Bernanke at the AEA: Everyone missed the big story

A lot of people are discussing Bernanke’s defense of “monetary ease” in 2002, but everyone seems to be missing the much bigger story.  Bernanke’s explanation for the Fed’s actions in 2002 show exactly how monetary policy failed in 2008.  In particular, Bernanke made the following three observations regarding 2002:

1.  Monetary policy needs to focus on the macroeconomy, not specific sectors.

2.  Monetary policy must be forward-looking, must target the forecast.

3.  Monetary policy must be especially aggressive when there is risk of liquidity trap (which would render conventional policy ineffective.)

In 2008 the Fed did exactly the opposite.  Between September and December 2008 the Fed focused on banking, not the macroeconomy, they adopted a backward-looking Taylor Rule, and they were extremely passive when the threat of a liquidity trap was already obvious.

In the 1920s Governor Strong argued that if the Fed were to try to stop the stock market boom, it would be like spanking all one’s children just because one had misbehaved.  Bernanke seems to feel the same way:

Monetary policy is also a blunt tool, and interest rate increases in 2003 or 2004 sufficient to constrain the bubble could have seriously weakened the economy at just the time when the recovery from the previous recession was becoming established.

So the Fed should focus on the macroeconomy, not a specific sector.  I agree. Unfortunately, they did exactly the opposite in the last 4 months of 2008.

The second quotation shows that Bernanke does not accept the policy implications of backward-looking Taylor Rules, he strongly believes that monetary policy must be forward-looking; policy must respond to looming inflation or deflation threats, not the previous year’s inflation rate:

Slide 4 shows that the version of the Taylor rule based on forecast inflation (in green dots) explains both the course of monetary policy earlier in the past decade as well as the decision not to respond aggressively to what did in fact turn out to be a temporary surge in inflation in 2008. This comparison suggests that the Taylor rule using forecast inflation is a more useful benchmark, both as a description of recent FOMC behavior and as a guide to appropriate policy.

There might be no better example of the distinction between forward and backward-looking Taylor Rules than economic situation on September 16, 2008, when the FOMC made its most momentous error.  To set the scene, recall that we had been in recession for nine months.  But the early stages of the recession were extremely mild, and the last unemployment reading had been about 6.2%.  The last CPI data that was available was for July, and due to the extremely high oil prices the headline inflation numbers were over 5%, far above the Fed’s implicit target.  Core inflation, however, was still well-behaved.  And most importantly, the precipitous fall in commodity prices between mid-July and mid-September had greatly reduced inflation expectations.  Two days before the FOMC meeting Lehman failed, and this further reduced inflation expectations in the TIPS markets.  By the time of the September 16th meeting, 5 year TIPS spreads had fallen to only 1.23%.  Now let’s compare the policy implications in September 2008 of a backward looking Taylor Rule, to the forward-looking policy that Bernanke says that he favors:

Backward-looking rule:  There is both a risk of recession and high inflation.  No policy action called for.

Forward-looking rule:  There is risk of recession, and risk that inflation will sharply undershoot the implicit 2% target.  Cut rates.

The FOMC statement said the risks of recession and inflation were roughly balanced.  The Fed took no action.  John Taylor won and I lost.

At this point you might wonder why I put so much emphasis on one meeting, after all the Fed did eventually get around to cutting rates almost to zero, so what’s the rush?  How much harm can be done from one botched meeting?  Normally a single error can be easily corrected, but let’s see what Bernanke says about policy where the economy is teetering on the edge of a liquidity trap:

The aggressive monetary policy response in 2002 and 2003 was motivated by two principal factors. First, although the recession technically ended in late 2001, the recovery remained quite weak and “jobless” into the latter part of 2003. Real gross domestic product (GDP), which normally grows above trend in the early stages of an economic expansion, rose at an average pace just above 2 percent in 2002 and the first half of 2003, a rate insufficient to halt continued increases in the unemployment rate, which peaked above 6 percent in the first half of 2003.  Second, the FOMC’s policy response also reflected concerns about a possible unwelcome decline in inflation. Taking note of the painful experience of Japan, policymakers worried that the United States might sink into deflation and that, as one consequence, the FOMC’s target interest rate might hit its zero lower bound, limiting the scope for further monetary accommodation. FOMC decisions during this period were informed by a strong consensus among researchers that, when faced with the risk of hitting the zero lower bound, policymakers should lower rates preemptively, thereby reducing the probability of ultimately being constrained by the lower bound on the policy interest rate.  (emphasis added.)

Of course this is precisely the mistake the Fed made in late 2008.  In mid-September they saw no need for a rate cut, even though rates were only 2%, and even though T-bill rates had briefly dipped close to zero in the financial crisis of early 2008, and even though Lehman had just failed, and even though TIPS showed a dangerous plunge in 5 year inflation expectations, and even though we had been in recession for 9 months.  In other words, things were already far more scary than in 2002.  By mid-October many observers argued that we were already in a liquidity trap and there was nothing more that the Fed could do.  Time for fiscal stimulus.  That sure didn’t take long!  I have to give credit to Bernanke, he is a superb at diagnosing problems.  To use a medical analogy, Bernanke is the guy you want diagnosing the patient and Krugman is the guy you’d send into the operating room.  In a crisis, you need someone with a sense of urgency and decisiveness, and as well as an understanding of what could happen if you don’t act in time.

I was lulled by all those academic articles (some by Bernanke) into thinking that the Fed had a backup plan if we went into a liquidity trap.  Krugman kept saying “it’s harder than it looks.”  I still don’t think it is that hard, but as a practical matter Krugman was right.  There was no backup plan.  The Fed acted as if there was little they could do once rates hit zero.  That makes Bernanke’s pre-emption doctrine all the more essential.

When I was young, I recall some teenagers who played a dare-devil game on top of those white mushroom shaped water towers that dot the Midwest.  They would start at the top, and see who dared slide the furthest down.  Of course if you make a mistake it is all over.  Once you start sliding, the slope keeps getting steeper.  (Kids, don’t play this game without a parachute.  And if you must play, wear sneakers!)  It turns out that the Fed was playing a similar game, seeing how far down they could let T-bill yields fall without enacting an aggressive policy of monetary ease.  In the first 10 days of October the stock market crashed by 23%.  I’m guessing that by October 10th Bernanke had the sickening feeling in his stomach that one of those kids would have had if they went “over the horizon” on a water tower.   There must have been some point when the Fed realized it was too late for monetary policy.  Why not go immediately from 2% to zero once the mistake was realized?  I’m guessing that they thought that would look too desperate, and admission that they had blown it in the September meeting.

The Fed had already suffered a similar embarrassment in December 2007, when they cut rates 1/4 at a meeting where many were looking for a 1/2 point cut.  The stock and bond markets panicked after the 2:15 pm announcement.  Stocks are easy to explain, but you might be surprised to hear that 3 month T-bill prices actually rose, i.e. yields fell.  How could yields have fallen on a tighter than expected Fed announcement that sharply depressed the stock market?  Easy, the markets knew the Fed blew it, and that this would push us into recession.  They also knew that the recession would force the Fed to make an embarrassing make-up call in the near future.  And they were right, we went into recession and the Fed panicked and cut rates an additional 75 basis points just a month later, and another 50 at the next scheduled meeting—a total cut of 125 basis points in 10 days.  In the fall of 2008 they needed an even more aggressive move.  They needed to cut rates so sharply that 3 month T-bill yields rose.  And they blinked.

Arnold Kling likes to make my belief that tight money was to blame seem very mysterious:

Like Scott Sumner, Krugman views the main problem as deflationary expectations. I am not sure where these expectations come from. In Sumner’s story, people just decide that the monetary authority is willing to see the price level drop.

If you think that low interest rates are easy money and high interest rates are tight money, then it does seem very mysterious.  But it is not mysterious to the stock market.  Unlike Kling, the stock market does believe monetary policy has a near-term impact on the economy.  It isn’t just coincidence that the Dow soared hundreds of points right after 2:15pm on September 18, 2007, or that it crashed hundreds of points right after 2:15pm on December 11th, 2007.  And it’s not just coincidence that 3 month T-bill yields fell on a tighter than expected Fed announcement that plunged us into recession.  Those market responses are sending us powerful signals about how Fed policy shocks (or even Fed errors of omission in the face of an increase in money demand during financial crises), can have a powerful effect on AD.  And the markets also understand how deflationary policies can dramatically worsen a financial crisis.

Most economists ignore these signals because they don’t fit in to their “low interest rates = easy money” worldview.  The markets were warning us all along, it’s just that we weren’t paying any attention.

Here’s how I read the recent AEA meeting speech.  Bernanke is in a box.  He feels he must defend the Fed from the neo-Austrian charge that monetary ease in 2002-04 blew up the housing bubble and led to the sub-prime fiasco.  But if he does so it leaves him open to criticism from people like me who point out that his explanation implies money was much too tight in late 2008.  Bernanke knows that for every Sumner, Woolsey or Hetzel, there are 1000 neo-Austrians.  In purely political terms he probably made the right move at the AEA.  History may not judge him so kindly.



21 Responses to “Bernanke at the AEA: Everyone missed the big story”

  1. Gravatar of Marcus Nunes Marcus Nunes
    5. January 2010 at 08:45

    Since only the SS´s et al, a relative small group, argues that MP was tight in late 2008, Bernanke concentrates in defending himself from the more popular accusation that MP was loose in 2002-04.
    It would be much harder for him to defend himself from the accusation that MP was tight in 2008, especially because he KNOWS that´s true. To repeat, from Bernanke´s 1999 article on Japanese MP:
    “Among the more important monetary-policy mistakes were 1) the failure
    to tighten policy during 1987-89, despite evidence of growing
    inflationary pressures, a failure that contributed to the development
    of the “bubble economy”; 2) the apparent attempt to “prick” the stock
    market bubble in 1989-91, which helped to induce an asset-price crash;
    and 3) the failure to ease adequately during the 1991-94 period, as
    asset prices, the banking system, and the economy declined
    “But I also believe that there is compelling evidence that the
    Japanese economy is also suffering today from an aggregate demand deficiency. If monetary policy could deliver increased nominal spending, some of the difficult structural problems that Japan faces would no longer seem so difficult”.
    “The argument that current monetary policy in Japan is in fact quite accommodative rests largely on the observation that interest rates are at a very low level. I do hope that readers who have gotten this far will be sufficiently familiar with monetary history not to take seriously any such claim based on the level of the nominal
    interest rate. One need only recall that nominal interest rates remained close to zero in many countries throughout the Great Depression, a period of massive monetary contraction and deflationary pressure”.
    And to contradict his recent answer to DeLong´s question (Why haven´t you adopted a 3% inflation target?) Bernake in 1999 regarding Japan wrote:
    “With respect to the issue of inflation targets and BOJ
    credibility, I do not see how credibility can be harmed by
    straightforward and honest dialogue of policymakers with the public.
    In stating an inflation target of, say, 3-4%, the BOJ would be giving the public information about its objectives, and hence the direction in which it will attempt to move the economy”.

  2. Gravatar of thruth thruth
    5. January 2010 at 08:49

    Marcus: those quotes are truly disturbing. All I can think is that there must be a ton of institutional friction at the Fed.

  3. Gravatar of Marcus Nunes Marcus Nunes
    5. January 2010 at 08:59

    Maybe the “frictions” have to do with the style of leadership. greenspan usually got his way. Remember the second half of the 90´s when everyone, Krugman included, was asking for rate increases because growth was above “potential” and Greenspan held firm arguing that productivity was rising?
    Being more of a “consensus” guy, Bernanke was not forceful enough with his arguments and lost the day!

  4. Gravatar of Marcus Nunes Marcus Nunes
    5. January 2010 at 09:02

    The whole Beranke paper is well worth reading (sounding very much like SS):

  5. Gravatar of Tom Hickey Tom Hickey
    5. January 2010 at 09:44

    Seems to me that what the Fed really missed was the looming balance sheet problem, due to their lax oversight as well as that of other regulators, not to mention the predatory practices and fraud in the mortgage market that was being amplified through securitization.

    I don’t see that there was or is too much the Fed can do through lowering rates or providing liquidity when what was and is still needed is clearing of toxic loans and additional bank capital to make up for the losses.

    Yes, adding liquidity and guarantees when the system locked up prevented a freeze of the system, but what saved the system was the faux stress tests that declared the banks essentially OK and FASB’s changing the accounting rules to mask insolvency. This crisis was Enron amplified.

    But this fix is temporary. The problems still remain. This is not a business cycle, but a financial cycle driven by Ponzi finance (in Minsky’s sense) and debt deflation. This has to be wrung out of the system.

    Beyond that, we are left with the TBTF doctrine and Congress is already working on pre-approving a 4 trillion bailout package for the next crisis. Seems like the TBTF’s are already eyeing ways that they can get their hands on that pot of gold after this one was so easy.

    Bernanke avoided mentioning any of this.

  6. Gravatar of David Pearson David Pearson
    5. January 2010 at 09:50

    If I’m not mistaken, the 10yr TIPS spread is at the same level as before the recession started.

    My question, which I’ve asked before, is, at what point would you change your mind that the Fed is “too tight” in its current policy?

    -2.7% TIPS spread (its high)

    Of course you would rather focus on the 1yr TIPS spread, but that is not the question. The question is, all else equal, is there a level of long-term inflation expectations that would start to trouble you, to signal that the Fed must act to restrain them?

    Economists are understandably loathe to go on record with such things, but to the extent that you are making a policy recommendation (the Fed should “loosen” now), its helpful to policy makers to anticipate when you might change that recommendation.

    The 10yr TIPS spread is only 2.4% now, so the set of levels above give you plenty of room before they are triggered.

    How about it?

  7. Gravatar of JimP JimP
    5. January 2010 at 09:56

    None of this really explains WHY Bernanke failed to do in 2008 what he clearly and obviously KNEW he should do? Why?

    Except thruth on “institutional friction”. Bernanke knew he did not have the votes. And he did not have an Obama statement openly supporting an aim for higher inflation. Neither of them had the backbone to fight for the correct policy. And they don’t have the backbone now either.

    Personally I see no other explanation.

  8. Gravatar of JimP JimP
    5. January 2010 at 09:59

    They are managing decline, as Kagan said in that article. They quiver and shake at the thought of the ever powerful China.

    They expect decline – and they are delivering it.

  9. Gravatar of thruth thruth
    5. January 2010 at 09:59

    David Pearson: “which I’ve asked before, is, at what point would you change your mind that the Fed is “too tight” in its current policy?”

    presumably as soon as NGDP is back to its trend path.

  10. Gravatar of David Pearson David Pearson
    5. January 2010 at 10:10


    You might be missing the point of my question. Of course a primary target for Scott will be NGDP, and an underlying assumption is that long term inflation rates will be LESS volatile as the Fed’s NGDP target gains credibility. So my question is, at what point does that underlying assumption become invalid? Perhaps the TIPS expectations range I gave above is too small to invalidate the assumption. How about 5%? Is there ANY long term inflation expectations level that would cause Scott to abandon his primary target of NGDP? That is the question, not whether NGDP should be the target.

    Of course, you can always make the question go away by saying, “I don’t expect long term expectations to become unanchored.” Fine, then all the more reason to become concerned if they do. Presumably, then the lower your forecast for volatility in expectations, the less tolerant you should be of exceeding that volatility.

  11. Gravatar of David Stinson David Stinson
    5. January 2010 at 10:18

    Hi Scott.

    Two questions:

    a) How do you define a liquidity trap?

    b) Do you believe we are still in a liquidity trap?

    The reason that I ask is that we don’t currently seem to be in what the textbooks would define as a liquidity trap: i) the stock market has obviously rallied significantly implying a declining demand for money, and ii) I gather that there has been a significant shift on the part of households from equity holdings to corporate debt. Both suggest a willingness to hold assets whose value is sensitive to interest rates.

    Finally, we often read that a steep yield curve is a good predictor of economic recovery and we currently have a steep yield curve. What should a yield curve look like in a liquidity trap?

  12. Gravatar of David Stinson David Stinson
    5. January 2010 at 10:18

    Actually, I guess that was three questions.

  13. Gravatar of David Stinson David Stinson
    5. January 2010 at 10:58

    Regarding David Pearson’s points, I wonder whether one should expect that the inflation risk premium has widened considerably since Sept 2008. Consequently, the TIPS spread, particularly at the five and ten year durations, would reflect not only the market’s expected value of inflation but also the widened risk premium.

  14. Gravatar of jj jj
    5. January 2010 at 11:32

    Scott presents a clear difference in FOMC response using forward- and backward-looking Taylor rules, but I don’t see why the different FOMC response should cause a different market response. Under the backwards rule, the market knew (in 2008) that if holding the rate caused deflation, then in 2009, the FOMC would look backwards at 2008 and cut the rate to counter the deflation. And wouldn’t that expectation prevent the deflation in the first place?

    I’d like to see the mathematical analysis comparing forward-looking, backward-looking, and level targeting rules. I’m sure it’s been done; anybody know where?

  15. Gravatar of The Bernanke Speech « The Everyday Economist The Bernanke Speech « The Everyday Economist
    5. January 2010 at 11:44

    […] the actions of the Federal Reserve in the early part of the decade. Scott Sumner makes a keen observation: Bernanke’s explanation for the Fed’s actions in 2002 show exactly how monetary policy failed […]

  16. Gravatar of pushmedia1 pushmedia1
    5. January 2010 at 13:18

    jj, Woodford’s “Interest and Prices” is a good reference.

  17. Gravatar of Don the libertarian Democrat Don the libertarian Democrat
    5. January 2010 at 13:23

    This is for Robin Hanson: I agree with you.

  18. Gravatar of scott sumner scott sumner
    5. January 2010 at 17:40

    Marcus, That’s a great article. Thanks. I got two new posts out of it. It did sound exactly like me.

    Thruth, I also found it disturbing.

    Tom Hickey, You said;

    “This is not a business cycle, but a financial cycle driven by Ponzi finance (in Minsky’s sense) and debt deflation. This has to be wrung out of the system.”

    This is where I differ from others. I think it is a business cycle, created by tight money.

    David Pearson, Very good question.

    1. I would continue to argue for more monetary ease even if I didn’t think we needed more AD. Even if expected NGDP growth was adequate, I’d advocate a combination of greater monetary ease and offsetting fiscal tightening.

    2. You are right, I do regard the next two years as the problem. I am not surprised that ten year TIPS are at 2.4%, that’s roughly the average inflation rate over the last couple decades. Now that the 2008-09 deflation is over, I think that is a reasonable forecast. But I still think we need to get closer to the old NGDP trend line (not all the way back.)

    At some point in 2010 I will probably shift emphasis from telling the Fed what to do, and start focusing on getting a more intelligent long range policy regime. That would have three components: Targeting the forecast, targeting NGDP, and level targeting. Right now the Fed isn’t doing any of those three things. So that will be my focus once I stop bashing them for tight money. But even Bill Woolsey, who favors only 3% NGDP growth, has advocated faster growth to get back to the trend line. Obviously at some point that argument loses force. In the 1999 article Bernanke recommended that the BOJ try to raise prices back to pre-recession levels. Today he probably would no longer make that recommendation, as the Japanese GDP deflator is about 15% below 1994 levels.

    In a perfect world I’d recommend the Fed aim for fast (around 6% to 8%) NGDP growth for a year or two, then tighten to get inflation expectations down closer to 2%. But they need an explicit target path to do this, and so far they refuse to set one.

    JimP, You might be right.

    David#2, Yes I did understand that was the implicit assumption in your question. I probably didn’t give you the answer you wanted, so let me try to be more specific. I’d throw in the towel if 5 year TIPS spreads reached around 2.75% or 3%. I think that would imply pretty fast NGDP growth, maybe too fast. Of course long run I’d try to get NGDP growth down to around 5% at most, which would anchor inflation. If you asked me next year I’d cite an even lower number as being worrisome, maybe 2.5% over 5 years.

    The wild card here is long term fear of inflation caused by monetization of the debt. Thus tight money in 1929-33 led to big debts which led to higher inflation after WWII. You could argue that a little monetary ease now could boost the economy, lower the budget deficit and reduce long term inflation fears. It’s certainly not a hypothesis policy makers should complacently rely on, but it’s possible.

    David#3: You can have a steep yield curve in a liquidity trap. I would define it as near zero T-bill yields, although I have always thought the term ‘trap’ was misleading, because they is really no trap at all. Monetary policy is still highly effective–you just can’t lower your interest rate target.

    I’ll take the others later.

  19. Gravatar of scott sumner scott sumner
    5. January 2010 at 18:12

    I see I got the two David’s mixed up in my previous reply.

    David Stinson, Yes, I think the risk premium has widened. Some of it also reflects rising oil prices that haven’t fully shown up in the TIPS prices (they use lagged CPI data.)

    jj, That’s a very good point, and it shows that the forward-backward distinction isn’t the only problem. There must be something else that goes wrong once rates hit zero. Perhaps the Fed has forgotten all those research papers on how to ease in a liquidity trap, or they feel that effective actions would be politically controversial. But I agree with your logic, my explanation can’t be the whole story.

    Don, I find your comment rather cryptic. Why here and not on Hanson’s blog?

  20. Gravatar of Don the libertarian Democrat Don the libertarian Democrat
    5. January 2010 at 19:06


    I’m referring to this post:

    I’m showing that it is possible for a commenter to simply agree with the blogger ( you ), as I do in this case ( and as I have done many times before on blogs, although sometimes I simply e-mail the blogger so as not to take up posting space ). Whether this qualifies as interesting or not is another question. Given that he wants comments to be rated/graded, I’m not long for the commenting-verse in any case.

    Take care,


  21. Gravatar of scott sumner scott sumner
    6. January 2010 at 14:40

    Don the Libertarian Democrat, Thanks for pointing that out. Even before blogs, I noticed that people were ruder in emails than face to face. I’m sure psychologists have an explanation. I try to be polite, but I believe that I end up being a bit ruder in comments than I am face to face. It is easy to become exasperated “I’ve answered that same question 100 times!” I forget that this time might be a newcomer who has never read the blog.

Leave a Reply