Posted by Daniel Hall on May 1, 2008
Yes, this McCain-Clinton idea to suspend the gas tax is stupid. I think the best quote I’ve seen so far comes from Len Burman via Greg Mankiw:
Yesterday I was on the NewsHour to talk about the gas tax holiday. I asked if there was another guest and the producer said, “We tried, but we couldn’t find anyone to argue the other side (that the gas tax holiday made sense).”
The more interesting part of the discussion is not the piling on — though certainly people should be pointing out how mind-bogglingly stupid this proposal is — but the discussion about tax incidence. Most commenters are arguing that producers will benefit more than consumers from the tax holiday, because summer oil supply is very tight. As Greg Mankiw describes it:
What you learn in Economics 101 is that if producers can’t produce much more, when you cut the tax on that good the tax is kept . . . by the suppliers and is not passed on to consumers.
Or here’s Paul Krugman:
Why doesn’t cutting the gas tax this summer make sense? It’s Econ 101 tax incidence theory: if the supply of a good is more or less unresponsive to the price, the price to consumers will always rise until the quantity demanded falls to match the quantity supplied. Cut taxes, and all that happens is that the pretax price rises by the same amount.
But Tim Haab is not so sure that supply is less responsive than demand. He nails the theoretical explanation. In fact, if you really want a great straightforward description of tax incidence, see this pair of posts he wrote just last month. Pay particular attention to the chart in the second post, which graphically depicts what he says here:
Only two cases are clear cut. If buyers are price insensitive and and sellers are price sensitive (top right panel), the buyer will bear the burden of the tax. If sellers are price insensitive and buyers are price sensitive (bottom left panel), sellers will bear the burden. In cases where both are sensitive or both insensitive, the results are unclear and depends on the relative sensitivities.
So when will supply of gas be perfectly inelastic? The most obvious answer to me would be when refineries are running at full capacity over the relevant range of prices. That is, regardless of the market price, refineries can’t keep up. Is that the case? The chart to the right gives U.S. refinery capacity and gasoline consumption in gallons per day from 2002-2007. On average, U.S. refineries produce at about 50% capacity. (emphasis added)
No. Just… no.
The problem is Tim compares refining capacity data to gasoline consumption. This is an apples-to-oranges comparison. Capacity is measured by the input — crude oil — and this is being compared to a specific output, gasoline. But refineries produce many things besides gasoline — jet fuel, home heating oil, etc. — and even if every U.S. refinery were optimized to make gasoline (which isn’t the case) there would still be losses from the refining process itself.
As it turns out the EIA* actually publishes weekly data on refinery capacity margins (or “percent operable utilization” as the EIA calls it). I downloaded the series to create the chart below. Note that with the exception of Katrina (September 2005) that refineries have run at 80-98% capacity over the last few years (most typically from 85-95%). Note that there is also a seasonal pattern: utilization usually declines in winter and peaks in summer. This would imply less elastic supply in summer. Note also, however, that refineries are entering this particularly summer with the lowest capacity utilization rates we seen in spring in awhile, around 85% instead of 90-95%.
What would I conclude from all this?
1. If you wanted to maximize the portion of this tax ‘refund’ that went to oil companies, you should make the gas tax holiday during summer, when supply is most inelastic.
2. Compared to previous years, producers would probably benefit less from the gas tax holiday this year, since capacity margins are down.
3. Despite the factual errors and the incorrect implication that summer gas supply is fairly elastic, Tim is probably right when he says, “maybe the tax holiday might have some effect on both consumers and producers.” Why is that? Remember, demand for gas is also very inelastic. In other words, which world are we in? I say the bottom right case in this chart.
This guy sums it up pretty well when he says,
the gains of the tax cut will be split evenly between producers and consumers… I’m not saying that the gas tax cut is a good idea. In fact, I think it’s horrible pandering that wont help anyone in the long-run.
*Public service announcement: when using energy data about the U.S. to back up any argument, please avail yourself of the EIA. If you can’t find what you need, look again. Still not sure? Check a third time. The EIA is amazing. Use it.