Rank three demand systems have been recognized to be best-suited for long-term demand projections due to their Engel flexibility. Using a fitted QUAIDS model, food demand in Vietnam is projected under scenarios that account for alternative growth rates in food expenditures, food prices and urbanization. Results indicate that at higher levels of expenditure growth, budget share of rice declines while budget shares of high-valued foods such as meat and drinks increase. Demand for rice is projected to decline in 2020 and further in 2030 both on a per capita basis and in total while demand for other foods increases. The projections also show that the effect of urbanization is more remarkable for rice while it is modest for non-rice food groups. Results of this study highlight the importance of considering the effect of income distribution and urbanization in long-term food demand projections.