Fixing the Game by Roger L. Martin

Sometimes the difference between a good book and an Aaargh! book is a single unexamined value. Fixing the Game by Roger L. Martin is an Aargh! book. The phrase that kept running through my head as I read was so close… so close.

The book is two things: an exceptionally clear and original analysis of the question of what ails modern capitalism, and an exceptionally woolly headed prescription for how to fix it. Unlike many books that are strong on analysis, the prescription isn’t bad because it is an anemic afterthought shoved into a last chapter (here, the prescription runs through the entire book, with a goodly fraction of the word count devoted to it). It is weak because its foundational assumptions about the psychology of capitalism are hopelessly idealistic.

That’s what makes the book so frustrating. It could have been so much more. Still the book retains a lot of its value because it is relatively easy to tease apart the parts colored by idealism from the parts that are not.

Fixing the Game is an early offering in what will undoubtedly turn into a whole what-ails-capitalism cottage industry.

Unlike the first wave of books that appeared shortly after the subprime meltdown, this book does not concern itself with investigative journalism or with blow-by-blow narration of the events that led up to the crisis. Instead, the book focuses on presenting an alternative theory of capitalism based on different assumptions.

The current theory, Martin argues (based on maximizing shareholder returns), has basically been proved false beyond all reasonable doubt. Fixes based on the current theory are doomed to fail in predictable ways.

The alternative theory is based on the idea of significantly decoupling the real market of business functioning from the expectations market (the stock market).

The narrative as a whole is based on a telling of the story of capitalism on two time scales: the decade of the 00’s (which spanned three significant financial scandals: the 2000 bubble collapse, the 2005 options backdating scandal and the subprime crisis) and  the post World War II half-century, which spanned a watershed shift in 1976 towards shareholder returns becoming the dominant performance metric for businesses.

The analysis is compelling, but the prescription, unfortunately, is both conceptually flawed and ideologically compromised.

Still, for a first book in what I hope will be a genre that spawns many more, it gets the conversation off to an excellent start.

Business as a Professionalized Sport

This is one book that can be forgiven its sports metaphor because it is so precise and so apt. If you are unfamiliar with the basics of the main American sports, you will unfortunately miss a lot of the nuanced arguments. This is not a generic metaphor where you can easily replace football with soccer and baseball with cricket.

Others have made the point Martin makes: that the main problem with capitalism is the manic focus on shareholder returns. Where the analysis breaks new ground is in explaining clearly how and why shareholder-returns is the problem, using an extended metaphor of business as a professionalized sport. Using this metaphor, and by holding up the NFL as an organization that manages its sport (American football) well,  the book illuminates an inherently murky subject bedeviled by far too much esoteric jargon.

The metaphor frames both professionalized sports and business as the interplay of two markets: the real market (games on the field or actual company performance) and the expectations market (the sports betting market and the stock market respectively).

The rest of the mapping falls out naturally. Executives become star players. The SEC and company boards become the governing forces. Customers become sports fans. Stock market speculators become gamblers. Boards become coaches. Owners become major shareholders.

The highlight of the analysis is the contrast between governance of football as practiced by the NFL and governance of business as practiced by the regulatory forces. Martin’s main claim is that much of the sustained performance of the NFL in governing American football well (as evidenced by its increasing popularity compared to basketball and baseball) can be attributed to just a handful of practices:

  1. Strict separation of the real market and the expectations market via prohibition of gambling on the part of players and coaches, and effective policing of the prohibition, to ensure that players pursue returns in the real market rather than the expectations market
  2. Continuous, deliberate evolution of the game rules on and off the field to keep game outcomes as unpredictable as possible. I’ll get to why this is a good thing later (Martin’s analysis is incomplete).
  3. An effective model of player free agency that maintains a healthy balance of power among players, owners and game administrators and salaries on earth (in contrast to the MLB).

By contrast, Martin argues, the game of capitalism (and sports other than football) violates all these three sound design principles for a sound regulatory environment.

  1. The real and expectations market are strongly and deliberately coupled via stock-based compensation for executives (and employees), whose performance is judged based on their ability to set and meet (i.e. “manage” under conditions of moral hazard) expectations in the expectations market. This would be like team captains offering spread estimates before games and apologizing for not meeting them, or trying to win by a certain margin.
  2. Knee-jerk upgrades to regulatory regimes via band-aid fixes, based on a false theory of capitalism and economic disease, implemented in the wake of crises, rather than proactive prevention based on a better theory of economic health.
  3. A very poor balance of power that favors executives over all other stakeholders (employees, stockholders, board members, customers), rather like star players in baseball, who earn outrageous salaries in relation to the value they add, compared to star NFL players.

If these seem like rather broad and tautological philosophical points, you’re going to be surprised by the elegance of some of the pieces of evidence marshaled in support of this analysis. For example, in the period 1983 – 1993, CEOs met stated earnings expectations 50% of the time. Between 94-97, a turbulent period widely regarded as less predictable, they hit expectations 70% of the time. Either leadership quality increased miraculously, or the CEOs had simply learned to manage expectations rather than realities (a tougher challenge).

Martin presents a wealth of both quantitative and anecdotal evidence showing in detail how expectations are actually managed. CEOs may be the conductors of these orchestras, but weak boards, craven analysts and greedy shareholders are equally complicit in this theater of absurdity. What makes the analysis particularly compelling is the detailed comparison to behaviors that are actually outlawed in the NFL (all varieties of throwing games to win in the gambling markets). As the saying goes, knowing how not to fail is not the same as knowing how to succeed. The NFL offers a compelling model of how to succeed.

If the capitalism game rewards executive behaviors that are somewhere between neutral and harmful, it also penalizes good behavior: companies that pay attention to customers and building up the value of actual assets see stagnant performance on the stock market (Apple is a recent  exception that is insufficiently analyzed in the book).

If you’re like me and don’t  have a detailed technical understanding of how stock markets work, the book is also invaluable for its exposition of a few key academic ideas that are central to the design of the modern game of capitalism (in particular the 1976 paper by Jensen and Meckling, The Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure, which led to the elevation of stock performance above all other metrics, and the 2005 paper by Erik Lie, On the timing of CEO stock option awards which precipitated the options back-dating scandal).

Overall the analysis is very persuasive and most importantly clear, thanks to the NFL analogy. Without the striking contrast of an existence proof for a system that actually works, the analysis might have been no more than a just-so story, easily countered by an alternative just-so story.

Problems with the Analysis

That said, given the ambition of the change agenda being proposed (nothing less than the re-invention of capitalism), the analysis is not nearly strong enough. In particular, the heavy dependence of the argument on the before/after comparison around the 1976 watershed year unnecessarily weakens the overall argument.

For example, one piece of evidence offered is that real average returns dropped from 7.5% for 1933-1976 to 6.5% for 1977-2010, with significantly increased volatility (both in everyday terms and in terms of the frequency of big bubble-bust cycles).

Even accepting these not-too-solid numbers as roughly right, causality is harder to determine than Martin admits. It is not clear that the rise of shareholder returns is entirely to blame. 1976 is also a very good watershed year for a lot of other things that changed.

  • American businesses suddenly encountered real competition for the first time since World War II for example, with many technology markets based on WWII innovations (automobiles, aircraft, radio, computing) maturing and turning into zero-sum mature market games.
  • The OPEC crisis in 1979 is another major factor that confounds the historical analysis. Oil is so fundamental to the industrial economy that significant events in the story of oil must be incorporated into any larger story.
  • The late 70s also marked a powerful transition from sloppy and inefficient operational cultures that had reigned for several decades, to lean and quality-focused cultures that have reigned (with their own problems) between 1980 and today.
  • This operational shift was accompanied by the birth of the modern strategy industry, also around 1979 (as Walter Kiechel documents in Lords of Strategy). While one subplot of the strategy story was devoted to managing share prices, there was a larger shift in management models as well. CEOs started actually “doing” strategy.
  • A few decades of middle-management Organization Men ruling the roost ended, and sociopath senior management types regained control, a transition that, for all its problems, was a positive one.
  • The era of lifetime employment began to crumble, with both employee and employer loyalty crumbing with it.

With so many things changing at once, a story that blames one factor for much of the malaise is suspiciously convenient. I do think the focus on shareholder returns was a major cause, but it is unlikely to have been  the only cause or even a primary cause. I suspect Jensen and Meckling merely codified operating principles that had already emerged in response to environmental shifts. The shareholder-value rulebook likely came after the actual subterranean shifts were already underway.

The restriction to two time scales (2000-2010 and 1933-2010 with a 1976 watershed) is also a source of weakness. The excesses of poorly regulated capitalism during the Gilded Age (which I am reading about right now) and the longer history of bubble phenomena, dating back to the South Sea Bubble of 1622, need to be comprehended in a truly credible version of the argument. Shareholder-returns-focus  seem like it might be the latest manifestation of a more fundamental force. A longer natural history of corporate performance metrics seems called for.

That said, most of these problems can be fixed with more words and research. The fundamental argument, appropriately qualified, is believable.

The deeper problem is with the proposed theoretical framework itself. In particular, some of the weaknesses arise from the simplistic characterization of the market as non-zero-sum and value producing (fan enjoyment and customer value respectively) and the expectations market as zero-sum and non-value-creating.

This isn’t really true or completely fair.

The so-called “real” market is non-zero-sum only during  expansion phases while a generation of new technologies (often birthed by war) matures. Innovation unfortunately is a punctuated equilibrium process rather than one that adds value steadily.

In mature markets, leading up to the next technological paradigm shift, the market behaves more like a zero-sum game where market share is won or lost driven by fluctuations in the distribution of natural resources, random fashion trends and advertising rather than fundamental increases in value.

It is  also unclear to me that a real-market entity is a good thing simply because it produces “real” things. Many sincere people work hard solving real problems and creating real products and services that customers use. Unfortunately, many of those products and services are arguably, bad for the world. From cigarettes to addictive junk food to television programming that panders and enslaves, the real market is hardly a cornucopia of humanity-elevating production. Indeed, some of these things are worse than hedge funds, which is saying a lot.

Equally, the zero-sum and negative-sum dynamics of the expectations market does play a legitimate role in driving the diffusion of certain kinds of information and creating an environment of Darwinian weeding out of weaknesses (or what we like to call creative destruction).

While that legitimate role does not excuse the existence of perverse incentive structures that lead to outsize, under-taxed profits (especially in pure expectations market games like hedge funds), the strong suggestion in the book that the real market is “good” and the expectation market “bad” is simply misguided.

Unlike sports betting, the stock market plays an actual information-theoretic governance role in the functioning of capitalism. Even hedge funds (for which Martin reserves his worst criticism), with a few rule changes can probably do some good through their apparently perverse amplification of market volatility.

The Prescription

Martin’s prescriptions are a curious mix of subconscious cynicism and conscious idealism.

On the one hand, we get careful and detailed accounts of exactly how the 2-20 model of hedge funds leads to cancerous dynamics, and how simple changes to those rules can make the game vastly healthier. Some of the ideas are quite radical (like banning normal options-based compensation packages) but supported by very cogent arguments.

But on the other hand we get the vacuous suggestions that company board members should view their work as public service and that the rules should be written to enable CEOs to lead “authentic” lives (apparently, while enjoying their fabulous wealth, they are secretly pining for “authenticity” and bemoaning their empty lives, beholden to investor meetings, and are to be pitied).

While many of the individual tactics are based on realistic psychology and ideas about checks and balances, removal of moral hazards and conflicts of interests, and strengthening of weak parties in a currently unequal balance of power, the big picture suggestions essentially amount to a forlorn appeal to higher motives and positive psychology principles.

The whole prescription rolls up into a surprisingly weak grand goal: to displace the formula “maximize shareholder value” with the equally problematic “maximize customer delight.”  Martin’s argument in favor of customer delight is based on one good reason and one bad one.

The good reason is his rather clever argument (which he explains at length) that customer delight is a “powerful” objective in that it subsumes other, weaker objectives.  If you pursue customer delight, he argues, you will automatically also grow shareholder value at a “reasonable” rate (“reasonable” is one of many weasel words in the prescription). Since optimizing a robust function of many variables tends to be difficult in complex systems, maximizing the most powerful single objective is the next best thing.

The problem with this argument of course is the one I have already mentioned: “customer delight” reveals nothing about social value. You could be a drug pusher delighting cocaine addicts.

I don’t have an alternative proposal. No single formulaic and static objective is going to work. What you need is a more powerful dialectic that keeps the the meaning and pursuit of economic value in a state of fluid contention among contending voices.

Much of the woolly headedness starts with a fundamental value-based assumption on page 94 and then runs through the book like a gigantic fault line:

The illegal and unethical behavior of business executives over the past few decades suggests that something is seriously out of whack in the corporate world. Assuming people would rather be ethical than unethical, how did we wind up with such pervasive unethical and illegal behavior?

Far from being a throwaway line, the entire prescription rests on this one assumption. In effect, Martin’s is a theory of executive behavior based on Douglas McGregor’s Theory Y: that left to themselves people will naturally behave well (as opposed to Theory X, that given a chance, people will lie, cheat and look for any old way to get ahead). You just have to remove the perverse conditions that make them behave badly, and perhaps offer them an uplifting sermon to get them started in a new direction.

This assumption, quite simply is not true, but that’s a longer story, going back to theories about “man in the state of nature” due to Rousseau and Hobbes, and involving more recent work in evolutionary psychology. I’ll talk about that story another day, when I review Francis Fukuyama’s The Origins of Political Order, but for now, let me just offer without proof, the assertion that humans are a mix of X and Y tendencies. Some have more of one tendency and some have more of the opposed tendency. Different environments confer advantages on people with different tendencies. Good governance systems are designed around the full range of likely human behaviors.

When you base an ambitious prescription on a Theory Y model of executive motivation, you necessarily end up with such non-solutions as promoting “authenticity,” shaky conflations of “real” with “non-zero sum” and an uncritical idolization of an idealized “real” market where real things are produced and executives nobly strive to delight customers and life improves for all.

Yes. There was a time when men were real men, women were real women and small furry creatures from Alpha Centauri were real small furry creatures from Alpha Centauri.

Is there a way to assemble the ideas in the book into a different prescription? One that is not based on notions of human perfectibility and shaky appeals to the nobler motives of executives who are doubtfully cast as “would rather be ethical” characters? I am not suggesting we should be developing systems based on assumptions of inherent corruptibility. Such systems tend to be too cautious, paranoid and costly.

I am suggesting that we think about systems based on realism rather than either idealism or cynicism.

A Different Dialectic for Capitalism

If an unreconstructed notion of “delighting customers” cannot fuel a richer dialectic for capitalism, where might we find a richer dialectic? To create a powerful dialectic, you need a collision of well-matched forces.

There are two strong clues in the analysis and NFL analogy.

The first is the idea that keeping the game unpredictable is central to good governance. This means making more games real contests. Organized sports become boring when one force consistently beats another (example, offenses consistently beating defenses or vice-versa) or when one player is allowed to accumulate too large an advantage. At a deeper level, when every game is a challenge, innovation rates increase.

The second is the idea that empowering of players via free-agency was central to creating this unpredictability in the NFL. You cannot create a system of rules that leads to mostly even and unpredictable contests if one side is able to accumulate significantly more resources than the other. A rule change can neutralize a specific idea like the West Coast Offense, but it cannot restrain a team stacked with more stars, and fueled by more money than the competition. Innovation stagnates because some teams lose hope and others win without trying.

Unpredictability is a time-honored goal in the design of sports. Thorstein Veblen noted, in The Theory of the Leisure Class, that the games of the leisure class evolve their rules to reduce outcomes to pure chance. Mechanisms like handicaps carry this logic to an extreme.

The purpose is not to create a contest where the best team or athlete wins, but one where the definition of “best” is kept fluid so that the audience has the most fun, due to the unpredictability. As a side-effect, the gambling markets remain inherently simple because there isn’t as much information for them to process: at most you have simple betting or spread betting. The gambling markets merely process superstitions, tacit outsider information, and other forms of information that are not of immediate value to players in the real market. Usable information is kept out of the expectations market to the extent possible, and people in possession of usable information are prevented from participating in the expectations market and encouraged to actually use it to improve performance.

This is a fundamentally sensible strategy. The best use of real information is improving the real game; feeding it back and improving real performance via innovation. Using real information to manipulate expectations is a tragic waste.

Chasing Unpredictability, Arming Free Agents

In the NFL, the game is changed when a tactical innovation provides makes a previously unpredictable element of the game predictable. Such an advantage is mostly used to win games, not to manipulate the gambling market. Rules are eventually changed to even the odds when such innovations occur. The spurs innovation and the game evolves faster than the gambling.

The fact that the stock market is a maze of incomprehensible regulation that changes with every new crisis, while the copyright and patent regimes have been evolving at a glacial pace, suggests that we are driving innovation in the wrong game.

To move the rapid innovation to the real market, and to keep the stock market game comprehensible and slowly evolving, you need a contest in the real market between forces that have fundamentally opposed objectives.

Martin’s prescription (and his proposed new model of capitalism) perpetuates a fundamental mistake: treating capitalism as a game between regulators (both central and board-level) and players — people with information and levers that allows them to manipulate both expectations markets and real markets.  To use the NFL analogy, this amounts to having the players and coaches fight the NFL governing body and team owners.

This isn’t a contest. The people with privileged information will always win. It cannot be made an unpredictable and even contest: those with privileged information will use that information to get ahead whatever the incentive structure.

You cannot neutralize an arms race with rules. You can only displace it with a different arms race.

So what you need instead is to displace the contest: make it a game between two parties with equal access to privileged information and fundamentally different motivations, who are forced to fight in the real market rather than the expectations market.

In the sports examples that Martin cites, this other party comprises individual players. Thoughtfully enabled free agency was a big part of what made the NFL formula work. Pitting individual players against teams creates a contest between equally powerful forces.

The natural conclusion is that you need to enable serious free agency in the working population with respect to their “teams” (corporations). From portable health insurance and IP regimes that favor individuals to protection for whistle-blowers and venture capital markets that are stacked in favor of entrepreneurs rather than VCs, plenty can be done to create a very interesting new dialectic for capitalism.

When individuals, ranging from good to slightly evil to totally evil, find that they can do more with their skills and information in the real market than in the expectations market, innovation will shift to the real game instead of being focused on the design of ever more incomprehensible derivative instruments and frenzied loophole-chasing.

Why does Martin miss this obvious argument and insistently return to authentic executives and delighted customers? How does he manage to miss the biggest source of real information in the business world, the large mass of employees?

I am not sure.  Possibly it is due to past failures to mobilize this class via a challenge to capitalism itself (i.e., socialism). Or perhaps it is something else.




Get Ribbonfarm in your inbox

Get new post updates by email

New post updates are sent out once a week

About Venkatesh Rao

Venkat is the founder and editor-in-chief of ribbonfarm. Follow him on Twitter


  1. Unfortunately, this analysis ignores the most intrinsic and fundamental difference between games and real life in that we get to choose what games we want to play and we only choose to play “fun” ones because the results don’t “matter”. This is in complete opposition to real life where “fun” is something to be avoided.

    There is the anecdote that if a UX designer were to design a game, it would be a giant green button labeled “Win” which, if you clicked on it, you would win the game. It’s then the game designer’s job to take this interface and progressively make it “worse” by adding in more difficulty in a way that makes it more “fun”. Similarly, if you had to achieve a high score in tetris every time you wanted to change the font in Microsoft Word, that would be a poor interface to word as the “fun” would be in direct opposition to your goals.

    The design of games is the complete opposite of the design of productive systems. I would argue that the primary purpose of businesses is not to maximize shareholder value, it’s to figure out how to make the game most unfair to the disadvantage of their competitors. Businesses talk about “building a moat” and “finding our sustainable advantage” and all that means is never approaching our competitors in a fair fight. As Sun Tzu once probably kind of said, “If you’re entering into a fair fight, you’ve already lost”.

    A better analogy than games might be war and, especially, the wars in the waning years of the British empire when the military, too influenced by the experience of sport, lined up their soldiers in tidy little rows, expecting a “sportsmanlike” opposition, ready to engage in a “fair” fight. You, of all people, should know how well that worked.

    I could continue down this vein of analogy: Look at martial arts vs street fighting or kendo/fencing vs sword battles. Attempts at sportifying real life activities sap the life and energy out of them.

    • The analogy is at the level of how games are governed, not how they are played, so it is stronger than you are suggesting. And the book’s point is that for better or worse, business has been sportified using “shareholder returns” as the score. Any set of laws is essentially a sportification.

      Your critique seems correct if it is only players (individuals who can directly affect outcomes) who matter. Mass-mobilization wars are an example. There, it makes sense for players to try and figure out how to gain an unfair advantage.

      Sportification doesn’t help any individual player and especially not the best ones in a “natural” setting. Sportification isn’t done for the benefit of players. It is done for the benefit of non-players. I view sportification as the creation of rules that create balance of power situations among players where natural dynamics would lead to concentration of power that would hurt non-players.

      Shareholder returns was enthroned as the source of sportification logic didn’t work. It turned out that the players could easily subvert this particular sportification solution, but I don’t think you can conclude that sportification itself is a bad idea (that comes down to saying that anarchy is preferable to rule of law basically, which basically ends up leading to monopolist despotism as it did in China in 300 BC or in America during the Gilded Age).

      What makes organized spectator sports and capitalism different from games where there are only players, is that there is a third party whose interests are best served by creating a game among the players. This is society at large that pays social costs not modeled in the game.

      In a way, sportification is a simple divide and conquer strategy adopted by non-players to control players. Game players (business executives and other sociopaths in business) simply have too much information in the game of economics. If they are pitted against information-weak regulators, investors and customers, they win. So you find a way to split the game players into opposed teams that fight each other to maintain a balance of power. Company vs. company partly achieves this (i.e. market competition with some antitrust protection), but we’ve learned it is not enough.

      Management vs. collectivized labor was another dimension of the game that didn’t work out, but (individualist) free agency vs. management might work out. Modern jobs are descendents of institutions such as enforced serfdom (there was a time in tsarist Russia where serfs would be whipped for even attempting to travel outside the manor lands… basically slavery). They still have a lot of serfdom-like characteristics.

      At its core, sportification of business is very simple. But not like your UX button. It isn’t about the user. The primal example of sportification is what the Joker does in The Dark Knight: after destroying the mafia leadership, he’s left with 2 foot soldiers. He breaks a pool cue into two pieces and gives each of them one, with the line, “I only have room for one more person in my gang, make it quick.” The Joker wins no matter who wins.

      Sportification is an inherently dangerous activity in this sense, since it is an attempt by the weak to control the strong by creating and managing a controlled level of conflict. You have to talk fast and convince two big bullies that they’d have more fun fighting each other than beating you up. The bullies only let you create and evolve a game because they are jealous of each other.
      The capitalism game needs to be run like that: so that society wins irrespective of which players win. Instead, we’ve handed a single class a gun and said “be noble and share.”

      • In the U.S., there is no outer ring of power where people are truly outside the game. The game is regulated lightly enough that the real action is inside the game, not outside it. The rule-makers are only a proxy by which sociopaths agree to be kept in check by one another just enough to prevent revolution by the masses, which would tear them all down, or dictatorship, which would overtax them or lock them out of opportunities by force.

        There is a constant scrum between, say, regulation and deregulation, but it’s not really the power struggle between masses and elite, weak and strong, that they’d have you believe in 10th grade civics class. Whoever the masses vote for is open to legalized bribery (lobbying) that subverts their power. Every organization with any strength grows until it hits the boundaries of its neighbors, then tries to stick parasitic tendrils into any place it can find value. Even the three supposedly separate branches of government are constantly elbowing for more influence over one another. Which is exactly how the founders planned it – they deliberately set up power to challenge power in perpetuity. Freedom of speech is so important because without it, the only recourse is reverting to guns.

        If anything, the main force reducing naked grabs for power is that there’s more money to be made in competitive sociopathy than in dictatorship – winning in a niche rather than in totality. If the U.S. fell under a horrible dictator, the economy would collapse from $10T to $1T, leaving only $500B annual at at 50% tax rate, most of which would go to military and bureaucracy. If the dictator could spend 1% of that on himself (palaces and such), he’d take home $5B in assets, say 10% of which or $500M would be liquid. All under constant threat of assassination, coup, and military takeover from outside. Meanwhile Bill Gates amassed $60B in liquid assets, under threat only from a near-miss antitrust case and one pie in the face. Sociopaths are smart enough to notice which is better to emulate.

      • Sports can only be sports if the intrinsic outcome of the game doesn’t matter. This allows you to ignore the outcome and focus on the journey. Take Kendo vs sword fighting for example. At the end of Kendo, people might be slightly bruised but nothing has fundamentally changed about the world, you engage in Kendo for the process of Kendo. With sword fighting, a success is that the other person is dead. You are engaging in sword fighting so that the other person can be dead and you are agnostic about whatever process is necessary to render that goal. You don’t care how much fun you had, in fact, you want the least exciting sword fight possible which probably involves stabbing the other person in the back.

        There are two ways of sportifying Kendo, one is to make the artificial rewards so much larger than the real rewards that actors submit to the sportification, if two nations decided to settle a dispute in single combat rather than war. The other is to reduce the real rewards to zero, by using wooden swords instead of steel ones.

        In the business context, that means either replacing real money with monopoly money in business transactions or finding an incentive that is larger than the entire global economy.

        • In sports, the intrinsic outcome of the game does matter. Winning games advances you in player rankings, improves statistics, and improves reputation. The downstream effects: higher base salary, more endorsements, more fame (including all the perqs that go with that such as lax law enforcement and more options for romantic pairing — or tripling etc.), the possibility of being inducted into some hall of fame or other. It also helps one to build a positive narrative about one’s life — an identity — which is more valuable than one might think. See here for example:–a-wall-street-trader-tells-all-1674614.html (gentleman couldn’t feel good about his identity as a trader, which ultimately led to his departure from trading to focus on the identity he did feel good about, writer)

          But I also think you’re making a rather confused argument by eliding the reason that kendo was actually sportified and by ignoring motivations in general. There was not a period in which people were saying to themselves “gee, I’d really like to get in a sword fight, but there’s a large risk of dying.” Kendo is not really a sportification of sword fighting. The reason you don’t see swordfighting much any more is that knifes are more concealable and guns are more deadly, so there simply isn’t any call for it. Kendo is a sportified version of swordfighting training. The difference between kendo and two people hitting each other with sticks (also a fun game) is that kendo incorporates enough technique into the rules that an abler swordfighter is also an abler practitioner of kendo. Winning a kendo match means something vis a vis one’s swordfighting ability.

          Also note, real swordfighting in both feudal Japan and feudal Europe had extensive codes of conduct that were enforced through public opprobrium. In both cases, swordfighting itself was sportified without removing the “death” part (and without displacing kendo as its own game). These rules served to make duels longer and more evenly matched — essentially less predictable (benefits the majority of courtiers who are not in the top 10% swordfighters, and more importantly benefits the lords who need these people alive to run his fiefdom) — so we can see that the NFL’s prescriptions can apply even in situations where the outcomes matter quite a bit.

          • When I’m talking about sword fighting, I’m not talking about the sport version but, rather, the version used in war. I originally had street fighting and karate in my example if that’s more clear.

            The death of the other person is intrinsic in sword fighting, it’s a *part* of the activity and an unavoidable consequence. Thus, you engage in sword fighting for the express purpose of the death of your opponent. Player rankings etc, are extrinsic factors in football. At the end of the day, all that ever gets accomplished in football is moving a ball up and down a field. Everything else is an extrinsic game system layered on top of that.

    • Your war analogy actually does apply to a changing game:

      If a game is constructed so that those who cheat win, but then the game is changed to make their cheating illegal in future, then many of the characteristics of war still appear.

      In fact, it is in your interest to cheat openly the moment you believe someone else is going to close your advantage: Stay ahead and be the last one to benefit.

      This also means that people are continuously finding flaws in their regulative environment and using them against each other rather than against the regulators. On this basis you might advocate a situation in MMO games where the first person to declare an exploit keeps the advantages they have secured, and everyone else from then on gets them removed.

      If you do this you are co-opting the regulated into the information-processing task of maintaining a level playing field. People don’t play fair, they play dirty in a way that makes the game more rigorously fair.

      There is a flaw with this approach however, letting someone “keep the spoils” of their tactics in the real world might lead to some persistent destabilising effects; instead of the fairly predictible supporter-based returns of being a successful team, the goal achieved by your cheating may be more systemically important and constitute a persistent advantage of it’s own.

  2. The NFL as a metaphor suffers from what Taleb calls the “ludic fallacy” – the belief that life is like a game. It’s a closed system with a small pool of leaders who understand that keeping things balanced inside the system means keeping all their various subordinates roughly equal in power, i.e. equally weak. Darwinian competition is kept at a comfortable simmer and everyone wins except the sociopaths. Sounds great and all, but who will do this for the stock market? The government? With a steady flow of Goldman Sachs employees to the Federal Reserve etc., the government as leader is being gamed. The SEC has no accountability, Supreme Court appointments are a political battleground, on and on. Wherever there is wealth or power, there’s someone gaming the system. There are only so many recursive layers of watchdogs and ultimately it peters out into being the responsibility of the general public, which is outraged at being robbed but can’t spend all its time worrying about laws that ultimately they are not in charge of crafting. The government has an agency problem as well, and there isn’t a higher body to which we may appeal for help, except ourselves. Which is why the author is writing books instead of laws.

    The way we manage sociopaths in the U.S. isn’t to outlaw them, it’s to encourage them, so that there are enough of them around that any given one can’t last for long before the pile of wealth he accumulates comes under attack from other sociopaths and is redistributed. The law is just there to slow things down enough that information can propagate faster than one person can build a dictatorship.

    • Exactly. I agree completely. The weak creating a balance of power among the strong is fundamentally a much harder problem than the strong creating a balance of power among the weak. Keeping Darwin at a “comfortable simmer” (nice phrase) is a problem of stabilizing a naturally unstable equilibrium.

      My proposal is exactly about encouraging more sociopathy, this time from a different class (non-executive employees).

  3. Venkat, great stuff as usual. It really gets me thinking about why people play games in the first place. What IS their real motivation? Is it the goal of entertainment, or are we trying our hand at being a psychic, hoping to sway the decision or perform the mental gyrations in mere moments to identify an outcome long before the coin has been tossed.

    Obviously, attempting to simplify any endeavor into a ruleset ignores the subtleties the surrounding ecosystem. Theory X vs. Theory Y ignores the grey scale in between and those who truly need micromanagement rather than those who are simply looking for a place in which to do their good work and the market local, national, and world market conditions under which the employment is engaged. Sane versus insane ignores the relativity of ones environment and how the mind creates coping mechanisms to endure them. Randomness vs. control, it would seem is similarly illusory and in the mind of the beholder.

    Thanks for keeping my grey matter moving!

  4. A syllogism:

    1. A principal role of government is to intervene in the case of market failure.
    2. If a business sector is too profitable, or a company (companies) is too successfully monopolistic (oligopolistic), that’s a sign of market failure.
    1+2=3. In those cases, the government should intervene.

    Hence my own modest proposal for reforming capitalism, absurd on the face of it in the U.S. but maybe workable in someplace like Singapore: establish standard guidelines for the nationalization of especially profitable companies, or maybe the for the creation of competing nonprofits. If it’s nationalization, then you want to pay the capitalists well enough that it’s like being acquired by Microsoft or Google (“Hurray!”).

    Or maybe the company isn’t actually acquired, just their IP is freed under eminent domain. (That wouldn’t work on the financial industry, though.)

    • The problem with nationalization is that the transition of a market to commodity status is a reversible one. Plenty of countries have nationalized banks and railways for instance. And then you suddenly realize that even those apparently finished games need to reopened to creative destructive innovation. So you need the reverse process of deregulation and privatization as well, to put sectors of the economy back “in play.”

      • I was kind of thinking that instead of nationalizing the railways, that the gummint should become *one* of the competitors. Open to endless abuse, obviously, but this is only to happen in markets that are already broken.

  5. I like a lot of this analysis, but you need to revisit this assumption:

    “Unlike sports betting, the stock market plays an actual information-theoretic governance role in the functioning of capitalism.”

    It’s a nice theory, one that they teach in most places, but when you dig through the assumptions used (esp. if you compare the actual % role of the stock market in financing businesses vs. the theory) it all starts to look a lot less simple.

  6. The book is two things: an exceptionally clear and original analysis of the question of what ails modern capitalism, and an exceptionally woolly headed prescription for how to fix it.

    Same can be said of Karl Marx :)

    PS Can’t wait for your review of Debt

  7. “The book is two things: an exceptionally clear and original analysis of the question of what ails modern capitalism, and an exceptionally woolly headed prescription for how to fix it. ”

    Same can be said of Karl Marx :)

    PS Can’t wait for your review of “Debt”

  8. @Allen K.: The standard guideline for nationalizing a company, in places that do it, IS that the company is especially profitable. That would be oil, in Saudi Arabia, Russia, Venezuela, etc. Also nationalized will be the means to control the nation, such as telecommunications, electric power, and so on. Even if the money and power is put to social purpose at first, the system is one coup away from becoming a military dictatorship. Governments that run businesses don’t run them like businesses, they tap them like a keg and spend the money in completely different places, usually the military and other means of ensconcing themselves further in power. They don’t care about profitability since they can easily become a state monopoly with no competition (inside the border). Where the business requires little creative input, such as emptying known oil wells, this might keep up production well enough for a long time, but where innovation is necessary, government ownership slows down decision-making so much that innovators go elsewhere.

    The problem with separating the real and expectations markets in business is that they have the same goal: making money. People don’t run businesses for reputation points that they can cash in elsewhere, like star athletes collecting endorsement contracts. The money from the two markets is intimately intertwined. A brand-new company will be all expectation and no reality, all investment money and no product. It’s in an owner/manager’s best interest to fuel a bubble in expectations when selling shares, to maximize capital on hand and have the best range of choices for later decision-making (not to mention paying themselves), rather than being starved for capital. Optimizing business decisions to make the most profit on capital comes much later and is arguably only meaningful in mature businesses. Business is inherently speculative until it succeeds.

  9. Odd, it seems that my comments were somehow trivialized and removed, while other comments managed to make it on the site and even replied to. Well, it appears you’ve lost a reader, Venkat. I’ll be in your spam folder.

  10. Carlota Perez, in “Technological Revolutions and Financial Capital” posits that there’s always a period in a technological revolution where financial “innovation” outstrips real-economy activity. And that this generally results in a market crash which, alongside the increasing income-inequality that also comes with early stage of TRs, leads to regulatory/culture changes (Turning Point) that rein things in.

    (The book isn’t really that long or turgid, but I’m finding it a slow read for some reason…)

  11. Too big a post to engage systematically but here are a few reactions and/or unaddressed points that interested me.

    1. Anyone who talks of replacing or, more annoyingly, reinventing capitalism discredits himself from the start. This capitalism that so many people hate can be summarized thus: People should be reasonably free to buy and sell things, including their labor, as they see fit — with some sensible restrictions to prevent their decisions from directly harming others. Given the utter failure of every other system to provide for human welfare, no sane person would really oppose capitalism as thus defined. They would, however, argue endlessly over what, specifically, constitute reasonable restrictions. Thus, all these would-be re-inventors are just arguing at the margins (and anyone who is doing more than that is just nuts).

    2. The NFL is nowhere near as good at maintaining competitive parody or building value for both owners and fans as you seem to believe.
    A. Though it does a much better job than any of the other sports in preventing every team from performing exactly the same every year, a handful of teams have dominated the game, even during the free agency period because for all the sport does to boost stupid franchises, a game with three things — a solid GM, a solid coach and a solid QB — will always win and there are no rule changes that can stop that. Three teams — the Colts, the Steelers and the Patriots — have dominated the AFC for more than a decade. The NFC has been a bit less predictable but the same teams tend to have winning records year after year.
    B. The NFL has engaged in short term profit taking that has damaged its luster. Every time a new TV contract gets signed, more commercial breaks are inserted into games. As a result, games now take a full hour longer than they did 40 years ago and, partly as a result, the NFL’s ratings have tumbled. Not nearly as much as baseball or basketball, but take a look at how many people actually watched Monday Night Football in 1980, and then keep in mind that there were 60 million fewer Americans to watch it. People under 30 don’t actually watch that much football because they grew up after the NFL had watered down its product too much (and they grew up with more entertainment options). An NFL game now contains less than 20 minutes of actual action in about 3 hours and 15 minutes — worse than baseball, and ratings are way down.

    3. Unpredictability is not the only thing that people want in sports, otherwise the biggest sport in America would be high-stakes coin tossing. They want something where skill and excellence are rewarded, but where the sides are pretty evenly matched as well. This cannot always be engineered.

    4. Using the players in sport as an analogy for the executives in business is entirely inappropriate because we are way better in figuring out the value of athletes than we are in figuring out the value of executives. This is particularly true of baseball, where we’re pretty confident we know how many wins a player is worth — although even in this most measured of sports, no one has ever calculated the truly important number, how much “fan enjoyment” a typical player creates. (For example, Josh Beckett is a good pitcher with a low ERA who helps the Red Sox win games. But he tends to take more than 30 seconds between pitches, which makes his games annoying to watch, even if you like the Red Sox. That should make him less valuable but, to my knowledge, doesn’t affect his contract.) We’re much less certain of how to value NFL and NBA players (I know nothing of hockey), but we’re still infinitely better able to value them than to value executives. People who try to study executive pay objectively come to conclusions that vary by several orders of magnitude, which means there’s nothing but worthless gut instinct to say if any executive is a true marvel who deserves tens of millions of dollars a year or an interchangeable cog who should make $250,000 tops.

    5. Any book that suggests that “the problem” is one thing or even largely one thing is being silly, unless it defines its one problem very broadly. For example, I would, in fact, argue that principal/agent conflicts are a massive part of the problem with how modern life functions but that’s such a broad observation as to be useless. There are dozens of different types of principle/agent problems and all of them need different fixes.

    6. Your assertion that people make “wrong choices” through expressed preferences in spending their time or money makes me deeply suspicious of anything else you have to suggest about changing the rules that govern our society and economy. Who the hell are you to say that people are making mistakes by choosing to watch mindless TV or even to smoke? Yes, people do make lots and lots of such mistakes, but it’s a pointless observation because every attempt in the history of man to have smarter, more educated folks shape the choices of those unknowing masses has ended disastrously. The fact that you have read as much as you have and still clearly want to go down that path of encouraging people to want the “right” things is weird and disturbing. I, too, would be happier if folks hadn’t destroyed the best of the West’s traditional communal culture to sit isolated at home, watch “Real Housewives” and get diabetes by eating disgusting and unhealthy food, but it’s not your place to tell them how to live (beyond making them pay at least some of the costs associated with the health aspects of their behavior.)

    • This capitalism that so many people hate can be summarized thus: People should be reasonably free to buy and sell things, including their labor, as they see fit — with some sensible restrictions to prevent their decisions from directly harming others.

      1. Buying and selling things, including labor, long predates anything one might call “capitalism.” By at least 3000 years or so (probably a lot more).
      2. This definition of capitalism doesn’t include anything about actual capital, which is fairly important in defining capitalism.

      Given that, the “discredited from the start” comment was probably going a little too far.

      On your point 6, I think Venkat was just saying that “pushing consumer pleasure buttons” is not a good rule to replace “increase stockholder value” because then you end up with heroin distributors dominating the economy instead of investment banks. Not that people shouldn’t be allowed to make stupid choices.

  12. I guess what I am wondering is why would you want to fix “capitalism” in the first place? (I used quotation marks in response to Scoop’s naive definition of capitalism which places too much emphasis on individual choice). However you define our current worldwide industrialized economic system, it is not worth saving given that it is killing the natural systems that support it. Is China capitalist? Is the IMF and World Bank capitalist? Does it matter?

    Did the Easter Islanders argue about better ways to chop down the trees used to build their monuments while the last trees were cut down? Probably, and that’s what you guys are doing here.

    • Well there’s a big difference between cutting down the last trees and the last-but-hundred. In other words there’s enough support system still functioning for now. It’s probably good to talk about both.

      Honestly I’d say that a permit auction system should fairly represent the constraints of natural support systems, if we could get one of the ground without people going ape about it collapsing their businesses. And providing we could allow transfer between countries, so that the chinese people don’t get completely shafted for taking most of the dirty end of our supply chain.

      So there are current vested interests and problem complexities standing in the way of sustainability, as well as a series of expectations about product churn (which Venkat tangentially mentions). Shifting capitalism to be more adaptable and internally stable seems logical.

      But I have to check my assumptions there, because part of the solution Venkat suggests is reducing the room to manoeuvre of executives and shareholders by setting up new layers of struggle. In other words reducing the internal dysfunctional stability of capitalism. To my mind this means that you probably better give some other actors in society greater stability instead, so someone has enough headspace to consider the environment!

    • Well there’s a big difference between cutting down the last trees and the last-but-hundred. In other words there’s enough support system still functioning for now. It’s probably good to talk about both.

      Honestly I’d say that a permit auction system should fairly represent the constraints of natural support systems, if we could get one of the ground without people going ape about it collapsing their businesses. And providing we could allow transfer between countries, so that the chinese people don’t get completely shafted for taking most of the dirty end of our supply chain.

      So there are current vested interests and problem complexities standing in the way of sustainability, as well as a series of expectations about product churn (which Venkat tangentially mentions). Shifting capitalism to be more adaptable and internally stable seems logical.

      But I have to check my assumptions there, because part of the solution Venkat suggests is reducing the room to manoeuvre of executives and shareholders by setting up new layers of struggle. In other words reducing the internal dysfunctional stability of capitalism. To my mind this means that you probably better give some other actors in society greater stability to compensate, so someone has enough headspace to consider the environment!

  13. Here’s my theory about why shareholder return was elevated from value, to metric, to god.

    In the 1970’s a great thing happened, computers advanced to the point that they could be used to make models. For generations man had desired to know the mathematical outcome of his assumptions, but he had not the pen, paper nor longevity to make it a reality. I imagine the first scientists to say “wait, it can perform how many calculations per second?” had about the same feeling in their gut and excitment in their mind as Trogg did when he said “wait, if I strike these two rocks together I can start fires?”

    Scientists and researchers from all fields, whether physics, health or economics, put in memetic, decided to “MODEL ALL THE THINGS!” But now we need to consider a big little idea called legibility. Modeling requires legibility. You can set a number to age, life expectancy, income, blood pressure, price to earnings ratio, shareholder return, etc. You can’t set a number to goodwill, happiness, the longterm value of a company or “customer delight.”

    So consider the recipe for failure with small additions to reflect what’s happened in the last 40 years:

    Here is the recipe:

    ■Model a complex and confusing reality, such as the economic dynamics of a city
    ■Fail to model all the subtleties of how the complex reality works
    ■Attribute that failure to the irrationality of what you are looking at, rather than your own limitations
    ■Come up with an idealized blank-slate vision of what that reality ought to look like (in this case, maximized shareholder return is the highest value)
    ■Argue that the relative simplicity and platonic orderliness of the vision represents rationality
    ■Use authoritarian power to impose that vision, by demolishing the old reality if necessary
    ■Watch your rational Utopia fail horribly

    Just in general, models need variables like shareholder return, they need metrics to stick to, they need a logic (even if illogical) to their construction and evaluation.