Agricultural Microinsurance: Global Practices and Prospects

This article is based on the book Agricultural Microinsurance: Global Practices and Prospects, published by the MicroInsurance Centre.

Introduction to agricultural microinsurance

Agricultural microinsurance is about providing insurance to small-scale farmers in developing countries. This in itself presents a number of particular challenges to insurers:

• Uncontrollable: Ideally the occurrence of an insured event should not be under the direct control of the insured person. However, this is not always the case with many kinds agriculture insurance.
• Unequivocal: Assessing agricultural loss can be very difficult and costly, as the loss could be caused by a combination of the insured-against events.
• Fraud: Farms are often physically remote, which creates opportunities for fraud.
• Moral hazard: Physical remoteness makes it hard for an insurer to check whether insured farmers are diligently taking care of their crops or livestock.
• Adverse selection: can have a destabilising effect on an insurance system, because the principle of risk-pooling will not work if only those negatively affected buy the insurance.
• Covariant risk: In agriculture, covariant risk is frequently an issue because droughts, pests and animal or crop epidemics are likely to affect many farmers at the same time.

All these factors, together with the costs of loss adjustment, can make agricultural indemnity insurance a very costly business, difficult to make profitable or indeed to break even. In fact hardly any agricultural insurance programs cover their costs (indemnity payments & administrative costs) from premiums. Almost all are subsidised, as agriculture is a much politicised sector.

Animal insurance: Livestock insurance can cover losses resulting from death, disease and accidental injury to livestock. It can cover an individual animal or a herd.

Crop insurance: Crop insurance covers the loss of crops due to one or more perils, and can be covered in a number of different way: yield loss (a lower-than-anticipated yield), quality loss (crops of a lower quality than anticipated), revenue loss (due to price fluctuations), or a combination of these. The two most common types of crop insurance are Named-peril crop insurance (policies payed out according to the actual damage that results) and, Multi-peril crop insurance (based on shortfalls on expected yield, rather than on the dam¬age caused by a particular loss event.)

Index-based insurance: is a way of providing protection against correlated risk such as extreme weather events. It is not strictly insurance, as individual losses are not assessed – instead it pays-out to all policy holders in a geographic area when certain conditions are reached in the proxy, or index.

Index insurance solves three of the most difficult challenges of agricultural insurance, and greatly reduces the prospects of fraud:

• Moral hazard: the farmer cannot influence an index that is based on weather.
• Adverse selection: whether farmers opt in or out, this will have no impact on the risk, because the risk would be based on the index, e.g. level of rainfall.
• Costs of loss adjustment: it is not necessary for a loss adjustor to visit the farm and calculate losses, as once the index trigger is exceeded, the payment is sent regardless of loss.
• Reduces the prospects for fraud.

Unfortunately index insurance is not quite a panacea, because it introduces new challenges. One challenge is basis risk, which can be described as the mismatch between the amount received because the index has been triggered and the amount actually lost by the client. Improved data collection and product design may be able to minimize basis risk however this typically makes the product more complex and more difficult for the low-income market to understand.

The landscape of agricultural microinsurance

The 2007 landscape survey undertaken by the MicroInsurance Centre, attempts to map the extent of agricultural microinsurance worldwide by studying specific areas of microinsurance such as regional distribution, types of microinsurance cover,  the risk carriers, the schemes and pilot programs, and finally the state of microinsurance regulations. The main conclusions of which are as follows:

• There are very few agricultural insurance schemes in developing countries with products that are accessible to poor farmers. The landscape survey found a total of 122 schemes worldwide, and not all of these are fully operational.
• Agricultural microinsurance is concentrated in Latin America.
• There are very few dedicated agricultural microinsurance schemes in developing countries. Those that do exist use existing agent infrastructures which end up perpetuating non-viable business models.
• Agricultural microinsurance is highly subsidized, and run on business models that are not sustainable.

Increasing access to agricultural microinsurance

Non-insurance intervention can play a direct role in mitigating risk, given the difficulties associated with agricultural insurance. The most obvious way to reduce risk is to prevent it from happening in the first place (vaccinating livestock, strengthening systems to prevent stock theft, planting more drought and pest-resistant crops.) Another common way of mitigating risk is for household members to share risk by pursuing multiple liveli¬hoods, including off-farm activities, and pooling their income. Other ways include sharecropping (farmer has a contract with a landowner to use land in return for giving the land¬owner a share of the farmer’s harvest), and the use of forward contracts, which reduces the risk of price fluctuation but agreeing contractually the price beforehand. Nevertheless, non-insurance options cannot mitigate the effects of catastrophic losses, other than intervention by governments and aid agencies; there are few real substitutes for insurance.

Interventions can take place at three levels: the macro level (supporting the policy environment), the meso level (supporting the infrastructure necessary to support agricultural microinsurance e.g. institutes to train insurance staff, reinsurers and agricultural support staff), and the micro level (improving the sustainability of risk carriers and distributors).

Macro level

Regulations: There is evidence that the lack of regulation or the existence of inappropriate regulation can impede the progress of microinsurance, although there has as yet not been any analysis of the impact of existing regulation on agricultural microinsurance. Evidence of regulatory impediments to the spread of agricultural microinsurance is mostly anecdotal. By and large there is a distinct lack of information regarding this area and more studies need to be allocated to examine either what regulations need to be put into place or what changes could improve current regulations.

Policy: There is invariably a need to subsidise agricultural microinsurance, and these come in various different forms (the distributor can be subsidised, the risk carrier can be subsidised, the government can take on the reinsurance risk at a subsidised rate…etc).

Another policy matter is whether to compel insurance companies to sell agricultural micro-insurance products. The Indian government has done this with microinsurance, with mixed results. On the positive side this policy has made India the world’s largest supplier of microinsur¬ance and an engine of innovation. On the negative side many insurers treat the policy as a “cost of doing business” and provide low quality, poorly serviced products.

Meso level

Microinsurance in general has a strong need for trained special¬ist staff (loss adjustors and actuaries), agricultural microinsurance necessitates additional skills, and for example in livestock insurance, relies on veterinarians for a number of functions such as managing risk, underwriting, loss adjustment and fraud control. To begin building specialist insurance capacity, development agents such as donors, gov¬ernments, development banks, NGOs and MFIs, need to undertake landscape studies at country level to establish the supply of agricultural insurance expertise and develop ways to improve it.

Another important area of intervention is improving data collection (livestock mortality, morbidity rates, weather…etc). This lack of data not only makes the design of products difficult but, also can act as a barrier to insurers considering entering the market.

Consumer education: A problem throughout the microinsurance sector, the lack of understanding and trust towards insurance will limit the demand and impede the functioning of a scheme, even when people do participate.

Micro level

Leveraging existing distribution infrastructure: In order to keep costs down it is clear that agricultural microinsurance products should be sold through an aggregator so the costs of selling and servicing agricultural microinsurance can be reduced by using its distribution infrastructure.

The availability of affordable reinsurance is patchy at best, and in tropical regions, which face frequent covariant risk events such as droughts, cyclones, plagues and floods, there is no real substitute for reinsurance. Studies need to be organised to understand what kinds of agricultural microinsurance risks commercial reinsurers would be prepared to cover, and which ones they would not, or else look at new means of providing reinsurance - by setting up a dedicated reinsurer, or subsidizing commercial reinsurance premiums, or working with government to accept some of this risk.

Better product design: Certain errors in product design are commonly repeated because products are not designed specifically for people on low income. In too many cases the insurer merely reduces the premium and benefit of conventional agricultural insurance, while keeping the other features constant, resulting in products that cover for minor losses and tend to be expensive and overly complex. The best microinsurance products are those designed through a process of close consultation with potential policy holders. More research needs to be done in terms of creating a list of the best and worst practices.

Reducing costs: We cannot avoid the conclusion that microinsurance is a low cost, high volume business, and unless costs are contained, agricultural microinsurance cannot be sustainable. A few interesting cost-contained strategies emerged in the case studies in Chapter 2.

• ADR-TOM, in its Burkina Faso plough oxen scheme, made very effective use of its policy holders in reducing the costs of risk management through group co-payments. Attaching the insurance product to existing group structures such as credit solidarity groups will greatly reduce cost.
• Technological advances are likely to bring costs down and increase access. The only significant technological innovation uncovered in the landscape survey was the use of sealed low cost weather stations that send their data via satellite for use in index insurance schemes.
• The cost structure of agricultural microinsurance simply does not allow for products to be sold individually though agents. Agricultural microinsurance needs more specialized aggregators who provide access to large numbers of potential policy holders and, are linked in some significant way to the livelihoods of the farmer policyholders.

Conclusion

Like traditional agricultural insurance, agricultural microinsurance is difficult and costly. For development agents wanting to help reduce the risks for low-income farmers, its adoption needs to be considered against a number of potentially easier to implement risk manage¬ment strategies. These include reducing the chance of the risk occurring in the first place (e.g. vaccinations in the case of livestock), or reducing the impact of the risk once it has occurred (e.g. through the provision of savings and credit facilities).

However useful these strategies are, they cannot deal with catastrophic risks such as droughts, cyclones, floods and plagues, and this is where agricultural microinsurance clearly does have a role to play. As the various interventions suggested show, there is an important role for development agents and other stakeholders to help it become an effective risk management tool.

Source:
Jim Roth & Michael J McCord, 2008: Agricultural Microinsurance – Global Practices and Prospects, published by the MicroInsurance Centre. Funded by the Ford Foundation in collaboration with the Microinsurance Network’s Agricultural Microinsurance Working Group.
Click here for the full report