We're Bullish on a Housing Recovery
Stronger household formation by millennials is key.
We believe U.S. residential construction growth will be robust over the coming decade, but the road to a fuller recovery has proved longer than expected. Household formation continues to build momentum, but slowly, especially among younger adults. We expect roughly 1.3 million starts in 2017, up from 2016’s 1.1 million.
A review of homebuilder and labor market data has us optimistic about residential construction in the coming quarters. Homebuilders have taken notice of the immense demand potential from millennials and have begun building smaller, more affordable homes to attract these first-time buyers. This means lower mortgage payments and down payments for strapped young adults. Labor markets have also tightened, catalyzing real wage growth since early 2015 for those aged 18-34. With job opening levels at prerecession highs across numerous industries, we expect further wage gains to support rising household formation in the coming quarters.
Our updated forecast assumes that momentum in single-family starts continues, thanks to firmer labor markets and early signs of a more affordable mix shift in single-family construction. We expect single-family starts of 935,000 units in 2017, up 19% from 784,000 in 2016. Single-family permits, which typically lead starts by about one month, have continued to rise in early 2017. We expect multifamily starts to slip to 375,000 from 2016’s 380,000. Multifamily permit growth has begun to flatline amid tighter credit and ample new supply.
Our long-term forecast is intact. We expect starts to peak at 1.9 million by 2021 before falling back to a demographically sustainable 1.5 million. Stronger household formation by the massive millennial generation is the key to our bullish forecast.
Housing Supply and Demand
Last year concluded with a strong quarter for home construction. Housing starts averaged a 1.22 million-unit pace during the fourth quarter before gaining momentum, averaging 1.27 million during January and February. We expect starts to average 1.31 million during 2017, an 11.9% increase from 2016.
We expect demand to grow more robust over the coming quarters as headship rates rise among young adults. Demand should be strongly supported by demographics as the 80 million-strong generation of millennials ages into its 30s. Headship rates (the percentage of adults who head a household) have stabilized among those aged 25-34 in 2016. Improving financial conditions should catalyze a recovery in younger adult headship, buoying adult household formation into 2017.
For much of the housing recovery, a meaningful portion of demand has been met by a drawdown of vacancies, diminishing the need for new homes. But with the nationwide vacancy rate (12.7%) reaching prerecession levels (12.7% in 2005), future demand will increasingly depend on new homes. While some have grown concerned about the dearth of starter homes available on the market, we think homebuilders have an incentive to pivot to that audience. A combination of higher mortgage rates, reduced lot availability, and labor constraints is poised to encourage homebuilders to increase shareholder returns by turning over a larger number of smaller, more affordable units every quarter.
Owner and Renter Affordability
Heavy student debt burdens, income erosion among younger adults, and tight underwriting continue to make homeownership less affordable than it would typically be in a low-rate environment for younger and lower-income groups. Although renting is less affordable than it has been historically, demand for rentals remains high relative to prebubble norms.
Tightening labor markets suggest cause for optimism. After an extended period of declining real wages, young adults began to see wage growth in recent quarters. As wages tick higher for those aged 18-34 and financial constraints unwind, we expect recovering rates of household formation. After roughly a decade of sliding headship rates, we’ve started to see signs of stabilization over the past two years in these age categories.
We expect further gains over the coming year. With job opening rates high relative to prerecession norms and unemployment rates already low, we expect upward pressure on real wages. Supply dynamics also appear attractive. A large stock of multifamily units is nearing completion, which should decelerate rental rate growth in 2017. Homebuilders moving down market on single-family homes should also help, partially offsetting what could be a 9% mortgage payment increase, assuming three quarter-point rate hikes in 2017. Credit availability remains a lingering concern, with the bulk of purchase mortgage approvals being granted to those with credit scores exceeding 700.
Credit availability is slowly improving but remains an obstacle for many potential homebuyers, particularly younger and less creditworthy borrowers.
We attribute the increasing availability of mortgage credit to several factors. The resolution of multi-billion-dollar litigation between major mortgage banks and government-sponsored enterprises and the subsequent easing of GSE representations and warranties policies closed the gap between GSE standards and banks’ actual willingness to lend. The finalization of mortgage lending rules created by the Wall Street Reform and Consumer Protection Act decreased regulatory uncertainty. Improvement in the overall economy has increased lender confidence as well.
The Federal Reserve Senior Loan Officer Opinion Survey has shown loosening standards for prime and GSE-eligible mortgages since 2012, but banks’ appetite for subprime and other riskier forms of mortgage lending remains subdued. Indeed, 58% of mortgage originations in the final quarter of 2016 were to borrowers with FICO scores above 760, with another 17% in the 720-759 range. While the home purchase closing rate increased from 71% in late 2015 to 77% in January 2017, some observers have pointed out that less qualified borrowers are leaving the applicant pool. The Urban Institute’s Housing Credit Availability Index has fallen 70% from its peak in March 2006.
We believe roughly half of young adults have credit scores below 640. Private-market subprime credit vanished in the wake of the global financial crisis, and Federal Housing Administration loans have failed to fill demand. Less than 10% of all mortgages are being originated to customers with such low credit scores, effectively eliminating a large pool of borrowers from the market. Without a further loosening of credit standards, they will be forced to wait for a combination of time and improving economic conditions to improve their creditworthiness.
Much has been made of potential regulatory reform in the wake of the U.S. presidential election. The fate of Fannie Mae and Freddie Mac remains uncertain, with many Republicans favoring a full privatization of the mortgage market. We do not believe the private market currently has the appetite or capital to provide credit at similar levels, let alone expand into riskier categories. Federal agencies guarantee more than $5 trillion in mortgage debt. Private-label mortgage debt outstanding has already fallen dramatically from the peak in 2007, when it accounted for more than 20% of the total. We currently see no near-term catalysts to reverse its decade-long downturn in importance.
We expect wood products companies to ride rising single-family starts to strong profit growth over the next several years. In the space, we anticipate comparably greater and more rapid profit uplift for downstream lumber and panel manufacturers than upstream timber real estate investment trusts. Accordingly, we regard lumber producers Canfor and West Fraser Timber (WFT) as better ways to play a continued housing recovery than timber REITs Rayonier (RYN) and Weyerhaeuser (WY).
Lumber and oriented strand board markets are likely to face meaningful capacity constraints in the not-too-distant future. As recently as October and November, North American lumber capacity utilization rates averaged roughly 90% despite starts activity of only a 1.21 million seasonally adjusted annual rate, well below a multidecade average of roughly 1.5 million units. Amid rising single-family starts, we expect North American lumber demand will expand to around 75 billion board feet through 2021 from roughly 60 billion today. As capacity utilization rises and small, high-cost mills are forced into operation, we expect benchmark prices to rise roughly 45% to more than $500 per thousand board feet by 2021.
By contrast, a decade of weak housing activity has led to an abundance of mature pine in the Southeast, which is likely to limit log price increases in the near term. Timberland owners Weyerhaeuser and Rayonier expect log prices to remain fairly flat in 2017. Although we expect log prices will rise as housing demand strengthens, accumulated supply will mean the eventual highwater mark is lower than it otherwise might be.
We believe lumber production will prove to be the bottleneck in the housing supply chain. While neither Canfor nor West Fraser has sustainable competitive advantages, high capacity utilization rates should lead to attractive pricing environments for incumbents, encouraging new capacity to come on line. We expect log prices to rise materially over the coming decade; however, fiber and hauling costs make up less than half of lumber production costs. Given that labor, energy, and overhead costs grow at a modest pace, we anticipate EBITA margins for Canfor and West Fraser should easily top 20% during the peak years of our housing forecast, between 2019 and 2022. Canfor remains our top pick for its high exposure to lumber production and 15% upside.
The public homebuilders finished 2016 strong and appear poised for continued growth in 2017. When analyzing the homebuilders, we keep a close eye on forward-looking indicators, such as backlog units and price, owned and controlled lot count, and active community count, to help us gauge the industry’s outlook. We think these metrics paint a positive picture for 2017 as industry backlog value (units x price) is up 11% year over year and lot count and community count edged higher, demonstrating that homebuilders remain confident and are investing for future growth.
If consistent growth in home deliveries, new orders, backlog, and land supply throughout 2016 was a bright spot for homebuilders, margin performance was the most significant challenge. Average GAAP gross margins declined from the prior year during each quarter in 2016. Gross margins have been pressured by higher land and labor costs due to tight supply for both inputs. Still, homebuilders have generally been able to leverage their fixed costs and improve their capital efficiency, which helped cushion operating margins and improve returns on invested capital in 2016
We cannot stress enough that investors should pay more attention to ROICs than gross margins. The former gives investors a holistic view of homebuilder profitability, capital efficiency, and value creation in one metric. Furthermore, homebuilders use ROIC to underwrite land deals. We expect input cost inflation to continue to pressure gross margins, and gross margin headwinds could be exacerbated as homebuilders shift their mix toward entry-level products. However, entry-level homes generally have lower cycle times and quicker absorption rates that drive ROICs higher despite potentially lower gross margins. We think homebuilders will continue to leverage fixed costs, which will support stable operating margins in 2017. We also expect industry ROICs to improve in 2017 as homebuilders continue to embrace lighter land strategies and faster-turning products.
It has been well publicized that land markets are tight, but the public homebuilders navigated the land markets and successfully increased lot supply in 2016. At the end of 2016, industry lot count increased 3% over the prior year, representing about six years of land supply, about in line with the land supply at the end of 2015 despite a 13% increase in home deliveries in 2016. Rising land costs are a headwind to gross margins, but the growth rate of cost per lot noticeably slowed in the fourth quarter, perhaps an early sign that land prices are easing.
Although we have a bullish housing outlook, we think an improving housing market is generally priced into the homebuilder stocks we cover, and we think these stocks are about fairly valued at current prices. In contrast to lumber producers, homebuilders enjoy less operating leverage and are less likely to benefit from constrained supply. Of the homebuilders we cover, PulteGroup (PHM) appears the cheapest at today’s price.
All five of these homebuilders have adjusted their strategies to attract first-time homebuyers. D.R. Horton (DHI) was a first mover with the introduction of its true entry-level product, Express Homes, in the spring of 2014. Since then, other homebuilders have followed suit. NVR (NVR) introduced its Simply Ryan product, and luxury home leader Toll Brothers (TOL) introduced a lower price point with its T Select product. PulteGroup recently increased its land spending on entry-level products as it seeks a more even balance among first-time, move-up, and active adult buyers. Lennar (LEN) is focused on attracting first-time buyers with higher-density condominiums and townhomes located closer to city centers. The market for existing entry-level homes remains very tight, and the first-time new construction market has been underserved for years, so we think this shift in strategy is sorely needed and should be welcomed with high demand.
Although this strategy shift is in the early stages, we think increased availability of new entry-level homes is already affecting the housing market as the median square footage of newly completed single-family homes fell in 2016 for the first time since the recession.
Home Improvement Market
In the home improvement space, we believe Bed Bath & Beyond and Williams-Sonoma remain materially undervalued, trading at 39% and 22% respective discounts to our fair value estimates. While investors fear the impact of rising interest rates on housing turnover, we believe the slowdown in home sales is unlikely to surface in the near term. In 2016, existing-home sales increased 3.8% to 5.45 million units, slightly above the 15-year average. This supports demand for housing-related products. Demand from first-time homebuyers and those trading up or down could be pulled forward as consumers attempt to purchase ahead of higher mortgage costs, keeping turnover lifted in the near term. Continued turnover of housing inventory should support spending across the home improvement space, bolstering sales at traditional home improvement retailers like Home Depot (HD) and Lowe's (LOW) but also at the supporting soft goods retailers like Bed Bath & Beyond. Moreover, an aging housing stock, rising wealth effect, and growing household formation all help support spending on products to improve or design the home. Paint volume continued to grow in 2016, with players in the contractor market such as Sherwin-Williams (SHW) outpacing players in the do-it-yourself market such as Valspar (VAL).
A confluence of steady economic recovery, demographic trends, and historically low levels of new rental supply in the years following the financial crisis have supported robust performance throughout the rental housing markets and our multifamily REIT coverage over the past several years. We think the demand outlook for rental apartments will remain robust in the near to medium term, assuming positive economic momentum. However, the environment has also spurred a large wave of construction, with delivery scheduled to rise in 2017 at almost 3 times the long-term average. Elevated deliveries have provoked intense competition among assets in lease-up, prompting rent concessions and incentives to build initial occupancy, which restrained market rent growth and affected Equity Residential’s (EQR) and AvalonBay’s (AVB) performance. With its significant urban Class A exposure, where much of the new supply has been delivered, we expect Equity Residential’s operations to remain pressured relative to AvalonBay, which focuses on operating and developing infill and suburban submarkets. However, we expect levels of new supply in their specific markets to moderate starting in late 2018, with performance for each firm eventually reverting to historical norms thereafter. Overall, we believe the multifamily asset class is characterized by price-taking and does not warrant an economic moat. Therefore, investors should continue to be sensitive to changes in market-level rental demand versus rental supply expectations.
Charles Gross does not own (actual or beneficial) shares in any of the securities mentioned above. Find out about Morningstar’s editorial policies.