A back-of-the-envelope sense of the socio-economic impact of reducing fossil fuel usage to fight global warming

I’ve been writing posts recently advocating for a Sky Trust style program to
1) cap carbon emissions from the burning of fossil fuels
2) auction the permits within that cap
3) pay out revenue from the auction to each American on a per-person basis.

One thing I’ve wondered about is, “what will the impact be on people for reducing the availability of fossil fuels by the amount necessary to aggressively fight global warming?”

The first part of the answer is to see how much reduction of fossil fuels will be necessary. The standard goal being targeted in currently proposed legislation is to cut carbon emissions by 80% by the year 2050. That gives us roughly 40 years (since any program will only begin in a couple years at best), and, depending on the precise details of the program will require annual reductions–for the first bunch of years, at least–of between 2% and 4%.

So let’s go with the 4% per-year reduction scenario, and let’s assume that reducing carbon emissions by 4% is effectively the same as reducing use of fossil fuels by 4%. What’s that mean in terms of the impact on the economy and everybody’s lives?

Well, a first (and rough) estimation is to compare what happened in past economic recessions when fossil fuel consumption fell. True, in those cases, the cause and effect weren’t quite lined up the same–much of the reduction in fuel use during a recession is a reaction to the recession, whereas in a policy of fighting global warming it would be a forced restriction on the availability of the fuels, with some level of economic fallout as a result.

On the other hand, some of the cases of reduced fossil fuel use in the past were during the oil embargo of the mid 1970s and following the Iranian Revolution in 1979. While there was a lot of vicious cycle going on in those years (spiked oil prices –> economic recession –> reduced demand for energy and fuel) (or, spiked oil prices –> inflation –> Federal Reserve Bank cranks up interest rates to fight inflation –> severe recession –> reduced demand for energy and fuel), it seems a decent rough initial approximation of what will happen under a policy to fight global warming.

The Department of Energy’s Annual Energy Review, Table 1.3 “Energy Consumption by Primary Energy Source, 1949-2006” shows when fossil fuel use has declined in past years. Years with declines are:

1952 -0.6%
1954 -2.7%
1958 -0.5%
1974 -3.4%
1975 -3.8%
1980 -4.2%
1981 -3.2%
1982 -5.4%
1983 -1.1%
1985 -0.6%
1986 -0.1%
1990 -0.8%
1991 -0.6%
2001 -2.2%
2006 -1.5% (preliminary data)

So that’s not terribly good news for the bulk of people who will have to adapt to reduced availability of fossil fuels under a cap-and-trade program. Things in 1975 and the early 1980s weren’t that pleasant, so I’m led to believe (I was too young to know what was going on very well myself–I remember seeing gas lines, but didn’t really grasp the situation, and had no concept of interest rate hikes or any of that sort of thing). And those were years with drops in fossil fuel use of around 4%, toward the upper end of how much of a drop might be mandated under a cap-and-trade program, at least for the first couple of decades.

But the news isn’t as bad as it might first seem. First of all, a good cap-and-trade policy will include, as noted at the top of this post, an auction of all emissions permits and the distribution of all auction revenues to all Americans on an equal, per-person basis. That will pad the economic hardship quite a bit. Imagine if, during the 1973 oil embargo or following the Iranian Revolution, all the excess profits received by the oil companies had not gone to corporate executives or stock holders or Saddam Hussein or the King of Saudi Arabia or any of the others with their fingers in the oil pot, but instead all that money had been divvied out back to the consumers who were paying high prices at the pump. People still would have had to adjust, but it would have been a heck of a lot easier to do so. That’s part of the beauty of a cap-auction-rebate version of a cap-and-trade program.

We’re still not to the end of the story. (And never will be, technically speaking, but that doesn’t mean this post will go on forever.) Another difference between those historical cases of reduced fossil fuel use and our hypothetical reduction under global warming policy is the much-vaunted reduction in the energy intensity of the economy today versus those past years.

What that means is that, today, less energy is used in the creation of $1 worth of economic activity than was needed to create the same economic activity 10, 20, or 30 years ago. On a per-dollar basis, the economy is much more energy efficient than it was in the past. For example, in the 20 years between 1985 and 2004, the energy intensity of the US economy fell by 10%. As a result, a reduction in the availability of fossil fuels should not have as severe an impact, relatively speaking, as the same size reduction that may have happened in the early 1980s. The economy, because of improvements in energy efficiency, should be able to absorb the reduction in energy availability more gently. (That’s one of the reasons given for the fact that the large increases in the cost of oil these past couple of years have not led to the kind of recession that similar increases caused in the 1970s.)

So what’s the take-home message? This, I guess: a cap-and-trade program to reduce global warming will probably have only a relatively mild negative effect on the economic lives of most Americans (using traditional economic measures of well-being, which definitely leave something to be desired), or maybe even just a neutral effect, or maybe even a mildly positive effect (due to the distribution of money from the permit auction). That’s pretty good news in a world of global warming.


  • Jonathan Teller-Elsberg

    One other thought on why the impact of a cap-and-trade-with-rebate system wouldn’t be very bad, or even a bit good economically speaking: a major factor in recessions is that businesses, feeling or fearing declines in consumer purchases, are less inclined to invest in new equipment and such. That lack of investment is bad for the industries that supply the equipment and such, and so part of the vicious cycle of recessions. That’s why low interest rates–intended to spur more investment to boost a recessionary economy–don’t always work very well. Even if the interest rate is low, why invest in new equipment if you doubt there will be customers for your product?

    Anyway, that shouldn’t be an issue after the establishment of a cap on carbon emissions for two reasons. First, just to stay even, businesses will be interested in investing in efficiency technologies and non-carbon energy sources. Second, because of the rebate aspect of the plan–and only if it IS a part of the plan–businesses will know that most Americans will have more disposable income than previously, and so they can expect more customers, not less. That’ll be a good incentive for investment, and so a boost to the economy in spite of rising energy prices.

  • You say: “What that means is that, today, less energy is used in the creation of $1 worth of economic activity than was needed to create the same economic activity 10, 20, or 30 years ago.”

    I say: Nowhere on your page or the referenced DOE page does it say these dollar figures are normalized for inflation. If inflation is not taken into account, the reduction is pretty much meaningless. If the dollar is worth half what it was in 1970, then the equivalent GDP is twice the 1970 GDP. If energy consumption and the economy remained the same, then it would look like we were using half as much energy, even though nothing had changed.

    It would be much more interesting if they had tracked energy price versus GDP. That way inflation would be corrected in both numerator and denominator. Of course, energy price is noisy, and hard to define, so this would be harder. Since the reporter is the US DOE, it’s unlikely they are interested in a true investigation, but are more likely to be interested in advocating an agenda.

  • Jonathan Teller-Elsberg

    Bill, actually the referenced DOE page explicitly _does_ say that dollar amounts used are adjusted for inflation. Quoting from the first paragraph on the linked page: “The nation’s output of goods and services, as measured by inflation-adjusted Gross Domestic Product (GDP), increased more than six-fold, from $1.63 trillion to $10.75 trillion over the period.” In that span of time, says the DOE, energy use increased by only three-fold, a clear, though rough, indicator of reduced energy intensity of economic activity.

    I’m not sure why you think tracking energy price versus GDP would be more interesting. Sure, it is an interesting question–but entirely of a different sort. If you track energy price versus GDP, what you’d basically be learning is an answer to the question: “How has inflation in energy price compared to overall inflation?” I’m not sure what the answer is, maybe energy prices, on average, have been rising faster than other prices, maybe about the same, maybe slower. Like you say, energy prices are noisy, so you’d have to do a lot of averaging over multiple years–unless, of course, what you were most interested in was the relative price behavior in a specific, narrow time period.

    But that information tells us nothing about the energy needed over time to perform all the country’s economic activity.

    What makes you say that defining energy price is hard to do? Isn’t it, well, just the weighted average price of a barrel of oil, a ton of coal, and a cubic foot of natural gas? What are you thinking about that makes it more complicated?

  • First, I apologize. I didn’t see that it was inflation adjusted. Since it’s the US DOE in the Bush administration, I’m inclined to think it’s more spin than substance, but perhaps the spin is more subtle.

    Second, I was thinking that if you track energy price versus GDP you have dollars in both the numerator and denominator, so in some sense the multiple caused by inflation cancels out. Is this multiple called the inflator? Now, the energy price inflator, if we may call it that, is different than the general economy’s inflator, so you may see variations caused by that difference. I’m not sure how to correct for that. In any case, this is less interesting because the numbers are inflation adjusted. Of course, there are different ways to adjust for inflation as well. The inflation numbers the Dept of Labor Statistics puts out (is that the right Dept?) don’t always seem to match people’s intuitive notions of inflation. I’ve read a lot of newspaper stories that point out the contradictions between the official low-inflation story and the story you get when you go to grocery stores and track prices on things people actually buy.

    Finally, defining energy price as a weighted average is hard to do because you have to define the weights. I’m not sure what criteria one would use, but I am sure that different weights would give you different answers. If one were looking for statistical validation of some ideologically driven claim, the weights would be a way to spin the answer.

  • Jonathan Teller-Elsberg

    Bill, I agree that government statistics (but not only theirs) should be considered carefully for reasons of spin. However, I think the spin comes more in the choice of what to calculate statistics about, rather than in intentionally messing up of the numbers. Beyond that, for all that the Bush administration has tried very hard to politicize all levels of the bureaucracy, there are lots of entrenched number crunchers who, I suspect, take pride in doing a good job and getting out quality numbers. I’m an optimist in that regard.

    You are right that inflation statistics are kept by the Department of Labor, in their Bureau of Labor Statistics division. You can find them at http://www.bls.gov/bls/inflation.htm.
    Several different methods for calculating inflation have been developed. Each has certain pros and cons, as you can imagine. I don’t begin to know the ins and outs of them, but some of the difference between what is reported as “inflation” in the normal news and what people feel like they experience can have to do with that. One difficult feature of calculating inflation is that the so-called “basket of goods” that people, on average, purchase is constantly changing. In addition, even things that seem to be unchanged don’t necessarily stay the same over time, due to technological changes. A computer today is kinda the same, but also kinda different from one bought 5 or 10 years ago. The retail price might be the same, but what the computer can do has changed. How can that change in quality be calculated into inflation? That’s one of the issues that means there’s a little bit of art mixed in with the mathematics.

    If you go to the Consumer Price Index page, you’ll see that there’s lots of variety in what gets called inflation. You can see what inflation has been for most all goods, or a subset of goods (like food and beverage, or medical care, or housing); or you can see what it was in a particular city or region. No one person will actually purchase the average things in the average location, so the inflation you, as an individual, experience will only be roughly measured by the CPI. But when trying to analyze overall trends, it sure comes in handy.

    As for a weighted average of energy prices, I actually think that wouldn’t be so hard for someone used to doing weighted average math–which definitely is NOT me! I’ve done enough of it to think that it can be done “easily,” but not enough recently enough to be able to do it instantly. I’d have to fiddle with the calculation to make sure I got it right. But I’m sure there are math and econ geeks out there that could do it in their sleep. I mean, you’re right that different weights would give you different answers, but the choice of weights wouldn’t have to be entirely arbitrary. Instead, you’d choose your weights based on the existing data of how much oil was used one year vs. how much natural gas that year vs. how much coal. That’d give you the weights for the prices of those energy sources for that year. Then you’d look at the next year and get new weights the same way.