Why do senior citizens get so many discounts? A lot of it is because they have the time to shop for bargains — so if you don’t give them a bargain, they’ll find someone else who will.
We talked about this and other forms of discounting (or, in economic jargon, “price discrimination”) in my Principles class just last week, emphasizing that it does no good to hand out discounts willy-nilly; instead you want to target them to the most price sensitive customers. That’s why you sometimes have to jump through hoops (like filling out a rebate form) to get your discount — the customers who are motivated to fill out a rebate form tend to be exactly the same customers who are most likely to look elsewhere for a good price.
We talked too about how the airlines have always strived to separate business travelers from leisure travelers so they can charge the business travelers more and the leisure travelers less — the leisure travelers being more likely to take the train (or just stay home) if they don’t get a bargain. The problem, though, is that it’s hard to tell who’s a business traveler and who’s a leisure traveler. So historically, there have been devices like discounts for those who stay over a Saturday night, which is something a business traveler is unlikely to want to do.
Now I can go back to my students and tell them something about the value of staying awake in their economics classes. Because someone who’d obviously absorbed this lesson well has started a new site called GetGoing that takes this idea and runs with it. Here’s how it works: You book two conflicting itineraries, say a trip to San Francisco and a trip to Atlanta on the same date. You are quoted prices that are typically about half what you’d get elsewhere. You agree to fly. And then it books one of the trips, chosen randomly.
Why is this profitable for the airlines? Because no business traveler is willing to risk landing in Atlanta when he’s got a client to meet in San Francisco. So the site can use discounts to lure lots of new vacation travelers without worrying that high-paying business travelers will try to claim those discounts. They even explain all this, right on the front page.
This really is a beautiful application of some pretty elementary economic theory to what I expect will be a highly profitable enterprise. I only regret that my timing was off in two senses: I learned about this a little too late to incorporate it into my Principles lecture, and a little too late to beat Alex Tabarrok to the punch.
Holy crap, that’s the most brilliant thing I’ve heard in a long while. My immediate question: are we making the pie bigger or redistributing it by inventing this method?
My first take is that we have to be making the pie bigger, since after all, the customer and the airline both are participating (I assume the airline companies are cooperating in this enterprise). Plus, there’s apparently some left over for the originator of the site, so there must have been efficiency gains somewhere along the line.
On the other hand, I don’t seem to recall price discrimination as increasing efficiency except in the case of monopoly/duopoly etc.
It’s pretty late, so I’ll think about it more, just wanted to post my thoughts now to hear people’s reactions.
My first thought was that you could game it by finding someone who has a business trip planned for the same time, and then entering both itineraries and splitting the savings for that one discounted ticket. But I guess they get your name when you place the order, and don’t let you transfer them.
Airlines would like to fly every plane 100% full, even if this requires selling some seats very cheaply. If this scheme results in more full planes, then it reduces the cost of flying for everyone including business travelers.
Wow! …. and they don’t actually have to choose randomly, do they?
If you have multiple clients across the USA you can use this to book one random itinerary and pay full fare on another and still save money.
Step 1: Ask Client 1 & Client 2 whether Date A or B would be okay.
Step 2: Book A or B
Step 3: Book destination that was not chosen.
Step 4: Confirm appointments with clients.
More elaborate schemes are also possible.
That said: this will work only with those clients who are not sensitive to a change in date and whom you could offer a discount. I doubt there is much business travel that works this way.
Don’t you need monopoly pricing power (or something close to it) to get price discrimination to work?
I know that Russ Roberts and John Lott wrote a paper claiming that people think they see price discrimination everywhere while in fact it’s quite unusual.
It is good that they are so open about what they are doing – people may be distrustful of price discrimination where it is hidden, such a supermarkets hiding the bargains. This can only work if the customers that pay the high price do not mind too much.
What I found slightly surprising was that they have 8 patents to protect this. A quick look shows this as one of them (I think).
http://worldwide.espacenet.com/publicationDetails/biblio?CC=US&NR=2008133320A1&KC=A1&FT=D&ND=3&date=20080605&DB=EPODOC&locale=en_EP
Gluhovsky (one of the co-founders) seems to have authored a great many patents, for many of which Yahoo is the applicant. Given the simple way this scheme *seems* to work, can 8 patents really protect the business? Are these sorts of things what patents should be for?
I like this.
Wondering if it’s gameable by selecting 2 destinations that are close to each other? Or if you wanted to take a trip down the north east corridor. You book a flight, to boston or washington. If you land in boston you take your trip the way down, if you land in washington you travel north.
@6
CC, a good definition of price discrimination due to Stigler is that the ratio of price to marginal cost differs over similar goods. So what you need as firm is at the very least a small degree of market power, where market power is defined as the ability to raise prices above marginal costs (profitably). This is necessary, it is however not sufficient.
To illustrate, just imagine that you’re an airline company offering economy class and business class seating, and a competitor enters offering only business class seating but at a substantially reduced rate. In this case you might have market power to raise prices profitably above marginal costs without price-discrimination, but once you set the wedge between economy and business this is sufficient incentive to trigger entry (Note: I don’t know whether this holds in the airline industry, it’s just the first example that came to mind).
In addition, there are other reasons why price-discrimination might not work (e.g. Coase-conjecture).
> “You are quoted prices that are typically about half what you’d get elsewhere.”
That would be great if it were true, but it’s not. When I looked at it, their price on the trip I was interested in (NYC to London) was more than 10% *higher* than the best price for the same trip on Kayak.com. (Possibly because Kayak was clever about combining sub-trips on different carriers, whereas GetGoing was only showing direct flights.) Other trips did seem to be discounted a small amount compared to what you find via Kayak or Hipmunk or whatever but not very much – maybe 10% to 20% off. If you just browse the site looking at roundtrips to Europe, even the amount of savings they *claim* to be giving you is pretty small.
Where did you get the claim “typically about half” from? Can you find ANY trips that show up on the A screen with quotes that appear to be “about half off”? (I couldn’t, but maybe I’m picking the wrong regions or the wrong time intervals…)
Glen @10
And their “price guarantee” is a complete joke. You have to find a lower price for the identical itinerary on the same airline on the same flight # in the same class within 24 hours, and they have to verify it. Why didn’t they add the same seat numbers?
What if Farnsworth McCrankypants objects to people flying to random destinations?
Ken B wins the thread.
You explanation is wrong where it says, “chosen randomly”. The choice is almost certainly not random. The web page does not claim that it is random.
12 – it is a logical extension; Farnsworth knows the opportunistic leisure traveler is more likely to watch porn in his hotel room. (Maybe extension was a poor word choice.)
From the original post:
“So the site can use discounts to lure lots of new vacation travelers without worrying that high-paying business travelers will try to claim those discounts.”
This doesn’t quite make sense. “The site” referred to here is GetGoing. The website states that GetGoing is a travel agent – and like any travel agent they buy seats from the airline on behalf of an individual.
The airline is selling seats to a business, it’s just that the business (GetGoing) is only buying and on-selling the non-changeable, non-refundable, non-transferable fares (which are usually the cheapest). Why would GetGoing care if their customers are a business or leisure travelers?
@Roger #14 –
The site says they flip a coin to select your final itinerary:
https://www.getgoing.com/support/booking-your-trip/#q48
You’re right, that page says that they flip a coin. They obviously do not do that. It would be more profitable to choose the flight that is less filled.
@Roger – I suspect GetGoing will choose the fare offering them the highest margin.
Assume the choice is random. Then can’t you game it by just going through the booking process until you get the flight you want?
(I’d check it myself, but for some reason the site is failing to load for me.)
Keith: You commit in advance to buying whichever flight you’re randomly assigned.
Steve: OK, fair enough. And I guess there must not be any refunds for cancellations either?
A variation on #5:
In the (limited) number of cases in which a firm has more than one individual who can perform the task related to the visit (sales presentation, customer service, whatever) and for which the firm need not specify the name of an individual who will be coming (e.g., “a sales representative” or “a customer service representative”, etc.), only one date need be agreed to by both clients (or in the case of an intra-firm meeting, one date agreed to by both locations).
Step 1: Ask Client 1 & Client 2 (or intra-company locations 1 & 2) whether a specified date would be okay.
Step 2: Book destination chosen by GetGoing.
Step 3: Book elsewhere for destination NOT chosen by GetGoing.
Yes, again, this would be a limited number of cases. I assume that most business trips require specific individuals to travel, and I assume these tickets are not transferable even within a firm.
@Roger 18
Not so clear! That would open them up to manipulation. It’s the old game theory point: sometimes flipping a coin is the best way to decide.
Just to amend my #19, when I say “for which the firm need not specify the name of an individual who will be coming”, I mean specify prior to booking the flight. The person could be specified to the visitee after booking the flight (i.e., before the visit), of course.
Here’s another example for your class:
From years of experience, direct return flights from
Japan to Germany are around 50% more
expensive than flights from Germany to Japan (sigh!).
I always thought this was racism, but now it seems more
reasonable that it is because Japanese pay what they
are asked, while Germans spend days comparing prices….
The business model appears based on selling low-cost, low-margin fares in a market with strong competition from other vendors – yet GetGoing attracted significant venture capital and holds a number of patents. Why? I found this paragraph in a report interesting:
“GetGoing shares its extensive data with participating airlines, giving airlines additional tools to analyze what would work best for their routes and inventory. GetGoing also uses both historical data from the airlines as well as customer-provided data to personalize destination recommendations for travelers.
http://www.tnooz.com/2013/03/06/news/move-over-priceline-and-hotwire-getgoing-comes-out-of-beta-with-a-fresh-spin-on-opaque-discounts/#rjksaX5jqUHpHWPs.99
Most users are signed up through FaceBook. GetGoing can access their FB data and that of their linked friends. What if the ‘novelty’ is actually new algorithms to mine the FB data, and link it to travel behavior? And the ‘up to 40% off’ and ‘Pick two, Get One’ are clever marketing strategies to get as many people as possible signed up, even if they don’t purchase anything? Participating airlines could be getting extremely valuable data for the cost of a few cheap fares. Yeah, totally brilliant.