I was dismayed to learn that AT&T is trying to buy T-Mobile for a whopping $39 billion.
AT&T can use the extra towers to improve reception in very crowded metropolitan areas, but the decrease in competition and likely resulting increase in price is a big problem.
People who sell a product charge what the market will bear, but if the market isn’t fully competitive—if customers have few options to take their money elsewhere—then customers can’t punish high prices or poor service, and providers charge more for less.
The wireless market is already not competitive for two important reasons. First, providers lock in customers with a combination of contract law and technology. They claim contracts and handset locks are necessary to recoup the costs of subsidized handsets, but why don’t they all charge less for month-to-month service on unsubsidized handsets? (T-Mobile is still alone in offering such a discount.)
Second, the industry is already an oligopoly, with so few major competitors that they already have the power some power to charge inflated prices. The standard measure of an industry’s competitiveness is the Herfindahl–Hirschman Index, or HHI.
To calculate an HHI, you take the square of the percentage of each firm’s market share. A firm with 20% share adds 400 points (20 x 20) to the HHI. According to Department of Justice antitrust guidelines (which, unfortunately, the DoJ and FTC have stopped following), if the HHI is over 1,000, the market is moderately concentrated—that is, not fully competitive. If the HHI is over 1,800, the market is highly concentrated and thus non-competitive. If a market is already over 1,000, then any merger raising the HHI by 100 points or more is presumptively a problem for competition.
To see how bad things are already, and how much worse they would be after the proposed merger, we should calculate the HHI for the wireless industry, both before and after. First, here are the ComScore market shares for each carrier as of March 2010:
Table 1: Market Concentration in the Wireless Industry, March 2010
|Carrier||Share, %||Share Percentage, Squared|
This is what a noncompetitive oligopoly market looks like. We already see this in a lot of important ways—suboptimal cell service, attrocious customer service, stubbornly high prices, and charges that are often exponentially larger than the marginal cost.
The prices for text messaging in particular are a great example of “price gouging” and illustrate the industry’s tacit collusion (pdf). The cost for the network provider of handling a text message is virtually zero, since the messages are small enough to fit into the “control channel,” or the tiny bit of data that your phone and cell network are exchanging even when you’re not talking or using mobile data.
In a truly competitive wireless market, a customer would drop a provider who charges up to $20/month for something that’s actually nearly free to provide. Imagine if McDonalds sold hamburgers at their current prices but charged $0.20 for each french fry—or $20 for all the fries you can eat. Potatoes are cheap, so we’d be offended and take our money elsewhere, because the fast food market is highly competitive.
In mobile telephony, however, there almost is no “elsewhere” to take our money, especially if you need reliable nationwide coverage. The number of players is small enough, and customers are locked in enough, that there is little opportunity to punish this price gouging. (Thankfully, free messaging-over-data via services such as Google Voice allow customers some opportunity for arbitrage, but expensive data plans and technological know-how limit this opportunity to to the most economically and technologically well-positioned customers.)
So the bad news of an uncompetitive market is already here. Now, let’s see what the market might look like after an AT&T/T-Mobile merger. Here’s that table, assuming that all T-Mobile customers stay with AT&T (and most will have to for some time, thanks to their two year contracts):
Table 2: Approximate Market Concentration Following AT&T/T-Mobile Merger
|Carrier||Share, %||Share Percentage, Squared|
|AT&T plus T-Mobile||37.2%||1384|
A substantial number of T-Mobile customers will switch to Verizon or Sprint, but the HHI would still be in the mid-2000’s, and no scenario makes this market more competitive than today’s market. In short, customers and regulators should be worried.
Now imagine what happens when it’s specifically T-Mobile that goes away. They have long been the cheapest option, offering the worst service among the big four in exchange for much cheaper prices. They’re the only company that has experimented with discounted pricing for month-to-month customers. Inexplicably, they’re still the only major US carrier to deploy UMA, which allows voice calling over wifi. (I’d love to use my Verizon minutes to make and receive calls over my home wifi router; instead, I’m forced to take the chance that I’ll drop yet another call in my first-floor apartment. Can you hear me now?)
T-Mobile offers several unique features in the otherwise troublesome wireless market, and AT&T is unlikely to keep many if any of them. Ma Bell just wants the customers, towers, and spectrum. If they wanted to sport UMA or cheaper pricing, they could have offered them years ago.
The current cell market is already highly concentrated, so we get service that is overpriced, with limited features and a quality of service that does not justify what we pay. If federal regulators allow AT&T to buy T-Mobile—which, unfortunately, is practically a given—the market will be even less competitive.
This merger means less choice and still-higher prices for something like the service we’ve long since been promised. If you have a lot of stock in the telecom industry, however, it’s a big win.