Stephen Gjerstad and Vernon L. Smith (GS) take that view.
Most notably, housing expanded rapidly by nearly 60% from 1922 to 1925, leveling out in 1926 and then began its long descent, not bottoming out until 1933. In 1929 new housing expenditure had returned to its 1922 level before any of the remaining expenditure categories had declined more than small temporary amounts…the 60% increase in the rate of new home construction expenditures from 1922-1925 was matched by a 200% increase (from $1B to $3B) in the net flow rate of mortgage credit.
This sound similar to the more recent episode. That is, much of the decline in housing construction was already behind us when the bottom dropped out of the rest of the economy. GS point out that the credit boom in the late 1920s did not raise house prices as it did more recently, because (they argue) in the 1920s the housing supply was more elastic.
Edward Leamer, in Macroeconomic Patterns and Stories, shows that spending on housing and consumer durables generates much of the variation in GDP growth relative to trend. However, in the typical recession, as GS also note, the decline in housing construction coincides more closely with and accounts for a larger share of the decline in GDP.
GS point out that the mortgage credit boom of 2002-2006 was fueled by foreign funds. This is in accord with Ben Bernanke’s famous “savings glut” hypothesis. They also write,
For the combined government-sponsored enterprises (GSEs), mortgage credit outstanding grew 10.9% per year from 1990 to 1997, 15.5% per year from 1997 to 2001, and 9.7% per year from 2001 to 2006. Private mortgage credit outstanding grew only 3.1% per year from 1990 to 1997, 6.1% per year from 1997 to 2001, but then it grew at 15.2% per year from 2001 to 2006.
Quantitatively, this supports those who wish to absolve Freddie Mac and Fannie Mae of blame for the mortgage credit bubble. However, the decline in credit quality at Freddie Mac and Fannie Mae was significant. Had they stuck to their lending standards of the early 1990s, history would have been much different.
GS go on to note,
Ten trillion dollars came off the value of equity in U.S. firms during the technology sector crash between 2000 and 2002, with hardly a dent in bank balance sheets. Output was largely unaffected as well, with only a mild recession in 2001. Similarly, the stock market crash on Oct. 19, 1987 yielded no recession. Only $2.2 trillion came off the value of homes between their peak in the first quarter of 2006 and the third quarter of 2007. In the last year before the crisis, the Federal Reserve undertook significant measures to enhance liquidity, and it also took the extraordinary step of bailing out a large financial institution (Bear Stearns), yet the financial system still buckled under the stress of a relatively small decline in housing asset values.
GS argue that the stock market crash was relatively benign, because investors were betting with their own funds. In the case of the real estate crash, Other People’s Money was being used.
This is important, because many people will want to look at the recent recession in terms of a wealth effect. Household wealth went down, so spending went down, so jobs were lost. As you know, I am a skeptic of the Keynesian focus on spending and jobs. I think of those as symptoms of patterns of sustainable specialization and trade, not as causal variables.
Instead, GS think that in the case of the mortgage meltdown, banks took a big hit. This made them reduce lending, and that reduction in lending produced the steep recession. Perhaps one can tie this to a PSST story. If businesses need bank loans in order to experiment with ideas that will generate new trade patterns, then a reduction in lending will cause problems. I hesitate to endorse such a story, because I do not see signs in the business sector of widespread unsatisfied demand for credit.
GS argue that the economy would have been better off with liquidation of financial institutions than with bailouts.
If all bad loans are written down to market value, the losses wipe out a bank’s capital, and bondholders are forced to take a haircut until liabilities equal assets, then the balance sheet of the bank is clean. When new investors provide new capital for the bank, that capital doesn’t need to be applied to fill in old holes in the balance sheet, and the return on new investment isn’t diluted by claims on it from the previous shareholders. The alternative in which loan losses aren’t fully recognized is much less favorable to new capital investors. For then new capital investments must be applied to filling in the old holes in the balance sheet.
I am skeptical of this. If liquidation works and bailouts fail, then why did we have the Great Depression? Note that Ben Bernanke takes the opposite view–bank lending is reduced by liquidation and better maintained by bailouts. And he has the Great Depression in mind. He did not just arrive at this position from getting beaten up by Henry Paulson–it was his view as an academic.
So, here is where we are. If housing is the primary cause of the Great Depression and the recent great recession, why is housing construction such a small share of GDP, and why is much of the decline in housing construction behind us before the overall slump gets going?
If loss of wealth is the primary cause, then why did the economy not collapse after the stock market decline of 2000-2001? If sluggish lending by bailed-out banks is the primary cause, then why did we have the Great Depression? And why does it appear, at least to me, that the demand for bank lending collapsed more than the supply?
Scott Sumner would tell a monetary contraction story. I would tell a story about the availability of new technologies to increase productivity and a sudden increase in ruthlessness on the part of businesses in applying those technologies. Meanwhile, new opportunities for entrepreneurs are repressed by supply-side factors–New Deal cartelization in the 1930s and increases in non-wage labor costs more recently.
“ruthlessness”?
“GS point out that the mortgage credit boom of 2002-2006 was fueled by foreign funds. This is in accord with Ben Bernanke’s famous “savings glut” hypothesis.”
As Jeffrey Hummel and others acknowledge, this is a version of F. A. Hayek’s macroeconomic mechanism of the boom and bust cycle. See Hayek’s Monetary Nationalism and International Stability and Monetary Theory and the Trade Cycle.
“Scott Sumner would tell a monetary contraction story.”
The bust side of F. A. Hayek’s macroeconomic mechanism of the boom and bust includes a mechanism explaining the most boom ‘secondary deflation’, ie the monetary contraction.
Milton Friedman and Scott Sumner don’t have a monopoly on this part of the causal story.
Hayek has been inconsistent in applying it or in estimating its size in real time at particular points in time, but its been there in his work from almost the beginning.
“If loss of wealth is the primary cause, then why did the economy not collapse after the stock market decline of 2000-2001?”
I’d argue, as you say GS do, that debt was the crucial difference this time around. Obviously the wealth effect is in play either way, but when you had much more debt and the ensuing leverage involved in the housing boom, you get the bank runs and other Diamond-Dybvig instabilities that are involved with the current banking system, which can make the collapse much more precipitous.
“If liquidation works and bailouts fail, then why did we have the Great Depression?”
There are so many other differences between the current situation and the ’30s, that I doubt liquidation would have the same catastrophic effects it is claimed to have had back then. For one, the FDIC was not around back then, so knock-on bank runs at the banks that lent to the liquidated banks are unlikely. Also, a certain number of banks were liquidated this time, it’s not like it didn’t happen, just a lot less. I am similarly skeptical that GS and Bernanke’s fears of diminished lending are relevant.
“If sluggish lending by bailed-out banks is the primary cause, then why did we have the Great Depression?”
This question doesn’t make sense. Obviously liquidated banks were not there to lend anymore, so those pushing the “sluggish lending” theory can claim that it was much worse during the Depression because there was no bailout, which Bernanke averted by bailing out banks and at least letting there be some lending, albeit much reduced from the boom.
“And why does it appear, at least to me, that the demand for bank lending collapsed more than the supply?”
And here is where we get to the crux of the matter. I would tell a story where home-owners used the boom to flip houses and take out HELOCs to inflate their consumption, effectively pulling future spending into the present. When the housing boom did not prove sustainable, they slowly realized that this new wealth was petering out, then finally had to cut back a couple years later, hence the delay. This led to the collapse in demand, by consumers for products and companies for lending, that largely persists to this day.
Regarding technology, this is the elephant in the room that is felt by its absence, ie where are the great new tech businesses that are pumping in revenue? There have only been two big successes for revenue in the last decade, Google with its search ad monopoly and Apple driving and leading the mobile boom, followed by Eastern companies like Samsung and Asus. The fact that the techies have been too dumb to come up with a real revenue model for online tech is what has been missing this decade.