Siddharth “Sid” Suri works at the intersection of computer science and behavioral economics. His work analyzes the relationship between social network topology and behavior using a variety of techniques including behavioral experiments, massive data analysis and theoretical modeling. Moreover, Sid has become one of the leaders in designing, building, and conducting “virtual lab” experiments using Amazon’s Mechanical Turk. His work has appeared in Science, PNAS, as well as top computer science venues. He won the Best paper award and a Top 10% paper award in ACM EC 2012.
Sid earned his Ph.D. in computer and information science from the University of Pennsylvania in 2007 under the supervision of Michael Kearns. After that he was a postdoctoral associate working with Jon Kleinberg in the computer science department at Cornell University. Then he moved to the Human & Social Dynamics group at Yahoo! Research led by Duncan Watts. Currently, Sid is one of the founding members of Microsoft Research, New York City.
Monopsony Online: Crowdworking and Market Power
We analyze crowdsourcing as a labor market through the example of Amazon Mechanical Turk (AMT), a popular, commercial site that allows anyone to post and complete small, paid tasks online. We consider how power dynamics between requesters (“employers”) and crowd workers (“employees”) set the terms for and expectations of employment. In theory, crowdsourcing could circulate work fairly and directly to individuals seeking microtasks. However, as practiced, commercial crowdsourcing services, like AMT, 1) systematically occlude the information workers need to choose appropriate employment opportunities and 2) implicitly make individuals bear the high costs of finding viable tasks to do. We frame the AMT labor market in terms of monopsony to diagnose this dynamic. Monopsony typically describes a situation where an employer has a greater degree of wage-setting power because of the limited employment opportunities available to a pool of workers. For this reason, evaluating monopsony online has important implications for how we think about digital work.
Our project therefore draws on ethnographic research and quantitative analysis of survey data to argue that market frictions give rise to the inequitable distribution of power among requesters and crowdworkers. We hypothesize market distortions on AMT are a result of 1) inadequate information about what we call the “goodness of tasks”; 2) high search costs imposed on workers; and, 2) reputation bias, which makes market entry prohibitive to new entrants. We conclude with insights from crowdworkers about how to reform online labor platforms to serve the needs and interests of all people dedicating their time and energy to crowdwork.