Monday, September 14, 2009

Dallas Stars and Dynamic Demand Models

I found this story the other day on the 'Bizjournals' network. Apparently the Dallas Stars (NHL) have an advanced ticket pricing system that can be updated daily based on things like standings, the opposing team, day of the week, etc. The topic is something I am very interested in: determing dynamic models of demand for sports leagues and how they could be beneficial to both fans and teams as well as league policy. I think this is an important advancement in non-season ticket leagues (basically all but the NFL); however, I'm not sure it's that new. Teams have been known to vary their prices based on the visiting team, as well as offer deals to fans when attendance is not so great--for example, special promotions. The idea is that the team now knows when demand for their product shifts in and out. Perhaps this easily updatable ticket pricing model is more advanced than past ones. I don't really know. I'm curious to see if after the season (or multiple seasons) we could see a change in Stars single game ticket buying behavior.

But that's not really my point here. The team claims that this type of pricing, "offers great benefits to the team and its fans". I don't totally disagree with that. The first part is definitely true: a pricing model like this could be very beneficial to the team. The reason for implementing something like this is not only to find a single ticket price--which could be lower and allow more fans in the gate, or higher, which could maximize profits for the team, especially in a relatively non-competitive atmosphere where the team is a price-setter of sorts--but also so that they have the ability to price discriminate. That's what airlines do when you go to their website and you see the different levels of prices. If given the choice, I take Southwest's "Wanna Get Away" rates. It's the same damn seats for about 1/6th of the price. But as you get closer to the day of the flight, prices increase dramatically, and you might be stuck with "Business Select". They know they can do this because business trips are almost invariably planned closer to the flight than a family vacation and thus are demanded very highly right away.

Airline pricing systems are obviously more complicated than just that--in fact, airline pricing is the most advanced in pretty much any industry--but that's probably the most telling example of what these models do. They find the groups that are willing to pay more, and sell tickets for more to them, ultimately cutting into consumer surplus--or the difference between what you paid and what you were willing to pay for a ticket. That's not necessarily a benefit to the fan since the model will attempt to charge each fan the maximum they are willing to pay. As I said before, the Stars can also gauge when their demand function shifts out (for playing well, playing a popular opponenent, or whatever reason) and raise prices based on this more quickly than they were able to beforehand. More on that later.

So how does this possibly help the fans? Well the other effect the pricing model and price discrimination can have is cut into deadweight loss--or the tickets that could have been sold at a lower price, but were only offered at a 'single' price that maximized profits for the team when they were less able to gauge variation in demand among different people. Deadweight loss happens when pricing is not truly competitive or output is artificially limited. A company (monopoly) ensures the maximization of profits from a single price that is set (this is a generalization in terms of ticket pricing), which does not allow for everyone to purchase the product (assuming they cannot perfectly price discriminate). I'd argue it is the case here to a certain extent given that the Stars are the only NHL team in Dallas and comparable entertainment venues aren't perfect substitutes. There are competitors: basketball, football, and other entertainment. However, we're going to assume that the Stars can sell more seats at essentially $0 marginal cost since the seats are already built. Setting a single ticket price significantly above the marginal cost will limit output--or the number people who can come to the game if the Stars charged a lower price, but still above their cost.

Cutting into the deadweight loss is good for both fans and the team. They can do this by lowering all ticket prices, or by offering lower prices only to people that weren't attending because they didn't value the tickets so highly. The latter allows the teams and fans can come closer to reaching an efficient outcome: the team makes more money AND more fans are happy because more of them were able to attend within their budget. Fans that were already there, however, may not be getting as good of a deal when price discrimination takes place over setting a single low price.

I'd like to state here that I'm assuming the number of seats at the stadium is the maximum that could be built. Or in other words, the marginal cost of building new seats is higher than the amount team could make off of them. We'll just say the 'last fan' values the 'last seat available' too low to consider build a seat for the next fan.

Since the team already sold its season tickets to 'high-demand' fans, they don't have to worry about giving those 'high-demand' fans tickets too cheaply. They likely already have an idea at how to suck as much money out of these fans as possible. So these fans are not necessarily better off--but not necessarily worse off. And everyone that really wants single game tickets will probably buy them at the preseason price to ensure they get to attend a game. So they're not really better or worse off either. The rest of the fans are a 'mystery' (kind of). This model reduces the mysteriousness of these fans in the same way the airline pricing does. The aim here is to target different groups of fans that are still willing to pay more than $0 to attend the game, with each group placing a different value on that attendance. Since there is no real marginal cost--a generalization--to putting butts in the seats, if the model is efficient, the stadium will sell out every game and each fan will more or less pay exactly the full amount they value that ticket. But is the 'perfect' gauge of demand better for fans in all cases? Not necessarily. While more fans get to attend those games, the in-season ticket buyers that were already in attendance--and valued the ticket higher than the price--are likely going to have to pay more to go.

This model also does something else: it estimates when the value the same people place on attending changes in some way. This refers to a "shift" in demand: when everyone suddenly places more (less) value on attending a game. This could happen because of winning, the quality of opponent coming to town, or a number of things of that sort. A pure shift out in demand will cause the team to raise ticket prices to (likely) all of the groups.

Because of the limit on the number of seats, the stadium or arena could have already been selling out at a lower price implemented before the team could quickly gauge the demand shift. In this case, all the advanced demand model will do is tell the Stars to raise their ticket prices to fans. The advanced model allows the team to ensure they're not charging too low of a price to anyone.

Going from the unadvanced to the advanced model in a sellout situation--and assuming no change in demand--no more fans are getting tickets. This reduces the consumer surplus and only the consumer surplus. The only thing happening here is that certain fans are having to pay higher prices for their tickets. The same thing happens if there is a shift in demand. Higher prices to everyone that suddenly values the tickets more. While more 'efficient' economically, I wouldn't consider that to be a benefit to the fan.

So do we see "great benefits to the team and its fans"? Well, it depends. If the team wasn't selling out, and now is, new fans could benefit by getting a chance to see the team for cheaper. That's good for both the team (who marginally spends $0 on getting new butts into the seat) and the new fan (who didn't get to go before, despite having some demand for doing so). However, single ticket fans that were originally getting a 'great deal' aren't necessarily getting that anymore. So while it's more efficient, it's not necessarily 'better' for the fans.

My gut here tells me the model will gauge more about the ‘shift’ in demand, rather than the different groups to price discriminate toward. If that is the case, we still may have pricing that limits output (attendance), despite the fact that the cost to fill the additional seats is $0. Perfectly discriminating is very difficult and I doubt the model will do it perfectly. In the end, I think there may be both increased attendance from those who value tickets lower, with increased prices to certain buyers that the team can pinpoint. Whether or not you feel this is better as a fan likely depends on which of those 2 groups you consider yourself part of.

No comments:

Post a Comment