Impact Evaluation of an Affordable Housing Program in India
In the absence of viable affordable housing policies, the percentage of India’s urban population that cannot afford a home will increase substantially in the next couple of years. Poor households will be most severely affected because, in the absence of access to affordable housing, these households will very likely continue to settle in slums or poor urban areas. There are several factors that hinder the construction of mass affordable housing in India. Construction is constrained on both the supply and demand sides. On the supply side, there is a lack of long-term financing options for implementers of the construction of affordable housing, particularly in low-income states like Odisha, Bihar, Madhya Pradesh, Chhattisgarh, and Rajasthan. On the demand side, there is a perceived risk of investing in the poorer segments of the urban population.
Program and Evaluation Details
The UK Government's Foreign, Commonwealth & Development Office (FCDO) and the Indian National Housing Bank (NHB) developed a program that focuses on increasing access to affordable housing through three key mechanisms. First, the program finances construction of affordable housing in eight low-income states in India. Second, the program finances home loans for low-income households. Third, the program provides technical assistance to the Indian government and several private sector partners in order to strengthen policies to promote affordable housing.
AIR is partnering with KPMG and IMRB in designing and implementing a mixed-methods research design to determine the impact of the affordable housing program. To identify the contribution of the program to policy change at the sector-level, we conduct in-depth interviews and focus groups with representatives of FCDO, NHB, housing finance companies, and the state-level housing board in Rajasthan, Chhattisgarh, and Madhya Pradesh. To determine the impact of the housing program on access to credit, housing, socioeconomic and health outcomes, we use a propensity score matching design that includes 2,760 households.
