Return to Op-Ed page

Cass R. Sunstein, Vanity Fair, in The New Republic (March 29, 1999)

Book Review of Luxury Fever: Why Money Fails to Satisfy in an Era of Excess by Robert H. Frank

Is it a problem when people spend $17,500 for a Patek Philippe
wristwatch? Should Americans be troubled when their fellow citizens
pay $5,000 per night for a suite at Little Nell in Aspen, or $14,000
for a Hermes Kelly handbag, or $1,000 for gray flannel pants? Robert
Frank thinks so. He believes that American society is suffering from
"luxury fever." This is not simply a problem of frivolous
expenditures by rich people with more money than they can handle. In
his view, there is a much deeper difficulty. Frank believes that
excessive spending for luxury goods-spending that does not, in fact,
make people happier-shows something important about human nature,
public policy, and the general question of what promotes wellbeing.
In Frank's view, luxury spending is an example of troublesome and
illunderstood behavior in which even reasonable people make
themselves, and each other, worse off-behavior that is "smart for
one, dumb for all."

Frank wishes to break the "consumption treadmill" by preventing
people from needlessly competing with each other (through
expenditures that are literally wasteful), so that "trillions of
dollars" can be diverted to better uses, both private and public.
With the reallocation of those trillions, he argues, we can increase
investment, produce much more equality, and also fund an aggressive
welfare state, without at the same time compromising economic
growth. He thinks that a better understanding of the consumption
treadmill shows that the conflict between equality and efficiency is
greatly overstated, even illusory. Indeed, he urges that, with a
one-line change in the tax code, we can achieve more of both at the
same time. This is a provocative and unusual book. It is also a lot
of fun, and in places it is hilarious. It deserves a close reading.

Frank begins by tracing what he sees as a recent boom in spending
for luxury goods among the rich. To be sure, people have always been
willing to spend a lot on luxuries; but Frank finds a significant
change, at least "at the top of the income pyramid." In general, he
says, luxury spending has been growing four times as fast as overall
spending. Frank documents a range of more particular trends, with
dramatic increases in the purchase not only of watches and handbags
costing thousands of dollars, but also of mansions, or "trophy
homes"; of luxury trips (with per diem spending of at least $350);
of luxury cars (those selling for more than $30,000); of huge
vacation residences; of expensive cosmetic surgery (chosen by more
and more, and by younger and younger, people); of increasingly large
primary residences (with the average home built in the United States
nearly double the size of its counterpart in the 1950s) .

What accounts for these remarkable increases in extravagance?
Despite sluggish growth in the American economy in the last
quarter-century, this has been a period of terrific gains for those
near the top of the economic ladder, who are enjoying "unprecedented
prosperity." There is also the matter of an important demographic
shift: the postwar baby boomers are now entering their peak earning
years. Yet Frank lays special emphasis on "the unusually rapid
growth in the incomes of top earners in every demographic
category"-a consequence, he thinks, of the spread of what he calls
"winner-take-all markets."

These are markets in which small differences in individual
performance produce enormous differences in economic reward. They
are most familiar in the world of sports; but those markets are
increasingly pervasive, in medicine, fashion, law,journalism,
design, accounting, and elsewhere. The rise of thousands of
"winners" has helped produce not only growing inequality, but also a
boom in luxury spending. Frank discussed this particular phenomenon
at length in The Winner-Take-All Society, written with Philip Cook,
which appeared in 1995, and he builds on that discussion here.

These are facts about the expenditures of the rich, but Frank finds
similar patterns for middle-income and lowerincome earners. The
personal savings rate in the United States is down nearly 90 percent
since 1980, and consumer debt is on the rise, with dramatic
increases in credit card debt and in bankruptcy. Frank proposes a
link between this phenomenon and the practices of the rich. "It is
natural for people at all income levels to experience new desires in
the presence of others who spend more than they do. And even apart
from any changes in what we consciously desire, our individual
spending decisions are often influenced by the fact that our menu of
available choices is so strongly shaped by what others spend." Thus
most people spend more, and for more luxurious items, because of the
trends set by those at the top.

The general increase in spending for luxury goods, and the sharp
decrease in saving. might not be much of a problem if people's
expenditures actually made them happier. Yet Frank thinks that this
is not the case. These expenditures have "come at the expense of not
only lower savings and greater debt but also many other things we
value." In a central part of his argument, he denies that
consumption as such buys happiness. What really matters is where
people stand in their society, not their absolute level of
consumption. In particular, cross-cultural evidence shows that
reported life-satisfaction "is essentially independent of per-capita
income," above a certain threshold level of affluence. (Frank does
not specify that level, but he suggests that people do need to have
minimally adequate shelter and nutrition, and that happiness is
lower in extremely Door nations.) In various nations, including
America and Japan, people do not seem significantly happier when per
capita income rises sharply. Average satisfaction levels are stable
over time even in the presence of substantial economic growth. Above
a certain level of income, your relative position (how you compare
to others) seems to matter a great deal, and your absolute position
(how much you have) seems to matter not at all. Frank adds the
suggestion, based on social science, that people can adapt a great
deal to good and to badand thus new consumer goods, even luxury
goods, are likely to do little to increase people's subjective sense
of well-being.

Frank is not arguing that money cannot buy happiness. What matters
is not what people spend. What matters is what they buy. The best
kind of spending involves what Frank calls "inconspicuous
consumption," which he defines to include "freedom from traffic
congestion, time with family and friends, vacation time, and a
variety of favorable job characteristics," such as adequate physical
space. Frank contends that inconspicuous consumption will provide
"gains that endure." In America, however, people tend to neglect
this point and to devote most of their resources to a kind of
wasteful and mutually destructive competition for more and better

This, then, is the ultimate source of the problem: the consumption
treadmill, a kind of "positional arms race" that Frank finds at the
heart of luxury spending. In his view, we tend to spend money
because we care about what we have compared to others. Once people
get on the treadmill, there is no obvious stopping point, and soon
they find themselves spending large amounts for goods that give them
little, temporary, or no pleasure.

To break the process, we would have to act collectively; but we
can't, and so we don't. Nor are those who purchase luxuries entirely
oblivious to this. "Many people are fully aware, at some level, that
we might be better off if all saved more and all spent less on
houses and cars. But that choice is simply not an item on any
individual family's menu. A family can choose the amount it spends
on its own house, but it cannot choose the amounts that others
spend." Thus luxury spending is a classic case of the struggle for
better relative position-better position in the hierarchy-leading to
a form of waste that people are unable to prevent on their own.

The irony is that individual purchasing decisions inflict a harm on
other people (by forcing the rest of us to want to keep up), and
they do so without making the purchasers better off. The most
obvious analogy is literally an arms race, where nations acquire
more weapons in order to keep ahead of their rivals, even though
"when all spend more on weapons, no one is more secure than before."
In a general attack on his fellow economists, Frank claims that this
sort of problem is not isolated but pervasive.

With respect to competition, Frank stands against Adam Smith and
with Charles Darwin, who saw that with respect to evolution,
competition can be "smart for one, dumb for all." Athletes find
themselves using anabolic steroids in order to get an edge on the
competition; students use "cram courses" to prepare for the SAT,
producing no change in relative position and hence waste; people
talk louder and louder to be heard at a cocktail party; and people
who care about status may spend more and more to look good, or not
bad, in front of their peers. In a bit of speculative,
seat-of-thepants Darwinism, Frank even offers an evolutionary
explanation for why people care so much about relative position,
suggesting that evolution would favor people who slightly
outperformed others, and cared about doing so.

Sometimes this kind of problem is solved through law. An important
example is the environment. In an unregulated market, individual
polluters can impose environmental damage on others; they will not
reduce pollution voluntarily, because they would bear the full
costs, but not the full benefits, of the reduction. Regulation, or a
tax on polluting activity, is a necessary corrective. Consider also
zoning laws, which prevent merchants from having to compete with
"ever larger and more garish" signs; Sabbath norms and blue laws,
which "limit the extent to which people can trade leisure time for
additional income"; and overtime and safety regulation, which
prevents people from "competing for more money at the expense of
leisure time and health." Notwithstanding these success stories,
American government has failed to provide a general check against
competition for ever more expensive goods. Thus Frank urges that his
"claim, reduced to its essence, is that the conflict between
individual and group is the single most important explanation of the
imbalance in our current consumption patterns."

What is the solution? Voluntary self-help won't work. People's
judgments are inevitably biased in the direction of conspicuous
consumption, partly because they have much more information about it
than about inconspicuous consumption, partly because conspicuous
consumption is simply so tempting. Nor does Frank believe that
ordinary social norms are likely to solve the problem. To be sure,
many norms do appear directed against conspicuous consumption: if
you buy a Porsche or Mercedes, you may be violating the norms of
your community. But the reach of such norms is quite limited.

And so Frank turns to law. If he is right, the most obvious
solution is a luxury tax targeted to specific goods which people
purchase as part of the consumption treadmill. But Frank rejects
this solution on several grounds. Such a tax will have to specify
the goods that count as "luxuries"; and the predictable consequence
is an interest-group struggle over which goods to tax most heavily.
In any case, people are all too likely to attempt to evade any
particular tax.

The best solution, in Frank's view, is much more general: a
progressive consumption tax. As he presents it, the tax would be
calculated very simply, essentially by subtracting savings from
income. For example, a family that reported $100,000 in income and
$10,000 in saw ings would pay tax on $90,000-whereas the same
family, saving $50,000, would pay tax on a mere $50,000. Frank
believes that such a tax would sharply diminish competition over
goods that do not make people happier. At the same time, a tax of
this kind would shift the use of dollars in more productive and less
wasteful directions.

A consumption tax would help ordinary citizens, because it would
increase the incentive to devote their resources to things that
actually make them better off. It would help the economy, Frank
says, because money spent on consumption would now be spent on
investment, which is better for productivity, at least in the long
run. Indeed, higher taxes would stimulate the economy by diminishing
people's incentives to participate in winner-take-all markets. Frank
thinks that we could obtain all these goods while also decreasing
inequality, because the sharply progressive consumption tax could be
used to redistribute resources from rich to poor-without (and this
is his most ingenious claim) at the same time making rich people
worse off than they are now. Rich people would not be worse off
because the only consequence would be to prevent them from competing
for luxury goods, which do not, by hypothesis, improve their lives.
For Frank, the result of this analysis is to eliminate the alleged
conflict between "equity" and "efficiency." It is entirely possible
to have both.

Frank concludes with a vigorous attack on "trickle-down economics"
and the claim that "we can't afford" certain social programs. He
believes that trickle-down economics has things backward, because "a
shift to a progressive tax on consumption would be more likely to
increase economic growth rather than to inhibit it." And while Frank
acknowledges that some social programs have been wasteful, he thinks
that many have done a great deal of good, and that we could use a
good deal more.

Though he emphasizes environmental protection and health care,
Frank is especially concerned about the absence of decent education
and job prospects for poor people. He thinks that federal employment
programs, funded by a progressive consumption tax, could find many
useful tasks for people with little training or experience. In sum,
a progressive consumption tax would break competition over relative
position without hurting those engaged in that competition, and in
the process it would release trillions of dollars for both private
and public investment. "The only intelligible reason for having
stuck with our current spending patterns for so long is that we
haven't clearly understood their sources and how painless it would
be to change them."

At the level of theory, Frank's claim can be reduced to a single
sentence: since people make many luxury-type expenditures largely or
only to improve relative position, we should adopt policies that
produce more savings, and more equity, while breaking a form of
competition that does little good for anyone, including the rich.
Such an argument is both plausible and ingenious, and it is
refreshing to find illuminating new arguments offered in the service
of a fairer social order. But I am not sure that Frank is right. He
overstates the wastefulness of purchases of luxury goods, and he
understates the difficulties associated with a progressive
consumption tax.

Status and relative position do count. But we do not know when they
count or how much they count. Consider overtime and occupational
safety laws, which Frank enlists in support of his general argument.
Critics complain that such laws are counterproductive, because they
end up taking money out of the hands of workers. Overtime laws give
employers a disincentive to hire people to do extra work, and
occupational safety laws take money out of wages and devote it
instead to (often small) safety improvements in the workplace. Frank
thinks that even if workers end up with less money, this is not a
problem, because workers' relative position will stay constant, and
relative position is what workers care about. But the question
remains: do the workers whose wages go down because of these laws
really care only, or mostly, about relative position? This is far
from clear, and Frank offers no direct evidence to support his
affirmative answer.

To be sure, Frank's principal concern is with luxury spending,
which may be distinctly associated with the acquisition of status.
At least some of the time, however, people buy expensive goods
because they like them. I have a friend who recently bought an Audi,
which is a $40,000 car; and for him there is an especially
attractive feature in this car, which is the presence of heated
seats. These seats are not observable, and they do not confer
status. In fact, my friend appears not to care about status at all;
he just likes the car. Many people buy extraordinarily expensive
stereo equipment simply because the sound is so good, and they are
willing to pay for expensive vacation homes because the surroundings
are so beautiful. Frank has no evidence to the contrary.

Frank says, correctly, that the "frame of reference" is set by what
others have, and he responds that people are usually wrong to think
that such goods will make them happier. People's powers of
adaptation are such that a fancy car, stereo, mansion, or diamond
will not produce "enduring gains." This, too, is plausible, but it
is not clear that Frank has established its truth. Much of his
evidence consist of studies of self-reported happiness. Recall that
people in rich countries report themselves as no happier than people
in much less rich countries, holding relative position constant. But
how much can be learned from people's statements about how happy
they are? (And on six-point scales!) Relative position may account
for people's statements about their happiness on such
scalesstatements made perhaps after a quick survey of how other
people are doingwithout also accounting for their actual happiness
in life.

Holding everything constant, do we really know that people in the
top 20 percent, or middle 20 percent, in a very rich country are no
happier, or better off, than people in the top 20 percent, or middle
20 percent, of a significantly less rich country? In any case,
subjective happiness is hardly all that matters. Indeed, it is far
from clear that happiness should be taken as a subjective concept.
Whatever their "happiness," on questionnaires or otherwise, people
who have more resources are able (other things being equal) to live
better lives, by improving their health, providing better education
for their children, protecting themselves from extremes of heat and
cold, taking more and better vacations, and generally facing less
anxiety for the future. Economic growth may be good for these
reasons alone; its principal consequence is not to increase spending
on luxuries.

Frank thinks that someone who buys a luxury good inflicts a kind of
harm, akin to pollution, on other people. With this claim, too,
there are problems. Many people are not much influenced by luxury
purchases; perhaps most of us, most of the time, are not adversely
affected at all. Indeed, some people affirmatively enjoy seeing
luxury goods, even in the hands of others. In this way, people who
buy wonderful clothing or watches may actually make their friends
and neighbors better off. Frank's analogy to pollution is
illuminating, but far from precise. And it is not obvious that any
harm inflicted by luxury spending is the kind of harm for which
government ought to respond with taxation. Should government really
tax people for creating envy in others?

Turn now to Frank's policy recommendation. For many decades many
economists, on both the right and the left, have favored a
consumption tax over an income tax, partly as a means of encouraging
savings. Some people support a consumption tax as a way of
encouraging more savings but firmly reject higher rates on the
wealthy, usually on the theory that higher rates will decrease
economic growth (hence trickle-down economics). And many
commentators have urged, long before Frank, that a progressive
consumption tax would be a big improvement. (In recent years,
Senators Pete Domenici and Sam Nunn have proposed a tax of this
general kind which they call the "USA tax.") Frank's distinctive
contribution is to argue that an understanding of the consumption
treadmill shows that a progressive consumption tax would make many
people better off while making few, if any, worse off.

The idea of a consumption tax raises two obvious questions. Isn't
such a tax likely to be regressive, to hurt the poor and to help the
rich? And how do we know what counts as "savings"? Obviously,
wealthy people can save a lot and are easily able to divert income
to savings, whereas lower income people, who have to spend what they
earn, will pay tax on every dollar. To maintain the distributional
burdens of the current income tax, a progressive income tax might
have to be combined with a wealth tax (not discussed by Frank), or
it might have to include extremely (and politically speaking,
prohibitively) high marginal rates on expenditures by high earners.
There are hard technical issues here.

In any case, Frank is wrong to suggest that a progressive
consumption tax would be "simple." For one thinK a consumption tax
would have to replace, or to be squared with, the rest of the
existing tax system, including the corporate income tax, the
existing system of deductions, the treatment of gifts and bequests,
the mortgage deduction, insurance and annuities, borrowed money, and
so forth. The transition problems would be serious. For another
thing, the distinction between "consumption" and "savings" can be
thin. Does a house count as savings? Stocks? Books? Stamps?
Diamonds? Problems of this kind have made tax specialists worry
whether government can draw the necessary lines.

Frank's apparent remedy is to treat as "savings" only those
investments that are put into a specially designated account. But
what are the permissible uses of such an account? There are also
serious problems of equity. A family that spends $20,000 on cookies
and candy will be treated the same as a family that spends $20,000
on repairing their roof and educating their children; and both will
be treated the same as a family that spends $20,000 on a party
dress; and all of these will be treated the same as a family that
invests $20,000 in real estate. It is not easy to defend the
judgment that everything should count as "consumption" except
deposits into a specifically designated savings account. No wonder
that the Domenici-Nunn proposal, though very much in Frank's spirit,
is several hundred pages in length.

This raises a deeper problem. If Frank is really concerned about
excessive luxury spending in particular, a consumption tax seems to
be a pretty crude remedy. Most people spend only a little of their
money on luxuries; for most of us, most of the time, expenditures
are far more likely for housing, food, and clothing. For this and
other reasons, it is not clear that a progressive consumption tax
would solve the problems that Frank identifies.

Recall that one of Frank's motivating ideas is that those who
purchase certain goods impose a harm on others. Even if it should
try, how could government possibly monetize any such harm? In any
case, a general consumption tax is not a well-tailored solution to
the perceived problem. A consumption tax would cover any and all
consumption, much of which inflicts no harm on others. It would not
tax the particular injury to others produced by particular acts of
consumption. Indeed, a consumption tax could produce an overall
increase in savings without much decreasing the particular kinds of
expenditures that trouble Frank. At the very least, this cannot be
assessed a priori.

Perhaps we will have to say what Frank does not say (but, I
suspect, believes): that higher taxes on the wealthy are justified
even if such taxes hurt the wealthy. If Frank has not shown as much
as he claims, however, he has nonetheless shown a great deal. A
progressive consumption tax would have some real advantages over the
income tax. People often do struggle wastefully over status and
relative position, each competing to the detriment of all. And
people do buy things that fail to give much pleasure. Even grown-ups
often lack a good sense of what kinds of expenditures, and
investments, will actually promote their well-being.

A familiar problem with unrestricted free markets is that they can
produce pervasive injustice. A less familiar problem is that free
markets often trap people, including the well-off, into wasteful and
continuing struggles for better position. It is in the very nature
of the problem that even reasonable people may be unable to
extricate themselves from those struggles without collective help.
In the face of struggles of this kind, free markets should not be
identified with freedom, properly understood.

Return to Op-Ed page