The Political Economy of Trade Policy
Robert E. Baldwin
The Journal of Economic Perspectives, Vol. 3, No. 4. (Autumn, 1989), pp. 119-135.
Stable URL:
http://links.jstor.org/sici?sici=0895-3309%28198923%293%3A4%3C119%3ATPEOTP%3E2.0.CO%3B2-S
The Journal of Economic Perspectives is currently published by American Economic Association.
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in
the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/journals/aea.html.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic
journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,
and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take
advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org.
http://www.jstor.org
Wed Oct 31 10:40:49 2007
Journal of Economic Perspectives- Volume3, Number 4-Fall 1989- Pages 119- 135
The Political Economy of Trade Policy
Robert E. Baldwin
International trade seems to be a subject where the advice of economists is
routinely disregarded. Economists are nearly unanimous in their general opposition
to protectionism, but the increase in U.S. protection in recent years in such
sectors as automobiles, steel, textiles and apparel, machine tools, footwear and
semiconductors demonstrates that economists lack political influence on trade policy.
The type of protectionism chosen does not follow economists' advice, either. A
frequently asked question on undergraduate trade exams is why a small country's
welfare losses are less when it curtails imports with a tariff rather than by negotiating
"voluntary" export-restraint agreements (VEAs) with foreign suppliers.' Even though
generations of students have correctly pointed out that the equivalent of the domestic
tax revenues raised by a tariff is transferred as a windfall gain to foreign countries
when VEAs are introduced, these agreements are now the preferred means by which
countries pursue protectionism. Moreover, if the purpose of protection is to redistribute
income to producers, production subsidies (financed by lump-sum taxes)
dominate both tariffs and import quotas on efficiency grounds, since the consumption
costs of protection are avoided. Yet governments generally prefer to assist industries by
providing import protection rather than production subsidies.
Economists have tended to attribute such disregard for their policy conclusions to
a lack of economic education. However, while many consumers still do not seem to
'As trade economists have long pointed out, if a country is large so that the import supply curves it faces are
upward sloping, import taxes can improve its terms of trade by decreasing the prices at which foreigners
supply their exports, An optimum set of such taxes coupled with lump-sum redistribution arrangements can
raise everyone's welfare in the country. However, if foreign countries are also large, retaliation on their part
can reduce all countries' welfare levels below what they were before the import taxes.
Robert E. Baldwin ts Hilldale Professor of Economics at the Uniuersity of Wisconsin, Madison,
Wisconsin.
120 Journal of Economic Persfiectiues
know where their economic self-interest lies when it comes to trade matters, it seems
clear that today's top policy leaders are quite aware of economists' conclusions on
most trade issues. Some economists, consequently, have begun trying to understand
the policy-making process that leads politicians to disregard the advice of economists
on issues of international trade.
Two broad approaches have been developed to analyze the political economics of
trade policy and the processes that generate protectionism.2One approach emphasizes
the economic self-interest of the political participants, while the other stresses the
importance of the broad social concerns of voters and public officials. This paper
outlines the nature of the two approaches, indicating how they can explain the above
anomalies and other trade policy behavior, and concludes with observations about
integrating the two frameworks, conducting further research, and making policy based
on the analysis.
The Economic Self-InterestApproach
Under the economic self-interest approach, an individual favors or opposes a
particular trade policy depending on whether the policy increases or decreases the
person's real income. Hillman (1989) provides a comprehensive survey of this ap-
proach.
In the standard two commodity, two factor Heckscher-Ohlin model with identical,
homothetic preferences and perfect competition, this perspective implies that if
the imported good is produced in a labor-intensive manner, a worker in an economy
facing fixed international terms of trade will favor an import duty, while a capitalist
will favor free trade.3 Consequently, under a majority voting rule, protection would
seem to be the outcome if there are more workers than capitalists. But if, as is
generally assumed in this model, income redistribution is a costless activity, free trade
will be chosen, since gainers under free trade (in this case, the capitalists) can always
more than compensate those who lose under this policy and still be better off than
under protection. The capitalists will tie the redistribution scheme to the voting
outcome.
he focus in this paper is on explaining the protection process, that is, on positive political economy.
However, there is also a body of literature that analyzes the welfare implications of lobbying for or against
protection and lobbying for the rents created by protection. Bhagwati (1988) describes the first type of
lobbying efforts as "downstream" activities and the second as "upstream" activities.
3 ~ h i sfo1low-s from the Stolper-Samuelson theorem, which states that the real return to the factor used
intensively in producing the import good (in this case, labor) will increase when the domestic price of the
import good increases. As Mayer (1984) has shown, if an individual supplies both labor and capital services,
the person will favor protection or free trade under the assumed factor-intensity conditions depending upon
whether his or her capital/labor endowment is less or greater than the economy's capital/labor endow-
ment.
Robert E. Baldwin 121
Costs of Redistribution and Voting
When the assumptions of the model are eased to allow for costs of redistribution
or voting, the conclusion that free trade will be chosen no longer holds. In the
situation described above, if redistribution costs plus the compensation to workers
required to secure a majority vote for free trade amount to more than the capitalists
gain from free trade, a compensation scheme tied to voting behavior will no longer be
undertaken by capitalists and popular vote will select the protection option.
The existence of voting costs also can lead to protectionism, especially in
situations where the benefits from free trade are quite small for many individual
voters. Consider a specific-factors model in which both capital and labor are
industry-specific (Baldwin, 1984). In these circumstances, the losses in real income to
individual workers and capital owners in an industry subject to a significant increase
in foreign competition are likely to exceed their individual costs of voting for
protection. In contrast, the fraction of income spent on the industry's product by
individual workers and capitalists employed in other sectors, who gain from the
decline in the import price, may well be so small that the higher costs of the protected
good for each are less than the per capita costs of voting against protection or
acquiring detailed information about its effects. Consequently, the comparatively
small number of people employed in the injured industry all vote for protection and
the large number of people working in other industries may not vote at all.
Money, Votes, and Coalition Politics
Protectionism may also be selected because of the well-known free-rider problem
associated with public goods (Olson, 1965). Trade policy has the characteristic of a
public good, since a beneficiary from a policy such as free trade cannot be excluded
from its benefits, even though the person does not contribute to the costs of obtaining
this policy. The decision to contribute campaign funds to officials favoring free trade
has the structure of a prisoners' dilemma game for individual consumers. If each
consumer believes his or her contribution is too small to affect the policy outcome, the
individual will conclude that it is best not to contribute, regardless of whether all other
consumers do or do not contribute. Thus, the dominant strategy for each consumer is
not to contribute in support of free trade, and the outcome is that all end up worse off
then if they had contributed.
As Olson points out, a common-interest group is more likely to overcome the free
rider problem and raise the funds needed for effective lobbying if the number of its
members is small or if the benefits from a particular policy are unevenly distributed,
since then some members of the group will have a significant economic stake in the
policy outcome. For a large firm with (say) 25 percent of the domestic industry's
market that is threatened by import competition, it may pay to lobby, no matter what
the other firms in the industry do.
However, in making this decision, such firms must also take into account the
anti-protection lobbying of large firms in other sectors who use the industry's output
as an intermediate production input. For example, as Destler and Ode11 (1987)
122 Journal of Economlc Perspecttves
document, the Caterpillar Company vigorously opposed the extension of steel quotas
in 1984, claiming to be the largest American steel user as well as the biggest exporter
of U.S.-made steel. Furthermore, in these authors' view, the active opposition of
copper users was the main reason for the failure of this industry to obtain import relief
in 1984.
In contrast to Olson, Caves (1976), in stressing the importance of the voting
strength of an industry in gaining protection, hypothesizes a negative relationship
between the level of protection and the degree of industry centralization both in
geographical and market share terms. He argues that a geographically dispersed
industry can obtain the support of a larger number of elected representatives than a
regionally centralized one, while a low level of firm concentration will attract greater
political support from customers and suppliers. However, although most countries
studied support the importance of the employment size of an industry as a determinant
of the level of protection, they also generally bear out Olson's contention that a
high level of industry protection is associated either with a small number of firms or a
high degree of market concentration (Anderson and Baldwin, 1987).4
The importance of an industry's voting strength for gaining protection is nowhere
better supported than in the textile and apparel industries. With nearly two million
employees in many firms scattered over a wide geographical area, these industries
have been highly successful in using their voting strength to gain import protection.
The approval of the Textile Bill of 1985, which tightened textile quotas, by both
branches of the Congress illustrates the industry's political clout. (However, the bill
was vetoed by President Reagan.) Tosini and Tower (1987) found that the proportion
of textile and apparel workers in the total work force of a congressional district or state
was the most significant variable explaining the pattern of voting.
In contrast to the textile and apparel sectors, which have managed to hold the
ratio of imports to consumption to less than 15 percent, the much smaller leather and
leather products industry has been unable to prevent the ratio from rising to over 25
percent. The import ratio for women's footwear, for example, rose from 5 percent in
1965 to 33 percent in 1985.
Differences in market structure seem to influence the manner in which industries
lobby for protection. As noted above, the textile and apparel sectors, which are
composed of a large number of employees and firms, rely mainly on the voting
pressure they can exert directly on individual members of Congress and the president
4 ~ a v e s(1976) finds industry protection to be negatively related in a significant manner to market
concentration. But Saunders (1980), in a later study of Canadian protection utilizing two-stage least squares
estimation, finds a positive (and nearly significant at the 10 percent level) relationship between concentration
and industry protection. When both the number of firms and the concentration ratio are included in
the same regression equation, the concentration variable has a negative (though insignificant) sign in some
of the countries studied, while the coefficienton the number of firms is generally significantly negative. One
possible reason for this may be a perception by elected officials that voters are skeptical about their
motivation in responding to the petitions of the dominant firms in a concentrated industry, but not to firms
in an industry where the number of companies is small but the concentration ratio is not high.
The Political Economy of Trade Policy 123
to gain their objectives (Bauer, Pool, and Dexter, 1972). They do not launch large
lobbying campaigns aimed at the general public. In contrast, the steel and auto
industries, which have also been very successful in securing import protection but
which are not as large in voting terms and are oligopolistically organized (thus
presumably having higher profits), spend large sums lobbying the general public as
well as public officials on the merits of their case for protection. Lacking the voting
power of textiles and apparel, they seem to believe it is necessary to convince the
public that granting protection is desirable on such grounds as "fairness," national
security, or the need for a temporary adjustment period, before legislators not
dependent directly on their votes will go along with protection. Industries like steel
and autos also tend to focus their lobbying efforts on the Senate, where the typical
member has more influence and faces the voters less often than in the House. Small
industries have little chance of receiving industry-specific protection, although they
can sometime free-ride on a politically powerful industry. They usually lobby through
general business associations and direct their efforts at securing general changes in the
trade laws that make it easier to obtain protection.
Whether it is an industry's voting power or financial strength that create
pressures for protection, these pressures are powerful enough that many presidents
have made protectionist concessions to improve the chances that Congress would
approve a general trade-liberalizing bill. President Eisenhower, for example, felt it
necessary to promise to impose oil import quotas to minimize the opposition of
congressional members from the oil and coal-producing states to the extension of the
Trade Agreements Act in 1955, while President Kennedy agreed to seek an international
quota system for cotton textiles in order to minimize opposition to the Trade
Expansion Act of 1962. President Nixon increased the likelihood of Congressional
approval of the Trade Act of 1974 by renewing steel import quotas and agreeing to
negotiate an international textile agreement covering man-made materials, wool, and
cotton.'
Triggers for Self-Interested Action
Olson (1983) has recently argued that, before a group organizes and undertakes
lobbying activities, a crisis like a sudden increase in imports or a decline in employment
may be necessary to focus attention on a group's common interests. This view is
supported by the success in resisting tariff cuts during the Tokyo Round of multilateral
trade-liberalizing negotiations (1974-1979) of industries in which employment
had been declining and import penetration ratios rising in recent years (Baldwin,
1985). The psychologists Kahneman and Tversky (1979, 1984) also find empirical
support for the view that individuals place a greater welfare weight on the loss of a
"'hen the bill first came before the House Ways and Means Committee in 1973, the textile and apparel
unions testified against the bill. However, by the time the bill reached the Senate Finance Committee in
1974, the Multifiber Arrangement had been approved, and they did not testify at all.
124 Journal of Economic Perspectives
given amount of income than on a gain of the same a m ~ u n t . ~This would seem to be
part of the explanation why protection is usually given only to industries in which
profits and employment are declining. Both workers and capitalists are prepared to
exert greater efforts to recoup income declines than to gain increases above historically
normal levels. In addition, firms may lobby more vigorously for protection when faced
with losses, because in bad times they are less concerned that greater protection will
attract new entrants to the industry.
One also expects vigorous efforts to secure protection (or resist trade liberalization)
in the face of significantly increased import competition by those industries
where a substantial part of earnings is due to industry-specific physical and human
capital or to the success of unions in raising wages. The extensive (and successful)
lobbying activities of the auto and steel industries, where fixed costs are substantial
and wages are 55 percent and 40 percent higher, respectively, than the average for all
manufacturing sectors, seems to be motivated by this consideration. Even union
leaders in the apparel industry, in which wages are low compared to manufacturing
generally but high relative to many service sector jobs, recognize the importance of
maintaining the rent component in their wages. Sol C. Chaikin, President of the
Ladies' Garment Workers, has often indicated that he would be glad to accept free
trade in the industry if someone could tell him where his members can find other jobs
at wages near what they are currently making.
However, when changes in protection as a result of international trade-liberalizing
negotiations or levels of protection across industries are analyzed, high levels of
human capital per employee in an industry are generally found to be associated with
high duty reductions and low levels of protection (Anderson and Baldwin, 1987; Ray,
1981). As is explained more fully later, these relationships support the social concerns
approach to trade-policy determination: specifically,the view that public officials seek
to protect workers whose skills and wages are relatively low from the market pressures
resulting from liberal trade policies.
If uncertainty is introduced into the self-interest model along with incomplete
insurance markets (du5 for example, to moral hazard and adverse selection), risk
averse economic agents may prefer protective policies that reduce the variability of
price or income fluctuations to free trade (Cassing, Hillman, and Van Long, 1986;
Newbery and Stiglitz, 1984). This may be why the developing countries have long
argued for international commodity stabilization schemes for most of the primary
products they export.
Greater certainty about the restrictive effects of quotas than tariffs seems to be a
major reason why protection-seekers generally lobby for quantitative restriction in
preference to tariffs. Also, their chances of obtaining this form of protection are
improved if they lobby for country-selective quantitative restrictions, such as VEAs.
When quantitative restrictions apply only to those countries that are the main sources
of increased imports, injured domestic industries (and government officials) do not
61am grateful to Joseph Stiglitz for calling my attention to the literature on this point.
Robert E. Baldwin 125
have to be concerned about counter-lobbying from other countries whose exports have
not been increasing and who would be hurt by overall restrictions.
The countries selected for export restraint are likely to lobby against protection,
but the windfall gains accruing to them from allocating the export quotas among their
own producers act to soften their opposition.' The importing government loses the
revenue that could be collected with a tariff but, since special interest groups recognize
the difficulty of securing general revenues through lobbying, this is not likely to lead to
objections from very many domestic interest groups. Furthermore, because VEAs are
made outside of the rules of the General Agreement on Tariffs and Trade (GATT),
government officials do not have to face the domestic political problem of reducing
protection in some other industry to compensate other countries for the increase in
protection in the injured industry or, if this is not done, of having foreign countries
retaliate by increasing protection against some other domestic industry.
The self-interest framework also can be used to explain why most governments
use tariffs and quotas more often than domestic subsidies for assisting injured
import-competing sectors. Subsidies are more transparent than tariffs and quotas,
since they appear in the government's budget and must be financed with taxes or by
borrowing. Consequently, industries seeking assistance are concerned that subsidies
will be more difficult to obtain politically than import protection and be provided for
a shorter time duration. Elected officials also seem to believe they will lose fewer votes
of those outside the injured industries and face less anti-protection lobbying pressures
if they supply assistance in the form of import protection rather than domestic
subsidies. The welfare loss to the general public is, of course, greater under import
protection, but poorer knowledge about the consequences of tariffs and quotas
compared to domestic subsidies means this does not translate into greater political
resistance.
A National Prisoner's Dilemma
The discussion to this point has dealt only with the domestic political economy of
trade policy, that is, the manner in which public officials and domestic interest groups
interact in determining a country's trade policy. But the manner in which public
officials interact with their counterparts in other governments also shapes the nature of
trade policies. Viewing these interactions in game theoretic terms has proved helpful
in understanding certain aspects of international political economy. For example,
Brander (1986) employs the structure of the prisoners' dilemma game to illustrate why
countries may end up with protection, even though they would all be better off under
free trade. Assuming we can use numbers to represent the net trading benefits
perceived by mercantilistically-minded political leaders, suppose two large countries,
A and B, each gains 400 (only the relative size of the numbers are important) if they
both follow a free trade policy. However, if one country protects and the other adopts
!As Krishna (1989) shows, quantitative restrictions such as VEAs also facilitate collusive behavior between
competing firms.
I26 Journal of Economic Perspectives
Figure I
A Perceived Prisoner's Dilemma in Trade Policy
Country E
Cooperate Defect
Cooperate 400, 400 50, 500
Country A
Defect 500.50 100, 100
free trade, the country imposing protection gains 500, while the other gains only 50,
since the protecting country gains employment and improved terms of trade at the
expense of the free trade nation. When both protect, each gains only 100.As is evident
from Figure 1, the dominant strategy for each country is to protect, since this yields
each country the highest gain no matter what strategy the other country follows. The
belief that they were involved in this sort of game may account for the protective
policies adopted by many industrial countries in the early 1930s, when the possibility
of increasing employment through unilateral protection was especially appealing.
The Social Concerns Approach
The basic idea of the social concerns model is that trade policies are explainable
mainly by the government's concern for the welfare of certain social and economic
groups and by its desire to promote various national and international goals. Among
the goals that have been most widely discussed are preserving the status quo
distribution of incomes, redistributing income, or building political power through
trade alliances. Proponents of the social concerns approach to analyzing trade policy
formation have not been very precise as to why governments seek to implement these
various goals, or why one goal is sought in one case and not in another. Most seem to
hold to the view that the general public endorses these objectives and that the
reelection possibilities of government officials are dependent on the extent to which
these goals are implemented. Presumably, they envision each person's voting behavior
as being guided by his or her social welfare function, possibly involving interpersonal
comparisons based on the individual's ethical postulates.
Since economists tend to prefer models that rely on self-interest,models of social
concerns are sometimes taken too lightly. But it should be recognized, as authors like
Kau and Rubin (1982) and Andreoni (forthcoming) point out, that many individuals
do not free ride even though it is in their narrow economic self-interest to do so, either
for ethical reasons or because they gain satisfaction from participating in these
The Political Economy of Trade Policy 127
activities. An illustration of this point is the refusal of many consumers to buy cheaper
foreign goods, even when the country of origin cannot be discerned by others, on the
grounds that foreign workers are being exploited or are taking jobs away from
American workers.
Concern for the Changing Distribution of Income
Corden (1974, p. 107) attempts to interpret trade policy with a conservative
social welfare function, namely that "any significant absolute reductions in real
incomes of any significant section of the community should be avoided." This means
that increases in income are given relatively low welfare weights and decreases very
high welfare weights. Corden believes that this particular set of social values is
especially useful in explaining the temporary protection often granted industries
seriously injured by rapid increases in imports.
The Reagan administration's pressuring of the Japanese into voluntarily limiting
their exports of autos to the United States so that the industry could have some time to
become competitive in the small car market seems to fit this objective of trade policy
especially well. Another example was the government's 1973 action of freezing the
price of domestically produced oil up to the output levels existing prior to the oil crisis
in that year. As Kalt (1983) finds, the pattern of voting in the Senate during the oil
crisis was motivated mainly by the desire to prevent a significant shift in income away
from oil consumers and in favor of domestic oil producers.
Other authors stressing the income distribution goals of government, like Cheh
(1974) and Lavergne (1983), argue that trade policies of governments are motivated
by a desire to minimize (or delay) adjustment costs, especially to workers. In
examining the Kennedy round of multilateral trade negotiations, Cheh found a
pattern of low tariff cuts in industries with high proportions of elderly workers,
declining employment, and rising import penetration ratios. These findings support
the adjustment-assistance hypothesis.
In contrast, Constantopoulos (1974) and Fieleke (1976) believe governments use
trade policy not just to maintain the status quo but to promote relative increases in the
standard of living of the lowest income groups. This view is supported by the finding
in most country studies that industries with high proportions of unskilled workers or
low wages tend to have high levels of protection and low tariff cuts in multilateral
trade negotiations (Anderson and Baldwin, 1987). U.S. industries for which this
relationship holds include textiles, apparel, leather and leather products, furniture and
fixtures, and miscellaneous industries such as toys and sporting goods, jewelry, and
musical instruments.
One incident illustrating the income distribution concerns of elected officials was
the threat in 1979 by many members of Congress to vote against approving the
government procurement code negotiated in the Tokyo Round, when they discovered
that preferences were no longer going to be given to small and minority domestic firms
in bidding for government contracts. Ambassador Strauss, the U.S. Trade Representative,
quickly negotiated an exception to the code's general principle of national
treatment and nondiscrimination that allowed set-asides for these firms.
128 Journal of Economic Perspectives
However, the use of trade policy to promote a more equitable distribution of
income can run into problems of inconsistency over time, as Staiger and Tabellini
(1987) argue. They consider the case of a country where free trade is the Pareto
optimal policy. In this case, the government will have an incentive to announce a
policy of free trade to achieve the optimal allocation of labor, but then to surprise the
workers with a tariff that improves the distribution of income by raising wages in the
import-injured industry. The original announcement of a free trade policy is, therefore,
not time-consistent. In this model, the import competition must cause price to
decline enough in the industry so that it is advantageous for some workers to move
even if they must accept a lower wage (because of their lower level of industry-specific
skills). But after the tariff is imposed, the workers who had moved to another industry
will not return to the industry of their initial employment because they would have
lost their sector-specific skills and thus will have to start again as unskilled workers.
However, workers will recognize that because the government cares about income
redistribution, the policy commitment to free trade is not credible. Workers will come
to expect the tariff and fewer will leave the import-threatened industry after a price
decline. Yet the very fact that they do not leave will make the tariff even more
necessary on income-distribution grounds, to prevent wages in the industry from
falling even below the level that could be earned as unskilled labor in other industries.
The final outcome is a distribution of income where the expectation of a tariff
prevents the economy from adjusting, thus lowering social welfare due to the production
and consumption distortions associated with protection. Then the enactment of
the tariff restores the income distribution only to a level that would have been
achieved by a policy of free trade. The loss of economic efficiency buys no gain in
equity.
Thus, according to Staiger and Tabellini, the inability of a country to precommit
to a liberal trade policy becomes the reason for protection. One policy implication of
this analysis, according to the authors, is that a commitment to a simple set of trading
rules, which preclude the idea of the government attempting or the industry expecting
a policy surprise, may often be superior to an activist but discretionary trade policy.
This analysis may help to account for the long-standing import protection of such
industries as textiles, apparel, and sugar. The preamble of the international Multifiber
Arrangement (MFA) for textiles has for many years included as an objective "the
reduction of trade barriers and the liberalization of world trade in these products"
(General Agreement on Trade and Textiles, 1974). However, producers in the
industrial countries know that the MFA also permits quantitative restrictions as a
safeguard against injurious import increases. Consequently, they have an incentive not
to undertake the difficult process of adjustment needed to bring about trade liberalization,
since they know they can expect continued protection if they do not adjust.
Deardorff (1987) argues that income distribution concerns also may explain why
governments seemingly prefer to handle import surges with VEA's rather than tariffs.
As he points out, an industry can be seriously injured by increased imports (and thus
qualify for protective action by the government under the safeguard provisions of the
G A n ) before the full extent of the price-reducing increase in imports is known.
Robert E. Baldwin 129
Consequently, if policy in trade matters is guided by Corden's conservative social
welfare function, namely, that income reductions for any significant group should be
avoided, the government will be reluctant to use a tariff to provide import relief
because of the uncertainty about the extent of the price reduction. Too low a tariff will
leave productive factors in the import-competing sector worse off than before the
import surge, while one that is set too high will reduce the income level of the
productive factors employed in other sectors below what they had been receiving. (For
simplicity, Deardorff assumes all factors to be sector-specific.)
In contrast, an externally allocated quota set at the import level prevailing prior
to the increase in imports-or, what is essentially the same, an agreement by foreign
suppliers to restrain their exports to these levels-holds welfare levels of both groups
of domestic factors at their pre-import surge levels. Furthermore, if the increased
imports are from only a few foreign countries, such a VEA can raise the welfare of
these exporting countries above pre-import surge levels due to the quota rents they
gain and, at the same time, maintain the income levels of other exporting nations.
However, as Deardorff recognizes, the motivation for trying to avoid harm to foreign
countries may be related more to national self-interest, namely the fear of retaliation
or loss of goodwill, than to a social concern not to take policy actions that reduce the
income levels of other countries.
Income distribution concerns may also be a reason why governments generally
prefer to assist an import-injured industry by protective measures rather than by
providing a production subsidy. By restricting imports and thereby raising the price of
the industry's product, the government puts the burden of helping the injured industry
on those consumers who benefitted from the initial decrease in the price of this good.
If the price returns to its level prior to the increase in imports and even if the tariff
revenue is not returned to the good's consumers or its equivalent is given to foreign
producers, these consumers are no worse off than initially. In contrast, a production
subsidy financed by general taxes leaves those taxpayers who do not consume the
import good worse off than initially.'
Trade Policy as Foreign Policy
The social concerns approach can be extended to cover various international
objectives of the government. In fact, foreign policy has been used as a justification
both for protectionism and for liberalizing trade."
Gilpin (1987) and others have argued that hegemonic states, such as the United
States after World War I1 and Great Britain during the latter part of the nineteenth
century, adopt liberal trade policies as a mean of inducing other countries to accept
%ince the sum of consumer and producer surplus is greater when a production subsidy rather than a tariff
or quota is used to restore income levels in the injured industry (even if thr tariff rrvenue or its equivalent is
retained in the domestic econom)), this argument assumes that income-redistribution measurrs by the
govrrnment are not feasible.
'~ovrrnmrnt policies such as national defense can be explained on thc basis of the narrow self-interest
motivations of citizens, whereas the concerns for the welfare of others discussed up to this point are based on
forms of altruism, such as an implicit social insurance contract.
130 Journal of Economic Perspectives
their political leadership in the world. The major international political objective of
the United States in the early postwar period was to limit the spread of Communism
by strengthening the free world economically, and these writers maintain that the
liberal U.S. trade and aid policies in this period were largely motivated by this
purpose. This may explain why the United States (and Great Britain in the nineteenth
century) did not take advantage of its economic power to improve its terms of trade
nor to prevent small industrial countries and developing nations from free-riding on
its trade concessions and those of other larger countries. National security considerations
still play an important role in trade policy determination, as the US.-Israel free
trade agreement indicates.
Income distribution concerns at the international and domestic levels also may
influence countries' foreign economic policies. For example, the comparatively large
amounts of untied aid and extensive duty-free treatment provided to developing
countries by a number of small industrial countries seems difficult to explain entirely
on self-interest grounds. Helleiner (1977) suggests that government officials adopt a
mercantilistic attitude in trade negotiations and try to obtain the greatest cuts in
protection from others in return for the least cuts by their own country. Minimizing
the domestic adjustment problems associated with trade liberalization could be one
motivation for this attitude.
Some Conclusions
As economists will readily recognize, protectionism does not seem to be just a
matter of which sectors have the best social welfare reasons for public assistance.
Within a range of social values, the particular social justifications for various trade
policies that are accepted by the electorate seem to be significantly influenced by
lobbying activities on the part of economic interest groups and the government itself.
In fact, economic self-interest almost always dominates a person's concern for the
welfare of other groups or the nation as a whole, when a significant part of an
individual's income is affected by a trade policy. One does not, for example, observe
workers or managers in import-injured industries opposing protection for their sectors
on the grounds that this may cause unemployment in other industries and reduce the
real income levels of low-income consumers. National and group concerns are likely to
dominate personal economic welfare considerations only when the economic self-interest
effects of a trade policy on an individual are small or unclear. Furthermore. the
larger the decline in individual economic welfare as a person chooses to support
various social goals that only benefit others, the less willing the person seems to be to
sacrifice additional economic welfare to promote additional desirable social objectives.
However, the economic self-interest approach cannot stand alone, either. In its
pure form, it requires an electorate that is ill-informed or unwilling to participate in
the decision-making process due to the high costs of political action or to the free rider
problem. But the considerable attention given to trade issues by the modern media,
the extensive efforts undertaken by politicians and others to determine the views of
The Political Economy of Trade Policy 131
voters, the widespread lobbying activities by all sorts of common-interest groups, and
the fact that a substantial number of citizens do vote in elections all suggest that the
economic self-interest model by itself is inadequate for explaining either general trade
policies or particular industry trade policies.
Many individual decisions on trade and other economic matters have only a
small effect on a person's real income, and the individual's various social concerns can
play an important role in shaping his or her decisions. To expand their already
substantial contributions toward understanding the policymaking process, economists
should integrate such social motivations into their microeconomic optimizing framework.
In doing so, they should also include in their models the various imperfections
that characterize real world economic and political markets.
I believe that the studies presented here establish that an analytical framework
including both economic self-interest and concern for the welfare of others as
motivating forces for political action is needed for understanding policy formation in
the trade field. The economic self-interest and social concerns approaches are not
contradictory; both motivations are included in the standard formulation of an
individual's social welfare function. Under Corden's conservative social welfare function,
for example, the typical voter is willing to accept the small decrease in his or her
own economic welfare and the welfare of other consumers that results from a tariff
increase on an imported product, if this prevents a significant real income decline for
workers in an important domestic industry. Consequently, declining industries are
more likely to receive protection than prosperous ones, not only because of the
self-interested lobbying from capitalists and workers who face income losses, but
because the typical voter is more willing to grant protection to a declining industry.
Directions fpr Further Research
A multitude of possible reasons for protection have been outlined. The problem
now is to weigh the reasons against one another, weed out those that are less relevant,
and combine the others into a more unified theory. To make progress along these
lines, there is a need to test the various hypotheses in the self-interest and social
concern models with variables that better reflect the key ideas behind the two
approaches.
Efforts should be made, for example, to obtain better measures of the extent of
political pressures from an industry than such indirect measures as the number of
firms in the sector, the industry's concentration ratio, or its growth rate. Information
on the size of an industry's lobbying expenditures on a particular issue, the volume of
testimony and public statements on the issue by its members, the extent to which
industry members make their views known to public officials through visits and letters,
and so forth would enable investigators to measure the political demands for protection
more accurately.
Similarly, as Bale (1977) has shown, better measures than growth rates in
employment or the proportion of older workers in an industry can be estimated to
measure the labor adjustment costs associated with increased imports. More complete
socioeconomic information on workers in an industry than just wages alone should
I32 Journal of Economic Perspectives
also be used to measure the conditions under which voters and policy-makers are
willing to grant protection on equity grounds. While obtaining improved data is costly
and time-consuming, the effort seems necessary if we are to make significant progress
in understanding the relative importance of the various political and economic
determinants of trade policy.
Policy Recommendations
If economists wish their advice to carry weight in the political process, they must
be more willing to examine the influences of institutions and procedures on policies.
O n the international level, for example, the ability of an organization such as the
GATT to obtain adherence to an agreed-upon set of trading rules and to settle
disputes among nations is crucial for the degree of order in world trading conditions.
Domestic examples of the importance of political institutions for the degree of
protectionism are easy to find. Since 1934, the willingness of the Congress to allow the
executive branch to handle most modifications of individual tariffs has had a major
effect in reducing protection over the years. since the president has traditionally been
more liberal on trade-policy matters than Congress, probably because of the special
foreign policy responsibilities of the president. Congress has rec~ntlyattempted to
push the executive branch in a more protectionist direction. For example, the
Omnibus Trade and Competitiveness Act of 1988 shifts responsibility for the enforcement
of U S . rights under trade agreements (so-called Section 301 responsibilities)
from the pres~dentto the US. Trade Representative, a political appointee subject to
Senate confirmation.
The manner in which the International Trade Commission (ITC) and the
president announce their import-injury decisions illustrates the importance of procedures
in shaping policy decisions. The ITC is charged with determining whether an
industry has suffered serious injury (or the threat thereof) as a result of increased
imports. and the public report it issues after its investigation deals only with this issue.
The possible effects of protection on the welfare of consumers and industries using the
product as an intermediate production input, as well as on the national interest
generally, are considered by the president only after receiving an affirmative injury
decision from the ITC. Furthermore, the report of the interagency committee established
in the executive branch to consider these and other implications of granting
protection is not published. In addition, Congress has prohibited members of the
International Trade Commission who vote negatively in import-injury decisions,when
the majority of members vote affirmatively, from participating in the Commission's
decision determining how much protection to recommend to the president. The
president usually issues only a very brief statement explaining whether protection is
granted and, if so. to what degree and in what form. Thus, while the public learns
about the economic problems faced by the petitioning industry and the benefits firms
will receive from protection, it learns very little about the social costs of the protection.
This government practice of giving the public only one side of the story biases the
entire process toward greater protection. .4 requirement that the interagency report be
published or that some federal agency (perhaps the Council of Economic Advisers?)
Robert E. Baldwin I33
investigate the costs of granting protection and publish the results of its study would
bring a much better balance to the procedures. If the decision-making process for U.S.
trade policy is made more open and transparent to the general public, it would tend
to reduce import protection or change the form of industry assistance.
Economists should also be careful to distinguish between the economic and
political feasibility of policies. A number of explanations for protection provided in
this paper rely on the inability of government to use theoretically first-best policies to
maximize social welfare for economic reasons. For example, in accounting for protection
in an environment of price uncertainty, it was assumed to be impossible to
establish complete insurance markets to deal with the uncertainty because of moral
hazard or adverse selection. Similarly, redistribution by means of lump-sum taxes and
subsidies was ruled out as being too costly in Staiger and Tabellini's explanation of
why social welfare-maximizing governments adopt protection and in Deardorffs
account of why governments use Pareto-inferior means of protection.
In other cases, certain policies may be economically feasible but appear to be
politically infeasible. For most of the industries receiving import relief, especially such
industries as textiles, apparel, sugar and steel that have been protected for many years,
there appear to be economically feasible policies other than protection for meeting the
income distribution concerns about those employed in these sectors, policies that
would raise social welfare above the level associated with protection. However, as
efforts to deal with the income problems of farmers by using income transfers indicate,
groups being assisted by the government often use their political power to block the
use of these alternative redistribution means to increase the likelihood they will
continue to receive assistance in the form they are used to.
If they are to make policy recommendations, economists need better empirical
and experimental studies of the economic and political feasibility of well-known policy
measures as well as greater efforts to devise feasible new policies and political means
for carrying out generally accepted economic and social goals. An illustration of
promising economic research along these lines is the recent work (Woodbury and
Spiegelman, 1987; Lawrence and Litan, 1986) aimed at devising adjustment-assistance
measures that provide the same income assistance to displaced workers as
existing programs, but encourage economic adjustment rather than unemployment
and immobilized resources. An example at the theoretical level of this sort of analysis
is the work of Dixit and Norman (1980; 1986),which demonstrates that a system of
taxes and subsidies on goods and factors can accomplish the same redistributive goals
as lump-sum taxes and subsidies, while requiring less information (and thus presumably
being less costly) and also avoiding the many practical problems of lump-sum
transfers.
Even if further research along these lines confirms that there are some real world
situations where protection would be the best available measure for maximizing social
welfare, economists should be very cautious about surrendering the efficiency argument
for liberal trade policies. A final insight from viewing how trade policy is made
is the readiness of protectionist entrepreneurs to twist the policy analyses of economists
to promote their own welfare at the cost of national welfare. For example, arguments
134 journal of Economic Perspectives
based on the so-called new trade theory emphasizing imperfect competition and
increasing returns are beginning to be made for providing government assistance to a
wide variety of industries, even though the empirical work done so far in this field
suggests that the case for government intervention on strategic grounds in imperfectly
competitive markets is quite limited (Richardson, 1989). Presuming that a liberal
trade policy promotes national welfare best and permitting exceptions only if based on
careful empirical documentation is likely to yield the highest national welfare in the
real world of p o l i ~ ~ m a k i n ~ . ' ~
References
Anderson, Kym, and Robert E. Baldwin, "The
Political Market for Protection in Industrial
Countries." In El-Agraa. Ali, M., Protectzon, Cooperalion,
Integralion and Developmenl. London:
Macmillan Press, 1987, 20-36.
Andreoni, James, "Why Free Ride? Strategies
and Learning in Public Goods Experiments,"
Journal of Public Economics, forthcoming.
Baldwin, Robert E., The Political Economy of
U.S. Imporl Policy. Cambridge. The MIT Press,
1985.
Baldwin, Robert E., "Kent Seeking and Trade
Policy: An Industry Approach," Weltz~irtschaftliches
Archiu, Heft 4, 1984, 120,646-677.
Bale, Malcolm D., "United States Concessions
in the Kennedy Round and Short-Run Labor
Adjustment Costs," Journal oj Inlernational Economics,
May 1977, 7, 145-148.
Bauer, Raymond A., Ithiel De Sola Pool, and
Lewis Anthony Dexter, American Business and Public
Policy: The Policy of Foreign Trade. Chicago:
Aldine-Atherton, 1972.
Bhagwati, Jag.dish, "Is Free Trade Passe After
All?" Department of Economics, Columbia University,
1988.
Brander, James A., "Rationales for Strategic
Trade and Industrial Policy." In Krugman, Paul
R., ed. Strategic Trade Policy and the Neiu International
Economics. Cambridge: MIT Press, 1986.
Caves, Richard E., "Economic Models of Political
Choice: Canada's Tariff Structure," Canadian
Journal of Economics, May 1976, 9, 278-300.
Cassing, James H., Arye L. Hillman, and R'go
Van Long, "Political-Influence Motives and
Choice Between Tariffs and Quotas," Journal of
Internalional Economics, November 1986, 19,
279-290.
Corden, W. M., Trade Policy and Economic I+%/fare.
Oxford: Clarendon Press, 1974.
Constantopoulos,Maria, "Labor Protection in
Western Europe," European Economic Reuzews, December
1974, 5, 313-328.
Cheh, John H., "United States Concessions in
the Kennedy Round and Short-Run Labor .4djustment
Costs," Journal qf International Economics,
November 1974, 4, 323-340.
Deardorff, Alan V., "Safeguards Policy and
the Conservative U'elfare Function." In
Kierzkowski, Henryk, ed., Protection and Competition
in Internalional Trade. Oxford: Basil Blackwell Ltd,
1987.
Destler, I. M., and John S. Odell, Anti-Proteclion:
Changing Forces in United Stales Trade Politics,
Policy Analyses in International Economics, No.
21, Washington: Institute for International Economics,
September 1987.
Dixit, Avinash and Victor Norman, Theory of
International Trade: A Dual Equilibrium Approach.
Cambridge: Cambridge University Press, 1980.
Dixit, Avinash, and Victor Norman, "Gains
from Trade without Lump-Sum Compensation,"
Journal oj International Economzcs, August 1986, 21,
111-122.
Fieleke, Norman, "The Tariff Structure for
10
In this journal, Krugman (1987) reached a rather similar conclusion.
Manufacturing Industries in the United States: A
Test of Some Traditional Explanation," Columbia
Journal of World Btuiness, Winter 1976, 11, 98-104.
General Agreement on Tariffs and Trade, Arrangement
Regarding Internalional Trade in Textiles,
Geneva: General Agreement on Tariffs and Trade,
1974.
Gilpin, Robert, The Polil~calEconomy oJ Inlernational
Relations. Princeton: Princeton University
Press, 1987.
Helleiner, G. K. "The Political Economy of
Canada's Tariff Structure: An Alternative
Model," Canadian Journal oJ Economics, May 1977,
4, 318-326.
Hillman, Arye L., The Political Economy of Prolection.
New York: Harwood Academic Publishers,
1989.
Kahneman, Daniel and Amos Tversky, "Prospect
Theory: An Analysis of Decision under
Risk," Economelrica, March, 1979, 47, 263-291.
Kahneman, Daniel and Amos Tversky,
"Choices, Values, and Frames," American Psychologirl,
April 1984, 4, 341-350.
Kalt, Joseph P., The Economics and Politics of Oil
Price Regulalion: Federal Policy in lhe Posl-Embargo
Era. Cambridge: MIT Press, 1983.
Kau, James B., and Paul Rubin, Congressmen,
Consliluenls, and Conlribulors. Boston: Martinus
Nijhoff Publishing, 1982.
Krishna, Kala, "Trade Restrictions as Facilitating
Practices," Journal of International Economics,
May 1989, 26, 25 1- 170.
Kmgman, Paul R., "Is Free Trade Passe?"
Journal oJ Economic Perspeclives, Fall 1987, 1,
131-144.
Lavergne, Real, The Poltlical Economy oJ L1.S.
Tarfls: An Emp~rical Analysis. New York:
Academic Press, 1983.
Lawrence, Robert Z., and Robert E. Litan,
2% Political Economy of TradePolicy 135
Saving Free Trade: A Pragmatic Approach. Washington,
D.C.: The Brookings Institution, 1986.
Mayer, Wolfgang, "Endogenous Tariff Formation,"
American Economic Reuiew, December 1984,
74, 970-985.
Newbery, David M. G., and Joseph E. Stiglitz,
"Pareto Inferior Trade," Review qf Economic Studies,
LI, 1-12, 1984.
Olson, Mancur, The Logic OJ Collective Aclion.
Cambridge: Harvard University Press, 1965.
Olson, Mancur, "The Political Economy of
Comparative Growth Rates." In Mueller, Dennis,
ed., The Political Economy of Growth. New Haven:
Yale University Press, 1983.
Ray, Edward J., "The Determinants of Tariff
and Nontariff Trade Restrictions in the United
States," Journal of Polilical Economy, February 1981,
89, 105-121.
Richardson,J. David, "Empirical Research on
Trade Liberalizations with Imperfect Competition:
A survey," OECD Economic Studies, Spring
1989.
Saunders, Ronald S., "The Political Economy
of Effective Protection in Canada's Manufacturing
Sector," Canadian Journal of Econom~cs, May
1980. 13, 340-348.
Staiger, Robert W., and Guido Tabellini,
"Discretionary Trade Policy and Excessive Protection,"
American Economic Review, December 1987,
77, 823-837.
Tosini, Suzanne, and Edward Tower, "The
Textile Bill of 1985: The Determinants of Congressional
Voting Patterns," Public Chotce, 1987,
54, 19-25.
Woodbury, Stephen A., and Robert G.
Spiegelman, "Bonuses to Workers and Employers
to Reduce Unemployment: Randomized Trials in
Illinois," American Economic Review, September
1987, 77,513-530.
You have printed the following article:
The Political Economy of Trade Policy
Robert E. Baldwin
The Journal of Economic Perspectives, Vol. 3, No. 4. (Autumn, 1989), pp. 119-135.
Stable URL:
http://links.jstor.org/sici?sici=0895-3309%28198923%293%3A4%3C119%3ATPEOTP%3E2.0.CO%3B2-S
This article references the following linked citations. If you are trying to access articles from an
off-campus location, you may be required to first logon via your library web site to access JSTOR. Please
visit your library's website or contact a librarian to learn about options for remote access to JSTOR.
[Footnotes]
3
Endogenous Tariff Formation
Wolfgang Mayer
The American Economic Review, Vol. 74, No. 5. (Dec., 1984), pp. 970-985.
Stable URL:
http://links.jstor.org/sici?sici=0002-8282%28198412%2974%3A5%3C970%3AETF%3E2.0.CO%3B2-P
4
Economic Models of Political Choice: Canada's Tariff Structure
Richard E. Caves
The Canadian Journal of Economics / Revue canadienne d'Economique, Vol. 9, No. 2. (May, 1976),
pp. 278-300.
Stable URL:
http://links.jstor.org/sici?sici=0008-4085%28197605%299%3A2%3C278%3AEMOPCC%3E2.0.CO%3B2-G
4
The Political Economy of Effective Tariff Protection in Canada's Manufacturing Sector
Ronald S. Saunders
The Canadian Journal of Economics / Revue canadienne d'Economique, Vol. 13, No. 2. (May,
1980), pp. 340-348.
Stable URL:
http://links.jstor.org/sici?sici=0008-4085%28198005%2913%3A2%3C340%3ATPEOET%3E2.0.CO%3B2-6
http://www.jstor.org
LINKED CITATIONS
- Page 1 of 3 NOTE:
The reference numbering from the original has been maintained in this citation list.
10
Is Free Trade Passé?
Paul R. Krugman
The Journal of Economic Perspectives, Vol. 1, No. 2. (Autumn, 1987), pp. 131-144.
Stable URL:
http://links.jstor.org/sici?sici=0895-3309%28198723%291%3A2%3C131%3AIFTP%3E2.0.CO%3B2-F
References
Economic Models of Political Choice: Canada's Tariff Structure
Richard E. Caves
The Canadian Journal of Economics / Revue canadienne d'Economique, Vol. 9, No. 2. (May, 1976),
pp. 278-300.
Stable URL:
http://links.jstor.org/sici?sici=0008-4085%28197605%299%3A2%3C278%3AEMOPCC%3E2.0.CO%3B2-G
The Political Economy of Canada's Tariff Structure: An Alternative Model
G.K. Helleiner
The Canadian Journal of Economics / Revue canadienne d'Economique, Vol. 10, No. 2. (May,
1977), pp. 318-326.
Stable URL:
http://links.jstor.org/sici?sici=0008-4085%28197705%2910%3A2%3C318%3ATPEOCT%3E2.0.CO%3B2-T
Prospect Theory: An Analysis of Decision under Risk
Daniel Kahneman; Amos Tversky
Econometrica, Vol. 47, No. 2. (Mar., 1979), pp. 263-292.
Stable URL:
http://links.jstor.org/sici?sici=0012-9682%28197903%2947%3A2%3C263%3APTAAOD%3E2.0.CO%3B2-3
Is Free Trade Passé?
Paul R. Krugman
The Journal of Economic Perspectives, Vol. 1, No. 2. (Autumn, 1987), pp. 131-144.
Stable URL:
http://links.jstor.org/sici?sici=0895-3309%28198723%291%3A2%3C131%3AIFTP%3E2.0.CO%3B2-F
http://www.jstor.org
LINKED CITATIONS
- Page 2 of 3 NOTE:
The reference numbering from the original has been maintained in this citation list.
Endogenous Tariff Formation
Wolfgang Mayer
The American Economic Review, Vol. 74, No. 5. (Dec., 1984), pp. 970-985.
Stable URL:
http://links.jstor.org/sici?sici=0002-8282%28198412%2974%3A5%3C970%3AETF%3E2.0.CO%3B2-P
Pareto Inferior Trade
David M. G. Newbery; Joseph E. Stiglitz
The Review of Economic Studies, Vol. 51, No. 1. (Jan., 1984), pp. 1-12.
Stable URL:
http://links.jstor.org/sici?sici=0034-6527%28198401%2951%3A1%3C1%3APIT%3E2.0.CO%3B2-T
The Determinants of Tariff and Nontariff Trade Restrictions in the United States
Edward John Ray
The Journal of Political Economy, Vol. 89, No. 1. (Feb., 1981), pp. 105-121.
Stable URL:
http://links.jstor.org/sici?sici=0022-3808%28198102%2989%3A1%3C105%3ATDOTAN%3E2.0.CO%3B2-N
The Political Economy of Effective Tariff Protection in Canada's Manufacturing Sector
Ronald S. Saunders
The Canadian Journal of Economics / Revue canadienne d'Economique, Vol. 13, No. 2. (May,
1980), pp. 340-348.
Stable URL:
http://links.jstor.org/sici?sici=0008-4085%28198005%2913%3A2%3C340%3ATPEOET%3E2.0.CO%3B2-6
Discretionary Trade Policy and Excessive Protection
Robert W. Staiger; Guido Tabellini
The American Economic Review, Vol. 77, No. 5. (Dec., 1987), pp. 823-837.
Stable URL:
http://links.jstor.org/sici?sici=0002-8282%28198712%2977%3A5%3C823%3ADTPAEP%3E2.0.CO%3B2-K
Bonuses to Workers and Employers to Reduce Unemployment: Randomized Trials in Illinois
Stephen A. Woodbury; Robert G. Spiegelman
The American Economic Review, Vol. 77, No. 4. (Sep., 1987), pp. 513-530.
Stable URL:
http://links.jstor.org/sici?sici=0002-8282%28198709%2977%3A4%3C513%3ABTWAET%3E2.0.CO%3B2-R
http://www.jstor.org
LINKED CITATIONS
- Page 3 of 3 NOTE:
The reference numbering from the original has been maintained in this citation list.