>

Econ 100b

Created 4/30/1996
Go to
Brad De Long's Home Page


Lecture Seventeen

Rules vs. Discretion; the Great Depression
(Economics 100b; Spring 1996)

Professor of Economics J. Bradford DeLong
601 Evans, University of California at Berkeley
Berkeley, CA 94720
(510) 643-4027 phone (510) 642-6615 fax
delong@econ.berkeley.edu
http://www.j-bradford-delong.net

February 26, 1996


Administration
Rules vs. Discretion
Great Depression of 1929-1933
Spending Hypothesis
Money Hypothesis
Destabilizing Deflation
Next Time: Using the IS-LM Model for Reading the Wall Street Journal


Administration

My conversations with David Romer...

Rules vs. Discretion

"Discretion" (or "authorities"): do what the Federal Reserve feels like.
"Rules": money growth = 3% + (Unemployment rate - 6%)
More realistic today: rule is: interest rate = 2% + inflation - (unemployment - 6%)

Reasons for rules:

Reasons for authorities:
We don't know enough to write a complete rule--there are always surprises
Better to choose authorities and give them the proper incentives...

Argument still goes on, with no sign of ending. But some facts (or semi-consensuses):


Great Depression

By far the most extraordinary economic disaster to have befallen America was the Great Depression of 1929-1939. From the unemployment rate rose from 3.2 to 25.2 percent. Real GDP fell by 30 percent; gross investment fell by more than 85 percent--and the nominal interest rate i (on prime private commercial paper) fell from 5.9 percent in 1929 to an average of 1.7 percent in 1933. On pages 275-281 Greg Mankiw runs through the Great Depression, dividing accounts of how the Great Depression happened into two groups--the "Spending Hypothesis" and the "Money Hypothesis".

Spending Hypothesis

The most important fact about the Great Depression as seen by proponents of the "Spending Hypothesis" is that throughout the 1930s, as production fell and unemployment rose, nominal interest rates continued to fall. A recession that is produced by monetary stringency--by a sudden tightening of monetary policy, as was the case in 1979-1983--sees very high nominal interest rates.

Since the Great Depression was not caused by monetary stringency--since, at least as measured by interest rates, there was no monetary stringency (and, in fact, the "real" liquidity of the economy did not fall between 1929 and 1933 because the price level fell as fast as the money stock. 1933's real money stock was less than 4% below 1929's)--it must have been caused by a spending shock.

Possible candidates for a spending shock:

But this is not fully satisfactory. Why should investment suddenly collapse by more than 85%?


Hence, Money Hypothesis

Best known advocates are Friedman and Schwartz, who say that monetary forces "caused" the Great Depression.

Using the IS-LM model, we might interpret the money hypothesis as explaining the Depression by a contractionary shift in the LM curve. Seen in this way, however, the money hypothesis runs into two problems.

So general conclusion: when Friedman and Schwartz write that monetary forces "caused" the Great Depression, they are using a peculiar definition of "cause": they think that monetary policy could have averted the Great Depression, but did not.


Destabilizing Deflation

But easy to see attractiveness of money hypothesis. Spending hypothesis is, what? That investment demand simply collapsed one day, for no reason, by more than 80%? Not fully satisfactory.

The line of explanation that I think is most likely to be true--that I wrote my first and third articles ever on, in fact--hinges on the fact that the interest rate in the "LM" curve is a nominal and the interest rate in the IS curve is a "real" interest rate. And here let me proudly announce my complete and enthusiastic agreement with Greg Mankiw's treatment of the issue, which I think is superb on pages 278-281.

From 1929 to 1933 the U.S. price level fell by a quarter--and it was this deflation that turned what in 1930 or early 1931 was a typical economic downturn into an unprecedented Great Depression.

So consider:

(1) Y = C(Y - T) + I(i - E(p)) + G

(2) M/P = L(i, Y)

and let's take an equilibrium, and suddenly shift E(p)--for which Greg Mankiw's textbook writes a little Greek letter "pi" with an "e" superscript. What do we find? Well, if we have plotted the nominal interest rate "i" on the vertical axis, we find we have suddenly shifted the IS curve down by a lot. If we have plotted the real interest rate "r" on the vertical axis, we find we have suddenly shifted the LM curve up by a lot.

What sudden shifts in expected inflation (or deflation) do is they break the link between shifts in i and shifts in r.

Thus between 1929 and 1933, money can be "loose" in the sense of nominal interest rates being low, yet credit can be "tight" in the sense of very high real interest rates--because of expected deflation.

In a sense the economy has become a snake that eats its own tail: there is deflation because people fear a Great Depression, yet the only reason to fear an Great Depression is the existence of general deflation.




Next Time: Using the IS-LM Model for Reading the Wall Street Journal


>

Econ 100b

Created 4/30/1996
Go to
Brad De Long's Home Page


Professor of Economics J. Bradford DeLong, 601 Evans
University of California at Berkeley
Berkeley, CA 94720-3880
(510) 643-4027 phone (510) 642-6615 fax
delong@econ.berkeley.edu
http://www.j-bradford-delong.net/