Our Bureau The price level of residential house based on advertisement data is found to be lower than the level of residential house price based on registration data, according to an article in Reserve Bank of India’s latest monthly bulletin.

The article’s authors, who picked up data on residential property prices from select web portals and processed it using big data tools to compile a residential house price index — WEB HPI — observed that the advertised prices are displayed at a discount to attract buyers whereas the final transaction takes place at a higher cost. “Overall, the WEB Index is found to be lesser than HPI, implying that the price level based on registration data is higher than the price level indicated in the advertisements.

“Further, the WEB Index graphs are smoother than the HPI. It means advertisement prices are not changed frequently, whereas, the HPI based on actual registration data is having higher price variation over time (as the actual sale varies significantly),” the article said Moreover, the registration prices are embedded with several underlying factors of the property such as floor rise, position, facing, and other factors, which may not be displayed in the advertisements, emphasised authors Indranil Gayen and Sasanka Sekhar Maiti from RBI’s Department of Statistics and Information Management.

HPI is compiled by the RBI every quarter based on the official data of property transactions collected from the registration authorities of respective State governments for 10 major cities — Mumbai, Delhi, Chennai, Kolkata, Bengaluru, Lucknow, Ahmedabad, Jaipur, Kanpur and Kochi. The current WEB-HPI is a derived index based on web-crawled data covering all the cities under HPI except Kochi.

Referring to the divergence between HPI and WEB indices, the authors said this could be due to the fact that the advertised prices are changed at a longer time horizon, whereas the actual prices of properties are much higher as various underlying factors are incorporated in it.

Declining trend

“Several reports on real estate market research state that there is a declining trend in the sale of the residential properties from 2013-14 onwards. As a result, the properties displayed in the advertisements may not be representative of the set of properties actually sold.

“As per the compilation methodology of HPI, the set of residential properties differ from one quarter to another, which is dependent on the actual sale during the underlying period,” the authors said.

Residential property price is an important piece of information for policymakers, especially for the central banks. In the Indian context, there are two established housing price indices — HPI compiled by RBI; and RESIDEX, which is compiled by the National Housing Bank. These indices are typically available with a lag of (at least) one quarter.

In order to obtain an early signal on residential house prices, an attempt has been made by the authors to compile an alternate residential HPI based on data available on the web portals of select real estate agencies in India. Big data tools were employed to collect, process, and store these data.