Friday, 6 March 2009

Measuring Economic Thinking

In previous posts, I discussed political ignorance and the measure of ignorance that I constructed using data from the New Zealand Election Survey. Let's now start looking at whether ignorance explains anything important about individuals' party and policy preferences. First, are the ignorant more or less likely to agree with economists on questions of economics? To answer that, we have to construct a measure of "economic thinking".

The New Zealand Election Survey includes eight policy preference questions that seem to me to address issues about which economists are generally on one side. They're not a perfect set of positive questions, like the ones Bryan Caplan is able to use in his work comparing the views of the public and economists on the economy, but they're not bad. While economists will disagree vehemently about the magnitude of the effects, most economists would agree that minimum wages reduce the creation of new jobs. In our survey, 983 respondents either strongly agreed or agreed that minimum wages reduce the creation of new jobs, 698 were neutral, and 1461 disagreed or strongly disagreed. So, I created a new variable with a score of +2 for those strongly agreeing, +1 for those agreeing, 0 for those neutral, -1 for those disagreeing, and -2 for those strongly disagreeing. Whaples' 1996 survey of economists found that 87% of labour economists agreed that minimum wages increase unemployment among the young and unskilled. This isn't identical to "minimum wages reduce the creation of new jobs", but it isn't far off.

Views as to the effects of minimum wages on job creation is my first measure of "economic thinking". What are the others?

"To solve New Zealand's economic problems, the government should control wages (prices) by law."
Most economists would violently disagree with both options. Here, 319 respondents strongly agreed that prices should be controlled, with a further 863 agreeing. 160 strongly agreed with wage controls, with a further 535 agreeing. The number disagreeing outweighed the number agreeing in both cases, but not by a lot.

"High income tax makes people less willing to work hard."
Economists are certainly divided as to the precise tax elasticity of the labour supply, and especially at current marginal tax rates, but few would disagree that high tax rates would induce a negative labour supply response: it would only run in the opposite direction if most workers were on the backward-bending portion of their labour supply curves. 916 respondents disagreed that high income tax reduces labour supply; a further 138 disagreed strongly.

"There should be a law to further reduce pay differences between women and men."
Most analyses of the matter have shown that gender pay differences disappear once you control for education, work experience, type of job and time outside of the workforce. Whaples (1996) found 83% of labour economists disagreed with such legislation. 773 respondents here strongly supported such legislation, with a further 1180 simply supporting. Only 900 either opposed or strongly opposed.

"The government should introduce import controls."
While some economists have made names for themselves by providing blackboard proofs of specific circumstances in which specific forms of protectionism might help particular countries (look up "terms of trade" argument), most economists would not expect these circumstances either to be identifiable ex ante or that actual tariff policy would ever be conducted in accordance with such theories. Fuller and Geide-Stevenson consequently find 92% of economists agreeing that tariffs and import quotas generally reduce welfare. Here, 518 strongly support import controls with a further 1075 simply supporting. Only 232 strongly opposed with a further 562 simply opposing.

"Immigration is good for the New Zealand economy."
Whaples found that 96% of labour economists agree that "the overall gains to American society from immigration exceed the losses." Here, 766 disagreed or strongly disagreed, with 1925 agreeing or strongly agreeing.

"The government should provide a job for everyone who wants one."
Economists tend to prefer transfers to the poor rather than job-creation schemes. Even in the context of current stimulus packages, most economists would agree that money should be spent only on projects that actually pass a cost-benefit analysis, rather than digging holes and filling them in again. This survey was conducted in the lead-up to the 2005 election, when such arguments would have been far from the front of voters' minds regardless. 738 respondents strongly agreed that it definitely is the government's responsibility to provide a job for everyone who wants one, with a further 1437 simply agreeing. 1060 said that government should not take on that responsibility, with a further 228 saying that it was definitely not the government's responsibility.

So, I have eight measures where I think economists' views are relatively one-sided. How to combine them? Again, I used a principal-component analysis to extract an underlying index of what I call "economic thinking". Like the ignorance measure, it provides an ordinal ranking of respondents' economic thinking, with mean zero and standard deviation one by construction.

Next up: What makes people think like economists? Sneak preview: Political ignorance doesn't help.

Footnote: My principal component measure loaded onto two factors. I re-did all of the analysis I'll later discuss using other ways of measuring economic thinking: first by using another principal component measure where I drop the two questions that load onto the second factor (effect of taxation; immigration), then by just using the sum of the measures. Not much changes.


  1. I don't know about the minimum wage bit: this

    was published in 1997 and probably has changed many economists opinion by now. You may be using outdated data.

  2. Alex: No the Card and Kruger results have little if any real effect, simply because they are not believed. A good summary as to why is the new book "Minimum Wages" by David Neumark, William L. Wascher.

  3. Where economists tend to disagree will be on the magnitude of the effects and whether it's worth the tradeoff: some will argue that a small decrease in job creation is worth paying given the large number of folks who'll see increased wages; others judge the costs of induced unemployment as outweighing the benefits for those who keep their jobs. Both arguments admit an increase in unemployment.

  4. Could you please clarify the government import control question? I'm confused by the "welfare" term. For me, it has particular social connotations. Do 92% of economists agree that protectionism has a negative impact on 'social' welfare? If so, how is 'social welfare' measured? Do 92% of economists agree that import controls have a negative effect on economic performance?

  5. Welfare means the surplus enjoyed by consumers and producers. We typically reckon that import controls make consumers worse off by more than they help particular domestic producers.