London, 14 December 2009 

Current debates about cutting energy consumption and carbon emissions often carry a strong undercurrent of asceticism. There is an almost missionary zeal to save the planet by reverting to a simpler and more satisfying past when energy consumption was lower (or at least encourage other people to make necessary sacrifices). 

But the post-war experience of the United States and other industrialised economies suggests it is possible to combine rising living standards with the same or lower energy use.

Berkeley Physics Professor Richard Muller has described the phenomenon as "comfortable conservation", the idea that it is possible to curb greenhouse emissions without sacrificing quality of life.

Given correct price signals, the advanced economies have made huge advances in using energy more efficiently, with most of the gains coming in the 1970s and 1980s in response to the dramatic rise in real energy prices after the oil shocks.

If China and other emerging markets match these gains in future it would avoid a substantial part of the projected emissions increase. Measuring efficiency changes over long periods of time is surprisingly difficult. Estimating energy use is straightforward. Output and consumption are harder, though, because the mix of both production and consumption changes too much over extended periods.

There is no agreed way to compare output and consumption today with the 1940s. Most observers would concur that adjustments should be made for population increase and inflation to provide real per capita numbers. But constant-dollar GDP, chained-dollar GDP, and manufacturing production all have strengths and weaknesses as indicators of real output or consumption.

Nonetheless, on any measure, the United States has experienced a revolutionary improvement in energy efficiency since 1945.

Energy Intensity Falls

The attached chartbook provides various indicators of U.S. energy consumption and efficiency between 1949 and 2008.

 See CHARTBOOK AT END:  

Total consumption from fossil fuels, nuclear, renewable sources and imported electricity tripled between 1949 and 2008, rising from 31 quadrillion British thermal units (BTUs) in 1949 to 99 quadrillion last year. But over the same period real economic output rose seven-fold, from $1.9 trillion to roughly $13.5 trillion.

In 1949, households and businesses consumed around 16,500 BTUs of primary energy for every dollar of GDP (in 2008 prices). By 2008, energy intensity had fallen more than half to just 6,900 BTUs.

The oil shocks mark a clear break point in the series. Before the 1970s, the economy was actually becoming steadily more energy intensive, with per-dollar consumption rising from 16,500 BTUs in 1949 to a peak of 21,300 BTUs in 1969, mostly as a result of widespread electrification, the spread of appliances such as refrigerators, and mass car ownership.

But the oil shocks spurred an enormous conservation effort that saw energy intensity halve in just 15 years to 9600 BTUs per dollar in 1984. Progress has continued, though more slowly, as the easy gains had already been achieved and the decline in real energy costs throughout the late 1980s and 1990s removed much of the urgency.

 .. By Any Measurement

Some observers will object that using constant-dollar GDP fails to account for substantial changes in both the mix of consumption (from merchandise to services, and between different types of product) and production (from manufacturing to services, and domestic production to imports, with energy intensive industries displaced offshore and replaced by imports).

For critics, the improvement is a mirage, caused by the hollowing out of the U.S. manufacturing base and its replacement by imported goods. But running the same analysis with chained-dollar GDP (to take account of changing consumption patterns) and estimates of manufacturing output (to isolate the effect of offshoring) shows similar reductions in intensity.

In chained dollars, energy consumption per unit of GDP more than halved from 17,300 BTUs in 1949 to 7,500 BTUs in 2008. In terms of manufacturing production, energy consumption has fallen almost 60 percent per unit of output. Using chained-dollar GDP or manufacturing indices tends to smooth the trend and remove the pronounced break in the 1970s. Both series show bigger efficiency improvements during the 1950s and 1960s followed by smaller gains in the 1970s and 1980s.

But that is simply because the massive escalation of energy prices caused an enormous and very rapid shift in consumption and production patterns; the composition as well as the level of output is very sensitive to energy prices. The end result is the same.

Improvements in efficiency are even more dramatic if consumption is adjusted for population growth. Despite reductions in intensity, the total amount of energy used has continued to climb from 75 quadrillion BTUs in 1973 to 99 quads in 2008.

But almost all the gains can be explained by the continued growth in the U.S. population. More people naturally use more energy. If we adjust the increase in energy use (31percent since 1973) for the rise in the resident population (44 percent) energy consumption per person has actually been flat or falling for the last 35 years.

Huge Efficiency Scope

They key point is that energy consumption does not have to rise linearly with economic output and living standards. Given the right price signals, and ongoing pressure to squeeze out efficiency improvements in a competitive market economy, U.S. households and firms have achieved enormous efficiency gains in the past 60 years. Even larger gains have been made in Japan and many European countries.

If energy prices rise high enough for long enough, either because of another price spike or as the result of carbon taxes or cap-and-trade programmes, usage will fall sharply.

While there is still scope for significant efficiency improvements in the advanced economies, the real prize lies in emerging markets.

If China, India and other developing countries can replicate the same efficiency trend, there is scope to raise their production and living standards significantly while mitigating the impact on energy consumption and greenhouse gas emissions.

The noticeable acceleration around the time of the oil shocks suggests price signals are absolutely crucial. Putting an appropriate price on emissions via a carbon tax or cap-and- trade programmes is therefore essential in the longer-term.

But a practical first step, especially in many developing countries, would be to lift regulated, often loss-making rates for electricity and other fuels. While they have worthy social goals, most of the benefits flow to middle class households rather than the poor, and they do nothing to provide efficiency incentives.

In the absence of proper price signals, funding from the advanced industrial economies and technology transfer will only achieve limited improvements. Only a rise in energy costs in the developing world will spur the roll out of more efficient appliances and vehicles essential to replicate the efficiency gains achieved in the United States and elsewhere, limiting future emissions increases.

Ends --

Attachments:

chartbook1

593 Kb