Environment, technology, and the social articulation of risk in West African agriculture

Economic Development and Cultural Change

Volumn 45, Issue 3, 1997, Pages 557-590

  Carter, Michael R ^a  

^a UNIVERSITY OF WISCONSIN MADISON   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

AGRICULTURAL RISK; HOUSEHOLD ANALYSIS; RISK MANAGEMENT; SOCIAL RELATIONS; TECHNOLOGY ADOPTION;

BURKINA FASO;

EID: 0031422464     PISSN: 00130079     EISSN: None     Source Type: Journal    
DOI: 10.1086/452291     Document Type: Article

References (62)

1. J.-P. Platteau "Land Reform and Structural Adjustment in Sub-Saharan Africa: Controversies and Guidelines," Food and Agriculture Organization (FAO) Economic and Social Development Paper no. 107 (FAO, Rome, 1992); and D. Bromley and J. P. Chavas, "On Risk, Transactions, and Economic Development in the Semi-Arid Tropics," Economic Development and Cultural Change 37, no. 4 (1989): 719-36.
2. J.-P. Platteau "Land Reform and Structural Adjustment in Sub-Saharan Africa: Controversies and Guidelines," Food and Agriculture Organization (FAO) Economic and Social Development Paper no. 107 (FAO, Rome, 1992); and D. Bromley and J. P. Chavas, "On Risk, Transactions, and Economic Development in the Semi-Arid Tropics," Economic Development and Cultural Change 37, no. 4 (1989): 719-36.
3. See M. Watts, "On the Poverty of Theory: Natural Hazards Research in Context," in Interpretations of Calamities, ed. K. Hewitt (Boston: Allen & Unwin, 1983); and M. R. Carter and F. Zimmerman, "Mediating Risk through Markets, Rational Cooperation and Public Policy: Institutional Alternatives and Trajectories of Agrarian Development in West Africa," IRIS Research Program Working Paper (University of Maryland, 1994). Although not specifically focused on agrarian and food security issues, A. Banerjee and A. Newman ("Occupational Choice and the Process of Development," Journal of Political Economy 101, no. 2 [April 1993]: 274-94) trace the impact of risk on the perpetuation and perhaps deepening of initial endowment differentiation or inequality.
4. See M. Watts, "On the Poverty of Theory: Natural Hazards Research in Context," in Interpretations of Calamities, ed. K. Hewitt (Boston: Allen & Unwin, 1983); and M. R. Carter and F. Zimmerman, "Mediating Risk through Markets, Rational Cooperation and Public Policy: Institutional Alternatives and Trajectories of Agrarian Development in West Africa," IRIS Research Program Working Paper (University of Maryland, 1994). Although not specifically focused on agrarian and food security issues, A. Banerjee and A. Newman ("Occupational Choice and the Process of Development," Journal of Political Economy 101, no. 2 [April 1993]: 274-94) trace the impact of risk on the perpetuation and perhaps deepening of initial endowment differentiation or inequality.
5. See M. Watts, "On the Poverty of Theory: Natural Hazards Research in Context," in Interpretations of Calamities, ed. K. Hewitt (Boston: Allen & Unwin, 1983); and M. R. Carter and F. Zimmerman, "Mediating Risk through Markets, Rational Cooperation and Public Policy: Institutional Alternatives and Trajectories of Agrarian Development in West Africa," IRIS Research Program Working Paper (University of Maryland, 1994). Although not specifically focused on agrarian and food security issues, A. Banerjee and A. Newman ("Occupational Choice and the Process of Development," Journal of Political Economy 101, no. 2 [April 1993]: 274-94) trace the impact of risk on the perpetuation and perhaps deepening of initial endowment differentiation or inequality.
6. The analyses of C. Paxson ("Using Weather Variability to Estimate the Response of Savings to Transitory Income Fluctuations," American Economic Review 82, no. 1 [March 1992]: 15-33) and C. Udry ("Risk and Insurance in a Rural Credit Market: An Empirical Investigation of Northern Nigeria," Review of Economic Studies 61, no. 3 [1994]: 495-526) are good examples of these studies of risk that broadly build on the permanent income approach of A. Deaton ("Household Savings in LDC's: Credit Markets, Insurance and Welfare," Discussion Paper no. 153 [Woodrow Wilson School, Princeton University, 1991]). Paxson examines the degree to which consumption in Thailand is insulated from shared or covariate production shocks (e.g., broad weather patterns), while Udry examines Nigerian data to see the impact on consumption and asset stocks of individual-specific or idiosyncratic shocks (e.g., bird and animal damage).
7. The analyses of C. Paxson ("Using Weather Variability to Estimate the Response of Savings to Transitory Income Fluctuations," American Economic Review 82, no. 1 [March 1992]: 15-33) and C. Udry ("Risk and Insurance in a Rural Credit Market: An Empirical Investigation of Northern Nigeria," Review of Economic Studies 61, no. 3 [1994]: 495-526) are good examples of these studies of risk that broadly build on the permanent income approach of A. Deaton ("Household Savings in LDC's: Credit Markets, Insurance and Welfare," Discussion Paper no. 153 [Woodrow Wilson School, Princeton University, 1991]). Paxson examines the degree to which consumption in Thailand is insulated from shared or covariate production shocks (e.g., broad weather patterns), while Udry examines Nigerian data to see the impact on consumption and asset stocks of individual-specific or idiosyncratic shocks (e.g., bird and animal damage).
8. The analyses of C. Paxson ("Using Weather Variability to Estimate the Response of Savings to Transitory Income Fluctuations," American Economic Review 82, no. 1 [March 1992]: 15-33) and C. Udry ("Risk and Insurance in a Rural Credit Market: An Empirical Investigation of Northern Nigeria," Review of Economic Studies 61, no. 3 [1994]: 495-526) are good examples of these studies of risk that broadly build on the permanent income approach of A. Deaton ("Household Savings in LDC's: Credit Markets, Insurance and Welfare," Discussion Paper no. 153 [Woodrow Wilson School, Princeton University, 1991]). Paxson examines the degree to which consumption in Thailand is insulated from shared or covariate production shocks (e.g., broad weather patterns), while Udry examines Nigerian data to see the impact on consumption and asset stocks of individual-specific or idiosyncratic shocks (e.g., bird and animal damage).
9. H. Alderman and C. Paxson, in "Do the Poor Insure? A Synthesis of the Literature on Risk and Consumption in Developing Countries" (Discussion Paper no. 164 [Woodrow Wilson School of Public and International Affairs, Princeton University]), distinguish risk management (income stabilization) from risk coping (consumption stabilization). While much of this article is about what they call risk management, its concept of socially constructed effective risk is related to but distinct from their concept of partially stabilized income variability against which households employ risk-coping strategies. In particular, this article's concept of effective risk signals ways in which social phenomena (property rights, the social capital of reciprocity rules) affect both risk management and risk coping.
10. R. Townsend, "Risk and Insurance in Village India," Econometrica 62, no. 3 (May 1994): 539-92.
11. While the economics literature is awash with the presumption that higher-yielding and more remunerative activities are riskier, this need not be the case, as T. Walker and J. Ryan, Village and Household Economics in India's Semi-Arid Tropics (Baltimore: Johns Hopkins University Press, 1990) evidence. However, in the West African area under study here, farmer field trials of improved varieties and management practices found that these were indeed typically associated with greater downside risk, especially the increased fertilization levels necessary to obtain high yields from improved varieties; see P. Matlon and H. Vierich, Annual Report of ICRISAT/Upper Volta Economics Program, 1982 (Ouagadougou, Burkina Faso: ICRISAT, 1982).
12. While the economics literature is awash with the presumption that higher-yielding and more remunerative activities are riskier, this need not be the case, as T. Walker and J. Ryan, Village and Household Economics in India's Semi-Arid Tropics (Baltimore: Johns Hopkins University Press, 1990) evidence. However, in the West African area under study here, farmer field trials of improved varieties and management practices found that these were indeed typically associated with greater downside risk, especially the increased fertilization levels necessary to obtain high yields from improved varieties; see P. Matlon and H. Vierich, Annual Report of ICRISAT/Upper Volta Economics Program, 1982 (Ouagadougou, Burkina Faso: ICRISAT, 1982).
13. While maize, a third cereal grown by nearly all households in the regions, dominates yields of the other cereal crops, the extent of its adoption is limited by resource constraints. As shown in table 1, maize cultivation takes place only on tiny plots, close to the household residence where relatively high inputs of labor and animal fertilizers can be applied.
14. The data collection was undertaken by the West African program of ICRISAT. For discussion of sampling methodology, see the "Guide to Data Collection and Encoding Procedures" (ICRISAT, West African Economics Program, Ouagadougou, Burkina Faso, 1986).
15. At the household level, the pooled data set thus contains 300 observations (100 households x 3 years). To further augment the range of the production data, 150 observations derived from a 3-year panel on an additional 50 households in southern Burkina Faso were utilized. The econometric analysis is actually carried out at the individual plot level, so that the total pooled observations from the three regions number 740 for millet, 1,162 for sorghum, and 528 for maize.
16. C. Udry, "Credit Markets in Northern Nigeria: Credit as Insurance in a Rural Economy," World Bank Economic Review 4, no. 1 (1990): 251-69 discusses in detail the various sorts of events that underlie idiosyncratic shocks.
17. This section thus analyzes a subset of what Alderman and Paxson call income stabilization strategies.
18. t], where the t subscripts on the expectations operator indicates an expectation taken over the long-term, intertemporal distribution of returns.
19. J.-P. Platteau, "Traditional Systems of Social Security and Hunger Insurance: Some Lessons from the Evidence Pertaining to Third World Village Societies," in Social Security in Developing Countries, ed. E. Ahmad et al. (Oxford: Clarendon Press, 1991).
20. S. Coate and M. Ravallion, "Reciprocity without Commitment: Characterization and Performance of Informal Risk Sharing Arrangements," Journal of Development Economics 40, no. 1 (1993): 1-24.
21. M. Fafchamps, "Solidarity Networks in Preindustrial Societies: Rational Peasants with a Moral Economy," Economic Development and Cultural Change 41, no. 1 (1992): 147-74; E. Ostrom, R. Gardner, and J. Walker, Rules, Games and Common Pool Resources (Ann Arbor: University of Michigan Press, 1994); and Carter and Zimmerman, "Mediating Risk" (n. 2 above).
22. M. Fafchamps, "Solidarity Networks in Preindustrial Societies: Rational Peasants with a Moral Economy," Economic Development and Cultural Change 41, no. 1 (1992): 147-74; E. Ostrom, R. Gardner, and J. Walker, Rules, Games and Common Pool Resources (Ann Arbor: University of Michigan Press, 1994); and Carter and Zimmerman, "Mediating Risk" (n. 2 above).
23. M. Fafchamps, "Solidarity Networks in Preindustrial Societies: Rational Peasants with a Moral Economy," Economic Development and Cultural Change 41, no. 1 (1992): 147-74; E. Ostrom, R. Gardner, and J. Walker, Rules, Games and Common Pool Resources (Ann Arbor: University of Michigan Press, 1994); and Carter and Zimmerman, "Mediating Risk" (n. 2 above).
24. D. Norman, "Rationalizing Mixed Cropping under Indigenous Conditions: The Example of Northern Nigeria," Journal of Development Studies 11, no. 1 (October 1976): 3-21.
25. These taxes also encourage defection from the reciprocity, as Carter and Zimmerman ("Mediating Risk") and Fafchamps discuss in detail.
26. For examples, see J. Scott and B. Kerkvliet, "How Traditional Patrons Lose Legitimacy," Cultures et Development (Summer 1973), pp. 501-40; and T. S. Epstein, "Productive Efficiency and Customary Systems of Reward in Rural South India," in Themes in Economic Anthropology, ed. R. Firth (New York and London: Tavistock, 1967); and M. Watts, Silent Violence (Berkeley and Los Angeles: University of California Press, 1983).
27. For examples, see J. Scott and B. Kerkvliet, "How Traditional Patrons Lose Legitimacy," Cultures et Development (Summer 1973), pp. 501-40; and T. S. Epstein, "Productive Efficiency and Customary Systems of Reward in Rural South India," in Themes in Economic Anthropology, ed. R. Firth (New York and London: Tavistock, 1967); and M. Watts, Silent Violence (Berkeley and Los Angeles: University of California Press, 1983).
28. For examples, see J. Scott and B. Kerkvliet, "How Traditional Patrons Lose Legitimacy," Cultures et Development (Summer 1973), pp. 501-40; and T. S. Epstein, "Productive Efficiency and Customary Systems of Reward in Rural South India," in Themes in Economic Anthropology, ed. R. Firth (New York and London: Tavistock, 1967); and M. Watts, Silent Violence (Berkeley and Los Angeles: University of California Press, 1983).
29. H. Binswanger and J. McIntire, "Behavioral and Material Determinants of the Relations of Production in Land Abundant Tropical Agriculture," Economic Development and Cultural Change 36, no. 1 (October 1987): 73-100.
30. G. Feder and R. Noronha, "Land Rights Systems and Agricultural Development in Sub-Saharan Africa," World Bank Research Observer 2, no. 2 (July 1987): 143-70.
31. As such, the function predicts mean yields given rainfall, and whatever endogenous variable input reactions households make to it does not identify the marginal impact of rain, holding variable input choice constant. In yield-rainfall space, the MAY function can be conceptualized as the outer envelope of the ceteris paribus production functions defined for given levels of variable inputs. The endogenous reactions production units make to rainfall reactions should flatten the sensitivity of yields to rainfall. J. P. Chavas, P. Kristjanson, and P. Matlon ("On the Role of Information in Dealing with Risk: The Case of Sorghum Yield in Burkina Faso," Journal of Development Economics 35, no. 2 [1990]: 261-80) and P. Matlon ("Indigenous Land Use Systems and Investments in Soil Fertility in Burkina Faso," in Searching for Land Tenure Security in Africa, ed. J. W. Bruce and S. Migot-Adholla [Dubuque, Iowa: Kendall Hunt, 1994]) discuss in detail endogenous reactions to realized weather information.
32. As such, the function predicts mean yields given rainfall, and whatever endogenous variable input reactions households make to it does not identify the marginal impact of rain, holding variable input choice constant. In yield-rainfall space, the MAY function can be conceptualized as the outer envelope of the ceteris paribus production functions defined for given levels of variable inputs. The endogenous reactions production units make to rainfall reactions should flatten the sensitivity of yields to rainfall. J. P. Chavas, P. Kristjanson, and P. Matlon ("On the Role of Information in Dealing with Risk: The Case of Sorghum Yield in Burkina Faso," Journal of Development Economics 35, no. 2 [1990]: 261-80) and P. Matlon ("Indigenous Land Use Systems and Investments in Soil Fertility in Burkina Faso," in Searching for Land Tenure Security in Africa, ed. J. W. Bruce and S. Migot-Adholla [Dubuque, Iowa: Kendall Hunt, 1994]) discuss in detail endogenous reactions to realized weather information.
33. While this calibration of the rainfall measure to the actual planting date of the field seems to introduce a sort of endogeneity to the rainfall measure, it is unlikely to be especially important because in the study area crops are planted as soon as the seasonal rains have been sufficient to soften the ground and make planting possible. The beneficial effect of having measured rainfall relative to planting date is to eliminate from the data the occasional early and late rains that are not especially relevant to crop performance. The intention here is to get an estimate of the effect of cropping season rainfall on yields that is as reliable as possible; including extraneous rains would tend to flatten out that relationship. The use of a single rainfall measure does suppress intraseasonal variation in the rainfall pattern. In effect, the specification utilized estimates the effect of rainfall presuming an average intraseasonal pattern. The results thus will tend to understate yield variability. However, short of a vast increase in the complexity of the estimation that the data seem disinclined to support (i.e., estimating a sequence of fortnightly rainfall probability functions and including a matching set of fortnightly rainfall variables in the yield functions), there seems to be no way around this problem.
34. Over the 3 years of the panel, changes in family demography should be relatively modest. The fixed effect specification should soak up most of the differences in yields based on different consumer/worker ratios. Any extent to which the fixed effect specification fails to do so should not bias the rain-yield and rain error variance parameters. Such failures would, of course, increase residual variation of the production function. However, as noted in the text, I took a conservative approach in interpreting this variation as an indication of idiosyncratic risk.
35. This use of the Simpson index follows B. Blarel, P. Hazell, F. Place, and J. Quiggin, "The Economics of Farm Fragmentation in Sub-Saharan Africa," World Bank Economic Review 6, no. 2 (May 1992): 233-54.
36. It was also impossible to identify any effect of toposequence on the yield-rainfall relationship.
37. This approach is likely to provide a conservative estimate of idiosyncratic risk faced by households since some of the measured inputs that statistically explain yield variation (say, labor devoted to replanting a field) are in fact costly responses to idiosyncratic shocks. However, as suggested in the text, the conservatism of this procedure is likely to be overwhelmed by problems of measurement error, which will still show up as residual variation that cannot be distinguished from true idiosyncratic shocks. Short of direct measurement knowledge of the distribution of the idiosyncratic shocks themselves, there appears no way out of this conundrum. Nonetheless, the estimates derived here should still be useful indicators of the approximate magnitude of the shocks involved.
38. The full production function regression results are available from me on request.
39. In deriving these numbers, a large field size (5 hectares) was assumed and the intercrop variables were held at their mean values for the sample.
40. Udry, "Credit Markets in Northern Nigeria" (n. 10 above).
41. This assumption rules out health risks and tends to overstate the capacity of an individual to self-insure against idiosyncratic risk. Fields scattered by separate individuals under reciprocity arrangements would, however, insure against the idiosyncratic risk of ill health.
42. The failure to find a significant negative impact of field scattering on yields is not necessarily surprising. Given the extremely low levels of fertilizer use, the major problem implied by field scattering is likely to be labor time lost to travel.
43. The data indicate that the average Sahelian production unit contains approximately seven consumers (0.90 cultivated hectares per consumer), and that the average Transition Zone unit has 10 consumers (0.43 hectares per consumer).
44. The risk exposure measure is not an indicator of the starvation risk faced by production units. Because individuals may enjoy entitlements to a broader portfolio than the household's agricultural output for the current year, the objective risk numbers indicate the probability that other entitlements and accumulated assets must be tapped to realize a minimal subsistence level of consumption. Using data drawn from the same villages studied here, T. Reardon, C. Delgado, and P. Matlon ("Determinants and Effects of Income Diversification amongst Farm Households in Burkina Faso," Journal of Development Studies 28, no. 2 [January 1992]: 264-96) find that own-farm agricultural income comprises only 50%-60% of total household income. Other income sources are not well diversified, however, in the sense that they depend on the same risk components that affect own-farm production.
45. The "safety first" approach to risk of J. Roumassett (Rice and Risk [Amsterdam: North Holland, 1975]) and James Scott (The Moral Economy of the Peasant [New Haven, Conn.: Yale University Press, 1976]) implies that households will not trade off subsistence security against higher expected returns. Under this approach, the objective risk exposure measures give a complete ranking of alternative portfolios.
46. The "safety first" approach to risk of J. Roumassett (Rice and Risk [Amsterdam: North Holland, 1975]) and James Scott (The Moral Economy of the Peasant [New Haven, Conn.: Yale University Press, 1976]) implies that households will not trade off subsistence security against higher expected returns. Under this approach, the objective risk exposure measures give a complete ranking of alternative portfolios.
47. Besides its intuitive appeal (see, e.g., the discussion in H. Varian, Microeconomic Analysis [New York: Norton, 1994]), the assumption here of moderate constant relative risk aversion matches the findings of experimental work on risk preferences summarized in H. Binswanger and D. Sillers, "Risk Aversion and Credit Constraints in Farmers' Decision-Making: A Reinterpretation," Journal of Development Studies 20, no. 1 (1983): 5-21.
48. Besides its intuitive appeal (see, e.g., the discussion in H. Varian, Microeconomic Analysis [New York: Norton, 1994]), the assumption here of moderate constant relative risk aversion matches the findings of experimental work on risk preferences summarized in H. Binswanger and D. Sillers, "Risk Aversion and Credit Constraints in Farmers' Decision-Making: A Reinterpretation," Journal of Development Studies 20, no. 1 (1983): 5-21.
49. Rather than analytically calculating density functions for increasingly complex portfolios, I used mean utility values for long simulated time series to approximate expected utility.
50. It might be suspected that such costs will rise with increased usage of fertilizer and other inputs that must be transported.
51. By way of comparison, D. McCloskey, "The Persistence of Common Fields in Europe," in Peasants and Their Markets, ed. W. Parker and E. Jones (Princeton, N.J.: Princeton University Press, 1975) suggests that field scattering in medieval England reduced risk at a cost of 8% of annual output.
52. The match between observed behavior and that which appears reasonable in light of the estimates of risk and risk-coping possibilities is reassuring. While this article provides an empirical foundation for modeling behavior that would be optimal given the estimated structure of risk, there is no attempt in it to undertake this modeling effort. Building on this article's decomposition of risk into environmental, technological, and social components, Carter and Zimmerman ("Mediating Risk" [n. 2 above]) do undertake an explicit simulation of dynamically optimal behavior.
53. Coate and Ravallion (n. 14 above) use this specification.
54. See, e.g., the discussions in Scott and in Fafchamps (n. 15 above). While sharing rules built around an average production outcome will potentially support greater risk reduction, its impact here on downside subsistence risk of interest will not be significantly different from that obtainable around subsistence norm-based sharing rules. Moreover, while sharing rules built around average production outcomes can admit Pareto-efficient full risk sharing, real-world welfare systems and my own introspection suggest that subsistence-based norms are more likely to be supported by social capital.
55. While social sharing barely increases the certain grain equivalence value of a household's portfolio (see the data in table 4), safety first analysts (e.g., Scott and Roumassett) see disaster probabilities of 4% to be quite high, while 1% appears as a much more manageable degree of risk exposure. The transfers from the compound head to the 10 subordinate production units average 210 kilograms per year in the Sahel, and 100 kilograms per year in the Transition Zone.
56. F. Zimmerman, "Structural Change and Evolving Institutions in Developing Countries: A Dynamic Programming Simulation Analysis" (Ph.D. diss., University of Wisconsin - Madison, 1994) discusses the evidence of the contemporary weakening of the compound structure.
57. See M. Eswaran and A. Kotwal, "Credit as Insurance in Agrarian Economics," Journal of Development Economics 31, no. 1 (August 1989): 37-53; and Carter and Zimmerman, "Mediating Risk."
58. See M. Eswaran and A. Kotwal, "Credit as Insurance in Agrarian Economics," Journal of Development Economics 31, no. 1 (August 1989): 37-53; and Carter and Zimmerman, "Mediating Risk."
59. Put differently, empirical findings of smoothed consumption - particularly those that are disembodied measures of the actual devices being used - are not once and for all statements that risk is nonproblematic. The creative destruction of institutions is clearly part and parcel of economic growth, and evolving individual vulnerability has clearly been central, especially during transitional regimes of the sort that A. K. Sen (Poverty and Famines [Oxford: Oxford University Press, 1980]) labels PESTs, or pure exchange systems in transition. Moreover, the cost of successful risk-management and risk-coping strategies is not static in a growing economy. Continued rational pursuit of customary modes of insurance as technology, prices, and markets evolve leaves poor households vulnerable to what M. Carter, B. Barham, and D. Mesbah ("Agro-Exports and the Rural Resource Poor in Latin America" Latin American Research Review 31, no. 1 [January 1996]: 33-65) call exclusionary growth.
60. Put differently, empirical findings of smoothed consumption - particularly those that are disembodied measures of the actual devices being used - are not once and for all statements that risk is nonproblematic. The creative destruction of institutions is clearly part and parcel of economic growth, and evolving individual vulnerability has clearly been central, especially during transitional regimes of the sort that A. K. Sen (Poverty and Famines [Oxford: Oxford University Press, 1980]) labels PESTs, or pure exchange systems in transition. Moreover, the cost of successful risk-management and risk-coping strategies is not static in a growing economy. Continued rational pursuit of customary modes of insurance as technology, prices, and markets evolve leaves poor households vulnerable to what M. Carter, B. Barham, and D. Mesbah ("Agro-Exports and the Rural Resource Poor in Latin America" Latin American Research Review 31, no. 1 [January 1996]: 33-65) call exclusionary growth.
61. F. Zimmerman and M. Carter ("Dynamic Portfolio Management under Risk and Subsistence Constraints in Developing," Countries," Agricultural and Applied Economics Staff Paper 402 [University of Wisconsin, November 1996], and "Rethinking the Demand for Institutional Innovation: Land Rights and Land Markets in the West African Sahel," Agricultural and Applied Economics Staff Paper 400 [University of Wisconsin, August 1996]) show that individualized risk exposure creates incentives for further modification of tenure rules and the activation of a land market.
62. F. Zimmerman and M. Carter ("Dynamic Portfolio Management under Risk and Subsistence Constraints in Developing," Countries," Agricultural and Applied Economics Staff Paper 402 [University of Wisconsin, November 1996], and "Rethinking the Demand for Institutional Innovation: Land Rights and Land Markets in the West African Sahel," Agricultural and Applied Economics Staff Paper 400 [University of Wisconsin, August 1996]) show that individualized risk exposure creates incentives for further modification of tenure rules and the activation of a land market.