February 11, 2010
For the past 20 years, Forbes Magazine (and previously Financial World Magazine) has put out its annual estimates of the thirty MLB teams' revenue, income, and value every April. For obvious reasons, these numbers can be incredibly valuable for anybody outside the industry; they are the only full set of financial estimates that are publicly available, so by default they're our best option for any kind of economic analysis.
The question is, how much stock can we actually put into these numbers? We know that over time, their aggregate numbers have mirrored MLB's own numbers. (Part of that could be due to their editors backwards-engineering, of course.) But what about for individual teams? It's hard to evaluate their revenue estimates, since MLB holds on to this data with a death grip. But we do know how much teams sell for, so it's not that hard to compare Forbes' estimates to the real-world results. If their numbers are solid predictive indicators, we can not only forecast what teams will sell for, but also evaluate the final number.
For example, take some of the last few franchise sales. Maybe it's just me, but $570 million sounds like a lot for the Rangers, as does $500 million for the small(ish)-market Padres. Did you realize that these are the second- and third-highest prices in the history of the sport, behind only the Cubs sale last year? (The Red Sox were sold along with NESN and Fenway Park for $700 million, of which the team accounted for about $400 million, depending on who you ask.)
Now obviously, MLB's economics are constantly changing, and they're significantly different now than they were even five or six years ago. The worth of $500 million today isn't the same as $500 million in 2004, when the Dodgers, who are obviously far more valuable than the Rangers or Padres, were sold for around $370 million. But can we figure out if these deals really were actually overpriced? Hypothetically, the Forbes numbers could serve as a good benchmark, but we need to validate that. To do so, we'll compare them to all of the recent sale prices on record, which we have thanks to Rod Fort's indispensableSports Business Data Pages.
Altogether, there have been 33 franchise sales since Forbes started running its estimates in 1990, not counting sales of minority stakes and expansion fees. Compared to their Forbes valuations, there are some prices that are way over (Tom Hicks paid a 77-percent premium for the Rangers in 1998), and some that are way under (David Glass got a 21-percent discount on the Royals in 2000).
A side note on the sale prices: some, like the Red Sox and Cubs, had to be adjusted to only account for the team. Others needed to be extrapolated out from the price paid and the percentage bought-so, for example, Disney paid $140 million for 75 percent of the Angels in 1999, which valued the team at $187 million. Also, there were a couple sales that were not in Rod's list, because the sale prices aren't readily available. Those include the Diamondbacks' and Rays' sales in 2005.
The average difference is around eight percent, so on a very general level, if you want to know how much your team would sell for if it was put on the market, you could simply add that eight percent to their Forbes valuation. The problem is, there's no mode here. In fact, there are only four sales within even three points in either direction of that eight percent mean. Take a look at the chart below; there's a bit of a cluster around +30 percent, and again around -15, but nothing that indicates some sort of inherent trend. Even worse, Forbes only gets within 20 percent on 22 of the 33 sales; that's a pretty big margin of error, especially now that we're dealing with purchases of $500 million and more.
I tried to see if there were any characteristics that made teams systematically overvalued or undervalued, which would help explain the lack of order in that chart. In other words, maybe if a team had won a lot of games that previous season, or had been successful over the past three years, or had won the World Series recently, prospective buyers would be willing to pay more. Unfortunately, the highest r-squared number for any of these is .02, so none of these seem to have any kind of significant impact. (An r-squared coefficient tells you how much variation in one factor is the result of another factor. Anything below .30 or .40 usually isn't telling you a whole lot.)
I also looked at MLB's financial performance, as well as the health of the U.S. economy as a whole, to see if either of those made any difference. Obviously, industry revenues and total GDP are going to have a pretty significant impact on sale prices, but that doesn't help us predict whether a given team is going to fetch more than its Forbes valuation. So instead, I looked at MLB's one-year, three-year, and five-year growth rates, in terms of revenue, along with the same growth rates for GDP. Still not much doing, though: None of these variables returned an r-squared value higher than .10, leaving us back at square one.
Now, let me throw out here the fact that I don't think the valuations are total bull. The answer is probably something as simple as the number of bidders for each team, which isn't an easy thing to predict, especially when the greater economy seems to have no impact whatsoever on demand. (Just look at the recent sale prices in this horrible atmosphere.) Plus, you never know when you'll have a bidder who simply wants the team more than anybody else, above and beyond any rational valuation, or an owner who has his back up against the wall and needs to get out at any cost.
But that doesn't really help us here. Without any actual empirical evidence to back up their methods, it's going to be hard to use them as a major piece of any serious research projects. That doesn't mean they're worthless. I think they can still be used on a very, very general level, and I'll still go over them in detail when they come out in April. But for anything more rigorous, these valuations are pretty much dead on arrival.