The positive relationship between access to credit and household welfare has been well explored and documented in the literature. The literature however sees just few research papers about Vietnam. One issue with most research papers in this area however is that while the conclusions are consistent, they mostly draw their findings from the analysis of cross-section data obtained from household surveys. Such an analysis represents a snapshot view and does not tell us much about the long-term impact of access to credit on household welfare. In this paper, we develop an econometric framework to test the impact of access to credit on household welfare in the long-run using data from two household surveys in Vietnam. The first survey covers a sample of 4,799 households, 150 communes and 300 villages over the country. The second survey, taken after 5 years, covers a sample of 5,999 households, 194 communes and 388 villages, including all households surveyed in the first survey. Our econometric framework considers and controls for the effect of endogenous credit and of the sample selection bias. We employ a three stage regression: the first stage controls for the sample selection bias; the second stage controls for the endogeneity of credit; and the third stage is to estimate the impact of access to credit on household welfare, where the inverse Mill's ratios and the predicted residuals, which are computed from the first and second stages, are included as explanatory variables. We find that access to credit has a positive and significant long-term impact on household welfare in terms of per capita expenditure, per capita food expenditure and per capita non-food expenditure. We also find that although both formal and informal credit contribute to household welfare, formal credit has a relatively higher impact than its informal counterpart. Our findings imply some policy recommendations. On the one hand, the findings encourage policies that improve the access to credit for poor households in rural areas. On the other hand, the low level of impact suggests that the subsidized credit should be reconsidered.