Tuesday, June 5, 2018

Housing: Part 302 - Austan Goolsbee was right

I saw a link to this March 2007 New York Times Op Ed by Austan Goolsbee, titled "‘Irresponsible’ Mortgages Have Opened Doors to Many of the Excluded" on Twitter. (HT: NT*)  The point of the tweet was that it isn't fair to judge people with gotcha's about things they said before the crisis, because most people were caught off guard, and this op-ed is an example of something that Austan Goolsbee, a respectable economist, wrote that everyone now knows was horribly wrong.

The interesting thing is that the op-ed that is used as an example of something obviously wrong was presciently correct.  It is something Goolsbee should, and I hope someday will again, be proud of.

Goolsbee wrote (in 2007):
Almost every new form of mortgage lending — from adjustable-rate mortgages to home equity lines of credit to no-money-down mortgages — has tended to expand the pool of people who qualify but has also been greeted by a large number of people saying that it harms consumers and will fool people into thinking they can afford homes that they cannot. 
Congress is contemplating a serious tightening of regulations to make the new forms of lending more difficult. New research from some of the leading housing economists in the country, however, examines the long history of mortgage market innovations and suggests that regulators should be mindful of the potential downside in tightening too much.
This is exactly what happened.  And we have paid dearly for it.

Compare this to statements from Case & Shiller in their 2004 paper "Is there a bubble in the housing market?":
(J)udging from the historical record, a nationwide drop in real housing prices is unlikely, and the drops in different cities are not likely to be synchronous: some will probably not occur for a number of years.  Such a lack of synchrony would blunt the impact on the aggregate economy of the bursting of housing bubbles. 
This is not what happened.

Yet, here we are in 2018, and Austan Goolsbee is supposed to be embarrassed by what he wrote while Case & Shiller are commonly credited for calling the bubble.

idiosyncraticwhisk.blogspot.com 2018
Source: Zillow.com: Median prices by zip code, sorted by IRS income
Here is a chart comparing home prices for the entire period from 1998, before the price run-up, to 2013, when prices were near the bottom.  Does this look like the picture of a housing market where a few high priced cities finally fell back to earth?  Or, does this look like a market where we took a billy club to low-tier lending markets until they relented?  As I have pointed out previously, there is usually very little difference within metro areas between price appreciation of high tier and low tier markets, on average, over time, because, contrary to conventional wisdom margin shifts in credit access just don't have that much of an effect on prices in functional markets.

idiosyncraticwhisk.blogspot.com 2018
Source: Zillow.com: Median prices by zip code, sorted by IRS income
How can I say that?  Well, in most cities, during the boom, there wasn't much difference between high tier price appreciation and low tier price appreciation (here shown from 1998 to 2006).  So, either there was a credit bubble in the 2000s and it didn't have much of an effect on low tier prices, or there wasn't a credit bubble in the 2000s.  (The reason the rise in low tier prices is limited to the Closed Access cities likely has little to do with credit markets, which I explain here.  But we can infer that it has little to do with credit markets just by looking at these graphs, since it would be odd if only a handful of cities were affected by the credit bubble and those cities were the only cities where long term prices in low tier neighborhoods held up well.  You could plausibly explain that away by arguing that in the Closed Access cities, a credit bubble led to price increases while in other cities it led to oversupply.  But for oversupply to explain this scale of a price drop, surely tenant vacancy levels would be high while subsequent low tier rent inflation was low - and both would be extreme.  Yet, in most cities, there is nothing.  Rents are rising and vacancies never rose.)

My point here is to note that in order for credit markets to really move prices between market tiers, the shift in credit access has to be extreme.  The credit boom in the 2000s that appeared extreme had little effect within metro area housing markets.  To knock 30% off of low tier prices in a city, compared to high tier prices, you really have to apply Godzilla level shifts to the market.

Goolsbee referenced this paper from Kristopher Gerardi and Paul S. Willen from the Federal Reserve Bank of Boston and Harvey S. Rosen of Princeton.  They write, "We find that over the past several decades, housing markets have become less imperfect in the sense that households are now more able to buy homes whose values are consistent with their long-term income prospects."  In hindsight, this was absolutely true and continued to be true during the bubble.  According to the Survey of Consumer Finance, the rise in homeownership was focused on households with high incomes, college degrees, and lucrative careers.  And, even the rise in distressed levels of debt (debt payments greater than 40% of income) that happened at the end of the boom, was among households with top incomes.  Those were the households that accounted for much of the small, early wave of defaults in 2007.  Middle and lower-middle income borrowers who are the type of borrower that we might think of as credit constrained in a functional market, accounted for much of the later, much larger wave of defaults that happened well after we went and did exactly what Austan Goolsbee told us not to do.

"Also, the historical evidence suggests that cracking down on new mortgages may hit exactly the wrong people....When contemplating ways to prevent excessive mortgages for the 13 percent of subprime borrowers whose loans go sour, regulators must be careful that they do not wreck the ability of the other 87 percent to obtain mortgages."  I didn't say that.  Austan said that, in 2007.  And it's even worse than he feared.  Originations to borrowers with FICO scores above 760 have continued at at least the same pace since 2007 as they had been issued from 2003-2007.  Originations below 720 have been cut by two-thirds, even though there had been no increase in low FICO score originations during the boom compared to high FICO scores.  Even lending to FICO scores between 720 and 760 has dropped by more than half.  We didn't stop at 87% of the subprime market.  We killed half of the prime market.  That's how much life you have to stomp out of credit markets to get low tier markets to collapse.  There isn't anything subtle going on here.

Austan.  You were right.  Claim your crown.

* The original tweet isn't there anymore, but the tweets and replies to it remain.


  1. But you can't say lending to people who cannot deal with toxic mortgages was a good thing either, Scott. Hey, I have a Socal example of the problem with density. Google is locating a campus near Manhattan Beach. Thoses prices are already too high. Why didn't Google locate in Claremont? Lots of smart people there. There is an argument to be made that sprawl lowers house prices. Lol.

    1. There are certainly better ways to construct a mortgage market, but I think their place as a cause of the bust is overstated.

    2. I agree. There was a liquidation by the Fed that made the housing crash turn into the Great Recession. MMers have proved that.

  2. I agree with this post.

    Seems to me the "easy money/credit did it" story got legs due to GOP-right-wing argument-framing after the bust.

    The GOP is much smarter than the Donks in framing arguments. After the bust, they framed and dog-whistled the villains as Clinton CRA rules and easy money for low-income urbanites.

    In the GOP version, the government strong-armed banks into lending to inner-city types, or enticed banks into sin through certain give-aways and rules.

    I think this narrative drove perceptions for a long time, maybe to the present.

    The GOP narrative also fit hand-in-glove with the Cato-Mises/Austrian crowd which sees the love of (easy) money as the root of all evil.

    As usual, no one wants to talk about property zoning, and restricted supply. There are no atheists in fox holes, and there are no libertarians when neighborhood property zoning is under review.

    Pre-bust 2008, what I saw in L.A. in mid-city was the middle-class "reclaiming" or "gentrifying" neighborhoods that had gone downscale after the deindustrialization and forced busing of the 1970s-1990s. I wonder if CRA played any role is this gentrifying.

    The process is continuing now.

    The narrative that the Fed choked the economy in 2008 will just not get legs. One of the hoariest and tallest totems in the macroeconomic pantheon is "Tightus Moneyus."

    As an aide, one of the oddities, repeated again and again in government, is how "experts" send us off the rails.

    From Vietnam to prisoner recidivism to monetary policy to many social welfare programs, the experts seem fogged by their preconceptions or biases.

    1. I think you're right, Ben. I would add in right wing animus toward Fannie & Freddie, which also seems to be based on exaggerated notions of their influence on housing markets and which helped fuel the private securitization boom, which was a form of lending that was much more systematically destabilizing when disruption came.

    2. Benjamin, I agree. And even those who fomented the lie that the CRA was the cause of the bubble, and then the crash, like Stossel and Cavuto, repudiated that view, but it was way too late. Fox did a great disservice to the American people.

  3. What we need is the ability for people to get mortgages early in an economic cycle. The problem is that typically credit markets open late in a cycle once people are fully employed and wages are rising. All too often, those same rising wages dent corporate profitability and start a downturn. Pretty sure if you charted mortgage availability it would often be at its highest in the few years preceding recession. I'm not sure how to fix this though.