2017: End of Year US Residential Real Estate Review

Brad Lookabaugh – Quantitative Portfolio Manager

Download a PDF of this report here.

Slow and steady

The latest S&P/Case-Shiller report showed home prices are continuing their steady climb, with the national index increasing 6.2% year-over-year for September 2017. Data for the first nine months of the year have now been released[1] and the index shows a year-to-date gain of 4.2%. National prices have risen in a remarkably stable fashion. This year, all monthly year-over-year readings been in a 0.6% range, and, dating back to October 2015, returns have remained within a range of 4.9% and 6.2%. The Case-Shiller 20-City Composite has been even less volatile. This index, which tracks home prices for a subset of the 20 largest metros in the country, also sports a 6.2% year-over-year increase for September 2017 and had risen at a rate between 5% and 6% for 25 consecutive months until this latest reading. To be sure, part of this low volatility is expected – the index is smoothed by design. Yet, one has to go back to the mid-1990s to find a comparable stretch of steady movements for the national index. The 20-City Composite, which only started in 2000, has no other comparable stretch. Even Zillow, which recently reported a 6.5% year-over-year increase of its national index for October, has had relatively lower variability.

The rare steady “grind” higher in residential real estate is evolving against a backdrop of worry and overbearing caution as the broader economic recovery enters its 10th year. The general narrative increasingly focuses on unaffordability, lack of supply, and impending demographic and industry structural changes. Certain coastal markets have had significant prices increases without commensurate gains in wages for the general populous. With a greater portion of homes out of reach, more people are opting not to “trade-up”. This chokes the supply of lower priced, starter homes that appeal to first-time home buyers. In addition to soft existing home sales, new home sales are repeatedly described as weak for this late in an economic recovery. The homeownership rate has precipitously declined as more people find renting the preferred alternative to buying a home. Despite being a banner year for home builder shares (the ISE Homebuilders Index is up nearly 60% YTD[2]), the companies blame lack of skilled labor and raw material issues as main production bottlenecks. Slow moving demographic tides are also under way. Many millennials are approaching prime home buying age. Baby boomers are approaching retirement in waves, yet many are working later in life. These trends alter historical supply and demand mechanics. Finally, the Federal Reserve is beginning to tighten its monetary policy, and the recently released tax proposal promises changes to mortgage interest deductions.

These topics and trends are important; their collective interactions and impact on the residential real estate market must be understood. Yet, with countless real estate and general economic data frequently released, it is easy to become absorbed in the recent past and the immediate future. Consistently focusing on the noise does little to accomplish the goal of comprehending structural market changes. This letter will provide a perspective on the market dynamics of not just this past year, but of the recovery more broadly. With the big picture in mind, we offer a fundamental view of the market for 2018 and beyond.

Fundamental forces of nature

One would not know it from headlines, but this housing recovery is built on a solid foundation. It has matured with broad price stability and its financial underpinnings have grown safer and more robust.

This price stability is evident on a national and metro level. The figure above shows the monthly nominal Case-Shiller national price index on a log scale, overlaid with several cumulative annual growth rate (CAGR) paths. These paths show where the index would be if the index grew at given annualized rates from the base period. The Case-Shiller has historically travelled along the 6% CAGR path. After peaking in 2008, the index was knocked into the 4-5% paths. This chart gives context to the recent low volatility; it has persisted for the past 5-7 years as the index increased in parallel to the 5% path. The national index behavior during the recovery since 2011 is quite normal relative to its 40-year history. In fact, it appears to be one of the healthier stretches in its history, since there is no evidence of a sharp increase, as seen in the late 1980s and 2000s.

Examining one level deeper, we see that most major metro areas that have had strong house price appreciation (HPA) have also had strong economic and population growth.

This next chart shows the relationship between nominal[3] GDP growth and HPA for each of the 20 metro areas in the Case-Shiller 20-City index during the first five years of this cycle[4]. The size of the bubbles indicates the total population growth rate over the period. There are several trends to highlight from this chart:

  • All 20 metros experienced about 4.5% annualized GDP growth, 6.0% HPA, and modest population growth. It is a reasonable benchmark to anchor to.
  • West coast markets have outperformed. Strong HPA in the Seattle and San Francisco markets are supported by its population and economic growth in this recovery. These markets also have more unique supply constraints due to geographical characteristics, congested transportation infrastructure, and regulatory concerns. Portland is a notable outlier, with high HPA and lower economic growth; however, its population has grown by over 7%, placing pressure on prices.
  • Markets with outsized population growth have seen strong HPA. Phoenix and Las Vegas were devastated by the crisis; they are liquid markets that began this recovery with a very low base. Despite having softer downturns and room to build, the influx of people into the Denver and Dallas areas has outpaced supply.
  • New York, Boston, Chicago, and Washington DC are markets that actually grew slower in this time period. While pockets of these areas have experienced strong HPA, in aggregate, they have had lower population growth, slower economic growth, and a large supply of land to expand.

The chart above demonstrates that the economic and demographic trends, coupled with supply constraints for certain markets, have supported the realized HPA for the top 20 metros between 2011 and 2016. Although we do not yet have visibility into the GDP or population changes for this past year, certain markets have had HPA shifts from the 2011-2016 trends (see table to the right):

  • Several markets have cooled after a strong run-up from the depths of the crisis, most notably Phoenix, Miami, Denver, and San Francisco.
  • Las Vegas and Portland have maintained their hot pace. Markets such as Detroit and San Diego have slowed slightly, but continue to rise at a healthy pace.
  • Other markets have strengthened. The one attracting most attention is Seattle; it is up over 9% YTD. The Charlotte metro had plenty of supply to soak up in the beginning of the recovery. Despite strong economic activity and nearly a 10% increase in population, the annualized 5-year return through 2016 was only 4.5%. This market has accelerated in 2017 as much of that open supply has been used up.
  • New York, Cleveland, DC, and Chicago continue to significantly lag the broader market.

This economic recovery has perplexed policy makers and investors alike. The stable, slowly increasing improvement over the last ten years has frustrated those that expect the “normal” periods of 4%+ real growth of yore. The recovery in residential real estate has not been much different. Lasting structural changes have formed in the economy at large, altering the way deep seeded markets were thought to have worked. This residential real estate bull cycle has characteristics that mirror those in the broader economic recovery. We have shown this from an economic perspective. There are also similarities from a pure financial standpoint.

Building machines

The aftermath of the Great Recession brought waves of regulation and changing moods in the lending industry. One side effect of tightening credit conditions and risk averse lending behavior by creditors is lost opportunities for potential borrowers, in this case home buyers or owners. However, an important outcome, particularly as a market re-orientates itself, is the creation of a strong financial base for the market to stand on and grow.

The current credit conditions have not changed in the way that the recovery would imply. Fannie Mae and Freddie Mac report that, of the loans they purchased between 2011 and 2016Q2, over 60% were originated for borrowers with FICOs above 750. This compares with roughly 40% for loans originated in 2007 and 33% for those from 1999-2004. Whereas over 30% of loans were originated with FICOs under 700 in a given year from 1999-2007, approximately one in ten have fallen into that category during this recovery[5]. This trend appeared in 2009-2010 as a “knee-jerk” reaction to the crisis, but it has persisted through the years. The wounds from the crisis run deep for lenders. In general, underwriting conditions have stayed conservative across the nation, but there are notable differences across metros. Some of the hotter markets (e.g. San Francisco, San Jose, Seattle) report higher average FICOs and lower average origination loan-to-value (LTV)[6]. While this exacerbates the lack of affordability, it also is evidence that these fast appreciating markets have not been fueled by risky lending.

Another feature of the financial crisis was the prevalence of riskier and exotic products. At the height of the mid-2000s boom, as much as 42% of new originations were adjustable-rate mortgages. Many of these had “teaser” rates and were processed for borrowers with unverified or lack of income or assets. Today, nearly 90% of all originations are 30- or 15-year fixed rate mortgages. Standard products have comprised the majority of originations during this recovery.

With home prices rising and credit standards remaining tight, it is not surprising that serious delinquency[7] and foreclosure rates have slowly fallen back to near pre-crisis levels. The percent of homes at or near negative equity (i.e. LTV approximately or greater than 100%) has declined from a high of around 30% of all residential properties in 2011 to under 7% today[8].

While the securitization industry continues to normalize, Fannie Mae and Freddie Mac continue to transfer risk from their enormous portfolios to other parts of the financial ecosystem. They have substantially reduced their balance sheets, resulting in a less vulnerable system. The balance of their combined portfolios is now less than 40% of its size in 2011[9]. Both are required to reduce their respective balances to $250B by the end of next year. Fannie Mae is already below this threshold.

Into the unknown

Several of the aforementioned concerns – rising mortgage rates, supply and affordability issues, and structural demand changes – will continue to evolve through the end of 2017 and into 2018.

Many expect mortgage rates to increase in 2018 as the Fed continues to tighten monetary policy and implements the plan to reduce its balance sheet. The concern is that this presumed increase in rates will have an immediate reduction in refinance volume and hurt purchase volume as potential home buyers face increasing monthly payments. Interestingly, mortgage rates have floated in a pretty tight range through the entirety of this cycle. We believe the reduction in mortgage-backed securities from the Fed’s nearly $4.5T balance sheet will have little impact on mortgage rates. This operation will take years to run its course, and there is a tremendous amount of demand for these securities from institutional investors around the world that will gladly step to fill the void created by the Fed’s exit[10].

The initial GOP tax proposal includes new limitations on the mortgage interest deduction. Specifically, it reduces the cap of eligible loan balances from $1mm to $500k. Many in the residential real estate industry have publicly claimed that this change will hurt potential and current home owners, home builders, and other industry stakeholders. While this will have an impact on refinances and, potentially, purchases, it will not be evenly spread across the nation. In addition, this may have long-term beneficial side effects if it results in further deleveraging by incentivizing people to refrain from buying bigger homes or continuously tapping into their home equity[11].

It is likely that lending standards will be relaxed in the near future. We’ve approached a point at which the tightened conditions have become excessive, with an overwhelming number of reasonably credit qualified people locked out of the market. We believe the market is at that inflection point. Standards will loosen a bit, but how credit quality evolves from here will be important. Just as the Fed is aiming to time it’s monetary policy changes perfectly, the mortgage industry must loosen the credit standard screws carefully. Precise execution of this will allow more people to enjoy homeownership and maintain the integrity of the recovery to date. These changes could come at a time when more people are returning to the market. The US homeownership rate has stabilized after continuously declining for more than ten years.

Month-to-month data releases and surprises muddy the waters, yet the economic progression, conservative lending practices, and continued deleveraging and distribution of risk have resulted in a clear, robust housing market. These developments have evolved since the recovery began in 2011 and have continued through this past year. We anticipate these trends to continue into 2018. While some of the hotter markets may soften and other laggards may fail to catch up, there are few fundamental forces that would lead to a material decline or significant increased volatility in home prices.

[1] Index results are published on a three-month lag to ensure sufficient collection of sales data.

[2] Through 2017-11-28

[3] Plotting real growth rates would simply translate the plot since both values would be adjusted by the same inflation series. The relationship would still hold.

[4] Although house price index data is available through September 2017, metro area GDP and population data for 2017 have not been released.

[5] Urban Institute, Fannie Mae, Freddie Mac

[6] Urban Institute

[7] 90 or more days late

[8] Mortgage Bankers Association, CoreLogic, Urban Institute

[9] Urban Institute, Fannie Mae, Freddie Mac

[10] The Fed has started to roll of $50B per month, of which $20B is in MBS, until its total balance sheet is reduced to $3T.

[11] At the time this was published, the mortgage deduction changes, as described here, were only included in the House bill. Both the House and Senate bill repeal the deductions for home equity debt.

Share this Post