Peak Watch

Questioning Peak Oil Economic Assumptions

Due to the lack of trumpeters, the end of the world has been postponed for three weeks
     —anonymous

There is a large and expanding economic literature that seeks to explain why the steep oil price rise since 2003 has not led to a recession. The common conclusion, arrived at by different models and analysis methods, is that the U.S. economy is now mostly immune to oil prices hikes and has been so since the mid-80's. Peak oil proponents assume a simple model of the relation between oil supply or price shocks and economic performance. Economists have called into question some aspects of that model. Those studying peak oil need to make cogent responses to ensure that their view is taken seriously. It is beyond the scope of this column to do a detailed analysis, but it is possible to lay out some of the current debate and the issues that need to be addressed.

Peak Oil Economics

Peak oil economics is largely based on hypotheses presented in the Hirsch report.¹ The basic model asserts that global oil supply disruptions create price shocks which lead, in turn, to recessions. (See pp. 26-32 of the report.) The first graph (left) shows supply shortfalls since 1954. The second graph (below left) shows the historical relationship between oil prices and U.S. recessions since 1969. Hirsch et. al. explain that "oil price increases have preceded most U.S.recessions since 1969, and virtually every serious oil price shock was followed by a recession. Thus, while oil price spikes may not be necessary to trigger a recession in the U.S., they have proven to be sufficient over the past 30 years."

The oil price shock of 2003-2007 has not caused a recession in the U.S. Thus the correlation cited by the Hirsch report is contradicted in the recent period. The supposition that oil price shocks lead to recessions was first explored in various works of Professor James Hamilton of the University of California, San Diego. Hamilton's work is referenced in San Francisco Reserve Bank's 2005 study Why Hasn't the Jump in Oil Prices Led to a Recession? and is cited, along with other sources, in the Hirsch report.

Current Economic Conditions

Before discussing the arguments made in more recent economics papers, it is useful to review the economic conditions relating to oil that exist today.

1. Oil prices nearing $100 a barrel are just now close to the inflation-adjusted levels of the 1970's and early 80's oil supply shocks. See the price graph at Hamilton's Econbrowser article $90 a barrel: Is it time to start worrying about the oil price shock of 2007?
2. Energy expenditures as a percentage of household income is moving toward levels last seen in the 1970's and early 80's, but have not reached those levels yet in the U.S. See Not Fixing to Walk, this column, October 21, 2007.
3. The recent oil price shock is said to be "demand-driven". There has been no supply shock in the 2003-2007 period except for that caused by the American invasion of Iraq. The global oil supply according to the EIA supply data has been a bumpy plateau ranging between 83.9 and 85.5 million barrels per day since the beginning of 2005. See The Perfect Storm, this column, October 31, 2007. There have been local exogenous shocks e.g. in the Niger Delta, or the Gulf of Mexico following Katrina and Rita, but these have not led to a significant sustained percentage drop in global supply. Production declines in regions like the North Sea and Mexico have been made up elsewhere so far.
4. Oil expenditures as a percentage of U.S. GDP are nearing levels last seen in the 1970's and early 80's according to Paul Segal's Why Do Oil Price Shocks No Longer Shock? (Oxford Institute for Energy Studies, October, 2007). The excerpt below and accompanying graph is from page 19 of Segal's new study. The graph shows petroleum expenditures as a share of GDP (percent) from 1970 to the present.
   

   Taking the average of the three years leading up the shock and comparing it with the level at the peak of the shock, the rise in petroleum expenditure as a share of GDP was 3.2 percent to 1974 and 3.6 percent to 1979. The rise from the average over 2001–03 compared to 2006 is 2.3 percent; if we assume that the average oil price for 2007 will be $80 then the rise will have reached 3.3 percent. Thus on this measure we have only just reached the level of increase in oil costs that were experienced in the 1970s. [emphasis added]
   

Summarizing the situation as we approach 2008, the ongoing oil price shock has occurred in a context of strong global demand since 2003, and flat supply since 2005. GDP has been growing on average 3.27% per year during the 2003-2007 period.  Sharply rising oil prices in these circumstances implies a very low short-run price elasticity of oil demand reflecting results reported in the University of California (Davis) study Evidence of a Shift in the Short-Run Price Elasticity of Gasoline Demand. Unlike the big prices shocks of the 1970's and early 80's, the world is enduring a demand shock in which oil prices alone have not affected economic growth.

The current run-up in oil prices during a surging economy appears to have no historical precedent. In the absence of supply significant shocks, and as oil prices, household expenditures on energy, and share of GDP taken up by oil expenditures rise toward or match levels last seen in 1970's and early 80's, the world is entering uncharted water.

Views of the Economists

Hamilton updated his views in Oil and the Macroeconomy (August, 2005), stating the basic problem and his observations at that time. His paper is quoted at some length here—

Nine out of ten of the U.S. recessions since World War II were preceded by a spike up in oil prices. One way to inquire whether this might be just a coincidence is with a statistical regression of real GDP growth rates (quoted at a quarterly rate) on lagged changes in GDP growth rates and lagged logarithmic changes in nominal oil prices... (p. 1)

If one estimates a log linear relation between GDP growth and lagged oil prices, the statistical significance of the relation falls as one adds more data (Hooker, 1996), suggesting at a minimum that a linear relation is either misspecified or unstable... The size of the effect [oil price shocks on GDP] is substantially smaller as well— whereas the 1949-80 regression would predict that GDP growth would be 2.9% slower (at an annual rate) four quarters after a 10% oil price hike, the 1949-2005 regression would predict only 0.7% slower growth... (p. 7)

As noted by Kilian (2005), civil unrest in Venezuela in December 2002 led to a drop in production of 2.3 million barrels a day, representing 3.4% of world production at the time. The net oil price series ... reflected a surge in crude oil prices 20% above their previous 3-year high. Nevertheless, there was no discernible drop in GDP. Another surge in [oil prices] of 18% occurred in 2004:III [3rd quarter], accompanied by a 1.3% increase in world production, and a third surge of 21% in 2005:I, accompanied by a 0.2% increase in production, with no recession as of the time of this writing... [and below, pp. 9-10]

It is clear from the last two examples in particular that demand increases rather than supply reductions have been the primary factor driving oil prices over the last several years. Insofar as these demand increases resulted from global income growth, one wouldn’t expect to see the sharp drop in consumer spending on other key items that accompanied the episodes in Table 1 [see the document]. At a minimum, the failure of a recession to result as of the time of this writing from the oil price increases of 2003-2005 suggests that there is not simply a mechanical relation, even a nonlinear one, between oil prices and output. The experience is consistent with the claim that the key mechanism whereby oil shocks affect the economy is through a disruption in spending by consumers and firms on other goods and that, if this disruption fails to occur, the effects on the economy are indeed governed by the factor share argument [oil expenditures as a share of GDP] [emphasis added]

Demand increases, not supply reductions, have been the main driver of the 2003-2007 oil price shock. Hamilton wrote this in 2005, and there is still no recession as 2007 winds down. Consumer spending, which makes up about 70% of U.S. GDP, has remained strong during the price run-up. Since consumer expenditures have not been curtailed, Hamilton surmises that there has been no recession because oil expenditures as a percent share of total GDP remains low, or at least below levels seen in the 1970's and early 80's. This "factor share" is now going up, as noted above.

Here is a partial list of recent academic studies or web articles that claim that oil price shocks alone are not sufficient to cause economies to tip over into recession. Some of these are referenced in Hamilton's recent Econbrowser story cited above, others are not.

1. Why Do Oil Prices No Longer Shock? — Paul Segal
2. The Macroeconomic Effects of Oil Price Shocks: Why are the 2000s so different
from the 1970s — Olivier J. Blanchard and Jordi Gali
3. Declining Effects of Oil-price Shocks — Munechika Katayama
4. The Economic Effects of Recent Increases in Energy Prices — Congressional Budget Office
5. Not All Oil Price Shocks Are Alike: Disentangling Demand and Supply Shocks in the Crude  Oil Market — Lutz Kilian
6. Be Not Afraid — Jerry Taylor and Peter VanDoren

The Segal (2007) paper (link op. cit.) provides a literature survey of economic studies, including Hamilton's original studies and some of the ones listed above, so we will start with the main results reported there and then present the conclusions of other works. Segal's historical literature survey and discussion supports these key findings—

We also saw, however, high oil prices have had no perceptible impact on the macroeconomy over the last few years, and that in the data oil price rises already stopped having an impact [on GDP] some time in the 1980s. At the same time, oil price rises stopped passing through to inflation, and this may hold the explanation: if oil price rises do not raise prices, then interest rates do not need to respond to them, and the impact on aggregate activity may therefore be minimal. (p. 22)

The Blanchard and Gali study explains the recent failure of higher oil prices to lead to recession as  follows—

We examine four different hypotheses for the mild effects on inflation and economic activity of the recent increase in the price of oil: (a) good luck (i.e. lack of concurrent adverse shocks), (b) smaller share of oil in production, (c) more flexible labor markets, and (d) improvements in monetary policy. We conclude that all four have played an important role.

Two conclusions clearly emerge from this analysis: First, there were indeed other adverse shocks at work in the 1970s; the price of oil explains only part of the stagflation episodes of the 1970s. Second, and importantly, the effects of a given change in the price of oil have changed substantially over time. Our estimates point to much larger effects of oil price shocks on inflation and activity in the early part of the sample, i.e. the one that includes the two oil shock episodes of the 1970s.

Our basic empirical findings are summarized graphically in Figure 1 [shown left]. The left-hand graph shows the responses of U.S. (log) GDP and the (log) CPI to a 10 percent increase in the price of oil, estimated using pre-1984 data. The right-hand graph displays the corresponding responses, based on post-1984 data. As the Figure makes clear, the response of both variables has become more muted in the more recent period. As we show below, that pattern can also be observed for other variables (prices and quantities) and many (though not all) other countries considered. In sum, the evidence suggests that economies face an improved trade-off in the more recent period, in the face of oil price shocks of a similar magnitude.

Segal, Blanchard and Gali share the point of view that the structure of the economy is now different than it was in the 1970's in so far as the effects of oil price shocks on GDP is more "muted". They cite various factors, including the lack of a contribution to core inflation, more flexible labor markets and greater efficiency of oil use in the economy. Lutz Kilian goes even further—

One of my key findings is that no two oil price shocks are alike. Nevertheless, there are some regularities. All the major real oil price increases since the mid-1970s can be traced to increased global aggregate demand and/or increases in oil-specific demand. The latter demand shifts are consistent with sharp increases in precautionary demand in the wake of exogenous political events in the Middle East. In contrast, disruptions of crude oil production play a less important role, suggesting that the traditional approach of linking oil price increases to exogenous shortfalls in crude oil production must be re-thought. The most recent build-up in the real price of oil, in particular, can be attributed almost entirely to positive global aggregate demand shocks. Moreover, when political events do affect oil prices, as during the Persian Gulf War, for example, my analysis suggests that it is less the actual physical supply disruptions than the increased precautionary demand for oil triggered by fears about future oil supply shortfalls (which may or may not be realized) that is driving the price of oil. [emphasis added]

Reading The Economic Impact of Oil Supply Shocks on the G7 Countries (Killian, 2006), it is evident that his analysis completely contradicts the Hirsch report's views on endogenous oil supply shocks, even those of the 1970's and early 80's. Hirsch et. al. state that "for purposes of this study, the 1973-74 and 1979 disruptions are taken as the most relevant, because they are believed to offer the best insights into what might occur when world oil production peaks." Killian asserts that "The 1973/74 oil supply shock did not play a major role in the 1974/75 recession."

In the contrarian cases listed here, economists have sought to divorce oil price shocks, in part or in whole, and regardless of whether they are supply or demand driven, with subsequent recessions.

Discussion from a Peak Oil Perspective

Peak oil theorists assume without question that oil price shocks, especially if these are due to major supply disruptions, lead to recessions. In the 2003-2007 period, a price shock caused by strong demand has not led to a recession. Here are some relevant considerations for thinking about the problem.

1. The oil supply has been a bumpy plateau since January of 2005. This may be the permanent global peak of production, a condition that some plausible supply scenarios suggest could last many years. Production might even remain in a more flexible 82-87 million barrels per day range for another decade or so, depending on demand levels set by price and productive capacity. We must also think about the price inelasticity of oil demand in this context.
2. If these economic studies are correct that oil price shocks do not alone inhibit GDP growth, and in the absence of a large endogenous supply shock in the Middle East or elsewhere, the U.S. economy could muddle through the coming decade at various levels of GDP growth. If a recession does occur soon, this will likely be due to conditions that curtail consumer spending such as the credit crunch following the collapse of the housing bubble, with rising oil prices and expenditures serving as an important contributing—not the sole—factor.
3. A monotonic, perhaps exponential, decrease in global oil production over several consecutive number of years, signaling that the bumpy plateau period is over, will cause the kind of supply shock that peak oil theorists believe will lead inexorably to shrinking economies (measured by GDP) and great human distress. Unlike previous oil (supply or price) shocks, oil shortages on the world market would be a permanent, not a temporary condition. It is this case, where no growth in the oil supply is possible, that the resiliency of industrial economies will be put to the test.

Contrarian economist Paul Sobel concludes like this—

It was also shown that the oil price has only just (as of late 2007) reached the magnitude of the shocks of the 1970s, in terms of total direct cost to the economy. While this may mean that the harm to the global macroeconomy is just around the corner, the existing empirical research would suggest otherwise.

One wonders if there is any oil price at which Sobel thinks that harm to the global macroeconomy will occur. In the case of peak oil (point #3 above), oil prices can only go up unless demand, including oil demand replaced by substitutes, matches or falls below yearly production declines. No doubt Sobel has not considered the peak oil scenario. Rest assured, there are oil prices high enough to harm the global macroeconomy.

Many of those concerned about peak oil take their economic assumptions—naively, peak followed by collapse—as axiomatic. So-called "doomers" assume the worst, but they have not done all of their homework. Peak oil theorists or commentators often ignore the economic literature that discusses the complex relationship between GDP growth and the oil markets. They do so at their peril. By default, the oil supply and price is seen as the single overwhelming determinant of economic performance. This simple model is belied by the demand-driven price shock of the 2003-2007 period. Many economists have also been taken by surprise, and they are scrambling around to come up with explanations.

Many of those studying peak oil need to develop well thought-out scenarios that include models of future economic activity based on their oil supply projections instead of running around telling everybody the sky is falling. The good news is that there may be time to develop such scenarios and implement solutions, like a robust national railroad system, that make sense. Time will tell.

Contact the author at dave.aspo@gmail.com

1. Peaking of World Oil Production: Impacts, Mitigation and Risk Management, by Robert L. Hirsch, SAIC, Project Leader, Roger Bezdek, MISI and Robert Wendling, MISI, February 2005.

Car Crazy

Well, if I had money tell you what I'd do
I'd go downtown and buy a Mercury or two
Crazy 'bout a Mercury, Lord I'm crazy 'bout a Mercury
I'm gonna buy me a Mercury and cruise it up and down the road
     — from David Lindley's version of Mercury Blues

The Paris-based International Energy Agency has issued its  World Energy Outlook 2007 — China and India Insights. The release provoked the usual firestorm of handwringing and protest. "I am sorry to say this, but we are headed toward really bad days," IEA chief economist Fatih Birol told TIME this week. "Lots of targets have been set but very little has been done. There is a lot of talk and no action." Lest the peak oil people feel justified by the bad news, however, Birol made sure to divorce the IEA's position from any possible connection with peak oil (see David Strahan's Supply Crunch is not  Peak Oil — IEA, November 7th, 2007). Oh my! Rather than explain the intricacies of everyone's position here, let's talk about the human love affair with cars as a way of understanding the IEA's annual report.

The thing to keep in mind about motorized vehicles is that almost everyone on Earth who can afford one wants one. The IEA's WEO 2007 oil fact sheet gives the usual demand-driven result: the world will want to consume 116 million barrels per day of oil in 2030. The desire for transport "is the main driver." More precisely, the would-be driver is the main driver in every sense of that word.

The WEO 2007's special focus is on fast-growing energy demand in China and India. The IEA predicts that an astonishing 42% of increased demand for oil will come from those two countries alone, and there is a simple reason for their growing thirst. The IEA expects that GDP will grow in Asia and household incomes will rise accordingly. When incomes rise, people want cars. When people want cars, they require fuel to run them. Therefore we will need 116 million barrels of the stuff everyday, mostly to satisfy our desire to drive. This is the oil part of the IEA's WEO 2007 in a nutshell.

What makes future growth in demand for cars predictable? To answer this question, it is necessary to look at the methodology and assumptions used by the IEA to construct the WEO and specifically, the transport sector module (p. 9). Things get a bit technical here, but don't worry.

As the largest share of energy demand in transport comes from oil use for road transport we have updated the WEM with a detailed sub-model based on an S-shaped Gompertz function, proposed in a paper titled Vehicle Ownership and Income Growth, Worldwide: 1960-2030 by Joyce Dargay, Dermot Gately and Martin Sommer [henceforth VOIG]. This model gives the vehicle ownership based on income (our GDP assumptions through to 2030) and 2 variables: the saturation level (assumed to be the maximum vehicle ownership of a country/region) and the speed at which the saturation level is reached. The equation used is:

Where V is the vehicle ownership (expressed as number of vehicles per 1000 people), y is the saturation level (expressed as number of vehicles per 1000 people), a and b are negative parameters defining the shape of the function (i.e. the speed of reaching saturation). The saturation level is based on several country/region specific factors such as population density, urbanization and infrastructure development...

Passenger car ownership is then calculated based on the vehicle breakdown from World Road Statistics 2006 from the International Road Federation plus other regional statistics. Using the equation above, changes in passenger car ownership over time are modelled, based on the average current global passenger car ownership. Both total vehicle stock and passenger vehicle stock projections are then derived based on our population assumptions. [links, emphasis added]

The function that produces the S-shaped curve provides a model for understanding market penetration over time for a new or existing technology, e.g. cars, ipods or cellphones. Examining the historical data from 1960 to 2002 in 45 countries, the VOIG economists found that the Gompertz function defined above was the best model describing past growth in car ownership. This function was then applied to the future based on predictions of GDP and population growth, the expected degree of urbanization or population density, how many roads will be built, etc.

The graph (first above left) shows an illustrative Gompertz function and its implied income elasticity. The function (on the left) shows the growth curve over time. The curve eventually flattens out—this is the point of saturation and no more growth in vehicle ownership per thousand people occurs as income rises. However, if a country's population expands, the total number of cars will continue to grow.

The income elasticity of demand for cars (on the right) is a bit tricky, but can be easily explained. The number of cars people buy—actually the percent change in the quantity of cars demanded—rises with income, but this growth rate is not steady (linear). As soon as people can afford one, they buy a car, and increasing income levels stimulate fewer car purchases thereafter. The VOIG economists explain—

The relationship between the growth of vehicle ownership and per-capita income is highly non-linear. Vehicle ownership grows relatively slowly at the lowest levels of per capita income, then about twice as fast as income at middle-income levels (from $3,000 to $10,000 per capita), and finally, about as fast as income at higher income levels, before reaching saturation at the highest levels of income.

The second graph (above left) shows the projected Gompertz curves and associated income elasticities for six different countries, the USA, France, and the Netherlands representing the developed (OECD) economies (left), and Indonesia, China and India on the undeveloped side. Saturation levels vary among the various countries, as do the income elasticities. For example, "USA and France have similar saturation levels but different low-income curvatures: USA reaches 200 vehicle ownership at per-capita income of $7,000 while France reaches it at $9,400." Thus Americans are slightly more car crazy, just a bit more eager to own a car when they can afford it. This is probably due to the relative size of the two countries and America's sacrosanct car culture.

The OECD economies are at or near their saturation levels, but that's not the case in the developing world. The saturation levels shown for the Asian countries are projections derived from the Gompertz model. The authors give their results—

We project that the total vehicle stock will increase from about 800 million in 2002 to over 2 billion units in 2030. By this time, 56% of the world’s vehicles will be owned by non-OECD countries, compared with 24% in 2002. In particular, China’s vehicle stock will increase nearly twenty-fold, to 390 million in 2030. This fast speed of vehicle ownership expansion implies rapid growth in oil demand...

By 2030, China will have 269 vehicles per 1000 people – comparable to vehicle ownership levels of Japan and Western Europe in the early 1970’s – and it will have more vehicles than any other country: 24% more vehicles than the USA. China’s vehicle ownership is projected to grow rapidly for two reasons: (1) its projected high growth rate for per-capita income during 2002-2030, 4.8% (which is actually much slower than its recent rapid growth), and (2) vehicle ownership is growing 2.2 times as fast as per-capita income, as it passes through the middle level of per-capita income ($3,000 to $10,000) with the highest income-elasticity of vehicle ownership. Similarly for India and Indonesia...

The Chinese are, and will be, snapping up automobiles as soon as their income levels permit it. The prediction (graph left) is that there will be over 2 billion vehicles on the Earth by 2030. And China will not yet be saturated. The countries graph above clearly shows that China's saturation level is predicted to reach 800 vehicles per thousand people, whereas their level of saturation will only be about 34% of that level (269 cars) in 2030. There's still plenty of room for growth! Not to mention Indonesia and India...

Elizabeth Kolbert, author of Field Notes from a Catastrophe, an excellent book on climate change for the non-scientist, recently published an interesting article in the New Yorker called Running on Fumes: Does the "car of the future" have a future?  Talking about "techno-optimists" like Iain Carson and Vijay Vaitheeswaran, who are promoting this car of the future, Kolbert had this to say—

Yet their [the techno-optimists] argument rests on an account of global trends that is nothing short of terrifying...

Were China and India to increase their rates of car ownership to the point where per-capita oil consumption reached just half of American levels, the two countries would burn through a hundred million additional barrels a day. (Currently, total global oil use is eighty-six million barrels a day.) Were they to match U.S. consumption levels, they would require an extra two hundred million barrels a day. It’s difficult to imagine how such enormous quantities of oil could be found, but, if they could, the result would be catastrophe. “Just consider the scale of the potential problem—for instance, the effect on global warming of seven hundred and fifty million more cars in India and China, belching carbon dioxide,” Carson and Vaitheeswaran write.

Techno-optimists believe that business as usual can continue indefinitely. Although one can imagine global gridlock, the ultimate global traffic jam in which every square meter of drivable surface on the Earth has a car sitting on it, the cornucopian principle that there are no practical limits to growth is still held sacred. All that's necessary to do is to change automobile technology by switching over to so-called hypercars that get 100 miles per gallon or some other nifty alternative. But Kolbert isn't buying it.

... improving gas mileage will take us only so far. Once the Chinese and the Indians really start driving, doubled or even tripled fuel efficiency won’t suffice. This is why Carson and Vaitheeswaran regard the Prius merely as a stopgap: the true car of the future has to accommodate everyone, which is to say six and a half billion, soon to be nine billion, people.

Hard-core techno-optimists insist that this goal, too, could be met, if only automakers and politicians would apply themselves to the task that up to now they’ve taken such pains to avoid. This is a comforting argument; unfortunately, though, it assumes precisely what’s at issue. After all, just because someone has never bothered to enter the New York City Marathon doesn’t mean that if he runs in it he’ll win.

Unknowingly using some standard peak oil energy arguments, Kolbert reports that she simply can not see where the energy is going to come from to power all these future cars, whether they are lightweight, carbon-fiber hypercars or not. She concludes with this—

If someone, somewhere, comes up with a source of power that is safe, inexpensive, and for all intents and purposes inexhaustible, then we, the Chinese, the Indians, and everyone else on the planet can keep on truckin’. Barring that, the car of the future may turn out to be no car at all.

Bravo! Elizabeth Kolbert, an incisive and gifted writer, has accomplished something the IEA and their economists have all failed to do—she has figured the problem out and suggested the most probable outcome. We are grateful for her insight, which provides a context for understanding the IEA's World Energy Outlook 2007. And on that cheerful note, have a Happy Thanksgiving.

Contact the author at dave.aspo@gmail.com

Not Fixing To Walk

"Toto, I've got a feeling we're not in Kansas anymore"
       — Dorothy, in The Wizard of Oz

With oil prices stuck in the $90's per barrel range, the official story is that the U.S. economy has shrugged off this year's price increases. National Public Radio's High Oil Prices Not Harming the Economy Yet  (November 12, 2007) duly reports that "while the price of oil affects just about everything that is made, transported, eaten and sold in the United States the cost hasn't had the impact on the economy many people expected." And it is true—the world has not ended yet, the economy has not ground to a halt. Things appear to be going swimmingly now, but the warning signs are there for all to read. What do the longer term trends tell us about the future?

Transportation accounts for about 66% of America's oil consumption. Oil prices may skyrocket, but unless consumers feel the pain at the pump, there is little immediate effect on people's everyday lives. As summer turned to fall, it seemed mysterious that gasoline prices did not rise as oil costs rose. Market conditions did not permit refiners to pass through their increased costs, so their profit margins—measured by the crack spread, the price differential between a barrel of oil and the refined products made from it—had dipped recently to about $5. This was not a situation that could go on forever regardless of where demand levels were. So the gasoline price is now jumping up.

The Houston Chronicle reports that Gas costs expected to follow oil price.

Crude oil's march to $100 a barrel is about to hit Americans square in their wallets, bringing higher gasoline prices that could set records by year's end...

This week, average regular gasoline prices topped $3 a gallon nationwide, the highest ever in November, despite a post-summer slowdown in driving that usually helps ease pump prices this time of year. But the full impact of the recent jump in crude prices still hasn't found its way to the pump, meaning gasoline prices are likely to keep climbing...

"There's no question there are more increases to come," said Michael Wittner, senior oil analyst with investment bank Societe Generale in London. The increases come as gasoline refiners are passing through higher crude costs to customers after those costs ate into refinery profits during the third quarter.The price of crude oil, which accounts for about half the cost of a gallon of gasoline, has risen nearly 30 percent since Labor Day...

On Friday, gasoline for December delivery rose 0.7 percent to $2.46 a gallon in New York Mercantile Exchange trading. To estimate what consumers will pay at the pump, a good rule of thumb is to add 75 cents to the wholesale price for retail markup, taxes and other handling charges, Wittner said.

Wittner's formula predicts that gas prices will rise to an average of $3.21/gallon in coming weeks. On November 13th AAA listed the average regular price as $3.105, up 0.244 from a month ago and 0.778 over last year's price on the same date (the graph left shows this year's trend).

As noted in The Perfect Storm (ASPO-USA, October 31, 2007), oil prices (and hence gas prices) are not yet high enough to provoke major changes in American driving behavior. It is not clear at what price this will happen—perhaps $120/barrel and $4.00/gallon will cause significant changes if these are possible. The price elasticity of gasoline demand (UC Davis study) is quite low, suggesting that Americans are "locked in" to their driving habits. If this conjecture is true, consumers have little choice but to absorb the pain until costs become prohibitive. As DeMarcus Sims told the Shreveport Times about buying gasoline, "you ain't got no choice. And I'm not fixing to walk."

The standard explanation for why increased energy costs, and particularly gasoline costs, have not altered consumer behavior much is that energy spending is still fairly low as a percentage of all household expenditures. This proposition has been true in the past, as the measured by the Chicago Federal Reserve in Household energy expenditures, 1982-2005. The graph (left) shows the overall historical trend. The Chicago Fed found that "energy consumption represented approximately 7% of expenditures between 1990 and 2004, a decrease from 11% in the early years of the 1980s and an average of 9% throughout the 1980s."

Note the spike in 2001 and the steady rise since 2003. The long data series says that everything is going fine, but as Dorothy told Toto, we're not in Kansas anymore. The Chicago Fed estimated 2005 expenditures as approximately 8.5% of the household budget, assuming a price elasticity in the low case of –0.23 for gasoline, which is not in accord with the UC Davis study (link op. cit.), which estimated elasticities of 0.034 to -0.077 for the 2001-2006 period, values much lower than that assumed in the Fed's letter.

The Consumer Federation of America (CFA) has been keeping a close eye on the emerging energy costs trend, and they are increasingly alarmed about the situation. The graph (left) from their study No Time to Waste (October, 2007) shows rising expenditures for energy since 2002. Household energy costs surpassed those for groceries after 2004.

Turning now to gasoline, the recent trend becomes even more compelling. Mark Cooper¹, the Consumer Federation's data analyst, was kind enough to give us his most recent data on expenditures (graph left). The numbers given are nominal dollars. Estimates have been made for 2007 and 2008. Cooper's analysis shows that gas costs have leaped 48.8% since 2002, an average annual increase of 9.76%. As exponential functions go, that's a doozy!

The CFA data is based on the Bureau of Labor Statistics' Consumer Expenditure Surveys and average yearly gasoline prices as compiled by the EIA. The 2006 average price was $261.8/gallon and has risen 52.5% since 1997. The labor statistics 2006 data indicates that transportation is the 2nd largest household expenditure after housing.

Effects on consumers of higher energy prices fall disproportionately on the working poor and other lower income wage-earners who are not "officially" below the poverty threshold. Who are these people? Besides the author, they are the ones cleaning up the building after hours, those who check you out of the grocery store or stock the shelves, the person who gave you that Supercut, the people who served up the large fries with that Big Mac, the people who ate the fries and Big Mac, the ones who sold you that shirt at the Gap, they are Wal-Mart's "associates", etc. Many of these people drive to and from work because public transport is unavailable and they can not afford to live close to their jobs. In America's "service" economy, everyone has the right to a lousy low-paying job. 

The Chicago Fed laid out the less-than-pretty picture (graph above left). While the 1990-2004 average energy expenditure was set at 7%, the number for the entire period was 9.6% for the working poor. It is not surprising that this group led all other categories in the amount budgeted for both gasoline and electricity. It is not hard to imagine the injurious effects of gasoline price increases since 2004 on this beleaguered class of Americans. In a "service" economy, these are the people who provide the most basic services. When Forbes tells you that it's easy Living with $3-A-Gallon gas, it's a sure bet they aren't thinking about the poor. At what point do the lives of lower-income people become nonviable? At the inevitable $4.00 per gallon?

Talk of a recession clouds the issues presented by steadily rising household energy costs. A recession, especially if it is severe, would temporarily interrupt the upward trend. (Driving tends to go down if you don't have a job to go to.) The horns of the current dilemma (Boston Herold, November 12, 2007) are higher oil prices combined with lower house values now that the housing bubble has officially burst.

The economy - already reeling from a major housing slowdown and a massive financial-sector credit crunch caused by the subprime-mortgage meltdown - is in a “very fragile” state that’s now facing yet another large blow in the form of skyhigh oil prices, said Mark Zandi, chief economist at Moody’s Economy.com.

It is a continuing mystery as to why sharply rising oil prices since 2003 have not resulted in an economic downturn. No one seems to know the answer. Back in 2005, the Federal Reserve Bank of San Francisco asked Why Hasn't the Jump in Oil Prices Led to a Recession?  Noting that prices are riding the rising tide of Asian demand, the San Francisco Fed, in agreement with the IEA's 2007 World Energy Outlook about the causes, had this to say—

Our discussion suggests that the answer to the question posed in the title has two parts. First, looking only at the correlation between some measure of the price of oil and output tends to exaggerate the role that oil price shocks played in the recessions of the 1970s, at least partly because one ends up ignoring the other things that were going on at that time. Second, an increase in the price of oil that reflects higher demand will not have the same effect as a decrease in supply. Here, though, it is useful to keep in mind that price increases that reflect higher growth in other countries will have the same effect on the U.S. as price increases that reflect a reduction in the worldwide supply of oil—unless  U.S. exports to these fast growing countries account for a significant share of U.S. output. [emphasis added]

So the Fed's 2005 study has concluded that if it walks like a duck and quacks like a duck, it must in fact be a duck—which leaves us where we started. What the Fed did not know in 2005 was that oil supply levels have been a bumpy plateau since they did their analysis. In any case, it appears that household energy expenditures are now rapidly playing catch-up with levels last seen in the 1970's and early 1980's. When those levels are eventually reached, a recession driven by energy-related inflationary pressures and lower overall demand by exhausted consumers seems inevitable. Welcome to the future.

Contact the author at dave.aspo@gmail.com

1. I'd like to thank Mark for giving me the new unreleased data set. The new analysis is an update to his The Impact of Rising Prices on Household Gasoline Expenditures published in 2005. The older document shows a very strong correlation between gasoline prices and household expenditures in the 1998-2003 period, which is hardly surprising. Thus projections can be made based on the expectation that this correlation remains strong in future years.

A Modest Proposal

Blessed is he who expects nothing, for he shall never be disappointed.
    — Jonathan Swift

In 1729, Jonathan Swift wrote a satirical essay, "A Modest Proposal," in which he suggested that poor Irish tenant farmers sell their children to be eaten as food to raise money to pay their exorbitant rents. Swift’s ironic proposal was immediately recognized as beyond the pale, but it served to highlight the gross economic inequities prevalent at the time. In modern times, proposals that U.S. consumers actually reduce their consumption of oil and its refined products have been viewed as almost equally unacceptable, as any politician advocating a serious policy leaning in this direction would no doubt tell you.

The modest proposal presented here calls for implementation of a gasoline tax and a floating cap on American oil consumption. These are two of three energy efficiency options examined in the Council on Foreign Relations (CFR) study National Security Consequences of U.S. Oil Dependency, published in October, 2006. Although to many it will seem akin to America eating its young, we must put on the table a serious solution to the looming, grievous threat to U.S. future prosperity fostered by our dependency on oil imports, if only to focus the debate. Politically palatable, and therefore inadequate, plans, such as President George W. Bush’s "20 in 10" initiative mandating a 20 percent reduction in gasoline consumption by 2017, are not strong enough to solve the imported oil dependency problem. The White House initiative depends on three weak measures: tighter corporate average fuel economy (CAFE) standards (5 percent); inefficient, limited corn ethanol production; and cellulosic ethanol production (together, 15 percent). A commercially viable cellulosic process does not yet exist.

The United States has about 4.6 percent of the world’s population and consumes nearly 25 percent of its liquid fuels every day. Ironically, the longer this huge, inequitable consumption imbalance goes on, the greater the jeopardy to America’s economic well-being. There are three primary reasons for this. First, America’s dependence on imported oil is an expensive, continuing drain on the economy. Moreover, real oil import costs are higher than the per-barrel cost, as reflected, for example, in military spending to protect the oil supply. Second, as oil imports increase with each passing year, our energy security is more and more threatened. Increased dependency raises American vulnerability to oil shocks, funds terrorism and places power in the hands of unfriendly, unreliable suppliers like Venezuela, shaky suppliers like Nigeria or other nations that use their fossil-fuels wealth to achieve geopolitical goals. Finally, there are serious costs associated with climate change due to the carbon dioxide emissions resulting from burning liquid fossil fuels.

The security threat will only get worse when production finally peaks or plateaus among non-members of the Organization of the Petroleum Exporting Countries (OPEC) in 2010, as predicted by such diverse sources as ExxonMobil and the International Energy Agency. Thereafter, the world’s "call on OPEC" will increase. However, increasing instability among the Persian Gulf states makes reliance on oil from the region ever more dangerous. U.S. foreign policy is now dominated by attempts to mitigate the oil-supply risks in a more competitive world.

Consumption is Out of Control

U.S. gasoline demand (graph left) averaged 9.295 million barrels a day (a four-week average) ending October 26, 2007. The tally is actually 1% lower than a year earlier, according to the EIA's This Week In Petroleum. The recent downward turn in demand follows on gasoline prices that are 23% higher than last year at the same time. Despite higher year-over-year prices except in the July-September period (graph below left), consumption has been consistently higher than in 2006 up until October's dip.

Price increases have not even been able to keep year-over-year demand flat, let alone decrease it, most of the time. The EIA has suggested that prices would have to be "in the neighborhood" of $3/gallon—where they are now—before demand would come down, but consumption was unaffected in May and June when prices exceeded that level. The dominant pattern seen in 2007 is part of a larger trend. A University of California (Davis) study by Jonathan E. Hughes, Christopher R. Knittel and Daniel Sperling, Evidence of a Shift in the Short-Run Price Elasticity of Gasoline Demand, found that American demand for gasoline is much less sensitive to price increases than it was in the 1970s. The study compared the price elasticity of demand (PE) for two comparable price periods, 1975-1980 and 2001-2006, where this economic concept is defined as follows:

Price Elasticity of Demand = (% change in quantity demanded) / (% change in price)

The higher the PE, the more price influences consumer demand; for higher prices, it indicates larger decreases in consumption due to rising costs. The study found that the PE value for gasoline ranged from -0.034 to -0.077 during the period 2001-2006, compared to -0.21 to -0.34 for 1975-1980 — a decrease of almost a factor of 10. The PE for the October, 2007 data cited above is -0.043, near the low end of the  more recent range. This result clearly shows that the effects of higher gasoline prices on demand in the 2001-2006 period were far less pronounced than during the 1970s, a disturbing trend that is reflected in the EIA’s 2007 data.

Low price elasticity does not indicate zero demand response to higher gasoline prices, however. An article published April 14, 2007, "Californians Buying Less Gas," in the Los Angeles Times, reports that consumption in the Golden State decreased 0.7 percent in 2006 in response to sustained prices well over $3/gallon. This was the first such drop in 14 years. The paltry decrease indicates that market processes alone are not yet up to the task of controlling consumption. Only enforced policy mandates will make a significant dent in demand.

The longer-term consumption trend from 1949 to 2004 shown in Figure 3 (above left) from the EIA’s Annual Energy Review 2005, depicts the ongoing threat to America’s oil security, as consumption reflects the net imports curve. According to the EIA data, net imports stood at 12.4 million barrels per day (b/d) in 2004, accounting for 60 percent of oil demand. By contrast, U.S. production peaked at 11.3 million b/d in 1970, when imports were only 3.2 million b/d. About 68 percent of total oil consumption is used for motor transportation.

The OCI: Implementing a Cap on Oil Imports

The CFR report lists five broad initiatives to improve U.S. energy security. Of these, the first – increasing energy efficiency – is the most important by far because it encompasses direct, immediate measures for actually capping and reducing oil consumption. Many good solutions fall under what can be termed the "different fuels/different vehicles" (DFDV) strategy. For example, in They’ve Had Us Over a Barrel Long Enough, published in the January/February 2007 issue of World Energy Monthly Review, Richard R. Loomis, Brian K. Tully and Susan Salter present a comprehensive overview of the many fuel and vehicle alternatives. The security threat is urgent, however. Stringent policies are needed to push consumers more quickly in the DFDV direction America needs to go. The CFR’s Task Force included three efficiency measures for reducing liquid fossil fuels demand:

1. Impose a federal excise tax on gasoline
2. Tighten and reform the CAFE rules
3. Place a cap on gasoline consumption and adopt a system of tradable vouchers.

As suggested by the recent California trend, only much higher prices than those set by the markets will significantly affect demand. Therefore, a tax of $0.50 to $1 per gallon should be imposed immediately. However, the gasoline tax alone will be insufficient to cut deeply into net imports.

Figure 3 suggests American oil consumption is following the wrong line; it should follow the production curve, not the net imports curve. A modest proposal reflecting this situation may very well be warranted, as follows: The United States should adopt an Oil Consumption Index (OCI) mandating that, at all points in the future, domestic production must provide at least 40 percent of total American oil consumption.

The OCI states specifically how an oil consumption cap is implemented by policy. Should domestic production fall below the required 40 percent, imports must be lowered accordingly until the OCI boundary condition is met. Should U.S. production exceed 40 percent of consumption, imports could increase just to the point where the OCI condition is not violated. The value of universally distributed gasoline vouchers may be tied to the total consumption cap number as defined by the OCI. The hardship of the gasoline tax can be partially offset by the voucher program, which would be constructed to ensure that the excise tax burden does not fall disproportionately on poor or rural communities. Penalties will be assessed for cheating since a black market will undoubtedly arise.

The OCI provides immediate motivation for consumers to change their discretionary driving behaviors. There are strong incentives to implement all DFDV technologies, especially for non-oil-based fuels or highly efficient vehicles. It would probably not be necessary to tighten CAFE rules since the markets under OCI will reward such efficiency. In addition, the United States can start to assume a leadership role in reducing carbon dioxide emissions. Gasoline tax revenue may be used to fund the infrastructure necessary to implement alternative DFDV strategies. As imports are reduced over time, the United States can offload unreliable or unfriendly suppliers. The oft-cited argument that the United States can drill its way out of its predicament will be put to the test: The OCI provides heavy incentives to increase domestic oil production. Whether the Arctic National Wildlife Refuge or the East and West Coast continental shelves will be opened up for exploration and production is an additional policy decision involving uncertain rewards that must be weighed against the environmental risks.

Complete energy independence is a myth, and the American way of life is negotiable. Consumers need to understand that they use much more oil than they produce and become fully aware of the perilous, unsustainable road on which the nation is embarked. The oil consumption index lays out a coherent road map for achieving greater energy security in the future.

Contact the author at dave.aspo@gmail.com

1. This is a reprint of an article I wrote for the May, 2007 issue of World Energy Monthly Review. A death in the family keeps me busy this week. I will return with a new column next week.