Singapore's Political Economy: Two Paradoxes

Economist Bryan Caplan, noted for his insights on public choice, visited Singapore in November 2008. His observations address how “economically efficient, but politically unpopular” policies might successfully be carried through a democratic system, and sheds light on the environment in which public policy is made and implemented in Singapore.

Date Posted

1 Jul 2009


Issue 6, 14 Jul 2009


Singapore, when compared to almost any other democratic country, has two deeply puzzling features.

Puzzle #1: It frequently adopts policies that economists would call “economically efficient, but politically unpopular”.1 For example, Singapore has near-unilateral free trade, admits large numbers of immigrants, supplies most medical care on a fee-for-service basis, means-tests most government assistance, imposes peak load pricing on roads, and fights recessions by cutting employers’ taxes. These are policies that could easily have cost politicians their jobs in many other democracies, yet they have stood the test of time in Singapore.

Puzzle #2: Even though Singapore follows the forms of British parliamentary democracy, it is effectively a one-party state. The People’s Action Party (PAP) has held uninterrupted power since the country gained Home Rule in 1959, has never received less than 60% of the popular vote,2 and has always enjoyed an overwhelming majority in Singapore’s Parliament—it currently holds 82 out of 84 seats.

To put these paradoxes in perspective, we need to review the Median Voter Model, the workhorse of studies in modern political economy. In the Median Voter Model, two political parties compete for votes by advocating a “platform”—a bundle of policies. Citizens in turn vote for the party with the platform closest to their ideal policy bundle.

Setting aside various complications,3 the Median Voter Model implies that competing political parties tend to offer policy platforms that centrist voters regard as ideal. This also implies that a rival party should be able to match a dominant party’s electoral success simply by mimicking its centrist platform. As a result, parties rarely achieve lasting political dominance.

Despite its abstractions, the Median Voter Model usually fits the facts of actual political behaviour in democracies around the world. However, it also highlights why Singapore’s political economy is so puzzling. Singapore persistently adopts policies that would have been overturned by the democratic process almost anywhere else on earth, but the same party keeps winning election after election by a landslide. Why doesn’t a rival party promise to abolish unpopular policies and soar to power? How, in short, is Singapore’s political-economic equilibrium possible?


One common conclusion is that Singapore must be—despite its Westminsterian pedigree—a thinly-veiled dictatorship, which informally suppresses political rivals and rigs its elections, which in turn allows the Government to unilaterally adopt unpopular (yet efficient) policies. This “Singapore as a thinly-veiled dictatorship” theory coheres neatly with Western stereotypes about the city-state, and elegantly resolves the two paradoxes. Unfortunately, this dictatorship thesis ignores three basic facts.

First, Singapore has several legal opposition parties;4 they may face minor indignities but hardly live in mortal fear of the PAP.5 Pressure from the dominant party is a feeble explanation for the opposition’s near-total failure to gain political office, given that many countries (such as Pakistan) demonstrate vigorous electoral competition despite far graver dangers.

Second, while there are unusual restrictions on political expression, these shield people from criticism, not policies. Opposition candidates who avoid personal attacks against PAP politicians can and do freely attack specific policies as ineffective or unfair. An opposition candidate could safely campaign on a platform to abolish Electronic Road Pricing or slash immigration. Indeed, an opposition candidate could safely campaign on a platform to rein in politically-motivated defamation suits. In the Median Voter Model, embracing these positions would quickly usher opposition politicians into power—assuming, of course, that the median voter genuinely wants the changes in question.

Third, there is no evidence that Singapore’s elections are corrupt. Indeed, international observers have consistently rated its government as one of the least corrupt in the world,6 with elections that are “free from irregularities and vote rigging”.7 The Global Barometer country report for Singapore finds that 86% of Singaporeans believe that their elections are either “completely free and fair”, or “free and fair, but with minor problems”.8 So while decades of one-party electoral dominance is frequently a strong symptom of electoral corruption, this is not the case in Singapore.

International observers have consistently rated Singapore’s government as one of the least corrupt in the world.

This is not to deny the many peculiarities of Singaporean politics.9 My point is that these peculiarities are largely irrelevant as far as the Median Voter Model is concerned. In Singapore, voters are free to vote for opposition candidates, and opposition candidates can safely advocate the elimination of unpopular policies. In the Median Voter Model, this is all you need for the will of the people to prevail.


Many of Singapore’s successful policies would be considered unpopular around the world, but they persistently survive the democratic test in Singapore. Once we reject the dictatorship hypothesis, the next obvious explanation for Singapore’s effective policies is that its electorate is unusually economically literate. In my book The Myth of the Rational Voter: Why Democracies Choose Bad Policies,10 I find that even in the relatively marketoriented US, the market mechanism is unpopular, especially in international and labour markets. Could Singapore be the exception that proves the rule—a country where the man-in-the-street embraces market mechanisms?

It is worth pointing out that this “unusual economic literacy” hypothesis largely fails for the country where it is most plausible: Hong Kong, which has been ranked the freest economy in the world since 1970.11 Under laissez-faire policies, Hong Kong enjoyed decades of remarkable economic growth. One would expect this excellent performance, combined with status quo bias, would lead to popular support for laissez-faire policies. It does not: Lau Siu-kai and Kuan Hsin-chi12 find that a majority of Hong Kong residents want to change many of its most distinctive policies.

Given Singapore’s many economic, political and cultural similarities to Hong Kong, it seems unlikely that Singaporean public opinion would be significantly better. Unfortunately, nothing comparable to the survey by Lau and Kuan exists for Singapore. But the 2002 World Values Survey13 data suggests little reason to believe that the Singaporean electorate is unusually economically literate. Singaporeans and Americans have virtually the same beliefs about the social benefits of competition,14 and Singaporeans are more likely to accept income inequality as a trade-off for better incentives.15 Yet Singaporeans are actually less sympathetic to private enterprise than Americans,16 and seem to favour a much more restrictive approach to immigration than do Americans.17

Many of Singapore’s successful policies would be considered unpopular around the world, but they persistently survive the democratic test in Singapore.

There is admittedly some indication that Singaporeans are unusually concerned about economic performance. The survey showed that 58.8% of Singaporeans say that “a high level of economic growth” should be their nation’s top priority; 48.6% of Americans say the same.18 Similarly, only 37.9% of Singaporeans—versus 65.2% of Americans—think it would be a “good thing” if people put “less emphasis on money and material possessions”.19 Once Singaporeans recognise the economic benefits of a policy, they seem more willing to support it. Still, at this point, the case for the “unusual economic literacy” hypothesis looks rather weak.


Even if Singaporean public opinion were unusually economically literate, it would still be hard to explain the PAP’s dominance. In the Median Voter Model, opposition parties’ best response would be to mimic the policies of the ruling party, leaving voters indifferent. Singaporean politics plainly doesn’t work this way; it seems to be in a political class of its own as long as we think of it primarily as a country.

The picture changes radically if we instead think of Singapore as a city. In the United States, big city politics is often about as lopsided as Singaporean politics. Democratic mayors have won without interruption since 1931 in Chicago20 and 1964 in San Francisco.21 While the Democrats have failed to monopolise the mayor’s office in New York City, they have near-PAP dominance of the New York City Council: Democrats hold 45 out of 48 occupied seats.22 Note that in the Median Voter Model, this cannot be explained purely by the liberalism of urban voters. After all, why can’t the Republican parties in Democratic cities simply move sharply to the left?23

In purely formal terms, the Median Voter Model can account for one-party democracy if you assume that voters have not only policy preferences, but party preferences as well.24 This means that even if two parties were to offer identical policies, some voters would still decidedly prefer one party to the other. This then allows the favoured party some “wiggle room” to deviate from the public’s policy preferences without courting defeat in the next election.25 Well-informed and well-meaning politicians could use this to persistently deliver economically efficient but politically unpopular policies.

Why precisely would voters have these “party preferences”? They might reflect group identification or loyalty. Voters might see one party as being “their party”, just as they see the local sports team as “their team”. Preferences may reflect deference—a belief in one party’s superior competence and/or intentions. This could stem from a successful track record; but voters could also be deferring to politicians’ current traits, such as intelligence and charisma. A final, more pessimistic interpretation is that they reflect resignation. A voter might favour one party over another not because he wants it to rule, but because he feels that resistance is futile. In the US, for instance, people who sympathise with a third party rarely vote for it because “it has no chance of winning”.

Which of these mindsets seem to fit the realities of Singapore? While empirical evidence is scarce, several different sources confirm the importance of deference in Singaporean politics. Singaporeans are markedly more satisfied with their national leaders and convinced of their good intentions than Americans.26

Critics seem to endorse the deference hypothesis, suggesting that the “success of the government in the economy” has lent it a significant “performance legitimacy”.27 Compared to Americans, Singaporeans show little interest in “giving people more say”; just 19.7% make this a top priority, compared to 32.6% of Americans.28

The secret to Singapore’s success seems to lie in its electorate’s “party preference” for a ruling party that takes economic reasoning seriously.

While the deference story does fit well, there is also considerable evidence that resignation is also at play. Singaporeans’ unusually low professed interest in politics is telling.29 The stereotype of the apolitical Singaporean appears to have much basis in fact.

Overall, it appears that the answer to the paradoxes of Singaporean political economy can be attributed to a range of “party preference” factors. More systematic research is necessary to figure out which variants best fit the facts, as well as the relative importance that deference, resignation and party loyalty play in Singapore’s politics and policies.


In the West, Singapore is widely perceived as a benevolent dictatorship. From this starting point, social scientists have little to learn from Singaporean political economy. The explanation for Singapore’s success is simply that it had the good fortune to be ruled by the smartest, nicest dictators on earth.

Once misconceptions about Singapore’s democratic credentials are corrected, however, the city-state looks “curiouser and curiouser”; it seems to contradict everything that experts think they know about democracy: How can any party honestly win election after election—much less a party committed to many economically efficient but unpopular policies?

There is little evidence to suggest that Singaporean voters are markedly more economically literate than voters in other countries. The secret to Singapore’s success seems to lie in its electorate’s “party preference” for a ruling party that takes economic reasoning seriously; this preference gives the party enough slack to implement policies that might not survive a direct popular referendum.

Understanding the paradoxes of Singapore sheds new light on political economy in general. While most democracies have frequent partisan turnover at the national level, subnational democratic politics are often as one-sided as in Singapore. In the broader world though, such forms of one-party democracy do not seem to depend on the delivery of remarkable economic performance. Is this because the relative importance of loyalty, deference, and resignation varies? Or did Singaporeans simply have the good fortune to put its trust in men who happened to deserve it?

Once political economists have a better understanding of one-party democracy, they will be ready to take a second look at national politics. Why exactly is it so difficult for one party in a democracy to stay on top at the national level? One interesting hypothesis is simply that people are more interested in—and therefore less resigned about—national politics. But this raises further questions: What determines whether a given democratic contest will catch voters’ interest? And under what circumstances does greater interest lead to worse policies?

The case of Singapore is a fascinating challenge to time-tested models of how democracy works. But more importantly, the mechanisms underlying Singaporean political economy are probably at work in every democracy. These mechanisms are not unique to Singapore, just uniquely visible.


Dr Bryan Caplan is Associate Professor of Economics at the Department of Economics, Center for Study of Public Choice, and Mercatus Center, George Mason University. Dr Caplan is the author of The Myth of the Rational Voter: Why Democracies Choose Bad Policies, published in 2007 by Princeton University Press. It was named “the best political book this year” by the New York Times and a Best Book of 2007 by the Financial Times. His articles have appeared in the American Economic Review, the Economic Journal, the Journal of Law and Economics, the Wall Street Journal, the New York Times and the Washington Post.


  1. See especially Ghesquiere, H., Singapore’s Success: Engineering Economic Growth (Singapore: Thomson Learning, 2007).
  2. Mutalib, H., Parties and Politics: A Study of Opposition Parties and the PAP in Singapore (Singapore: Marshall Cavendish Academic, 2004): 5.
  3. See e.g. Cooter, R., The Strategic Constitution (Princeton, Princeton University Press, 2000): 17-49. One seemingly strong assumption—that preferences are one-dimensional —has received a surprising degree of empirical confirmation. See e.g. Poole, Keith, and Howard Rosenthal, Congress: A Political-Economic History of Roll Call Voting (New York: Oxford University Press, 1997).
  4. These include the Workers’ Party of Singapore, the Singapore Democratic Alliance, and the Singapore Malays National Organisation. The only illegal party is the Communist Party of Malaya. See generally Mutalib (2004), and Mauzy, D., “Electoral Innovation and One-Party Dominance in Singapore”, in Hsieh, J., and D. Newman, eds., How Asia Votes (New York: Chatham House Publishers, 2002): 246.
  5. See generally Mauzy (2002): 241-245; Mutalib (2004): 239-267. As Mauzy and Milne observed: “The Singapore government has not committed any serious violations of civil rights. There have been no extrajudicial killings or political disappearances, and there are currently no political detainees.” In: Mauzy, D., and R. Milne, Singapore Politics Under the People’s Action Party (New York: Routledge, 2002): 128.
  6. The World Bank’s Governance Matters data set, for example, gives Singapore stellar scores in “Rule of Law” and “Control of Corruption”. See:
  7. See:
  8. Tan, E. and Wang, Z., “The State of Democracy in Singapore: Rethinking Some Paradoxes”. Paper presented at conference entitled, “The Asian Barometer Conference on The State of Democratic Governance in Asia” organised by Asian Barometer, Taipei (Taiwan), 20-21 June 2008.
  9. In most democracies, the members of the ruling party respond to their opponents’ verbal abuse with more verbal abuse—not lawsuits.
  10. Caplan, B., The Myth of the Rational Voter: Why Democracies Choose Bad Policies (Princeton: Princeton University Press, 2007).
  11. See:
  12. A majority of Hong Kong residents want to change many of its most distinctive policies: 57.6% favour a minimum wage, 68.4% favour price controls for necessities, 74.7% want more progressive taxation, and 75.5% want to “protect local industry against foreign competition”. In: Lau, Siu-kai and Kuan Hsin-chi, “Public Attitude toward Laissez Faire in Hong Kong”, Asian Survey 30 (1990): 770.
  13. See:
  14. World Values Survey variable identifier E039.
  15. When asked whether “incomes should be made more equal” or whether “we need larger income differences as incentives” (higher scores on a 1-10 scale indicate greater support for incentives), Singaporeans’ average answer was 6.88, versus 5.72 for Americans (World Values Survey variable identifier E035).
  16. When asked whether private ownership or government ownership should be increased (higher scores on 1-10 scale indicate greater support for government ownership), the average answer in Singapore was 4.75, versus 3.62 for the US (World Values Survey variable identifier E036).
  17. Only 4% of Singaporeans favour open borders, and just 24% are willing to admit immigrants “as long as jobs [are] available”; the comparable numbers in the United States are 12.4% and 44.8% according to World Values Survey variable identifier E143. Mauzy and Milne confirm this pattern: see Mauzy and Milne (2002: 152 and 2002:191).
  18. World Values Survey variable identifier E001.
  19. World Values Survey variable identifier E014.
  23. See generally Schleicher, D., “Why is There No Partisan Competition in City Council Elections?: The Role of Election Law”, Journal of Law and Politics 23 (2007): 419-473.
  24. Lindbeck, A., and Weibull, J. W., “Balanced-Budget Redistribution as the Outcome of Political Competition”, Public Choice 54 (1987): 273-297; Caplan, B., “When Is Two Better Than One? How Federalism Amplifies and Mitigates Imperfect Political Competition”, Journal of Public Economics 80 (2001): 99-119.
  25. Why then is one-party democracy so much more common in cities than countries? The most plausible explanation is that small, dense populations tend to be less heterogeneous—not only in their policy preferences, but in their party preferences as well. As Mutalib (2004: 272) suggests, “A larger land area and population base would allow greater avenues and wider opportunities for political dissension and other sectional or geographical interests to be articulated.”
  26. World Values Survey variable identifiers E125 and E128.
  27. See Mutalib (2004): 239.
  28. World Values Survey variable identifier E003.
  29. Only 3.2% of Singaporeans say they are “very interested” in politics, and another 32.8% say that are “somewhat interested”; in the US, the corresponding numbers are 18.3% and 47.2% (World Values Survey variable identifier E023). The World Values Survey consistently finds that compared to Americans, Singaporeans are extremely reluctant to engage in even low-level political participation, such as signing a petition or attending a rally (World Values Survey variable identifiers E025 and E027).

Back to Ethos homepage