Content uploaded by Richard Toye
Author content
All content in this area was uploaded by Richard Toye
Content may be subject to copyright.
The Origins and Interpretation of the
Prebisch-Singer Thesis
John Toye and Richard Toye
The Prebisch-Singer thesis is generally taken to be the proposition that
the net barter terms of trade between primary products (raw materials)
and manufactures have been subject to a long-run downward trend. The
publication dates of the first two works in English that expounded the
thesis were nearly simultaneous. In May 1950, the English version of
The Economic Development of Latin America and Its Principal Prob-
lems, by Raúl Prebisch, appeared under the UN’s imprint. In the same
month Hans Singer published an article, “The Distribution of Gains be-
tween Investing and Borrowing Countries,” in the American Economic
Review. The continuing significance of the “Prebisch-Singer thesis” is
that it implies that, barring major changes in the structure of the world
economy, the gains from trade will continue to be distributed unequally
(and, some would add, unfairly) between nations exporting mainly pri-
mary products and those exporting mainly manufactures. Further, in-
equality of per capita income between these two types of countries will
Correspondence may be addressed to JohnToye, Department of Economics, University of Ox-
ford, Manor Road, Oxford OX1 3UQ, United Kingdom; e-mail: john.toye@economics.oxford.
ac.uk. The authors are grateful for the generous assistance they have received from José-
Antonio Ocampo and the staff of ECLAC and ILPES in Santiago; from Aurora Tang-Keko
at the UN Archive in New York; and from the United Nations Intellectual History Project and
its directors, Louis Emmerij, Richard Jolly, and Tom Weiss. We especially thank Sra. Adela
Prebisch for permission to quote from the Prebisch papers in her possession. Valuable com-
ments and materials were received from James Boughton, Edgar J. Dosman, Valpy FitzGerald,
James N. Miller, John Shaw, and two anonymous referees. Translations are the authors’ own,
and all errors remain their sole responsibility.
History of Political Economy 35:3 © 2003 by Duke University Press.
438 History of Political Economy 35:3 (2003)
be increased by the growth of trade, rather than reduced. This could be,
and has been, taken as an indicator of the need for both industrialization
and tariff protection.
Prebisch and Singer identified two types of negative effects on pri-
mary producers’terms of trade. One effect occurs because of systemati-
cally different institutional features of product and factor markets, such
as cost-plus pricing and the unionization of labor in industry. Another
negative influence is that of technical progress, both from the asymmet-
ric distribution of its fruits, but also from its asymmetric impact on future
demand, favorable to that of industry while unfavorable to that of agri-
culture. The empirical significance of the thesis has been much disputed
and continues to be controversial after more than fifty years. One recent
investigation has claimed that these two effects have operated strongly
in the forty years after the Second World War, and that they have indeed
outweighed the positive influences on primary producers’terms of trade
arising from capital accumulation and the growth of industrial produc-
tion. This particular study suggested that the economic mechanisms that
disfavor primary product producers, which were specified by Prebisch
and Singer, have had significant impacts, even though the net secular
decline of primary producers’ net barter terms of trade has been found
to be relatively small, at around 1 percent a year (Bloch and Sapsford
1998).
The Prebisch-Singer thesis contradicted a long tradition of contrary
belief among economists. The nineteenth-century English political econ-
omists believed that the terms of trade of industrial manufactures relative
to agricultural produce would tend to decline. This belief underpinned
their pessimism about the sustainability of rapid population growth. That
manufactures’ terms of trade would decline, and that rapid population
growth was therefore unsustainable, were two propositions that caused
political economy to be dubbed the “dismal science.” This basic frame-
work of ideas remained remarkably stable throughout the entire century
and a quarter from Robert Malthus to the early works of John Maynard
Keynes. Although, by the late 1940s, this proposition was rarely stated
explicitly, when Prebisch and Singer came to reverse the classical ex-
pectation of declining terms of trade for manufactures, their conclusions
were immediately controversial, and are still so regarded by some to-
day.1
1. For an extended discussion, see Toye 2000, chap. 1. When criticizing the views of Pre-
bisch and Singer, Jacob Viner (1952, 114) and Gottfried Haberler (1988, 39–40) both made
reference to this doctrine, although without endorsing it themselves.
Toye and Toye / The Prebisch-Singer Thesis 439
Prebisch is frequently credited with having formulated the declin-
ing terms of trade thesis before Singer did. Joseph Love (1980, 58–59)
claimed that “Prebisch clearly seems to have reached his position earlier
than Singer.” Other authors have held that Singer discovered the the-
sis independently and simultaneously. Cristobal Kay (1989, 32) wrote
that “Singer...reached his conclusions independently from Prebisch
and around the same time [so that] the thesis on the deterioration in
the terms of trade is known in the economic literature as the ‘Prebisch-
Singer thesis.’” This second view was indeed that held by Singer him-
self.2Our account of the events surrounding the United Nations Eco-
nomic Commission for Latin America (ECLA) conference in Havana
in May 1949 reveals that Prebisch did not discover independently that
the terms of trade of primary products were secularly declining, but re-
lied wholly on the previous work of Singer. The false impression that he
had made the discovery (either first or simultaneously) was the conse-
quence of political tensions between the developed and the underdevel-
oped countries that had welled up at Havana, and the way in which the
top administrators of the United Nations secretariat responded to those
tensions.
For at the time Prebisch and Singer made their respective contribu-
tions, both were working for the UN. Prebisch was working for ECLA
in Santiago de Chile and Singer was in the Department of EconomicAf-
fairs (DEA) at UN headquarters in NewYork. It might seem odd that the
United Nations, whose role it was to find solutions to world economic
problems in order to promote peace, should be the cradle of such a con-
troversial doctrine, one that lent itself so readily both to the economic
nationalism of the underdeveloped countries and to the polemics of the
Cold War. Here we tell exactly how this came about, and show how, in
fact, it was the UN’s own eagerness to disclaim responsibility for the
doctrine that brought Prebisch into the limelight. The consequent wide-
spread belief that Prebisch had first made the discovery in turn invested
the thesis with an enhanced significance in hemispheric politics. More-
over, the UN’s failed stratagem for distancing itself from Prebisch’s ideas
had the unintended consequence of making the world organization itself
appear as a nursery of Latin American economic radicalism.
2. Singer recalled, “My discovery that Prebisch thought along lines that were so congenial
to me came after I had drafted my paper.” This suggests a belief that Prebisch had arrived at
both the secular decline thesis and an economic explanation of it independently. John Toye and
Richard Toye, interview with Singer, 12 May 2000.
440 History of Political Economy 35:3 (2003)
Precursors of the Prebisch-Singer Thesis
By the time of World War II, the belief had already begun to gain ground
that agricultural countries had better reasons than industrial ones to be
pessimistic about their economic prospects. The experience of the inter-
war years had appeared to demonstrate this. As the Swedish economist
Gustav Cassel noted in a League of Nations study in 1927, “From 1913,
a very serious dislocation of relative prices has taken place in the ex-
change of goods between Europe and the colonial world” (Love 1991,
2). The world crisis of 1929 drew further attention to such questions,
particularly in Latin America.As SanfordA. Mosk noted in 1944, when
reviewing trends in the continent’s economic thought, “The relatively un-
favourable price position for raw materials and foodstuffs that prevailed
in the interwar period, and especially during the depression of the 1930s,
profoundly affected the outlook of Latin Americans” (Whitaker 1945,
143). This perception had already led to the claim, which had become
increasingly commonplace in the region, that primary product exporters
were at a disadvantage in international trade, compared to exporters of
industrial products (see, for example, Simonsen 1939, 15).
Primary-commodity exporting countries like Brazil and Argentina
were starting to see their future economic security in terms of promot-
ing industrialization. The war years intensified such resolve, and also
raised confidence that an industrialization drive, particularly if organized
by the government, could succeed.3Charles Kindleberger (1943b, 349)
bolstered this conviction, by suggesting as early as 1943 that industri-
alization was the path of the future, invoking Engel’s law of demand
against the classical orthodoxy on the terms of trade. In that year he
wrote that “inexorably...theterms of trade move against agricultural
and raw material countries as the world’s standard of living increases
(except in time of war) and as Engel’s law of consumption operates.” It
is well established that another of Kindleberger’s articles (1943a), with a
similar pro-industrialization message, based both on the differing elas-
ticities of demand for primary and manufactured products and on the
special “institutional organisation of production in industry,” was read
by Prebisch.4
3. Central European thinkers had advocated state-led industrialization in the 1920s on the
grounds of unequal exchange between groups of nations at the center and the periphery of the
world economy. This idea, particularly as used by Werner Sombart, was introduced to Latin
America by Ernest Friedrich Wagermann in his Evolución y ritmo de la economía mundial
(1933). See FitzGerald 1994, 94–95.
4. See the discussion in Love 1994, 421, 421 n. 84.
Toye and Toye / The Prebisch-Singer Thesis 441
Although many North American neoclassical trade theorists reacted
very critically to the Prebisch-Singer thesis in the 1950s, others among
them were not immune from this emerging current of opinion. Paul Sam-
uelson has been frequently caricatured as the high priest of an over-ab-
stract and ideological neoclassical orthodoxy.5Yet, remarkably, in 1948
he himself asserted the tendency of the terms of trade of primary pro-
ducers to decline. He wrote at the end of his famous article on the equal-
ization of factor prices as a result of trade: “[Now] the terms of trade are
abnormally favourable to agricultural production. Without venturing on
rash prophecy, one can venture scepticism that this abnormal trend of
the terms of trade, counter to historical drift, will continue” (Samuel-
son 1948, 183–84; emphasis added). Up to this time, however, anticipa-
tion of the Prebisch-Singer thesis remained in the category of remarks
en passant, or obiter dicta made during the course of the demonstration
of other, quite distinct, propositions. It is nevertheless surprising that one
of these stray anticipatory remarks was from the pen of Samuelson, the
economist who was later set up as the archenemy of structuralism, and
that this fact has been hitherto generally overlooked.
Prebisch’s Intellectual Formation
Prebisch was born in Argentina in 1901. By the time he entered the UN,
he had had a long career of public service in his home country, culminat-
ing in his tenure as head of its central bank. He was forced to resign this
post in 1943, as a result of a Perónist stratagem, and went on to spend
five years at the University of Buenos Aires, struggling as an isolated
intellectual to write a book, titled Money and the Rhythm of Economic
Activity, that was never completed or published.6He later recalled being
offered the post of executive secretary to ECLA just after it was estab-
lished in February 1948 (Love 1994, 414; Magariños 1991, 128). He said
that at this time he “emphatically refused.” His motives were that he did
not want to give up his university post, and also that he doubted that
an international organization would take seriously the point of view of
underdeveloped countries (Magariños 1991, 128–29). Gustavo Martínez
Cabañas, a Mexican, was appointed executive secretary instead, but he
5. See, for example, Cardoso [1980] 1984, 23, and Furtado 1987, 101. This view is also to
be found in Love 1980, 63, which was endorsed by Kay (1989, 5) and Love (1994, 422).
6. For an exposition of Prebisch’s outline of this planned book, drafted in October–
December 1943, see Dosman 2001.
442 History of Political Economy 35:3 (2003)
only took up his post in January 1949. Thus, Eugenio Castillo, a Cuban,
who was his deputy, was in effect running the commission before this
date.
Later in 1948, the Argentine government barred Prebisch from teach-
ing (Ferrer 1990). He definitively resigned his university post in Novem-
ber and began to consider working outside the country.7The managing
director of the International Monetary Fund (IMF), Camille Gutt, and
his deputy Edward M. Bernstein visited BuenosAires in November 1948
and offered Prebisch a senior post in the Fund. This followed up an offer
of a short-term assignment in Washington, made in the previous Janu-
ary. When in late December 1948 Gutt cabled that the terms of the offer
would have to be changed, Prebisch replied that he was “quite willing to
join the Fund on the basis proposed.” However, the executive board of
the IMF decided not to proceed. The U.S. government had reversed its
favorable position on the appointment, because it wanted to improve re-
lations with Perón, and Brazil also voted against Prebisch.8On 11 March
1949 Maurice L. Parsons, director of operations at the IMF, wrote an
extremely apologetic letter to Prebisch, expressing personal regret at the
Fund’s failure to secure his services.9
Simultaneously with Gutt’s approach, Eugenio Castillo of ECLA
asked Prebisch for help in preparing an economic survey of LatinAmer-
ica. Prebisch was at first unwilling to commit himself, because he much
preferred to accept the IMF offer.10 It was only on 10 January 1949
that he agreed to work for ECLA as a consultant on a short-term con-
tract.11 His allotted tasks were to coordinate and pull together in final
form the commission’s planned Economic Survey of Latin America,
which was to be presented to the ECLA conference in Havana in May
1949. Up to this point, he had made only one public contribution on the
terms of trade issue that needs to be noted. The depression of the 1930s
7. Letter from R. Prebisch to E. Castillo, 23 November 1948, Prebisch Papers (in the pos-
session of Sra. Prebisch).
8. This episode is examined in more detail in Dosman 2001.
9. Letter from J. Marquez to R. Prebisch, 23 January 1948; letter from R. Prebisch to E.
Bernstein, 17 December 1948; letter from M. L. Parsons to Prebisch, 11 March 1949, Prebisch
Papers. James Boughton, currently in-house historian of the IMF, passed on to us this comment
from Jacques Polak: “I have absolutely no recollection of any such offer, but it would not be
wholly out of character for Bernstein to make it without discussing it with me (a mere division
chief at the time).”
10. Letter from R. Prebisch to E. Castillo, 23 November 1948; letter from Francisco Coire
to R. Prebisch, 24 December 1948, Prebisch Papers.
11. Letter from R. Prebisch to E. Castillo, 10 January 1949, Prebisch Papers.
Toye and Toye / The Prebisch-Singer Thesis 443
had created exceptionally unfavorable terms of trade for exporters who
relied on agricultural products and raw materials. By 1932, export prices
had fallen in Argentina to 37 percent of what they had been in 1928,
and her net barter terms of trade were down to 68 from a 1928 value
of 100, indicating a less precipitous fall of import prices (Thorp 1998,
105, table 4.1). Prebisch had documented this when was he was direc-
tor of research at the National Bank of Argentina. He published an arti-
cle in 1934 arguing that “it is a well-known fact that agricultural prices
have fallen more profoundly than those of manufactured articles,” and
that Argentina had to export 73 percent more than before the depression
to obtain the same quantity of manufactured imports (Prebisch [1934]
1991, 341). However, Prebisch was merely noting a fact, and did not
provide any theoretical analysis of it (Magariños 1991, 63–64). He saw
it as a feature of depression economics, that is, as a short-run cyclical
problem. He believed that the remedy was to be found in expansionist
economic policies, not, as the Prebisch-Singer thesis would later imply,
in major changes in the structure of the international economy.12
Prebisch attended the World Economic Conference in 1933, and The
Means to Prosperity, which Keynes published at this time, powerfully
affected his hitherto orthodox thinking (Magariños 1991, 100).13 His
views must also be seen in the context of the emerging current of opin-
ion in Latin America, previously mentioned, that asserted that primary
producing nations were at a particular disadvantage in relation to indus-
trialized countries. Before 1949, Prebisch played only a marginal role in
promoting this discourse. During his years at the University of Buenos
Aires, the focus of Prebisch’s research was on the international business
cycle, in the tradition of Wesley Mitchell (1927) and Joseph Schumpeter
(1939), and on the prospects for the use of Keynesian countercyclical
policies.14 It was in this context that he first used the terms cyclical cen-
ter and periphery.15 In his introductory class in 1945, he referred to the
12. He believed that “the normal oscillation in the economic life of an agrarian country
can be supported by the monetary system, if that system is in good condition” (Prebisch 1991,
2:566).
13. In March 1933, the Times of London published four articles by Keynes, which, ex-
panded by an introductory and concluding chapter, were published in pamphlet form as The
Means to Prosperity. The slightly enlargedAmerican edition is in Keynes 1973–89, 9:335–66.
14. “Prebisch’s main concern was the international propagation of the business cycle,” ac-
cording to Furtado (Kay 1989, 9).
15. In 1980, Prebisch did not recollect Sombart’s earlier use of these terms, according to
Love (1980, 63). However, FitzGerald (1994, 95) thinks that it is “most unlikely” that Prebisch
“was unaware of the origins of the contemporary center-periphery concept.”
444 History of Political Economy 35:3 (2003)
maintenance of full employment as the supreme responsibility of the
United States, as “the monetary and economic center of the world,” al-
though he did not use the term periphery in this particular lecture (Pre-
bisch 1945, 529). In an intervention in a meeting of American central
bankers in 1946, he used both terms together, to argue that the respon-
sibilities of the “cyclical center” had been too much emphasized, and
that the “countries of the periphery” themselves must resolve disequilib-
ria with internal causes (Prebisch 1946, 163–64).16 In 1948, he reverted
to his previous (1945) theme that the main responsibility for carrying
out a countercyclical investment policy rested with the cyclical centers
(Prebisch 1948b, 161).
A glimpse of Prebisch’s overall research program at this time can be
found in a letter to Eugenio Gudin.
I believe that the cycle is the typical form of growth of the capital-
ist economy and that this is subject to certain laws of motion, very
distinct from the laws of equilibrium. In these laws of motion the dis-
parity between the period of the productive process and the period of
the circulation of the incomes therefrom holds a fundamental impor-
tance. So I have tried to introduce systematically the concept of time
into economic theory and also that of space, which in the ultimate in-
stance resolves itself into a problem of time. It is precisely the concept
of space that has led me to study the movement in the center and the
periphery, not with the aim of establishing formal distinctions but to
point out transcendent functional differences.17
He also believed that, more generally, economic theory required “reno-
vation” in order to bring it nearer to reality.18
Love (1980, 57) has stated, on the evidence of a transcript of lec-
tures given by Prebisch in 1948, that “Prebisch implicitly already had
his opinions about the direction of Latin America’s long-range terms of
trade, since he had argued in the classroom in 1948 that the benefits
of technological progress were absorbed by the center.” Furthermore,
16. Prebisch did not define cyclical center, but we infer from the context that he meant
those economies from which major depressions that can affect the rest of the world originate.
In 1945, this could have only referred to the economy of the United States.
17. Letter from R. Prebisch to Eugenio Gudin, 20 December 1948, original in Spanish,
Prebisch Papers.
18. Letter from R. Prebisch to E. Castillo, 23 November 1948, original in Spanish, Prebisch
Papers.
Toye and Toye / The Prebisch-Singer Thesis 445
“Prebisch had formulated the elements of his thesis before the appear-
ance, in 1949, of the empirical base on which the thesis rested in its
first published form—the UN study, Relative Prices” (Love 1980, 65;
for more on both statements, see Love 1994, 417–18). Part of Prebisch’s
1948 lectures did discuss the case where one country (call it A) experi-
ences technical progress in some of its economic activities (manufactur-
ing) and then trades with another country (B) that does not experience
technical progress. Prebisch argued that country A can retain for itself
the fruit of technical progress, and specifically asserted that, historically,
both Great Britain and the United States had done so. His argument about
the conditions under which this happened was, however, confused, and
in the course of it he did not actually use the terms center and periphery
(Prebisch 1948a, 87–98). The argument is not sufficiently well specified
to permit the claim that it could have only one logical implication for
the net barter terms of trade of country B. In the context, one would cer-
tainly not have been surprised if Prebisch had asserted that country B’s
terms of trade would continuously decline, but the crucial point is that
he did not do so.
Before he became aware of the UN data, Prebisch never explicitly
stated the thesis that Latin America’s terms of trade had been subject
to long-term decline, as opposed to the sharp short-term decline that he
noted in 1934.19 Nor does Love claim that he did; he claims only that “he
implicitly already had his opinions.” Prebisch had clearly by 1948 ar-
rived at the idea that the fruits of technical progress could be distributed
unequally, an idea that he would later refine and integrate into his ex-
planation of the secular decline phenomenon. But his chief concern at
the time was still the study of the business cycle, as the typical form of
growth of the capitalist economy; of the secular decline phenomenon
itself, he remained unaware. The following year, however, he publicly
stated that secular decline was taking place. To understand how this hap-
pened, why priority of discovery was subsequently claimed for Prebisch
and how this priority claim then affected the meaning that contempo-
raries attached to the thesis, it is necessary to trace a complex sequence
of political events and administrative decisions. We start with Singer’s
arrival at the United Nations.
19. Prebisch did, however, use the phrase “persistent fall in the international prices of our
exports” in the 1943 outline of his unpublished book, according to Dosman 2001.
446 History of Political Economy 35:3 (2003)
Singer Starts Work on the Terms of Trade
Singer, the younger of the two, came to the UN via a rather different
route than Prebisch, although his story also included escape from perse-
cution. He was born in the German Rhineland in 1910, and studied under
Joseph Schumpeter in his Bonn period. After the Nazi seizure of power
in 1933, Schumpeter used his contacts with Keynes to place Singer in
a scholarship to undertake a Ph.D., on secular trends in land values, at
Cambridge. In 1940, the Nazis put Singer on their “Special Wanted List
GB,” to be arrested by the Gestapo in the event of a German invasion
of Britain (Schellenberg 2000, 242). Singer’s doctoral work led him to
wartime employment in the Ministry of Town and Country Planning, but
he intended to return to academia after the war. However, David Owen,
the head of the UN DEA, invited Singer to join the United Nations. In
1946, Owen sent a formal request to this effect to Glasgow University,
Singer’s employer, and Singer quite reluctantly agreed to go on a two-
year leave of absence, one that ultimately turned into a twenty-two-year
period of UN service.20
He arrived in New York in April 1947, knowing there only Michal
Kalecki and Sidney Dell, while Owen was absent at the international
trade negotiations in Geneva. He found himself unrestricted in his choice
of research subject and immediately began to work on trade problems,
although his previous economic background had not been specifically
related to trade issues. Singer later recalled:
A strong influence among the early colleagues in the United Nations
was that of Folke Hilgerdt, the Swedish economist who had already
shaped the League of Nations publications on the Network of World
Trade. Working with him was Carl Major Wright, a Danish econo-
mist who was particularly interested in the relationship of primary
commodity prices to trade cycles and economic growth in industrial
countries. Two other staff members in the trade section were Walter
Chudson (United States) and Percy Judd (Australia), the latter being
very expert in the economics and details of commodity agreements.
Discussions with these four must have drawn my attention quickly to
problems of terms of trade. (Singer 1984, 280)
20. “I was not very happy....Iwent—though I really didn’t want to go. For me that was a
step down and I was quite looking forward to settling down in Glasgow.” Hans Singer interview
with Richard Jolly, 13 October 1995.
Toye and Toye / The Prebisch-Singer Thesis 447
The official stimulus for this work was the report of first session of the
UN Sub-Commission on Economic Development. The members of the
subcommission were elected on 5 June 1947. Considering problems of
economic development in underdeveloped countries, the report of its
first session contained the following comment:
The recent rise in the prices of capital goods and transport services
has made the task of economic development particularly difficult in
the case of the under-developed and the least developed countries. The
Sub-Commission therefore considers it important that a careful study
be made of the prices of capital goods and of the relative trends of
such prices and of prices of primary products, so that it may be in a
position to make appropriate recommendations concerning the prob-
lem. (United Nations 1949b, 1)
As a result, the UN secretariat began to study the terms of trade. The task
was to address a short-term problem. The original objective was not to
discover the historical drift of the terms of trade, or what had happened
over the long run. The problem was that, during the war, a number of un-
derdeveloped countries had run export surpluses that they subsequently
wished to use to import capital goods for development. In the interval,
the prices of capital goods had risen, so the export surpluses were worth
less in terms of imports than they had been when they were earned. This
provoked the question of whether underdeveloped countries’ terms of
trade could be expected to continue to deteriorate in this way, and the
implication of this for their economic development. This was the offi-
cial purpose of the research on which Singer embarked.
Singer worked under the general guidance of Folke Hilgerdt, who,
as director of the UN Statistical Office, provided a key link between its
work and the statistical work of the former League of Nations on trade.
Hilgerdt was the principal author of a series of studies on commerce
and commercial policy, which the Economic, Financial, and Transit De-
partment of the league issued as part of its program of studies on post-
war problems. The final volume, Industrialization and Foreign Trade
(1945), included an appendix on the statistics of international trade be-
tween 1871 and 1938. Appendix tables 7 and 8, when read in conjunc-
tion, show that between those dates the price index of manufactured
articles fell significantly less than that of primary products. However,
nothing was made of this in the summary of findings of the report. The
statistical base for this study was available for Singer’s research (League
448 History of Political Economy 35:3 (2003)
of Nations 1945, 154–67, 116–21). Moreover, Hilgerdt expressed puz-
zlement to Singer over the behavior of the British terms of trade data
(Singer 1991, 9).
In his Ph.D. dissertation, Singer had studied problems of the very
long run. Unlike Hilgerdt and Wright, he was not interested in cyclical
effects on the terms of trade produced by booms and slumps in indus-
trial countries. He, being more influenced by Gunnar Myrdal, focused
on structural differences between industrial and nonindustrial countries,
and their long-term effect on the evolution of the terms of trade between
them. His overarching concern was that of distributive justice. His ques-
tion was not whether gains from trade existed, which he did not doubt,
but the “fairness” of the distribution of those gains between the coun-
tries that traded. If there were power differences between countries—
disparities in market power or in technological power—did trade, and
changes in the terms on which it was conducted, become a mechanism
of “un-equalizing” growth between countries globally? His interest in
the commodity terms of trade was thus a derivative of the larger ques-
tion of worldwide un-equalizing growth. That question itself was framed
by the historical context of the process of decolonization, as in the tran-
sition of India to independence in 1947. Were the colonial powers, he
wondered, willing to relinquish control of their colonies only because
the international economic system would now spontaneously generate
the same world division of labor that had previously been enforced mil-
itarily and politically?
The results of Singer’s research were presented in a UN document
titled Post-war Price Relations between Under-developed and Indus-
trialized Countries that appeared on 23 February 1949 (United Nations
1949b).21 This was an advance version of the terms of trade study, sub-
ject to final checking of the data, which was made available to the Sub-
Commission on Economic Development. The document was retitled Rel-
ative Prices of Exports and Imports of Under-developed Countries for
its general circulation in late 1949, with the subtitle A Study of Post-war
Terms of Trade between Under-developed and Industrialized Countries.
It was a remarkable piece of work. It included an attempt to see what
historical statistics indicated about the long-term trend in the agricul-
ture versus manufactures terms of trade, although its origin lay in devel-
oping countries’ concern with future relative prices as industrialization
21. It should be noted that this is available in the UNOG Library in Geneva, but not in the
Dag Hammarskjöld Library at the United Nations in New York.
Toye and Toye / The Prebisch-Singer Thesis 449
Table 1 Ratios of U.K. Imports to Exports, 1876–1948 (1938 =100)
Period (or year) Current Year Weights Board of Trade Index
1876–80 163
1881–85 167
1886–90 157
1891–95 147
1896–1900 142
1901–5 138
1906–10 140
1913 137 143
1921 93 101
1933 98 96
1938 100 100
1946 108
1948 117
drives gathered pace. It showed that the terms of trade of underdevel-
oped countries had improved between 1938 and 1946–48. This recent
improvement was, however, placed in a much longer historical perspec-
tive, showing that between 1876 and 1948 they had seriously deterio-
rated. The historical section contained the report’s most dramatic finding.
It was that “from the latter part of the nineteenth century to the eve of the
Second World War, a period of well over half a century, there was a sec-
ular downward trend in the prices of primary goods relative to the prices
of manufactured goods” (United Nations 1949c, 7). Singer recalled that
he and Hilgerdt together spotted this trend in the data.22 The cumulative
effect of secular decline was calculated to be substantial.
By 1938, the relative prices of primary goods had deteriorated by
about 50 points, or one-third, since (the 1870s) and by about 40 points,
somewhat less than 30 per cent, since 1913. (23)
The statistical evidence for this downward trend was given in table 5 of
the report, of which a simplified version is presented in our table 1.
What was the significance of this secular decline? The report was
careful not to suggest that if a country’s terms of trade improved, its
welfare necessarily increased. It might or might not, depending on the
circumstances in which the rise in export prices takes place. If the price
rise was a result of a failure of supply, it might not leave the exporting
22. John Toye and Richard Toye, interview with Hans Singer, 12 May 2000.
450 History of Political Economy 35:3 (2003)
country better off. Nevertheless, in general, an improvement in terms
of trade would increase the availability of resources for development.
A secular decline for underdeveloped countries meant a loss of capac-
ity to absorb foreign financing for development, and thus to respond to
the “added incentive towards industrialization” (United Nations 1949c,
16, 127). A further, far more controversial, implication was also drawn,
that “the under-developed helped to maintain...arising standard of liv-
ing in the industrialized countries, without receiving, in the price of their
own products, a corresponding equivalent contribution towards their own
standards of living” (126). This carried a clear message of historical in-
justice, and this message was, as we shall see, very shortly to be rejected
by the subcommission.23
Singer had already announced this message in a seminar that he gave
to the graduate faculty of the New School of Social Research, NewYork,
on 23 December 1948. There he said that “Marxist analysis, in which
rising standards of living for given groups and sections are somehow
held to be compatible with general deterioration and impoverishment, is
much truer for the international scene than it is for the domestic.” He at-
tributed the growing inequality in the distribution of world income to the
change in price relations between primary materials and manufactured
goods (Singer 1949, 2–3).24
How Prebisch Made Use of the Singer Study
The creation of ECLA had resulted from “an act of audacity” by Hernán
Santa Cruz, the Chilean representative on the UN Economic and Social
Council (ECOSOC) (Magariños 1991, 136). He felt that Latin Amer-
ica was being unjustly ignored by the great powers, and conceived the
idea of a UN commission to deal with the region’s economic and social
problems, on the model of the UN Economic Commission for Europe.
Receiving no orders to the contrary from his government, he took the
initiative and submitted his proposal as an item for the UN ECOSOC
agenda (Santa Cruz 1995). On 1 August 1947, he formally introduced
his resolution to a meeting of ECOSOC. Although the United States and
other industrial countries much disliked the idea, their opposition was
23. “I thought if you look at foreign trade from the point of view of the poor countries,
exporters of primary products, what does it look like?And it appears an unequal system that is
weighted against them.” Hans Singer interview with Richard Jolly, 11 October 1995.
24. We are most grateful to John Shaw for drawing this document to our attention.
Toye and Toye / The Prebisch-Singer Thesis 451
not sufficiently determined, and in early 1948 ECLA duly came into
being.25
One of the justifications for establishing ECLA was to provide better
information about economic conditions in the region as a whole.26 Origi-
nally, it was assumed that this could be achieved merely by collating sta-
tistics submitted by the individual Latin American governments. When
most governments failed to provide the required figures, the commission
realized that it would have to collect them itself, for the first Economic
Survey of Latin America, which was due to be presented to the ECLA
conference in Havana in May 1949.Additionally, the first ECLA confer-
ence, in June 1948, had passed a resolution asking for the preparation of
“a study of the movements of import and export prices, the determining
factors of such movements, and the consequences thereof on the balance
of payments.”27 By the autumn of 1948 it became clear that these tasks
were beyond the unaided capacity of the ECLA office in Santiago.28 The
weak statistical abilities of the fledgling commission thus form an im-
portant background factor in the preparations for Havana.
Accordingly, Louis Shapiro, a statistician in the Department of Eco-
nomic Affairs, was sent from the UN office in NewYork on deputation to
Santiago for three months between December 1948 and March 1949, in
order to organize these tasks. It was during this mission that he received
from Hans Singer the provisional draft of his study on the terms of trade.
From Santiago, Shapiro wrote Singer a letter of acknowledgment dated
5 January 1949:
Thank you most kindly for your letter of 17th December 1948, and
for the enclosed provisional draft of the General Part of your study
25. On 25 February 1948, a motion establishing the new body was passed, with no votes
against, and four abstentions (Byelorussia, Canada, the United States, and the USSR). As Leroy
Stinebower, who was present on behalf of the United States, later recalled, “The forces of glob-
alism were being overwhelmed—or at least over shouted—at that moment by a lot of region-
alism stuff....eveninmyworst dreams I didn’t think that regionalism would go as far as it’s
gone in this world.” Leroy Stinebower Oral History, 9 June 1974, pp. 65–66, Harry S. Truman
Library, Independence, Missouri.
26. “The Economic Commission for Latin America...shall arrange for such surveys, in-
vestigations and studies to be made of the economic and technical problems (of the region) as
it may deem proper and participate in the same.” Extract from the Official Records of the UN
Economic and Social Council, second year, fifth session, proposal addressed to the secretary-
general by the delegation of Chile, document E/468, 12 July 1947.
27. Resolution of the first session of ECLA, 24 June 1948, UN document no. E/CN.12/71.
28. Celso Furtado (1987, 54–56) tells the story of how he, as a new recruit to ECLA, was
urged by the executive secretary, Martínez Cabañas, to bring to Santiago as many documents
on Brazil as possible for the purpose of strengthening the survey.
452 History of Political Economy 35:3 (2003)
on the Terms of Trade. I have also received via the pouch drafts of
the country sections of Terms of Trade for which many thanks. I have
read quite carefully the General Part and find it most admirable.Your
note on the methodology and the statistical “caveats” are especially
noteworthy. I have passed this on to Mr Castillo who also agrees that
this is an excellent piece of work. ECLA plans to include a substan-
tial statistical section on the terms of trade in the forthcoming Survey
of Latin America and will, with your permission and clearance, rely
heavily on your data.29
The International Monetary Fund (IMF) had also promised to send
ECLA a study on the terms of trade. Shapiro reported to Singer that
he had not been encouraged by its progress and content when he had
visited Washington, but indicated that ECLA would study carefully the
IMF work in conjunction with Singer’s own, and that Castillo was “in
complete agreement with this procedure.” Shapiro also asked to be sent
revised drafts of the “general part” and of the Latin American country
sections as they were completed.30
A version of the IMF study was available by March 1949, although
it was not published as an IMF staff paper until the following year. It
dealt mainly with the period 1938 to 1946 and included no data at all
from before 1925. It found that the terms of trade of Latin America as
a whole had improved between 1938 and 1946. The IMF study made
no comment, however, on the UN study’s thesis of secular decline since
the 1870s (Ahumada and Nataf 1950). Nevertheless, the IMF study left a
larger imprint on the Economic Survey of Latin America than did Singer’s
study. In the event, the survey’s one mention of secular decline in the
terms of trade of primary producers was extremely brief (United Na-
tions 1949a, xix, 216–20), and almost certainly inserted after Prebisch
had completed and circulated the document that became The Economic
Development of Latin America (Prebisch 1950). Thus Singer’s work did
not create any real impact in ECLA until it had reached the hands of
Prebisch himself.
How did it get there? The second route by which the Singer study was
transmitted to Latin America was via Martínez Cabañas. Before arriving
29. Letter from L. A. Shapiro to H. W. Singer, 5 January 1949, UNArchive, NewYork. As
the reader will recall, Eugenio Castillo (“Mr Castillo”) was the deputy executive secretary of
ECLA.
30. Letter from L. A. Shapiro to H. W. Singer, 5 January 1949, UNArchive, NewYork.
Toye and Toye / The Prebisch-Singer Thesis 453
at ECLA, Prebisch had gone to Mexico to deliver lectures at the Univer-
sidad NacionalAutónoma de México, arriving in mid-February.31 In late
February or early March, Francisco Coire, head of the Latin American
section of the DEA in NewYork, sent him both the Singer study and the
IMF study.32 On 5 March, Martínez Cabañas wrote from New York to
Prebisch in Mexico drawing both these works to his attention.
Our friend Coire has informed me that he has already sent you two
studies on questions relating to foreign trade: one drawn up by Sr.
Singer which is to be found under the number E/CN.1/Sub3/W5 with
the title Post-war Price Relations in the Trade between Undeveloped
and Industrialised Countries....theother study is from the Interna-
tional Fund and refers more concretely to the theme of Foreign Trade.
He reported that the conclusions of the Singer study had been “much
debated.” He pointed out to Prebisch that both studies had a bearing on
a problem—the terms of trade—which was “one of the most important
of those that will be treated in the general study that we are going to
present at the Havana Conference.” He repeated to Prebisch that he (i.e.,
Prebisch) would have the final responsibility for drawing up that report.33
Thus it was the Martínez Cabañas visit to NewYork, rather than the early
version seen and favorably commented upon by Castillo, that made the
effective link from Singer to Prebisch.
Prebisch must have arrived in Santiago soon after 9 March, when he
had received a telegram from Castillo asking him to come immediately.34
This was shortly after Shapiro’s return to New York. It seems that before
long Prebisch turned his mind to the question of the terms of trade, as
one would have expected, given the strong urgings of Martínez Cabañas.
Then, on 1 April 1949, Prebisch sent a request, through Castillo, to Sha-
piro in New York for three types of additional data—additional, that is,
to the data contained in United Nations 1949b, which it seems clear that
Prebisch had now read.35 Castillo did not explain the reasons behind Pre-
bisch’s request, but the wish for data starting in 1873, the year the British
31. Letter from E. Castillo to R. Prebisch, 16 February 1949, Prebisch Papers.
32. Letter from G. Martínez Cabañas to R. Prebisch, 5 March 1949, original in Spanish; F.
Coire to R. Prebisch, 8 April 1949, Prebisch Papers.
33. Letter from G. Martínez Cabañas to R. Prebisch, 5 March 1949, original in Spanish,
Prebisch Papers.
34. Telegram from E. Castillo to R. Prebisch, 9 March 1949, Prebisch Papers.
35. “1. Wholesale prices indices of the United Kingdom from 1873 to 1947, divided, if
possible, into indices of manufactured products and raw materials. 2. Price indices for exports
454 History of Political Economy 35:3 (2003)
Great Depression began, suggests an interest in the respective experi-
ence of the United Kingdom and the United States as “cyclical centers,”
a problem he returned to in the Economic Survey of Latin America in
the following year (United Nations 1951).36 Prebisch was also anxious
to see an early draft of Kalecki’s study on inflation and Coire’s draft of
part 2 of the survey.37
Celso Furtado, an ECLA staff member, later recalled how Prebisch
initially worked very much on his own, and then a month after his ar-
rival (presumably in mid-April), circulated within ECLA a first draft of
an introduction to the Economic Survey. Furtado described this draft as
a digest (Fr. mouture) of the papers that Prebisch had brought with him
from Buenos Aires. The subjects covered were disequilibria in the bal-
ance of payments, the declining U.S. import coefficient, capital controls,
low saving leading to domestic inflation, and the limits of industrializa-
tion. Thus, the first draft of Prebisch’s introduction did not cover the
terms of trade, according to Furtado’s account of it.38 He did employ
his “center-periphery” terminology and acknowledged the importance
of industrialization. It seemed to Furtado (1987, 65) that “this text con-
tained extremely interesting ideas, but the author placed himself on the
defensive.”
Furtado (1987, 65–66) recalled as follows:
We had hardly started to discuss the document, when it was suddenly
discarded, without any explanation. Prebisch’s new text was not cir-
culated for discussion. I suppose that it was ready on the eve of the
Havana Conference, because it was sent to us typewritten, in its final
version, shortly before we left. It was a much longer text including ta-
bles and charts. The tone had changed, now it was a manifesto urging
Latin American countries to launch into industrialisation. One could
discover there a definite taste for a polished and polemical style.
This account suggests that mid-April 1949 was the decisive point in Pre-
bisch’s drafting process. At this point, however, Prebisch had not had any
and for imports for the United Kingdom and for the United States for the same period. 3. United
States national income from 1910 to 1929.”
36. Letter from E. Castillo to L. A. Shapiro, 1 April 1949, UN Archive, New York.
37. Letter from E. Castillo to F. Coire, 1 April 1949, UN Archive, NewYork.
38. This first draft does not appear to have survived among the manuscript papers and cor-
respondence currently in the possession of Sra. Prebisch, although these have not been fully
cataloged.
Toye and Toye / The Prebisch-Singer Thesis 455
reply to Castillo’s data requests of 1 April. The reply was not sent until
27 April, and it was the third channel of transmission of Singer’s study.
Shapiro scribbled the words “in Singer Paper” over Prebisch’s second
request, and this was the only one he could fulfill (apart from the U.S. na-
tional income figures for 1910–29).39 However, it is almost certain that
this data arrived too late to have any influence on Prebisch’s introduction.
There are no traces of it in the finished product, which, as far as the terms
of trade are concerned, contains only the U.K. part of the data in table 5
of United Nations 1949b, slightly reformulated. Prebisch spliced one of
the U.K. series down to 1913 with another of the U.K. series thereafter,
and put them on a base of 1876 −80 =100, instead of 1938 =100 as
originally (Spraos 1980, 107; 107 n. 2; 111–12, table 1 and the single-
asterisk note).
If it was not the arrival of additional information, then, what was it
that stimulated Prebisch to abandon his original draft? This remains un-
clear. Around this time Prebisch received Coire’s draft of part 2 of the
Economic Survey. Only one chapter of this was used in the finished prod-
uct (United Nations 1949a, 247 n. 1), and is now apparently lost, but
from Coire’s remarks to Prebisch it seems clear that it had a strong pro-
industrialization message and had a polemical tone in places.40 This, or
perhaps further reflection on the implications of the Singer study, may
have given Prebisch extra inspiration.
Prebisch dealt with the whole issue of secular decline extremely brief-
ly in his new text. The introduction that summarizes its argument does
not even mention the terms of trade. The subject is then handled in the
first three pages of chapter 2. Prebisch’s only comment on the terms of
trade statistics, albeit one that was to resonate through the subsequent
critical debate, was that “it is regrettable that the price indexes do not
reflect the differences in quality of finished products” (Prebisch 1950,
8). This short but crucial section powerfully reinforced his other main
arguments—that the international division of labor was an “out-dated
schema,” and that “industrialization is the only means by which the Lat-
in-American countries may fully obtain the advantages of technical prog-
ress” (Prebisch 1950, 1, 16).
39. Letter from E. Castillo to L. A. Shapiro, 1 April 1949, and Shapiro’s reply, 27 April
1949, UN Archive, New York.
40. Letter from F. Coire to R. Prebisch, 8 April 1949, Prebisch Papers.
456 History of Political Economy 35:3 (2003)
Anonymity of Authorship in the UN
As has been seen, it is not possible to sustain the claim that Prebisch
was the first to discover the phenomenon of the secular decline in the
terms of trade of primary producers. On the ground that Prebisch’s con-
tribution was complete by May 1949, while Singer’s paper was not pre-
sented to the American Economic Association meeting until December
1949, Love (1980, 58–59) concluded that “Prebisch clearly seems to
have reached his position earlier than Singer.” This was faulty reason-
ing, given Singer’s authorship of the UN study, something of which Love
does not seem to have been aware. John Spraos has commented on the
general lack of awareness of Singer’s authorship, and he correctly attrib-
uted it to the UN’s rule that authors of its publications remain anony-
mous.41
Anonymity is indeed the general fate of the authors of UN publica-
tions.42 Since the UN has employed many distinguished economists, it
has become a standard task for their biographers to try to disentangle
those sections of UN publications that they authored.43 What is odd in
the present case is not that Singer’s authorship should have been invis-
ible, but that Prebisch’s should have been visible. What was originally
intended to be the introduction to the ECLA survey for 1948 eventually
appeared under Prebisch’s name as The Economic Development of Latin
America and Its Principal Problems. Had the piece been retained as the
introduction to the survey, its author would, like Singer, have remained
anonymous. In the event, however, it appeared as a separate UN publica-
tion, but with the author personally identified. The UN rule of anonymity
was applied to Singer and Prebisch asymmetrically. This came about in
the following way.
The Sub-Commission on Economic Development held its third ses-
sion, from 21 March to 11 April 1949. It discussed Singer’s study, Post-
war Price Relations. It accepted, somewhat grudgingly, the statistical
41. “It is an interesting fact, perhaps not widely known because of the anonymity of UN
staff publications, that the principal author [of Post-war Price Relations and Relative Prices]
was Hans Singer who became the second twin in the ‘Prebisch-Singer thesis’through a subse-
quent signed article” (Spraos 1980, 107 n. 3).
42. Alexander Loveday, who directed the economic research of the League of Nations,
wrote critically of “the rigidity of their [the UN’s] anonymity rule,” which he thought ham-
pered the employment of temporary experts. See Loveday 1956, 296.
43. See, for example, the efforts of Jerzy Osiatynski to identify the UN writings of Michal
Kalecki, in Osiatynski 1990–97, 7:552–75.
Toye and Toye / The Prebisch-Singer Thesis 457
evidence, but rejected the lessons that had been drawn from it. Its report
said the following:
The Sub-Commission is constrained to point out that the study under
review contains certain conclusions in regard to the price relationship
between developed and under-developed countries which, in its opin-
ion, do not represent a correct picture of the actual position.As a result
of the discussion, the Sub-Commission agreed that while the docu-
ment contained an adequate study of relative price trends of primary
commodities and manufactured goods, it was necessary to broaden
the scope of the study into that of the terms of trade between under-
developed and industrialized countries, including prices and quanti-
ties traded, and in extending it, to cover the most recent movements in
these fields. (United Nations 1949d, 12)
The most controversial of the suggestions that Singer had made in inter-
preting his findings, was, of course, that underdeveloped countries were
helping to maintain a rising standard of living in industrialized countries
without receiving any equivalent compensation. This was potentially po-
litically explosive. While it appealed to the underdeveloped countries, it
appealed not at all to the developed. It seems plausible to suggest that the
subcommission used the (acknowledged) fact that the picture presented
by the study was in some ways incomplete as an excuse for disclaiming
its radical conclusions.
It is probable, given the slow rate of circulation of UN documents, that
Prebisch was not aware of the subcommission’s report when he wrote.
Be that as it may, in the final version of his introduction, he repeated
the implication to which it seems the subcommission had objected by
quoting the relevant passage from Singer’s study (Prebisch 1950, 10 n.
3). Worse, by using his terminology of center and periphery, he further
dramatized it:
The enormous benefits that derive from increased productivity have
not reached the periphery in a measure comparable to that obtained
by the peoples of the great industrial countries. Hence, the outstanding
differences between the standards of living of the masses of the former
and the latter and the manifest discrepancies between their respective
abilities to accumulate capital. (1)
When he presented this version in Havana, it received the acclaim of
the delegates of the Latin American countries (Magariños 1991, 130).
458 History of Political Economy 35:3 (2003)
However, what was music to the ears of the delegates of Latin Ameri-
can countries would have displeased the industrial countries, especially
the United States. This fact appears to have caused some consternation
among UN high officials in NewYork, who were anxious to distance the
UN from Prebisch’s introduction. Accordingly, after the Havana confer-
ence was over, it was submitted to the secretary-general as an “essay”
commissioned in the process of “fostering research.”44 It was then pro-
posed to the UN publications committee to break the rule that authors of
UN publications should not be identified by name. This course of action
was designed to ensure that Prebisch took “credit (and responsibility) for
the report...in order to emphasize that the views expressed...were
those of the author and not those of any UN organ.” The proposal was
presented “as an exceptional one, unlikely to recur but in the present cir-
cumstances very desirable.”45 Prebisch’s suspicion that no international
organization would feel comfortable with the viewpoint of the underde-
veloped countries was thus confirmed.
The UN’s tactic backfired. The Spanish original had been issued in
May 1949, but, as Furtado has noted, it was some time before both this,
and the English translation, were eventually published in NewYork by
the United Nations, being circulated, as he put it, “with the slowness
characteristic of official documents” (Furtado 1987, 80). Meanwhile,
however, a Portuguese translation of the Spanish original, undertaken
at Furtado’s own urging, was published in Brazil in September 1949. It
is at this point that the history of Prebisch’s enormous influence began,
spreading out from Brazil eventually to become worldwide.46 The pub-
lication of the English version of The Economic Development of Latin
America and Its Principal Problems merely strengthened that effect in
North America and Europe. In fact, this was how Singer now discov-
ered the existence of Prebisch. “I believe it was between presenting my
paper to the AEA in December 1949 and its publication in the summer
44. Gustavo Martínez Cabañas to Trygve Lie, 9 October 1949, letter of transmittal printed
in Prebisch 1950.
45. H. E. Caustin to A. D. K. Owen, 12 October 1949, UN Archive, DAG-17 box 33; see
also Magariños 1991, 129.
46. Its first incarnation was in Spanish, as UN document E/CN.12/89, dated 14 May 1949.
It appeared in the July–September 1949 issue of El trimestre economico. At the instance of
Furtado, it was published in Portuguese in the Revista Brasileira de economia in October 1949.
It was republished in Spanish in Santiago in April 1950, and again in English later in 1950 in
Lake Success, New York, as UN document no. E/CN.12/89/Rev.1. Its subsequent publication
history until 1986 is to be found in United Nations 1987, 52.
Toye and Toye / The Prebisch-Singer Thesis 459
of 1950 that I discovered that Raúl Prebisch, my colleague at the UN,
had developed very similar opinions and had also put the problem of
poor terms of trade for primary products into the centre of thinking of
the Economic Commission for Latin America” (Singer 1997, 141). The
main result of identifying the author was that the polished and polemical
Prebisch rapidly gained greater recognition in Europe and North Amer-
ica as a “UN economist” than did the more understated Singer, who had
published under his own name only in academic journals.
Prebisch’s Contribution: The Economic
Mechanics of Secular Decline
Given the evidence outlined above, it seems clear that, if the Prebisch-
Singer thesis is defined as the statement of the phenomenon of secular
decline in the terms of trade of primary products, if anyone can be said to
have anticipated Singer, it would be Kindleberger or Samuelson, rather
than Prebisch. After all, both of them, unlike Prebisch, made explicit
published remarks on the issue. Prebisch, however, made a contribution
distinct from that made by Singer. This was to advance a cyclical-cum-
structural mechanism to explain the decline, one more complicated than
the purely structural interpretation of Singer.
Unlike the 1948 lectures, which singled out restrictions on labor im-
migration from countries not experiencing technical progress (type B),
the new mechanism was based on institutional factors that permitted the
retention of productivity gains by labor in countries where there was
technical progress (type A). Prebisch (1950, 12) further argued, char-
acteristically, that “the existence of this phenomenon cannot be under-
stood, except in relation to trade cycles and the way in which they occur
in the centres and at the periphery, since the cycle is the characteris-
tic form of growth of the capitalist economy, and increased productiv-
ity is one of the main factors of that growth.” He suggested that, even
though in the boom primary product prices typically rise faster than in-
dustrial prices, deterioration in the commodity terms of trade of the pe-
riphery is nevertheless possible, if, in the slump, primary commodity
prices decline steeply enough, compared with industrial prices. The ex-
planation offered of why primary product prices declined severely in the
slump compared with industrial prices was “the well-known resistance
to a lowering of wages” at the center. By contrast, “the characteristic
lack of organization among the workers employed in primary production
460 History of Political Economy 35:3 (2003)
prevents them from obtaining wage increases comparable to those of the
industrial countries, and from maintaining the increases to the same ex-
tent” (13).47
However, instead of leaving matters there, Prebisch also made an-
other argument that seemed to undermine his first explanation. He iden-
tified industrial production and primary production with groups of coun-
tries, described as “center” and “periphery.” He then argued that the
differing strength of organized labor at the center and the periphery was
not the crux of the matter, because even if workers at the periphery were
able to resist wage decreases as strongly as industrial workers were, ad-
justment would take place by another process. The high prices of pri-
mary products would force a contraction of industrial production, which
in turn would cut the demand for primary products. Recalling the experi-
ence of the Great Depression of the 1930s, Prebisch commented that “the
forced readjustment of costs of primary production during the world cri-
sis illustrates the intensity that this movement can attain” (13–14). This
was the germ of the idea, later taken up by dependency and world sys-
tem theorists, that the “center” was able to drain resources away from
countries on the “periphery,” regardless of their respective states of la-
bor organization.
As has been seen, Prebisch had arrived at his explanation in April and
May 1949. Prebisch and Singer probably arrived at their respective ex-
planations of primary commodity terms of trade decline independently,
although again Singer was first. Singer had presented his explanation, in
embryonic form, in a paper originally presented in December 1948 and
published in March 1949, then more fully in another paper to the Ameri-
can Economic Association conference in December 1949.48 This further
paper explored factors that had “reduced the benefits to under-developed
countries of foreign trade-cum-investment based on export specializa-
tion in food and raw materials.” The first of these was that the secondary
and cumulative effects of foreign export enclave investment were felt in
the investing country, not the country where the investment was made.
The second was that countries were “diverted” into types of economic
47. The dislocation to relative prices caused by monopolistic tendencies in the labor and
manufactures markets of Europe had been the theme of Gustav Cassels’s 1927 League of Na-
tions study Recent Monopolistic Tendencies in Industry and Trade (Love 1991, 2).
48. The basic idea of a structural difference between countries where increased efficiency
of production leads to higher incomes and those where it leads to falling product prices was
already in Singer 1949, 2–3.
Toye and Toye / The Prebisch-Singer Thesis 461
activity that offered less scope for technical progress and internal and
external economies. The third factor, “perhaps of even greater impor-
tance,” was the movement of the terms of trade. Singer, while conceding
that the statistics in Relative Prices (United Nations 1949c) were open
to doubt and objection in detail, regarded the general story that they told
as “unmistakable” (Singer [1950] 1975, 48).
Singer’s proposed mechanism of secular decline was less complicated
and based less on business cycles than Prebisch’s. It was an asymmetric
process whereby (1) the gains from technical progress in manufacturing
are distributed to the producers in the form of higher incomes, while (2)
the smaller gains from technical progress in primary commodity produc-
tion are distributed to the consumers in the form of lower prices. On this
basis, the industrialized countries have the best of both worlds, as pro-
ducers of manufactures and as consumers of primary products. The un-
derdeveloped countries have the worst of both worlds, as consumers of
manufactures and as producers of primary products. Thus the benefits of
foreign trade are shared unequally, and traditional foreign investment in
plantations and mines did, after all, “form part of a system of ‘economic
imperialism’ and of ‘exploitation,’” albeit not in the classical Marxist or
Leninist sense (49–51).
Prebisch’s interpretation of the secular decline, although possessed
of its own ambiguities, gave an illusion of greater concreteness than
Singer’s. Instead of two sets of countries defined by the types of prod-
ucts that they exported and imported, Prebisch’s concept of center and
periphery seemed to have a spatial, even geographical, reality to it. His
introduction of economic cycles into the mechanism allowed the short
and medium term to be integrated with the long term, and countered
the static quality of the purely structural approach. In general, this more
complex schema opened the door to broader analyses of the economic
conjuncture and policy recommendations on the issues of immediate
concern to Latin American economists. Although fertile in these ways,
Prebisch’s interpretation itself was still very succinct, perhaps reflect-
ing the novelty of the secular decline thesis even to him, and it there-
fore remained obscure on a number of crucial questions (Cardoso [1980]
1984, 27, 29–30; Furtado 1987, 66–67). It was also quite different from
the theoretical model that Prebisch published in 1959, which resembled
the standard neoclassical model of a small open economy, but with the
exception of a small number of special assumptions (Prebisch 1959).49
49. See the commentary in Flanders 1964 and FitzGerald 2000, 61–69. As FitzGerald in-
dicates, there are two stages in Prebisch’s explanation of terms of trade decline, the first based
462 History of Political Economy 35:3 (2003)
The ambiguities notwithstanding, Prebisch’s “heresies,” boldly laid out
“a la Bernard Shaw” (as he put it), proved as appealing to the underde-
veloped countries of LatinAmerica as they were anathema to UN head-
quarters in NewYork (Magariños 1991, 129).
Summary and Concluding Reflections
The United Nations ECLA had been born against the wishes of the Unit-
ed States. In preparing for its first major conference it lacked the statis-
tical infrastructure to discharge one of its primary tasks, to survey the
common economic and technical problems of the region. This provided
the cue for statistical help from UN headquarters in New York. During
the preparations for the conference, Singer’s work was transmitted to
ECLA by three different channels between December 1948 and April
1949. An early version arrived in December 1948, but does not seem
to have been followed up by Eugenio Castillo. The second transfer was
via Martínez Cabañas and Coire to Prebisch before the latter arrived in
Santiago, and this was the critical route.At Martínez Cabañas’s prompt-
ing, Prebisch “latched onto the terms of trade idea,” in the words of
Victor Urquidi,50 and meshed it with his own framework of thought in
April/May 1949. The additional data he requested from New York (the
third route) arrived too late to be useful. The document that became The
Economic Development of Latin America and Its Principal Problems
thus contained—as far as the terms of trade was concerned—merely a
small part of the original UN study data, minimally reformulated.
Thus the conventional view, that Prebisch achieved priority over Sing-
er in stating the thesis of secularly declining terms of trade for primary
producers, is not based on an accurate chronology of ideas. The near si-
multaneity of the dates of the first English publications of the thesis by
Prebisch and Singer cannot be relied upon as a means of dating the two
men’s contributions. The key events did not take place in 1950 at all,
but in 1948 and 1949 in the run-up to and the aftermath of the ECLA
conference at Havana. Prebisch’s presentation to the Havana conference
was a resounding success, turning him into a champion of the inter-
ests of the underdeveloped countries. Senior UN officials attempted to
on a neo-Ricardian model of price formation at the center and technical progress, and the 1959
version, in which there is no functional distribution or technical progress, just certain nonstan-
dard characteristics of exogenous demand.
50. In an interview conducted by Tom Weiss for the UN Intellectual History Project.
Toye and Toye / The Prebisch-Singer Thesis 463
distance themselves from Prebisch by identifying him as the author of
the views to which the developed countries took exception. In so doing,
they unwittingly heightened his prestige.
The common belief that the thesis of the deterioration of the terms
of trade was first picked up from Latin America by North American and
European scholars during the 1950s is therefore inaccurate (Kay 1989,
8). What happened in the 1950s was in fact the transmission of the the-
sis back again. The more detailed chronology of events given here has
also indicated a further unsatisfactory aspect of the conventional wis-
dom, namely, the general lack of precision about what constitutes the
Prebisch-Singer thesis. Is it to be understood simply as a claim about
a long-run downward trend in the terms of trade of primary producers,
that is to say, the statistical phenomenon of secular decline? Or, is it the
delineation of an economic mechanism that could account for a long-run
secular decline? Or is it both? If the thesis is the empirical fact of a sec-
ular decline, it is clear that Singer had priority, and Prebisch’s work was
wholly derivative. If it is the specification of a theoretical mechanism
to account for a secular decline, Singer also had priority, but Prebisch’s
contribution was independent—and also more elaborate, if convoluted.
In terms of the dissemination of these ideas, however, it is clear that
Singer benefited from the existence of Prebisch as much as Prebisch had
previously benefited from the existence of Singer. Let us ask what would
have happened to the ideas of each, had the other been absent. If Singer
had not overcome his personal reluctance to join the UN, Prebisch may
well not have integrated declining terms of trade into his text. However,
if Prebisch had gone to work for the IMF and not ECLA, the impact of
Singer’s study in Latin America might have amounted to no more than
the very faint mark that it left on the 1948 Economic Survey. From the
viewpoint of publicity and political repercussions, therefore, it was in-
deed the “Prebisch-Singer thesis.”
The thesis promptly entered the Cold War battlefield. The initial U.S.
response was an attempt to close down ECLA in 1951.51 Having failed,
successive administrations slowly learned that Prebisch was in fact more
pragmatic than he was polemical. By 1961, President Kennedy proposed
51. At the 1951 ECLA meeting in Mexico, a U.S. proposal to close the commission down
was narrowly averted after Prebisch offered a stout defense of its work (and after a timely
intervention from President Vargas of Brazil). See Magariños 1991, 138–141; Furtado 1987, 94;
and “Progress Report Made by the Executive Secretary to the Fourth Session,” UN document
no. EC/CN.12/220, 29 May 1951.
464 History of Political Economy 35:3 (2003)
to the UN a “development decade” and launched the Alliance for Prog-
ress in the hope of a more constructive U.S. relationship with LatinAmer-
ica. Mr. M.V. Lavrichenko, deputy head of the Department of the USSR
Ministry of Foreign Affairs, argued in the Second Committee of the Gen-
eral Assembly, that ECOSOC “should devote more of its attention to
such urgent economic problems as the prevention of the economic plun-
dering by the imperialist Powers of the countries of Asia, Africa and
Latin America.” In support of this claim, he reminded his listeners that “a
United Nations survey published in 1949 concluded that, in the course
of almost half a century, there had been a steady drop in the prices of
raw materials in comparison with those of industrial goods.”52 Thus he
demonstrated the long-lasting controversial power of the thesis that Pre-
bisch and Singer had between them articulated, and that the United
Nations had attempted to bury, yet inadvertently had ended up making
world-famous.
References
Ahumada, J., and A. Nataf. 1950. Terms of Trade in Latin American Countries. IMF
Staff Papers 1.1:123–35.
Bloch, H., and D. Sapsford. 1998. Prebisch and Singer Effects on the Terms of Trade
between Primary Producers and Manufacturers. In Development Economics and
Policy, edited by D. Sapsford and J. Chen. London: Macmillan.
Cardoso, F. H. [1980] 1984. Les idées à leur place. Paris: Éditions A. M. Métailié.
Dosman, E. J. 2001. Markets and the State: Theory and Practice in the Evolution of
the “Prebisch Manifesto.” Revista de la CEPAL/CEPAL Review, no. 75:87–102.
Ferrer,A. 1990. The Early Teachings of Raúl Prebisch. Revista de la CEPAL/CEPAL
Review, no. 42:27–33.
FitzGerald, E. V. K. 1994. ECLA and the Formation of Latin American Economic
Doctrine. In Latin America in the 1940s: War and Postwar Transitions, edited by
D. Rock. Berkeley: University of California Press.
. 2000. ECLA and the Theory of Import Substituting Industrialization in
Latin America. In Industrialization and State in Latin America: The Post-war
Years, edited by E. Cardenas, J. A. Ocampo, and R. Thorp. London: Macmillan.
Flanders, M. J. 1964. Prebisch on Protectionism: An Evaluation. Economic Journal
74:305–26.
Furtado, C. 1987. La fantaisie organisée: Le développement est-il encore possible?
Paris: Publisud.
52. Official Record of the UN General Assembly, sixteenth session, second committee,
720th meeting, 11 October 1961, pages 22–24.
Toye and Toye / The Prebisch-Singer Thesis 465
Haberler, G. 1988. International Trade and Economic Development. San Francisco:
International Center for Economic Growth.
Hagemann, H., ed. 1997. Zur deutschsprachigen wirtschaftswissenschaftlichen Em-
igration nach 1933. Marburg: Metropolis-Verlag.
Harris, S. E., ed. 1943. Postwar Economic Problems. London: McGraw-Hill.
Kay, C. 1989. Latin American Theories of Development and Underdevelopment.
London: Routledge.
Keynes, J. M. 1973–89. The Collected Writings of John Maynard Keynes. 30 vols.
London: Macmillan.
Kindleberger, C. P. 1943a. International Monetary Stabilization. In Postwar Eco-
nomic Problems, edited by S. E. Harris, 375–95. London: McGraw-Hill.
. 1943b. Planning for Foreign Investment. American Economic Review, Pa-
pers and Proceedings 33:347–54.
League of Nations. 1945. Industrialization and Foreign Trade. League of Nations
Series II. Economic and Financial 1945.II.A.10.
Love, J. L. 1980. Raúl Prebisch and the Origins of the Doctrine of Unequal Ex-
change. Latin American Research Review 15.3:45–72.
. 1991. A New Look at the International Intellectual Environment of the Thir-
ties and Forties. Hans Singer Papers in possession of John Toye, Prebisch box.
. 1994. Economic Ideas and Ideologies in Latin America since 1930. In
Economy and Society,pt.1ofLatin America since 1930: Economy, Society, and
Politics.Vol.6ofThe Cambridge History of Latin America, edited by L. Bethell.
Cambridge: Cambridge University Press.
Loveday, A. 1956. Reflections on International Administration. Oxford: Clarendon
Press.
Magariños, M. 1991. Diálogos con Raúl Prebisch. Mexico: Banco Nacional de Com-
mercio Exterior/Fondo de Cultura Económica.
Meier, G. M., and D. Seers, eds. 1984. Pioneers in Development. Washington, D.C.:
The World Bank.
Mitchell, W. C. 1927. Business Cycles. New York: National Bureau of Economic
Research.
Osiatynski, J., ed. 1990–97. Collected Works of Michal Kalecki. 7 vols. Oxford:
Oxford University Press.
Prebisch, R. 1945. Introducción al curso de economía política. Revista de ciencias
económicas 33:525–37.
. 1946. Responsabilidad de los paises de la periferia. In Memoria: Primera
reunion de técnicos sobre problemas de Banca Central del Continente Ameri-
cano. Mexico City: Banco de México, S.A.
. 1948a. Apuntes de economía política (dinámica económica): Clases dic-
tadas en el Curso Universitario de 1948. Buenos Aires: Faculty of Economic
Sciences, University of Buenos Aires.
. 1948b. Dictamen del Dr. Raúl Prebisch acerca de los anteproyectos sobre
Banco Central y bancos. Revista de hacienda 13.28:146–78.
466 History of Political Economy 35:3 (2003)
. 1950. The Economic Development of Latin America and Its Principal Prob-
lems. UN document no. E/CN.12/89/Rev.1. Lake Success, N.Y.: United Nations.
. 1959. Commercial Policy in Under-developed Countries. American Eco-
nomic Review, Papers and Proceedings 492:251–73.
. [1934] 1991. La inflación escolástica y la moneda Argentina. In vol. 2 of
Obras 1919–1948, 337–50. Buenos Aires: Fundación Raúl Prebisch.
. 1991. Obras 1919–1948. 3 vols. Buenos Aires: Fundación Raúl Prebisch.
Samuelson, P. 1948. International Trade and the Equalisation of Factor Prices. Eco-
nomic Journal 58 (June):163–84.
Santa Cruz, H. 1995. The Creation of the United Nations and ECLAC. Revista de la
CEPAL/CEPAL Review, no. 57:17–33.
Schellenberg, Walter. 2000. Invasion 1940: The Nazi Invasion Plan for Britain. Lon-
don: St. Ermin’s Press.
Schumpeter, J. 1939. Business Cycles: A Theoretical, Historical, and Statistical
Analysis of the Capitalist Process. London: McGraw-Hill.
Simonsen, R. C. 1939. Brazil’s Industrial Evolution. São Paulo: Escola Livre de
Sociologia e Política.
Singer, H. W. 1949. Economic Progress in Underdeveloped Countries. Social Re-
search: An International Quarterly of Political and Social Science 16.1:1–11.
. [1950] 1975. The Distribution of Gains between Investing and Borrowing
Countries. In The Strategy of International Development: Essays in the Econom-
ics of Backwardness. London: Macmillan.
. 1984. The Terms of Trade Controversy and the Evolution of Soft Financing:
Early Years in the UN. In Pioneers in Development, edited by G. M. Meier and
D. Seers. Washington, D.C.: The World Bank.
. 1991. Comment by Hans Singer on “Raúl Prebisch, 1901–1971: The Con-
tinuing Quest,” by Edgar J. Dosman and David H. Pollock. Hans Singer Papers
in possession of John Toye, Prebisch box.
. 1997. The Influence of Schumpeter and Keynes on the Development
of a Development Economist. In Zur deutschsprachigen wirtschaftswissen-
schaftlichen Emigration nach 1933, edited by H. Hagemann. Marburg: Metrop-
olis-Verlag.
Spraos, J. 1980. The Statistical Debate on the Net Barter Terms of Trade between
Primary Commodities and Manufactures. Economic Journal 90:107–28.
Thorp, R. 1998. Progress, Poverty, and Exclusion: An Economic History of Latin
America in the 20th Century. Baltimore: Johns Hopkins University Press.
Toye, J. 2000. Keynes on Population. Oxford: Oxford University Press.
United Nations. 1949a. Economic Survey of Latin America 1948. UN document no.
E/CN.12/82. Lake Success, N.Y.: UN Department of Economic Affairs.
. 1949b. Post-war Price Relations in Trade between Under-developed and
Industrialized Countries. UN document no. E/CN.1/Sub.3/W.5. Lake Success,
N.Y.: UN Department of Economic Affairs.
Toye and Toye / The Prebisch-Singer Thesis 467
. 1949c. Relative Prices of Exports and Imports of Under-developed Coun-
tries. Lake Success, N.Y.: UN Department of Economic Affairs.
. 1949d. Report to the Economic and Employment Commission on the Third
Session of the Sub-Commission Held from 21 March to 11 April 1949. UN docu-
ment no. E/CN.1/65. Lake Success, N.Y.: Sub-Commission on Economic Devel-
opment.
. 1951. Economic Survey of Latin America 1949. UN document no.
E/CN.12/164/Rev.1. NewYork: United Nations Department of Economic Affairs.
. 1987. Raúl Prebisch: Un aporte al estudio de su pensiamento. UN docu-
ment no. LC/G.1461. Santiago de Chile: CEPAL.
Viner, J. 1952. International Trade and Economic Development. Oxford: Oxford
University Press.
Whitaker, A. P., ed. 1945. Inter-American Affairs 1944. New York: Columbia Uni-
versity Press.