This paper was prepared for John B. Taylor and Michael Woodford, Editors

21

Kusko, Poterba and Wilcox (1997) showed, using data on 10,000 401k plan participants in a

manufacturing firm, found that barely 20% of participants directed any of their own balances into an

S&P index fund, while nearly 25% of participants directed all of their discretionary balances into a

fund invested completely in the own company stock.

21

stock market in one’s own country, as would be suggested by economic theory.

21

A prominent reason that most people appear apathetic about schemes to protect them

from price level uncertainty in nominal contracts is that they just do not seem to think that

past actual price level movements are any indicator of future uncertainty.  In a questionnaire

I distributed (1997a) to a random sample from phone books in the U.S.A. and Turkey, the

following question was posed:

We want to know how accurately you think that financial experts in

America (Turkey) can predict the price level in 2006, ten years from now.

Can you tell us, if these experts think that a “market basket” of goods and

services that the typical person buys will cost $1,000 (100 million TL) in

2006, then you think it will probably actually cost:

(Please fill in your lower and upper bounds on the price:)

Between $__________ (TL) and $__________ (TL)

The median ratio between high and low was 4/3 for U.S. respondents and 3/2 for

Turkish respondents.  Only a few respondents wrote numbers implying double- or triple-

digit ratios, even in Turkey.  The ratios not far from one that most respondents revealed

would seem to suggest excessive confidence in the predictability of price levels.  Note that

in Turkey the CPI increased three-fold between 1964 and 1974, 31-fold between 1974 and

1984, and 128-fold between 1984 and 1994.  But, Turkish respondents appear to connect the

price level movements with prior political and social events that may be perceived as having

largely predicted the price movements, events that are themselves not likely to be repeated

in the same way.  While these people have apparently learned to take certain steps to protect

themselves from price level uncertainty (such as not investing in long-term nominal bonds),

they do not appear to have a well-developed understanding of the potential uncertainty of

the Turkish Lira that would allow them to deal systematically with such uncertainty.  For

example, they have shown relatively little interest in government indexed bonds.

Magical Thinking

B. F. Skinner (1948) in what is now regarded as a classic experiment fed starved

experimental pigeons small quantities of food at regular fifteen-second intervals with no

dependence whatsoever on the bird’s behavior.  Even though the feeding was unaffected by

their behavior, the birds began to behave as if they had a “superstition” that something in

their behavior caused the feeding (see also McFadden, 1974).  Each pigeon apparently

conditioned itself to exhibit a specific behavior to get the food, and because each bird

---

22

exhibited its characteristic behavior so reliably, it was never deconditioned:

One bird was conditioned to turn counter-clockwise in the cage, making

two or three turns between reinforcements.  Another repeatedly thrust its

head into one of the upper corners of the cage.  A third developed a  “

tossing”  response, as if placing its head beneath an invisible bar and lifting

it repeatedly.... (1948, p. 168)

Arbitrary behaviors that are so generated are referred to with the term “magical thinking”

by psychologists.

A wide variety of economic behaviors are likely to be generated in exactly the same way

that the arbitrary behaviors of the pigeons are generated.  Thus, for example, firms’

investment or management decisions that happened to precede increases in sales or profits

may tend to be repeated, and if this happens in a period of rising profits (as when the

economy is recovering from a recession) the notion that these decisions were the cause of

the sales or profit increase will be reinforced.  Because firms are similar to each other and

observe each other, the magical thinking may be social, rather than individual, and hence

may have aggregate effects.

Roll (1986), with his hubris hypothesis concerning corporate takeovers, argued that

managers of bidder firms may become overconfident of their own abilities to judge firms,

because of their luck in their first takeovers.  This overconfidence can cause them to overbid

in subsequent takeover attempts.

 

The tendency for speculative markets to respond to certain news variables may be

generated analogously.  The U.S. stock market used often to be buoyed by positive news

about the economy, but in recent years it appears to tend to be moved in the opposite

direction by such news.  This new “perverse” movement pattern for the stock market is

sometimes justified in the media by a theory that the good news will cause the Federal

Reserve to tighten monetary policy and that then the higher interest rates will lower the

stock market.  But the whole belief could be the result of a chain of events that was set off

by some initial chance movements of the stock market.  Because people believe these

theories they may then behave so that the stock price does indeed behave as hypothesized,

the initial correlations will persist later, and thereby reinforce the belief.

Quasi-Magical Thinking

The term quasi-magical thinking, as defined by Shafir and Tversky (1992), is used to

describe situations in which people act as if they erroneously believe that their actions can

influence an outcome (as with magical thinking) but in which they in fact do not believe

this.  It includes acting as if one thinks that one can take actions that will, in effect, undo

what is obviously predetermined, or that one can change history.

For example, Quattrone and Tversky (1984) divided subjects into a control and

experimental group and then asked people in both groups to see how long they could bear

to hold their hands in some ice water.  In the experimental group subjects were told that

---