Advantages and disadvantages of big data used as economic signals

Such new forms of data have been extremely useful for tracking economic activity on a regular basis during pandemics. They give us a quick sense of what’s happening in the economy, rather than waiting for government statisticians to match general quarterly estimates of activity in the economy to structured surveys as well as administrative data. These established processes can sometimes come in too slow to understand a rapidly evolving situation. The last two years have come across a variety of big data gaps.

However, it is also important to be careful when using big data to make anything more than tentative estimates about economic growth. In this context, let’s stick with mobility data for a while.

There have been a number of studies that show how in normal times Google mobility data syncs up with the underlying economic activity. But what about such uncertain times?

Economists at the Organization for Economic Co-operation and Development (OECD) took a closer look at the link between mobility trends and economic growth. They examined data from 51 countries in the second and third quarters of 2020 and 43 countries in the fourth quarter of that year. The researchers found that the impact of mobility indicators on economic growth weakened with each successive quarter. There are two possible reasons for this. First, policy makers have learned to target specific types of economic activity rather than impose an outright ban on movement. Second, both citizens and enterprises have learned to adapt to new forms of work and leisure.

What both the factors have in common is that we as a society have embraced the pandemic. The relationship between mobility indicators and economic growth has not remained the same throughout. In more technical parlance, the regression coefficients have changed as economic agents have largely learned to live with the virus. OECD data indicates that a 10 percentage point change in mobility was associated with a 2.2 percentage point change in economic growth in the second and third quarters of 2020, but only a 0.9 percentage point change in economic growth in the fourth quarter of that year. The year. This is a sharp decline.

This also means that one analyst trying to estimate the impact of changes in dynamics data on quarterly economic activity based on coefficients for the first and second quarters of 2020 will have to deal with another analyst working with coefficients from the fourth quarter. You will get very different results. This fact matters when trying to estimate the impact of the third wave on the Indian economy, especially when mobility data is considered an important element.

There are challenges with some other types of big data as well.

Consider night light, which is now being used by some economists as a proxy for economic activity. Input data on the lights placed on past sunsets comes from several satellites that can help pick up the intensity of light generated by humans in an area. In a recent working paper, Ayush Patnaik, Ajay Shah, Anshul Tayal and Susan Thomas of the research firm xKDR have highlighted that clouds interfere with the way data on night light is captured by satellites. Can hover kilometers above the ground. Four researchers show that there is a downward bias in readings during cloudy months, and have created an algorithm to at least partially correct this downward bias.

Another problem with interpreting big data is the context in which it is read. For example, in the months of the pandemic, consumer demand in many categories has shifted from services to goods, either due to lockdowns or fear of exit. The e-way bill generated is a very useful advance indicator of economic activity when goods move across the country. But there is no need to generate such bills for services. Therefore a wide shift in demand from services to goods is likely to lead to a sharp rise in e-way bills, which can be explained by overall economic activity. Similarly, a change in demand for services may show that e-way bill growth has slowed down. This does not mean that the economy has slowed down.

The easy availability of new forms of big data is certainly an opportunity for economic analysts. The examples above—on mobility data, night lights and e-way bills—have been used to show that such data still needs to be handled with care when it is used to make broader assumptions about the economy as a whole. is made to install. Some of these data points have also been aggregated into easy-to-follow indices of current economic activity. They have been extremely useful during pandemics, but collecting data with varying reporting frequencies as well as potentially different seasonal patterns is a difficult problem.

The use of big data in economic analysis is welcome. there is no turning back. However, there are still a slew of analytical puzzles that need to be solved.

Niranjan Rajadhyaksha is a member of the Academic Board of Meghnad Desai Academy of Economics