Implementation of EU Acts in the Czech Parliament: A Slow and Laborious Quest with an Uncertain Outcome

The Czech Republic is a bicameral democracy. The dominant chamber of parliament is the Chamber of Deputies, to which the government is accountable and which can hold a vote of no confidence. Two hundred deputies are elected on a proportional basis in 14 regional constituencies and, once elected, organise themselves into political clubs. Club membership generally corresponds to party affiliation. Individual deputies also have the right to initiate legislation alongside the Government Council, the Senate and the regional councils. Unlike the Senate, which is subject to a 30-day time limit, the Chamber of Deputies has no time limit for considering bills, so most of the time is spent on the three readings of the bill in the Chamber. The Chamber of Deputies can also overrule the Senate in most cases and can override the veto of the President of the Republic. In conclusion, the Chamber of Deputies is the dominant political and legislative arena, so we will focus on it. The second important fact for this paper is that the Czech Republic is an export economy that depends on the functioning of the internal market. International trade accounts for 72% of GDP, and 80% of exports go to EU countries. The functioning of the internal market is, therefore, of vital interest. Despite this, the Czech Republic is one of the countries with the highest transposition deficit in the internal market—the Czech transposition performance in 2020 exceeds the 2007 European Council targets by 40%. The Czech transposition deficit, which also includes partial or incorrect transposition, reached 1.7% in the same year, exceeding the European Commission’s target by 240%. ⁽²⁾

The Lisbon reform of primary law has led to a formal strengthening of the position of national parliaments. Subsidiarity control is only one of the tools available to parliamentarians. It is a domestic issue of how much of the European agenda is subject to parliamentary scrutiny and how much discretion government representatives have in the deliberations of the Council and the European Council. In his comparative study of CEE countries on parliamentary participation in EU policy-making, Karlas classifies the Czech parliament as one of the weaker and less active. This is not so much because of its formal powers but rather because of the time it devotes to the European agenda, which is usually discussed only in committees. ⁽³⁾ However, the Chamber of Deputies has a very dominant position in the implementation process, where the Government has only limited tools to influence the direction and speed of the legislative process. It is not uncommon for a proposal prepared by the Government’s experts to be completely rewritten and replaced by deputies’ amendments in the Chamber. The weaker ex-ante position of the Chamber of Deputies in European policy-making is compensated for by its ex-post control over the implementation process. It is the process of implementation that has the potential for national political representation to define the process and direction of European integration. This creates a space for revolt and opposition. For this reason, the Czech Parliament has been blamed for infringements and delays in implementing EU law. The legislative process in the Czech Republic is relatively long even compared to other countries. ⁽⁴⁾ On average, it takes nine months for a law to be enacted and even more for the EU law implementation. And that is if it is enacted at all…

In the field of compliance research, several theories justify the shortcomings in the implementation of European law. These include internal political obstacles, national attitudes towards European integration, the capacity of public administrations, or the number of actors involved and their capacity to intervene in the content of the implementing act. ⁽⁵⁾ Despite the factual correctness of these claims, however, these reasons cannot justify Member State against infringement proceedings under Article 258 TFEU. ⁽⁶⁾ Research is also increasingly focusing on substantive issues related to the choices and decisions needed to implement EU law. ⁽⁷⁾ Member States can follow the path of over-implementation or, conversely, the limited path of minimum implementation. ⁽⁸⁾ This is particularly evident in the use of opt-out clauses. ⁽⁹⁾ Furthermore, Member States also have to choose their implementation strategy ⁽¹⁰⁾ in terms of the form and frequency of implementation of EU law—older Member States such as Ireland often implement EU law by executive acts, while others choose the form of parliamentary law. Italy, for example, chooses a combination of these routes, with part of the EU law being implemented by the single omnibus legge europea, and the remaining acts being implemented by the Government through its ordinances. ⁽¹¹⁾

Kaeding’s research into the transposition of transport directives between 1957 and 2004 for the UK, Germany, the Netherlands, Spain and Greece found that 61% of transposed acts were delayed. The average delay ranged from 8 months to 6.5 years. ⁽¹²⁾ The length of delays varies between Member States, with an average delay of 10 weeks in Spain and 60 weeks in Greece. Kaeding considers the number of transposition acts, the number of actors involved with veto power and the detail of the transposed directives to be the main variables affecting the speed of transposition. ⁽¹³⁾ A more hidden cause of delay is the choice of type of national transposition act, which determines who is involved in the transposition process, what powers the various actors have and what procedural deadlines apply, which in turn determines the time needed for transposition. There is also a role for national discretion, the extent and purpose of which changes with changes in the political leadership of the state, which in extreme cases can set the transposition process back several steps.

Administrative capacity, i.e. a state’s own ability to absorb and transpose EU policy and its implementing acts into its legal system, inevitably varies from one Member State to another. Capacity is determined by the organisation of the administration, its hierarchy, the division of tasks and responsibilities, experience, human and financial resources and tradition. In principle, Börzel et al. distinguish two basic criteria—the resources available to the public administration and the capacity to use them effectively. France or Italy have sufficient administrative resources but often cannot use them effectively enough. This is why the Member States with the highest administrative capacity but without the political strength to resist possible pressure from Brussels have been the subject of the least number of procedures by the Commission. In the period 1986–1999, it was mainly the more eurosceptic Denmark and the Netherlands that were subject to many times fewer infringement proceedings than the pro-integration Belgium and Portugal. ⁽¹⁴⁾ Paradoxically, higher administrative capacity leads to higher compliance even in more integration-reluctant countries. ⁽¹⁵⁾ Several rounds of enlargement to the East have brought in countries that have only recently emerged from authoritarian regimes, so it can be assumed that they will not have the necessary capacity to implement EU law in a timely and correct manner. As a result, there is likely to be a greater likelihood of states failing to transpose directives on time. Börzel and Sedelmeier, on the other hand, point to the complex process of rebuilding the institutional framework in these countries, which was linked to pre-accession reforms and the need to incorporate the acquis. Capacity building and related reforms were supported by conditionality instruments, financial programmes and post-accession transition periods. The combination of these factors meant that the new Member States had even lower transposition failure rates than the original Member States. ⁽¹⁶⁾ However, Falkner and Trieb’s research places the new Member States, including the Czech Republic, in the category of “the world of empty legal letters”, i.e. states that have formally transposed directives but have not changed the practical reality. ⁽¹⁷⁾ Similarly, Voermans speaks of paper implementation culture; a state in which practice differs significantly from what is reported by the Commission. ⁽¹⁸⁾ Additionally, I would also draw attention to the European Commission’s rethinking of its approach to launching infringement proceedings in 2004 leading to more selective process where many infringements might be consciously omitted. ⁽¹⁹⁾ This leads to situation where the number of infringements does not reflect reality. Pro example, between 2010 and 2020 the Commission issued 279 formal notices in non-notification cases again the Czech Republic, but only three cases were subsequently referred to the Court.

According to Martha Thomas, recession and other economic difficulties may reduce the willingness of Member States to adopt directives on time, the content of which may impose additional administrative and economic burdens. ⁽²⁰⁾ Another problem for timely implementation may be the different political cycles at the EU and national levels. In the last year before an election, ruling parties are focused on campaigning and the chances of meeting EU obligations on time, let alone not being elected, are not a priority for the new government. New governments, on the other hand, try to make up for these deficits and build up a good reputation in Brussels—especially if there is a period of government discontinuity. ⁽²¹⁾ Conversely, a government’s reputation can be damaged by a Commission action, so especially in the early stages, when the cost of avoiding conflict is lower, the government will seek to resolve a potential (albeit inherited) problem quickly. At the beginning of an electoral cycle, the cost to the government of tabling and debating a proposal is generally lower than at the end—the government has just received a mandate from the electorate, there are no associated scandals and any intra-coalition conflicts are in their infancy. In contrast, voters are more likely to take into account newer proposals presented at the end of an electoral cycle in their pre-election deliberations. This is consistent with the findings from the Czech Republic, where the number of non-implemented EU acts increases in the last year of the electoral period, peaks in the first year of the new government and then declines rapidly.

Unfortunately, in-depth research into the causes of transposition delays is still lacking in Czech academia. Therefore, in this article, I would like to focus at least on the validity of the accusation of the Czech Parliament as a source of delayed transposition. The analysis is based on the data collected for the 6^th, 7^th and 8^th parliamentary terms (here and in the following “EP” and “electoral period”), which corresponds to the time period 2010-2021. The length of this period makes it possible to eliminate some of the peculiarities of political developments and, at the same time, it is a period that is not affected by the additional implementation of the acquis communautaire after accession in 2004. The third reason for choosing this period is that it is the post-Lisbon period. This means that the European framework of primary law remains consistent throughout the period. The data have been obtained manually from the website of the Chamber of Deputies.

First, it is necessary to define whether an implementing bill is something different from a regular bill. In the Czech legislative process, we can find a specific category of bills called European bills. This is a collective term for bills that 1) implement European law, 2) remove shortcomings in previous implementation, or 3) amend the existing implementing law within the limits of national autonomy. Usually, these are mixed proposals, including new, non-EU-related material, based on a national political decision—pure implementing acts are the exception. European laws do not have a specific legislative procedure but follow the same legislative procedure as any other proposed bill. It is merely an informative label.

The large number of bills submitted reflects the wide range of subjects with the right to propose legislation. 618 bills were introduced in the 6^th EP, 640 in the 7^th EP and 735 in the 8^th EP. However, the high number of proposals is offset by a relatively low success rate of 54%, 34% and 43%, respectively. The majority of proposals are not even rejected but are not debated at all and die out at the end of a particular parliamentary term. For example, the bill to amend the Population Register Act was tabled a week after the Chamber was established, but in four years it has not even passed its first reading. ⁽²²⁾ Traditionally, the Government has the highest success rate with its proposals, which are responsible for introducing about a third of the bills. Government bills pass in 77%, 50% and 68% of cases. Still, it is an underperformance compared to Saigh’s thesis that in parliamentary democracies, governments typically propose at least 70% of the legislative agenda and about 80% of what they propose is passed. ⁽²³⁾

The approved law usually also varies from its original tabled content—only 9% of proposals that make it to the third reading do not have amendments tabled. This can lead to a situation when an MP’s amendment is adopted during the legislative process that is contrary to EU law, whereupon a new bill should have been tabled to correct the error once the legislative process was completed to correct it. For example, an approved amendment to the Food and Tabaco Act, which set a mandatory proportion of Czech-originating vegetables sold in supermarkets—a blatant manifestation of a quota-style measure—was voted down only in the Senate. The Government is the sole initiator of European bills implementing EU law prepared by individual ministries. Only its proposals are labelled with the European label. The ministries involved in the preparation of EU legislation prepare to implement measures, which in the Czech Republic are usually in the form of law and therefore require parliament assent. In the period 2009–2019, approximately 370 legislative directives were adopted on the EU level, of which 317 were transposed at least partially in a statutory form.

The time consumption of the Czech legislative procedure is gradually increasing. While in the 6^th EP (2010–2013), the length of the legislative procedure for Government proposals slightly exceeded seven months on average, in the following years, the average length of the whole process increased to nine months. This includes procedures in the CoD, in the Senate, and in the President’s considerations and formal publication. Given the constitutionally prescribed time limits for both Senate procedures and a president’s considerations of veto, the majority of this time falls on three readings in the Chamber of Deputies—now roughly seven months. Fewer than half of the Government’s proposals are European bills, raising the question of whether the CoD treats these bills differently from purely “domestic” bills. It seems logical that bills with an EU element should be debated more quickly—failure to implement EU law on time is sanctioned by infringement procedures and possible financial penalties. Similarly, Robert Zbíral concludes from data on the 5^th EP that European bills are less discussed than national ones, ⁽²⁴⁾ so it should be possible to approve them faster. However, this hypothesis is not confirmed in later EPs.

In the first place, it should be noted that about a third of the proposals that are labelled as European bills are not approved – explicitly rejected by Parliament or not discussed. Mostly the latter. The ratio between non-adopted implementing and non-implementing Government proposals was initially 1:2 in favour of implementing acts in the 6^th EP but moved to 1:1 with subsequent terms. Adopted EU bills were debated roughly 14 days faster than domestic laws in the 6^th EP, only 10 days faster in the 7th EP, but in the 8th EP, they were debated on average for 11 and half months—two and a half months longer than non-implementing Government proposals on average.

These data, especially in the 8^th parliamentary term, are affected by several extremely controversial implementations, such as the transposition of the amendment to the so-called Firearms Directive. ⁽²⁵⁾ The directive was adopted on 17 May 2017 with a 15-month transposition period. The interministerial comment procedure for the implementing bill took place in November of the same year, and in February 2018, the government approved the proposal and submitted it to the Chamber of Deputies. ⁽²⁶⁾ In addition to the transposition procedure, the Czech Republic filed an action for annulment of the Amending Weapons Directive, which the Court of Justice rejected in December 2019. ⁽²⁷⁾ The first reading of the transposition bill in the Chamber of Deputies took place only after 28 months of its scheduling on 3 June 2020. Subsequently, the adoption was relatively quick—the second and third readings took place in October 2020, so the Senate assessed in December, and the President signed the law before the end of the year. It should be noted at this point that a challenge to the validity of an amending directive by a Member State does not exempt it from the obligation to transpose on time. Although this is an exemptional case, mainly due to the saliency of the public debate, the above data suggest a worrisome trend that could potentially affect the Czech Republic’s capacity to meet its EU membership obligations. ⁽²⁸⁾

The Czech legislative process consists of three readings. The first reading serves as a filter to eliminate non-viable proposals, and MPs can send the bill to the relevant committees, reject it or send it back for revision. After the committees have made their recommendation, the second reading takes place, during which MPs can propose their amendments. The amendments are discussed again in the committees that recommend viable amendments. Amendments are voted on in the third and final reading. With the exception of the first reading, most measures have time limits, although these can be extended. The legislative process has tools for speeding it up—both extraordinary and ordinary. I consider the procedure under Art. 90(2) of the Rules of Procedure of the Chamber of Deputies to be an ordinary acceleration of the process—the sponsor of a bill can propose that the Chamber of Deputies give its assent to a bill at first reading. In this case, there are no second and third readings of the bill, no amendments can be tabled, and the bill is voted on as proposed. However, this procedure can be blocked by 50 MPs or two political groups. This rule is quite strict, and therefore, fast-tracking can only be used for consensual proposals. Between 2010 and 2021, there were 119, 132 and 160 requests for fast-track procedures, but due to congestion in the Chamber, only 101, 126 and 87 proposals were put to the vote; the rest were not even debated. The accelerated procedure was adopted only in one-third of the cases voted on for both EU and non-EU bills. If fast-track is rejected, the bill continues through the ordinary procedure.

The Art. 90(2) of the Rules of Procedure was originally proposed to speed up the adoption of EU-related legislation in 1999, but as a result of political opposition, the original motion was amended, and its applicability was extended to all legislative proposals. As a result, during 2010–2021, 97 bills were approved under the fast-track procedure, of which only a small fracture of 11 were European bills. On the other hand, for 61 European bills, the Chamber did not agree to fast-track them. Fast-track motions are usually submitted by the Government after elections when the new Chamber of Deputies is dealing with a transposition deficit inherited from the previous election period or at the end of an election cycle when there is insufficient time for the lengthy full legislative process. In addition to political criteria, gold-plating, avoiding alteration clauses or unrelated domestic content in the European bills hinders cooperation with opposition clubs. A case in point is the proposal of the Data Processing Act, ⁽²⁹⁾ which was submitted with a (unsuccessful) motion for approval under the fast-track procedure. Later, the relevant Chamber Committee even called the Ministry of Interior to prepare a comprehensive amendment and rewrite the bill in its entirety to limit unnecessary administrative burdens. ⁽³⁰⁾

The second option is to speed up implementation through the legislative emergency procedure. According to Art. 99 of the Rules of Procedure, a state of legislative emergency is declared by the President of the Chamber of Deputies at the request of the Government in situations where the fundamental rights and freedoms of citizens, the security of the state or significant economic interest is endangered. If the Chamber of Deputies agrees that such a significant risk exists, there is no first reading and the third reading may immediately follow the second reading. The whole process is significantly shortened and, in the case of coordination with the Senate and the President’s Office, only single days take place between the submission of the proposal and its publication in the statute book. ⁽³¹⁾ In such short periods of time, the transparency and coherence of parliamentary deliberations suffer substantially. MPs have hours to familiarise themselves with often radical legislation and, thanks to the limitations of the rules of parliamentary debate, are unable to gain clarity even through discussion. ⁽³²⁾ This is a significant limitation on the functioning of parliamentary democracy, so this regime must be used only for genuine crises. The Constitutional Court confirms this, approving that the declaration of a legislative emergency might take place “only if it is interpreted in the most restrictive, constitutionally friendly and responsible manner.” ⁽³³⁾ This is particularly restrictive of the opposition’s rights. However, this motion has not been used to implement European law.

As we can see, the Czech Parliament has practically two tools at its disposal to speed up the whole procedure. However, their fundamental problem is that they are both too fast, while ordinary deliberations are too slow. A Government which relies on the confidence of the majority in the CoD can indeed attach a question of confidence to a proposal, but it does so only a few times in a matter of high political priority proposals; never related to EU law implementation. European bills are thus subject to the ordinary procedure, and the Government’s will to sacrifice political capital to push them through the Chamber is limited.

The result is a rather bleak outlook for the timely implementation of EU Acts. From the European bills data sample, I am now eliminating those bills that correct implementation errors or amend implementation within the limits of national autonomy. I also exclude laws that are only relatively loosely related to European legislation, e.g. regulating the efficiency of tax collection, such as the Electronic Sales Registry Act or the Single Collection Point Act. Furthermore, codexes are excluded from the sample. By doing so, I am working with a data set of 87, 95 and 63 European bills, which are truly implementing proposals. Additionally, European bills often implement a package of EU acts; for this reason, I will use only the latest implementation deadline for each bill and even the latest date in the case of a split deadline. In the case of directives, the date of the transposition period end has been used, and in the case of regulations, the date of the beginning of its applicability is used. The implementation deadlines are taken from the ISAP ⁽³⁴⁾ and EUR-Lex databases.

The result of this assessment is not positive. Out of the European bills approved in the 8^th EP, in only 10 cases did the national implementation law enter into force before the end of the implementation period, while in 87 cases, it was late. However, it is not just the Chamber to blame; the Government is also a major contributor to this problem. Again, if we assume that at least nine months are needed for the full deliberations in both houses, the signature of the President and publication in the law book, the Government submitted the implementing proposals on time in only 10 cases.

Let’s say that the above-described findings can be justified by the fact that in the 8th EP, the minority Government without support in the CoD governed. It is much harder to gather support for every individual bill, and every proposal is subject to bargaining and trade-offs. Moreover, it can be assumed that the Government does not have the implementation of EU law among its priorities, ⁽³⁵⁾ so it will rather devote political capital to its political priorities. Therefore, the situation should be better in the 7^th EP of a majority Government consisting of three political parties with a stable majority of its own. The same method was used to identify 95 European bills that were passed during this period that implemented EU law—transposing directives, adapting regulations or even Court of Justice decisions. The Government submitted only 16 proposals in time, nine months in advance so that 15 implementing laws were also adopted in time—before the end of the transposition period or before the regulation became applicable. However, 80 laws did not enter into force until after the deadline set by the relevant EU acts. This period is characteristic of delayed drafting in the Government, as 34 bills were sent to the Chamber after the end of the transposition period, which is the highest rate among the periods under review.

For most of the 6^th EP, there was a stable Government majority and fewer parliamentary clubs in the Chamber of Deputies, which made daily business simpler. Therefore, the legislative procedure took, on average, two months shorter than in the two subsequent election cycles, only seven months. However, in 2011, the new Government inherited from its predecessors an exceptionally high transposition deficit of 1,9 %, which it managed to reduce to a historic low of 0,1 %. On the other hand, the overall picture is similar – bearing in mind the methodological limits described above, 12 implementing acts entered into force on time compared to 66 delayed acts. Despite the faster legislative procedure, the slowness in the Government maintained, with only 15 proposals tabled on time, in this case, seven months before the end of the implementation period. By contrast, 27 proposals were tabled after the end of the implementation period, however the inherited deficit from the 5^th EP plays a part—19 delayed cases lay in the first year of Government.

The problem of the low priority of European bills and the inadequate instruments for fast-tracking has already been described above. The root of the problem, however, is the late submission of the proposal to the Parliament. So now we look upstream in the legislative process:

The drafting procedure is governed by two Government’s Decision – Legislative Rules ⁽³⁶⁾ setting general rules of drafting and special Methodological Guidelines on EU law implementation. ⁽³⁷⁾ Both documents are only Government internal documents without the general force of law. According to the Legislative Rules, the first moment that we have a central record of each bill is the moment when the bill is tabled for the inter-ministerial comment procedure. At this stage, ministries and other commenting authorities can comment on the bill proposal. The process normally takes at least 20 working days but may exceptionally be shortened or even waived. Once possible conflicts have been resolved, the draft is referred to the Government’s advisory body, the Legislative Council of the Government, which comprises independent legal experts who further comment on the draft. Once the comments have been incorporated, the proposal is submitted to the Cabinet. This pre-parliamentary phase should be taken into account when planning the implementation schedule. If we use a similar sample of European bills for 2013–2021, the average length was around 190 days, i.e. 6.5 months. On average, the Cabinet will debate and submit the bill to Parliament within 14 days.

Data concerning the drafting of the implementation bill by the ministries are not publicly available and are decentralised within the ministries. By default, the works include the drafting itself and the inter-ministerial comment procedure. The specific processes vary between ministries and are determined by the structure of the ministries—sometimes, policy departments have their own legal experts, while others leave the legal aspect to the legislative departments. The frequency with which specific departments deal with the European agenda also plays a role. The Ministry of Finance, the Ministry of Agriculture and the Ministry of Industry and Trade handle most of the European legislation, but there are also differences between the individual departments within the ministries.

If we were to try to quantify the time taken by ministries to prepare a European bill, we would use the available data. The first point will be the date of uploading the draft into the inter-ministerial comment procedure, i.e., after the ministerial drafting is done, and the second point will be the end of the implementation period of the implemented act. By doing this, we will be able to identify how much time is left for other actors outside the ministry in charge. Bearing in mind that the Governmental phase takes an average of seven months, and the subsequent parliamentary phase requires nine months, it should be noted that in the 7^th EP, there was an average of 14.5 months left until the end of the transposition period, but in the 8^th EP it was just around eight months. Moreover, several proposals were submitted for the interministerial comment procedure after the transposition period had already expired—23 proposals were submitted for the 7^th EP, and 15 were submitted for the 8^th EP. Thus, although an undisciplined Parliament is often blamed for late transpositions, the problem already arises at the Government and the civil service level because the drafts are already tabled after the given deadline.

The Czech Republic has a very special relationship with the implementation of EU law. The problem of the time-consuming preparation and approval of the European bills has been described above. Although we can characterise the Czech parliamentary culture as improvised and with little discipline and the organisation of the work of the Chamber of Deputies in particular as haphazard and inefficient, the roots of the problem lie elsewhere. Implementation delays start at the ministerial level and worsen with the next stages of the legislative procedure. Timely implementation of EU law is giving way to other political and private interests—for example, the already delayed consideration of the Data Processing Act implementing the GDPR has been repeatedly postponed due to the minister’s leave. There are several possible solutions to this situation—changing the Rules of Procedure of the Chamber of Deputies, creating a special procedure for European acts deliberations with strict time limits or skipping the first reading entirely, and strengthening the supervision and coordination of the Cabinet over the implementation of EU legislation. The root is, however, always the same—lack of the political will and commitment to fulfil the obligations arising from the Czech membership in the EU.