Most of what we know about international migration is based on migration from developing nations to developed nations. In fact, in 2007 the World Bank estimated that about 74 million, or nearly half of migrants from developing countries, live in other developing countries (Ratha and Shaw, 2007). Since the end of the 1990s, with stronger economic and political ties between China and many African nations, migration from China to Africa has increased dramatically. It is now reported that close to 1 million migrants from the PRC are living in Africa (Park 2009; Li 2010). Chinese migrants in Tanzania fall into two main categories: “state-led” migrants employed by state-owned enterprises (SOE) on one-to-three year contracts linked to infrastructure projects, and “independent” migrants who are not associated with the government. Using the results from the Chinese in Africa Health Study (CAHS) pilot, conducted in the summer 2013 in Dar es Salaam, Tanzania, this paper will assess the risk preferences of the two types of migrant groups in order to better understand the risk profiles of people from China who choose to move to lower-income countries in Africa, and how risk preferences relate to health decisions and behaviors.

CAHS is a pilot study designed to establish the feasibility of a network-based sampling approach among Chinese migrants in Africa. 147 Chinese migrants were recruited in Dar es Salaam with an innovative network-based  sampling approach (Mouw and Verdery 2012).  In addition to their own socio-demographic characteristics, migration history, risk preferences, mental health, and health behaviors, respondents also provided information on the demographic characteristics of 1,308 Chinese alters living in Dar es Salaam. These data provide a description of socio-demographic characteristics, social networks, and health behaviors of a hard-to-reach population of migrants about which very little is known.

One question in the health economics literature is the extent to which risk preferences predict health behaviors, particularly risky behaviors such as smoking, drinking, or use of preventive care. Risk preferences have been found to be negatively associated with being overweight or obese, seatbelt non-use, cigarette smoking, and heavy drinking (Anderson and Mellor 2008). Using CAHS data, we hypothesize that the independent migrants may be less risk averse than SOE migrants. Risk preferences were measured based on hypothetical gamble questions. Preliminary regression results show “state-led” migrants are relatively more risk averse compared to independent-led migrants when we control for age, gender, years of education, hukou status, marital status, and ability to speak English. When controlling for the same variables and migrant status, people with insurance are less risk averse than those without insurance. Ever smokers are also less risk averse than never smokers, and frequency of alcohol use is negatively correlated with risk aversion. Characterizing risk profiles of different migrant groups may help policymakers design more effective interventions that can encourage more preventative behaviors among Chinese migrants in Africa and target segments of the populations most in need of these interventions.