46 | REAL WORLD MACRO

Article 2.2

GDP AND ITS DISCONTENTS

BY ALEJANDRO REUSS
April 2013

E conomists have been thinking for a long time about what it means for a country
or its people to be rich or poor. That was one of the main questions Adam Smith,
the British philosopher often described as the “father of modern economics,” took on
in his most famous book The Wealth of Nations (1776). At the very outset, Smith made
a point of defining the “real wealth” of a country as consisting in the “annual produce
of the land and labour of the society.” (Note that Smith was using the word “wealth”
in a way that is closer to the colloquial meaning of the word than to its current tech-
nical meaning in economics. He was actually defining a country’s income rather than
its wealth.) That definition might seem uncontroversial now. Many economists would
certainly respond that of course it’s the production of goods and services that makes a
country wealthy. But Smith had an important axe to grind. He was arguing against the
view, widespread in his day, that a country’s wealth consisted in the accumulation of
gold and silver—an aim that led to a set of policies (especially promoting exports and
suppressing imports) known as “mercantilism.” In his own time, Smith was a maverick.
The kind of approach that Smith advocated, of counting up the total quantities
of goods and services produced in a country in a year, is now a central part of macro-
economic measurement. When economists tabulate a country’s gross domestic product
(GDP), they’re trying to measure the “annual produce … of the society” more or less
as Smith proposed. GDP attempts to add up the total value, counted in money units,
of the goods and services produced within a country in the course of a year. This ap-
proach, while a big advance over the view that a country’s wealth consisted primarily
of its hoards of precious metals, however, is more problematic and controversial than it
might appear at first glance. Economists and other social scientists have, in various ways,
criticized the ways that GDP is counted and used as a measure of a country’s “wealth”
or “development.” Here, we’ll focus on three key critiques: 1) the distributional critique,
2) the feminist critique, and 3) the environmental critique. The first is really a criticism
of the approach of looking at the total (or average) production of goods and services for a
society as a whole, and ignoring the distribution of access among its members. The other
two argue that GDP is misleading because it fails to count all goods and services (focus-
ing narrowly on those that are easiest to put prices on).

What is GDP Per Capita?

Gross domestic product (GDP) per capita is the standard measure of average in-
come used by mainstream economists, and it has become widely used as a measure
of economic well-being. Gross domestic product is a measure of the total value of
all the goods and services produced in a country in a year, which we can also think
of as the total incomes of all the people in that country. A country’s total GDP is a
very poor measure of how “rich” or “poor” its people are. A country can have a very
 CHAPTER 2: MACROECONOMIC MEASUREMENT | 47

high total income, even if the average income is low, just because it has a very high
population. China, for example, now has the highest total income of any country
in the world, even the United States. Its average income, however, is about one-
sixth of that of the United States, in terms of real purchasing power. China ranks so
high in total income because it is the largest country (by population) in the world.
By the same token, a country can have a very large average income, but have a low
total income, because it has small population. Developed countries have relatively
high levels of income per capita. The top 20 countries, by this measure, include 13
European countries, the United States and two other British offshoots (Australia
and Canada), and Japan. Two of the remaining three members of this exclusive list,
Qatar and United Arab Emirates, are small, oil-rich countries.
This problem, unlike those spotlighted in the three critiques we’ll discuss below, is
easy to solve. Instead of stopping at total GDP, we can calculate a country’s GDP per
capita. The phrase “per capita” simply means per person. (“Capita” comes from the Latin
word meaning “head,” so “per capita” means “per head.”) To get GDP per capita, we just
divide a country’s GDP by its population. This gives us the average GDP for that coun-
try, or a measure of the average income. (Other measures of a country’s total income,
such as Gross National Product or Gross National Income are similar to GDP, so GNP
per capita or GNI per capita are similar to GDP per capita.) Income per capita gives us a
better picture of the standards of living in a country than total income.

What’s Wrong with GDP Per Capita?

Mainstream economists and policymakers have treated increasing GDP per capita
as virtually synonymous with development, so it’s important to discuss GDP in
more detail. Here, we will focus on three major criticisms of GDP per capita as a
measure of well-being or “development”:

The Distributional Critique

Average income can be misleading. Average (mean) income is one estimate of the
“middle” of the distribution of income in a country. Most people, however, do not
get the average income. Most get less than the average, some get more (and a few get
much, much more). A relatively small number of people with very high incomes can
pull the average up by a great deal, making the average less representative of most
people’s standard of living.
Figure 1, for example, shows the income distribution for Brazil in 2007. The popu-
lation has been ranked by income, and then divided into five equal parts (or quintiles).
Each bar represents the difference between the average income for one of these quintiles
and the average income for the country as a whole. The bar furthest to the left represents
the difference between the average income of the lowest-income quintile and the overall
average. The next bar represents this difference for the next-lowest-income quintile, and
so on, all the way up to the bar at the far right, which represents this difference for the
highest-income quintile. (The lowest-income quintile is called the “first” quintile, the
next-to-lowest is called the “second” quintile, and so on, up to the highest-income, or
“fifth,” quintile.) The GDP per capita for Brazil in 2007 was about $9,800. Notice that
the average income for each of the bottom four quintiles is less than the GDP per capita
48 | REAL WORLD MACRO
Brazil: Income Distribution, 2007
(Difference between each quintile's
INCOME DISTRIBUTION, BRAZIL, average income
2007 (DIFFERENCE and
BETWEEN EACH
QUINTILE’S AVERAGE overall average
INCOME ANDincome)
OVERALL AVERAGE INCOME)

20000 Highest
Current international dollars

15000

10000

5000

0
Next-to-highest

-5000 Middle
Next-to-lowest
-10000 Lowest

Quintile

Source: World Bank, World Development Indicators: Income share held by lowest 20%, second 20%,
third 20%, fourth 20%, highest 20%; GDP per capita, PPP (constant 2005 international $); GDP, PPP
(constant 2005 international $) (data.worldbank.org/indicator).

(or average income) for the society as a whole, as indicated by the bars extending down.
The average income for Brazil as a whole is more than six times as much as the average
income for the first (lowest-income) quintile, almost three times as much as the average
income for the second quintile, and more than one-and-a-half times as much as the av-
erage income for the third quintile. Even the average income for the fourth quintile is a
little less than the average income for the whole country (so many people in the fourth
quintile have incomes below the national average, though some have incomes above it.)
More than two-thirds of Brazil’s population, then, have incomes below the
country’s per capita income—many of them, far below it. The reason GDP per
capita for Brazil is so much higher than the incomes of most Brazilians is that the
income distribution is so unequal. The average income for the fifth (highest-income)
quintile is almost three times the average income for Brazil as a whole.

The Feminist Critique

GDP only counts part of the goods and services produced in a country. Earlier, we said
that GDP was “a measure of the total value of goods and services” produced in a coun-
try. This is true, but it is a very flawed measure. GDP only includes the value of goods
that are produced for sale in markets, ignoring goods and services that people produce
for their own consumption, for the consumption of family members, and so on. In
developed economies, most individuals or households have money incomes that allow
them to buy most of the things they need to live. They also, however, produce goods
and services for themselves, family members, and others. For example, people care for
and educate their children, cook meals for themselves and other members of their fam-
ily, clean their own homes, drive themselves and family members to work, school, and
errands, and so on. These kinds of goods and services count as part of GDP when
someone is paid to do them (for example, when we pay tuition to a school, the bill at a
 CHAPTER 2: MACROECONOMIC MEASUREMENT | 49

restaurant, the fee to a professional cleaning crew, or the fare to a taxi driver), but not
when people do it for themselves, family members, or others free of charge. One could
add many other examples, but the first lesson here is that GDP undercounts the total
output of goods and services. Since so much of the labor that produces these uncounted
goods and services is done by women, feminist economists, such as Marilyn Waring,
the author of If Women Counted: A New Feminist Economics, have been in the forefront
of this critique of GDP as a measure of economic development or well-being.
In some developing economies, the uncounted goods and services may form a larg-
er part of the overall economy than in developed countries. Many people may have small
farms and grow their own food. Some people weave their own cloth and make their own
clothes. Some people build their own shelters. As economies “develop” economically,
they may become more “monetized.” This means that people produce fewer goods for
their own consumption, for their families, or to trade for other goods (barter), relative to
the total amount of goods and services. Instead, they start selling either goods they pro-
duce or selling their own labor for money, and buying the things they need. An increase
in GDP over time may, in part, reflect an increasing output of goods and services. But
it may also reflect, in part, that some goods went uncounted before (because they were
not produced for sale in markets) and are now being counted. This means that GDP (or
GDP per capita) may exaggerate the growth of economies over time.

The Environmental Critique

GDP does not account for changes in the natural environment. We can think of parts of
the natural environment as providing people with valuable “natural services.” Until re-
cently, economic measurement has almost completely ignored natural services. Once we
start thinking about the environment's services, it becomes obvious how critical they are
for our well-being. A forest, for example, absorbs carbon dioxide from and provides oxy-
gen to the atmosphere, provides flood control, reduces soil erosion, provides habitat for
wildlife, offers natural beauty and outdoor recreation, provides some people with sources
of food and fuel (especially in lower-income countries), and so on.
If GDP only counts human-produced goods and services, then, it is undercounting
the total goods and services. If a forest is cut down for timber, and the wood is sold in a
market, this adds to GDP. However, the value of the services that the forest provided are
not deducted from GDP as conventionally measured, since these are not sold in mar-
kets and do not have prices. Cutting down a forest may both add something (harvested
wood, which can be used, for example, to build houses or make furniture) and subtract
something (natural services) from the well-being of society. There is no way to say, in
general, whether what it gained is greater or less than what is lost. However, as long as we
think that the services the forest provided were worth something, we can say for certain
that what GDP measures as being gained is greater than what it is really gained—since
GDP only counts what is gained and ignores what is lost.

If Not GDP, then What?

Part of the power of GDP per capita is that it boils everything down to one easy-to-
digest number. It is easy to create a table comparing the GDPs of many countries.
(Obviously, it would be harder to compare many countries in more complex ways,
50 | REAL WORLD MACRO

including a bunch of descriptive numbers for each.) This is also at the core of the
weaknesses of GDP per capita. When we calculate a total or average of anything,
we are, in effect, throwing out the information we have about variation between
different individuals. This problem is at the heart of the first critique: Calculating
total GDP or GDP per capita means excluding information about income distribu-
tion. In addition, calculating the total output of goods and services, when a modern
economy includes thousands and thousands of different kinds of goods, requires
some unit in which we can measure the output of each one. (We can’t add together
pounds of potatoes and pounds of steel, much less goods and services that can’t be
measured in pounds at all, like electricity or haircuts.) GDP has accomplished this
by measuring everything in terms of monetary units. This leads to the second and
third critiques. Monetary measurement has led to a blind spot for goods and services
that do not have market prices (household production, environmental services) and
are not easy to measure in money terms.
There are three major possibilities. One is to go on calculating GDP per capita,
but to do a better job at capturing what GDP misses. For example, some scholars
have tried to put a dollar values on nonmarket production (like subsistence farming
or household production) and add these to GDP to get a more accurate estimate.
Another is to come up with an alternative one-number measure to compete with
GDP. Two important ones are the genuine progress indicator (GPI) and the human
development index (HDI). The GPI incorporates, in addition to market production,
measures of both nonmarket production and environmental destruction into a sin-
gle summary figure (in money terms). It does not address the distributional critique.
Calculated by the United Nations Development Programme (UNDP), the HDI com-
bines GDP per capita, average educational attainment, and average life expectancy
into a single numerical index. It addresses neither the feminist nor the environmental
critique, and it does not explicitly address the distributional critique. However, more
equal societies tend to rank better on HDI than on GDP per capita, because they tend
to achieve higher average education and life expectancy. (The UNDP also calculates
an inequality-adjusted HDI, which explicitly penalizes inequality.)
Finally, a third approach is to abandon the quest for a single summary mea-
surement. Some environmental economists oppose attempts to incorporate envi-
ronmental changes into GDP or other monetary measures, which requires reducing
environmental services to money values. This implies, they argue, that some quan-
tity of produced goods can substitute for any environmental good, which is not
true. They propose instead “satellite accounts” that measure environmental changes
alongside GDP. Widely used measures of income inequality also exist, and can en-
hance our picture of an economy. Measurements of median income, access to basic
goods (like health and education), economic inequality, nonmarket production, en-
vironmental quality, and other factors all should figure, in some way, into our un-
derstanding of economic life. We may just have to accept that we need to take into
account multiple measures, and that no single-number “bottom line” will do. q