Game Theory, Anyone?

Turbulent..  Challenging.  Unprecedented.  Once-in-a-career event.  Paralyzing.  Opportunity. 

However you want to characterize the period we’ve experienced and are still working our way through and out of–and shall be, I predict, for a few more years–if it has served a salutary purpose, and it has, it’s been opening firms’ senior leadership’s minds to the possibility of “thinking different.”

Welcome to Game Theory.

Game theory, codified it not invented by John von Neumann and Oskar Morgenstern in 1944 (Theory of Games and Economic Behavior), has grown to encompass the analysis of interactions between individual actors in complex socioeconomic contexts which tend to resemble markets:

• Interactions among the players are repetitive; that is to say, it’s not a one-time only encounter or a sort of sudden-death overtime.  This is important because it introduces the notion of maintaining and enhancing one’s reputation.  Scorched-earth tactics and burning bridges are, shall we say, suboptimal.
• Each actor is presumed to be rational, at least insofar as they can see their own self-interest–
• But their own self-interest anticipates others’ reactions to their own choices, decisions, and “moves.”

Grossly oversimplified presentations of game theory–more by way of caricatures than presentations–have become standard fodder for MBA courses, typically in the form of the classic “prisoner’s dilemma,” with unrealistic but theoretic-model-friendly assumptions such as: the prisoners can’t communicate; the game is never repeated; each can give only one answer at one point in time, etc.

Needless to say, the real world involves immeasurably more dynamic,more multi-player, and more protracted in time, considerations than the textbook prisoner’s dilemma. 

So what use can game theory possibly be?

Our reliable friends at McKinsey have attempted to answer this question, in “Making Game Theory Work for Managers,” which advertises itself as nothing less than an attempt to “generate answers representing the best compromise between risks and opportunities in all likely futures.”

How successful are they?

Here are some of the dilemmas faced in trying to adapt theoretical game theory to the senior leaders’ real-world role:

• Striking the appropriate balance between simplification of a problem to make it manageable vs. retaining enough complexity so that it’s relevant.
• The extremely detail-oriented nature of any particular hypothetical exercise in game theory –our good professors call this “sensitivity to initial conditions.”
• The preference of theory to generate a single monolithic predicted outcome rather than an array of more and less probable, more and less favorable, possibilities. 

(Digression:  If I were asked the classic nasty/aggressive interview question, “What’s your greatest fault?,” and had been administered truth serum, I suspect I’d blurt out that I don’t like to state the obvious.  Because….it’s obvious.  I much prefer to dwell in the land of nuance and greys rather than black and white.  Of course, this is a signal failing if you cannot assume, as I do but is often not the case at all, that your interlocutor shares the same premises you do as to exactly what’s “obvious” and what isn’t.)

The new and more dynamic model discussed in the McKinsey piece claims to improve upon the artificially constrained textbook model as follows:

 Instead of predicting a single outcome, with all factors balanced, the model first generates a narrow set of strategic options that can be adjusted to account for changes in various assumptions. Instead of solving an individual game, the model automatically involves a sequence of several games, allowing players to adjust their actions after each of them, and finds the best path for different combinations of factors. As one result, it supports executive decisions realistically by presenting managers with the advantages and disadvantages of the strategic options that remain at each stage of the progression. In a second step, the model finds the “best robust option,” considering its upside potential and downside risks under all likely scenarios, assumptions, and sensitivities as time elapses. This approach is different from attempts to look for equilibrium in an artificially simplified world.

Are you thinking, about now, that these are generalizations that have little but platitudinous application to any issues you’re actually facing today?

In fact the authors essentially admit as much by saying that “The best way to understand the model is to examine it in action,” and proceed to present their case study of having worked with the deregulation of the European railway network.  Starting this very month (January 2010), cross-border passenger service will be fully open to competition in the EU.  Germany, Italy, Sweden, and the UK have expanded on that by opening long-distance domestic passenger rail service to competition as well.

The first lesson to take away from this is that every market is very much its own.  Every market, that is, is highly specific.  Context matters.  History matters.  (In the case of rail, of course, geography matters.)  So think about what follows not in terms of one-to-one correspondence with challenges you might be facing, but as illustrative of a way of thinking about moves you might make and competitors and clients might make, in turn. 

Indeed, if there is one single notion I’d like to implant in your thinking with this column, it’s the power of dynamic as opposed to static analysis. 

By that I simply mean that if you take Action X, the marketplace, clients, and your competition do not stand still.  In the military’s inimitable phrase, “The enemy gets a vote.”  (Dwight Eisenhower, or George Patton, or Douglas MacArthur [take your pick–attributions vary] said that “no battle plan survives its first encounter with the enemy.”)   Static analysis would assume the environment is, well, static. Guess again.

So, to more on the European passenger rail market.

What might entrants to this newly deregulated industry anticipate?

Price wars are certainly a possibility.  On the other hand, network effects are very important to customers in rail service.  Not can you get me point-to-point, but how thick and dense is your network?  The ability to get to an extremely wide variety of destinations on one carrier (presumably for a favorable price) is not to be gainsaid. 

The analysts posit four main avenues of attack for new entrants:

• Meet incumbents on their own terms, by providing nearly identical service.
• Attack, by providing more frequent or cheaper service.
• Specialize, with a niche service such as high frequency at peak hours.
• Or differentiate themselves with an offering focusing on, say, leisure travelers who are very price-sensitive, but who don’t care about cheaper, slower, older rolling stock, or conversely going for the high-end with premium, “business class only” high-speed luxury service.

As for the incumbents, they too have a range of options:

• Ignore the newcomers.
• Try to mimic their offerings and hope to prevail through greater brand-name recognition.
• Take the offensive by undercutting cheaper competitors on price, over-delivering vis-a-vis luxury competitors on service, and reinforcing networks and “hub and spoke” models.

Then there’s a third level of dynamic change going on. How is the market, overall, evolving?  Again, there’s an array of possibilities:

• Overall demand changes:  If rail service improves (in the eyes of travelers), car and plane trips will, relatively speaking, decline.  Trains will gain market share.
• Cost differentiation.  While newcomers may initially have cost advantages (fewer legacy costs), incumbents often enjoy economies of scale.  Which side of the balance beam prevails is highly context-specific.
• Network advantages.  Incumbents almost invariably have more mature and extensive networks (office platforms and practice areas).  This is difficult for newcomers to surmount unless customers demonstrate a preference for the boutique approach.
• And price sensitivity.  What, in economese, is the “price elasticitiy” of demand for your services?  You better hope that it’s very low indeed (that is to say, that clients are highly insensitive to rates and fees, and that they perceive your firm’s services as valuable with little regard to cost.

So what can we conclude?

Intriguingly, one of the most powerful and “robust” (as academics like to say) findings of the McKinsey study was this:

“When we run the European passenger rail model through an array of different situations, a critical factor appears to be the way demand reacts to liberalization. Will the new offerings seduce travelers to take trains rather than cars or jetliners, or will overall demand remain stagnant, leaving rail companies to battle for an unchanged pool of customers?”

Why do I highlight this?

Because we tend not to think this way.

But what if changes to the BigLaw business  model, including the possibility of increased demand for BigLaw services in lieu of substitutes, could actually increase our market share, as it were?  What are those “substitutes?”  In-house counsel, most obviously.  But also, and increasingly, outsourcing vendors located everywhere from downtown Manhattan to Bangalore and Fargo, North Dakota.  Why should our instinct be to run up the white flag in the face of this brave new competition?  We shouldn’t be so shy, or callow, or scared.

---

The basic message is clear:  Think about what you might do.  Then think about what other firms will do.  Then think about what clients will do. 

Repeat.