Crop insurance gone wrong – Why subsidised crop insurance won’t work

Farmers have been applying excessive amounts of fertiliser under the subsidy scheme, therefore if they were to face the market prices the coupon will not 

compensate their fertiliser requirements fully – Pic by Shehan Gunasekara

On 22 September 2016 I wrote about the importance of crop insurance programs, different ways of implementing crop insurance schemes and the most important things to be considered in implementing a successful program. However I limited myself from evaluating or constructively commenting on the crop insurance programs on the ground as of today. 

There are several crop insurance programs operating to back up the investments of farmers. The majority of these programs are backing up paddy farmers, however some programs do cover other field crops as well. The Government along with only a handful of private sector companies is managing crop insurance programs in Sri Lanka but it was all by the Government initially. The crop insurance program by the Government has some interesting ups and downs. 

Agriculture insurance program for Sri Lanka started in 1958, targeting the ‘Maha’ season. The initial program targeted 26,000 acres of paddy land in five major districts and it started as a pilot initiative. By 1974, nearly 16% of the total area cultivated with paddy had come under insurance in both seasons. Policymakers then decided to expand the coverage of the program therefore the scheme was expanded to other crops like green gram, cowpea, chilli, soya bean and even livestock, especially cattle and poultry. Year 1973 stands alone as the year where the Crop Insurance Board (CIB) was established under the Parliamentary Act No. 27 of 1973. The mandate of the Crop Insurance Board is to operate a comprehensive agricultural insurance scheme for the benefit of the farmers in respect of rice, other field crops and livestock. 

Since 1974, the scheme has been funded with finances from the Government-consolidated fund and also under self-finance schemes. The Agricultural and Agrarian Insurance Board Act No. 20 of 1999 came into effect from 16 August 1999 mainly to look after and administer the self-financing component of the agricultural insurance scheme. 

The Agricultural and Agrarian Insurance Board (AAIB) was established under the following main two objectives: 

(1) To launch an insurance scheme and a social security benefit scheme covering the farming and the fisheries sectors and providing old age benefits through principal activities such as granting pensions and social security benefits for the farmers and the fishermen.

(2) To achieve the desired goal of bettering the farm and farm life, ensuring their effective participation in the overall national production and the enhancement of their lifestyles. 

Back then the insurance program was voluntary, and research shows that it had a very low participation rate. However, the management of the crop insurance program was transferred to the National insurance Trust fund (NITF). In 2012, the previous Government tied the crop insurance program to the fertiliser subsidy. 

At that time the paddy farmers were given a significant subsidy on the fertiliser (close to 90% of the fertiliser cost was subsidised). The Government then made mandatory crop agriculture insurance on all farmers by indirectly increasing the price of a 50kg bag of subsidised fertiliser by Rs.150 to Rs.500 from the earlier price of Rs.350 as a compulsory premium for the crop insurance scheme. All farmers were automatically enrolled in the insurance scheme at agrarian centres countrywide when they purchase subsidised fertilisers. 

However, in year 2015 the current Government removed the fertiliser subsidy by introducing a coupon scheme. The coupon is worth Rs. 25,000 per year. Therefore, if a farmer is doing two cultivations “Yala” and “Maha,” there will be only Rs. 12,500 per season. While the introduction of the coupon system brings in many arguments, the objective of the article is not to comment on that but let me highlight a few points since we are here. 

Farmers have been applying excessive amounts of fertiliser under the subsidy scheme, therefore if they were to face the market prices the coupon will not compensate their fertiliser requirements fully. The coupon system is based on the fertiliser recommendation by the Department of Agriculture. At the same time a land that is being used to excessive fertiliser might not be productive if the fertiliser application is reduced suddenly. Finding organic fertiliser as a substitute has many issues and I have written plenty about that.

However the most important point I want to make in this article is the sustainability of the crop insurance program. The previous system was in a way self-financed, though it was mandatory (I have many questions on attaching the insurance program on to a subsidy program, but records show that it was self-financed). The insurance premium collected from all the farmers who were under the fertiliser subsidy allowed the insurance program to be self-sufficient, making claims for paddy farmers whenever the harvest was damaged by natural causes. Therefore in theoretical terms, the program was an “Indemnity based system”. 

When the Government introduced the coupon system, they did not made any attempt to incorporate the insurance premium in to the coupon system, basically eliminating the premium collection system. For a moment let’s assume that the decision to not build the insurance premium into the coupon system is intentional. This asks two main questions: 

(1) Was it a good idea to incorporate the insurance premium in to an input subsidy scheme in the first place?

(2) What should have been done when the subsidy system changed in to a coupon system?

My attempt here is to answer these two questions. 

Was it a good idea: Piggybacking insurance on an input subsidy scheme?

To begin, I argue that buying insurance is a personal choice. It reflects to a certain extent the risky nature of the activity and the perceptions of the person. Therefore, a farmer can be either a risk loving person, a risk neutral person or a risk averse person. Therefore the decision to buy insurance will vary from farmer to farmer. 

If the farmer is cultivating with a very low probability of risk to natural disasters the likelihood of that farmer buying insurance is low. However, the previous system basically forced the farmer to pay an insurance premium rather than making it a voluntary system. The forced system, especially when it was tied to an input subsidy might result in popular insurance problems such as “adverse selections and moral hazards”. Collectively we call them the “principal-agent problem”, which is a result of the information asymmetry. 

Crop insurance, when heavily subsidised, can even have important negative social impacts. Subsidies for risk management have similar effects as subsidies on any other farm input; it encourages over use. And since the reduction in production costs is partly paid for by the subsidy, the dead weight loss of the subsidy is always greater than the combined benefits to producers and consumers. 

In practice, this implies that risk management subsidies reduce risk costs to farmers to below their true social value, leading to excessive risk taking (e.g. growing unsuitable crops in high risk regions) and increased exposure to future drought losses by the farmer. Subsidies not only create dependence on future drought assistance from the Government, but also lower social welfare. The implication is that, wherever possible, Government interventions should be limited to risk management programs that decision-makers pay for themselves. 

There could be couple of reasons why an insurance scheme piggybacked on an input subsidy would fail in the long run. Private insurance companies typically do not insure yield losses due to pest and diseases, and prefer to write insurance against specific and insurable perils; it is rare to see a private company that writes multiple risk or all risk insurance, and the ones that have had very short lives. Also public insurers are often mandated to extend their insurance to small farms, and this can add enormously to administration costs. 

By the same token, programs that offer contracts on a field-by-field basis (as in the US) are also expensive to administer. Furthermore inappropriate incentive problems arise within insurance institutions when the Government underwrites most of their programs. When insurers know that the Government will automatically cover most losses, they have little incentive to pursue sound insurance practices when assessing losses. In fact, they may find it profitable to collude with farmers in filing exaggerated or falsified claims. 

Finally, governments undermine public insurers for political reasons. In such a case the important question to ask is, “Why should farmers purchase crop insurance against major calamities (including drought) if they know that farm lobbies can usually apply the necessary political pressure to obtain direct assistance for them in times of need at no financial cost?” 

Without a well-diversified insurance portfolio, crop insurers are susceptible to covariability problems, and face the prospect of sizeable losses in some years. Since public insurers are rarely able to obtain commercial insurance or contingent loan arrangements, this specialisation increases their dependence on the Government. 

What could have been done during the transition?

When the new Government came in to power the administration of the agriculture insurance program that was tied to the fertiliser subsidy was again given back to the Agriculture and Agrarian Insurance Board (AAIB). However, when the fertiliser subsidy was replaced by the coupon system, it was not quite clear on the steps that were taken to ensure the longer-term sustainability of the agriculture insurance program. 

One might argue different steps that could have been taken for example: Building the insurance premium into the coupon system, just like what was there with the subsidy system. In there, the farmer could be either asked to pay the usual insurance premium when the coupon money was issued, or the coupon money can be issued after deducting the insurance premium. We could think some other initiatives as well. Since fertiliser companies will be increasing their sales with increased fertiliser prices it is possible to encourage them to contribute for a “Crop Insurance Levy”. 

At the moment, all the financial institutions including micro-finance companies pay a “Crop Levy” that is collected to a fund and being managed by the National insurance trust fund (NITF). This money obviously can be put to better use, but it is quite possible that the Government might use that to support the agricultural insurance program for the moment. 

In the short term it is a viable idea, but not in the long run. In the long run such money should be used to promote the private sector in participating in crop insurance programs. As explained earlier, one of the main reasons for the private organisations not to take part in crop insurance is the risk involved. Therefore, money collected through a crop levy is a suitable option to back up the risk of the private agricultural insurance initiatives. It is in a way sustainable since you collect the levy from the same organisations where the risk is being backed up. 

However, in my opinion none of these are longer-term solutions. It is important that the private sector gets involve in the agriculture insurance and farmers are making a choice on whether to buy insurance or not, rather than being forced. Involvement of the government sector increases the possibilities of adverse selection. In an ideal situation, and I believe Sri Lanka can easily reach there; you need following criteria for a successful crop insurance program. These information are not my own synthesis, people have done many research before recommending these points. 

The ideal crop insurance program should have following characteristics: 

(1) It is affordable and accessible to all kinds of rural people, including the poor 

(2) It compensates for catastrophic income losses to protect consumption and debt repayment capacity 

(3) It is practical to implement given the limited kinds of data available 

(4) It can be provided by the private sector with little or no government subsidies 

(5) It avoids the moral hazard and adverse selection problems that have bedevilled crop insurance programs.

If you are focused on these points above, what is needed is an area-based index contracts, such as regional rainfall (and other weather) insurance as oppose to what we have now, an indemnity-based system. It might take time for us to move away from a subsidised crop insurance program to a market-based program. As mentioned earlier the current system is a good source of the economic problem “the adverse selection” (I am not saying that any other system would not have this issue, but here the probability is high. Also acting against it is relatively impossible). 

When adverse selection is present several things would happen: 

(1) Higher overall price as the Government insure themselves against high risk farmers taking out insurance.

(2) Low-risk farmer may not want to buy because it is relatively too expensive – leading to a missing market. This is why you would see the protests from farmers for being forced.

(3) Government may invest considerable time in identifying which groups of farmers are higher risk. 

The remedies against adverse selection are straight forward, to avoid adverse selection, firms need to try and identify different groups of people and then calculate different premiums, presumably a higher rate form high-risk farmers. However, it is clear that this would not work in a system that is heavily subsidised and mandatory. 

What is the final take home message? I am not suggesting that the Government should simply give up the current crop insurance system though it is forced and heavily subsidised. This is because something is needed to protect farmers until a proper market system is developed. Therefore in the short run, getting money from the “Crop Levy” and protecting farmers against the risk is acceptable. 

However we need to start working on private sector engagement (By 2007 there was only one private sector organisation doing crop insurance. I don’t think that figure has changed much). One way to do this is to provide protection against the risks of the private insurers (the Central Bank to a certain extent covers the cost of small loans given by financial institutions. This need to be extended to the insurance sector may be using the “Crop Levy”). 

Because we are concerned about food security, self-sufficiency in rice, seasonality, natural disasters aggravated by climate change and consumer sensitivity to commodity prices, one could argue a forceful system is better for Sri Lanka. However, I argue otherwise, agricultural insurance has to be based on the risk, the opportunity cost of cultivating in a risky area or being exposed to a natural disaster (which is the insurance premium) should match the willingness to pay for insurance. Then farmers will buy insurance. In a mandatory system, especially if the premium is a flat rate that is heavily subsidised, a farmer with a lower perceived risk will be at a disadvantage. 

(Dr. Chatura Rodrigo is an economist in Agriculture and Environment. The author can be contacted through [email protected] or 94 77 986 7007).