When will everything be back to normal in housing?
The lingering effects of the housing bust still distort market action eight years after the crash.
A normal market would be characterized by a steady or increasing values for each of the following:
- home ownership rate,
- home sales,
- first-time homebuyer participation,
- purchase originations
- new home construction, and
- household formation
Perhaps one or two of these metrics may be low or declining due to unusual and short-lived circumstances, but a healthy, “normal” market would see strength in these areas. Unfortunately, we witness the following:
- 20-year lows in home ownership
- 6-year lows in home sales
- 30-year lows in first-time homebuyer participation
- 20-year lows in purchase mortgage originations
- New home construction less than 50% of normal
- Household formation less than 50% of normal
Doesn’t the word recovery imply improvement? When people recover from illness, their health improves; shouldn’t a housing recovery show improvement in multiple areas, not just price? So far, price is the only indicator of improvement, assuming high prices are better than affordable prices, of course.
Several concrete signs lately point to a broad-based housing recovery across the nation: Home sales are increasing, prices are rising, credit access is gradually easing and private capital is slowly coming back. But not everything is back to normal: credit availability is still very tight, delinquencies are still elevated and taxpayers still back the overwhelming majority of the mortgage market. With such contradictions and no single indicator of market health, how will we know when the market has recovered enough?
Since realtors and builders want to lower lending standards to increase commissions and sales, the credit-is-too-tight meme refuses to die. Despite industry spin, mortgage lending standards are not tight, which is alarming considering that the US taxpayer backs most of the loans.
In 2001, home prices were rising moderately, mortgage credit was easier to obtain (but by no means loose), underwriting was sound, private capital was abundant and default and foreclosure rates were low. This was a healthy housing market, but how does today’s market compare?
2001 is a reasonable benchmark year, but not the best. The market conditions in 1998 or 1999 would be a better comparable. By 2000 lending standards loosened up enough that California began to elevate prices above the stable relationship between the cost of owning and renting.
- The government has a much larger share of the market. In 2001, the Federal Housing Administration, Department of Veterans Affairs, and the government sponsored enterprises together backed about 52 percent of all first-lien mortgage originations while private capital financed the remainder. In 2014, the government-backed share was 71 percent after peaking at nearly 90 percent in 2009. Obviously we have come a long way from 2009, but the government share is still too high and unsustainable.
As the chart above shows, securitizations without government backing abruptly stopped in 2007, and that market has seen almost no recovery over the last eight years. Realistically, to lure private capital to the mortgage market, interest rates must rise, and since that won’t help reflate the housing bubble and bail out the banks, policymakers keep the current regime in place.
- It’s much harder to qualify for a mortgage. The Housing Finance Policy Center’s Housing Credit Availability Index measures the expected default risk of all mortgages originated in a given year – a measure that precisely reveals how hard it is for borrowers to qualify for a mortgage. The default risk of mortgages originated in 2001 (and from 2000 to 2003 in general) was about 12.5 percent. As the housing bubble began inflating along with riskier lending, the expected default risk increased to nearly 17 percent. Post-crisis however, as lenders switched to originating only the safest mortgages, the expected default risk fell rapidly to just 4.6 percent in 2013. Access to credit has since started to improve slowly with default risk increasing to 5.7 percent in early 2015. While this level is encouraging, it’s less than half of the risk the mortgage market took in 2001. Indeed, today’s lending environment continues to remain extremely tight compared to the cautious standards of the 2000 to 2003 period.
The graphic above does not tell the full story. The elimination of destabilizing loan products is not a tightening of standards, it’s a recognition of how dangerous these products are even in small doses.
- There are fewer mortgages being originated. Approximately 4.7 million first-lien purchase mortgages were originated in 2001. In contrast in 2013, the latest year for which HDMA data are available, only 3 million such mortgages were originated. Our previous research shows that a big reason for this decline is the extraordinarily tight lending standards of the post-crisis period. Specifically, the mortgage market would have made an additional 1.2 million loans in 2013 if the cautious lending standards of 2001 had been in place.
This basic premise is wrong. Qualifying more people by lowering standards or encouraging the use of toxic loan products is not the answer, it’s the recipe for inflating another housing bubble.
- The rate of seriously delinquent mortgages is much higher. When a mortgage remains unpaid for more than 90 days or goes into foreclosure, it is considered seriously delinquent. Although the rate of seriously delinquent mortgages has declined recently as house prices have recovered, it still remains high relative to 2001. Loans that are more than 90-days delinquent or in foreclosure comprised 4.2 percent of all outstanding mortgages in the first quarter of 2015, down from nearly 10 percent in 2009, but still elevated from 2.4 percent in 2001. This reflects the hangover from the foreclosure crises, as new origination is pristine.
People were considered crazy if they suggested in 2006 that mortgage delinquency and foreclosure rates would increase ten-fold and even nine years later would be more than triple historic norms. The mess isn’t cleaned up yet.
We continue to move in the right direction, but progress has been uneven. So how will we know when the mortgage market has recovered enough and is healthy once again? When the government share of the mortgage market is closer to 50 percent, when the expected default risk is closer to 12.5 percent, when first-lien mortgage originations are closer to 5 million, and when the rate of seriously delinquent mortgages is closer to 2.4 percent. In other words, when it looks more like 2001.
Trulia’s economists added some additional metrics:
A normal housing market was characterized by a balance between supply and demand. The supply of housing was controlled by millions of individual homeowners with sufficient equity to sell whenever they pleased without obtaining a bank approval, and banks owned almost no REO. Historically, lenders had very little direct impact on the housing market through short sale approvals or REO sales.
The demand for housing also came from millions of individuals, about two-thirds of whom borrowed around 80% of the purchase price to acquire property. New household formation from high-paying jobs prompted builders to provide more rental and for-sale products wherever job growth occurred, and low unemployment kept house prices steady even in areas of temporary economic distress.
The new normal in housing is very different from the old normal.