See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/4825947

The agricultural transformation

Article in Handbook of Development Economics · January 1988

Source: RePEc

CITATIONS READS

583 8,745

1 author:

C. Peter Timmer
Harvard University, emeritus
204 PUBLICATIONS 11,375 CITATIONS

SEE PROFILE

All content following this page was uploaded by C. Peter Timmer on 30 May 2019.

The user has requested enhancement of the downloaded file.

Chapter 8

THE AGRICULTURAL TRANSFORMATION

C. PETER TIMMER*
Harvard University

Contents
1. Introduction 276
2. The process of agricultural transformation 279
2.1. Evolving stages 280
2.2. Agriculture and economic development 283
2.3. The role of the agricultural sector 288
3. Why agriculture is different 291
3.1. Decision-making in agriculture 292
3.2. Characteristics of agricultural production functions 294
3.3. The farm household as both producer and consumer 299
3.4. What difference does the difference make? 300
4. Transforming agriculture 302
4.1. The sources and dynamics of technical change 302
4.2. Unresolved issues 313
5. Agricultural development strategy 321
5.1. Policies for "getting agriculture moving" 321
5.2. Alternative strategies for maintaining the transformation process 323
5.3. Agricultural policy and structural change 327
References 328

*I would like to thank the participants at the authors' workshop for helpful reactions to my initial
ideas for this chapter. Particular thanks go to Larry Westphal, Pranab K. Bardhan, David Dapice,
and Scott Pearson for serious and critical readings of the first draft. As always, my deepest debt is to
my wife and editor, Carol, for her patience and persistence in helping me make my manuscripts
readable and for her mastery of the wonderful new technology that permits me to lose half the
manuscript with the push of a button and for her to get it back after considerable effort and anguish.

Handbook of Development Economics, Volume I, Edited by H. Cheneo' and T.N. Srinivasan

© Elsevier Science Publishers B.V., 1988
276 C.P. Timmer

1. Introduction

The agricultural transformation has been a remarkably uniform process when

viewed from outside the agricultural sector itself. As documented by Clark
(1940), Kuznets (1966), Chenery and Syrquin (1975), and the patterns reported in
the introductory chapter to Part II of the Handbook, the share of agriculture in a
country's labor force and total output declines in both cross-section and time-
series samples as incomes per capita increase. The declining importance of
agriculture is uniform and pervasive, a tendency obviously driven by powerful
forces inherent in the development process, whether in socialist or capitalist
countries, Asian, Latin American, or African, currently developed or still poor.
It is at least slightly puzzling, then, that a second uniform and pervasive aspect
of the development process also involves agriculture- the apparent requirement
that rapid agricultural growth accompany or precede general economic growth.
The logic of the classical model of economic growth requires it:
Now if the capitalist sector produces no food, its expansion increases the
demand for food, raises the price of food in terms of capitalist products, and so
reduces profits. This is one of the senses in which industrialization is dependent
upon agricultural improvement; it is not profitable to produce a growing
volume of manufactures unless agricultural production is growing simulta-
neously. This is also why industrial and agrarian revolutions always go
together, and why economies in which agriculture is stagnant do not show
industrial development [Lewis (1954, p. 433, emphasis added)].
The historical record to which Lewis alludes supports the strong link between
agricultural and industrial growth, at least in market-oriented economies. The
English model is often held up as the case in point:
Consider what happened in the original home of industrial development, in
England in the eighteenth century. Everyone knows that the spectacular
industrial revolution would not have been possible without the agricultural
revolution that preceded it. And what was this agricultural revolution? It was
based on the introduction of the turnip. The lowly turnip made possible a
change in crop rotation which did not require much capital, but which brought
about a tremendous rise in agricultural productivity. As a result, more food
could be grown with much less manpower. Manpower was released for capital
construction. The growth of industry would not have been possible without the
turnip and other improvements in agriculture [Nurkse (1953, pp. 52-53)].
Despite a significantly different view in the current literature about the impact
of the English agricultural revolution on labor productivity, the key importance
Ch. 8: The Agricultural Transformation 277

of the increase in agricultural output has not been challenged [Timmer (1969),
Hayami and Ruttan (1985)]. Nor is this importance restricted to the lessons from
the currently developed countries. In surveying the statistical link between
agric~altural and overall economic growth in currently less-developed countries,
the World Bank reached the following conclusions:

The continuing importance of agriculture in the economies of the developing

countries is reflected in the association between the growth of agriculture and
of the economy as a whole. Among countries where the agricultural share of
GDP was greater than 20 percent in 1970, agricultural growth in the 1970s
exceeded 3 percent a year in 17 of the 23 countries whose GDP growth was
above 5 percent a year [see Table 8.1]. During the same period, 11 of the 17
countries with GDP growth below 3 percent a year managed agricultural
growth of only 1 percent or less. Agricultural and GDP growth differed by less
than two percentage points in 11 of 15 countries experiencing moderate
growth. There have been exceptions, of course, but they prove the rule: fast
GDP growth and sluggish agriculture was a feature of some of the oil- or
mineral-based economies such as Algeria, Ecuador, Morocco, and Nigeria.
The parallels between agricultural and GDP growth suggest that the factors
which affect agricultural performance may be linked to economy-wide social
and economic policies . . . . Expanding agricultural production through techno-
logical change and trade creates important demands for the outputs of other
sectors, notably fertilizer, transportation, commercial services, and construc-
tion. At the same time, agricultural households are often the basic market for a
wide range of consumer goods that loom large in the early stages of industrial
development-textiles and clothing, processed foods, kerosene and vegetable
oils, aluminum hoUoware, radios, bicycles, and construction materials for home
improvements [World Bank (1982, pp. 44-45)].

The need for rapid agricultural growth and for the decline in the agricultural
sector's share of output and the labor force are not contradictory, of course, but
the apparent paradox gave rise to a widespread misperception that agriculture is
unimportant-that it does not require resources or a favorable policy environ-
ment - because its relative share of the economy declines.
So long as market forces provide the primary direction to the sectoral alloca-
tion of resources, how academics perceive this process is irrelevant to the process
itself. When government planners intercede, however, they do so within a
framework of objectives and constraints, and this framework is ultimately
conditioned by the prevailing academic understanding of how economic growth
proceeds. The mainstream paradigm of the 1950s suggested that agriculture could
and should be squeezed on behalf of the more dynamic sectors of the economy.
This strategy could be successful if agriculture was already growing rapidly (as in
278 C.P. Timmer

Table 8.1
Growth of agriculture and G D P in the 1970s

Agricultural G D P growth
growth Above 5 percent 3 - 5 percent Below 3 percent

Above 3 percent Cameroon Malawi a Bolivia Liberia

China a Malaysia Burma a Nicaragua
Colombia Paraguay Mall a Senegal
Dominican Rep. Philippines Somalia a
Guatemala Thailand Tanzania a
Indonesia Tunisia
Ivory Coast Turkey
Kenya Yemen Arab Rep.
Korea, Rep. of
1 - 3 percent Costa Rica Bangladesh~ Burundi a
Ecuador Central African Rep. ~ Sierra Leone a
Egypt El Salvador Zaire a
Lesotho Haiti a
Honduras
India a
Pakistan a
Sri Lanka a
Sudan a
Upper Volta a
Below 1 percent Morocco Togoa Angola a
Nigeria Chad a
Congo. Rep.
Ethiopia a
Ghana
Madagascar a
Mauritania a
Mozambique a
Nepal a
Niger a
Uganda a

a Low-income countries.
Source: World Bank (1982, p. 45).

Western Europe and Japan) or if it started with a large surplus relative to the
subsistence needs of the rural population (as in the USSR). But if the agricultural
sector started with traditional technology and yields and living standards near
subsistence, the "squeeze agriculture" paradigm created economic stagnation, not
growth. In those cases, major attention was needed to induce an agricultural
transformation if the industrial revolution was to have any real hope of success.
Upon closer examination, it is not paradoxical that agricultural growth leads to
agricultural decline. At least two mechanisms, now relatively well understood and
Ch. 8: The Agricultural Transformation 279
documented, account for this process of structural transformation. 1 Engel's Law
alone, in a closed economy with constant prices, explains a declining share for
agriculture (and low farm incomes unless some farmers leave agriculture) no
matter how fast the sector grows. Because growth is led by demand patterns in
market economies, a less-than-unitary income elasticity for the products of the
agricultural sector guarantees that gross value of sales by farmers will grow less
rapidly than gross domestic product. As Lewis implies in the previous quotation,
if agricultural output fails to grow rapidly enough, rising prices might actually
garner farmers a higher share of consumers' expenditures. But this reflects lower
real incomes, not the result of economic growth.
If the terms of trade are not to rise in favor of agriculture, farm productivity
must r i s e - a n agricultural revolution is needed. The second factor that explains
the joint agricultural growth and relative decline is seen in the rapid growth in
agricultural productivity, measured by output per laborer or output per hectare,
in all the successfully developed countries. Technical change in agriculture in all
of the OECD countries proceeded at such a pace that the long-run terms of trade
declined for farm products. Lower prices thus exacerbated the sluggish demand
growth due to low income elasticities; the combination put pressure on agricul-
tural resources to move out of farming and into the more rapidly growing sectors
of the economy. Such intersectoral movements of resources have been painful in
all societies that have undergone successful structural transformation, and all
societies have found mechanisms to cushion the adjustment process.
The paradox over the agricultural transformation occurs at this point. Just as
countries learn how to institutionalize the process of rapid technical change in
agriculture, its product no longer has high social value. The resulting low incomes
for farmers create powerful political pressures to slow the process of structural
change, and the seemingly inevitable result is massive distortion of the price
structure [Johnson (1973), Anderson and Hayami (1986), World Bank (1986)].
Nearly all rich countries protect their agricultural sectors from international
competition, and countries no farther along in the development process than
Malaysia, Indonesia, Zimbabwe, and Mexico protect key food-producing sectors
during periods of depressed world prices.

2. The process of agricultural transformation

From both historical and contemporary cross-section perspectives, the agricul-

tural transformation seems to evolve through at least four phases that are roughly

XFor a very useful summary of the literature that documents the agricultural transformation
process itself and also attempts to explain it in terms of the prevailing models of economic
development, see Johnston (1970).
280 CP. Timmer

definable. The process starts when agricultural productivity per worker rises. This
increased productivity creates a surplus, which in the second phase can be tapped
directly, through taxation and factor flows, or indirectly, through government
intervention into the rural-urban terms of trade. This surplus can be utilized to
develop the nonagricultural sector, and this phase has been the focus of most
dual economy models of development. For resources to flow out of agriculture,
rural factor and product markets must become better integrated with those in the
rest of the economy. The progressive integration of the agricultural sector into
the macro economy, via improved infrastructure and market-equilibrium link-
ages, represents a third phase in agricultural development. When this phase is
successful, the fourth phase is barely noticeable; the role of agriculture in
industrialized economies is little.different from the rote of the steel, housing, or
insurance sectors. But when the integration is not successfully accomplished - and
most countries have found it extremely difficult for political reasons- govern-
ments encounter serious problems of resource allocation and even problems
beyond their borders because of pervasive attempts by high-income countries to
protect their farmers from foreign competition. Managing agricultural protection
and its impact on world commodity markets thus provides a continuing focus for
agricultural policy makers even when the agricultural transformation is "com-
plete".

2.1. Euoloing stages

The four phases in the agricultural transformation call for different policy
approaches. In the earliest stage of development the concern must be for "getting
agriculture moving", to use Arthur Mosher's vivid phrase [Mosher (1966)]. A
significant share of a country's investable resources may well be extracted from
agriculture at this stage, but this is because the rest of the economy is so small.
Direct or indirect taxation of agriculture is the only significant source of
government revenue.
Building a dynamic agriculture requires that some of these resources be
devoted to the agricultural sector itself. As the section on agricultural develop-
ment policy at the end of this chapter explains, these resources need to be
allocated to public investment in research and infrastructure as well as to
favorable price incentives to farmers to adopt new technology as it becomes
available. As these investments in agriculture begin to pay off, the second phase
emerges in which the agricultural sector becomes a key contributor to the overall
growth process through a combination of factors outlined by Johnston and
Mellor (1961).
As the empirical literature on structural patterns of growth emphasizes, there is
a substantial disequilibrium between agriculture and industry at this early stage
Ch. 8: The Agricultural Transformation 281
of the development process [Kuznets (1966), Chenery and Taylor (1968), Chenery
and Syrquin (1975)]. Indeed, differences in labor productivity and measured
income (as opposed to psychic income) between the rural and urban sectors
persist to the present in rich countries, although the gap is narrowing and now
depends on agricultural prices for any given year. 2
The process of narrowing the gap gives rise to the third environment for
agriculture, in which it is integrated into the rest of the economy through the
development of more efficient labor and credit markets that link the urban and
rural economies. This integration is a component of the contribution process; the
improved functioning of factor markets merely speeds the process of extracting
labor and capital from those uses in agriculture with low returns for those in
industry or services with higher productivity. The improved markets have welfare
consequences as well, because they lessen the burden on individuals trapped in
low-income occupations. The gain has costs, however. As agriculture is integrated
into the macro economy, it becomes much more vulnerable to fluctuations in
macro prices and level of aggregate activity and trade [Schuh (1976)] and much
less susceptible to management by traditional instruments for the agricultural
sector, such as extension activities and specific programs for commodity develop-
ment and marketing.
This vulnerability and complexity create the fourth phase in the agricultural
transformation, the treatment of agriculture in industrialized economies. As the
share of the labor force in agriculture falls below about 20 percent and the share
of food expenditures in urban household budgets drops to about 30 percent,
low-cost food is not as important to the overall economy nor is it as expensive in
relative terms to increase in price [Anderson (1983)]. A host of political problems
arise if low farm incomes, induced by rapid technical change and low farm-gate
prices, are allowed to push resources out of agriculture. Farmers do not want to
leave, especially if they must sell their farms under duress at low prices; and
urban-based unions do not want to see them coming to the cities in search of
industrial jobs. A nostalgic memory of farming as a " w a y of life" leads many
second- and third-generation farm migrants living in cities to lend political
support to higher incomes for agriculture, even at the expense of higher grocery
bills (which m a y be barely noticeable). By this stage of the process, the share of
the farm-gate price of the commodity in the consumer's market basket is small
because of processing and marketing costs. Commodity price supports become

2The structural rigidities in the economy that give rise to this substantial disequilibrium obviously
mean that neoclassical models based solely on perfect markets and rational actors will fail to predict
accurately the impact of government interventions. However,purely structural models that assume an
absence of market response might be equally far from the mark. A messy amalgam of structural
rigidities, imperfect markets, and decision-makersinterested in their own, but vaguelydefined, welfare
seems to characterize the actual starting point from which government interventions must be
evaluated.
 282 C.P. Timmer

/
f
~ -" ~ ~ ~.
o --

--6 ~ . - @ 6
"E E
g ~ ' 0; /
o o ,~ g,,-~ g~
~-~ / ~ tO
- _ ~
~ E ~ ~ 0 3

_~ ~ . ,'T

5 9~- = .~a /"

LZ
/ V- _gE. -=o
/ / o E..~
/
/
I
. . . . . . i /
/
/ 0
/
i /
/
.:..:~ .E ~ E o m'-~
/
I .1:: ~I.E

.~> ~ ~,.= ~ ~ ,.o

-L
6
\
\
\
~E
~g \
\
g~
\

il ' _ _ _ J I- . . . .
J=
o

L~
Ch. 8." The Agricultural Transformation 283

the primary vehicle for supporting farm incomes, and the subsidies have devastat-
ing effects on resource allocation. Farmers invest heavily in land and machinery
when farm prices are high, only to produce surpluses that are impossible to sell
profitably [Johnson (1985), Cochrane (1979)]. Eventually, the biadgetary and
distortionary costs of this approach become so high that even the European
Community, Japan, or the United States must face choices over how to rational-
ize agricultural returns with their social profitability.
The economic environments for agriculture created by these four phases are
shown schematically in Figure 8.1. The financial and labor resource flows out of
agriculture over time (or as incomes increase in a cross-section sample) are
impressionistic. Whether the trough between the "Mosher environment" and the
"Johnston-Mellor environment" in Figure 8.1 drops into negative ground or
always remains positive presumably depends on alternative sources of financial
resources at this stage in development. Urban or overseas remittances, petroleum
revenues, or foreign assistance might temporarily fill the gap left by a declining
relative contribution from agriculture. 3 But as agricultural productivity begins to
rise, labor and financial flows to the rest of. the economy increase. The
"Schultz-Ruttan environment" begins as the absolute population in agriculture
starts to decline, and the "D.G. Johnson environment" begins as the agricultural
labor force drops to a fairly small proportion of the overall labor force. Whether
financial resources continue to flow out of agriculture at this stage in the process
depends almost entirely on government price policy and its resulting impact on
farm investment. Policies to cushion the impact on farmers of successful structur-
al change need not inevitably rely on price interventions that impede the
adjustment process, but price supports have been the most popular in the United
States, Western Europe, and Japan for plausible political reasons [Anderson and
Hayami (1986)].

2.2. Agriculture and economic development

This overview of the agricultural transformation raises two basic issues to be

discussed in this chapter: the contribution or role of agriculture in economic
development, and the conditions or factors that lead to the modernization of the
agricultural sector itself. Obviously, many other important topics are not treated
here. One is the changing control over resources in the rural sector, which
determines who gains and loses during the agricultural transformation. Only the

3It is also important to distinguish subsectors within agriculture. An export crop subsector
producing rubber or coffee might continue to provide financial resources to the rest of the economy,
some of which could be returned to the foodcrop subsector in order to foster its development. Much
of the discussion in this chapter is concerned with modernizing the foodcrop subsector while
recognizing the important role played by the other agricultural subsectors.
284 C.P. Timrner

structuralist and radical political economy literature deals directly with the
distribution of income and power in rural areas as an integral component of
agricultural development. A major theme of "neo-neoclassical analysis" since the
mid-1970s, however, has been the incorporation of such issues into rational actor
models of rural household decision-making [see Bardhan, Chapter 3 in this
Handbook]. While much of the dynamic and macroeconomic perspective of the
radical models is lost in the household models, much is gained in the form of
testable hypotheses about the impact of new technology or pricing policies on the
structure of rural markets and distribution of output in the short run. 4
The historical record after the Second World War suggests that many countries
saw an opportunity to pursue a "jump strategy" and move directly from the early
stages of the Mosher environment in Figure 8.1 to the later stages of the
Johnston-Mellor environment, thus bypassing the necessity to invest in agricul-
tural development.

... the most significant comparison.., is that between the levels of productiv-
ity in the under-developed countries and the western countries at the period
when the latter began to industrialize . . . . IT]he present average level of
agricultural productivity in African and Asian countries (between them repre-
senting four-fifths of the Third World population) is 45 percent below that
reached by the developed countries at the start of the industrial revolution. In
fact it is at the same level as that of the European countries before their
agricultural revolution.
Now, most under-developed countries wish, consciously or unconsciously, to
by-pass this stage just when other structural conditions of development are
making a "take ofF' more difficult than it was when most European countries
and the United States were imitating England's example. What makes the
failure to admit or even to recognize this problem all the more serious is that
the problem itself is intractable. Leaving aside mental attitudes, landownership
and political considerations, it cannot be stressed too forcibly that an increase
in the area cultivated per agricultural worker is one of the essential conditions
of an increase in productivity. But in view of the population explosion it is
impossible to assume, even on most hopeful assumption, that the reduction in
cultivated area per worker will be anything but slight [Bairoch (1975, p. 42)].
A jump strategy sees the extraction of resources from agriculture for economic
development as being in conflict with the investment of public and private
resources in its modernization. This has been especially true in countries with
systems of planned resource allocations designed to force the pace of economic
development. As more and more countries adopted the paradigm of central

4See Hart (forthcoming) for an eloquent complaint that such micro models effectively "gut" the
Marxian analysis of its vision of class interactions providing the driving force to rural dynamics.
Ch. 8: The Agricultural Transformation 285

planning to direct these resource allocations, the separate issues of contribution

and modernization became key analytical issues as well. Unfortunately, the
economics profession was ill-equipped to address them because all previous
examples of agricultural modernization had taken place within more or less
market-oriented settings (except in the Soviet Union, where agricultural modern-
ization remains quite incomplete). The behavior of backward agricultural systems
under the new planning context became a topic of much theorizing and debate,
but only in the 1960s and 1970s did the empirical record become both long and
varied enough to draw reasonably firm conclusions.
It is worth summarizing briefly what the empirical record showed by 1960
when the results of Kuznets' decade-long study of the quantitative aspects of
modern economic growth started to be widely available. The historical record
began as early as the late eighteenth century in the United Kingdom and 1839 in
the United States and as late as 1880 in Japan and 1925 in the USSR. For all
countries for all time periods observed, the share of agriculture in the total labor
force declined, sometimes sharply, as in Sweden, the United States, and Japan,
and sometimes more gradually, as in the United Kingdom, Belgium, Italy, and
Australia. The share of agriculture in national output showed slightly more mixed
patterns than those of the labor force. The share was nearly stable or even rose
slightly over some periods in the United Kingdom, France, the United States,
and Australia. The more general tendency of the share in output to decline is
clear, but the share of the labor force always declined more rapidly. The obvious
result was that labor productivity in agriculture rose more rapidly than in the
economy as a whole when measured over the long periods of time required for
sustained economic growth to cause substantial changes in the structure of an
economy. Although agricultural productivity per worker was nearly always less
than the level of national productivity, its faster rise meant that the gap tended to
narrow.
Three clear exceptions to this trend in Kuznets' data are Italy, Japan, and the
USSR, all of which are latecomers to the process of sustained growth and are
countries in which state intervention into the industrialization process was much
more active than in the early developers. The failure of agricultural productivity
per worker to rise as fast as national productivity in these three countries might
thus be seen as an early signal that the patterns in the less-developed countries
seeking to start down the path of modern economic growth might be significantly
different from the historical path followed by the Western countries and docu-
mented by Kuznets. Table 8.2, drawn from a paper by Hayami (1986), shows that
the recent productivity record for the rapidly growing East Asian economies
confirms a strongly different pattern from that in North America and Western
Europe. Even the more slowly growing developing countries (Philippines and
India) have a mild reversal of the "traditional" pattern in which growth in labor
productivity in agriculture exceeds that of labor productivity in manufacturing.
286 C.P. Timmer

Table 8.2
International comparison in the growth rates of labor productivity
in agriculture and manufacturing, 1960 (1958-62 averages)
to 1980 (1978-82 averages)

Labor productivity
growth rate (%/year) a
Agriculture Manufacturing (1)-(2)
(1) (2) = (3)

Developed countries:
United States 6.3 3.2 3.1
United Kingdom 5.5 2.6 2.9
France 6.4 4.2 2.2
G e r m a n y (F.R.) 7,7 4.1 3.6
Japan 5,3 6.7 - 1.4
Developing countries:
Korea 4,0 7.5 -- 3.5
Philippines 3.2 3.5 b - 0.3
India 1.3 2.1 - 0.9

a Calculated from the ratios of the real output index to the employ-

ment index.
bGrowth rate from 1960 to 1975.
Sources: FAO, Production Yearbook; U N , Yearbook of Industrial Sta-
tistics; ILO, Yearbook of Labor Statistics; OECD, Labor Force Statis-
tics. Hayami (1986, p. 10).

This "premature" growth in manufacturing productivity (or, alternatively, the

neglect of efforts needed to raise agricultural productivity) is especially troubling
in historical perspective, as the quote from Bairoch previously indicated. Table
8.3 reproduces Bairoch's historical comparisons of "net agricultural production
by male labor employed in agriculture expressed in 'direct' calories". Only Italy
in 1840 had a lower productivity level than that of Africa and Asia in modern
times. The gap in agricultural productivity on average between European coun-
tries beginning their industrial revolutions and Africa and Asia is, as Bairoch
already noted, about 45 percent. "A gap of about 45 percent is sufficiently wide
for us to be able to assert that agricultural conditions in the currently developed
countries before the beginning of the industrial revolution must have been very
different from those of the under-developed countries of Asia and Africa today"
[Bairoch (1975, pp. 40-41)].
Based on data only up to the early 1970s, Bairoch's pessimism reflects the
widespread neglect of agriculture in many development efforts in the 1950s and
1960s, as well as the shortfalls in food production that triggered the world food
crisis in 1973-1974. A similar pessimism based on a quite different reading of the
historical record is provided by scholars working in the Marxian tradition and
following the insights of Lenin on the changing class structure of agriculture as it
becomes more capitalistic under the pressures of modernization [Baran (1952),
Ch. 8." The Agricultural Transformation 287

Table 8.3
Comparisons between levels of agricultural productivity

Country and "stage" of Index number of

development Period agricultural productivity

Developed countries:
Recent position
France 1968/72 100.0
United States 1968/72 330.0
Position before or during "take-off'
France 1810 7.0
Great Britain 1810 14.0
Sweden 1810 6.5
Belgium 1840 10.0
Germany 1840 7.5
Italy 1840 4.0
Russia 1840 7.0
Switzerland 1840 8.0
United States 1840 21.5
Spain 1860 11.0

Less-developed countries:
Recent position
Africa 1960/64-1968/72 4.7
Latin America a 1960/64-1968/72 9.8
Asia 1960/64-1968/72 4.8
Middle East 1960/64-1968/72 8.6

Total for all less-

developed countries: 1960/64-1968/72 5.5

a Excluding Argentina.
Source: Bairoch (1975, p. 40).

de Janvry (1981), Griffin (1979), Lenin (1899)]. The failure of the Marxist-Leninist
prediction that peasant (family) agriculture disappears under the competitive
pressures of modern corporate agriculture has led to a rethinking of the inevita-
bility of all countries following a path through capitalism to socialism and
eventually to communism. To explain the failure, the dependency school empha-
sizes relationships between the metropolitan (developed) center and the periphery
(underdeveloped) countries in the third world. A single process of global eco-
nomic growth occurs in a zero-sum context, in which the growth of the center is
at the direct expense of the periphery. Class relationships in the urban-based
governments of the periphery explain the perpetuation of economic policies that
favor only a small urban elite (and possibly landlords). In Latin America,
de Janvry (1981) and colleagues have extended the analysis to explain agricultur-
al policy and performance on the basis of a process of marginalization. Their
model argues that agricultural laborers and independent peasants gradually lose
288 C.P. Timmer

control of the resources needed to raise their living standards as large landowners
invest in capital-intensive farming techniques and displace peasants from the
market. The rural masses are too dispersed to mobilize effectively, and they suffer
a process of gradual immiseration.
Hayami and Ruttan provide a useful summary of three theories of develop-
ment and their implications for agriculture:

The implications of dependency theory for agricultural development stand in

sharp contrast to the growth-stage and dual-economy theories. The growth-stage
theories attempt to explain the process of transformation from a primarily
agrarian to an industrial economy. In the dynamic dual-economy models
incorporation of peasants into the market results in the disappearance of
dualism. The dependency perspective attempts to explain why the periphery
remains trapped in a backward agrarian state. In the dependency view incorpo-
ration of rural areas into the market is the source of marginalization-it
perpetuates rather than erodes dualism [Hayami and Ruttan (1985, p. 37)].

Although Hayami and Ruttan do not find the dependency theory very useful
for designing policies that foster the process of agricultural development, one of
the main questions asked by scholars of the dependency school remains un-
answered: why has agricultural development played a strongly positive role in the
overall development process in so few countries? Why have so many opportuni-
ties identified by agricultural scientists and economic planners been missed? Most
neoclassical scholars will agree that they do not have answers to these questions.

2.3. The role of the agricultural sector

The debate over the role of agriculture in the process of economic development
extends at least as far back as the Physiocrats in the eighteenth century. The
biblical advice to store during seven good years to be ready for seven lean years
certainly reflects a concern for agricultural planning. Clark (1940) and Kuznets
(1966) provided the general facts about the role of agriculture during the growth
process available to economists and planners at the beginning of the drive for
economic growth in the less-developed countries. These facts formed the basis for
the prevailing neoclassical view that agriculture was a declining sector, a "black
box" in Little's phrase (1982), which contributed labor, food, and perhaps capital
to the essential modernization efforts in industry. No policy efforts on behalf of
agriculture's own modernization were needed because the sector declined natu-
rally. Most interpretations of the Lewis model (1954), especially the Fei-Ranis
versions (1964), which became the main teaching paradigms, ignored the factors
needed to modernize traditional agricultural sectors so that they could play
Ch. 8: The Agricultural Transformation 289
positive contributory roles in the development of the rest of the economy. The
structuralist views of Prebisch (1950) about declining terms of trade for tradi-
tional products and the importance Hirschman (1958) attached to linkages to
" m o d e r n " economic activities further diminished any apparent rationale for
actively investing in the modernization of agriculture itself. As Hirschman wrote
in 1958, "agriculture certainly stands convicted on the count of its lack of direct
stimulus to the setting up of new activities through linkage effects - the superior-
ity of manufacturing in this respect is crushing" [Hirschman (1958, pp. 109-110)].
A final reason for the neglect of agriculture has recently been clarified by Sah
and Stiglitz (1984). The Soviet debate in the early 1920s over industrialization
policy revolved around whether turning the terms of trade against agriculture (the
" p r i c e scissors") would speed the rate of accumulation for investment by the
state. Preobrazhensky (1965) argued successfully that it could. Sah and Stiglitz
show the precise conditions under which he was right and the welfare conse-
quences that flowed from implementing such a policy. Although the conditions
that must hold for their analysis to be valid are very stringent, a robust result is
that the agricultural terms of trade should be lowered only if the state has a low
rate of time discount, that is, it favors investment over current consumption.
Forced-pace industrialization campaigns in such circumstances then rely on the
state's capacity to extract surpluses from agriculture even in the face of stagnant
or falling agricultural production.
It is easy to see why agriculture was neglected as a source of growth in early
strategies of economic development. The historical record shows that it always
declines in relative importance in growing economies. It is the home of tradi-
tional people, ways, and living standards- the antithesis of what nation builders
in developing countries envisioned for their societies. Moreover, agriculture was
thought to provide the only source of productivity that could be tapped to fuel
the drive for modernization. Surplus labor, surplus savings, and surplus expendi-
tures to buy the products of urban industry, and even surplus foreign exchange to
buy the machines to make them, could be had from an uncomplaining agricultur-
al sector. Nothing more was needed to generate these resources than the promise
of jobs in the cities and a shared nationalistic pride in the growing power of the
state. Despite how simplistic these promises sound in the mid-1980s, the
success of the Soviet approach caused them to be very appealing when first
uttered by such charismatic leaders of the developing world as Sukarno,
Nkrumah, Nasser, and Nehru. The unique features of agriculture as a sector were
simply not widely understood in the 1950s. Nor was it accepted that the
development of a modern agriculture was necessary as a concomitant to develop-
ment of the rest of the economy.
Some of these factors began to be recognized by the 1960s, and a more positive
emphasis was placed on "role" rather than the more forced concept of "contribu-
tion" of agriculture. The classic article by Johnston and Mellor (1961) listed five
290 C.P. Timmer

roles for agriculture in economic development:

(1) increase the supply of food for domestic consumption;
(2) release labor for industrial employment;
(3) enlarge the size of the market for industrial output;
(4) increase the supply of domestic savings; and
(5) earn foreign exchange.
Although the second, fourth, and fifth roles are certainly consistent with the
earlier "extractive" views of agriculture, Johnston and Mellor insisted that all five
roles are equally important. Agriculture in the process of development is to
provide increased food supplies and higher rural incomes to enlarge markets for
urban output, as well as to provide resources to expand that urban output.

It is our contention that "balanced growth" is needed in the sense of simulta-

neous efforts to promote agricultural and industrial development. We recognize
that there are severe limitations on the capacity of an underdeveloped country
to do everything at once. But it is precisely this consideration which under-
scores the importance of developing agriculture in such a way as to both
minimize its demand on resources most needed for industrial development and
maximize its net contribution required for general growth [Johnston and
Mellor (1961, pp. 590-591)].

Others, especially Nichols (1963), Schultz (1953), and Jorgenson (1961), also
emphasized this interdependence between a country's agriculture and its in-
dustry. Myint (1975) stressed a curious inconsistency between the "closed econ-
omy" model implicit in this domestic interdependence and the fifth role, earning
foreign exchange, which obviously implies the country is open to international
trade. This trade perspective returns in the 1970s and 1980s to dominate thinking
about appropriate development strategies, but it was largely ignored in the 1960s,
perhaps because of the dominance of the "Indian model" in development
thinking, in which sheer size keeps the importance of foreign trade quite small,
even apart from the "inward looking" strategy being pursued.
Despite the early insistence by agricultural economists that the agricultural
sector must be viewed as part of the overall economy and that the emphasis be
placed on the sector's interdependence with the industrial and service sectors
rather than on its forced contributions to them, the notion of agriculture as a
resource reservoir has persisted in general development models. Reynolds empha-
sized an important but usually overlooked distinction between static and dy-
namic views of the resource transfers:

In most development models, modern industry is the cutting edge of economic

growth, while agriculture plays the role of a resource reservoir which can be
drawn on for supplies of food, labor, and finance to fuel the growth of urban
Ch. 8." The Agricultural Transformation 291

activities. It is argued that this is both a logical necessity and a matter of

historical experience, illustrated by the case of Japan.
In commenting on this view, I must emphasize a distinction that is often not
clearly drawn: (1) It is one thing to assert that, in an economy where
agricultural output is not rising, the agricultural sector contains potential
surpluses of labor time, food output, and saving capacity requiring only
appropriate public policies for their release. This we may term the static view
of resource transfer. (2) It is quite a different thing to assert that, in an
economy where agricultural output is being raised by a combination of
investment and technical progress, part of the increment in farm output and
income is available for transfer to non-agriculture. This we may term the
dynamic view of resource transfer. The model-building implications of this
approach are different, and its policy implications are decidedly different
[Reynolds (1975, pp. 14-15)].

The welfare consequences of the two views are also sharply different. Forced
extraction of resources from a stagnant agricultural sector almost always creates
widespread rural poverty, sometimes famine. Market linkages that connect a
dynamic agricultural sector to rapidly growingindustrial and service sectors offer
an opportunity for rural inhabitants to choose in which sector they wish to
participate. There are certainly losers in this process: high-cost producers in
unfavorable ecological settings who cannot compete with low-cost producers in
favored locales who have access to new technology; or newly landless laborers
who have lost their tenancy access to land when commercial relationships replace
patron-client relationships. But new technology and market linkages create more
opportunities than they destroy if both the agricultural and nonagricultural
sectors are growing together. An emphasis on finding the policy environment that
creates such mutual growth is needed. For agriculture, that environment must
call forth rapid technical change. Experience since the mid-1960s has demon-
strated how to do that, but the key has been to understand why the agricultural
sector is different from the industrial and service sectors [Hayami and Ruttan
(1985), Timmer et al. (1983)].

3. Why agriculture is different

The early purposeful neglect of agriculture can be partly attributed to develop-

ment economists who were remote from any real understanding of what makes
the agricultural sector quite different from either manufacturing or services [Little
(1982)]. In developing countries, the agricultural sector is different from other
productive sectors of an economy, particularly in its large contribution to
292 C P . Tirnrner

national income and the large numbers of participants in the sector. Both the
agricultural transformation itself and the contribution of agriculture to the rest of
the economy depend on three important features discussed here: the peculiarities
of the agricultural production function, the importance of home consumption of
output for the sector, and the role of the agricultural sector as a resource
reservoir. These features are more evident in traditional societies, and their
distinctiveness erodes during the process of economic modernization. The design
of agricultural policy, in both poor and rich countries, is complicated by these
features, but a recognition of them is essential to a full understanding of the
contribution agriculture might realistically be asked to make to a country's
development effort. 5

3.1. Decision-making in agriculture

The sheer size of agriculture in most poor countries' economies, with over 50
percent of national output and up to 80 percent of the labor force in agricultural
activities, distinguishes the sector from all others in the early stages of develop-
ment. When directly related input and output industries and marketing activities
are included, "agribusiness" seldom declines to less than 20 percent of any
country's economy. Hence the sector remains the largest single "industry" in
absolute size even in rich countries.
In most countries, if the available arable land were divided equally among the
farm population, the resulting average farm size would be "small" by comparison
with United States or European standards. Farms of less than a hectare char-
acterize China, Bangladesh, and Java; even in Japan average farm size is still only
slightly greater than one hectare. The average in India is only about 1 to 2
hectares, and in Africa and Latin America farms tend to be less than 10 to 20
hectares in size. Average farm size in the United States is well over 100 hectares
and over 50 hectares in the United Kingdom.
The available farmland, of course, is usually not equally divided among all the
potential farmers. The conditions of land tenure and the size distribution of
farms are important characteristics of a country's agricultural decision-making
environment. A country with a unimodal distribution of f a r m s i z e s - a large
number of small, family-operated farms capable of supporting the family mem-
bers above a subsistence level, with only a fringe of smaller and larger farms
around this modal n o r m - h a s the potential to use agricultural development
strategy as a means of reducing rural poverty at the same time that it increases
agricultural production. Cotmtries with bimodal distributions of farm sizes - many

SAn effort to formalize the impact of agriculture's distinct features, especially the behavioral and
material determinants of production relations, is in Binswangerand Rosenzweig(1986).
Ch. 8: The Agricultural Transformation 293

very small farms on a minority of the land with a few very large, estate-like farms
that occupy most of the arable land and produce most of the food surplus
available for urban m a r k e t s - f a c e much more difficult dilemmas over how to
reduce the impact of rural poverty while using traditional output-increasing
strategies of agricultural development [Johnston and Kilby (1975)].
In both private and collective agricultures, decision-making is conditioned
primarily by the nature of incentives to work rather than by the pace and design
of the work itself, and these incentives are difficult to structure in an efficient
manner unless the cultivator owns the land. In situations where ownership and
operation are separate, a host of complicated contractual arrangements that
strive for second-best efficiency outcomes have evolved in different settings
[Bardhan, Chapter 3 in this Handbook, Binswanger and Rosenzweig (1981, 1986),
Stiglitz, Chapter 5 in this Handbook].
Farming is an undertaking that involves many decisions. What crops to plant,
what inputs to use, when to plow, to seed, to cultivate, to irrigate, to harvest, how
much to keep for home consumption, how much to sell and how much to store
for later sale are the farming decisions that occupy the daily routine of most
agricultural producers. What is unique about agriculture is that literally millions
of individuals and households are making these decisions themselves. Changing
agricultural production decisions to increase food output is an entirely different
process from changing decisions about how much steel or cement to produce. In
most countries a dozen or so individuals could take direct action which would
lead to a 10 percent increase in steel output in a year or so, and their decisions
would be decisive.
Nowhere, not even in socialist countries, can a similar small group of individu-
als decide to raise food production by 10 percent. A small group of planners, or
the president and the cabinet, can decide they w a n t food production to rise by 10
percent. They can tell the food logistics agency, the ministry of agriculture, the
newspapers, and agriculture extension agents that they want food production to
rise by 10 percent. But they cannot increase food production 10 percent by
themselves. They must also convince the millions of farmers in their country to
want to increase food production by 10 percent and make it in their self-interest
to do so.
The vast number of agricultural decision-makers implies that there are simply
too many to reach directly with either pleas for cooperation or police power.
Farmers must see the benefits of higher output for themselves because there are
too many opportunities to let high yields slip beneath the hoe or in a late
fertilizer application, even under the watchful eyes of a guardian. Farming is a
subtle combination of skilled craft and brute force. The brute force alone will not
achieve high yields.
In traditional agriculture with static technology, farmers learn these skills by
repeated trial and error. The lessons of parents and grandparents remain rele-
294 C.P. Timmer

vant. But when new technology becomes available, farmers do not automatically
acquire the requisite skills to deal with disequilibrium [Schultz (1964, 1975)].
Government interventions can have a high payoff, particularly investment in
extension services, general education (especially rural primary education that
includes instruction in farming skills), and rural infrastructure to lower the costs
of exchanging inputs and outputs, which become essential ingredients in speeding
the adoption of new agricultural technology.
The scope for effective government intervention is conditioned by the efficiency
with which farms allocate the resources at their disposal to produce crops,
relative to alternative uses of these resources, the technical ability of farmers to
achieve the maximum output from a given set of inputs, and the impact of
alternative forms of land tenure on both allocative and technical performance of
farmers. Given the large number of farmers within a typical developing country,
government extension agents cannot teach each individual farmer new agricultur-
al techniques. Price policy for farm crops and agricultural inputs, on the other
hand, is an intervention that reaches most farmers quite directly while being
amenable to effective government control. Consequently, knowing the role of
relative prices in influencing the behavior of farmers is extremely important. The
effectiveness of prices in changing producer decisions also depends on farmers'
allocative and technical efficiency and on the form of tenure contract for the land
they farm [Streeten (1986), Krishna (1984)]. It is a mistake to think that farmer
responsiveness to price is somehow immutable and is given exogenously to the
agricultural sector. Even if all farmers were narrow-minded profit-maximizers of
their available production functions, there would be substantial scope for altering
both the production function and the economic environment in which the
maximization takes place. In a world in which risk management involves the
establishment of patron-client relations, in which substantial bargaining may go
on within the farm household over task assignments, the division of income, and
gender-specific access to nutrients, and in which the access of farm members to
labor and credit markets may change radically over time even within fairly stable
agricultural technology and prices, the decision-making process itself must also
be treated as a variable.

3.2. Characteristics of agricultural production functions

One unusual feature of the agricultural production function is the efficiency cost
of separating labor and management. Knowing what the right inputs are, how to
combine them, and how to tend the process is the major function of manage-
ment. In owner-operated farming, this management skill is combined with the
farm household's own labor power, which is also an important ingredient in
growing crops. Several unique features of agricultural production functions
Ch. 8: The Agricultural Transformation 295
contribute to the decision-intensity of farming, to the productivity of the family
farm, and to the search for reasonably efficient substitutes for direct land-
ownership where the family farm is not prevalent. Seasonality, geographical
dispersion, and the role of risk and uncertainty are the most important.

3.2.1. Seasonality

No agricultural region of the world has an absolutely constant year-round

climate. Winter and summer create distinct growing seasons in the temperate
zones. Wet and dry seasons, or the monsoon season, create conditions when
planting is appropriate, when harvesting would be difficult, or simply when some
crops will not thrive. Climatic variations cause agricultural production to follow
distinct seasonal patterns even in most tropical areas, but seasonality is not a
fixed and rigid constraint. Rice will grow in the dry season if irrigation water is
provided, and tomatoes will grow in Siberia in January under artificial lights in a
warm greenhouse. Seasonality is important to farmers because it is generally
cheaper to let nature provide many of the essential inputs for agricultural
production- solar energy, water, carbon dioxide, temperature control, and essen-
tial nutrients from natural soils. But it is not always economical to let nature
dictate the agronomic environment. One of the major tasks of government policy
is to invest in socially-profitable interventions, such as irrigation and drainage,
that increase farmers' control over the crops that can be grown in particular
regions and time periods.
Seasonality also tends to create high premiums to timely performance of such
critical agricultural tasks as plowing, planting, cultivating, and harvesting. Even
though the available labor pool might be more than adequate to provide the
required number of workers per hectare over an entire year for all the crops being
grown, if certain tasks must be performed very quickly at specific times to ensure
maximum yields, important labor bottlenecks might occur in the midst of an
average surplus labor pool. Such bottlenecks can meet with two responses. One is
to work out long-term contracts with laborers that gives them preferential access
to farm employment in the off-season (or access to land to operate as a tenant
farmer, or to credit, etc.) in return for working on the landowner's farm during
the peak seasons [Bardhan (1984)]. Alternatively, because such arrangements tend
to impose high supervisory requirements on the owner's time, they frequently
induce individual farmers to mechanize specific tasks- plowing or harvesting-
even when much rural unemployment exists over the course of the year. In such
circumstances, a tractor that pays for itself in both private and social terms by
timely plowing also has a very low marginal cost of operation for other tasks as
well, and labor displacement can be much more widespread than would be
indicated by the removal of the plowing bottleneck alone.
296 C.P. Timmer
Two features of seasonality are important in designing agricultural policy.
First, seasonal aspects of agricultural production frequently constrain yields
because of input bottlenecks. Labor (and its supervision) is most often the
constraining factor, but fertilizer, seeds, credit, or irrigation water supplies must
also be available in highly-specific time periods. When fertilizer reaches the
village godown a month after the proper application time, it might as well not
have arrived at all. Government authorities responsible for the management of
agricultural input supply distribution are frequently unaware of or insensitive to
the extreme importance of timely input availability. Suppliers whose incomes
depend on providing inputs to farmers when and where needed are much more
responsive to shifts in weather, cropping patterns, and new technologies than are
agencies trying to allocate inputs within the guidelines of five-year plans and
supplies available from a planned industrial sector. Modern agriculture that uses
industrial inputs as the basis for high yields is a dynamic enterprise quite unlike
factories. Input and output markets must function efficiently, reacting to weather
changes, alterations in cropping patterns, and technical change if production is to
grow rapidly. Centrally planned allocations of industrial products to the agricul-
tural sector are almost never in the right place at the right time, or even the right
product.
Second, there are often very high private economic returns to eliminating
seasonal bottlenecks in production. When these private returns are at least partly
generated by higher and more stable yields of agricultural products, society is
also likely to gain. But if the private gains come from displacing hired labor that
has few alternative production opportunities, the social gains might be small or
even negative. The seasonal dimensions to agricultural production complicate the
planning process considerably. Most agricultural data are published on an annual
basis, and there is an inevitable tendency to think about the sector in terms of the
same annual growth performance criteria that are used to evaluate the steel or
cotton textile industries. Such an annual approach hides two important roles for
government analysis and intervention: in the appropriate provision of inputs
when and where they are needed, and in the full analysis of the social impact on
agricultural production of private investments to reduce seasonal bottlenecks.

3. 2.2. Geographical dispersion

Agriculture is the only major sector that uses the land surface as an essential
input into its production function. Like seasonality, this widespread use of land is
due to the largesse of nature. It is almost always cheaper to let farms capture the
free solar energy and rain than it is to stack a hundred stories of hydroponic
"fields" on top of each other and provide the light, nutrients, and water from
industrial sources. This wide geographical dispersion of agricultural production
has an important economic consequence. Transportation becomes essential if any
Ch. 8: The Agricultural Transformation 297
output is going to leave the farm for consumption by others or if inputs, such as
modern seeds, fertilizer, pesticides, or machinery, are to be used on the farm to
raise output.
In combination, seasonality and geographical dispersion create the need for a
marketing system that can store the product from a short harvest period to the
much longer period of desired consumption and can move the commodity from
the farm where it was grown to the many households where it will be consumed.
Both of these functions require that the commodity change hands and that
exchange of ownership take place. This transaction can happen only when both
parties agree on the terms of the exchange or the price for the commodity at the
point of sale. In socialist economies the terms of exchange are often set by the
state. But all other marketing services must still be provided if the food grown by
farmers is to be eaten by consumers.
The necessary growth of marketing services is an often overlooked component
of the agricultural transformation. As Kuznets (1966) pointed out, farmers are
caught in a double squeeze by Engel's Law. The income elasticity for overall food
expenditures is less than one, implying a declining share of national income for
agriculture if commodity prices are stable. But a rising share of the consumers'
food expenditure is devoted to marketing costs, and so farmers receive a
declining share of food expenditures, thus compounding the decline in their share
of national income. As discussed below, technical change has proceeded so
rapidly in agriculture in the past century that farm commodity prices have tended
to fall relative to prices for other goods and services produced by growing
economies. Technical change is also a major factor explaining the rapidly falling
share of national income captured by agriculture directly.

3.2.3. R i s k and uncertainty

Farmers the world over talk primarily about two topics: the weather and prices.
On these two variables ride the rewards for the whole year's effort in farming. A
failed monsoon, a flood, or a hailstorm can wipe out the crop. A bumper harvest
can cause large losses if the price falls too low. No other industry, even
construction or tourism, is so dependent on the whims of nature and volatile
markets to bring in a profit on the investment of time and money that goes into
farming. Farmers who repeatedly make good decisions in the context of rapid
changes in their economic environment tend to survive and thrive. Those who do
not frequently fail; they move to urban areas in search of jobs or become
impoverished landless laborers dependent on the rural economy for their incomes
and access to food. Socialist-managed agricultures can cushion much of the
welfare shock to individuals by sharing risks, but the importance of rapid and
effective decision-making remains as the key to dynamic efficiency in agricultural
systems.
298 C.P. Tirnmer

The fact that weather is uncertain causes farmers to behave differently than
they would if weather were always known. This general uncertainty usually leads
farmers to choose crops that will resist weather extremes, particular varieties of
crops that are more tolerant of weather variations, and lower levels of inputs than
would be optimal in a certain world due to the risk of losing the investment
altogether. Equally important, farmers' reactions to weather variations as they
actually occur also have aggregate consequences [Roumasset, Boussard and Singh
(1979)]. A late monsoon might cause millet instead of wheat to be planted, good
rains might permit a second or third rice crop, and high temperatures and
humidity can lead to serious pest and disease problems that force farmers to
change crop rotations. Each adjustment by farmers can spill over into rural labor
markets, causing serious shortages if planting must be done suddenly when the
weather breaks or the harvest brought in before a flood. A particularly "dry" dry
season might mean the second crop is not planted or harvested, and an im-
portant, perhaps critical, source of wage income is eliminated for many rural
workers. The reduced crop output might not be the most important consequence
of such a crop failure. A famine could result because of the failed income
opportunities [Sen (1981)].
Fluctuations in aggregate production are magnified at the level of marketings
available for consumption by nonfarm households because farm-household con-
sumption tends to vary somewhat less than production. In years of poor weather,
net marketings decline proportionately more than production. Similarly, in good
years the percentage increase in marketings is usually substantially larger than
the production increases. These wide fluctuations simply add to the difficulty of
stabilizing domestic food prices and provisioning urban areas.
Price uncertainty also adds to the farmer's difficulty in deciding what crops to
grow and how many inputs to use in growing them. Unlike the handful of
manufacturers in large-scale industries, farmers are unable to set their output
prices and later adjust production and inventory levels to meet the price targets.
Unlike consumers, who know with near certainty the price they must pay for a
given quantity and quality of a commodity at the time they buy it, farmers must
make major decisions about purchases of inputs well in advance of knowing what
prices their resulting output will bring. At the time many key farming decisions
are m a d e - t h e allocation of land to various crops, fertilizer applications, hiring
labor for weeding- the farmer can only guess at the prices for the output.
Reducing weather and price uncertainties is an important role for government
interventions. Dams and drainage ditches can reduce the impact of rainfall
variations, disaster insurance can provide a new start even if heavy investments
are wiped out, and research on more adaptable but still high-yielding plant
varieties can reduce the risks of new technology. Similarly, reducing price
uncertainty is a major government role, which can be accomplished with better
price forecasting information, the use of import and export policy to provide a
band of prices within which domestic price formation can take place, or a more
Ch. 8." The Agricultural Transformation 299

aggressive floor and ceiling price policy implemented with a government-operated

buffer stock program. Of course, not all stabilizing efforts are worth their costs,
and some fluctuations are necessary if changes in output are to be accommodated
by changes in demand, even allowing for changes in stock levels. The relative
costs and benefits of commodity price stabilization have been the subject of
extensive theoretical analysis. Price stabilization schemes for world markets
perform poorly in both theory and practice [see, especially, Newbery and Stiglitz
(1981)], but the merits of domestic price stabilization programs that use trade as
well as buffer stocks to achieve their goals depend very much on the local
circumstances of dynamics of supply and demand [Streeten (1986), Timmer
(1986)].

3.3. The farm household as both producer and consumer

Truly subsistence households produce to meet their own consumption needs and
do not need the market for either buying or selling. To such households price
signals are not only irrelevant, they are unseen. Few such households remain in
today's world, not because farm families no longer consume produce from their
own fields, but because most farm families now buy and sell inputs and output in
rural markets. They are aware of and react to market prices in making a wide
variety of household decisions. Most farm households still retain some or most of
their farm production for home consumption, and this role of home consumption
is a further distinguishing feature of the agricultural sector. Few steelworkers or
even textile workers take their products home for household use.
Only under highly restrictive and unrealistic assumptions about the complete-
ness of markets and access of all farm households to them can production and
consumption decisions be analyzed separately [Singh, Squire and Strauss (1985)].
In rural areas of developing countries, the need to make connected production
and consumption decisions within a single household obviously complicates life
for the farm household; the value of additional time spent in food preparation or
tending the children must be balanced against the productivity of an additional
hour weeding the rice, driving the ducks, or tending the home garden. Where it
exists, the opportunity to spend some of that time working for cash on a
neighbor's farm or in a rural wage-labor market places a lower bound on the
value of household-farm time, and the value of leisure ultimately places a limit on
the willingness to work, especially at low-productivity tasks. For households with
inadequate land to grow surplus crops for sale and with limited outside employ-
ment opportunities, however, the marginal value of leisure time might be low
indeed, possibly near zero. Even tiny increments to output can be valuable for
very poor households.
The importance of joint household-farm decision-making also raises complex
questions for analysts in search of ways to organize data and research issues into
300 C.P. Timmer

manageable and comprehensible frameworks for analysis. These complex ques-

tions have recently become the focus of a revived interest in models of household
economies. The "new household economics" provides a powerful perspective on
joint decision-making about food production, food consumption, investment in
human capital, and even fertility and other demographic decisions. By showing
how all these decisions are related to each other because of the time constraint,
and hence to the economic environment surrounding the household, the house-
hold economics models provide analysts with a conceptual understanding of the
complicated lives that rural people live [see Schultz, Chapter 13, in this Handbook,
Evenson (1981), Rosenzweig, Chapter 15 in this Handbook]. At the same time,
most such models grossly simplify the actual complexity of rural household
decision making. The key issue is nearly always the functioning of rural labor
markets because it determines the perception of the opportunity cost of labor in
each household. In a survey of tire theoretical and empirical literature on the
functioning of rural labor markets, Binswanger and Rosenzweig offer the follow-
ing conclusions:
Progress toward a richer, integrated theoretical framework that can deal with
the complexities associated with market failures as well as the determination of
wages and other contractual terms has been hampered by the evolution of
theory along two, mutually inconsistent paths. The rural wage determination
models developed so far assume the complete absence of a land rental or sales
market; that is, they take land distribution as exogenously given. The contrac-
tual choice models, on the other hand, treat the wage rate as exogenously
given, while concentrating on land and credit market transactions; thus they
have little to say about the determination of earnings or employment. The
strength of contractual choice models lies in their clarification of the efficiency
and equity implications of contracts and in their identification of the underly-
ing causes of the market imperfections that lead to the contracts. These models
also suggest the difficulties associated with policy intervention in single-tenancy
or credit markets that is aimed at curing symptoms or apparent deficiencies in
such arrangements. Without this integration of all the major interrelated
markets- land, labor, credit- into a single, coherent rural model, however, we
will be severely handicapped in attempting to predict the consequences of
economic development in the rural sector [Binswanger and Rosenzweig (1981,
pp. 54-55)1.

3.4. What difference does the difference make?

Two important implications flow from the distinctive characteristics of agricul-

ture relative to industry, and both are treated extensively in sections that follow.
First, if agricultural decision-making is in fact based on rational assessments of
highly heterogeneous environments, substantial knowledge of micro environ-
Ch. 8: The Agricultural Transformation 301
ments is necessary to understand the impact of policy interventions or technical
change on the agricultural sector. Designing new technology and fostering its
widespread adoption is primarily a public sector activity because of the relatively
small scale of individual farmers, but the success of any given technical innova-
tion depends on the private decisions of those same multitudinous farmers.
Understanding the source, dynamics, and impact of technical change in agricul-
ture is thus a major part of understanding the agricultural transformation, a
process vastly complicated by the smallness of scale, geographic dispersion, and
heterogeneity of the environment, both economic and ecological, that is char-
acteristic of agriculture in developing countries.
The second important implication of agriculture's distinctiveness is how it
conditions the role of public policy, particularly that other than the design and
implementation of research leading to technical change. The vision dies hard of
agriculture as a resource reservoir to be tapped indiscriminately, without reinvest-
ment or adverse consequences for growth, on behalf of the urban economy.
Although a few countries have a record of sustained progress in agriculture and
concomitant overall economic growth, the list is short. Only eight countries listed
in the Worm Development Report, 1986 have growth rates for agricultural GDP of
3 percent per year or greater for both the 1965-73 and 1973-84 periods, along
with growth rates for total GDP of 4 percent per year or greater for the same t~vo
periods: Kenya, Pakistan, Indonesia, Ivory Coast, Philippines, Thailand, Brazil,
and Mexico. Sri Lanka and Turkey came close; Malaysia would probably have
been included had data been available for the earlier period. Because population
growth in several of these countries is near or more than 3 percent per year, even
these excellent aggregate performances leave the rate of growth per capita at
levels that permit a doubling of incomes in a quarter of a century at best.
It has obviously been difficult to find the right mix of policies to sustain
agricultural growth. Much of the reason traces to a failure of policy-makers to
understand the characteristics of agriculture that make policy design so com-
plicated. They face yet another paradox: the essentially private-sector nature of
agricultural decision-making at the same time that the environment for that
decision-making is heavily dependent on sound government interventions into
agricultural research, rural infrastructure, and market relationships. The distinc-
tive characteristics of agriculture argue that governments intervene into agricul-
tural decision-making at great risk, for they can easily cause farmers to withdraw
from making investments and producing for the market, which are essential to
mobilizing resources for overall economic growth. And yet, intervene they must.
The environment for transforming agriculture is a public good created by wise
but active public intervention.
It is easy to get the mix wrong, even to have the elements backward. Some
governments have tried to dictate farm-level decisions on inputs and outputs
while totally ignoring both the investments in research and infrastructure needed
to create a healthy agriculture and the pricing environment that will mobilize
302 c.P. Timn,er
peasants on behalf of higher productivity. But enough success stories have been
accumulated for some general lessons to be propounded. The dimensions of
successful technical change are discussed next, followed by a review of overall
policies for agricultural development.

4. Transforming agriculture

Agricultural output can increase along a given supply curve or with a shift in the
supply curve to the right. The scope for increasing output along a fixed supply
curve by continuing to raise prices is extremely limited even in fully commercial
and technically advanced farming systems; nearly all long-term growth in crop
and livestock production comes from investment that expands capacity and from
technical change that increases output-input ratios. The importance of prices for
transforming agriculture is not in triggering the short-run response of farmers,
although this is sometimes quite dramatic in situations where severe distortions
are eliminated, but in conditioning the investment climate and expectations of all
decision-makers in the rural economy about the future profitability of activities in
the sector. Positive expectations lead to rapid investment in technical change
when it is available.

4.1. The sources and dynamics of technical change

Technical change is the source of most growth in productivity in the long run,
since continued investment in capital that embodies traditional technology very
quickly faces low marginal returns [Schultz (1964), Hayami and Ruttan (1985)].
As late as the 1920s, most of the agricultural innovations in Europe and the
United States arose on the farm and were gradually diffused by word of mouth
and by agricultural colleges. Such on-farm innovation continues, but the scientific
revolution in agriculture has made the discovery of technical innovations much
more dependent on knowledge and capital investment. Very few farmers even in
the United States have the resources to carry out significant agricultural research
programs, and most such research is conducted by publicly-funded centers for
agricultural research and by a handful of large agribusiness concerns, which are
involved primarily in developing hybrid seed technology, chemical technology
(herbicides and insecticides), and agricultural machinery. 6 The small scale of
operations and limited financial resources of most farms mean that little im-
portant agricultural research is conducted by farmers.

6The revolution in biotechnologymight change the concentrationof agricultural research in the

near future. Numerous small companies, many associated with faculty members of universities, are
engaged in geneticmanipulationof important agriculturalcrops and animals,althoughthe impact on
farm productivityhas not yet been significant.
Ch. 8." The Agricultural Transformation 303
Diffusion of new technology is also a matter of policy concern, especially
because not all farm households have equal access either to the knowledge to use
new technology or to the agricultural and financial resources needed to make it
productive on their own farms. Some inputs are lumpy and cannot be used
efficiently on farms of even average size in many parts of the world. Large-scale
tube-wells and tractors might contribute significantly to higher productivity even
on small farms if institutional arrangements could be found to separate the
service flows that such inputs can provide from the ownership of the assets
themselves.
The evidence suggests that truly profitable innovations spread quickly no
matter what the government does. Wherever the entrepreneurship exists and the
economic environment permits, rental arrangements and tractor-hire services
frequently emerge spontaneously [Goldman and Squire (1982)]. However, the
location-specific nature of much new agricultural technology, especially seed
technology, means that large areas of a country might be bypassed by the
diffusion process unless government research and extension workers are actively
engaged in the on-farm testing and evaluation of new technology. Adapting a
general agricultural technology to a specific seed strain or technique that fits
individual farming environments is a major responsibility of local research and
extension stations.
An important concern of government policy is the impact of technical change
on agricultural employment and rural income distribution. Historical evidence
shows enormous variation in both the short-run and long-run impacts of innova-
tions. The issues cannot be addressed satisfactorily by looking only at an
individual farm or even at the agricultural sector [Scobie and Posada (1978), Hart
(forthcoming), Hayami (1984)]. The primary effect of higher-yielding varieties of
wheat and rice, for example, has probably been on food intake of nonagricultural
workers. In addition, agricultural innovations tend to be embodied in inputs that
must be provided through markets. An increased role for market relationships
might threaten the risk management aspects of established patron-client rela-
tionships and thus have complicated effects on the entire rural economy and
eventually on the urban economy as well.
Most technical change in agriculture involves improvements in the biological
processes by which plants and animals grow and yield output useful to society or
in the mechanical functions that are necessary for the biological processes to
carry on more efficiently than in a natural setting. Primitive agriculture uses
natural biological materials and processes in combination with human labor and
management to bring in a crop or livestock product. Modern agriculture uses
scientific knowledge to reshape the biological materials so that each plant and
animal is more productive, and it increasingly substitutes machines for human
labor.
Biological-chemical innovations, such as hybrid seeds, fertilizers, and pesti-
cides, all tend to be yield-increasing and thus save on land. Mechanical technol-
304 C.P. Timmer

ogy can also have a yield effect when it permits more timely cultivation and an
extension of multiple cropping, cultivation of heavy soils, or the use of water
pumps on dry lands, but most mechanical technology is designed to make
agricultural work less physically burdensome and to save on the amount of labor
needed to produce a unit of output.

4.1.1. A simple model of changing agriculturalproductivity

Productivity in agriculture traditionally is measured in one of two ways: in

output per hectare, or output per agricultural worker. Despite the focus by
agricultural scientists on the former measure, from a welfare perspective the latter
measure is clearly the relevant one. Output per hectare is important only as a
vehicle for raising output per worker. In land-scarce environments facing rapid
population growth and limited absorption of labor by industry, of course, raising
output per hectare might be the only way to raise labor productivity. Most
analyses treat both measures, and the model here, derived from Hayami and
Ruttan (1985), does as well.
Figure 8.2 plots agricultural output per unit of land area in logarithmic units
on the vertical axis. Hayami and Ruttan convert agricultural output into wheat
units, Bairoch (1975) uses "direct" calories, and the World Bank reports agricul-
tural value-added and contribution to GDP in its annual Worm Development
Report. For the purposes of this discussion, the vertical axis is simply crop yields
per hectare.
The horizontal axis measures agricultural output per worker on a logarithmic
scale. Most econometric analyses of changes in agricultural productivity use
output per worker as the dependent variable, and the workforce is traditionally
defined as male workers in agriculture for the reason, indeed a rather lame one,
that women play very different roles in agricultural production in different parts
of the world and national statistical offices are not very consistent in how they
treat the matter. For the purposes here, the workforce is measured as the entire
economically active agricultural population. Because both axes are measured in
logarithms, 45 ° lines trace out constant ratios of land per worker. Productivity
changes over time can be traced out by connecting the coordinates at the
beginning and the end. Figure 8.2 illustrates a variety of possibilities.
From the point of view of improving the welfare of rural workers, only
movements to the right - toward higher output per worker - can help. Even then,
the distribution of output among workers, landowners, and owners of other
factors of production will determine whether or not the higher productivity has
widespread welfare effects. Straight movements to the right are likely to be
relatively rare. As Figure 8.2 notes, such movements imply a declining agricultur-
al workforce and no changes in yields, normally in conjunction with new
Ch. 8: The Agricultural Transformation 305

Y
A population growth matched by
increased yields (through higher
labor inputs and technical
change), but no improvement
in rural living standards
population growth A
- - constant
faster than technical
change in raising
yields, with igration from
deterioration / agriculture, or new
in rural living s t a n - ~ / land opened for culti-
dards
~n vation with technical
change raising crop yields;
ing standards depend on
E t of rural landlessness
al wages

/
population growth faster
than land opening / / declining agricultural work
or tech- force with no changes in
nical change, / yields (but new mechanical
plus problems with soil / technology needed to maintain
erosion or bad p o l i c i e s / output with fewer workers)
leading to lower o u t - /
put and deterio-
rating rural /
,< livingstan- / /
d a r d s /
opening new land on the
frontier with lower quality
soils and no technical change

/ Y
L
Agricultural output per worker (logarithmic scale)

Figure 8.2. Various possibilities for changing land and labor productivities in agriculture.

mechanical technology to maintain levels of output with fewer workers per

hectare.
What might have been a typical path while new continents were being
colonized in the seventeenth and eighteenth centuries, but which is virtually
unseen now, is rising labor productivity with falling land productivity. Lower-
quality soils, distance from input and output markets, and low demand for
technical innovations do not prevent extensification of agriculture at the frontier
from raising living standards- and hence inducing migration- even while yields
are falling.
The far more common pattern is a movement upward to the right, as the
productivity of both land and labor increases. If the movement is exactly in a 45 °
306 C.P. Timmer

direction, agricultural land per worker remains constant, and yields must rise if
labor productivity is to rise. A striking difference between currently developed
countries and poor countries is their paths relative to this 45 ° line. As Bairoch
(1975) noted, and evidence from Hayami and Ruttan (1985) to be presented
shortly indicates, most developed countries increased land per worker even in the
early stages of their development, whereas only a few less-developed countries are
able to do so. The reasons are obvious. Either new lands must be opened faster
than population growth, or out-migration from agriculture must proceed fast
enough to cause an absolute reduction in the agricultural work force. Only a
handful of countries can meet either of these conditions.
In countries with very limited agricultural land resources and rapid rates of
growth in population, often the best that could be done since the early 1960s was
to maintain constant labor productivity by increasing crop yields at the same
pace as expansion of the rural workforce. This combination generates a vertical
growth path, which might alternatively be described as running fast technologi-
cally to stand still economically. But some countries have not even done this well.
Their populations have grown faster than the pace of technical change on farms,
and their productivity path is an arrow up and to the left, reflecting lower
standards of living in rural.areas.
The most dismal situation, however, is movement downward to the left,
reflecting deterioration in both measures of agricultural productivity. Output per
hectare and output per worker fall in such circumstances. The reasons might be
extremely rapid growth in population with expansion into ecologically unstable
agricultural areas, or such bad policies that farmers retreat from even the
technology that they used previously. None of the countries in Hayami and
Ruttan's analysis fits this last pattern, but no countries from sub-Saharan tropical
Africa were in their sample.

4.1.2. The historical record

Hayami and Ruttan (1985) assembled evidence for changes in productivity of

agricultural labor between 1960 and 1980 (see Figure 8.3). Three patterns are
obvious. Nearly all countries in their sample showed improvement in both
dimensions of productivity - only Bangladesh had a decline in labor productivity,
and only Chile had a decline in land productivity. Most developed countries had
faster increases in labor productivity than in land productivity, thus presenting
patterns of change "flatter" than the 45 ° lines of constant area per worker - hence
farm sizes had increased. Most developing countries had patterns of productivity
change steeper than the 45 ° lines, implying decreased area per worker and
smaller farm size.
Hayami and Ruttan see three basic patterns of agricultural development in this
historical record (see Figure 8.4). The Asian path requires strongly rising land
Ch. 8: The Agricultural Transformation 307

(Y/A
LABOR PER UNIT OF AGRICULTURAL OUTPUT LOG.SCALE)
(L/Y) i ,4

//%',
/ // ,

_J
<
Ta ," . -"
(D ,'" Ne ,'" "J
g ,,"" Ja ,,,,J -'" ~
<
ra 10 E. " " / Su / •'" -Be -J ~
Z
5
Ba "'Ph
<
}-
uJ
1-
re-
~,~,"~-~ Sp Ir " US
F-

2
F-
0
"'MeSy Ve oq'"" <
. /'~e/l,4 Ch ..~7,Ar +<o~,~':'" ~t-
tr

b /, uJ
(.J
rr" "" Par Li " -"
"'" ~'~_f'"
"
Aus a
Z
(.9
< 0_1

I I , (Y/L)
10 100 1000

Key: AGRICULTURAL OUTPUT PER MALE WORKER (LOG.SCALE) (A/Y)

Argentina Ar Finland Fi Mexico Me Sri Lanka Sr

Australia Aus France Fr Netherlands Ne Surinam Su
Austria Au Germany, F.R. Ge New Zealand NZ Sweden Swe
Bangladesh Ba Greece Gr Norway No Switzerland Swi
Belgium Be India In Pakistan Pak Syria Sy
(& Luxembourg) Ireland Ir Paraguay Par Taiwan Ta
Brazil Br Israel Is Peru Pe Turkey Tu
Canada Ca Italy It Philippines Ph United Kingdom UK
Chile Ch Japan Ja Portugal Po United States USA
Colombia Co Libya Li South Africa SA Venezuela Ve
Denmark De Mauritius Ma Spain Sp Yugoslavia Yu
Egypt Eg

F i g u r e 8.3. I n t e r n a t i o n a l c o m p a r i s o n of labor and land productivities in agriculture, the 1960 data

p o i n t s c o n n e c t e d to the 1980 points by arrows. Source: H a y a m i and R u t t a n (1985, p, 121).
308 C.P. Timmer

Technology
front;eronthe
rnetaproductiofuncti
n on
Aconstan
t
~ ~ a n path ~ L
t.o

E
Europeanpath
&
o,

\
~8465~/ Newc°ntinenta'path

Sub-Saharan
TropicalAfrica

/ Y

LogarithmiScal
c e
Figure 8.4. Patternsof changein agriculturalproductivity.

productivity in early stages to cope with small farm size and rapid growth in
population, but eventually labor productivity grows rapidly as the rest of the
economy absorbs rural workers and raises wages. This is the "Korea-
Taiwan-Japan" model, but Pakistan, Philippines, Indonesia, Sri Lanka, and even
Egypt might also have access to this path.
At the other extreme, the path of productivity change in the newly opened
continents with surplus land is almost uniformly in the direction of higher labor
productivity, and this has been true in the United States, Canada, and Australia
since the mid-nineteenth century. It was only after the higher commodity prices
caused by the world food crisis in 1973-74 that land productivity rose faster than
labor productivity in the newly settled continental areas.
Ch. 8: The Agricultural Transformation 309

The European path falls nicely between the land-scarce and the land-surplus
paths. In Figure 8.3, many countries are clustered here, and Hayami and Ruttan
note that the paths for Denmark and the United Kingdom create an envelope
that contains the entire European experience:
Denmark, which has remained relatively specialized in agricultural production
among European countries, has attained a high labor productivity in agricul-
ture by increasing output per unit of land. In contrast, the United Kingdom,
which initiated the Industrial Revolution, has attained a relatively high level of
agricultural efficiency mainly by enlarging agricultural land area per worker in
response to the absorption of labor in nonagricultural occupations. France,
which traditionally followed an agrarian policy designed to protect the peasant
family farm (la petite exploitation familiale) from external competition and
internal social change, achieved higher output per hectare than the United
Kingdom but slower growth in output per worker than either the United
Kingdom or Denmark until the formation of the European Economic Com-
munity (EEC). Since 1960, stimulated by increased demand for the protected
EEC market, output and productivity of French agriculture have expanded at a
very rapid rate [Hayami and Ruttan (1985, p. 130)].
Connecting the most advanced countries along each productivity path reveals
the technology frontier (see Figure 8.4). Hayami and Ruttan describe this frontier
as a metaproduction function, arguing that tlae underlying technologies that
describe it are potentially available to all countries at a point in time. The
technology actually developed and disseminated depends on relative factor
scarcities. "Induced innovation" leads scientists to develop mechanical technolo-
gies to raise labor productivity in labor-scarce societies (for example, the new
continents), whereas scientists in land-scarce societies, such as those in Asia,
develop biological-chemical technologies to raise output per hectare. The poten-
tial of induced innovation to solve the agricultural problems of the currently
developing countries will be discussed below. First, some special problems of the
African experience must be placed in the context of productivity.
In addition to the three stylized productivity paths generated by the sample of
countries in the analysis by Hayami and Ruttan, Figure 8.4 shows the growth
path in agricultural productivity for Africa from 1965 to 1984, calculated from
recent World Bank data. The definitions used are not identical to those used by
Hayami and Ruttan, but the pattern shown is robust and perplexing. Between
1965 and 1973, Africa's productivity performance was very much like that of new
continental areas: slow growth in land productivity and more rapid growth in
labor productivity. Because of rapid growth in population, this increase in labor
productivity reflected significant progress in increasing overall agricultural out-
put.
310 C.P. Timmer

Something quite unique in historical experience occurred in Africa between

1973 and 1984. For the entire continent between the Sahara and South Africa,
the productivity of both land and labor declined. In Hayami and Ruttan's
sample, only one country experienced a decline in each measure separately, and
none declined in both. In Africa, an entire continent made up of more than thirty
countries suffered a decline in both (although a few individual countries saw
growth in both measures). The reasons for this startlingly poor performance are
only beginning to be analyzed and understood, but it is virtually certain that a
complex combination of bad weather, inadequate and inappropriate agricultural
technology, and poor economic policies are to blame. The interplay between
technology and agricultural development policy is the key issue here. Some
analysts (but not Hayami and Ruttan) have seen induced innovation as an
automatic market solution to a country's development problems. The African
experience shows clearly that such is not the case.
4.1.3. Sources of productioity differences

Differences in agricultural productivity can stem from a variety of factors:

different endowment of internal resources, such as land and livestock; different
use of technical inputs, such as fertilizer and mechanical power; different
investment in human capital through general and technical education; and
different size of farms, which might generate economies or diseconomies of scale.
Table 8.4 shows examples from the effort by Hayami and Ruttan (1985) to
explain differences in productivity of agricultural labor according to differences
in these factors' contributions to output. The contribution of each factor to
productivity is based on econometric analysis of the same data set that generated
Figure 8.3. Hayami and Ruttan estimated a production function by pooling their
cross-section data for 44 countries for three time periods (1960, 1970, and 1980)
and used the Cobb-Douglas output elasticities to account for differences in labor
productivity. The results contradict Bairoch's pessimism about the potential of
developing countries to raise their labor productivity in agriculture in the context
of diminishing land per worker. In the three low-income countries in Table
8.4 - India, Philippines, and Peru - roughly half the difference in labor productiv-
ity relative to that of the United States was due to differential use of technical
inputs and investment in human capital (general and technical education). Even
the internal resource constraints are not completely binding because investment
in livestock has an elasticity of output about double that of land. Scale economies
are significant in Europe and newly opened continents but not on the small-scale
farms characteristic of Asia and Africa.
Hayami and Ruttan conclude this part of their analysis on a positive note:
The perspective implied by the results of this analysis for agricultural develop-
ment in the less developed countries is essentially encouraging. It is clear that
Ch. 8: The Agricultural Transformation 31i

o£.

,-q
V 'I' ~'''V ,I

g~

o
,--4
o

~'-~
i

"~. ~ o

,..o ~ o

Ce
~ ~ , ~ ~ ~ ~
.,~ e
.,.~

.=_
O O O 0 0 0 0 0 0 0 0 0 O 0 0 O O

e~ ~00
P. ,"'4

:5 6'9

L~
.~
e..

o L~ o ~ & ~ = =
",3
<

&
312 C.P. Timmer

agricultural output per worker in the LDCs, especially the poorest ones, can be
increased by several multiples by adequate investments in education, research,
and the supply of modern technical inputs, even if land area per worker
continues to decline because of growing population pressure in the rural
sector... It is especially encouraging to find that the agricultural production
function of the LDCs is neutral with respect to scale. This implies that the
low-income LDCs will not be too severely handicapped by the declines in the
land-man ratio and farm size, relative to the older developed countries, at least
over the next decade or two [Hayami and Ruttan (1985, p. 157)].

4.1.4. Problems with technical change

Biological-chemical innovations were discovered and introduced in land-scarce,

labor-abundant societies, such as Japan and Western Europe, whereas mechani-
cal innovations were developed and used in land-rich, labor-scarce societies, such
as the United States, Canada, and Australia. Such induced innovation suggests
that each society develops an agricultural technology appropriate to its resource
endowments and agricultural needs. This process might not continue to yield
appropriate results, however, in the context of a much more interdependent
international agricultural system. Perhaps more troublesome, the examples
Hayami and Ruttan used to illustrate the relevance of their iriduced innovation
hypothesis are all large countries, which are easily able to justify the overhead
expenses of a modern agricultural research and extension system. The means to
develop appropriate technology for small poor countries, such as Chad, Haiti, or
even Laos, remains to be seen.
Because most new agricultural technology is embodied in a physical input a
bag of fertilizer, a new seed, a tractor, or an irrigation p u m p - it can be effective
in a farmer's field only if a purchase (or rental arrangement) is made. Several
consequences flow from this simple fact. For small farmers to participate in the
benefits of technical change, not only must it be workable on their small farms
(combines, for instance, usually are not), but they must also be able to purchase
the input that carries the new technology. If a new seed-fertilizer package has a
200 percent rate of return, even borrowing from a village moneylender at 10
percent per month might be profitable. But for the full benefits of modern
technology to reach small farmers, it might be essential that formal rural credit
systems be accessible to the farm household with only half a hectare or less. 7
Equally important, if new technology is embodied in inputs, a marketing and
distribution system is necessary for farmers actually to be able to purchase the
inputs. Many traditional agricultural societies have a long history of small-scale

VThe dangers of subsidizing this credit are now well recognized. See Adams and Graham (1981)
and Gonzalez-Vega(1977).
Ch. 8: The Agricultural Transformation 313

marketing of surplus output to urban regions in exchange for consumer items,

such as cloth, kerosene, or pots and pans, needed by farm households. There is
no similar experience with large-scale movements of inputs, such as fertilizer or
modern seeds, to those same dispersed farm households. The embodied nature of
agricultural technology means that farmers cannot just be told about it. The
marketing system must also deliver the inputs when needed.
A further characteristic of embodied agricultural technology is that comple-
mentary fixed capital investments are often required to achieve the maximum
benefits from the innovation. Usually this investment takes the form of better
water control, land-leveling, and drainage. Sometimes much better control of
seed bed preparation or more sensitive and faster harvesting techniques to avoid
shattering and other harvesting losses is also needed; these might require tractors
with modern implements or, for harvesting, combines or threshers. Shorter-
maturity cereal varieties often are ready to harvest while the rainy season is still
under way and solar drying is difficult or impossible. In such cases mechanical
dryers and added storage capacity are essential.

4.2. Unresolved issues

Many questions are unresolved or still contentious in the agricultural develop-

ment profession. Most involve the relationship between technical change and the
policy environment needed to make it effective. The problem in defining the
relationship is partly due to the lack of understanding of household decision-
making, especially in environments where linked contracts among labor, land,
and credit complicate analytical models. Moreover, economists have had a
difficult time modeling the interface between micro decisions and macro out-
comes because neither micro competitive models nor macro policy models
provide an adequate basis for analysis of decision-making in this grey area.
Serious disputes over the long-run sustainability of modern, input-intensive
agriculture are unresolved. The following discussion does not answer these
questions but does frame them in the context of the previous discussion of
technical change and the discussion yet to come of agricultural development
strategies.

4.2.1. Evolution in thought

Research in the 1970s into the links between technical change and the decision-
making environment at the farm level led to three quite significant changes in
thinking about agriculture and development. First, and no doubt the most
important for the long run, agricultural decision-makers- farmers and
traders began to be thought of as an integral part of the rest of the economy,
314 C.P. Timmer

connected to it by rational decision-making in the face of new technologies or

income and price changes. Acceptance of this principle by economic modelers
and policy-makers led to a fundamental shift in attitudes about how agriculture
should be treated in the development process- not as an isolated appendage but
as a key component essential to the health of the overall organism.
Second, new importance was given to developing and choosing appropriate
technologies, whether in agriculture directly, in processing, or in the industrial
sector. If millions of farmers, or thousands of traders and small industrialists,
were making rational decisions about investment, the nature of the technology set
facing them would be at least as critical as the set of prices in determining the
consequences of economic growth for employment and income distribution. In
addition, farm-level decision-making was obviously conditioned by farm-level
constraints and opportunity costs for household resources, especially unskilled
labor. Technical packages that were inappropriate in the face of those
constraints - especially packages imported largely intact from Western agricultur-
al systems- were not adopted, and agricultural development failed to take place.
Sparked by the world food crisis in 1973-74, but guided by this earlier under-
standing of the importance of fitting technology to field-level conditions, nearly
all the international centers for agricultural research devoted a significant share
of their budgets to discovering the nature of farm-level constraints on the
adoption of new technology and to developing specific crops as well as entire
farming systems that dealt more effectively with these constraints [see IRRI
(1978), CIMMYT (1984)].
The third major change in thinking in the 1970s also was sparked by the world
food crisis, but it similarly had its roots in the new understanding of micro
decision-making and the key role of technical change in agriculture. Early
agricultural development strategies were aimed at providing resources for urban
industry by helping "early adopters" of new farm technology. The consequences
of this strategy for in, ome distribution and rural welfare prompted the develop-
ment of the "basic needs" movement and efforts to promote "growth with
equity" [Chenery et al. (1974)]. In some sense this concern for equity was an
almost inevitable consequence of renewed emphasis on improved technology,
which better-off farmers tended to adopt earlier, and on better price incentives
for higher output, which clearly benefited larger farmers with a higher proportion
of marketed output. Indeed, the focus on price incentives to achieve production
results often overlooked the potentially serious consequences for poor consumers
who were net purchasers of food, many of whom were very small farmers or
landless laborers in the countryside [Timmer (1979)]. Out of the concern for
promoting equity and meeting basic needs came a major revival in interest in
demand analysis. Although Indian planners had used income elasticities for
staple foods disaggregated by income class in the earliest five-year plans, no one
had attempted the empirical disaggregation of price elasticities by income class
Ch. 8: The Agricultural Transformation 315
until after the world food crisis in the mid-1970s. Because food prices can change
from year to year by much larger relative amounts than incomes per capita,
knowing such disaggregated price elasticities became critical to judging the
welfare impact of the wide price fluctuations characteristic of the 1970s. ~ As
Eicher and Staatz (1984) summarized it, the 1970s were primarily a period when
agricultural economists renewed their microeconomic roots, and the decade
brought forth a rich, and often confusing, harvest of new empirical evidence. It
remains for this evidence to be synthesized in the 1980s to provide better
understanding of the development process and agriculture's role in it.

4.2.2. Farm decision-making

One of the difficulties in understanding decision-making in agriculture is that

farmers face remarkably diverse ecological and economic settings. Corporate
businessmen in California or Sao Paolo make their living from agriculture, but so
too do near-subsistence peasants in India or Guatemala. Despite differences in
scale of operation and location, however, private agriculture is a markedly
homogeneous industry in the kinds of decisions that must be made day in and
day out and in the uncertainties that surround those decisions. The corporate
soybean farm in Sao Paolo or the rice farm in California has more in common
with the wheat-growing peasant operation in the Punjab than with U.S. Steel or
Volkswagen of Brazil. Much of the daily work done on these farms is at the
initiative of the individual workers, and the incentives they face to perform this
work in a timely and careful fashion strongly influence the quality and quantity
of agricultural output.
Modeling farm decision-making is relatively simple under two extreme sets of
assumptions: if the household is entirely self-sufficient and faces no markets, and
if the household faces a complete set of perfect markets. Even these simple
settings can be complicated by risk and uncertainty, by bargaining among
members of the household over access to resources and output, and by non-
pecuniary externalities in the welfare function as reflected by investment in the
maintenance of a "moral economy" [Hart (forthcoming), Jones (1986), Scott
(1976)]. As emphasized previously by Binswanger and Rosenzweig, however, it is
the reality of interlocking land, labor, and credit markets and the explicitly
limited access of some households to some markets that challenge model-builders
who hope to capture the complexity of rural life and thereby be able to predict
the outcome of changes in policy and technology or the commercialization of
rural transactions. At a conference organized by Binswanger and Rosenzweig

8An attempt to generalizefrom early empiricalresults about the relationshipbetween income level
and the magnitude of the pure substitution term in the Slutsky matrix is in Timmer (1981); a review
of the literature on disaggregated demand parameters is in Waterfield (1985) and Alderman (1986).
316 C.P. Timmer
(1981) to examine the empirical record on these issues, what was striking was the
sheer diversity of arrangements at the micro level. While this is not cause to reject
model-building as an approach to understanding decision-making of rural
households, it does caution against making general predictions without specific
empirical foundation to the model. Building such empirical foundations to
household decision-making models has occupied a substantial part of the agricul-
tural economics profession since the early 1970s [Singh, Squire and Strauss
(1985)]. The results show clearly the merits of treating the household as a
combined producing and consuming unit, but, as Binswanger and Rosenzweig
note, the dynamic aspects of the household's interactions with its environment
are only beginning to be revealed:
Explanations of the long-term changes associated with development must be
found, ultimately, in models that explicitly treat the reproductive and techno-
logical behavior that leads.to the long-term evolution of supply and demand.
Attention has recently turned to the study of decisions that have long-term
consequences-decisions about human capital investment, fertility, health,
technical change, and agricultural intensification. Such decisions, however, are
themselves conditioned by the outcomes and institutional arrangements in
rural factor markets. The integration of market and household behavioral
models within an explicit dynamic framework enveloping all sectors of an
economy has yet to come [Binswanger and Rosenzweig (1981, pp. 55-56)]. 9

4.2.3. Micro-macro links and structural change

Trying to explain the declining share of agriculture during the process of

structural transformation by analyzing the decision-making of farm households is
a bit like trying to explain evolution by studying the molecular biology of plants
and animals. The explanation for evolution, of course, must ultimately have its
basis in molecular biology, and, likewise, structural transformation must be based
on micro decision-makers. But our capacity to move from one level to the other is
very limited. Aggregation of micro outcomes does not trace out macro growth
paths very well, primarily because of the difficulty in specifying investment
functions and the introduction of technical change. Even the reverse causation,
where macro settings influence micro decision-makers, has only recently been
incorporated into models of agricultural sector performance [Chilchilnisky and
Taylor (1980), Schuh (1976), Taylor (1980), and Timmer, Falcon and Pearson
(1983)].

9Such all integration obviously takes us outside the realm of this chapter into the other topics
treated in this Handbook. In particular, see the chapters by Bardhan (Chapter 3), Behrman and
Deolalikar (Chapter 14), Bell (Chapter 16), Birdsall (Chapter 12), Rosenzweig(Chapter 15), T. Paul
Schultz (Chapter 13), Sen (Chapter 1), Stiglitz(Chapter 5), and Williamson(Chapter 11).
Ch. 8." The Agricultural Transformation 317
One specific attempt to measure the impact of macro prices on structural
change in agriculture was reported in Timmer (1984). The share of agriculture in
G D P was the dependent variable in the model, which used the same variables for
income and population size as those used by Chenery and Syrquin (1975).
However, the rural-urban terms of trade were also added as an explanatory
variable. The terms of trade as well as the per capita income variable were
explained by a simple four-equation, recursive structural model, with income per
capita depending on lagged income per capita, lagged "real" foreign exchange
rate, lagged investment, and the share of oil imports in GDP. The real exchange
rate (in purchasing power parity) was then explained by lagged income per
capita, the current account balance, real cereal prices in world markets, and the
oil import share. In the next step, investment as a share of G D P was explained by
current income per capita, the foreign exchange rate, the current account balance,
and the oil import share. Finally, the rural-urban terms of trade were determined
by the foreign exchange rate, real prices for noncereal agricultural products in
world markets, real cereal prices in world markets, and the oil import share. It
was then possible to estimate a Chenery-Syrquin equation with agricultural share
of G D P as the dependent variable and predicted values of each of these
dependent variables in the structural model as independent variables in the
model of structural change.
The model was estimated for seven countries in the Asia-Pacific region for the
years 1960 to 1980. x° All variables in all equations were significant and of the
right sign, confirming the logic of the structural model. More interesting was
the separate importance of the foreign exchange rate and the oil import share in
determining the rural-urban terms of trade in these countries, especially because
the oil import share was also a highly important variable in explaining the foreign
exchange rate itself. The two oil price shocks in the 1970s thus opened a window
of opportunity to trace the effects of a major macro perturbation as it rippled
through the economy, including the agricultural economy. The results of estimat-
ing the model and simulating changes in the oil price (holding import or export
volumes constant in the short run) confirmed the notion that the agricultural
sector is strongly influenced by variations in macro prices. The positive effect of
currency devaluations on the rural-urban terms of trade was confirmation that
rural goods and services tend to be more "tradable" than urban goods and
services. No logic requires this result, of course, and many economists would tend
to think the opposite-bulky, low-value agricultural commodities are naturally
protected by high marketing costs and should therefore be less tradable than
urban industrial goods. This view fails to reflect two considerations. First, strong

1°The countries were Indonesia, South Korea, Malaysia, Mexico, Philippines, Sri Lanka, and
Thailand. They were chosen to have a balance among oil importers, oil exporters, and a country
approximately self-sufficient.An oil exporter naturally has a negative oil import share in the model.
318 c.P. Timmer

substitutions are possible among agricultural commodities in both production

and consumption. Sweet potatoes might not be tradable directly, but if they
compete for resources and customers with rice that is tradable, the sweet potato
economy will behave as if they were a tradable commodity. Second, extensive
protection is provided to the urban industrial sector in most developing coun-
tries, protection that has the effect of converting the sector from tradable to
nontradable. Consequently, these empirical results showed that the strong tend-
ency of developing countries to maintain overvalued exchange rates (even for
their existing degree of industrial protection) not only impedes efficient resource
allocation and rapid growth (because of the importance of the foreign exchange
rate in the per capita income equation), but also significantly biases income
distribution against the rural sector.
Oil imports force countries to remove some of that bias. As pressures build to
create incentives to export in order to pay for the oil imports, the rural sector
receives improved terms of trade since it produces many of those exportable
goods. The effect is symmetrical for oil exporters. As oil prices rise for exporting
countries, the terms of trade deteriorate for their agricultural sectors. Since the oil
share is also a significant factor in exchange rate determination, the ultimate
impact of oil prices is even larger. Consequently, "Dutch D i s e a s e " - t h e decline
in employment and output in labor-intensive export sectors in countries experi-
encing a boom in resource prices-is at least as much a rural problem as an
urban industrial one. The agricultural difficulties of Nigeria, Venezuela, Mexico,
and Indonesia (until 1978, when macroeconomic management changed in order
to cope with the problems created by high oil prices) can be seen to have
common macroeconomic roots. Likewise, the increased supplies of agricultural
commodities in world markets and the reduced demand for them in the early
1980s must also have at least part of their explanation in sectoral responses to the
oil price changes of the 1970s. As oil prices fall in the 1980s, some of the pressure
to export agricultural commodities should be reduced (although servicing the
debt incurred while oil prices were high attenuates this effect to some extent). As
a consequence, agricultural commodity prices in world markets should recover
somewhat relative to oil prices.
This type of analysis - conducted within a general-equilibrium perspective even
if not within a formal, computable general-equilibrium model-reinforces the
early and partial results obtained in the 1970s from analysis of choice of
technique in production in developing countries: macroeconomic policies, espe-
cially with respect to macro prices-wage rates, interest rates, and foreign
exchange rates - significantly influence these choices and consequent employment
and output levels, as well as income distribution. The link from macro policy to
agriculture is quite strong. In the other direction, the general-equilibrium conse-
quences of agricultural adjustments to shifts in these policies seem to be quite
Ch. 8: The Agricultural Transformation 319

significant, but these are not yet understood in other than the roughest theoretical
and empirical way.

4.2.4. Resources for growth and sustainabili(y

Parallel to the incorporation of agriculture into macroeconomic and general-equi-

librium analysis has been the growing acceptance of what was a highly controver-
sial and widely denied argument in the 1970s: that rapid economic growth with
broad participation of the entire population for sustained periods of time was
necessary for a country to deal successfully with widespread poverty and hunger.
The desirability of such growth was seldom questioned; the controversy was over
the adequacy of the world's resource base to sustain such rapid growth for more
than a handful of special cases (the " G a n g of Four": Singapore, T-aiwan, South
Korea, and Hong Kong). The more radical segment of the basic needs movement
adopted a small is beautiful philosophy that called for substantial changes in the
lifestyles of the rich in order that the poor could share more equitably in a
limited standard of living for the entire world [Schumacher (1975), Lappe (1971)].
Although this perspective has certainly not disappeared, a decade-long decline
since the mid-1970s in basic food prices on world markets, to historic lows in real
terms, and the monthly efforts by OPEC in the mid-1980s to prop up oil prices
against a seemingly inexorable market determined to lower them, have changed
the nature of the debate. The issue is not whether the global resources are
available for economic growth, but whether they can be managed appropriately
to generate and sustain that growth. The record after the world food crisis in
1973-74 reveals that farmers and societies respond vigorously to apparent food
shortages, whether in response to prices in world markets or to a perceived
vulnerability to uncertain external market supplies. Technical change in agricul-
ture, at least in the United States and Western Europe, has accelerated in the
1980s after stagnating in the 1970s, and it is difficult not to see this as a form of
induced innovation in the Hayami-Ruttan sense.
The sustainability of this technical change has been repeatedly challenged,
especially after the first oil shock in the early 1970s. Lester Brown has been one
of the most articulate and influential of these challengers, and his State q( the
World, 1984 contains a succinct statement of the concern:
Although the economic crisis of the eighties is exacerbated by economic
mismanagement, its roots lie in the depletion of resources, both nonrenewable
and renewable. During the fifties and sixties the world economy steadily
boosted its use of oil, a finite resource, putting it on a path that by definition
was not sustainable over the long run. The depletion of oil reserves, and its
effect on world oil prices, is the most immediate threat to world economic
320 C.P. Timmer

stability, but the depletion of soil resources by erosion might be the most
serious long-term threat. The unprecedented doubling of world food supplies
over the last generation was achieved in part by adopting agricultural practices
that led to excessive soil erosion, erosion that is draining the land of its
productivity. After a point agriculture can no longer be sustained and the land
is abandoned.
Sustainability is an ecological concept with economic implications. It recog-
nizes that economic growth and human well-being depend on the natural
resource base that supports all living systems. Technology has greatly ex-
panded the earth's human carrying capacity, most obviously with advances in
agriculture. But while the human ingenuity embodied in advancing technology
can raise the natural limits on human activity, it cannot entirely remove them.
A sustainable society is one that shapes its economic and social systems so that
natural resources and life-support systems are maintained. Today, we study the
archaeological sites of earlier civilizations that failed to do so, depleting their
soils, mismanaging their irrigation systems, or otherwise embarking on an
unsustainable development path [Brown (1984, pp. 1-2)].
Of course, none of these civilizations possessed the scientific capacity of
modern societies to create new technologies specifically designed for the resource
shortages that emerge over time. Unless this capacity suddenly erodes dramati-
cally, it seems likely to provide solutions to future shortages of resources in
similar fashion to those of the past.

4.2.5. Role of gouernment

At the same time that planners have learned that resource management rather
than resource constraints per se is the primary bottleneck to economic growth,
they came to view prices generated in international markets as important signals
about relative scarcity of various resources and to regard trade as the most
efficient vehicle to alleviate significant imbalances of resources in a given country.
The importance of market signals and trade has led to a growing consensus
around a market orientation and the use of private incentives as the most
effective way to achieve economic growth, at least in agriculture. Millions of
decision-makers have turned out to be too many to reach from central planning
offices, because agricultural diversity is too great for information to reach those
offices effectively. Agriculture is itself changing too rapidly for planners to keep
up. This rapid change is reflected primarily in international markets, and agricul-
tural economies that are cut off from those markets miss key signals about the
efficiency of domestic resource allocation.
An emphasis on the role of international markets and trade is easily carica-
tured into an argument for free trade and "getting prices right" by setting them
Ch. 8." The Agricultural Transformation 321

at whatever the border price happens to be [Volrath (1985), for example]. The
extent to which a country's internal decision-makers face international market
signals is one of the key policy instruments available to a government to influence
income distribution as well as efficiency of resource allocation. Because dynamic
efficiency is more important for economic growth than static optimization, a
concern for the long-run impact of prices on expectations, investment, and
technical change is entirely legitimate. Free trade provides no guarantee that
dynamic efficiency will be achieved, and the record of East Asia cannot be
offered as evidence that free trade leads to rapid growth. That same record does
suggest, however, the importance of an export orientation for industry in combi-
nation with growing incomes in the rural sector. For an agricultural development
strategy to be relevant for the 1990s, it must incorporate the factors responsible
for that record.

5. Agricultural development strategy

Several lessons have been learned since the mid-1960s about the functioning of
the agricultural sector and its potential role in the development process. The
agricultural sector has been seen in a general-equilibrium perspective, and the
importance of macroeconomic policy for agricultural performance has been
recognized. Rapid economic growth has been considered necessary to deal with
the human welfare concerns that stem from poverty and hunger, and such growth
is feasible because of the potential for technical change. Market-oriented systems
with private incentives have shown superior performance in achieving this
growth. Policy analysis has tended to concentrate on one of three dimensions of
government intervention into the agricultural growth process: stimulating tradi-
tional agriculture into growth; maintaining agricultural growth to generate re-
sources for the rest of the economy; and protecting the welfare of farmers from
their own high productivity during the final and painful stages of structural
change in industrialized societies.

5.1. Policies for "getting agriculture moving"

It has become increasingly recognized that in order for agriculture to play a

multiplicity of positive roles, it needs resources and favorable development
policies, not heavy taxation and neglect. By the 1970s, agreement was being
reached on the nature of resources needed to develop agriculture, and some
progress was being made in identifying the policies needed to make those
resources effective. Heavily influenced by Schultz's book, Transforming Tradi-
tional Agriculture, and the increasingly widespread evidence that farmers re-
322 C.P. Timmer

sponded rationally to economic incentives, strategists in the late 1960s and early
1970s focused on two complementary agendas: understanding the microeconomic
setting of farm-level decision-makers in order to create incentives for investments
in higher output, and generating the stream of technical innovations that would
be profitable for individual farmers to adopt in order to produce that output.
With high food prices in world markets in the mid-1970s, providing better
incentives to rural producers was a "simple" matter of liberalizing trade policy
and permitting international price signals to be more freely transmitted to the
domestic economy. Because rural economies had been discriminated against for
so long by the industrialization strategies, rural incomes were very low relative to
urban incomes. Goals in terms of both equity and efficiency were furthered by
raising agricultural prices to their world levels, a point stressed by Schultz and his
colleagues (1978) and now pursued by Western aid agencies.
The emphasis in the late 1970s and early 1980s on market liberalization as a
means of providing adequate price incentives to agricultural producers has run
into serious problems in the mid-1980s. Allowing domestic food and agricultural
prices to be determined by world prices creates serious difficulties for both
producers and consumers because commodity prices in world markets are much
more variable than prices ft)r industrial products. Since the mid-1970s when this
strategy was articulated, many prices, especially for grains, have collapsed to
historic lows, and there is relatively little prospect of recovery in the foreseeable
future or at least within the vision of planning agencies and policy-makers. Once
the painful decision is made to raise price incentives to farmers, it is not easily
reversed, especially because the medium-term consequences for income distribu-
tion would be sharply negative. 11 To provide farmers with positive price incen-
tives then requires agricultural price protection, which might possibly lead to the
same type of high-cost, inefficient agricultural sector that presently exists for
industrial sectors in these developing countries.
The appropriateness of an incentive-led strategy for agricultural development,
as opposed to a market-liberalization strategy, depends on whether the argument
for protection has any merit: that providing adequate price incentives to farmers
through protection from international competition will encourage an infant
industry to grow up and produce at low cost. 12 In the 1950s and 1960s, import
substitution for industrial products was used to justify the use of price protection
for domestic (infant) industry, through tariffs, quotas, or bans on cheaper foreign
goods. As industries matured, their goods would be able to compete with foreign

l i T h e very short-run consequences for poor consumers would be positive, just as they were
negative when price incentives were adopted in the first place [Timmer (1979)].
"a2Protection has also been justified on the basis of price stabilization, i,e. that "low- prices in
world markets would rise to "normal" or "trend" levels before long. But an analysis of alternative
trends to be used to defend this proposition shows that the current price is a better predictor of future
prices, for at least five years into the future, than estimated trends [see Schwartz (1987)].
Ch. 8: The Agricultural Transformation 323
goods in local markets and be exported, and trade barriers could come down. But
for various reasons that had to do with "X-efficiency" and political economy,
only a few countries were able to make this transition [see Pack, Chapter 9, and
Westphal, Chapter 20, in this Handbook]. The industrial sectors in most develop-
ing countries are high cost and inefficient, and they remain heavily protected. The
agricultural sectors in these countries have borne much of the burden imposed by
the industrial and trade policies, in the form of high-cost inputs and overvalued
exchange rates that (implicitly) subsidize imported foodstuffs and tax rural
exports [see S. Lewis, Chapter 30 in this Handbook].
Protection for the rural sector carries clear benefits but equally clear costs and
risks. Protection would maintain the momentum in agricultural production
achieved through higher price incentives, and it would also support income levels
in rural areas. But if the lessons from industrial protection in the 1960s are
applicable to agriculture in the 1980s, planners should be cautious. If the
agricultural sector is fundamentally different in its response to protection from
that of the industrial sector, or if world markets for its output are sufficiently
different because of price instability, then short-run protection might be ap-
propriate. The answer is complicated by the realization that the economies most
successful in translating import substitution into export-led growth-Japan,
South Korea, and Taiwan-also have adopted the highest rates of agricultural
protection [Anderson and Hayami (1986)].

5.2. Alternative strategies for maintaining the transformation process

The lessons from the Asian success stories do not define a single strategic
approach to agricultural development. The agricultural sector is a means to an
e n d - n o t an end in itself. Three sharply different paths for appropriate policies
toward agriculture are open if the goal is to speed the overall process of
development. The first path has parallels to the philosophy of the 1950s, in which
benign neglect of agricultural policy was thought to be sufficient for stimulating
the process of economic growth. This perspective grows out of the recognition of
the role of well-functioning markets and decision-makers operating in a world of
"rational expectations". In this view, most policy is irrelevant to farmers in more
than a very transitory sense, and this is especially true of price policy:
One lesson that we should be able to learn from observation of the world is
that the absolute incomes earned by farm families in various countries have no
relationship to farm prices. Even stronger, the relative incomes of farm fanfilies
have no relationship to farm prices, except as benefits of higher prices have
been capitalized into the value of land and land has been acquired by gift or
inheritance [Johnson (1985, p. 43)].
324 C.P. Timmer
In this world, agricultural incomes are determined by employment opportuni-
ties outside agriculture, the agricultural sector must decline in proportional
output terms and absolutely in the labor force, and the long-run decline in basic
agricultural commodity prices due to technical change simply emphasizes that
society is best served by getting resources out of agriculture as rapidly as
possible. Although the clearest case for this view of the world is in the OECD
countries, a host of middle-income countries, and even some quite poor countries,
are also facing the problem of declining real incomes in the agricultural sector
under the impact of rapid technical change domestically and lower world prices
for the resulting output. This perspective is obviously consistent with the view
that open economies will show better performance than those with substantial
trade barriers.
A sharply different path has been sketched by Mellor and Johnston (1984).
Building on their earlier stress on balanced growth (1961), Mellor and Johnston
call for an "interrelated rural development strategy" that improves nutrition in
one dimension while it fosters the broader growth process in the other. The
approach calls for a major role of government in strategic design and program
implementation, a role that is in marked contrast with the free-market approach
sketched out previously:

We have, therefore, emphasized that improvements in nutrition [one of Mellor

and Johnston's key objectives for agricultural development] require a set of
interacting forces: accelerated growth in agriculture; wage goods production; a
strategy of development that structures demand towards high employment
content goods and services; increased employment; and increased effective
demand for food on the part of the poor. Agricultural growth not only satisfies
the need for food to meet nutritional requirements (which is the other side of
the wage-goods coin), but fosters a favorable employment-oriented demand
structure as well. Agriculture's role in generating a structure of demand,
favorable to rapid growth in employment, is central [Mellor and Johnston
(1984, pp. 567-568, emphasis added)].

Mellor and Johnston go on to summarize their earlier argument that agricul-

ture can play this multiplicity of roles only if a unimodal development strategy is
followed, that is, one in which a broad base of smallholders are the central focus
of agricultural research and extension services and the recipient of the bulk of
receipts from agricultural sales. The authors see the dualism inherent in bimodal
strategies- those placing modernization efforts primarily on large, "progressive"
farms while neglecting the "backward" smaUholders- as the major obstacle to
putting their set of interacting forces in motion:
The most common barrier to the interrelated strategy indicated is pronounced
dualism in capital allocations-too much to industry and the unproductive
Ch. 8: The Agricultural Transformation 325
elements of the private sector rather than to agriculture, and to capital-inten-
sive elements within those, as well as to large-scale and therefore capital-inten-
sive allocations within agriculture. The outcome of the strategy will depend
upon national-level decisions about macroeconomic policies, exchange rates,
interest rates, and investment allocations among sectors and regions, not just
within agriculture itself. Indeed, the whole strategy fails if it is viewed simply
as the responsibility of agriculture ministries [Mellor and Johnston (1984, p.
568)1.

This interrelated strategy must be directed by government planners; there is

relatively little concern or role for the private sector, other than small farmers.
The analysis leading to the strategy remains heavily influenced by closed econ-
omy considerations, and little attention is given to either domestic marketing
activities or their relationship to international markets. Three key elements are
suggested as essential to meeting all objectives of agricultural development - mas-
sive investment in human capital through nutrition, health, and family planning
services in the countryside, creation of the complex, rural organizational struc-
tures seen in Japan and Taiwan that provide services to small farmers while also
serving as a voice for their interests, and investment in rapid technical change
appropriate to these small farmers in order to raise agricultural output and rural
incomes simultaneously.
Notably missing in this list of key elements is significant concern for the
structure of incentives for agriculture relative to industry's or for the country's
tradables relative to those of foreign competitors. Although it is realized that the
macroeconomic setting is no doubt important to agriculture, it remains outside
the scope of appropriate strategy for agricultural development. Not surprisingly,
given the argument in Johnston and Clark (1982), the intellectual foundation for
this strategy lies in rural development, not in a vision of agriculture linked to the
macro economy and world markets by powerful market mechanisms. It is this
latter vision which provides the third potential path for agricultural development
strategy for the rest of the 1980s and into the 1990s.
The third approach contrasts with both the "free market" and "interrelated
rural development strategy" approaches. It calls for government policy interven-
tions into outcomes in domestic markets but uses markets and the private
marketing sector as the vehicle for those policy interventions. This "price and
marketing policy" approach recognizes widespread market failures in agriculture
as well as extensive government failures in implementation of direct economic
functions. ~3 The strategic dilemma is how to cope with segmented rural capital
and labor markets, poorly functioning land markets, the welfare consequences of
sharp instability of prices in commodity markets, the pervasive lack of informa-

13This is a theme of both the Bardhan (Chapter 3) and Stiglitz(Chapter 5) in this Handbook.
326 c.P. Timmer

tion about current and future events in most rural economies, and the sheer
absence of many important markets, especially for future contingencies involving
yield or price risks. One powerful lesson emerged from the postwar development
record: direct government interventions through state-owned enterprises to cor-
rect market failures frequently make matters worse by inhibiting whatever market
responses were possible in the initial circumstances, without providing greater
output or more efficient utilization of resources. The agricultural sector in
particular is vulnerable to well-intended but poorly conceived and managed
parastatal organizations that attempt a wide array of direct economic activities,
including monopoly control of input supplies, capital-intensive state farms, and
mandated control over crop marketing and processing. As Bates (1981) has
demonstrated, these direct controls and agencies have a strong political economy
rationale for a government that tries to reward its supporters and centralize
power and resources in the hands of the state [see also Lipton (1977)].
The answer to the dilemma over making matters worse, in the "price and
market policy" approach, is to gain a much clearer understanding of the
necessary interaction between the public and private sectors. Government inter-
vention into agriculture for political reasons has an ancient history. One major
claim of monarchs to the throne was their capacity to keep food prices cheap and
stable, as Kaplan (1984) made clear and as several modern governments have
discovered to their demise. Political objectives for the performance of
agriculture-its capacity to feed the population regularly and cheaply, or its
ability to provide fair incomes to farmers caught in the painful pressures of
successful structural transformation- are inevitable and, in some long-run sense,
highly desirable.
The "price and marketing policy" path argues that these objectives are best
served by making carefully designed interventions into the prices determined in
markets, not by leaving markets alone or by striving to reach the objectives
through direct activities by the government [Timmer (1986)]. If the "free market"
approach incurs heavy political costs as markets relentlessly redistribute incomes
to the winners in the course of economic development, and the "interrelated rural
development strategy" incurs heavy managerial and administrative costs as the
government plays an active and direct economic role, the "price and marketing
policy" approach incurs heavy analytical costs.
These analytical costs come from the need to understand each country's path
of structural change, the workings of factor and commodity markets, and the
potential impact of macro and commodity price interventions on these markets
and ultimately on the structural path itself. It requires that government interven-
tion be based on an empirical understanding of economic responses to a change
in policy and the political repercussions from them. There is an important role
for models in illuminating where to look for these responses, but the models
themselves cannot provide the answers. This is especially true as attempts are
Ch. 8: The Agricultural Transformation 327
made to build into the models the response of policy itself to changes in the
economic environment [see Roe, Shane and Vo (1986)]. Such endogenous policy
models might reveal some of the historical factors that accounted for policy
shifts, but they seldom provide a sense of when the degrees of freedom for
policy initiative are about to expand. Frequently, this is in times of crisis.
Policy-makers often embark on bold experiments in such times, and the payoff
would be very high if sufficient analytical understanding already existed in order
for them to anticipate the response to a policy change.
All three strategic approaches recognize the importance of government invest-
ments in agricultural research and rural infrastructure. Even here, however, there
are likely to be significant differences in emphasis. The free-market approach is
likely to put a relatively greater share into research, the rural development
strategy into human capital investments, and the price and marketing approach
into rural infrastructure that lowers marketing costs. Investments in all three
areas are obviously desirable. The issue is at the margin: where are scarce
resources to be invested? In addition, different countries have different starting
points and different needs, so no single strategic approach makes sense for all
countries. But it is difficult to see how countries can develop their rural sectors
without relatively efficient marketing systems and adequate financial incentives
for their farmers. Accordingly, significant elements of the price and marketing
approach seem destined to be incorporated into all successful agricultural devel-
opment strategies, even if they emphasize the free market or rural development
approaches in other dimensions.

5.3. Agricultural policy and structural change

Hayami and Ruttan have asked why agricultural growth has not been faster and
more evenly spread around the world:
We indicated that the basic factor underlying poor performance was neither
the meager endowment of natural resources nor the lack of technological
potential to increase output from the available resources at a sufficiently rapid
pace to meet the growth of demand. The major constraint limiting agricultural
development was identified as the policies that impeded rather than induced
appropriate technical and institutional innovations. As a result, the gap widened
between the potential and the actual productive capacities of LDC agriculture
[Hayami and Ruttan (1985, p. 416)].
This emphasis on the relationship between policy and agriculture's role in
structural change has provided the organizing theme for this chapter. A progres-
sion of topics has followed from understanding why the agricultural sector is
different from the industrial and service sectors and how the differences condition
328 C.P. Timmer

the nature of effective policy interventions. The factors needed for inducing the
agricultural transformation, to "get agriculture moving", involve a complex mix
of appropriate new technology, flexible rural institutions, and a market orienta-
tion that offers farmers material rewards for the physical effort they expend in
their fields and households and for the risks they face from both nature and
markets.
The role of the government has been analyzed throughout this chapter, first, as
it fosters the transformation process through its investments-in both budgetary
and policy t e r m s - in agricultural development, and, second, as it tries to cope
with the problems of success. A recurrent theme of this chapter has been that a
successful structural transformation is painful for the agricultural sector in all
societies; nearly all rich countries protect their farmers at the expense of domestic
consumers and taxpayers and of foreign producers. The rapidly growing econo-
mies of East and Southeast Asia are facing this issue in an acute fashion, well
before their overall economies can bear the fiscal burden of heavy agricultural
subsidies [Anderson and Hayami (1986)]. The experiences of the currently
developed countries with respect to the social, political, and economic stresses
caused by a declining role for agriculture have important lessons for latecomers
about to encounter these same stresses [Reich, Endo and Timmer (1986)]. There
is a world of difference, however, between those countries growing rapidly
enough to be feeling the consequences for income distribution of the relative
decline of the agricultural sector and those countries in which the agricultural
transformation itself has yet to begin in a significant way. The contrast between
Asia and Africa in this regard is striking. Many development specialists feel that
reversing Africa's declining food production per capita and declining real in-
comes per capita is the most important challenge for the rest of the century. For
the agricultural development profession, the difficult question is whether the
lessons from Asia in stimulating the process of agricultural transformation can be
transferred to the vastly different African setting. Many policy experiments are
now under way; analysis of the record generated in the 1980s by these experi-
ments will provide new insights in the 1990s into determining which models of
development can best stimulate and explain the process of structural transforma-
tion.

References

Adams, D.W. and Graham, D.H. (1981) 'A critique of traditional agricultural credit projects and
policies', Journal of Development Economics, 8:347-366.
Alderman, H. (1986) 'The effectof food price and income changes on the acquisition of food by
low-income households', IFPRI Research Report. Washington, D.C.: International Food Policy
Research Institute.
Anderson, K. (1983) 'Growth of agriculturalprotection in East Asia', Food Policy, 8:327-336.
Ch. 8: The Agricultural Transformation 329

Anderson, K. and Hayami, Y., with associates (1986) The political economy of agricultural protection:
East Asia in international perspective. London: Allen and Unwin.
Baran, P.A. (1952) 'On the political economy of backwardness', Manchester School of Economic and
Social Studies, 20:66-84.
Bairoch, P. (1975) The economic development of the third world since 1900. Berkeley: University of
California Press.
Bardhan, P.K. (1984) Land, labor, and rural poverty. Cambridge: Cambridge University Press.
Bates, R.H. (1981) Markets and states in tropical Africa: The political basis of agricultural policies.
Berkeley: University of California Press.
Binswanger, H.P. and Rosenzweig, M.R. (1981) Contractual arrangements, employment and wages in
rural labor markets: A critical review. New York: Agricultural Development Council, and India:
International Crops Research Institute for the Semi-Arid Tropics.
Binswanger, H.P. and Rosenzweig, M.R. (1986) 'Behavioral and material determinants of production
relations in agriculture', The Journal of Development Studies, 22:503-539.
Brown, L.R. (1984) 'Overview', in: L.R. Brown, et al., eds., State of the world, 1984: A Worldwatch
Institute report on progress toward a sustainable society. New York: W.W. Norton.
Chenery, H.B. and Syrquin, M. (1975) Patterns of development, 1950-1970. London: Oxford
University Press.
Chenery, H.B. and Taylor, L. (1968) 'Development patterns among countries and over time', Review
of Economics and Statistics, 50:391-416.
Chenery, H.B. et al. (1974) Redistribution with growth. London: Oxford University Press.
Chilchilnisky, G. and Taylor, L. (1980) 'Agriculture and the rest of the economy: Macro connections
and policy restraints', American Journal of Agricultural Economics, 62:303-309.
CIMMYT Economic Staff (International Maize and Wheat Improvement Center) (1984) 'The
farming systems perspective and fartners participation in the development of appropriate technol-
ogy, in: C.K. Eicher and J.M. Staatz, eds., Agricultural development in the third world. Baltimore,
MD: Johns Hopkins University Press.
Clark, C. (1940/1957) The conditions of economic progress, 3rd ed. London: Macmillan.
Cochrane, W.W. (1979) The development of American agriculture: A historical analysis. Minneapolis,
MN: University of Minnesota Press.
de Janvry, A. (1981) The agrarian question and reformism in Latin America. Baltimore, MD: Johns
Hopkins University Press.
Eicher, C.K. and Staatz, J.M., eds. (1984) Agricultural development in the third worM. Baltimore, MD:
Johns Hopkins University Press.
Evenson, R. (1981) 'Food policy and the new home economics', Food Policy, 6:180-193.
Fei, J.C.H. and Ranis, G. (1964) Development of the labor surplus economy: theo~ and policy.
Homewood, IL: Irwin.
Goldman, R.H. and Squire, L. (1982) 'Technical change, labor use, and income distribution in the
Muda irrigation project', Economic Development and Cultural Change, 30:753-775.
Gonzalez-Vega, C. (1977) 'Interest rate restrictions and income distribution', American Journal of
Agricultural Economics, 59:973-976.
Griffin, K. (1979) The political economy of agrarian change: An essay on the green revolution, 2nd ed.
London: Macmillan.
Hart, G. (forthcoming) 'Interlocking transactions: Obstacles, precursors or instruments of agrarian
capitalism', Journal of Development Economics.
Hayami, Y. (1984) 'Assessment of the green revolution', in: C.K. Eicher and J.M. Staatz, eds.,
Agricultural development in the third world. Baltimore, MD: Johns Hopkins University Press.
Hayami, Y. (1986) 'Agricultural protectionism in the industrialized world: The case of Japan',
prepared for a conference held at the East-West Center, Honolulu, February 17-21.
Hayami, Y. and Ruttan, V. (1985) Agricultural development: An internationalperspective, revised and
expanded edition. Baltimore, MD: Johns Hopkins University Press.
Hirschman, A.O. (1958) The strategy of economic development. New Haven, CT: Yale University
Press.
International Rice Research Institute (IRRI) (1978) Economic consequences of the new rice technology.
Los Banos, Philippines: International Rice Research Institute.
330 C.P. Timmer
Jones, C.W. (1986) 'Intra-household bargaining in response to the introduction of new crops: A case
study from North Cameroon, in: J.L. Moock, ed., Understanding Africa's rural households and
farming systems. Boulder, CO: Westview.
Johnson, D.G. (1973) World agriculture in disarray. New York: St. Martins Press.
Johnson, D.G. (1985) 'World commodity market situation and outlook', in: B.L. Gardner, ed., U.S.
agricultural policy: The 1985 farm legislation. Washington, DC: American Enterprise Institute for
Public Policy Research.
Johnston, B.F. (1970) 'Agriculture and structural transformation in developmer/t countries: A survey
of research', Journal of Economic Literature, 3:369-404.
Johnston, B.F. and Clark, W.C. (1982) Redesigning rural development: A strategic perspective.
Baltimore, MD: Johns Hopkins University Press.
Johnston, B.F. and Kilby, P. (1975) Agriculture andstructural transformation: Economic strategies in
late developing countries. New York: Oxford University Press.
Johnston, B.F. and Mellor, J.W. (1961) 'The role of agriculture in economic development', American
Economic Review, 51:566-593.
Jorgenson, D.W. (1961) 'The development of a dual economy', Economic Journal, 71:309-334.
Krishna, R. (1984) 'Price and technology policies', in: C.K. Eicher and J.M. Staatz, eds., Agricultural
development in the third world. Baltimore, MD: Johns Hopkins University Press.
Kaplan, S.L. (1984) Provisioning Paris: Merchants and millers in the grain and flour trade during the
eighteenth century. Ithaca, NY: Cornell University Press.
Kuznets, S. (1966) Modern economic growth. New Haven, CT: Yale University Press.
Lappe, F.M. (1971) Diet for a small planet. New York: Ballantine.
Lenin, V.I. (1899, reprinted 1964) The development of capitalism in Russia. Moscow: Progress
Publishers.
Lewis, W.A. (1954) 'Economic development with unlimited supplies of labor, Manchester School of
Economic and Social Studies, 22:139-191.
Lipton, M. (1977) Why poor people stay poor: Urban bias in world development. Cambridge, MA:
Harvard University Press.
Little, I.M.D. (1982) Economic development: Theo~, policy, and international relations. New York:
Basic Books.
Mellor, J.W. and Johnston, B.F. (1984) 'The world food equation: Interrelations among development,
employment, and food consumption', Journal of Economic Literature, 22:531-574.
Mosher, A.T. (1966) Getting agriculture moving: Essentials for development and modernization. New
York: Praeger.
Myint, H. (1975) 'Agriculture and economic development in the open economy', in: L.G. Reynolds,
ed., Agriculture in development theory. New Haven, CT: Yale University Press.
Nichols, W.H. (1963) "An 'agricultural surplus' as a factor in economic development", Journal of
Political Economy, 71:1-29.
Newbery, D.M.G. and Stiglitz, J.E. (1981) The theory of commodity price stabilization: A stuctv in the
economics of risk. Oxford: Clarendon Press.
Nurkse, R. (1953) Problems of capital formation in underdeveloped countries. New York: Oxford
University Press.
Prebish, R. (1950) The Economic Development of Latin America and Its Principal Problems. Lake
Success, NY: U.N. Dept. of Economic Affairs.
Preobazhensky, E. (1965) The new economics. Oxford: Clarendon Press.
Reich, M.R., Endo, Y. and Timmer, C.P. (1986) 'The political economy of structural change: Conflict
between Japanese and United States agricultural policy', in: T.K. McCraw, ed., America versus
Japan. Boston, MA: Harvard Business School Press.
Reynolds, L.G., ed. (1975) Agriculture in development theory. New Haven, CT: Yale University Press.
Roe, T., Shane, M. and De, H.V. (1986) 'Price responsiveness of world grain markets: The influence
of government intervention on import price elasticity', Technical Bulletin no. 1720, Washington,
DC: International Economics Division, Economic Research Service, U.S. Department of Agricul-
ture.
Roumasset, J.A., Boussard, J.M. and Singh, I.L, eds. (1979) Risk, uncertain(v and agricultural
development. New York: Agricultural Development Council.
 Ch. 8." The Agricultural Transformation 331

Sah, R.K. and Stiglitz, J.E. (1984) 'The economics of price scissors', American Economic Review,
74:125-138.
Schuh, G.E. (1976) 'The new macroeconomics of agriculture', American Journal of Agricultural
Economics, 58:802-811.
Schultz, T.W. (1953) The economic organization of agriculture. New York: McGraw-Hill.
Schultz, T.W. (1964) Transforming traditional agriculture. New Haven, CT: Yale University Press.
Schultz, T.W. (1975) 'The value of the ability to deal with disequilibria', Journal of Economic
Literature, 13:827-846.
Schultz, T.W. (1978) Distortions of agricultural incentives. Bloomington, IN: Indiana University Press.
Scobie, G.M. and Posada, R.T. (1978) 'The impact of technical change of income distribution: The
case of rice in Colombia', American Journal of Agricultural Economics, 60:85-92.
Scott, J.C. (1976) The moral economy of the peasant. New Haven, CT: Yale University Press.
Schumacher, E.F. (1975) Small is beautiful. New York: Harper and Row.
Schwartz, R.J. (1987) 'Optimal trends for forecasting prices: An empirical assessment of three grains',
unpublished Ph.D. thesis, Harvard University, Cambridge, MA.
Sen, A.K. (1981) Poverty and famines London: Oxford University Press.
Singh, l.J., Squire, L. and Strauss, J. (1985) 'Agricultural household models: A su~,ey of recent
findings and their policy implications', Yale University Economic Growth Center Discussion Paper.
Taylor, L. (1980) Macro models for developing countries. New York: McGraw-Hill.
Streeten, P. (1986) What price food? Washington, DC: Economic Development Institute of the World
Bank.
Timmer, C.P. (1969) 'The turnip, the new husbandry, and the English agricultural revolution', The
Quarterly Journal of Economics' 83:375-395.
Timmer, C.P. (1979) 'Issues of production and consumption: A review of T.W. Schultz, ed.,
Distortions of agricultural incentives', Science, 205:385-386.
Timmer, C.P. (1981) "Is there 'curvature' in the Slutsky matrix?", Review of Economics and Statistics,
62:395-402.
Timmer, C.P. (1984) 'Energy and structural change in the Asia-Pacific region: The agricultural
sector', in: R. Bautista and S. Naya, eds., Energy and structural change in the Asia-Pacific region:
Papers and proceedings of the thirteenth Pacific trade and development conference. Manila: Philippine
Institute for Development Studies and the Asian Development Bank.
Timmer, C.P. (1986) Getting prices right: The scope and limits of agricultural price poli~y. Ithaca, NY:
Cornell University Press.
Timmer, C.P., Falcon, W.P. and Pearson, S.R. (1983) Food policy analysis. Baltimore, MD: Johns
Hopkins University Press for the World Bank.
Volrath, T. (1985) 'Developmental consequences of unrestricted trade', Foreign agricultural economic
report no. 213. Washington, DC: Economic Research Service, U.S. Department of Agriculture.
Waterfield, C. (1985) 'Disaggregating food consumption parameters', Food Policy, 10:337-351.
World Bank (1982) Worm development report 1982. New York: Oxford University Press.
World Bank (1986) World development report 1986. New York: Oxford University Press.

View publication stats