some economists, together with policymakers at the White House and Federal Reserve, are warning that mortgage-lending standards have become too restrictive, years after carelessness by lenders inflating the housing bubble.
… the Urban Institute, a think tank in Washington, estimated that around 200,000 fewer mortgages were made in 2012 due to credit standards that were more stringent than they were before the housing bubble.
Don’t say that like it’s a bad thing!
In mortgage lending, a type I error is making a loan that defaults. A type II error is turning down a loan that would have been repaid. Some remarks.
1. When home prices rise, it very hard to make a type I error and very easy to make a type II error. If the house price has gone up, the borrower’s equity is presumably positive, and borrowers who get in trouble will sell their homes and pay off their loans.
2. During the bubble, Congress and HUD officials were convinced that lenders were making type II errors due to racial bias and anachronistic conservative lending standards.
3. After the crash, Congress and government officials were convinced that lenders had been making type I errors due to greed, predatory lending, and lack of regulatory oversight.
My own view is that lenders make type I errors and type II errors all the time. Still, I would rather have errors made by people for whom credit risk assessment is a profession, without the amateur meddling that comes from the political sector.
Journalists believe that every economic statistic has a ”good” sign and a “bad” sign. Higher prices are bad. Higher unemployment is bad. More mortgage lending is good. More house construction is good.
Economists usually believe it is best if the number reflects the least distortion possible. So it’s hard to know for sure whether for 200,000 mortgages should have been made. What is “too stringent”?
The credit flowing to housing could also go towards productive things rather than bidding up house prices, so there is also that.
Yes, but voters who own houses will be happier if the values of their homes keep going up. People who might benefit from more productive uses of credit generally don’t know who they are.
I think 2 is wrong. They thought this because the investment banks were grabbing market share away from the gses, or the gses were obstacles to the investment banks gaining even more, depending on political position.
Incentives also matter. Examples from real life: Banker A, below board level, gets bonused on loan production with no take back for loans which go bad after 1 year. Bank loan committee relies on Banker A for data about loans. Bank produces large number of loans which go bad and fails. Banker A has received bonuses for writing bad loans, including loans to cover bad loans. Paper work was produced by Law firm B. Bank fails. Banker A is still employed by takeover bank who the FDIC guaranties 80% of losses. Banker A and Law firm B are now paid by bank to “recover” losses from bad loans. Rather than settle/take back collateral they chose to not foreclose, run up legal bills, and increase “losses” which FDIC will pay. Bank gets a 4:1 return on “loss”, Banker A gets more bonus, Law firm B gets more fees, borrowers are better off not paying interest and delaying bankruptcy, FDIC doesn’t care/is asleep. Rinse and repeat. FWIW- Bankers and Lawyers brag about playing the incentives. (Classic agency problem also). Type 1 errors vs Type 2 can be balanced only if the credit professionals and enablers have skin in the game. ie, if you get paid by the loan you must pay back when it goes bad. risk vs reward.