New Research: Social Networks to Labor Networks
Posted: May 3, 2013 at 5:02 am, Last Updated: May 6, 2013 at 6:45 am
Social networks, best known for the antics of Facebook and Twitter, show a serious side when similar connections are used to look at where people work to give insight into economic complexities, according to new research by a George Mason University student.
Omar Guerrero took a cue from social networks and epidemiology (the study of patterns, causes and effects in health and disease) and dived into the labor pool to study how workers move between firms, move up within companies or are handed pink slips.
“We thought maybe we could apply some of the techniques from social networks to labor networks,” says Guerrero, a doctoral student in the Department of Computational Social Science of the Krasnow Institute for Advanced Study at George Mason.
Workers moving from one firm to another create links between companies. Lessons culled from these “labor flow networks” could guide economic and employment policy, Guerrero says.
A few years ago this type of detailed data didn’t exist, nor did the technology to examine it. A native of Mexico City, Guerrero turned to Finland to find a complete economy drawn in numbers. The Finns register with the government whenever they change jobs, move or experience other such life-impacting events.
By studying the “labor flow networks” for the entire economy of Finland, Guerrero has teased out lessons that can be applied to economic policy and employment.
It’s a novel way of looking at an old problem, says Rob Axtell, a Krasnow professor and chair of the Department of Computational Social Science. “We have a sense that we can bring a new perspective to this area,” he says. “It’s a little like there’s a virgin forest over there and we’re going to explore it.”
Current economic models don’t mimic crisis well, Axtell says. When entire industries are hammered, such as the automotive or financial sectors during the recent recession, the normal flow of workers is blocked.
That means the conventional view of “here’s a pool of unemployed and here’s a pool of vacancies” doesn’t show what’s actually happening in the economy, Axtell says. “In the pool picture you don’t care if it’s the financial sector in New York or the auto sector in Detroit, you just say ‘here’s a pool of unemployed.’ But Omar’s network picture gives you a completely different mental model for how unemployment works.”
The labor flow network approach also details who’s hiring.
“Not all firms are created the same,” Guerrero says. “Some grow very fast. Some are big and grow very slowly. We think that with networks we can show this. With networks we can capture the sizes, the growth experiences. We can also capture the way labor goes through them and try to associate this labor flow with the experience a firm has regarding growth and its size.”
This dynamic could be helpful to policy makers who want to kick-start employment by targeting companies with high-growth potential, says Guerrero. About 10 percent of firms account for three-quarters of a country’s employment growth, he adds.
Guerrero discovered several indicators in his research that point to a company hiring more workers in the future. For example, neighborhood connectivity is a good sign and applies to small or large firms. “How important am I in my neighborhood?” Guerrero says.
Another factor is closeness — that’s how close firms are to each other in their field. “If I am working in a firm and I want to get to another firm, on average, how many other firms would I have to work at before getting there?” Guerrero describes.
Many industries have mainstays that people pass through in their career. In the financial sector, it’s the big New York-based financial houses such as Goldman Sachs & Co. or the now defunct Lehman Brothers.
Guerrero is creating computer models that mimic a real economy by using what he’s learned in analyzing the Finnish data. It’s a high-fidelity representation of an economy that has promise well beyond Finland, Axtell says. As other countries begin to release dense amounts of economic data, the labor flow networks likely follow the same pattern, he says.
Policy makers can use Guerrero’s computer models to show how a tax cut or other incentive could impact businesses and employment, Axtell says. It could help economies worldwide navigate complexities and avoid bubbles.
“It’s very hard to navigate bubbles,” Axtell says. “People pay too much at the peak and then they’re underwater later on. It’s hard to live in a world where processes are too variable. Omar’s model in the long run could be relevant for a wide class of economics, which could lead to better functioning social systems and better functioning economies. That’s the long-term goal.”
Guerrero came to George Mason after reading Axtell’s pioneering work while working on his master’s degree in computational economics at the University of Essex (U.K.). Guerrero is headed to Oxford University in the fall for his postdoctoral work.
Write to Michele McDonald at firstname.lastname@example.org