National Bureau of Economic Research Working Paper 15396, October. Hongbin Cai, Yuyu Chen, Hanming Fang, and Li-An Zhou (2009).

In the first study of a randomized microinsurance program, Cai et al. find that microinsurance provision significantly increased the number of sows raised by rural farmers in southwest China, and contend that though currently understudied, microinsurance may be as important a tool as microfinance in poverty alleviation.

Rural farmers in developing countries face significant risks relative to their income, yet many do not have access to insurance.  Likewise, many rural farmers would benefit from expanding in scale or using more technology, but do not have access to credit.

International development institutions have done a meritorious job extending credit in developing countries.  Grameen Bank is one prominent example.  However—and curiously—insurance provision has not received the same attention.  There have been few studies of microinsurance schemes upon which development institutions can base their programs; this paper seeks to redress that gap with an examination of a large, randomized program providing insurance to rural farmers for raising sows in southwest China.  The results are notable, as it is the first study of the causal impact of a randomized microinsurance program.

Pork is a large industry in China, comprising 48.4% of livestock production. 

However, large pig farms are rare, and pigs are generally raised in small numbers by rural families in their backyards.  Due to the small scale, rural Chinese hog farmers often face significant risks, including natural disasters and new and spreading hog diseases.  The authors report that hog mortality is high due to “backward breeding technology, weak swine farm infrastructure, poor vaccination and veterinary drug abuse” (8).  These problems have hurt production in recent years, leading to a 60% price increase between 2006 and 2007.  Consequently, the Chinese government began to subsidize hog raising, such as by offering subsidized insurance for sow mortality.   Coverage for 1,000 yuan was offered at a price of 12 yuan (down from the unsubsidized price of 60 yuan).

Methodologically, assessing the impact of insurance schemes is more difficult than assessing the impact of credit schemes because people choose to take up insurance based on their risk preferences, which may also affect their investment choices, effort, and output.  The authors set up their randomized experiment carefully to navigate this problem, and vary the insurance packages offered to isolate the true causal effect of insurance on output.

Animal husbandry workers (government employees charged with offering insurance to rural farmers through the government program) working in 480 villages were randomly assigned to one of three incentive packages: a control group offered a fixed reward of 50 yuan for participating in the study, a low incentive group offered a fixed reward of 20 yuan plus 2 yuan per sow insured, and a high incentive group offered a fixed reward of 20 yuan plus 4 yuan per sow insured.  These incentives were considerable considering the monthly salary of animal husbandry workers is only 15 yuan.

Looking at the average number of sows insured out of the average number of sows total in each village, in control group villages, 15 out of 29 sows were insured, compared to 22 out of 28 in low incentive villages and 27 out of 30 in high incentive villages.

The authors find that offering insurance significantly increased sow raising.  Specifically, low and high incentive villages insured approximately 9.6 and 12.0 more sows than control villages.  Subsequently, this led to 7.3 and 9.1 more sows being raised in low and high incentive villages in the next year.  The authors therefore conclude that farmers are underinsured through informal insurance options. 

One puzzle emerging from the study was the low insurance take-up rate in control villages.  Despite the fact that farmers had to pay only 12 yuan to insure 1,000 yuan worth of loss, the take-up rate remained only around 50%.  The authors argue that this low take-up rate has to do with trust, an issue that is particularly important in the microinsurance case relative to the microfinance case, since with microinsurance individuals are required to pay a certain cost up front and trust that they will receive the benefits later.  The authors report that “The issue of trustworthiness of government policies is particularly relevant in China since governments at all levels often renege on their policy promises, and from the viewpoint of Chinese farmers, local bureaucrats at townships are always searching for ‘legitimate’ reasons to ask them for money and sometimes even cheat them into paying unnecessary fees.  More importantly, if local governments fail to deliver their promises in the contract, there is virtually no way for farmers to sue the government in the court” (24).  In this case, the insurance policy itself represents a risk.  Therefore, farmers’ trust level for government-provided insurance may be an important barrier when it comes to insurance uptake.

Overall, this study represents an important first examination of the impact of microinsurance on rural farmer productivity.  It finds that formal insurance significantly increases the number of sows raised by farmers.  The authors conclude that microinsurance “may be as important as microfinance in poverty alleviation, and microinsurance can supplement and strengthen the effects of microfinance by protecting the farmers from the inherent risk of entrepreneurial activities” (32).