Of all the terms presented in my recent glossary of airline-ese, two of the most familiar were "gate hold" and "holding pattern." Congestion, logjam, gridlock; if you haven't noticed, flying has become more delay-prone than ever. How this has come to be, and what can be done about it, aren't always understood. Step 1 is to lay blame where it's due. Say what you want about short-staffed airlines and an inefficient ATC (air traffic control) system, the baseline cause is better explained in simple logistics: There are too many planes carrying too few people.
Twenty years ago, here at my hometown airport, Boston's Logan, the typical rush-hour takeoff queue was a mix of wide-body jets and assorted smaller types. A typical 10-plane queue went something like this: DC-10, 727, DC-9, 767, L-1011, 757, A300, 727, 767, DC-10. That's 10 aircraft carrying about 1,800 seats. Nowadays, that 10-plane queue is a 20-plane queue, and it might look like this: CRJ-100, A319, MD-80, 757, 737, EMB-135, CRJ-100, Dash-8, A320, ERJ-170, EMB-145, A319, 767, 737, A321, BE-1900, MD-80, 757, CRJ-700, 737. For those who aren't hip to the various makes and models, those are mostly small planes. That's twice the number of aircraft carrying the same 1,800 seats.
Over the past quarter-century, remarkably low ticket prices have encouraged an ever-increasing number of fliers, to the point where twice as many people now travel by air than did in 1980. Meeting that demand, you're tempted to think, would have been an easy matter of increasing capacity: Instead of flying a 200-seat 767 from New York to Los Angeles, make it a 747 instead, with 450 seats. But that's not how it happened. Indeed capacity has grown, but the trend has been toward smaller planes, not bigger ones -- and more of them, departing more often to more cities. As the number of fliers has doubled, so has the number of planes carrying them.
The airlines believe that the best way to exploit markets is to offer as many flights as possible to as many destinations as possible. Technology has helped them do it. For instance, the advent of efficient long-haul twin jets brought about the fragmentation of transoceanic hubs and spokes, making it not only technically feasible but profitable to connect smaller markets with smaller planes -- at surprisingly small fares. Who would have imagined, in the days when transatlantic travel was dominated by the 747, that we'd soon be flying nonstop to places like Pisa and Venice, Italy, or Edinburgh, Scotland, in a plane half that size, for $299 each way?
At the other end of the scale, nimble and sophisticated regional craft like those made by Embraer and Bombardier have revolutionized short-haul flying. In 1980, going from New York to Miami, you stepped aboard a 275-seat L-1011 or, at the smallest, a 160-seat Boeing 727. Today, don't be shocked if you're riding in a 50-seat regional jet -- owned and operated not by the mainline carrier from whom you bought the ticket, but one of its code-share affiliates.
Correspondingly, along the major trunk routes, carriers are willing to offer more flights than ever. Instead of five departures a day between LaGuardia and Chicago, why not 10? If the demand isn't there to fill 10 big planes, farm out the difference to regional jets. Isn't that what travelers want -- the opportunity to choose from as many flights as possible? Surveys say yes, and frequency has become one of the holy grails of airline marketing. Frequency sells tickets.
Unfortunately, it doesn't always mean you'll get where you're going on time -- or even close to on time. It all goes out the window when flights don't actually leave or arrive when they're supposed to. Between some city pairs, more than 50 percent of flights are routinely tardy by 30 minutes or more.
Two weeks ago I was on a plane going from Washington Reagan to LaGuardia. It was a clear, warm, windless day. We pushed back shortly after 10 a.m., only to find ourselves in a holding bay at the far end of the airport, hit with a 90-minute takeoff delay. Ninety minutes' wait, for a flight that would barely last 40. When the controller was asked about the reason, he responded dryly: "The usual. Volume."
Volume. What he meant is that the Northeast corridor had become saturated -- now an almost daily occurrence, even in good weather. Throw in storms or low visibilities, and waits can extend for hours. Airlines will tell you this is an ATC issue. But is it? True, the en route sectors of airspace could be better utilized, for instance by taking greater advantage of GPS technology and reducing the horizontal distance limits between aircraft. But in the end, you can squeeze only as many arrivals into and out of a major airport as its runways will allow. The airspace issue ultimately becomes an airport issue.
Nationwide, volume delays account for only about 12 percent of late arrivals, but the northeastern United States, with its shoulder-to-shoulder chain of busy metropolitan areas, gets hit especially hard. The results can be staggering: A single afternoon thunderstorm might entail several thousand passengers' missing international connections out of New York's JFK or Newark, N.J., costing each of them a full day of travel.
"Carriers have made effective adjustments at many busy airports," says John Heimlich, chief economist for the Air Transport Association, the industry's most powerful trade group. "Look at Dallas and Atlanta, where schedules have been evenly spread to better absorb the number of flights. Unfortunately, this doesn't work at airports where market share is fragmented fairly evenly among five or six carriers. LaGuardia is the classic case. It's extremely hard for any one airline to consolidate and stay competitive at the same time. Unfortunately, what's good for that carrier might not be optimum for everybody taken together."
Airport authorities, the Federal Aviation Administration and some independent advocates like the Reason Foundation have considered a number of potential fixes, such as implementing peak-period landing fees and encouraging the growth of outlying airports in lieu of saturated hubs. The former is vociferously opposed by airlines, and even if implemented would have only a small impact. Cheap fares mean that carriers are able to pass fees along to consumers with minimal repercussions. And if demand falls along a certain route, or from a certain hub, because Southwest is siphoning customers to Islip and Manchester and Providence, will the legacy carriers respond by slashing flights? No, they'll just shift to smaller planes. It has already happened. At airports like Boston and LaGuardia, the propagation of regional jets has resulted in almost 50 percent of the takeoffs and landings accounting for a paltry 20 percent of passengers. (After decades of needless controversy, BOS has at long last opened a 5,000-foot relief runway to be used exclusively by regional planes.)
LaGuardia, JFK, Washington Reagan and Chicago O'Hare are unique among U.S. airports in that the number of arrivals and departures is regulated. Only so many planes are allowed to come and go each hour, with every flight filling a so-called slot. Obviously the system needs tweaking, either by reducing the allowable number of peak-period slots or by encouraging airlines to consolidate departures. "The problem will not be easily solved without direct government intervention," voices Perry Flint, editor of Air Transport World magazine. Flint points out that the FAA has issued a notice of proposed rulemaking to address congestion at LaGuardia. The regulation would penalize airlines that operate aircraft deemed too small by the agency, taking slots away from offending airlines. "To some degree, airlines are currently using RJs [regional jets] to hold onto slots that do not support large jets and that they otherwise would lose."
"You need to look at the big picture," adds Heimlich. "Seventy-five to 80 percent of all flights are on time. That's a very strong number."
That is true, but it neglects the reality that certain regions, and certain airports, are more delay-stricken than others. If you ask me, it would behoove any airline to scrutinize particular routes, and particular airports, and begin to make changes at those most affected. Let's try an informal poll. You're a frequent flier who travels often between Chicago and New York -- the country's heaviest domestic market. Instead of a choice of a dozen daily 737s or A320s, a third of which arrive an average of 30 minutes late, how would you feel about a choice of half a dozen 767s instead, with all of them arriving on time? Apply this line of thinking systemwide, and I suspect most ATC bottlenecks would be eliminated. (Not only that, but the larger economies of scale would save millions of gallons of fuel while reducing emissions.)
Maybe I, for one, should be careful what I wish for. Fewer departures would mean fewer planes and, in turn, fewer pilots. No matter, as expecting the airlines to voluntarily shrink their operations -- even if, in the long term, it meant greater profits -- is foolish. Along with their customers, they are resigned to a certain acceptable level of inconvenience. Barring a total redesign of some of the nation's busiest facilities -- about as far-fetched a proposal as constructing a new civilization on Venus -- we'll be living with this hassle indefinitely. Skies and runways are clogged, flights are late. We pout and shrug our shoulders.
- - - - - - - - - - - - - - - - -
Top five U.S. airports, percent on-time arrival performance (2007)
1. Oakland (OAK) 79.9 percent
2. Houston (IAH) 78.1 percent
3. Baltimore (BWI) 77.9 percent
4. San Diego (SAN) 77.5 percent
5. Atlanta (ATL) 76.7 percent
Bottom five U.S. airports, percent on-time arrival performance (2007)
1. Newark (EWR) 55.0 percent
2. New York (LGA) 58.0 percent
3. Chicago (ORD) 58.5 percent
4. New York (JFK) 60.0 percent
5. Philadelphia (PHL) 64.9 percent
World's 10 busiest airports, annual takeoffs and landings
1. Chicago (ORD) 970,000
2. Atlanta (ATL) 958,000
3. Dallas (DFW) 711,000
4. Los Angeles (LAX) 650,000
5. Las Vegas (LAS) 605,000
6. Houston Intercontinental (IAH) 562,000
7. Denver (DEN) 559,000
8. Phoenix (PHX) 555,000
9. Philadelphia (PHL) 535,000
10. Minneapolis-St. Paul (MSP) 532,000
Source: Bureau of Transportation Statistics, Air Transport World
- - - - - - - - - - - - - - - - -
Do you have questions for Salon's aviation expert? Send them to AskThePilot and look for answers in a future column.