Monday 25 February 2013

Minimum wages and climate change

Stephen Gordon wants to construct a Venn diagram showing the proportion of people who accept the evidence on minimum wages (it's a poor way of helping the poor) and who also accept the evidence on global warming (the place does seem to be getting warmer); he expects the intersection to be disappointingly thinly populated.

I can't help him out exactly, but I can add a bit.

The 2008 New Zealand Election Survey asked whether the government should control wages and whether strong action is needed on global warming, but had no questions on the employment effects of minimum wages. It's not the best: maybe some folks want maximum wages but don't like minimum wages, and maybe some (like me) are happy to take the evidence on global warming but are less convinced we need to invest massive resources in mitigation today - I'd have been somewhere between neutral and support on that question.

Here's the raw cross-tab.
Those who accept the science on government wage controls should oppose or strongly oppose them; those who deny the science on government wage controls will be more likely to support or strongly support them.

There are 1,471 of 2,892 respondents who accept the science on wages and 542 of 2,892 who reject it. Among those accepting the science on wages, 46.6% support or strongly support government action on global warming while 27% oppose or strongly oppose it. Among those neutral or opposed to the science on wages, 60.4% support or strongly support government action on warming while 15.2% oppose or strongly oppose it. These hit the 7+ t-stats. So disagreeing with the science on wages seems to predict stronger support of climate policy.

Among those supporting or strongly supporting strong government action on global warming, 45.8% accept the science on wages and 22.9% reject it. Among those neutral or opposed to strong action on climate, 59.7% accept the science on wages while 12.9% deny it. The t-stats on these across groups are higher than 7.

So if those supporting government action on climate are more likely to have supported the science, those supporting climate science seem significantly less likely to accept the science on wages and significantly more likely to reject the science on wages. And supporting the science on wage controls correlates with lower support for government action on climate and stronger opposition to it.

It's not a pure test because it's not anti-science to say that the scientists are right about mean expected warming over the next century but still to oppose "strong action" because you don't think it passes cost-benefit analysis. But you'd expect that there'd at least be a positive correlation between accepting the science and wanting action - it would seem odd to want action on climate change while thinking there's no warming.

The survey also has a measure of self-reported ideology: 0 left, 9 right, 5.4 mean. Another fun fact: dropping all the "don't know" respondents, mean self-reported ideology is 5.5 among those accepting wage science and 5.2 for everyone else; mean reported ideology among those wanting strong action on climate change is 4.8 and 6 for everyone else. The t-stat on group differences in ideology on climate is 11.9; on wages, 3.6. So the ideological divide on climate policy seems greater than that on wage controls.

I'd previously put together a couple of factor scores pulling together responses on social questions to get a measure of social liberalism and one on economic questions to get a measure on economic liberalism. I'd left the climate action question out of both factor analyses because supporting "action" on climate is neither pro- nor anti-market.* So I have a mean zero, SD 1 measure on social liberalism (higher is more liberal) and on economic liberalism (higher is more liberal).

A couple quick and dirty specifications have social liberalism strongly predicting support for climate policy, economic liberalism strongly predicting opposition to climate policy, household income not affecting preferences, and education predicting increased support for climate policy. In the ordered logit specification, a standard deviation increase in social liberalism predicts a 0.41 standard deviation increase in support for climate policy; a standard deviation increase in economic liberalism predicts a 0.6 standard deviation decrease in support for climate policy; a standard deviation increase in education predicts a 0.17 standard deviation increase in support for climate policy. There aren't any other measures in there that could capture generalised attitudes towards science, alas.

There are plenty of reasons why economic liberals could come out less in favour of strong action on climate change. A few candidates, some better than others:
  1. Accept the science, but reckon future mitigation is more likely to pass cost-benefit, or that other projects are more worthwhile (a la Lomborg). Or, in stronger form, accept the science, be sceptical about the prospects for policy to fix things, and recognise that a warmer world could well be a better world up through, say, three degrees of warming. David Friedman makes the best argument along these lines** (his earlier blog post here). I don't think anybody who grew up in Manitoba can deny that there are some positive effects from a gradual warming.
  2. Accept the science, see a need for policy, but reckon that the "strong action" mentioned in the question means something more than the standard economic advice of a revenue-neutral carbon tax that can ramp up over time.
  3. Accept/agnostic on the science, but see that most of the folks shouting loudest for climate action are a bit nuts on other economic issues and be hanged if you'd ally with them - heck, some seem to think that reduced economic growth is a feature rather than a bug of some climate policies. And the same bunch that shouts about global warming also reckoned that peak oil was a serious concern - which was utterly insane given that, if peak oil had been right, it would have been a part of the solution to warming! As David Friedman put it: even if there were zero evidence of global warming, many of the proponents of anti-warming policies would still support those policies, but on other grounds. 
  4. Reject the science: the loudest proponents are completely wrong on the economic issues you know something about, and really seem to have worked backwards from "policies I support" to "the data must have said X" in assessing things on those margins, so you can't reject that they've done the same here. Note that this is stronger than the explanation immediately prior: it says that the scientists are part of some kind of conspiracy.
  5. Reject the science: macroeconomic models are a bit nuts, and climate change models have all the nuttiness of some of the big macro models but with even more uncertainty about cloud feedback loops. If economists can barely get a consensus on the government spending multiplier, how can we trust coefficients on climate sensitivity? And if sensitivity were scary bad, how did the planet ever manage not to turn into Venus a few million years ago? Sure, the models look ok over the period of calibration, but their out-of-sample predictions of warming in the 2000s weren't all that great. Until the models can figure out why warming leveled off in the 2000s, should we really trust what they say about 2150? 
  6. Pure mood/expressive affiliation, or that in combination with that pretty much every other prediction of global doom has been rather wrong.
I'm personally somewhere between 1 & 2: a low carbon tax capable of being ramped up over time could slow the pace of warming, giving more time for adaptation, and helping to guard against the scarier warming scenarios.***  I suspect that some of the opposition to climate science among those who are not climate scientists and who are not in a position to personally evaluate the quality of the literature comes from 4 & 6.

There's a reasonable contingent of pro-market people**** who are happy to take the science on climate and figure a revenue-neutral carbon tax isn't all that bad. Would that more of the pro-climate-policy crowd would come over to the intersection of Stephen Gordon's Venn diagram. It's mildly frustrating that the New Zealand Green Party excoriates those opposed to rather strong action on global warming as anti-science while rejecting the consensus views of economists on economic policy.

* Club Pigou is pro-market; Club ban-everything-that-emits isn't. Note that the appropriate domain of Club Pigou ought to be bounded.

** The first half-hour of the linked Friedman video provides a wonderful exposition of how economists think about externality and policy; strongly recommended.

*** I take David Friedman's point on that the Nordhaus / Weitzman insurance argument for climate policy is flawed where it considers fat tail risks of doing nothing while ignoring fat tail risks of doing something. But if the main potential low-probability high-cost risk of emitting too little is another ice age, it seems easier to ramp up CO2 emissions if things look like they're heading that way than it would be to remove CO2 that's already been emitted.

**** Did I mention Club Pigou?


  1. Great post. Can you do the same on the intersect between people who accept scientific evidence on global warming and people who accept scientific evidence on the safety and environmental impact of GM crops?

  2. I wish I could. The 2005 NZES included questions on GMOs but not global warming; the 2008 NZES replaced the GMO questions with a warming one.

  3. Haha, great idea. I suspect there'd be a relatively small intersection in that Venn diagram.

  4. Nice post. I think its becoming clear that Positive Feedback causing Radiative Forcing is not happening so we wont be seeing 3-6 degree rises anytime soon. .7-1.2 degree per century is about right. CO2 is the wrong target, if they want to make a difference then Black Carbon (soot) should be the target. That really is altering our environment. However it is concerning to see that these Green policies all require an increase in the price of energy. I would have thought that being able to provide the cheapest energy possible would be beneficial to our economy and community. In NZ it should be the basis of our competative advantage. Cure "poverty" perhaps.
    Especially since it could (ironically) start to get colder. The solar cycles are stating to match the pattern for a Maunder Minimum.....

  5. You say "It's mildly frustrating that the New Zealand Green Party excoriates those opposed to rather strong action on global warming as anti-science while rejecting the consensus views of economists on economic policy." You say But in truth the scientists who do climate change are held in vastly greater esteem for their scientific credentials than economists. Dont recall a consensus view from economists before the global financial crisis (or if there was a consensus, it was that all is well with the world). And I dont imagine too many economists who possibly advise against a minimum wage actually want to live in a country that does not have one.

  6. There are some things on which economists have a pretty strong consensus. They're mostly microeconomic policy issues rather than macro. But they're often strongly inconsistent with policy as it exists. For example:

    The scientists who do climate change are held in esteem by those who agree with their conclusions and are dismissed as a clubbish cartel actively working to suppress dissenting views by those who disagree with their policy conclusions.

  7. I used to worry more about sunspots. Then warming continued, albeit more slowly, despite a Dalton minimum - or what folks were telling me was one anyway. If we can simply level off or mildly increase despite a solar cycle saying things ought to be cooling, that makes me more rather than less worried. I have no clue though how folks project Maunder minima.