The Impact of Formal Insurance Provision on Farmers' Behavior: A Randomized Trial in Rural Zambia
Ken Miura (Brown University)
Takeshi Sakurai* (University of Tokyo)
Abstract
This study presents empirical evidence of changes in farmer behavior after offering them weather index insurance contracts. To quantify the impacts of insurance on agricultural decision-making, this study makes use of the random allocation of free insurance to small-scale farmers in rural Zambia, while endogenous insurance demands are also investigated. Our empirical results show that the provision of insurance leads farmers to sow maize seeds earlier—a practice known to increase maize yield, but which is riskier in terms of rainfall variability. In addition, it is found that insured farmers enlarge the maize field size and use more fertilizer; that is, the provision of insurance encourages farmers to invest in maize production in a risky environment. We also report suggestive but interesting evidence that weather index insurance could substitute for small-livestock holdings, a conventional self-insurance tool used to mitigate income variation.
Junichi Ito* (Kyoto
University)
Haruka Asahi (Kyoto University)
Shinichi Kitano (Kyoto University)
Abstract
"Direct Payment Scheme for Hilly and Mountainous Areas" was introduced in 2000, with the aim of preventing any further abandonment of farmland in the areas and helping farmers to sustain continued agricultural production, and thereby enhancing multi-functionality of agriculture. Members in rural community decide whether or not to participate in the scheme, taking account of economic benefits accrued to them. This article, referring to the institutional mechanism as the government's indirect intervention, elucidates the economic consequences based on a theoretical model. While the direct intervention under which the government determines the cultivated areas maximizes social surplus, the indirect intervention leaves room for improving Pareto efficiency, suggesting that the former is superior to the latter in terms of policy effectiveness. However, rural communities are able to receive a rent under the indirect intervention. Our theoretical model proposes a hypothesis that communities with steep terrain are more likely to participate in the scheme, while those with small size and aging of the population are less likely to join it. The probit analysis verifies the hypothesis. PSM estimators do not provide deterministic assessment for treatment effects of the scheme; they are significant when PSM is applied to cross-sectional data, but not significant for the difference-in-difference.
川崎賢太郎* （農林水産政策研究所）
要 旨
本報告では、農家の死亡率に関する二つの分析結果を報告する。
第一に、市町村別のデータを用いて、職業と死亡率の関係を解明する。集計バイアスやセレクションバイアスに注意しながら計量分析を行った結果、農業者比率の高い市町村では、60歳以上の死亡率が有意に低いことがわかった。この結果は、農家の方が非農家よりも長寿であることを示唆している。死因別に見ると、老衰・循環器疾患・ガンなどが減る一方、交通事故（農機具事故含む）が増えることがわかった。
第二に、気温と死亡率の関係を解明する。両者の関係は、温暖化への懸念を背景に近年疫学や経済学分野で注目を集めており、通常よりも気温の高い日や低い日には死亡者が増加することがわかっているが、職業別の分析はほとんどなされていない。非線形性や順応効果等に注意しながら分析を行った結果、農家の方が非農家よりも高温の影響を受けやすいことがわかった。これらの結果を基に、猛暑や寒波が死亡者の増加を通じて耕作放棄地や休耕地といった農地のロスをどれほど増やすかについても議論される。
Altruism or Exchange? An Experimental Evidence on the Motives behind Private Transfer in Sri Lanka
Takeshi Aida* (National
Graduate Institute for Policy Studies)
Yasuyuki Sawada (University of Tokyo)
Abstract
This study develops a new approach to the classical question whether private transfer is motivated by altruism or exchange. By combining artefactual field experiment and standard household survey data, we can directly measure the degree of altruism and trust (i.e., exchange) and analyze their impact on the co-movement of consumption. We find that higher altruism toward lower income partners and their income difference reduce the difference in consumption, which supports the altruism hypothesis of a motive of private transfers.
Masahiro Shoji* (Seijo University)
Abstract
Using data from a unique household survey and an artefactual field experiment conducted in rural Bangladesh, this study tests the impact on trust of the extrinsic incentive to enforce cooperation in the risk sharing arrangement. The incentive is characterized by the patterns of exogenous income shocks in the real world and trustee’s risk preference, and trust is elicited experimentally. The incentive can facilitate trust by increasing one’s cost to take selfish behavior and opportunities of intensive communication. The empirical result is in line with the hypothesis. This result also suggests why the social proximity fails to predict trust in some of previous studies.
Mutual Interaction of Microinsurance and Informal Insurance: A Field Experiment in Cambodia
Seiichi Fukui (Kyoto
University)
Mitsuo Inada* (Kyoto University/JSPS Postdoctoral Research Fellow)
Fumiya Nakao (Kyoto University)
Abstract
This study investigates mutual interaction of formal insurance protection and informal insurance arrangements in developing countries. Although introducing formal insurance protection in developing countries is expected to play a important role for alleviating unexpected negative shocks for poor households, there is a trade-off between formal insurance and informal insurance arrangements. Investigating how we contribute to the poverty reduction by combining informal insurance arrangements with formal insurance protection is a significant policy question. Using a unique data from field experiments in Cambodia, our empirical result are consistent with crowding out effects that amounts of income transfer diminish once households can access to a formal insurance. We also find that traditional mutual cooperation within communities have a negative impact on the purchase of formal insurance protection. Furthermore, our results suggest that households with less risk averse and higher time discount rate tend to choose no insurance. It is reasonable that risk-loving households choose no insurance and households with smaller time discount rates buy a formal insurance.
Yasuharu Shimamura* (Kobe University)
Abstract
This paper explores the potential effect of infrastructure, particularly irrigation systems and rural paved roads, in relation to the global food price crisis that occurred in 2007-2008 and thereafter affected poor households in rural Indonesia. Our empirical analysis shows that there exists a certain threshold for the proportion of local paved roads that divides villages according to either lower or higher degree of local market integration to the surrounding markets. This study furthermore shows that irrigation systems in the villages with lower market integration mitigated the adverse effect of food price surge for food buyers at the expense of producers’ benefit.
タイ国家計の貯蓄行動―マイクロデータ利用による貯蓄関数の計測―
新谷正彦* （西南学院大学）
要 旨
本報告の目的は、タイ国の家計費調査の個別結果表を用い、恒常所得仮説とライフ・サイクル仮説を合わせた貯蓄関数の計測結果の報告にある。
チャールズ・ホリオカ(2010)は、アジア諸国について、人口の高齢化による家計と政府の貯蓄の低下の関係を実証し､貯蓄の低下の予測をおこなっている。その理論背景は、ライフ・サイクル仮説である。
世帯主の年齢によって、1990-2011年の期間の調査年のタイ国の家計費調査の平均値で貯蓄率を観察すれば、タイの家計の消費・貯蓄行動に対しても、ライフ・サイクル仮説が妥当しそうである。しかし、個別サンプルを観察すれば、支出の変動より所得の変動が大きく、ライフ・サイクル仮説のみで、タイの家計の消費・貯蓄行動を説明できないと思われる。したがって、恒常所得仮説とライフ・サイクル仮説を合わせた貯蓄関数の計測を試みた。その計測結果は、統計的にも、経済理論的にも良好であった。すなわち、Lusardi
(1998)を参考に、所得関数を計測し、そのパラメータを用いて各サンプルの恒常所得と一時所得とを推計し、この推計値を用いて貯蓄関数の計測をおこなった。
貯蓄関数の計測結果は、大きな所得変動に対して小さな消費の変動、すなわち、消費の平滑化を説明するものである。ホリオカの帰結に対する貯蓄関数の計測結果の意味づけは、今後の課題である。