A Blog by Jonathan Low

 

Aug 5, 2018

Why Did Anyone Think MoviePass Could Have Worked?

The problem with calculating the 'average' cost of anything - like movie tickets - often ignores crucial data from submarkets. JL

Felix Salmon reports in Slate:

(The) expectation was that while MoviePass might find its first adopters in markets with $15 movie tickets, it would ultimately catch on nationwide, at which point the average price of a MoviePass ticket would converge toward the average price of a normal ticket. But that didn’t happen—partly because MoviePass died long before the convergence could occur, but also because the actual marginal attractiveness of MoviePass in urban areas was hugely greater than it was elsewhere.
Where did MoviePass go wrong?
The easy answer is: It was selling dollars for a dime, and you can’t do that for very long until you go bust. For a mere $10 a month, MoviePass members could see as many movies as they wanted to—which is exactly what they did. The company burned cash until it didn’t have any cash left to burn and fizzled out.
But in the spirit of Chesterton’s fence, it’s worth looking at the thesis behind MoviePass, and to try to make a distinction between the calculated risks that just didn’t pan out, on the one hand, and the crazy ideas that were never going to work, on the other.
What’s more, we have a great text to work from: MoviePass CEO Mitch Lowe’s February interview with Recode’s Peter Kafka. It’s recent enough that Kafka could ask all the obvious questions, but long enough ago that Lowe could still claim with a straight face that his business model was going to succeed. (Shares of Helios and Matheson Analytics, MoviePass’ parent company, were trading at a split-adjusted $1,185 per share at the time—well down from their brief high above $8,000, but vastly higher than the $0.80 at which they closed on Tuesday.)
Lowe explained succinctly how the service worked:
We send you a MasterCard debit card in the mail with your name on it. You decide you want to go to a movie. You find the movie, you find the theater, you find the showtime you want to go to. Then when you get within a hundred yards of the movie theater, you check in and magically that credit card works at that theater for about 30 minutes.
Reading between the lines, you can already see part of the business model: making it just hard enough to buy a ticket that very few people would buy a ticket and then not go see the movie, and even hard enough that a lot of people wouldn’t use MoviePass for big popular movies. (To see a movie that’s going to sell out, MoviePass users have to get to the theater hours before the screening to buy their tickets—and if it sells out before the day of the screening, they’re out of luck altogether.) This part of the MoviePass service seems to have worked as intended. MoviePass managed to save a bit of money when members bought tickets to blockbusters in advance. The company also didn’t suffer people buying lots of tickets to movies they never bothered to go see, or people using MoviePass’ money on anything other than movie tickets. (For instance, one possible hack would be to use a theater’s loyalty program to get a free ticket, or get a discounted ticket in some other manner, and then use the MoviePass debit card at the concession stand. By all accounts, MoviePass was good at preventing those kind of shenanigans.)
Then there was an exchange about MoviePass’ costs:
Kafka: In New York City, every movie ticket price is, I don’t know, 15 bucks.
Lowe: Yeah. It’s crazy.
Kafka: It’s crazily expensive.
Lowe: Yeah. Just one movie and you’ve saved money.
Kafka: Right. That’s the too good to be true part. Even the rest of the country, an average ticket is what, nine bucks?
Lowe: No, the average price across the country is $8.73.
Note how Lowe emphasizes that the average movie ticket in America is cheaper than Kafka thinks. (A whole 27 cents cheaper!*) Investors, take note: A MoviePass member who goes to one movie a month is profitable for the company.
This turns out to have been one of MoviePass’ clear errors. The average price of a movie ticket across the country might be $8.73, but that doesn’t mean that the average price that MoviePass paid was $8.73. After all, the more expensive movie tickets are, the more attractive MoviePass becomes—and unsurprisingly, MoviePass was always more popular in expensive urban markets than it was in rural areas.
Lowe’s expectation was that while MoviePass might naturally find its first adopters in markets with $15 movie tickets, it would ultimately catch on nationwide, at which point the average price of a MoviePass ticket would rapidly converge toward the average price of a normal ticket. But that didn’t happen—partly because MoviePass died long before the convergence could occur, but also because the actual marginal attractiveness of MoviePass in urban areas was hugely greater than it was elsewhere.
The big idea behind MoviePass was always that it would get people to start going out to the movies again, rather than staying in and watching Netflix. Lowe’s logic was simple: If you don’t have a MoviePass subscription but do have a Netflix subscription, the cost of seeing the movie is the cost of admission, while the cost of watching Netflix is zero. If you have a subscription to both, then the cost of both is zero, and you might well prefer the big-screen communal experience of moviegoing.
In reality, however, the calculus is much more complicated than that. For an urban childless millennial, the cost of seeing a movie with MoviePass might well be substantially cheaper than Netflix, once you compare the price of popcorn to the price of Seamless and the price of a Diet Coke to the price of the bottle of wine you’d inevitably drink in front of the TV if you stayed home. By going out to the movies, you’re actually saving money—and your liver.
For a suburban couple with kids, by contrast, going out to the movies is still vastly more expensive and logistically exhausting than staying in with Netflix, once you’ve dealt with the car and the parking and the sitter and the fact that the kids still need to be fed, which means that the marginal cost of feeding yourself at the same time is almost certainly smaller than the amount you’ll spend at the concession stand.
To put it another way: Yes, part of the reason that urban tickets cost more is simply a function of the cost of real estate. But another part of the reason is that in urban areas, the cost of the ticket is much more likely to be all or most of the total cost of going out to the movies. Whereas in the suburbs, tickets need to be cheap just because all the other costs are so high. Which in turn means that MoviePass defrayed a much higher proportion of the cost of urban moviegoing than it did of suburban moviegoing. That meant that MoviePass naturally became attractive primarily to city dwellers who racked up $15 a pop every time they used their card.
The idea that MoviePass would be particularly attractive to the consumers who bled the most money from the company has a name: adverse selection. And Lowe seems never to have heard of that concept:
Here’s the trick: 89 percent of American moviegoers only go to four or five movies a year. When they join MoviePass, they double their consumption and go to about 10 a year. That’s a little bit less than one a month. They balance out the 11 percent of the population that go 18 times before joining MoviePass and then after go three times a month. It works out. Over time, it actually works out to be about one movie per month per subscriber.
This makes perfect sense, mathematically—if 100 percent of American moviegoers decide to sign up for MoviePass. On the other hand, if the 11 percent who go three times a month sign up and the other 89 percent don’t, then the entire business model implodes. Naturally, the 11 percent will be the first moviegoers to sign up. Which means that MoviePass’ parent company needed to have deep enough pockets to sustain substantial losses while it was waiting for the other 89 percent to get the memo. It didn’t.
One problem seems to have been that Lowe was so excited about ancillary revenue that he took his eye off the core premise:
Our goal is to get to breakeven with the subscription and the cost of goods. Then we have all these different ways that we make your life better as a customer. We know how to market films to you. You know, the studios are incredibly inefficient the way they market small films. Over the last three weeks, we bought one in every 19 movie tickets in the country, but when we promote a film, we’re buying one in 10, so we’re lifting. These are for subjective $50 million box office films. The studios are paying us to be a more efficient marketer of films.
Sure, if you’re breaking even on the core movie-ticket product, then I’m sure you can make good money by getting movie studios to pay you to send your subscribers to their movie rather than someone else’s. But that’s a big if.
Similarly, if you’re breaking even on the core movie-ticket product, then you have a strong negotiating position with respect to the movie theaters. They know that you’re not going away, and that if you start directing your members to rival chains, that will mean they start losing profits. They will then have a strong incentive to start kicking back some of the revenue you’re sending them, not least because, by all accounts, MoviePass members spend more money at the concession stand than people who paid out-of-pocket for their ticket.
On the other hand, before you’ve found your financial footing, the chains have no incentive to embrace a model that’s likely to lead them to competing with each other for MoviePass favors. Lowe was under no illusions on that front:
If you think back when Orbitz and Travelocity, Expedia were founded, the airlines and the hotels wished they didn’t exist. They want that one-to-one relationship with customers. That’s exactly the way the big theaters look at us. They go, “Geez, we
one-to-one relationship. We don’t want a middleman.”
When Lowe started trying to wring concessions out of chains while he was losing an enormous amount of money, then, the chains could clearly see how weak his negotiating position was and sensibly refused to engage.
What’s more, Lowe knew early on that his initial projections were massively off-base.
When we sold half of the company to a data analytics company, in the contract we put in a clause where we would get a $2 million bonus if we hit 150,000 subscribers in 18 months. I thought, “Okay, that gives me about three or four months of padding.” I thought maybe it’d take us a year. We got there in two days. This is way beyond my expectations.
This might have been Lowe’s biggest mistake. He and his backers were looking forward to getting 150,000 subscribers in a year, and in reality they got there in two days. When a new product is hugely successful, managers invariably celebrate, as Lowe did here—but they should also take the opportunity to ask whether they got some of their calculations wrong, and if so, which ones.
One thing he failed to foresee was that, in the era of Reddit, when that many people flock to a single product in such a short amount of time, a community will naturally spring up around them. Lowe’s metaphor of choice was that MoviePass was like an all-you-can breakfast buffet: You might load up and go over the top for the first day or two, but ultimately you’d settle down to a more reasonable and sustainable consumption level.
But going to a lot of movies isn’t bad for you; it’s fun, especially if you do it with friends. If a whole group of friends has MoviePass memberships, it starts feeling silly for some subset not to go out to the movies on any given night. When you’re in your group-text app, a movie night becomes the zero-cost default option, cheaper and more pleasant than squeezing into a cramped apartment and ordering pizza. Far from getting tired, going to the movies becomes a habit. Even online communities count, if everybody is going to see a certain film and then talk about it. In a sense, MoviePass built an all-you-can-eat buffet that got better the more you used it, and the economic problems with such a buffet are pretty obvious.
Those economic problems became obvious to the stock market pretty early on, and ignoring market signals was Lowe’s final big mistake. Lowe was far too sanguine about losing money:
There are actually dozens and dozens of businesses like ours that invest in building a large subscriber base. Whether it’s Netflix that buys $8 billion of content a year—and believe me, they have to borrow the money to do it.
There’s a reason that Netflix can borrow billions of dollars a year—which is that it has a soaring share price. Bondholders, reasonably or not, have convinced themselves that if Netflix ever has trouble paying the interest on its debts, it can always issue new shares. A multibillion-dollar market capitalization gives companies a lot of options when it comes to raising cash. Lowe, by contrast, merely averred, unconvincingly, that “we have a deep-pocketed backer that is prepared to fund us all the way.”
The fact is that such backers are mythical beasts: They simply don’t exist. An investor will make a contrarian bet for a short time, but ultimately will always want that bet to start getting ratified by the markets. If after a few months no one else comes along for the ride, and instead everybody starts betting against you, then at some point you stop reaching into those pockets, no matter how deep they might be.
In order to keep the funding coming, Lowe needed to be able to point to a rising share price and a growing group of investors willing to fund his model. When he couldn’t do that, he was toast.

0 comments:

Post a Comment