Exciting Research Results from the First-Ever Economic Graph Challenge
March 9, 2017
At LinkedIn, our mission is to connect talent to opportunity at a global scale. One way we realize it is by analyzing the data within LinkedIn to better understand the global workforce and provide more granular insights into the economy. In turn, this allows us to more accurately recommend to our members what skills, connections, and more could be most useful to them in the future. Much of this work is represented by our Economic Graph, a digital map of the global economy identifying connections between people, skills, and companies.
We conceived the Economic Graph Challenge (EGC) in 2014 as an opportunity for external researchers to dig into our highly-structured dataset to unveil insights that would further the goal of creating economic opportunity. Out of hundreds of submissions, we ultimately selected 11 finalists and their proposals. Those teams have spent months working with LinkedIn to uncover valuable insights. As we now wrap up the first edition of the Challenge, we’d like to highlight a few of the most interesting findings from this research.
Gender bias in self-promotion by MBAs
One team analyzed whether male and female graduates of top-ten MBA programs from 2011-2016 promoted themselves equally on their LinkedIn profiles. The team discovered that women and men were comparable in the number of skills and honors they included on their profiles, but that women were less likely than men to include job descriptions or summaries. We don’t know why women in top MBA programs are less inclined to include these more descriptive parts of their LinkedIn profiles, but this could be a fertile avenue for future research.
This team dug into the costs and benefits of restricting worker mobility using non-compete clauses. Such clauses can be a double-edged sword: while they potentially limit workers’ ability to seek out better employee-job matches in the short term, they may also encourage capital investment at existing firms by helping firms retain talent. Using data on employment history and court decisions on the enforceability of non-competes, the researchers concluded that both of these effects indeed occurred. Non-competes simultaneously reduce worker departures to entrepreneurship opportunities (and thereby reduce the creation of new firms), but do also lead employers to invest more.
Collaborations with Indiana University and the MIT
Additionally, two EGC teams unveiled such promising research that we decided to form longer partnerships with their institutions: Indiana University and MIT.
The team from Indiana University designed an algorithm that identifies fine-grained geo-industrial clusters called “microindustries” (e.g., electric vehicle manufacturers in northern California, or Milanese fashion houses) based on workers’ firm-to-firm transitions. Using Bloomberg S&P 500 data, the team showed that talent flows into and out of microindustries correlate with changes in market cap. This demonstrated that our dataset captures real-world economic phenomena at a level of detail and scale that rivals best-of-breed methodologies from agencies like the Bureau of Labor Statistics.
The MIT collaboration has a research agenda of measuring the value of human capital and the distribution of that value between firms, employees and consumers. In the first stage of that research, the team is using LinkedIn data as well financial data to determine the value of human capital embodied in firms. In future stages of the work, they plan to use LinkedIn salary data to measure how much of the value of human capital accrues to the workers themselves and test different theories of investment in human capital.
How you can get involved in Economic Graph research
Next week, we plan to announce a new program for even longer-term partnerships with select economists and researchers. This is an exciting opportunity for us at LinkedIn, and builds on our past work to share key insights about the global economy with our members, the cities they live in, and the world at large.
If you are a researcher, economist, or member of a non-profit that would like to propose a new Economic Graph research project, visit https://engineering.linkedin.com/data/economic-graph-research/ to learn about a special new program that we are launching on March 15.