Daniel O’Connor (Vice President of Public Policy at the Computer & Communications Industry Association (CCIA))*
Not surprisingly, as the Internet matures, and online platforms play an increasingly important role in the world’s economy, the question of how to regulate these relatively new online platforms has received significant attention. As regulators and competition policy experts turn their attention to online platforms, it is important for them to take into consideration how these platforms function, the considerations they must take into account, and how they differ from the traditional companies and markets from Econ 101. (See CCIA and CPI’s ebook for a more in depth analysis of online platforms and regulatory policy)
Unlike traditional markets, where companies buy and sell products or services, platforms often function as matchmakers that connect different groups of producers and consumers. Although this phenomenon is not unique to the online world, as shopping malls, stock exchanges and newspapers also function as platforms, platforms are particularly common on the Internet. Their business viability often comes from solving traditional frictions in the economy, thus making it easier for third parties to engage in mutually beneficial interactions. In facilitating these interactions, the platform operators have to balance interests of all platform users in order for their platforms to grow. This can create political friction, as the different constituencies on a platform often have conflicting interests in how the costs and benefits of the platform are distributed. Parker, Van Alstyne, and Choudary sum up this dynamic in their book, Platform Revolution:
“Multi-sided platforms involve interests that don’t always align. This makes it difficult for platform managers to ensure that various participants create value for one another, and make it likely that conflicts will emerge that governance rules must resolve....This is a juggling act that even giants and geniuses get wrong.”
Instead of being traditional markets subject to a standard supply and demand function, these platforms bring together different groups of interdependent users. For example, Uber is not a company with a one-way relationship with riders or drivers, instead it provides the platform that connects suppliers of trips to potential customers and provides tools that allow for transparent pricing, easy payments, and clear ratings. Instead of higher prices indicating increased market power for Uber, they instead indicate a mismatch of supply and demand and the need to entice more drivers onto the platform at a given moment in a particular location. And, as illustrated by the Uber example, the different constituencies served by the platform have different interests. Riders want cheaper rides and strict monitoring of the drivers to limit the chances they get into a car with a bad or unsafe driver. Drivers, want higher prices for rides, and often chafe at the strict rules and performance targets that Uber requires for them to continue offering rides on the platform. Unsurprisingly, riders and drivers have both legally and politically challenged Uber for how the costs and benefits of the Uber platform are distributed: with riders suing the company over its surge pricing and drivers complaining that the performance rules are too stringent. Like other gripes about online platforms, these complaints tend to focus on the effects one constituency, not on all users of the platform. If riders get their way, the price of Uber rides will fall, benefiting them, but harming drivers, some of whom are already complaining about the prices being too low. If Uber rules skewed more towards drivers, then riders would be harmed by higher prices and lower quality drivers on the platform.
Regulators need to be wary as well. The political economy of platforms means that the temptation exists for constituencies on the platform to politically organize and shift costs to other groups of the platform’s users, helping one group but doing more total harm than good. This is not a theoretical concern, as it is exactly what happened with Australia’s intervention in the credit card interchange fees (aka operators of a payment platform) responding to the complaints of merchants.
It is the job of a platform operator to balance these interests to keep all users of the platform happy. If prices are too low, or rules on drivers too onerous, drivers can easily flee to a competing platform like Lyft. If prices are too high, or drivers of too low quality, users can do the same. Platform operators have to carefully balance interests of all users of the platform. An open source mobile platform, such as Android, has to balance the interests of smartphone users, application developers, and mobile device manufacturers. Ebay has to the balance the interest of buyers and sellers. Facebook has to balance the interests of users with multiple types of preferences and privacy, security and decorum with that of the advertisers that are thinking of associating their brands with the platform.
How a platform balances interests between different groups of platform users is not easy, and is usually the key determinant of whether a platform succeeds or fails. Facebook thrived because it learned lessons in social network management from Friendster and MySpace, which failed to balance interests in a way that pleased enough of their users and attract mainstream advertisers. Although some users might want more freedom to post provocative content or remain anonymous, Facebook chose to have strict rules governing the nature and content of participation in its social network partly because it saw how MySpace’s anonymity and lack of moderation of content drove many users away and turned off the bulk of corporate advertisers.
Furthermore, the supply and demand functions of platform users are interlinked, which confound traditional economic assumptions, as David Evans and Richard Schmalensee have noted:
“The old formulas—including the ones we have taught generations of undergraduates in Econ 1—do not give the right answers for multi-sided platforms. The math is simply wrong. Traditional economics holds, for example, that it’s never profitable to sell products at less than cost. The new multi-sided economics shows that even paying some customers rather than charging them anything can be profitable in theory.”
Different groups of users on a platform have different preferences and intensities of preferences. Traditional economics teaches that pricing below marginal cost is not possible for an extended period of time, and traditional antitrust enforcement has viewed such activity as potentially an indication of predatory pricing designed to drive competitors from the market. However, with platforms, it often makes sense to subsidize one side of the market (usually the side with the most elastic demand curve) in order to produce the greatest overall utility. Search engines offer free searching to users, and make money by charging advertisers. Newspapers often make content available to readers very cheaply (and sometimes for free) in order to increase the appeal of the platform to advertisers. Online platforms that connect buyers and sellers, such as Ebay or Etsy, usually allow shoppers to peruse their websites for free and make money on commissions from the seller when a product or service is sold on their platform.
Although the price structures and rules implemented by online platforms should not be exempt from regulatory oversight, it is important for regulators to take a wide view of the all constituencies of platform users and weigh the costs and benefits of rules and price structures across the entirety of the platform. Regulators should not lose sight of the likely ripple effects on other users of the platform, if they contemplate actions aimed at addressing the complaints of one constituency. Given that operating a platform usually requires a precise and tricky balancing act to keep all users happy enough, even slight changes to that equation can throw off the platform’s equilibrium. Indeed, platform markets are complex and should be approached with care.
* Daniel O’Connor is the Vice President of Public Policy at the Computer & Communications Industry Association (CCIA), where he works with government and industry leaders on competition, intellectual property, international trade and global Internet policy. He leads CCIA’s efforts on antitrust policy and covers business and technological developments in high-tech markets. O’Connor has published many articles discussing the single digital market, Android competition and platform regulation.