We both went to Swarthmore College. Maybe that explains why we are on the same page on what to do about Freddie and Fannie, namely nothing. He writes,

In the years immediately following the financial crisis, there were few people who would have bet that, in 2018, the GSEs would still be in a conservatorship in which they are effectively publicly owned companies. Nonetheless, inertia has proven to be a powerful force. In this case, I would argue that it has been a force for good in the housing market. The current system is one that minimizes the problem of moral hazard in housing finance, which was so important in the run-up in prices in the housing bubble years. It also is the most efficient mechanism for financing mortgages, as it requires fewer resources to be wasted on housing finance.

I disagree about getting rid of mortgage-backed securities. I approve of risk being transferred to the private sector.

Go back to what I wrote in November.

the current system works about as well as one could hope. With the backing of the Federal government, agencies like Freddie Mac and Fannie Mae can keep the market supplied with adequate capital to provide 30-year fixed-rate mortgages. At the same time, by using securitization and credit risk transfers, the agencies off-load the greatest proportion of risk to private entities.

Of course, if you gave us carte blanche to change the system, Baker and I would go in very different directions.

Scott Sumner writes,

OK, if was obvious that home prices were wildly excessive in 2006, why is that not also true today? Nominal house prices are now far above 2006 levels, and even in real terms they are rapidly approaching the 2006 peak

My thoughts:

1. Housing starts have been in the toilet for a decade. Looking at these charts, single-family housing starts were at an annual rate of over 1.2 million starting in 2001, and they were over 1.6 million from 2003 through 2005. Single-family starts plummeted to around 500,000 per year from 2008 through 2012, and they barely topped 800,000 last year.

Over the last decade national average rental costs have risen faster than inflation. As I recall, they were rising more slowly than overall inflation from 2001 through 2007.

In short, compared with 2006, high housing prices today look more like a supply problem and less like a bubble.

Update: Commenter Handle points to the price/rent ratio chart, which clearly tells you that we are not at 2008 levels.

2. The U.S. has multiple housing markets. Looking at the Core-Logic price index page, the cities that have seen big price increases in the past decade include San Francisco, Boston, Denver, and Seattle. San Francisco and Boston are notoriously supply constrained. Denver and Seattle also are more likely experiencing long-term increases in demand relative to supply. The “sand states,” where housing performed the worst during the crisis, have lower prices than they did ten years ago, in some cases much lower.

In short, today we are seeing high prices in areas where income growth has been very strong. In 2005 and 2006, we saw higher prices driven by looser credit conditions.

In testimony before a House committee, Mike Lea writes,

The standard mortgage instrument in other countries differs significantly from the US FRM. The standard product in Canada, Germany and many other European countries is a short to medium term fixed rate mortgage sometimes referred to as a rollover. The rate is fixed for a 1 to 5 year period (up to 10 years in Canada and Germany) after which the rate is reset at the current market interest rate. The loans have amortization terms of 25-30 years. The borrower can select the same or a different fixed rate term at reset. This feature allows borrowers some protection against potential interest rate shocks (e.g., if the reset rate is high and the borrower expects it to fall she can select a one year fixed rate term; conversely if she believes rates are low and likely to rise she can opt for a 5 or 10 year fixed rate term). There is a prepayment penalty during the fixed rate period (a yield maintenance penalty that removes the financial incentive for refinance).

…Mortgage loans are recourse obligations in all of the countries surveyed and default rates have been and are significantly less than in the US. With recourse lenders have the right to pursue deficiency judgments against borrowers providing a significant deterrent to mortgage default.

If you are at all interested in the subject, read the whole thing.

If we want a housing finance system that is as safe as that of, say, Canada, then we need to evolve away from the 30-year fixed-rate mortgage. But if you take the 30-year fixed-rate, no-recourse mortgage as a given, then my line is that our current housing finance system is about as safe as we are going to get.

Back when they were planning these hearings, they solicited my testimony. They ended up not soliciting it for the actual hearing. I am not complaining, because I don’t glory in giving this sort of testimony. But below is what I wrote. Continue reading →

He responds to my post.

So, in the highest priced cities, middle class buyers were an insignificant part of the market, but when prices in those cities shot up and then collapsed, our main policy response was to prevent middle and lower-middle class households from being homeowners in places like Texas, where they never posed a problem.

Read the whole thing. The main take-away is that if you thought in terms of just one housing market, as policy makers in Washington did, you could not possibly make good decisions.

Commenter Handle is skeptical of the revisionist view that investor loans rather than subprime lending fueled the housing bubble and bust.

how does debt to high risk borrowers stay constant while everyone knows that underwriting standards dropped a lot? Why wouldn’t a drop in standards have scooped up a lot of marginal borrowers and expanded the debt in that category?

Many possibilities.

1. The loosening in underwriting standards may not have been as deep and widespread as people have come to believe.

2. Perhaps looser standards did not draw in many borrowers, because people self-rationed. Most people fear taking on a lot of debt, even if lenders are offering it to them.

3. Perhaps the authors of the revisionist papers are deceiving themselves by the way that they look at trends in debt among homeowners. If debt was going up for some borrowers and down for others, then on average you might not see a debt increase. But the debt might have increased among borrowers less able to carry it. Note that the ability to carry debt depends on many factors, including factors that are not observable to economists researching the issue.

4. An increase in house purchases by affluent speculators and an increase in house purchases by sub-prime borrowers are not mutually exclusive explanations of the boom and bust. It could be that both phenomena together were important. The cycle would have been much less extreme if either of those phenomena had not taken place.

5. Perhaps the biggest shock was not the loosening of standards prior to 2008 but the tightening of standards subsequently. As the “subprime crisis” unfolded, politicians hit lenders with new rules and, more important, harsh rhetoric about “predatory lending” that discouraged lenders from offering a mortgage loan to anyone who actually needed one (lending to people with plenty of assets was still ok). Perhaps if underwriting standards had merely reverted to those of 2002, home prices would have stabilized at a higher level.

I lean toward (3) and (4). Maybe someone (Kevin Erdmann?) can talk me into (5).

Stefania Albanesi, Giacomo De Giorgi, and Jaromir Nosal write,

A broadly accepted view contends that the 2007-09 financial crisis in the U.S. was caused by an expansion in the supply of credit to subprime borrowers during the 2001- 2006 credit boom, leading to the spike in defaults and foreclosures that sparked the crisis. We use a large administrative panel of credit file data to examine the evolution of household debt and defaults between 1999 and 2013. Our findings suggest an alternative narrative that challenges the large role of subprime credit in the crisis. We show that credit growth between 2001 and 2007 was concentrated in the prime segment, and debt to high risk borrowers was virtually constant for all debt categories during this period. The rise in mortgage defaults during the crisis was concentrated in the middle of the credit score distribution, and mostly attributable to real estate investors.

“Real estate investors” means buyers of houses who did not intend to occupy them but were instead buying them for speculation. The non-owner-occupied phenomenon (also known as investor loans) was noted by Andrew Houghwout and others in 2011, and it was something that I suspected back in October 2008. Kevin Erdmann also has pointed out that it was not low-income borrowers who drove the boom and bust.

Alex Tabarrok writes,

Over the entire 20th century, housing prices never once roce above 131, the 1989 peak. But beginning around 2000 house prices seemed to reach for an entirely new equilibrium. In fact, even given the financial crisis, prices since 2000 fell below the 20th century peak for only a few months in late 2011. Real prices today are now back to 2004 levels and rising. As I predicted in 2008, prices never returned to their long-run 20th century levels.

As a matter algebra, house price = (rent) x (price/rent). If this time is different, it is due to a combination of higher rents and higher price/rent ratios. As I read the graph of prices against rents at the Economist, it looks like prices have been rising faster than rents, so an increase in the price/rent ratio is certainly a contributing factor, although more in other countries than in the U.S.

I think that in many parts of the world, including portions of the U.S., price/rent ratios are getting very high. If I were Alex, I would not stand in front of the sign that says “mission accomplished” claiming to have debunked the house price bubble.

About a month ago, the Harvard joint center for housing studies released a report. Lots of interesting stuff.

housing completions in the past 10 years totaled just 9.0 million units—more than 4.0 million units less than in the next-worst 10-year period going back to the late 1970s. Together with steady increases in demand, the low rate of new construction has kept the overall market tight, leaving the gross vacancy rate at its lowest point since 2000

Yes, Kevin Erdmann, you are right. We have really not built enough housing since 2007. Also, I don’t understand why I can’t make money investing in REITs. Rents are rising, interest rates are low, . . .

On average, 45 percent of renters across the nation’s metropolitan areas can afford the payments on a median-priced home in their market area, but the shares range from less than one in ten in the high-cost markets concentrated on the Pacific Coast as well as in Florida and the Northeast, to two-thirds or more in low-cost metros in the Midwest and rural South. In areas where homebuying is well out of reach for a large majority of renters, there is much less potential for increases in homeownership.

This is a fascinating divergence. The ratio of rent to price is comparable to an interest rate. Having vastly different rent-price ratios across regions is sort of like having vastly different interest rates across regions, except that there is no way to arbitrage against it.