Changes in risk attitudes and vulnerability to idiosyncratic and covariate shocks - Evidence from panel household data in Thailand and Vietnam

In this paper we aim to explain temporal variation in individual risk attitudes by an exogenous measure of idiosyncratic and covariate shocks. We estimate the impact of idiosyncratic and covariate shocks in terms of variation in consumption, at the individual and at the aggregate level, using multilevel modeling. The empirical basis for our analysis is a panel data set of 2812 respondents from rural Thailand and Vietnam that was collected in 2008 and 2010. 

We find that on average, risk attitudes change over time in Thailand and Vietnam. In addition, we show that idiosyncratic shocks trigger changes in risk attitudes of respondents from Thailand, whereas covariate shocks affect risk attitudes of respondents from Vietnam. The impact is also different for poor and non-poor respondents. The results suggest the existence of a negative feedback loop, where shocks increase poor individuals’ risk aversion and consequently their likelihood of remaining in poverty. 

We conclude that respondents living in persistent poverty in Vietnam perform better in insuring idiosyncratic shocks through risk-sharing and safety nets. In Thailand, such mutual insurance mechanisms across individuals are not likely to perform well.

Click here to view paper.