Estimation of an endogenous switching regression model with discrete dependent variables: Monte-Carlo analysis and empirical application of three estimators

The performances of alternative two-stage estimators for the endogenous switching regression model with discrete dependent variables are compared, with regard to their usefulness as starting values for maximum likelihood estimation. This is especially important in the presence of large correlation coefficients, in which case maximum likelihood procedures have difficulties to converge. Monte-Carlo simulations indicate that an estimator that corrects for conditional heteroskedasticity of the residuals is superior in almost all instances, and especially when maximum likelihood is problematic. This result is also obtained in an empirical example in which off-farm work participation equations of farm women are conditional on farm work participation status. First version received: July 1995/final version received: March 1998     

Intergenerational Succession in Small Family Businesses: Borrowing Constraints and Optimal Timing of Succession

Small family businesses differ from non-family businesses in that their functioning is not independent of the life cycle of the owner-operator, and in that other family considerations sometimes lead to sub-optimal managerial decisions from the point of view of the business. This is why a smooth intergenerational succession is essential to the profitability of the business, and to the welfare of the family as a whole. Succession within the family involves first of all the choice of a successor. The choice is affected by birth order, age differentials, and qualifications of potential successors. Choosing a successor means reaching an agreement about the timing of succession and income distribution before and after succession. This paper focuses on the decision of the business-operating family when to bring in the designated successor. A utility-maximizing time is shown to differ from the income-maximizing time only in the presence of binding borrowing constraints. Such constraints are likely to enhance an earlier succession in order to use the successor's accumulated off-business assets to ease the constraints and to increase future business income due to earlier accumulation of business-specific human capital by the successor. An additional model shows that the successor will not be willing to wait indefinitely for the formal ownership transfer of the business, because of the risk of being disinherited in some future period. The consequences of possible strategic behaviors of both the owner and potential successors on the results of these models is discussed informally.     

Off-Farm Work Participation of Israeli Farm Couples: the Importance of Farm Work Participation Status

This article claims that coefficients of off-farm work participation equations for farm residents might be estimated inconsistently if selectivity based on farm participation is ignored. Participation equations will be different for farm residents who do not work on farm, especially with respect to the dependence of reservation wages on farm attributes. I estimate an endogenous switching regression model in which farm and off-farm participation equations are estimated jointly while the off-farm participation coefficients are different for those who work on farm and those who don't. Using the 1981 Israeli Census of Agriculture data, I reject the hypothesis of insignificant selection bias and the hypothesis of equal coefficients in the two subsamples.     

Labor Supervision and Institutional Conditions: Evidence from Bicol Rice Farms

This article analyzes labor supervision activities reported in a survey of rice farmers in the Bicol region of the Philippines. We present a theoretical model that relates supervision intensity to institutional conditions. We estimate a supervision intensity equation and control for the decision to hire labor, the decision to supervise hired labor, and the type of the labor contract. The empirical estimates use different village-level spatial and demographic indicators as proxies for institutional conditions. The results are consistent with the prediction of the theoretical model that supervision increases with weaker institutional conditions.     

Plot size and maize productivity in Zambia: is there an inverse relationship?

This article examines the relationship between maize productivity and plot size in Zambia. It offers a unique empirical approach. First, it focuses on maize, which is the major crop on small and medium size farms in Zambia, but also accounts for the endogenous determination of the size of the plot devoted to maize. Previous studies have used total farm size or harvested area. Second, it corrects for selectivity in maize cultivation. Third, it controls for differences in land quality and weather conditions across districts. Finally, it offers a structural interpretation of the above framework by modeling farm decisions as a sequential, two‐stage process, in which land is first allocated to the different crops based on the information set of the farmers at the time of planting, and the yield is affected by subsequent application of inputs, the quantities of which may depend on additional information that is revealed after planting. We use this recursive structure and the differences in the information sets over time to identify the model. The results show that the endogeneity of plot size is very important in this analysis. When considering plot size as an exogenous explanatory variable, we find a monotonic positive relationship between the yield of maize and plot size, indicating that economies of scale are dominant throughout the plot size distribution. However, when we correct for the endogeneity of plot size, we find that the inverse relationship dominates the economies of scale in all plots up to 3 ha, which constitute 86% of our sample. These results suggest that market imperfections should be targeted by any policy aimed at increasing maize productivity in Zambia.     

Simultaneous estimation of work choices and the level of farm activity using panel data

We analyse the simultaneous determination and evolution overtime of work choices and the level of farm activity, using apanel of Israeli farm households. We estimate jointly a multinomialchoice model of work activity and an endogenous switching regressionof farm size, accounting for unobserved heterogeneity and correctingfor simultaneity bias. Our results imply that some family farmstend to expand over time and specialise in farming, whereasother farm households downsize their farming operation and increasetheir engagement in the off-farm labour market. Therefore, thesize distribution of farms is converging towards a bimodal distribution.     

Farm work, off-farm work, and hired farm labour: estimating a discrete-choice model of French farm couples' labour decisions

We model jointly several types of discrete-choice labour decisionsof farm couples: farm work, off-farm work and hired farm labour.Rather than estimating a set of reduced-form participation equationsthat is subject to an internal inconsistency problem, we estimatea 16-choice multinomial model. We find that farm labour of operatorsand of spouses are substitutes. Hired farm labour increaseswith the farmer's qualifications, perhaps substituting for thecouple's labour inputs. Other adults in the household substitutefor the farm labour input of the farm couple and hired workers.We find that the organisation of work by French farm householdshas changed over time, in the direction of a more equal sharingof tasks and responsibilities between members of the farm family.     

Trade and the rate of income convergence

To the extent that trade policy affects trade flows between countries, the ramifications can be far-reaching from an economic growth perspective. This paper examines one aspect of these ramifications, namely the impact of changes in the extent of trade between countries on changes in the rate of reduction in the size of the income gap that exists between them. Export and import data are used as the criteria for determining bilateral trade between major trade partners, resulting in the creation of 127 pairs of countries on the basis of export data and 134 pairs on the basis of import data. An increase in trade between major trade partners - and, in particular, increased exports by poorer countries to their wealthier partners - is shown to be related to an increase in the rate of convergence between the countries.