Network Neutrality and Bank of America's charges

There has been much discussion over the last few days about Bank of America’s decision to charge consumers for debit card transactions. It’s an odd fee in that you pay $5 for the first transaction each month and then the rest of the month is free. While this is related to the Durbin reforms that reduced the fee that banks could get from merchants for processing these transactions (which is per transaction value), it illustrates some interesting issues with respect to two-sided markets that have received some discussion on this blog. By the way, in Australia, a controversy emerged this week when banks decided to raise debit card fees to merchants. In both cases, consumer groups were outraged because consumers would eventually pay. This just illustrates something we must always be careful about in two-sided markets — just because a price to a group rises does not mean that they end up actually paying.

The controversy over bank fees is related to a ‘more digital’ issue — net neutrality. That issue involves the possibility that ISPs could charge different content providers different fees. So if Verizon noticed that its consumers used Netflix alot, they could charge Netflix more to access their customers. The claim is that some content providers impose more costs on ISPs and so should pay more to cover common infrastructure costs or manage congestion. Of course, if Netflix was charged more, it would pass those costs onto customers in the form of higher fees or drop off the ISP in question. While a Netflix may be able to use competitive leverage to keep those charges lower, smaller content providers may not be so lucky.

My co-author, Stephen King, suggested a way out of this dilemma for debit card transactions: set the fee to merchants at zero and require consumers to be charged explicitly at the point of sale — just like they are for ATM transactions. The idea is that consumers might hold several debit cards and so if their bank was more expensive, they could switch right there at the counter. The point is that the consumer is causing the costs on the system when they choose how to pay and so should see that. And if enterprising merchants want to absorb that fee, they can do that too; just as a consumer’s bank can do for ATM transactions. The point is that no bank can impose a fee on merchants.

What if this happened for ISPs?  In this situation, Netflix and other smaller content providers would be off the hook for payments. In the King-version of that, no content provider would pay for any transit. Instead, consumers that pulled content off the web would be responsible (through their ISPs) for the chain of payments that result. To be sure, Netflix would be more expensive for consumers because of the bandwidth consumed and so Netflix will have to reduce its prices to compensate but if the consumer’s ISP charges too much for that, the consumer will be induced to switch — not because of some tussle with Netflix or some exclusive content deal — but because infrastructure costs are too high. When it comes to the intensity of use, it is consumers who are causing the costs and so it makes sense for them to pay. It also gives them a big opportunity to fruitfully apply competitive leverage.

That said, there are caveats to all this. If there is a single ISP in areas, this may lead to very poor outcomes but then again, things aren’t great for consumers there and at least that ISP will be unable to leverage the other side of the market. And then there is the issue of uploads as opposed to downloads. However, that doesn’t stop us from going through the thought experiment and seeing what the world where one price is eliminated looks like. That said, I don’t expect any ‘thank you’ cards from consumer groups for floating this suggestion.