Why greed is bad
This weekend’s Wall Street Journal has an essay entitled “Greed is Good” about the role of bonuses in the financial services industry. The article doesn’t really argue that greed is good–it doesn’t really consider the topic of greed at all, leaving me to wonder if the Journal came up with the catchy title–but the framing invites the question, Wall Street incentive structures aside, of whether greed actually is good.
What is greed?
Any answer has to begin with a definition of greed. The Oxford English Dictionary suggests “Inordinate or insatiate longing, esp. for wealth”, which isn’t very helpful since it simply forces us to define “inordinate”. I’d suggest the following: longing for wealth becomes greed when it causes someone to sacrifice the welfare of others to enhance his own. A simple prisoner’s dilemma game provides an illustration. Consider two business partners, Albert and Bernard. Both have a choice to contribute their efforts and capital ($2 each) toward a new innovation, or extract the assets of the their partnership through an unfortunately un-prosecutable fraud. The innovation requires the contributions of both partners, so if either partner extracts, the business collapses. The partnership has accumulated assets of $4, so when both partners extract (EE), they both take half ($2) and the business collapses. When one partner contributes and one extracts, the extracting partner takes the accumulated assets ($4) and the contribution of the other partner ($2) for a total of $6, and the other partner is left with a loss of $2. When both contribute, the innovation yields a net return of $6, so the partners each have payoffs of $5. The image below depicts the game, the payoffs to each player, and the total wealth.
Under standard assumptions, both players will choose to extract, and the outcome will be EE. Consider the case, however, when Bernard expects Alfred to contribute. Bernard will still be better off by extracting, but doing so would be greedy: it increases Alfred’s payoff at the expense of a considerable drop in social welfare (the partners collectively lose $6 in wealth as a result of failing to innovate) and a crushing loss for Bernard (from the $5 he expected, evidently naively, to -$2). All of this for a very modest increase in Alfred’s payoff.
Thus defining greed enables us to draw a distinction between enlightened self interest and greedy self interest. Enlightened self interest increases one’s payoff while also increasing–or, at least, not decreasing–the payoffs of others. Greedy self interest sacrifices the welfare of others for one’s own benefit. One might also distinguish between weak greed, where one redistributes wealth toward oneself while leaving total social welfare unchanged, and strong greed, where social welfare suffers a real or opportunity loss. Such a model would suggest that there be nothing wrong with insatiable desire for wealth–indeed, it might indeed fuel innovation and economic development–provided that one factors the effects on the welfare of others into one’s decisions.
So what’s wrong with greed?
Prominent economists have argued that economic actors–at least corporate actors–should take any legal actions that maximize profits. This amounts to a claim that “greed is good”, and tells both Albert and Bernard that they are justified in choosing to extract, provided they can find a legal way to do so. If society desires to prevent the extract/extract outcome, then laws should be passed to forbid it.
The problem with this position is that legal systems are costly to build and use, difficult to change, and inevitably imperfect. No legal system will rule out all greedy, or even all strongly greedy, actions. Even given perfect laws, the high cost of litigation will effectively permit actions that are prevented in theory. If Alfred can afford a team of talented lawyers and Bernard cannot, the chances are that Alfred can figure out a way to expropriate Bernard with impunity.
Value systems that condemn greed as immoral will pressure ambitious actors toward socially constructive ways of acquiring wealth. Desirable contribute/contribute equilibria will be relatively more probable. Conversely, a society that draws no distinction between greed and enlightened self interest, even legimitimating greed by loudly proclaiming it good on the pages of prominent national media, will increase the likelihood of destructive contribute/extract and extract/extract types of equilibria.
The damage caused by greed goes beyond the outcomes of particular games. As Fukuyama notes (and the Albert-Bernard example suggests), economic activity rests on a foundation of Trust. When Albert contributes and Bernard extracts, Albert feels betrayed; he will surely be less likely to contribute next time. If greed actions become common, everyone will expect others to extract instead of contribute, and protect themselves by extracting. Collaborative action will become increasingly difficult. Social capital will erode.
[A note on attribution: I'm not aware of whether others have developed these ideas, but they seem straightfoward enough that I'd expect someone has. If there's anyone whom I should be acknowledging, please let me know.]