Amidst the continued tightening within the labor market, the recent ongoing volatility in the investment markets, and uncertainty over interest rates and global economics, I wanted to take a step back to see where we are in the housing sector and assess the progress we’ve made in several key areas since the end of the last recession. It’s no secret that housing has improved, but what is the data really saying about where we’ve come from and where we’re going?
Sales. Existing home sales, reported by the National Association of Realtors, are currently running at a 5.36 million seasonally-adjusted annual rate, based on the October data. Since the beginning of 2014, the annual rate has increased approximately 10%, and the rate of existing home sales has increased some 55% off July 2010 trough levels (3.45 million). Home sales have been in a general uptrend since 2010, although a modest correction toward the end of 2013 did stunt growth somewhat. For all this improvement, though, existing home sales are just now regaining levels not seen since the second quarter of 2007.
New single family home sales, a much smaller component of the housing market, are running at a 495K annual rate as of October, up 76% from the May 2010 post-recession trough, and up slightly more than 12% from the beginning of last year. Although builder sentiment has certainly picked up considerably in recent years, the level of new home sales is running well below the average dating back to 1970, and remains entrenched close to historical recession levels. Highlighting the immense impact of the financial crisis on the residential real estate industry, the annualized volume of new home sales dropped roughly 80% from the peak to the trough. New home sales are still 64% below peak levels.
Single family homes “for sale” are up 59% from July 2012 trough levels, but remain 60% below their May 2006 peak and still hover close to 1970 levels.
nventory. As of October, housing supply was 5.6 months, somewhat below the roughly 7.5 month average dating back several decades, and well below the 14.3 month peak in February 2009. Total homes counted in supply was 2.1 million, down from roughly four million in the middle of 2007 and, although in somewhat of a multi-year decline, supply nevertheless appears to be in a stable range. Homes sitting vacant in inventory have certainly declined with the drop in foreclosure activity, and are now about 8% below third quarter 2009 levels and near 1.9% of all units, according to the Census Bureau. Rental inventory is also down considerably in recent years, from a recession peak of more than 11% to a current level of 7.3%, despite robust multi-family construction growth.
Starts. Housing Starts have hit a post-recession peak at 1.06 million (versus 478K in April 2009), but remain well below the 2.1 million February 2006 peak AND remain near troughs for prior cycles (1966, 1975, 1982, and 1991). Single family starts as of the end of the third quarter of 2015 were 203K, up a robust 160% dating back to the first quarter of 2009; conversely, multi-family starts were up an even greater 495% to close to a 27-year peak.
Prices. The housing price index produced by the Federal Housing Finance Agency (FHFA) ended the second quarter about 18% above its trough level in 2011 and about 10% higher than the official end of the most recent recession. The 20-city Case Shiller index has trended positive since early 2012 and is up almost 28% from the end of the recession. Since the beginning of 2014, the index is up about 7%, and directionally, continues to move higher. Recent research by Moody’s reported that 38% of metro areas had prices above their pre-recession peak at the end of September, up from 30% a year ago.
Affordability. The Housing affordability index from the National Association of Realtors still supports industry growth, although the index is down almost 25% from its early 2013 peak due primarily to continued home price growth, offset by persistently low mortgage rates. However, from the trough in housing affordability in July of 2006, the index has gone up almost 60% and is up 26% from the beginning of the recession.
Household Net Wealth. Household total financial assets were up 3.4% year-over-year during the second quarter, which continues to exceed the rate of growth of household liabilities (2.4%). Consequently, net wealth for all US households continues to grow post-recession, ending 2Q at a combined $85.71 trillion. Household net wealth is up an astounding 56% from the 1Q09 trough ($55.04 trillion), more than eclipsing the massive (financial crisis-driven) 19% decline from the third quarter of 2007 through the first quarter of 2009. Real estate, which accounts for about 29% of that net wealth, ended the second quarter 36% off trough levels dating back to 2011, and is up about 10% from the beginning of 2014.
Home equity has doubled to $12.1 trillion since the trough was reached in national housing prices in 2011. As the Wall Street Journal reported a couple weeks ago, homeowner equity as a share of real estate values is nearing pre-downturn levels. The WSJ noted however, that home equity loans, lines of credit and cash out refi’s, remain at depressed levels although higher year-over-year.
According to the Federal Reserve, home equity as a percent of real-estate values as recently as the second quarter was 56%, still below the pre-recession level. This indicates mortgage debt remains higher than it probably should be considering the stark rise in property values and the general improvement in the economy and in employment.
Cash-out refinancings are up almost 50% from a year ago, but these are being done by borrowers with much more existing equity in their homes and by borrowers interested in taking on less risk. In fact, the WSJ reported that resulting debt following the average refinance during the three months ended August was 68%, well below cyclical peaks. Home equity per homeowner has nearly doubled since the trough in 2011. However, Moody’s reported that the impact of the rise in home equity on the economy was about a third of what it was before the credit crisis.
Mortgage Market. According to the Federal Reserve, household mortgages outstanding ended October at $9.413 trillion, down 12% from the $10.7 trillion peak during the first quarter of 2008 and up only slightly (0.4%) from the most recent second quarter trough level ($9.37 trillion). Only recently has mortgage volume turned positive, following a slow thaw in credit standards as indicated by senior loan officer surveys (it is important to note that credit remains very tight with borrowers with weak credit).
The delinquency rate on single family mortgages at the end of the third quarter was 6.23%, reflecting a continued drop from the interim 2012 peak of 11.97% and the post-crisis peak of 12.50% during the first quarter of 2010. Meanwhile, CoreLogic indicated that the percentage of homeowners underwater in their mortgage had dropped from 21% at the end of 2011 to just below 9% at the middle of the year.
Demographics. According to recent work by Yardeni Research, the yearly change in the number of households in this country is running about 1.5%, as of September. This is up meaningfully from the roughly 0.5% annual level of chance following the recession; however, household growth is still well below historical growth rates and still in secular decline.
Yardeni’s work also shows a strong correlation between the growth in households and the growth in the civilian working age population, which we already know from analyzing the unemployment rate that secular forces related to an aging workforce are at play. In fact, the correlation between households and the civil labor force participation rate is equally strong, and if the secular trends on participation are indeed valid, then one might expect continued weakness in the growth of total households as a percent of the working age population.
The silver lining is that 2015 has witnessed the strongest household growth in more than ten years, and may reflect a modest reversal should the economy continue to improve. Interestingly, and perhaps not too surprising, despite the more recent uptick in household formation, the vast majority of these are renters, not homeowners, according to the Census Bureau and Haver Analytics. This naturally impacts the demand for entry-level housing. Moreover, a growing percentage of the younger population cohorts are renting now, and as Yardeni points out, renting households now account for more than 36% of all households from a 2004 trough level of below 31%.
Looking Behind the Drywall. The prognosis for housing, on balance, appears favorable. With economic growth expected to be ~2.5% next year, and unemployment falling below 5%, the rate of job growth and new household formation should drive demand for housing and at least some housing price support. There’s nothing to suggest that housing starts won’t continue to increase over the next couple of years, but the rate of house price growth does appear unsustainable.
A 1.0-1.25% growth rate in the roughly 118 million U.S. households, assuming it is near-term sustainable, implies an improving rate of housing starts at worst consistent with recent trends, despite the rise in renters. To be sure, the stability in household formation is by no means a sure thing, but the good news is that housing demand over the last few years has outstripped supply in many major markets – even if that housing supply imbalance has led to generally unsustainable home price inflation.
The National Association of Home Builders (NAHB) survey remains firmly in positive territory. However, many builders are once again now reporting labor shortages, which implies that the growth in the housing stock may (continue to) fail to keep up with demand in some areas.
Prices could be hard to predict, though, and it is hard to tell what impact the next Fed tightening cycle will have on the buyer motivation. However, a slow tightening trend, which has been widely telegraphed, could have a fairly benign impact on housing. More importantly, slower incremental gains in employment as the economy gets closer to “full” employment, coupled with the potential for new wage pressures and its associated impact on inflation and expectations, could drive the ten year treasury higher, as well as the mortgage rates tied to it. This would reduce affordability, and perhaps dampen the growth in demand.
The 2016 outlook from Merrill Lynch is more or less consistent with my assessment, which I believe is reasonable at this juncture. They expect almost 1.3 million housing starts, a 5% increase in existing home sales, 10% growth in new home sales and price growth of only ~2%. Importantly, one of the more insightful points which Merrill makes is that housing, which usually leads the domestic economic cycle, has indeed lagged this time around, despite the Fed’s zero interest rate policy (ZIRP). Getting off the “zero bound” leads us somewhat into uncharted waters, and with elevated prices in many markets, it will be important to monitor the laws of unintended consequences as we move ahead.
charts courtesy St. Louis Federal Reserve, Yardeni Research
