Week 3:
Money illusion (Shafir 1997)
Goal: the money illusion phenomenon is the tendency of people to think in nominal rather than real
monetary values when considering different options. This paper dives into this phenomenon using
survey questions regarding people’s reactions to variations in inflation and prices. It therefore sheds
a light on the psychological principles that underly money illusion. It ultimately proposes that people
often think about economic transactions in both nominal and real terms, however that the money
illusion arises from an interaction between these two different interpretations resulting in a bias
towards a nominal evaluation. The paper illustrates the interaction between money illusion and
other decision factors such as loss aversion, risk attitudes and fairness concerns to show the
importance of this illusion in the economy.
There are three classes of anomalous observations that point towards the effect of money illusion on
people’s choices and the economy:
1. Price-/wage stickiness: the first observation is that prices and wages are ‘sticky’, meaning
that they don’t change as fast as would be expected in reaction to changes in
quantities/markets.
2. Absence of indexed contracts/laws: the second observation is the general lack of indexed
contracts/laws in many places where economic theory would predict them to be. Indexing
contracts/laws is the practice of linking these to a nominated price index (such as inflation)
and periodically adjusting for them.
3. Confusion in common discourse and media reports : in common discourse and media reports
it is commonly found that real monetary values are not taken into consideration, but
nominal monetary values are used. This phenomenon sometimes disappears when inflation
is high (thus the difference between real and nominal monetary values is great), but even
then it sometimes persists.
Multiple representations: a psychological account: Next follows an explanation from psychology for
the money illusion phenomenon.
Cognitive psychology research has shown that alternative representations of the same
situation/information can lead to systematically different responses (such as framing effect). Tt is
more useful to frame projects in terms of final assets, as opposed to changes in wealth because then
these framing effects would disappear. For example, when the same problem is framed as a change
of gaining 15.000 or losing 10.000, as opposed to a chance of 240.000 and a chance of 265.000
(status quo: 250.000) people would more often avoid the risky option and stick with the status quo
even though the expected value of the risky option is higher, because they then experience loss
aversion. People tend to adopt the frame that is presented to them (e.g. wealth versus change in
wealth) and proceed to evaluate the options they have using that frame.
In addition to this, in other situations people actually use multiple representations at the same time,
as opposed to a single representation. In this situation the actual response is a mixture of the
different representations, weighted by its relative salience. The paper suggests that this mechanism
underlies the money illusion phenomenon. Economic transactions can be represented in nominal or
real terms (thus multiple representations are possible). Due to the nominal representation being
simpler, more salient and often sufficing in the short run people tend to think of economic
transactions in nominal terms and leaving less room for the representation in real terms to affect
decision-making.
, This leads to people finding a 2% wage increase in times of 4% inflation less aversive than a simple
2% wage cut, seeing as the nominal representation is more salient and affects decision-making
more.
Furthermore people base their judgment on reference points and in situations where this reference
point can often be nominal (e.g. historical buying price of a house) people anchor onto this reference
point and loss aversion occurs relative to a reference point, without keeping in mind the real values.
Ultimately the reliance on a nominal evaluation is due to the ease, universality and salience of the
nominal representation.
Experimental studies on the money illusion: Next follows a section of different experimental studies
on the money illusion in different contexts using survey questions.
Money illusion in earnings: the degree of satisfaction people have with their income
depends not only on its buying power, but among other things on how it compares with an
earlier salary or with the salaries of co-workers (both are reference points).
For example, when presented with a situation in which two equally educated people have a
job at a firm, person A gets 36.000 at a firm with an average starting salary of 40.000 and
person B gets 34.000 in a firm with an average starting salary of 30.000, people judge person
B to have more job satisfaction and person A to be more likely to leave the firm, even
though person A has a higher absolute salary.
This type of observation is also found in relation to nominal vs real monetary values of
salaries. Three different groups were posed the same situation and group 1 had to rate the
two raises on purely economic terms, group 2 had to rate who they thought would be
happier and group 3 had to indicate who was more likely to leave their present job for
another position. The problem is as follows:
Group 1 (for which the economic terms are emphasized) rated Ann as doing better in
economic terms than Barbara (which is actually true).
Group 2 rated Barbara as being more happy with her job.
Group 3 rated Ann as being more likely to leave her job for another job.
What we can thus observe is that when economic terms are emphasized, the majority of
people correctly evaluate the scenario in terms of real rather than nominal terms (which
groups 2 and 3 don’t do, their evaluation rests upon nominal evaluation).
Money illusion in transactions: these studies evaluate people’s assessment of specific
economic transactions. In several experiments using transactions with differences in real vs
nominal losses/gains, people preferred the situations with nominal gains and didn’t pay
Money illusion (Shafir 1997)
Goal: the money illusion phenomenon is the tendency of people to think in nominal rather than real
monetary values when considering different options. This paper dives into this phenomenon using
survey questions regarding people’s reactions to variations in inflation and prices. It therefore sheds
a light on the psychological principles that underly money illusion. It ultimately proposes that people
often think about economic transactions in both nominal and real terms, however that the money
illusion arises from an interaction between these two different interpretations resulting in a bias
towards a nominal evaluation. The paper illustrates the interaction between money illusion and
other decision factors such as loss aversion, risk attitudes and fairness concerns to show the
importance of this illusion in the economy.
There are three classes of anomalous observations that point towards the effect of money illusion on
people’s choices and the economy:
1. Price-/wage stickiness: the first observation is that prices and wages are ‘sticky’, meaning
that they don’t change as fast as would be expected in reaction to changes in
quantities/markets.
2. Absence of indexed contracts/laws: the second observation is the general lack of indexed
contracts/laws in many places where economic theory would predict them to be. Indexing
contracts/laws is the practice of linking these to a nominated price index (such as inflation)
and periodically adjusting for them.
3. Confusion in common discourse and media reports : in common discourse and media reports
it is commonly found that real monetary values are not taken into consideration, but
nominal monetary values are used. This phenomenon sometimes disappears when inflation
is high (thus the difference between real and nominal monetary values is great), but even
then it sometimes persists.
Multiple representations: a psychological account: Next follows an explanation from psychology for
the money illusion phenomenon.
Cognitive psychology research has shown that alternative representations of the same
situation/information can lead to systematically different responses (such as framing effect). Tt is
more useful to frame projects in terms of final assets, as opposed to changes in wealth because then
these framing effects would disappear. For example, when the same problem is framed as a change
of gaining 15.000 or losing 10.000, as opposed to a chance of 240.000 and a chance of 265.000
(status quo: 250.000) people would more often avoid the risky option and stick with the status quo
even though the expected value of the risky option is higher, because they then experience loss
aversion. People tend to adopt the frame that is presented to them (e.g. wealth versus change in
wealth) and proceed to evaluate the options they have using that frame.
In addition to this, in other situations people actually use multiple representations at the same time,
as opposed to a single representation. In this situation the actual response is a mixture of the
different representations, weighted by its relative salience. The paper suggests that this mechanism
underlies the money illusion phenomenon. Economic transactions can be represented in nominal or
real terms (thus multiple representations are possible). Due to the nominal representation being
simpler, more salient and often sufficing in the short run people tend to think of economic
transactions in nominal terms and leaving less room for the representation in real terms to affect
decision-making.
, This leads to people finding a 2% wage increase in times of 4% inflation less aversive than a simple
2% wage cut, seeing as the nominal representation is more salient and affects decision-making
more.
Furthermore people base their judgment on reference points and in situations where this reference
point can often be nominal (e.g. historical buying price of a house) people anchor onto this reference
point and loss aversion occurs relative to a reference point, without keeping in mind the real values.
Ultimately the reliance on a nominal evaluation is due to the ease, universality and salience of the
nominal representation.
Experimental studies on the money illusion: Next follows a section of different experimental studies
on the money illusion in different contexts using survey questions.
Money illusion in earnings: the degree of satisfaction people have with their income
depends not only on its buying power, but among other things on how it compares with an
earlier salary or with the salaries of co-workers (both are reference points).
For example, when presented with a situation in which two equally educated people have a
job at a firm, person A gets 36.000 at a firm with an average starting salary of 40.000 and
person B gets 34.000 in a firm with an average starting salary of 30.000, people judge person
B to have more job satisfaction and person A to be more likely to leave the firm, even
though person A has a higher absolute salary.
This type of observation is also found in relation to nominal vs real monetary values of
salaries. Three different groups were posed the same situation and group 1 had to rate the
two raises on purely economic terms, group 2 had to rate who they thought would be
happier and group 3 had to indicate who was more likely to leave their present job for
another position. The problem is as follows:
Group 1 (for which the economic terms are emphasized) rated Ann as doing better in
economic terms than Barbara (which is actually true).
Group 2 rated Barbara as being more happy with her job.
Group 3 rated Ann as being more likely to leave her job for another job.
What we can thus observe is that when economic terms are emphasized, the majority of
people correctly evaluate the scenario in terms of real rather than nominal terms (which
groups 2 and 3 don’t do, their evaluation rests upon nominal evaluation).
Money illusion in transactions: these studies evaluate people’s assessment of specific
economic transactions. In several experiments using transactions with differences in real vs
nominal losses/gains, people preferred the situations with nominal gains and didn’t pay
