Incentives: The Unsocial Capital

A few months ago, I had to get a car door repaired while in India. The car was eventually repaired to my satisfaction, but it was a harrowing experience dealing with different players with different incentives. The lock on one of the doors was jammed. The repair person I was assigned at the car dealership said that he could take care of it without having to replace the part … provided I “took care of him.” Separately, his boss told me that because it was a car door in question, I shouldn’t take any chances and that I should get the part replaced. But the new part would still have to be replaced by the employee who wanted me to deal privately with him and I wasn’t sure he would do it right. Eventually, I went with the manager and got the part replaced even though it wasn’t needed. The experience underscored the importance of individual incentives: the manager’s regarding sales of new parts and the repairman’s desire to go beyond his fixed salary.

COLLATERAL ...

During the past few years, banks in the United Kingdom and the United States have been giving out loans rather freely. People bought houses, whose prices went up, and the banks felt secure about their loans. The banks also felt secure because they were passing on the risk through securitisation – bundling the loans up as collateralized mortgage obligations (CMOs) – so the risk of a homeowner not paying back would be incurred by the institution that bought the CMO.

These CMOs had a decent risk premium over U.S. bonds or U.K. gilts so they were attractive to the portfolio managers in the same banks. Portfolio managers want to make returns over U.S. treasuries, and buying something “as safe as houses” seemed like a really good idea. So they bought as many of these CMOs as they could, including those based on subprime mortgages, while the mortgage people in the same banks kept loaning out as much money as they could. The result is that the whole world’s economy is disrupted on account of a relatively small number of people.

... AND COLLATERAL DAMAGE

Damage can also take the form of significant loss in life depending on the incentives of a few people. In Northern Ireland, certain groups are trying to re-create the problems that we thought were well behind us. You could ask how does killing two British soldiers and a policeman help Northern Ireland? But the real question to ask is how it benefits the leaders of these groups. You might ask why we went into Iraq if you cannot convince yourself of the pious explanations forwarded about “freeing” Iraqis. There is more democracy there now than before the war, but with more than a million Iraqis dead in six years, it is hard to avoid asking the question about the incentives of the decision-makers.

We could raise questions about senselessness in Darfur, in the Israeli attack on Gaza that invited charges of war crimes by Israeli groups as well as by Amnesty International, and in Hamas’ rocket attacks on Israeli towns, but again we would be asking the wrong questions. The real questions should be about the handful of decision-makers and their personal incentives – holding on to power in Darfur, about a three-way election in Israel, or about Hamas leaders drawing the world’s attention to their own “valour” in standing up to the prolonged siege of Gaza. The rest of us feel empowered only by being in the herd buying their slogans.

CEO COMPENSATION

Past mergers and acquisitions cast long shadows over the present economic problems. Why do companies continue to acquire other companies despite the historical fact that the acquiring company does not gain shareholder value in four out of five cases and many times loses it badly? It turns out that even in a bad acquisition the CEO of the acquiring company personally benefits a lot.

Companies pay CEOs well and CEO salary has grown massively over past decades, not just in absolute terms but also as a multiple of the lowest paid employee in the company. But does more money really attract the most talented? Or, more likely, does it attract the greediest? Worse of all, does colossal compensation attract people who are not only greedy but also smart?

CONCLUSION

Reflecting on my years in supply chain consulting, the reason things don’t work is not that people are stupid but because people are so smart! Sales people are incentivized by revenues, others by cost, and yet others by margins or unit volumes. The result is mostly a mess but sometimes things work and we call it “best practice.”

What does this mean for analytic modellers and modeling? Should we adopt messy agent-based simulation models – but where’s the theory? – or should we continue with theory-based elegant models of monolith companies or contrived supply chains with one quadratic-risk-averse supplier with one logarithmic risk-averse customer? It seems pointless (and difficult) to model social capital, but shouldn’t we figure out how to incorporate the unsocial capital of incentives in our models?