Hamilton went back to 2003, when crude oil was around $30 a gallon and forecast what an oil shock like the one we experienced in 2007-08 (when oil peaked around $140) would do to GDP. He graphed the result through the end of 2008 and, lo and behold, it was damn close to actual GDP. As though there were no such thing as a collaterized debt obgligation in the first place! Here's the graph (the orange dotted line is Hamilton's projection given oil prices; the black line is actual GDP):
Perhaps you'll join me in thinking: Huh? Are we really to
believe that this whole thing was caused by oil shocks? I mean, it
certainly makes you appreciate the mess Detroit is in, but really. How
anti-climactic. It makes this crisis seem so ... 1970s. What about real estate, subprime mortgages and defaults? Hamilton says the housing industry had been tightening up long before the recession -- "subtracting 0.94% from the average annual GDP growth rate over 2006:Q4-2007:Q3." And housing is factored into Hamilton's analysis. It was just one of a handful of multipliers that always turn down during oil shocks.
The Real Time Economics Blog at WSJ moves the theory forward with a pretty interesting bit of revisionist history. The grand retelling goes something like this. Cheap gasoline from the 1990s into this decade encouraged families to set up their homes farther from the cities where they worked. But as the price of gas began to increase, it put a big strain of these families' commutes. With gas rising from $2 to $4, the price of these long drives doubled, straining those families' most expensive payments, namely: mortgages. When families realized they could not afford their exurban commutes, they sold their homes for a big loss. Voila: Their mortgage crisis became a bank crisis and the rest is our living history.
Hamilton concludes.
Eventually, the declines in income and house prices set mortgage delinquency rates beyond a threshold at which the overall solvency of the financial system itself came to be questioned, and the modest recession of 2007:Q4-2008:Q3 turned into a ferocious downturn in 2008:Q4.My head's still spinning a bit, but it's interesting to think about the political consequences of a report like this being mainstreamed. If the idea somehow stuck that an oil shock was responsible for the financial crisis, it could be a significant catalyzer for the push toward energy reform. Today we're seeing a great national movement to change Wall Street because the general consensus is that Wall Street caused this crisis. Whether Hamilton's theory is wacko or brilliant, just imagine what a national movement to revolutionize America's energy consumption would look like. What if we had oil parties instead of tea parties, demanding more government investment in alternative fuels and subsidies for green technologies. That would really be something.
The problem with that last bit of fantasy is that green energy would cost far more than even last years oil. What would *that* do to the economy?
Great post but I have to agree with the comment above. Hamilton's findings would seem to indicate that the economy can't function well at higher relative energy prices. If that's the case then the road to an economy that uses fewer fossil fuels may be longer and more torturous than expected.
Tom (and mgoodfel) - You're both right. If we switched over to green energy right now, the economy would stop in its tracks. My suggestion is that the country needs more of an impetus to shift a broader buffet of energy sources so that we won't be shocked into a recession every time China decides to expand. And the belief that oil caused this crisis -- whether or not that's true -- would, I think, create just that impetus. I'm not going to be able to make this point better than Fareed Zakaria, so I'll just link to his cover story from two weeks back here:
http://www.fareedzakaria.com/articles/articles.html
and quote this little sample:
"The debate about oil vs. natural gas vs. biofuels vs. alternative energy is wholly unrealistic. If we are going to sustain and support this kind of population and economic growth, we'll need everything ... The key is to free ourselves at every level of the energy chain. That means, first of all, finding sources of energy that are abundant, cheap and don't have hidden costs—environmental, social or military."
Oil prices may have been the straw that broke the camel's back but lets not forget why there is still a debate on why oil prices went up as far as they did. Some say it was supply demand fundamentals, a popular blame was the declining value of the dollar, but there was also curious hings occurring on wall street at the same time. I found it suspicious how Goldman Sachs analysts releasing predictions of $200 crude could on the same day break intraday change records and shut down NYMEX because of too much volatility.
Am I the only one that thinks the two curves don't look all that much alike at all?
No you are not, I had the same thought both the shape of the curve and the time lag. $4.00 gasoline was not the reason people could not make payments on NINJA loans. While I agree we must break our dependence on fossil fuel nuclear, being the only currently practical one in my thinking, $4.00 a gallon gasoline did not "cause" the financial meltdown.
>> With gas rising from $2 to $4, the price of these long drives doubled, straining those families' most expensive payments, namely: mortgages. When families realized they could not afford their exurban commutes, they sold their homes for a big loss. Voila: Their mortgage crisis became a bank crisis and the rest is our living history.
==
I don't know of any families in my area who sold out at big loss and moved in-town due to gas prices. And, I live in metro Atlanta - a strong example of the exurban lifestyle. I'm sure there may have been some, but I don't see this virtual "stopper pulled from the bathtub" effect that is noted. I'd have to see the data on housing "abandonments" and some associated correlation analysis before I bought this story. I'm not saying it can't happen. I just don't think it's happened yet. When gas prices go to $8 or $10 a gallon, perhaps.
The danger today is that we might find an energy resource that might suffice for the next couple of decades. We might then come up with a new way of living that would be appropriate to this new resource, and suffer similar consequences when it becomes scarce inevitably.
We will need to think, somewhat along the lines of what Zakaria suggests - in terms of a portfolio of energy resources, some of might be expensive and only available in the longer term. However, society should continue to invest in such future resources, even when other affordable resources are available. Just as we consider alternative energy resources, we must also build portfolios of options in other sectors - an appropriate mix of public and private transportation for example.
The wrenching situation we are finding ourselves in now, is due to the fact, that we need to make substantial changes in order to adjust to a new energy reality, whether that means moving to smaller houses, or reducing commutes.
Further, as Zakaria also suggests, let us design the future options without hiding costs or considering impacts on the whole system.
A desperate attempt to get Wall street off the hook. A more plausible theory for me would be this: when we stop sending billions of dollars annually to oil despots and start investing the money in our own real economy, people will have good-paying jobs and will be able to afford their mortgages and fuel-efficient vehicles.
Thank you for this quite brilliant and entertaining article! I do agree with the comments here (and consequently with the author's''Huh?' question), despite the miniscule saving rate in the US, I don't think that a gasoline price spike of 2 dollars per gallon could have, by itslef, triggered the epic downturn we have witnessed recently (than again I live in Paris and use a bicycle to commute)...
But turning to oil in an attempt to understand this downturn better does seem reasonable. After all, the price volatility of crude a year ago was a first sign of 'confidence fragility' on the markets, and had its roots far away from supply/demand economics... I remember the OPEC refusing vehemently to increase production output as they insisted there was plenty of oil available, still the barrel got to wear its prettiest dress at $140 only to find itself stripped naked again a few months later due to overflowing storage tanks.
In terms of an 'energy movement' evolving from this, personally I would love to see that happen although it seems rather unlikely. I guess what we need to understand first is that green/new energy is not 'still too expensive' but rather 'old energy' is grossly under-priced due to the market's inappropriate factoring-in of environmental, economic and political risks linked to the rapidly depleting resource that is oil (i.e. another question of rebuilding a model to reflect a changed landscape).
To close, I suggest that, at the heart of any kind of energy movement, would have to lie a change in our lifestyles, but that's a whole different story, isn't it?
HL
I think -- assuming investigation gets this far -- that it will turn out that the oil shock was, in turn, caused by Wall St, specifically by ibank traders who may or may not have been operating from 85 Broad St, NYC. The rumor on the Street last summer -- which I heard several weeks before the banking crisis started -- was that Goldman's traders drove the forward price of oil up and then shorted the hell out of it.
Sorry, it just doesn't add up.
Let's say the commute is 40 miles each way in a 20 mpg car: uses 2 gallons at an increase of $2/g per trip or $8/day for 5 days a week being $40/week and $80/paycheck or $160/month. I just don't see the cost breaking the back of everyone out there -- budgets could adjust in other areas easily to accommodate.
And then when gas dropped back down, wouldn't they be able to afford their payments and the slide come to a screeching halt?
There may be the correlation between oil prices and the financial crisis posited in Hamilton's article (although the evidence presented here is thin, and I agree that the two curves are not convincingly similar).
Even if the correlation exists, and is quite strong, it's almost incredible that the oil prices were the fundamental reason for the implosion of the financial system. At most, the spike in oil prices would have been the proximate cause in the same sense that an over-inflated balloon floating in a room containing a collection of medieval weapons would be pricked by a halberd before it landed on a mace, or any of the other myriad pointy things.
The fundamental cause of the financial crises was the pyramiding of increasingly leveraged financial debt instruments that were based in the idea that the housing market would be one of continually-increasing prices. This was absolute nonsense, as is the idea that any economic sector can increase indefinitely, but the financial operators were too busy making huge amounts of money in transaction fees to care whether or not the bubble would eventually burst. Which it was always going to.
Interestingly, and as something of an aside, I'm perplexed with the popular notion that the failure of the financial markets was their inability to identify, quantify, and manage risk. I believe that it's clear that the people pulling the real wealth out of the system didn't care about the risk of the overall system collapsing; their risk was that they wouldn't be able to make enough money before it did. I wonder of anyone has looked into where the money pulled of the economy by the financiers really went. Who really got rich, phenomenally and obscenely rich, from all of this?
In any case, Hamilton may well have identified a correlation, even the proximate cause of the collapse. But it doesn't matter. Something was always going to be the prick that burst the bubble.
Correlation is not causality. The current crisis was triggered when rising interest rates on adjustable rate mortgages made those mortgages unaffordable.
Banks, or more properly suckered bond investors, made too many loans to people who were barely able to afford the original monthly payments, much less increased payments. Combined with loans to people who clearly could not repay the original terms, the foreclosure rate skyrocketed.
When the housing market collapsed, the charade of borrowing on debt came tumbling down and the deception and legalized thievery purpetrated by the financial industry was exposed.
Unlike the 1970's crisis, gas supply remained plentiful prior to the crash and the price only doubled.
James Hamilton is probably correct though more likely the oil shock was the "straw the broke the camel's back" is most likely or the "prick the burst the bubble." Oil shocks in the past have had the same affect. The dot-com bubble is a good example, and the recession of the early 1990s with the S&L failures.
This theory is probably not valid. Although there is no more urgent priority than developing renewable energy supplies and incentivizing utilities to switch to the same, the most likely scenario in the circumstances Hamilton sketches would be for the cash-strapped families to sell their gas guzzling SUV and buy a hybrid! We know this actually a happened last year because Priuses were disappearing from dealers faster than they could arrive.
Selling the house would be a last resort. I wonder if Hamilton is married and if he has ever talked to his wife about selling the house. Perhaps if he had, he would realize that for most families this truly would be a last resort.
They will buy a new car readily, but selling the house is a draconian solution and people will do that only if there is no other alternative.
An even more troubling and all too plausible thesis on how oil price volatility may be the cause of the financial crisis is provided at:
http://ideas.wikia.com/wiki/Volatility_in_the_Price_of_Oil_since_Hubbert%27s_Peak_and_Investment_Risk
This wiki was posted January this year. The quantitative analysis in this essay appears to illustrate a considerably higher precision of alignment between volatility shocks and of other key economic variables than that provided in the Hamilton model.