Social media enables promising new approaches to measuring economic activity and analyzing economic behavior at high frequency and in real time using information independent from standard survey and administrative sources. This paper uses data from Twitter to create indexes of job loss, job search, and job posting. Signals are derived by counting job-related phrases in Tweets such as “lost my job.” The social media indexes are constructed from the principal components of these signals.
The University of Michigan Social Media Job Loss Index tracks initial claims for unemployment insurance at medium and high frequencies and predicts 15 to 20 percent of the variance of the prediction error of the consensus forecast for initial claims. The social media indexes provide real-time indicators of events such as Hurricane Sandy and the 2013 government shutdown. Comparing the job loss index with the search and posting indexes indicates that the Beveridge Curve has been shifting inward since 2011.
This paper develops new measures of flows in the labor market using social media data. Specifically, we use Twitter data to produce and analyze new weekly estimates of job flows from July 2011 to early November 2013. We present methods for validating such novel economic measures and articulate principles for assessing the usefulness of time series derived from social media (Section I). We do this first by comparing our estimates with official data. Our Twitter derived job loss index tracks initial claims for unemployment insurance (UI) and carries incremental information relative to both lagged UI data and the consensus forecast (Section II). We also propose social media indexes to measure concepts with weaker analogues in official statistics—job search and job posting—and then use these measures to study shifts in the relationship between posting and job loss (Section III).
Social media provide an enormous amount of information that can be tapped to create measures that potentially serve as both substitutes and complements to traditional sources of data from surveys and administrative records. The use of social media to construct economic indicators has a number of potential benefits. First, social media data are available in real time and at very high frequency. Such timely and high-frequency data may be useful to policymakers and market participants who often need to make decisions prior to the availability of official indicators. The fine time-series resolution may be particularly helpful in identifying turning points in economic activity.
Second, social media data are potentially a low-cost source of valuable information, in contrast to traditional surveys that are costly for both the respondent and the organization collecting the data. Third, social media offer a distinctive window into economic activity. They represent naturally-occurring personal communication among individuals about events in their everyday lives without reference to any particular economic concept. Like administrative data, but unlike surveys, social media challenge economists to map the observed information into the economic concept being measured. Fourth, social media can be used to answer questions we would have liked to ask in surveys had we known about events in advance. In ordinary survey design, we frame the questions and then collect the data. Social media allows us to reverse this order and generate ex post “surveys.” For example, we use the indexes to examine the impact of two shocks to the labor market, Hurricane Sandy in October 2012 and the October 2013 government shutdown.
This paper implements social media indexes for job flows. Why do we focus on job flows? Substantively, job flows are of central interest to economists, market participants, and policymakers. Practically, the weekly frequency of the official UI claims data makes them a good benchmark for testing the performance of our social media measures.
See full article on Using Social Media to Measure Labor Market Flows in PDF format here.
Via personal.umich
