By and large, consumers have benefited handsomely from airline deregulation. The Brookings Institution estimates we are saving some $12 billion a year in lower fares.

But not everybody is pleased with deregulation. Flight delays at major airports, and growing concerns about an obsolete air traffic control system, are the cause of many a complaint.The underlying problem is that the infrastructure on which airlines depend - airways and airports - is not keeping pace with the growth in air traffic.

Most recently attention has been focused on the problem of "fortress hubs." At certain major airports, a single airline has ended up with the vast majority of gates and service. Studies show that at places like Pittsburgh, where US Air dominates, and St. Louis, dominated by TWA, originating flights cost up to 50 percent more than standard industry fare levels.

In other words, even though aviation as a whole is much more competitive than before deregulation, localized monopolies are emerging at these fortress hubs.

The simplistic answer to this problem is that Congress must spend more of the funds in the Aviation Trust Fund to increase airport and airway capacity.

But this conventional wisdom misses the point in two important ways. First, it's unlikely that Congress or the administration will agree to significant spending increases because of the federal deficit. Second, simply sending more funds through existing bureaucratic channels is not going to give us the capacity increases we need.

Government bureaucracies are simply not equipped to run fast-moving high-tech enterprises effectively or efficiently. The Federal Aviation Administration has always lagged far behind the state-of-the-art, even when it had billions to spend.

Airports are essentially businesses - potentially very profitable ones. But they are not free to be run like businesses as long as they must operate as wards of the state.

The FAA owns the control tower and employs the controllers; if Chicago's O'Hare has a shortage of controllers, the airport cannot go out and hire more. Nor can an airport raise its prices to finance new runways and taxiways.

We won't solve the airport capacity shortage until we turn the major airports into real businesses. Congress should permit airports to opt out of the federal grant program - in return for which, the airport would take over its control tower and landing aids from the FAA and have the right to charge market prices for use of the airport.

That would give O'Hare, La Guardia, Los Angeles, Dallas/Fort Worth and the others the ability to add controllers, radars, and other high-tech gear that would increase their hourly capacity even without building more runways.

And that would solve the "fortress hub" problem. Today, airports' major source of funds for new gates is municipal bonds.

But, incumbent airlines - like US Air in Pittsburgh - usually have a veto power over new bond offerings. If airports could charge per-head passenger fees and market prices for landings, they would have alternative sources of funds, free of airline veto power. That would let them add the new gates that would let new airlines begin service, undercutting the monopoly prices now being charged by fortress-hub airlines.

The best way to do this is to make the airports private, as Margaret Thatcher has done in Britain. By selling its airport, a city would gain a major billion-dollar windfall. And the real estate would be added to local property tax rolls, producing hefty annual income on into the future.

The feds would gain, too, from corporate income taxes paid by the for-profit airport companies.

But, the big winners would be air travelers. Making air traffic control and airports private would give us the increased capacity and modern equipment that's needed to keep pace with the growth of a competitive air travel business. That's the best way to be sure we'll keep on saving $12 billion a year in lower air fares.