Well then, would $100 a barrel worry you?

A week ago I reviewed the reasons why $90-a-barrel oil by itself would not be enough to cause an economic recession. As oil prices charged up to $96 on Friday, a reporter asked me at what price I’d change my mind.

Last week I described recent research by Olivier Blanchard and Jordi Gali and Munechika Katayama, among others, that concludes that the U.S. economy today may be substantially less vulnerable to oil price increases than it appeared to be in the 1970s. I noted that one common basis for all such claims is the observation that the fraction of our income that is spent on energy purchases is significantly lower today than it was in the 1970s. But I also noted that this is due in large part to the decline in the relative price of energy between 1981 and 1997, and that recent price increases have been reversing that trend.

One hundred times the ratio of (1) annual average of monthly WTI price per barrel (from FRED) times annual U.S. crude oil consumption (from EIA) to (2) annual nominal GDP (from BEA Table 1.1.5). 2007 values based on (1) averages so far in 2007, (2) 2007 average GDP and oil consumption combined with a price of $100 a barrel, and (3) 2007 averages combined with a price of $150 a barrel.

The U.S. consumed 20.7 million barrels of oil per day for the first six months of 2007, which would be 7.5 billion barrels over a year. If prices today were still at the 2004 average oil price of $40 a barrel but the quantity of oil used and nominal GDP were at their values so far in 2007, $40 oil would imply an annual expenditure of about $300 billion, which would be just a little over 2% of our current $14 trillion GDP. But so far in 2007, we’ve seen an average oil price of $68 a barrel, which means an expenditure of $500 billion, over 3.5% of GDP. The above graph extends these calculations into some hypotheticals– What if the oil consumption and GDP stayed where they have been so far this year, but oil prices averaged $100 or $150 over the year? It turns out that a price of $150 a barrel would put us back up to an expenditure share as high as it’s ever been historically. Consumers cannot continue to ignore oil price increases for much longer.

But even if we return to those historical expenditure shares, an oil price increase need not have the same potential to produce a recession as it may have had in the 1970s. I believe that a key reason that we saw economic recessions following the oil supply disruptions of 1956, 1973, 1978, 1980, and 1990 was that these events were associated with sudden changes in consumer spending on items such as domestically manufactured automobiles, and that these demand shifts were a key cause of the subsequent economic downturns. Even if we return to a point where we are spending as much on energy as we were in 1981, the domestic auto companies are not as important to the U.S. economy as they were then. Moreover, gradual oil price increases that occur in the absence of the dramatic geopolitical developments are unlikely to exert the same disruptive effects on consumer confidence as these earlier historical episodes.

Despite the fact that the demand effects of what has happened so far should prove to be smaller than in some other episodes, the economy already faces substantial challenges from the housing downturn and threat of further financial instability. It would not take much to turn the resulting slowdown into an outright recession. For this reason, I’m watching auto sales and consumer sentiment particularly closely,

The automobile sales data released last week do not look too troubling to me. Sales of light trucks, which includes the SUV category that I would expect to respond most dramatically to gasoline price increases, were basically the same in October 2007 as they had been in October 2006 for both domestics and imports.

Data source: Wardsauto.com

Car sales (excluding light trucks) were actually up modestly on a year-to-year basis.

Data source: Wardsauto.com

Source: Conference Board

Consumer confidence, as measured by both the Conference Board and the Michigan-Reuters surveys, appears to be weakening, though it is always hard to be sure about such trends given the noise in these indicators.

Data source: FRED and
Market Watch

In my opinion, the key question as to whether an oil price increase would push the economy into a recession remains the context of the price change. The oil price increases over the last few months were not associated with any actual disruption in petroleum supplies, and do not have the same potential to change consumer sentiment and spending patterns as dramatically as occurred in many of the earlier historical oil shocks. For this reason, even if oil does go above $100, my biggest concern remains the housing sector and financial problems.

But if the tanks start rolling or missiles start flying in the Middle East, my worry factor is going to soar along with the price of oil.

Technorati Tags: ,


48 thoughts on “Well then, would $100 a barrel worry you?

  1. Ivan Kitov

    “But I also noted that this is due in large part to the decline in the relative price of energy between 1981 and 1997, and that recent price increases have been reversing that trend.”
    If to plot CPI and core CPI in the USA between 1960 and 2007 three different periods can be easily distinguished (http://inflationusa.blogspot.com/2007/06/some-observation-facts-about-cpi-and.html)
    1960 to 1980 – both evolve in sync;
    1980 to 2000 – price for food and fuel grew at much lower rate than for other goods and services;
    2003 to present – fast recovery.
    The miracle is not the current price recovery, the miracle is the slow growth between 1980 and 2000.
    Who and how was able manage such a deviation?
    I would guess that oil and food are normal goods and should follow the evolution of overall inflation in the midterm run.

  2. E. Poole

    Are not the prices of gasoline and heating fuel the ones we should be monitoring for indications of consumer and business behavioural change?

    Earlier in 2007, there was considerable public moaning about high gasoline prices driven by tight refining capacity.

    Source: http://www.oilintel.com/spothome.cfm?loc_id=7
    The refining margins as suggested by the above chart have declined precipitously over the summer.
    US domestic retail gasoline prices are only just catching up to late spring 2007 peaks.

    Source: http://www.eia.doe.gov/oil_gas/petroleum/data_publications/wrgp/mogas_home_page.html
    When retail gasoline prices break through the peak experienced in early 2007, I suspect that we’ll observe a marked slowdown in the growth of aggregate demand.

    At one point, high petroleum product prices will leave much machinery & equipment ‘obsolete’. The subsequent economic restructuring should cause a recession. I suspect that price point is much higher than current levels.

    P.S. I gather we may use HTML tags for style but not for imbedding graphic objects. Please consult Source URLs for relevant pictures.

  3. John Thacker

    Another thing is that the price of coal has not risen nearly as much as oil, and it is a substitute for some uses. With oil above $90/barrel and coal around $50 a ton, there’s a shift towards using more coal in places where it can be used. The US exports a lot of coal, too, and those exports are up with the increase in the price of oil. This things combine to reduce the impact of the high oil price.

  4. Ed Sanders

    Since there was no disruption in supply, what do you believe to be the cause of suppliers increased pricing power?
    Growing demand in the face of inelastic supply seems like a big part, but it would seem the declining value of the dollar relative to other currencies also is playing a role.
    While our price for oil has skyrocketed, the price paid in much of Europe has risen far less, obviously the same can be said for Canada, and a number of other countries.
    Considering our trade imbalance there are obviously inflationary implications from this situation, but I’m not sure how significant they are, any insight?

  5. Buzzcut

    There are a couple of off brand gas stations on my way to work that I use to judge gas prices.
    They were at under $2.60 a week or two ago.
    They were over $3.15 a day or two ago.
    Last night one of them was under $3.00.
    Gas prices are nowhere near where one would think that they should be for $96 a barrel. Whatever is driving that crude price, it sure isn’t US gasoline demand, if actual gas prices are any indication.

  6. M1EK

    There is relatively little ability to substitute oil consumption with coal, JT, unless you’re referring to gasification, which is dubious for a lot of reasons beyond the economic.

  7. Buzzcut

    Coal for transportation is problematic, but not impossible:
    The Stanley Steamer could have used coal directly as a fuel!
    How much would it cost to make synthetic fuels from coal. Beats me. Here’s a DOE study done by Bechtel and AMOCO(!?!)

    How much is $19 a barrel converted from ’94 dollars to ’07 dollars? Double it, add 50% as a safety factor, and you’re still in the ballpark.
    Electric cars are essentially a way to provide coal powered mobility.
    All these things are possible if oil gets expensive enough!

  8. M1EK

    Electric cars, yes, except for the battery problem (cell-phone-scale lifetimes were exceeded in hybrids due to narrowing the charge band; not really an option for all-electric). You will know the day of the electric car has arrived when Toyota or Honda starts pushing one – and, yes, the Volt is just FUD.
    The rest of it is propeller-head nonsense – and politically a non-starter even if it ever became economically feasible.

  9. Buzzcut

    F-T is not propeller headed nonsense. It is a viable process, in production in South Africa at this time.
    As for the electric car, look, even v1.0 of GM’s EV1, which utilized lead-acid batteries, had a range of 60 miles. That’s not great, but it would take care of most commuters.
    I actually saw an EV1 on the road once, in Buffalo of all places. Very cool vehicle. I wonder how 10+ years of technology change would affect the cost of that vehicle if GM reconsidered production?

  10. M1EK

    F-T, tar sands, oil shale, are all part of the same nonsense. Those that have used F-T have often done so for reasons other than the purely economic, and in times where the political implications didn’t exist (or in some cases where they were denied access to oil for trade).
    The EV1′s battery was not going to live a long, healthy life. That’s part of the reason why GM crushed them. Again, it’s the narrow-band charging strategy that Honda and Toyota figured out that enables them to ensure that their batteries will last the life of the car, which is really what you need to sell one of these in the foreseeable future. (In some distant future, perhaps people wouldn’t mind having to replace the main battery in 3 years, but that’s unlikely to be anytime soon enough).

  11. Buzzcut

    Tar sands are nonsense, huh? That’s why every big refinery in the midwest is upgrading to be able to refine them?!?
    BP near Chicago is spending $3.8B, Marathon near Detroit is spending $1B and ConocoPhillips near St. Louis is spending an even $4B.
    And the pipeline from Alberta to the midwest is getting a $1B expansion.
    Tar sands aren’t nonsense. They’re here and now!

  12. PR

    Truck sales?
    Please explain to me how sustained high oil prices could not trigger massive inflation as a result of skyrocketing transportation costs of food, every single consumer good, electricity and fuel oil to heat and light each home and business.
    Aren’t we even more vulnerable, as the US economy depends less on manufacturing and more on retail? Any large dip in retail sales will trigger massive layoffs of retail workers, who then will not be able to afford but the barest consumer goods, and certainly not any sort of health care.
    Where am I wrong in my assumptions behind this scenario?

  13. Anonymous

    I started a control systems business in 1972 having industrials along the old Mission San Jose (East 14th) corridor of the East Bay as customers. During the shortage and consequent they bought generators, built bulk fuel storage facilities, … Today, all these firms are gone. It began with the exodus of International Truck, Peterbilt, Catapillar, … Now, from Richmond to Hayward, East 14th looks like Wycliff Ave. in Cleveland.

  14. Obscurity

    I’m guessing you don’t live in the Northeast US. Stockpiles of gasoline in 5 of the 6 refining districts in the US are at near record levels – meaning the supply (short term) of gasoline is huge compared with demand – making the price of crude irrelevant to the end consumer.
    Of course, in the northeast, the stocks are at record lows. So there prices are much more closely correlated with the price of oil.
    It may be a good idea to look at “crack-spreads” which is the divergence between the futures prices on crude and futures prices on gasoline. They typically widen this time of year as a lot of refining capacity switches from gasoline production to heating fuel production.

  15. Barchaser

    Just thought I would comment. I’m an armchair economist and a retired OCC Bank Examiner. Love the website, and I look forward to regular reading. Most of the articles I’ve read so far, confirm what I already believe. Keep it up.
    Gary, Seattle

  16. PR

    Thanks for the tip. After reading that post and re-reading the current post, I see you’re saying that the rate of oil price increase matters more than the instantaneous price, and the technological and political situations will determine the long-term risk.
    Can we increase efficiency (to use as close as possible to Stuart’s zero barrels per GDP dollar) fast enough (or switch to different energy sources fast enough) to an prevent economic slowdown? Will the political situation be stable enough to guarantee a slow increase?
    Are these questions an econobrowser can answer?

  17. JDH

    PR, I believe that the historical relation between oil shocks and recessions results from demand effects such as I described above. But the issues with peak oil, assuming it develops gradually (much like the last 4 years) are different, and have to do, just as you say, with whether technology can adapt sufficiently quickly. And for that question, I fear this Econbrowser does not have a very clear answer.

  18. benamery21

    Actually, there are a lot of places coal can be substituted for oil in proven fuel switching applications, they just aren’t in the U.S., and aren’t short-run options.
    Most oil exporting countries use oil for uses which were abandoned by oil importers eons ago (1970-1985 in the US!) as low value. People actually burn oil in powerplants. At $100/bbl you better believe you’d make a profit on building a coal plant, colliering coal from the U.S., and selling the resultant oil on the market. A 1000MW coal plant with all the bells and whistles would cost about $1.5B — burn about $210M worth of coal per year (at $73/ton which ought to allow delivery pretty much anywhere) — and displace about $1B/yr of crude consumption (at $100 per barrel).

  19. benamery21

    In 2004 36% of Mexico’s electricity was generated using fuel oil. The Middle East is also over 1/3rd.

  20. Avo

    The futures market says oil will be about $82/barrel in 5 years, so obviously there’s nothing to worry about.

  21. kio

    Simple extrapolation of the current evolution of the difference between CPI and core CPI shows an interception with zero line somewhere in 2009, i.e. core CPI will be equal to headline CPI.
    Current values are: 208.490 and 211.628 for the headline and core (less food and energy) CPI, respectively
    in the point of interception, oil price will be around $120. then, if we are back to the period between 1960 and 1980, oil price will be evolving together with other goods and services, i.e. along headline CPI curve.
    In the case (less likely) the difference will continue to evolve the same way as does currently, one will observe elevated oil prices.
    Another opportunity consists in some periodic repetition of the cycle between 1980 and 2009. Then, oil price will be falling during the next 20 years since 2009.
    I guess the first way is the most probable.

  22. M1EK

    “The futures market says oil will be about $82/barrel in 5 years, so obviously there’s nothing to worry about.”
    Avo, you didn’t give enough of a hint to treat this as sarcasm, so I’ll have to assume it’s not – the 5-year futures predicted precisely zero of the last couple of oil spikes, nor did they predict the generally rising trend.

  23. Footwedge

    I have grave doubts about the supposition that energy isn’t that important in our economy any more. The argument about

  24. Peaknik

    The International Energy Agency just published its World Energy Outlook, in its executive summary they say this:
    “Although new oil-production capacity additions from greenfield projects are expected to increase over the next five years, it is very uncertain whether they will be sufficient to compensate for the decline in output at existing fields and keep pace with the projected increase in demand. A supply-side crunch in the period to 2015, involving an abrupt escalation in oil prices, cannot be ruled out.”
    Although the IEA continues to be optimistic (albeit less with each passing WEO, and, hey, next year’s WEO is going to revise its ascendancy on the ultra optimistic USGS 2000 world oil assessment) about the peak of world production, I don’t think it can’t be ruled out. As JHD has said before, oil prices (and futures) should sky-rocket if we are truly peaking.
    But, what if markets are wrong about this, what if the unthinkable (for economists, at least) happens?
    Bob Hirsch said we need 20 years to adapt. Do we have enough time?

  25. Avo

    M1EK: it was sarcasm. I’m one of those who has been criticizing the utility of the futures market as a predictor since JDH first brought up this notion two years ago.

  26. Stuart Staniford

    I took a stab at estimating how fast US oil efficiency in the auto sector could improve in The Auto Efficiency Wedge. Unfortunately, so far there has been almost no movement in the aggregate fleet fuel economy as a result of the high oil prices of the last few years:
    However, at least the automakers are developing better models, and congress is talking about increasing the CAFE standards, so perhaps we are at least placing improvements in the pipeline.

  27. Jack

    Avo’s comment reminds me that I would explicitly like to personally thank Dr. Hamilton for discussing oil futures trading 2.5 years ago as a challenge for those who bought into the peak oil theory. His articles (although probably not intending to do so) gave me enough info to get a commodities account and jump into the fray. So far, so good. It’s also been fun to watch and learn as his opinions have gradually changed over the past few years. His blog is always on my daily read list.

  28. Buzzcut

    Stuart, you knew the automakers were serious about the energy crisis when GM brought out a passenger car diesel in 1977 (engineering on which began pre-1974, amazingly. GM was a great engineering company back in its heyday).
    Despite its awful reputation now, the Oldsmobile diesel was spectacularly successful. GM sold over 1M of them from ’77 to ’86. That’s a lot of cars and very comporable to Prius sales rates.
    And what do you know, GM is bringing out an SUV/light truck diesel for 2010.
    And, of course, there’s all the hybrid models out there and in the pipeline.
    I think that the automakers are more serious about fuel efficiency than they have been in 25 years. If it hasn’t effected the fleet average yet, it should very soon.

  29. Joseph Somsel

    Car purchasers and auto manufacturers would be a whole lot more enthusiastic about diesels if the regulators would cut them some slack about pollution rules.
    Diesels are almost outlawed for passenger cars here in California due to rules that seem overly restrictive to many observers. Building a legally sellable car in this state to use diesel for improved fuel economy seems to require heroic engineering.

  30. Hal

    Avo et al, would you market skeptics agree with the thrust of this alternative point:
    Either there’s nothing to worry about in terms of oil supply, or: the futures market is giving away vast sums of money to anyone who wants it, so you can be rich as Croesus with little risk, and so again there’s not much to worry about personally.
    In other words, whether you believe the market or not, doesn’t the mere existence of those prices force you to be some sort of an optimist?

  31. Buzzcut

    GM’s 2010 diesel allegedly meets California regs.
    No doubt about it, the NOx regs in California are damned difficult to meet with a diesel. But it appears that a few companies are able to meet the standard using urea injection (part of the selective catalytic reduction system, or SCR).
    The problem is that the EPA doesn’t like SCR. They’re worried that the urea tank could go dry, and the driver could just keep on driving. Being an aftertreatment process, SCR has no impact on how a diesel runs. There is no “NOx sensor” to tell the ECU that the engine is not emissions compliant.
    Of course, the same is true of all the aftertreatment systems on a diesel, and the EPA is happy with particulate traps and oxidation catalysts.
    What’s the biggest irony? All the aftertreatment kills the fuel efficiency!

  32. Joyce

    This may be OT a bit but I just want to mention that that VW has diesels in cars in the U.S. I think consumers are just not generally educated enough to know that diesels aren’t the smelly, hard to start cars of yesteryear. I love mine.

  33. Avo

    Hal, no one is going get “rich as Croesus with little risk” off peak oil. Assuming a margin requirement of 5%, and an eventual price of $300/barrel, you get a factor of 60 in profit from here. But that assumes no cash left over for a margin call if the market moves against you. (I’ve never heard anyone claim oil prices will go straight up.) And at very high prices, you have risk of severe economic dislocation and political risk (e.g., perhaps the futures market will get shut down; even today, Fox News is running stories on the evil oil speculators who drive the price up).
    A factor of 60 is very nice; it would turn $100K into $6M, but that’s hardly “rich as Croesus”.
    You could have done much better if you’d bought Iomega in June 1994. In less than 2 years, it was up by more than a factor of 100. Without any leverage at all.

  34. Buzzcut

    OT is the way to be…
    You made a great investment in your VW. You might not know this, but VW did not make an ’07 diesel. I was looking into used ’06 Jetta diesels, and they’re actually selling for MORE than what they stickered for new, with as much as 30k miles on them!
    I broke down and leased a new Saab. It gets economy car mileage, if not the insane mileage of the Jetta diesel.
    Mercedes is really the only choice for passenger car diesels in ’07, and they’re only available in 45 states. Spending $50k on an E320 wasn’t in my budget, sadly.
    Also, here in Chicago, diesel fuel is often more expensive that even Amoco Ultimate.

  35. vorpal

    james, I have difficulty believing that having less industry as a percent of GDP makes us less vulnerable to oil prices.
    We are more insulated from price changes for 2 reasons:
    1) We import goods, which are subsidized via depressed currency exchange rates
    2) Our automobiles are more efficient than in the 70′s
    Reason No.2 genuinely helps our resiliency against the oil price.
    Reason No.1 is only helpful as long as the exchange rates are suppressed. This is a temporary reprieve.
    If today’s cars are twice as efficient as the 70′s then (very) approximately twice the oil price of the 70′s would generate similar results. This could be caused by aforementioned currency adjustments.

  36. Buzzcut

    If today’s cars are twice as efficient as the 70′s then (very) approximately twice the oil price of the 70′s would generate similar results.
    Not only that, but even accounting for inflation, Americans are much richer than in the 1970s. So to reach the same % of income, energy prices would have to be much higher than even their inflation adjusted peaks.

  37. doug r

    Maybe a chart of oil price vs disposable income would be helpful. Kick in the reality of falling housing values as the general perspective and the probable effect of $90 oil might be clearer.

  38. Mike Laird

    Lots of comments and ideas about oil! JDH, my opinion is that an analysis of percent of GDP is flawed – how much is the question. The flaw is that discretionary income today is a lot less discretionary than it was in, say 1980, the last big recession and the peak on your chart. Today, most homes have many more fixed expenses than in 1980. Examples include: much more credit card debt per household, multiple cell phone contracts, higher cable TV contracted expense, internet connection contracts, and in many homes (20%?) adjustable mortgage rates that are going up. The net effect of homeowners 20 year move to more fixed expenses is to reduce truly discretionary funds. As oil prices rise, the crunch will come sooner now than it did in 1980. I believe the tipping point is much lower than $150 a barrel, as your simple exercise would indicate.

  39. fred schumacher

    The most troubling fact in the posting is that light truck sales have not slid. Americans have still not gotten the message that high prices are now permanent. Perhaps it was Katrina masking a long-term trend that makes them think that the present situation is an aberation rather than the new norm.
    Just as oil supply has now reached inelasticity, American disposable income has done so too. Incomes have been stagnant for a long time. Debt loads are way up. There is no way for middle-income Americans to increase their cash flow. They’re tapped out.
    Auto payments and insurance come out of monthly fixed expenses. Fuel cost comes out of disposable income. Continuous high fuel prices will start having an effect on auto buying decisions. Companies with the right products will benefit.
    For the last 20 years I’ve driven mostly minivans. They are, by far, the most versatile vehicle ever invented. I’ve averaged a consistant 24 miles per gallon, making my fuel cost, in the old days, of 6 cents per mile, for a total ownership and operating cost of 11 to 15 cents per mile (I only drive used vehicles). I used to average 45,000 miles per year, commuting between North Dakota and Minnesota.
    After Katrina, I bought a plain-jane 1998 Neon with 5-speed. I average a consistant 38 miles per gallon (44 on the highway). My fuel cost now is 8 cents per mile, with a total cost of 14 cents per mile. I drive the Neon 80% of the time and the van 20%. My transportation cost is about what it was before.
    Maybe I’m just a quick learner, but Americans will, ultimately, be forced to change their habits. For the sake of the planet, I hope fuel costs stay high.

  40. Scott

    If you’re going to start calculating what the US economy spends on crude oil,
    you need to use better numbers.

    The US consumes 20 million bpd only if you count refinery gain and other stuff.
    We actually buy about 15
    to run through refineries. You don’t pay for
    the refinery gain in volume. It’s part of the crack
    spread. It also indicates that we’re a lot better at getting more
    out of a barrel
    of crude
    than in
    the past.

    Second, WTI prices overstate what refineries really pay for crude. WTI is
    a light sweet crude. Heavy sour crudes trade at a discount to WTI. Increasingly
    US refineries process heavier sour crudes. Back off your price by about $10
    a barrel and you’re in the ballpark.

    Heavy sour crudes used in the US include North Slope oil, imports from Venezuela,
    and Mexican Mayan crude among others.

    If WTI averages $90 for all of 2007 (it won’t), we would spend about 3.3%
    of GDP on crude oil, up from about 2.6% in 2006. That’s a change of 0.7%, half
    of what it was in ’73 and ’80.

  41. JDH

    Excellent points, Scott. There are a number of ways one could do this calculation (e.g., retail vs. wholesale). I was trying to get just a quick rough summary, and I believe that the broad conclusion I was drawing from that exercise is valid.

  42. Aaron Krowne

    While auto companies aren’t as important to the economy as they were in the oil embargo era, now driving itself is more important, since Americans live so far-flung. I’m not sure the net influence is therefore any lower. It has simply moved from producer/employment-centric to consumption-centeric.

  43. Darren

    Hal: “Either there’s nothing to worry about in terms of oil supply, or: the futures market is giving away vast sums of money to anyone who wants it, so you can be rich as Croesus with little risk, and so again there’s not much to worry about personally.”
    In an environment of extreme social unrest, being rich as Croesus from $300 oil is not necessarily a good thing.
    As others have pointed out, it’s hardly risk free to short the market… you can be right too soon. Anyone who’d shorted NASDAQ in January 1999 would have been absolutely right but would unfortunately be bankrupt due to margin calls.
    That being said, I like being long oil at $80, I’m going to have to look at that.

Comments are closed.