Hard cash and climate change: repost from 2005
While thinking about decarbonizing transport, I dug out this old post from 2005. It’s interesting to see how the debate has evolved (or not) since then.
The big change has been that the prospects for technological alternatives like alternative energy sources and electric vehicles have improved dramatically. As regards transport, I don’t see much reason to change the analysis I presented in 2005. Unfortunately, while some progress has been made along the kinds of lines I suggested, it’s been very limited compared to the radical changes in electricity generation. So, we are only at the beginning of the process of decarbonizing transport.
Tim Worstall gets us past that pesky NYT paywall to link approvingly to a John Tierney column arguing that the way to encourage energy conservation in the US is not to fiddle with standards but to raise prices. Broadly speaking I agree. At a minimum, getting prices right is a necessary condition for an adjustment to sustainable levels of energy use. Nevertheless, the rate of adjustment and the smoothness with which adjustment takes place can be greatly enhanced by the adoption of consistent pro-conservation policies, or retarded by the adoption of inconsistent and incoherent policies.
This is as good a time as any to restate the point that, given a gradual adjustment, very large reductions in energy use and CO2 emissions can be achieved at very modest cost. Rather than argue from welfare economics this time, I’ve looked at the kind of adjustments that would be needed to cut CO2 emissions from motor vehicle use (one of the least responsive) and argued that price increases would bring this about over time, without significant pain.
With the price of gasoline in the US passing $3/gallon and most of the remaining sceptics now conceding the reality of human-caused climate change, it seems like a good idea to re-examine some fundamental assumptions in the debate over climate change. Rather than focus on the short-run arguments about the Kyoto protocol, it seems more useful to focus on the question of whether anything can really be done to stop climate change.
A common estimate is that to stabilise the global climate, we would need to reduce emissions of carbon dioxide by 60 per cent, and proposals to achieve this by 2050 have been put forward. Assuming only a limited role for alternative energy sources, it seems reasonably to look at a 50 per cent reduction in primary energy use.
It’s a widely-held view that the kinds of changes required to stabilise the global climate must imply a fairly radical reduction in our material standard of living. This view is shared by radical environmentalists, who see such a reduction as a good thing, and by opponents of such changes most of whom, at least in developed countries are on the free-market right.
The fact that radical environmentalists view the modern economy as critically dependent on unsustainable patterns of energy use is not surprising. On the other hand, supporters of the free-market generally praise the flexibility of dynamism. Currently, energy use accounts for about 6 per cent of GDP. The suggestion that reducing this proportion to, say, 3 per cent, is beyond our capacity seems to represent a very pessimistic view of our economic potential.
There’s a standard economic technique for giving a rough estimate of the economic cost of such a shift. Begin with the assumption that in the long run, the demand for energy is sufficiently flexible that a 10 per cent increase in costs will eventually produce a 10 per cent reduction is usage, relative to the underlying trend. Although energy use responds slowly to price changes in the short run this is a fairly conservative estimate of price responsiveness over periods of a decade or more.
Given this assumption, halving energy use would require a 100 per cent increase in prices (by coincidence this is about the change that’s been seen in US gasoline prices in the last few years). A standard economic calculation suggests that the reduction in economic welfare associated with such a tax would be somewhere between 50 and 100 per cent of the revenue raised, or between 1.5 per cent and 3 per cent of GDP. That’s about one year’s worth of economic growth. Remember that this estimate is not for the modest first steps required under Kyoto, but for a reduction in emissions on the scale required to stabilise climate.
Is such a broad-brush estimate reasonable? One way to check is to look in detail at the kinds of changes that would be needed to achieve such a reduction in the most sensitive single category of energy use, that of private motor vehicles.
Consider changes over twenty years, a period long enough for the vehicle fleet to turn over, and for people and firm to make adjustments to home and work locations, commuting and shopping patterns, and so on.
First, a significant reduction could be achieved simply by improvements in the technical efficiency of fuel use. The motor vehicle industry, although technologically mature, still exhibits steady improvements in the efficiency of engines and other aspects of vehicle design. When fuel prices are low, much of the effort is allocated to improving performance.
When fuel prices are high, and policy is oriented towards reducing energy use, innovations that improve fuel economy are favoured. Over 20 years, and with support from publicly funded research, it seems reasonable to anticipate a 20 per cent improvement in fuel economy, for all types of vehicles, relative to the ‘business as usual’ trend.
Second, some shift towards alternative fuels could be anticipated. While radical alternatives such as ethanol and hydrogen and alternatives to internal combustion such as electric cars have so far proved disappointing, an increase in the effective cost of petrol would encourage greater use of existing alternatives such as LPG and diesel, which are more efficient in terms of carbon emission.
Yet further improvements could be achieved with measures to reduce traffic congestion, including purely technical innovations such as more sophisticated management of traffic lights and market innovations such as congestion charges.
Next, the mix of vehicles in the fleet would change over time. The gain from this source can be illustrated by a simplified example. Suppose that half of fleet uses 10l/100km, and half uses 5l/100km, yielding an average of 7.5l/100km. If the proportions changed to 25:75, the average would fall to 6.25,and fuel use would fall by 15 per cent. Most of this change would arise as a result of consumer responses to changing prices. However, existing policies that favour the use of large, inefficient vehicles (such as the special treatment of SUVs in US fuel economy regulations) should be scrapped, and replaced by policies pointing in the opposite direction.
A small further saving, say 5 per cent, could be achieved through discretionary decisions on which vehicle to use for a given trip. Given high fuel prices, a household with a small car and a 4WD might be more inclined to use the small car when dropping the kids off at school, for example.
A similar small change, say a 5 per cent reduction in fuel use, could be achieved through improved driving habits. These include stricter adherence to speed limits on open roads, and avoiding excessive acceleration and braking in urban areas.
So far, we’ve considered changes which involve no change at all in travel patterns (with the exception of congestion pricing, which would actually improve things), and only marginal adjustments in lifestyle. The biggest single change, in the fleet mix, would do little more than restore the mix prevailing in, say, 1980. Yet taken together, these changes would be sufficient to reduce energy use by between 30 and 40 per cent and CO2 emissions by an even larger amount.
Now consider some changes in travel patterns. The most important single variable is the distance travelled by each person. To get an idea of feasible magnitudes let’s consider a 20 per cent reduction in distance travelled. For commuting, the biggest single use of time, this could be achieved if people chose to live a little closer to work, to rearrange schedules to allow a four-day week, or to telecommute one day each week. Similar savings could be made on shopping and leisure travel with only modest costs.
The fuel cost of travel also depends on the extent to which people share cars. The average occupancy of cars has declined steadily reaching about 1.1 persons per vehicle for commuting trips in the US in 2000, and about 1.5 persons per vehicle for all trips. A partial reversal of this trend, raising occupancy to 1.65 persons would reduce fuel use by 10 per cent for a given number of person-km travelled.
Finally, there’s public transport and alternatives to cars like bicycles and walking. Doubling the share of these would reduce the number of vehicle trips by around 10 per cent, though the reduction in fuel use would be smaller since mostly short trips would be avoided.
Adding all of these modest changes together would yield a reduction in fuel use of more than 50 per cent Some of these changes would be imperceptible, others would require marginal adjustments over a couple of decades. Taken all together, they would be barely noticeable relative to the changes in lifestyle that most people experience over such a period.
You might think that adding together a whole lot of small changes in the same direction is stacking the deck in some sense. But this is the way markets work. An increase in the effective cost of some commodity generates adjustments on many different margins, all in the direction of economising on that commodity.
It is also the way coherent public policy works. If a goal of reducing energy use or CO2 emissions is properly embedded in public policy, it will be reflected in modest shifts in many different dimensions of policy, producing a significant aggregate impact.
The combination of price responsiveness and public policy can be seen working together in the reduction in tobacco use over the forty-odd years since the link between smoking and cancer was first officially recognised in the US in the Surgeon-General’s report of 1964. At the time, the proportion of men who smoked was 52 per cent and smoking among women was rising rapidly as older social taboos lost their effect. In 2000 the proportion who smoked was down to 25 per cent for men, and 20 per cent for women and was declining for both groups.
Admittedly, the health risks of smoking are borne mainly by the smoker, so the link between giving up and receiving benefits is direct and personal. Against this, nicotine is possibly the most addictive drug known to humanity. Giving up smoking requires an effort far greater than the modest changes discussed above.
The reduction in smoking was achieved by a combination of higher taxes, aggressive public information campaigns and public policies that gradually limited smoking in various public places, but without any radical changes or any element of compulsion comparable to Prohibition of alcohol or of the many drugs that are currently illegal.
What is true for driving and smoking is even more so for other forms of energy use, particularly in business and industry. Given a consistent upward trend in prices and a coherent set of public policies, massive reductions in energy use would follow as surely as night follows day.