Wednesday, August 13, 2008
Posted by Jeff Lipshaw
Bill Henderson cross-posted his usual massively insightful and deftly written analysis on "faculty free agency" over at ELS, and I posted an equally long comment over there. I repeat here some thoughts on the problem and the solutions (slightly edited).
1. I'm not sure that law firms are going to give us a better model for structuring faculty incentives for the following reasons. The real question is whether the institution created by the people involved in the institution is greater than the sum of the parts. The underlying assumption in Bill's analysis is that cooperation and coordination indeed do create value that none of the individuals alone can create. What we have is something of a Prisoner's Dilemma game: individual incentives suggest that law professors AND law partners maximize for themselves, but institutions (and perhaps the professors or partners themselves) would be better off if they cooperated. As we know, one way out of the Prisoner's Dilemma is repeat play until the participants learn to trust each other NOT to maximize individually.
2. For precisely the reasons Bill cites (the incommensurability of teaching and service), it is very difficult to demonstrate a payoff to professors that warrants cooperation even after repeat plays of the game. Law firms are a little better, but my intuition, based on long experience, is that the payoffs of cooperation are almost as difficult to perceive for the partners. My theory is that it's because, indeed, it really is very hard to build any such value. The brand "Skadden" or "Weil Gotshal" is so hard to build, and once built, so independent of the contribution of any single partner (collective action problem at work?), that it just doesn't take us anywhere.
3. In my experience (and intuition), complex business models will actually take us farther in terms of the analysis of the ideal cooperation. I don't want to confuse this analytically with the corporate model of organization, as my co-author Larry Ribstein would insist (see Ribstein & Lipshaw, Unincorporated Business Associations, 4th ed., coming to a bookstore near you in 2009), but let's use the model for ease of explanation. There's no doubt in the corporation, we are building a value proposition greater than the sum of the inputs, physical and human capital together, that expresses itself in a measurable output that integrates ALL of the inputs (contra law school for the reasons you point out). That is, the market value of the enterprise not only exists but is measured every day in terms of a share price (and if the corporation is public, you can actually see it change minute by minute!). Whether or not the systems actually work (I don't want to argue the CEO compensation issue here), the goal is to align the interests of the individuals in the organization with that unified measurement of value, and usually by tying compensation to increases in that value.
4. Compensation within a multi-divisional public company is an iterative process of balancing individual initiative with cooperative goals. It's really tough to get right. I don't think you do it by long-term contracts (there's an involuntary servitude aspect to this I'd want to think through). Compensation is the carrot; I think your idea of a contract is the stick. You can encourage cooperation by incentives - tying compensation to sticking around, for example, by vesting it over a period of time - but you still have the problem of the individual versus the collective. If you tie the compensation entirely to the group output, you run the risk of having your stars leave if they are performing but the rest of the company sucks. If you tie it to individual performance, and not the collective output, you get precisely the cooperation issue we are discussing.
My reaction generally was that, even in the corporation, algorithmic solutions did not work. You needed a spirit of cooperation that preceded the compensation, or individuals would nevertheless figure out a way of skewing things to favor their own interests over the collective. Despite all the best HR theory, we still had to deal with the "wooden nickel" problem by means of leadership, not analysis. What do I mean? Take corporate overhead allocations. Business unit leaders hated them because they were a cost to the business they could not control, and cost them money in terms of their own compensation. My position (as GC and hence one of those cost centers) was that focusing on reducing the cost was adding value, but that merely arguing the allocation was not. That is to say, trading costs between units in terms of allocation was merely trading wooden nickels.
To conclude, I have some skepticism over the possibility of a rule-based solution to the problem - rule based in the sense of writing a contract to deal with it or rule based in the sense of developing compensation algorithms. To continue on a theme that I hope is apparent in my writing, the solutions here have to do with something like leadership, and that has a way of not being reducible to rule-following.