Evaluating the Accuracy of Homeowner Self-Assessed Rents in Peru

Attributing a rental value to the dwellings of homeowners is essential in various contexts, including distributional analysis and the compilation of national accounts, consumer price indexes (CPIs), and purchasing power parity indexes. One of the methods for making the attribution is to use homeowner estimates of the market rental value they would pay (receive) for their dwellings if these were rented. This is known as homeowner self-assessed rent. However, homeowner estimates may not be accurate because of the way questions aimed at soliciting such information are phrased, the sentimental attachment of the homeowners to the properties, lack of information about rental markets, and other reasons. Yet, researchers and practitioners often neglect to ascertain the accuracy of homeowner assessments. This study argues that comparing unconditional or conditional means may be misleading if one has not ascertained whether the observable characteristics of homeowner and tenant dwellings are similar. Using Peruvian data from 2003 to 2017, the study tests the accuracy of self-assessed rental values with matching estimators. In Metropolitan Lima, homeowners typically provide accurate estimates of the rental market values of their dwellings. In rural areas, market rental values are underestimated by homeowners in more instances. The direction and magnitude of the inaccuracies in Metropolitan Lima and in rural areas are comparable and range between ?25 percent and ?20 percent.