What can tropical deforestation models tell us?


Related stories

Because of growing concern about tropical deforestation, researchers have sharply increased their efforts to model the factors influencing forest clearing in recent years. The Center for International Forestry Research (CIFOR) is currently reviewing the literature in this area, and has prepared the following set of tentative conclusions based on some 120 different models analyzed so far. The models include global, national, regional, spatial, and household regression models, computable general equilibrium models, linear programming models, and analytical models. David Kaimowitz and Arild Angelsen are currently preparing a full report on this review, which will be available later this year.

As an individual actively involved in forest policies we thought you might be interested in seeing these conclusions. But more importantly, we would like your feedback on whether you feel they adequately reflect the literature you have had access to. We would be particularly interested to know if you are aware of any modeling work which contradicts or modifies these conclusions. Comments should be addressed to: David Kaimowitz (d.kaimowitz@cgnet.com).

The discussion which follows analyzes the factors found to affect forest clearing in these models. It should not be taken to imply all deforestation is inappropriate. How much and which forest ought to be cleared in each country is a matter for national policy makers to decide.

We would also like to emphasize that inevitably we have been forced to synthesize the models’ most general conclusions regarding tropical deforestation. We have tried, where possible to make reference to the specific circumstances under which these conclusions are likely to hold. Even so each particular country and region is unique and the patterns of deforestation in each country can only be partially attributed to the variables analyzed in the models.


Population affects deforestation through: 1) direct land hunger of rural families; 2) labor markets; 3) demand for agricultural and forest products; and 4) induced technological change.

Population density and the percentage of land area in forest are strongly correlated at the national level. However, this correlation often disappears when other variables are included in the models, telling us that rather than population explaining forest cover a third set of factors may be simultaneously affecting both. Population variables and recent deforestation are not as strongly correlated. Models which use FAO deforestation data to examine the relation between population and forest cover give spurious results since the FAO deforestation estimates themselves are partially based on population data.

At the local and regional level, population is endogenous and determined by infrastructure availability, soil quality, distance to markets, and other factors. At these levels, population often shows no statistically significant relation to deforestation once these other variables are taken into account.

Except at the global level of analysis, migration is probably a more relevant demographic issue related to deforestation than natural population growth. Many government policies affect migration including infrastructure investment, colonization policies, subsidies for specific activities, and macro-economic policies.

To the extent people migrate to forested areas because it is economically attractive for them to do so, population levels in those areas cannot be considered an independent variable with regard to deforestation. It is more appropriate to consider population exogenous in areas with limited migration.

Greater population increases labor supply, which tends to lower wages, and that in turn can lead to higher deforestation. However, many factors can intervene at each step of this chain, and invalidate this general conclusion.

Few models focus explicitly on the relation between population and the demand for agricultural and forest products. This aspect is relatively less important in contexts where per capita income is growing or declining rapidly or where agricultural and forestry products are strongly tradable. Because of globalization, the population – demand relation has become less important at the national and regional levels. New agricultural and forestry export prospects may lead to rapid deforestation in low population countries where small domestic markets previously limited deforestation.

If technology becomes more labor intensive as population density increases, deforestation should increase less than proportionally to rural population growth.

Income / Economic Growth

Higher national per capita income is associated with greater deforestation in developing countries. It is not clear whether deforestation declines as countries become richer. Results regarding the impact of rapid economic growth on deforestation are contradictory. Conclusions regarding this variable should be made with particular caution since the only available models are based on global regression analyses with poor data.


No clear conclusion can be made regarding this variable. Some studies find an association between high external debt and greater deforestation. Others find no clear connection. They are all based on global regression analysis and poor data.

Agricultural Prices

Higher agricultural prices stimulate greater forest clearing. They make agriculture more profitable and finance the clearing of additional land. The effect is stronger when agricultural output and labor supply are elastic. In the long-term general equilibrium price adjustments dampen some of the initial effects, so the net effect on deforestation is likely to be lower.

Policies which seek to improve the terms of trade for agriculture, such as currency devaluations, trade liberalization, reductions in agricultural export taxes, agricultural price subsidies, and reduced fiscal spending on non-agricultural sectors tend to raise prices received by farmers, and hence increase deforestation. Thus, successful structural adjustment policies (SAPs) often increase pressure on forests, and policies such as over valued exchange rates, industrial protectionism, and urban biases in spending can reduce deforestation.

These conclusions are based mostly on static partial equilibrium analyses. In more dynamic and general equilibrium models, policies which improve the terms of trade in favor of agriculture in the short run may reduce urban demand for food stuffs, making the ultimate impact on deforestation inconclusive.

Pro-export policies designed to increase agricultural and forest product exports are likely to have stronger deforestation effects than policies that promote production for the domestic market. This is because an increased supply of agricultural exports is less likely to put downward pressure on prices, and dampen the initial effects of the policies. Similarly, pro-agricultural policies are likely to have stronger deforestation effects in the contexts of globalized agricultural markets and trade liberalization.

Since different crops / livestock products use distinct technologies, changes in the relative prices of different agricultural products may affect forest clearing more than changes in the general profitability of agriculture. This makes it impossible to predict how specific policies will affect forest clearing without looking at the changes they generate in prices for specific products and the pressure supplying each of those products puts on forests.

Off-Farm Employment / Wages

More off-farm employment and higher wages should decrease deforestation because agriculture and forestry become less profitable. One counter effect is that wage increases stimulate demand for certain agricultural and forestry products associated with deforestation.

Higher agricultural wage rates limit the deforestation effects of agricultural booms because they increase the costs of agricultural production. Thus, institutional mechanisms, such as unionization and rural minimum wages, which favor wage increases limit deforestation. Greater off farm employment can simultaneously reduce deforestation and diminish poverty.

Transportation Infrastructure

Greater access to forests generally leads to more deforestation. This is valid with regard to roads, forest fragments, coastal countries, and islands. The simple correlation between roads and deforestation, however, overstates the real causal relation because roads are partly endogenous. Nevertheless, no policy designed to reduce inappropriate deforestation can be considered comprehensive unless it includes clear guidelines regarding this issue.

Land Tenure and Land Markets

Deforestation is greater under open access regimes than when there are full property rights, and even greater when forest clearing allows people to obtain additional property rights. Paradoxically, however, making land tenure more secure in places where property rights were obtained by forest clearing gives an additional incentive to clear forests.

High land costs, when caused by land taxes or one-time price increases, discourage deforestation. However, continuously rising land prices, combined with the possibility of obtaining additional property rights by clearing land, encourage deforestation for speculative purposes.

Input Prices

Higher input prices reduce deforestation by making agriculture less profitable, but increase it by provoking substitution from more to less intensive technologies. The net effect is indeterminate. Studies from Africa suggest the second effect is slightly larger, implying that high input prices there encourage deforestation. In this context, the elimination of input price subsidies may lead to greater forest clearing.

Technological Change

Technological changes, such as new crop varieties, which increase yields without changing the demand for labor or capital, increase deforestation in the short run. They may reduce deforestation in the medium run if they lead to lower prices and that, in turn, discourages bringing new areas into production. If new technologies are more labor intensive, their short-run effect on deforestation is indeterminate. Since agriculture becomes more profitable, this stimulates increases in area, but since each unit of production requires more labor, and this can bid up wage rates, dampening the initial effect. Another relevant factor is whether the technological change is only applicable in already cleared lands or can also be applied to currently forested lands. Where labor supply is inelastic (as it is likely to be in the short run before migration can occur) and where the new technology only applies to already cleared land technological change is likely to reduce deforestation. The empirical evidence on this issue is weak, but tends to show that technological change has reduced deforestation.

Agricultural research in export crops is more likely to promote deforestation; research in non-tradables with inelastic demand less so. Investing in agricultural research designed to improve production in areas not threatened by deforestation is more likely to reduce pressure on forests, than supporting research designed to increase agricultural productivity in forested areas.

Interest rates / discount rates

High interest rates and discount rates reduce investment both in forest clearing and in forest management. These results, however, are based only on analytical models, with no empirical evidence.

Timber prices

Increased timber prices reduce land clearing for agriculture if there is not an open access situation and there are no capacity constraints on logging. In open access situations, higher timber prices simply reduce the costs of conversion. Timber trade restrictions shift timber production from high to low cost areas and to later periods. They don’t necessarily reduce total logging, but at any given moment they increase the area in forest which has not been logged. Policies which lower timber prices or increase costs from regulation reduce logging, but also investment in avoiding encroachment and in forest management. Having a more elastic labor supply will tend to dampen these effects. Global regressions give mixed results regarding whether logging is correlated with deforestation and are based on poor data.

Concession Policies

Concessionaires will only protect already logged forest from encroachment when they expect to log the entire concession area and the discounted value of future timber harvests is greater than the cost of managing the forest and avoiding encroachment. Under certain plausible assumptions there is no simple direct relation between concession duration and the probability a concession will be sustainably managed. Like the results regarding interest rates, these conclusions are based only on analytical models, not empirical evidence.

Environmental factors

Forests tend to be cleared more in drier, flatter, higher fertility areas, with adequate drainage.

(Visited 42 times, 1 visits today)
Copyright policy:
We want you to share Forests News content, which is licensed under Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0). This means you are free to redistribute our material for non-commercial purposes. All we ask is that you give Forests News appropriate credit and link to the original Forests News content, indicate if changes were made, and distribute your contributions under the same Creative Commons license. You must notify Forests News if you repost, reprint or reuse our materials by contacting forestsnews@cgiar.org.

Further reading