Miron and Rigol on bank failures and output

I only met Bernanke once, as I recall it was in the 1990s.  He presented a paper at Bentley on the role of bank failures in the Great Depression.  I only asked one question: “How much would output have declined if the Fed had successfully stabilized NGDP, but the banking panics had still occurred.”  I don’t recall the exact answer, my sense is that he said output would have declined, but obviously much less than it actually did.  Bernanke certainly thought the monetary channel was very important, and saw his credit channel paper as supplementing the Friedman and Schwartz explanation.

I’ve always been skeptical about the credit channel.  Yes, it certainly matters to some extent, but given the decline in NGDP, and the bad supply-side policies, we got roughly the Depression I would have expected. So I don’t think the bank failures add much, except that they obviously help explain why money was so tight. After all, under the gold standard the hoarding of currency and reserves is even more harmful than under fiat money, as the Fed cannot offset as easily.

In the current recession a policy of NGDP targeting would have prevented a steep rise in unemployment, even if lots of banks had failed.  Of course, if we had been doing NGDP targeting, the number of bank failures would have been only a small fraction of what actually occurred.  Most of the failures were linked to loans to developers than went bad when the economy tanked, not the subprime mortgage mess.

Mark Sadowski sent me a study by Jeffrey Miron and Natalia Rigol that shows that much of Bernanke’s results hinge on a single month’s data, March 1933.  And that’s a very odd month to draw any implications from, as not only was there an extraordinary number of bank failures, but there was also a national bank holiday. So the entire banking system was shut down for much of the month.

We reconsider this issue by reporting regressions that drop March, 1933 from the sample entirely (along with appropriate lags). Columns (5)-(8) of Table 2 show the results. In this specification, the bank and business failure variables still enter negatively, consistent with Bernanke’s hypothesis, but the 7 coefficients are no longer statistically significant. Thus, exclusion of the Bank Holiday does not reverse Bernanke’s results but it weakens them substantially.

.  .  .

To the extent U.S. experience during the Great Depression – and especially the view that bank failures play a significant, independent role during that period – formed the intellectual foundation for Treasury and Fed actions, however, our results suggest at least a hint of caution. If the Great Depression does not constitute evidence for Too-Big-to-Fail, then what historical episodes do provide that evidence? We leave that question for another day.

In a world of no Too-Big-to-Fail, no FDIC, and NGDPLT, the optimal bank regulatory regime is no regulation at all.

Of course we don’t live in that world.  What’s the optimal regulatory regime in a world with TBTF, with FDIC, and without NGDPLT?  How the hell would I know?

The good news is that if we ever get to NGDPLT, we won’t need TBTF and FDIC.


Tags:

 
 
 

47 Responses to “Miron and Rigol on bank failures and output”

  1. Gravatar of TravisV TravisV
    17. September 2013 at 18:32

    Excellent support for the argument that recession –> financial crisis, not financial crisis –> recession.

    At any rate, isn’t there good reason to believe that the optimal regulatory regime is “something like Canada’s”?

  2. Gravatar of Morgan Warstler Morgan Warstler
    17. September 2013 at 18:43

    OK,

    in the past I have always just said if we did NGDPLT in 2005 or 2008, we’d have either seen:

    2005: bankers eat it by not making loans

    2008: bankers it eat by having to unwind loans.

    And then today I started into what if fed bought / sold random shares of stock to stay on a 4.5 NGDP level target…

    For me, perhaps thru mental defect, it is easier to visualize the hot potato, when the Fed is either giving the nation’s shareholders money for stock OR giving them stock for money.

    It’s easier for me to grok this, bc on both side of the trade I just expect the fed to lose money.

    And I PUKE when i think of fed letting “money traders’ banks make money on both sides of trade.

    But my mental gag reflex eases up, when I think “cantillon effect” and the tad bit $ goes to shareholders directly.

    There’s market research here for MM crowd.

  3. Gravatar of MMdoubter MMdoubter
    17. September 2013 at 18:46

    What about counterparty risk? That was a big reason why the markets froze up post-Lehman. Would NGDPLT dissuade the financial system from forming hubs to which many spokes connect? (I don’t think so – that’s market microstructure driven by economies of scale, barrier-to-entry considerations and what-not)

    Now, in the current system the Fed’s LoLR facilities are meant to tackle that. Can we do away with those if we have NGDPLT? Would a big counterparty collapse immediately signal a fall in the NGDP futures market? Will the Fed’s response to shore up the NGDP market be enough to mitigate the counterparty exposures+risk premia that banks will face in funding themselves? What if there is a lag between a big-bank collapse and NGDP futures sell-off?

    Will the blunt injection of money via the NGDP futures market prevent big banks from hoarding capital at the cost of making smaller banks insolvent (as their funding costs become higher and higher)?

    Also, in the good times, will NGDPLT prevent banks from using leverage to enhance RoE? If so, how?

  4. Gravatar of Morgan Warstler Morgan Warstler
    17. September 2013 at 18:47

    I now put forward a new MP rule:

    The transmission method “where money enters leaves system” for MP, should be based entirely on Human Gag Reflex (HGR).

  5. Gravatar of dtoh dtoh
    17. September 2013 at 19:36

    1. I think you confuse things by referring separately to the credit channel, wealth effect, and asset price channel. When there is an increased exchange of financial assets for real goods and services by the non-financial sector, that results in an increase in NGDP. Other than expectations, this is the only significant channel. (And when I say exchanging financial assets, it includes everything: selling Treasuries out of your portfolio, new share issuance, increased line of credit draw downs, consumer auto loans, increased credit card balances, etc. etc.)

    2. If the financial sector (FED included) is less willing (either from a quantity or price perspective) to take financial assets in exchange for money, then you get a marginal drop in NGDP. All kinds of things can cause this: shocks, counter-party risk, bank failures, bad FED policy, etc. But in the end it’s all the same thing. If the private financial sector becomes less willing to do the exchange, the Fed has to step in and do more…. otherwise you will get a drop in NGDP.

    3. And if the private financial sector knows the FED will act, it is much more unlikely that it will seize up in the first place.

    Honestly, the fact that this is so incredibly trivial and yet it is still being debated makes me think the economics profession is a moron magnet.

  6. Gravatar of dtoh dtoh
    17. September 2013 at 19:40

    Scott, you said;

    “Of course we don’t live in that world. What’s the optimal regulatory regime in a world with TBTF, with FDIC, and without NGDPLT? How the hell would I know?”

    Simple…. you just have the Fed regulate minimum and maximum asset to equity ratios in the private financial sector.

    Banks are engaging in over-speculative investments in mortgage back securities?…. Lower the max asset to equity ratio on MBOS.

    Economy in the doldrums?….. Raise the min asset to equity ratio on all assets.

    This too is trivial.

  7. Gravatar of TESC TESC
    17. September 2013 at 20:05

    Dr. Sumners,

    It sounds like a great idea, but why wouldn’t we need FDIC? How can we justified getting away with it under NGDPLT?

  8. Gravatar of Saturos Saturos
    17. September 2013 at 22:51

    David Laidler prefers inflation targeting: http://www.econtalk.org/archives/2013/09/david_laidler_o.html

    I still am pretty happy with the inflation rate as the target of policy. I base this a lot on Canadian experience. We’ve been targeting the inflation rate since 1991; we’ve done it pretty successfully. We didn’t have a big asset market crisis here. We had not had a recession until 2008 since 1991. So inflation targeting worked pretty well for us. And it worked I think because it was a very explicit policy target agreed between the government and the Bank of Canada. It wasn’t an informal thing, as it was in the Fed. It was discussed continually. And as time passed, the targets were hit and it gained in credibility. So I would see no reason to go from that to a nominal GDP target. I don’t like nominal GDP as a target for policy, for the simple reason is: that’s a variable that’s measured with a lag and it’s subject to a lot of revision. And I don’t see how you can run forward-looking monetary policy targeting the behavior of a variable that you don’t get a good reading on for 18 months after it’s happened. With inflation targeting based on a Consumer Price Index, you get timely data and it’s not subject to revision. The indices are not perfect, but they are well understood by policy makers and the general public understands them as well. If I tell my wife the Bank of Canada is targeting nominal GDP, she’ll just look at me and wonder what on earth I’m talking about. If I tell her they are telling her they are targeting the rate at which the cost of living goes up, she understands that. And knowing that there is that target out there affects the behavior of ordinary consumers and producers, not just financial markets.

  9. Gravatar of sdfc sdfc
    17. September 2013 at 23:08

    The good news is that if we ever get to NGDPLT, we won’t need TBTF and FDIC.

    Except for the small matter of the fall in the money supply.

    Or are you saying that if enough reserves are injected banks will always be able to honour their deposits with currency?

  10. Gravatar of Petar Petar
    17. September 2013 at 23:19

    @mm doubter, heres my take
    there is interesting paper on counterparty risk in the moneymarkets on the ecb web (Heider, Holthausen & Hoerova I think). basically it shows that counterpary risk stemmed from development in asset side of agents balance sheets, and that level/dispersion of that risk is very important for the existence of multiple equilibria, some of those leading to a collapse of the money market. So when you got generally tanking asset prices in wake of NGDP collapse, you could see why suddenly there are more assets to “sour” and counterparty risk leading to liquidity hoarding can happen. That would all be fine if money market wasnt important for ecb in the sense of traditional interest rate channel. If they did QE, they wouldnt need money market that much – they would focus on nominal stability instead of trying to save their prefered mt. channel. Same as Fed had problems with primary dealer system, ECB was trapped by its own operational framework relying on banks and repos. so basically it was failure to stabilize ngdp – moneymarkets functioned (with somewhat higher spreads) from 8.2007. to lehman run-up – NGDP was halfway to bottom by then

  11. Gravatar of Saturos Saturos
    17. September 2013 at 23:28

    The Economist on tapering: http://www.economist.com/news/leaders/21586313-combine-small-cut-bond-purchases-clear-commitment-support-economy-more-if

  12. Gravatar of W. Peden W. Peden
    17. September 2013 at 23:35

    Saturos,

    It seems Laidler’s argument is based on comparing apples with oranges: “NGDP targeting is a more confusing concept than a colloqial explanation of inflation targeting.”

    One might as well say that inflation targeting is harder for the public to understand, because “Targeting the hedonic calculus-weighted CPI measure of inflation with due regard for output gaps and proper interpretation of the impact of supply-side shocks and distortions to the CPI through factors like discrepancies between the total index value and core CPI is more confusing that targeting average incomes.”

    The difference, of course, is that “targeting average incomes” is a synonym for “NGDP targeting”, whereas Laidler can only make inflation targeting make sense for the public by describing it in a misleading way, because what I said above is (a) a correct description of inflation-targeting and (b) not the same as “targeting the cost of living”.

    And the point that CPI data is “not subject to revision” is just plain wrong and David Laidler must know that it’s wrong. And the claim that the general public understands the CPI is the biggest piece of ivory-tower nonsense I’ve heard in a long time.

    (Note: I am a fan of David Laidler.)

  13. Gravatar of W. Peden W. Peden
    17. September 2013 at 23:39

    Oh, and as Scott has covered before, what David Laider means by “the cost of living” and what his wife almost certainly understands by “the cost of living” are two different things, so even Laidler’s colloqial explanation of inflation targeting only seems to work because of an equivocation.

  14. Gravatar of nickik nickik
    17. September 2013 at 23:54

    @sdfc

    As other banking people had said. The emergeny lending of the fed is not as importend nowdays because of overnight markets. It should always be possible to get lones from overnight markets if the NGDP keeps on track.

  15. Gravatar of Saturos Saturos
    17. September 2013 at 23:55

    Does “opportunistic disinflation” explain the Fed’s endgame?
    http://www.theatlantic.com/business/archive/2013/09/fed-favorite-janet-yellen-is-no-dove-and-thats-a-good-thing/279773/

    (potential to answer two comments at once here Scott.)

  16. Gravatar of sdfc sdfc
    18. September 2013 at 00:13

    Nickik

    No one said anything about emergency lending by the Fed. My comment was in regard to the destruction of deposits when banks fail.

  17. Gravatar of Benjamin Cole Benjamin Cole
    18. September 2013 at 01:25

    “Most of the failures were linked to loans to developers than went bad when the economy tanked, not the subprime mortgage mess.”–Scott Sumner

    Dang it!

    That’s what i have been saying. Though I would rephrase, “Most of the bank failures were linked to loans to developers, real estate investors and speculators that went bad when the economy tanked…”

    They were an awful lot of “trophy office buildings” that should not have been bought.

    But Scott Sumner, earlier you said the housing bust preceded the general economy bust. The data shows it did, but obviously was highly amplified by the subsequent overall economic contraction (brought about by too-tight money). And remember, commercial properties equally tanked.

    This suggest a root cause: too-tight money after people were leveraging up to buy real estate.

    As for Fannie and Freddie, shut ’em down, bunt I think they are not too important. The home mortgage tax deduction also encourage people to over-allocate to housing to take risks to acquire housing…but it is improper to ever mention that!

    This goes back to my latest meme-trope: Real estate is “sui generis,” as banks will lend on real estate, thinking it is collateral. Buyers can leverage, and we are talking huge numbers between residential and commercial. Enough to tank the banks, and thence the economy, if the Fed is too timid in reflating.

    Thus, some sort of regulations might be in order to real estate lending, as in minimum good-sized down payments.

    A NGDP targeting regime that results in general mild real estate inflation will also prevent large scale busts.

    Real estate is the Achilles Heel of banks. Even the 1990s dot.com bust hardly made the economy flinch.

  18. Gravatar of Michael Michael
    18. September 2013 at 02:54

    MMDoubter wrote:

    “What about counterparty risk? That was a big reason why the markets froze up post-Lehman. Would NGDPLT dissuade the financial system from forming hubs to which many spokes connect? (I don’t think so – that’s market microstructure driven by economies of scale, barrier-to-entry considerations and what-not)”

    I think the idea is that NGDPLT can begin to counteract the problems of moral hazard.

    The implicit “public” rationale for engaging in bailouts is, more or less, to prevent the failure of institutions that are “too big to fail” from causing a catastrophic fall in national income.

    However, such institutions are not unaware that they are likely to be bailed out if they blow up. This means that they can, in effect, use the expetation of public backing when needed as a taxpayer-funded insurance policy – especially if they are interconnected in a variety of complex ways. And the more interconnected, the greater the systemic threat, the greater the chance of a bailout and therefore the greater potential return on investment, particularly for unsecured bondholders.

    Under NGDPLT, failure of TBTF institutions won’t cause a catastrophic fall in national income (it might cause some temporary inflation instead if it is big enough to cause supply shocks ), so there would be no “public” rationale for bailouts.

  19. Gravatar of Brian Donohue Brian Donohue
    18. September 2013 at 03:44

    benjamin,

    You say “This suggest a root cause: too-tight money after people were leveraging up to buy real estate.”

    You also say: “A NGDP targeting regime that results in general mild real estate inflation will also prevent large scale busts.”

    So, if we’re talking about the roots of the leveraging up, what about the 6.5%+ NGDP growth from late 2002 to early 2006 and associated real estate inflation?

  20. Gravatar of ssumner ssumner
    18. September 2013 at 05:03

    travis, Sounds right.

    MMdoubter, You are gliding over several issues. I mentioned NGDPLT and no TBTF or FDIC. You responded with NGDPLT, but no assumption of moral hazard. That makes a huge difference in terms of the structure of the banking industry.

    In addition, discussion of “would it matter” needs to be more precise. Do you mean would it matter for banking? Or for RGDP? With stable NGDP, I’d expect RGDP to be fairly stable, even if banking was in crisis. I’d add that it is unlikely the interbank loan market would have frozen up in 2008 if we had been doing NGDPLT.

    dtoh, You said;

    “If the private financial sector becomes less willing to do the exchange, the Fed has to step in and do more…. otherwise you will get a drop in NGDP.”

    I made the same point in my post, just using different language.

    As far as the “simple regulations”. I’m told by people smarter than me than any and all capital regs can be evaded by banks using accounting tricks.

    TESC, We don’t need it because bank failures would no longer hurt the economy.

    Saturos, I’ve demolished those arguments many times, as you know, but he obviously hasn’t read my demolition. Otherwise I’m a big fan of Laidler.

    sdfc. No, I’m not saying they’ll always be able to honor their deposits. I’m saying a fall in M doesn’t matter as long as expected NGDP is on target.

  21. Gravatar of ssumner ssumner
    18. September 2013 at 05:13

    Saturos, They are right that she is not an inflation dove. As for “opportunistic disinflation,” obviously that’s a bad idea.

    Fortunately we don’t need any more disinflation, so it’s now a moot point.

    As for The Economist, I’d say a good compromise would be to do a slight taper, and lower the unemployment threshold for raising rates to 6.3%.

  22. Gravatar of Rien Huizer Rien Huizer
    18. September 2013 at 05:32

    Scott,

    Wishing away a piece of “sticky” financial infrastructure is symptomatic of utopianism. I thought you were an -intelligent- libertarian. People are not risk neutral and hence, lots of fears that politicians can and will exploit, NGDP etc targeting or not. And no one will trust the government to keep the Fed mandate (even if it is NGDP targeting based) intact when that would be opportune. Not even the benign rulers of the PRC would allow such a constraint on their discretion , let alone the self interested anarchists in DC.

  23. Gravatar of Brian Donohue Brian Donohue
    18. September 2013 at 05:40

    OT: Tyler Cowen talks Summers step-down on Linked in (he uses the word bubble(!)):

    http://www.linkedin.com/today/post/article/20130916210334-1776606-after-larry-summers-where-to

    Read the comments Scott. The American people are making the same inference (QE = money to rich) that I suspect Obama harbors. Cranks of all political stripes agree.

    There is enough of a beachhead within the profession at this point- you need to set your sights on a larger stage like Morgan keeps telling you- you could go down in history.

  24. Gravatar of ssumner ssumner
    18. September 2013 at 05:45

    Rein; You said;

    “Wishing away a piece of “sticky” financial infrastructure is symptomatic of utopianism. I thought you were an -intelligent- libertarian. People are not risk neutral and hence, lots of fears that politicians can and will exploit, NGDP etc targeting or not.”

    And here’s what I said:

    “Of course we don’t live in that world.”

    I thought you were an intelligent commenter who knew how to read. Guess not.

    Brian, You said;

    “Read the comments Scott. The American people are making the same inference”

    Big mistake, equating blog comment sections with “the American people.” What percentage of Americans have heard of “QE” How many know what it is?

    Two percent?

  25. Gravatar of Mark A. Sadowski Mark A. Sadowski
    18. September 2013 at 05:55

    W. Peden,
    “And the point that CPI data is “not subject to revision” is just plain wrong and David Laidler must know that it’s wrong.”

    In David Laidler’s defense the CPI is not revised in Canada.

    But just because CPI isn’t revised doesn’t mean it’s accurate. It’s still based on a sample. All it means is that StatsCan either can’t or won’t revise it. And all of this means Canada’s unrevised CPI are almost certainly far less accurate than Canada’s revised NGDP.

    Laidler also says:
    “I don’t like nominal GDP as a target for policy, for the simple reason is: that’s a variable that’s measured with a lag…”

    In the US the inflation target is the PCEPI. In months in which there are advanced quarterly GDP estimates released (in the US RGDP and NGDP is released simultaneously) the PCEPI for the last month of the quarter is typically released *after* NGDP.

    For example the advance estimate of 2013Q2 GDP was released on Wednesday July 31. The PCEPI wasn’t officially released until Friday August 2.

    (It’s worth noting if you’re mathematically inclined its a simple matter to back out the PCEPI inflation rate from the GDP release before the inflation rate is officially released.)

    What about the fact NGDP is released quarterly and price indicies are released monthly? This is strictly a convention. Private forecasting firms produce monthly estimates of NGDP. There’s nothing at all preventing government agencies from producing monthly estimates of NGDP.

    In any case one should not be targeting past levels of NGDP. One should target future levels of NGDP.

  26. Gravatar of benjamin cole benjamin cole
    18. September 2013 at 05:58

    Brian
    I can’t tell if you agree with me or not.

  27. Gravatar of Brian Donohue Brian Donohue
    18. September 2013 at 06:33

    benjamin,

    I’m just tracing back the roots of causality…unless the “leveraging up to buy real estate” was itself uncaused.

  28. Gravatar of Brian Donohue Brian Donohue
    18. September 2013 at 06:36

    Scott,

    On a superficial level, Americans connect “the Fed” and “Wall Street”. I mean, they’re all bankers right?

    I’m not convinced myself on this point. I recall Goldman Sachs, saying (presciently?) several years ago, that the needed QE was about $3 trillion. I don’t trust Goldman Sachs to do anything that isn’t 100% in their interest.

  29. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    18. September 2013 at 06:55

    Slovenia believes in Sinatra-nomics;

    http://www.dw.de/slovenia-insists-it-can-save-its-own-banks/a-17088243

    ‘At an informal meeting in the Lithuanian capital Vilnius, Slovenian Finance Minister Uros Cufer had to explain to his 16 eurozone counterparts exactly how his country intends to get out of a banking crisis that is threatening the country’s financial system. Eurogroup chief Jereon Dijsselbloem had rather suddenly put Slovenia on the meeting’s agenda. For years, Slovenia’s banks have been sitting on bad mortgages estimated at around 7 billion euros ($9.3 billion).

    ‘The government now wants to liquidate two of the smaller banks and then inject an extra billion euros into the system, Cufer explained. “We still have enough money in the coffers,” he said curtly. But Dijsselbloem, also the Dutch finance minister, made clear that this can only be the beginning of a more a thorough restructuring program.’

  30. Gravatar of W. Peden W. Peden
    18. September 2013 at 08:10

    Mark A. Sadowski,

    One could target unrevised NGDP, so if that’s what Laidler’s basing his argument on then it makes no sense.

    “In any case one should not be targeting past levels of NGDP. One should target future levels of NGDP.”

    Exactly.

  31. Gravatar of MMdoubter MMdoubter
    18. September 2013 at 08:22

    Scott,

    “…if we ever get to NGDPLT, we won’t need TBTF and FDIC”
    This statment has no assumption of a moral-hazard. It says, and correct me if I’m wrong, NGDPLT => (no FDIC/TBTF). I fail to see how.

    “In a world of no Too-Big-to-Fail, no FDIC, and NGDPLT, the optimal bank regulatory regime is no regulation at all.”
    This is not equivalent to the first statement I’ve quoted above.

    Parsing this further:
    “…the optimal bank regulatory regime is no regulation at all”.
    Too imprecise. What’s the objective function we’re talking about here?

    “With stable NGDP, I’d expect RGDP to be fairly stable, even if banking was in crisis”
    Only if banking is no longer your monetary policy transmission conduit. You have made no such assumptin here.

    “I’d add that it is unlikely the interbank loan market would have frozen up in 2008 if we had been doing NGDPLT.”
    How? What would have been the exact mechanism to prevent this under an NGDPLT regime?

    (My main concern – how does a blunt monetary-policy action help individual banks with vastly heterogenous funding mechanisms and costs?)

  32. Gravatar of nickik nickik
    18. September 2013 at 08:49

    @sdfc

    My point is this. Maybe I misunderstood you. If you actuall have your money in a bank that is actually going bust, its your problem.

    The NGDP level and overnight lending market should prevent bank runs on solvent banks.

  33. Gravatar of dtoh dtoh
    18. September 2013 at 09:34

    Scott, you said;
    “As far as the “simple regulations”. I’m told by people smarter than me than any and all capital regs can be evaded by banks using accounting tricks.”

    I don’t think so. I was involved developing structured products for banks to get around Basel I, when it came out. Not that easy. Even if it was easy, as long as the Fed has the ability to set the ratios by asset class and adjust for specific banks, they could just zap anyone who was playing games. I guarantee this is an easy way to solve the all the financial regulation problems.

  34. Gravatar of TESC TESC
    18. September 2013 at 16:32

    @ MMdoubter,

    ”Only if banking is no longer your monetary policy transmission conduit. You have made no such assumptin here.”

    That is an understament. All the way back from 2009 Sumners has stated that MP on NGDPLT can be conducted through purchases to the general public. Check Sumners on EconTalk in 2009.

  35. Gravatar of ssumner ssumner
    18. September 2013 at 18:29

    Brian, You said;

    “On a superficial level, Americans connect “the Fed” and “Wall Street”. I mean, they’re all bankers right?”

    Wrong. Bernanke?

    MMdoubter, You said:

    “What’s the objective function we’re talking about here?”

    Utilitarianism.

    You said:

    “Only if banking is no longer your monetary policy transmission conduit. You have made no such assumptin here.”

    It is not now and will not be in the future.

    You said:

    “How? What would have been the exact mechanism to prevent this under an NGDPLT regime?”

    The main cause of the crisis was falling asset prices. The main cause of the falling asset prices was falling NGDP expectations.

    dtoh, Maybe, but the conventional wisdom is that the Basil 1 standards were evaded.

    TESC, Even today I’m told that many OMOs don’t directly involve banks.

  36. Gravatar of sdfc sdfc
    18. September 2013 at 19:58

    Scott

    It’s difficult to see how you can have higher income growth with a declining money supply.

    Nickick

    A failing bank is going to have trouble funding itself on the interbank market no matter how low the cash rate is. Deposits are money so widespread bank failures are everyone’s problem.

  37. Gravatar of Benjamin Cole Benjamin Cole
    18. September 2013 at 20:06

    Brian-

    If you are still reading, it would be interesting to know what caused consumers to buy houses and large institutions to make major wagers on huge office buildings (and obtain aggressive financing) all in the 2006-7 period.

    I was familiar with the California story on the institutional side; buyers kept saying there were barriers to entry in the CA market (local building regs and snags), and that CA population was still growing, and that per capita income in CA was very high, and that foreign money was always looking in CA etc. Around any bull market you find a story being concocted by marketeers.

    It may be we have a “global glut” of capital, and so huge amounts of capital are looking for a home (no pun intended). This drives down yields, and makes property look more attractive (rents and possible appreciation).

    I think a soft landing, or plateau could have been easily engineered by the Fed if they did not tighten but maintained a steady growth path,

    instead, the Fed tightened at precisely the wrong moment, pulling the rug out from heavily leveraged poetry buyers….

  38. Gravatar of Benjamin Cole Benjamin Cole
    18. September 2013 at 20:08

    Um, the last sentence should read “property buyers.”

  39. Gravatar of sdfc sdfc
    18. September 2013 at 21:05

    That’s one hell of a highwire act you’re expecting of central banks.

  40. Gravatar of Brian Donohue Brian Donohue
    19. September 2013 at 03:41

    benjamin,

    Still here. I understand there were lots of shenanigans going on with other aspects of policy in the first half of last decade, but NGDP was running at a 6.5%+ clip and CPI was approaching 3.5% in 2004-2006. Perhaps a little ‘monetary offset’ then would have gone a long way.

    So yeah, Keynesians have this theory about deficit spending during recessions and running surpluses during the good times, but empirically, they only seem able to come through on the first half of the theory.

    MM doesn’t really have a track record to evaluate. Was Scott urging the Fed to rein things in a bit in 2005? I dunno.

  41. Gravatar of MMdoubter MMdoubter
    19. September 2013 at 04:15

    Scott:

    “Utilitarianism.”

    Excellent. Now if you could provide some mathematical proof supporting your claims.
    (Woodford backs his brilliant ideas with math – and that’s why he gets to go to Jackson Hole)

    “It is not now and will not be in the future.”

    I disagree. The Fed SLO surveys show a very strong correlation between lending standards and credit spreads (especially HY spreads) and overall output. Easing of lending standards came directly as a result of (a) cleaner balance-sheets of banks freeing up more capital to lend and (b) improved expectations of aggregate demand which, in turn, were fuelled by higher consumption expectations as the households deleveraged. Deleveraging was made easier by lower uncertainty regarding collateral prices (because of targeted asset-purchases and maturity transformation by the Fed, and *not* expectations of nominal income) and lower interest rates. The Fed’s main transmission mechanism has been reduction of uncertainty – *not* raising expectations of nominal income (that’s just a by-product).

    And this is the main reason I disagree with NGDPLT as an optimal policy. As long as (1) the financial system is NOT made less susceptible to sunspot collapses and (b) uncertainty about prices/returns or output-gaps are NOT managed *individually*, risk-premia arising due to volatility will always make money tighter and make nominal shocks more severe.

    NGDPLT does reduce uncertainty about nominal income but having a central-bank loss function which includes both inflation and output gaps is even better. Why discard two pieces of info which have orthogonal components in favor of a policy loss-function (NGDPLT) which scrambles both of those pieces irretrievably?

    “The main cause of the crisis was falling asset prices. The main cause of the falling asset prices was falling NGDP expectations.”

    I agree with the first sentence. Only partly with the second sentence.

  42. Gravatar of ssumner ssumner
    19. September 2013 at 06:36

    sdfc, Velocity can rise enough to offset. It’s unusual, but does occur on occasion.

    MMdoubter, You said:

    “NGDPLT does reduce uncertainty about nominal income but having a central-bank loss function which includes both inflation and output gaps is even better. Why discard two pieces of info which have orthogonal components in favor of a policy loss-function (NGDPLT) which scrambles both of those pieces irretrievably?”

    Because inflation doesn’t measure what economists assume it measures. NGDP growth is much more closely correlated with the so-called “welfare costs of inflation” than inflation itself. If you really want to look at inflation, I’d suggest wage inflation, which correlates much better with demand shocks and unemployment than price inflation.

    Nobody has a clue as to the current output gap, economists are all over the map on that one. The experts keep revising the trend line.

    You said:

    “I agree with the first sentence. Only partly with the second sentence.”

    Let me be clear that the earlier fall in housing prices (2006-08) was an exogenous shock. I meant the fall in almost all asset prices in late 2008 was driven by the collapse in NGDP expectations. That’s when the financial crisis got really bad.

    You said:

    “Excellent. Now if you could provide some mathematical proof supporting your claims.
    (Woodford backs his brilliant ideas with math – and that’s why he gets to go to Jackson Hole)”

    When Christy Romer endorsed NGDPLT in the New York Times she cited a paper by an obscurce Bentley professor, which had no math, as providing the “logic” behind NGDPLT. So I don’t need any lessons on how to win friends and influence people. I’ve already succeeded beyond my wildest dreams.

    Math is very overrated. The implications of almost all macro models is clear from the assumptions built into the models. Krugman says he was stunned to find that (in 1998) the math showed that temporary monetary injections were ineffective. I already knew that in 1993. But if you insist on dressing things up with math, here you go:

    1. The Fed drives NGDP by controlling the base:

    NGDP = MB*V(i) where V is a positive function of nominal rates.

    And employment is negatively related to the ratio of nominal average hourly wages and NGDP/person

    E = f(W/[NGDP/person])

    where nominal wages adjust slowly and monetary policy drives NGDP/person.

    Is that enough math? I have longer versions but that’s the essence. You do realize that Nick Rowe has shown that NK models don’t really prove anything? You have to assume there is a sensible equilibrium, it doesn’t come out of the NK model.

    A far as your last point, I agree that RGDP is correlated with lending standards, but that’s mostly because the Fed stupidly allows NGDP to be correlated with lending standards. They need to stabilize NGDP, and let the financial system take care of itself.

  43. Gravatar of 1930s Bank Failures and Output: DYTVSC? | askblog 1930s Bank Failures and Output: DYTVSC? | askblog
    20. September 2013 at 03:29

    […] Scott Sumner has more. […]

  44. Gravatar of Rien Huizer Rien Huizer
    20. September 2013 at 04:32

    Scott,

    Thanks for the reprimand but devoting a post to something you yourself do not believe to be part of the real world is in my book, utopian. I did read the whole piece, but looking for meaning.

  45. Gravatar of TESC TESC
    21. September 2013 at 13:29

    @ Sumner

    “TESC, Even today I’m told that many OMOs don’t directly involve banks.”

    Could you please give me a link or something on OMOs that do not involve banks? That is a very interesting topic and I cannot find anything substantive in the net so far.

    Thanks

    @ MMdoubter

    “Excellent. Now if you could provide some mathematical proof supporting your claims.”

    How about you look at the correlation between NGDP and unemployment rate? I do not think that you could say that this correlation is caused by something else. When there is less spending, there is less employment. I am a Math major, but my Real Analysis class did not prepared me for this. Psychology seems was more “util” and a simple Supply/ Demand graph will let you know. Decreases in demand, decrease the quantity supplied, which entails less employment.

  46. Gravatar of ssumner ssumner
    22. September 2013 at 06:13

    TESC, I’m told that OMOs are done with New York bond dealers. Some of those bond dealers are banks, and some are not. Perhaps someone can confirm.

  47. Gravatar of TESC TESC
    22. September 2013 at 06:50

    I got it. Your mention of “dealer” gave me the cue to look on the net. They are call “Primary Dealers”. They can be banks or securities broker-dealers firms. They definitely do not have to be banks.

    Thanks

Leave a Reply