-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
1/21
Agricultural Expansion, Resource Booms and
Growth in Latin America: Implications for
Long-run Economic Development
EDWARD B. BARBIER *University of Wyoming, Laramie, WY, USA
Summary. — This paper examines a number of key issues concerning agricultural land expansion,resource booms and economic growth in Latin America. The structural characteristics of agricultural development, represented by cropland share of total land area, agricultural export
share of total exports, and cereal yields, appear to influence agricultural land expansion in theregion. Long-run agricultural land expansion may also be correlated with a ‘‘boom and bust’’pattern of economic development. Contributing factors include distorted land and resourcemarkets, ineffective property rights, and the failure to target policies to improve the efficient andsustainable management of natural resources. 2003 Elsevier Ltd. All rights reserved.
Key words — agricultural land expansion, Dutch disease, economic growth, Latin America,
Resource booms
1. INTRODUCTION
Recent research suggests three worryingtrends with regard to the role of naturalresources in long-term economic developmentin Latin America. First, natural-resource com-modity booms in the region appear to beaccompanied by declining GDP per capita.This may be the result of a potential ‘‘Dutchdisease’’ effect; i.e., the impact of the ‘‘boom’’ isto divert economic resources from more inno-vative, manufacturing sectors to less innova-tive, primary sectors. Second, rapid land usechanges are occurring in Latin America, char-
acterized mainly by the conversion of tropicalforests and other natural habitat to agriculture.Empirical evidence suggests however, that overthe long run agricultural land expansion maybe correlated with lower, rather than higher,levels of GDP per capita. Finally, excessiveexploitation of land and natural resourcesacross the region is being promoted by policiesthat distort land and resource markets, inef-fective property rights and the failure to targetpolicies to improve the efficient and sustainablemanagement of natural resources.
This paper explores the economic determi-nants of these three key trends concerningagricultural land expansion, resource booms
and growth in Latin America. Specifically, thepaper focuses on recent empirical and theoreti-
cal work that addresses the following questions: ––What are the main factors determininglong-run agricultural land expansion inLatin America compared to other tropicalregions?
––Given the importance of natural resource-based sectors for most economies in LatinAmerica, what has been the impact of
www.elsevier.com/locate/worlddev
World Development Vol. 32, No. 1, pp. 137–157, 2004 2003 Elsevier Ltd. All rights reserved
Printed in Great Britain0305-750X/$ - see front matter
doi:10.1016/j.worlddev.2003.07.005
*Earlier versions of this paper were presented at The
World Bank Session on ‘‘Measurement Issues in NaturalResources, Development and Poverty’’ at the Confer-
ence on ‘‘Natural Capital, Poverty and Development.’’
The Munk Centre for International Studies, University
of Toronto, Canada, September 5–8, 2001 and at the
Workshop on ‘‘Growth and Environment: Macro-Eco-
nomic Perspectives,’’ Inter-American Development
Bank, Washington, DC, May 29, 2001. I am grateful for
comments provided by Walter Arensberg, Kirk Hamil-
ton, Glenn-Marie Lange, Ramn Lpez, Diego Rodriguez
and an anonymous referee, as well as to participants at
the above two meetings. The paper benefited from
research assistance provided by Seong-Hee Kim and
editorial assistance by Margie Reis. Final revision
accepted: 23 July 2003.
137
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
2/21
price-induced ‘‘resource booms’’ on eco-nomic growth in the region?
––Is there also an inherent ‘‘boom and bust’’pattern of economic development associatedwith agricultural land expansion, and if so,
do economic policies in Latin America exac-erbate this problem?
The outline of the paper is as follows. Section2 provides a brief overview of land use anddeforestation trends in all tropical regions aswell as in Latin America. Section 3 discusses acrosscountry analysis of the effects of growth,income per capita and other macroeconomicfactors on agricultural land expansion indeveloping regions, including Latin America.Section 4 summarizes recent theories and evi-dence as to why resource booms in Latin
America may not lead to higher growth rates,and examines the relationship between long-run agricultural land expansion in LatinAmerica and economic development. Section 5looks at the role of current economic policies ininfluencing these relationships. The concludingsection discusses the possible policy implica-tions for Latin America.
2. LAND USE AND DEFORESTATION
TRENDS IN ALL TROPICAL COUNTRIESAND LATIN AMERICA
Table 1 displays global tropical deforestationtrends over 1980–2000. There appears to be asubstantial difference in deforestation, particu-
larly in Latin America, between the 1981–90and 1991–2000 periods. Over 1981–90, the areaof tropical forests cleared on average each yearin Latin America was 7.4 million ha, almost asmuch as the area deforested in Asia and Africa
put together. Within Latin America, the mainsource of total deforestation occurred in SouthAmerica, around 6.2 million ha annually. But,the fastest rate of deforestation was takingplace in Central America and Mexico, with1.5% of the forests in the region being clearedannually. Over 1991–2000, tropical deforesta-tion appears to have slowed globally, andespecially in Latin America. Deforestation inthe region has declined by over 40% to 4.3million ha annually. The sharpest decline hasoccurred in South America, where annual
deforestation has fallen to 3.4 million ha, or0.4% per year. Central America and Mexicostill has the fastest rate of deforestation, at1.4% per year, and approximately 1 million hais being cleared annually in this region (mainlyin Mexico). In the Caribbean, the modestdeforestation across the region appears to havebeen offset by an increase in forest area in Cubaby 28,000 ha.
Total deforestation across Latin Americaremains concentrated in a few countries: Brazil(2.2 million ha annually, or a 0.4% deforesta-
tion rate), Mexico (0.6 million ha, or a 1.1%rate), Peru (0.3 million ha, or a 0.4% rate),Colombia (0.2 million ha, or a 0.4% rate), andBolivia (0.2 million ha, or a 0.3% rate). This isnot surprising, given that these five countriescontain 86% of the total tropical forest area
Table 1. Global tropical deforestation trends, 1980–2000
Region Land area
(million
ha)
Forest cover Annual
deforestation
1981–90
Estimated
annual deforestation
1991–20
1980
(million
ha)
1990
(million
ha)
Million ha % per
annum
Million ha % per
annum
Africa 2,236.1 568.6 527.6 4.1 0.7 5.3 1.0
Asia and Pacific 892.1 349.6 310.6 3.9 1.2 2.5 0.8
Latin America and
Caribbean
1,650.1 992.2 918.1 7.4 0.8 4.3 0.5
C. America and
Mexico
239.6 79.2 68.1 1.1 1.5 0.97 1.4
Caribbean 69.0 48.3 47.1 0.1 0.3 )0.02 )0.1
Trop. South
America
1,341.6 864.6 802.9 6.2 0.7 3.37 0.4
Total 4,778.3 1,910.4 1,756.3 15.4 0.8 12.2 0.7
Source: FAO (1993), except for estimated annual deforestation 1991–2000 (FAO, 2001).
WORLD DEVELOPMENT138
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
3/21
found in Latin America. In fact, Brazil aloneaccounts for 60% of Latin America’s tropicalforests. Thus the slowdown in deforestation inBrazil over 1991–2000 is largely responsible forthe decline in overall tropical deforestation in
Latin America compared to the 1981–90period.
The available evidence suggests that in mostdeveloping economies the decline in forest andwoodlands is mainly the result of land conver-sion, in particular agricultural expansion forcrop production (FAO, 1997). Land expansionoccurring in tropical regions appears to berelated to structural features of the agriculturalsectors of developing economies, such as lowirrigation and fertilizer use as well as poor cropyields. Low agricultural productivity and input
use reflect poor agricultural intensification anddevelopment, which in turn mean more pres-sure is put on conversion of forests and othermarginal lands for use in crop production.
Table 2 indicates both the current and futuredependence of developing countries on agri-cultural land expansion for crop production.Over 1970–90 increased harvested areaaccounted for 48% of the additional crop pro-duction in Latin America. This dependence onexpanding harvested land was the highest forany developing region over this period.
Although improved cropping intensity andyields are expected to reduce Latin America’sdependency on agricultural land expansionover 1990–2010, about 29% of the contributionto total crop production increases in the regionwill be derived from expansion of cultivatedland.
Fischer and Heilig (1997) combined theresults of the FAO (1995) study summarized inTable 2 with recent UN population projectionsto estimate the demand for additional culti-vated land in developing countries in 2050.
Their results are indicated in Table 3. Alldeveloping countries are expected to increasetheir demand for cultivated cropland consid-erably, leading to extensive conversion of for-ests and wetlands. In South America, cultivatedland area is expected to increase by over 50%by 2050, with about 70% of the new landcoming from deforestation and wetland con-version. Agricultural land will expand byaround 40% in Central America and theCaribbean, but 80% of this new land is likely tobe from forest and wetland conversion.
Pressures on the land and natural resourcebase in developing countries are also likely tocome from the overall demands of economic
development. For example, many low-incomeand lower middle-income economies are highlyresource dependent (Barbier, 1999). Not only dothese economies rely principally on directexploitation of their resource bases through
primary industries (e.g., agriculture, forestry,fishing, etc.) but also over 50% or more of theirexport earnings come from a few primarycommodities. Natural capital––the value of thenatural resource endowment of a country––isparticularly important in the developing world.For low-income countries dependent on exportrevenues from primary commodities (otherthan petroleum), 20% of their national wealthcomprises natural capital (World Bank,1997). 1 In the poorest countries, agriculturalcropland comprises around 80% of the natural
capital wealth. The share of natural capital intotal wealth is 11% in the Caribbean, 9% inSouth America and 6% in Central America.Clearly, exploitation of the natural resourceand land base of these countries is essential fortheir long-term economic development efforts.
3. FACTORS DETERMININGAGRICULTURAL LAND EXPANSION
The discussion of the previous section sug-
gests that the major cause of forest loss indeveloping countries is conversion to agricul-ture. Thus a crosscountry analysis of agricul-tural land expansion should also provideinsights into the factors influencing tropicaldeforestation. Equally, previous studies of tropical deforestation may be able to suggestsome of the possible effects of growth, incomeper capita and other macroeconomic factors onagricultural land expansion in developingregions, including Latin America. Four distinctanalytical frameworks have been proposed in
the economics literature for motivating cross-country estimations of the causes of agriculturalland conversion and tropical deforestation: theenvironmental ‘‘Kuznets’’ curve hypothesis,competing land use models, forest land conversionmodels, and institutional models (Barbier &Burgess, 2001). As the following brief reviewindicates, these analytical frameworks enable usto focus on certain key economic factors thatmay determine tropical agricultural landexpansion and to choose the appropriate vari-ables to include in our crosscountry regres-
sion. 2The environmental Kuznets curve (EKC)
hypothesis states that an environmental ‘‘bad’’
AGRICULTURAL EXPANSION AND RESOURCE BOOMS 139
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
4/21
Table 2. Trends in crop production and harvest area in developing regions
Region Crop production
1970–90 Contribution (%) of
increases in
1990–2010 Contribution (%) of
increases in
1990–2010 C
in
Yields Harvested area Yields Harvested area Arable land
Sub-Saharan Africa 53 47 53 47 64
Near East and North Africa 73 27 71 29 31
East Asiaa 59 41 61 39 82
South Asia 82 18 82 18 22
Latin Americab 52 48 53 47 60
All developing countries 69 31 66 34 62
Source: FAO (1995).a Excludes China.b Includes the Caribbean.
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
5/21
first increases, but eventually falls, as the percapita income of a country rises. There are anumber of recent theoretical models explainingwhy such an inverted-U relationship betweenincome and environmental ‘‘bads’’ might hold(e.g., Andreoni & Levinson, 2001; McConnell,1997; Stokey, 1998). Although the EKC modelhas generally been applied to pollution prob-lems, there have been a number of recentstudies that have also examined whether thishypothesis also holds for global deforestation(e.g., Antle & Heidebrink, 1995; Cropper &
Griffiths, 1994; Koop & Toole, 1999; Panayo-tou, 1995; Shafik, 1994). The basic EKC modelfor deforestation is usually
F it F it 1 ¼ F ðY it ; Y 2it ; zit Þ
¼ a1Y it a2Y 2it þ zit b þ eit ; ð1Þ
where F it F it 1 is the change in the forest stockover the previous period (which is negative if deforestation is occurring), Y it is per capitaincome and zit is a 1 n vector that includesother explanatory variables, such as population
density or growth and other macroeconomicvariables. 3
The application of the EKC model (1) toexplain deforestation trends across countrieshas produced mixed results. When the model istested for both temperate and tropical coun-tries, it is inconclusive (Antle & Heidebrink,1995; Panayotou, 1995; Shafik, 1994). Whenapplied to just tropical countries, the inverted-U relationship tends not to hold for all coun-tries but may apply to specific regions. Forexample, Cropper and Griffiths (1994) find
some evidence that the EKC model is relevantto Latin America and Africa. But, for each of these regions the turning point––the per capita
income level at which the deforestation rate iszero and is about to decline––is generally twoto four times higher than the average per capitaincome for that region. Nevertheless, as Table 1has indicated, tropical deforestation in mostdeveloping regions, especially Latin America,has slowed down recently, and one possibleexplanation may be the EKC hypothesis.
Other empirical analyses have taken as theirstarting point the hypothesis that forest loss intropical countries is the result of competingland use, in particular between maintaining the
natural forest and agriculture (e.g., Barbier &Burgess, 1997; Ehui & Hertel, 1989). As indi-cated in Tables 2 and 3, the evidence acrosstropical regions––and in particular LatinAmerica––is that substantial conversion of forest and woodlands to agriculture is occur-ring. From an economic standpoint, given thetime and effort required to reestablish tropicalforest (where this is ecologically feasible) suchconversion implies that potential timber andenvironmental benefits from forestland areirreversibly lost. Therefore, competing land use
models usually include some measure of the‘‘price’’ or opportunity cost of agriculturalconversion and deforestation in terms of theforegone benefits of timber production andenvironmental benefits from forestland
F it F it 1 ¼ ADðvit ; zit Þ; o A
D=ovit
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
6/21
forestland conversion decision of agriculturalhouseholds. There have been several suchapplications to Latin American countries (e.g.,Anderson, Granger, Eustaquio, Weihold, &Wunder, 2002; Barbier, 2002; Barbier & Bur-
gess, 1996; Chomitz & Gray, 1996; Nelson,Harris, & Stone, 2001). Such approaches modelthe derived demand for converted land by ruralsmallholders, and assume that the householdseither use available labor to convert their ownland or purchase it from a market. This in turnallows the determinants of the equilibrium levelof converted land to be specified. In suchmodels, the aggregate equilibrium level of cleared land across all households is usuallyhypothesized to be a function of output andinput prices and other factors affecting aggre-
gate conversion
ADit ¼ ADð p it ;w Lit ; wit ; xit ; zit Þ;
o AD
o p it > 0;
o AD
ow Lit 0; ð3Þ
where p is the price of agricultural output, w L isrural wage (labor is a key component in landclearing), w is a vector of other inputs, x arefactors influencing the ‘‘accessibility’’ of forestareas (e.g., roads, infrastructure, distance to
major towns and cities), and as before zit rep-resents other economic explanatory variables.
Although both the competing land use andforestland conversion models appear to workwell for specific tropical forest countries, it isdifficult to obtain time series data acrosscountries for key price data in the respectivemodels, i.e., vit in Eqn. (2) or p it and especiallywit in Eqn. (3). Crosscountry data on important‘‘ x’’ variables, such as rural road expansion androad building investments, are also hard tofind. As a result, crosscountry analyses of Eqns.
(2) and (3) have tended to leave out prices and‘‘ x’’ factors, or employed proxies. For example,in their empirical estimations for deforestationacross all tropical countries, Barbier and Bur-gess (1997) employed roundwood productionper capita as a ‘‘proxy’’ for vit in Eqn. (2), aspreferred measures of the ‘‘opportunity cost’’of conversion (e.g., land values, timber rents)are not available across countries. Similarly,Southgate (1994) used annual populationgrowth, agricultural export growth, crop yieldgrowth and a land constraint dummy to explain
annual agricultural land growth across LatinAmerica over 1982–87. He found that popula-tion and agricultural export growth were posi-
tively related to land expansion, whereas yieldgrowth and the land constraint were negativelyrelated. Other studies have also demonstratedthat structural agricultural, economic andgeographic factors, which vary from country to
country, are significant in explaining the dif-ferent land conversion trends across countries(e.g., Barbier & Burgess, 1997; Deacon, 1999).These ‘‘structural’’ variables include agricul-tural yield, cropland share of land area, agri-cultural export share, and arable land percapita, which capture country-by-country dif-ferences in agricultural sectors and land usepatterns, as well as other exogenous macro-economic variables, such as population growth,rural population density, GDP growth, realinterest rates and debt. These factors may be
particularly significant explanatory variables ina crosscountry analysis, if the difficulty inobtaining crosscountry time series data on keyvariables, such as rural wages, roads, otherinput prices, makes it impossible to includevariables representing agricultural returns or‘‘accessibility’’ of forest lands in the model.
Finally, recent empirical analyses at both thecountry and crosscountry level have exploredthe impact on tropical deforestation of institu-tional factors, such as land use conflict, securityof ownership or property rights, political sta-
bility, and the ‘‘rule of law’’ (e.g., Alston, Li-becap, & Mueller, 1999, 2000; Deacon, 1994,1999; Godoy et al ., 1998). The main hypothesistested is that such institutional factors areimportant factors explaining deforestation
F it F it 1 ¼ F ðqit ; zit Þ; ð4Þ
where qit is a vector of institutional factors andand zit represents other economic explanatoryvariables.
Although such models have demonstratedthe importance of institutional factors in
determining deforestation, it is unclear howmuch weight should be given to such factors inpreference to explanatory variables identifiedby other approaches to crosscountry analysesof forest loss. Nevertheless, the failure toinclude institutional factors in a crosscountryanalysis of land use change may mean thatpotentially important explanatory variableshave been omitted.
In sum, the four main analytical frameworksmotivating crosscountry analysis of tropicaldeforestation and land use change emphasize
the following key variables: (a) from Eqn. (1),the inclusion of per capita income and incomesquared terms to test for a possible EKC rela-
WORLD DEVELOPMENT142
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
7/21
tionship; (b) in the absence of adequate cross-country data for the price and ‘‘ x’’ variables inEqns. (2) and (3), the inclusion of the synthesismodel should include certain ‘‘structural’’variables ðsit Þ, such as agricultural yield, crop-
land share of land area, agricultural exportshare, and arable land per capita, to capturecountry-by-country differences in agriculturalsectors and land use patterns; and finally, (c)from Eqn. (4) the inclusion of key institutionalfactors thought to influence land expansion anddeforestation.
Thus a possible model for a crosscountryanalysis of the possible effects of growth,income per capita and other macroeconomicfactors on agricultural land expansion indeveloping regions might look like:
Ait Ait 1 ¼ AðY it ; Y 2it ; sit ; zit ; qiÞ;
for country i at time t ; ð5Þ
where Ait Ait 1 represents expansion in agri-cultural land area, Y it is per capita income, s it isa vector of ‘‘structural’’ variables, such asagricultural yield, cropland share of land area,agricultural export share, and arable land percapita, and zit represents other exogenousexplanatory variables, such as population den-sity or growth, GDP growth and other macro-
economic variables. Finally, as institutionalfactors ðqiÞ tend to be invariant with time, twoversions of the model can be tested, one with-out and one including qi.
Eqn. (5) was estimated through a panelanalysis of tropical agricultural land expansionin Latin America over 1961–94, with thedependent variable being the percentage annualchange in agricultural land area. 4 The EKCvariables ðY it ; Y
2it Þ are represented by gross
domestic product (GDP) per capita in constantpurchasing power parity (1987 $) and by GDP
per capita squared, respectively. The structuralvariables ðsit Þ are cereal yield, cropland share of total land area, agricultural export share of total merchandise exports and arable land percapita. The additional explanatory variablesðzit Þ are population and GDP growth. Thesource of data used for these variables was theWorld Bank’s World Development Indicators,which has the most extensive data set for keyland, agricultural and economic variables fordeveloping countries over the period of analy-sis.
Table 4 indicates the results for LatinAmerica both without institutional factors, qi,and with the inclusion of three institutional
variables: a corruption index, a property rightsindex and a political stability index. Theseindices were obtained from the Levine– Loayza–Beck data set used in Beck, Levine, andLoayza (2000) and Levine, Loayza, and Beck
(2000), which are available from the EconomicGrowth Research Group of the World Bank.The corruption and property rights indices aredirectly from the Levine–Loayza–Beck data setand are averaged over 1982–95. The politicalstability index was created as a composite indexof the average number of revolutions and coups(averaged over 1960–90), average number of assassinations per million population (averagedover 1960–90), and an index of ethnic frac-tionalization (averaged over 1982–95). As theseindices were not available for all tropical
countries in the original sample, the inclusionof these institutional factors reduced the samplesizes of the regressions considerably. In addi-tion, the three indices are time invariant, andwith their inclusion in addition to the originalexplanatory variables of the model, fixed effectsregressions cannot be run. 5
Both one-way and two-way fixed and randomeffects models were tested for the sample of Latin America countries. Table 4 displays theresults for the preferred models and the relevantstatistics. In the table, the parameter estimates
are reported as elasticities, in order to facilitatecomparison of the effects of the different vari-ables, which are in different units. In theregression without institutional indices, only thestructural variables appear to explain agricul-tural expansion in Latin America. Growth inagricultural land area increases with croplandshare of total land area and agricultural exportshare but declines with cereal yield. But, neitherper capita income nor GDP per capita squaredis significant. Thus, the results of this regressionlead to rejection of the EKC hypothesis for
explaining the pattern of agricultural landexpansion in Latin America.
In comparison, the inclusion of institutionalfactors has a considerable influence on theanalysis. First, the EKC hypothesis can nolonger be rejected, although the estimated EKCturning point of $4,946 is one third larger thanthe average per capita income of $3,675 for thesample of Latin American countries. Thusthere is some evidence for Latin America thatagricultural land expansion will start to slowdown as per capita income rises in the region,
but clearly most countries are still well belowthe ‘‘turning point’’ level of income for this tooccur. Second, of the institutional variables,
AGRICULTURAL EXPANSION AND RESOURCE BOOMS 143
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
8/21
political instability appears to have a significantand positive impact on agricultural expansionin Latin America. 6 Third, some structuralvariables are still important in explaining landexpansion. Growth in agricultural land area inLatin America increases with cropland share of total land area and agricultural export share.
Finally, population growth appears also to beassociated with greater land expansion in theregion.
The above results for the synthesis modelprovide interesting additional insights intoagricultural land expansion in Latin America.First, the pattern of agricultural development,as represented by such structural variables ascropland share of total land area, agriculturalexport share of total exports, and to some
extent, cereal yields, appears consistently toinfluence agricultural land expansion. Thiswould suggest that policies that influence the
Table 4. Panel analysis of agricultural land expansion in tropical Latin America, 1961–94
Dependent variable: agricultural land expansion
(% annual changea)
Elasticity estimatesb
Without institutional factors With institutional factors
Explanatory variables ( N ¼ 319) ( N ¼ 233)
GDP per capitac )0.388 7.943
(PPP, constant 1987 $) ()0.109) (2.595)**
GDP per capita squared 0.050 )3.499
(0.036) ()2.328)*
GDP growth 0.009 )0.066
(% annual change) (0.127) ()0.862)
Population growth )0.734 1.421
(% annual change) ()1.234) (2.271)*
Cereal yield )2.632 )0.623
(kg per hectare) ()2.028)* ()0.929)
Cropland share of land 5.565 0.789(% of land area) (2.485)** (2.343)*
Agricultural export share 0.462 0.383
(% of merchandise exports) (2.098)* (2.948)**
Arable land per capita )0.263 0.358
(hectares per person) ()0.163) (0.83)
Corruption index 0.249
(high¼ 0, low¼10) (0.345)Property rights index )0.363
(high¼ 5, low¼1) ()0.425)Political stability index 0.494
(high¼ 0, low¼1) (2.220)*
Kuznets curve No Yes(Turning point estimate) ($17,359) ($4,946)
F -test for pooled model 3.323**
Breusch–Pagan (LM) test 11.23**
Hausman test 15.52*
Durbin–Watson statistic 2.045
Adjusted R2 0.128
Preferred model One-way fixed effects OLS
a Mean for the sample without institutional factors is 0.75%; mean for the sample with institutional factors is 0.66%.bt -Ratios are indicated in parentheses.
c Mean for the sample without institutional factors is $3,654; mean for the sample with institutional factors is $3,675.
PPP is purchase power parity.*Significant at 5% level.
**Significant at 1% level.
WORLD DEVELOPMENT144
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
9/21
structure of agricultural production and overalldevelopment in the region could have asubstantial influence on the future demand forcultivated land. For example, substantialincreases in cereal yields and improved agri-
cultural intensification could mean that theincreases in cultivated land required by LatinAmerica by 2050 may not be as large as pre-dicted in Table 3. Second, although the exis-tence of an EKC effect for agriculturalexpansion in Latin America appears to behighly sensitive to the model specification, therelatively low estimated EKC turning point forLatin America suggests that broader structuraleconomic changes and development in theregion could start to slow agricultural landgrowth as well. The fact that political stability
may also be an important institutional influ-ence on slowing land expansion indicates thatpolitical development may confer importanteconomic and environmental benefits to theregion. Finally, the positive relationshipbetween the share of agricultural exports intotal merchandise exports and land expansionin Latin America confirms that the increasedresource dependence of an economy may becorrelated with greater exploitation of its nat-ural resource and land base. As will be dis-cussed in the next section, a price-induced
‘‘resource boom’’ may not always lead towidespread economic benefits, even in resource-dependent economies.
4. RESOURCE BOOMS AND ECONOMICGROWTH: IMPLICATIONS FOR LATIN
AMERICA
So far, we have established that agriculturalland expansion and natural resource exploita-tion by primary sector activities are funda-
mental to economic development in LatinAmerica. But, if per capita income is to besustained or increased in these economies,especially with population increases, then anydepreciation of natural resources must be offsetby investment in other productive assets. Thisimplies that natural resource rents must bereinvested in the economy by both managingnatural resources so as to maximize rents andchanneling those rents into productive invest-ments elsewhere in the economy. Although itwould seem that the windfall profits generated
by resource price booms would be beneficial tothis process, as we will discuss further below,this may not be the case for resource-abundant
developing countries, especially in LatinAmerica.
In sum, even though an economy is endowedwith abundant natural resources, the countrymay not necessarily be exploiting this natural
wealth efficiently and generating productiveinvestments. Or, as Wright (1990, p. 666) hassuccinctly put it: ‘‘there is no iron law associ-ating natural resource abundance with nationalindustrial strength.’’ In fact, recent evidencesuggests that resource-abundant countries,especially developing economies, may not bebenefiting economically from this comparativeadvantage. Many low-income and lower mid-dle-income economies that can be classified ashighly resource dependent today also currentlydisplay low or stagnant growth rates (Barbier,
1999). Crosscountry analysis has confirmedthat resource-abundant countries––i.e., coun-tries with a high ratio of natural resourceexports to GDP––have tended to grow lessrapidly than countries that are relativelyresource poor (Sachs & Warner, 1997). Econ-omies with a high ratio of natural resourceexports to GDP in 1971 also tended to have lowgrowth rates during the subsequent period1971–89 (Sachs & Warner, 1995).
Such recent empirical evidence might beconsidered surprising, given the commonly held
view that abundant natural resources ought tobe the basis for economic expansion for thosecountries fortunate to have such a richendowment. For example, as historical prece-dent, empirical evidence clearly shows that theorigins of rapid industrial and economicexpansion in the United States over 1879–1940were strongly linked to the exploitation of abundant nonreproducible natural resources,particularly energy and mineral resources(Romer, 1996; Wright, 1990). In particular,during 1880–1920, the intensity of US manu-
facturing exports in terms of nonreproducibleresources grew both absolutely and relative tothe resource-intensity of imports. There is alsoevidence however that there were other factorsthat made this historical situation in the UnitedStates unique. For example, Wright (1990)maintains that, over this era (a) the UnitedStates was not only the world’s largest mineralproducing country but also one of the world’slargest countries and markets; (b) high inter-national transport costs and tariff barriers formanufactured goods compared to highly effi-
cient and low cost domestic transportationmeant that the United States was a vast freetrade area for internal commerce and industrial
AGRICULTURAL EXPANSION AND RESOURCE BOOMS 145
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
10/21
expansion that benefited from ‘‘economic dis-tance’’ from the rest of the world; and (c)because of the quantities of resources that wereavailable combined with the large internalmarkets for goods, increasing investment in
basic technologies for extracting and processingnatural resources was highly profitable. AsWright (1990, pp. 665 & 661) suggests: ‘‘theabundance of mineral resources, in otherwords, was itself an outgrowth of America’stechnological progress,’’ and in turn, ‘‘Ameri-can producer and consumer goods were oftenspecifically designed for a resource-abundantenvironment.’’
It is doubtful however that the unique cir-cumstances that allowed the United States toachieve ‘‘congruence’’ between intensive
resource use and basic processing and manu-facturing technologies over 1879–1940, andthus attain rapid economic expansion, areapplicable to resource-abundant developingeconomies today. For one, after 1940, thisunique ‘‘congruence’’ had clearly ended for theUnited States, largely due to changes in theinternational economy, even though the UnitedStates still had abundant resources. As Wright(1990, p. 665) points out:
the country has not become ‘‘resource poor’’ relative to
others, but the unification of world commodity mar-kets (through transportation cost reductions and elim-ination of trade barriers) has largely cut the linkbetween domestic resources and domestic industries. . .To a degree, natural resources have become commodi-ties rather than part of the factor endowment’ of indi-vidual countries.
As some researchers have pointed out, thechanged international conditions during thepostwar era may have also affected the role of primary-product export promotion as the
‘‘engine of growth’’ for developing economies.During this era, the main source of economicgrowth in developing countries has not beenprimary-product based exports but labor-intensive manufactured exports (Findlay, 1996;
Findlay & Wellisz, 1993). 7
Not only are the conditions for ‘‘congru-ence’’ between resource abundance, technolog-ical progress and industrial expansion lackingin developing economies today, but some havealso maintained that increased economicdependence on resource exploitation may be
detrimental to innovation and growth. Forexample, recent explanations of the limitationsof resource-based development have focused on
the poor potential for such development ininducing the economy-wide innovation neces-sary to sustain growth in a small open econ-omy. Matsuyama (1992) has shown that tradeliberalization in a land-intensive economy
could actually slow economic growth byinducing the economy to shift resources awayfrom manufacturing (which produces learning-induced growth) toward agriculture (whichdoes not). Sachs and Warner (1995) also arguethat the relative structural importance of trad-able manufacturing versus natural resourcesectors in an economy is critical to its growthperformance, i.e., when a mineral or oil-basedeconomy experiences a resource boom, themanufacturing sector tends to shrink and thenontraded goods sector tends to expand. This
phenomenon is often referred to in the litera-ture as the ‘‘Dutch disease’’ effect. 8
Sachs and Warner (1999) have recentlyexamined evidence over 1960–94 for 11 majorLatin American economies to test the hypoth-esis that any natural resource booms occurringin these countries my have had a positive impacton their growth performance. 9 First, theauthors note that the main structural feature of these economies is that they have remained byand large exporter of primary commodities ormanufactured products based on these com-
modities. Second, they suggest that a significantresource boom occurred in only four of the 11countries (Bolivia, Ecuador, Mexico and Ven-ezuela), and mixed evidence of a boom inanother three (Chile, Colombia and Peru).Sachs and Warner conclude however, that inonly one of these seven countries (Ecuador) dida resource boom have a positive and lastingeffect on GDP per capita. In two countries(Chile and Colombia) there appears to be noeffect of a resource boom on economic devel-opment, and in the remaining four cases
(Bolivia, Mexico, Peru and Venezuela), theresource boom actually produced a negativeimpact on GDP per capita. On balance,resource booms appear to frustrate economicgrowth in Latin America, most likely through aDutch disease effect.
If natural resource booms are not importantcatalysts for economic development in poorercountries, then perhaps the process of resourceexploitation and land expansion occurring inthese economies is not as economically benefi-cial as it could be. That is, the structural eco-
nomic dependence of a small open low or lowermiddle income economy on exploiting its nat-ural resource endowment––in particular its
WORLD DEVELOPMENT146
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
11/21
dependence on processes of agricultural landexpansion––may not be leading to sustainedand high rates of economic growth. This maybe occurring because natural resource assets,including land, are not being managed so as to
maximize rents and/or whatever rents are beinggenerated in the economy are not being chan-neled into productive investments elsewhere inthe economy.
Brander and Taylor (1997, 1998) providesome theoretical support for this perspective.They note that overexploitation of manyrenewable natural resources––particularly theconversion of forests to agricultural land–– occurs in developing countries if propertyrights over a resource stock are hard to define,difficult to enforce, or costly to administer.
They demonstrate that opening up trade for aresource-abundant economy with an openaccess renewable resource may actually reducewelfare in that economy. As the resource-abundant country has a comparative advantagein producing the resource good, the increaseddemand for the resource good resulting fromtrade openness leads to greater exploitation,which under conditions of open access leads todeclining welfare in the long run. Brander andTaylor conclude that, as the problem lies withthe ‘‘open access’’ nature of exploitation in the
resource-abundant economy, then the first-bestpolicy would be for the developing country toswitch to more efficient resource managementpolicy through simply establishing propertyrights. As however, they acknowledge, and aswe discuss further in Section 5 below in the caseof Latin America, there are many policy andinstitutional distortions that currently workagainst such solutions in developing countries.Consequently, Brander and Taylor (1997, p.550) argue in favor of ‘‘second best approa-ches’’ such as the country imposing ‘‘a modifiedHartwick’s rule’ (see Hartwick, 1977) underwhich an exporting country that experiencedtemporary gains from selling a resource goodon world markets might re-invest those pro-ceeds in an alternative asset.’’
As we discuss further in Section 5 with respectto Latin America, current policies in resource-abundant developing economies appear not tobe ensuring that any resource rents earned arere-invested efficiently into other productiveassets in the economy. If this is the case, irre-spective of what may happen to the country’s
terms of trade or commodity prices, any initial‘‘economic boom’’ associated with land con-version or increased resource exploitation is
invariably short-lived as the extra rents gener-ated are eventually dissipated. Once the landexpansion and increased exploitation of newresource ‘‘reserves’’ comes to an end, or poorerquality land and resources are brought into
production, then some economic retrenchmentis inevitable. What we should therefore observeis that economic development in a resource-dependent small open economy displays aninherently ‘‘boom and bust’’ pattern.
Again, Brander and Taylor (1997) show thata small, open and resource-abundant economythat produces both a resource and a manufac-turing good in the long run will have such apattern of development. That is, the economywill experience early gains from trade, followedby a period of declining utility. With the spe-
cific case of Latin America in mind, in whichraw materials are often inputs into semi-pro-cessed or processed exports, Loopez (1989) alsodevelops a two-good model of a resource-richopen economy in which the open accessrenewable resource serves as an input into an‘‘enclave’’ export processing sector. Loopezshows that improvements in the terms of tradeincreases the rate of open access resourceextraction and real income to increase in theshort-run, but inevitably permanent incomefalls in the long run.
As argued throughout this paper, the classiccase of open access resource exploitation inmany developing countries is conversion of forest to agriculture. If agricultural landexpansion in these small open economies isassociated with a ‘‘boom and bust’’ pattern of economic development, then there are twopossible consequences. First, economies thathave increased their agricultural land base sig-nificantly over the long run are likely to havelower levels of GDP per capita then economiesthat have tended to reduce their dependence on
agricultural land expansion. For the lattercountries, a shrinking agricultural land basemay be evidence that tradable manufacturingand other dynamic sectors have become struc-turally more important in the economy relativeto natural resource sectors and that agricultureitself has become a more capital-intensive,productive and innovative sector. 10 Second,for those countries that are dependent onagricultural land expansion, further increases inagricultural area will tend to produce onlymodest increases in GDP per capita. Beyond a
certain point, additional increases in landexpansion will be associated with lower GDPper capita, because of the ‘‘boom and bust’’
AGRICULTURAL EXPANSION AND RESOURCE BOOMS 147
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
12/21
pattern of resource-dependent developmentdescribed above.
A fairly straightforward way of empiricallyverifying the above phenomenon is to empiri-cally estimate a relationship between GDP per
capita and some measure of long-run agricul-tural expansion. For example, if the latterindicator was some index, ait , then the abovehypotheses suggest that there may be a cubicrelationship between per capita income, Y it , andthis indicator of long run agricultural landchange:
Y it ¼ b0 þ b1ait þ b2a2it þ b3a
3it : ð6Þ
Note that b0 > 0, b1 < 0, b2 > 0, b3 < 0 andjb1j > b2 would imply that (i) countries with
increased long-run agricultural land area wouldhave lower levels of per capita income thancountries with decreased agricultural land areaand (ii) per capita income would tend to fluc-tuate with long-run agricultural land expan-sion.
Eqn. (6) was estimated through employing apanel analysis of tropical developing countriesover 1961–94. Per capita income, Y it , is againrepresented by gross domestic product (GDP)per capita in constant purchasing power parity
(1987 $). The indicator ait is an agriculturalland long-run change index, created by dividingthe current (i.e., in year t ) agricultural land areaof a country by its land area in 1961.
The results of the analysis for all tropical
countries and Latin America are shown inTable 5. For both regressions, the estimatedcoefficients are highly significant and also havethe expected signs and relative magnitudes.Thus the estimations provide some empiricalevidence that agricultural land expansion intropical countries and the Latin Americanregion conforms to a ‘‘boom and bust’’ patternof economic development. This is seen moreclearly when the regressions are used to projectrespective relationships between long-run agri-cultural land expansion and GDP per capita,
which are displayed in Figure 1.As indicated in the figure, an increase in
agricultural land expansion in the long run isclearly associated with a lower level of percapita income than decreasing agricultural landarea. For all tropical countries, the turningpoint is a long run agricultural change index of 1.2. For Latin America the turning point is 1.3.Although continued agricultural land expan-sion beyond these points does lead to a slightincrease in GDP per capita, this impact is
Table 5. Panel analysis of per capita income and long run agricultural expansion, 1961–94
Dependent variable: GDP per capita (PPP, constant 1987 $) a
Parameter estimatesb
All countries Latin America
Explanatory variables ( N ¼ 1; 135) ( N ¼ 401)14,393.37 11,525.92
(23.69)** (3.62)**
Long run agricultural land area change index (ait )c
)24,293.31 )17,535.19
()19.04)** ()2.55)**
a
2
it 15,217.53 12,156.88(11.182)** (2.42)
a3it )2,896.32 )2,692.56
()6.59)** ()2.20)**
F -test for pooled model 168.01** 151.13**
Breusch–Pagan (LM) test 6,576.23** 2,239.33**
Hausman test 6.85 1.68
Adjusted R2 0.368 0.145
Preferred model One-way random effects One-way random effects
a Mean for all countries is $2,986, and for Latin America $3,675. PPP is purchase power parity.b t -Ratios are indicated in parentheses.c Mean for all countries is 1.18, and for Latin America 1.26.
*Significant at 5% level.
**Significant at 1% level.
WORLD DEVELOPMENT148
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
13/21
short-lived. For all tropical countries, percapita income starts to fall once the land areaindex reaches 2.3; for Latin America this occurssooner at an index of 1.7. Finally, the averagelong-run agricultural land change index for alltropical countries is 1.18, whereas it is 1.26 forLatin America. As Figure 1 indicates, a com-parison of these averages suggests that agri-cultural land expansion is associated withslightly higher levels of per capita income than
in other tropical countries.
5. DO ECONOMIC POLICIES IN LATINAMERICA FURTHER AGRICULTURAL
LAND EXPANSION?
The above analysis suggests that agriculturalland expansion and resource exploitation inLatin America may not contribute to economicdevelopment because either insufficient rentsare being generated from conversion and
depletion of natural resources or the rentsearned are not being invested in productiveassets or both. Ill-defined and lack of enforce-
ment of property rights for land create ‘‘openaccess’’ conditions for exploiting land, whichmay also pose a problem. There are threepossible ways in which current policies in theregion may exacerbate these problems. First,existing land ownership and property rightregimes may encourage overuse and conversionof land. Second, government policies maycontribute to unequal resource and land access,which leads to increased frontier land conver-
sion by poorer farmers. Third, recent structuralagricultural and trade liberalization reformsmay not generate the necessary incentives tobreak the structural dependence of LatinAmerican economies on continued agriculturalland expansion. Specifically, these largelymacroeconomic policy reforms are not aimedat ensuring that any rents earned from resourceconversion and depletion are re-invested effi-ciently into other productive assets in theeconomy. Nor do these reforms address the‘‘first best’’ policy solution identified by
Brander and Taylor (1997), namely encourag-ing more efficient resource managementthrough correcting distortions in property
0
2000
4000
6000
8000
10000
12000
14000
16000
0 0.5 1 1.5 2 2.5 3
Agricultural land long-run change index
GDP percapita(constant1987 $)
All Tropical Latin America
1.26 1.18
Figure 1. Projections of agricultural land expansion and GDP per capita in tropical countries, 1961–94.
AGRICULTURAL EXPANSION AND RESOURCE BOOMS 149
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
14/21
rights and unequal access to resources, partic-ularly land.
The general open access conditions of unoc-cupied forest land is now recognized as a keycondition underlying frontier agricultural
expansion in developing countries (Pearce,Barbier, & Markandya, 1990). In Latin Amer-ica, the problem is made worse by existing landownership rules. In particular, land titling reg-ulations which essentially acknowledge forestclearing as evidence of effective occupation andownership for both agriculture and livestockraising have been documented as a major factorin frontier agricultural conversion in CostaRica, Ecuador, Honduras, Panama and otherLatin American countries (Kaimowitz, 1995;Mahar & Schneider, 1994; Peuker, 1992;
Southgate, Sierra, & Brown, 1991; Sunderlin &Rodrııgez, 1996). For example, in Costa Rica,occupation of public lands has resulted in 60%of farms lacking land title, and often competingclaims for land (Peuker, 1992). This has pro-vided an incentive to undertake activities on theland, such as clearing land of trees, whichclearly demonstrate possession. Title to landcan be obtained after 10 years of possession,and a claimant can title up to 100 hectares (ha)of land if the property is devoted to agricultureand up to 300 ha if it is devoted to cattle
ranching. The process has proved to be highlysusceptible to fraud with respect to time of occupation, the area of the land to be titled,and the actual use of the land.
Land titling regulations can also affect theinitial ‘‘opening up’’ of unoccupied frontierlands. It is often short-term extractive opera-tions––such as timber concessions, miningconcerns, large-scale commercial ranching andfarming––that are more likely to be involved ininitial frontier development. Usually it is fairlystraightforward for governments to allocate
large tracts of frontier land to commercialconcerns and individual operators for extrac-tive purposes, and often their activities receivesubsidies or other fiscal incentives of somekind. Short-term land speculators may also beencouraged in this way. Generally, the objec-tive of these extractive and speculative opera-tions is to maximize short-term resource rents;long-term investment in frontier economicdevelopment is not a major priority––particu-larly if it is difficult to acquire long-termproperty or use rights or to control illegal
occupation. As a consequence, once sufficientrents are extracted, land abandonment andselling-off is common.
As noted above, however, once the frontier is‘‘opened’’ by large-scale activities, the lack of secure property rights and general open accessconditions prevailing on the frontier inevitablyencourage rapid expansion of frontier agricul-
tural activities by small-scale farming andlandless households in search of new lands. Forexample, Schneider (1994) envisions this pro-cess as a perpetually ‘‘moving frontier.Extending roads into previously isolated forestsinitiates the land grab, which is then acceleratedby the absence of property rights. As furtherroad-building and large infrastructure projectsare established in new areas, the process repeatsitself.
In Latin America, inequalities in wealthbetween rural households also have an impor-
tant impact on land degradation and defores-tation processes. Such problems are exacerbatedby government policies that favor wealthierhouseholds in markets for key resource, such asland. First, poorer households are often unableto compete with wealthier households in landmarkets for existing agricultural land. Theresult is two segmented land markets: thewealthier rural households dominate the mar-kets for better quality arable land, whereas thepoorer and landless households either trade inless productive land or migrate to marginal
lands.Second, although poorer households may be
the initial occupiers of converted forestlandthey are rarely able to sustain their ownership.As the frontier develops economically andproperty rights are established, the increase ineconomic opportunities and potential rentsmakes ownership of the land more attractive towealthier households. Because of their betteraccess to capital markets, they can easily bidcurrent owners off the land, who in turn maymigrate to other frontier forest regions or
marginal lands.For example, in Colombia distortions in the
land market prevent small farmers fromattaining access to existing fertile land (Heath& Binswanger, 1996). That is, as the marketvalue of farmland is only partly based on itsagricultural production potential, the marketprice of arable land in Colombia generallyexceeds the capitalized value of farm profits. Asa result, poorer smallholders and, of course,landless workers cannot afford to purchase landout of farm profits, nor do they have the non-
farm collateral to finance such purchases in thecredit market. In contrast, large landholdingsserve as a hedge against inflation for wealthier
WORLD DEVELOPMENT150
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
15/21
households, and land is a preferred form of collateral in credit markets. Hence the specu-lative and nonfarming benefits of large landholdings further bid up the price of land, thusensuring that only wealthier households can
afford to purchase land, even though much of the land may be unproductively farmed or evenidled.
Similar to Colombia, tax and credit policiesin Brazil generally reinforce the dominance of wealthier households in credit markets and thespeculative investment in land as tax shelters(Mahar & Schneider, 1994). Because poorerhouseholds on the frontier do not benefit fromsuch policies, their ability to compete in formalland markets is further diminished. This rein-forces the ‘‘sell out’’ effect of transferring
frontier land ownership from poorer initialsettlers to wealthier and typically urban-basedarrivals, forcing the poorer households to driftfurther into the frontier (Schneider, 1994).
According to Loopez (2003, p. 271) such pol-icies in Latin America over the past 50 years aresymptomatic of the general economic policyfailure in the region that has
focused on the generation of an expensive and oftenincoherent system of short-run incentives to promoteinvestment in physical capital. . . by undertaxing capi-
tal income and wasted in massive subsidies to the cor-porate sector in a futile effort to promote investmentand economic growth.
This has had two overall consequences on theland degradation and deforestation process inthe region. First, as described above, theresulting market and tax distortions promotethis process directly, in a deliberate strategy of ‘‘wasting natural resources as a way of enticinginvestors’’ (Loopez, 2003, p. 260). Second, LatinAmerican governments are dissipating scarce
revenues and financial resources ‘‘instead of concentrating their efforts in raising enoughpublic revenues to finance the necessaryinvestment in human and natural capital andthe necessary institutional capacities to effec-tively enforce environmental regulations’’(Loopez, 2003, p. 271).
Since 1990, many Latin American countrieshave embarked on substantial reform of theiragricultural sector, including trade liberaliza-tion. These reforms have included tarifficationwith bound tariffs, eliminating quota restric-
tions, removing export taxes, and reducing oreliminating the role of state trading agencies (deJanvry, Key, & Sadoulet, 1997). An assessment
of eight Latin American and Caribbeancountries over 1985–94 highlights the majorimpacts of these policy changes (Valdees,1996). 11 The study found that the initiation of trade liberalization reforms coincided unex-
pectedly with a drop in border prices of mostagricultural commodities below their long-termtrend, which combined with the appreciation of exchange rates resulted in sharper than expec-ted drops in real domestic producer prices.Although the level of direct taxation of exportshas declined in some countries, tariffs on agri-cultural imports continue and are still higherthan those for nonagricultural imports. Tradeand price regimes still prevail that lead toconsiderable overpricing of protected interme-diate inputs and distortions in the product
output mix. The effects of the policy reforms onnet agricultural income transfers have beensomewhat uneven. 12 On the whole, the evi-dence suggests that government agriculturalpolicies across Latin America still tend to dis-criminate against exportable agricultural com-modities.
Given the sector-wide effects of such majoreconomic reforms, it is difficult to assess theresulting production responses of households,except on a case study basis. A recent analysisof Mexico’s experience indicates some of the
complex linkages between agricultural pricingpolicies, smallholder responses and expansionof frontier activities.
Until the liberalization reforms in the early1990s in Mexico, the livestock sector benefiteddirectly from preferential loans and subsidizedbeef prices, and the agricultural sector frommaize price supports and subsidized fertilizerinputs. A panel analysis for 1970–85 across allMexican states confirmed that planted agricul-tural area was highly correlated over this periodwith the relative maize-fertilizer price ratio. In
addition, livestock numbers were positivelycorrelated with beef prices and credit dis-bursements, suggesting that the removal of these subsidies would have a direct and nega-tive effect on agricultural and livestock expan-sion (Barbier & Burgess, 1996). The potentialimpact of agricultural policy reform on theexpansion of planted area and livestock num-bers are good examples of the ‘‘first order,’’ ordirect, effects of changes in pricing on theincentives for frontier expansion and forestconversion by rural households. The authors
suggest however, that there are also likely to besome ‘‘second order,’’ or indirect, effectsresulting from economy-wide and sectoral
AGRICULTURAL EXPANSION AND RESOURCE BOOMS 151
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
16/21
reforms that may produce opposite incentiveeffects, potentially even outweighing the firstorder impacts. In particular, rural migration toforested areas may increase as a result of theimpacts of economic reform on the returns and
value of existing agricultural land.For example, as a result of the liberalization
reforms and structural changes as part of joining the North American Free TradeAgreement (NAFTA), Mexico is expected toundergo a long transition to sustained eco-nomic growth accompanied by substantialreturn migration to rural areas (Levy & vanWijnbergen, 1992b). The positive inducementto convert forest land, based on increases in therural labor force and falling rural wages, mayoutweigh the incentives to reduce agricultural
and cattle expansion due to increasing realGDP per capita (Barbier & Burgess, 1996).
The effect of the substantial reduction in theproducer price of maize in Mexico as a result of the trade liberalization has been a reduction inoutput and thus planted area. This has alsoprovoked a large fall in land values for rainfedland, which is expected to decline to nearly one-quarter that of irrigated land, thereby makingsubsistence farmers, rainfed farmers who arenet sellers of maize, and landless rural workersworse off (Levy & van Wijnbergen, 1992a).
Although subsistence farmers will benefit fromlower consumer prices, they will be doublyaffected by the loss in value of their rainfed landand in employment opportunities as daylaborers. The overall lack of employment andincome opportunities could induce ruralworkers and subsistence farmers to migratetoward frontier forest areas, or to convertremaining forestland that is available to themlocally. These ‘‘second order’’ effects of tradeliberalization on deforestation could outweighthe initial impacts of, say, a reduction in maize
producer prices on planted agricultural area(Barbier & Burgess, 1996).
6. BREAKING THE CYCLE: A ROLE FORTARGETED ECONOMIC POLICIES?
The evidence presented in this paper hasindicated that the abundance of land and nat-ural resources available in most Latin Ameri-can countries does not necessarily mean thatexploitation of this natural wealth will lead to
sustained economic growth and benefits for theregion. Although the focus of this paper wasmainly on the process of agricultural land
expansion, the results are probably generalizedacross the natural resource sector of LatinAmerican economies.
First, the process of agricultural landexpansion appears to be driven mainly by the
structural characteristics of the agriculturalsectors of Latin America economies––relativelow crop yields, high ratio of cropland to totalland area and a large share of agriculturalexports in overall merchandise exports.Although there is some evidence that agricul-tural land expansion is likely to slow down asLatin American economies develop and grow,as long as the structural characteristics of agriculture in the region remain fundamentallyunchanged, cultivated land area could increasein Latin America by as much as 50% by 2050,
with 70% of the new land originating fromdeforestation and wetland conversion (seeTable 3).
Second, the evidence from resource pricebooms in Latin America suggests that there islittle indication that such booms generate sus-tained economic growth in the region. To thecontrary, they may lead to Dutch disease effectswhereby scarce resources are diverted frommore innovative industrial sectors to naturalresource exploitation. There is also evidencethat economies that have increased their agri-
cultural land base significantly over the longrun are likely to have lower levels of GDP percapita then economies that have tended toreduce their dependence on agricultural landexpansion. Continued land expansion may alsobe correlated with a ‘‘boom and bust’’ patternof economic development.
Finally, current policies and reforms in LatinAmerica appear not to be overcoming thestructural characteristics in agriculture andeconomic development that lead to the abovetrends. That is, agricultural land expansion and
resource exploitation in Latin America may notbe contributing substantially to overall eco-nomic development because current policies arefailing to ensure that sufficient rents are beinggenerated from conversion and depletion of natural resources and that any rents earned arere-invested efficiently into other productiveassets in the economy. To overcome thisproblem may require more targeted policiesaimed at correcting critical structural imbal-ances in Latin America economies that inhibitthe efficient exploitation of natural wealth for
sustained long-term development. 13That is, as the recent policy reforms imple-
mented in Latin American countries have very
WORLD DEVELOPMENT152
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
17/21
broad, economy-wide and sectoral impactswith unknown (and possibly negative) aggre-gate impacts on land and resource use strategiesof rural households, it may be necessary tocomplement these reforms with specific, tar-
geted policies to generate direct incentives forimproved rural resource management. Themain aim of such policies would be to eliminatedistortions that reduce the economic returns of existing smallholder agricultural lands, improvethe access of poorer rural households to creditand land markets, and eliminate any remainingpolicy biases that favor relatively wealthyfarmers and landowners.
For example, given the evidence presented inthis paper, policies that raise the returns onexisting agricultural land would more likely
lead to idle land being brought into production,rather than more forestland conversion. In thepost-reform era, however, most agriculturaland livestock policies in Latin America are not‘‘targeted,’’ but instead have the overall objec-tive of raising the production of key agricul-tural and livestock products, regardless of whether this increased production comes fromexisting or frontier agricultural land. If thereturns to both existing and convertible forestland can be raised, then there will be littlechange in the amount of forestland converted
to agriculture.There is also substantial scope in many
Latin American countries to increase nonpricetransfers to the agricultural sector (Valdees,1996; de Janvry et al ., 1997). Such transferscould reduce significantly the incentives forland degradation and forest conversion inLatin America if they take the form of subsi-dies targeted to improve access by the ruralpoor to credit, increased research and exten-sion to disseminate information on the con-servation and improvements of smallholder
land, and public investment in irrigation andother infrastructure facilities to improve theproductivity and returns to existing small-holder land. For example, in Mexico there issome evidence that an investment program inland improvements to increase the productiv-ity of rainfed land could potentially mitigatethe negative distributional implications of NAFTA on the maize sub-sector (Barbier &Burgess, 1996; Levy & van Wijnbergen,1992a). Such a program could involve invest-ments not only in irrigation infrastructure for
1.1 million ha of rainfed land but also indrainage, land leveling, ditch-clearing and soilconservation. Moreover, improvements in the
returns to existing agricultural areas andsmallholdings could lead to an expansion of rural farm and off-farm employment oppor-tunities, thus reducing migration by landlessand near-landless households to frontier areas.
This suggests that a land improvementinvestment program for existing rainfedfarmers, particularly in states and regionsprone to high deforestation rates, could pro-vide both direct and indirect incentives forcontrolling deforestation by increasing thecomparative returns to farming existingsmallholdings as well as the demand for rurallabor.
Improving the extension of effective creditmarkets and services to reach poor ruralhouseholds while continuing to eliminate sub-
sidies and credit rationing that benefit mainlywealthier households may be important inreducing resource degradation in many LatinAmerican countries. There is evidence thateconomic reforms in Colombia, as well as inother Latin American countries (e.g., Brazil,Chile and Ecuador) may have significantlyreduced credit subsidies to agriculture (Valdees,1996). While this may have ended credit sub-sidies and rationing that had benefited mainlywealthier farmers and select agricultural com-modities, it did not increase rural credit for
land improvements, purchases, and investmentsby poorer farmers. According to Heath andBinswanger (1996), distortions in the creditmarket have clearly formed a major part of theincentives for rural households in Colombia tomigrate to both marginal upland areas andequally fragile land in the forested Amazon-Orinoco Basin.
As we have seen, institutional failures thatpromote insecure tenure or ownership of landalso have an important influence on resourcemanagement decisions of rural households.
Tenure insecurity may mean that the incentivesto invest in land improvements are lacking, aswell as making it extremely difficult for farmersto obtain formal credit for such improvements.Thus an important inducement for many poorsmallholders to invest in improved land man-agement is to establish proper land titling andownership claims on land currently occupied bythese smallholders. To improve land tenuresecurity in areas where frontier expansion isoccurring, it may be necessary to develop moreformal policies for smallholder settlement, such
as preferential allocation of public land, withfully demarcated ownership and tenure rightsto smallholders.
AGRICULTURAL EXPANSION AND RESOURCE BOOMS 153
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
18/21
NOTES
1. As a comparison, natural capital comprises only 5%
of wealth in North America, and 2% in the Pacific
OECD and Western Europe.
2. A major problem facing crosscountry analyses of
deforestation is the unreliability of any international
data set on forest cover, especially post-1990. The
United Nations Food and Agricultural Organization
(FAO), which has been the international agency respon-
sible for compiling forest area data across all countries,
based its 1990 Global Forest Resource Assessment on
population growth projections in order to overcome an
inadequate forest database for some countries and
regions. This means that the FAO country forest cover
data are inappropriate for crosscountry analyses of
deforestation that use demographic factors as explana-tory variables. This in turn means that crosscountry
analyses that employ the FAO forest cover data since
1990 and use demographic factors as explanatory
variables need to be treated with caution. An alternative
source of forest area data is the FAO Production
Yearbook. This includes data on forest, crop and
pasture land, but does not specify land area under
‘‘closed broadleaved forest.’’ The data are drawn from
national government responses to surveys rather than
using primary data sources and are generally considered
to be less reliable than the Global Forest Resource
Assessment data. The inappropriateness of using theFAO 1990 assessment based country forest cover esti-
mates in crosscountry deforestation analyses has been
pointed out by Barbier and Burgess (1997, 2001),
Cropper and Griffiths (1994) and Deacon (1999). Given
the problems with recent FAO forest stock data high-
lighted above, some have argued that crosscountry
studies should concentrate on explaining agricultural
land expansion, Ait Ait 1 rather than deforestation
across tropical countries (Barbier & Burgess, 2001).
3. Strictly speaking, deforestation is defined as (minus)
the percentage change in forested area, or
ð F it 1 F it Þ= F it 1. Deforestation is however, clearly
related to the change in forest stock variable, F t F t 1,
in Eqn. (1). In fact, various crosscountry analyses have
tended to use either specification as the dependent
variable to represent forest loss. To simplify notation,
F t F t 1 is used in Eqn. (1) and subsequent equations as
a short-hand expression for deforestation.
4. Tropical countries are those countries with the
majority of their land mass lying between the tropics.
5. Including the three time-invariant institutional indi-
ces in a fixed effects regression leads to collinear
regressors (Baltagi, 1995). As the institutional indices
are in themselves ‘‘weighted’’ country-specific dummy
variables, including the indices in an ordinary least
squares (OLS) regression will essentially imitate a fixed-effects model. Of course, the estimated coefficients on the
institutional variables may also be including the influ-
ence of other slow-changing factors that vary across
countries.
6. The failure of the property rights index to be
significant may reflect the fact that this variable indicates
the degree of protection of private property rights across
all sectors of the economy rather than the security of
land tenure in the agricultural sector. The failure of the
corruption index to be significant may be the result of
two countervailing influences. First, increased corrup-tion is associated with greater rent-seeking behavior,
which in turn may induce expansion of large-scale
commercial agricultural investments, leading to greater
land conversion. On the other hand, Loopez (1998) has
argued that reduced corruption and improved bureau-
cratic efficiency may actually facilitate the implementa-
tion of land and credit policies that stimulate a ‘‘race for
property rights’’ to convert forest and other common
resource land to agriculture.
7. From their case study analysis of five open devel-
oping economies, Findlay and Wellisz (1993) conclude
that over the postwar era it was economies with
relatively no resources, such as Hong Kong, Singapore
and Malta, that were among the earliest and most
successful exporters of labor-intensive manufactures. In
contrast, resource-rich Jamaica and the Philippines have
done relatively poorly, whereas Indonesia and Malaysia
have done comparatively better by balancing primary
exports with rapid expansion of labor-intensive manu-
factures.
8. Originally, the ‘‘Dutch disease’’ phenomenon wasassociated with the macroeconomic implications of an
economy’s over-dependence on a single, traded natural
resource sector (e.g., oil), which emphasized the enclave
character of the sector as the predominant source of
foreign exchange availability (e.g., see Neary, van
Wijnbergen, & van Wijnbergen, 1986). As the conse-
quence of a resource price boom (e.g. oil price shock),
expansion of the resource-based sector would be accom-
panied by a change in the real exchange rate, and the rest
of the economy would decline relatively. The more
recent treatments of the ‘‘Dutch disease’’ phenomenon,
such as by Matsuyama (1992) and Sachs and Warner(1995) discussed here, focus less on the economic
implications of a resource boom via real exchange rate
WORLD DEVELOPMENT154
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
19/21
movements but via internal economic distortions caused
by the shift of resources from a more innovative sector
(e.g., manufacturing) to a less innovative sector (e.g.,
agriculture, minerals). This latter representation of the
‘‘Dutch disease’’ is more appropriate for characterizing a
small open economy, in which real exchange ratedetermination is not considered.
9. The countries are Argentina, Bolivia, Brazil, Chile,
Colombia, Ecuador, Mexico, Paraguay, Peru, Uruguay
and Venezuela.
10. In the small open economy model of Brander and
Taylor (1997), if the country specializes in the manufac-
turing of goods in the long run, it gains unambiguously
from trade.
11. The countries are Argentina, Brazil, Chile, Colom-
bia, the Dominican Republic, Ecuador, Paraguay and
Uruguay.
12. These transfers were measured as the change in
value-added resulting from price and trade interventions
on inputs and final products, measured at the actual
level of production, plus the net value of nonprice
related transfers (Valdees, 1996). For those countries with
substantial income outflows, input subsidies and non-price transfers (investment in irrigation, research and
extension, credit subsidies and other government expen-
ditures) did not compensate, or compensated very little
for the substantial income outflows.
13. Such targeted policies may have also the scope to
reduce rural poverty substantially. See de Janvry et al .
(1997) for further discussion. The proposed investment
strategy would also represent a substantial shift in the
overall policy framework in Latin America away from
futile efforts to promote physical and financial capital
accumulation, which have benefited mainly of wealthyurban investors, at the expense of protecting and
enhancing overall natural and human capital, which
benefit mainly the rural poor. See Loopez (2003) for
further discussion.
REFERENCES
Alston, L. J., Libecap, G. D., & Mueller, B. (1999).Conflicts and land use: the development of prop-
erty rights and land reform in the Brazilian Ama-zon frontier. Ann Arbor: University of MichiganPress.
Alston, L. J., Libecap, G. D., & Mueller, B. (2000). Landreform policies, the sources of violent conflict, andimplications for deforestation in the Brazilian Ama-zon. Journal of Environmental Economics and Man-agement, 39(2), 162–188.
Anderson, L. E., Granger, C. W. J., Eustaquio, J. R.,Weihold, D., & Wunder, S. (2002). The dynamics of deforestation and economic growth in the BrazilianAmazon. London: Edward Elgar.
Andreoni, J. R., & Levinson, A. (2001). The simpleanalytics of the environmental Kuznets curve. Jour-
nal of Public Economics, 80, 269–286.Antle, J. M., & Heidebrink, G. (1995). Environment and
development: theory and international evidence. Eco-nomic Development and Cultural Change, 43(3), 603– 625.
Baltagi, B. H. (1995). Econometric analysis of panel data.Chichester: John Wiley.
Barbier, E. B. (1999). Poverty, environment and devel-opment. In J. C. J. M. van den Bergh (Ed.), Thehandbook of environmental and resource economic(pp. 731–744). London: Edward Elgar.
Barbier, E. B. (2002). Institutional constraints anddeforestation: an application to Mexico. EconomicInquiry, 40(3), 508–519.
Barbier, E. B., & Burgess, J. C. (1996). Economicanalysis of deforestation in Mexico. Environment and Development Economics, 1(2), 203–240.
Barbier, E. B., & Burgess, J. C. (1997). The economics of tropical forest land use options. Land Economics,
73(2), 174–195.Barbier, E. B., & Burgess, J. C. (2001). The economics of tropical deforestation. Journal of Economic Surveys,15(3), 413–421.
Beck, T., Levine, R., & Loayza, N. (2000). Finance andthe sources of growth. Journal of Financial Econom-ics, 58(October), 261–300.
Brander, J. A., & Taylor, M. S. (1997). Internationaltrade and open-access renewable resources: the smallopen economy. Canadian Journal of Economics,30(3), 526–552.
Brander, J. A., & Taylor, M. S. (1998). Open accessrenewable resources: trade and trade policy in a two-country model. Journal of International Economics,
44, 181–209.Chomitz, K. M., & Gray, D. P. (1996). Roads, land
markets and deforestation: a spatial model of landuse in Belize. The World Bank Economic Review,10(3), 487–512.
Cropper, M., & Griffiths, C. (1994). The interaction of population growth and environmental quality.American Economic Review. AEA Papers and Pro-ceedings, 84(2), 250–254.
Deacon, R. T. (1994). Deforestation and the rule of lawin a cross-section of countries. Land Economics,70(4), 414–430.
Deacon, & Robert, T. (1999). Deforestation and own-ership: evidence from historical accounts and con-temporary data. Land Economics, 75(3), 341–359.
de Janvry, A., Key, N., & Sadoulet, E. (1997). Agricul-tural and rural development policy in Latin America:
AGRICULTURAL EXPANSION AND RESOURCE BOOMS 155
-
8/9/2019 Edward Barbier (2004) Agricultural Expansion, Resource Booms and Growth in Latin America Implications for Long-r…
20/21
new directions and new challenges. WorkingPaper No. 815. Department of Agricultural andResource Economics, University of California atBerkeley.
Ehui, S. K., & Hertel, T. W. (1989). Deforestation andagricultural productivity in the Côote d’Ivoire. Amer-
ican Journal of Agricultural Economics, 71(August),703–711.
Findlay, R. (1996). Modeling global interdependence:centers, peripheries and frontiers. American Eco-nomic Review, 86 (2), 47–51.
Findlay, R.& Wellisz, S. (Eds.). (1993). The political economy of poverty, equity and growth: five small openeconomies, a World Bank comparative study. NewYork: Oxford University Press.
Fischer, G., & Heilig, G. K. (1997). Population momen-tum and the demand on land and water resources.Philosophical Transactions of the Royal Society SeriesB, 352(1356), 869–889.
Food and Agricultural Organization (FAO) (1993).Forest resources assessment 1990: tropical countries.Rome: FAO.
Food and Agricultural Organization of the UnitedNations (FAO) (1995). World agriculture:towards
2010––an FAO study. Rome and New York: FAOand John Wiley & Sons.
Food and Agricultural Organization of the UnitedNations (FAO) (1997). State of the World’s Forests1997 . Rome: FAO.
Food and Agricultural Organization (FAO) (2001).Forest resources assessment 2000: preliminary results.Rome: FAO.
Godoy, R., Jacobson, M., De Castro, J., Aliaga, V.,
Romero, J., & Davis, A. (1998). The role of tenuresecurity and private preference in neotropical defor-estation. Land Economics, 74(2), 162–170.
Hartwick, J. (1977). Intergenerational equity and theinvesting of rents from exhaustible resources. Amer-ican Economic Review, 6 , 972–974.
Heath, J., & Binswanger, H. (1996). Natural resourcedegradation effects of poverty and populationgrowth are largely policy-induced: the case of Colombia. Environment and Development Economics,1, 65–83.
Kaimowitz, D. (1995). Livestock and deforestation inCentral America in the 1980s and 1990s: a policy
perspective. EPTD Discussion Paper, No. 9, Envi-ronment and Production Technology Division, Inter-national Food Policy Research Institute,Washington, DC.
Koop, G., & Toole, L. (1999). Is there an environmentalKuznets curve for deforestation? Journal of Devel-opment Economics, 58(July), 231–244.
Levine, R., Loayza, N., & Beck, T. (2000). Financialintermediation and growth: causality and causes.Journal of Monetary Economics, 46 (1), 31–77.
Levy, S., & van Wijnbergen, S. (1992a). Maize and thefree trade agreement between Mexico and the UnitedStates. The World Bank Economic Review, 6 (3), 481– 502.
Levy, S., & van Wijnbergen, S. (1992b). Transition problems in economic reform: agriculture in theMexico-US free trade agreement. Policy Research
Working Paper No. 967. Washington, DC: TheWorld Bank.
Loopez, R. (1989). Exchange rate determination innatural resource-rich economies. Revista de AnaalisisEconoomico, 4(1), 91–105.
Loopez, R. (1998). Where development can or cannot go:
the role of poverty-environment linkages. In B.Pleskovic & J. E. Stiglitz (Eds.), Annual Bank conference on development economics 1997 (pp. 285– 306). Washington, DC: The World Bank.
Loopez, R. (2003). The policy roots of socioeconomicstagnation and environmental implosion: LatinAmerica 1950–2000. World Development, 31(2),259–280.
Mahar, D., & Schneider, R. R. (1994). Incentives fortropical deforestation:some examples from LatinAmerica. In K. Brown & D. W. Pearce (Eds.), Thecauses of tropical deforestation. London: UniversityCollege London Press.
Matsuyama, K. (1992). Agricultural productivity, com-parative advantage, and economic growth. Journal of Economic Theory, 58, 317–334.
McConnell, K. E. (1997). Income and the demand forenvironmental quality. Environment and DevelopmentEconomics, 2, 383–399.
Neary, J., van Wijnbergen, P., & van Wijnbergen, S.(Eds.). (1986). Natural resources and the macroeconomy. Cambridge, MA: MIT Press.
Nelson, G. C., Harris, V., & Stone, S. W. (2001).Deforestation, land use, and property rights: empir-ical evidence from Darieen, Panama. Land Economics,77 (2), 187–205.
Panayotou, T. (1995). Environmental degra