What both progressives and conservatives missed

Progressives and conservatives disagree as to what caused the crash of 2008.  Progressives tend to blame the instability of unregulated capitalism, prone to bubbles and crashes.  Conservatives blame moral hazard created by government insurance and/or policies that tried to get more low income people into housing.  But both seem to see the sub-prime crash as the proximate cause of the crisis of late 2008.  I think they are both wrong.

In my view the sub-prime crisis of early 2006 to mid-2008 caused only a small rise in unemployment, and manageable levels of debt default.  Rather the severe crisis was caused by a sharp fall in AD, a monetary policy failure, which began in mid-2008.  Here’s what my theory explains, which the standard progressive and conservative explanations fail to explain:

1.  In late 2008 the crisis spread to the housing market of states that never experienced a housing bubble, like Texas.

2.  In late 2008 the crisis spread to commercial real estate, which was less affected by pro-housing government policies.

3.  The crisis spread to state and local government debt and pensions.

4.  The crisis spread to the sovereign debt of numerous countries, both with and without banking system collapse.

A fall in nominal spending, if severe enough, will cause all of those problems.  Housing bubbles, the GSEs, the CRA, etc, cannot explain why all those problems suddenly developed after late 2008.

I’m depressed by almost all discussions of the current crisis, as they all start with the premise that “it goes without saying” that the Great Recession was triggered by financial crisis.  No, the Great Recession caused the financial crisis, just as the Great Depression of the 1930s caused a very similar pattern of financial crises.  Ditto for the early 1980s, to a much lesser degree.  We didn’t know it at the time, but we now know that NGDP and employment started falling sharply before the severe financial crisis of the fall of 2008.

Only the original sub-prime crisis, as well as a few other housing crises in places like Iceland, were exogenous events not caused by tight money.  Those weren’t even close to being severe enough to have caused our current problems.

A severe drop in NGDP growth will cause a major recession, and it will severely damage balance sheets.  Many people think balance sheet distress causes recession.  They are confusing cause and effect.

[The following was written a few days ago, before my truth about wealth post.  I now have doubts about this distinction, but perhaps it’s still useful.]

In 1929 we were rich, and by 1933 we were much less rich.  We didn’t falsely believe we were rich in 1929, we really were rich.  We became poorer over the next 4 years.

In 2006 we thought we were really rich.  Some of this was a mirage, as Tyler Cowen likes to point out.  But not all.

Between 2006 and 2008 we gradually became more aware of the amount of wealth we actually had in 2006.  We had built unneeded houses.  By mid-2008 we had a fairly accurate understanding of our wealth.  Then between 2008 and 2009 we lost a lot of wealth that we really did have in 2008.  That’s a very different problem.  That’s a smaller example of 1929-33.

In 1929 we were producing lots of things like cars, homes, home appliances, roads, office buildings, factories, food, clothing.  These are exactly the things a rapidly developing country should be producing.  We were rich.  Then for 10 years we stopped producing much of this output.  We became poor.  After the war we started up again, and found we needed lots more of exactly what we were producing 1929; cars, homes, home appliances, roads, offices, factories, food, clothing, etc.  And we became rich again.

Is it possible to have a major “balance sheet recession” or “reallocation recession” without a sharp drop in NGDP growth?  Maybe–but I don’t expect to live long enough to see one.



34 Responses to “What both progressives and conservatives missed”

  1. Gravatar of David Pearson David Pearson
    11. July 2011 at 05:13


    The timing of events in the Great Depression and 2008 crisis are materially different.

    In 1929-1933, we had a crash in NGDP expectations followed, years later, by a financial crisis. Expectations crashed in 1929. The Bank of the United States failure — the first of a major bank — did not occur until 1931. A general run on banking system liabilities did not occur until 1933.

    In 2007-2008, there was no crash in NGDP expectations until almost the end of the period. Before that, Bear Stearns, Fannie and Freddie, and other large shadow banks (SIV’s, Dexia, Hypobank, Indymac, Countrywide) experienced runs on their liabilities. There was a spate of downgrades of AAA-rated securities — supposedly bulletproof except in an NGDP crash environment — before NGDP expectations crashed. Gorton’s work shows that concerns over AAA-rated shadow bank collateral were the proximate cause of the generalized run on shadow bank liabilities that we call “the financial crisis” of 2008.

  2. Gravatar of David Pearson David Pearson
    11. July 2011 at 05:22

    The situation today is analogous: we have seen a spate of downgrades of AAA-rated securities. This time it is the sovereign debt of the European periphery. These securities serve as collateral backing shadow bank liabilities, including those of U.S. money market funds and TBTF bank-written CDS. Meanwhile, the German stock market is near recent highs — no evidence of crashing NGDP expectations on a Europe-wide basis.

    The run on the U.S. shadow banking system occurred when Lehman showed markets the limits of the implicit “Fed Put”. If the debt of the European periphery is similarly seen as no longer guaranteed by the EU/ECB, European banks (I call them shadow banks because of their “universal banking” structure) will experience runs on their liabilities. If this happens, will we look back and say this crisis was caused by a crash in NGDP expectations?

  3. Gravatar of Morgan Warstler Morgan Warstler
    11. July 2011 at 05:49

    Pshaw. My narrative EASILY covers all your bases.

    1. The crisis was caused by Dems lending to bad borrowers. Fannie and Freddie are at the heart of this.

    2. But behind it is FDIC (free bank marketing) + the ability to sell off loans as fast as they were written.

    Without loans being bundled, CDOs, swaps, and the raft of financial innovations don’t exist.

    3. IF you want to find blame the GOP, the answer lies in failing to destroy public employee unions far sooner.

    There can never be “good times” in public employment. Those jobs need to pay no more than, and perhaps slightly less than – without any better benefits than the average US worker receives.

    THEN, when there is a crisis, the fall off in state and local revenue THAT COMES FROM a crash (1 & 2) in the housing market caused by far smaller impact on the overall economy.


    As to commercial, that’s a WIN for me Scott, not you. They got bail out, they were forced to muddle through and EAT losses immediately. A such they are in a stronger position.

    The problem is that 2 is basically justified in the same way that 1 is justified: it makes credit extension easier.

    The answer to 1 and 2 is just simply a higher cost of borrowing. Raise the cost of borrowing, reduce the risk of default, reward capital formation.

  4. Gravatar of Morgan Warstler Morgan Warstler
    11. July 2011 at 05:50

    commercial got NO bailout.

  5. Gravatar of marcus nunes marcus nunes
    11. July 2011 at 05:57

    I´m “intruding”, but think readers may find this complement to your post illustrative of your arguments!

  6. Gravatar of John John
    11. July 2011 at 06:24

    Your theory is very interesting but is there anywhere you address the argument that tight money and contractions in the money supply might be a result of unsound economic conditions rather than the cause of them? How would you address the charge that your theories confuse cause and effect? What does injecting the right amount of green pieces of paper do to create new products an capital goods instead of just bidding up prices?

  7. Gravatar of Morgan Warstler Morgan Warstler
    11. July 2011 at 06:24

    OK, it turns out the economist with Paul Ryan was Cochrane:


    And altho Matty missed it, the most important thing they could have been discussing was Cochrane’s endorsing of futures targeting:


    And perhaps we can mention again his punch line:

    “My CPI-targeting proposal, like a gold standard, is ultimately a way of communicating and enforcing the fiscal commitments needed to stop inflation or deflation in every theory. It takes tax revenue to buy back dollars when fighting inflation. Seeing the fiscal costs of defending the dollar will help force the Congress to address our chaotic fiscal policy before it spills into inflation that the Fed is powerless to control. Fiscal discipline, clearly communicated, is ultimately, is the Big Stick that controls inflation.”

    That was and still is the money graph.

  8. Gravatar of Morgan Warstler Morgan Warstler
    11. July 2011 at 06:34

    And not to exclusively poke at Matty, this is the best post he has written in a long while:


  9. Gravatar of Contemplationist Contemplationist
    11. July 2011 at 06:50

    Morgan your partisanship is tiring. I have no love for the Democratic Party, but the low-income housing bonanza was as much a plan of Bush & Co. as Barney Frank.

  10. Gravatar of Scott Sumner Scott Sumner
    11. July 2011 at 07:06

    David, I don’t agree, I think NGDP expectations were falling well before Lehman (which is the only financial crisis that mattered–Bear Stearns didn’t) We don’t have NGDP futures markets, but TIPS spreads fell fast in the 2 months prior to Lehamn–and the fall in NGDP expectations triggered Lehman.

    You are also wrong about Germany. Their stock market reflects booming exports to China, the domestic economy (NGDP) fell sharply in 2009, even more so than the US. In any currency zone, including the US, some regions will do much better than others.

    The problems in the eurozone are not caused by a current crash in NGDP expectation, but one that occurred in late 2008, and put NGDP on a 10% lower track. They have been suffering fiscal problems ever since, just kicking the can down the road with bailouts.

    Morgan, Yeah, the “Dems” are to blame for Greece and Ireland and Spain and Italy and Portugal.

    Marcus, Thanks for the link. That’s a much more complete story.

    John, Ever since David Hume there is a mountain of evidence that “pieces of green paper” affect real output. Read Friedman and Schwartz’s Monetary History, for example.

    Morgan, Half my collection is vinyl, the other half CDs. CDs are cutting edge for me.

    Contemplationist. I agree.

  11. Gravatar of David Pearson David Pearson
    11. July 2011 at 07:43

    “I think NGDP expectations were falling well before Lehman.”

    Correct, they were. They had not, however, “crashed”. Perhaps it is more accurate to say, “a moderate correction in NGDP expectations led to the financial crisis”.

  12. Gravatar of John John
    11. July 2011 at 07:45

    I wasn’t disputing that monetary interventions have real effects, what seems doubtful is that any of the effects of “loose” money are good. Also, there were larger deflations before the Great Depression or Great Recession but none of the resulting downturns deserved the Great title.

  13. Gravatar of David Pearson David Pearson
    11. July 2011 at 08:00


    One thing I wonder: would your view of the Great Depression be the same if localized bank runs had begun before the ’29 stock market crash? What if the Bank of the United States had failed in March of ’29, the culmination of a series of these local runs.

    The above is an accurate analogy of the events of 2007-2008. The ABCP market crashed in September 2007, runs on SIV’s began, several financial institutions of significant size failed, unprecedented downgrades of AAA-rated collateral occurred, a major bank failed (Bear), and then, six months later, we had a stock market crash.

  14. Gravatar of Gregor Bush Gregor Bush
    11. July 2011 at 08:00

    Nothing drives me crazier than someone telling me that a weak recovery is “natural” becasue we’re in a balnce sheet recession. They say it as if balance sheets are exogenous to nominal spending and monetray policy.

    On an unrelated note, I was surprised to see that you agreed with Arnold Kling that stock RETURNS cannot exceed GDP growth without either the P/E ratio rising or earnings to GDP rising. I’m fairly sure he’s wrong for reasons that I and a number of other commenters pointed out on his blog. Care to do a post on this?

  15. Gravatar of David Pearson David Pearson
    11. July 2011 at 08:04

    I forgot to add two of the biggest financial institutions in the U.S. — Fannie and Freddie — failed before the stock market crash as well. While they had no clearing function, this would be analogous to the failure of several major banks in the summer of 1929.

  16. Gravatar of SHocking SHocking
    11. July 2011 at 08:24

    OR you could consider Brad Delong’s Keynesian synthesis that the collapse in the value of supposedly safe assets (the AAA MBS) led to a reevaluation of the quality of a broad spectrum of savings vehicles (including state, corporate, and sovereign bonds), which required the massive exchange of safe assets for cash facilitated by the US treasury, which would then be rolled into worthwhile and much-needed infrastructure projects. Seems like that solves both the AD mess and the safe-asset problem, which isn’t so much an issue with across the board NGDP decline as it is a collapse of specific risk-bearing asset classes. Although I mostly agree with the monetarist narrative, I think that it’s pretty oblivious to some of these severe financial market symmetry problems that emerged.

  17. Gravatar of thruth thruth
    11. July 2011 at 08:54

    Re Causation: I’m sympathetic to David Pearson’s focus on financial institutions, but I don’t think timing needs to be tied to failure of institutions per se. My barometer of the stress in the financial system is spreads on short-term commercial paper (which I know Scott doesn’t like). These were highly elevated from 2007 on. Those higher spreads were symptomatic of the failure of short-term money markets (e.g. repo transactions that could no longer be made because the collateral backing them is no longer considered safe) which, I suppose could be thought of as either (i) the destruction (privately created) money or (b) a run to (conventional) money. The Fed’s discount window activity during the early period was nowhere near enough to make up for those lost money market transactions, as evidenced by the failure to restore financial system spreads to normal levels, and hence we had a self-reinforcing monetary contraction. (The more the Fed got behind, the more the balance sheets of the financial sector deteriorated, the more private money got destroyed). Ultimately, I think it was the combination of TARP, the various guarantees and the various Fed actions (specifically, QE1 and TALF) that arrested the destructive process.

    I’m still sympathetic to Scott’s view that if we had a Fed that targeted NGDP expectations maybe none of this would have happened, but we don’t, so it did.

  18. Gravatar of marcus nunes marcus nunes
    11. July 2011 at 08:55

    Rambles on and on but fails to recognize that today´s fiscal problems result from “reneging” on MP 3 years ago!

  19. Gravatar of Benjamin Cole Benjamin Cole
    11. July 2011 at 09:04

    People: Just listen to Scott Sumner. He is the best economic commentator going.

  20. Gravatar of Richard A. Richard A.
    11. July 2011 at 09:14

    We have been in a nominal GDP deficient recession. A true “balance sheet recession” or “reallocation recession” would cause a loss in real but NOT nominal GDP. Since the FED controls GDPn, they are the ones who caused the Great Recession.

  21. Gravatar of David Pearson David Pearson
    11. July 2011 at 09:22

    If holding constant NGDP expectations eliminated systemic risk, then we would have no need for macroprudential policies. Financial institution failures would re-distribute wealth but cause no deadweight economic loss. In fact, the whole concept of “systemic risk” would be a fallacy: all risk at that level would result solely from Fed policy errors. Regulators would adopt as their sole focus safeguarding investors against fraud, but impose no limits on financial institution leverage or illiquidity.

    If policymakers concluded the above, would we have less severe financial crises, or more?

  22. Gravatar of Morgan Warstler Morgan Warstler
    11. July 2011 at 10:15

    David, I thought that level targeting on NGDP was supposed to make it impossible for the Fed to make policy errors. They don’t get to have any choice on what actions they take, the trend is the trend, and immediate course correction to trend is always the choice.


    Scott, all of Europe, especially the PIGS have been generally ruled by one-marshmallow-eaters.

    That many northern states were becoming conservative before the crisis is ALMOST as interesting the way Mundell is forcing the southern states to become conservative as well.

    C, I blame Greenspan more than Bush for the housing situation. Bush I blame for growth of public employee unions.

  23. Gravatar of Contemplationist Contemplationist
    11. July 2011 at 10:30

    Morgan you are blind, deaf and dumb if you ignore Bush’s role in the housing situation.

    You can start with this:


    and this:

  24. Gravatar of Morgan Warstler Morgan Warstler
    11. July 2011 at 13:44

    C, Steve Sailor is a horrible racist bigot. And I’m 20/20, hyper-tuned, and get my point across if you take time to read it.

    Look, as I have said over and over, I get that after 2000, Greenspan and Bush set about to make sure that Bush got re-elected… basically, offset the Internet Crash.

    I don’t care about that.

    THAT IS WHAT THE FED IS FOR to help make sure that Republicans win and Democrats lose. If you are going to put a friggin Central bank into play it MOST CERTAINLY will serve the interests of owners of hard assets – that’s what monetary policy is – if you have to have it – your goal is to make sure that people who have stacks of capital value don’t have it evaporated away.

    What I care about is that after Bush got elected in 2004, he didn’t get far more aggressive about cutting back on F&F. But to be fair, he did try and stop the most egregious shit coming out of Frank and Company.

    And what I do care about is after 9/11, Bush let the government worker rolls explode and let their union dues recycle into Dem party.

    It isn’t enough to be libertarian and say pox on both their houses, to be a real libertarian, you’ve got to be willing to hack the system and short circuit BY ANY MEANS NECESSARY the growth of “free shit for voting.”

    The sad but true singularly most effective strategy of those that prefer liberty and property rights IN TODAY’S SYSTEM, is make sure ALL the money gets spent on the folks who have the $ (defense, agriculture, pharma, etc) so that there can be no handouts to those that don’t have any when Dems are finally in power.

    It is the only viable way to make sure that re-distributed wealth doesn’t happen. It thankfully disheartens generations of Dem voters. It isn’t nice, it isn’t pretty – but since 1980, its the only thing that has started to roll back the tide since 1913.

  25. Gravatar of TheLobsterMan TheLobsterMan
    11. July 2011 at 15:52

    Saying “AD fell” is no more explanatory than “the gods are angry.” What caused AD to fall? Changes in preferences? A fall in income? And if we lost real wealth between 2008 and 2009, where did it go?

  26. Gravatar of Contemplationist Contemplationist
    11. July 2011 at 18:24

    Ah nice to know you don’t care about the truth, Morgan. And “racist” is a slur and abuse. Would you care to dispute Bush’s own speech(es)? I thought not. Good day to you, sir

  27. Gravatar of Lorenzo from Oz Lorenzo from Oz
    11. July 2011 at 18:24

    Morgan: Steve Sailor is a horrible racist bigot. None of which is remotely germane to whether he has a point about Dubya and housing. (Which he does.) Even if we agree the Dems were more guilty (given that Dubya did make some attempt to reign in Fannie and Freddie), it gets us nowhere. The problem is bad ideas: the way to really win is to get people who had mistaken ideas to change them. You bang on endlessly about Scott pitching to conservatives. But they are one part of one Party: the way to really win is to be able to make pitches to as many people as possible in both Parties.

  28. Gravatar of Lorenzo from Oz Lorenzo from Oz
    11. July 2011 at 18:25

    I was quoting with “Steve Sailor is a horrible racist bigot”, but the italics disappeared.

  29. Gravatar of Morgan Warstler Morgan Warstler
    12. July 2011 at 00:14

    Steve Sailor is a racist.

    If you cant find another site online to link what is a reasonable argument – it means you don’t spend enough time at non-racist web sites.

    C, re-read that.

    And Lorenzo, I gave an exact response on Bush, I said WHY Sailor had no point; which is different for 2000-2004 and then 2004-2008. And that means morally allowing for F&F before 2004, not after.

    I’m not trying pitch as many people as possible, I’m trying to subvert the will of the masses where they would be whipped up into a frenzy to the benefit of Dem leaders.

    The masses are not a legitimate thing. When the masses are removed from the picture and policy is built for those who save, invent, trade, etc. things are fine.

    I know you want to pretend there is “one party” but it is simply not true. We can best best be described using bi-polar theory A (top 1%) and B (90-99%) compete, and C (1-89%) plays for its interest.

    And morally, I’m concerned with all of those private employees who at some point in their earning lives will touch the top 90-99%…

    That’s a huge swath of “Likely Voters” almost all of whom routinely vote against the masses.

  30. Gravatar of ssumner ssumner
    12. July 2011 at 09:06

    David, you said;

    “Correct, they were. They had not, however, “crashed”. Perhaps it is more accurate to say, “a moderate correction in NGDP expectations led to the financial crisis”.’

    That’s certainly very plausible, given that heavy losses in subprime lending had already weakend the banking sytem in late 2007 and early 2008. But don’t forget that “the” financial crisis was an ongoing affair. It got worse in October and November as NGDP expectations fell sharply. The worst of the crisis (according to the IMF) was in March and April 2009, which is precisely when NGDP expectations were weakest. As NGDP expectations gradually improved, estimated losses to the US banking system became much smaller.

    John, I agree that loose money is not helpful. I prefer a stable predictable monetary policy, which allows slow but steady growth in NGDP.

    Even in the 1800s, the economy did much better in years with inflation, than years with deflation (like 1873-75, and 1893-96.)

    David, Regarding your second comment, none of the factors you cite had a major effect on the macro economy. Unemployment was reasonably low in mid-2008, and expected to stay low in 2009.

    As far as a bank panic in 1929, it might have increased currency demand, and made the depression worse.

    Gregor, I read that as a typo, in the rest of the post he clearly indicated stock prices, not returns. But that’s a good point. I didn’t notice that.

    David, I thought Fannie and Freddie were bailed out. Did they actually fail?

    Shocking, Undoubtedly some of both. There was some re-evaluation, but the quality of many assets really did get much worse.

    thruth, I don’t doubt that the financial system was a part of how things actually played out. Let me put it this way; monetary contraction (falling NGDP) was a necessary and sufficient condition for a severe recession. Financial distress was neither.

    Marcus, yes, many people make that mistake.

    Thanks Benjamin.

    RichardA, I agree.

    more to come . . .

  31. Gravatar of ssumner ssumner
    12. July 2011 at 09:14

    David, You asked;

    “If holding constant NGDP expectations eliminated systemic risk, then we would have no need for macroprudential policies. Financial institution failures would re-distribute wealth but cause no deadweight economic loss. In fact, the whole concept of “systemic risk” would be a fallacy: all risk at that level would result solely from Fed policy errors. Regulators would adopt as their sole focus safeguarding investors against fraud, but impose no limits on financial institution leverage or illiquidity.
    If policymakers concluded the above, would we have less severe financial crises, or more?”

    I would think much less, as without deposit insurance and TBTF, banks would take far fewer risks, just as they took far fewer risks in the 1920s.

    Morgan, That’s an interesting defense of Mundell, I wonder how he’d defend the euro.

    The Lobsterman, Bad monetary policy caused AD to fall, it’s the Fed’s job to stabilize AD. Suppose the helmsman fell asleep at 2:00am. The ship goes of course. Who’s fault is it–the wind, or the sleeping sailor? I say the sleeping sailor.

  32. Gravatar of TheLobsterMan TheLobsterMan
    12. July 2011 at 17:19

    It’s not enough to say bad monetary policy caused AD to fall. What effect of bad monetary policy? I assume you mean that wages were too sticky and failed to adjust downwards because there was insufficient inflation. But that’s shifting the question from “What caused AD to fall?” to “What caused equilibrium wages to change so much and so quickly?”

  33. Gravatar of Greg Greg
    13. July 2011 at 08:06

    Why can’t we assign equal responsibility to both causes?

    Consider this scenario:

    Because of sticky prices in the face of monetary tightening, the economy was headed for a correction before the housing bubble popped. Indications were extremely subtle and went undetected by the commentariat at large. It just so happened that the tight money policies started taking effect at roughly the same time as the entire subprime mortgage security trade found itself undercollateralized.

    How is it not reasonable to guess that one enhanced the other? Sounds like a perfect (and perfectly plausible) storm to me.

  34. Gravatar of ssumner ssumner
    13. July 2011 at 11:50

    Lobsterman, I think you are confusing real and nominal wages. Wages relative to NGDP did not need to change, but when NGDP fell sharply below trend, obviously nominal wages needed to fall. Example: Suppose the price level fell in half. Even if there was no problem in the labor market, nominal wages would have to fall in half to keeep real wages unchanged.

    Greg, Yes, the factors interact.

Leave a Reply