Wind and solar power are subsidized by just about every major country in the world, either directly or indirectly through tax breaks, mandates, and regulations.

The main rationale for these subsidies is that wind and solar produce, to use the economic term of art, “positive externalities” — benefits to society that are not captured in their market price. Specifically, wind and solar power reduce pollution, which reduces sickness, missed work days, and early deaths. Every wind farm or solar field displaces some other form of power generation (usually coal or natural gas) that would have polluted more.

Subsidies for renewables are meant to remedy this market failure, to make the market value of renewables more accurately reflect their total social value.

This raises an obvious question: Are renewable energy subsidies doing the job? That is to say, are they accurately reflecting the size and nature of the positive externalities?

That turns out to be a devilishly difficult question to answer. Quantifying renewable energy’s health and environmental benefits is super, super complicated. Happily, researchers at the Lawrence Berkeley Lab have just produced the most comprehensive attempt to date. It contains all kinds of food for thought, both in its numbers and its uncertainties.

(Quick side note: Just about every country in the world also subsidizes fossil fuels. Globally, fossil fuels receive far more subsidies than renewables, despite the lack of any policy rationale whatsoever for such subsidies. But we’ll put that aside for now.)

Here’s how much wind and solar saved in health and environmental costs

The researchers studied the health and environmental benefits of wind and solar in the US between 2007 (when the market was virtually nothing) and 2015 (after years of explosive market growth).

 

 

 

 

 

 

 

Specifically, they examined how much wind and solar reduced emissions of four main pollutants — sulfur dioxide (SO2), nitrogen oxides (NOx), fine particulate matter (PM2.5), and carbon dioxide (CO2) — over that span of years. The goal was to understand not only the size of the health and environmental benefits, but their geographical distribution and how they have changed over time.

To cut to the chase, let’s review the top-line conclusions:

  • From 2007 to 2015, wind and solar in the US reduced SO2, NOx, and PM2.5 by 1.0, 0.6, and 0.05 million tons respectively;
  • reduction of those local air pollutants helped avoid 7,000 premature deaths (the central estimate in a range from 3,000 to 12,700);
  • those avoided deaths, along with other public health impacts, are worth a cumulative $56 billion (the central estimate in a range from $30 to $113 billion);
  • wind and solar also reduced CO2 emissions, to the tune of $32 billion in avoided climate costs (the central estimate in a range from $5 to $107 billion).

So, if you add up those central estimates, wind and solar saved Americans around $88 billion in health and environmental costs over eight years. Not bad.

That number is worth reflecting on, but first let’s talk a second about how they came up with it.

Uncertainties abound in measuring positive externalities

Tallying up these benefits is difficult for all sorts of reasons.

First, you have to figure out which sources are displaced, when and where, which meant researchers had to build a power system model that covered the country and produced hourly estimates.

Second, you have to figure out just how much of the primary pollutants — SO2, NOx, PM2.5, and CO2 — were avoided by displacing that power generation. To do that, researchers used EPA’s AVoided Emissions and geneRation Tool (AVERT) model. (Don’t ask.)

Third, you have to figure out the avoided impacts, and their value, of the local air pollutants (SO2, NOx, and PM2.5) that were prevented. To do that, researchers used a “suite of air quality, exposure and health impact models” from EPA and elsewhere. (Not all pollutants or impacts were included — impacts from other parts of the power plant lifecycle, like mining, were excluded, for instance. See the paper itself for many more caveats.)

Fourth, you have to figure out the avoided impacts, and their value, of the carbon dioxide emissions that were prevented. To do that, you need to know the “social cost of carbon” (the total quantified benefits of an avoided ton of CO2). Researchers used a wide range of estimates for the SCC.

In all those steps, there are uncertainties and ranges, some having to do with the limitations of models, some having to do with the limitations of our understanding of the impacts of pollution, some having to do with difficult-to-quantify intangibles like the value of a human life.

These uncertainties explain the wide range of estimates involved: premature mortalities range from 3,000 to 12,700; local pollution impacts from $30 to $113 billion; CO2 climate impacts from $5 to $107 billion. (It’s worth saying that there are good reasons to think most SCC estimates are lowballing — certainly $5 billion is ludicrous.)

These ranges reflect the simple fact that different models weigh things differently, from the physiological impacts of pollution to the value of missed work. This is part of what muddies the politics of environmental regulation: Costs are specific and concentrated; benefits are uncertain and diffuse.

Wind and solar benefits vary over time and from place to place

If you dig into the paper, you find that the most interesting data has to do with the variations in benefits across regions and over time.

It’s complex, but in a nutshell, the health and environmental benefits of wind and solar vary depending on what other sources are being displaced, and how much, and when.

For example, fuel shifting (coal to gas) and various pollution regulations have meant that the average pollution of conventional power plants fell over the years of the study. If conventional plants are emitting less, then displacing them avoids less. So on average, early wind and solar displaced more local pollutants per-MWh than later.

It’s slightly different with CO2. The average CO2 emissions of the power sector fell as well, thanks to fuel shifting, but not as fast — not fast enough offset the explosive growth of wind and solar. So the amount of CO2 displacement per-MWh has remained roughly steady.

Here’s what that looks like graphically — these are the benefits over time. CO2 is on the upper left:

 

 

 

 

 

Wind and solar’s positive effects on local pollution have, on a per-MWh basis, fallen over time as other power plants have cleaned up somewhat. But their positive effects on CO2 pollution have remained steady. If it isn’t already, CO2 displacement will soon become wind and solar’s most valuable positive externality.

Wind and solar effects also varied widely by region, because some regions have cleaner power sectors than others. In California, wind and solar are mostly displacing natural gas. In the upper Midwest and mid-Atlantic regions, which rely more heavily on coal, wind and solar have greater impact.

Here’s what that looks like graphically. The first two rows show the marginal benefits of wind and solar by region; the bottom two rows show the total benefits by region:

 

 

 

 

 

 

A few things jump out here.

First, wind power has had enormous air-quality benefits in the upper Midwest and the mid-Atlantic. Yuge!

Second, poor California is building the shit out of renewable energy, but it’s mostly displacing natural gas, which has relatively low levels of local pollution. That means the state is getting relatively little air-quality benefit from wind and solar — its shift to renewables is mostly benefiting the climate (look at that solar spike at the bottom).

Third, the Southeast, one of the regions that benefits most from every marginal addition of wind and solar (thanks to the prevalence of coal power), has built the least. That’s dumb.

Subsidies are roughly in line with air and climate benefits, but only roughly

In something of a curious coincidence, the central-estimate health and environmental benefits of wind and solar in 2015 — 7.3¢/kWh and 4.0¢/kWh respectively — are “comparable to estimates of total federal and state financial support” for wind and solar.

So for all the various subsidies and tax breaks for wind and solar, we’re getting roughly what we paid for. (If you believe the central estimates. Of course, “central” does not mean “most likely,” so we still don’t really know exactly what we’re getting, but we’ll put that aside too.)

However, while the absolute level of subsidies might match the absolute level of benefits, they do not line up on a granular level. The health and environmental benefits of wind and solar vary widely by time and region, but most policy incentives for wind and solar do not. Federal tax incentives treat all wind and solar projects the same. And when subsidies do vary, as in state-level policy, it’s rarely connected to their varying benefits.

The conclusion the researchers draw from this subsidy mismatch is that “addressing air quality and climate change through policies directly supporting wind and solar is not necessarily the most cost-effective approach.”

That’s true, as far as it goes, though I think there are still plenty of good reasons to support wind and solar. What’s fun, though, is to think about what it might look like if state and federal supports for wind and solar did vary by time and region.

How might that work? Take the same pot of money and instead of flat, capacity-based subsidies, offer time-varying, per-KWh subsidies based the pollution-intensity of the power being displaced. That would be computationally difficult, but not theoretically impossible. (If you want to really nerd out, the Brattle Group proposed something roughly similar for energy markets in a white paper.)

Time- and location-sensitive subsidies would attract wind and solar investment to the regions where it will do the most air-quality and atmospheric good, increasing their impact. And as a bonus, those regions often overlap with regions badly in need of blue collar jobs and regions where the fight against climate change could use a political boost, so it could increase their sociopolitical impact as well.

California wouldn’t like that much. But the upper-Midwest sure would!

 

 

 

 

 

A final word on costs and benefits

In this case, as in all such cases, it is somewhat misleading to simply compare total subsidies with total health and environmental benefits. The total amounts are not all that matters. It also matters how costs and benefits are distributed — i.e., equity matters as well.

To put it bluntly: A dollar in federal taxes is not equivalent to a dollar of avoided health and environmental costs. The latter dollar is worth more than the former dollar.

Why is that? Simple: Federal taxes come disproportionately from the wealthy, via our progressive federal income tax, but health and environmental benefits disproportionately help the poor. And as any good economist will tell you, the same dollar is worth more to a poor person than it is to a rich person.

This is something that often gets lost in discussions of environmental regulations. It’s not just that their total benefits almost always exceed their direct costs. It’s that those benefits are uniquely egalitarian and progressive.

In the case of climate change, any reduction in CO2 emissions benefits everyone on Earth (egalitarian), while disproportionately helping the poor, who suffer earliest and most from climate impacts (progressive).

In the case of local air-quality benefits, cleaner air benefits everyone in the region who breathes (egalitarian), while disproportionately helping the poor, who are more likely to live in close proximity to fossil fuel power plants (progressive).

In terms of equity, converting a dollar of wealthy people’s money into a dollar of health for low-income communities seems like a good deal to me. And if you can get multiple dollars of low-income health benefit for every dollar of high-income taxes, well, that’s a no brainer.

Everybody breathes. Any dollar of federal income taxes used to produce a dollar of air and climate benefits is a net gain for justice.

Get in Touch