David Emanuel Andersson

The Social Importance of Tolerance May 16, 2009

Inglehart likes to argue that postmaterialist values are strongly associated with democracy (see my preceding blog entry). Inglehart further contends that postmaterialists – according to his four-item, eight-item, or twelve-item indices – will be increasing as a proportion of the population as long as there is sustained economic growth and peaceful external relations. This should reflect a stable cohort effect where values are established during childhood socialization processes.

While I agree with the thrust of Inglehart’s arguments and hypotheses, I think that the latest wave of the world values survey demonstrates that other measures are better at identifying cohort effects than any index of postmaterialism.  For example, the percentage of American WVS respondents who identified strong defense forces as their top priority – one of the key materialism indicators – was 15.5% in 1990, 14.3% in 1995, 16.0% in 1999, but 29.0% in 2006. Meanwhile, “protecting freedom of speech” – a key postmaterialism indicator – was 22.6% in 1990, 21.2% in 1995, 25.4% in 1999, but then went down to 17.8% in 2006. This is not what we would expect from a stable cohort effect — instead I suspect that the perceived insecurity of adulthood rather than childhood explains the greater proportion of materialists in the United States in 2006 than during the 1990s.

However, Inglehart’s charts indicate that intolerance is as strongly associated with modernism as postmaterialism is with postmodernism. But tolerance seems less susceptible to external shocks such as terrorist attacks or preparations for war. For example, the percentage of Americans that would not like neighbors of another race than their own was 8.7% in 1990, 6.6% in 1995, 8.0% in 1999 and 4.1% in 2006. A similar downward trend is for the most part evident for the percentage wishing to avoid gay neighbors: 38.6% in 1990, 29.5% in 1995, 23.3% in 1999 and 26.0% in 2006. And I think the greater stability of the tolerance measures makes a lot of intuitive sense: why would an economic recession or a war make me less willing to have gay neighbors, assuming that I grew up in a non-homophobic environment? On the other hand, it is plausible that a recession would focus my attention on the prospects for economic growth and that a war would make me worry about the nation’s defense capabilities (admittedly the second scenario is not plausible in my case; I have always been an admirer of the defense policies of Costa Rica, the Norwegian territory of Svalbard and the autonomous Finnish island of Åland).

To back up my hypothesis, I have spent the last week estimating associations between tolerance and various measures of institutional performance that I believe are relevant for anyone with broadly democratic values: freedom of the press (Freedom House), political rights and civil liberties (Freedom House), corruption (Transparency International) and the Democracy Index of the Economist magazine. I also looked at a performance measure that appeals beyond committed liberals or democrats: the Human Development Index of the United Nations, which reflects per capita GDP, life expectancy, and literacy; to complete the picture I then looked at a favored measure among classical liberals: the Economic Freedom Index of the Fraser Institute. Finally, I combined all these indices into an aggregate index that should reflect combined performance in terms of civil liberties, democracy, rule of law, economic freedom, and material standard of living (I call it the Socio-economic Development Index, or SDI).

There is a very predictable pattern that is revealed when perusing all these indices: the best-performing ten countries are almost always the Nordic countries, the Netherlands, Switzerland, Canada, Australia, and New Zealand, closely followed by countries such as the US, the UK, Germany, Belgium, France, and Spain. The exception is the economic freedom index, where the top 10 consist of English-speaking countries, Switzerland, and two well-known Asian city states, closely followed by the Nordic countries with the notable exception of Sweden.

Anyway, the correlations between a special Tolerance Index (acceptance of other-race, immigrant, and gay neighbors and general acceptance of homosexuality, divorce, and prostitution) and the various performance measures turned out as follows, including all 83 countries that took part in the WVS at least once between 1995 and 2008 with the  relevant questions included:

Socio-economic Development Index: .807

The Economist’s Democracy Index: .791

Corruption Perceptions Index: .727

Freedom of the Press Index: .704

Political Rights and Civil Liberties Index: .704

Human Development Index: .667

Economic Freedom Index: .564

Partial correlation analysis revealed that the effect of the Tolerance Index remained highly statistically significant after controlling for Inglehart’s Postmaterialist Liberty Aspirations Index (e.g. .596 as the partial Tolerance Index -SDI correlation), while the reverse was not true (only  .165 between Inglehart’s index and the SDI when controlling for the Tolerance Index, which is insignificant at the one-tailed .05 level). The Tolerance Index also remains significant after controlling for per capita GDP (e.g. .486 between the Tolerance Index and the Democracy Index when controlling for per capita GDP in purchasing power parity dollars).

Another interesting result is that the two measures of tolerance toward gays were almost as strongly correlated with the various freedom and democracy measures as the Tolerance Index itself. For example, the simple correlation between acceptance of gay neighbors and the aggregate SDI index is .762, which is a remarkably close association.

Inglehart emphasizes that the evolution of values, economies, and political systems are interdependent, and that consequently there is neither a Weberian priority given to values as ultimate deteminants, nor a Marxian priority of economic relationships. I agree. But what I think this simple analysis shows is that tolerance is closely associated with freedom, democracy, the rule of law, and economic well-being. This is not a new insight, but I think that the relationship is much stronger than most people (including me) would have dared hope for.

I have to admit that I am not a value-neutral analyst in this case. I actually think that tolerance is the greatest of the social virtues. And it’s great to have a good evidential argument against social conservatives: whether your priority is freedom of speech, democracy, free markets, the rule of law, or economic development it follows that social tolerance is your friend, not your enemy. And the tolerance that matters most is sexual tolerance, closely followed by multicultural and racial tolerance. In other words, intolerant goals can probably only be attained at the cost of living in a less free, less democratic, more corrupt, more regulated, and less prosperous society.

This conclusion actually echoes Richard Florida’s arguments, whose main point has been that economically dynamic and culturally creative regions require tolerant values, as well as talented individuals and high technology. Florida is of course also famous for his “gay index,” which measures regional concentrations of gay residents rather than national attitutes toward gays or homosexuality. While I was originally a little skeptical (I thought that the notion of a “gay index” is more likely to be noticed than other indices, and is therefore an attractive self-promotion strategy), I now think that Florida is really onto something with his “3T” message. And this is all to the best.

Meanwhile, these indices and results can be used to offer location advice in a globalized world:

For tolerant free-market types: Switzerland

For tolerant welfare-state types: Sweden

For intolerant free-market types: Singapore

For intolerant welfare-state types: France (ok, intolerant by OECD standards only, not by global standards)

For homophobes: Jordan, Pakistan, and Bangladesh should all be ideal, although Iran is a close runner-up.

For tolerant people who don’t like snow: Spain

In case you are wondering, the top ten countries according to the tolerance index are Andorra, Sweden, the Netherlands, Switzerland, Spain, Iceland, Denmark, Australia, Germany, and New Zealand. The bottom ten (of 83 surveyed nations) are – beginning at the bottom of the list – Bangladesh, Jordan, Iran, Pakistan, Turkey, Indonesia, Georgia, Rwanda, Nigeria, and Morocco.