We close this section with a brief mention of “money illusion,” the con-
fusion between real and nominal values first discussed by Fisher (1928),
and more recently investigated by Shafir et al. (1997). In financial markets,
Modigliani and Cohn (1979) and more recently, Ritter and Warr (2002),
have argued that part of the variation in P/D ratios and returns may be due
to investors mixing real and nominal quantities when forecasting future
cash flows. The value of the stock market can be determined by discounted
real cash flows at real rates, or nominal cash flows at nominal rates. At
times of especially high or especially low inflation though, it is possible that
some investors mistakenly discount realcash flows at nominalrates. If in-
flation increases, so will the nominal discount rate. If investors then dis-
count the same set of cash flows at this higher rate, they will push the value
of the stock market down. Of course, this calculation is incorrect: the same
inflation which pushes up the discount rate should also push up future cash
flows. On net, inflation should have little effect on market value. Such real
versus nominal confusion may therefore cause excessive variation in P/D
ratios and returns and seems particularly relevant to understanding the low
market valuations during the high inflation years of the 1970s, as well as
the high market valuations during the low inflation 1990s.
4.2.2.preferencesBarberis, Huang, and Santos (2001) show that a straightforward extension
of the version of their model discussed in section 4.1 can explain both the
equity premium and volatility puzzles. To do this, they appeal to experi-
mental evidence about dynamic aspects of loss aversion. This evidence sug-
gests that the degree of loss aversion is not the same in all circumstances
but depends on prior gains and losses. In particular, Thaler and Johnson
(1990) find that after prior gains, subjects take on gambles they normally
do not, and that after prior losses, they refuse gambles that they normally
accept. The first finding is sometimes known as the “house money effect,”
reflecting gamblers’ increasing willingness to bet when ahead. One interpre-
tation of this evidence is that losses are less painful after prior gains because
they are cushioned by those gains. However, after being burned by a
painful loss, people may become more wary of additional setbacks.^22
To capture these ideas, Barberis, Huang, and Santos (2001) modify the
utility function in Eq. (11) to
E (15)
C
t t bCtttX z
t01
01
0 1ρ
γυγ
γ−
+
=∞−+∑ ̃(,).
34 BARBERIS AND THALER
(^22) It is important to distinguish Thaler and Johnson’s (1990) evidence from other evidence
presented by Kahneman and Tversky (1979) and discussed in section 3, showing that people
are risk averse over gains and risk-seeking over losses. One set of evidence pertains to one-shot
gambles, the other to sequences of gambles. Kahneman and Tversky’s (1979) evidence suggests