For several decades, US policy in telecommunications and electronic mass media has focused on deregulation and the encouragement of competition. Of these actions, none was more important than the Telecommunications Act of 1996. What has been the impact of these initiatives? And what are the implications for deregulation and antitrust policies in the future?

To provide an empirical answer, I have looked at the market concentration trends in the US information sector, over 20 years and across 72 separate industries including long-distance telecommunications, broadcast TV, newspapers and internet service providers. For the purposes of this analysis, I then grouped these media industries into two categories, those that are not regulated, such as newspapers, and those that are either regulated (such as local phone services) or closely affected by regulation (fibre optic cables, for example).

The results were obtained using the Hirschmann-Herfindahl Index, which is calculated by adding up the squared market shares of all the companies in an industry. For example, if company A accounts for 40 per cent of an industry, and companies B and C for 20 per cent each, with the remainder fragmented among numerous tiny firms, then the “HHI” is 1,600+400+400=2,400 - a concentrated industry. In contrast, if there are ten firms, each with 10 per cent of the market, then the HHI is 1,000, indicating an unconcentrated industry.

The results show that the concentration level in unregulated information industries - the lower line - is fairly low, and well inside the government’s threshold of an HHI of 1,000 for unconcentrated industries. This level has increased only very gently over time. In regulated industries, on the other hand, concentration is high, now over 2,000, well above the government’s threshold of 1800 for highly concentrated industries. This concentration declined in the 1980s and early 1990s but rose pronouncedly after 1996, the year of the landmark Telcommunications Act. (On a positive note, it is still lower now than it was right after the massive AT&T divestiture in 1984, which split up the company into eight pieces.)

A new way to communicate
Michael Powell, chairman of the US Federal Communications Commission, on how to lay the foundations for a lasting recovery in telecoms. Read

Why is this so? In general, regulated industries are concentrated - which is the reason, after all, why the government intervenes in the first place to protect the public from the negative results of market power. But the data also show that not only is concentration high for regulated information industries, it has increased since the deregulatory 1996 Telecommunications Act. This was certainly not the intended effect. But the Act, by encouraging entry that had been prevented before, has created pressures on companies to merge in order to re-establish stability in their markets. Given the economic characteristics of their markets - high fixed costs, low incremental costs - competition leads to prices that are too low to cover total costs. The result is chronic instability. The business response is to re-establish the market power that allows for higher and more stable prices.

Where does this leave antitrust and regulation policy in the future? The conventional view is that antitrust enforcement is a substitute for regulation. With regulation now being phased out, antitrust would thus become more important. But this does not seem likely. To enforce a competitive market structure through aggressive governmental antitrust policy would leave the industry unstable.

	New economy policy forum
	Archive of recent articles and explanation of how the forum works Go there

Could private antitrust cases take the role of governmental ones? Such private actions by competitors and their customers have recently proliferated in the US. As a matter of public policy, such lawsuits are usually a bad idea. In telecoms, there are already ample safeguards in place: the FCC, state public utility commissions, courts and consumer protection agencies, plus federal and state legislatures. Adding another level of redress that is likely to be driven by class-action lawyers in search of treble damages will add little to achieving justice and will rather contribute to delay and uncertainty for many participants in the network environment. In the end, courts will recognise this and defer to the specialised administrative agencies to resolve these conflicts.

Thus, neither public nor private antitrust actions are likely play a central role in telecoms. Does this mean letting market power dominate? Not really. It simply means that some direct regulation will remain. Antitrust enforcement will not prove to be a substitute for direct regulation. Nor will deregulation prevail, because the industry will find its equilibrium in some form of oligopoly rather than competition. And with this, inevitably, comes some regulation of the negative effects of oligopoly, primarily at the consumers’ level. This will reverse the trend of deregulation which has been policy gospel for a long time. Thus, the bursting of the dotcom and of telecoms bubbles will be followed by the puncturing of the deregulation bubble, too.

The writer is professor of economics and finance at Columbia University and director of its Columbia Institute for Tele-Information

If you feel you can contribute to a debate on this subject, e-mail techforum@ft.com