EconLog small logo  

Markets, Morals, and United Flight 3411

1470 Economists Say Immigratio... The public only thinks it like...


by Steven Horwitz

Presumably most readers are now broadly familiar with the recent incident on United flight 3411 from Chicago to Louisville. The short version is that the operator, Republic Airlines which is a United regional carrier, needed four volunteers to give up seats to make room for a crew it needed to transport to Louisville for a morning flight. Flight 3411 was boarding when the crew was assigned the seats, so passengers were seated when the ground crew offered compensation for folks who would leave the plane and take a later flight. When there were not enough volunteers, the gate agents used United's protocol for involuntary bumping from a flight. One passenger, a doctor, refused to leave when his name was called. Reports indicate he became belligerent and the Chicago Aviation Police were called.
They very forcibly dragged him off the plane, causing significant injuries to his mouth and elsewhere in the process, as passengers filmed the incident. The videos went viral, United's public relations response was terrible, and everyone on social media instantly became airline experts. Boycotts of United were called for and its stock took a tumble.

I have weighed in on the incident on Facebook myself, and I have my own take on it, but here I'm more interested in exploring a variety of questions this incident raises for classical liberal political economy. How can we use those ideas to understand what happened? What questions and issues does this incident pose that are challenges for those ideas? I will offer some views of my own in what follows, but it's these questions that interest me most.

Let's start with some of the economics here. First, why do airlines overbook flights and have to bump people? From the airlines' perspective, they would like every flight to leave the gate completely full. The marginal costs of adding the last few passengers is very low (just a tiny bit of extra fuel and food and drinks), so the net gain from selling those seats is pretty high. Given that some passengers might have refundable tickets and change their plans, and that others will miss flights due to missed connections or any of a bunch of other reasons, it makes sense for airlines to oversell flights and have "spare" passengers to fill in. The cost to doing so is that if too many people do show up, some folks will have to give up their seats.

The late Julian Simon helped to develop the system for dealing with such situations, which was to offer passengers compensation to give up their seats. This is economics in action, ensuring that those who value the seats the most will get them, while those with "flexible travel plans," will be compensated for their flexibility. United offered up to $800 plus hotel and meals to passengers on flight 3411 but did not get enough takers.

The response from economists is "well, why stop at $800 then?" If supply curves slope up, raise the price and elicit more volunteers. There are a few answers to this. One of them involves price controls. The Department of Transportation limits the compensation airlines are required to provide to involuntarily bumped passengers to 4 times the fare to a maximum of $1350. It makes no sense for the airlines to offer more than that for voluntary boarding denials because if no one takes $1350 voluntarily, the airline doesn't have to pay more than that if they have to force someone off the plane.

But, you might respond, given what happened and all the negative publicity for United, wouldn't it have been worth it to go above the maximum? That maximum is all they are required to provide, but they could provide more. I think the answer here takes us into just the sorts of questions that this case raises.

I'm guessing that the gate agents just assumed that the people involuntarily denied boarding would comply and thus get the required compensation. They could not have anticipated the reaction of the one passenger, nor could they have predicted that security would bloody him and that the videos would go viral. Of course it would have been cheaper in retrospect for United to have offered more than the maximum, but it's not clear why they would have thought that was a wise thing to do in the situation they found themselves at the moment of choice. Economic choices are always made in the face of uncertainty and are driven by our expectations. What we know after the fact can't explain why we chose the way we did, knowing what we knew then. The job of the economist is to start our explanations from the perceptions of the choosers. This is the subjectivism and methodological individualism at the heart of sound social science.

Economists might also ask why the doctor didn't offer to personally compensate someone else for giving up their seat. Suppose he had said "I'll pay $2000 to anyone who agrees to leave the flight so that I can get home to my patients." First, it's not clear that doing so is legal, as the airlines might see that as an illegal resale of a ticket. Second, the transaction costs of enforcing that contract would be high if the doctor didn't have the $2000 in cash. How would the other party ensure the doctor paid him or her?

These two points nicely illustrate the issues involved in the Coase Theorem. In situations of apparent externalities, such externalities can be eliminated through negotiation and exchange if transactions costs are low and property rights are clearly defined and well-enforced. Those two conditions were not present here, preventing an exchange from solving the problem.

What this aspect of the incident nicely shows is that prices, markets, and exchange can be very effective at diffusing conflict and allocating scarce resources such as airplane seats if they are allowed to work and the institutional conditions (low transaction costs and clear property rights) are sufficiently present. The doctor clearly placed a high value on taking that particular flight and someone else would have been willing to give up a seat at a high enough price. Normally raising the offer induces enough people to give up their seats, but in the face of a maximum compensation requirement and no reason to expect the violence that ensued, the pricing process couldn't get the job done. Raising or eliminating that maximum compensation requirement would be a sensible policy response to this fiasco.

A second part of this incident of interest to classical liberal political economy is the role of contracts. As noted above, the inability to know if a contract between the doctor and another passenger for direct compensation was either legal or sufficiently low cost to enforce was a problem preventing a better outcome. The other two relevant contracts were the contract between the passengers and United and then the contract between United and the Republic Airlines crew they had to transport.

The latter contract was such that United was obligated find those four crew members seats on that flight even though they claimed the seats after the flight began to board. One can rightly blame United for finding itself in a situation where a crew was so out of place at the last minute, but, given that, it had to accommodate them.

This contract, of course, is also framed by various federal regulations about crew rest. Why couldn't the crew take a later flight or even drive the 5 hours to Louisville? Most likely because that would not give them enough required rest for their early morning flight. United clearly made a decision to inconvenience four passengers on the flight to Louisville rather than delay or cancel a whole plane full of them the next morning. Again, not expecting a violent outcome, that seemed like a reasonable decision at the time.

This aspect of the flight is also a nice example of Bastiat's "the seen and the unseen." Suppose United had decided to leave four passengers on the flight, preventing the crew members from getting seats. We might well applaud United's surface humanity because we'd see how it valued those customers and wanted to get them to Louisville on time. What we might not see is the canceled flight the next morning, which was the direct, if unseen, result of that decision. All of the people in Louisville inconvenienced by the cancellation would likely not know that United's "humanity" the night before was the cause. Good social analysis tries to see the unseen and recognize the opportunity cost of choices. Had United flown without the crew for Louisville, the visible gains to four passengers would have been more than outweighed by the unseen losses to far more the next morning.

Firms face difficult decisions like this all of the time, and had United been able to entice enough volunteers, or had the doctor not refused to disembark, this would have been a complete non-story of the sort that happens from time to time in air travel. The regulatory environment made it more difficult for the airline and its passengers to have the flexibility necessary negotiate solutions that could have avoided the conflict and violence that resulted. One might think that the maximum compensation rule or the rest time regulations are good ideas, but even well-intended regulations can have unintended consequences and outcomes like this reflect the often hidden costs of such regulations.

The other contract is the one stipulating United's obligations to its passengers. There is much debate on social media about what is meant by "boarding" in this context. United claimed that this situation was covered by the clauses in the "Contract of Carriage" that discuss involuntary boarding denials. Others have argued that because the passengers were already on board, this was covered by the "Refusal of Transport" section. If the former is the case, then United acted properly in denying seats to four passengers, including the doctor. If the latter is the case, the argument is that creating seats for crew movement is not listed among the acceptable reasons for refusing transport. If so, then the crew member's instruction to leave the plane was not a lawful instruction and the doctor's refusal to obey is not a violation of the obligation to obey crew instructions, or so claim some legal experts.

I'm not going to take a stance on this debate here. The debate does point out the importance of clear and complete contracts for markets to work for mutual benefit. Of course no contract can be totally complete or require no interpretation, and that's what judges are for. Almost certainly, the doctor will sue and courts will sort out exactly what is meant by "boarding," among other things. And whatever those decisions are, they will have the salutary effect of creating clearer expectations for both passengers and airline employees that will, one hopes, avoid these sorts of situations in the future - or at least give us guidance about how best to deal with them. Clearer legal expectations can provide room for negotiated solutions, or at least clarity about which party has which rights.

Without the ability to use prices and negotiation to resolve the conflict at the heart of this incident, it is not surprising that force was necessary. Given United's (reasonable at the time) decision to inconvenience four passengers rather than a whole flight the next morning, four people were going to have to get off of that plane. When there was no other voluntary mechanism for the doctor to keep his seat by buying out another passenger, and when he refused to disembark, United was going to have to use some sort of force to get him to comply with what they thought (perhaps wrongly) were the conditions of his contract to travel.

The problem, of course, was that the Chicago Aviation Police who responded used excessive force in resolving the issue. I have not seen a video that shows what happened before they began to drag the doctor out, but it's hard to believe that the kind of force they used was necessary and appropriate. And United certainly has no interest in the reputation it has now gained as a company that allows cops to bloody its customers. The cultural reasons why police might feel entitled to use that degree of force are beyond the scope of this post. However, it's worth noting that once exchange, prices, and negotiations cannot be used, conflicts over scarce resources will involve force, and quite possibly violence.

The lesson here from classical liberal political economy is that clarifying the institutional questions about the nature of the Contract of Carriage, the ability to auction off a seat directly to a customer, and the unintended consequences of various regulations, would help to avoid the need to use force in the future. In other words, the rules of the game matter. Before we blame the players for their possible errors, we need to know if the rules were sufficiently clear to the players and plausibly enforceable. The debate over denied boarding versus refusal to transport as well as the lack of clarity over whether the doctor could have auctioned off his seat reflect uncertainty about the rules that make positive-sum game outcomes that much less likely.

This whole incident raises a question that classical liberals must often confront. Suppose, for the moment, that United acted completely within the law throughout this whole process, including asking the police to remove the passenger. Suppose that the passenger was clearly in the wrong here. At what point, if at all, does a firm's legal right to take certain actions run into moral or ethical limits on what they should do. In other words, just because the airline could do what it did, should it have done so? To be fair, it could not have predicted the violence used by the police, who clearly had no moral limits on their behavior, making it surely the most outrageous part of this whole incident. However, is there some moral or ethical calculation that United could have engaged in that could have prevented this?

It does seem like all of their other options, up to the point of calling in the police, would have caused more harm to more people than the path they chose. If one's ethics are broadly utilitarian, it's not clear what the other options were. If one's ethics are more deontological, what was the point at which the airline should have behaved differently? One can criticize United's competence with respect to crew placement, but where was their ethical failure here, if any, and what would the consequences have been of making that more ethical choice? These are questions that political economy can help us answer, but perhaps cannot answer on its own.

Finally, two brief observations. First, stuff happens. There's no doubt that it took a perfect storm of events to create this whole fiasco: Republic Airlines having crew out of place, a flight that was full, a compensation scheme that didn't get any takers, a regulatory environment that made alternative solutions unobtainable, a passenger who refused to be passive, and, especially, law enforcement officials who clearly over-reacted with with excessive force. Millions of people fly every year and fewer than 1 out of every 10,000 get involuntarily denied boarding. It took a highly improbably set of circumstances to produce this awful scene, and we should think carefully about whether the best response to "black swans" is a regulatory one. Fixing the rules of the game through the legal system and removing regulations that contributed to the problem would seem better options.

The beating that United has taken on social media and in the stock market is well-deserved. Whatever their degree of complicity in creating the problem, their response to it has been an utter disaster. Firms need to get out in front of these sorts of problems and "run to" them rather than running from them. The public relations response just confirmed the narrative from the video that this is a firm that does not care at all about what happens to its customers. In a world of instant, near-zero cost communication, a firm's reputation is everything. A better, more empathetic response would have gone a long way in stopping the cascade of negative publicity. Even those of us who like markets and who can understand why United made some of the choices they did as the incident itself unfolded should feel free to join the criticism of the mistakes it did make, both along the way and in the aftermath. Negative feedback is central to the market punishing those who make mistakes.

One thing is for sure: the odyssey of flight 3411 will rank up there with the scalding McDonald's coffee cup incident as lessons in business behavior that will be studied repeatedly. The McDonald's case did have the positive effect of leading stores to reduce the temperature of the coffee they sell, which has reduced the number of burned customers. Hopefully, the numerous lawsuits that will surely follow the events of flight 3411 will have a similarly positive effect by clarifying the rules of the game that govern such situations, and thereby enable the positive-sum exchanges and negotiations of the market to substitute for the violence when situations like this arise again.

Steven Horwitz is currently the Charles A. Dana Professor of Economics at St. Lawrence University in Canton, NY, an Affiliated Senior Scholar at the Mercatus Center in Arlington, VA, and a Senior Fellow at the Fraser Institute of Canada. He is the author of three books, including most recently Hayek's Modern Family: Classical Liberalism and the Evolution of Social Institutions. He has written extensively on Hayek and Austrian economics, monetary theory and history, and American economic history. In the Fall of 2017, he will become the Schnatter Distinguished Professor of Free Enterprise at Ball State University in Muncie, Indiana.

COMMENTS (35 to date)
Phil writes:

With a title including the word "morals", the amount of force used by the police is certainly within scope. It, too, is an issue of marginal benefit and marginal cost.

A more fundamental moral question is why the police were used to resolve a civil (contractual) dispute in the absence of a crime. A business should not be allowed to use the coercive power of the state to resolve a business dilemma.

drobviousso writes:

> In other words, just because the airline could do what it did, should it have done so? To be fair, it could not have predicted the violence used by the police, who clearly had no moral limits on their behavior, making it surely the most outrageous part of this whole incident.

Great essay, as I'd expect from you. But I want to disagree with this point a bit.

"it could not have predicted the violence used by the police." What else were they called for, if not for the threat or application of force. Police usually get compliance by the threat of force, but its a legitimate threat they back up.

If the airline *does not* recognize the purpose of calling the police is to call for the application of force, they are fooling themselves. If the airline *does* recognize that, they chose to apply force instead of pay a market clearing price.

That, to me, is the nut of the ethical failure. 'I'd rather hit you than pay you' is an unethical position. But I'm sure its quite difficult for United employees are basically saying this based on their human intuition. I don't think anyone explicitly said "I'd rather break this guy's jaw than pay out another hundred bucks," but United acted in a way that had that effect. That's an unintended consequence of the complexity of their auction system, but its an auction system they designed and put into place.

Hazel Meade writes:

Phil gets at the key, point but I'm not sure I agree.

Presuming that United was in the right and they had the right to remove the passenger from the airline, what are they supposed to do if the passenger refuses to leave? The airplane is private property and when someone is trespassing illegally on private property it is within one's rights to call the police to make them leave. Also contract enforcement is a legitimate function of government. If a judge rules that a person has to pay another person a certain amount, what happens if the first person refuses to pay up? Presumably, at some point, police get involved.

I think maybe the one misstep they may have made is that they should have had their own private security guards to deal with this first, and only as a last resort relied on the TSA. This way the agents could be trained to deal with passengers in a customer-service friendly manner and they can be fired if they handle a situation badly. TSA agents are trained to deal with potential terrorists and aren't accountable to united.

Steve Horwitz writes:

I agree Hazel.

Given that airport cops are probably not trained in de-escalation techniques (and should be), getting them involved in a customer service issue was a problem. I don't know what the feasible alternative was, given the doctor's adamant refusal to play by what appeared to be the rules.

bill writes:

I recommend the documentary: Hot Coffee

John writes:

Apparently Delta has already found a market solution for buying out seats , retail gift cards that enough people value more than travel vouchers.

Maniel writes:

Prof Horwitz,

From a UAL perspective, agents on the ground had a limited time to resolve the seating shortage without disrupting the flight schedule. If I had been in the decision maker’s situation, I would have gone through a written (or mental) checklist, e.g., 1) how many seats do we need?, 2) available tools (monetary compensation), 3) backup plan (involuntary removal), 4) passenger rights, 5) available resources for removal (UAL employees, TSA, other). As we all agree, two of the unknowns were passenger cooperation and removal-personnel tact and skill.

As Monday-Morning quarterbacks, we are well placed to make a better decision now, but we also agree that the question on the table is how to address the next situation. I suggest considering a new tool for the toolbox: passenger-removal insurance. It might well be unpopular, but passengers who did purchase it (ahead of time) could not be subject to involuntary removal. The argument for it is that makes explicit that, because they overbook, the airlines reserve the right to remove passengers with tickets and assigned seats.

Hazel Meade writes:

I made the same comment over at Marginal Revolution yesterday. The airlines could offer "Guarenteed Boarding" for an additional fee, for a limited number of seats per flight. Then if you really don't want to be the person who gets bumped, you can pay slightly more. Ticket prices would probably adjust so that non-guarenteed tickets would be a little less expensive, which would be better for the relatively poor who don't mind being reaccomodated.

JFA writes:

" The Department of Transportation limits the compensation airlines are required to provide to involuntarily bumped passengers "

This means that there is *minimum* that UA is required to compensate the passengers. The logic of your argument (UA didn't offer more because they were only required to offer x amount) actually calls for more regulation and increased required payouts.

Capt. J Parker writes:

It is a mystery to me why there is such tacit acceptance of the airline's practice of overbooking. In any other line of business such a practice would be called breach of contract or possibly fraud. Yeah sure, there's an overbooking notice somewhere if you click through three links on the airline's website. But, when I book a flight I am presented with a series of webpages that give me every reason to think that I'm entering into an understanding where I agree to pay money and present myself at the departure gate at the correct time and the airline agrees to fly me to my destination. It seem to me rather bizarre to assume United was exercising it's property rights in unseating passengers and not consider that United was committing a breach of contract that was in effect a breach of the passengers property rights.

Also, the more details that emerge about UA 3411 the more I think that the flight was not overbooked. United decided at the last minute to unseat boarded passengers because a problem with United's operations lead to a crew shortage at UA 3411s destination. Something is broken with the air travel market if service providers can breach implied contracts because of their own faulty operations and suffer no consequences. Libertarians need to hope that the reason for such a consumer unfriendly broken market is that the airline's are collectively hiding behind government rule making.

David Belkin writes:

"Had United flown without the crew for Louisville, the visible gains to four passengers would have been more than outweighed by the unseen losses to far more the next morning."

Well, it's the trolley problem, isn't it? Most individuals favor throwing the switch to divert the runaway trolley from the track where it will kill five to the spur where it will kill one, while most oppose throwing the fat man from the bridge to stop the trolley, even though the net gain in lives (utility) is the same. My sense is that bumping passengers before they've boarded has the feel of throwing the switch, while forcing a passenger to deplane has the feel of throwing the guy off the bridge - even when the opportunity cost of inaction (cancellation of the morning flight) is the same in either case and is 'seen.'

CMOT writes:

First, the passenger was totally passive, not belligerent in any way. Chicago Police Department put out a completely and obviously false report on the incident, as they have for every other police brutality case captured on video.

Second, the CAA police force consist mostly of officers who had to leave the Chicago Police Department due to corruption and brutality problems, and is run by persons who committed atrocities against detainees at Abu Ghraib. Yes, really.

So "To be fair, it could not have predicted the violence used by the police ..." is false. It was easily predictable.

Ken P writes:

They were unwilling to pay the market price for the seats they wanted so they worked around it with an eminent domain like approach. Fine print contract stipulations are not morally persuasive to me especially when dealing with a near monopoly situation like airlines.

United Airlines basically made a bet on the demand for seats, lost and then backed out of their bet.

John Hayes writes:

Hi Stephen, I think there are a lot more dynamics here than immediately apparent and I want to talk about the principal-agent problem between the airline and the employees running the auction.

Much has been made of the decision to stop auctioning at $800 but this is probably due to the approval chain within the airline. The airline (principal) would like to pay the lowest price to recall a seat, while the employee conducting the auction (agent) would like to finish with the least effort. To balance these goals, the airline will create "approval friction" making it harder to produce higher offers. In theory, this induces the employee to work harder selling the lower offers because it's just easier.

I wouldn't be surprised if exceeding $800 on a regional flight (ticket price

As opposed to making higher offers "higher friction", the airline could also reduce the effort of running successful auction by improving the sales skills of their employees. The first thing they should do is change the auction format, the Dutch auction is, slow and difficult to execute and shrinks the buyers market to those willing to draw the attention of 100s of people. A silent Vickery auction would produce a better price for the airline, even rounding up to the second lowest bid. Collecting bids individually would also be difficult enough to forge that airline management could reduce the toil required to approve higher levels.

Auctioning seats has been a great success, now do it better.

andy writes:

I don't quite understand the idea behind unintended consequences of regulation; if there were no regulation regarding the compensation, then either the airline couldn't refuse you the seating or they could refuse the seating without any compensation (I think that was actually the case?).
The only 'unexpected effect' seems to me that in such situation the refusals would be more frequent, people would get more used to it, the contract would be 'clearer' and there would be better mechanisms in place (e.g. non-refusal fee).

Mark Bahner writes:
This means that there is *minimum* that UA is required to compensate the passengers. The logic of your argument (UA didn't offer more because they were only required to offer x amount) actually calls for more regulation and increased required payouts.

It seems to me that at least United has probably learned from this incident that going above the minimum is good business policy. It would be interesting to see someone calculate, several months from now, what this incident cost United (e.g. from lower stock price, employee time spent dealing with the press, payouts to the passengers, etc.). My guess is that it will be over $2000. :-)

It's hard for me to believe *someone* wouldn't have given up their seat for $2000.

R Richard Schweitzer writes:

This treatise begins with an erroneous premise for the instance:

This was not a case of "overbooking;" that is, the contracting of transport of more passengers than available capacity.

This was a case of logistical malfeasance in providing for crew placements and their transfer/ transport.

That said, legally, the contract of carriage is a revocable license of transport, entirely at the discretion of the carrier.

Thomas B writes:

I found this incident really quite encouraging.

The public reaction was almost universally, "you sold the seat, you should have offered enough to get someone to volunteer their seat".

This is important for two reasons. First, it shows that - the massive progress made by Progressivism notwithstanding - the reflex of turning to market-based, mutually voluntary solutions is alive and well. Second, it also shows that the public has a firm grasp of the difference between Law and Regulation - or, in this case, between the implied contract and the Small Print - and comes down heavily on the side of implied contract, which is analogous, essentially, to rejecting regulation that conflicts with common law.

Finally, in my experience McDonalds has NOT learned from the hot coffee experience. I often buy McDonalds coffee at the drive through, and more often than not the coffee is scalding hot, undrinkable until I have gone far down the road. I have no doubt that if I should spill it in my lap, I would be burned. PRO TIP: when they ask if you'd like milk or sugar, ask them to put an ice cube into the coffee.

Mark Bahner writes:

Another interesting thought: The distance from O'Hare to Louisville is 317 miles. Within 20 years at maximum, and perhaps within 15, it will be possible to go from O'Hare to Louisville in under 3 hours by autonomous taxi, at a cost of about $100 (to the owner of the vehicle).

So the United people could say, "We'll give you $1000, and we'll pay for your autonomous taxi to get you to Louisville about 2 hours later than you'll get there by plane. How about it?"

bill writes:

A - the airlines could include one more box in the ticket purchase form: "if we need to bump you, how much do you want us to pay you?"

B - based on the responses, the airline may prefer to increase or decrease the amount it oversells any particular flight. (ie, if everyone wants $10,000, they might not overbook)

C - over time, the airlines might adjust the cancellation fees and flight change fees based on your answer to A above. If you're willing to be bumped for $500, maybe you can still cancel for $100, but if you want $10,000 to be bumped, then the fee to cancel would be the full price of your ticket.

I liked the analogy above of involuntary bumping is like eminent domain.

John Smith writes:

I strongly disagree with the article. The doctor had it coming. I personally am glad that the police used force on him, once he disobeyed a law enforcement officer.

You are entitled to verbally object to a LEO. You are not allowed to physically resist orders issued by a LEO. If you disagree, sue him later and let the courts sort it out.

Chris L writes:

R Richard Schweitzer has a good insight. This was not overbooking. Overbooking is normally handled at the gate where this sort of thing couldn't happen. Once everyone is on the plane the company has to overcome the natural propensity toward "this seat is mine". Also there is greater time pressure. The Airline created this problem with the logistical errors so should rightfully pay for the mistake. Hence the normal auction policy should be thrown out and simply increase the bidding as necessary to get a quick and satisfactory voluntary result. There is no legal limit on compensation for voluntary bumping.

Jim Glass writes:

United offered up to $800 ... but did not get enough takers. The response from economists is "well, why stop at $800 then?" ... The Department of Transportation limits the compensation airlines are required to provide to involuntarily bumped passengers to 4 times the fare... It makes no sense for the airlines to offer more than that for voluntary boarding denials because ... the airline doesn't have to pay more than that if they have to force someone off the plane.

"It makes no sense" only if the airline puts near-zero dollar value on the goodwill of its customers -- and near-zero cost on converting it to bad will. Which is just terrible in a highly competitive service business. Booting passengers involuntarily by force *always* will produce bad will.

Why Delta Air Lines Paid Me $11,000 Not To Fly To Florida This Weekend

So Delta gets good publicity from Forbes probably worth more than $11,000 in advertising, while United gets millions of dollars of negative advertising in return for opportunistically invoking a reg to save $800+. Clever!

In other words, just because the airline could do what it did, should it have done so? To be fair, it could not have predicted...

To the contrary, the result was utterly, banally predictable: Bad will among customers -- the booted one plus those who saw, those talked to, etc. Competent customer service businesses know that bad rep spreads fast, one unhappy customer quickly becomes 10 people who've heard bad things about you. This is why they are so relentless in pushing "satisfaction guarantees". The only thing unpredictable here was the extreme *degree* of the bad publicity. But even that was predictable, as in our social media world if you peeve off customers enough times this will happen sooner or later.

the gate agents ... could not have anticipated the reaction of the one passenger

Again to the contrary, they should have been trained that given the huge number of customers they deal with such events are a dead certainty to happen, and how to handle them. "The customer is always right", "Leave the customer satisfied" are marketing fundamentals that go back to the 19th Century which top service businesses teach endlessly.

The profit-maximizing strategy for the airline is to bid as high as it takes to get passengers to leave overbooked flights happily, then overbook to the extent that maximizes revenue net of that projected cost (easily done). Passengers who take "buy outs" definitionally are happy and become walking advertisements for the airline. (The Forbes story, several personal anecdotes I could tell.)

Airline managers who instead think the auction process is a burden imposed on them by regulation, that they can luckily escape at the 4x level to stick a shiv in randomly selected customers, are incompetent. (United's already gone bankrupt once.)

The law & econ of all this is quite interesting, but from the business side this is all one needs to know.

Jim Glass writes:

It is a mystery to me why there is such tacit acceptance of the airline's practice of overbooking

My son is a college senior who has a class in which they discussed this episode at length, massively bashing on United of course. This idea came up. 'Why do they even allow overbooking?"

Son is an econ major and asked the class: "Just for the record, how many people here would be willing to pay $40 more for every flight to get rid of overbooking?" Zero people replied.

Donald Greer writes:

So this is a perfect example of something I've started pointing out whenever somebody says something like, "There ought to be a law."

My standard response has become, "How many people are you willing to injure or kill to enforce that law?"

It seems obvious to me that any law, no matter how innocuous, must be enforce with the threat (and possible use) of force, and given enough such enforcements, somebody is going to have a bad day.

If blue lives actually mattered (as many busybodies like the mutter), then people would argue for the reduction in generally meaningless laws that cops have to enforce and get into fights and sometimes die over. And of course, in this case as in many others, it's not the cops, but the public at large that was the "victim" of the cumulative laws which created this situation and made it a law enforcement matter.

ConnGator writes:

Hazel, the customer did NOT consider himself trespassing, he had a legal right (ticket) to be there.

I understand there is fine print saying it could be revoked, but from his point of view the airline took his money, promised to fly him home, then broke that contract. That is why he refused to leave.

All involuntary bumping should be banned. If airlines overbook they should have to increase pay until they can persuade someone nonviolently to take a later flight.

Peter Gerdes writes:

The fact that people are demanding higher payouts isn't clearly evidence they favor consensual market based solutions. After all the customers all knowingly choose to buy tickets that provided for such a contingency (arguably the contingency was for before boarding but that doesn't matter see below) even though the market offers some options (jetblue) with no overbooking and some airlines (delta) with fewer involuntary removals.

Seems to me people aren't acting so much pro/anti market as having populist intuitions favoring simple property rules rather than complex agreements about contingencies.

A few quick points:

1) Whether or not United was technically in violation of it's Contract of Carriage the doctor was legally bound to vacate united property at their demand. If the order violated the contract he would have the right to sue for breech of contract but not to occupy United property without permission.

Same principle as if you sell me a $5 ticket to watch the super bowl at your super awesome home theater system. If you decide your going out instead of hosting at the last minute I don't gain the right to break into your house.

2) Seems to me the fact that the 4 seats were going to airline employees is irrelevant. Nothing about overbooking entails that the tickets must be sold rather than issued to airline employees.

Peter Gerdes writes:

IMO the right solution to this whole mess is for airlines to offer overbooking insurance when you buy a ticket (as they offer trip insurance).

Customers could choose to pay more in exchange for a guarantee they will never be involuntarily barred from a flight (or some lesser amount to ensure a minimum level of compensation to be involuntarily barred).

This has the nice feature that unlike regulation it would probably be a net win rather than a net cost. By more enabling more efficient price discrimination the price of a ticket without overbooking insurance would drop. Moreover, airlines would have greater predictability about the costs of overbooking and thus could more efficiently fill flights.

Fred Anderson writes:

I see two alternatives that no one seems to have discussed as yet. (Most probably because everyone else grasps their economic inferiority better than I do.) However . . .

United could have chartered an air taxi to ferry the flight crew to Louisville. It would have been perhaps double or triple the cost of displacing four passengers, but obviously far less disruptive.

United is headquartered in Chicago. Surely they have a corporate jet (or two) based there that could have been pressed into service.

And whatever happened to the past practice of selling standby seats -- discounted tickets that only guarantee passage if they have the empty space. Now this flight was fully booked, so that wouldn't have helped. But an airline could have a policy that the last 2, 3 or 4 seats could be sold as standbys -- callable if they needed them (as here they did).

Tom West writes:

A - the airlines could include one more box in the ticket purchase form: "if we need to bump you, how much do you want us to pay you?"

Well, I can see almost everyone putting $10,000 as a sort of lottery ticket - "Yay! I got bumped!".

To restrain that, I'd consider adding that your ticket price goes up $1 for every $100 of compensation that you want if you are bumped after the first $1,000.

Paul Bogle writes:

At Peter Gerdes

At the very least we should try to make the analogies equate. You said:

"Same principle as if you sell me a $5 ticket to watch the super bowl at your super awesome home theater system. If you decide your going out instead of hosting at the last minute I don't gain the right to break into your house."

The passenger was flying from the west coast to Louisville changing planes in Chicago. He flew the first leg, changed planes and was boarded on the second leg. As such I will modify your words to equate what happened as I see it.

"Same principle as if you sell me a $5 ticket to watch the super bowl pregame show at your super awesome home theater system, and you sell me a $5 ticket to watch the super bowl game at your super awesome home theater system. I watch the pregame but right before kickoff of the game you decide you're going to give my seat to what you consider a more valuable person. You first offer me $5 to leave, then $10 but when I refuse you call a sheriff's deputy you happen to have a close relationship with to come and explain the situation to me."

Dao may be acting like a prick or a child or a Rosa Parks depending on your point of view, but he paid for his seat in advance and was invited into his seat on that airplane by the airline.

Paul Bogle writes:

Can someone explain to me how:

"The Department of Transportation limits the compensation airlines are required to provide to involuntarily bumped passengers to 4 times the fare to a maximum of $1350."

is any different in effect to:

"The Department of Transportation limits the fare airlines are required to charge to a maximum of $1350"

This is a figleaf, not a regulation.

Chris L writes:

I'm not quite sure what you're asking but the $1350 limit is in addition to the cost of the ticket. So if your ticket costs $5000 and you are bumped more than 2 hours you can get $6350 in cash.

I've been trying to think of why there is a limit and the only thing I can think of is it's there to appease the airlines to give them greater power during the voluntary bumping negotiations. It probably was put in due to some lobbying from airline groups or something.

Frank Sappell writes:

Why would the DOT impose a limit on compensation? Does the federal government have a legitimate interest in interfering with a transaction between two private parties? The purpose of this limit is to contain the airlines' financial liability, should they lose the gamble that no-show passengers won't outnumber the overbooked ones. The amount of overbooking is a business decision, a classic risk:reward judgment that is carefully and cunningly calculated. Airlines will occasionally make the wrong call (not often, fortunately). When this happens, the regulators have no role in protecting the carriers from the consequences of their own bad judgment.

When overbooking occurs, the airline has already been paid in full for every seat it can provide; now it wants to be paid for seats it cannot provide. Even when sufficient passengers take the incentive bait, the airline's morality is exposed: it would happily inconvenience its paying passengers, for the sake of selling what isn't theirs to sell.

Hazel Meade writes:

Just in case this isn't clear to everyone, when a passenger gets "bumped", he still gets a seat on a flight, in addition to the compensation offered. It will simply be a different flight, later in the day or the next day (in which case the airline typically offers to pay for a hotel for the night).
This isn't a matter of the airline taking his money and then not delivering the goods. This is more like delivering the goods *late*, while offering store credit to compensate the customer for the inconvenience. I've seen one case (just from my flying experience) where overbooking cascaded to the next flight, but the airline tends to give the bumped passenger priority and offer compensation to someone else.

Again, we can argue about contract terms but ultimately United gets to decide who flies on it's planes. Airlines can and have kicked people off flights for all sorts of reasons not limited to overbooking. They can throw you out because they don't like the shirt you have on, or because you showed up drunk, or because you aren't wearing shoes. It's their airplane and every contract has terms for breaking it, so they can always break it and pay the cost.

Comments for this entry have been closed
Return to top