Resumé

This paper analyzes the consequences of the 1992 change in the voting rule in the Finnish Parliament. Before this reform, one third of all Parliament members could delay a law proposal for reconsideration by the Parliament. This rule was abolished in 1992 which meant that the Finnish Parliament finally adopted a simple majority rule to decide on new legislation. The empirical part of this article analyzes the effects of the reform on the parliamentary parties' voting power. The voting power of the big parties increased compared to that of the small parties. However, the variation among smaller parties was greater. The biggest losers were medium size parties. Considering the government and the Parliament as institutions, the emphasis clearly moved to the government. Considering parties in the government coalitions as a whole (adding up their share in the government and in the Parliament), the picture was quite clear. The opposition lost at least some of its voting power. This change was clearest in the case of the party government model in which the opposition lost its voting power completely.

Introduction

The aim of this paper is to analyze the consequences of the 1992 reform in the voting rule in the Finnish Parliament, Eduskunta. In practice the reform meant that the Finnish Parliament finally adopted a simple majority rule to decide on new legislation. More precisely, the goals of this paper are to see how the voting power of the Parliament parties of various sizes changed, how the distribution of power between the Parliament and the government changed, and what effects the new rule had on the power distribution between the government and the opposition parties. First, the old and the new voting rules of the parliament and the reasons for the change are introduced. In the empirical part, the effects of the change are analyzed with the Shapley-Shubik index of voting power and three different theoretical models depicting the government-parliament relations. Finally, some possible implications of the constitutional change are suggested.

---

*Mikko Mattila, Department of Political Science, University of Helsinki, P.O. Box 54, FIN-00014 University of Helsinki, Finland.

Constitutional Change

The voting procedure of the Finnish Parliament with its postponement rule has comparatively been a curiosity. Until the change in 1992, one third of all parliament members (67 MPs) could postpone an ordinary legislative proposal¹ for reconsideration by parliament (Anckar 1992, 161). This reconsideration did not take place in the next annual parliamentary session, but in the session after that. A postponed legislative proposal became law only if it was adopted unchanged by the Parliament. The postponement rule originated from the 1906 constitution and was also upheld in the 1928 Parliamentary Act which still regulates the work and procedures of the Parliament. The intention of the rule was to protect the constitution from a possible socialist take-over, but particularly in the early 1990s it was used by the leftist parties to block all attempts to cut public welfare spending. This deferment rule was repealed from the beginning of September 1992. However, the old minority rule is still maintained in order to protect the basic legislated rights to social assistance. Constitutional amendments also need to be passed by a two-thirds majority (Sundberg 1993, 420-21).

One of the original reasons for these qualified majority rules was to ensure that the Parliament would not pass ill-considered legislation which would be based on the opinion of a temporary majority of the Parliament. The qualified majority requirement has since been seen as a major institutional contributor to the "consensus" policies of the 1970s and 1980s in Finland (Heiskanen & Martikainen 1988, 373). The need for prompt economic policy measures particularly required extensive bargaining between the government and opposition parties.

Although quite rarely used by the opposition, the postponement of a legislation has usually been considered a defeat for the government. Therefore, it is surprising how rarely this option has been used by the opposition in practice. During the period 1917-86, the average number of deferred law proposals was 1.5 per year (Helander 1990, 57). However, the indirect consequences of the rule were more important. The postponement rule meant that the government had to ensure a majority of two thirds to be sure that its legislative proposals were accepted by the Parliament. This meant that every controversial law proposal had to be negotiated with the opposition before the final approval voting in the Parliament. If no agreement was reached in these negotiations, the government did not usually put the proposal before Parliament or, if already tabled, withdrew it. The practical consequence of this was that when the opposition could work as a united group it could veto all government legislation.

A spatial example will clarify the importance of opposition parties with
veto power (see e.g. Tsebelis 1995, 89-90). Figure 1 depicts a simple
bargaining situation in a parliament with three parties. Parties A and C form

the government coalition while party B is in opposition. The points in the figure depict each party's most preferred position in two policy dimensions X and Y. The SQ point is status quo, i.e. the present situation. The gray area shows all the possible combinations of X and Y preferred by both government parties A and C. If these two parties have enough votes in parliament to make decisions without the consent of the opposition party B, a new status quo will be found in the gray area. However, if B has enough votes to block legislative proposals the status quo point will not be moved, because B will be worse off in the gray area than at SQ. Thus the opposition veto power has the major consequence that changing the existing legislation is extremely hard because there has to be an agreement on it, not only inside the government coalition but also between the government and the opposition.

The example in Figure 1 is of course simplified. In reality the opposition in Finland consists of a number of parties usually from both ends of the political spectrum. In order to use the veto power, at least 67 MPs from opposition parties must agree to defer a government legislative proposal. However, this does not change the fact that the government has to take opposition opinion into consideration when deciding what kind of proposal to put before the Parliament.

A Voting Power Index for Multi-Stage Decision Making

Measuring voting power is by no means a simple task. A wide variety of alternative power indices have been introduced in the literature. The properties of these indices have also been under careful scrutiny by a number of authors (e.g. Nurmi 1987). One of the most widely used voting power indices is a measure introduced by Shapley & Shubik (1954). This measure is based on the idea of forming all possible coalitions parties can build and then calculating for each party the number of times these parties are in a crucial position in a coalition, i.e. if they leave the coalition it is no longer a winning coalition. The Shapley-Shubik measure has been applied in a variety of decision making settings (see e.g. Herne & Nurmi 1993; Pappi et al. 1995; Widgren 1995).

The Shapley-Shubik voting power for party i (ø,) is calculated as the
following equation:

where s is the number of parties in a coalition S, n is the number of parties in the parliament, and a, is 1 if the coalition S of which i is member is a winning coalition, but is not a winning coalition without i, and zero otherwise (Coleman 1986, 196). The summation is taken over all possible coalitions in which i is a member.

The Shapley-Shubik measure has some properties which make it very useful in the analysis of voting situations (ibid., 193-95). One is that if the number of votes in the decision making body adds up to one, then the sum of the power measure of all members also equals one. Another useful feature of the Shapley-Shubik measure is that these measures can be added if there are two decision phases, e.g. in a case where a decision has to pass two chambers. This property is based on the fact that it can be shown that if a game is formed by adding two games, then the Shapley-Shubik value of the new game is the sum of the two original games (see, e.g. Friedman 1991, 267-70). This feature will be used next when the voting power indices for parties are calculated.

König & Bräuninger (1996) extend the Shapley-Shubik voting power index to multi-chamber voting situations, comparing voting procedures in the US and German parliaments. Their methodology can be applied in the Finnish case, although the Finnish Parliament is unicameral. However, if the agenda setting power of the government is included, the voting situation can be analyzed as a two-chamber problem. Before the examination of the Finnish case, a brief example of how the Shapley-Shubik values in the multichamber situation are calculated is shown.²

The Shapley-Shubik power measure can be calculated with the use of all the possible sequences players can be put into. For example, if there are three players in a voting situation, these players can cast their votes in 3! = 6 different orders. The Shapley-Shubik value can then be calculated by looking at each sequence and seeing which player is pivotal. A pivotal player is a player who is in a position in a sequence to cast the decisive vote, if all the players preceding him/her vote "yes" and all the players after him/her vote "no." In the example of three players, the second player in a sequence is pivotal. Because the player before him/her votes "yes" and the one after him votes "no," the second player can decide the outcome. The total Shapley value for a player is the number of sequences in which this player is pivotal divided by the number of sequences.

Table 1 shows an example of how this simple way of calculating Shapley values can be extended to a multi-chamber situation. In this example there is a parliament with two chambers, and a simple majority in both chambers is needed to pass new legislation. In this parliament the party discipline is strong and the parties always vote as a group. The first chamber of the parliament is controlled by party A, which has all the votes. The second chamber consists of three parties B, C and D which all have one third of the votes. These four parties can be arranged into 4! = 24 different sequences, as shown in Table 1. In each sequence the pivotal party is marked with a bold letter. Party A, which controls the first chamber, is in a pivotal position if at least two of the parties in the second chamber vote before A. For example in the sequence BCAD, party A is pivotal, because if A votes "no" the law will be rejected (remember that in a sequence all players preceding the pivotal player vote "yes" and all players after the pivotal player vote "no"). A party in the second chamber is pivotal if party A in the

first chamber and one party in the second chamber vote before it. In the voting sequence ABCD, party C is pivotal, because its decision defines the final outcome. The overall voting power is the number of pivotal positions divided by the number of sequences. In this case party A's voting power is 0.5 and the voting power of B, C and D is 0.167.

Using the addition rule mentioned earlier, it is possible to add up parties' power measures. Using the same example in Table 1, it is possible to calculate the power of each chamber, which is simply the total voting powers of all parties in that chamber. In the example the voting power is 0.5 for both chambers. Furthermore, in a situation where a party is represented in both chambers, its total voting power is simply the total of its voting power in each of the chambers. If player A in the first chamber belongs to the same party as player B in the second chamber, the total voting power of this party is 0.5+ 0.167 = 0.667.

Empirical Analysis

König & Bräuninger (1996) and Pappi et. al (1995) use the multi-stage Shapley-Shubik procedure to calculate voting power in a parliament, using three sets of different theoretical assumptions about the relationships between the parliament and the government. They call these models the legislative model, policy leadership model and party government model. The first of these models does not take into account the role of government as an agenda setter and as a veto player. These assumptions are included in the policy leadership model. The party government model assumes that the government takes a leading position as an agenda setter and that the government parties in the parliament are always loyal to the government. Each model adds new features to the analysis, and thus the last two models are probably closer to the real situation of Finnish political decision making than the legislative model. However, the legislative model can be used as a baseline model and the others compared to it.

In the following analysis these models are adapted to the Finnish institutional setting. It is assumed that the legislative process consists of two stages. First, the government decides whether a legislative proposal is put before the parliament or not.³ In the second stage, the legislative proposal goes through the parliamentary procedures and, if not rejected before, goes to a vote in which it is decided whether the proposal will be enacted as a new law or not. Thus, a proposal must negotiate two hurdles before it is accepted as a new law.

The first stage of this process (the government decision) can be seen as a
voting situation in which each party in government has as many votes as it
has ministers in the government. The government decisions are made

according to the simple majority rule (Nousiainen 1988, 228). If the proposal passes this vote, it will be put before the Parliament. The Shapley-Shubik measures for this first stage are calculated using the number of votes (i.e. ministers) each government party has. Of course, the opposition parties have no voting power at this stage.

The second stage involves the parliamentary voting. Since party discipline is strong in the Finnish Parliament, it is reasonable to think of parliamentary parties as unitary actors in parliamentary voting situations. This stage is analyzed in two ways. First the voting power indices are calculated using the old qualified (two-thirds) majority rule and then by the simple majority rule.

Data

The empirical data for the analysis consists of the last four parliamentary election results and government coalitions formed after the elections. The period covers the years from 1983 to the present. The governments included are Sorsa's fourth government (1983-87), Holkeri's government (1987-91), Aho's government (1991-95), and Lipponen's government (1995-). All of these were majority governments in the sense that they controlled more than half the votes in the Parliament. However, only Lipponen's government controlled more than two thirds of the votes. The actual change in the Parliament Act took place during Aho's government in 1992. However, using more government coalitions and parliaments gives more reliable results concerning the overall effects of the reform.⁴

The dependent variable in the following empirical analysis is the change in voting power as a percentage when moving from the qualified two-thirds majority rule to the simple majority rule. This change is analyzed in relation to party size and between government and opposition parties.

Legislative Model

This model is the simplest of the three consisting of only one stage of decision making. The agenda setting power does not feature in this model. The parties in the Parliament simply vote on bills and the voting power is calculated as the "standard" Shapley-Shubik value for all parliamentary parties. Before the abolition of the postponement rule, a party is pivotal if 133 representatives (out of 200) vote "yes" before it. After the abolition of the postponement rule a party is pivotal if 100 "yes" votes are cast before it.

Figure 2 shows a scatterplot of the effects of the change from qualified to simple majority for parliamentary parties of different size in the legislative model. The plot includes the parliamentary parties during the period mentioned earlier (1983-present). The y-axis shows as a percentage how much the Shapley-Shubik voting power index changes as a result of the new

voting rule. The x-axis shows the size of parliamentary groups in seats. The curve in the picture is a lowess curve⁵ depicting the "average" relation between the change in voting power and the size of the group. The figure shows that the relationship between the size of a party and the change in voting power is not linear. Clear "winners" in the change are big parties, especially those with more than 35 MPs. Most of the biggest "losers" are small parties, many of them losing even 40-60 percent of their voting power. However, there are also some small parties that gained more voting power as a result. Interestingly, the curve seems to "slump" when the party size is between 10 and 25. These medium size parties (usually the Swedish People's Party, the Leftist Union and the Greens) seem to be the overall biggest losers.

Table 2 shows what happens to the combined voting power of the
government coalition parties as a result of the change. With the exception of
Aho's government the changes are not big. Aho's government gained about

10 percent more voting power. The reason for small overall changes is that although large government parties gained more voting power, the smaller coalition partners lost some of theirs, and the final results add up only to small overall changes. However, the legislative model is the most unrealistic of the three models analyzed in this article. The next two models take into account the leading position of the government coalition in the decision process.

Policy Leadership Model

The policy leadership model takes the agenda setting power of the government into account. Now the voting power is calculated as a "twochamber" system, where the first chamber is the government voting which determines whether the proposal is sent to the Parliament at all, i.e. the government has veto power on every proposal. The second chamber is the actual vote in the Parliament. Before the abolition of the postponement rule, a government party is pivotal if more than 134 "yes" votes are cast in the Parliament and the party is pivotal in the government voting. A Parliament party is pivotal if more than half of the ministers in the government vote "yes" and 133 "yes" votes are cast in the Parliament. After the abolition of the postponement rule the situation is the same, except that the number of "yes" votes required in the Parliament is 100.

The relationship between party size and the change in voting power in the
policy leadership model is depicted in Figure 3. The dotted curve shows the
relationship for government parties and the ordinary curve for opposition

parties. The curves show that as in the legislative model, particularly the small opposition parties are losers after the change. However, if a small party is in government it may either lose or increase its voting power. Among the large parties the difference between government and opposition parties is not very great, although large government parties seem to manage a little better than large opposition parties. The same slump as in the legislative model among parties of 10 to 25 members is also shown in the curves of Figure 3.

The voting rule change may have significant effects on the distribution of power between government and parliament as institutions. This can be analyzed simply by adding up the voting power indices for parliament and government groups. Table 3 shows that the voting power has clearly shifted in favor of the government. During all four government periods the share of the voting power of the Parliament has declined. For example, if the old twothirds majority rule still existed, the Parliament's share of voting power during the Lipponen government would be 0.47 compared to 0.29 with the simple majority rule.

The combined voting power of government parties also increased as the rule changed for the first three governments. However, the combined voting power for the Lipponen government coalition declined slightly because it includes three medium size parties (the Left Union, the Greens and the Swedish People's Party) which all lost some of their voting power.

Party Government Model

Of the three models, the party government model comes closest to the real situation in a "normal" policy making process in Finland. This model also emphasizes the government's control of the coalition parties in the Parliament. The model is similar to the policy leadership model, except that the government parties in the Parliament decide to vote as one group and always back the government decision. In this situation it seems that all government parties form one big party which is always loyal to the government. Now the voting power indices can be calculated as in the case of the policy leadership model. However, the voting power of the "government" party must be somehow distributed to individual agents. This division is made in the same proportion as the government parties' voting power at the government decision making stage.

The results for the party government model are more extreme than for the other models. The reason is that opposition parties have lost all of their voting power in practice because all government coalitions in the analysis are majority governments, controlling more than half the votes in the Parliament. When the simple majority rule is used they also control all the voting power. Thus a scatterplot is not necessary to analyze the results.

A summary of the results for the party government model are shown in
Table 4. With the exception of the Lipponen government, the results show
that the power of government (as compared to the Parliament) has increased

as a result of the voting rule change. Since Lipponen's government controls 143 seats in the government, i.e. more than a two-thirds majority, the voting rule change had no effect in this case. The total voting power of government parties in Table 4 shows that the opposition lost all voting power.

Conclusion and Possible Implications

The analysis in this article was based on a theoretical index of voting power, namely the Shapley-Shubik index, which has some drawbacks. For example, these findings apply only if it can be assumed that all party groups are always united in their voting. Furthermore, in the political reality of the Finnish Parliament, the power of the opposition has depended on its ability to reach agreement among the opposition parties. When these parties represent both extremes of the political left-right dimension, such a consensus is sometimes difficult to attain. However, with these reservations in mind, the results give some indication of how the distribution of power between the government and the opposition and between the government and the Parliament has changed as a result of the abolition of the postponement rule.

The empirical analysis was based on three theoretical models which all depicted the government-parliament relationship in different ways. All three models showed the same kind of change in the voting power distribution. Some clear patterns emerged, although the changes were not always linear. It seems that the voting power share of the big parties increased compared to that of small parties. However, the variation among smaller parties was greater. It seemed that the biggest losers were medium size parties (about 10-25 parliament seats). Some of the smaller parties were even able to increase their share of the voting power.

Considering the government and the Parliament as institutions, the emphasis shifted more clearly to the government, which was able to increase its share of the voting power. For the government parties as a whole (adding up their share in the government and in the Parliament), the picture was quite clear. The opposition lost at least some of its voting power. This change was clearest in the case of the party government model in which the opposition lost its voting power completely.

It is probably too early to analyze the implications of these institutional power shifts empirically. However, based on the theoretical literature and findings from other countries, some conclusions can be drawn on their possible effects.

1) The size of government coalitions will become smaller. In their comparative analysis of government coalitions, Lane & Ersson (1994, 242) found that the government coalitions in Finland are frequently over-sized. However, one might note that if the postponement rule is taken into account,

Finnish governments have almost always been under-sized. Following the abolition of the postponement rule, one can expect the size of Finnish government to conform to the normal rule of minimum-winning coalitions, i.e. having more than half of the votes in the Parliament, but not as much as before.

2) The duration of government coalitions will be longer. The comparative studies (e.g. Lane & Ersson 1994, 304) have found that political stability as measured by government durability has been low in Finland. One theoretical factor that has contributed to this is the ideological diversity of the government coalition which can have negative effects on government durability (Warwick 1992; Budge & Keman 1993, 158-88). The larger the coalitions are, the more ideologically diverse they are, and consequently the shorter their life-span. Thus the average durability of Finnish governments will increase.

3) There will be a reduction in the overall number of bills. According to Döring's (1995a; 1995b) and Henning's (1995) theory and empirical results, the overall number of bills will decline when the government's agenda setting power increases. The reason is that when the acceptance of government proposals in the parliament requires less resources from the government, it will substitute non-conflictual proposals (which require few resources) with conflictual proposals (requiring far more resources). Because the majority of proposals are non-conflictual in nature, the overall number of bills will decline.⁶

4) The number of controversial laws will increase. This implication is directly linked to the previous one. The increased voting power of government parties means better resources for producing more politically controversial laws. Extensive negotiations with the opposition are no longer needed to ensure that a controversial law proposal will pass.

5) The power of interest groups will decrease. Several empirical studies (e.g. Laumann & Knoke 1987; Pappi et al. 1995; Knoke et al. 1996) have shown how various interest groups are able to exert influence on political decision makers through institutionalized connections or networks. If the number of important players decreases when the voting power is concentrated more on few key government parties, the potential "access channels" for interest groups are reduced. It is no longer enough to lobby opposition parties in order for interest groups to block important and unwanted legislative proposals. A possible side effect of the constitutional change may thus be an overall reduction in the influence of interest groups.

The implications of the constitutional reform can also be seen more broadly. As mentioned earlier, the minority protection rules have contributed to a consensual way of making policy in Finland (Heiskanen & Martikainen 1988). Accordingly, the abolition of these rules and the coincident severe economic depression have contributed to the demise of Finnish consensus

policies in the 19905. The political decision makers in the government
parties no longer need support from opposition parties and interest groups to
be able to implement their policies.

---

NOTES

1. The postponement rule was not applied to tax laws applicable for longer than one year, international treaties, and other minor exceptions. However, tax laws that raised taxes had to be accepted by a two-thirds majority.

2. In the analysis of voting power, both Shapley-Shubik and Banzhaf indices are often used. However, Laakso (1978; 1980, 253) has showed how the Banzhaf index behaves against "common sense" when considering different parties' situations regarding different decision rules. According to Laakso, the Shapley-Shubik value does not behave in this paradoxical manner. König & Bräuninger (1996, 337) also prefer Shapley-Shubik values to Banzhaf values.

3. Of course, individual parliamentarians may also propose laws. Usually these proposals are minor changes in some law and it is very rare that they are accepted by the Parliament. For example, according to Andeweg & Nijzink (1995) only 1.2 percent of private member bills were passed by the Parliament from 1978 to 1982.

4. The Shapley-Shubik values in the empirical analysis part are calculated by Thomas König and Thomas Bräuninger's Indices of Power (IOP) program.

5. A LOWESS curve ("locally weighted scatterplot smoother") is used to facilitate visual analysis of scatterplots. It is basedon an average function, where the impact of an observation on the curve's slope is greater, the closer it is to the curve (Schnell 1994, 109-13).

6. According to a study by Wiberg (1994), Finland is one of the most productive law producers in the western world. One reason may have been the two-thirds voting rule.

REFERENCES

Anckar, D. 1992. "Finland: Dualism and Consensual Rule," in Damgaard, E., ed.,
Parliamentary Change in Nordic Countries. Trøgstad: Scandinavian University Press.

Andeweg, R. B. & Nijzink, L. 1995. "Beyond the Two-Body Image: Relations between
Ministers and MPs," in Döring, H., ed., Parliaments and Majority Rule in Western Europe.
Frankfurt: Campus Verlag.

Budge, I. & Keman, H. 1993. Parties and Democracy. Coalition Formation and Government
Functioning in Twenty States. Oxford: Oxford University Press.

Coleman, J. 1986. Individual Interests and Collective Action. Cambridge: Cambridge
University Press.

Döring, H. 1995a. "Fewer Though more Conflictual Bills: Parliamentary Government Acting
as a Monopolist," in Döring, H., ed., Parliaments and Majority Rule in Western Europe.
Frankfurt: Campus Verlag.

Döring, H. 1995b. "Institutions and Policies: Why We Need Cross-National Analysis," in
Döring, H., ed., Parliaments and Majority Rule in Western Europe. Frankfurt: Campus
Verlag.

Friedman, J. W. 1991. Game Theory with Applications to Economics (2nd ed.). Oxford: Oxford
University Press.

Helander, V. 1990. "Lepäämäänjättämismekanismi. Tutkimus VJ 66.7 §:n käytöstä
itsenäisyyden aikana," Valtio-opillisia tutkimuksia 45. Department of Political Science.
Turku: University of Turku.

Heiskanen, I. & Martikainen, T. 1988. "The Finnish Public Sector: Its Growth and Changing
Role in 1960-1984," in Lybeck, J. A. & Henrekson, M., eds., Explaining the Growth of
Government. Amsterdam: Elsevier.

Henning, C. H. C. A. 1995. "A Formal Model of Law Production by Government as a Natural
Monopoly," in Döring, H., ed., Parliaments and Majority Rule in Western Europe.
Frankfurt: Campus Verlag.

Herne, K. & Nurmi, H. 1993. "The Distribution of A Priori Voting Power in the EC Council of
Ministers and the European Parliament," Scandinavian Political Studies 16, 269-84.

Knoke, D., Pappi, F. U., Broadbent, J. & Tsujinaka, Y. 1996. Comparing Policy Networks.
Labor Politics in the U.S., Germany, and Japan. Cambridge: Cambridge University Press.

Laakso, M. 1978. "Valtaindeksien paradoksit," Research Reports A 49. Institute of Political
Science. Helsinki: University of Helsinki.

Laakso, M. 1980. "Electoral Justice as a Criterion for Different Systems of Proportional
Representation," Scandinavian Political Studies 3, 249-64.

Lane, J.-E. & Ersson, S. O. 1994. Politics and Society in Western Europe. Guildford: Sage.

Nousiainen, J. 1988. "Finland," in Blondel, J. & Miiller-Rommel, F., eds., Cabinets in Western
Europe. Basingstoke: McMillan.

Nurmi, H. 1987. Comparing Voting Systems. Reidel: Dordrech.

Pappi, F. U. & König, T. & Knoke, D. 1995. Entscheidungsprozesse in der Arbeits- und
Sozialpolitik. Der Zugang der Interessengruppen zum Regierungssystem über Politikfeldnetze:
Ein deutsch-amerikanischer Vergleich. Frankfurt: Campus Verlag.

Schnell, R. 1994. Grapisch gestützte Datenanalyse. München: Oldenburg.

Shapley, L. & Shubik, M. 1954. "A method for evaluating the distribution of power in a
committee system," American Political Science Review 48, 787-92.

Sundberg, J. 1993. "Finland," European Journal of Political Research 24, 419^23.

Tsebelis, G. 1995. "Veto Players and Law Production in Parliamentary Democracies," in
Döring, H., ed., Parliaments and Majority Rule in Western Europe. Frankfurt: Campus
Verlag.

Warwick, P. 1992. "Economic Trends and Government Survival in West European
Parliamentary Democracies," American Political Science Review 86, 875-887.

Wiberg, M. 1994, "Law Production in Finland: Strategic Considerations," in Wiberg, M., ed.,
Rationality in Institutions. Annales Universitatis Turkuensis B 206. Turku: University of
Turku.

Widgren, M. 1995. National interests, EU enlargement and coalition formation: four essays on
national influence in the EU. Helsinki: The Research Institute of the Finnish Economy.