Carole L. Jolly and Barbara Boyle Torrey
What are the effects of population growth on land use change? Despite the interest in and importance of this question, there is a relatively small body of carefully designed research that begins to provide answers to it. In order to make progress in this field we need to understand why careful research on this topic is so scarce, examine the work that has been done, and propose ways to encourage research in an area that may be critical to the future of many countries.
WHY THIS IS A DIFFICULT QUESTION TO ANSWER
One difficulty in developing general answers to the question posed is that much of the existing research has focused on case studies, in which results often depend on idiosyncracies of physical and human endowments. Consequently, although case studies are useful in illuminating particular intricacies of the population and land use relationship, they are not readily compared. In order for them to be comparable, a general framework of analysis would have to be developed and applied systematically to countries or areas in very different circumstances.
A second difficulty in studying the relationship between population growth and land use change is the challenge of precisely matching demographic and ecological data that generally are not collected over the same geographic regions. Demographic surveys are usually conducted by political region, such as a district or country; land use data are more often col-
lected or a particular ecosystem or landscape, which can cross political boundaries. Even when demographic and land use data are gathered for the same area, no area is a closed system because migration complicates population dynamics and climate change complicates land use dynamics.
A third difficulty is that much of the research conducted on population growth and land use change has been weak in identifying and quantifying the set of causal connections between demographic and land use changes, making definitive conclusions difficult. Instead, much of the research has focused on documenting associations between land use changes and demographic variables rather than identifying the specific causes for particular changes.
Billie L. Turner highlighted several additional obstacles to research in this area in his presentation. He stressed that the data are very poor for global comparative studies of land use. Almost all land use data measure conversion of land to human use; there are few data on modification of land use. Other obstacles are the different views among social scientists about what constitutes proof of population and environment linkages. Finally, Turner noted that there is no agreement among scientists about the level of observation suitable for studying population and land use dynamics.
Research on population growth and land use change has been made more complicated by the use of two conflicting paradigms, one based in natural science (or classical economics) and the other in neoclassical economics. The natural science paradigm places more emphasis on the finiteness of resources than on technological and institutional change and the accumulation of physical and human capital. This paradigm views population growth as a threat to the inherent limits of arable land to provide food, shelter, and sustenance. The neoclassical economics paradigm emphasizes the accumulation of both physical and human capital and the substitution of abundant factors for scarce ones. This paradigm suggests that population growth can be the impetus for technological and other changes that mitigate or even eliminate the effects of natural resource limits on economic well-being. Empirical research is capable of indicating which of these paradigms has more explanatory power, but the research base is thin and has not yet led to a body of knowledge on which public or scientific consensus has developed.
Classical economists, beginning with Malthus, stressed the difficulty of maintaining a steady or increasing standard of living given a finite resource base and a growing population. Malthus argued that food production could only grow at a linear rate while populations grew geometrically; thus population growth would ultimately outstrip the ability of the economy to meet the demand for food (Malthus, 1798). Although agricultural production has so far met and often exceeded populations' growing needs for food, there remains a concern among many natural scientists that the ecological limits
to food production provide little scope for future expansion (Ehrlich and Holdren, 1971; Ehrlich et al., 1977; Brown et al., 1991; Raven, 1991).
Much of the empirical research done under the natural science paradigm has focused on estimating carrying capacity to identify areas of population pressure. Research has also analyzed the effects of increasing human populations on species loss and ecosystems. A large study, undertaken by the Food and Agriculture Organization (FAO), analyzed the ''population supporting capacities'' of 117 developing countries and concluded that over half of the countries could not achieve food self-sufficiency by the year 2000 with low levels of inputs (mainly labor). Almost a third of the countries could not meet their needs even with an intermediate level of inputs (some fertilizer tools and simple conversation techniques). Nineteen countries could not meet their needs even at very high levels of inputs (advanced technology, complete mechanization, and all necessary conservation measures [Higgins et al., 1983]).
Neoclassical economists are also concerned with whether an economy can provide an increasing or steady standard of living given a finite resource base and a growing population. To determine whether output growth can keep up with population growth, attention is focused on two factors: technological advance and the substitution of scarce factors with more abundant ones (Stiglitz, 1979; Simon, 1981). Under well-functioning markets, as land resources become scarce, incentives will increase for people to develop technologies to farm previously unused land (extensification) and to increase production per existing unit of land (intensification). People will also substitute more abundant resources, such as fertilizer and labor, for land. In her widely cited work, Ester Boserup (1965, 1970, 1981) documented how, in response to greater population density, farmers reduced their fallow periods, began to use the plow, and implemented multiple cropping cycles to make their land more productive. Although she acknowledged that land degradation could occur as hills were cultivated or fallow periods excessively shortened, Boserup concluded that technology, such as terracing and fertilizer, could minimize the damage.
Under the neoclassical economics paradigm, land degradation can be the result of several factors, the most important being inefficient markets and lagged responses to population growth. For example, government subsidies have made it economically viable for people to settle the Amazon rain forest, where soil quality is poor and sustained agricultural production difficult (Schmink and Wood, 1987). Artificially low producer prices for agricultural products and lack of property rights in many countries in Africa have given farmers few incentives to conserve their land for future use (Lele and Stone, 1989). As populations increase and land resources become scarcer, farmers need to farm more intensively. But farmers and governments may not react quickly enough to develop technologies to farm their land more efficiently, resulting in land degradation.
THE INCREASING IMPORTANCE OF POPULATION GROWTH AND LAND USE CHANGE
The difficulty of doing research on this issue is commensurate with its increasing importance. The world's population is likely to double in the next 60 years, even if fertility rates fall in virtually every developing country (United Nations, 1991; World Bank, 1992; Bureau of the Census, 1991). This is because, regardless of how effective family planning programs may be in reducing fertility, the adult population of the next 20 years has already been born and its increased demand for food is inevitable. The earth's land resources will have to become more productive in the next 60 years if the growing needs for food are to be accommodated.
Today's population dynamics are unique. Although there is a negative correlation between density and natural increase (births minus deaths; see Table 1), many countries that are already very densely populated are still experiencing high rates of population growth (see Figures 1 and 2). Most of the research on population and land use change has focused on levels of population and population densities. However, rates of change in high-density countries may also be critical variables.
The increases in the world's population will not be evenly distributed around the globe. Some of the poorest and least-developed regions will be the ones required to adjust most rapidly to growing numbers. In most of the developed countries, there are likely to be only small increases or even decreases in total population size. For other countries, however, particularly in Africa, there will be a doubling or tripling in the number of people. Regions also differ in their levels of migration, both internal and international. These differences will in turn affect the degree of urbanization and the spatial distribution of people.
TABLE 1 Number of Countries by Population Density and Rate of Natural Increase (excluding city-states), 1991
In studying the population and land use relationship, it is essential to consider both numbers of people and their behavior. Population growth influences land use patterns in combination with consumption behaviors and productive activities of the world's peoples. It is also essential to consider countries or regions in broader international context. Land use patterns in the developing world are sometimes strongly influenced by international trade and the high purchasing power of residents in the developed countries. The Committee on Population's Workshop on Population Growth and Land Use Change in Developing Countries was organized to look at some of these complex issues.
Major points of both the workshop papers and discussions are summarized below. If a condensed paper is included in this volume, it is indicated.
M. Gordon Wolman's opening remarks, which are included in this volume, raise many of the questions that were pursued over the 2-day workshop. He stresses that people have always used and changed the way they
have used land from prehistoric time to the present. However, today's land use is qualitatively and quantitatively different from that in the past. Land use change has occurred at much faster rates as well as on much larger scales since 1900. It is difficult to evaluate these changes, partly due to the poor quality of available land use data and the difficulty of determining whether climate change or human activity has had the greatest effect on land use. Wolman questions whether the absolute size of the population or the rate of population increase has been more important in affecting land use change. He concludes that, given that the pace and types of land use change and population growth in the past are very different from those of today, the study of the past may not be the key to understanding the future.
In assessing the effects of changes in land use on the environment, Isaak S. Zonneveld's paper, which is included in this volume, discusses the importance of maintaining ecological integrity as land use changes. He describes the emergence of landscape ecology to study land form, function, and change. Zonneveld explains how and why land use change is measured and describes the strengths and weaknesses of remote sensing.
Steward Pickett's commentary, summarized in this volume, builds a bridge between the natural scientists' perspectives of Wolman and Zonneveld and the social science perspectives of the other papers. He discusses how
ecologists tend to view the world from a micro perspective, whereas the social scientists tend to explain the world from a more macro perspective. Ecologists are shifting from studying ecosystems with little human intervention to analyzing the human dimensions of the ecological changes they document. He concludes by noting that "the battle to include humans in the scope of ecology is evidenced by the persistence of George Perkins Marsh's 'great question': 'Whether man is of nature or above her?'"
The discussions and papers by Hans Binswanger, Richard Bilsborrow and Martha Geores, and Uma Lele focus on the strengths and limitations of the neoclassical paradigm as stated by Ester Boserup (1965, 1970, 1981). Binswanger's discussion of his published research with Prabhu Pingali (1985) quantified how farmers responded to population growth through land intensification and extensification. Binswanger reviewed many of the innovations that farmers in developing countries make in the face of population pressures, such as irrigation, use of fertilizers, and multiple cropping. He also discussed the differences between the development of agriculture in the United States, which had relatively low population density, and in Japan, with high population density, to illustrate the different innovations farmers may make when faced with very different resource constraints.
Binswanger and Pingali's research concludes that, in growing populations, moderate increases in agricultural production are feasible with low-level land investments. However, a rapidly growing population will out-strip this agricultural production capacity without the intervention of science-and industry-based inputs, such as high-yielding seeds (Binswanger and Pingali, 1985).
Many factors can slow the speed of Boserupian responses. Uma Lele's paper and discussion outline the primary impediments operating in Africa. She notes that population is not distributed according to land quality because of high rates of vector-carried diseases in some areas and the placement of colonial infrastructure and European settlements. Macroeconomic policies that discriminate against agriculture distort market prices and thus people's responses to increased population size. The development of Western-style land tenure systems in Africa has sometimes led to the concentration of rights of access in certain groups and removed indigenous means of determining usage. In addition, poor credit markets and lack of technology suited for African agriculture have made the land intensification process more difficult.
Using cross-sectional data collected at the national level to quantify some of the relationships that Boserup's and Binswanger and Pingali's research identifies, Bilsborrow and Geores's paper finds a strong relationship between overall population density and the percentage of agricultural and pasture land used in production. Increases in fertilizer use have occurred widely across developing countries in recent decades as populations have
grown, and this increased use has contributed to generalized increases in land productivity. However, the relationship between percentage changes in population and land used are not significant. Furthermore, the cross-sectional analysis among countries shows no relationship between density of population and the rate of depletion of forests. They document the poor quality of the data and the difficulty of doing cross-sectional analysis using data from developing countries. But they also suggest that this kind of research could be useful in identifying the key relationships to be pursued in more depth within countries.
Many of the conclusions of the Binswanger, Bilsborrow, and Lele presentations were illustrated in the case studies that were presented. (Four of these studies have been summarized and are included in this volume.) In fact, given the problems of making international comparisons and the limitations of current theories about how land use changes with increasing population density, the case studies of specific regions presented at the workshop became critical in better defining the issues and problems. Nigeria, India, Mauritius, Thailand, and Honduras were the countries discussed.
Rapid increases in population in northern Nigeria since the 1960s and the resultant land use changes are the subject of Michael Mortimore's paper, which is summarized in this volume. He demonstrates that the small-scale farmer invests considerably more in land improvement in the high-population-density area than in the low-density area to meet growing demands for food. In the high-density area, soil fertility is being managed on a sustainable basis. Given these results, he concludes that "population growth, and high population density, are compatible with sustainable resource management under smallholder conditions."
The effects of population growth and density on agricultural land use in India are examined in Robert Evenson's paper, which is included in this volume. He estimates the effects of population change on several agricultural investments during 1955–1987. The first stage of analysis shows that some of the investment in agricultural research, agricultural extension services, and rural infrastructure and net cropped area were induced by either population growth or increases in population density. However, many of these changes were also partly the result of strategic planning by the government. The second stage of analysis estimates the real returns to labor and to land by farmers as a function of population growth, technology flows, and infrastructure. Although population-induced changes in technology and infrastructure have a positive effect on real wage incomes, they are not sufficient to offset the negative effects of population growth.
The unique situation of Mauritius is studied by Wolfgang Lutz and Einar Holm in their paper, which is also summarized in this volume. Mauritius is an island country with high population density (and a low population growth rate) that has become increasingly integrated into world markets,
making the effects of population change on land use highly indirect. Today, Mauritius imports much of its food and pays for it with its exports of textiles and sugar. As a result, since 1965, land previously used for sugar cane production has been progressively turned into urban space.
Theodore Panayotou's paper discusses Thailand's recent population growth and land use change. Between 1960 and 1990, the population doubled and the forested areas decreased 50 percent. He argues that the size of the agricultural population is the single most important determinant of the expansion of cultivated land. One of the contributing factors to land degradation in Thailand was that 40 percent of the agricultural land is held without land title, thereby reducing incentives for farmers to manage their land sustainably. Correlations suggested that deforestation in Thailand was related to the increase in population, the price of wood, and the poverty and low education level of the population.
The paper by Billie DeWalt, Susan Stonich, and Sarah Hamilton, included in this volume, evaluates the evidence linking population growth and the degradation of land and water resources in Honduras. They conclude that the deterioration of these resources is attributable more to unequal resource distribution and the patterns of commercial agricultural expansion than to population growth. They suggest that similar relationships exist in other Central American countries.
Vernon Ruttan's paper, included in this volume, returns to some of the issues raised by Wolman at the beginning of the workshop. He steps back from the focus of the case studies and discusses the implications of population growth and environmental change for agricultural production. He summarizes the historical concerns about sustainable growth. He then goes on to discuss institutional change and the problems of the differences in costs between private and social environmental services. He argues that scientific and technical constraints to increased agricultural production mean that agricultural research needs to be reorganized. "A serious effort to develop alternative land use, farming systems, and food systems scenarios for the 21st century should be initiated."
Samuel Preston concluded the workshop by summarizing the general themes that were discussed and refined by the other participants on the effects of population growth on land use.
In the long run, population growth almost certainly affects land use patterns. The effects of population growth occur mainly through the extensification and intensification of agricultural production. Different population growth rates and different population densities probably produce different sets of land use changes. The evidence is partly logical. More people need
more food, a situation that will affect how land is used for agriculture. The evidence is partly historical, as illustrated in M. Gordon Wolman's paper, which shows that land use patterns over the last 6,000 years are associated with the expansion of the human population. These associations have existed for over three decades in Thailand as documented in Theodore Panayotou's case study. There is also cross-sectional evidence, as illustrated in Richard Bilsborrow and Martha Geores's paper, that notes a correlation between a country's population density and the percentage of its arable land that is used in production. Finally, evidence based on Ester Boserup's model and research shows how population increases induce people to cultivate additional land or to farm their present land more productively, as demonstrated in Robert Evenson's paper.
Most of the changes in land use associated with very rapid population growth are likely to be disadvantageous for human beings. The changes that Boserup and others have described show that as populations grow, the technology required to maintain output is more expensive and requires more investment and labor. These are mainly the direct, on-site costs. There are also indirect, off-site costs that may be as great or greater than the on-site costs. They include salinization resulting from irrigation and contamination of common property resources (resources that are commonly owned, but without rules or regulations governing their use) from fertilizer use.
Many of the workshop participants thought the most important way to offset these costs associated with rapid population growth was through institutional change, such as property rights and agricultural research, as noted in Vernon Ruttan's paper. Whether these effects completely or partially offset the problems created by rapid population growth cannot be determined for the general case. But it is difficult to demonstrate an instance in which the offsetting effects are great enough to make the population as well off in terms of its land use patterns as it would have been with slower population growth. Evenson's paper shows that North India is not such an instance.
Population growth is not the only, or in many cases, the most important influence on land use. Other influences include technological change and changes in production techniques—which can be exogenous or, in some situations, endogenous or partly induced by population growth. National and international markets for goods and agricultural products clearly influence patterns of land use, as do government regulation and tax policies or the absence thereof. Income inequality was repeatedly mentioned by the workshop participants; clearly, it is an important factor in land use patterns in Honduras, as demonstrated in the Billie DeWalt, Susan Stonich, and Sarah Hamilton case study. Inequality itself, however, is in part influenced by rates of population growth.
In analyzing the question of whether population growth is the most
important factor affecting land use, there are two issues—one is historical and one is prospective. What has been the relative influence of population growth on land use in the past? If the goal is to alter patterns of land use in the future, would reducing fertility be one of the most cost-effective ways to intervene?
Family planning programs put in place today will have their primary effect on the margin, which is the new births averted in the future. Looked at another way, however, population momentum simply means that if population policy is to have a major effect on future population growth patterns, it must begin sooner rather than later.
Because the effects of population growth on land use depend on many factors, case studies that clearly delineate the relative role of these factors are needed. Among these conditioning factors are markets for agricultural and forestry products, land tenure systems, soil quality, climate, and capital markets. The workshop case studies suggest that population growth is most likely to result in land degradation when land is held in common without rules governing its access, when production is mainly for subsistence, and when the soil is fragile and rainfall light. Under these conditions, fast population growth clearly creates potential for producing soil degradation. Parts of Africa may fit this pattern, but the northern Nigeria case study by Michael Mortimore shows that the farmers have adapted quite well to the doubling of their population.
With clear property rights, robust soils, and efficient markets, population growth is less likely to result in land degradation. Under these conditions, rapid population growth, which results in larger markets for agricultural products, gives land owners incentives to protect soil quality, which they are able to do by borrowing in relatively efficient capital markets. At the same time, land ownership provides collateral for the borrowing needed to invest in the protection of soil. Most real situations are somewhere between these two extreme scenarios, and more research is needed on the role of these conditioning factors in different areas.
Rapid population growth is likely to make the survival of other members of the animal and plant kingdom more difficult. Accompanying rapid population growth in the past has been greater species loss and a higher attrition within species than would have occurred in the absence of human expansion. It is difficult to place an exact value on this loss, particularly with regard to its importance for humans. There are widely varying views as to the weight that humans should attach to the welfare of other species. It is clear that the preferences for those species are not presently being reflected in any market mechanism, although the survival of several species is debated in the political process.
Better data are needed to analyze more definitively the relationship between population growth and land use. It was clear in the workshop
discussion that many of the questions posed could not be answered with the available data. As Bilsborrow and Geores argued, additional time series data on land use variables are especially needed.
The research on resource limits has been less fruitful than research to determine the underlying dynamic relationships between population growth and land use change. Although many workshop participants recognized that all resources are not perfectly substitutable for one another, they agreed that conducting research on the effect of population growth on land use in terms of dynamics and interactions of factors was more useful than discussing the relationship in terms of fixed limits. Such an approach would mirror the continuing change in the many factors affecting land use and their interactions.
It was noted in conclusion that the workshop focused primarily on changes in the states or faces of the earth—how much deforestation is occurring, how much soil is being lost, etc. What needs more attention are changes in biogeochemical flows in which land use plays a major role. For example, feeding 10 billion people will require increased fertilizer use. If fertilization is intensified, methane production will increase, which may interfere with attempts to sustain the biosphere as we know it. Land use is an important factor, but it is related to many others that change with population growth.
Despite some careful research examining the relationship of population growth and land use change, we are still reviewing the beginning of an applied discipline rather than reviewing a mature body of research. The growing importance of these relationships to many countries is inconsistent with the lack of broad, systematic research attention. This lack of a thorough body of research, however, is due in part to some of the problems mentioned in this introduction. At this stage, methodological suggestions are easier to make than substantive conclusions.
Studies that rely on cross-sectional data cannot capture the dynamics between changes in population growth and land use change, as discussed by Bilsborrow and Geores. But historical time-series data rarely exist over the right time frame and scale to be useful. Pickett and Zonneveld illustrated the differences in scale that natural scientists and population experts use. Although correlations can be illustrative, causality is elusive. And, it is correlations and not documented casual relationships that have reinforced the work of the natural and social scientists in their different paradigms. Only carefully designed research to elucidate causality between population growth and land use change will help researchers address the fundamental differences in their paradigms.
The problems of scale and time horizon require that further research be focused on carefully designed and coordinated case studies, such as have been included in this volume. Indicators of how to measure and judge land use change need to be developed. Mortimore's case study of northern Nigeria suggests what some of the physical indicators might be; DeWalt's case study utilizes some of the social indicators. Only when there is a much larger number of sophisticated case studies will we be able to generalize about how current and future population growth rates in the world are likely to change land use.
This is the beginning of an important research area, not its culmination. Clearly the workshop papers and discussion raised more questions than they answered. The current population growth rates in some of the developing countries make this research not only important but essential to their ability to accommodate their future populations. Therefore, the questions raised by these workshop papers will be used to direct the subsequent work of the Committee on Population and, we hope, others in this area with the purpose of stimulating new research on the relationship between population growth and land use change in the future.
Binswanger, H.P., and P.L. Pingali 1985 Population growth and technological change in agriculture. Pp. 62–89 in T.J. Davis, ed., Proceedings of the Fifth Agriculture Sector Symposium: Population and Food. Washington, D.C.: World Bank.
Boserup, E. 1965 The Conditions of Agricultural Growth: The Economics of Agrarian Change Under Population Pressure. Chicago: Aldine Press.
1970 Present and potential food production in developing countries. Pp. 100–113 in W. Zelinsky et al., eds., Geography and a Crowding World. New York: Oxford University Press.
1981 Population and Technological Change. Chicago: University of Chicago Press.
Brown, L.R., A. Durning, C. Flavin, H. French, J. Jacobson, N. Lenssen, M. Lowe, S. Postel, M. Renner, J. Ryan, L. Starke, and J. Young
1991 State of the World 1991: A Worldwatch Institute Report on Progress Toward a Sustainable Society. New York: W.W. Norton.
Bureau of the Census 1991 World Population Profile: 1991. Report WP/91. Washington, D.C.: U.S. Department of Commerce.
Ehrlich, P.R., and J.P. Holdren 1971 Impact of population growth. Science 171:1212–1217.
Ehrlich, P.R., A.H. Ehrlich, and J.P. Holdren 1977 Ecoscience: Population, Resources, Environment. San Francisco: Freeman.
Higgins, G.M., A.H. Kassam, L. Naiken, G. Fischer, and M.M. Shah 1983 Potential Population Supporting Capacities of Lands in the Developing World. Rome: Food and Agriculture Organization.
Lele, U., and S. Stone 1989 Population Pressure, the Environment and Agricultural Intensification: Variations on the Boserup Hypothesis. Madia Discussion Paper 4. Washington, D.C.: World Bank.
Malthus, T.R. 1798 An essay on the principle of population. Reprinted, pp. 15–139 in P. Appleman, ed., An Essay on the Principle of Population. New York: W.W. Norton., 1976.
Raven, P.H. 1991 Winners and losers in the twentieth-century struggle to survive. Pp. 259–266 in K. Davis and M.S. Bernstam, eds., Resources, Environment, and Population: Present Knowledge and Future Options. A supplement to Vol. 16, 1990, Population and Development Review. New York: The Population Council and Oxford University Press.
Schmink, M., and C.H. Wood 1987 The political ecology of Amazonia. Pp. 38–57 in P.D. Little and M.M. Horowitz, eds., Lands at Risk in the Third World: Local Level Perspectives. Boulder, Colo.: Westview Press.
Simon, J. 1981 The Ultimate Resource. Princeton, N.J.: Princeton University Press.
Stiglitz, J.E. 1979 Neoclassical analysis of resource economics. Pp. 36–66 in K. Smith, ed., Scarcity and Growth Reconsidered. Baltimore, Md.: The Johns Hopkins University Press.
United Nations, Department of International Economic and Social Affairs 1991 World Population Prospects 1990. No. 120. New York: United Nations.
World Bank 1992 World Development Report 1992: Development and the Environment. New York: Oxford University Press.