« Puebla | Main | $500 a barrel for oil? »
Questions you dare not ask
Cowen's Third Law used to read "All propositions about real interest rates are wrong," so I hesitate to tread on this ground. The question is, when inflation comes, why doesn't the expectation of that inflation lead to proportional increases in nominal interest rates, thus keeping the real rate constant? The studies I've seen all indicate a less than one-to-one Fisher effect. I can think of a few hypotheses:
1. People systematically underestimate the forthcoming rate of price inflation. (Still?)
2. People have generally adaptive expectations, but they will adjust quickly and rationally to big enough jolts. And maybe inflation rises slowly, but deflations come all of a sudden. So it seems there are more periods when people's expectations are lagging an inflation than a deflation, and that will produce the data pattern stated above.
3. People have generally rational expectations in a game against nature, but they are playing a game against the Fed. The Fed is smarter than the people. And the Fed has studied Newcomb's Paradox so it can, on average, figure out when a dose of inflation will surprise people (in a positive way, of course, socially speaking). So every now and then we get these surprise bursts of inflation, but no comparable surprise bursts of deflation, which of course would not help output any. In this set-up the Fisher effect won't fully hold.
4. A Mundell-Tobin effect is operating, so real rates of return are falling because the inflation moves people out of currency and into capital.
5. The new money enters through the loanable funds market, thereby depressing real interest rates.
6. Sometimes things just don't work out the way you think they ought to.
Once you consider the tax system, you realize how much the cards are stacked against our attempt to explain this. Many people can deduct their nominal interest payments from their taxes, and that implies we should see a more than one-to-one Fisher effect from inflation. But we don't.
You'll also note that under most of these explanations the specified dose of inflation does not have a significantly negative effect on private savings. If the inflation is expected, the nominal interest rate adjusts. If the inflation is not expected, it doesn't scare off savings.
Do you have other ideas? I believe the incomplete Fisher effect is a result which holds both across time and across countries, but maybe you know the latest paper which I don't.
Posted by Tyler Cowen on December 23, 2007 at 05:46 AM in Economics | Permalink
Comments
People don't expect the inflation to continue, which is betting on a political result.
At the long term end, when that happens, your bonds rise in value, producing a payoff.
Posted by: Ron Hardin at Dec 23, 2007 5:52:15 AM
Is it possible that the expectation is that the change in inflation is only temporary. In which case, consumption is unchanged. This would require a decrease in interest rates to allow temporary expansion of the money supply. If the inflation turns out to be permanent, then interest rates would have to rise to counteract inflation, but they would lag because of the initial reversal.
Posted by: Rob at Dec 23, 2007 6:52:24 AM
What are the two first laws?
Posted by: Guan Yang at Dec 23, 2007 8:44:48 AM
I am going to reveal my ignorance here, but it would seem to me that the answer is obvious if you assume that inflation occurs because of excessive increases in money supply.
The banks are the first beneficiary of new money creation. This new money has the same value as the older sound money (for lack of better word. This enables them to lower real interest rates because they no longer have to pay as high a price for money. In my little ignorant hillbilly mind, I have always thought of this effect as being akin to seigniorage.
Maybe there are good empirical reasons for discounting this theory. If so, I await my education.
Posted by: Ape Man at Dec 23, 2007 9:12:51 AM
Apeman makes a good point, when the church bells in Puebla woke me up at 5 a.m. I realized: "I forget to mention the loanable funds channel!" Sadly the computer was in Yana's room at the time. I added it to the list.
Posted by: Tyler Cowen at Dec 23, 2007 10:15:31 AM
Suggestion for next time: Wikipedia links for Fisher effect, Newcomb's Paradox, Mundell-Tobin effect, for the benefit of the interested uninitiated.
Posted by: Rafael M at Dec 23, 2007 11:01:38 AM
Tyler, there is evidence to support the Fisher effect. See Atkins and Coe (2002, Journal of Macro), Shome et al (1988, Journal of Finance).
The crucial thing is that you have to take into account the risk premium when there is uncertainty. This could be derived from a model such as Lucas' (1978 Econometrica).
Obviously, results will depend on the methodology used, but answering your question: evidence in favour of the Fisher effect does exist.
Posted by: jpf at Dec 23, 2007 11:27:35 AM
The demand for loans falls, especially home mortgages, because people care about monthly payments, not just investment return.
Posted by: joan at Dec 23, 2007 11:29:10 AM
Two stories. First, who said real interest rates are constant? Inflation shows persistent comovements over the business cycle, and so does productivity. Second, risk premia play a far more important role than we have considered so far.
Posted by: pinus at Dec 23, 2007 12:20:51 PM
A variation of the Malthus trap?
Posted by: Wwren at Dec 23, 2007 12:31:48 PM
In order to make the appropriate analogy, let us consider the typical day of a stock:
(a) Most of the time it tricks up.
(b) Once in a while it crashes down.
When it trickles, it is beating expectations, just like how people "underestimate" price inflation.
It's not that people "underestimate" price inflation in the sense of expected value, it's just that the way we weight (a) and (b) means that so long as (b) is not happening, it appears that we are underestimating (a). Our genuinely rational weighting of (b) provides the illusion while (b) is absent.
While this may appear to be "systemic", it is simply "good math" [that makes us wobble as we walk down the hallway].
Posted by: infopractical at Dec 23, 2007 1:30:35 PM
When bonds and other contracts are written in nominal terms, high inflation means there no such thing as a long term bond. The high nominal interest rate on a bond is partly a repayment of principal, so that by the time the bond matures, it's remaining real principal is but a small fraction of the original amount lent. Since the real yield curve normally slopes up, shortening the effective terms of all loans this way implies lower observed real interest rates, i.e., nominal rates that do not go up as much as inflation does.
I don't know how large this effect is, but it is one theoretically plausible explanation for the effect you postulate.
Posted by: Jeff Hallman at Dec 23, 2007 4:15:54 PM
With regard to the theory that real interest rates don't significantly affect the savings rate, has anyone noticed that the US national savings rate is zero?
Posted by: Savings Rate at Dec 23, 2007 5:18:04 PM
If the U.S "Savings Rate" is really "Zero" then how come we accumulate more durable (and transferable) assets than any other economy this side of the Sultanate of Brunei (which is one huge SWF)?
Posted by: R. Richard Schweitzer at Dec 26, 2007 2:02:03 AM
hi,I University majoring in the legal profession.After graduation,I 徵信 the work of the strong interest.Has worked in several徵信社.Has a wealth of experience. Now I immigrants France,Hope to continue to engage in the work of徵信 credit.
now, is to wake up every day to drink 咖啡, shopping. I hope that early awareness of Boles.
thanks,thank very much.
Posted by: Tonny at Mar 18, 2008 2:40:35 AM
Posted by: Alii at Apr 3, 2008 10:28:26 PM
