What would happen if feds cap airline change fees?

Once upon a time you bought a plane ticket from point A to point B and it included a seat assignment (unless you were on Southwest), luggage (checked or on-board), and maybe even a snack or meal.  Now everything has been unbundled, with separate charges for seats, luggage, priority boarding and so much more.

Congress is considering an intervention by capping the amount airlines can charge for changing a flight reservation.  Currently American, Delta and United all charge $200 to change a reservation for a domestic flight.  According to WSJ, US airlines collected $2.9b in change fees last year.

What would happen if Congress put an upper limit of, say, $150 on change fees?  Standard economic analysis would interpret this as a price ceiling that would have unintended side-effects.  Airlines have already warned that they would raise fares and other fees in response, along with making fewer tickets changeable.  They also point out that customers who want more flexibility can pay for it when they buy their ticket by paying a higher fare.

But here's another thought.  The US domestic airline market is now very far from the competitive ideal of economics textbooks.  Price ceilings imposed on monopolists lead to lower prices AND increased output as long as the price provides a competitive rate of return.

My take: passengers and airlines have both benefitted from airline deregulation in the late 1970s.  Fares are much lower, more planes are flying and those planes are full.  What would really help customers is more competition.  What if we let foreign airlines provide domestic service?

United Airlines incident: price controls strike again

Airlines make a choice regarding how many tickets they sell on a flight.   Because airline seats are perishable commodities and the cost of servicing an extra passenger is zero, airlines want each plane to fly with a full passenger load.  Theaters face the same challenge.  Yet when you buy theatre tickets, how often do you find someone else in your seat?  

One reason airlines rely on overbooking is that US Department of Transportation regulations encourage it, as pointed out in this HBR online piece.  The regs allow bumped passengers to be paid 200-400% of the price of their ticket (one-way, I might add) with an overall cap of $1350.  So a passenger who bought a heavily discounted ticket might only receive $400-500 in compensation for being bumped, well below what a true volunteer might demand.  

Airlines could manage passenger loads in different ways, such as penalizing no-shows who do not contact the airline in advance and are not on a connecting flight.  If airlines insist on overbooking, then the most efficient (in the economics sense of the word) compensation mechanism would be an auction where passengers bid for the right to be bumped.  In the case of the infamous Chicago to Louisville flight two weeks ago, the bid price for being bumped would rise until there were four true volunteers.  That would no doubt be quite a bit more than United actually spent, but I bet they sure wished in retrospect that they had paid those four passengers enough to get them to exit the plane without assistance.  

Is the TSA a lose-lose proposition?

The Transportation Security Administration has been in the headlines lately as wait times in Chicago O'Hare, Charlotte, and other airports have hit three hours and passengers have missed flights.  Although some passengers will enroll in PreCheck (I just did and it is pretty easy to do) and be able to get through faster, many are starting to ask whether the TSA can pass a basic cost-benefit analysis.

The TSA spends about $5.5b per year on airport security and employs 47k security officers.  According to NYT, TSA claims that tighter budgets have forced personnel cutbacks, whereas TSA critics have responded that the agency has failed to plan and failed to prioritize.

Although the cost side is pretty clear in terms of budget and passenger wait time, the benefits are much harder to measure.   I have yet to hear of a TSA screen confiscating materials that were to be  used in a terrorist attack.  Think of the movie that could have been made, with Melissa McCarthy as the heroic screener catching Murray Abraham with a ticking bomb!

Does fear of apprehension by the TSA deter would-be terrorists?  The evidence on this front is not encouraging, as reported in today's WP:

Agency watchdogs have documented that undercover security operatives managed to smuggle 67 illegal weapons or simulated bombs past TSA security on 70 tries last year, that TSA officials were unable to properly vet 73 aviation employees who had links to terrorism, thereby allowing them access to secure areas, and the senior managers have a long history of bullying whistleblowers who identify potential problems.  

Another potential benefit is peace-of-mind for passengers.  It could be that even if TSA is totally useless for preventing terrorist acts, passengers may value the service and be reluctant to fly in its absence.  The agency is partially funded by passenger fees.  In the future it would be worthwhile to (1) conduct research to measure the value of peace-of-mind and (2) figure out a way for air passengers to pay the entire cost.

Airlines and antitrust

On the peak holiday travel weekend the Justice Department announced that it was launching an antitrust investigation into the airline industry.  The four largest airlines in the U.S. now have 80 percent of the market.  Three of those airlines were involved in mergers, all of which were approved by the Justice Department!  The feds seem concerned that whenever an airline exec says the word "discipline" at an industry conference, it is secret code for "price fixing" or "capacity limiting."

A recent WSJ piece looked carefully at recent trends in air travel capacity.  It turns out that there are 12% more domestic seats for sale now than two years ago, hardly what you would expect for an industry with high fixed costs and (now with lower fuel prices) more modest variable costs.  Airlines are cutting back on flights but adding more seats to each flight by (1) reducing space between seats and making the seats smaller and (2) replacing small regional jets with larger aircraft.

If the feds are seriously searching for a factor limiting capacity in the industry, they might want to take a look at airports.  When was the last time a new airport was built in a major city?  When was the last time a new runway was added or more gates were added in the average city?  If local governments fail to invest in airport capacity, it will be hard for the airlines to put enough seats in place to meet demand.

Why airlines cancel flights

Many of you will be on a plane this Labor Day weekend; 1.5% of you should expect your flight to be cancelled.  Amy Cohn, an industrial engineering professor at Michigan (a great university with unfortunate choices in school colors and mascot), explains the logic behind flight cancellations in this New Republic piece.

Simple economics would focus on marginal revenue and marginal cost.  That is, if the savings in jet fuel and labor hours offset the costs of rebooking passengers then it makes sense to bump the flight.  For instance if there are 8 am and 10 am flights from Raleigh to Detroit and both are half-empty, it makes sense to cancel one.

Cohn points out that network effects need to be considered.  The 8 am flight to Detroit is likely to go on to as many as eight other locations by the end of the day, all of which can be messed up by a cancellation.  Also, flights now are much fuller than they were 20 years ago, so rebooking can be quite expensive for the airlines in terms of bumping would-be passengers who would have paid a hefty premium to book a last minute seat on a later flight.

Have airline change fees gotten out of hand?

If you run an airline, load management is a key determinant of whether you make money.  Left to their own devices, customers would gladly reserve seats and then decide at the last minute whether to show up or not.  An empty airline sheet is like yesterday's newspaper - no one is going to buy it.  To create an incentive for passengers to book flights that they plan to take, airlines came up with cancellation fees 20 years ago.   In 1991 they generated $51m in revenue; today they generate more than $1b.  (Aside: these fees are not subject to federal air service taxes or local fees for airline use.  Ditto for baggage fees and other incidental charges.  We have seen airlines shift their revenue flow significantly over the last five years to these untaxed revenue sources.)

Airlines now charge between $150 and $200 to change a reservation.  For an international flight, this is a charge that is hard to avoid, especially if you already have completed one leg of your journey.  But for domestic flights, passengers can compare the cost of a one-way ticket home and the cost of changing their reservation.  WP reports that more and more domestic customers are booking their own return flights and, if warranted, pulling no shows at the gate. 

In a nutshell, the change fees may have gotten so high that airlines are now facing a problem with no shows.  Given the choice between paying $200 for changing a reservation and paying $150 for a one-way return, the no show approach makes economic sense.  Better to pay $150 and never engage with the airline that sold the ticket than to spend 30 minutes on the phone and pay more. 

My take is that companies usually think through their pricing structure strategically, but in this case they may need to reconsider.  They may need to think about changing the penalty for changing a booking to a percentage of the cost of the flight. 

Will feds let US Airways-American merger fly?

Sometime this week American Airlines is expected to merge with US Airways, with the combined airline using the American brand and US Airways CEO Doug Parker expected to run the enterprise.  The new American would be the world's largest airline, leaping over United (#1) and Delta (#2).  Currently American is #3 and US Airways is #5 in terms of US market share. 

Economically the deal makes lots of sense.  US Airways has a strong domestic presence but lacks American's international routes.  The two airlines do not have very many overlapping routes, but they would most likely eliminate some redundant hubs.

But will the deal pass the sniff test at the Department of Justice?  Delta was allowed to acquire Northwest in 2008 and United was allowed to acquire Continental in 2010, so one would think that this merger would be allowed to proceed based on these precedents.  On the other hand, DOJ has sued to stop the acquisition of small fry Corona by big-boy Ambev, as well as AT&T's bid to buy T-Mobile. WSJ reports that the European Union also would have to sign off because of the large number of trans-Atlantic routes involved.

Stay tuned.  To paraphrase Bette Davis in "All About Eve," fasten your seat belts, it's going to be a bumpy ride. 

Where does NYT get business writers?

On today's front page, NYT leads with a story about declining service to smaller airports, blaming deregulation, consolidation and capitalism in general.  As many news stories do, the lead paragraph includes an airline horror story: the 12 hour ordeal of one Josh Hunter as he took Continental/United  from Mobile to Cincinnati on Easter.  Actually he flew from Mobile to Houston to Cleveland to Columbus and then drove to Cincinnati, with 3.5 of the 12 hours spent in the car.  

Could there be a quicker way to go?  Of course!  Go to Kayak.com and you quickly see that you can take Delta to from Mobile to Atlanta to Cincinnati in 3.5 hours of flying and layover time.  So why did Josh take the longer route?  Maybe it had something to do with price?  And maybe he bought on Priceline or one of its ilk? 

To be balanced, let me just say that the NYT article does raise some more valid concerns.  The price charged for flying in and out of smaller airports has been going up (the cheapest Mobile to Cincinnati round trip fare in Kayak for the days I sampled was $524).  And some airports have lost service altogether, although I am not sure the citizens of Sioux City, Iowa have a constitutional right to air service.

One economic head-scratcher for me: why can I go from RDU to San Francisco for $282 on May 17-21 whereas the cheapest fare from any New York City area airport to San Francisco is $458?

Southwest-AirTran merger

Southwest Airlines has historically followed a strategy of gradual expansion to new markets.  Now they have thrown out the old playbook and radically increased their potential market share by buying AirTran.  This will create a number of challenges:

• Southwest has specialized in city-to-city flights and has avoided the hub and spoke system employed by the major carriers.  Most of AirTran's operations are based in Atlanta with many passengers transferring from one plane to another.  How to merge the two business models?
• Airline mergers are usually messy businesses, especially as unionized employees jockey for position on seniority.  Southwest's business model relies heavily on high employee engagement.  How will they keep everyone aligned with the mission and vision?
• Southwest currently flies only one type of plane, the Boeing 737.  Airtran uses 717s as well as 737s.  How can the two types of planes be meshed into an integrated system that relies on fast turnaround time on the ground and employee familiarity with only one type of aircraft?

The upside for Southwest is greater access to some major markets: Atlanta, New York and Washington DC.  A greater presence for Southwest in those markets is also good news for consumers. 

One final thought: after approving the United-Continental merger, surely the Antitrust Division of the U.S. Department of Justice will not block this merger on the grounds that it would limit competition?

Proper consequences for airline overbooking

I do not normally link to WSJ editorials, but they ran one yesterday on airline overbooking that contains a very important economic lesson.  Right now each airline has its own bumping policy; most provide up to an $800 voucher if they cannot get volunteers to accept smaller bribes.  Such a voucher is obviously inadequate compensation; otherwise, someone would have volunteered to accept it.  One of the most basic principles of economics -- gains from trade between well-informed adults -- is being violated.  Bumped passengers should be allowed to collect enough to offset the inconvenience.

How much more?  The Obama administration is proposing that the upper limit for vouchers be bumped up to $1300.  Although this will help some passengers, the WSJ rightly argues that there should be no upper limit at all.  Instead airlines should conduct proper auctions to ration the available seats.  This would do two good things.  First, passengers would choose whether to take a later flight in response to the airline's offer; no one would be forced off a flight.  Second, airlines would cut back on overbooking.  It could very well cost an airline more than $800 to properly compensate the last overbooked passenger.  Right now airlines have every incentive to sell a ticket on the same date of an overbooked flight for an amount over $800, knowing they will end up ahead.