Electric car subsidies are hated by economists, loved by activists. Why?

Earlier this month, over at the Energy Institute at HAAS, economist James Bushnell had an interesting post on the way economists and climate hawks clash on the subject of electric cars. It also offered a window into broader clashes between those groups.

As Bushnell points out, economic research — see here, here, here, and here — is generally not supportive of current subsidies for electric vehicles (EVs). It finds that the environmental benefits of EVs are, for now at least, marginal, and worth less than the subsidies paid for them. Current subsidies are also regressive, going primarily to upper-income taxpayers.

From the point of view of many economists, this makes electric car subsidies bad policy. If the goal is to reduce greenhouse gas emissions, policy ought to seek out the cheapest reductions first. (And what's the best way to find the cheapest reductions? You guessed it: a price-based instrument like a carbon tax.)

Even if the goal is narrowly to subsidize the environmental benefits of EVs, the subsidies ought to vary from place to place and time to time. All the research finds that the benefits of EVs vary based on the price of electricity, the source of electricity, grid congestion, and other situational factors.

Economists believe current EV subsidies are excessive and inefficient. And they are, as Bushnell says, somewhat bewildered that this research doesn't seem to dim the enthusiasm of climate hawks for EVs. In fact, it only seems to draw their hostility.

He runs through a few responses and tries to summarize them with a distinction:

Cost-based versus target-based climate policy

Economists take a cost-based approach, he says. They tally up the environmental benefits of a policy — using, among other things, the "social cost of carbon," economists' best estimate of the damage wrought by a single ton of carbon emissions — and compare them to the costs. If benefits exceed costs, it's good policy; if not, it's not.

Climate hawks, he says, favor a target-based approach. They decide what temperature threshold to avoid (say, 2° Celsius over preindustrial levels), what level of emission reductions to achieve (say, 80 percent below 1990 levels by 2050), or what percentage of power ought to come from carbon-free energy (say, 50 percent by 2050) and work backward from there. To commit to a target is to say, in effect, that it is worth any cost to get there. That doesn't make any sense to most economists.

So let me try to articulate, perhaps more fully than Bushnell does, why a climate hawk might view the economists' conclusions with skepticism. We'll start with electric cars.

Electric cars versus the bean counters

In some trivial sense, all decision-making is based on weighing costs and benefits. The problem comes in the fact that costs (materials, electricity, disposal) tend to be relatively easy to quantify, while benefits are much more difficult. The benefits used in cost-benefit analysis are the ones that can be quantified, but there's little reason to think that all benefits, or even the bulk, can be quantified in advance with any confidence.

Take electric cars. Here are a few reasons — merely suggestive, not meant to be thoroughly argued — climate hawks might think electric cars are more important to support than their quantifiable environmental benefits indicate.

1. The lines are diverging. It may be that the environmental benefits of EVs are, at the current movement, marginal. But we can see which way the wind is blowing, as it were. Fossil fuels are going to become more problematic. The science of climate change has only gotten more grim in recent decades, a trend that's likely to continue, raising our estimate of oil's true cost. Oil is finite (recent hysteria notwithstanding), and the cheapest stuff has already been found, while demand in developing countries is rising. And of course civic and political pressure will challenge oil in coming years, as well. Meanwhile, we know the grid is rapidly greening; we know batteries are rapidly getting cheaper and better; we know EVs are at the front end of a long innovation curve. A snapshot of benefits today is less significant than the long-term benefits of accelerating those trends.
2. Electric cars will enable and accelerate the greening of the grid. Right now, the potential of "variable renewable energy" — wind and solar — is limited by the fact that it's intermittent and uncontrollable. There are several measures that can be taken to loosen those limitations, but two of the most important are a) shifting more energy demand to electricity, and b) getting more energy storage on the grid. A large-scale shift to electric cars does both, which could help unlock the full potential of renewable energy much faster.
3. Everything will have to electrify eventually. It's becoming more and more clear that neither biofuels nor diesel is going to substantially reduce transportation emissions anytime soon. The fastest and possibly only route to deep decarbonization is greening the grid and shifting as much energy use as possible to electricity. Electric transportation's benefits relative to liquid-fueled transportation will grow rapidly and last as long as people use cars. From the long-term perspective, then, we capture as many of those benefits as possible by electrifying as soon as possible. That justifies accelerating EVs' movement along the innovation and commercialization curves.
4. Electric cars are "sticky" and politically popular. Electric cars capture people's attention. They are a potent symbol of the clean-energy future, one of the few that individuals can show off to their friends and neighbors. They make low carbon into a form of status signaling. It's easy to attract political support for them, whereas getting support for broader policies like carbon taxes is much more difficult. Sometimes you take what you can get.

For any of these reasons, someone might think that, taking everything we know into account, it makes sense to defend the electric car subsidies that now exist, even while pushing for other policies.

But the differences run deeper than that.

Burying value questions in equations

How is it that economists calculate the environmental benefits of electric cars? Among other things, it involves assigning numeric values to a variety of things many people might fairly regard as contestable moral questions.

For example, how much is a human life worth? You can't calculate the benefits of saving one without a number. How much is it worth to avoid a sickness? How much are intact ecosystems worth? How much are other species worth?

Or: How much is a life this year worth compared with a life ten years from now, or 50 years from now? How much weight we give future costs and benefits relative to the present is measured by our "discount rate." Discount rates are particularly important in climate change discussions, where the connections between cause and effect are measured in decades, sometimes centuries. The choice of discount rate can make the difference between a model that counsels urgent action and one that counsels delay.

Or: what is proper value to assign low-probability, high-impact risks, so-called "fat-tail risks"? (Think, for instance, of the possibility of 6 degrees Celsius warming, with large parts of the Greenland and Antarctic ice sheets melting.) These risks tend to break traditional cost-benefit analysis. If risk = likelihood X damage, and the damage is effectively unbounded, the risk is infinite even at low probabilities. It breaks the model. So most models ignore fat-tail risks.

What is the "correct" value of a life or a species? What's the correct discount rate, or the correct way to account for fat-tail risks? These are all ultimately questions of values, about how to assess risk, how much to sacrifice today to ensure a better tomorrow, how to equip our descendants for success. They cannot be answered by science. And they are not for economists to decide, at least not outside the academy.

Add on to that thicket of value questions another layer of complexity in the form of political economy. What activists, advocates, and politicians can accomplish through politics is only tangentially related to what economists do and don't approve. In US politics, in particular, it is basically a miracle when anything happens, much less anything positive. For political economy purposes, it is important to think not only about policies that reduce the most carbon emissions, but also about policies that create the most constituents for future policies.

Balancing all these considerations requires the old-fashioned virtue of wisdom, the ability to assess and synthesize a wide range of factors, only some of which can be quantified. It helps to have an understanding of the relevant economic research, but it also helps to have a good grasp of politics, culture, and history. Spreadsheets are important, but so are intuition and empathy.

Creating positive social change is difficult — the "strong and slow boring of hard boards," as Weber put it. It's a draining and frustrating business, requiring a level of persistence, a willingness to eat shit and keep going, that most members of the thinking-and-writing tribes (like yours truly) do not grasp.

Those who devote their lives to it can be forgiven for reacting with irritation when economists wander in with their models and expect to end the conversation. There are more things in heaven and earth, Horatio, than are dreamt of in your economics.