It is a great pleasure for me to be here
today. I can think of no better place than the Paris School of Economics
for what I wish to discuss today — facts and fictions about
international trade economics.
Economists have long analysed and helped us
understand trade, why nations needed it to prosper, and what governments
had to do to reap the gains while managing the costs. The many theories
you and your predecessors have developed leave no doubt about the
importance of trade to growth and economic development.
But if the economics of trade policy are
clear, the politics of trade are highly complex. Trade policy, like so
many other areas of policy, has ramifications on how resources are
distributed, and this inevitably creates competing interest groups
within society. Pressures exerted by such groups mean governments must
balance these interests in ways that do not necessarily conform to what
economic analysis might prescribe.
The public debate that inevitably accompanies
contested policy formulation challenges the notion that open trade
brings overall societal benefits. At the same time, contested policy
provides a fertile field for the growth of urban legends and falsely
premised ideas with popular appeal.
In my comments today, I wish to identify and
address some of these fallacies. We need to bring sound economic
analysis to centre-stage in this debate. Secondly, I want to locate
trade policy in a wider policy setting because it is at our own peril
that we take trade policy and all its political complexities out of
their proper context.
Fallacy #1: Comparative advantage does not
At the outset, let me recognize Paul Krugman’s
intellectual contribution to international trade theory — the so-called
“new trade theory” — in which he shows that, even in the absence of
productivity differences between two countries, trade benefits them
both. He focuses on the presence of increasing returns to scale, in
which a firm’s average cost per unit declines as production increases
and underscores that consumers value variety in consumption. While the
new trade theory reduces the role played by comparative advantage, it
identifies new sources of benefits from trade that were not emphasized
or recognized by the classical economists. More trade benefits all
countries because specialization in production reduces average cost and
consumers gain access to a wider variety of products. In contrast,
traditional theories of trade assume the variety of goods remains
constant even after trade-opening.
We often hear the claim that the principle of
comparative advantage and mutually beneficial exchange may have been
true in the past, but it no longer applies in the 21st century — where,
among other changes we see the seemingly inexorable rise of countries
like China and India.
There are those who now call into question
Ricardo's theory that differences in relative productivity between
countries lead to their specialization in production and to trade. Doubt
has arisen that this specialization based on comparative advantage
results in higher total output, with all countries benefiting from the
There is a much-cited paper by Paul Samuelson
in the Summer 2004 issue of the Journal of Economic Perspectives which
showed theoretically how technical progress in a developing country like
China had the potential to reduce the gains from trade to a developed
country like the United States. This paper appeared to be a dramatic
about-face against the idea that open trade based on comparative
advantage is mutually beneficial.
I emphasize the word “appeared” because
subsequent analysis by Jagdish Bhagwati, Arvind Panagariya, and T. N.
Srinivasan contradicted this view. In that paper, starting from autarky,
China and the United States open up to trade and experience the usual
gains based on comparative advantage. In the following part of the
paper, Samuelson considers how technological improvements in China will
affect the United States. In the case where China experiences a
productivity gain in its export sector, both countries benefit. China
gains from the higher standard of living brought about by the increase
in productivity while the United States gains from an improvement in its
terms of trade. In the case where China experiences a productivity gain
in its import sector, there is a narrowing of the productivity
differences between the countries which reduces trade; and as trade
declines, so too do the gains from trade.
So what Samuelson has showed is not that trade
along lines of comparative advantage no longer produces gains for
countries. Instead, what he has shown is that sometimes, a productivity
gain abroad can benefit both trading countries; but at other times, a
productivity gain in one country only benefits that country, while
permanently reducing the gains from trade that are possible between the
two countries. The reduction in benefit does not come from too much
trade, but from diminishing trade. Furthermore, even in this case,
Samuelson himself does not prescribe protectionism as a policy response
since, as he put it ,“what a democracy tries to do in self defense may
often amount to gratuitously shooting itself in the foot”.
In my view, the analysis by Bhagwati,
Panagariya and Srinivasan should convince us that the principle of
comparative advantage, and more generally, the principle that trade is
mutually beneficial, remains valid in the 21st century.
Fallacy #2: It is unhealthy for trade to
grow faster and faster compared to output
After the Second World War we have seen a
marked expansion of international trade, with trade growing much faster
than world production. The ratio of international trade to the value of
world GDP has risen from 5.5 per cent in 1950 to over 20 per cent today.
There are those who argue that this trade expansion poses a danger to
the health of the global economy.
I would argue this is a reflection of the
international fragmentation of production and the rise of trade in
intermediate products. Reductions in transport costs, the information
technology revolution, and more open economic policies have made it
easier to “unbundle” production across a range of countries. The parts
and components that make up a final product are manufactured in
different countries around the globe, many of which are developing
countries. These intermediate goods may cross national borders several
times before they are assembled as a final product. Some of what passes
for international trade is in reality intra-firm trade, exchanges of
intermediate inputs and goods for processing between establishments
belonging to the same company. By allowing each country that is a member
of the supply chain to specialize in the part or component in which it
has a comparative advantage, the internationalization of supply chains
creates enormous economic benefits.
This growth in the trade of parts and
components means that import statistics will overstate the degree of
competition that comes from one’s trade partners. In international trade
theory, trade in goods is seen as a substitute for the movement of
factors of production. Thus, a country's imports of goods from its trade
partner are seen as additional supplies of the partner country's labour
and capital, which competes with the importing country's own workers and
entrepreneurs. But the share of value added by factors of production of
the origin country in traded products is considerably lower than in the
Take the example of an iPod assembled in China
by Apple. According to a recent study, it has an export value of $150
per unit in Chinese trade statistics but the value added attributable to
processing in China is only $4, with the remaining value added assembled
in China coming from the United States, Japan, and other Asian
countries. Now I know we tend to use statisticians like lamp posts —
both for support and for illumination — but I find these statistics very
enlightening. The degree to which a given volume of imports implies
competition between the factors of production in the country of origin
and the importing country's factors of production will be overstated.
Focusing on gross values of trade or imports coming from a particular
country also understates the degree to which the importing country’s own
firms benefit from trade because part of their output is incorporated in
the imported good.
So bilateral trade statistics may not
correctly reflect the origins of traded products. To go back to the
example of the $150 dollar iPod imported from China, it turns out that
less than 3% ($4 out of $150) of the value of the product reflects
China’s contribution, with the bulk of the product’s value being made by
workers and enterprises in the United States, Japan and other countries.
And yet current trade statistics would attribute $150 to Chinese
Relying on conventional trade statistics also
gives us a distorted picture of trade imbalances between countries. What
counts is not the imbalances as measured by gross values of exports and
imports, but how much value added is embedded in these flows. Let us
take the bilateral trade between China and the United States as an
example. Based on data from the Institute of Developing Economies (IDE-Jetro)
and WTO estimates, in 2008 the domestic content comprised 80% of the
value of the goods exported by the US. The comparable figure was 77% in
the case of Japan, 56% for Korea and 42% for Malaysia and Chinese
Taipei, meaning that about half the exported value originated from other
countries. Using conventional trade statistics would overestimate the US
bilateral deficit vis-à-vis China by around 30% as compared to measuring
in value added content. The figures would reach more than 50% when the
activity of export processing zones are fully taken into account. Does
it make sense for us to start measuring trade in value added rather than
on gross value as is the case today!
Fallacy #3: Current account imbalances are
a trade problem and ought to be addressed by trade policies.
Imbalances are a natural and widespread
economic phenomenon, if we only care to observe it. If we examine the
flows of spending and income between Aix en Provence and the city of
Paris, or between the 16ème and 5ème arrondissements of Paris for that
matter, the flows are not balanced. But we do not think this is a
problem and make no big deal of it. Yet we worry about imbalances
I think one legitimate reason why one may be
more concerned about imbalances between countries is that they may be
governed by a different set of economic institutions and regulations.
Fiscal and monetary policies are likely to be different. Regulations,
particularly in the financial sector, could also differ even among
neighbours. Exchange rate systems may vary significantly as well. These
differences may introduce “distortions” and thus provide a legitimate
reason to pay greater attention to imbalances among countries. Another
reason for paying attention to imbalances is when the amounts are large,
in absolute and relative terms, and there is fear that they could have
significant spillover effects on other countries.
Current account imbalances between countries
are primarily a macroeconomic phenomenon, a sign of international
differences in aggregate savings and investment behaviour and have
little to do with trade policy. A current account deficit of a country
reflects dissaving by domestic residents — an excess of total
expenditures, both private and public, over national income. A current
account surplus, on the other hand, represents savings by domestic
residents with national income exceeding total expenditures.
Put in another way, a current account surplus
is created if domestic savings exceeds domestic investment, which means
that the country is a creditor to the rest of the world. A current
account deficit ensues if domestic savings is less than domestic
investment, in which case, the country is drawing on savings provided by
The growth of global imbalances in the last
decade reflects in part the greater integration of financial and capital
markets. This has made it easier to accommodate big differences in
savings propensities and investment opportunities across countries.
Domestic residents can afford to save less because citizens of other
countries are saving more, and with few restrictions on capital flows,
those foreign savings can be made available at lower cost to domestic
In the absence of policy distortions,
imbalances provide a means to better allocate capital internationally.
Imbalances are a sign that savings in one country are being deployed or
used in another country. If investment prospects are plentiful in a
country, but its residents are unable to generate a sufficient amount of
saving to exploit them, foreign savings can fill the gap. The domestic
economy benefits from being able to undertake the profitable investment
project while foreign investors get a higher return than what they can
obtain in their own country.
Given that current account deficits and
surpluses originate in differences in savings propensities and
investment opportunities across countries, trade restrictions will not
permanently reduce deficits since they do not alter the underlying
conditions driving the imbalances. In fact restrictions could make
matters worse. First, they can trigger retaliatory action by those
affected by the import restrictions. Second, import restrictions create
economic inefficiency, not least in the country applying the
restrictions. As Abba Lerner taught us a long time ago, policymakers
often fail to recognize that one undesirable, but irrefutable
consequence of applying import restrictions is that such restrictions
also discourage exports.
Given that differences in savings behaviour
between countries lie at the root of current account surpluses or
deficits variations in exchange rates will not correct chronic
Exporters, moreover, often do not fully pass
through the effect of an exchange rate appreciation or depreciation to
the selling price in export markets. There is an extensive economic
literature that documents this “less than full pass through” to prices
of exchange rate changes in a wide variety of industries from
automobiles to photographic film to beer. This may explain why empirical
studies about the impact of exchange rate changes on imbalances tend to
show only limited or ambiguous effects and why adjustment in exchange
rates may have little effect on imbalances.
My arguments assume that exchange rates are
allowed to adjust freely to their equilibrium levels as determined, for
example, by long-term inflation differentials as in the so-called
purchasing power parity theory, or by interest rate differentials as in
the classic Mundell-Fleming explanation. I know that the lack of
exchange rate flexibility in some countries has been raised as a problem
in the context of global imbalances. The only thing I can say about this
issue is that it is difficult to talk about the optimal exchange rate
system for a country — whether completely flexible or fixed — without
addressing the much broader question of what the appropriate
international monetary system must be.
Fallacy #4: Trade destroys jobs
The problem with the argument that trade
destroys jobs is that it sees only the threat posed by imports to jobs
but does not take into consideration how jobs may be created in the
export sector as a consequence of trade opening. It also fails to take
into account that trade opening can increase the rate of economic
growth, and therefore improve the ability of the economy to create new
I recognize that traditional trade models
assume full employment so that more trade in those models does not
result in the creation of new employment. Jobs are merely reallocated
from shrinking to expanding sectors. Nevertheless, even if trade opening
does not necessarily create new jobs, the re-allocation of labour is
valuable. It means workers are moving from sectors where their marginal
product is lower to those where it is higher, resulting in productivity
gains to the economy and increased output.
More open trade can increase the rate of
economic growth by increasing the rate of capital accumulation, speeding
up technological progress through innovation or knowledge creation, and
improving the quality of institutions. A country that opens up trade may
become more attractive to foreign investors with more foreign capital
flowing to its shores. Since technology is often embodied in goods,
trade can be a very effective means of diffusing technological know-how.
Another way in which trade can increase productivity is
“learning-by-exporting”, whereby participation in world markets enables
producers to lower costs or move up the value-added chain over time.
This diffusion of technology through trade is important because of the
skewed distribution of research and development expenditures around the
world, which is even more skewed than the distribution of world income.
For example, the G-7 countries accounted for 84 per cent of global R & D
expenditures in 1995 but only 64 per cent of global GDP. Finally,
international trade can have a positive impact on the quality of a
country's institutions, such as the legal system, which in turn improves
economic performance. A faster growing economy will be able to absorb
more workers than a slow-growing or stagnant economy, a point that may
be particularly important for poor countries with large or rapidly
Trade opening can today contribute to economic growth and through that
to creating much needed jobs. The stimulus package of the Doha Round is
there to be taken.
Fallacy #5: Trade leads to a race to the
bottom in social standards.
Some have argued that more trade will drive
governments in rich countries to lower their social or labour standards.
More trade would hurt workers in rich countries. The problem with this
argument is that there is very little empirical basis for it. It is
difficult to find examples where countries have lowered social or labour
standards in response to trade competition.
One variant of this argument has been used by
Emmanuel Todd who claims that open trade between developing countries
such as China and industrial countries is the reason for the economic
crisis. In his view, the competition coming from low-wage countries has
put pressure on wages in industrial countries and caused a deficiency in
I would argue that differences in wages
largely reflect differences in labour productivity. There is a fairly
close correlation between wages and productivity across countries with
some estimates going so far as to suggest that 90 percent of wage
differences between countries can be explained by productivity
differences. While wages in many developing countries may be low, labour
productivity in these countries is also a fraction of western levels.
A more fundamental problem with Todd’s
analysis of the causes of the economic crisis is his presumption that
trade is a zero sum game, that what one nation gains, another country
loses. But we know from economic theory that trade is mutually
beneficial and that there are overall gains to countries who are
involved in trade. Hence, trade should increase incomes of trading
countries and, consequently, increase demand rather than lead to a
deficiency in demand.
The rise of China and India meant the
integration of tens, if not hundreds, of millions of Chinese and Indian
workers into the global economy. What is astonishing is that this
integration occurred at the same time unemployment rates in OECD and
Euro zone countries were declining. Between 1998 and 2008 for example,
the average unemployment rate in the Euro area fell from 9.9% to 7.4%
while the corresponding reduction in the OECD countries was from 6.6% to
5.9%. Surely, if the zero sum argument is correct, the rise of China and
India would have thrown millions of workers in developed countries out
Finally, some also argue that trade has been
responsible for the observed widening of the gap in the wages of skilled
and unskilled workers in industrial countries. However, much of this gap
can be explained by skill-biased technological change rather than by
trade. Skill-biased technological change refers to improvements in
technology, such as the information technology revolution, which
increased demand for skilled workers relative to unskilled workers. In
the United States, this technological change combined with the slowdown
in the relative supply of college-equivalent workers led to the
expansion of wage differentials between skilled and unskilled workers.
While industrial countries have experienced greater trade openness
during the same period, trade has not played a big role in rising wage
inequality. Empirical studies that have attempted to measure the
contribution that trade has made to inequality give estimates of only
between 4% and 11% to 15%.
Fallacy #6: Opening up trade equals
There is a tendency to confuse trade opening
with deregulation of the economy. One way to distinguish between the two
is that trade opening refers to a reduction in trade barriers or a
reduction of those measures that discriminate against foreign goods and
services. A country that opens up to trade does not compromise its
ability to regulate, provided of course those regulations do not
discriminate in an unjustifiable manner against foreign goods and
services, thus contradicting opening commitments.
There are many good reasons why governments
intervene in the economy and why such intervention needs to continue
even while trade is being opened. The production or consumption of a
good may generate pollution (an environmental externality) as a
by-product. The market for a good or service may be characterized by
information asymmetry, with sellers being more informed about the
quality or safety of their product than buyers. Firms may be restricting
competition by colluding or abusing their dominant position. There may
be moral hazard, i.e. excessive risk taking, in the banking system
because of the existence of deposit insurance. In all these cases,
regulations may form part of the public policy response to the
underlying market failures. Moreover, there are very legitimate
distributional reasons for governments to intervene in markets with a
view, for example, to improving poor people's access to public services
or to ensure equitable access to such services across all regions of a
country. WTO rules do not constrain governments' ability to pursue such
To see the difference between trade opening
and deregulation, I would like to take financial services in the North
American market as an example. The North American Free Trade Agreement
(NAFTA) opened up cross-border trade in financial services between
Canada and the United States, but otherwise allowed each country to
retain its system of financial governance. I believe this is one reason
why despite freer trade in financial services, the financial systems of
Canada and the United States had radically different experiences during
the financial crisis. Lax regulations in the U.S. mortgage market
contributed to the resulting sub-prime crisis. Meanwhile, on the other
side of the border, for Canadian banks, which were more tightly
regulated, sub-prime loans in 2006 accounted for less than 5% of new
mortgages in Canada, compared to 22% in the United States. They were
hardly touched by the crisis. In the midst of the global crisis in late
2008, they remained liquid and had an average tier 1 capital ratio that
was significantly above the regulatory minimum.
The issues I have touched upon today are
crucial to get the economics of trade right. But they can also
contribute to a more informed political debate on trade. We need debate,
and informed choices, because trade has an impact on our lives, whether
as workers or as consumers.
But I also wish to avoid getting stuck in the
old policy rut that simply promotes greater trade openness as a matter
of virtue or as the secret of a nation's economic success. Debunking
fallacies about trade is not enough. We should always remind ourselves
that trade is but one element of the story, an element from which it is
possible to expect too much as well as too little.
The benefits from open trade will only be
realized if the overall policy context is right. This is why I believe
that what I have referred to as the “Geneva Consensus” is so important.
The essence of this consensus is that trade openness can — and I stress
CAN — contribute to economic well-being and political harmony in
important ways, but only if other conditions are met. What are these
Firstly, we need sound macroeconomic policy,
and not a doctrine that sees trade policy as a quick fix for
over-arching economic fundamentals.
Second, trading opportunities created by
openness are worth little, and perhaps even unwelcome, if price signals
do not reach their destination because this is made impossible by a lack
of physical infrastructure and functioning markets. These elements are
part of a basic development agenda, one in which the international
community certainly has a role. This is why I have placed so much
emphasis on the Aid-for-Trade initiative. This programme is firmly
grounded in the belief that facilitating trade in the broadest sense is
part of a grand international bargain. I should emphasize that arguing
for pre-conditions to make trade openness work is not to make the case
for eschewing trade openness. On the contrary, it is the case for
creating conditions to embrace openness.
Finally, governments have a responsibility to address the distribution
of the benefits from trade. If trade opening is perceived as benefiting
a small privileged group, quite possibly at the expense of others in
society, then it loses its political legitimacy. Spreading the benefits
from trade is no easy matter, especially for poor countries. Challenges
arise both in terms of adjustment costs associated with changes in
relative prices, as well as in a more fundamental sense which has to do
with creating social infrastructure and personal opportunity.
These are just a few facts that I wanted to
share with you and I would like to thank you very much for your
Let the public debate commence now.