5 Agricultural development and economic growth

Agriculture has been transformedin the past three decades. Outputhas increased at nearly twice therate of earlier periods, fueledpartly by the developing world'sgreatly increased capacity to pro-duce food and partly by changesin the developed countries. Aboveall else, there has been unprece-dented technical change in agri-culture throughout the world.

Alongside this remarkable andsometimes underestimatedachievement stands the "worldfood problem." Hundreds of mil-lions of people in the developingworld are still without enoughfood. Population growth, the ef-fects of which are often exacer-bated by a highly unequal incomedistribution, has sharply reducedboth the per capita benefits of in-creased food production and theassociated increases in income percapita.

This paradoxof poverty in themidst of plentyhas long plaguedpopular understanding of the roleof agriculture in economic devel-opment. On the one hand, it hasled to a sense of hopelessnessabout the world's malnourished;on the other, to technological ov-erconfidence. Overanxiety aboutfood crises has alternated withtaking agriculture for granted,even neglecting it.

None of these extreme reac-tions is appropriate or helpful.Agriculture does indeed face spe-cial problems, of which two inparticular stand out. The first isbiologicalthe close dependenceon the natural environment. Thisresults in output levels that fluc-tuate unpredictably from seasonto season, and growth that variesgreatly from one region to an-other depending on natural re-40

sources and the extent to whichtheir potential has already beendeveloped. The second is eco-nomicthe problem of the radi-cal adjustments farmers face in theprocess of agricultural develop-ment. While its output may con-tinue to grow, agriculture's shareof both GDP and employment in-variably decreases as economiesdevelop and industrialize. In-deed, at later stages of develop-ment, the absolute size of the la-bor force in agriculture declines,ultimately to a very small fractionof the total.

In most ways, though, agricul-ture is not unique. Farmers, likeother workers, respond to incen-tives. In buying food, consumersare influenced by relative pricesand available income, as they arewhen they buy other things. Ag-riculture's success or failuretherefore owes much to factorsthat influence all economic activ-ity. If it lags, it is because of in-adequate investment, lack of in-centives, and inappropriatepolicies, as well as problems oftechnology, climate, and soil.

This chapter describes some keyelements in the worldwide ex-pansion and transformation ofagriculture. It discusses wheregrowth has taken place, where ithas not, and the problems whichpatterns of growth have pro-duced or failed to solve. In par-ticular it highlights the followingfeatures of agricultural develop-ment:

Progress has been slowest inthe low-income countries. Thesecountries account for most of theworld's poor, the vast majority ofwhom work on the land.

There are strong links be-tween agriculture and overall eco-

nomic growth. Few countries haveachieved sustained economicgrowth without first, or simulta-neously, developing their agri-culture.

As it proceeds, developmentdictates a declining role for agri-culture. Initially dominant, iteventually accounts for only asmall fraction of net output, em-ployment, and income. Success-ful adjustment to agriculture'schanging role determines not onlythe pace and pattern of devel-opment as a whole, but also theseverity of the problems ruralpeople face in the developmentprocess.

The international economicenvironmenttrade, technology,commercial capital, and devel-opment assistancehas contrib-uted to major changes in worldagriculture. Most changes havebeen beneficial, but some seriousproblems remain: notably the ag-ricultural protection policies of thedeveloped countries, which limitthe market access of developing-country producers, at high costboth to governments and to con-sumers in the developed world.

The pattern of recent growth

While the rate of growth of agri-cultural output over the past threedecades was about double that ofearlier years, it was deceleratingduring the period. Worldwide,output rose at 3.1 percent a yearin the 1950s, 2.6 percent in the1960s, and 2.2 percent in the 1970s.The broad pattern was one of de-clining growth rates in the non-market industrial economies,while rapid growth was main-tained in developing countries and

industrial market economies.(Table 5.1 and Figure 5.1 showthe pattern for the 1960s and1970s.)

In the industrial market econo-mies agricultural growth has beenled by rising demand for high-value products such as meat,poultry, dairy produce, fruit, andvegetables. This stimulated rapidgrowth in the production of ce-

reals, especially in North Americaand Australia, because changes inanimal husbandry led to heavydemand for feed grains. Today,grain fed to livestock accounts for70 percent of their grain con-sumption. Between them, live-stock and feed grain account forover 65 percent of gross agricul-tural output (see Box 5.1 andChapter 6).

Source: FAQ.Note: Production data are weighted by world export unit prices. Decade growth rates are based on midpoints of five-year averages,except that 1970 is the average for 1969-71.

Figure 5.1 Agricultural and food output, by major world regionand by country group, 1963-70 and 1970-78

Growth rate (percentage)5

4

3

Agricultural outputFood output

1963-70 70-78 63-70 70-78 63-70 70-78

Africa MiddleEast

Note: Excludes China.Source: FAO.

LatinAmerica

63-70 70-78 63-70 70-78 63-70 70-78 63-70 70-78

Southeast SouthAsia Asia

SouthernEurope

Industrialmarketeconomies

63-70 70-78 63-70 70-78 63-70 70-78

High-income Nonmarket Worldoil exporters industrial

economies

Rapid economic growth hasdrawn people away from agricul-ture. Agriculture's share of the la-bor force in most developedcountries is now only one-thirdof what it was in 1960, averagingonly 6 percent in the developedmarket economies in 1980. As aresult of a major shift toward la-bor-saving technology, labor pro-ductivity (output per worker) has

41

Table 5.1 Growth rates of agricultural and food output by major world regions (excluding China),1960-80

Agricultural output Food output

Total Per capita Total Per capita

Region and country group 1960-70 1970-80 1960-70 1970-80 1960-70 1970-80 1960-70 1970-80

Developing countries 2.8 2.7 0.3 0.3 2.9 2.8 0.4 0.4Low-income 2.5 2.1 0.2 -0.4 2.6 2.2 0.2 -0.3Middle-income 2.9 3.1 0.4 0.7 3.2 3.3 0.7 0.9Africa 2.7 1.3 0.2 -1.4 2.6 1.6 0.1 -1.1Middle East 2.5 2.7 0.0 0.0 2.6 2.9 0.1 0.2Latin America 2.9 3.0 0.1 0.6 3.6 3.3 0.1 0.6Southeast Asia 2.9 3.8 0.3 1.4 2.8 3.8 0.3 1.4South Asia 2.5 2.2 0.1 0.0 2.6 2.2 0.1 0.0Southern Europe 3.1 3.5 1.8 1.9 3.2 3.5 1.8 1.9

Industrial market economies 2.1 2.0 1.1 1.2 2.3 2.0 1.3 1.1Nonmarket industrial economies 3.2 1.7 2.2 0.9 3.2 1.7 2.2 0.9Total world 2.6 2.2 0.7 0.4 2.7 2.3 0.8 0.5

Box 5.1 Food versus feed?"Sheep eat men" was the seventeenth-century slogan of English peasants dis-possessed when common land was en-closed for private pastures. Today, some600 million tons of cereals are fed to an-imals every year. This amount could feed2.5 billion people, more than twice thenumber in poverty. Moreover, grain con-verted to meat loses 75 to 90 percent ofits calories and 65 to 90 percent of itsprotein. Accordingly, a lively school ofthought maintains that feeding grain tocattle, pigs, and chickens takes food awayfrom the hungry.

The issue is not a simple one, how-ever. The main use of cereals for feedoccurs in developed countries, where overtwo-thirds of the grain consumption isaccounted for by animals. This patternof grain use reflects the high incomes ofconsumers in the developed Countries. Ifcereals were not fed to livestock, cerealproduction in these countries would de-cline sharply.

A massive transfer of purchasing powerfrom rich countries to poor ones wouldbe needed to shift consumption from an-imals to people. Even if this unlikelyprospect were realized, the longer-term

consequences are problematic. Most ofthe world's hungry are farmers or live infarm areas where basic cereals are grown.It would be difficult if not impossible toinject large amounts of cereals into theseareas without reducing incomes and pro-duction in the very regions where in-creased income is most needed. Directfood assistance has an important role toplay, but it needs to be carefully bal-anced with other forms of assistance thatstimulate food production and incomes.

As development proceeds, the inter-action between livestock producers andgrain farmers plays an increasingly im-portant part in sustaining agriculturalgrowth. The rapid growth in demand forfeed is one factor responsible for the dra-matic increase in cereal yields in the in-dustrial countriesto nearly 3.5 tons ahectare. In the 1930s cereal yields werealmost identical in developed and de-veloping countries, about 1.1 tons perhectare. Rapidly developing middle-in-come countries are already following thepattern established by agriculture in thedeveloped world; about 25 percent of thegrain they consume is used for livestockfeed.

frequently risen faster in agricul-ture than in other sectors. Agri-culture has greatly increased itsuse of other inputs such as fertil-izer and machinery, which nowrepresent nearly half the value ofgross output. Agricultural valueadded has therefore risen lessrapidly than gross outputat lessthan 1 percent a year during the1970s in many countries,

In the nonmarket industrialeconomies of eastern Europe andthe USSR, output grew rapidly inthe 1950s and 1960s because vir-gin land was opened up and theuse of fertilizer and machinery in-creased. Between 1970 and 1980,however, agricultural growthslowed sharply in the two largestcountries, the USSR (averaging 1.9percent a year) and Poland (1.3percent a year). This slowdown

42

came at a time when the demandfor meat and other livestock prod-ucts was rising rapidly. Despitegreatly increased imports of feedgrains, agriculture has been slowto adapt to this changing patternof demand. Some of the reasonsfor this failure lie in historySo-viet agriculture suffered heavilyfrom war and neglect for thirty-five years after the Revolution of1917while others lie in the dif-ficulties of an overly centralizedsystem.

Agricultural output in the de-veloping countries considered asa group grew at historically highrates, with growth of just under3 percent a year during both the1960s and the 1970s (see Table 5.1).But the rate of growth of popu-lation was also unprecedentedlyhigh. The outcome was a modest

(0.3 and 0.4 percent a year) in-crease in agricultural and foodoutput per capita.

As is usually the case, these ag-gregates conceal substantial dif-ferences. In particular, there wasan acceleration of growth of ag-ricultural output in Southeast Asia;from being near the average fordeveloping countries (0.3 percentper capita a year) in the 1960s, itrose to 1.4 percent a year in the1970s. The Latin American coun-tries also increased their growthof agricultural output per capita,up to 0.6 percent a year in the1970s from 0.1 percent in the1960s.

By contrast, the yearly rate ofgrowth of agricultural output inAfrica declined (from 2.7 percentin the 1960s to 1.3 percent in the1970s)and the rate of popula-tion growth accelerated. Thesechanges meant that output percapita grew at 0.2 percent a yearduring the 1960s, but then fell by1.4 percent a year in the 1970s.Part of the decline was associatedwith a slowdown in the produc-tion of nonfood crops (tropicalbeverages and fibers); but thegrowth of food output per capitawas also transformed from amodest increase in the 1960s toa decline (-1.1 percent) in the1970s.

In South Asia, the balance be-tween population growth and ag-ricultural growth remained essen-tially unchanged over the twodecades. In spite of growth ratesof agricultural output which av-eraged 2.5 and 2.2 percent, re-spectively, output per capita grewat only 0.1 percent in the 1960sand about zero percent in the1970s. This performancedisap-pointing, since South Asia hadbeen one of the major benefici-aries of the Green Revolution andof massive investments in irriga-tion and fertilizeris a salutoryreminder of the consequences of

high and sustained rates of pop-ulation growth.

Of course, these growth rateswere not uniform within each re-gion, nor were they uniformwithin countries. In India, for ex-ample, not all regions are en-dowed with the fertile soil, effec-tive flood control, and irrigationand drainage systems that havepermitted some areas to showspectacular increases in agricul-tural production by adopting thenew crop technologies. In the1960s, agricultural growth ex-

ceeded 5 percent a year in abouta third of the country's 281 dis-tricts, but output actually fell in afifth of the districts and rose atless than 1 percent a year in an-other fifth.

Such regional differences in ag-ricultural growth rates are a com-mon phenomenon in most coun-tries, both developed anddeveloping. Within each of theEEC countries, for example, thereare regions of increasing and de-creasing agricultural income dur-ing the 1960-80 period. These re-gional disparities tend to beaccentuated during early devel-opment stages, however, becauseof the proportionately greater im-pact of natural conditions, thevariable effect of technology, andthe limited capacity of govern-ments to support agriculturalgrowth in less favored regions. Inthe United States, for example,which provides both national andstate support for agriculture, ag-ricultural output increased in everystate between 1930 and 1970, al-though growth rates still variedsignificantly between 1.7 and 6percent.

Low or negative growth in ag-ricultural production per capita inSouth Asia and Africa presents amajor challenge to domestic poi-icymakers and the internationalcommunity alike, The unprece-dented advances in plant breed-

ing, in the use of fertilizer, andin investment in irrigation overthe past twenty years have re-sulted in a tie in the race betweenpopulation growth and agricul-tural production in South Asia. InAfrica, where no comparable ad-vances in agricultural technologyhave been realized, the race is atpresent being lost. The challengeis to accelerate the pace of biolog-ical innovation, to provide thepolicy environment and the com-plementary resources that makethe new technologies effective inincreasing output and incomes,and to reduce the rapid rate ofgrowth of population.

The role of agriculture indevelopment

Economic growth is invariably ac-companied by a declining shareof agriculture in total output, in-come, and employment. At ad-vanced stages of development, thelabor force in agriculture starts todecline absolutely. Two mainforces, which together produce the

major structural transformationsshown in Figure 5.2, are at theheart of the development proc-ess. First, the pace of agriculturalgrowth is limited by the growthof demand for its output, 90 per-cent of which is food in all but ahandful of countries (see Chapter6). In turn, the growth of demandis limited by the tendency for theproportion of household spend-ing on food to decline as incomesrise. In India, with a per capitaincome of $240, for example,households spend 60 to 70 per-cent of their income on food; ina $10,000 per capita economy suchas Canada, they spend less than20 percent on food and are ableto enjoy variety and quality to agreater extent.

This tendency, known as En-gel's Law, is observed in alleconomies as they grow. The lim-its to agricultural growth imposedby the growth in domestic de-mand can, however, be raised ifincome growth is broadly dif-fused in the economy and coun-tries export their surplus produc-

Figure 5.2 Cross-country share of employment and outputin agriculture, 1980

Percentage100

80

40

0

U.U

Nepal Niger

... .:1 U

UU-I

-

Cameroonvory Ldätt

-----aNiraragua- .. _. ! . . . IUU. I

20 U U

U. UU U

UU

U

100 1,000

GNP per capita (dollars)

Note: Excludes high-income oil exporters.

U U

EmploymentOutput

Greece

UU

U

U

U

U

Il Iu.IIilUU.

10,000

43

U U

U 'IU

Pakistan-.U U_..

U

Burma60 ...

U.

tion. (This option is available onlyto a limited extent, because En-gel's Law operates on a globalscale, as well as for individualcountries.)

In the case of nonfood agricul-ture, competition from industrialsubstitutes grows as investmentand innovation lower the price ofindustrial goods and services rel-ative to agricultural raw mate-rials. Common examples are theuse of synthetics instead of nat-ural fibers (cotton, wool, jute) forclothing, the substitution of oil orcoal for wood, and the switch fromtimber to steel and cement in con-struction. Substitution is seldomtotal, however, so demand fornatural products usually contin-ues to grow. To benefit, produc-ers have to concentrate on mar-keting as well as on efficientproduction.

With development, farmers be-come increasingly removed fromconsumers. As subsistence farm-ing gives way to marketed pro-duction, the scope for specializedmarketing, transport, processing,and packaging is enlarged. In theindustrialized economies, theseactivities produce much morevalue added and employmentthan does agriculture itself; thefarm-gate value of food is nowonly one-third of the price con-sumers pay for it.

The second main force account-ing for the structural change inagriculture is increased agricul-tural productivity, made possibleby technological innovations andaccumulated public and privateinvestment. Employment ex-pands and diversifies as a con-sequence of the relative decline inthe agricultural work force. In theearly stages of development, al-most everybody works in agri-culture. Most of the activities laterclassified as "industrial" and"service" are initially carried out,if not within agriculture, at least

44

in rural areasthe manufactureof clothing, furniture, and tools;construction of housing and otherbuildings; transport, processing,and marketing; education, medi-cine, social and cultural activities.As development progresses, theseactivities become differentiatedand separated from agriculture,until only core agricultural pro-duction functions remain. Whilesome nonagricultural activities arestill carried out in rural areas,modern industry and servicesopen up new and qualitativelydifferent opportunities, many ofwhich require relatively high skillsand education. Although thisprocess is one of the most desir-able benefits of growth, it is alsothe most difficult to handle in hu-man terms (see Chapter 7).

The transfer of labor from ag-riculture (where the amount ofcapital per worker and averageproductivity is relatively low) toindustry and services (where cap-ital per worker and average pro-ductivity are relatively high) is thekey to raising incomes and out-put. The higher average produc-tivity of labor in the nonagricul-tural economy is reflected inincome differences between it andagriculture. In most countries, in-comes are lower in agriculturethan elsewhere, as shown by thefact that farming's output share isconsistently lower than its em-ployment share (Figure 5.2). Thisincome difference provides theincentive for people to move outof agriculture into nonagriculturalactivities in urban centers. Theagricultural shares of output andemployment are roughly equal-ized only at a comparatively latestage of development.

The process of change can bevery rapid in historical terms: inJapan and the less developed partsof Europe, the agricultural laborforce represented 40 to 50 percentof total employment as recently

as 1950. Developed countries nolonger depend heavily on agri-culture: it accounts for only 4 per-cent of output and 6 percent ofemployment in industrial marketeconomies, and 15 and 17 per-cent, respectively, in plannedeconomies. In low-income coun-tries, by contrast, agriculture stillaccounts for 30 to 50 percent ofoutput and an average of 70 per-cent of employment. Even in mostmiddle-income countries, 10 to 20percent of output and over 40percent of employment are de-rived from agriculture.

The continuing importance ofagriculture in the economies of thedeveloping countries is reflectedin the association between thegrowth of agriculture and of theeconomy as a whole. Amongcountries where the agriculturalshare of GDP was greater than 20percent in 1970, agriculturalgrowth in the 1970s exceeded 3percent a year in 17 of the 23countries whose GDP growth wasabove 5 percent a year (Table 5.2).During the same period, 11 of the17 countries with GDP growth be-low 3 percent a year managed ag-ricultural growth of only 1 per-cent or less. Agricultural and GDPgrowth differed by less than twopercentage points in 11 of 15countries experiencing moderategrowth. There have been excep-tions, of course, but they provethe rule: fast GDP growth andsluggish agriculture was a featureof some of the oil- or mineral-based economies such as Algeria,Ecuador, Morocco, and Nigeria(see Box 5.2).

The parallels between agricul-tural and GDP growth suggest thatthe factors which affect agricul-tural performance may be linkedto economy-wide social and eco-nomic policies. In the many low-income countries where agricul-ture predominates, effective poi-icies, institutions, and investment

Table 5.2 Growth of agriculture and GDP in the 1970s

Above 3 percent CameroonChina*ColombiaDominican Rep.GuatemalaIndonesiaIvory CoastKenyaKorea, Rep. of

'Low-income countries.Source: World Development Indicators.

programs for agriculture are vir-tually synonymous with effectiveoverall management. This is notsurprising since there are manyimportant growth linkages be-tween agriculture and the rest ofthe economy. Expanding agricul-tural production through tech-nological change and trade cre-ates important demands for theoutputs of other sectors, notablyfertilizer, transportation, com-mercial services, and construc-tion. At the same time, agricul-tural households are often thebasic market for a wide range ofconsumer goods that loom largein the early stages of industrialdevelopmenttextiles and cloth-ing, processed foods, keroseneand vegetable oils, aluminumholloware, radios, bicycles, andconstruction materials for home

Malawi'MalaysiaParaguayPhilippinesThailandTunisiaTurkeyYemen Arab Rep.

improvements. In most middle-income economies, agriculturalpolicies are only slightly less in-fluential. Despite the similar re-sources and history of the coun-tries, agricultural productivity hasincreased twice as fast in Came-roon and Liberia in recent yearsas it has in neighboring Guineaand Ghanaand four times as fastin Tunisia and Colombia as inMorocco and Peru.

Managing agriculturaldevelopment

The process of transformation iswell advanced in many middle-income countries, where agricul-ture's contribution to output andemployment has already declinedto modest levels and the agricul-tural labor force is starting to

Bangladesh Burundi'Central African Rep.* Sierra Leone'El Salvador Zaire*Haiti'HondurasIndia'Pakistan'Sri Lanka'Sudan'Upper Volta'

shrink in absolute terms. In thesecircumstances, mechanization andother labor-saving techniques(such as the use of herbicides) playa major role in sustaining agri-cultural growth. Education andtraining are needed to help peo-ple move into industrial and serv-ice jobsespecially people frombackward areas. In some of theadvanced middle-income econo-mies, marketed farm produce isincreasing sharply, often severaltimes faster than the overall rateof increase in agriculture. Heavyinvestment in roads, railways,ports, and other links in the mar-keting chain are required, to-gether with policies and institu-tional arrangements to ensureefficient transport, processing, andstorage. Similar considerationsapply to the timely availability,

45

Bolivia LiberiaBurma' NicaraguaMali' SenegalSomalia'Tanzania*

1-3 percent Costa RicaEcuadorEgyptLesotho

AgriculturalGOP growth

growth Above 5 percent 3-5 percent Below 3 percent

Below 1 percent Morocco Togo' Angola'Nigeria Chad'

Congo, Rep.Ethiopia'GhanaMadagascar'Mauritania'Mozambique'Nepal'Niger'Uganda'

variety, and cost of industrial in-puts such as chemical fertilizer.Finally, farmers and traders relymore on efficient (and increas-ingly complex) financial arrange-ments to provide credit for in-vestment and working capital (seeChapter 6). Among the middle-income economies, rural povertybecomes a less severe, and hencemore manageable, problem. Inbetter-off rural communities,nonfarm employment is increas-

Box 5.2 Oil and agriculture: The Nigerian experienceOil exportsabout 2 million barrels a dayin 1980have become Nigeria's majorforeign exchange earner. But oil has beena decidedly mixed blessing for agricul-ture.

In the early 1960s, agriculture wasthriving. Small farmers, who made upmore than 70 percent of the Nigerian workforce, expanded their production of foodand cash crops. Agricultural exports werebooming and small farmers producedabout 90 percent of them. Nigeria wasthe world's leading exporter of ground-nuts, groundnut oil, palm kernels, andpalm oil, and the second largest cocoaexporter (after Ghana). Income gener-ated in the farming sector supported thedevelopment of many local manufactur-ing and service industries and providedmost of the foreign exchange needed tobuild roads and other infrastructure.

Progress in agriculture was cut shortby the oil boom in the mid-1960s andwas further disrupted by the three-yearcivil war, which began in 1967. As a re-sult, farming has yet to recover. Agri-cultural growth has been close to zeroover the twenty years from 1960 to 1980,despite economy-wide growth of 4.8 per-cent a year. In the 1970s, the volume ofagricultural imports quadrupled whileagricultural exports were halved. By thelate 1970s, Nigeria was a net importer ofagricultural produce.

Oil income has depressed agriculturein the following ways:

A substantial revaluation of the nairaand high local inflation made exportingincreasingly unprofitable, while food im-ports became increasingly competitive

with local production. Higher incomesled urban Nigerians to change their eat-ing habits in favor of imported foods.

An immense urban boom attractedlarge numbers of young Nigerians intothe towns. Because of rapid overall laborforce growth, the number of farm work-ers may not have fallen absolutely, butthe rate of increase was very slow andthe proportion of older workers in-creased sharply.

The considerable effort made to sus-tain agriculture through public invest-ment and new development schemes wasgreatly hampered by the dearth of trainedand experienced manpower, which waslargely drawn into other, more buoyantsectors.

Oil clearly offers a major opportunityto accelerate the process of structuraltransformation. But the Nigerian expe-rience suggests lessons for other coun-tries in whose economies oil has a majorpotential role. Oil revenues must be usedto support an efficient transformationprocess. This means that close attentionmust be paid to manpower constraints,that public sector financial discipline mustbe tight, and that the impact of oil onthe exchange rate must be carefully mon-itored. Public investment needs to focuson expanding the economy's absorptivecapacityconcentrating on removingbottlenecks in transport, power, and otherinfrastructure sectors. Expenditures oneducation and training need to expandto meet the human resource require-ments of a more advanced and diversi-fied post-oil economy.

46

ingly important; for many, farm-ing itself becomes a part-time oc-cupation. Where problems dopersist, many countries have tar-geted investment toward ruralhealth, nutrition, and infrastruc-ture.

While middle-income countriesexpanded their agricultural out-put by 3.1 percent a year in the1970s, low-income countriesmanaged only 2.1 percent. As aconsequence, agricultural output

per capita increased by 0.7 per-cent a year in middle-incomecountries, while because of thepoor performance of the Africaneconomies it actually declined inlow-income countries by 0.4 per-cent a year.

The challenges of agriculturaldevelopment are clearly muchharder to meet in countries at ear-lier stages of development, whenthe economy itself is largely ag-ricultural. Agriculture merits sup-port and stimulusyet it has toprovide them itself, being the onlyreal source of income, savings, andgrowth. At the same time, coun-tries have other competing prior-ities, including industrial and ur-ban development. Against thisbackground, two sets of policy is-sues assume particular impor-tance. The first is domestic: whatincentives should be given tofarmers and how should theychange over time? The second setis international: how can govern-ments in the developed countriesassist agricultural progress in thedeveloping countries?

Relative prices and other incen-tives are often biased against thefarmer in developing countries.Overvalued exchange rates, tax-ation of agricultural exports, heavyprotection of fledgling industry,and inefficient parastatal process-ing and marketing agenciesallcontribute to this bias. Some ofthe issues involved in trying tomeasure the effects of these fac-tors are discussed in Box 5.4.

The relatively low prices re-ceived by farmers have seriouslyhampered growth in many de-veloping countries. Where priceshave not been kept artificially low,and where other conditions forgrowth have been favorable,farmers have responded by in-creasing output. The responsive-ness of farmers to incentivesincontrast to the outmoded andmistaken view that peasants are

set in traditional wayshas beenobserved in societies with diversesocial systems and levels of de-velopment (see Box 5.3).

Pricing alone is neither a com-plete explanation of agriculturalperformance nor the sole key toprogress, however. Other areasof government policy are impor-tant:

Comparative advantage. Manycountries have traditionally en-couraged specialization in agri-culture to exploit the comparativeadvantage given to them by na-ture. In recent years, countriessuch as the Ivory Coast, Malay-sia, and the Philippines haveachieved rapid agricultural growthby encouraging export crops whilealso expanding food output.

The importance of trade for theagricultural growth of developingcountries is evident from the shareof agricultural output which is ex-ported. Based on the FAQ's pro-

duction data (see Table 5.1), in1979, out of ninety developingcountries, exports accounted for50 percent or more of total agri-cultural output in ten countries;more than 20 percent in thirty;and more than 10 percent in fifty.Among countries exporting morethan half their total output, Jor-dan, Malaysia, Mauritius, and theCentral American and Caribbeancountries figure prominently.Among those exporting betweena half and one-third of total out-put, nearly half are in sub-Sa-haran Africa. For developingcountries in these two categories,the export market is a major fac-tor determining agriculturalgrowth. Since these are oftencountries where agriculture is adominant sector, agricultural ex-ports are an important factor intheir overall economic growth aswell.

Exporting is a rewarding but

also a risky business. Between1961 and 1978, African cocoa pro-ducers lost market share to LatinAmericans; South Asian domi-nance in tea was eroded by com-petition with African and Chineseexporters and fell from 80 percentof world exports to under 50 per-cent; Latin American coffee pro-ducers supplied only 51 percentof the market in 1978, comparedwith 62 percent in 1961; and thesmall, sugar-producing islands ofthe Caribbean have been unableto expand into new sugar marketsand have lost ground to East Asianand Latin American producers.

Specialization involves heavyinvestment and a commitment toresearch. Brazil's exports of soy-beans are a notable recent ex-ample of exploiting the compar-alive advantage of a crop that wasalready being grown in the coun-try (see Box 5.5). But greater dif-ficulties are posed for very small

Box 5.3 Improving incentives in Chinese agricultureAs part of a general shift toward greaterflexibility, agricultural policies in Chinastarted changing in 1977. The system ofrigid production planning in collectiv-ized agriculture was relaxed, and house-hold and other private agricultural activ-itie were encouraged. The approach stillsets local sales targets for the main crops,but allows production teams more lati-tude in their decisions on crops, land,and input use. Deregulation of local mar-kets helps promote the private sale andexchange of produce.

Incentives to promote the productionof priority commodities have been in-creased. Between 1977 and 1979 statepurchasing prices for all the importantcommodities were raised by 20 to 30 per-coot. In addition, farmers are being en-couraged to sell crops and livestockproducts to state agencies via entitle-ments to special allocations of grain orfertilizer (at relatively favorable prices)along with normal sales proceeds. Extraproduction is particularly favored: grainsales above quota attract a premium of

50 percent over base prices.Within the production cooperatives,

incentives have also been sharpened. Thechanges involve improvements to "workpoint" systems so that less is appor-tioned to workers as basic rations andmore as a return for the quality andquantity of individual work. Various typesof contractual arrangements are beingencouraged under which small groups ofworkers manage collective land and otherassets under contract to the productionteam, with bonuses for above norm per-formance.

Although it is too early to assess thelong-term impact of these changes, theyhave already resulted in a dramatic up-surge in agricultural output:

Between 1977 and 1980, the growthof output averaged close to 7 percent ayear.

Food-grain output increased by some

37 million tons, with growth averagingclose to 5 percent a year, though in both1977 and particularly 1980 bad weatherwas responsible for poor wheat crops.

More remarkable was the upsurgeamong some of the other major fieldcrops. Cotton output increased 23 per-cent in 1979-80, and oilseeds by 92 per-cent from 1977 to 1980; in both cases therewas a sizable expansion of cultivated areaas well as yield increases. Large in-creases are also reported for sugar, silk-worm cultivation, and jute. There is evi-dence that these increases reflectsubstantial shifts in cropping patternsbased on area specialization and localcomparative advantage.

Meat production (almost entirelypork) increased over 50 percent duringthe past three yearsa response both toimproved producer prices and to policiesthat give livestock producers greaterfreedom to use grain for animal feed.

China's planners did not anticipate sobig a response. Nor did they adjust con-sumer sales prices for the leading items;therefore subsidies, already quite a largeitem in the budget, have shot up toaround $5 billion a year.

47

poor countriesfor example, Bu-rundi, Gambia, Guinea, andUgandathat lack a research ca-pability and are almost totally de-pendent (for 75 to 95 percent oftheir foreign exchange) on one ora few tropical export crops. Ex-ceptional efforts will be requiredto achieve greater stability and di-

48

versity in the largely agriculturaleconomies, including supportfrom the international commu-nity to assist them in this difficulttask.

Technological progress. In thepast, agricultural research rele-vant to developing countries'needs was confined largely to

tropical export crops. Some, suchas cane sugar, have been throughseveral Green Revolutions over thepast century. In general, though,research has become critical onlyin modern times; it has been ap-plied to considerable effect incountries such as China, India,Indonesia, and the Philippines.

Box 5.4 Nominal protection and foreign exchange valuationDetermining the incentives or disincen- ual crop and country data since year-to- Nominal protection coefficients (NPCs)tives farmers face in different countries year changes can be large. But the data take into account tariffs, quotas, andis a rough science. The protection meas- confirm the high level of protection in nontariff barriers that protect farmers asures shown in the figure compare the many developed countries, where many well as the impact of export taxes or re-domestic prices that farmers are paid with producer prices are 50 to 100 percent strictions that penalize farmers. Protec-prices they could expect to face if the above world market levels, in contrast to tion is also a function of the country'scountry in question were freely trading the situation of most developing coun- exchange rate as it is influenced by policy.on the world market. Too much signifi- tries, where prices are often well below Protecting domestic industries, for ex-cance should not be attached to individ- the world market, ample, reduces industrial imports below

'2

o.0

.9 °

0

V

Adjusted nominal agricultural protection coefficients,

3.0 BAR

by country, late 1970s

WHT

2.0

ALFU

MET

EARU

%WHT .1.5 RIC UMZE U

WlITBAR4

U SOR

BEF RIC

WHT

RIC

UU PLY U UU U SUG U SUGMZE BEF U U U U MZE U BEF

E-'mi, UUBEF

1.0m WHT

. .USES

MZEU WHTUUU

MZE WHTma.

U COT U MZEMZE

RIC UU SOR UU SNF U COC U COC

iU

0.66

0.5

0.33.

U WHI UUUUUU

U BEF URYE BAR WHT

WHT C01U UMZE UCOT

BEFU U UU

WOLCOT MZE U

MZEUUBEF U COT

UWHT RIC UU BEF

COFU

URIC

MZE' ---USOR

UWHT U COT U FRT

-----'..-- ...,.... ....

UHZT URIC

- ,,,,,, ...........

5) 0C CN CN

to

0 >, .9 C5) a 5)

-' °

0 1-' HC

>

0 0 rO 00 N'

St -

U

05)

Note ALF is alfalfa; BAR, barley; BEF, beef; BEN, beans; COC, cocoa; COF, coffee; COP, copra; COT, cotton; FRT, fruits; GRT, groundnuts; GUA, gum arabic;HZT, hazelnuts; LVK, livestock; MET, meat; MLT, millet; MZE, maize; OLO, olive oil; OLS, ollseeds; PLY, poultry; POT, potatoes; PPR, palm products; PRK, pork;RIC, rice; RUB, rubber; RYE, rye; SES, sesame; SNF, sunflower; SOR, sorghum; SUC, sugarcane; SUG, sugar; TOB, tobacco; WHT, wheat; WOL, wool.

Much remains to be done for cropsgrownin semi-arid areas and forrain-fed annual crops generally(see Chapter 6). Although re-search is one of the cheaper in-vestments that governments canmake, it requires highly skilledpeople to direct it. Foreign as-sistance has played a major role

in this area, especially through theestablishment of international ag-ricultural research centers andthrough the financing and initialstaffing of national research ef-forts in poorer countries.

Mobilization of rural resources.Motivating and organizing farmcommunities to help in collective

works and infrastructure is widelyrecommended, though less oftensystematically practiced. China isthe outstanding example of whatcan be done to improve roads,land, and irrigation through com-munity efforts that use little pub-lic money. There and in someother countries, land reform has

49

their probable level in an unprotected count for over- or undervaluation of the elusive. Research at the World Bank andmarket. This tends to overvalue a coun- exchange rate, elsewhere suggests, however, that thetry's exchange rate. Agricultural export. The NPCs are not adjusted for one levels of traded inputs in agriculturalers then have a harder time competing other important factorthe protection outputs are relatively low in developingin the world market, since their costs are given to tradable inputs used in produc- countries, so that the qualitative resultsincreased in local currency while their tion, chiefly fertilizer and machinery. Such of EPCs and NPCs are similar.dollar export receipts exchange for less effective protection coefficients (EPC5) are

local currency than otherwise. The NPCs more difficult to compute because datashown in the figure are adjusted to ac- on input uses in agriculture are more

Import-competing commoditiesExport-competing commodities

Little trade orboth imported and exported

U2.0

RIC

PI'R

SOC

SUG

1.0

U MZEGRT

MLT

RIC SOR WHT

MZERIC

UWHT

COT MZE HNX

OL!WHTUN I UI UU

MZE0 66 U URIC URIC URIC

URIC UUSUG UWHT

0.5 SUG

RUBU

H

BCOP TOB SUGuUu U URIC

U U

UCOT UI USOR

RIC

HH H'JHCOF 8 8'U U

uU

MZECOT

UBEF

COFU UU COT UUSES_---_UCOT

COF TOB 8U U

UU URIC

UU

UCOC UGUA

0 c .5 . . .5 c -

C)

c .

provided an early stimulus forcommon action. This is not sur-prising, since collective action andindividual benefit become moreclosely and more directly alignedas holdings of land and other as-sets become more nearly equal.Labor is the most plentiful andcheapest resource available in low-income countries, while capital isscarce and expensive. Using therural work force for infrastructureimprovement is an importantmeans of converting labor intocapital. Management is often thekey constraint (see Chapter 7).

Public sector investment andsupport. Agriculture attracts a su-prisingly small share of govern-ment spendingbetween 5 and10 percent of most developingcountries' central governmentbudgets. Nevertheless, publicsector outlays on critical bottle-necks have helped to stimulateand sustain growth by the devel-opment of large-scale irrigation,research, and rural infrastructure.

This approach has been particu-larly useful in areas where the po-tential for rapid growth alreadyexists, and special programs oftencombine several elements into adevelopment package. In somecountriesIndonesia and the Re-public of Korea, for examplefer-tilizer subsidies have been usedto encourage growth. The effec-tiveness of such techniques shouldnot disguise the fact that they canbecome very costly; removingthem once they have achievedtheir objective can be difficult.

In some cases, however, the in-tervention of the public sector hasbeen counterproductive. In-stances of ineffective parastatalorganizations involved in themarketing of agricultural prod-ucts and the supply of inputsabound. Some governments havetaken bold steps to improve theefficiency of distribution systemsby opening up the sector to co-operatives and private traders (seeBox 5.6).

Foreign capital. Much early ag-ricultural development in, for ex-ample, Argentina, Brazil, Malay-sia, and East and West Africa waslargely based on private invest-ment, often involving foreignmanagement and control. Privateforeign investment and loans arestill important in some middle-in-come economies. In the low-in-come economies, little externalprivate investment or lending hasrecently gone into agriculture. In-stead, foreign aid has become themain external source of foreignexchange and technical support,particularly after 1973 (see Table5.3).

In the mid-1960s, while nearlya quarter of all official develop-ment assistance (ODA) was foodaid ($1.4 billion out of $5.9 bil-lion), project and technical assist-ance for agricultural programs wassmall. The food crisis of the early1970s, coupled with a growing re-alization that poverty and mal-nutrition were persistent prob-

Box 5.5 Brazilian soybeans:Brazil's expansion of soybean productionis a dramatic agricultural success story.Before 1960 production was insignifi-cant, but by the end of the 1970s soy-beans surpassed coffee as the principalBrazilian crop in value, covered some 17percent of the cropped area, and consti-tuted 14 percent of world soybean out-put Over this period Brazilian soybeanproduction grew at 27 percent a year inthe 1960s entirely through expansion ofthe area planted, and in the 1970s boostedby yield improvements of more than 2percent a year. Brazilian exports of soyand soy products (oil, meal, and cake)rose from $53 million in 1969 to $2.3 bil-lion in 1980, a remarkable 42 percent ayear rate of growth. Soybeans make upmore than 10 percent of Brazil's annualexports. As soy oil was substituted forimported vegetable oils, domestic con-sumption grew almost 28 percent a yearover the 1970s, and there was also grow-ing demand for soy meal, mainly forchicken feed.

A number of factors helped to stimu-late this impressive growth:

Rapid expansion of world demandfor protein for human and animal con-sumption; stagnation in the output of al-ternative protein sources (such as the Pe-ruvian anchovy); the fact that Brazil'ssoybean crop comes to market duringthe US winter, when world supplies areseasonably low; and the US embargo onits soya exports in 1973 (which stimu-lated Japanese importers to find alter-native sources).

Farmers typically double-crop soy-beans in the summer and wheat in thewinter. New wheat varieties that matureearlier enabled soybean-wheat double-cropping to expand to other areas. Cul-tural practices for the two crops are sim-ilar, and soybeans benefited from sup-port aimed at promoting wheatproduction (for example, the fivefold realexpansion of subsidized agricultural creditin the 1970s, about 20 percent of whichwas for soybeans).

Research was a critical factor. Morethan sixty soybean cultivars brought toBrazil from the United States during the1960s contributed to the developmentof domestic cultivars by the NationalCenter for Soybean Research. Of the forty-

eight cultivars distributed to farmers atthe end of the 1970s, twenty-six had beendeveloped in Brazil. It has been esti-mated by the Brazilian Agricultural Re-search Enterprise (EMBRAPA) that two-thirds of yield increases have come fromgenetic improvements. The other thirdhas come from improving soil manage-ment practicesproper application offertilizers, correctives, nitrogen-fixingbacteria, and herbicidesalso derivedfrom intensive and continuous research.That national yields grow apace with in-creases achieved at the research centersattests to the successful diffusion of theresearch results and their adoption byfarmers.

50

Creating a comparative advantage

Box 5.6 Private complements public: The Bangladesh experimentIn the mid-1970s, Bangladesh planners

estimated that fertilizer use had to ex-pand by 15 percent a year to attain thetargeted 4 percent yearly agriculturalgrowth rate. At that time, the Bangla-desh Agricultural Development Corpo-ration (BADC), a public corporation under

the Ministry of Agriculture, handled allmarketing of farm supplies as well as allprocurement of fertilizer. It had done well,but the strains had begun to tell.

In 1978 BADC established the NewMarketing System (NMS) to reduce re-strictions on private traders and thusmove toward a more open system fordistributing fertilizer. BADC wouldgradually withdraw from retailing and,except in remote areas, sell mainly towholesalers at "primary distributionpoints Private dealers and cooperativeswould be permitted to buy from all BADCwarehouses. Private movement of fertil-izer would be unrestricted except in theborder zones.

With bilateral assistance from theUnited States, BADC began the NMS in

the Chittagong Division, which coversone-fourth of the area of the country andaccounts for a third of total fertilizer con-sumption. It increased official dealers'margins, permitted farmers to buy fromany trader, and made it easier to becomea trader. It developed a private dealercredit system from commercial banks.BADC set up a monitoring system forthe NMS and took steps to cut down itsinternal transport and storage problems.

The NMS enjoyed reasonable success.Fertilizer sales increased, and forty-fivegovernment warehouses were put out ofbusiness, leaving the government activemainly in remote areas which did notattract wholesalers. Retail prices droppedbelow official prices around the primarydistribution points. The new fertilizerwholesalers showed they could movefertilizer cheaply and effectively fromsurplus to deficit areas, selling to bothfarmers and retailers.

On the basis of this pilot experience,the NMS was adopted nationwide. As of

mid-1980, major accomplishments of theNMS included:

BADC's fertilizer points of sale werereduced by 55 to 60 percent; about one-third of the original 130 warehouses wereclosed.

Farmer access to fertilizer points ofsale greatly increased in the ChittagongDivision.

Farmers paid less for fertilizer underthe NMS.

A new class of entrepreneurs hademerged.

The net results of the program havebeen almost entirely positive. The pri-vate distribution system is more flexibleand can supply farmers at lower costs,and much of BADC's managerial and la-bor force can be released for other tasks.The costs of constructing and maintain-ing the marketing and storage infrastruc-ture have been substantially reduced. Thegovernment is now collaborating moreactively with the private sector in otherareas.

Table 5.3 Official assistance (commitments) to agriculture, 1973-80

lems in large parts of the world,produced a major shift toward ag-riculture. Flows of ODA and otherofficial assistance for agriculturemore than doubled in real termsbetween 1973 and 1980. The shareof agriculture and rural develop-ment in the lending of the mul-

tilateral institutions rose to nearly30 percent.

The share of aid in agriculturalinvestment varies widely fromcountry to country. In China, byfar the largest country, virtuallyno external assistance was re-ceived during the past twenty-five

years. Similarly, in India, thoughlarge amounts of aid have beenreceived, it represents only 3 to 5percent of total spending on ag-riculture. Its use is concentratedon selected areas such as irriga-lion, and its effectiveness is largelydetermined by overall agricul-

51

(millions of dollars, constant 1979 prices)

Item 1973 1974 1975 1976 1977 1978 1979 1980

Official development assistance (ODA)DAC 1,594 2,819 2,359 2,246 3,279 3,633 4,304 3,773Multilateral agencies 1,533 1,833 1,530 1,814 2,139 2,761 2,503 2,969OPEC (bilateral and multilateral) 69 218 640 378 461 307 243 179Total ODA 3,196 4,870 4,529 4,438 5,879 6,701 7,050 6,921Percentage change +82 -6 -2 +32 +14 +5 -2

Other official flows (005)DAC' 351 275 137 395 159 403 329 222Multilateral agencies 902 1,610 2,944 2,150 2,816 3,275 2,319 2,621OPEC (bilateral and multilateral) 63 90 333 221 80 49 99 48Total OOF 1,316 1,975 3,414 2,766 3,055 3,727 2,747 2,891Percentage change +50 +60 -19 +10 +22 -26 +5

Grand total (all donors) 4,512 6,845 7,943 7,204 8,934 10,428 9,797 9,812Percentage change +62 +16 -9 +24 +17 -6 (.)

Source: OECD.a. Includes EEC.

tural investment. Elsewhere, es-pecially in the poorer Africancountries, aid accounts for 20 per-cent of public investment in ag-riculture. There too, the effective-ness of aid still depends on thegovernment's interest in, andcommitment to, agriculture.

The need for policy and in-vestment changes is not uniformamong developing countries.Many have been quite successfulin managing their agriculture andadapting to structural changes.They need little advice, althoughthey may need continued finan-cial support to sustain well-de-signed policies and programs.Others need to make majorchanges. They are not alone. Thedeveloped countries have foundit extremely difficult to managethe last stages of structural changein agriculture; their capacity to dealmore effectively with these chal-lenges will be an important factorin developing countries' successor failure.

The international environment

The international environment cansupport or restrict agricultural de-velopment in developing coun-tries in two main ways. First,through the mechanisms de-scribed in Part I of this Report,the international economy pow-erfully conditions the overallgrowth prospects of developingcountries. In turn, more buoyantoverall growth simultaneouslyprovides both strong support foragricultural growth and the re-sources to permit an expansion ofproductive capacity.

Second, trading opportunitiesfor agricultural products permitagricultural growth beyond thelimits set by the expansion of do-mestic demand alone. Agricul-tural exports remain a majorsource of foreign exchange formany developing countries, and

52

therefore an important determi-nant of overall economic growth.Patterns of agricultural trade havevaried markedly over the pastthree decades, reflecting in partthe growth of the world economyand in part the differential agri-cultural growth performance de-scribed earlier in this chapter.

During the period of fastestgrowth in the world economyroughly, from 1955 to 1973theexpansion of international tradewas dominated by manufacturedgoods, fuel, and minerals. Worldtrade grew at 8 percent a year,trade in agricultural goods by lessthan half that. By 1973, devel-oping countries' agricultural ex-ports had fallen from 60 percentof the value of their total exportsin 1955 to only 30 percent. Never-theless, agricultural goods re-

Figure 5.3 Index and growthrates of world exports, by volume

volume index (1972-74 av = 100)150

100 All goods

50 Manufactures

Average annual growth rate (percentage)10

Manufactures

All goods1963-73

5

Agriculturalcommodities

0

Agriculturalcommodities

1965 70 75

Sourc': GATT.

1973-79

80

mained the single most importantcategory of exports for more thantwo-thirds of the low- and mid-dle-income economies.

The early 1970s proved a turn-ing point in the agricultural tradeof developing countries. Since1973, the growth of world tradehas declined to 4.7 percent a year,in line with the slowdown of theworld economy. But the growthof agricultural trade has acceler-ated to 4.8 percent a year, largelybecause the demand for food hasincreased rapidly (see Figure 5.3).The extra purchasing power of theoil-exporting countries boostedtheir share of world food importsfrom less than 6 percent in 1973to 10 percent in 1978. Over thesame period, strong demand formeat and poultry in eastern Eu-rope, coupled with a poor supplyresponse from domestic agricul-ture, boosted the region's maizeand wheat imports from 3 to 14percent of its total food imports.Eastern Europe now accounts forabout one-third of world importsof maize and wheat. China hasalso become a major grain im-porter. In total, the world graintrade has expanded from about 30million tons in the mid-1950s toaverage 130 million tons a year inthe 1976-80 periodnearly one-half because of rising imports bydeveloping countries (includingChina) and one third because ofimport growth by the USSR andeastern Europe.

The extra demand has been metmainly by developed countries.The biggest relative increases inexports have come from the EEC,which produced trade surplusesin a number of commodities as aresult of continued agriculturalgrowth and stagnant demand inthe home market. By contrast, thelarge rise in North American grainexports only slightly increasedtheir importance in world foodtrade. Overall, the net food deficit

1965 70 75 80

of the industrial market econo-mies narrowed from 16 percent ofworld food trade in 1965 to lessthan 5 percent in 1978.

The oil-importing developingcountries gave greater emphasisto agricultural exports during the1970s. By 1978, their trade sur-plus on food amounted to $21 bil-lion, sharply up from $6 billion in1973. Performance has variedwidely by region and country.Some countries have been able topenetrate expanding markets; ex-amples include Brazil's successfulsoybean exports and Thailand'snew trade in cassava chips forlivestock feed. A number of low-income countries sharply in-creased their food exports to Mid-dle Eastern markets after 1973 (seeFigure 5.4). These markets nowaccount for more than 20 percentof low-income countries' food ex-ports, compared with 4 percentin 1973.

These were the successes. Inother cases, however, especiallywhere exports were based on tra-ditional tropical crops, inelasticdemand limited increases in for-eign exchange earnings. Al-though volume expanded in the1970s, the terms of trade deteri-orated considerably. Had agricul-tural prices been moving in linewith prices of manufacturedgoods, this higher volume of ex-ports would have given low-in-come oil importers an additional$1.2 billion of foreign exchange in1978 and middle-income ones anextra $5.7 billion-enough to fi-nance 15 percent and 27 percentof their respective current ac-count deficits (see Table 5.4).

Slowing demand in protecteddeveloped-country markets hasaltered the shape of developingcountries' agricultural trade. Whentrade was limited to a few com-modities consumed almost exclu-sively in the developed world, ex-port volume and prices were

Figure 5.4 Export growth ratesi, by country groupAverage annual gi'owth rate (percentage)

10

0

-10

a. Constant 978 prices.b, Newl iiidustrializing countries.

Goods and nonfactor services

Food and beverages 71965-70 70-78 65-7)) 7(1-78 65-71) 70-78 65-71) 7)1-75 65-7)) 70-78 65-71) 70-78

Industrial All develop- Low-income Middle-income Middle-income Middle-income

economies ing countries oil importers oil importers oil importers oil exportersexcept NlCc5

Table 5.4 Oil-importing developing countries' food and beveragetrade, 1970 and 1978(billions of 1978 dollars)

1978 value1970 valueDifferenceValue change due to:

Volume growthPrice decline

determined by fluctuations in in-dustrial countries' growth and byproduction changes in the devel-oping countries. Today, devel-oped countries remain the majormarkets for food exports, takingabout 70 percent of middle-in-come countries' exports and 50percent of low-income countries'exports. But developing coun-tries' food exports to these mar-kets grew much more slowly be-tween 1965 and 1978 (1.1 percenta year) than their exports to oil-exporting developing countries(6.3 percent a year), to high-in-come oil exporters (8.9 percent ayear), and to centrally plannedeconomies (4.9 percent a year).These three groups now repre-sent the most dynamic and open

5.98 3.04 2.94 38.79 20.90 17.895.03 2.91 2.12 30.38 16.65 13.730.95 0.13 0.82 8.41 4.25 4.16

2.22 0.24 1.98 16.19 6.36 9.83-1.27 -0.11 -1.16 -7.78 -2.11 -5.67

markets for the agricultural ex-ports of developing countries,weakening the historical link be-tween commodity trade and thegrowth of industrial countries.

The trade and agricultural pol-icies of developed countries influ-ence the options open to the de-veloping countries in importantways. In contrast to the situationin most developing countries, ag-riculture in the developed worldis heavily protected and sup-ported. The aim has been to sup-port farm incomes in the face ofrapid growth of earnings else-where in the economy. In someinstances, governments have alsosought to retain a farm economycapable of supplying most of theircountries' food. Countries in the

53

Low-income oil importers Middle-income oil im porters

Item Exports Imports Net Exports Imports Net

EEC have protection levels be-tween 30 and 80 percent abovethe commodity prices in worldtrade (see Box 5.4). The UnitedStates has less explicit protection,but its budgetary expenditures asa percentage of agricultural valueadded (38 percent) are similar tothose of the EEC (39 percent). Ja-pan has the most heavily pro-tected farmers, largely because ofimport restrictions (the domesticprice of rice, for example, is twiceits import price, and wheat andbarley are protected even more).Developed countries also subsi-dize their agriculture throughprograms of research, technicalassistance, credit, and income-maintenance schemes. By 1980,public spending on agriculture bythe EEC countries, Japan, and theUnited States was about $80 bil-lion; the EEC accounts for closeto half the total.

These protectionist policies havehad mixed consequences for thedeveloping countries. An impor-tant positive feature is that thesepolicies have stimulated researchand technical development in ag-riculture, which benefit develop-ing countries through the inter-national transfer of technology.More controversial are the lowerinternational prices for somecommodities associated with de-veloped-country disposal of ag-ricultural surplus, notably grainsand milk products. While clearlybenefiting importing developingcountries in the short run, in thelonger term the lower prices maydepress farm incentives for localfood production and stimulatepermanent changes in prefer-ences for foods which cannot belocally produced. The benefitstend to accrue largely to a rela-tively affluent urban minority ofconsumers, while adverse effectsare felt by the poor rural majority.

For developing-country export-ers, the impact of agricultural

54

protection is clearly negative inseveral respects (see Box 5.7). De-veloping-country exporters havebeen hard hit by this legislatedinability to expand into tradi-tional industrial-country markets.For example, the share of devel-oping-country sugar exports goingto the developed word has fallenfrom 88 percent in 1970 to 64 per-cent in 1978. Although they havebeen able to expand their sharein other developing-country sugarimport markets, this has not beensufficient to maintain the value ofsugar exports in real dollar terms,relative to 1970. Moreover, strongcompetition from the industrialcountries' surplus has capturedincreasingly large shares of thegrowing market in those devel-oping countries that import sugar;46 percent in 1978 compared with24 percent in 1970.

By comparison with trade inmanufactures, little progress wasmade in reducing agriculturalprotection in the most recent(1975-78) Tokyo Round of multi-lateral trade negotiations. That lack

of progress is compounded by thefact that nontariff measures areoften used to exclude developingcountries' agricultural exports.

Governments in industrialcountries are sensitive to the pro-tectionist demands of their ownfarmers, and this factor exacer-bates the uncertainties and risksof agricultural trade. For exam-ple, the EEC is considering waysto reduce the budgetary costs ofthe common agricultural policynot by reducing protection, butby extending tariff barriers to im-ported vegetable oils. Such meas-ures, particularly if they lead tohigher domestic production, asthey have done with sugar, arebound to harm efficient produc-ers in the developing world andto force consumers in developedcountries to pay even higher pricesfor food.

Protection also creates substan-tial volatility in world prices. Byinsulating domestic producers andconsumers from external marketrealities, governments are placingthe burden of balancing global

Box 5.7 Measuring the impact of agricultural protectionThe International Food Policy ResearchInstitute (IFPRI) has completed the mostrecent study of the impact of protectionon agricultural trade. Based on data ofthe mid-1970s, their study concluded thata 50 percent reduction in OECD devel-oped-country trade barriers (affectingsome 99 agricultural commodities) wouldincrease agricultural exports from somefifty-six developing countries by about 11percent, equivalent to $3 billion of extrasales at 1977 prices. Long-term benefitsmight considerably exceed these esti-mates, if a more liberal trade environ-ment (that exporters had confidencewould be sustained) encouraged new in-vestment and expanded specialized pro-duction over a longer period.

The distribution of benefits from tradeliberalization reflects the uneven patternof protection, the patterns of conipara-

tive advantage of the various develop-ing-country producers, and their exportpolicies. Roughly a third of extra exportrevenues would accrue to sugar produc-ers; beverages, tobacco, and meat prod-ucts account for another third. Close to60 percent of the additional revenueswould accrue to Latin American produc-ers, over 20 percent to countries in Asia,and 10 percent each to sub-Saharan Af-rica and the North Africa and MiddleEast region. Benefits would also be con-siderable for smaller exporting countriesnot included in the sample of fifty-six(which were selected on the basis of apopulation of 4 million or more in mid-1975), for example, Cuba, Jamaica, andMauritius (as sugar producers), Cyprus(wine), Zimbabwe (tobacco), and CostaRica (meat, coffee).

supply and demand on a rela-tively small part of the market.Since it is disproportionately thedeveloping countries who partic-ipate in the open world market,they have to bear the bulk of thisburden. Fluctuating commodityprices make planning more diffi-cult and wrong investment deci-sions more likely. One legacy ofrecord sugar prices in 1974 was aproliferation of costly import-sub-stituting schemes for growingsugar. They continue to act as adrain on budgets and, as heavydebt-service payments fall due, onforeign exchange.

A considerable degree of insta-bility of agricultural commodityprices is, however, inherent intheir underlying demand andsupply relationships. Neither de-mand nor supply are very sensi-tive to short-run price changes;on the other hand, the demandfor agricultural raw materials inparticular varies significantly withfluctuations in industrial countrygrowth, and the supply of theseproducts is often quite variabletoo. It can be affected by weatherconditions, disease, and pests inmajor producing areas, and by thelong gestation period which sep-arates plantings of new tree cropsfrom the period of maximumyield. Commodity prices there-fore often rise and fall dramati-cally and impart a high degree ofinstability to the economies ofcountries heavily dependent onthese commodities for their ex-port earnings.

In addition to various protec-tion measures taken by the con-suming countries to insulatethemselves from price fluctua-tions, efforts to deal with un-stable commodity prices havetaken the form of long-term bi-lateral trade agreements and mul-tilateral commodity agreements.The latter have met with limitedsuccess for a variety of reasons

Box 5.8 The "food crisis" and its food security aftermathBetween 1972 and 1974 the world priceof wheat and rice roughly trebled from$60 to $200 a ton and from $130 to $500a ton, respectively; fertilizer prices shotup still faster from $50-$75 to $300 a ton.These unprecedented increases in theprices of the world's most basic foodsand of a critical input needed to increasefood production were dramatic manifes-tations of the "world food crisis." Theywere seenalong with anxieties aboutland availability and reports of hungerand famineas harbingers of an im-pending collapse of the world food sys-tem, and of a future beset by dramati-cally higher and more unstable foodprices.

By 1976, however, the prices of wheat,rice, maize, and fertilizer had fallen asdramatically as they had risen. Wheatprices in particular were below their pre-1972 level. It was clear that a fundamen-tal shortage of food was not in the offing.Prices of cereals and most agriculturalcommodities are now not expected to bemuch above present levels until well intothe 1990s.

In retrospect, the factors causing thefood crisis were avoidable. Over the pe-riod from 1955 to 1972, large publicly held

grain stocks in the developed countriesand heavy food aid shipments had cre-ated an environment in which some de-veloping countries had little incentive toproduce food and fertilizer. At the sametime, developed-country domestic poli-cies led to the atrophy of efficient grain-marketing mechanisms. The sharp risein grain stocks in 1968 led to an abruptreduction of wheat production by thegrain-exporting countries between 1969and 1971. In 1972, world grain produc-hon declined; simultaneously, large grainpurchases were made by the USSR.

Domestic markets in a number of im-porters, notably those in western Europeand the planned economies, were insu-lated by governments from the effect ofworld prices. Consequently, the burdenof adjustment fell most heavily on thosewho were not protected, including manydeveloping economies and the grain-ex-porting countries (where high grain pricesdramatically reduced the feeding of grainto livestock). Grain stocks were not suf-ficient to contain prices, which rose rap-

idly. Fearing a shortage, many countriesbought more grain than they needed,driving prices still higher.

The food crisis taught some painful butimportant lessons. A repeat of the 1972-74 situation seems unlikely owing to thefollowing responses to it:

Much greater emphasis has beengiven to increasing food production, theessential long-term source of food secu-rity.

More widely held and carefullymonitored stocks have been established.

Grain markets are now functioningmore effectively than in the past. Thefeeding of grain to livestock and deci-sions to buy, to sell, to store, and to tradegrain are more responsive to pricechanges.

Countries have imposed limits on theamounts of grain they will export beforereviewing the grain market.

Countries are giving more thoughtto the broader impact on food securityof production and acreage adjustmentspreviously considered to be of purely do-mestic concern.

Governments have also reached anumber of international agreements thatimprove world food security:

An International Emergency Food Re-serve of 500,000 tons was established in1976, and contributions of 588,000 tonswere made to it in 1981.

A new Food Aid Convention was ne-gotiated in 1980 which raised the mini-mum annual contribution of food aid from

4.2 to 7.6 million tons.The IMP Food Facility was established

in 1981 as an extension of the IMP Coin-pensa tory Fund Facility to provide financial

assistance to offset fluctuations in coun-tries' food import bills either because ofshortfalls in domestic production orhigher world food prices.

Whether these changes, together withincreases in world stocks, provide "ad-equate world food security" is still beingargued; further mechanisms to promotestability and security may emerge. Butthe food crisis had the additional effectof increasing awareness among plannersand policymakers that a properly func-tioning international world food marketalso contributes substantially to im-

proved world food security.

55

incomplete participation by allproducing and consuming coun-tries, lack of agreement on targetprices, limited financial supportfor buffer stocks, or lack of com-mitment by participants to the ac-tions required in the face of pricechanges. These problems are fur-ther complicated by the tendencyof most commodity prices to riseand fall together. Efforts to di-versify production within tropicalproducts as a means of achievinggreater stability have therefore notbeen very successful. And onlylimited overall stability can begained from a single commodityagreement. While more effectivecommodity agreements may beachieved, including a commonfund to support such agreements,the long-term solutions are diver-sification into other exports, suchas manufactures, and into agri-cultural products (including foods)for a more stable domestic mar-ket. In the short term, domestic

56

economic management policies toreduce the internal impact of fluc-tuating external prices can con-tribute to longer-term diversifi-cation.

Volatile prices make govern-ments more receptive to the ideaof agricultural self-sufficiency andmore prone to intervene in pro-duction and trade. The issue offood security provides anotherimportant example. To the extentthat the 1972-74 world food crisismade governments more awareof the instability caused by insu-lated agricultural markets, itswider impact may have been fa-vorable (see Box 5.8). Generally,though, governments tend to reactto instability by further protectingtheir own markets or by creatingor revamping international regu-latory mechanisms. These do in-deed mitigate some of the ad-verse effects of price cycles, butat substantial real costs, both eco-nomic and political. They do not

tackle the roots of the problem,which lie in the distorted struc-ture of trade.

To sum up, the challenge forpolicymakers in developing anddeveloped countries alike is tobreak out of the web of restrictivemeasures that impede agriculturein some countries and oversti-mulate it in others; to create atrading system in which compar-ative advantage plays a more im-portant role in production andtrade decisions; and to encouragemarket mechanisms that reducethe risks of participating in thesystem. The restructuring of in-centive policies at home and thecreation of such a trading systemwould promote the effective long-run deployment of the world'sagricultural resources and helpmeet the developing economies'pressing short-term needs for im-proved access to overseas mar-kets for their agricultural exports.