How much are gasoline prices weighing on consumers?

On Friday Reuters reported:

Rising gasoline prices beat down U.S. consumer sentiment in early January, overshadowing an improved job outlook and passage of temporary federal tax breaks, a survey released on Friday showed. A year-end surge in gasoline prices ratcheted up consumer inflation expectations to their highest in more than two years, according to the latest data from Thomson Reuters and the University of Michigan. The surveys’ preliminary January reading on the overall consumer sentiment slipped to 72.7, below 74.5 in December. It fell short of a 75.4 reading predicted by economists polled recently by Reuters.

The correlation between consumer sentiment and gasoline prices is reasonably strong. In the graph below, sentiment is plotted in black and the blue line corresponds to the negative of the inflation-adjusted gasoline price, plotted as a negative number in order to emphasize the feature that increases in gasoline prices are usually accompanied by a drop in consumer sentiment.



Consumer sentiment and negative of real gasoline prices, 1978:M1-2011:M1. Black line (scale on left): Reuters/Michigan index of consumer sentiment, from FRED and contemporary news accounts. Blue line (scale on right): negative of the average U.S. retail gasoline price in 2010 dollars per gallon. Gas prices 1978-1989 are U.S. city average retail price of unleaded gasoline from EIA. Data from 1990-2011 are regular gasoline, all formulations, from EIA. Deflated using the 2010:M12 CPI from FRED.
gas_sent_jan_11.gif



I obtained a quick summary of this relation by regressing consumer sentiment in month t (denoted st) on a constant, sentiment the previous month, and the difference between the average real gasoline price in months t and t-1 and what the price had been in month t-2. That 2-month change in the real gasoline price is denoted xt, and is measured in 2010 cents per gallon. Here are the results of that regression with standard errors in parentheses. A 10 cent per gallon increase in gasoline prices shaves a half point from the consumer sentiment index.




sent_eq_jan_11.gif


The average real U.S. gasoline price increased from $2.73/gallon in September to $3.08 so far in January. If you dynamically simulate the above equation, you end up predicting a cumulative effect of rising gasoline prices over the last 3 months of a little more than a 2 point drop in consumer sentiment. In other words, if gas prices had not risen, we might have expected to see consumer sentiment about two points higher than it currently is.

All of which is consistent with the assessment I offered last month. Rising oil prices are not enough to derail the recovery, but they are perhaps starting to have some effects.

StumbleUponLinkedInReddit

21 thoughts on “How much are gasoline prices weighing on consumers?

  1. Matt Young

    I used to think is was the consumer, now I think it is the producer who is more sensitive. Is there an equivalent producer sentiment?

  2. MarkS

    “Rising oil prices are not enough to derail the recovery, but they are perhaps starting to have some effects”.
    What recovery? When 10% of GDP is federal debt and the FED tops it off with another $1.4 Trillion in bank subsidies, I fail to see where real “recovery” is occurring.
    In the short-term, rising oil prices will contract real economic activity in other sectors, as it pumps-up inflationary pressures in food, energy, and transportation. The Current Account will take another hit, and additional pressure will be placed on the dollar.
    In the long run, high oil prices will force increased efficiency and a migration to other energy sources. The ten-thousand dollar question is: Will America have pissed-away so much of its wealth on imported oil before this adjustmant, that it loses control of its domestic industries, and spirals into continuous decline as a nation state?

  3. rjs

    this shouldnt be a surprise at all…according to a Pew survey on Dec 15th, 29% are having trouble affording food, and 48% are finding it difficult to pay for electric and heating bills…so any price increase of another necessity would just beat them down more…
    however, the underclass is not driving consumer sales; retail sales were driven by double digit increases at Tiffany’s, Saks Fifth Ave, Louis Vuitton, Nordstrom and other high end retailers, punctuated by increases of 35% in sales of cadillacs and 29% in porsches, it appears that was all driven by the same top 1% who will be getting the big obama tax cuts . . .
    we need different consumer sentiment indicators for the two classes of consumers, the one that matter & the ones that dont count…

  4. Ed Hanson

    Interesting statistic about Saks Fifth Ave. They are in the process of closing their 8th store nation wide, this time its Denver. So much for booming sales.

  5. Mr. S

    RJS’s notion of a multi-tier consumer sentiment indicator segregated by income is interesting…. I’d really like to see that actually. I imagine significant pressure to get rid of it would ensue though because it would fuel an already class-war currently underway.

  6. Ken Houghton

    Mr. S., that’s the best joke of the day.
    JDH – I think you ran the wrong study. My gasoline expenses went up about $10 from Nov to Dec. Even granting there are a lot of people who drive much more than I do, actual cost of gas probably only went up $30-50 for the average two-car household.
    What went up a lot was the heating bill. (Indicatively, 4-6x the range quoted above.)
    You’re using gasoline as a proxy for fuel costs. I can cut back my driving some; I can’t voluntarily freeze my kids.
    Gasoline is more discretionary, so it may affect “sentiment” more, but the inability to fill the tank is causally related to needing to reallocate to home heating.

  7. westslope

    Thanks JDH. If you regress first-differenced sentiment on first-differenced gasoline prices, does the relationship still hold?

    I wonder if any other economies demonstrate the same close relationship between petrol prices and consumer sentiment? I would guess that there is a similar relationship between food prices and consumer sentiment in many countries.

  8. Nemesis

    “Rising oil prices are not enough to derail the recovery, but they are perhaps starting to have some effects.”
    Only if one perceives the US economy through the lens of Keynesian aggregate demand, i.e., total gov’t spending/GDP of ~36%, including personal transfers.
    Again, take a look at oil consumption as a pct. of private (less gov’t, including personal transfers which are redirected from primarily the private sector via the gov’t to individuals) real GDP and private final sales, and household debt service/disposable income, and one sees that the US private sector is flat line or even slightly negative.
    Oil consumption/private GDP is well above the private sector recession threshold since the early 1970s.
    There will be no sustained growth of private investment, payroll, and income growth under such conditions, and thus equity prices are grossly overvalued in this context.
    “What recovery? When 10% of GDP is federal debt and the FED tops it off with another $1.4 Trillion in bank subsidies, I fail to see where real “recovery” is occurring.”
    IOW, as I infer that MarkS implies, after oil consumption and debt household service, the private sector is not growing, despite the gov’t having borrowed and spent an equivalent of 30% of private GDP in two years, which will reach 40-45% of private GDP by the end of ’11 and 100% by ’15-’16.
    What happens to real private final sales when the inevitable fiscal contraints occur at recessionary oil and gasoline prices while household debt service for the bottom 80-90% remains high with high labor underutilization?
    And apart from ag exports, US “exports” so many are hoping will drive growth are primarily US supranational firms shipping capital equipment to their foreign subsidiaries and contract producers to produce intermediate and finished goods for the US and world markets, i.e., “trade”.
    As US aggregate and private sector growth again decelerates in ’11, so, too, will the growth of US “exports” to their foreign producers, causing China-Asia and Canada and Australia to slow or eventually contract from bursting real estate and commodities price bubbles.
    Global “trade” will inevitably and “unexpectedly” slow or contract, which is what is implied already by the BDI and Shanghai Index.
    Real retail sales per capita.
    Moreover, as to the point about retail sales, adjusted for gov’t-reported price inflation and population growth, real retail sales are no higher than private payrolls, industrial production, private fixed investment, and the ECRI WLI in the late ’90s.
    And the current levels of payrolls, production, retail sales, and investment are very likely to be a secular/structural no-growth trend line for the next decade (as was the case for Japan since the late ’80s and early ’90s), with per capita growth contracting 0.7-1%/yr.
    The Bernanke Fed flooding the banking system with free reserves to prop up equity and corporate bond prices only ensures that equity valuations are much higher than otherwise would occur at this point in the secular debt-deflationary regime, guaranteeing low, or negative, real total returns to equities 2-3, 5, and 10-20 years hence.
    However, if, as I contend, larger secular and super-secular self-similar rhythms (and here) of the evolving capitalist system are at work, which are related to energy units/time, net energy, energy intensity, demographics, mass-social herding effects, and compound interest, then we are in no way “in control” or “managing” conditions; rather, we are reacting, consciously or otherwise, to the inevitable structurally deterministic forces over which we have no control as they are inexorably entrained in time and space.
    If so, and the Bernanke Fed, banksters, and corporate and political leaders are aware of this, which they surely are, the best they can do is attempt to mitigate the more harmful effects of the forces bearing down on us.
    Unfortunately, there is no historical template to use to navigate a debt-deflationary depression AND peak production, and eventual inexorable depletion of, the dominant energy source AND population overshoot; therefore, the risk is high for harmful unintended consequences at unprecedented scale.

  9. aaron

    Ken, fuel economy is affected by the weather. It declines quite a bit. The amount of driving we do also goes down though, that usually sends prices lower.

  10. Ivars

    What recovery? Just a short relapse before next dip. That is the law of financial anti bubbles ( bursting of bubbles)- timescales are unique for each crisis, but not the overall log-periodic shape of deco-operation process:
    see here by Johanssen, Sornette:
    http://arxiv.org/PS_cache/cond-mat/pdf/9901/9901268v1.pdf
    Good thing is, for some reason, the USA already seems to be in a plato region, so all consequent dips will not bring DJIA below 6500 of March 2009- they will stay close to it or little above in next dip in jan-march 2012, then in next in Sept-Dec 2012, and then in last, in Sep-Dec 2013. After that, some growth may start, but in the meantime, what had happened to GDP?Is there any correlation between GDP and DJIA- then it would be possible to predict when and how severe next recession will be. It seems recession will last from Q4 2011 till Q4 2013, 2 years, but I do not know how to calculate the dip in GDP over this time.
    If average DJIA in 2010 was 11000, 2011 will be 10000, 2012 would be 8000 and in 2013 8500, what would be the GDP drop during these years? Then, given the drop in GDP on top of current unemployment ( or the one after Q3, 2011) consequences to unemployment?
    Any ideas what the equation is, between quarterly average DJIA change and quarterly GDP growth/drop , similar to one between gas prices and consumer index?
    Or may be it already exists, where can I look it up?
    Thank You in advance.

  11. Steven Kopits

    What happens to real private final sales when the inevitable fiscal contraints occur at recessionary oil and gasoline prices while household debt service for the bottom 80-90% remains high with high labor underutilization?
    Exactly.

  12. John Cardillo

    “Rising oil prices are not enough to derail the recovery, but they are perhaps starting to have some effects”.
    The U.S. imports nearly 10 million barrels of oil per day. At this rate every $25 increase in the price of a barrel of oil drains ~$90 Billion from the US Economy per year. Thats almost .5% of GDP. The real impact on GDP is probably a lot higher when you consider that the exported currency is no longer circulating in the U.S. Economy.

  13. Steve

    Still waiting on my parachute, ground is getting real close. This ones gonna hurt! LOL! I agree, we are going to see recession once again, later this year. Still standing by my prediction, no growth until the consumer debt load is reduced by 75%.

  14. Nemesis

    Very good, Jeffrey.
    See the following links about China’s oil consumption, import, and GDP per capita trend rates and why China’s growth mathematically and geopolitically cannot continue:
    http://imperialeconomics.blogspot.com/2010/07/chinas-oil-consumption-and-imports.html
    http://imperialeconomics.blogspot.com/2010/10/chinas-gdp-per-capita-global-gdp-and.html
    http://english.peopledaily.com.cn/90001/90778/7268950.html
    China’s oil consumption (54% of which is now imported) is closing in on the level of US imports. At the trend rate of oil imports and consumption, China will reach parity with the US no later than the late ’10s to early ’20s, with the US and China consuming around two-thirds of total peak petroleum production, leaving the rest of the planet to get by on the other third of production, which is about half of what they consume today.
    This is an imminent global resource war and post-Oil Age global economic collapse scenario, yet most economists are not paying attention, or are not paid to discuss it publicly.
    At the trend rate of depletion of US oil reserves, i.e., declining production, since US peak oil production in 1970 and 1985, per capita US oil production will have fallen 61-62% by ’15 and 66-67% by ’20 (75-80% in CPI terms), which is a 25-33% incremental decline in the next 5-10 years. That oil consumption per capita correlates highly with private GDP per capita, a similar decline in GDP per capita over the coming 5-10 years is quite likely.
    With China growing GDP and oil consumption at 8-10% at peak global oil production, a net liquid fossil fuel energy zero-sum regime has emerged, prohibiting growth in the developed countries while setting China-Asia on a collision course with the structural effects of Peak Oil and with the US imperial military that cannot afford to allow China continue to grow.
    The irony is that imperial US supranational firms have been the primary driver of China’s growth and development since the late ’80s to mid-’90s, with US firms investing tens of billions of dollars in plant and equipment, trade credits, infrastructure, and technology transfer.
    Our firms, desperate for investment opportunities from which to profit in a situation of falling rate of profits conditions in the developed world, have essentially funded China’s rise to compete for the remaining mineral resources of the planet that both the US and China must have, and cannot allow the other to monopolize.
    China’s runaway growth is now a direct national security threat to the US and by extension a global economic security threat, not unlike Japan, Germany, and Italy in the 1930s. China cannot continue to grow, either geopolitically or thermodynamically.
    The China Crash ahead will be simply astounding.

Comments are closed.