MSU Agricultural, Food, and Resource Economics
The majority of previous studies on agricultural risk management use static models and, for the most part, ignore use of borrowing and lending as an alternative method of managing risk. This study examines the interaction between credit, insurance, and liquidity constraints using a simple dynamic model for a risk averse farmer who uses revenue insurance to manage risk and also borrows and lends subject to a credit constraint. Theoretical and numerical results are provided to support the hypothesis that liquidity constraints can have a large impact on optimal insurance decisions.
Three theoretical results are derived with the following implications. First, with no liquidity constraints, a risk-averse farmer will choose full coverage of actuarially fair insurance, even if borrowing and lending is allowed. Second, with no liquidity constraint a positive premium loading reduces optimal coverage level below full coverage. These two results show that in a dynamic model with no liquidity constraints, insurance choices are not influenced by the desire to smooth consumption, as long as complete and wellfunctioning credit markets exist that permit efficient consumption smoothing to take place. Third, even if insurance is actuarially fair, a binding liquidity constraint reduces optimal coverage below the full coverage level. Implying that, a binding liquidity constraint may cause farmers to purchase insurance less often than would be expected in the absence of the constraint.
The numerical model was solved for a representative farm from Adair County in Iowa and provides the following implications. First, with complete and well functioning credit markets: (i) the maximum allowable coverage of actuarially fair insurance will always be optimal; (ii) at moderate premium loading (e.g. 30%), the maximum of 85% coverage allowed in practice will still be optimal; and (iii) at relatively high (e.g. 60%) premium loading the maximum allowed coverage will no longer be optimal except for highly indebted farmers. Second, a liquidity constraint causes a reduction of coverage below the maximum allowed level, even for actuarially fair insurance. A binding liquidity constraint limits (or eliminates) the insurer's ability to borrow for current expenditures including consumption and insurance. This causes him/her to not insure out of current wealth because current consumption is too valuable. Finally, an area-based insurance scheme exposes insurers to residual uninsurable risk which may preclude them from purchasing insurance, even if it is actuarially fair and there is no liquidity constraint. Hence, subsidies may be necessary to encourage the maximum allowable coverage.
This study has two main conclusions: (1) as long as complete credit markets exist and the farmer can borrow and save freely, consumption smoothing has no effect on insurance decisions if insurance is moderately priced and there is no residual uninsurable risk; (2) if residual uninsurable risk and/or a liquidity constraint exist then consumption smoothing can have a significant impact on the optimal insurance decision and, in some cases, self-insurance will be preferred over formal insurance.