This paper is based on a long-term research program with Rachel Kranton on the implications of identity

(2001).  Following many simple quantity theory models (Akerlof and Yellen (1985), Ball and Romer (1989),

Blanchard and Kiyotaki (1987)) the payoff of each individual subject in each trial depended on the price he chose,

the average price chosen by other subjects, and the money supply.  An equi-proportionate change of all prices and

the money supply would result in no change in the real payoffs, so that the equilibrium is money-neutral.   After a

number of initial trails in each individual experiment, the quantity of money changed.  Fehr and  Tyran found that

the approach to the new equilibria after the change in the money supply differed according to whether payoffs were

denoted in real or in nominal terms.  Money illusion complicated the approach to the new equilibria not just because

individual subjects themselves might have money illusion, but especially because they might impute it to others. 

More than half of their subjects thought that others would interpret high nominal payoffs as indicators of high real

payoffs.  The study shows not only the role of nominal framing in determining the nature of equilibria, but that small

amounts of nominal framing, even if imputed to others, can result in significant violations of money neutrality. 

Significantly neither frictions nor preferences could be implicated in the nature of this violation of monetary

neutrality.  There were no frictions, and the experimenters, by giving the payoffs, have determined that preferences

are only denoted in real terms.  

40

High Inflation.  Natural rate theory has not only limited economists’ theoretical

imaginations regarding trade-offs between inflation and unemployment when inflation is low; it

has also limited their theoretical imaginations when inflation is high.  At high inflation the

psychological reaction to wage increases that just match inflation may be quite different from the

reaction when inflation is low.  In that case employees are fearful of it and watch it closely. 

Contrary to respondents’ reactions to Barbara and Ann, and to the responses from Shiller’s

questionnaires, they may feel worse off if they receive a wage increase that just matches

inflation.  In this case higher inflation will shift the Phillips Curve outwards, not inwards.  The

usual natural rate models with rational expectations (Barro and Gordon (1983) and Rogoff

(1987)) will then have sold credibility short: it will be even more important in the presence of

overreaction than in the presence of rational expectations.  The exploration of such areas has

been unfortunately shut off by macroeconomists’ concentration on models without overreaction. 

Such overreaction gives added reason why economic policy should aim for moderate inflation

targets. 

Prices

---

57

See also Blinder and Choi (1990) and Blinder, Canetti, Lebow and Rudd (1998).  

41

In addition to long-run trade-offs between inflation because of employees’ preferences

regarding the nominal value of their wages (or of their wage increases), such trade-offs may

occur because customers have views regarding what nominal prices should be.   Models by Iwai

(1981), Rotemberg (1982), and Caplin and Leahy (1991) produce a trade-off between long-run

inflation and long-run unemployment.  Their key assumption is that there are real costs to

nominal price changes.  There would be no such trade-off if instead there were real costs to real

price changes.  In that case the assumptions of natural rate theory would not be violated.  These

models then pose the question why there should be such real costs from nominal price changes. 

The usual answer is that there are “menu” costs in making these changes known.  But such costs

are usually thought to be trivially small.  An alternative possibility is that customers think that

firms should not raise prices.  In that case nominal price increases, or increases of greater size,

are likely to induce some form of customer retaliation: for example, customers can switch from

one supplier to another.  Higher nominal inflation will then act as if customers have higher

elasticities of product demand, which result in lower firm mark-ups, and thereby reduces

equilibrium unemployment. 

There is evidence that suggests that firms do not like to make price changes, especially in

customer markets.  Like wage changes, price changes also agglomerate at zero.  Dennis Carlton

(1986) has shown that prices are often sticky for significant periods of time.

57

  Furthermore,

prices seem to be especially sticky in customer markets.  Alan Kackmeister has compared price

changes at the end of the 19

th

 century to such changes at the end of the 20

th

 Century.  Price

changes of specific goods at retail stores were recorded over a 28-month period from June 1889

---

58

I derive this result from Kackmeister’s data in the following way.  He finds that in the 19

th

 century that

only 5 percent of items changed their prices per month.  This means that the average spell of constant prices would

have been 20 months (the inverse).  But that is a biased statistic for the average length of time between price changes

for an item on the shelf.  The difference between the average spell of employment or unemployment and the average

spell being experienced by an individual suggests a rule of thumb ratio for four to one.  Using this ratio as a rule of

thumb suggests that the spell between price changes averaged over the individual items on the shelf would be 80

months. 

42

to September 1891 and then revisited for the same commodities for a comparable 28-month

period from June 1997 to September 1999.  In the 19

th

 century, when customer-dealer relations

were much more personal, the average spell of constant price for an individual good was

approximately 80 months.

58

  Such constancy of prices for individual items is consistent with our

theory that the consumers had a notion of the price that they ought to pay at stores where they

are continued and knowing customers.  Kackmeister suggests that the decline in long-term

customer relationships is one factor responsible for greater frequency of price change today.

Nakamura and Steinsson (2005) suggest a reason why customers think that firms should

not change prices.  The view consumer purchases as habit-forming.  Thus, by buying a particular

brand, or patronizing a particular store, consumers are putting themselves in a position where

they can be exploited.  It places the firm in a position where it can take advantage of the

consumer by raising prices.   Firms then make an implicit contract with their customers: that they

will not change their prices unjustifiably.  Since such an implicit contract is easier to make (and

enforce) regarding nominal prices than real prices, the implicit guarantee is in nominal terms. 

Nakamura and Steinsson also supply some data which suggests strikingly that firms do behave

this way.  The prices of goods in store 126 (chosen for its completeness of data) of Dominicks

Finer Foods chain indicate that many goods go on sale; but, remarkably, when they go off-sale,

they usually tend to return to the exact same nominal price.  Such behavior is consistent with the

---