Why add a cost to GHG instead of subsidizing renewables?

Currently, the U.S. pays wind producers 2.3 cents/kWh. The federal government pays 30% of the cost of solar panel (photovoltaic, or PV) installations, and allows rapid depreciation—one solar manufacturer says this reduces the cost of installed systems by 70 – 75%. Additionally, a number of states subsidize wind, either through mandating renewables, as California does, or through direct subsidies. For example, windy Iowa adds another 1.5 cent/kWh for wind, as well as exempting wind from taxes. States subsidize solar purchase, as Iowa does, and mandate solar, as Minnesota does (1.5% of electricity by 2020). California is in a class by itself among states without huge hydroelectric capacity, mandating 33% renewables by 2020. In addition to purchase subsidies, California is one of a number of states using net metering (paying solar producers the retail value rather than the wholesale value of their electrons).

Do these subsidies really help, or are there better ways to reduce greenhouse gas (GHG) emissions? At the bottom, I partially address solar subsidies. This post focuses on why economists generally prefer correct pricing to subsidies.

Bottom line

Failing to pay the costs of pollution is not free; we still pay for pollution, but indirectly. Pollution costs us trillions of dollars each year, averaging about $500/person, although those who use more fossil fuel energy are more responsible. Increasing the price by that cost encourages us:

• to switch to technologies and behaviors that pollute less, and
• to waste less.

Subsidies and mandates for renewables often, not always, lower pollution, but at a higher cost. Subsidized energy displaces the most expensive energy, which is not always the most polluting energy. Additionally, subsidies don’t encourage building where it will do the most good—as discussed below, the location chosen to build a windmill may well be different if motivated by correct pricing of fossil fuels rather than subsidies/mandates.

Raising the price for energy could hurt the poor, but proposed plans cover the extra costs of the poor and middle class, at least, by returning money independent of consumption. On the other hand, pollution subsidies (failure to include the cost of pollution in the price) go disproportionately to the rich. The poor are expected to suffer disproportionately from climate change as they suffer disproportionately from pollution.

This analysis discusses pollution costs from greenhouse gases (GHG), but ignores substantial costs from other pollutants. This is partly because it’s easier to add a cost on carbon dioxide, as we know how much CO2 is released per unit of energy and so have an idea how much damage it will do. Less is known about particular plants: how much of other pollutants they produce, and what damage those pollutants do.


If we don’t like the pollution from coal and other fossil fuels, why not just subsidize what we do like? Well economists have some reasons…

We really should pay the costs of pollution

Fossil fuels produce enormous amounts of pollution, and as long as we burn them, as long as we use the atmosphere as a sewer, the costs of pollution should be paid by the polluter. Those who fly or drive, or use coal or natural gas or other fossil fuels for electricity and heating, are incurring a cost, but not paying the cost. We see it in countries such as Saudi Arabia, where people pay only 1/5 of the cost of energy, that people overconsume when they aren’t charged a fair price, and the public money is less available where it is needed. So too do others pay the costs of pollution, including climate change (lives lost, health costs, lower yields for agriculture, increased storm damage). And when costs are artificially low, people forget we can insulate, turn off the lights, use fuel efficient cars, and take the bus—we are wasteful.

How much damage do fossil fuels do, subsidized by the rest of us?

As discussed in an earlier post, the current social cost of fossil fuels is $41/ton carbon dioxide; many respected economists argue it should be higher. All agree that it will rise over time. International Monetary Fund says the air pollution cost, not counting greenhouse gases, averages $57.50/ton of greenhouse gas (even though it’s not the GHG damage they are looking at).

I have seen no plans recommending incorporating damages from non-climate air pollution into the costs of using fossil fuels. One reason may be the complexities. Carbon dioxide does equal harm no matter where it is emitted, and the amount of pollution can be readily calculated from the amount of fuel. Other air pollution varies by plant, and the effect of that pollution varies with location and weather.


It does not always make sense for regulators to put a cost on pollutants. Adding a cost is usually cheaper than direct regulation, as long as emissions are easy to monitor. On the other hand, regulators would would have to deal with too much information to make good and timely decisions about what to charge for leaking natural gas (methane). Where monitoring is difficult, regulation may achieve the same goal more cheaply. In the case of natural gas leaks, infrared cameras can detect the existence of a leak cheaply, and the cost of fixing the leak is often paid for by the value of the gas saved.

Subsidies for renewables are not good at reducing greenhouse gas emissions, part 1

Suppose you have three sources of electricity:

Source Variable operating cost ($/MWh) Cost of emissions ($/MWh) Variable social cost ($/MWh)
Wind $0 $0 $0
Coal $20 $23 $43
Natural gas $30 $11 $41

Notes: $10/megawatt hour = 1 cent/kilowatt hour. The variable social cost includes both the operating cost (private cost) and the cost of greenhouse gas emissions. In this example, the price paid to pollute is about $20/ton GHG emissions (half what economists argue for). The phrase variable social cost sometimes includes unpaid costs as well.

Let’s assume in this simplified example that those providing power bid (offer) the variable cost, the cost of fuel, workers, etc. for the electricity, as the price at which they will sell electricity. They don’t included fixed costs, such as the cost of the plant. Selling price is the highest bid price of all electricity actually bought. If wind only is needed, utilities pay $0/MWh. If coal is also needed, the wholesale price rises to $20/MWh, 2 cents/kWh, and both wind and coal providers are paid that rate. If even more electricity is needed, natural gas is brought online and utilities now pay 3 cents/kWh to every supplier. Clearly, there is a fixed capital cost not included in the variable cost; power suppliers cover this portion of their cost when the selling cost of electricity is greater due to other sources being brought online, for example, expensive peak power for high demand periods. In addition, wind is directly subsidized.

If instead of subsidizing wind, fossil fuels paid $20/ton-CO2 of their pollution cost, then natural gas would be brought online first after wind, at a cost of 4.1 cents/kWh, paid to both wind and natural gas providers. If more electricity is needed, then coal is brought online, and all are paid 4.3 cents/kWh.

To summarize, subsidized wind displaces the more expensive natural gas. If instead, a sufficient greenhouse gas cost is added, wind now displaces coal. Economists say that adding a cost is more efficient at reducing GHG emissions.

Subsidies for renewables are not good at reducing greenhouse gas emissions, part 2

Suppose there are mandates to build wind, perhaps through renewables portfolio standards. Or there are direct subsidies, perhaps for each MW of wind built, or MWh produced. You the entrepreneur are looking at two locations. In each location, the capacity factor is 30%, that is, over a year, windmills produce 30% as much electricity as they would if they ran 24/7 at full tilt. In one location, more power is produced at night, in the other, slightly more electricity is produced during the day. With a direct subsidy, neither location appears to have an advantage.

If the wind decision is made after a cost is added to GHG emissions, it makes sense to add wind where GHG reductions are greater. In California, night electricity is disproportionately nuclear; in Pennsylvania, it is coal. The more expensive electricity during the daytime in both PA and CA, the one that would go offline if the wind blows, is natural gas. All else being equal, it is better to build windmills where the wind blows at night in PA, and during the day in CA.

Sometimes renewables subsidies are largely a waste of money. Because of renewables mandates, California builds wind in the Pacific Northwest. There, especially in the spring, wind competes with hydro. Some 40% of Danish wind backs up Norway’s hydro. Neither reduces GHG emissions.

Washington wind
Wind in Washington state

Note: wind, solar, and nuclear don’t play the same role in electricity supply

Nuclear power provides baseload electricity, the minimum electricity required during the day. It doesn’t ramp up and down as demand changes. Solar helps with the increased demand during daylight. From the examples above, it is clear that wind is most useful where there is fossil fuel baseload power, which then ramps down when the wind blows. Neither wind nor solar can supply baseload power, because they are not available 24/7, and solar cannot even be depended on during the daytime in most locations.

What about the poor? They don’t want to see the cost of energy go up.

Although this argument may be made most often by those with enough money who don’t want to pay more themselves, the argument has merit. If the price of energy increases, it will hurt those operating on tight budgets. Yet legislative proposals to date make sure the poor don’t suffer:

• All proposals for adding a cost to greenhouse gas emissions include refunding some (or all) money to consumers, a dividend independent of use. This refund would be high enough to compensate low-income consumers for price increases. Unlike direct subsidies for heating oil or other forms of energy, the rebate encourages me to reduce my own energy use, although it is sufficient to cover increased costs completely for low-income energy users.

• Subsidies currently go disproportionately to those who drive and fly more, and use considerably more electricity and heat, namely, the wealthy. Those who are poor or who have other health problems are more likely to pay these subsidies with their health.

• The poor will pay disproportionately for climate change. Not addressing climate change as rapidly and as cheaply as we can does the poor no good.

But aren’t subsidies better than nothing?

Subsidizing renewables is better than doing nothing.

But doesn’t it make sense to subsidize renewables to bring down the price?

Economists do support targeted subsidies, along with heftier funding of research and development. Carbon capture and storage is often cited as an important energy source needing deployment subsidies, and research and development. Similarly, new nuclear gets the same production tax credit as wind, although only for the first 6,000 MW of new build (about 5 reactors). This is because costs go down with learning—there is evidence that the 2nd reactor at Vogtle will be cheaper than the 1st. The fourth reactor of the same kind may be 20% cheaper than the first.

But what about solar subsidies?

Economists generally support pricing fossil fuels correctly over subsidizing better alternatives to fossil fuels. They sometimes support subsidizing solutions, e.g., solar and carbon capture and storage, in order to bring down the price, as a long term strategy. As discussed, solar electricity helps with a different niche than nuclear does—solar helps imperfectly with increased demand during the day, while nuclear supplies baseload power.

While economists at International Energy Agency and elsewhere agree that our current method of subsidizing solar is not sustainable, I have not seen alternative methods suggested, nor have I seen an analysis of how much subsidies make sense for how many years. (Please send me to analysis addressing either of these two questions.) And subsidies would need to be large for quite a while—a 2013 look at The Economics of Solar Electricity found the cost of south facing solar panels at a good angle ranged from 19 cents/kWh in Tucson to 26 cents in San Francisco to 29 cents in Boston and Trenton, with prices much higher if the panels don’t face south or if the angle is worse. (Solar actually costs more, as some costs were not included.) U.S. solar displaces natural gas, which wholesales for about 7 cents/kWh. The price of solar has dropped quite a bit with the shift to large manufacturers, and prices will drop further as soft costs (costs of purchase other than hardware) also fall. Still, it it will be a while before solar becomes cheaper than nuclear and fossil fuels.

Renewables subsidies have been more open-ended and larger than other subsidies to deploy energy in the richer countries. Do they make sense? I will post when/if I understand the issues.

The series:
Fossil Fuel Subsidies
Can we address climate change fairly cheaply?
• Why add a cost to GHG instead of subsidizing renewables?
Tax or cap and trade?

Comments are closed.