by Richard Robb
Applying Peirce’s rules, we can see why institutional investors hesitated to buy low-priced, complex, illiquid bonds. The fourth, purposeful rule—one applies the scientific method; the belief corresponds to data in the world—couldn’t justify these investments. There were no data on similar events because nothing like this had ever happened before. Many looked to the Great Depression for precedent, but conditions were so different back then (asset-backed securities hadn’t even been invented) that it couldn’t provide much guidance.
The rules I’ve categorized as for-itself—those that deal with acting in character —would have been equally useless:
1.A new belief X is consistent with the things one already knows. Senior executives started with the view that bonds like this were dangerous.
2.An authority to which one has committed tells one that X is so. During the crisis, few authorities were to be trusted, and fewer still recommended fishing on the bottom in structured finance.
3.X is the style of thing one is inclined to believe. Most would be inclined to believe the opposite—that complex securities were “toxic.”
Institutional investors had little reason to adopt new beliefs and even less to take for-itself, out-of-character leaps into complex asset-backed bonds. Even if a hedge fund manager could convince analysts at a pension fund through intensive interactions that this was the time to throw the rules aside, the analysts could not then transfer this belief and knowledge up through their governance structure. Under normal conditions, it’s hard enough for institutions to coordinate on for-itself hunches, no matter how well-informed, well-intentioned, and clear-headed each individual may be. It’s nearly impossible when investors are in the grip of black swanitis.
The Best Trade Ever
The reasons behind this near-impossibility dawned on me during the financial crisis, after making one of many pitches to try to raise money to take advantage of the bargains I saw.
I had just finished a presentation to the managers of an endowment fund in Nashville, Tennessee. It seemed like nothing I could say and no analysis I could present would convince them. In my colleague’s opinion, “We could have levitated over the conference table and asked, ‘How’d you like to tap into these powers?’ We could have pointed at a chair and turned it into gold. They still wouldn’t have invested.”
I believed that values in credit markets were extraordinary. (Of course, some deals that appear lucrative from a distance are not; at the time I believed I knew which were which.) Yet I simply couldn’t beam my beliefs into investors’ heads. Gradually, I saw that investors’ reluctance had little to do with the usual factors: aversion to risk, information asymmetries, aversion to blame, aversion to ambiguity, herd mentality, or irrational fear. Investors could not fully adopt my conviction about any given opportunity before it disappeared. They could not act on a hunch that I (and others) might just be right. To do so, an investor would have to suspend her belief in reasonably efficient markets—usually, low asset prices signify risk, but not in the collapsing markets of 2008. She would have to rely on experts to make sure she wasn’t missing anything. She’d have to dismiss the surface resemblance between the investments I was proposing and discredited sectors such as subprime mortgages. Then she’d have to persuade her investment committee, which would, in turn, have to convince other committees up the chain. After all that, she might have to wait for years before my beliefs were justified.
I understood this disconnect after rereading, and finally appreciating, the story of the best speculative bet of all time. It wasn’t some hedge fund shorting the U.S. subprime mortgage market, but Joseph and Pharaoh acquiring grain when its value was low and selling much later when its value was high.
According to the Book of Genesis (41:1–57), Joseph interpreted Pharaoh’s dream of seven fat cows followed by seven lean ones as a prediction for seven years of bountiful harvests followed by seven years of drought. Acting on inside information of the highest order, Joseph spent seven years collecting massive quantities of grain on behalf of Pharaoh. When the drought began and famine spread, Joseph sold and bartered the grain. Because the value of grain had shot up, these sales enabled Pharaoh to acquire all the money in Egypt and vastly expand his lands.
The story of Joseph and Pharaoh showed me what I was missing as I tried to make sense of investor behavior. Investment management involves more than information about which asset to buy or sell. To take a leap with other people’s money and hold on for the long term, an investment manager must have a truly extraordinary ability to communicate her beliefs through a chain of actors and inspire those actors to maintain their resolve.
After interpreting Pharaoh’s dream, Joseph went on to offer some unsolicited advice: hire a wise and discerning man (hint, hint) to oversee the grain-hoarding project. Why wise and discerning? Why not someone skilled at logistics or experienced with managing large enterprises? If everything Joseph predicted came true, the stockpile would grow for seven years with no drought in sight, testing the resolve of everyone in Egypt. After a few years, rival advisers, the military, priests, farmers, and perhaps Pharaoh himself might question the bet. Even Joseph might start to wonder whether the drought was really coming or worry that he had missed some hidden conditions. As the memory of prophecy dimmed, he might question whether he had misinterpreted the dream.
He did have a good track record, but track records come and go. In prison, he correctly predicted the fate of two other inmates: that Pharaoh’s cupbearer would be restored to his position in three days while the baker would be hanged. That was out of the ordinary, for sure, but had he just gotten lucky? Had his luck now run out? At least one prophecy was yet to come true: Joseph’s childhood dream that he would reign over his brothers.
Pressure must have mounted on Joseph to cut losses. Seven years is a long time to wait. Only a wise manager could maintain his nerve that long and inspire others to remain confident. Joseph and Pharaoh could not have fully transmitted the force of their belief to the people of Egypt, no matter how hard they tried. Their interior perspectives, molded by their experience, could not be shared. Verbal explanations would fail to communicate what prophecy is like. Instead, Joseph and Pharaoh relied on skill and authority to keep stakeholders at bay. Pharaoh not only hired Joseph but elevated him to a level seemingly unnecessary for an ordinary grain buyer. Upon appointing him, Pharaoh gave Joseph a signet ring, put a gold chain around his neck, clothed him in fine linen, and paraded him in front of the Egyptians in a chariot to show that Joseph was now his unequivocal number two. This authority allowed Joseph to accomplish his for-itself bet.
Joseph needed wisdom to hold onto his authority, which could have been snatched away at any moment. No matter how flexible the Egyptians may have been, it would have been impossible for Joseph and Pharaoh to convey their conviction; Joseph had a divine revelation and Pharaoh was there when it happened. This was unique. There were no words and certainly no mathematical models to describe the leap they had to take.
If my partners and I had the wisdom of Joseph or the authority of Pharaoh, no investor would have exited from our funds in 2008–2009 and we would have raised vast sums to buy at distressed prices. If we had inspired no confidence at all, everyone would have wanted to redeem from existing funds and we would have been unable to raise any fresh money. The reality was somewhere in between: about half of our investors redeemed, and new investments started to trickle into our funds in 2011.
From my point of view, investors who stayed on the sidelines in late 2008 through 2009 missed the chance to double or triple their money over two years without taking much risk. But I doubt any of them are kicking themselves now. As we’ve discussed, beliefs are inherently personal. So too, then, is a leap outside of them—an out-of-character act. Few institutional investors could take the sort of leap that bold bottom fishing required, no matter how hard the ardent believers tried to persuade them.
Years ago, I asked a well-known venture capitalist to give a short pr
esentation to a class on economics and finance. The class had covered a fair amount of theory that I was hoping she would reinforce with colorful examples. We chatted ahead of time about what she might say, and she stayed on message right up to the end. But she closed her talk by saying that really, when she looks an entrepreneur in the eye, her gut tells her what to do. I could feel the students thinking, “Oh, I can do that.” They wondered why they’d bothered with all the theory I’d been trying to teach them as a semester’s progress melted away.
Since then, I’ve hardly ever invited guest speakers. But the venture capitalist was right, in a way. She is in the business of betting on hunches that can’t be fully communicated. Her investors and limited partners get that. Every time she launches a new fund, her investors enter into a binding legal agreement to leave her alone for seven years—same as Joseph—to prove what she can do. It’s the twenty-first-century signet ring.
PART III
People
Can’t buy me love.
—JOHN LENNON AND PAUL MCCARTNEY
6
Altruism
The Good Samaritan’s good deed in the biblical parable satisfies the definition of altruism: a voluntary action undertaken at cost to the actor to advance someone else’s interests. The story begins with a man, beaten and robbed, lying half-dead on the road from Jerusalem to Jericho. Two men pass by without offering aid—indeed, without pausing to see whether he’s alive or dead. But not the Samaritan. He stops, dresses the man’s wounds, takes him to an inn, ministers to him, and as he’s leaving, gives the innkeeper money to care for him. The Samaritan voluntarily incurred a cost of time and money for which he could not have expected a reward. It was an expression of concern for the welfare of a stranger. Does that make the behavior for-itself? Before jumping to that conclusion, let’s take a look at the different categories of altruism.
Some prosocial dealings can easily be explained in terms of rational choice. At one end of the spectrum, other people are instrumental to our purposes. We work together to achieve common goals, engage in mutually beneficial activities, and perform favors in expectation of future favors. At the other end of the spectrum, our connection to others is so tight that their interests become our interests. The actions we take on their behalf have a purpose, and rational choice can explain how we optimize our desires alongside those we love or who love us. In the middle of the spectrum, however, we confront others as more than objects yet less than extensions of ourselves. Rather than purposeful, this category of altruism is spontaneous and for-itself.
A Taxonomy of Altruism
We can distinguish five main categories of altruistic behavior: (1) selfish altruism, when an individual appears to subordinate his interests while actually promoting them; (2) manners and ethics, when an individual observes social norms or adheres to established moral principles; (3) care altruism, when one person cares directly about the well-being of another; (4) mercy, when a person performs a sporadic altruistic act that defies rational explanation; and (5) love altruism, which describes acts that transcend all preferences and do not stand in relation to them. The purposeful choice model makes room for the first three types of altruism but not the last two.
Inherently Selfish Altruistic Behavior
Although “selfish” has a negative connotation, selfish altruism need not be sinister. It can be as innocent as doing a favor for a friend or business associate in anticipation of payback. (If the two parties have an explicit agreement, we wouldn’t call it altruism—that’s just doing business.) Selfish motives may be in play when people claim to be acting for the public good by volunteering, engaging in charity work, or donating money. If these activities are undertaken in order to appear generous, gain trust, or otherwise serve private interests, they are selfish.
The selfish altruist’s most important tool is repeat dealing. It’s widely known that if the prisoner’s dilemma game is played only once, players seeking to maximize their payoff will choose the inefficient noncooperative solution. But if they play repeatedly and care about the present value of their payoff, they’ll learn to cooperate. If one player cheats, the other will switch to the noncooperative strategy, playing tit-for-tat until cooperation is restored. Each cooperative play, considered in isolation, looks selfless because either player could do better by cheating. But in the long run, cooperation leaves both better off.
In contexts like business, reputation can link all players together, even those who don’t face off more than once. When each round is played with a new opponent, cooperative behavior can profit a self-interested person if subsequent opponents are aware of his reputation and have their own reputations to protect. Viewing each round in isolation, individuals once again appear altruistic, but under the surface each may be pursuing the optimal strategy for personal gain.1
Cynics may argue that all altruism is, at heart, selfish. This view can’t be taken seriously. If, for example, private gain were the sole reason for volunteering, then everyone would know it was a ruse, it would fail to convey the intended message, and no one would do it. Recognizing that some altruism is selfish does not commit us to the depressing view that all altruism is.
Manners and Ethics
We may not consider polite behavior altruistic, but it meets our definition. Although instilled in us from an early age, manners can be understood as desires in the purposeful choice model and traded for a price. Adherence to basic social norms is predictable: I normally don’t make phone calls on trains but will break that rule as a simple function of factors including the call’s urgency and expected length, the number of people in the car I’d disturb, and the quality of reception. Adherence to norms is rewarded with personal satisfaction and social acceptance; the violation of norms incurs a cost.
Acting in accordance with an ethical code can also be understood as purposeful. While rules would dictate the same behavior each time the same circumstances arise, people sometimes act against their ethical codes, suggesting that these codes are not hard and fast rules. Rather, adherence to an ethical code is fundamentally a desire that must be balanced against other desires. This desire varies in intensity from person to person. When considering the right course of action, the depraved hardly care at all, the majority sometimes give way to expedience (placing a price on violating a principle, but a small one), and the most upright place a high price on rule breaking or reject expedience altogether.
If we accept that people prefer to abide by manners and principles when it doesn’t cost them too much, we can interpret behavior in experiments like the “ultimatum game” as purposeful. This game is played by two anonymous people. The “allocator” proposes how to split a fixed pot, say ten dollars, between himself and the “receiver.” The receiver can either accept or reject her share. If she accepts, she gets to keep it. If she rejects, neither the receiver nor the allocator gets anything. If both players in a single-round game care only about money and trust the rules, the allocator should propose $9.99 for himself and $0.01 for the receiver. The receiver should accept, since $0.01 is better than nothing.
In experiments, though, that’s not usually what happens. The data show that most allocators in developed countries propose roughly even splits. In a carefully designed study, researchers found that splits cluster around 50/50 in Ljubljana and Pittsburgh and around 60/40 in favor of the allocator in Tokyo and Jerusalem. Low offers tended to be rejected in all four cities, suggesting that receivers trade financial gain for a chance to punish flagrant inequality.2
The cost of conforming to social norms in most of the ultimatum game literature is fairly low—a few dollars here or there. But an experiment conducted in poor villages in India with pots of up to 160 days’ wages found that higher stakes lead to more lopsided splits in favor of the allocator and less frequent rejections: as the price (to the allocator) of fair play and (to the receiver) of punishing unfairness went up, players tended to prioritize wealth maximization.3
Care Altruism
Bona
fide care for someone else’s well-being results in utility functions that are, in economics jargon, “interlocking.” If a child’s well-being is a direct input into her parent’s well-being, then the parent demonstrates what I’ll call care altruism; she desires that the child be happy. This can, but does not necessarily, work in both directions. Parents may or may not care about their children, and children may or may not care about their parents.
Care altruism can be split into two categories. In observed care altruism, the income of the benevolent person (the one who cares) is sufficiently high and the care is sufficiently intense that action results. In unobserved care, the benevolent person cares but not enough to incur a cost to do anything about it.
OBSERVED CARE ALTRUISM
Gary Becker’s elegant Rotten Kid Theorem shows how one benevolent family member can theoretically cement a household into a single optimizing unit. In his model, the family acts as if it shared one utility function and one budget even if the children are “rotten,” that is, concerned exclusively with their own consumption. They instinctively advance their own interests by maximizing household wealth so that the benevolent family member feels richer and obtains more of everything, including consumption for her children, rotten or not.4
The redistribution of wealth within the household arises automatically out of the preferences of the benevolent family member. (For the sake of simplicity, let’s follow Becker and designate the mother as that family member.) The mother is not motivated by a desire to affirm her self-image as nurturing; nor is she concerned about equity or justice. She simply wants the children to be happy because their satisfaction enters directly into her own. In this setup, redistribution is never used as a form of discipline, since punishment would impose a deadweight loss on the family that everyone would have to absorb.
