«Der Open-Access-Publikationsserver der ZBW – Leibniz-Informationszentrum Wirtschaft The Open Access Publication Server of the ZBW – Leibniz ...»
2001) have shown that adverse selection reduces the consumption of insurance by low-risk individuals or businesses, and results in the transfer of income from low-risk to high-risk insured. MIYAZAKI (1977) and WILSON (1977) demonstrate that, when it is impossible or highly-expensive to distinguish between low- and high-risk insurance applicants, the insurer prices insurance contracts at an average premium for all individuals. That results in undercharging high-risk customers and overcharging low-risk customers for similar contracts.
Past experience suggests that most popular crop insurance schemes, particularly multi-peril yield insurance and revenue insurance, are rather prone to adverse selection and moral hazard.
GOODWIN (1993) illustrates the effects of adverse selection on the actuarial performance of the US crop insurance program, demonstrating that only farmers whose risk is above average are likely to purchase insurance. The results of a study conducted by JUST et al. (1999) suggest that participating farmers tend to be those with higher-than-expected indemnities, as farmers with lower-than-expected indemnities are priced out of the program. They conclude that the domination of high-risk farmers in the insurance market can lead to market failure.
MIRANDA (1991) argues that area-yield insurance offers numerous advantages over individualyield crop insurance. Because information regarding the distribution of the area yield is generally available and more reliable than information regarding distribution of individual yields, insurers could more accurately assess the actuarial fairness of premiums under an area yield policy, thereby significantly reducing adverse selection problems. The use of an insurance product based on an index should eliminate the problem of asymmetric information between government and insurance companies, as well as between insurance companies and farmers, since all involved parties have symmetric information regarding the contract, and problems of moral hazard and adverse selection can be reduced significantly. However, SKEES and REED (1986) show that the potential for adverse selection depends on a farmer’s subjective assessment of the expected yield and the variability of the yield. They argue that premium rates based only on the mean crop yields of a region can lead to adverse selection, particularly when the variance of yield fluctuates considerably between farms. This aspect might be even more serious in a transition country, where farm productivity and production technologies could be rather heterogeneous in the initial stage. In this view, weather-based index insurance products provide some advantages because of the objective nature of the parameters that trigger indemnity payments. VARANGIS et al. (2002) argue that the weather can be independently Raushan Bokusheva verified, and therefore is not subject to the possibility of manipulation. Pre-conditioned, reliable assessment of area-yield based insurance can have similar benefits to weather-based index insurance.
3.2 Incentives for farmers and insurance companies to participate in crop insurance Realization of the law of large numbers is closely connected to incentives for farmers to buy insurance. If insurance is voluntary, then farmers’ participation in crop insurance would depend on, among other factors, how well it is suited to their needs. According to the conducted farm survey in Kazakhstan, features of insurance contracts such as sensitivity to changes in weather conditions (60.8 percent of the respondents), timing of contract fulfillment (44.6 percent) as well as the possibility of selecting a reasonable coverage (28.4 percent) and regional differentiation in contract design (24.5 percent) were referred to as main preconditions for the farmers' participation in crop insurance. Additionally, the farmers mention the cost of insurance as an important factor of their willingness to buy insurance. In this view, most farmers would tend towards insurance against only a group of the most serious natural hazards they face, as opposed to multi-peril insurance, provided that it would lower insurance costs. According to survey results, drought represents the most important natural hazard to grain production in the region, therefore, weather-based index insurance is likely to be accepted by farmers there.
However, since other important risks cannot be insured under this insurance product, farmers with multiple risks may desire another insurance scheme to provide coverage against their further risks. On the other hand, insurance contracts that are designed to protect against losses from a multitude of hazards may present challenges in terms of accurately assigning a probability of loss and determining an appropriate insurance rate (GOODWIN, 2001). This issue is even more critical if only limited historical yield data is available, as is the case in transition countries, where, due to restructuring, new entities have been emerging. Using regional data, however, may not accurately reflect the true likelihood of losses for individual farmers. As MIRANDA (1991) suggests, area-yield crop insurance provides incentives to farmers whose yields strongly correlate with the aggravate area yield. As the farm survey results demonstrate, this applies for most large farms in the investigated regions. Therefore, this insurance product can find acceptance by large farmers in Kazakhstan as well.
Furthermore, farmers, who in addition to high yield-variability face high price risk, could be interested in a revenue insurance scheme. In the context of an underdeveloped market infrastructure, price risk is of great importance to Kazakh farmers. According to the farm survey results, 64.4 percent of the interviewed farmers would like to have income insurance (HEIDELBACH et al., 2004).
Another important aspect of insurance market development associated with insurability is readiness of the private insurance sector to extend their services to agriculture. As results of structured interviews with insurance experts in Kazakhstan show, insurance companies are strongly distrustful to business in agriculture. Most of them do not possess any expertise in providing agricultural insurance. Those small parts of insurance companies, which do have some experts in the field, do not believe that risks in Kazakh agriculture can be privately insured. Additional aspects that hold them from involvement in the crop insurance market are high administrative and transaction costs, problems with monitoring and controlling moral hazard, and heavy regulation of the crop insurance market. Considering that both, area-yield insurance and weather-based-index insurance possess some advantages compared to traditional insurance products with regard to the above-mentioned problems, they could serve Crop insurance in transition: a qualitative and quantitative assessment of insurance products 15 as an "lead-in" for private insurance during the initial stage of development in the private insurance market in a transition economy. However, area-yield crop insurance, as well as weather-based-index insurance, does not solve the problem of risk pooling when systemic risk is present. In this case, an engagement on the side of either state or financial markets is inevitable for dealing with the problem.
3.3 Effects on farmer’s production patterns
An important issue treated in the literature concerns effects of insurance on farm productivity and production practices (CHAMBERS and QUIGGIN, 2002; COBLE et al., 1997; SMITH and GOODWIN, 1996). Reducing farmers' risk through insurance has been identified as affecting land use and inducing changes in production decisions. The effects of crop insurance on production pattern changes originate from the fact that under crop insurance, risk-averse farmers will behave as if they were risk-neutral (CHAMBERS, 1989). In view of the problem of marginal production areas with less productive farms in Kazakhstan and some other transition countries, this effect of insurance can be even more serious and severely distort factor allocation. Crop insurance can motivate farmers to choose a riskier bundle of outputs, inputs, and production practices that make farming more risky. Regarding this general problem, the literature concerns the optimal design of insurance contracts. CHAMBERS (1989) considers a contract-based approach, where insurance is designed with respect to an incentive compatibility constraint based on the agent’s first-order conditions for choice of inputs. MIRANDA (1991), MAHUL (1999) and BOURGEON and CHAMBERS (2003) examined the design of area-yield crop insurance with regard to the farmers "beta"-coefficient relating a farmer's yield to the risk pool’s yield.
On the other hand CHAMBERS and QUIGGIN (2004) argue that by having access to fair insurance, the producer does not need to engage in costly self-insurance. In the framework of statecontingent approach the authors show that by looking for a cost-minimising bundle of risk management tools and the technology to reach the optimal level of state-contingent income, the producer will be required to equalise the rate at which the risk management tool and technology balance out the state-contingent incomes. In this context the challenge is to apply this approach to empirical investigations into crop insurance design and pricing.
3.4 Feasibility and financial viability
Feasibility of an insurance scheme plays an important role considering applicability and viability of an insurance product. From this point of view, index-based insurance schemes provide some important advantages over other insurance schemes. Primarily due to their capacity to reduce transaction costs on the insurance market. For instance, in the case of transition countries where many small farms have emerged, area-yield crop insurance could allow to manage to some extent the problems of limited data availability. On the other hand, as serious differences in farm productivity could be present during transition, using area-yield as a reference value for risk pooling should be considered with caution. Thus, weather-based insurance can be viewed as a more advanced insurance product under these circumstances.
Like other crop insurance products, weather-based insurance cannot solve the problem of systemic risk pooling. However, due to similarities with weather derivatives, weather-based index insurance can prepare farmers for the potential adoption of such advanced financial instruments. An important precondition regarding the establishment of a weather-based index insurance product is the development of hydro-meteorological services and the provision of reliable and affordable weather information for insurance market participants. This issue Raushan Bokusheva underlines the importance of institutional frameworks. As most transition economies experience high budget restrictions, policy-makers have to pay attention to the insurance schemes which can be run privately, without any subsidization, or only on a small scale. Most attention must, however, be paid to the institutional accompaniment of the development of rural financial markets, in particular the crop insurance market.
At the initial stage of insurance market development, a great deal of attention must be paid to educating potential customers on insurance matters. In light of bad experiences with insurance during the Soviet era, farmers in most transition countries are skeptical about crop insurance.
Hence, pilot projects must be started to convince farmers of the advantages of their participation in the initial stages of crop insurance market development. In this regard, a strong engagement of government and public agencies must be present.
To summarize, in the view of a less-developed financial market in a transition economy, crop insurance can be considered as a possible instrument of a farmer’s income stabilization. The analysis shows that area-yield insurance and weather-based index insurance provide more advantages compared to multi-peril crop insurance and revenue insurance also in the
transition context. These advantages include:
• AYCI and WBII are introduced to manage systemic risk;
• Since only systemic risk is to be insured, insurers can more accurately assess the actuarial fairness of premiums, and thus reduce the adverse selection problems;
• Both schemes have relatively low transaction costs;
• AYCI is better applicable given prevailing data limitations;
• WBII is less bureaucratic, and thus provides less scope for corruption;