Friday, March 23, 2007
Underwriting Technology To Blame?
Nice to see this mention of my former employer in the New York Times, in its reporting on underwriting technology in the mortgage market.
And here's my take on the NYT article:
The graphic (right, from NYTimes.com) shows an assembly line of homes stamped with "Approved." The implication is that underwriting technology helped to foster the current problems with subprime lending. You might as well say that automated switchboards fostered crank calling, e-mail fosters spam, and so on. It's true, but not the most useful starting point for fixing the problem. There's nothing particularly virtuous about a process that takes six weeks and a ton of paperwork, compared to a 30-second credit check that incorporates the same data to get to the same result.
The article quotes Professor Nicolas Retsinas of Harvard's Joint Center for Housing Studies: “Before there was A.U., down payment mattered a lot. Where we’ve crossed the line in recent years is to say, we don’t need down payment.”
Yes, it's true that down payments don't matter as much as they used to, and it was the underwriting model that made lenders comfortable enough to extend loans to people without down payments. But I'm not so sure that's where we "crossed the line."
In the old days, a lender would hold a mortgage loan on its own books. Accordingly, the lender was risk-averse as a counterweight to the borrower. Neither the borrower nor lender wanted the loan to default, but it was the lender who was singularly exposed if that were to happen. Thus, the lender would tend to put the brakes on the borrower's aspirations.
In the current environment, the lender’s incentive is to originate loans. They're not compensated on whether the loan carries through to maturity, but rather to whether they can get borrowers to sign the contract. Both sides get a short-term "win" when the deal goes through, but the borrower's the only one who has to live with the consequences. The mortgage itself is sold into the secondary market and then pooled into collateralized mortgage obligations (CMOs). The CMOs are then broken into tranches of varying risk profiles. The investors holding the high-risk tranches are compensated for the risks they take, and so they are risk-neutral.
This topic can be difficult to explain, but let me try an analogy:
Let's say there's a 26-mile marathon. Most entrants have trained for the event and will have no problem finishing. Some will finish quickly, some will limp through given enough time, and others will drop out.
Suppose you want to place a bet on whether a single runner completes the race. If you had your own money riding on the outcome, you’d be pretty careful about it – you'd have to know his or her training history, past results and current fitness level, and you’d want to know what the temperature and humidity would be on race day. You’d take a pair of calipers to measure the racer’s body fat. And you’d want to know if someone’s lying to you. If they say they’re running 70 miles per week, you’d ask the neighbors just to make sure.
Now, suppose the organizers said to you, "Why are you betting on a single racer? It’s too risky! Instead, I’ll pay you $x for every person you can convince to enter the race."
Why would the organizers make such an offer? Because they don’t need EVERYONE to finish in order to make the numbers. They just need, say, five-sixths of the entrants to get through the race in order to maintain credibility as race organizers. You see, the organizers also operate a special betting parlor, where big bettors can wager on how what percentage of runners will actually finish. They make enough from the betting action to cover the minimal costs of bringing exhausted and defeated runners to the medical tent.
Eventually, you’d start to use the same calipers, the same neighbors and the same weather report not to disqualify people who shouldn’t be racing, but instead to tell potential racers, "Hey, you’re in about the same shape as so-and-so down the street, who ran a great race last year! Go for it!"
Bottom line: The CMO market pooled the risk but not the responsibility. The diffusion of mortgage risk across the capital markets is where we "crossed the line," rather than as a consequence of the enabling technology. Lenders were risk-averse, but now they are risk-neutral. Therefore, if there are negative externalities stemming from foreclosures by risk-seeking borrowers and their mortgage agents, we may have an argument for regulation to promote risk-averse decision-making.
-----
Some of my earlier coverage on the mortgage market from Bank Systems & Technology.
“Without the technology, there is no way we would have been able to do the amount of business that we did and continue to do,” Scott Berry, executive vice president for artificial intelligence at Countrywide Financial, told a trade publication, Bank Systems & Technology, in the summer of 2004.Here's the original article by Judy Ward.
And here's my take on the NYT article:The graphic (right, from NYTimes.com) shows an assembly line of homes stamped with "Approved." The implication is that underwriting technology helped to foster the current problems with subprime lending. You might as well say that automated switchboards fostered crank calling, e-mail fosters spam, and so on. It's true, but not the most useful starting point for fixing the problem. There's nothing particularly virtuous about a process that takes six weeks and a ton of paperwork, compared to a 30-second credit check that incorporates the same data to get to the same result.
The article quotes Professor Nicolas Retsinas of Harvard's Joint Center for Housing Studies: “Before there was A.U., down payment mattered a lot. Where we’ve crossed the line in recent years is to say, we don’t need down payment.”
Yes, it's true that down payments don't matter as much as they used to, and it was the underwriting model that made lenders comfortable enough to extend loans to people without down payments. But I'm not so sure that's where we "crossed the line."
In the old days, a lender would hold a mortgage loan on its own books. Accordingly, the lender was risk-averse as a counterweight to the borrower. Neither the borrower nor lender wanted the loan to default, but it was the lender who was singularly exposed if that were to happen. Thus, the lender would tend to put the brakes on the borrower's aspirations.
In the current environment, the lender’s incentive is to originate loans. They're not compensated on whether the loan carries through to maturity, but rather to whether they can get borrowers to sign the contract. Both sides get a short-term "win" when the deal goes through, but the borrower's the only one who has to live with the consequences. The mortgage itself is sold into the secondary market and then pooled into collateralized mortgage obligations (CMOs). The CMOs are then broken into tranches of varying risk profiles. The investors holding the high-risk tranches are compensated for the risks they take, and so they are risk-neutral.
This topic can be difficult to explain, but let me try an analogy:
Let's say there's a 26-mile marathon. Most entrants have trained for the event and will have no problem finishing. Some will finish quickly, some will limp through given enough time, and others will drop out.
Suppose you want to place a bet on whether a single runner completes the race. If you had your own money riding on the outcome, you’d be pretty careful about it – you'd have to know his or her training history, past results and current fitness level, and you’d want to know what the temperature and humidity would be on race day. You’d take a pair of calipers to measure the racer’s body fat. And you’d want to know if someone’s lying to you. If they say they’re running 70 miles per week, you’d ask the neighbors just to make sure.
Now, suppose the organizers said to you, "Why are you betting on a single racer? It’s too risky! Instead, I’ll pay you $x for every person you can convince to enter the race."
Why would the organizers make such an offer? Because they don’t need EVERYONE to finish in order to make the numbers. They just need, say, five-sixths of the entrants to get through the race in order to maintain credibility as race organizers. You see, the organizers also operate a special betting parlor, where big bettors can wager on how what percentage of runners will actually finish. They make enough from the betting action to cover the minimal costs of bringing exhausted and defeated runners to the medical tent.
Eventually, you’d start to use the same calipers, the same neighbors and the same weather report not to disqualify people who shouldn’t be racing, but instead to tell potential racers, "Hey, you’re in about the same shape as so-and-so down the street, who ran a great race last year! Go for it!"
Bottom line: The CMO market pooled the risk but not the responsibility. The diffusion of mortgage risk across the capital markets is where we "crossed the line," rather than as a consequence of the enabling technology. Lenders were risk-averse, but now they are risk-neutral. Therefore, if there are negative externalities stemming from foreclosures by risk-seeking borrowers and their mortgage agents, we may have an argument for regulation to promote risk-averse decision-making.
-----
Some of my earlier coverage on the mortgage market from Bank Systems & Technology.
- Regulators Warn on Home Equity Lending – You can't say that the regulators were asleep at the switch. The current raft of problems was anticipated and expected.
In a guidance issued by the OCC, the Federal Reserve, the FDIC, the OTS and the NCUA, the regulators found that "in many cases, institutions' credit risk management practices for home equity lending have not kept pace with the product's rapid growth and easing of underwriting standards."
- Online Advantage – Whatever happened to RESPA reform and the Guaranteed Mortgage Package? If regulators really want to clean up the incentive structure, this is the place to start. Put the incentives out there in the open, so that borrowers can better understand how everyone in the chain is being compensated.
- Open Skies – One of my first cover stories about the mortgage industry, containing an overblown comparison to airlines and a glossary of terms. It also talks about loan-level pricing, which will become more prominent in the emerging environment.
Labels: finance, mortgage, NYT, wall street
Subscribe to Posts [Atom]
