Friday, December 3, 2010

Yes Tyler, Low Interest Rates Mattered Alot.

Tyler Cowen is wondering whether the Fed's low interest rates in the early-to-mid 2000s really were that important to the credit and housing boom of the early-to-mid 2000s. He is having is doubts after seeing this excerpt from Michael Woodford's JEP article:
It is popular to attribute the credit boom (at least in part) to the Federal Reserve having kept the federal funds rate “too low for too long,” but comparison of the path of the funds rate in Figure 5 with the measures of credit growth in Figure 1A shows that the increase in lending was greatest in 2006 and the first half of 2007, after the federal funds rate had already returned to a level consistent with normal benchmarks. Instead, the fact that spreads were unusually low precisely during the period of strongest growth in lending... indicates that an outward shiftof the supply of intermediation schedule XS was responsible.
A question to Tyler Cowen and Michale Woodford: aren't you just the least bit curious about the chronological ordering of the events in the above paragraph? First, the federal funds rate is held below the neutral interest rate level for an extended period. Second, spreads decline and a credit boom ensues.  Hmm? Let see, that sounds a lot like the risk taking channel of monetary policy.  Here is how Leonardo Gambacorta of the BIS summarizes this monetary transmission channel:
Monetary policy may influence banks’ perceptions of, and attitude towards, risk in at least two ways: (i) through a search for yield process, especially in the case of nominal return targets; and (ii) by means of the impact of interest rates on valuations, incomes and cash flows, which in turn can modify how banks measure risk.
In terms of the recent credit boom, this channel says that keeping short-term interest rate inordinately low was key in driving down credit spreads and spurring on the credit boom.  In fact, Woodford's colleague at Princeton, Hyun Song Shin, and his coauthor Tobias Adrian have several papers on the risk taking channel. Here is an excerpt from a WSJ article from late last year that discusses how this channel was important to the recent credit boom-bust cycle: (My bold below)
Mr. Adrian and Mr. Shin find low rates feed dangerous credit booms, and thus need to be a factor in Fed interest-rate calculations. Small additional increases in rates in 2005, they say, might have tamed the last bubble. "Interest-rate policy is affecting funding conditions of financial institutions and their ability to take on leverage," says Mr. Adrian. That, in turn, "has real effects on the economy."
For a step-by-step account of how this channel worked during the credit boom see this post by Barry  Ritholtz.  Finally, let me end with George Selgin who provides an assessment of the above paragraph by Woodford: 
Let's see: you have excessively low, even negative, rates for several years, and partly for this reason house prices rise exceedingly rapidly. That rapid appreciation in turn stimulates a greater demand for credit, so rates start to come up. At the peak of the lending boom, rates are at or near their "benchmark" levels again. Therefore low rates couldn't be to blame for the boom.

Wow. Now that I know the facts, I am really sorry to have been one of those naive economists who thought the Fed had something to do with it.


  1. But did nominal income rise above trend during the same period? Perhaps a little, but not much according to the charts I have seen. Why is that? What does that say about monetary policy, bubbles, and the best level of nominal income targeting?

    It troubles me. What does this say about Sumner's 5 percent NGDP target? Because the Fed, albeit unintentionally, kept pretty close to that for years before the crisis, and even appeared to make up for previous undershooting (e.g. following the dot-com bubble). What does this say about nominal income targeting in general?

  2. Nobody can seriously doubt that monetary policy was an important strand in the crisis. But I think at least as important is a failure of regulation.
    Which ties in to Mr.Kelly's comment on NGDP. Surely what it says is that regulation is just as important as monetary policy.

  3. Lee & ECB:

    First, A big difference between then and now is that the neutral rate is low now, while back then is was not. There was a productivity boom pushing up the neutral rate just as the Fed was lowering rates. Second, how one measures trends is tricky. One can any trend they want. Third, I agree with ECB that nominal income stabilization may at times be a necessary but not sufficient condition to prevent the excesses from building. Of course, had such a rule been in place, the eventually bust would have been far milder.

    Yes, we need radical financial reform. What do we do given it is unlikely to happen?

  4. yes radical reform indeed. But how do we get it? Pessimistically, I look to the great economist Mancur Olson for the answer: a stagnation induced by powerful rent-seeking special interest groups (the financial oligarchs in the US case)is usually only broken by extraordinary events such as a complete system failure (Soviet Union, Britain in 1970s) or war.
    Things have to get worse before they get better.

  5. David,

    Do you mean "natural" or "neutral"?

    If you mean that "the" natural rate of interest was significantly above the market rate of interest, then why did this not show up as accelerating growth in nominal expenditures?

    The Fed increased the supply of loanable funds, but it appears additional spending by on housing and whatnot was offset by lower spending on other goods, i.e. the boom appears to have been (mostly) backed by real savings!

  6. Lee,

    Income growth was in fact considerably faster between 2001 and 2007 than it had been before then--have a look at the figures from DB's earlier post on QE2.

    Also, so far as I'm aware, the profession treats the "neutral" and "natural" interest rates as two names for the same thing.

  7. Thanks George,

    I'll check it out. Perhaps I have been looking at the wrong charts, or just looking at them the wrong way. The apparent discrepancy has been bothering me for some time.

  8. I have discovered a paper called "Where the Fed Goes Wrong: The 'Productivity Gap' and Monetary Policy ". Perhaps a a couple of you may have heard of it? That may have the answers I am looking for.

  9. By the way, in case you haven't noticed yet, Jeffrey Tucker, at the Mises Institute, discovered your article on QE2 and decided that "conservatism means nothing at all". I kinda knee-jerk defended you in the comments. Again, I apologise, and hope not to have misrepresented you.

    Check it out ... or not:

  10. "First, the federal funds rate is held below the neutral interest rate level for an extended period."

    How do you define the "neutral interest rate level"? As far as I know, the neutral interest rate can only be defined with reference to the equilibrium real interest rate -- which is, of course, a real price determined by global capital markets.

    Do you know what that interest rate was? Are you sure that the Fed was below the neutral level? How can you be?

  11. The Adrian and Shin article hardly proves that the housing boom and subsequent financial crisis was the fault of real interest rates. It simply shows that more risky projects are funded when the risk-free interest rate is lower. (Revelation of the century, right?) And all this in a model where financial intermediaries never default -- the entire model is built around a VaR constraint that banks *always* must have enough capital to pay off their lenders.

    Why were these funds directed toward insanely dubious investments in housing, rather than all sectors of the economy? Why did housing prices skyrocket in Phoenix and Las Vegas despite the lack of any reasonable fundamentals to support them? "Low interest rates" are not an answer. Massive principal-agent failures and some form of market irrationality are.

    Now, it's possible that the Fed's policy interacted with these market failures in a way that ultimately worsened the crisis. But the Fed certainly did not "cause" the crisis, unless you weaken the definition of causality to such a point that any entity that indirectly accentuated other market failures is responsible for "causing" the crisis. And that is a ludicrous definition.

  12. My problems in thinking about this are that the Beckworth explanation for the boom/bust is identical to the Austrians' explanation of the 1920s. The Fed kept interest rates "artificially" low instead of allowing them to rise with the natural rate during the boom.

    Sumner has a nice post explaining the daylight between himself and Beckworth. But where is the daylight between Beckworth and the Austrians here?

  13. Re-reading Sumner's critique, it's not clear to me how Dr. Beckworth handles the contradiction-- do interest rates matter or not? As the anonymous commenter above points out-- if interest rates were "artificially" low (using ABCT terminology) how do you determine the size of the artifice? If Sumner is right, then monetary policy wasn't terribly loose. Could you do a post clearing this up?

  14. Anonymous 1 & JDApp:

    There are plenty of paper showing the real federal funds rate was below the neutral real federal funds rate. I will put up a post later showing some.

    Anonymous 2:

    Yes, there were many factors coming together to create the credit and housing boom. However, all these other contributors would have been far less consequential, if at all, had it not been for the Fed. Again, I refer you Barry Ritholtz's step-by-step explanation of how the Fed's low rates enabled the other factors:

  15. Thanks for the response (and the new post). But it doesn't clarify the issue for me. I don't understand how one can argue that interest rates are not good indicators of the stance of monetary policy and yet also argue that the Fed should have raised interest rates because they are the indicator that monetary policy was too loose.

  16. JD Tapp:

    I haven't got the second post up where this will be addressed. Here is the answer: policy interest rates by themselves are not a good indicator of monetary policy, but relative to the neutral rate they are. Thus, one cannot look just at the federal funds rate (ffr) in 2003 or 2010 and conclude whether money is tight or loose. One can, though, look to the ffr relative to the neutral ffr to conclude what the stance of monetary policy is.

    Thus, in 2003 the ffr was held below the neutral rate for a sustained time. Monetary policy was too loose. Now in 2010 the ffr is probably close to or above the neutral rate. Monetary policy is too tight.

  17. Perhaps of your interest