The convergence of the proliferation of “online labor platforms” and the lingering employment crisis has catalyzed attention to “online labor platforms,” as indicated by this NY Times article “The Boom in Online Freelance Workers.” Innovation and disruption go hand-in-hand–certainly where online platforms are emerging in the domain of work.
A March 2012 research paper by Ajay Agrawal, et al, “How Do Online Platforms Flatten Markets for Contract Labor?” develops in-depth analysis of one potential area of disruptive effects: namely the flatten of contract labor rates across advanced and emerging economies. Here is the abstract:
Despite the internet’s ability to “flatten” markets, we find evidence of a significant penalty for job applicants from less developed countries (LDCs) on a major online platform for contract labor; developed country (DC) employers are less likely to hire LDC applicants, even after controlling for many observable characteristics including education, experience, and posted wage. However,platform-specific experience, which increases the likelihood of success for all applicants, has a disproportionately large benefit for LDC applicants. We attribute this DC employer behavior to the standardized and easily verifiable attributes of this platform-specific information, which disproportionately benefits applicants from LDCs whose off-platform experience and education is costlier for DC employers to evaluate. Our interpretation is consistent with another finding that LDC employers, whom we posit are more easily able to evaluate off-platform LDC experience, neither penalize LDC applicants nor disproportionately reward their platform-specific experience. Furthermore, we find evidence that DC employers with more experience on the platform are significantly more responsive to platform-specific experience, implying that even this information is costly to interpret since it requires investing in experience on the platform.
The paper was published electronically on the internet earlier this year and is available here: OLLaborPlat.