Michael Rosen reviews The Blind Side, a book on college and pro football by Michael Lewis. Rosen uncovers the reason why blind-side (usually left) offensive tackles are so important in football, and why they are so well paid.
They are valuable because they protect teams’ franchise players, their (usually right-handed) quarterbacks from sacks, fumbles, and most importantly, injury. They are paid well because there is an unregulated market for their services, and unregulated markets generally work very well.
It is reported that the top five offensive left tackles were paid an average of $7.5 million per year in 2005, second only to quarterbacks at $11.9 million per year. Rosen credits Lewis with noticing that it was free agency that enabled offensive linemen to finally be paid in accordance with their market value:
There was no new data to enable NFL front offices to value left tackles—or any offensive lineman—more precisely. The only thing that happened is that the market was allowed to function. And the market assigned a radically higher value to the left tackle than had the old pre-market football culture.
Lewis is spot-on about free agency. An unregulated market for the services of blind-side offensive lineman requires teams to compete. Competition appears to have driven prices up, which economic theory says competition does not do. But the theory is not wrong; it’s the analysis that’s superficial. Before free agency, teams captured and kept to themselves the value of blind side offensive tackles. What competition did is it transfered this value (what economists call “rent”) from teams to players.
Either Lewis or Rosen (or perhaps both) leave the misimpression that no new data became available after free agency to value blind-side tackles. “These guys don’t even have statistics that measure their success; the only relevant stats are negative – penalties incurred and sacks yielded.”
This cannot be true. Without valid, reliable and meaningful data, agents for the most valuable blind-side tackles would not have been able to prove their clients’ worth. But they did. Without such data, general managers would not have been able to discern which tackles were worth a lot more money and which were not. But they could.
If the necessary data did not exist prior to free agency, player agents made the necessary investments to collect them on behalf of their clients. At the same time, NFL general managers had to analyze these data, and probably develop alternative data of their own, to develop their own value rankings and counter what they thought were unsupportable agent claims.
Note that each side in this brand new market transaction — player agents and NFL general managers — had strong incentives to collect self-serving data, and to represent all data in ways that favored their own negotiating positions. But each side also had equivalently powerful incentives to equip themselves to rebut the claims of their negotiating opponents, and that means being able to rip apart analytically dubious and false claims.
This new market had no neutral source of valid and reliable data. The government surely didn’t collect and disseminate data on the contributions of blind side offensive tackles. Yet the data were produced. Interested parties, each with an intense conflict of interest, competed to discern which facts would persuade their negotiating opponent of the correctness of their case. Deceitfulness could be practiced by either side, but only at the heavy price of a loss of trust. The negotiation of NFL player salaries is a series of repeated transactions in which deceit causes distrust, and loss of trust translates into lost value for both players and teams. One can expect that it has been severely punished by the market, and without regulation. Today the market for NFL players, including blind-side offensive tackles, exists and apparently functions rather well. That means both sides have data — and quite likely, plenty of it — to guide private decisions in the market for football players.
In regulatory decision making the typical situation is very different. First, parties to the decision do not have equal bargaining power. It is the government agency that has the power to decide. That’s neither surprising nor troubling, for that’s what regulatory agencies do — they make decisions given the terms of their statutory authority.
It is the second difference that poses a serious problem. Regulatory agencies also are monopsonists (i.e., sole buyers) in the market for data that will be used to support their decisions — data such as risk assessments and regulatory impact analyses. Monopsony power enables many agencies to also be monopolists (i.e., sole producers) of risk assessments and regulatory impact analyses. It is hard to design a market with more examples of market failure in one place.
There is an inherent conflict between the role of analyst and the role of decision maker. The analyst seeks to enable informed decision making but is agnostic about policy choice. The regulator seeks to make policy choices but is agnostic about what analysis might be needed to justify them. Giving analysts the authority to make decisions invites them to hide their preferences within their analyses. Giving decision makers the authority to perform analysis invites them to direct that analyses support their decisions. Much of the conflict and controversy over risk assessment and benefit-cost analysis can be pinned on the fact that regulatory agencies perform both functions.Conflict will persist until the roles are separated.
This is analogous to the market for NFL players before free agency. General managers had essentially unfettered discretion to decide how much players were worth. They could prevent players from negotiating with other teams, thus eliminating competition for players’ services. When they made a salary offer, a player could take it or leave it. They controlled both the decision and what information was relevant to making it.