The Congressional Budget Office issued a briefing paper yesterday concluding that climate pricing will slightly reduce employment overall in the United States, because green jobs gains won’t quite offset other job losses. Unfortunately, this paper will likely be misquoted and misunderstood repeatedly. It’s not what it seems.

Properly understood, it should give us confidence that the clean-energy path is our best option by far for a high-employment future. Too bad CBO didn’t just say that. You could read the summary and never guess. I delved into the study itself and here’s what I learned.

The paper is essentially a summary of three general equilibrium modeling results on the industry-by-industry employment impacts of putting a price on carbon. All the studies in question are reputable research. One is from Resources for the Future. Another is from the Brookings Institution. The CBO paper is a fair synopsis of them.

But the models are, by their authors’ admission, all loose and imprecise. Each of them makes a long list of counterfactual assumptions about the economy and politics, in order to limit itself to a set of variables small enough in number to allow modeling.

  • In short, they pile simplifying but false assumptions one atop the other:

    • None of these models is an assessment of any actual climate bill before the US Congress. Such bills are elaborate pieces of policy with far too many moving parts to put into an econometric model. Instead, they attempt to model a crude policy that puts a rising price on carbon emissions.
    • The models assume that there is no first-mover advantage in developing clean-tech industries. They assume that carbon-constrained economies won’t develop expertise and exportable products that laggard nations will have a hard time catching up to. (Tell that to Denmark, for example, which embraced ambitious clean-energy policies years ago and now make a fortune selling windmills to the world.)
    • The models ignore the considerable benefits to the US economy of trimming the trade deficit, by weaning the nation off imported oil. Trimming trade deficits will also help to rebalance world financial and currency markets, reducing the pressures that contributed to the Great Recession.
    • The models discount the possibility—probability, many would say—of game-changing technological breakthroughs. Once carbon has a price and the ingenuity of tens of millions of profit-motivated people and companies turns to wringing carbon out of the economy, new approaches may emerge. These new techniques and technologies have a chance of launching whole new industries.
    • The models assume that US action on climate change will have no effect on other nations’ actions—that if the United States puts a price on carbon, other countries will be no more likely to do so. (Tell that to Canada. Ottawa has been saying for months that it will consider comprehensive climate policy after the United States does.)
    • The models also assume that US climate policy will include no border adjustments or other measures to levelize competition with any trading partners who do not match the US carbon price. Consequently, they assume that climate pricing will put the United States at a competitive disadvantage. (Tell that to the US Senators from every manufacturing state. There won’tbe a US climate policy that doesn’t somehow protect US manufacturers from a near term hit to their price competitiveness.)
    • The models also essentially assume that the economy is already as energy efficient as it could be, ignoring the massive, cost-effective savings potential documented by dozens of studies.
    • The models assume that reducing greenhouse emissions brings no ancillary economic upsides, such as the benefits of huge reductions in conventional air and water pollution that will accompany a post-carbon economy.
    • The models ignore the shock-absorbing benefits of a clean-energy economy. Price spikes in world oil markets are a major contributor to recessions, which are the biggest job killers around. Getting off the fossil-fuel rollercoaster protects jobs.
    • Finally, and perhaps most tellingly, they assume away all the impacts of climate change itself. That is, because they cannot model the impacts of climate change on employment (which could be either utterly devastating or merely debilitating), they simply ignore them. This assumption essentially invalidates the conclusions of the CBO briefing, because the “base case” (no price on carbon + status quo economic trends) cannot exist. If we do not reduce emissions, we will reap a future of both ecological and economic decline. Our choice is not between business as usual and a carbon-priced economy. Our choice is between (a) unchecked climate change and (b) less-rapid climate change in a carbon-priced economy.

    Of course, for the modelers themselves, the main point of such models is not to predict specific net outcomes. Modelers know better. The main point of such models is to forecast the broad outlines of change. For example, the models are on pretty solid ground, and agree with one another, when they say that employment in coal mining, oil drilling, and oil refining will decline sharply. Employment in energy-intensive industries such as transportation and chemicals manufacturing will also likely diminish. Employment in clean energy, clean tech more generally, and the gargantuan service sector (more generally still) will increase.

    So, here’s the good news. Despite the almost preposterously anti-clean-energy assumptions baked into the models CBO studied, they still basically show that the US economy will hardly lose any jobs at all on net. Employment will be “slightly lower,” CBO says, and only for the early decades of the clean-energy transition. In time, even as emissions fall toward 80 percent reductions by 2050, the nation will have returned to full employment.

    Take away some of the assumptions above, and “slightly lower” employment surely starts to look like “increased employment.”

    Here’s Robert Pollin of the University of Massachusetts and his colleagues, summarizing their survey of an earlier batch of economic modeling studies on cap and trade (which also made many of the same assumptions listed above):

    According to all the forecasts—including the worst-case scenario developed by the most pessimistic forecasters, the American Council on Capital Formation/National Association of Manufacturers—the impact of a cap-and-trade system on US GDP growth will be negligible. According to most forecasts, it will be almost indiscernible. . . .

    Even assuming this most severe negative effect of a carbon cap on economic growth [the National Association of Manufacturers’ forecast], it would still only require, over the course of 23 years, an additional 14 months for the US economy to reach the same level of GDP under a carbon cap as against the baseline scenario.

    I take the CBO paper’s real point as something similar: the worst possible outcome from pricing carbon would be to reduce overall employment slightly in the United States in the early years of the transition to clean energy. The more likely outcome by far is an increase in employment.

    Notes: Those interested in modeling should read David Roberts’ excellent piece, this study by Robert Pollin and his colleagues at the University of Massachusetts (especially around
    page 41), and
    Sightline’s Green Collar Jobs primer.