An alert commenter points to a more recent study of baseball team performance.
Having a larger Gini coefficient (as you’d see in a stars-and-scrubs roster) is ever so slightly associated with better outcomes over the rest of the season. However, the effect wasn’t large enough to be statistically significant, so this analysis says a team should probably just be indifferent about which approach it uses to build a roster.
I believe that the study that Turchin cited looked at inequality in terms of salary, whereas this study appears to look at inequality in terms of players’ contributions to wins. In any case, this study finds the opposite result, although it is not statistically significant.
Even the original study cited by Turchin does not quite say what he claims it says. The abstract reads,
in the latter part of the 1990s and continuing into the 21st century, the greater the team payroll and the more equally this payroll is distributed among team members, the better the on-field performance of the team.
I interpret this as saying that teams with higher total salaries for players were winning more. Given an aggregate salary level, it was better to have it more evenly distributed among team members. But that might indicate that, other things equal, it was better to have a balanced roster than a stars-and-scrubs roster.
Turchin uses this one study, which he interprets as showing that unequal salaries per se cause poor performance, to argue that inequality will lead to social collapse. Seems like quite a stretch.
Baseball is effectively a game of summing up various individuals and getting your raw total. There isn’t much room for gains or losses due to cooperation, etc. Player X’s performance will have little to no effect on player y’s. Not sure baseball the best game to use to support or dispute this theory.
Would be interesting actually to look at individual performance of players across various teams.
See if they do better or worse when moving from teams that differ in Gini coefficient.
Baseball actually well suited for this given how isolated each player is, and how minimal in theory, the effects of teammates should be. (I suppose you might get harder pitches based on who’s batting around you but the environmental effects are far less than in any other major sport)
Meaning the same player going from Team A with Gini coefficient X, and then seeing how he does when switching to Team B with Gini coefficient Y.
As great as baseball stats are, you can only explain society so much. In general the best way to develop a team is still the Branch Rickey way and using Free Agency to plug holes in the team and pay enough keep a core group of players together. (So the team doesn’t develop a core group of players and they all leave.) No team can regularly win if they overpay a bunch of 32+ year veterans and sooner or later the team collapses.
Anyway, the Baseball team metaphor works better on the designing a firm versus society. I find the best firms have a combination of talent that grew up in the organization and plugging wholes with ‘Free Agents’.
Using baseball as the basis of an economic model that is supposed to generalize is a terrible, terrible idea, and if any economist wants to use the time period of the late-90s to present he has to account for several structural changes and trends. Important issues, in no particular order
1) Baseball is probably the worst sport to use as the basis of team comparisons. Relative to any other team sport, the contribution of “teamwork” is very limited. For the most part, player performance is context-independent, and player production can be considered fungible. There is nothing in baseball that compares to the unit-wide performance of an offensive line, a basketball defensive scheme, etc. Baseball is overwhelmingly about individual matchups, and the reason the sport was ahead of the others in quantifying player ability was precisely because it is so much more easily separable, identifiable, and discrete.
2) WAR as a measure has a high degree of skew among players (a distribution chart can be seen here: http://www.fangraphs.com/library/misc/war/). With a 25-man roster, the most likely reason for a team to have a low GINI as measured by WAR is not that they are ‘evenly balanced’ but that they are mediocrity-and-scrubs teams, i.e. just plain worse.
3) The contract structure of baseball is such that it does not become free market until 6 or 7 years of playing time, when a typical player is 29 or 30. Statistically, players peak at around age 27, meaning that for most players their best years will be past them when they are available as free agents. Of the aggregate WAR produced (which is a finite-sum game), less than half (I want to say ballpark ~40%) is produced by players who have accumulated enough service time to have ever been a free agent. Most of the stars in baseball are 1) being currently paid less than their free-market value and 2) have not yet had the ability to become a free agent. For example, of the top-20 position players last year as measured by WAR, only 2 are/were available on the free agent market this offseason, only 2 more have ever in their careers been available as free agents, and only 2 more have enough service time that they could have been free agents had they not signed extensions well-in-advance of free agency with their present teams. The numbers are similar for pitchers. The ability to have the stars in which to construct a stars-and-scrubs roster is vastly more dependent (on a league-wide basis) on a team’s ability to draft and develop them than it is to trade for or sign them in free-agency.
Major league sports teams aren’t very egalitarian in terms of pay (although league minimum salaries are typically approaching a half million, so nobody is poor), but they’re pretty egalitarian in terms of working conditions: e.g., stars and scrubs take showers together, they fly together, get similar hotel rooms, run laps together, etc.
Salary inequalities are probably less divisive than working condition inequalities.
Silicon Valley pushed hard for working condition egalitarianism. In an era when “the key to the executive washroom” was a standard joke in sitcoms about corporate executives, I can recall spending a day interviewing at Intel in 1982 and being told that nobody is allowed a private office. Even the famous founders Robert Noyce and Gordon Moore had to have cubicles. I can recall walking by the cubicle of either Noyce or Moore. Of course, it wasn’t exactly like everybody else’s cubicle, it was 30 feet by 20 feet, and when I stood on my toes to look inside, the founder appeared to have original Monets hanging from the grey fabric walls of his cubicle. But that’s not the point, the point is that even Noyce and Moore didn’t get offices.