by the consumer provisions. ^*63 The authors of the DCFR also hold that if a consumer, upon entering into a contract, knowingly gives the other party impression of being an undertaking (or ´business´), the provisions protecting the consumer should not be applied as the person's conduct is contrary to the principle of good faith. ^*64

EU consumer acquis does not impose a general duty to act in good faith. The Green Paper addresses the option to provide for such a principle and notes that the inclusion of such a principle, which would act as a safety net, would fill in any future regulatory gap and ensure that the acquis remains future proof. ^*65 However, neither the already revised directives nor the new consumer rights directives foresee such a good faith clause. The principle of good faith and fair trading has been expressed in the DCFR. Direct contradictory conduct is addressed in Article I.-1:103 (2) under which conduct of a person is contrary to good faith if it contradicts his previous conduct or representations.

The authors of the article hold that the approach according to which a natural person can objectively be regarded, upon entering into a contract, as a consumer is justified and the concluded contract should be treated as a consumer contract. But if, upon concluding a contract, a person creates an impression of being an undertaking, the application of the consumer provisions is not justified. The authors build their opinion, firstly, on the fact that it is important for an undertaking to know whether or not the other party is a consumer already before entering into the contract as this may determine the extent of the undertaking’s pre-contractual and contractual obligations. For example, if an undertaking enters into a contract with a consumer he may be required to disclose more pre-contractual information, on must also take into account some specifics in using contractual legal remedies or restricting liability by agreement or in using standard terms. Therefore, an undertaking must be able to rely, upon concluding a contract, on the conduct and representations of the other party. Secondly, the authors hold that the duty to act in good faith should not only rest with undertakings but also with consumers―treating the consumer as the weaker part of a transaction does not mean that he should be allowed to behave contradictorily.

3. Conclusions

To date, the EU consumer acquis has employed an approach to incorporate the definition of a consumer in each separate legislative instrument and these definitions do not coincide. While the Commission notes in its Green Paper that the definition of a consumer is one of the issues which should be uniformly regulated in all of the directives to be revised, the two directives adopted in the consumer acquis revision process do not follow the principle. This conclusion is based on the observation that the definitions of a consumer in the new consumer credit directive, timeshare contracts directive and consumer rights directive do not match. As far as the consumer rights directive is concerned, there is also a reason to speak about a significant change in the concept of consumer as, although firmly holding on to the principle that a consumer is a natural person, unlike the EU consumer acquis to date, any natural person who acts primarily for a purpose not linked to his trade or business is also considered a consumer.

Hence, it can be said that the revision of the directives has not led to a greater consistency in the consumer acquis as far as the notion of consumer is concerned and if, until now, it could be said that the EU consumer acquis proceeded from a narrow definition of a consumer, two clearly differing doctrines are emerging at the legislative level. Member States who to date have limited themselves to just transposing the minimum criteria set out in the directives and have used one general definition of a consumer in the process, are now forced to decide whether to widen the definitions for certain types of contracts (which renders the idea of one general definition questionable) or amend the general definition (which involves a significant change in the addressees of the provision and its application instances, which requires a thorough analysis as well as legal and political decisions). The legislators of the Member States continue to face the challenge of tackling the problem of taking over EU legislation into their national laws where the instruments do not coincide. And paradoxically the problem related to the regulation of the definition of a consumer arises out of the revision of the EU consumer acquis whose goal is to achieve better coherence of acquis.

Notes:

^*1 See, e.g., J. Karsten, A. R. Sinai. The Action Plan on European Contract Law: Perspectives for the Future of European Contract Law and EC Consumer Law. – Journal of Consumer Policy 2003/26; G. Howells, R. Schulze. Modernising and Harmonising Consumer Contract Law. Munich: Selliers 2009; H. Schulte-Nölke. Perspectives for European Consumer Law. Munich: Selliers 2010.

^*2 Position of the European Union adopted at first reading on 23 June 2011 with a view to the adoption of Directive 2011/.../EU of the European Parliament and of the Council on consumer rights, amending Council Directive 93/13/EEC and Directive 1999/44/EC of the European Parliament and of the Council and repealing Council Directive 85/577/EEC and Directive 97/7/EC of the European Parliament and of the Council (P7_TC1-COD(2008)0196). Available at http://www.europarl.europa.eu/sides/getDoc.do?type=TA&reference=P7-TA-2011-0293&format=XML&language=ET#BKMD-21.

^*3 Article 2 of the original doorstep selling directive (85/577/EEC) and Article 2 (2) of the distance contracts directive (97/7/EC); Article 2 (b) of the unfair terms directive (93/13/EEC); Article 1 (2) a) of the consumer sales directive (99/44/EC); Article 2 (e) of the electronic commerce directive (2000/31/EC); Article 2 e) of the price indication directive (98/6/EC); Article 2 (1) f) of the new timeshare directive (2008/122/EC); Article 2 (D) of the distance marketing of consumer financial services directive (2002/65/EC); Article 2 (a) of the unfair commercial practices directive (2005/29); Article 4 (11) of the new payment services directive (2007/64/EC); Article 2 (4) of the package travel directive (90/314/EEC); Article 3 (a) of the original consumer credit directive and Article 3 (a) of the new consumer credit directive (2008/48/EC).

^*4 Council Directive 85/374/EEC of 25 July 1985 on the approximation of the laws, regulations and administrative provisions of the Member States concerning liability for defective products. – OJ L 210, 7.8.1985, p. 29.

^*5 Council Regulation (EC) No. 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters. Said act replaces the Brussels I convention of 27 September 1968.

^*6 Regulation (EC) No. 593/2008 of the European Parliament and Council of 17 June 2008 on the law applicable to contractual relations (Rome I). – OJ L 177, 4.7.2008.

^*7 Such a definition can be found in the consumer sale, unfair terms, e-commerce, distance marketing of consumer financial services and payment services directives as well as in the new consumer credit directive.

^*8 M. Ebers. The notion of ‘consumer’. – H. Schulte-Nölke, C. Twigg-Flesner, M. Ebers. EC Consumer Law Compendium. Comparative Analysis, 2008. Available at www.eu-consumer-law.org/consumerstudy_part3a_en.pdf (1.7.2011).

^*9 Council Directive 90/314/EEC of 13 June 1990 on package travel, package holidays and package tours. – OJ L 158, 23.6.1990, pp. 0059–0064.

^*10 In transposing the directive, the Estonian legislator managed to avoid terminological confusion in Law of Obligations Act by using the term ‘traveller’ instead of ‘consumer’ in its provisions dealing with package travel contracts (§866 ff.).

^*11 Tarbijakaitseseadus. – RT I 2004, 13, 86; 2010, 31, 158 (in Estonian).

^*12 C. Joustra. Consumer Law. – A. Hartkamp et al. Towards a European Civil Code. Kluwer Law International 1998, pp. 143–144.

^*13 Research Group on the Existing EC Private Law. Principles of the Existing EC Contract Law. Contract II—General Provisions, Delivery of Goods, Package Travel and Payment Services. Munich: Sellier, European Law Publishers 2009, p. 56.

^*14 Belgium, Lithuania, Portugal and Slovakia use a similar approach. See M. Ebers. The notion of ‘consumer’. – H. Schulte-Nölke, C. Twigg-Flesner, M. Ebers. EC Consumer Law Compendium. Comparative Analysis 2008, p. 456.

^*15 Võlaõigusseadus. – RT I 2001, 81, 487; RT I, 4.2.2011, 2 (in Estonian).

^*16 CCSCd, 23.3.2006, 3-2-1-8-06, paragraph 15; 8.12.2009, 3-2-1-126-09, paragraph 12.

^*17 Võlgade ümberkujundamise ja võlakaitse seaduse eelnõu seletuskiri ( Explanatory memorandum to the Draft Act on Restructuring of Debts and Debt Protection). Available at http://www.riigikogu.ee/?page=eelnou&op=ems&emshelp=true&eid=1004095&u=20110530145824 (1.4.2011) (in Estonian).

^*18 Research Group on the Existing EC Private Law. Contract II—Principles of the Existing EC Contract Law. Sellier. European Law Publishers 2009, p. 55.

^*19 Clause 2 1) of the original draft Consumer Protection Act (SE 14, 1.4.2003) defined consumer as follows: a natural person who intends to purchase or purchases and uses goods and services for a purpose which is not directly related to his economic or professional activity.

^*20 One exception to the rule is, e.g., the Electronic Communications Act (elektroonilise side seadus. – RT I 2004, 87, 593; RT I, 23.03.2011, 1 (in Estonian)) whose §2 55) sets out that a consumer is an end-user who is a natural person and who mainly does not use electronic communications services in his or her economic or professional activities. The current regulation does not specify the criteria of predominant use.

^*21 There are 20 of such Member States. See M. Ebers. The notion of ‘consumer’. – H. Schulte-Nölke, C. Twigg-Flesner, M. Ebers. EC Consumer Law Compendium. Comparative Analysis, 2008. Available at www.eu-consumer-law.org/consumerstudy_part3a_en.pdf (30.5.2011).

^*22 Communication of the Commission to the European Parliament and Council. European Contract Law and the revision of the acquis: the way forward. Brussels 11.10.2004. COM(2004) 651 final. – OJ C 14, 20.1.2005, p. 3 ff.

^*23 Directive 85/577/EEC to protect the consumer in respect of contracts negotiated away from business premises; directive 93/13/EEC on unfair terms in consumer contracts; directive 97/7/EC on the protection of consumers in respect of distance contracts; directive 1999/44/EC on certain aspects of the sale of consumer goods and associated guarantees.

^*24 Green Paper on the Review of the Consumer Acqui. Brussels 8.2.2007, COM(2006) final, p. 3.

^*25 Green Paper, pp. 11 and 15.

^*26 Directive 2008/48/EC of the European Parliament and of the Council of 23 April 2008 on credit agreements for consumers and repealing Council Directive 87/102/EEC. – OJ L 133, 22.5.2008, pp. 66–92.

^*27 Directive 2008/122/EC of the European Pa.rliament and of the Council of 14 January 2009 on the protection of consumers in respect of certain aspects of timeshare, long-term holiday product, resale and exchange contract. – OJ L 033, 3.2.2009, pp. 0010–0030.

^*28 Proposal of the EC Commission for a Directive of the European Parliament and of the Council on consumer rights COM(2008) 614 final, 8.10.2008.

^*29 C. von Bar, E. Clive, S. Nölke (eds.). Principles, Definitions and Model Rules of European Private Law. Draft Common Frame of Reference (DCFR). Munich: Selliers 2009.

^*30 Doorstep selling directive and distance contracts directive; in addition, the unfair terms directive and consumer sale directive will be amended; the Commission originally intended to replace all four directives with a new directive.

^*31 M. W. Hesselink. Towards a Sharp Distinction between B2B and B2C? On Consumer, Commercial and General Contract Law after the Consumer Rights Directive (June 8, 2009). – Centre for the Study of European Contract Law Working Paper Series 2009/06.

^*32 Council Directive 93/13/EEC of 5 April 1993 on unfair terms in consumer contracts. – OJ L 095, 21.04.1993, pp. 0029–0034.

^*33 ECJ, 22.11.2001, joined cases C-541/99 and C-542/99, Idealservice, paragraph 17. – ECR 2001, pp. I-09049.

^*34 Council Directive 85/577/EEC of 20 December 1985 to protect the consumer in respect of contracts negotiated away from business premises. – OJ L 372, 31.12.1985, pp. 0031–0033.

^*35 ECJ, 14.3.2003, C-361/89, Patrice Di Pinto, paragraphs 19 and 23. – ECR 1991, p. I-01189.

^*36 See CCSCd, 3-2-1-126-09, paragraph 12; 3-2-1-111-10, paragraph 11; 3-2-1-118-05, paragraph 34. In the latter case the Supreme Court clearly noted that a consumer is also regarded as a natural person under §34 of LOA.

^*37 I. Kull, M. Käerdi, V. Kõve. Võlaõigus I: Üldosa (Law of Obligations I. General Part). Tallinn 2004, p. 122 (in Estonian).

^*38 Commissions Second Progress Report on the Draft Common Frame of Reference, COM(2007) 447 final, 25.7.2007, p. 3.

^*39 Proposal of the EC Commission for a Directive of the European Parliament and of the Council on consumer rights COM(2008) 614 final, 8.10.2008.

^*40 European Economic and Social Committee. Opinion on the Proposal for a Directive of the European Parliament and of the Council on consumer rights (16.7.2009), p. 5.4.1. INT/464 – CESE 1190/2009 – 2008/0196 (COD). Available at http://www.europarl.europa.eu/oeil/FindByProcnum.do?lang=2&procnum=COD/2008/0196 (1.7.2011).

^*41 Draft European Parliament Legislative Resolution on the proposal for a directive of the European Parliament and the Council on consumer rights (COM(2008)0614 – C7-0349/2008 – 2008/0196(COD)). Ordinary legislative procedure—first reading. Amendment 59 (Article 2—point 1). Available at http://www.europarl.europa.eu/sides/getDoc.do?type=REPORT&mode=XML&reference=A7-2011-0038&language=EN (1.7.2011).

^*42 Research Group on the Existing EC Private Law. Contract II—Principles of the Existing EC Contract Law. Sellier. European Law Publishers 2009, p. 59.

^*43 G. Howells, R. Schulze. Modernising and Harmonising Consumer Contract Law. Sellier European Law Publishers 2009, p. 10.

^*44 European Community Note (9.10.2010—working document). The Proposal for a Directive on Consumer Rights: Scope, relationship with national general contract law and relationship with ohter community legislation. Available at http://ec.europa.eu/consumers/rights/docs/note_CDR_en.pdf (1.7.2011).

^*45 Ibid.

^*46 Article 9 (b) ii of directive 85/374 is exceptional in that it speaks about the main purpose of use.

^*47 ECJd 17.3.1998, C-45/96, Bearish Hypotheses- und Wechselbank AG v. E. Dietzinger. – ECR 1998, p. I-1199, paragraph 22. Here, parallels may be drawn regarding the Supreme Court’s case-law on suretyship issued for the purpose of guaranteeing the obligations of a legal person dealt with above in the text.

^*48 ECJd, 3.7.1997, C-269/95, Francesco Benincasa./.Dentalkit Srl.

^*49 ECJd, 20.1.2005, C-464/01, Gruber. – ECJ Cases 2005, p. I-00439.

^*50 M. Ebers. The Notion of ‘consumer’. – H. Schulte-Nölke, C. Twigg-Flesner, M. Ebers. EC Consumer Law Compendium. Comparative Analysis, 2008. Available at www.eu-consumer-law.org/consumerstudy_part3a_en.pdf (30.5.2011).

^*51 OLG Celle, 7 U 193/06. Available at http://app.olg-ol.niedersachsen.de/efundus/volltext.php4?id=4389 (1.4.2011).

^*52 P. Varul et al. Võlaõigusseadus I. Kommenteeritud väljaanne (Law of Obligations Act. Commented Edition). Tallinn: Juura 2006, §34 comment 2 (in Estonian).

^*53 Green Book, Annex I, clause 4.1.

^*54 DCFR, p. 92.

^*55 DCFR, p. 94.

^*56 G. Howells, R. Schulze. Modernising and Harmonising Consumer Contract Law. Sellier European Law Publishers 2009, p. 12; M. B. M. Loos. Review of the European Consumer Acquis. Sellier European Law Publishers 2008, p. 11.

^*57 European Community Note (9.10.2010—working document). The Proposal for a Directive on Consumer Rights: Scope, relationship with national general contract law and relationship with other community legislation. Available at http://ec.europa.eu/consumers/rights/docs/note_CDR_en.pdf (1.4.2011).

^*58 Proposal for a directive of the European Parliament and of the Council on consumer rights (COM(2008)0614 – C6-0349/2008 – 2008/0196(COD). Ordinary legislative procedure: first reading. Amendment 59 Article 2-point 1. Available at http://www.europarl.europa.eu/sides/getDoc.do?type=TA&language=EN&reference=P7-TA-2011-0116 (1.7.2011).

^*59 Seletus kiri võlaõigusseaduse ja teiste seaduste muutmise seaduse eelnõu juurde (Riigikogu XI koosseisu eelnõu 761 SE III) ( Explanatory Memorandum to the Draft Act Amending the Law of Obligations Act and Other Acts (draft 761 SE III of the 11th composition of the Riigikogu)), p. 12. Available at http://www.riigikogu.ee/?page=eelnou&op=ems&emshelp=true&eid=1033413&u=201101312101 (in Estonian).

^*60 BGH NJW 2005, 1045 ff.

^*61 Tsiviilseadustiku üldosa seadus. – RT I 2002, 35, 216; RT I, 6.12.2010, 12 (in Estonian).

^*62 E.g., decisions of the Civil Chamber of the Supreme Court 3-2-1-37-00 (acceptance of unsatisfactory performance, creating an impression that the performance was satisfactory, and later invoking breach of contract); 3-2-1-66-05 (failure to submit timely penalty claims and invoices and subsequent claiming of damages) and 3-2-1-32-06 (invoking the nullity of standard terms by the very user of the standard terms).

^*63 ECJd, 20.1.2005, C-464/01, Gruber, paragraphs 52 and 53. – ECR 2005, p. I-00439.

^*64 The approach employed in the Gruber case is considered too harsh as the court deemed wrong impression given by the consumer out of carelessness to be sufficient to preclude the consumer provisions. See C. Von Bar, E. Clive. Draft Common Frame of Reference (DCFR). München: Sellier 2009, p. 103.

^*65 Green Paper, p. 18.

The jurisdiction and arbitration clauses contained in consumer contracts can significantly limit the constitutional right of consumers to have recourse to the courts for protecting their rights. This particularly applies if such a clause is contained in the standard terms of the contract prepared by the other party, as a result of which the consumer cannot influence the substance of the contract. In such cases, the consumer has in fact been forced to agree that the disputes arising from the contract will be settled in the arbitral tribunal or court chosen by the seller or supplier if the consumer wishes to acquire the desired goods or service. However, such contracts are relatively frequent in the practice of the European countries, which has given rise to the judgements of the European Court of Justice such as Oceano Grupo ^*1 , Mostaza Claro ^*2 , Asturcom ^*3 , Pannon ^*4 and Pénzügyi Lízing. ^*5

The objective of this paper is to analyse how the European Court of Justice has sought to protect consumers against the jurisdiction and arbitration clauses contained in standard terms and what duties would arise from the above-mentioned judgements for the judges of Member States, including Estonia. The paper demonstrates that although the European Union legislator does not have general competence to regulate the civil proceedings of the Member States, the judgements of the European Court of Justice examined have considerable impact on the civil proceedings of the Member States, including the principles of procedural autonomy and the adversary principle of the parties.

2. Jurisdiction and arbitration clauses as unfair contract terms

There is no general prohibition in Estonian law on arbitration or jurisdiction clauses in consumer contracts. However, according to §104 (3) 1) of the Code of Civil Procedure ^*6 (hereinafter referred to as the CCP), the agreement on jurisdiction between the business and the consumer is valid only if concluded after the arising of the dispute. Section 105 of the CCP still enables the court to also accept a jurisdiction agreement concluded before the arising of the dispute on the precondition that the consumer (defendant) responds to the action without contesting jurisdiction and also in cases where the defendant does not respond to the action but participates in a court session without contesting jurisdiction. Yet, Estonian law, somewhat surprisingly, imposes on arbitration agreements only the requirement of a written form (CCP §719 (2)) ^*7 , but does not prohibit entry into them before the arising of a dispute.

A legal situation is considerably more complicated if the agreement on jurisdiction or arbitration is contained in standard terms. The issues related to the standard terms contained in consumer contracts are governed by Council Directive 93/13/EEC on unfair terms in consumer contracts ^*8 (hereinafter referred to as the Directive). Article 3 (1) of the Directive provides that a contractual term which has not been individually negotiated shall be regarded as unfair if, contrary to the requirement of good faith, it causes a significant imbalance in the parties’ rights and obligations arising under the contract, to the detriment of the consumer. Article 3 (3) of the Directive refers to the Annex to the Directive that contains ‘an indicative and non-exhaustive list of the terms which may be regarded as unfair’ ^*9 and according to its clause 1 q, a term which has the object or effect of excluding or hindering the consumer’s right to take legal action, particularly by requiring the consumer to take disputes exclusively to arbitration, can be regarded as unfair.

The provisions of Directive 93/13/EEC have been harmonised in Division 2 of Chapter 2 of the Law of Obligations Act ^*10 (hereinafter referred to as the LOA). According to LOA §42 (1) and (3) 10), a standard term is unfair, hence void, if it deprives the other party of the opportunity to protect his rights in court or unreasonably hinders such opportunity from being exercised. It cannot be unambiguously inferred from the wording of the provision whether and what kind of agreements on arbitration and jurisdiction should be considered as unfair according to Estonian law, while comments on the Law of Obligations Act do not ensure full clarity either. ^*11 Here we have to note that the European Court of Justice is not competent to decide on the unfair nature of a particular arbitration or jurisdiction clause ^*12 , but this can only be done by a national court that has to take into account national legislation and the circumstances of the dispute when forming its opinion, while also observing the general criteria developed by the European Court of Justice for assessing the unfairness of standard terms. ^*13

Opinions vary across different Member States about whether and on what conditions the arbitration clause contained in the standard terms must be considered as unfair for the purposes of the Directive. For example, in German law, the arbitration clauses contained in consumer contracts are not regarded a priori as unfair but must be assessed against, e.g., the distance of the arbitral tribunal from the residence of the consumer and the cost of the procedure for the consumer. ^*14 In Spanish law, however, submission to arbitration other than consumer arbitration, except in the case of arbitration bodies established by statutory provision in respect of a specific sector or circumstances is considered unfair. ^*15

As mentioned above, there is no universal position in Estonian law on when an arbitration clause contained in standard terms of a consumer contract can be regarded as a unfair standard term for the purposes of LOA §42 (3) 10). Neither have such standard terms become common in practice in Estonia, yet. As the first ones are already in place, sooner or later we must develop our own position on this issue.

The author agrees with Advocate General Trstenjak, delivering her opinion in the Asturcom case, that there are serious doubts about the independence and neutrality of the courts of arbitration since arbitrators may possibly have a personal interest in conducting the arbitration proceeding and hence disregarding the voidness of the arbitration clause. ^*16 As the remuneration of arbitrators depends on the conducting of arbitration proceedings as a rule, the arbitrators need not always be willing to establish the voidness of the arbitration clause and refuse to conduct the proceedings. Also, the basis for solving a dispute in arbitration proceedings is the fact that the parties have, of their free will, given up on the adjudication of the matter in national court and submitted the settlement of disputes to the competence of the court of arbitration. The development of the consumer’s free will in the case of standard terms is rather doubtful because of the unequal negotiation position of the parties. Besides these problems, unlike, e.g., in German law ^*17 , the decisions of national arbitral tribunals existing on a permanent basis in Estonia are automatically enforceable (CCP §753 (1)) and hence are subject to the supervision of the national judicial system only on the consumer’s request. ^*18 Since according to European law—as indicated above—the unfair nature of standard terms must be inter alia assessed in the context of domestic law, the author is of the opinion that the latter fact serves as a reason to consider the arbitration clauses contained in standard terms in Estonian law usually as unfair and hence void. A contrary position would seriously undermine the objective set out in the Directive—to effectively protect consumer against unfair standard terms.

It is somewhat more difficult to assess the unfair nature of jurisdiction clauses as here the consumer is not deprived of the opportunity to have recourse to the national judicial system but the consumer’s opportunity to settle the dispute in the court prescribed by law is simply precluded or limited. Although according to Estonian procedural law, the jurisdiction clause contained in standard terms is usually void (CCP §104 (3) 1)) ^*19 , it need not be the same way in the law of other Member States and this has given rise to several references for preliminary ruling lodged with the European Court of Justice. Emphasising that the assessment of the unfair nature of a specific provision lies with the competence of courts of the Member State, the European Court of Justice has still concluded that, e.g., a standard term contained in an agreement entered into between a consumer and undertaking, which confers the jurisdiction for the dispute only to the court in the territorial jurisdiction of which the seller or supplier has its principal place of business could be regarded as unfair. ^*20 Such a clause compels the consumer to accept that the dispute can be settled only in a court that may be located far from the consumer’s place of residence, which in turn complicates the appearance of the consumer before the court. In the case of disputes concerning limited amounts of money, the costs relating to the consumer’s entering an appearance could be a deterrent and cause him to forgo any legal remedy or defence. ^*21 In the context of Estonian law, it may be important to consider these guidelines of the European Court of Justice in the cases in which the agreement on jurisdiction was indeed reached after the arising of the dispute but still on the standard terms drafted by the operator (e.g., the operator has prepared an annex to the contract after the arising of the dispute, containing an agreement on jurisdiction described above).

At the same time, it cannot be precluded that the consumer does desire the settlement of the dispute in the (arbitral) tribunal prescribed by the standard terms, and in such a case, it would be unreasonable to deny the consumer this option from the point of view of procedural economy. That is why in its judgement on Pannon, the Court of Justice assumed the position that the court had to inform the consumer about the invalidity of such a clause and ask whether the consumer wished to settle the dispute in that court regardless thereof. ^*22 Taking into account the principle of interpretation in conformity with the Directive ^*23 , the mere participation of the consumer in the proceeding (cf. CCP §105) is not enough to approve jurisdiction but it is necessary that the court first explain that the standard term prescribing jurisdiction is void and hence the consumer is not required to participate in the proceedings. It is only when the consumer participates in the proceeding regardless of this that the national court may proceed with the discussion of the subject of the matter.

3. Duties of the court in establishing unfairness of standard term

The European Court of Justice has time after time expressed the opinion that the national court may assess the unfairness of standard terms ex officio. ^*24 In later judgements it has already been established that the national court does not only have the right thereto but also an obligation. Hence, the European Court of Justice observed in the Pannon case that the national court had to examine the possible unfairness of the jurisdiction agreement ex officio even if the consumer had not relied on that ^*25 while the position was reasserted in the case Pénzügyi Lízing. ^*26 This also applies in the case in which the court assesses if the case falls within its jurisdiction or not. The point of departure of Estonian law is, in principle, similar, as according to CCP §75 (1), the court must verify the jurisdiction of the matter ex officio anyway. ^*27 The Supreme Court of Estonia has repeatedly stressed that the court must examine the validity of the standard terms even if the parties have not relied on it. ^*28

A distinction must be made between the above question and the question of whether the court must ex officio also examine whether an agreement on arbitration or jurisdiction is a standard term at all (which may not be clear at all if the contract contains just a few clauses). In other words: if the consumer does not at all during the proceedings rely on the fact that standard terms have been imposed on him, does the court still have to examine that issue of its own motion? This question was answered in the case Pénzügyi Lízing. ^*29 Here the court, referring for the preliminary ruling, asked the European Court of Justice whether the national court was required to examine, of its own motion, the unfairness of a contractual term where it had available to it all the legal and factual elements necessary for the task, or the examination of unfairness of a contractual term also meant that the national court was obliged to undertake, of its own motion, an investigation with a view to establishing the factual and legal elements necessary to assess whether a term is unfair. More precisely, the court requested from the European Court of Justice an answer to the question that if the national court itself observed, where the parties to the dispute had made no application to that effect, that a contractual term was potentially unfair, could it undertake, of its own motion, an examination with a view to establishing the factual and legal elements necessary to that examination where the national procedural rules permitted that only if the parties so requested?

When answering the question, it is necessary to first recall the principle of procedural autonomy of the Member States, i.e., that outside a few special cases, the European Court of Justice does not have the general competence to regulate national civil procedures. Traditionally, the adversary principle and the principle of passivity of the court apply to the civil procedures of the Member States, i.e., the parties determine the extent of claim as well as the circumstances and evidence serving as its basis. ^*30 The European Court of Justice has recognised the principle of passivity of the court in the van Schijndel judgement, establishing that the court could act of its own motion in a civil suit only in special cases if required by public interest. ^*31

In the case Pénzügyi Lízing, however, the European Court of Justice arrived at the conclusion that in order to ensure the effectiveness of consumer protection ‘in the exercise of the functions incumbent upon it under the provisions of the Directive, the national court must ascertain whether a contractual term which is the subject of the dispute before it falls within the scope of that Directive’ ^*32 . Hence, the European Court of Justice assumed a position that the national court was required ex officio to examine if standard terms were concerned. The Court stressed that the national court had such an obligation also in case the national procedural law provided otherwise. ^*33 Hence, at this point, the court can no longer proceed from the request of the parties and the adversary principle is replaced by the inquisitorial principle that is in fact not characteristic of civil procedure. Such an approach is not actually new to Estonian judicial practice, although the position of the Supreme Court has varied on this issue over the years. The initial prevailing opinion in judicial practice was that the court was obliged to examine if a term could be a standard term. ^*34 In 2006, the Supreme Court adopted a contrary position, stating that ‘the party who wishes that the court apply to the terms of the contract the regulation of the standard terms in the Law of Obligations Act, must pursuant to LOA §35 (1) and (2) plead and prove the circumstances based on which the term of the contract can be qualified as a standard term according to LOA §35 (1)’. ^*35 It was only recently that the Supreme Court returned, emphasising that ‘upon the reopening of the matter, the court must of its own motion assess also whether the above-mentioned contract clauses are standard terms’ ^*36 . Hence, the present position of the Supreme Court on this issue coincides with that of the European Court of Justice. ^*37

At least to date, the European Court of Justice has not adopted a position whether the obligation to examine ex officio the unfairness of standard terms extends to arbitral tribunals. In legal literature, the unwillingness of the European Court of Justice to express its opinion about this issue has been ascribed to the fact that it is an extremely sensitive issue from the point of view of legal policy relating to the interrelationships between European Union law, national procedural law and arbitration law. ^*38 The author thinks that it would be reasonable for arbitral tribunals to also observe the principles provided in the judgements of the European Court of Justice to preclude—as will be soon demonstrated—the possibility to assess the unfairness of arbitration clauses and as a result annul the judgements of arbitral tribunals later on.

4. Possibility to assess the unfairness of arbitration clauses in post-judgement stage

As a result of the request for preliminary ruling submitted by two Spanish courts, the European Court of Justice had to decide whether it could derive from European Union law that the voidness of unfair arbitration clauses could also be taken into account after the arbitral tribunal had made an award in the case. Moreover, could such an opportunity exist even if the national procedural law of the Member States did not allow it? And if yes, then if and under what circumstances should the court annul an award made by such an arbitral tribunal?

The case of Mostaza Claro gave rise to the question whether in a situation in which the consumer had not at all participated in the arbitration proceedings, but later on submitted an action for the annulment of the award of the arbitral tribunal, a court could ex officio assess the unfair nature of the arbitration clause and annul the award of the arbitral tribunal as a result. In this case, the settlement of disputes related to the contract was subordinated to an arbitral tribunal in the standard terms of a mobile telephone contract entered into with the consumer. The consumer presented her objections to the action in the arbitral proceedings but did not contest the competence of the arbitral tribunal or claim the unfairness of the arbitration agreement. The arbitrator found against her. After that, the consumer contested the arbitration decision before the court, applied for the annulment of the arbitration award and for the first time claimed that the arbitration agreement had served as an unfair standard term and thus the arbitration agreement had been void from the start. ^*39 The court that had received the action for annulment referred to the European Court of Justice for a preliminary ruling and posed the following question:

May the protection of consumers under Council Directive 93/13/EEC […] require the court hearing an action for annulment of an arbitration award determine whether the arbitration agreement is void and to annul the award if it finds that the arbitration agreement contains an unfair term to the consumer’s detriment, when that issue is raised in the action for annulment but was not raised by the consumer in the arbitration proceedings? ^*40

The European Court of Justice stressed that in the case of standard terms, the consumer was in a weaker position vis-à-vis an operator and such an imbalance might only be corrected by positive action unconnected with the actual parties. It was the nature and importance of public interest underlying the protection which the Directive conferred on consumers that justified, according to the European Court of Justice, that the national court could determine of its own motion the unfair nature of the term and compensate this way for the imbalance which existed between the consumer and the seller or supplier. ^*41 As a result, the European Court of Justice assumed the position that the national court seised of an action for annulment of an arbitration award had to determine whether the arbitration agreement was void and annul that award where that agreement contained an unfair term, even though the consumer had not pleaded that invalidity in the course of arbitration proceedings, but only in that of the action for annulment. ^*42

The Estonian court must also consider the positions provided in the Mostaza Claro judgement when interpreting CCP §751 (1) 2). Namely, according to CCP §751 (1) 2), the court shall annul a decision of an arbitral tribunal made in Estonia if the party proves that the arbitration agreement is null and void pursuant to the laws of Estonia or another state, based on whose law the parties agreed to assess the validity of the arbitration agreement. Hence, the possible unfairness of an arbitration agreement contained in standard terms in accordance with LOA §42 (3) 10) must be assessed also when deciding on the action for annulment of the decision of an arbitral tribunal. However, it is questionable if indeed the consumer should prove the voidness of an arbitration clause, as it derives from the wording of CCP §751 (1) 2), or the duty of the court to identify the voidness of standard terms ex officio deriving from the earlier judgements of the European Court of Justice ^*43 should be recognised here as well. The author supports approval of the latter position. This is also supported by the conclusion of the Mostaza Claro judgement, according to which the court ‘must determine whether the arbitration agreement is void’. Hence, CCP §751 (1) 2) is obviously not in accordance with European Union law in that part and, based on the principle of interpretation in conformity with the directive, the provision should be interpreted so that in such a situation as set out in the provision, the court determining the voidness of the arbitration award must determine ex officio whether the arbitration agreement contained in standard terms is void.

While the Mostaza Claro case arose from the problem that the consumer did not rely on the voidness of the arbitration clause during arbitral proceedings, the behaviour of the consumer was even more passive in the Asturcom case. Namely, throughout the arbitration proceedings, the consumer had not expressed any objection to the competence of the court of arbitration or the claim lodged against her. The consumer did not contest the arbitration award in a national court during the period prescribed for that in Spanish law either. Hence, the arbitration award entered into force and was submitted to a national court to be declared enforceable. When deciding on the enforceability of the arbitration award, the court had the question of whether, based on the Directive on unfair terms, it could determine of its own motion whether it was an unfair standard term to the detriment of the consumer and if yes, if it could annul the arbitration award on these grounds. To obtain an answer, the court referred the matter for preliminary ruling to the European Court of Justice. ^*44

This case saw the collision of three important principles that are also recognised by the European Court of Justice: the principle of protection of consumers, the principle of efficient arbitration proceedings and the principle of res judicata, and the European Court of Justice had to seek a reasonable balance between these principles. When expressing its opinion, the European Court of Justice referred to its previous judicial practice, according to which the Member States are generally free to apply national procedural rules (principle of procedural autonomy of the Member States), but the freedom is restricted by a) the principle of effectiveness and b) the principle of equivalence. ^*45 The principle of effectiveness means that the national procedural order of a Member State may not render the exercise of individual rights conferred by European Union law impossible or excessively difficult. According to the principle of equivalence, the national procedural rules of a Member State must not place individuals in a less favourable situation upon exercising the individual right arising from European Union law than those governing domestic actions. ^*46 The European Court of Justice noted that Spanish procedural rules were in line with the principle of effectiveness but identified an opportunity to allow for the limited examination of the validity of arbitration clauses via the principle of equivalence.

The European Court of Justice namely established that pursuant to the principle of equivalence, it was required that the conditions of applying the legal provision of the European Union arising from national law would not be less favourable than the conditions of the application of the national provisions of the same ranking of its own motion. The Court further stated that Article 6 of the Directive on unfair terms served as a provision that was extensively based on public interest (protection of consumers against unfair terms), and that is why Article 6 of the Directive had to be regarded as a provision of equal standing to national rules which rank, within the domestic legal system, as rules of public policy. ^*47 Hence, Article 6 of the Directive, pursuant to which the Member States must ensure that unfair terms are not binding on the consumer, was promoted to the rank of public policy (ordre public).

The procedural laws of many Member States provide for the possibility of annulment of arbitration awards provided that the arbitration award is in conflict with the rules of public policy but not all the Member States allow for its assessment ex officio. ^*48 Against this background, the European Court of Justice established the following rule: inasmuch as the national court or tribunal seised of an action for enforcement of a final arbitration award is required, in accordance with domestic rules of procedure, to assess of its own motion whether an arbitration clause is in conflict with domestic rules of public policy, it is also obliged to assess of its own motion whether that clause is unfair standard term. The Court emphasised that the national court or tribunal was also under such an obligation where, under the domestic legal system, it had a discretion, i.e., right to consider of its own motion whether such a clause was in conflict with national rules of public policy. ^*49

Hence, there cannot be an unambiguous answer to the question whether a court determining the enforceability of an arbitration award can avoid declaring the arbitration award enforceable because of the unfair nature of the arbitration award based on European Union law and the answer will depend on the national procedural law of the Member State. If the national procedural law of a Member State does not allow for the annulment of an arbitration award that is in conflict with national rules of public policy of the court’s own motion, the court cannot do that based on the provisions of consumer protection of the European Union either. ^*50 In an opposite case, the national court must annul the arbitration award. ^*51

According to CCP §751 (2) 1), the court shall annul a decision of an arbitral tribunal based on the request of a party or at the court’s initiative if the court establishes that pursuant to Estonian law, the dispute should not have been adjudicated by an arbitral tribunal. The court has a similar right if the decision of the arbitral tribunal is contrary to Estonian public order or good morals (CCP §751 (2) 2). According to CCP §753 (2), the court shall dismiss a petition for declaring a decision of an arbitral tribunal to be subject to enforcement and shall annul the decision if a cause for annulment of the decision of the arbitral tribunal exists. Thus, considering the rule provided in the Asturcom case and the principle of interpreting in conformity with the Directive, an Estonian court must refrain from declaring the decision of an arbitral tribunal enforceable and annul the decision of its own motion. The positions provided in the Asturcom and Pénzügyi Lízing cases could give grounds to infer an even more extensive right of examination the court: namely, upon the receipt of a request for the enforcement of an arbitration award the court must ex officio examine whether it was a consumer dispute ^*52 , whether the arbitration clause was null and void and if yes, annul the arbitration award of its own motion. According to CCP §753 (1), this can be considered only in the case of a decision by a non-permanent arbitral tribunal, since a decision made in a proceeding of an arbitral tribunal operating in Estonia on a permanent basis is subject to recognition and enforcement without separate recognition and declaration of enforceability by the court.

5. Conclusions

Although the European Court of Justice can interpret general criteria used by the Community legislature in order to define the concept of unfair terms, the European Court of Justice is not competent to decide which arbitration and jurisdiction clauses can be regarded as unfair standard terms in a particular case. This falls within the competence of a national court that must consider, inter alia, the circumstances of a particular case and the context of national law upon the assessment of a standard term. The author supports the opinion that in the context of Estonian law, a standard term that deprives the consumer of the opportunity to have recourse to the courts and instead forces the consumer to settle potential disputes in an arbitral tribunal, must be generally regarded as unfair according to LOA §42 (3) 10) and hence void. The question of regarding an jurisdiction agreement entered into with the consumer as unfair does not play an important role in Estonian law since the agreements on jurisdiction contained in standard terms are, as a rule, void in consumer contracts according to CCP §104 (3) 1) because they were entered into before the arising of the dispute.

In the cases discussed in the paper, the European Court of Justice has significantly interfered with the principle of procedural autonomy and the adversary principle of the Member States. Namely, the European Court of Justice has emphasised several times that in certain cases the national court has to abandon its passive role and examine certain circumstances of its own motion, as it would otherwise be impossible to achieve the objective of consumer protection provided in the Directive on unfair terms. Hence, a national court has a duty to ex officio verify whether the terms of the contract are standard terms. The court must also assess the possible unfairness of standard terms of its own motion. A national court is under certain circumstances obliged to annul an arbitration award made on the basis of an unfair arbitration clause and do this even though the consumer has not contested the competence of the arbitral tribunal during the arbitration proceedings.

The arbitral tribunals should also observe the principles provided in the judgements of the European Court of Justice in order to avoid annulment of arbitration awards later on. This in turn means that it would generally be reasonable for operators to avoid using arbitration clauses in standard contracts entered into with consumers, and negotiate those terms individually.

Notes:

^*1 Case 27.6.2000, joined cases C‑240/98–C‑244/98, Océano Grupo Editorial and Salvat Editores. – ECR 2000, p. I‑4941.

^*2 Case 26.10.2006, C‑168/05, Mostaza Claro v. Centro Móvil Milenium SL. – ECR 2006, p. I‑10421.

^*3 Case 6.10.2009, C‑40/08, Asturcom Telecomunicaciones SL v. Cristina Rodríguez Nogueira. – ECR 2009, p. I‑9579.

^*4 Case 4.6.2009, C-243/08, Pannon GSM Zrt. v. Erzsébet Sustikné Győrfi. – ECR 2009, p. I‑4713.

^*5 Case 9.11.2010, C‑137/08, Pénzügyi Lízing Zrt v. Ferenc Schneider. – OJ C 13, 15.1.2011, p. 2.

^*6 Tsiviilkohtumenetluse seadustik. – RT I 2005, 26, 197; RT I, 30.12.2010, 2 (in Estonian).

^*7 Failure to comply with a formal requirement does not influence the validity of an agreement if the parties agree to the resolution of the dispute by an arbitral tribunal, CCP §719 (3).

^*8 Council Directive 93/13/EEC of 5 April 1993 on unfair terms in consumer contracts. – OJ L 95, 21.4.1993, p. 29.

^*9 Case 1.4.2004, C-237/02, Freiburger Kommunalbauten GmbH Baugesellschaft & Co. KG v. Ludger Hofstetter and Ulrike Hofstetter, paragraph 20. – ECR 2004, p. I-03403.

^*10 Võlaõigusseadus. – RT I 2001, 81, 487; RT I, 4.2.2011, 2 (in Estonian).

^*11 P. Varul, I. Kull, V. Kõve, M. Käerdi. Võlaõigusseadus I. Kommenteeritud väljaanne (Law of Obligations Act I. Commented Edition). Tallinn: Tallinn 2006, p. 156 (in Estonian). Neither does the Estonian Constitution discuss the question whether and to what extent a contract may limit the constitutional right of individuals to have recourse to the courts for the protection of their rights. See E.-J. Truuväli et al. Eesti Vabariigi põhiseadus: Kommenteeritud väljaanne (Constitution of the Republic of Estonia. Commented edition). 2nd, revised edition. Tallinn: Juura 2008, p. 162 ff. (in Estonian).

^*12 Mostaza Claro (Note 2), paragraph 22; Pénzügyi Lízing (Note 5), paragraphs 43, 44.

^*13 Freiburger Kommunalbauten (Note 9), paragraphs 21, 22; Pénzügyi Lízing (Note 1), paragraph 44.

^*14 K. Hilbig. Absoluter Verbraucherschutz bei unzulässigen AGB-Schiedsvereinbarungen? – Zeitschrift für Schiedsverfahren 2010, p. 76. For the law of other Member States, see, e.g., N. Reich. More clarity after ‘Claro’? Arbitration clauses in consumer contracts as an ADR (alternative dispute resolution) mechanism for effective and speedy conflict resolution, or as ‘deni de justice’? – European Review of Contract Law 2007/3, pp. 44–49.

^*15 Mostazo Claro (Note 2), paragraph 11. E.g., in French law, arbitration agreements are not allowed in consumer contracts (see N. Reich (Note 14), p. 47) and in Austrian law, the arbitration clauses contained in standard terms are prohibited (see K. Hilbig (Note 14), p. 75).

^*16 Opinion of Advocate General Verica Trstenjak in Case 14.5.2009, C-40/08, Asturcom Telecomunicaciones SL gegen Cristina Rodríguez Nogueira. Available at http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:62008C0040:DE:HTML (13.4.2011).

^*17 According to §1060 (1) of the German Code of Civil Procedure (ZPO), an arbitral award can be enforced only if it has been declared enforceable by court.

^*18 The CCP does not explain the notion of ‘permanent arbitral tribunals’ and does not establish any criteria for an arbitral tribunal to be considered permanent. Perhaps all the arbitral tribunals that are not ad hoc arbitral tribunals can be considered as permanent arbitral tribunals, hence in principle also an arbitral tribunal having one arbitrator, existing with, e.g., an undertaking, which adjudicates a couple of times a year.

^*19 A regulation analogous to that of Estonia also applies in German law, for details, see C. Meyer. Missbräuchliche Gerichtsstandvereinbarungen in Verbraucherverträgen: Anmerkung zu EuGH Urteil Pannon. – Zeitschrift für Gemeinschaftsprivatrecht 2009, pp. 221–223.

^*20 Pannon (Note 4), paragraph 44; the same also in Oceano (Note 1), paragraph 24. For details, see T. Pfeiffer. Prüfung missbräuchlicher Klauseln von Amts wegen (Gerichtsstand) – Günstigkeitsprinzip nach Wahl des Verbrauchers. – Neue Juristische Wochenschrift 2009, pp. 2367–2369. It must be noted in addition that according to Article II.-9:409 of the Draft Common Frame of Reference, such agreements on jurisdiction are unfair standard terms. According to the DCFR, this is the only standard term included in the black list or automatically void; in the case of all other standard terms, their unfair nature is only presumed, see DCFR Article II.-9:410.

^*21 Oceano (Note 1), paragraph 22.

^*22 Pannon (Note 4), paragraphs 33, 35.

^*23 For this, see, e.g., case 13.11.1990, C-106/89, Marleasing. – ECR 1990, p. I-4135, paragraph 8; ALCSCd, 16.6.2010, 3-3-1-36-10, paragraph 19.

^*24 See Oceano (Note 1), paragraph 29; Case 21.11.2002, C-473/00, Codifis, paragraph 38. – ECR 2002, p. I-10875.

^*25 Pannon (Note 4), paragraph 32.

^*26 Pénzügyi Lízing (Note 5), paragraph 56.

^*27 It is perhaps important to emphasise that this also applies if the case derives from an order for payment procedure, but the consumer lodges a general objection without separate contestation of jurisdiction.

^*28 See CCSCd, 3-2-1-155-05, paragraph 19; 3-2-1-2-08, paragraph 13; 3-2-1-56-08, paragraph 13; 3-2-1-2-11, paragraph 12.

^*29 Pénzügyi Lízing (Note 5), paragraph 25.

^*30 See for details M. Ebers. From Oceano to Asturcom: Mandatory Consumer Law, Ex Officio Application of European Union Law and Res Judicata. – European Review of Private Law 2010/4, pp. 825–828.

^*31 See case 14.12.1995, joined cases C-430-431/93, van Schijndel. – ECR I-4705, p. 21.

^*32 Pénzügyi Lízing (Note 5), paragraphs 49, 56.

^*33 Pénzügyi Lízing (Note 5), paragraph 51. The excess interference by the Court of Justice in the procedural autonomy of the Member States has also been criticised. See T. Pfeiffer. EuGH: Kompetenzen des EuGH bei der Auslegung der Klauselrichtlinie und die Pflicht der nationalen Gerichte zur Amtsermittlung. Kommentierte BGH-Rechtsprechung Lindenmaier-Möhring (LMK) 2010, 311868.

^*34 See CCSCd, 3-2-1-59-05, paragraph 17; 3-2-1-140-05, paragraph 12. A similar position is upheld in the Comments on the Law of Obligations Act (Note 10), pp. 126–127.

^*35 See CCSCd, 3-2-1-150-06, paragraph 17; the same position was repeated in decision 3-2-1-45-07, paragraph 11.

^*36 See CCSCd, 3-2-1-2-11, paragraph 12.

^*37 The position of German law has been rather similar. It is namely held that in principle the consumer has to plead and prove that it was a standard term; yet the obligation is considered fulfilled already after the terms of the contract have been presented to the court in a printed or otherwise reproduced form. See Münchener Kommentar zum Bürgerliches Gesetzbuch. Band 2. Schuldrecht Allgemeiner Teil. Kommentaator: E.-M. Kieninger. 5. Aufl. 2007, §305, margin No. 43.

^*38 C. Mak. Judgement of the Court (First Chamber) of 6 October 2009, Asturcom Telecomunicaciones SL v. Cristina Rodrigues Nogueira, Case C-40/08. – European Review of Contract Law 2010/4, pp. 443–444.

^*39 Mostaza Claro (Note 2), paragraphs 16–18.

^*40 Ibid., paragraph 20.

^*41 Ibid., paragraphs 25, 26, 38.

^*42 Ibid., p 39. The author supposes that the unfairness of an arbitration clause should be taken into account also when a consumer has recourse to the courts against an operator and the operator contests the competence of the court, relying on the existence of the arbitration clause. However, N. Reich (Note 14), p. 43, holds a different opinion.

^*43 For this, see Chapter 3.

^*44 Asturcom (Note 3), paragraphs 20–27.

^*45 Mostaza Claro (Note 2), paragraph 24; Asturcom (Note 3), paragraph 38.

^*46 Mostaza Claro (Note 2), paragraph 24. For the principles, see—particularly in the context of the Asturcom case—H. Schebesta. Does the National Court Know European Law? A Note on Ex Officio Application after Asturcom. – European Review of Private Law 2010/4, pp. 857–859 and 877–878.

^*47 Asturcom (Note 3), paragraphs 49, 52.

^*48 This is disputable, for example, in Spanish law that is why the Spanish court finally decided to declare the arbitration award enforceable. See C. Mak (Note 38), pp. 446–447.

^*49 Asturcom (Note 3), paragraphs 53, 54.

^*50 M. Ebers (Note 30), pp. 840, 846.

^*51 Asturcom (Note 3), end of paragraph 59.

^*52 This could be inferred, inter alia, from the prerequisite provided in paragraph 32 of the Pannon case: ‘where it has available to it the legal and factual elements necessary for that task’, which later in the Pénzügyi Lízing case were defined as the obligation of the court to examine of its own motion if the term falls within the scope of the Directive, see Chapter 3 above.

If a buyer has received from the seller goods that do not conform to the contract, the buyer has, as a rule, recourse to several remedies. According to the Estonian Law of Obligations Act ^*1 (hereinafter referred to as the LOA), if relevant prerequisites are fulfilled, the buyer can have recourse to the following remedies:

–     claim the performance of the contract through replacement or repair of the thing (LOA §222),

–     withhold performance (LOA §111),

–     claim compensation for damage (LOA §115),

–     terminate the contract (LOA §116 (1)) or

–     reduce the price (LOA §112).

The Estonian regulation of sales contracts has been largely founded on the United Nations Convention on Contracts for the International Sale of Goods ^*2 (hereinafter referred to as the CISG), but also on the German, Dutch, Swiss, Italian, etc. law provisions concerning contracts of sale of goods. ^*3 Directive 1999/44/EC ^*4 (hereinafter referred to as the Consumer Sales Directive) has served as the basis regarding contracts entered into with consumers. ^*5 At the same time, the provisions of the Consumer Sales Directive have not always been adopted solely in regard to the regulations on consumer sales but often also in regard to the provisions that are applied to all sales contracts. ^*6

This article focuses on the question of whether the buyer has (both in cases where the buyer is a consumer and where he is not) the right to freely choose between the remedies available to him in the case of defective goods or whether his choice is limited according to the LOA. This question is of interest, above all, with regard to the attempts to harmonise the regulation of consumer rights in the European Union. Namely, the text of the new draft Directive on consumer rights ^*7 (hereinafter referred to as the draft Directive) published in 2008 prescribed the hierarchy of remedies of the buyer like the currently applicable Consumer Sales Directive does. ^*8 Moreover, according to Article 4 of the draft Directive, the Directive was aimed at maximum harmonisation. The issue of the hierarchy of the buyer’s remedies has been excluded from the current text of the directive adopted by the European Parliament. ^*9 However the establishment by a directive of such a hierarchy in all the European Union Member States has been considered for a long time. In several Member States the latter would have brought about a reduction in the level of protection of consumer rights. ^*10

At the same time, on 3 May 2011, the text of the draft version of an Optional Instrument ^*11 (hereinafter referred to as the Optional Instrument) was published, which the European Commission could use as a ‘toolbox’ in the further regulation of European contract law. ^*12 The latest text of the draft version of the Optional Instrument was published on 19 August 2011. ^*13 According to its Article 107 (3) 1), the consumer may use any remedies mentioned in the instrument without granting the seller an option to cure. Hence, it is recommended by the Optional Instrument to provide for the absence of hierarchy of the consumer’s remedies and in addition to that also for the right of the consumer not to grant the seller a possibility to cure his non-performance.

This article shows that according to the LOA, both a buyer who is a consumer and a buyer who is not a consumer have in case of defective goods the right to freely choose a remedy suitable for him subject only to right of the seller to cure the non-performance. This means that the LOA lacks a hierarchy of the buyer’s remedies. The hierarchy of remedies would exist if the buyer was able to use other remedies only after he claimed from the seller performance through replacement or repair (LOA §222 (1)), or granted an additional time for the performance of the obligation (LOA §114 (1)). ^*14 It will also be demonstrated below that the rights of the buyer and the seller are balanced provided that the remedies used by the buyer are not in hierarchy but the buyer’s opportunity to resort to any of the remedies is limited solely by the seller’s right to cure his non-performance.

2. General rule on the buyer’s choice of remedy

A general rule on the freedom to choose a remedy is contained in LOA §101 (2). According to that rule, in the case of non-performance, the obligee may resort to any remedy separately or resort simultaneously to all remedies which arise from law or the contract and which can be invoked simultaneously, unless otherwise provided by law or the contract. Hence it could be concluded that a buyer who has received from the seller defective goods should, as a rule, be able to independently decide to which of the above-mentioned remedies he wishes to resort. However before assuming such a position, it must be identified if, in the case of the sale of defective goods, the law contains special provisions providing for a hierarchy of remedies. In addition it should be noted that the parties are free to agree upon the sequence of using remedies in the contract and thus the right of the buyer to choose the remedy may be subject to such contractual restraints. The latter topic is however not addressed in this article.

The comments on the LOA note about the hierarchy of remedies that in the case of lack of conformity of a thing, the buyer can only request from the seller the performance of the contract under LOA §222 (1) and as a rule other remedies are secondary. ^*15 The EC Consumer Law Compendium sets out that Estonia has adopted the hierarchy of remedies that enables the consumer first to request only replacement or repair. ^*16 However, it remains questionable on what bases the positions described have been assumed, as the LOA contains no specific provision establishing such hierarchy.

As mentioned before, the freedom of the obligor to choose a remedy can be limited by the law itself, as provided in LOA §101 (2). ^*17 The regulation of termination of the contract (LOA §116 (1)) and reduction of price (LOA §112 (1)) in the case of sales contracts will be analysed below in order to trace any references to a hierarchy of remedies. One of the arguments supporting the existence of a hierarchy of remedies has been claimed to be the fact that the Consumer Sales Directive that served as one of the sources when drafting the relevant provisions of the LOA provides for a hierarchy of remedies in the case of defective goods. ^*18 Pursuant to Article 3 of the Directive, the buyer’s primary remedies are the claim for replacement and repair, between which the consumer can choose ^*19 , and only after the replacement or repair become impossible or have not been carried out in timely manner or have brought about significant inconvenience for the consumer can the consumer resort to other remedies. ^*20 The Directive does not provide for the seller’s opportunity to cure the non-performance. ^*21 This article limits itself only to the regulation of the termination of the contract and the reduction of the price, because termination and reduction of the price are secondary remedies according to the Consumer Sales Directive. At the same time, the Consumer Sales Directive does not regulate issues related to compensation for damages. Because of this and the volume of the article, the aspects of compensation for damage are not addressed here. Besides, the provisions governing the buyer’s remaining remedies in the Chapter of the LOA on the sales contract lack references to the potential hierarchy of remedies. ^*22

3. Buyer’s right of termination in system of remedies

3.1. Regulation of the buyer’s right of termination in LOA

According to LOA §116 (1), a fundamental breach of contract serves as a precondition for the termination of a contract. There are no other preconditions (e.g., grant of an additional term before termination) arising from LOA §116 (1).

LOA §116 (2) provides an illustrative list of circumstances considered as a fundamental breach of the contract. ^*23 LOA §116 (2) 1)–4) list the types of non-performance by an obligor that are regarded as fundamental. Thus, for example, according to LOA §116 (2) 2), a breach of contract is fundamental if strict compliance with the obligation which has not been performed is the precondition for the other party’s continued interest in the performance of the contract, and according to LOA §116 (2) 3), if non-performance of an obligation was intentional or due to gross negligence. LOA §116 (2) 5) sets out that a breach of contract is fundamental also if the other party fails to perform any obligation thereof during an additional term for performance granted by the obligee or gives notice that he will not perform the obligation during such term. Hence, LOA §116 (2) 5) enables the obligee to terminate the contract also upon a non-fundamental breach if the obligor does not eliminate the breach during the specified term. ^*24

Although in the event indicated in LOA §116 (2) 5), the obligee must essentially require performance of the obligation before termination of the contract, this does not serve as a general prerequisite for the right to termination. It only provides for an additional possibility for the termination of the contract in cases where grounds of fundamental breach of the contract would in normal circumstances be denied. Hence, if it is not possible to prove that the obligor has failed to perform an obligation in a manner that would allow to consider his breach fundamental as defined in LOA §116 (2) 1)–4), the obligee may nevertheless turn the obligor’s breach of contract into a fundamental breach by requiring performance of the obligation within an additional term. LOA §116 (2) 5) is of considerable practical importance as it enables the obligee to free himself of the uncertainty about whether the non-performance by the obligor can be regarded as fundamental. ^*25 Taking into account that the wording of the situations of fundamental breach of contract listed in LOA §116 (2) 1)–4) are rather general, it is often advisable in practice to grant the obligor an additional term for performance before terminating contract. Despite the abovementioned, the author of this article is not of the opinion that the objective of LOA §116 (2) 5) is to establish a hierarchy among the remedies, but the idea is rather to give the obligee an additional opportunity to demonstrate the fundamental nature of the non-performance by the obligor.

In the event of fundamental breach of contract as defined in LOA §116 (2) 1)–4), the buyer should be entitled to terminate the contract by submitting a timely declaration of termination and the previous claim for performance does not serve as a prerequisite for the right of termination (there is no hierarchy between the two remedies). The buyer’s immediate right of termination is in certain cases limited only by LOA §116 (4), according to which termination of a contract without granting an additional term for performance is prohibited if the damage suffered by the non-performing party in the case of the termination would be disproportionate in relation to the expenses incurred in the performance or preparation for the performance of the obligation. However, terminating the contract without granting an additional term is permitted in the event of non-performance of an obligation specified in §116 (2) 2) ^*26 or if the other party gives notice that the party will not perform the obligation thereof (the second sentence of LOA §116 (4)). Consequently, a buyer who wishes to terminate the contract must generally consider whether the termination would cause disproportionate damage to the seller in relation to the expenses incurred in the performance or preparation for the performance of the obligation. If the exercise of the right of termination would cause such damage and the situations specified in the second sentence of LOA §116 (4) are not present, the buyer must grant the seller an additional term for performing the obligation (essentially demand performance before terminating the contract). Since it can be difficult for the buyer to foresee the damage incurred by the seller, it has generally been considered advisable to grant an additional term before termination. ^*27 Yet, it has been noted in legal literature about LOA §116 (2) and (4) that the effect of the provisions on each other remains unclear. ^*28

Based on the above, the author is of the opinion that according to the regulation of the termination of the contract contained in the general part of the LOA, the buyer is not, as a rule, obliged to grant an additional term for the seller to perform the obligation and, as a result, there is no hierarchy between the claim for performance and the right of termination. Nevertheless such a hierarchy may arise in regard to specific circumstances emerging in practice and in such cases it may be necessary to grant the obligor an additional term for performance based on LOA §116 (2) 5) or (4).

However, there is often no need for a buyer receiving goods that lack conformity to demonstrate a fundamental breach of the seller based on LOA §116. Namely, in the case of a sales contract, LOA §223 (1) and (2) complement the list provided in LOA §116 (2) with additional cases of a fundamental breach of the contract. ^*29 According to LOA §223 (1), the seller is deemed to be in fundamental breach of the sales contract also if

1)    the repair or replacement of a thing is not possible or fails;

2)   the seller refuses to repair or replace a thing without good reason, or

3)   the seller fails to repair or replace a thing within a reasonable period of time after the seller is notified of the lack of conformity.

Pursuant to the third alternative of LOA §223 (1), the seller is already in fundamental breach of his obligations if he fails to give notice of his intention to cure at his own initiative within a reasonable period of time (LOA §107 ^*30 ). The same position has been expressed by the Supreme Court, noting that according to LOA §223 (1) it is not only the case where the claim of the buyer for repair or replacement of the goods has not been fulfilled by the seller that constitutes a fundamental breach. ^*31

Pursuant to LOA §223 (2), in the event of consumer sale, any unreasonable inconvenience caused to the buyer by the repair or replacement of a thing is also deemed to be a fundamental breach of the contract by the seller. The situation of such significant breach of contract as described above can only occur if the buyer has demanded from the seller performance of the obligation (LOA §222 (1)) or the seller has undertaken to cure the breach at his own initiative (LOA §107).

If the breach of contract is fundamental, as defined in LOA §223 (1) or (2), then LOA §223 (3) enables the buyer to terminate the contract without granting the seller an additional term for performing the obligation. This provision once more reinforces the principle arising from LOA §116, according to which the buyer is not obliged to grant the seller an additional term for performing the obligation as a precondition for terminating the contract in the case of a fundamental breach of contract. ^*32

Based on the above, one can assume the position that there is generally no hierarchy between the claim for the replacement or repair of the goods and the termination of the contract as remedies in the event of a fundamental breach of the contract. Provided that the preconditions for resorting to these remedies are met, the buyer can choose termination as the first remedy to be used. The buyer must naturally take into account the principle of good faith set out in LOA §6 and General Part of the Civil Code Act ^*33 (hereinafter referred to as the GPCCA) §138. This means that one cannot terminate the contract when one acts in bad faith while exercising the right to terminate. ^*34 At the same time the buyer’s obligation to act in good faith when terminating the contract certainly does not make the claim to perform the obligation (LOA §222 (1)) a primary remedy and the right of termination a secondary remedy.

3.2. Buyer’s right to terminate in CISG, DCFR ^*35 , Consumer Sales Directive, German Civil Code ^*36 and Optional Instrument

The above position is also supported by the sources serving as the basis for the regulation of the termination of the sales contract in the LOA. For example, CISG Article 49 (1) a) provides for the buyer’s opportunity to avoid the contract if the object of the contract is defective. According to this, the buyer may declare the contract avoided if the failure by the seller to perform any of his obligations amounts to a fundamental breach of the contract. The wording of the provision, or the provision in conjunction with the other provisions of the CISG, does not suggest that the buyer should have previously claimed replacement or repair of the goods. Hence, the CISG is also subject to the rule that the termination may be the first remedy to be used by the buyer ^*37 , while the seller can limit the opportunity of the buyer to resort to it only by exercising his right of cure (CISG Article 48). ^*38

Chapter 4 of Part A in Book IV that governs contracts for sale in the DCFR does not provide for major differences compared to the general regulation of remedies in the DCFR. ^*39 Article III.-3:502 (1) of the DCFR gives grounds for termination; according to it, the creditor may terminate the contract if the debtor’s non-performance of a contractual obligation is fundamental. In the case of deficiencies that represent a fundamental breach of the contract, the buyer can resort to termination of the contract as a primary remedy and need not first claim replacement or repair or grant an additional term for the performance of the obligation. Hence, the rule is that the buyer can choose any remedy if the necessary preconditions exist. ^*40

The regulation of BGB §440 ^*41 and Articles 3 (3) and (5) of the Consumer Sales Directive that served as the basis for drafting the relevant provisions on the contract of sale in the LOA differ from those described. The main difference here is that pursuant to both the Consumer Sales Directive ^*42 as well as BGB §440 ^*43 the previous claim for the replacement or repair of the goods and the failure to satisfy the claim by the seller generally serves as a precondition for the buyer’s right of termination. ^*44 Thus termination is a secondary remedy according to the Consumer Sales Directive as well as the BGB. However as the Consumer Sales Directive requires minimum harmonisation ^*45 , the regulation of the LOA diverging from the Directive in favour of the buyer is allowed with regard to consumer sales and in accordance with the EU law.

The above once more confirms the author’s opinion that pursuant to the LOA, the buyer who received non-conforming goods need not as a rule claim performance before exercising the right of termination if the breach is fundamental. Namely, in the BGB and the Consumer Sales Directive, the hierarchy of remedies arises directly from the wording of the provisions concerned. As demonstrated, the wording of the LOA does not point to such a hierarchy. Hence, the approach chosen in the LOA is similar to that of the CISG ^*46 , and according to this, the buyer usually has the freedom to choose between different remedies.

Article 115 of the Optional Instrument does not prescribe the previous claim for performance as a precondition for the exercise of the buyer’s right of termination either. Unlike the regulation of the LOA, according to Article 115 (2) of the Optional Instrument, the consumer may terminate the contract in case of non-conformity of the goods, unless the non-conformity is insignificant. In case of non-consumer sales, subsection 1 still provides for fundamental breach as a precondition for termination. Thus under the Optional Instrument in the case of a consumer sale, the buyer enjoys a considerably wider opportunity of terminating the contract than permitted by the legislation and model act described above.

The absence of hierarchy between the buyer’s right of termination and the claim for performance as described above helps appropriately balance the rights of the buyer and the seller. For example, it is possible that a fundamental breach by the seller is so serious that the buyer has lost trust in the seller. In such a case, it is reasonable to enable the buyer to terminate the contract immediately without compelling him to submit an additional claim. ^*47 The buyer may also incur additional damage or inconvenience over time while he is waiting for the replacement or repair. ^*48

4. Buyer’s right to reduce the price in LOA’s system of remedies

The general regulation concerning the reduction of the price is in LOA §112. According to subsection 1, a party may reduce the price, if the party accepts defective performance. This section does not require the obligee to grant an additional term (LOA §114) for the obligor to perform the obligation before reducing the price. Yet, it has been established that the obligation to grant an additional term may in certain cases arise from the principle of good faith (LOA §6, GPCCA §138). ^*49 However, granting an additional term for the performance of an obligation is not, as a rule, a precondition for the obligee to reduce the price.

When it comes to reducing the price, an additional regulation ^*50 is provided by LOA §224, according to which the buyer may not reduce the purchase price:

1)    if the seller repairs the thing or replaces it (clause 1);

2)   if the buyer unreasonably refuses to accept the proposal of the seller concerning the repair or replacement of the thing (clause 2), or

3)   upon the purchase of a used thing which is sold by public auction (clause 3).

According to legal literature clauses 1 and 2 essentially preclude the reduction of the price by the buyer if the seller is willing to repair or replace the non-conforming goods and does it within a reasonable period of time. ^*51

LOA §224 2) precludes the buyer’s right to reduce the price if the seller has offered cure (LOA §107). Clause 1 precludes the buyer’s opportunity to reduce the price if the seller has cured the non-performance or the buyer has claimed replacement or repair and the seller has satisfied the relevant claim. Hence, the provision does not oblige the buyer to claim replacement or repair of the goods before reducing the price. In other words, the wording of LOA §224 does not presume that the buyer has claimed the replacement or repair of the goods prior to the reduction of the price. However, proceeding from the above, one must agree with the position taken in the comment on the LOA that the seller’s right to replace or repair (to cure within the meaning of LOA §107) the non-performance has priority over the right of the buyer to reduce the price. ^*52

The opinion that the buyer need not claim replacement or repair of the non-conforming goods before reducing the price is also supported by CISG Article 50 that serves as an example for drafting the relevant provisions of the LOA. Pursuant to the second sentence of Article 50, the buyer’s right to reduce the price is precluded if the seller cures any failure to perform his obligations or offers cure and the buyer refuses it without good reason. This means that even if the buyer has made a declaration of price reduction along with the notice of lack of conformity of the goods, the seller may still replace or repair the goods. ^*53 Therefore, CISG Article 50 does not impose the previous claim for replacement or repair as a precondition for reduction of price (Article 46 (2) and (3)). ^*54

Just as in the LOA, price reduction is not a secondary remedy in the DCFR either. As in Estonian law, the right to price reduction in the DCFR is limited only by the seller’s right to cure the non-performance (through replacement or repair of the goods). ^*55 The regulation found in the Optional Instrument is generally similar. Article 121 of the Optional Instrument does not impose on the buyer the obligation to claim performance before reduction of the price. The regulation of the BGB on the other hand differs from the above by setting out the priority of the claim for performance over the right of price reduction. ^*56 The relevant regulation of the BGB, however, has not served as a source for drafting LOA §224. ^*57

The regulation according to which the price reduction is not a secondary remedy must be regarded as appropriate. In such a case, the buyer retains the opportunity to keep the defective goods, if he so wishes (and if the seller does not wish to cure his non-performance) and start to use them immediately without having to claim the repair or replacement of the goods, which can be time-consuming.

5. Impact of the seller’s right to cure on buyer’s right to terminate the contract and reduce the price

From the seller’s point of view, the buyer’s freedom to choose without limitations a remedy suitable for him brings about a degree of uncertainty and can disproportionally damage the seller’s rights. As demonstrated before, according to the Consumer Sales Directive ^*58 and BGB ^*59 , the seller is protected against the negative consequences that could be inflicted on him through granting priority to the claim for performance. In both instruments, the legislator has aimed at keeping the contract unchanged, i.e. as it was concluded between parties, for as long as possible (pacta sunt servanda). ^*60

This has also been the goal of the Estonian legislator when drafting the LOA ^*61 , but the means to achieve the goal have been different. Namely, the LOA enables the seller to cure non-performance (LOA §107). ^*62 According to that, the seller who has delivered defective goods can replace or repair the non-performance as long as the buyer has not terminated the contract or claimed damages in lieu of performance. LOA §107 (1) 1)–3) and (2) impose additional limitations on the exercise of the right to cure. Allowing cure for the obligor is an expression of the general principle of good faith ^*63 and like in the LOA it has also been provided for in the CISG ^*64 and DCFR. ^*65

If the buyer wishes to reduce the price based on the defects of the goods sold, his right to resort to the remedy is limited by the seller’s right to cure his non-performance (LOA §224).

The relationship between the seller’s right to cure and the buyer’s right to terminate the contract is different. As noted above, LOA §107 (1) provides that the seller may cure the non-performance, as long as the buyer has not terminated the contract. Thus, at the moment the buyer has terminated the contract on the grounds of the defects of the goods, the seller is deprived of the right to cure the non-performance. ^*66 However, as long as the buyer has not made a declaration of termination (but has, for example, notified the seller of the defects under LOA §220), the seller as a rule has the right to cure his non-performance. Also, according to Estonian law, the buyer’s right to terminate the contract can depend on the seller’s right to cure through the notion of fundamental breach of contract set out in LOA §223 (1) and (2). As indicated in Section 3.1 of this article, the seller can in certain cases preclude the buyer’s opportunity to rely on fundamental breach of the seller’s obligation if the seller cures the non-performance during a reasonable period of time.

Hence, the LOA usually prioritises the seller’s right to perform his obligation according to the contract. By granting the seller the opportunity to cure the breach the buyer’s recourse to different remedies is limited. Unlike claiming performance and granting an additional term for performing the obligation, in the case of cure of the breach the initiative to eliminate the breach originates from the seller. ^*67 That is why the seller’s right to cure does not, in the opinion of the author of the article, introduce hierarchy to the system of buyer’s remedies. Due to the nature of right to cure, it cannot be considered interchangeable with the use of a remedy by the buyer. ^*68

For example, let us imagine a situation in which the buyer buys goods that turn out to be defective and informs the seller of the defect (LOA §220). The notice about the defect does not contain any other information. By submitting the notice, the buyer has in fact given the seller an opportunity to cure the non-performance but has not claimed performance from him (replacement or repair). In order to consider the claim of performance submitted, the buyer’s relevant expression of will would be required. However, this cannot be identified merely from a notice about defects. Consequently, there is no hierarchy between the buyer’s remedies in the LOA, yet the buyer’s right to choose any remedy is as a rule limited by the seller’s right to cure his non-performance.

The consumer sales regulation provided in the Optional Instrument differs from the description presented above. Article 110 (2) provides for the seller an opportunity to cure his non-performance. At the same time, according to Article 107 (3) 1), the seller’s right to cure is precluded if the failure concerns a consumer sales contract. However, if the buyer is not a consumer, the seller has been granted an opportunity to cure his non-performance (subsection 2). Thus, in the case of a consumer sale, the consumer is pursuant to the Optional Instrument absolutely free to opt for any remedies of those specified in Article 107. The author of this article believes that this may have increased the level of protection of consumer rights but the seller has been forced to a considerably unfavourable position. This is especially true, considering the fact that according to the Optional Instrument, a consumer is entitled to terminate the contract already in the case of an insignificant breach of the contract (see Section 3.2 of this article).

The opportunity to cure after the failure to perform an obligation gives the seller as a rule in the case of non-performance an opportunity to perform his obligation and thereby preclude the buyer’s right to resort to different remedies. The author of this article finds that the system is justified and appropriately balances the rights of the buyer and the seller. ^*69 It would be disproportionate to impose on the buyer an obligation to submit claim for performance against the seller before using other remedies because the seller has failed to perform his obligation. However, if the buyer is allowed to resort to any remedy, this can considerably damage the seller’s interests. ^*70 Hence, it is appropriate to limit the buyer’s rights, for example, through granting the seller the right to cure. ^*71 The advantage of the right to cure over prioritising the claim for performance is the fact that it compels the seller to act quickly to cure the non-performance, which also brings prompter legal clarity for the buyer about whether the seller intends to perform his obligation or not. ^*72

Through cure, the seller has an opportunity to limit the buyer’s options to use remedies but he must do it at his own initiative. Since he is also the party in breach of the contract, it should be in his own interests to do anything to perform his obligation as required (cure the non-performance). ^*73 At the same time, the permission to cure should not bring about considerable damage to the interests of either parties (including the reduction in the level of protection of the rights of a buyer who is a consumer) because both parties have the performance of the contract as their primary interest. Permitting cure is also economically justified. ^*74

6. Conclusions

The analysis of the LOA provisions contained in the article showed that according to the regulation of a sales contract, the remedies that the buyer could use in the case of lack of conformity are not in hierarchical relationship to each other. There is no hierarchy among the buyer’s remedies regardless of whether it is a consumer sales contract or any other sales contract.

The buyer’s freedom of choice does not unreasonably damage the seller’s rights either. Namely, according to LOA §107, the seller as a rule has the right to cure his non-performance as long as the buyer has not terminated the contract with the seller. By this, the Estonian legislator has provided as a rule for the seller an opportunity to limit the buyer’s choice of remedies if the seller so wishes. Pursuant to LOA §107 (3), the buyer may usually not use other remedies apart from withholding performance as long as the seller cures his non-performance within the framework prescribed by law (above all, during a reasonable period of time after giving notice of his intention to cure). The regulation described thus grants the seller an opportunity to perform his obligation after the initial failure to do so.

It may be concluded that both the buyer’s and seller’s interests have been taken into account when drafting the regulation of the sales contract in the LOA and that the interests have been successfully balanced through a system of remedies, complemented by the seller’s right to cure. Keeping in mind the developments of the European law specified at the beginning of the article, it would thus not be correct to impose on the buyer an obligation by all means to claim performance from the seller (also through the obligation to grant an additional term) before resorting to other remedies, as it had been prescribed by the draft directive on consumer rights for a long time. The demand for granting an additional term for performance as a prerequisite for using remedies is justified in legal systems in which the obligor is not entitled to the opportunity to cure (e.g., BGB). At the same time the value of the solution offered in the Optional Instrument, according to which a buyer who is a consumer could terminate the contract without granting the seller an opportunity to cure, should be seriously considered. This would ensure a great freedom of choice for the consumer but would probably leave the sellers’ interests unprotected to a considerable degree.

Notes:

^*1 Võlaõigusseadus. – RT I 2001, 81, 487; RT I, 4.4.2011, 3 (in Estonian).

^*2 RT II 1993, 21/22, 52.

^*3 116 SE I. Võlaõigusseadus (Law of Obligations Act. Draft). Available at http://web.riigikogu.ee/ems/saros-bin/mgetdoc?itemid=991610001&login=proov&password=&system=ems&server=rag-ne11 (28.3.2010) (in Estonian).

^*4 Directive 1999/44/EC of the European Parliament and of the Council of 25 May 1999 on certain aspects of the sale of consumer goods and associated guarantees. – OJ L 171, 7.7.1999, pp. 12–16.

^*5 116 SE I. Law of Obligations Act. Draft (Note 3).

^*6 See, e.g., LOA §222 (demand for performance of contract as a remedy), §223 (fundamental breach of contract by the seller).

^*7 COM(2008) 614, 8.10.2008.

^*8 Proposal for a Directive of the European Parliament and the Council on Consumer Rights COM (2008) 614 final. Explanatory Memorandum; Proposal on Consumer Rights Directive, preamble, No. 4, p. 3.

^*9 European Parliament legislative resolution of 23 June 2011 on the proposal for a directive of the European Parliament and of the Council on consumer rights (COM(2008)0614 – C6-0349/2008 – 2008/0196(COD)). Available at http://www.europarl.europa.eu/sides/getDoc.do?pubRef=-//EP//TEXT+TA+P7-TA-2011-0293+0+DOC+XML+V0//EN&language=EN#BKMD-21 (25.7.2011).

^*10 K. Lilleholt. Notes on the Proposal for a New Directive on Consumer Rights. – European Review of Private Law 2009/3, p. 338.

^*11 A European contract law for consumers and businesses: Publication of the results of the feasibility study carried out by the Expert Group on European contract law for stakeholders’ and legal practitioners’ Feedback. Available at http://ec.europa.eu/justice/policies/consumer/docs/explanatory_no-te_results_feasibility_study_05_2011_en.pdf (24.5.2011).

^*12 Ibid., p. 8.

^*13 Available at http://ec.europa.eu/justice/contract/files/feasibility-study_en.pdf (5.10.2011).

^*14 E.g., M. Schmidt-Kessel is in the same position concerning the furnishing of the notion of hierarchy with the content. The Right to Specific Performance under the DCFR. – G. Wagner (ed.). The Common Frame of Reference: A View from Law and Economics. Sellier 2009, p. 85.

^*15 P. Varul et al. Võlaõigusseadus II. Kommenteeritud väljaanne (Law of Obligations II. Commented Edition). Tallinn: Juura 2007, p. 37 (in Estonian).

^*16 H. Schulte-Nölke, M. Ebers (eds.). EC Consumer Law Compendium. The Consumer Aquis and its Transposition in the Member States. Sellier 2008, p. 427.

^*17 P. Varul et al. Võlaõigusseadus I. Kommenteeritud väljaanne (Law of Obligations Act I. Commented edition). Juura 2006, p. 321 (in Estonian).

^*18 See M. Kingisepp. Tarbijakaitseõigus Euroopa Liidus ja Eestis (Consumer protection law in the European Union and Estonia). Läte 2011, p. 232 (in Estonian).

^*19 M. Loos. Consumer Sales Law in the Proposal for a Consumer Rights Directive. – European Review of Private Law 2010/1, p. 37; C. Twigg-Flesner. Fit For Purpose? The Proposals on Sale. – Modernising and Harmonising Consumer Contract Law. G. Howells, R. Schulze (eds.). Sellier 2009, p. 159; E. Grabitz, M. Hilf, M. Nettesheim. Das Recht der Europäischen Union. München 2009, Article 3, margin No. 20; R. O. Sale. Italienische Rechtsprechung zum Gemeinschaftsprivatrecht – Der Verbrauchsgüterkauf zwischen Haftung des Verkäufers für Vertragswidrigkeit und Schutz des Verkäufers nach der Geltendmachung der dem Verbraucher zustehenden Rechtsbehelfe. – GPR 2010/3, p. 146.

^*20 M. Loos (Note 19), p. 36; C. M. Bianca, S. Grundmann. EU Sales Directive. Commentary. Intersentia 2002, p. 159, 168; Directorate General for International Policies. The Potential Impact of the Consumer Rights Directive on Member States’ Contract Law, 2009, p. 25. Available at http://www.directsellingeurope.eu/public/Analysis_and_background_do-cuments/European%20Parliament%20-%20study_ConsRights_ImpactMS_09.pdf (7.4.2011).

^*21 M. Schmidt-Kessel. Remedies for Breach of Contract in European Private Law. – R. Schulze. New Features in Contract Law. Sellier 2007, p. 195.

^*22 Here LOA §222 serves as an exception, providing for in its subsection 2 that when choosing between the replacement and repair of the thing, the buyer may in first place claim the repair if the given contract of sale is not a contract of consumer sale. However, the provision does not have a wider impact on the buyer’s right to use different remedies.

^*23 V. Kõve. Draft Common Frame of Reference and Estonian Law of Obligations Act: Similarities and Differences in the System of Contractual Liability. – Juridica International 2008 (XIV), p. 204.

^*24 It is not, however acceptable, to allow for termination under LOA §116 (2) 5) if the breach is of inferior importance in respect of the entire contract. P. Varul et al. (Note 17), p. 402.

^*25 P. Varul et al. (Note 17), p. 402.

^*26 According to LOA §116 (2) 2), breach is a fundamental breach of contract if strict compliance with the obligation which has not been performed is the precondition for the other party’s continued interest in the performance of the contract.

^*27 P. Varul et al. (Note 17), p. 402.

^*28 V. Kõve (Note 23), p. 204.

^*29 P. Varul et al. (Note 15), p. 63.

^*30 According to LOA §107 (1), a party who fails to perform a contractual obligation may cure the non-performance, including improving or replacing defective performance, as long as the other party has not terminated or cancelled the contract or demanded compensation for damage in lieu of performance, provided that:

1) cure is reasonable in the circumstances, and

2) cure does not cause unreasonable inconvenience or expenses to the injured party, and

3) the injured party has no legitimate interest in refusing cure.

According to subsection 3 of the same section, the injured party may withhold performance as of receipt of the notice of cure until completion or failure of the cure. During the time for cure, the injured party may use other remedies only if these are not inconsistent with the cure.

^*31 CCSCd, 3-2-1-11-10, paragraph 11. – RT III 2010, 11, 80 (in Estonian).

^*32 See also P. Varul et al. (Note 15), p. 64.

^*33 Tsiviilseadustiku üldosa seadus. – RT I 2002, 35, 216; RT I , 06.12.2010, 12 (in Estonian).

^*34 P. Varul et al. (Note 17), p. 402.

^*35 C. von Bar, E. Clive, H. Schulte-Nölke (eds.). Principles, Definitions and Model Rules of European Private Law: Draft Common Frame of Reference. Interim Outline Edition. München: Sellier 2009.

^*36 Bürgerliches Gesetzbuch, adopted on 18.8.1896. – RGBl., p. 195.

^*37 P. Schlechtriem, I. Schwenzer. Commentay on the UN Convention on the International Sale of Goods (CISG). 3rd ed. Oxford University Press 2010, Article 45, comment 12; T. Bachmann. Die elektive Konkurrenz. Berlin: Duncker & Humbolt 2010, p. 238; A. Mittmann. Einheitliches UN-Kaufrecht und Europäische Verbrauchsgüterkauf-Richtlinie. Peter Lang 2004, p. 135.

^*38 K. Lilleholt (Note 10), p. 338; B. S. Markesinis, H. Unberath, A. Johnston. The German Law of Contract. A Comparative Treatise. 2nd ed. Hart Publishing 2006, p. 389.

^*39 M. Loos (Note 19), p. 27.

^*40 Ibid., p. 28.

^*41 P. Varul et al. (Note 15), p. 63.

^*42 M. Loos (Note 19), p. 37; E. Grabitz, M. Hilf, M. Nettesheim (Note 19), Article 3, margin No. 20.

^*43 H. P. Westermann. Münchener Kommentar zum BGB. 5. Aufl. München: Verlag C. H. Beck 2008, §440, margin No. 3, 5.

^*44 Ibid., §434, margin No. 47; H. Amann, G. Brambring, C. Hertel. Der Schuldrechtsferorm in der Vertragspraxis. München: Verlag C. H. Beck 2002, p. 114; S. Lorenz. Selbstvornahme der Mängelbeseitigung im Kaufrecht. NJW 2003, 1417; U. Verweyen. Die Käufrechtsbehelfe des UN-Kaufrechts im Vergleich zu denen des neuen internen deutschen Handelskaufsrechts aus Sicht eines deutschen Warenexporteurs. Peter Lang 2005, p. 72.

^*45 H. Schulte-Nölke, M. Ebers (eds.) (Note 16), p. 427.

^*46 H. Sivesand. The Buyers Remedies for Non-Conforming Goods. Should There be Free Choice or are Restrictions Necessary? Vol 2. Sellier 2005, p. 127.

^*47 M. Loos (Note 19), p. 41.

^*48 Ibid., p. 40.

^*49 P. Varul et al. (Note 17), p. 374.

^*50 P. Varul et al. (Note 15), p. 65.

^*51 Ibid.

^*52 In relation with the contractor’ liability for work that lacks conformity with the contract, this has been stated by the Supreme Court in its decision 3-2-1-148-08, paragraph 12. – RT III 2009, 10, 73 (in Estonian).

^*53 P. Schlechtriem, I. Schwenzer (Note 37), Article 50, comment 7.

^*54 T. Bachmann (Note 37), p. 232; P. Schlechtriem, I. Schwenzer (Note 37), Article 47, comment 2.

^*55 C. von Bar, E. Clive, H. Schulte-Nölke (eds.) (Note 35), p. 1341.

^*56 H. P. Westermann (Note 43), §441, margin No. 1; T. Bachmann (Note 37), p. 233; B. S. Markesinis, H. Unberath, A. Johnston (Note 38), p. 510.

^*57 P. Varul et al. (Note 15), p. 65.

^*58 E. Grabitz, M. Hilf, M. Nettesheim (Note 19), Article 3, margin No. 15.

^*59 B. S. Markesinis, H. Unberath, A. Johnston (Note 38), p. 514.

^*60 E. Grabitz, M. Hilf, M. Nettesheim (Note 19), Article 3, margin No. 20; B. S. Markesinis, H. Unberath, A. Johnston (Note 38), p. 514.

^*61 P. Varul et al. (Note 17), p. 341.

^*62 See Note 29.

^*63 P. Varul et al. (Note 17), p. 341.

^*64 Article 48.

^*65 Article III.-3:202.

^*66 P. Varul et al. (Note 17), p. 343.

^*67 P. Huber. CISG – the Structure of Remedies. Rabels Zeitschrift für ausländisches und internationales Privatrecht. Bd. 71. 2007, p. 21.

^*68 M. Schmidt-Kessel (Note 14), p. 194.

^*69 In relation to the DCFR, the same position has been assumed by C. von Bar, E. Clive, H. Schulte-Nölke (eds.) (Note 35), p. 812.

^*70 V. Mak. Performance-Oriented Remedies in European Sale of Goods Law. Article Publishing 2009, p. 184.

^*71 H. Sivesand (Note 46), p. 117.

^*72 Ibid.

^*73 V. Mak (Note 70), p. 202.

^*74 Ibid., p. 149.

One of the purposes of the European Union (hereinafter referred to as the EU) is a functioning uniform internal market. However, a common market is largely based on contract law. The EU contains 27 different contract laws, making cross-border activity complicated and expensive. The European Commission is about to develop the Common Frame of Reference (hereinafter referred to as the CFR) for European contract law, one part of which is to cover insurance contracts (in the Principles of European Insurance Contract Law ^*1 —hereinafter referred to as the PEICL ^*2 ). The 1.7.2010 Green Paper from the European Commission on policy options for progress toward a European contract law for consumers and businesses ^*3 was presented for public discussion with the aim of consultation concerning the possibilities for further developing the field of European contract law. In its Green Paper, the European Commission in essence proposes seven approaches for improving the coherence of European contract law, including an optional European Contract Law (i.e., the ‘28th regime’ ^*4 or the ‘second regime’), which would be an optional instrument for consumers and undertakings to apply in their contractual relationships. ^*5 This optional right would be an alternative to the national contract laws in force. Specialists in European insurance law have found that the 28th regime idea should be guided from within the field of insurance law ^*6 ; a similar conclusion was stated in the European Economic and Social Committee’s opinion on the European Insurance Contract. ^*7 The Green Paper solicited 319 distinct positions in the course of public consultations. ^*8 As to the answers given in relation to the questions on the PEICL presented for the Green Paper, it may be assumed that the central issue from the point of view of the insurance sector is not whether to support the optional instrument and whether it should be applied to B2B and/or B2C contracts but whether the PEICL protects the consumers/policyholders too radically ^*9 and whether a win–win situation is involved. The objective of this article is to analyse the PEICL with respect to the question of whether the policyholder is protected too radically, studying the differences of the PEICL from the regulation of insurance contracts ^*10 in the Estonian Law of Obligations Act ^*11 (hereinafter referred to as the LOA). On account of the question asked in the title of this piece and in view of criticism of the PEICL by some of the European interest groups, the author limits this analysis in accordance with the norms of the PEICL that are more beneficial to the policyholder than is the regulation of the LOA. ^*12

2. Pre-insurance-contract legal relationships

2.1. The pre-contractual information duty of the policyholder

There are two main ways to regulate the pre-contractual information duty in insurance contracts:

1)    the insurer shall present a questionnaire to the policyholder and the policyholder answers all of the questions;

2)   the policyholder shall inform the insurer of everything relevant. ^*13

According to Article 2:101 (1) of the PEICL, the applicant, when concluding the contract, shall inform the insurer of circumstances of which he is or ought to be aware, and which are the subject of clear and precise questions put to him by the insurer. ^*14 Therefore, the PEICL prefers the person responsible to be the insurer. The objective of the PEICL regulation is to present an insurer with a right to ask for information on the circumstances of risk and to decide either to enter into an insurance contract or to decline it on the basis of the information given. ^*15 The selection of the question method stipulated in Article 2:101 is mostly explained by the fact that it is considerably more difficult for a policyholder to assess what kind of information is relevant in evaluation of insurance risk. Imposing an obligation on the insurer to ask clear and precise questions is more likely to reduce unnecessary transaction costs and rule out later disputes between the insurer and a policyholder. Therefore, in present insurance practice, the method of asking questions is more deep-seated than the so-called own-initiative rule that was a valid law in most European countries until recently. ^*16

In Estonia, according to the LOA’s §440 (1), upon entering into a contract, the policyholder shall inform the insurer of all circumstances known to the policyholder that, on account of their nature, may influence the insurer’s decision to enter into the contract or to enter into the contract on the agreed terms (material circumstances). Material circumstances are presumed ^*17 to be circumstances concerning which the insurer has directly requested information in a format that can be reproduced in writing.

The biggest difference between the PEICL and the relevant part of the LOA’s regulation is that in the Law of Obligations Act, the own-initiative rule has been chosen and the LOA covers only information that is known to the policyholder, whereas in the PEICL, information that is known or should be known to the policyholder is also important. ^*18 The author considers the PEICL approach preferable, since the so-called own-initiative rule brings about too many disputes in practice, because the policyholder may not always know of what the insurer needs to be informed.

Article 2:102 of the PEICL specifies that when the policyholder is in breach of the duty to inform, the insurer shall be entitled to propose a reasonable variation of the contract or to terminate the contract. To this end, the policyholder has a right to reject the proposal within one month after receipt of the notice. In the latter case, the insurer shall be entitled to terminate the contract within one month. As a rule, the right of withdrawal is limited in the PEICL to only cases of a wrongful breach. However, the PEICL also provides a right of withdrawal if the insurer is able to prove that it would not have concluded the contract if it had known the information concerned.

According to the LOA, providing wrongful information on the material circumstances has two consequences. If a policyholder has wrongfully violated the obligation to inform, the insurer may withdraw from the contract (LOA §441 (1)). If the violation of the obligation to notify on the part of the policyholder is not due to the fault of the policyholder, the law (LOA §460) provides that the insurer has a right to increase the insurance premium. ^*19

Therefore, there are two main differences between the LOA and the PEICL regulation. Firstly, the LOA—unlike the PEICL—does not enable the insurer to demand changing of the insurance premium (i.e., changing of the contract) if the policyholder has wrongfully violated its duty to inform. Secondly, the LOA does not provide a right of withdrawal if the insurer is able to prove that it would not have concluded the contract had it known the information concerned. The author holds that the LOA’s strict and favourable regulation with regard to termination of the contract is not justified by any means, nor is it in the best interests of either of the parties to the contract. ^*20

If a policyholder violates the pre-contractual obligation to inform and this is discovered after an insured event, the insurer has a right under §442 (2) of the LOA to withdraw from the contract on the basis specified in §441 of the LOA also after the insured event has occurred. However, the insurer is not freed from its obligation to fulfil the contract if the circumstances in relation to which the information was not provided had no bearing on the occurrence of the insured event and do not preclude or restrict the validity of the insurer’s performance obligation. The second sentence of the LOA’s §442 (2), considering assessment of exclusion or limitation, provides that, among other things, account shall be taken also of the ratio of the insurance premiums paid to those insurance premiums that should have been paid if information concerning the circumstances had been provided. ^*21

When one is contrasting §442 and §460 of the LOA with the PEICL’s Article 2:102 (5), it can be concluded that withdrawal from the contract by the insurer due to the policyholder’s violation of the duty to inform and also obligation to pay insurance premiums in the situation in which an insured event has taken place are almost identical.

Article 2:104 of the PEICL regulates the situation in which an insurer has been led to conclude the contract by the policyholder’s fraudulent breach of the pre-contractual duty to inform. The PEICL does not directly define fraudulent behaviour, and the comments refer to Article 4:107 (2) of the Principles of European Contract Law ^*22 (hereinafter referred to as the PECL), according to which a party’s representation or non-disclosure is fraudulent if it was intended to deceive. ^*23 According to Article 2:104, an insurer has as many as three ways to respond to fraudulent violation. The first of these is to do nothing and let the contract continue. The second and third derive from Article 2:101 and enable the insurer to change the contract (and withdraw from the contract if it fails) or to withdraw from the contract. It is worth emphasising that the PEICL does not demand that the contract be invalid and gives the insurer an opportunity to decide whether it wishes to withdraw from the contract or would rather give preference to an increase in the insurance premium. If the insurer wishes to withdraw from the contract, this is done retroactively and presents the parties with an opportunity to demand reversal of decision. ^*24 Nevertheless, the PEICL’s Article 2:104 deviates from the PECL regulation from the point of reversal of decision and maintains the insurer’s right to keep those insurance premiums already paid and to demand payment of the insurance premiums due. ^*25 Provision of said right has been justified by a need to prevent fraudulent behaviour and not let insurers develop the following way of thinking: ‘If the fraud succeeds, I’ll benefit and if it doesn’t, I’ll lose nothing (except in the case in which no insured event has taken place).’ Fraud that had no influence on the decision of the insurer (because the information was irrelevant or the insurer was aware of its incorrectness) also has no consequences. ^*26 No situation similar to the one of Article 2:104 of the PEICL has been regulated in the Law of Obligations Act; only §441 (5) of the LOA refers to the right of the insurer to terminate the contract due to deception on the basis of §94 of the General Part of the Civil Code Act. The author holds that this rigid regulation of the LOA, favouring termination of contract and not enabling the insurer still to offer insurance cover to the policyholder, is also not justified.

2.2. The policyholder’s pre-contractual duty to warn

According to Article 2:202 of the PEICL, the insurer shall warn the applicant of any inconsistencies between the cover offered and the applicant’s requirements of which the insurer is or ought to be aware. The insurer shall, analogously, warn the applicant that cover will not begin until the contract is concluded, if he is or ought to be aware of the fact that the applicant mistakenly believes that the cover commenced at the time the application was submitted (Article 2:203). In both cases, not providing said warning may create a situation in which the insurer is responsible for all losses due to this violation and the policyholder has a right to terminate the agreement. The Law of Obligations Act has no such direct consumer-protecting regulation, and the author considers this to be a problematic issue in Estonian law. Article 2:502 of the PEICL states that if the terms of the insurance policy differ from those in the policyholder’s application or any prior agreement between the parties, such differences in the later materials shall be deemed to have been given assent by the policyholder; unless he objects within one month of receipt of the policy, he is deemed to have accepted the changes. Section 436 of the LOA includes a similar regulation, but in Estonian law, the term for the policyholder’s presentation of an objection is limited to 14 days.

3. Contractual legal relationships in insurance

3.1. The period of insurance

According to the PEICL, in Article 2:601, the duration of the insurance contract is one year (except for personal insurance); however, the parties may agree on a different period if indicated by the nature of the risk. The EU tries to avoid the US mistakes in which insurers conclude too many short-term contracts that enable them to change the insurance premium often. ^*27 An example of consideration of this ‘nature of the risk’ is travel insurance, which is generally shorter in duration, since the risk of an insured event taking place is set in a shorter time period than one year.

After the one-year period, the contract shall be extended according to the PEICL’s Article 2:602 unless:

–     the insurer has given written notice to the contrary at least one month before the end of the contract period stating the reasons for its decision or

–     the policyholder has given written notice to the contrary by, at the latest, the day the contract period expires or within one month after having received the insurer’s premium invoice, whichever date is later. In the latter case, the one-month period shall start to run only if it has been clearly stated on the invoice in bold print.

On the basis of §453 of the LOA, it is presumed that the insurance period is one year, but since this is a dispositive norm, the parties may agree on a shorter or a longer insurance period.

Therefore, according to Estonian law, insurers basically have an opportunity to enter into short-term-insurance-period contracts and, in so doing, change the insurance payments. However, short-term (i.e., with a term of less than a year) insurance contracts (excl. travel insurance) are not widespread, in the author’s experience. When comparing the insurance period regulations of the LOA and the PEICL, the author holds that, both for consumer protection reasons and for purposes of optimising the expenses of the insurer, the PEICL regulation should be preferred. The corresponding regulation of the LOA does not impose a duty on the insurer to ensure constant control for the policyholder via constant insurance protection (in relation to the case of an automatically extended contract, the PEICL states that the client shall receive notification for the next period, or a declaration of cancellation, but according to the LOA, the insurer has no obligation to inform the client of the contract ending). This also damages mortgagees, since it has to check annually whether a policyholder has entered into a new contract. The interest of an insurer in the case of an automatically extended insurance contract may lie in cancellation of new pre‑contractual negotiations and reduction of costs thereby.

3.2. Withdrawal from an insurance contract after conclusion of the contract

According to the PEICL’s Article 2:303, the policyholder shall be entitled to avoid the contract by giving written notice within two weeks after receipt of the insurance contract documents. This principle is not applicable to contracts that last one month or less, and to contracts that have been extended; also it pertains to certain particular types of contract (i.e., civil liability insurance, group insurance, and immediate insurance cover). LOA §433 (1) also provides a two-week withdrawal right for a policyholder. At the same time, the imperatively provided right of withdrawal in Estonian law is limited to contracts that last more than one year, and life insurance contracts. Therefore, the biggest difference between the PEICL and the LOA’s regulation lies in the fact that the PEICL provides policyholders with a right to withdraw in two weeks also from contracts with a validity period of 31 days to one year. The corresponding regulation of the PEICL is based on Article 6 of Council Directive 2002/65/EC of the European Parliament and of the Council. ^*28 Although non-life insurance contracts of less than one year are not commonplace in Estonia (except in the field of travel insurance), the author holds that, for purposes of consumer protection, the Law of Obligations Act should be changed in relation to its discrepancy with Article 6 of Council Directive 2002/65/EC of the European Parliament and of the Council.

3.3. Payment of the insurance premium

In comparison of the Law of Obligations Act and the PEICL’s regulation as to payment of periodic premiums, it becomes evident that the system currently in place in Estonia is formally more liberal from the insurer’s point of view and more disadvantageous to the policyholder. Namely, the PEICL, in Article 5:102 (1), states that, as the first operation, the insurer shall issue an invoice stating the precise amount of the premium due as well as the date of payment and the place for payment; after the premium falls due, the insurer sends a reminder to the policyholder of the precise amount of premium due, granting an additional payment term of at least two weeks and describing the legal consequences of not making the payment, and once the additional period has expired without payment having been made, the insurer has a right to release itself from the obligation to fulfil the corresponding duties. The Law of Obligations Act does not require that an invoice be issued by the insurer. In practice, there have been cases in Estonia in which insurers present all subsequent premiums in one invoice and no separate invoices are issued for subsequent payments. ^*29 In the interests of legal clarity, the author holds that the PEICL regulation should be preferred.

Article 5:104 of the PEICL provides that if an insurance contract is terminated before the contract period has ended, the insurer shall be entitled to premiums in respect of only the part of the period prior to termination. ^*30 Section 459 of the LOA (in an imperative provision in accordance with §427 (1) of the LOA) provides that if a contract is terminated prematurely during a term of insurance by cancellation or withdrawal or for any other reason, the insurer is entitled to only those insurance premium for the time up to the termination of the contract. However, most of Estonia’s insurance undertakings state the so-called operation costs in their standard terms. ^*31 Therefore, it can be asked whether collecting such additional amounts is legitimate or not. If one takes the position that the insurer has a right to establish operation costs, it would lead to a situation in which technically it would be possible to state in the standard terms that the amount of the so-called operations cost is as high as any insurance premium payment or the remaining insurance premiums through to the end of the period. It would be possible thus to create a situation in which the policyholder is obliged to pay insurance premiums for the whole insurance period regardless of the contract having been terminated. The author finds that, to address this issue, it would be reasonable to specify the terms of both the LOA and the PEICL additionally so as to avoid differing interpretations such that the insurance premium would always be returned on the pro rata temporis principle to a day’s accuracy and with prohibition of deduction of any expenses.

3.4. Transfer of property

Section 494 of the LOA sets forth imperatively that if a policyholder transfers an insured thing, all the policyholder’s rights and obligations arising from the insurance contract transfer to the acquirer of the thing. At that point, an insurer may cancel the insurance contract with respect to the acquirer of the thing within one month of becoming aware of the transfer of the thing if the insurer gives at least one month’s notice of the cancellation (see §495 (1) of the LOA), and the acquirer of an insured thing may cancel the insurance contract by the end of the current period of insurance within one month from acquiring the thing (§495 (2) of the LOA). Thus, the LOA does not enable the acquirer of an insured thing to choose an acceptable insurer before the new period of insurance (which may theoretically mean that said person would have to remain in a relationship with an unwanted insurer for a year). If the insurer is not notified of the transfer of the thing in time, the insurer shall be released from its performance obligation according to §496 (2) of the LOA if an insured event occurs more than a month after the time when the insurer should have received corresponding notice. The corresponding regulation of the PEICL is considerably more flexible and concentrates more on the policyholder. Except in cases of inheritance and of agreement among the parties (insurer, policyholder, and acquirer) (Article 12:102 (3)), the insurance contract shall be terminated one month after the time of transfer, unless the policyholder and transferee agree on earlier termination (Article 12:102 (1)). The flexible approach of the PEICL is justified by the fact that an insurer cannot be forced to accept a policyholder whom it does not like and the acquirer may have just cause for not entering into a binding agreement that may not protect its economic interests. ^*32 The author agrees with these explanations and therefore considers the regulation in the PEICL to be more reasonable.

3.5. Agreed value

Insurable value is of central importance in property insurance contracts. The PEICL’s Article 8:101 (2) provides that the insurer shall not be obliged to pay more than the amount necessary to indemnify losses actually suffered by the insured. In cases of agreed value, the agreed value shall be applied even if it is greater than the actual losses, with the proviso that no deception occurred or false information was given in agreement upon the agreed value. However, according to §480 (3) of the LOA, the agreed value shall not be deemed to be insurable value if, at the time of the occurrence of the insured event, it differs significantly from the actual insured value. In that case, the actual insured value applies. Therefore, the biggest difference between the PEICL and the LOA is in the fact that in cases of agreed value, the PEICL enables compensation for more than the actual value, where no deception or false information was involved in the agreement upon the agreed value. The author considers the corresponding PEICL regulation to be justified—the insurer is the professional in concluding an insurance contract, having, through its practice, fuller awareness of the value of the assets. Conventionally, an insurance payment is related to insurance value (this is not a linear relationship). By leaving the policyholder to bear the responsibility in cases of an agreed value determined in consequence of a mistake (or with knowledge but without intent to deceive), the balance of contractual rights is severely disrupted, since where the right to increase the insurance premium addressed in §481 of the LOA applies, it is difficult, if not even impossible, for the policyholder to calculate or check the amount of the actual insurance premium.

4. Legal relationships before an insured event

4.1. The policyholder’s non-compliance with precautionary measures

According to the PEICL’s Article 4:102 (1), a clause providing that in the event of non‑compliance with a precautionary measure the insurer shall be entitled to terminate the contract shall be without effect unless the policyholder or the insured has breached its obligation with intent to cause the loss or recklessly ^*33 and with knowledge that the loss would probably result. This blanket standard term according to the PEICL is used by insurers in Estonia. Since in the Law of Obligations Act, an insurer’s withdrawal from the contract has been regulated in a mostly imperative manner and ignoring of the precautionary measures is not a basis for withdrawal, whether such a standard term is applicable under Estonian law is called into question. There is no judicial practice concerning this matter at the level of the Estonian Supreme Court.

Article 4:103 (1) of the PEICL provides that a clause stating that non-compliance with a precautionary measure exempts the insurer partially or completely from liability shall have effect only to the extent that the loss was caused by the non-compliance of the policyholder or insured with intent to cause the loss, or recklessly and with knowledge that the loss probably would result. Therefore, the PEICL considerably limits the insurer’s opportunity to exercise legal remedies in the case of non-compliance with precautionary measures—that is, use of conditions that require the policyholder to behave in one or another way before an insured event takes place. It expressly states those conditions in which the insurer is freed from the obligation to compensate for the damage if the policyholder does not meet some prerequisite. It also addresses conditions for confirmation of doing (or not doing) something upon the breach of which the insurance cover terminates. When we compare the conditions of Estonian insurance undertakings with this provision ^*34 , it emerges that the former are hostile toward the policyholder in comparison to the PEICL regulation. Numerous Estonian insurance undertakings ^*35 present a long catalogue of precautionary measures and their catalogue of cases of the insurer’s exemption from liability states that non-compliance with any requirement of the precautionary measure catalogue (regardless of the form of guilt) causes the insurer to be exempted from liability. From the standpoint of protection of the policyholder’s interests, the author finds the PEICL approach to be significantly more reasonable than the Estonian regulation is. The PEICL’s Article 4:103 (2) states that in cases of a clear clause providing for reduction of the insurance money according to the degree of fault, the policyholder or the insured, as the case may be, shall be entitled to insurance money in respect of any loss caused by negligent non-compliance with a precautionary measure. If the insurance contract contains no corresponding clause, the policyholder shall be entitled to the insurance money in its full extent. The commented edition of the PEICL ^*36 emphasises that, in line with the main philosophy of insurance, an insurance contract is concluded not only to cover an incidental risk of accident but also for negligent conduct. Clause 452 (2) (1) of the LOA provides that the insurer is released from the performance obligation upon the occurrence of an insured event if the policyholder has violated an obligation or if the violation has affected the occurrence of damage or the extent thereof. The PEICL’s Article 9:101, which provides for consideration of the policyholder’s fault in causing a loss, is analogous with its Article 4:103; i.e., only intent and recklessness may be taken as a basis for reduction of denial of indemnity, and reduction on account of recklessness requires a corresponding clear provision in the terms of the insurance policy. Thus the PEICL principle concerning consequences of intent or recklessness goes further than the Law of Obligations Act regulation that only presumes a cause-and-effect relationship between violation and loss (or guilt). Problems of non-compliance with precautionary measures are the main reason for which an insurer refuses indemnity. T herefore, the author considers the PEICL approach justified in comparison with the LOA.

5. Legal relationships after the insured event

5.1. Notice of an insured event

According to Article 6:101 of the PEICL, delay in notification of an insured event may reduce the money payable only to the extent in which the insurer proves that it has been prejudiced by undue delay. If a deadline for notifying of an insured event is set in the insurance conditions, it shall be reasonable and no less than five days.

In the standard terms of Estonian insurance undertakings, notice of an insured event is given a deadline of 2 ^*37 –5 ^*38 work days for reporting as soon as possible. ^*39 If a policyholder violates the obligation, stated in §448 (1) of the LOA, to inform the insurer immediately of the insured event, the result according to §449 (2) of the LOA is to immediately release it from its performance obligation, given that the violation was intentional. The author finds that, in essence, almost any delay in notice is intentional (unless the policyholder is ill, etc.) and, therefore, the insurer is almost always formally released from the performance obligation according to Estonian law in cases of late notice. In the event of unintentional violation, §449 (1) of the LOA, analogously with the PEICL’s Article 6:101 (3), enables expressing the indemnity in the extent to which the insurer was prejudiced by undue delay. Since §448 of the LOA is a dispositive provision, the insurers have, in essence, an opportunity to set unreasonable deadlines for policyholders (i.e., the two-work-day deadline referred to above). The Supreme Court of the Republic of Estonia, in its decision 3-2-1-56-01 (before the Law of Obligations Act entered into force), affirmed the insurer’s right to refuse an indemnity in a situation in which the policyholder notifies the insurer of the insured event after the deadline set in the standard terms and the delayed notice has no further impact on verification of the insurer’s performance obligation. Therefore, the author considers the PEICL’s regulation significantly more preferable as to reasoning and for consumer protection. ^*40

5.2. Indemnity and the limitation period for claims

As to making the performance obligation of an insurer executable, the Law of Obligations Act’s regulation is not as clear as the PEICL’s. Namely, §450 (1) of the LOA states that becoming collectable depends on completion of the process of determining the extent of the insurer’s performance, but if the corresponding activities have not been completed by one month after notice of the insured event, subsection 3 of the same section states that the policyholder may only request—in essence—a minimal advance payment that the insurer should pay. In practice, it enables avoiding advance payments to an insurer that acts with malice or making these payments to only an insignificant extent by also contesting the minimum extent of the loss to be indemnified. The insurer’s obligation to perform a contract also comes into effect if, two months after notifying the insurer of the insured event, the policyholder requests an explanation from the insurer as to why the process of determining the extent of performance has not yet been completed and the insurer fails to respond to the enquiry within one month (LOA §450 (2)). However, the insurer can deviate easily from this obligation by acting maliciously and replying on the circumstances of the formal procedure. According to Article 6:103 (1) of the PEICL, the insurer shall take all reasonable steps to settle a claim promptly. Unless the insurer rejects a claim or defers acceptance of a claim through written notice giving reasons for its decision within one month after receipt of the relevant documents and other information, the claim shall be deemed to have been accepted (6:103 (1)). Therefore, under the PEICL, making the insurer’s performance obligation executable is extended only when the insurer has not received the necessary information and documentation—receipt of these is ensured by the co-operation obligation of the policyholder stated in Article 6:102 of the PEICL. ^*41

Unlike the LOA’s insurance contract regulation, the PEICL directly stipulates the right of a policyholder to demand compensation for losses due to delay in payment of the indemnity (Article 6:105 (2))—i.e., in a situation in which the insurer acted in bad faith. In many European legal orders, a principle applies that the insurer is not responsible for losses caused by delay in indemnity and the insurer shall pay only the interest. However, in the modern legal literature, one finds that the insurer should compensate for all forecast losses that arise through delay in indemnity since the business activity of the insurer is to offer protection to the policyholder and the insurers are aware that delays in indemnity cause some sort of loss. ^*42 Although formally the policyholder could issue an analogous claim to the insurer for the compensation of damage on the basis of the general part of the Law of Obligations Act, the author holds that a corresponding regulation concerning insurance contracts (even as a reference to the general part) would preventively protect the interests of policyholders better. ^*43 Subsection 475 (1) of the LOA sets forth the rule that the limitation period for claims arising from an insurance contract is three years and that period commences at the end of the calendar year during which the claim falls due. However, §475 (3) of the LOA states that if the insurer denies the application for performance of its obligation, the insurer shall be released from the performance obligation if the policyholder does not file an action for compulsory performance of the obligation within one year, where the insurer has, in its response, informed the policyholder of the legal consequences of the expiry of the one-year term. Article 7:102 of the PEICL provides that action for insurance benefits shall be prescribed after a period of three years from the time when the insurer makes or is deemed to have made a final decision on the claim, or 10 years from the occurrence of the insured event. This enables emphasising that the corresponding regulation of the LOA ^*44 is significantly more to the disadvantage of the policyholder than is the PEICL. When we consider the applicable Estonian regulation, it can be pointed out also that the right of claim of the insurer against the policyholder (unpaid insurance premiums, subrogations, etc.) is three years (according to the regulation in §146 (1) of the General Part of the Civil Code Act ^*45 ), whereas the right of claim of the policyholder applies in certain cases (LOA §475 (3)) for only one year—such inequality in the limitation periods ignores the equality of the parties. It is important to emphasise that Article 7:101 of the PEICL sets a shorter deadline for the right of claim of an insurance premium by the insurer: one year. In the PEICL’s commented ^*46 edition, it is pointed out that the shorter deadline can be justified by the fact that policyholders generally pay the insurance premiums in parts (whether monthly or quarterly).

6. Conclusions

As an answer to the question stated in the title, the author finds that the PEICL as the European ius commune is considerably more consumer-oriented and flexible than the LOA is and limits the insurer’s liberation from the performance obligation. However, this is not excessively radical; it balances the interests and opportunities of the parties reasonably. A win–win situation cannot be found in simply optimising the insurer’s expenses by unifying the insurance products into a ‘pan-European insurance product’. While one achieves a ‘win’ in terms of savings in expenses, the other party too has to win something. As a rule, the policyholder cannot be compared with the insurer in its knowledge and economic capabilities; therefore, protection of its rights needs heightened attention. The author holds that both saving on expenses and simplicity of provision of the service in the pan-European activity should be the arguments for insurers choosing the PEICL even in a situation in which the national law would be more favourable to them.

Notes:

^*1 Principles of European Insurance Contract Law: Project Group ‘Restatement of European Insurance Contract Law’, Common Frame of Reference. Chapter III, Section IX1. Insurance Contract. Available at http://restatement.info/cfr/Draft-CFR-Insurance-Contract-17122007-FINAL.pdf (1.3.2011).

^*2 The general part (I part) and the non-life insurance part (II part) of the PEICL were first published and presented to the European Commission in December 2007, the III and the IV part are not complete yet, but will include life insurance and civil liability insurance (Project Group ‘Restatement of European Insurance Contract Law’. Available at http://www.restatement.info/ (1.3.2011).

^*3 European Commission, COM(2010)348 final. Available at http://eur-lex.europa.eu/Notice.do?mode=dbl&lang=et&lng1=et,enlng2=bg,cs,da,de,el,en,es,et,fi,fr,hu,it,lt,-lv,mt,nl,pl,pt,ro,sk,sl,sv,&val=518792:cs&page=1&hwords=Euroopa+le-pingu%C3%B5iguse+loomise+v%C3%B5imalused+tarbijate+ja+ettev%C3%B5tjate+jaoks%7E (1.3.2011).

^*4 ‘Internal’ Contract Law of the 27 Member States vs. the 28th regime, that is, a 2nd regime for all Member States. The European optional procedure will become a part of the national law of the Member States similarly to other sources of the European law. The 2nd regime would give the parties an opportunity to choose between two regimes of national contract law, one of which is being enforced by the legislator of the Member State and another by the European legislator. This is an alternative to traditional approximation of legislations.

^*5 Also see the opinion of the European Economic and Social Committee on ‘28th regime—an alternative allowing less lawmaking at Community level’ (own-initiative opinion). – OJ C 21, 21.1.2011, p. 26.

^*6 J. Basedow, J. Birds, M. Clark et al. Principles of European Insurance Contract Law (PEICL). Sellier European Law Publishers 2009, p. lxi. Also see the opinion of the Project Group ‘Restatement of European insurance contract law’ on the Green Paper. Available at http://ec.europa.eu/justice/news/consulting_public/0052/contributions-/284_en.pdf (8.6.2011).

^*7 OJ C 157, 28.6.2005, p. 1.

^*8 European Commission. Available at http://ec.europa.eu/justice/news/consulting_-public/news_consulting_0052_en.htm (8.6.2011).

^*9 German Insurance Association, European Financial Services Round Table and Allianz SE find in their response to the Green Paper in supporting the idea of PEICL that since PEICL is not based on a win-win situation and protects the policyholders too radically, it needs further development (available at http://ec.europa.eu/justice/news/consulting_public/0052/contributions-/163_en.pdf (8.6.2011), http://www.efr.be/documents%5Cpublication%5C72.2011.02.%20EFR%20signed%20letter%20on%20European%20Cont-ract%20Law%2031.01.2011.2011.pdf (8.6.2011) and http://ec.europa.eu/justice/news/consulting_public/0052/contribu-tions/7_en.pdf (8.6.2011)). However, the Belgium Consumer Protection Board has pointed out that in comparison with the Belgian law, PEICL has advantages in only four issues, whereas in 16 issues the national law protects the consumer better (H. Heiss, M. Lakhan. Principles of European Insurance Contract Law: A Model Optional Instrument. Sellier European Law Publishers 2011, p. 96).

^*10 Although the number of actions against insurers filed to courts of first instance is not high, the insurers tend to lose the disputes in litigations. In accordance with the judicial decision registry held by the Ministry of Justice of the Republic of Estonia (available at http://www.kohus.ee/kohtulahendid/index.aspx (10.6.2011)) and the specifying request for information submitted by the author to the Ministry of Justice (the author possesses the 9.6.2011 reply), the insurers won 28% of the cases in relation to the proceedings of the court of first instance in 2009, and on 72% of the cases, the policyholders won (also including the compromises made during the judicial proceedings, i.e., situations in which the position of the policyholder improved as a result of the judicial proceedings; the statistics does not include the actions by the policyholders that the court did not accept due to judicial shortcomings by the plaintiff). In 2010, 16.66% of the solutions were positive for the insurers and 83.34% to the policyholders. In 2009, the number of actions against the insurers increased by 7.5% in comparison with the previous year, and in 2010, this rate was 23.25%. In 2009, 43 actions were filed against the insurers and 42 decisions made, in 2010, 53 actions were filed and 39 decisions made.

^*11 In Estonia, insurance contracts are regulated by Chapter 4 of the Law of Obligations Act (Võlaõigusseadus. – RT I 2001, 81, 487; RT I, 8.7.2011, 6 (in Estonian)). In English, the wording of 2009 is available at http://www.legaltext.ee/et/andmebaas/tekst.asp?loc=text&dok=X30085K3&keel=en&pg=1&ptyyp=RT&tyyp=X&query=v%F5la%F5igusseadus (10.6.2011).

^*12 There are situations in which the LOA is more beneficial to the policyholder than the PEICL, but they are not as relevant to the regular policyholder—i.e., the LOA does not enable to not return the insurance premium, unlike Article 2:104 of the PEICL; according to the LOA, the parties may also conclude an insurance contract for an unspecified term. The LOA (§§499–504) is also more beneficial in protection of the mortgagee’s rights than the PEICL. There are also norms in which it is debatable which regulation is more beneficial to the policyholder.

^*13 The difference between these two regulations lies in the party that should bear the risk for ascertaining all of the relevant circumstances as to the insurance contract.

^*14 J. Basedow et al. (Note 6), p. 77.

^*15 Ibid.

^*16 Ibid., p. 78.

^*17 In legal literature, it has been found that in the LOA, a middle ground between two extreme regulations has been found: on one side, the policyholder shall inform the insurer on his own initiative of all circumstances known to the policyholder which are relevant in the terms of entering into the agreement, on the other, he must inform of the circumstances on which the insurer has directly requested information. See J. Lahe. Kindlustusõigus (Insurance Law). Tallinn 2007, p. 48 (in Estonian).

^*18 However, the author holds that a position should be taken that this difference is merely apparent, since according to the Law of Obligations Act, a negative consequence is only brought about by informed behaviour (LOA §441).

^*19 The condition for the right of increasing the insurance premium is that the insurer has no right to withdraw from a contract in relation to wrongful violation of the obligation to notify by the policyholder.

^*20 The insurer should be able to choose in each separate case in accordance with the gravity of the violation of the policyholder, whether it wishes to withdraw from the contract or it is possible to continue with the contract by increasing the insurance premium. Exclusion of the request to change the insurance premium in case of wrongful violation leads to termination of the contract by the parties. The author holds that the law should rather direct the parties to continue with the contract on different terms. The PEICL enables that. This is why the PEICL should be preferred in this situation, wherefore it enables the insurers to adequately react to the material circumstances. A change has also been made to the Switzerland’s new Insurance Contract Act, enabling the insurer to choose on the basis of the circumstances whether it wishes for a contract to be terminated or the insurance premium to be increased and the contract continued on new terms. See F. Hasenböhler. Pre-contractual Obligation to Provide Information under Private Insurance Law, p. 118. Available at http://www.kpmg.ch/en/docs/20080505_Pre-contractual_obligation_to_provide_information.pdf (3.1.2011).

^*21 P. Varul et al. Võlaõigusseadus II. Kommenteeritud väljaanne (Law of Obligations Act. Commented Edition). Tallinn 2007, p. 479 (in Estonian).

^*22 Principles of European Contract Law. Available at http://frontpage.cbs.dk/law/commission_on_european_contract_law/-PECL%20engelsk/engelsk_partI_og_II.htm (1.3.2011).

^*23 J. Basedow et al. (Note 6), p. 89.

^*24 This right is provided in Article 4:115 of the PEICL.

^*25 H. Cousy. The Principles of European Insurance Contract Law: Duty of Disclosure and the Aggravation of Risk. – ERA-Forum, September 2008, Vol. 9, No. 0, p. 130.

^*26 J. Basedow et al. (Note 6), p. 90.

^*27 Ibid., p. 142.

^*28 OJ L 271, 9.10.2002, pp. 0016–0024.

^*29 In judgement No. 3-2-1-51-06, the Supreme Court has provided that ‘in insurance relations, it is reasonable to presume that before the valid policy ends, the insurer shall notify the policyholder of its contractual obligations, issue a corresponding invoice and a warning’.

^*30 In the commented edition of the PEICL, it is pointed out that if an insurance premium has been prepaid, returning of the money shall take place on the pro rata principle, since modern information technology enables virtual expenseless calculation on the principle of pro rata temporis (1), and due to lessening of the peril, an insurance premium is no longer necessary from the point of view of the insurer’s solvency (2), the insurance risk is divisible on the basis of days/months/years in an economic sense (3), ‘preservance’ of an insurance premium is not justifiable as a ‘contractual penalty’ (4) and keeping such premium could be viewed as punishing of the policyholder, which is unjustifiable (5). See J. Basedow et al. (Note 6), pp. 203–204.

^*31 For example, in Article 28.4 of the QBE Insurance (Europe) Limited Estonian branch home insurance conditions it has been stated that upon termination of a contract, the QBE has a right for the insurance premiums for the time until the termination of the contract, as well as operations expenditures that make up 20% of the insurance premium calculated for the insurance period. See QBE kodukindlustus. Koduvara määratud riskide kindlustus (QBE home insurance. Household insurance). Available at http://www.qbeeurope.com/documents/estonia/Kodukindlustuse%20tingimused.pdf (1.3.2011) (in Estonian).

^*32 J. Basedow et al. (Note 6), p. 276.

^*33 Recklessness is a lighter form of guilt than intent, but more severe than gross negligence and it originates from the Montreal Convention 1999 on international carriage. See J. Basedow et al. (Note 6), p. 247.

^*34 For example, according to the home insurance conditions of ERGO Kindlustuse AS KT.0645.11 Article 17.1.4, the policyholder shall adhere to the instruction manuals of devices. The consequence of non-compliance with this obligation according to Article 20.1.1 of these conditions is partial or complete freeing of the insurer from the performance obligation, regardless of the form and extent of the guilt of the policyholder. Available at http://www3.ergo.ee/HtmlPages/Kodu_tingimused_KT_0645_11/$file/Kodu_tingimused_KT_0645_11.pdf (1.3.2011) (in Estonian).

^*35 For example, ERGO Kindlustuse AS vehicle insurance conditions KT.0056.1 (ERGO Kindlustuse AS sõidukikindlustuse tingimused KT.0056.1). Available at http://www3.ergo.ee/HtmlPages/S%C3%B5idukikindlustusetingimused/$file/s6idukindlustuse_tingimused.pdf (1.3.2011) (in Estonian); Seesam Rahvusvahelise Kindlustuse AS home insurance conditions 1/2011 (Seesam Rahvusvaheline Kindlustuse AS kodukindlustuse tingimused 1/2011). Available at http://www.seesam.ee/uploads/files/household-insurance/Kodukindlustuse%20tingimused%20(01.01.2011)_EST.pdf (1.3.2011) (in Estonian).

^*36 J. Basedow et al. (Note 6), p. 177.

^*37 For example, Salva Kindlustuse AS general vehicle insurance conditions SÜ-11 (Salva Kindlustuse AS sõidukikindlustuse üldtingimused SÜ-11), Article 23.5. Available at http://www.salva.ee/public/files/Kaskotingimused%20%28S%DC-11%29.doc.pdf (1.3.2011) (in Estonian).

^*38 For example, conditions of Codan Forsikring A/S Estonian branch’s car insurance A101/2011 (Codan Forsikring A/S Eesti filiaali autokindlustuse tingimused A101/2011), Article 8.3.5. Available at http://rsa.all-tec.ee/public/eesti/RSA_Autokindlustus_tingimused_A101_2011_EST.pdf (1.3.2011) (in Estonian).

^*39 For example, If P&C Insurance AS general insurance conditions ÜU 20021 (If P&C Insurance AS Kindlustuse üldtingimused ÜU 20021), Article 58. Available at http://tingimused.if.ee/ViewPDF.aspx?ID=1d6efdf9-0353-4a03-ade8-f47da4370d20 (1.3.2011) (in Estonian).

^*40 On certain cases, delayed notice may theoretically help to save the possible expenses for the insurer—for example, in a situation in which the windshield of a vehicle gets hit by a stone, replacing the windscreen immediately may not be in the best interests of the insurer since another analogous incident may happen during the insurance period, besides, delayed notice would not change anything as to investigation of the insured event (extent of damage does not change).

^*41 Article 6:102 of the PEICL states that if the policyholder, insured or beneficiary, shall not co-operate with the insurer in the investigation of the insured event (i.e., does not guarantee access to the insured location, does not provide information, proof or documentation of the insured event), the insurance money payable may be reduced to the extent that the insurer proves that has been prejudiced by the breach. In the event of any breach committed with intent to cause prejudice or recklessly, the insurer shall not be obliged to pay the insurance money. The analogous obligation stated in §448 (2) of the LOA is more restrictive to the policyholder—according to Estonian law, the policyholder shall give reasonable amount of information which is necessary to determine the obligation to perform the contract. Willing non-compliance with this obligation frees the insurer immediately of the obligation to compensate according to §449 (2) of the LOA and in case of other form of guilt, the insurer may reduce the amount to be paid in an extent of the loss incurred to it.

^*42 J. Basedow et al. (Note 6), p. 221.

^*43 The author implies that willing delay in a loss adjustment process enables the insurer to earn additional profit (i.e., as interests) at the expense of unpaid indemnities.  The author holds that the result of breach of the implied covenant of good faith and fair dealing cannot simply be paying a fine for delay to the policyholder, since the policyholder has a justified expectation that by buying an insurance cover, the results of the possible loss event will be covered in a fast loss adjustment process and thus, if compensation for the loss is delayed due to behaviour of the insurer in bad faith, obligating the latter to compensate the loss due to such behaviour is justified.

^*44 Subsection 12 (3) of the German Insurance Contract Act (Versicherungsvertragsgesetz) that was used as a ‘model’ for the corresponding norm of the LOA has been revoked.

^*45 Tsiviilseadustiku üldosa seadus. – RT I 2002, 35, 216; RT I, 6.12.2010, 1 (in Estonian).

^*46 J. Basedow et al. (Note 6), pp. 223–224.