The NYT rightly criticized Donald Trump's decision to pull the U.S. out of the Paris climate agreement, but part of its criticism is not right. It dismissed the idea that reducing greenhouse gas emissions would lead to job loss as "nonsense" that comes from "industry-friendly sources." While the claim that reducing greenhouse gas emissions will lead to job loss may be nonsense, it is, in fact, the result that comes from standard economic models that are used all the time to project the impact of regulation policy, tax policy, health care, and trade policy.

These models are all full employment models, which means that everyone who wants to work at the market wage for their skills has a job. The way that reducing greenhouse gases reduces employment is by reducing the real wage. For example, if gas and electricity cost more, and wages have not risen to account for this increase, the real wage will be less. In these models, at a lower real wage fewer people will decide to work.

So, if complying with our Paris commitments causes the real wage to be 1.0 percent lower, then this may lead 0.5 percent fewer people to want to work, which translates into roughly 800,000 fewer people working. (These numbers are hypothetical, not taken from actual models.) So when Trump is citing models showing job loss associated with reducing greenhouse gas emissions, he is actually relying on mainstream economics (there is still a considerable range in this modeling, as some is almost deliberately dishonest).

There is one other point worth making on this topic. The military spending that Trump is so fond of also kills jobs in these models. Pre-Iraq War, we were on a path to be spending around 2.0 percent of GDP on the military. Instead, we're looking at 3.3 percent now. A decade ago, CEPR contracted with Global Insight, one of the main econometric consulting firms, to project the impact of a sustained increase of 1.0 percentage point of GDP increase in military spending. It cost 700,000 jobs after two decades, mostly in construction and manufacturing.

In short, people may well want to reject the projections from these models — their track records have been pretty bad — but Trump is not just making this stuff up. And, the same sorts of models are widely used in other contexts (can you say "Trans-Pacific Partnership?").