Skip to main content


Online “beauty contest” auctions for IT freelancers may favor hiring locally over offshoring

Hema Yoganarasimhan

Hema Yoganarasimhan

The online auction has become a favored mechanism of transaction in the emerging “gig economy” for IT services.

It virtually erases geographic borders from this new marketplace for temporary skilled labor. Professional matchmaking sites such as Upwork, Freelancer and  Guru allow a small business in Iowa, say, to hire a developer in Bangladesh, a programmer in Bolivia, a designer in Kenya, an editor in Malaysia.

So, is this global online labor market just another vector for the outsourcing of American jobs to cheaper overseas markets?

Not necessarily, according to new research by Hema Yoganarasimhan, an assistant professor of marketing at the University of Washington Foster School of Business.

Yoganarasimhan’s sophisticated economic modeling of online labor auctions—which factor a combination of price, reputation and speed of delivery—indicates that the most efficient and profitable form of exchange is one that promotes more localized hiring.

“These auction-based transactions might actually favor workers closer to home,” she says. “Freelancers in developed countries tend to be more skilled and therefore able to charge a premium that businesses are willing to pay for the quality and convenience. This translates to higher commissions for the freelancing sites.”

Brave new labor market

Online labor markets have flourished since the financial crisis of 2007-08 triggered the Great Recession and a spike in unemployment worldwide. The aftermath—workers without jobs and firms without workers—has given rise to this new labor market.

This era of just-in-time staffing suits leaner organizations. And some skilled programmers, developers, designers, writers and multimedia producers prefer the flexibility of temporary gigs, or simply have given up on finding full-time positions altogether.

Either way, the new “workforce in the cloud,” as the Economist calls it, is booming. The consulting firm Staffing Industry Analysts forecasts that the online labor market will reach $5 billion by 2018.

Naturally, a number of businesses have emerged to broker this market. The largest online freelance matchmaker, Upwork, counts 12 million registered freelancers and five million clients. Its three million jobs posted annually are worth a total of $1 billion.

That’s big money potentially being leeched from the American labor force, as firms are tempted to hire low-cost IT personnel from developing countries rather than from the more expensive talent pool at home.

This would probably be inevitable—if cost were the only concern.

Beauty contests

But it is not. Online labor auctions operate like a “beauty contest,” in which judges make their determinations based on multiple criteria that are unknown to contestants (who try to optimize the winning combination of beauty, presentation, talent and poise).

This kind of black-box calculation is used in many procurement decisions, such as the awarding of military contracts, welfare-to-work projects and cellular spectrum licensing, as well as online ad sales, real estate transactions, and even decisions about who to hire, date and marry.

“We see this in many kinds of procurements for differentiated products or services where price is not the only consideration,” Yoganarasimhan says.

In the arena of IT project work, hiring decisions come down to an unspecified combination of price, reputation, speed of delivery and other factors. And it’s up to freelance tech workers to optimize the combination if they want to win the job.

Why local?

Yoganarasimhan found the temporary IT labor market a particularly interesting and accessible application of the beauty contest auction to study. “No one had looked at what’s the optimal way to run this kind of an economy,” she says.

By examining the bidding strategies of freelancers and the decision strategies of hiring firms in 4,000 labor auctions over a two-year period, she was able to establish patterns and model the most efficient and profitable way to run such a labor exchange.

Auction sites take a cut of every transaction. So while buyers and sellers deal with their own decisions, their brokers look to optimize the mix of transaction volume and commission percentage.

Yoganarasimhan’s model shows that lower-price bids from freelancers in developing regions such as India and Southeast Asia generate more transactions but lesser commissions. On balance, fewer transactions from higher-priced workers churn greater profits for sites like Upwork, Freelancer and Guru.

She recommends that these brokers focus on growing their buyers side and attracting more local sellers from developed countries that can satisfy hiring firms with quality and speed that compensates for higher prices.

Market forces, Yoganarasimhan believes, may turn the tide of offshoring, at least in the market for freelance IT services.

“People push back against these auction services because they are promoting offshoring and bad for the US economy,” she says. “But we find that these markets actually thrive on local workers, who live in the same developed countries as hiring firms.”

Estimation of Beauty Contest Auctions” is published in the August 2016 issue of Marketing Science.