Research Highlights Article
January 30, 2024
Expanding markets
What happens to incumbents when more sellers are added to online platforms?
A food delivery worker cycles in New York City.
Source: Julia Justo, CC BY-SA 4.0
From retail to travel to financing, online platforms now permeate almost every aspect of modern day economies. As of 2021, seven out of the top ten most valuable companies were platforms.
What does the expansion of these markets mean for sellers already operating on them?
In a paper in the American Economic Journal: Microeconomics, author Oren Reshef provides the first empirical evidence of what happens when an online two-sided market expands. He shows that expansions tend to favor incumbents, but only for the highest-quality sellers.
Reshef’s findings are based on analyzing comprehensive transactions data from the Yelp Transactions Platform (YTP)—an online marketplace for takeout and delivery from local restaurants—over a two-year period.
Similar to other two-sided platforms, such as eBay or Booking.com, YTP mediates transactions between platform users and sellers and assembles information, such as ratings and reviews
I think it shows the value of having a larger platform. And in particular, when you make the platform larger, you're also making it better, on average.
Oren Reshef
In 2018, Yelp expanded their platform by signing a deal with Grubhub, which was the largest food delivery service in the United States at that time. As a result of the deal, all restaurants affiliated with Grubhub were added to YTP, causing a sudden, substantial jump in the number of restaurants on the platform. In just a few weeks, the number of restaurants on YTP went from less than 60,000 to more than 80,000 without directly adding new users.
In theory, this supply-led shock could have helped or hurt Yelp’s incumbent sellers. On the one hand, increased competition could put significant pressure on incumbents’ bottom lines. On the other hand, network effects could benefit incumbents by drawing more consumers onto the platform.
“When you increase the number of restaurants—the number of sellers—you get more buyers joining,” Reshef told the AEA in an interview. “If there's more things to find, more people will search.”
Since all cities were equally affected by the partnership, Reshef could tease out the net impact of the Grubhub expansion by comparing incumbents in markets across different US cities before and after the new partnership..
He found that the number of weekly users in markets experiencing new entries increased over 36 percent compared to markets that remained unaffected by the partnership. And, on average, incumbents in treated markets saw increases in weekly revenue of up to 4.5 percent compared to incumbents in untreated markets.
“Overall, I found that the average effect is positive but weak; incumbents were doing a little bit better, but not by very much,” Reshef said. “The pie was getting larger, but their slice of the pie—their market share—was decreasing.”
However, these overall averages hid considerable diversity.
The revenue of the most highly rated businesses in markets that were affected by the partnership increased by up to 15.8 percent compared to high-quality businesses in unaffected markets. The growth in revenue was driven primarily by their ability to attract new customers and retain existing ones. In contrast, low-quality incumbents actually experienced a decrease in their revenue—in some cases, as much as 9.2 percent.
Weekly revenue
The chart below shows estimates of percentage changes of weekly business revenue before and after the Grubhub expansion for both high-rated and low-rated firms. The estimates come from comparing businesses in cities that experienced almost no change in the percentage of businesses available on YTP to those that experienced meaningful changes. The vertical bars represent 95 percent confidence intervals.
These results indicate that online expansions increased the return to higher ratings. It stands to reason that restaurants would look to invest in the quality of their wares in response to these changes, which is in fact what Reshef observed after the partnership. The average ratings of incumbents in treated cities rose by roughly 1 percent more than they did in unaffected cities, which correlates to a 3 percent growth in revenue.
Reshef’s findings show that the net effect of the countervailing forces behind market expansions depends crucially on the quality of incumbent businesses. As online platforms grow, understanding how their design influences markets will continue to be crucial to buyers and sellers who might otherwise be skeptical of the additional competition.
“I think it shows the value of having a larger platform,” Reshef said. “And in particular, when you make the platform larger, you're also making it better, on average.”
♦
“Smaller Slices of a Growing Pie: The Effects of Entry in Platform Markets” appears in the November 2023 issue of the American Economic Journal: Microeconomics.