Finance & Real Estate

2020-06-05 Andrea Beltratti, Alessia Bezzecchi

The Importance of Big Data for Real Estate

Having a reliable statistical basis is indispensable for operators in the real estate sector. Possessing more precise information can help understand the areas of particular difficulty and help economic policy (and the financial sector as well, given the enormous direct and indirect exposure of banks) to better understand the crisis to try to limit it and transform it into future growth

In addition to the loss of numerous human lives, the COVID-19 pandemic is causing significant economic harm to our country. An additional aspect, which is just as important, though less discussed, regards the role of data. It is sometimes said that data is the "new oil," to reference the potential wealth linked to the creation and use of sources of reliable data. This is not really a new discovery: still today history books speak of how the great civilizations of the past, from the Egyptians to the Assyrians, distinguished themselves precisely for the systematic collection of empirical evidence on commerce and nature. There are few doubts of the fact that a more careful management of data could have at least cushioned the losses inflicted by COVID-19. The broad discussion on the possibility to use an app to detect and record citizens' movements and interactions has raised a great debate in Italy, with a certain delay compared to what has happened in other countries, that for some time now have been attempting to rationally and consciously address the costs and benefits linked to the use of data.

A central element regards knowledge of the current situation. Over the course of time, academic studies have tried to anticipate official statistics, that are published at increasingly short intervals, but rarely more than monthly, to reach the frequency of weekly or daily studies (called "nowcasting"); the goal is to estimate what is happening in "real-time" to certain sectors or to the economy as a whole. In February of this year, light was shed on the daily indicators of Chinese economic activity, illustrated for example in various contributions by the Financial Times, that made it possible to examine the progress of the situation in China and understand what could happen elsewhere if the pandemic struck with the same intensity as in Wuhan.

The SDA Bocconi and Assoimobiliare Real Estate Innovation Lab was born precisely with the aim of providing a stronger empirical basis for the analysis of this central sector of our economy. Unfortunately, throughout the world, information on real estate is of lower quality than what we see in other markets, due to the structure of the industry, meaning the transactions that are limited to a part of the stock, which can have very variable characteristics from one period to another. In Italy, probably due to the presence of a culture more interested in the deduction of Descartes than the induction of Hume, we are still a bit behind other countries.

How can we assess the state of the sector in this moment of crisis? In the absence of official data, we can refer to various sources and make some hypotheses. One of these, in real-time, consists of the statistics of the searches conducted on Google Trend, whose significance has been demonstrated, among other things, by the results reached by Wu and Brynjolfsson, who used them to predict price and sale dynamics a quarter ahead of time. They showed, for example, that in the United States an increase of one percentage point in the index correlates with an increase of 67,220 homes in the subsequent quarter.[1] Many other researchers have proven the strength of these statistical results in an international context outside of the United States. We have considered the index of searches relating to the real estate or commercial property sector for Italy. The statistics analyzed, that are available starting with 2004, show the existence of a strong drop in the interest towards these keywords, both in relation to the average of the first four-month period of various previous years, and as regards the dynamic between January and April of 2020.

Another source of data, specifically relating to the commercial real estate sector, is that of the average data of investment volume calculated by the principal brokers. A linear regression between the annual percentage value of investment volume and the growth of real GDP in Italy, in the period from 2007 to 2019, produces a sensitivity (measured by the beta coefficient of the regression) equal to 3.85. In theory, the coefficient can be used as a multiplier of any scenario of a decrease in GDP to estimate the impact on the investment volume in the commercial sector, even though the low statistical significance of the relationship implies a broad margin for error in the prediction.

The available data thus allows for a very partial response to initial demand. We know that the sector is suffering from the crisis, but it would be useful to have more exact information to understand the areas of particular difficultly and to help economic policy (and also the financial sector, given the enormous direct and indirect exposure of banks) to better understand the crisis to attempt to mitigate it and transform it into future growth. The advantages of having a reliable statistical basis are indispensable for everyone, and in particular for operators in the sector.

 

Andrea Beltratti is a Professor in the Department of Finance of the Bocconi University, where he teaches Economics of the Real Estate Market and Equity Portfolio Management, and Academic Director of the Executive Master in Finance (EMF) at the SDA Bocconi School of Management.

Alessia Bezzecchi is Associate Professor of Practice in Corporate Finance & Real Estate at the SDA Bocconi School of Management, where she is Program Director of the Executive Master in Finance (EMF) and of the Executive Program in Real Estate Finance and Real Estate (EPFIRE).



[1] L. Wu, E. Brynjolfsson, "The future of prediction: how google searches foreshadow housing prices and sales," SSRN, 2013.

Real estate_big data