Nowcasting and the Public Interest

We recently wrote about the ways tools on Google such as Trends and Correlate can be used by just about anyone with a computer and an internet connection to better understand the world around them. We can safely say that Blawgconomics was not the first to find these tools interesting as there has been quite a bit of research and press, necessarily over the past few years, on how search data can be used to forecast or 'nowcast' any number of interesting things.

One of the latest examples of research in this vein (and a strong indicator that it isn't only amateur economists who are finding a use for internet data) is a report just recently released by the Bank of England entitled 'Using internet search data as economic indicators.' From the conclusion:

This article has considered the potential usefulness of internet search data as economic indicators. There remain some limitations of these data: there is only a short backrun, there is no information on the actual volume of searches, and as the index is based on a subsample the backrun of data can change. However, even in their current form, initial results suggest these data can be useful. In line with studies for other countries, internet search data can help predict changes in unemployment in the United Kingdom. These appear to be as useful as existing indicators. For house prices, the results are somewhat stronger: search term variables can outperform some existing indicators over the period since 2004. There is also evidence that these data may be used to provide additional insight on a wider range of issues which traditional business surveys might not cover. (emph. added)

The writers go on to say that the Bank will continue to monitor what is likely to become an increasingly valuable source of information.

There would, of course, be problems if governments and central banks were to rely solely on internet search trend data in making critical economic decisions (we can just see the headlines about the Central American republic trying desperately to hedge against The Bieber Effect based on the reports of a mid-level e-conomist now). However, used correctly, internet searches could provide very interesting data points for analysts trying to explain economic events.

The cynical among readers might be thinking 'Why not?...The masters of the economic universe could hardly do worse.' Blawgconomics would rather focus optimistically on the new jobs which could be created for those with a working knowledge of both economic theory and web search tools (we hope it isn't too gauche to note that such opportunities can of course always be forwarded to editor (at) blawgconomics.com, but we digress...).

No comments:

Post a Comment