Will Internet search trends offer early signals about the timing of the next recession? No one really knows, but the possibilities are intriguing. Several researchers in recent years have considered the idea that tracking Internet activity offers a window into the future.
A 2009 paper from two Google researchers considered the prospects, asking: "Can Google queries help predict economic activity?"
Economists, investors, and journalists avidly follow monthly government data releases on economic conditions. However, these reports are only available with a lag: the data for a given month is generally released about halfway through the next month, and are typically revised several months later.
Google Trends provides daily and weekly reports on the volume of queries related to various industries. We hypothesize that this query data may be correlated with the current level of economic activity in given industries and thus may be helpful in predicting the subsequent data releases.
We are not claiming that Google Trends data help predict the future. Rather we are claiming that Google Trends may help in predicting the present. For example, the volume of queries on a particular brand of automobile during the second week in June may be helpful in predicting the June sales report for that brand, when it is released in July.
Earlier this year, a pair of New York Fed economists wrote:
Internet search counts possess useful information, not available in other variables, to now-cast or forecast the trajectory of some financial market data. While this predictive power is by no means universal—as we observe above, for a number of markets, Internet search data do not provide explanatory power beyond that of more traditional forecasting methods—the basic message is of a useful addition to the economist’s toolkit.
A related line of research also considers the possibilities of using Internet search data to analyze/predict investment returns. For example, a studypublished last year in The Journal of Finance finds a connection between Google searches and equity prices (here's a working-paper version of the analysis). Another recent study reports that "asset prices are positively related to the growth rate of Google’s search, trading volume and the level of Google search clicks."
Given this small but growing corner of research, perhaps it's no surprise to find that Google searches on the word "recession" spiked in late-2007 and early 2008—right about the time that the Great Recession began, according to the NBER. The NBER didn't formally announce the onset of the recession until 12 months after the fact. More timely warnings were available, however, although all came with the usual caveats that harass real-time analysis of the business cycle. But that's a topic for another day.
Meantime, the recession-search indicator via Google Trends looks quite tame at the moment. You can find support for thinking positively from more traditional economic indicators, such as the Chicago Fed National Activity Index. Based on CFNAI's update through April, recession risk appears quite low.
Some analysts argue otherwise, and with the euro crisis still raging and perhaps worsening--again--it's easy to imagine a less accommodating future. But if the outlook for the U.S. economy is set to turn darker, one might expect that we'll get wind of the change relatively early via selected word-search queries at Google Trends.
Should we take Internet searches seriously as an early warning sign of changes in the business cycle. Yes, according to the latest study on this intriguing research topic:
In this paper, we propose a novel business cycle surveillance system that utilizes the query logs of search engines for business cycle modeling. The system employs an effective feature selection technique to identify query terms that are representative of business cycles…. Experimental results based on a five-year dataset show that the proposed system can classify the status of business cycles accurately, and the selected query terms reveal interesting human behavior patterns in different business cycles. Unlike economic variables, query logs are readily available through online Web services, so our system can provide business cycle information in a timely manner.
The challenge, of course, is matching in-sample methodologies with out-of-sample results. Studying business cycles in search of superior forecasting techniques is littered with systems that worked well in the past but stumbled when applied afterward in real time with data as it's released.
The true test for Internet searches as an early warning indicator, in short, starts… now. The results won't be known for several months (or years, if the U.S. economy avoids a recession). Meantime, the sky's the limit for the (in-sample) possibilities.