Dani Rodrik
In
the pantheon of economic theories, the tradeoff between equality and
efficiency used to occupy an exalted position. The American economist
Arthur Okun, whose classic work on the topic is called Equality and Efficiency: The Big Tradeoff,
believed that public policies revolved around managing the tension
between those two values. As recently as 2007, when New York University
economist Thomas Sargent, addressing the graduating class at
the University of California, Berkeley, summarized the wisdom of
economics in 12 short principles, the tradeoff was among them.
The
belief that boosting equality requires sacrificing economic efficiency
is grounded in one of the most cherished ideas in economics: incentives.
Firms and individuals need the prospect of higher incomes to save,
invest, work hard, and innovate. If taxation of profitable firms and
rich households blunts those prospects, the result is reduced effort and
lower economic growth. Communist countries, where egalitarian
experiments led to economic disaster, long served as “Exhibit A” in the
case against redistributive policies.
In
recent years, however, neither economic theory nor empirical evidence
has been kind to the presumed tradeoff. Economists have produced new
arguments showing why good economic performance is not only compatible
with distributive fairness, but may even demand it.
For
example, in high-inequality societies, where poor households are
deprived of economic and educational opportunities, economic growth is
depressed. Then there are the Scandinavian countries, where egalitarian
policies evidently have not stood in the way of economic prosperity.
Early this year, economists at the International Monetary Fund produced empirical results that
seemed to upend the old consensus. They found that greater equality is
associated with faster subsequent medium-term growth, both across and
within countries.
In high-inequality societies,
where poor households are deprived of economic and educational
opportunities, economic growth is depressed.
Moreover,
redistributive policies did not appear to have any detrimental effects
on economic performance. We can have our cake, it seems, and eat it,
too. That is a striking result – all the more so because it comes from
the IMF, an institution hardly known for heterodox or radical ideas.
Economics
is a science that can claim to have uncovered few, if any, universal
truths. Like almost everything else in social life, the relationship
between equality and economic performance is likely to be contingent
rather than fixed, depending on the deeper causes of inequality and many
mediating factors. So the emerging new consensus on the harmful effects
of inequality is as likely to mislead as the old one was.
Consider,
for example, the relationship between industrialization and inequality.
In a poor country where the bulk of the workforce is employed in
traditional agriculture, the rise of urban industrial opportunities is
likely to produce inequality, at least during the early stages of
industrialization. As farmers move to cities and earn higher pay, income
gaps open up. And yet this is the same process that produces economic
growth; all successful developing countries have gone through it. In
China, for example, rapid economic growth after the late 1970s was
associated with a significant rise in inequality. Roughly half of the
increase was the result of urban-rural earnings gaps, which also acted
as the engine of growth.
According to Dani Rodrik, we have to distinguish between good and bad inequality.
Or
consider transfer policies that tax the rich and the middle classes in
order to increase the income of poor households. Many countries in Latin
America, such as Mexico and Bolivia, undertook such policies in a
fiscally prudent manner, ensuring that government deficits would not
lead to high debt and macroeconomic instability.
On
the other hand, Venezuela’s aggressive redistributive transfers under
Hugo Chávez and his successor, Nicolás Maduro, were financed by
temporary oil revenues, placing both the transfers and macroeconomic
stability at risk. Even though inequality has been reduced in Venezuela
(for the time being), the economy’s growth prospects have been severely
weakened.
Latin
America is the only world region where inequality has declined since
the early 1990s. Improved social policies and increased investment in
education have been substantial factors. But the decline in the pay
differential between skilled and unskilled workers – what economists
call the “skill premium” – has also played an important role. Whether this is good news or bad for economic growth depends on why the skill premium has fallen.
If
pay differentials have narrowed because of an increase in the relative
supply of skilled workers, we can be hopeful that declining inequality
in Latin America will not stand in the way of faster growth (and may
even be an early indicator of it). But if the underlying cause is the
decline in demand for skilled workers, smaller differentials would
suggest that the modern, skill-intensive industries on which future
growth depends are not expanding sufficiently.
It is good that economists no longer regard the equality-efficiency tradeoff as an iron law.
In the advanced countries, the causes of rising inequality are still being debated.
Automation and other technological changes, globalization, weaker trade
unions, erosion of minimum wages, financialization, and changing norms
about acceptable pay gaps within enterprises have all played a role,
with different weights in the United States relative to Europe. Each one
of these drivers has a different effect on growth. While technological
progress clearly fosters growth, the rise of finance since the 1990s has
probably had an adverse effect, via financial crises and the
accumulation of debt.
It
is good that economists no longer regard the equality-efficiency
tradeoff as an iron law. We should not invert the error and conclude
that greater equality and better economic performance always go
together. After all, there really is only one universal truth in
economics: It depends.
