Thursday, March 20, 2008

Character and Capitalism

(Naked Capitalism) Steve Waldman is on a roll. He has an excellent piece today arguing that despite contemporary notions otherwise, capitalism and character (meaning moral fiber) have not and need not be contradictory.

Although Waldman makes a good case, the barriers to the return of character in commerce are more profound than he lets on.

A colleague of mine, Amar Bhide, a professor at Columbia School, did some field work in the early 1990s on the role of trust in business and was very disheartened with what he found. Power and economic self interest trumped considerations of morality. Business owners were willing to accept being screwed by customers because they felt they needed them. They resented it deeply, they grumbled, but they accepted that they lacked the leverage to demand better treatment (I will need to locate his article, which I think ran in the Harvard Business Review, but the title was something along the lines of "Why Be Honest?" and it concluded, with considerable reluctance, that there wasn't much upside in behaving well).

What makes it possible to have values is enforcement mechanisms, which usually boil down to prevailing standards. In a small town, if a store owner is unpleasant, tries to short change customers, or kicks his dog, word gets out and business suffers. But many of us mange our affairs in a impersonal way (think of Internet purchases). We have limited interactions with people. We might have a relationship with a firm, yet the account manager changes every couple of years. The standards thus reflect what is considered acceptable for the industry as well as the company. Great conduct in the credit card industry (if there even is such a thing) is not the same as great conduct from a hotel. Where is there opportunity for character to enter into either of these interactions? Even if someone goes the extra mile for you, you might never see that person again. Their effort will probably go unrewarded, or worse, might be against policy.

Now I am digressing a bit; Waldman meant character in a narrower Victorian sense, as being a person of one's word. But even then, the drive to efficiency that has become pervasive in American businesses has made it well-nigh impossible for that to be operative. The reliance on FICO scores in place of more nuanced credit decisions is merely the logical result of processes that have been in play for many years.

One of my favorite banking industry factoids is that despite the widespread belief that bigger banks are more efficient, every study of commercial banks ever done has found that they have a slightly increasing cost curve once a certain size threshold has been achieved. Where the increasing costs (per dollar of assets) kicks in varies, but the highest point I saw (this was some years ago) was $5 billion in assets. One study found the break point was $100 million.

Now this flies in the face of most logic. Banking is all about automated, repeatable transactions. Moreover, big banks enjoy tremendous cost advantages in funding (big bond issues and other capital markets sources are much cheaper than relying primarily on deposits). Big banks are also more likely to be in pure fee businesses like M&A and loan servicing which require nothing in the way of assets beyond desks and phones.

My pet theory is that old-fashoned lending, the know-your-borrower types, is much more efficient on an all-in basis than the interpersonal, multi-layered credit scoring processes used in every type of loan product in a large bank. Yes, it costs you a lot more to source the loan. But they probably make it up in lower loan losses (a combination of lower default and more success with loss mitigation, since if you know the borrower, he will probably feel a greater obligation to repay. In addition, if he does experience real duress, your knowledge of him and the community will enable you to make a more informed assessment of his ability to repay, again improving the odds of a successful restructuring.

No academic has ever investigated my pet theory, but as of the last time I read the literature, no one had come up with an explanation. But there is anecdotal evidence. My colleague Doug Smith (no relation) in an article in Slate, pointed out that not-for profit lenders had provided subprime loans, yet experienced losses no worse than on prime loans, precisely because they screened borrowers in the traditional manner, in person, and assessed, among other things, their understanding of the loan and their degree of commitment.

So why, when we have had a resounding failure of impersonal methods of assessment, am I pessimistic about the revival of interest in character? Because it appears our society is unwilling to go back on the false economy of preferring anonymous, rule-based approaches. These methods do broaden the market of possible counterparties, which is seen an entirely beneficial, when there are hidden costs in having only superficial proxies about the people you interact with. If anything, I see ample evidence this attitude is becoming more pervasive.

Some examples: it has become common practice to put vendors to large corporations through a formal approval process. I have found them to be inappropriate and often misguided. For instance, one company required its vendors (which included big ticket law firms and consulting firms) to certify that their employees had gone through drug screening recently (the logic apparently was that if they had a drug habit, they might steal Big Corp's secrets, as if they learned anything that might indeed be that easily traded upon). If you work for a good firm, the idea that you have to get tested to work on XYZ account is a non-starter.

Now a firm like McKinsey can usually escalate matters to a level of management where can get an exemption from this nonsense (a McKinsey partner told me that on a study where it came up, there was absolutely no useful intelligence the firm would be party to, plus, as pointed out, "We pay people well enough for them to afford their own drugs."). But a smaller vendor would be unable to escape this and other cookie cutter demands that are often both intrusive and irrelevant. I've pretty much written off having large corporations as clients (I used to work for them quite frequently). It isn't worth the indignity of going through that process. In fact, it seems destined (ex the firms that can wriggle their way out of it) to assure negative selection: only firms who really need the business will put up with these demands.

Similarly, it's become increasingly common to use credit reports as part of the new hire screening process, even for very low level jobs. Again, I'm not sure what that proves, I can think of plenty of reasons why an otherwise conscientious person might have a bad credit record (say needing to support an ailing parent and becoming overextended). In close calls where there was a particular reason to think it might be helpful, I could see resorting to it, but to use it as a proxy for character is silly.

From Waldman:
Via the indefatigable Mark Thoma, our attention is drawn to an odd piece by Robert Skidelsky. I was left mostly bewildered by the article, but I was intrigued by the author's discussion of the virtues that are and are not inculcated by market capitalism:

Consider character. It has often been claimed that capitalism rewards the qualities of self-restraint, hard-work, inventiveness, thrift, and prudence. On the other hand, it crowds out virtues that have no economic utility, like heroism, honor, generosity, and pity. (Heroism survives, in part, in the romanticized idea of the “heroic entrepreneur.”)

The problem is not just the moral inadequacy of the economic virtues, but their disappearance. Hard work and inventiveness are still rewarded, but self-restraint, thrift, and prudence surely started to vanish with the first credit card. In the affluent West, everyone borrows to consume as much as possible. America and Britain are drowning in debt.

One thing to remember is that there is no such thing as "capitalism". In the real world, there are actual practices and institutions, the details of which bear consequentially on both moral and economic outcomes. There are infinity of possible capitalisms, and at any given moment we are living just one. A stylized graph of supply and demand always hides more than it reveals.

The capitalism we are living right now is rather a nightmare, due to a credit, um, event. So it seems a propos to remember that credit analysis traditionally includes an explicitly moral component. Remember the "5 Cs of Credit"? Character, capacity, capital, collateral, and conditions. Character.

Here's a famous bit of financial history, as recounted by Jean Strouse in the New York Times:
Asked by the lawyer for a congressional investigating committee in 1912 whether bankers issued commercial credit only to people who already had money or property, [J. P. Morgan] said, "No sir; the first thing is character." The skeptical lawyer repeated his question and Morgan, in Victorian terminology, elaborated on his answer — "because a man I do not trust could not get money from me on all the bonds in Christendom."

If you think Morgan, the arch plutocrat, was just telling a nice sounding, self-serving lie, think again. Think about a world in which there was no SEC, FDIC, or Federal Reserve; in which there was no technology sufficient to prevent a person from simply disappearing, changing his name, starting over somewhere else. Morgan invested vast sums, and though he was a powerful man, he could always be taken. When parting with a dollar, he could not be so lazy as to presume a courtroom would ensure its repayment. Morgan had to trust.

Since Morgan's day, in pursuit of efficiency and safety, we've built up institutions designed to automate and certify the evaluation of character. When we lend money, we don't ask to meet the person who promises to repay us. We look for a nod by a regulator, the AAA brand provided by S&P or Moody's or Fitch, perhaps a FICO score. But those are not markers of character at all. We don't take them to be. We understand that banks engage in regulatory arbitrage, finding ways to stretch their balance sheet as far as possible for yield despite whatever regulatory regime is in place. We know that credit issuers (and bond insurers) do what they need to and no more for their rating, that perfectly dishonorable individuals attend to their FICO scores to maintain access to credit. Actual character is completely washed out of these proxies. The capitalism we have is one that presumes that all actors are sharks, that business is business, and that it is irrational to take any less than you can get away with unless you will "incur costs" from decertification. I'm not sure that J.P. Morgan would be willing to lend to any of us, and it's not because we're worse people. We just live in different times, a different world.

We shouldn't go back to the world as it was at the turn of the century. When character evaluation was a personal exercise, it necessarily depended upon social connections, whether someone you know and trust can vouch for someone you don't yet know, whether you can be sure that disgrace and dishonor would be costly. And we definitely should not adopt a moralistic attitude towards debt nonrepayment right now, just when a throng of irresponsible lenders are demanding "responsibility" from borrowers whose calls they would not even take a year ago. (For the record, I think that "jingle mail" is perfectly acceptable under present circumstances, and that the recent "bankruptcy reform" was a cruel mistake.)

But I do think that it's an interesting technical question, going forward, whether we couldn't set things up so that the criteria by which investors decide where to put their money map more closely to what we would recognize as trustworthiness or character. "Abolish the SEC and the Fed and the ratings agencies!" is not a sufficient proposal. Crises due to misplaced trust long predate those institutions, and are a large part of why they came to be in the first place. Morgan was successful not because he did what everybody did, but because he did what almost nobody did, despite the lack of ratings by S&P to stand-in for due diligence. Investors have always been hopeful and lazy in good times.

T.S. Eliot once wrote, "It is impossible to design a system so perfect that no one needs to be good." Perhaps the art is to come up with a system, however imperfect, under which being good is the best way to succeed.

No comments: