It was almost exactly 40 years ago that FedEx founder Fred Smith had the idea of flying a small fleet of Dassault Falcon 20s into Memphis, Tenn., from various cities in order to unload and redistribute packages and fly off again. While his idea has long become the dominant paradigm across the legacy airline world, new threats are putting it to the test.

Smith was able to connect many more markets with each other than if he had tried to link all of them directly. This made little difference for the cargo, and when passenger airlines adopted the concept later, passengers may have minded having to make connections at first, but they soon discovered that airlines flew to destinations they would otherwise not have been able to serve. Some highly successful airlines, most importantly Pan American Airways, tried operating without a hub (Pan Am did not even have a domestic network) but failed when they suddenly had to compete with the likes of Delta Air Lines, which had the advantage of collecting a lot of additional passengers to whom Pan Am did not have access.

Hubs began to develop after the airline industry deregulation in 1978 as airlines gained the freedom to arrange their route networks as they pleased. Later, aircraft technology played a key role in increasing hubs' market penetration even further. Bombardier and then Embraer introduced 50-seat regional jets that allowed airlines to offer service to smaller communities they had not been able to serve with larger aircraft. The boom of the 50-seaters lasted many years, but high fuel prices have now made them uneconomical to fly in most markets and, while regional feeder airlines in the U.S. still operate large numbers of them, they clearly have no future.

That is part of the reason that the hub system has become a double-edged sword for the traditional hub operators. High fuel prices and competition from low-cost carriers (LCC) are making it difficult to sustain the short-haul connections that feed the long-haul services on which the hub operators so depend—in fact they have given up on many smaller markets, particularly in Europe and North America. At the same time, carriers such as Emirates and Turkish Airlines have established thriving hubs.

“The hub has to be understood as a factory to combine itineraries,” says Michael E. Levine, a former senior U.S. airline executive who is now a researcher and senior lecturer at the New York University law school. “It is an extremely valuable device.” The combination of many different passenger itineraries into one feeder flight makes more services economically feasible. But the aggregation can also reduce the number of spokes and city pairs. “Hubs are vulnerable to aggregate changes in demand and production costs,” says Levine. This does not occur in the direct-services networks operated by LCCs, so they can give up routes more quickly.

Demand has been weak in the developed economies for quite a number of years, but legacy carriers have tried to keep the hubs intact while pulling out of hub bypass services. Production costs have become the more serious threat, and with that Levine predicts “profound change” for hubs in the U.S. and Europe. The sharp rise in fuel costs has led to the loss of many smaller feeder markets that could previously be served with 50-seat jets. In Europe, even the future of 70-seat jets looks questionable, and the regional airline industry has been shrinking fast. Airlines are moving up to larger aircraft, but there is not an infinite number of markets that can sustain them and the larger the markets grow, the more likely they are to attract LCC competition.

Using larger aircraft also increases vulnerability, and the aggregate effects of pulling services described by Levine become greater, too. The loss of feed affects the nonstop services originating at the hubs, as well, because the connecting itineraries are missing, and could affect long-haul operations, although “people are willing to accept more inconvenience for long-haul,” Levine says.

Reducing the number of spokes at hubs is also problematic because hundreds of possible city pairs go with a spoke in large hubs, and “big hubs work better than small hubs,” Levine points out.

One of the most important lessons is that a hub is not a hub. British Airways' operation at London Heathrow Airport offers lots of connections simply because it is so big, but capacity constraints prevent the airline from building meaningful traffic flows between certain arrivals and departures; they just happen by coincidence. Delta Air Lines has so many banks at its Atlanta operation that there are hardly any valleys between the peaks left, and the sheer number of flights brings a lot of connectivity quality. There are hubs with sophisticated planning such as Frankfurt, Amsterdam or Paris, where the dominant carriers are trying to learn from past mistakes. And there are the big Asian bases like Singapore or Tokyo that rely on a relatively small number of high-density routes and are therefore essentially big airports with some hub elements.

There are the lucky ones, like the Persian Gulf area's Dubai or Doha, geographically placed to offer almost exclusively long-haul-to-long-haul connections, avoiding the malaise of high unit costs on short-haul feeders.

While hubs are great for achieving broad market reach, they are also a very expensive way of operating. It is much cheaper to serve a market with one aircraft that makes just one high-fuel-burn takeoff and landing, requires a single payment of navigation and airport fees, and keeps flight distances as short as possible without a detour to make a connection.

LCCs have long taken advantage of this structural weakness by building point-to-point services with sufficient untouched demand. Many hub-and-spoke carriers have opted out of these markets because their overall unit costs were too high to even consider going head to head in the direct-services business.

“Low-cost carriers make feed from some destinations more difficult,” says Markus Franke, who runs Franke Aviation & Transportation Consulting. “Serving secondary markets over hubs becomes all the more unattractive for the legacy carriers and can affect long-haul feed. In a worst-case scenario, the entire traffic structure of a small or mid-sized hub without differentiation could collapse.”

So far, that is mainly a problem for secondary hubs such as Vienna or Copenhagen that have a large share of short-haul traffic. Whether because of high fuel prices or competition, some airports, have been “de-hubbed” (such as Memphis by Delta), and Levine expects more to follow.

Bigger hub-and-spoke operations are still better protected against LCCs. “If I have 40 short-haul destinations, that gives me 1,600 itineraries,” says a major airline CEO. “If a low-cost carrier moves into 100 of these markets with nonstop flights, I still have 1,500 left. And don't forget the long-haul destinations on top.” But the fuel-cost challenge eating into the regional feed affects all of them.

Fundamental hub economics still work, nevertheless. It is the implementation that causes trouble and needs changing. Some hub carriers have come close to failure because their networks were too big for what the local and connecting market required. “The hub concept is by no means dead,” says George Hamlin of Hamlin Transportation Consulting. “It just has to be managed properly.”

Hubs and aircraft choices are interconnected and so are airline business models and aircraft development. “As much of the long-haul market out of Europe can only be served by the hub-and-spoke model, there will always be demand for bigger aircraft,” says Alexis von Hoensbroech, Lufthansa's head of network planning.

From an equipment perspective, what needs to happen for the concept to be threatened? “As the long-haul point-to-point demand out of Europe is usually below what you need to fill a 250-seat aircraft, you could only get around the hub-and-spoke system if you had, for example, a 100-seater at the unit-cost level of a 250-seater,” says von Hoensbroech. “The stimulation potential for long-haul demand is much lower than for short-haul.” Such an aircraft would also be a great tool to address the high seasonality that is typical for direct routes, argues Strickland.

“Hubs could be threatened if there were aircraft that were incredibly more efficient than what's in use and on the horizon today,” says Franke.

Levine says some of these things might happen soon. “The [Boeing] 787-8, flying 200 people efficiently over 8,000 nautical miles, will be the first real test case for hubs that are heavily dependent on long-haul,” he asserts.

There are arguably two airlines that are reinvigorating the hub concept again. Emirates flies almost exclusively long-haul to long-haul, has no night curfew and sits in the middle of the world's fastest growth markets. Because it is already so big, opening just one more route means it can offer literally hundreds more city pairs at additional costs that become ever more marginal. And the hub has by now developed such momentum, creating enough growth, cash generation and profits that financing further growth is relatively easy.

However, there are constraints that even Emirates has to be concerned about. Airbus and Boeing cannot deliver quickly enough aircraft that are large enough for what the airline has planned, and the Dubai airport is approaching its capacity limits, even taking into account various multibillion-dollar expansion projects that will be completed in 3-4 years. The success of the new hubs will therefore not be limitless. And Levine points out the region's relatively small home market and political instability.

“The rise of Turkey and the Turkish airline industry creates an alternative hub for access to the Middle East and India with much stronger fundamentals,” he says, adding that “any development of aviation infrastructure in India would significantly reduce [Persian] Gulf hub demand.” Levine is concerned that huge widebody overcapacity could become an issue, with major consequences for manufacturers and aircraft market values.

Some effects of Emirates' hub limitations are becoming apparent: the airline recently revealed plans to open so-called fifth-freedom routes from Thailand, Singapore or Japan to the U.S., taking advantage of open-skies arrangements that would allow it to fly some two-stop connections.

The Turkish Airlines concept is distinctly different from Emirates', as Turkish operates predominantly narrowbodies on 3-4-hr. flights into Europe, Africa and Central Asia. That approach drives down unit and overall trip costs and allows the airline to serve markets that are either too small (for Emirates) or too distant to be served directly from Europe using narrowbodies (for European carriers). Turkish has a sizable long-haul fleet and will most likely add many more aircraft soon.

Tap the icon in the digital edition of AW&ST for a look at the world's 20 largest hubs, or go to AviationWeek.com/hubs

The World's 10 Busiest Air Routes
Region Route Passengers (millions) Growth 2012 Rank (change from 2011)
Asia Jeju—Seoul 10.156 2% 1 (=)
Asia Sapporo—Tokyo 8.211 8% 2 (+2)
Latin America Rio de Janeiro—Sao Paulo 7.716 -1% 3 (-1)
Asia Beijing—Shanghai 7.246 7% 4 (+3)
Southwest Pacific Melbourne—Sydney 6.943 -2% 5 (=)
Asia Osaka—Tokyo 6.744 -11% 6 (-3)
Asia Fukuoka—Tokyo 6.640 -3% 7 (-1)
Asia Hong Kong—Taipei 5.513 2% 8 (=)
Asia Okinawa—Tokyo 4.584 12% 9 (new)
Africa Cape Town—Johannesburg 4.407 -1% 10 (-1)
Source: Amadeus