Inequality: Bad for Your Health
An interview with Ichiro Kawachi
This article is from the January/February 2008 issue of Dollars
& Sense: The Magazine of Economic Justice, formerly
available at http://www.dollarsandsense.org/archives/2008/0108kawachi.html
How do you stay healthy? That’s a no brainer, right?
Eat the right foods, exercise, quit smoking, get regular medical
checkups. Epidemiologist Ichiro Kawachi wants
to add a new item to the list: live in a relatively egalitarian
society. Kawachi, a professor of social epidemiology
at the Harvard School of Public Health, has carried out a wide
range of research studies on the social and economic factors that
account for average health outcomes in different societies. Among
the most novel conclusions of this body of research is that people
in societies with high levels of economic inequality are less
healthy than those living in more equal societies, regardless
of their absolute levels of income.
Health policy is at least on the table in this election year.
The conversation, however, is almost entirely limited to whether
and how to ensure universal health insurance coverage. The work
of Kawachi and his colleagues suggests that the public debate
about health really needs to be much broader, encompassing a wide
range of public policies—in many cases economic policies—that
do not explicitly address health but that nonetheless condition
how long and how robust our lives will be. Their work traces the
multidimensional connection between an individual’s health
and the qualities of her social world, many of which can shift
dramatically when the gap between rich and poor widens.
Kawachi spoke with Dollars & Sense in November.
Dollars & Sense: Your research looks at
the relationship between economic factors and health, especially
whether living in a more unequal society, in itself, has a negative
effect on health outcomes—and you have found evidence that
it does. But I want to start by being really clear about what
this hypothesis means. There seems to be such a complicated web
of possible relationships between income and health.
Ichiro Kawachi: Let’s start with how your
own income affects your health. Most obviously, income enables
people to purchase the goods and services that promote health:
purchasing good, healthy food, being able to use the income to
live in a safe and healthy neighborhood, being able to purchase
sports equipment. Income enables people to carry out the advice
of public health experts about how to behave in ways that promote
But in addition to that, having a secure income has an important
psychosocial effect. It provides people with a sense of control
and mastery over their lives. And lots of psychologists now say
that sense of control, along with the ability to plan for the
future, is in itself a very important source of psychological
health. Knowing that your future is secure, that you’re
not going to be too financially stressed, also provides incentives
for people to invest in their health Put another way, if my mind
is taken up with having to try to make ends meet, I don’t
have sufficient time to listen to my doctor’s advice and
invest in my health in various ways.
So there are some obvious ways in which having adequate income
is important for health. This is what we call the absolute income
effect—that is, the effect of your own income on your own
health. If only absolute income matters, then your health is determined
by your income alone, and it doesn’t matter what anybody
else makes. But our hypothesis has been that relative income might
also matter: namely, where your income stands in relation to others’.
That’s where the distribution of income comes in. We have
looked at the idea that when the distance between your income
and the incomes of the rest of society grows very large, this
may pose an additional health hazard.
D&S: How could people’s relative income
have an impact on health, even if their incomes are adequate in
IK: There are a couple of possible pathways.
One is this ancient theory of relative deprivation: the idea that
given a particular level of income, the greater the distance between
your income and the incomes of the rest of society, the more miserable
you feel. People are sensitive to their relative position in society
vis-à-vis income. You may have a standard of living above
the poverty level; nonetheless, if you live in a community or
a society in which everyone else is making so much more, you might
feel frustrated or miserable as a result, and this might have
deleterious psychological and perhaps behavioral consequences.
So that’s one idea.
Another hypothesis about why income distribution matters is that
when the income or wealth gap between the top and bottom grows,
certain things begin to happen within the realm of politics. For
example, when the wealthiest segment of society pulls away from
the rest of us, they literally begin to segregate themselves in
terms of where they live, and they begin to purchase services
like health care and education through private means. This translates
into a dynamic where wealthy people see that their tax dollars
are not being spent for their own benefit, which in turn leads
to a reduced basis for cooperation and spending on public goods.
So I think there is an entirely separate political mechanism that’s
distinct from the psychological mechanism involved in notions
of relative deprivation.
These are some of the key ways in which income inequality is
corrosive for the public’s health.
D&S: When you talk about relative deprivation,
are you talking primarily about poor people, or does the evidence
suggest that inequality affects health outcomes up and down the
income ladder? For instance, what about the middle class? I think
for the public to understand the inequality effect as something
different from just the absolute-income effect, they would have
to see that it isn’t only poor people who can be hurt by
IK: Exactly, that’s my argument. If you
subscribe to the theory that it’s only your own income that
matters for health, then obviously middle-class people would not
have much to worry about—they’re able to put food
on the table, they have adequate clothing and shelter, they’re
beyond poverty. What the relative-income theory suggests is that
even middle-class people might be less healthy than they would
be if they lived in a more egalitarian society.
D&S: That’s what I was wondering about.
Say we compared a person at the median income level in the United
States versus Germany, both of whom certainly have enough income
to cover all of the basic building blocks of good health. Would
this hypothesis lead you to expect that, other things being equal,
the middle-income person in the United States will likely have
worse health because economic inequality is greater here?
IK: Yes, that’s exactly right. And that’s
borne out. Americans are much less healthy than Europeans, for
example, in spite of having higher average wealth.
D&S: But, unlike most other rich countries,
the United States does not have universal health care. Couldn’t
that explain the poorer health outcomes here?
IK: Not entirely. There was a very interesting
paper that came out last year comparing the health of Americans
to the health of people in England, using very comparable, nationally
representative surveys. They looked at the prevalence of major
conditions such as heart attack, obesity, diabetes, hypertension.
On virtually every indicator, the top third of Americans by income—virtually
all of whom had health insurance—were still sicker than
the bottom third of people in England. The comparison was confined
to white Americans and white Britons, so they even abstracted
out the contribution of racial disparities.
Health insurance certainly matters—I’m not downgrading
its importance—but part of the reason Americans are so sick
is because we live in a really unequal society, and it begins
to tell on the physiology.
D&S: Has anyone tried to compare countries
that have universal health care but have differing levels of inequality?
IK: There have been comparisons across Western
European countries, all of which pretty much have universal coverage.
If you compare the Scandinavian countries against the U.K. and
other European countries, you generally see that the Scandinavians
do have a better level of health. The more egalitarian the country,
the healthier its citizens tend to be. But that’s about
as much as we can say. I’m not aware of really careful comparative
studies; I’m making a generalization based on broad patterns.
D&S: It sounds like there is still plenty
of research to do.
D&S: You have already mentioned a couple
of possible mechanisms by which an unequal distribution of income
could affect health. Are there any other mechanisms that you would
IK: I think those are the two big ones: the
political mechanism, which happens at the level of society when
the income distribution widens, and then the individual mechanism,
which is the relative deprivation that people feel. But I should
add that relative deprivation itself can affect health through
a variety of mechanisms. For instance, there is evidence that
a sense of relative deprivation leads people into a spending race
to try to keep up with the Joneses—a pattern of conspicuous,
wasteful consumption, working in order to spend, to try to keep
up with the lifestyle of the people at the top. This leads to
many behaviors with deleterious health consequences, among them
overwork, stress, not spending enough time with loved ones, and
Very interestingly, a couple of economists recently analyzed
a study of relative deprivation, which used an index based upon
the gap between your income and the incomes of everybody above
you within your social comparison group, namely, people with the
same occupation, or people in the same age group or living in
the same state. What they found was that the greater the gap between
a person’s own income and the average income of their comparison
group, the shorter their lives, the lower their life expectancy,
as well as the higher their smoking rates, the higher their utilization
of mental health services, and so on. This is suggestive evidence
that deprivation relative to average income may actually matter
for people’s health.
D&S: It’s interesting—this part
of your analysis almost starts to dovetail with Juliet Schor’s
IK: Absolutely, that’s right. What Juliet
Schor writes about in The Overspent American is consumerism. It
seems to me that in a society with greater income inequality,
there’s so much more consumerism, that the kind of pathological
behavior she describes is so much more acute in unequal societies,
driven by people trying to emulate the behavior of those who are
pulling away from them.
D&S: Your research no doubt reflects your
background as a social epidemiologist. However, it seems as though
many mainstream economists view these issues completely differently:
many do not accept the existence of any causal effect running
from income to health, except possibly to the degree that your
income affects how much health care you can purchase.
IK: Yes, there is a lot of pushback from economists
who, as you say, are even skeptical that absolute income matters
for health. What I would say to them is, try to be a little bit
open-minded about the empirical evidence. It seems to me that
much of the dismissal from economists is not based upon looking
at the empirical data. When they do, there is a shift: some economists
are now beginning to publish studies that actually agree with
what we are saying. For example, the study on relative deprivation
and health I mentioned was done by a couple of economists.
Another example: some studies by an erstwhile critic of mine,
Jeffrey Milyo, and Jennifer Mellor, who in the past have criticized
our studies on income distribution and health in the United States
as not being robust to different kinds of model specifications—a
very technical debate. Anyway, most recently they published an
interesting study based on an experiment in which they had participants
play a prisoners’ dilemma kind of game to see how much they
cooperate as opposed to act selfishly. One of the things Mellor
and Milyo found was that as they varied the distribution of the
honoraria they paid to the participants, the more unequal the
distribution of this “income,” the more selfishly
the players acted. They concluded that their results support what
we have been contending, which is that income inequality leads
to psychosocial effects where people become less trusting, less
cohesive, and less likely to contribute to public spending.
D&S: That’s fascinating.
IK: Yes, it’s very interesting. So watch
this space, because some of the recent evidence from economists
themselves has begun to support what we’re saying.
D&S: In other parts of the world, and especially
in Africa, there are examples of societies whose economies are
failing or stagnating because of widespread public health issues,
for example HIV/AIDS. So it seems as if not only can low income
cause poor health, but also that poor health can cause low income.
I wonder if your research has anything to say about the complicated
web between income and health that those countries are dealing
IK: There’s no doubt that in sub-Saharan
Africa, poor health is the major impediment to economic growth.
You have good econometric studies suggesting that the toll of
HIV, TB, and malaria alone probably slows economic growth by a
measurable amount, maybe 1½ percentage points per year.
So there’s no question that what those countries need is
investment to improve people’s health, in order for them
to even begin thinking about escaping the poverty trap. The same
is true in the United States, by the way. Although I’ve
told the story in which the direction of causation runs from income
to health, of course poor health is also a major cause of loss
of income. When people become ill, for example, they can lose
their jobs and hence their income.
What I’ll say about the developing world is that in many
ways, the continuing lack of improvement in health in, for example,
the African subcontinent is itself an expression of the maldistribution
of income in the world. As you know, the rich countries are persistently
failing to meet the modest amount of funding that’s being
asked by the World Health Organization to solve many of these
problems, like providing malaria tablets and bed nets and HIV
pills for everyone in sub-Saharan Africa. If you look at inequality
on a global scale, the world itself could benefit from some more
redistribution. Today the top 1% of the world’s population
owns about a third of the world’s wealth. So, although certainly
the origins of the HIV epidemic are not directly related to income
inequality, I think the solution lies in redistributing wealth
and income through overseas development aid, from the 5% of the
world who live in the rich countries to everyone else.
D&S: Leaving aside some of the countries
with the most devastating public health problems, poor countries
in general are often focused just on economic growth, on getting
their per capita GDP up, but this often means that inequality
increases as well—like in China. Do you view the inequality
effect as significant enough that a developing country concerned
about its health outcomes should aim to limit the growth of inequality
even if that meant sacrificing some economic growth?
IK: It depends on the country’s objectives.
But I’d ask the question: what is the purpose of economic
growth if not to assure people’s level of well-being, which
includes their health? Why do people care about economic growth?
In order to lead a satisfying and long life, many people would
say. If that’s the case, then many people living in developing
countries may feel exactly as you suggest: they would prefer policies
that attend to egalitarian distribution over policies that are
aimed purely at growth.
Amartya Sen has written about this; he has pointed to many countries
that are poor but nonetheless enjoy a very good level of health.
He cites examples like Costa Rica and the Kerala region in India,
which are much, much poorer than the United States but enjoy a
high level of health. It really depends on the objectives of the
country’s politicians. In Kerala and Costa Rica, their health
record is very much a reflection of how their governments have
invested their income in areas that promote health, like education
and basic health services—even if doing so means causing
a bit of a drag on economic growth.
China also had this record until perhaps ten years ago. Now they’re
in this era of maximizing growth, and we’re seeing a very
steep rise in inequality. Although we don’t have good health
statistics from China, my guess is that this is probably going
to tell on its national health status. Actually, we already know
that improvement in their child mortality rates for children under
five has begun to slow down in the last 20 years, since the introduction
of their economic reforms. In the 1950s and 1960s, the records
seemed to suggest quite rapid improvements in health in China.
But that’s begun to slow down.
D&S: Certainly your research on the health
effects of inequality could represent a real challenge in the
United States in terms of health care policy. In many ways we
have a very advanced health care system, but many people are not
well served by it. What effect do you think your work could or
should have on U.S. health policy?
IK: Regardless of whether you believe what I’m
saying about income inequality, the most basic interpretation
of this research is that there are many things that determine
people’s health besides simply access to good health services.
We spend a lot of time discussing how to improve health insurance
coverage in this country. In the current presidential debates,
when they talk about health policy, they’re mostly talking
about health insurance. But it’s myopic to confine discussions
of health policy to what’s going to be done about health
insurance. There are many social determinants of health and thus
many other policy options for improving Americans’ health.
Investing in education, reducing the disparities in income, attacking
problems of poverty, improving housing for poor people, investing
in neighborhood services and amenities—these are all actually
health policies. The most fundamental point about this whole area
of research is that there are many determinants of health besides
what the politicians call health policy.
D&S: Besides doctors and medical care.
IK: Yes, that’s right. I used to be a
physician, and physicians do a lot of good, but much of health
is also shaped by what goes on outside the health care system.
That’s probably the most important thing.
The second thing is the implication that income certainly matters
for health. So policies that affect peoples’ incomes, both
absolute and relative income, may have health consequences. For
instance, I think the kinds of tax policies we have had in recent
years—where most of the benefits have accrued to the top
1% and the resulting budget deficits have led to cutbacks of services
to the rest of us, especially those in the bottom half of the
income distribution—have been a net negative for public
health, through the kind of political mechanism I have described.
D&S: It’s almost as though there should
be a line for health care in the cost-benefit analysis of any
change in tax policies or other economic policies.
IK: Absolutely. There’s an idea in public
health called the health impact assessment. It’s a technique
modeled after environmental impact assessments, a set of tools
that people are advocating should be used at the Cabinet level.
The idea is that when, say, the treasury secretary suggests some
new economic measure, then we can formally put the proposal through
a modeling exercise to forecast its likely effects on health.
Health certainly is very sensitive to decisions that are made
elsewhere in the Cabinet besides what goes on in Health and Human
D&S: What about global health policy? Are
groups like the World Health Organization paying attention to
the kind of research that you do?
IK: Yes, they are. Maybe seven or eight years
ago, the WHO had a commission on macroeconomics and health, headed
by Jeffrey Sachs. The idea was, by increasing funding to tackle
big health problems in the developing world, we can also improve
their economic performance and end poverty. That commission posed
the direction of causality from health to income. In the last
three years, the WHO has had a new commission on the social determinants
of health, headed by a social epidemiologist from England, Michael
Marmot. That group is looking at the other direction of causality—namely,
from poverty to ill health—and considering the ways in which
government policies in different areas can improve people’s
social environment in order to improve their health. I think they
are due to report next year with some recommendations as well
as case examples from different countries, mostly developing countries
whose governments have tried to tackle the economic side of things
in order to improve health outcomes.
D&S: Right now the United States is continuing
on this path of becoming more and more economically stratified.
Your work suggests that that doesn’t bode well for us in
terms of health. I wonder—this is very speculative—but
if we stay on this path of worsening inequality, what do you predict
our health as a country is likely to look like in 20 or 30 years?
IK: We’re already in the bottom third
of the 23 OECD countries, the rich countries, in terms of our
average health status. Most people are dimly aware that we spend
over half of the medical dollars expended on this planet, so they
assume that we should therefore be able to purchase the highest
level of health. I teach a course on social determinants of health
at Harvard, and many of my students are astonished to discover
that America is not number one in life expectancy.
I predict that if we continue on this course of growing income
inequality, we will continue to slip further. That gains in life
expectancy will continue to slow down. Life expectancy is increasing
every year, probably because of medical advances, but I suspect
that eventually there will be a limit to how much can be delivered
through high-tech care and that our health will slip both in relative
terms, compared to the rest of the OECD countries, and maybe even
in absolute terms, losing some of the gains we have had over the
last half century. For example, some demographers are already
forecasting that life expectancy will drop in the coming century
because of the obesity academic. Add that to the possible effects
of income inequality, and I could easily imagine a scenario in
which life expectancy might decline in absolute terms as well
as in relative terms. It’s likely that we have not yet seen
the full impact of the recent rise in inequality on health status,
because it takes a while for the full health effects to become
apparent in the population.
The interview was conducted by D&S co-editor
Amy Gluckman and intern Alissa Thuotte.