How important was the financial panic as a cause of the Great Recession?

Question

The collapse of the housing bubble was certainly a primary cause of the Great Recession.  The unwinding of the bubble (1) depressed aggregate demand through its adverse effects on consumer wealth and residential construction, and (2) triggered a financial panic, including runs on wholesale funding and indiscriminate fire sales of even non-mortgage credit.  I have argued that the panic, which choked off credit supply and stoked uncertainty and risk aversion, was the more important of these two channels of effect on the real economy.  Paul Krugman has argued the opposite; he says that while the financial panic may have made the descent into recession steeper, the overall depth of the recession would have been more or less the same if there had been no panic  (See below for relevant links.)

The key evidence for the panic having had a central role in the severity of the recession is timing.  The recession got much worse when the crisis escalated in the fall of 2008, and the recovery began shortly after the panic was contained in the spring. Paul acknowledges the timing but says that the recession would ultimately have gotten as bad, just less quickly if not for the panic. Here I offer a few comments in response.  See my paper and blog posts for details.

(1)    It’s not obvious that restoring equilibrium in the housing market required a recession as large as we saw. Indeed, for the first two years of the housing bust, starting in early 2006, house prices and construction fell significantly without much evident damage to aggregate output. A more gradual unwind of housing prices and construction might have been better absorbed by the economy and by the policy response.  In particular, if the panic had not accelerated the downturn, the Fed might not have faced the zero lower bound on interest rates so early or for so long.
(2)    On the magnitude of the effects, in August 2008 the Fed staff did “worst-case” forecasts that assumed house prices would fall by as much (it turned out) as they actually did. These conditional forecasts incorporated the best available estimates of wealth effects, impacts of changes in residential construction, etc., yet they radically underestimated the level of unemployment that would actually occur. So the direct effects of the housing bust appear quantitatively too small based on the best available models.
(3)    The international recession was deep and highly synchronized.  Trade collapsed.  It’s hard to see that as the result only of a U.S. housing bust. Given the global nature of finance, it’s easy to explain in terms of panic.
(4)    It’s true, as Paul emphasizes, that even though the panic was over by 2009 the recovery was slow.  Financial stress on the part of homeowners certainly explains some of that.  But, as Paul has noted, other factors delayed recovery, including the zero lower bound on interest rates, an attenuated fiscal response, and slow adjustment of wages.  These factors are relevant whatever the source of the recession.  Krugman in recent writings has also emphasized hysteresis effects on productivity growth.  A panic-induced credit crunch could have had such effects by affecting business startups, venture capital, R and D spending, and capital investment.

Paul and I agree that stopping the panic was important to avoid risking further damage to the economy.  We do disagree on how much of the damage that actually did occur was the result of the panic.

What do readers think?

References

Krugman original post: https://www.nytimes.com/2018/09/12/opinion/botching-the-great-recession.html
Bernanke BPEA paper: https://www.brookings.edu/wp-content/uploads/2018/09/BPEA_Fall2018_The-real-effects-of-the-financial-crisis.pdf
Krugman response to Bernanke’s BPEA paper: https://www.nytimes.com/2018/09/14/opinion/the-credit-crunch-and-the-great-recession-wonkish.html
Bernanke response to Krugman: https://www.brookings.edu/blog/ben-bernanke/2018/09/21/the-housing-bubble-the-credit-crunch-and-the-great-recession-reply-to-paul-krugman/
Krugman’s response to Bernanke’s response: https://www.nytimes.com/2018/09/22/opinion/steeper-versus-deeper-wonkish.html
Krugman post on hysteresis: https://www.nytimes.com/2018/09/30/opinion/the-economic-future-isnt-what-it-used-to-be-wonkish.html

Answer 1

I don't know.  The timing of escalation raises another possibility for the financial crisis and great recession:

FACEBOOK.  This is clearly a real business cycle shock.  In late 2006, everybody in the world over 13 could get a facebook account.  Consequently, usage accelerated dramatically in 2007 and 2008.   Everybody became less productive at work, because everybody started checking facebook several times a day at work instead of doing their job.  

https://en.wikipedia.org/wiki/History_of_Facebook

This was compounded by the release of the iPhone in 2007 which let everybody check facebook all the time, even if facebook was blocked on their work computer.  

While I'm joking, I actually think facebook did and does have real, yet perhaps difficult to measure effects on productivity.

Answer 2

In Krugman's last post, he ends by asking what might have caused "massive decline in GDP from what we used to expect, a full decade after the financial crisis"?. My explanation would point to labor share. The 2008 crisis was the only recession since the 1950's that was not proceeded with a rise in labor share. So going into the recession, labor share had not buffered aggregate demand. Then after the 2008 crisis, labor share still has not rebounded unlike past post-recession periods.
The drop in labor share is holding down capacity utilization, which is holding down inflationary pressures... which in turn holds down pressures to raise interest rates. So the Fed rate has stayed low while liquidity among capital has increased and liquidity among labor has not. But should interest rates stay low because firms are increasing their profits from not sharing income in labor's wages?
It seems interest rates are rising now because capital has become strong.  But even while capital is strong, labor income is still weak. Labor is generally following behind the economy. The Federal Reserve seems to have little power to increase the liquidity among labor, specifically wages.
I would say that the depth and breadth of the crisis was a function, in large part, of the drop in labor share and the inaction on the part of the government and Federal Reserve to reverse the drop.

Answer 3

I have to disagree with both of you Sir. I think the real state bubble is only the peak of the iceberg and not the real reasons that exist beneath it. I agree that the real state bubble meant the straw that breaks the camel's back and that the credit crunch was anticipated or even born for this main reason but I'm sure that the period of banking optimism that repeats again and again on the peak of a positive cycle is a normal happening everytime the economy (or even the positive cycle) reach a high peak of growth. This growth tend to end everytime in an importer country and I think that this time a lot of people were dreaming about get a job and own a house due to economic optimism. I know the banking system failed giving credit without control and a lot of people were not going to pay the debt but in my opinion at least half of them would have been able to pay the bank debt if the economic growth had kept being continuous. Anyway the credit extended to the whole economy was going to be unpaid by the companies and the workers losing their jobs due to the unwavering recession that happens after the maximun level of a positive cycle.

I think that the spread of the crisis was due to the synchronization of the business cycles among countries due to economic imports and exports that make grow an economy if the economic allies increase their demand in the country besides the banking bankrupt.

We can't deny that the negative economic cycles repeats themselves again and again on the capitalism system. In the US the crisis came back again when the country changed from an exporter country into a importer country. This is due to a fact that I know and the next economic crisis is coming soon. I have sensitive information about negative business cycles and the main reason that leads to them. We can find false reasons in every crisis that are true but not the main problem there.

I have developed a theory very simple that depicts the growth and recession of every importer country and why the underdeveloped countries can't grow. I'm looking for the right people to share it with because it can anticipate a negative cycle. Contact me via e-mail if you are interested Mister Bernanke. I have developed as well a system called Sustainable Capitalism that allows a permanent growth and full employment.

I'm sorry if my English is not perfect I was born in Spain though I'm American.

Answer 4

One picture I have always found very illuminating is this one:

<a href="https://www.evernote.com/l/AAGtqvEDKrFJNonNJYkATlbQsFBuifaDdnQB/image.png">https://www.evernote.com/l/AAGtqvEDKrFJNonNJYkATlbQsFBuifaDdnQB/image.png</a>

<iframe><img src="https://www.evernote.com/l/AAGtqvEDKrFJNonNJYkATlbQsFBuifaDdnQB/image.png" /></iframe>

<https://www.evernote.com/l/AAGtqvEDKrFJNonNJYkATlbQsFBuifaDdnQB/image.png>

Real exports, real private nonresidential fixed investment, real residential fixed investment, real government purchases, all as deviations from their 2007:IV shares of real potential GDP.

From 2005 through the end of 2007 the housing bubble breaks—and real housing investment relative to potential real GDP falls by 3.3%-points of real potential GDP. Yet there is no recession. Expenditure is smoothly switched from residential investment to exports and non-residential investment. Consumption is not noticeably weak in spite of the impact of diminished housing wealth on households.\

Thus my belief that if the financial crisis had been managed—if the Bagehot Rule had been followed, and if there had been authorities to lend freely at a penalty rate on collateral that was good in normal times—and if 2008 had passed without a crash, then our proves would have been over. It was not the case that the economy in November 2008 "needed a recession" as John Cochrane liked to claim, "because people pounding nails in Nevada needed to find something else to do". The expenditure-switching had already happened. All that needed to be done was to keep demand for safe assets from exploding—and that is what lending freely at a penalty rate on collateral good in normal times is supposed to do.

And then, of course, in late 2008 things go to **** in a handbasket. And fiscal support is inadequate by orders of magnitude. And then after 2008 nothing is done to restart financial intermediation in the housing sector at its normal pace. And after 2010 we see fiscal shock after fiscal shock after fiscal shock...

And so here we are, 10% poorer than we thought a decade ago we would be now. And with lots of scarring  on lots of lives and organizations...

Brad DeLong

Answer 5

You have the better of this argument with Krugman. The timing arguments are strong evidence that the financial crisis "proper" (the systemic run or panic) bore primary responsibility for the greatness of the Great Recession.

For example, overlay the CDS-bond basis (the best measure of "unduly" elevated bond yields) on top of the monthly change in U.S. private employment during the relevant period and they fit almost like a glove in terms of timing and acuity.

I also think Japan's experience in the 1990s provides unexpected support for "panic primacy." Japan's massive bubble burst in 1990-91 but it didn't have a macro contraction until late 1997/early 1998, coincident with the broader Asian financial crisis (which was obviously orthogonal to Japan's bubble). Japan had a major run on its financial sector's short-term debt at that time. In fact, under Romer & Romer's new measure of financial distress, Japan's panic in 1997-98 was the single biggest financial disruption in any of the 24 OECD countries they examined from 1967 (when their measure starts) through the eve of the 2007-2008 crisis. (See here: https://pubs.aeaweb.org/doi/pdfplus/10.1257/aer.20150320) This shadow banking panic and its connection to Japan's macroeconomic woes has been underappreciated.

P.S. I made a poor man's version of your timing argument in chapter 4 of my book on financial regulation, where I described a "panic-induced financing crunch" as the key driver of the Great Recession: https://www.press.uchicago.edu/ucp/books/book/chicago/M/bo22438821.html

Answer 6

Does it follow from the panic-induced side of the argument that had the Fed front-run the Lehman collapse whereby possibly averting the mass panic, the recession would have been better contained?

As demonstrated by the severe LIBOR/OIS spread spiking in Sept 2008, the wholesale dollar funding market completely froze amid the Lehman bankruptcy.  That said, this spread showed first sign of distress in the latter part of 2007, when it reached 100bps (vs ~10bps prior).  While the Fed was still rather concerned with inflation at the time, as a thought experiment, what would have happened had it started QE a year sooner?