Over the last couple of days, my Twitter feed has been swamped with discussion about green jobs. Yesterday, Worthwhile Canadian Initiative blogger and Laval Professor Stephen Gordon (@stephenfgordon) posted a piece on the Globe and Mail’s Economy Lab arguing that presenting, “the employment opportunities generated by a set of proposals…as an additional benefit of the policy agenda..(is) a mistake.” I expected this would generate some push-back when I read the headline, but didn’t expect that Grist.org blogger David Roberts (@drgrist) would be one of those weighing in. He did, with this post. I think they agree on more than either might believe. Here’s why.
First of all, both acknowledge that the green jobs angle has likely gone too far and has lost meaning. Roberts says that, “(green jobs) ended up being taken far too literally by both advocates and detractors, leading to a series of bean-counting skirmishes that cast more heat than light.” He follows-up on this later in his post with, “If the job-creating — or at least non-job-destroying — features of a policy can make the difference between passage and failure, then there is a perfectly rational bias in favor of environmental policies that create jobs.” Gordon’s position? “(The) emphasis on green jobs may simply be a reasonable response to concerns about the employment effects of environmental policy.” Sounds like they are both on about the same page.
Gordon’s main point is that labour, or human capital, is a resource. The more of it you use on one thing, the less of it you have available for some other use. That might sound communist, but it isn’t. It does mean that jobs are a cost, not a benefit. In a market system, if government policies are creating green jobs, other segments of the economy are necessarily facing a tighter labour market than they otherwise would. The contractors working to install solar panels might otherwise be roofers. The electricians hooking up solar systems are not going to be available for other work. Gordon is correct that where policies reward, “hiring more people to produce less energy,” this will decrease aggregate labour productivity in the economy overall.
Roberts’ response is not really contradictory to Gordon’s article. “Golly, Mr. Gordon, you mean “more jobs per kilowatt hour” implies “lower labor productivity”? Shucks, I bet none of the dozens of economists who have written about this ever thought of that!” In fact, Roberts directly states that, “jobs are ‘a cost’…but it is in the nature of wise investing to accept temporary costs in the name of lasting gains.” If there is a disagreement, I expect it would be on whether there is any correlation between the number of jobs created and the wiseness of the investment, but Gordon (and almost any economist) will get behind deploying resources today for a long term gain, as long as the gains justify the investment. Both clearly agree that jobs are not the main goal, but rather a means to an end…i.e. a cost.
Roberts’ next point is effectively that the opportunity cost of labour is really low in a recession…a point with which Gordon is more than familiar. If the economy is far from full employment, a government policy which creates jobs will not have significant negative effects on other employers unless it is drawing labour out of a niche market which remains tight during the recession. There may also be significant social gains to increasing employment in a downturn. That said, there remain important questions about the opportunity costs of the public funds used to create those jobs. As Roberts correctly points out, citing Keynes, “burying money in holes and paying people dig it up would be better than nothing (in terms of creating economic activity).” Just because something is better than nothing doesn’t mean it’s better than some other use of the same dollar. If you are climate-concerned, ask yourself if you would like government dollars funding solar panel installers or home insulation contractors and you’ll be on the right track. If you believe McKinsey, the same number of dollars deployed insulating houses would have a far bigger impact on emissions than those same dollars deployed installing solar panels. Bottom line: if you are going to use scarce resources (labour or government revenue), you need to be able to do better than, “it’s better than nothing,” or “this policy uses more labour than that one.” Scarce resources be invested in the opportunities with the greatest payoff. Gordon and Roberts might disagree on how to value different outcomes, but as a former colleague used to say, “you can’t win an argument about preferences.”
Finally, both Roberts and Gordon discuss the actual point of environmental policy – the environmental benefits. Roberts says that “economists frequently seem to forget (that) environmental policy isn’t…a pure economic drain. It actually creates social and health benefits.” Sure, there are some people with economics degrees who will make this argument, but there are few economists with this narrow a lens. In fact, this is exactly what Gordon was trying to argue for – that an environmental policy should be judged on the basis of its ability to mitigate environmental harms, and thus create social (including health) benefits rather than on the basis of the number of resources the policy demands in order to deliver an outcome. Gordon confirmed his agreement with this point later on Twitter saying, “health benefits should be a goal of environmental policy.”
So, where’s the disagreement? Both agree that jobs are a cost, and that these valuable resources should be deployed to achieve the greatest long term social good. Both agree that mitigation of environmental harms has significant social benefit, and that the cost of using labour in a recession is lower than in a boom. Sounds to me like both are advocating for just about the same thing – as Roberts puts it, “human well-being and sustainable prosperity.” Gordon might call that maximizing an objective function, but that’s about the extent of the disagreement I can find.