This week, the International Monetary Fund put out a big report claiming the world spends $5.3 trillion a year subsidizing energy, mainly fossil fuels.
That's an eye-popping number. And it led to plenty of outraged headlines: Look at all that money we're forking over to Big Oil! That's $10 million every minute! What's more, the IMF noted that if we yanked away all these "subsidies," global carbon dioxide emissions would drop 20 percent — a huge step toward dealing with climate change.
The only catch? The IMF is using a very unusual definition of the word "subsidy" here. This report doesn't say what most people might think it says.
When many people hear the phrase "fossil fuel subsidies," they probably think of government policies to underwrite oil, gas, and coal production or consumption. These sorts of direct subsidies are indeed very real, and total around $500 billion each year worldwide — with big effects on fossil-fuel use. But that's actually not what the IMF report is looking at.
Instead, the IMF report is using "subsidy" to refer to the fact that users of oil, gas, and coal don't pay the full price for all the environmental damage these fossil fuels cause. If we take that into account, the IMF estimates, then fossil fuels are arguably underpriced by roughly $5.3 trillion. In other words, dirty energy only seems cheap if you don't properly account for all the destruction it does.
This pricing question is a hugely important one — but it's pretty unconventional to call this a "subsidy." Economists typically refer to these unpriced social costs as "externalities." The IMF has just decided to rebrand them. And the big question is whether that rebranding is useful or just plain misleading.
The IMF's two types of fossil fuel "subsidies"
The IMF uses the word "subsidy" to refer to two very different things, so let's separate them out:
1) $492 billion in "pre-tax" subsidies for consumption: This is the stuff that's conventionally known as a "subsidy" for fossil fuels. In 2015, governments are expected to spend $492 billion to lower the market price of petroleum, natural gas, coal, and electricity for their citizens. This includes everything from price controls at the pump to subsidies for the companies drilling oil or mining coal. The biggest offenders here include Saudi Arabia, Iran, and Russia.
Economists broadly agree that these sorts of subsidies are needlessly wasteful. They eat up budgets, lead to higher levels of pollution, and aren't very effective ways of helping the poor. That's why countries like Egypt, Ghana, and Indonesia have begun paring back these policies in recent years. (This is always a painful process, since people don't like it when energy costs go up, but the recent drop in oil prices has eased the transition.)
Trouble is, these subsidies are fairly small-bore in the scheme of things. Scrap them all, the IMF figures, and global carbon dioxide emissions would fall by maybe 2 percent. Not bad! But not world-changing, either.
2) $5.3 trillion in "post-tax" subsidies: This is the more unconventional part of the IMF's analysis. The IMF report essentially argues that governments should be taxing energy — fossil fuels, mainly — above their market price in order to account for air pollution, global warming, and other social damages they cause.
Economists usually refer to these social damages as "externalities." When I buy a gallon of gasoline and burn it in my car, I'm inflicting costs on my neighbors. The smog I create will be paid for by others in the form of lung damage. The global warming I'm producing will be shouldered by others. As such, many economists would argue, I should pay an extra tax at the pump so that I'm paying my fair share of these social costs.
The IMF tried to tally up all these social costs and concluded that fossil fuels were underpriced by a whopping $5.3 trillion worldwide. This mispricing is considered a "post-tax" subsidy, and it's much bigger than those more conventional "pre-tax" subsidies discussed above:
If governments increased taxes on fossil fuels to correct this mispricing, they could raise about $2.9 trillion in revenues. (Because these taxes would lead people to use less oil, gas, and coal, governments wouldn't reap the full $5.3 trillion.)
If that ever happened, the IMF noted, global carbon dioxide emissions would fall about 20 percent — a huge drop. Consumers would see higher energy bills, yes, but they'd also have cleaner air, less global warming, etc., and overall welfare would be improved by about $1.8 trillion.
So what's not to like?
The problems with the IMF analysis
If you dig a bit deeper into the IMF analysis, however, problems emerge. As Tim Worstall points out at Forbes, the IMF ended up including all sorts of oddly disparate things in its count of post-tax "subsidies."
First, the IMF included the damages from global warming caused by fossil fuels. That's fair enough, and constituted about one-quarter of the $5.3 trillion. Then the IMF considered conventional environmental damage from oil, gas, and coal — e.g., particulate matter from coal plants that can cause respiratory disease. Also fair, and that made up about half the $5.3 trillion.
But then things get weirder. For instance, about 39 percent of the "social cost" of gasoline in the IMF analysis comes from accidents, traffic fatalities, and congestion. It is true that these things are externalities. But they're caused by automobile use, not by gasoline use per se. If we switched over to solar-powered electric cars tomorrow, we'd still have traffic accidents and congestion. It's strange to argue that this is some sort of "subsidy" for gasoline specifically.
On top of that, Worstall points out, the IMF considers it a "subsidy" that energy isn't taxed under Europe's value-added tax system at the same rate as other inputs to goods. But now we're really stretching the bounds of the word. Food is also "subsidized" by this measure. So are renewables like wind and solar, for that matter. What are we even talking about at this point?
Is this really a good use of the word "subsidy"?
Setting aside the specific quibbles with the IMF's analysis, there's also a broader question: Does it makes sense to refer to our failure to enact carbon taxes and the like as a "subsidy"?
There's an argument that, no, it's not appropriate (see Tim Worstall or William Connolley). After all, we already use the term "fossil fuel subsidies" to refer to government policies that explicitly reduce the market price of oil, gas, and coal. And there's a real push to scrap those subsidies. So it might just be needlessly confusing and distracting to fold in a grab-bag of other stuff under the term.
After all, when headlines exclaim that "The world is spending $5.3 trillion on fossil fuel subsidies," many people might think that this money is somehow going directly to oil, gas, and coal companies. I suspect they don't read that sentence as "The world should raise energy taxes by $5.3 trillion to account for the damages caused by fossil fuels." But the latter is basically what is meant.
Yet, on the other hand, you could argue that the IMF's unconventional use of "subsidy" here only sounds odd because we have a distorted view of how markets actually work. Generally speaking, we tend to think of markets as if they exist in some naturally free state — and any government spending that favors certain sectors is an artificial "subsidy" of sorts.
So, for instance, we think of the United States as "subsidizing" solar power through tax credits and renewable portfolio standards. But we don't typically think of society as "subsidizing" coal power by being willing to breathe in particulate matter and pay the resulting hospital bills.
The IMF report is effectively arguing that we should think of the latter as a subsidy. There's no such thing as a pure market that's not already shaped by the various policies and choices we make. We've set up a variety of political and social arrangements over the years that have made it artificially cheap to burn fossil fuels for energy. Calling these arrangements a "subsidy" may seem awkward, but it's not totally arbitrary.