Is Electoral Fairness Economically Efficient?

Imagine if had a different Table 1: Percent of Seats Won and Popular Vote Received in method of electing its political repre- 1997 Federal Election sentatives, or a different system of po- litical governance, in place of the Federal Seats Percent Votes Percent electoral and political system it has. Party Won of Total Received of Total Would there be any political and eco- nomic effects from alternative electoral Liberal 155 51.5 4,994,277 38.5 1 and governance systems? Reform 60 19.9 2,513,080 19.4 Bloc 44 14.6 1,385,821 10.7 Types of Parliamentary Quebecois government New 21 7.0 1,434,509 11.0 There are two major forms of parlia- Democrats mentary organization: majoritarian and Progressive 2 20 6.6 2,446,705 18.8 consensual. Majoritarian government Conservative is normally characterized by a one-party executive, executive dominance over the Other* 1 0.3 211,482 1.8 legislature, two dominant political par- * A number of political parties including the Canadian Action Party, the Marxist- ties, and an electoral system dispropor- Leninist Party, the Green Party, Independents, and non-affiliated voters are all in- tional to the popular vote (Lijphart, cluded in the “Other” category. 1994).3 It is designed to “create exclusive, Source: (1998), Official Voting Results of the 36th General Election, strong, decisive, and accountable par- 1997. Available on the Internet at www.elections.ca. ties and executives that are said to pro-

Patrick Basham is The Fraser Institute’s duce more responsible public policies” consensus among different political ac- Director of the Social Affairs Centre. He is (Crepaz, 1996). Countries that have tors” (Crepaz, 1996). Countries that completing his Ph.D. in Political Science majoritarian government include Can- maintain typical consensual systems of from Cambridge University. ada, Australia, and the United Kingdom. government include Finland, the Neth- Jason Clemens is the Director of Fiscal erlands, and Switzerland. Studies at The Fraser Institute. He has a Consensual government is generally Masters Degree in Business Administra- characterized by multi-party coalition Electoral systems tion from the University of Windsor. executives, executive-legislative bal- Debbie Roque is a third year Political ance, multiple parties, and more or less There are a number of different systems Science student at the University of British proportional representation (Lijphart, by which countries elect political repre- Columbia. She is currently completing a 1994).4 Consensual government is “de- sentatives. The two most frequently em- Co-operative Studies term at The Fraser signed to facilitate inclusion and pro- ployed are plurality (first-past-the-post) Institute. vide access, accommodation, and and proportional representation (PR).

by Patrick Basham, Jason Clemens, & Debbie Roque

May 2000 / FRASERForum 11 COVER story

Plurality is most often associated with would have been subsequent pressure A trade-off between majoritarian government. Plurality re- on the federal Liberal Party (38.5% representation and fers to first-past-the-post election sys- popular vote) to form a coalition with effective economic tems (as seen in Canada, the United the New Democrats (11.0% popular States, and the UK) wherein seats are al- vote) and another party commanding policy? located to candidates obtaining the larg- at least 0.5 percent of the popular vote. Traditionally, the relationship between est share of votes in a particular district. Put another way, the dynamics and majoritarian and consensual govern- compromises required to form a gov- On the other hand, PR allocates politi- ment—and by extension between PR ernment would be very much different cal representation according to the per- and plurality—was viewed as a trade- under PR than those existing today un- centage of votes received by a party on a off. Consensual government and PR 5 der first-past-the-post. municipal, regional, or national level. electoral systems were seen as more ac- The allocation of political representa- curately representing the electorate, Another interesting effect of the first- tives is, therefore, much more propor- particularly minority groups, although past-the-post system in a multi-party tional to the party’s popular vote than is this improved representation was environment is that it rewards geo- the case under plurality. achieved at the cost of a reduction in graphically concentrated voting. For government effectiveness. Conversely, instance, the federal Progressive Con- The political effects of majoritarian government and plurality different electoral servative (PC) Party ranked third in electoral systems were seen as able to systems popular votes but fifth in seats received. achieve relatively effective government The discrepancy between seats and at the expense of electoral representa- The general effect of the first-past-the- popular votes arose because the PC’s tion. post system of electing political repre- votes were much more geographically sentatives is to exaggerate the popular dispersed than either the separatist Bloc But in the mid-1970s, existence of the election results. For instance, in the Quebecois or the Western Canada- trade-off was disputed. A.T. Baylis 1997 Canadian federal election, the centred Reform Party. (1989) was one of the first scholars to Liberal Party garnered 51.5 percent of assess types of government according to the available seats with only 38.5 per- Another example of the effect of con- both political and economic perform- cent of the popular vote (see table 1). In centrated voting is illustrated by the ance. His conclusions were stark: politi- fact, the most recent Canadian federal disparity between the performance of cal regimes characterized by executive election result would have been star- the Bloc Quebecois, which ran solely in power-sharing maintained higher lev- tlingly different if Canada had used PR the province of Quebec, and the federal els of gross national product (per cap- as opposed to first-past-the-post. . Nationally, the ita) and lower unemployment rates. NDP received 0.3 percentage points Furthermore, as other scholars have Table 1 illustrates some of the effects PR more in popular votes than the Bloc could have had on the outcome of the pointed out, regimes characterized by Quebecois, but elected less than half as 1997 election. For instance, with the as- power-sharing outperformed other sistance of any other party garnering a many MPs. types of regimes based on political vari- mere 0.3 percent of the popular vote, it ables such as protests, riots, political is entirely possible that a coalition be- The system of government and the strikes, armed attacks, and political tween the Reform Party (19.4% popu- method by which citizens elect their deaths. In addition, there was “strong lar vote) and the Progressive representatives clearly affect political evidence that the more consensual the Conservative Party (18.8% popular outcomes. But do different political constitutional structure is, the more vote) would have been formed to estab- and electoral systems influence eco- favourable will be the macroeconomic lish a minority government.6 There nomic outcomes? outcomes” (Crepaz, 1996).

12 FRASERForum / May 2000 Do different political systems of government. Such countries Government outlays systems produce include Austria, Ireland, Portugal and and receipts France. Note that only France is catego- different economic rized as a hybrid-plurality, and only the outcomes? Examining government outlays and re- US is a presi- dential-plurality. ceipts suggests that electoral systems af- Data compiled by the Organisation for fect economic outcomes while systems Economic Cooperation and Develop- Economic growth of governance may not. Countries with ment (OECD) is ambiguous regarding PR electoral systems generally have the hypothesis that different electoral Countries maintaining first-past-the- more interventionist governments. systems and types of government pro- post election methods record higher Parliamentary-PR and hybrid-PR duce different economic outcomes. Us- levels of growth in real Gross Domestic countries recorded much higher levels ing eight economic variables (real GDP Product (GDP) over the period cov- of government—48.7 and 51.6, respec- growth, government outlays, govern- ered. For example, parliamentary- tively, in outlays, and 52.0 and 48.5, re- ment receipts, total average annual plurality countries record average real spectively, in receipts. Plurality coun- growth of both outlays and receipts, GDP growth of 3.0 percent while tries (parliamentary, hybrid, and presi- unemployment, inflation, and eco- hybrid-plurality countries experienced dential) scored much lower on the size nomic freedom) we tested the relation- 2.9 percent average real GDP growth. of government, indicating some type of ship between economic outcomes, The rates recorded by PR countries are relationship between the system of elec- types of government, and electoral sys- much lower: 2.7 percent and 2.1 per- tion and the subsequent size of govern- tems between 1983 and 2000. Table 2 cent for parliamentary-PR and hybrid- ment. shows the results of the analysis. Of the PR respectively. five categories listed in table 2, This is important since it may partially parliamentary-plurality and parliamen- It is also interesting to note that the explain the real GDP growth results ex- tary-PR have been discussed extensively presidential-plurality country, namely amined previously. If, indeed, PR elec- above. Countries with hybrid govern- the US, outperformed all other catego- tion systems lead to larger government, ance systems are those that incorporate ries in terms of average real GDP then it would be expected that, given both parliamentary and presidential growth with a rate of 3.4 percent. the relationship between economic

Table 2: Electoral Systems and Economic Outcomes (1983-2000) Classification Average Outlays Average Receipts Average Average Average Average Annual (% of Annual (% of Annual Unem- Annual Economic Real GDP) Growth of GDP) Growth of ploy- Change Freedom GDP Outlays as a Receipts as ment in Prices Score Growth %ofGDP a%ofGDP Rate Parliamentary- 3.0 40.0 -0.6 37.8 0.3 8.8 5.4 7.9 Plurality Parliamentary-PR 2.7 48.7 -0.5 52.0 0.0 8.1 4.8 7.7 Hybrid-Plurality 2.9 44.7 -0.5 40.9 0.4 7.7 6.2 7.1 Hybrid-PR* 2.1 51.6 0.1 48.5 0.3 10.4 4.9 7.2 Presidential- 3.4 32.9 -0.7 29.5 0.5 5.9 4.4 8.6 Plurality** Source: OECD Economic Outlook (December 1999). *Hybrid-PR refers to France. **Presidential-Plurality refers to the .

May 2000 / FRASERForum 13 COVER story growth and the size of government, mentary and hybrid forms of technology-driven economic change as those countries would experience lower government ranges between 0.5 and 0.8 both inevitable and desirable—and rates of real GDP growth.7 points. largely, if imperfectly, transmits these values through its political culture? Again, it is interesting to note that the Again, the US continues its superior presidential-plurality country (US) re- performance. It maintains a 0.7 point Conclusion corded a much smaller government in advantage over the next highest ranking terms of both outlays (32.9 percent of category of countries, parliamentary- Further and more refined analysis of GDP) and receipts (29.5 percent). plurality (7.9). the political and economic effects of particular electoral systems and models The results for the average annual Why is the United of governance should be encouraged to growth of outlays and receipts indi- States unique? better assess respective systemic out- cated very little difference between ei- comes. Only by continuing this re- ther government or electoral systems. Perhaps the most interesting finding is search will we ascertain whether or not the relative performance of the United a trade-off really exists between govern- Unemployment States. As noted earlier, the United ance, electoral fairness, and economic States maintains unique electoral and outcomes. The results for unemployment vary political systems. Arguably, it consti- greatly; there is no obvious relationship tutes its own category of political sys- References between either the type of government tem and, therefore, makes substantive or the electoral system and the rate of analytical comparisons quite difficult. Baylis, A.T. (1989). Governing by Committee. However, the positive economic and Collegial Leadership in Advanced Societies. unemployment. Once again, however, Albany: State University of New York Press. the US (presidential-plurality system) political effects of the American-style presidential system that can be inferred Crepaz, Markus M.L. (1996). “Consensus Ver- maintains a substantially lower average sus Majoritarian Democracy: Political Insti- unemployment rate compared to other from this data is clearly an area in need tutions and Their Impact on Macroeconomic countries. of additional research. Performance and Industrial Disputes.” Com- parative Political Studies 29(1): 4-26. Is the US’s relative affluence a conse- Elections Canada (1998). Official Voting Results Economic freedom quence of a constitutional system of of the 36th General Election, 1997. Available on government which explicitly limits the the Internet at www.elections.ca. We also tested The Fraser Institute’s government’s size and scope and Finer, S.E., (1975). Adversary Politics and Elec- pioneering work on economic free- which, through the separation of pow- toral Reform. London: Anthony Wigram. 8 dom against types of government and ers among the executive, legislative, and Finer, S.E., (1982). “Adversary Politics and the electoral systems. There seems to be a judicial branches of government, inten- Eighties.” Electoral Studies 1: 221-30. relationship between levels of eco- tionally tangles the passage of legislation, Lijphart, A. (1984). Democracies: Patterns of nomic freedom and the system of gov- thereby minimizing the opportunity Majoritarian and Consensus Government in ernment, but not between the type of for government involvement in the Twenty-one Countries. New Haven, CT: Yale electoral system. economic affairs of the nation? University Press. Lijphart, A. (1993). “Constitutional Choices for The economic freedom index scores of Or, more elusively, is American afflu- New Democracies.” In L. Diamond & M.F. Plattner, (eds), The Global Resurgence of both plurality and PR parliamentary ence an evolutionary by-product of a Democracy (pp 146-158). Baltimore: Johns systems differ by only 0.2 points. Simi- social culture that inherently favours Hopkins University Press. larly, the index scores for hybrid sys- private economic solutions over public Lijphart, A., and Crepaz, M.M.L. (1991). “Cor- tems differ by just 0.1 percentage point. ones, that more enthusiastically re- poratism and Consensus Democracy in However, the difference between parlia- wards success, and that views Eighteen Countries: Conceptual and Empiri-

14 FRASERForum / May 2000 2 cal Linkages. British Journal of Political Sci- Research. Available on the internet at Presidential systems incorporate aspects of ence, 21:235-246. www.cepr.org. both majoritarianism and consensualism. For Linz, J. (1990). “The Perils of Presidentialism.” Powell, B. (1982). Contemporary Democracies. further information on presidential systems of Journal of Democracy 1(1): 51-69. Participation, Stability, and Violence. Cam- democracy see: J. Linz (1992). “The Perils of bridge, MA: Harvard University Press. Presidentialism” in A. Lijphart (ed) Parliamen- Mainwaring, S., and M. Shugart, (1997). “Juan tary vs Presidential Government and S. Mainwar- Linz, Presidentialism, and Democracy.” Wayne, Stephen J. (1992). The Road to the White ing and M. Shugart (1997). “Juan Linz, Comparative Politics 29: 449-471. House, 1992, New York: St. Martin’s Press. Presidentialism, and Democracy” Comparative Organisation for Economic Co-operation and Politics 29. Development. (1999). OECD Economic Out- Notes 3 look. For further information see: A. Lijphart (1984), 1 Democracies, ch 1: The Westminster Model of Persson, T., and G. Tabelli, (1999). The Size and The following article is based on a paper sub- Democracy. Scope of Government: Comparative Politics mitted to Simon Fraser University by Jason Cle- 4 With Rational Politicians. Discussion Paper mens, one of the co-authors, as part of his For further information see: A. Lijphart (1984), No. 2051 for the Centre for Economic Policy on-going studies. Democracies, ch 2: The Consensus Model of Democracy. 5 It is important to note that there are myriad Fences for Improving Governance variations on the basic concept of electing repre- continued from page 10 sentatives in accordance with the percentage of the popular vote received rather than according organizations and turn them into instruments for excessive government inter- to which candidate received the most votes in a vention, inflation, and other damaging policies that they were supposed to cor- particular district. 6 rect. Many people in England believe that membership in the European A Progressive Conservative-Reform coalition Monetary Agreement will lead to even more influence of the European Commu- would have required the support of one of the nity bureaucracy in Brussels. This bureaucracy has increased regulations and so- six “Other” parties in order to establish a coali- cial policies on its members, which many Britons think are undesirable. Hence, tion minority government. 7 Britain has not joined the European Monetary Agreement. See William Mackness (1998), Canadian Public Spending: The Case for Smaller More For this reason, every effort must be made to ensure that treaties and interna- Efficient Government, The Fraser Institute; P. tional organizations designed to put fences around Canadian politicians are Grossman (1998), Government and economic themselves endowed with strong constitutions and charters which assure their growth: A non linear relationship, Public independence and commitment to the causes they were designed to pursue. Of Choice; E. Peden (1989), “Government Size, Productivity, and Economic Growth: The course, there are no fail-safe treaties and constitutions, but vigilance and clauses Post-war Experience, Public Choice; G. Scully for opting out of the agreements can avoid gross abuses and effects on members. (1991), Tax Rates, Tax Revenues and Economic Growth, NCPA Policy Report No. 98, National Notwithstanding the cautionary note in the preceding paragraphs, it remains Center for Policy Analysis, Dallas, Texas; V. true that fences to protect the public from selfish politicians and to protect politi- Tanzi and L. Schuknecht (1995), Growth of cians from pressures by special interest groups are powerful policy instruments Government and the Reform of the State in to improve governance. Such fences would do in the field of economic rights Industrial Countries, IMF Working Paper 130, Washington, DC; and V. Tanzi and L. Schuk- what the Charter of Rights and Freedoms has done for human rights. Such fences necht (1997), ”Can Smaller Government would improve living standards in Canada. Their potential should be examined Secure Economic and Social Well-Being?" in closely by those interested in improving governance in Canada. Herbert Grubel, ed., How to Spend the Fiscal Dividend: What is the Optimal Size of Govern- ment? The Fraser Institute. Notes 8 See, for example, James Gwartney and Robert 1 Lawson with Dexter Samida, Economic Freedom Herbert Grubel, ed., How to Use the Fiscal Surplus (Vancouver: Fraser Institute, 1998), p. 134. 2 of the World, 2000 Annual Report, Vancouver: H. Grubel, The Case for the Amero, Critical Issues Bulletin (Vancouver: Fraser Institute, 1999).  The Fraser Institute, 2000. 

May 2000 / FRASERForum 15