The Credit Crisis and Trust Networks
What do analyses of social networks and trust have in common with the subprime-mortgage credit crisis? Quite a bit, it turns out.
The NYTimes on Sept. 2, in “Can the Mortgage Crisis Swallow a Town?” introduces the Egglestons of Maple Heights, Ohio. They’re a hardworking family trying to sell their house in a market where prices are falling—because too many neighbors are selling—because banks are foreclosing—because owners bought exotic mortgage-crack they can no longer afford.
The same article describes a business victim: Mark Stefanski, CEO of Third Federal Savings and Loan, a Cleveland thrift that his parents founded in 1938.
Unlike most of his competitors, Mr. Stefanski resisted the urge to cash in on the subprime lending boom…
“The model has shifted,” says Mr. Stefanski. “It became very lucrative. But it was totally irresponsible for the sake of greed.” Not that Mr. Stefanski didn’t notice the profits to be had. “Absolutely, we were tempted,” he acknowledges. “We arm-wrestled and talked, but we decided not to change the model. We felt it wasn’t the right thing to do.”
Mr. Stefanski is no social worker. He lives in an affluent suburb of Cleveland and earned nearly $2 million last year. But he does not hide his feelings about just what went wrong in places like Maple Heights. “The whole system was based on raping the public,” he says, matter-of-factly. “Not everyone should own a home — just those who can afford it.”
Third Federal has a branch in Maple Heights, Mr. Stefanski says, and in the past, “we owned Maple Heights.” But in recent years, he says, “The predators just jumped on it.”
Third Federal’s share of the mortgage market in northeastern Ohio fell to a low of about 11 percent by 2001 from more than 30 percent in the early 1990s.
Back to trusted social networks.
There’s a lot of interest in using digital technologies to help connect people. The economic gains of lowered transaction costs from trust are real—it isn’t just a blogosphere fad.
The problem with trust is—at heart it’s an analogue function in a world of digital suitors.
Put too many degrees of separation between you and another, and trust falls down. The decay rate of trust from my LinkedIn friend to my friend’s friend is huge.
As my friend David Krathwohl says, “I’ll trust a seller’s eBay ranking to buy a book; that doesn’t mean I’ll introduce him to my daughter.”
Back to mortgages.
When a bank owns the mortgage, the bank earns money—or not—when the borrower pays back the loan—or defaults. When the owner of the mortgage is four times removed—when the mortgage is sold, then securitized, then re-sold as part of a package of collateralized debt obligations (CDOs)—that incentive is removed. The nth "owner" of the mortgage—the CDO buyer—doesn’t re-check the credit risk.
Bank regulations once addressed loose lending; there is no such regulation on the underlying credit risk for CDOs. As the Sept. 2 NY Times Magazine article "Subprime Time" describes, they depended on ratings agencies.
Like addicts seeking a faster high, banks and ratings agencies switched to the fast fees of transaction income over the longer returns of relationship banking. Ditto the ratings agencies. Ditto the securitizers.
The new system made for a more liquid market; sounds good, right? It spread risk; another good. And it substituted efficient markets for local cottage industries. More good.
But the whole thing also rested on turning relationships into transactions. This is the dark side of business process re-engineering, Web 2.0, and globalization—everything can be sliced and diced and outsourced to others in the name of efficiency and liquidity. But at each point, a diet of transactions-only starves relationships—and therefore trust. (For a far more financially literate discussion of this dark side of liquidity, see Equity Private’s Liquid Reflections of August 31).
Degrees of separation matter. Not everyone reacts like Mr. Stefanski, and he can’t carry the load alone.
Too many degrees and you get the musical chairs game: I’ll be gone, you’ll be gone, let’s just do this deal and get out before the music stops. Greed thrives, ethics starve. As with all relationships deprived of the currency of the Real, things can get pretty transactional pretty fast.
It’s ironic. What sounds like sound principles of insurance (spread the risk), banking (liquidity), and networking (spread the contacts), end up producing endemic greed, visiting disaster on consumers, and punishing ethical businesspeople.
Trust doesn’t travel well digitally. A song loses no data when it’s digitally copied—but digital replication of trust loses heart. A "trust network" based solely on transactions is a network devoid of all but the narrowest version of trust—a track record with no memory of what it is supposed to be tracking.