Access to credit is universally believed to be an effective tool to help the poor out of poverty. Yet the evidence for this has not considered all settings, especially the peri-urban areas of rapidly industrialising Asian countries. In these areas human capital is the main asset of the poor, so it is important to understand the input of credit on human capital. Therefore, this thesis begins with Chapter 2 showing the importance of human capital in income generation in Vietnam during the economic transition. The thesis then examines factors affecting credit participation and credit constraints for the poor in the peri-urban areas, and investigates whether credit participation impacts the poor’s education and healthcare spending and benefits their children’s schooling. Chapter 2 employs five large datasets of Vietnam Household Living Standard Surveys (VHLSS) conducted in 1998, 2002, 2004, 2006 and 2008 by Vietnam General Statistics Office (GSO) to examine the rate of return to schooling in Vietnam over the period of 1998-2008. The chapter finds that the rate of return has increased quickly during the recent economic reform and reached around 9-10 percent. The chapter clearly indicates an increasing importance of education in earnings during the later part of the economic transition in Vietnam. Therefore, human capital investment, including healthcare and education, is needed to help the poor escape poverty since they rely heavily upon labour income, especially in urban and peri-urban areas. One of the typical solutions to improve the poor’s human capital is to provide access to credit resources, however, there are many barriers blocking the poor’s access to credit. Chapter 4 uses a novel dataset collected by the author from peri-urban areas of Ho Chi Minh City, Vietnam in 2008 to examine how the poor use their loans, and factors affecting their credit participation and credit constraints. The chapter finds the presence of many commercial banks in the areas does not help the poor, who rely heavily on informal credit. Loans in the peri-urban areas are mainly used for non-productive purposes, which stresses the importance of consumption smoothing motives. Further, households in more rural wards have a higher probability of borrowing than more urban households, thanks to better community relationships and interpersonal trust. Competition by borrowing neighbours adversely affects the opportunity for borrowing in urban wards where the poor households’ borrowings rely more on subsidized credit funds. A closer look at specified microcredit sources reveals that household behaviours differ in each market segment. Furthermore, the poor are highly credit-constrained. Wealthier households, in terms of asset holdings and phone possession, appear less credit-constrained. However, except in the most rural part of the study area, the likelihood of credit constraints increases with distance to the nearest banks, which suggests that supply-side intervention could help in overcoming credit constraints. Overall, the poor in urban wards are more credit-constrained because of exclusion by commercial banks and weak interpersonal trust. Given that a sizeable fraction of the poor have participated in credit activities, there is a debate about whether microcredit has positive impacts on education and health for borrowing households. To provide evidence for this debate, Chapter 5 mainly uses the Propensity Score Matching (PSM) method to examine the impact of household credit on education and healthcare spending by the poor in the peri-urban areas. In addition to matching statistically identical non-borrowers with borrowers, my estimates also control for household pre-treatment income and assets, which may be associated with unobservable factors affecting both credit participation and the outcomes of interest. The PSM estimates show significant and positive impacts of borrowing on education and healthcare spending. However, multiple ordered treatment effect estimates reveal that only formal credit has significant and positive impacts, while informal credit does not have significant impacts. Whether the effects of credit are homogenous across distributions of outcome variables is another question of interest. This question asks whether the impact is the same along the outcome distribution, such as for households with already high consumption versus those with low consumption, or already high healthcare spending versus the low spenders. Chapter 6 employs a Quantile Treatment Effect estimator (QTE) and finds heterogeneity in the impacts on household budget shares for education and healthcare. Finally, household credit for the poor was examined and found to have a positive influence on current expenditure on education. However, to test whether the credit to the poor has longer term effects on education, in Chapter 7 there are results for estimating the impact of the credit on child schooling. Probit, Negative Binomial (NB) and PSM estimates roughly indicate no strong evidence of an effect, especially of informal credit, although formal credit may have a positive impact on child schooling.