More Than Just a Pause in Housing

Dan Oppenheim
Print Friendly, PDF & Email

SEE LAST PAGE OF THIS REPORT

FOR IMPORTANT DISCLOSURES

Dan Oppenheim , CFA

(415) 889-5617

doppenheim@ssrllc.com

October 16th 2018

More than Just a Pause in Housing

  • Housing turnover to fall 10% in 2019 and 5% 2020. We expect existing homeowners to delay move-up purchases, given the significant and rising spread between in-place mortgage rates (generally below 4.0%) and prevailing mortgage rates (4.90%). Sustained increases in mortgage rates of 100 basis points or more have been rare in recent decades (seen in 1987-89, 1994, 1999, 2005-2006 and 2013), with falling turnover and slowing price appreciation following each time.
  • Housing turnover fuels home improvement; sputtering ahead. Home improvement spending is highly correlated with housing turnover (0.72), with spending lagging turnover by four-to-five quarters. As such, we would not be surprised to see flat comp sales in both 2019 and 2020, down from 5.3% growth expected in 2018. We do not believe this is factored in to the valuations of the home centers, and see risk to the valuations of both Home Depot and Lowe’s, with the stocks just 10.6% and 10.4% below their respective highs. This is better reflected in valuations of building products companies, but still do not view the group as attractive.
  • Market-by-market analysis shows increase for-sale competition and easing of pricing power. Our analysis of trends in key housing markets suggests rising inventory trends to continue, and a significant downshift in pricing power. In addition, we believe consumer credit suggest a consumer environment that is more stretched than is generally perceived.
  • Stability helps owners of rentals; stronger cash flow growth likely. We believe the lower for-sale turnover, worsening for-sale affordability (similar to 2004 levels), and lower construction (housing starts will likely fall 5% in 2019) will lead to decreased rental turnover, better pricing power, and improving cash flow growth for both multi-family and single-family REITs.
  • Homebuilders to see decreased demand and pricing power, with risk to margins. We expect a challenging environment for homebuilders as existing home inventory increases and presents significant competition. Builders focused on the entry-level will likely be the least impacted.
  • Mortgage insurers to benefit from fewer prepayments. Higher mortgage rates should limit prepayments and the continued, but slower home price appreciation will mute loses.
Exhibit 1: SSR’s Preference Among Housing-Related Sectors
Source: SSR analysis

Executive Summary

Housing turnover to fall 10% in 2019 and 5% 2020. We expect existing homeowners to delay move-up purchases, given the significant and rising spread between in-place mortgage rates (generally below 4.0%) and prevailing mortgage rates (4.90%). As a result, we expect housing turnover to fall 10% in 2019 and an additional 5% in 2020. Sustained increases in mortgage rates of 100 basis points or more have been rare in recent decades (seen in 1987-1989, 1994, 1999, 2005-2006 and 2013), with falling turnover and slowing price appreciation following each time (Exhibit 2). Overall, there was a 13% median decline in single-family existing home sales from two months after the start of rising rates (existing home sales typically close one-two months after contracts are signed and many active buyers continue even as rates start to increase) to two months after the peak of rates.

Rising mortgage rates will serve as a deterrent to turnover. The trend of flat or declining mortgage rates over the past 30 years had a positive impact on turnover, as lower mortgage rates enable buyers to afford more, with lower rates partially offsetting the impact of a higher price if the buyer chose to move up. However, the rising mortgage rates now will act as an impediment to turnover as the higher mortgage rates mean that existing homeowners – the majority of home buyers – would face higher mortgage payments from moving to a house with the same mortgage amount, as they would leave behind the lower rate mortgage (only homes purchased recently have in-place mortgages with rates above 4%) and would take on a mortgage at 4.9%, with the higher mortgage payments on a $200,000 mortgage representing 2.7% of after-tax income for a household with $60,000 of gross income.

Decelerating price appreciation likely in coming years. We believe that the combination of decreasing demand and rising supply will lead to a slowing of price appreciation, which is also evidenced by the greater percentage of listings with price reductions. We expect home price appreciation to slow to 2% in 2019 and 2020, down from 5.6% in 2017 and likely 4.6% in 2018 (Exhibit 6). Our expectation of modest positive pricing is based on the challenges presented from the rising listings and decreased buyer pool, but is offset by the limited new construction activity.

Lower housing turnover to impact home center sales. We expect the reduced level of housing turnover to lead to slower growth in home center sales, as much of repair and remodeling activity takes place immediately prior to, or in the first year after the sale of a house. Existing home sales tend to lead trends in home improvement spending by four quarters, and we expect to see slower home improvement activity in late 2018 and into 2019 (Exhibit 7). We do not believe that this is factored into expectations for home centers.

Housing market quickly shifted away from undersupply, with inventory rising. The housing market pivoted in 2018, with existing home inventory rising to 4.3 months of supply in August from a multi-decade low of 3.2 months of supply in December 2017. We expect further growth in inventory to continue basis on our analysis of new listing trends across key markets. We anticipate that this will result in slowing price appreciation.

Sellers more likely to reduce prices, highlighting the shift in the market. We believe that one can gain a sense of pricing momentum by looking at the percentage of listings with price reductions. The current level of price reductions (percentage of listings with reductions) is the highest level of we have seen since the housing collapse a decade ago. Overall, we saw that 26% of listings from April-August 2018 had price reductions, well above the 18% average for those months from 2012-2017 and modestly above the 23% from April-August 2017.

Consumer credit trends indicate less ability for future home purchases. We believe worsening housing affordability, higher consumer loan originations and balances (autos and student debt), and challenged consumer loan performance (auto and student loans) points to limitations on future mortgage borrowing.

Slowing reflected unevenly across sectors, with greatest risk for home centers. We do not believe that the slowing turnover and likely easing home improvement activity is reflected in home centers stocks and see risk to both Home Depot and Lowe’s. Building products companies will also see slower sales, but stocks incorporate more of this. We prefer those more tied to repair and maintenance (MAS) than to big-ticket remodeling (FBHS and MHK). We view the residential REITs (multi-family and single-family rental companies) as better positioned than homebuilders, as the REITs will likely benefit from the lower turnover and fewer residents moving out to purchase homes. Entry-level homebuilders such as D.R. Horton and those focused on boosting returns (MDC, Meritage, and NVR) are best positioned within the homebuilding sector. Among the REITs, we prefer AvalonBay, Equity Residential, Camden Property Trust, Invitation Homes, MAA, and UDR. Mortgage insurers will benefit from limit prepayments (less refinancing) and the continued, but slower home price appreciation. We favor Essent and MGIC.

More than Just a Pause in Housing

Housing turnover to fall 10% in 2019 and 5% 2020. We expect existing homeowners to delay move-up purchases, given the significant and rising spread between in-place mortgage rates (generally below 4.0%) and prevailing mortgage rates (4.90%). As a result, we expect housing turnover to fall 10% in 2019 and an additional 5% in 2020. Sustained increases in mortgage rates of 100 basis points or more have been rare in recent decades (seen in 1987-1989, 1994, 1999, 2005-2006 and 2013), with falling turnover and slowing price appreciation following each time. Overall, there was a 13% median decline in single-family existing home sales from two months after the start of rising rates (existing home sales typically close one-two months after contracts are signed and many active buyers continue even as rates start to increase) to two months after the peak of rates.

Exhibit 2: Sustained Increases in Mortgage Rates Have Been Rare; and Almost Always Slow Sales
Increase in 30-yr mortgage rate (bp) Decline in pace of existing home sales
March-October 1987 226 -13%
March 1988-April 1989 116 -20%
October 1993-December 1994 247 -14%
January 1999-May 2000 181 -1%
June 2005-August 2008 124 -42%
May-September 2013 105 -7%
Source: National Association of Realtors and SSR analysis

Turnover currently in-line with normal levels, with room to fall. Even after the recent slowing of sales, we estimate housing turnover (total home sales divided by the housing stock) of 4.4% in 2018 and 4.3% in 2019, which is in-line with the historical 4.5% average. Most view the current level of turnover as being depressed since this 4.4% level is a full 35% below the levels seen during the housing boom of 2004 and 2005, when turnover peaked at 6.7%. However, aside from those years, turnover generally ranged from 3.5% to 5.0% so that this current 4.4% level is still above the midpoint of that more typical range of 3.5-5.0%. A 10% decline in turnover in 2019 would bring turnover down to 4.0%.

Exhibit 3: Housing Turnover Slowing as Higher Mortgage Rates Discourage Activity by Existing Owners
Source: Census Bureau, National Association of Realtors and SSR analysis

Consumers are now more sensitive to changes in mortgage rates. We believe that there is a greater impact from a given change in rates now than in decades past as a result of the higher home prices and mortgage amounts. That is, when mortgage rates increased in 1987, 1988-1989, 1999-2000, 2005-2008, and 2013, the increase in the monthly payment from a 50 basis point increase in mortgage rates represented between 0.9 and 1.1% of median gross monthly income (average of 0.96% of median gross monthly income), whereas it represents 1.34% of median gross monthly income in 2018.

Exhibit 4: 50 Basis Point Increase in Mortgage Rates Has Greater Impact than in the Past
Source: Bureau of Labor Statistics, Census Bureau, National Association of Realtors and SSR analysis

Flat or declining mortgage rates lifted turnover in past years. Importantly, most of the past 30 years were years of flat or declining mortgage rates, with mortgage rates above 10% 30 years ago (and higher still in the years before that), but below 5% now, despite the increases over the course of 2018. This trend of declining mortgage rates has had a positive impact on turnover, as lower rates enable buyers to afford more, with lower rates partially offsetting the impact of a higher price if the buyer chose to move up. This led turnover to generally stay in a range of 3.5-5% of existing homes, with above-normal turnover in 2004-2005 (peak of the housing boom) and below-normal turnover in 2007-2011 (the housing crash).

But rising rates now create a disincentive to move for owners of existing homes. Most discussion of the impact of higher mortgage rates focuses on affordability, an issue which is most significant for the first-time buyer. For existing homeowners, however, who make up the majority of home buyers, higher mortgage rates mean that moving would require leaving behind a mortgage rate that is likely lower than the prevailing mortgage rate of 4.9%, as we estimate that more than 85% of outstanding mortgages have coupons at or below 4%, with only those who bought in recent years having mortgage coupons above 4%. This lower coupon level on existing mortgages serves as a deterrent to mobility, as even a lateral move (to a home for which the homeowner would take on a mortgage of a similar amount) would lead to a higher monthly payment. More specifically, if a homeowner has a $200,000 mortgage at 4%, making a lateral move and taking on a similar $200,000 mortgage at 4.90% would lead to a $106.63/month increase in the monthly payment ($1,061/month, up from $955/month). While this may not seem like a lot, that difference equates to $1,280/year. (A family with a $200,000 mortgage would often have income of $60,000.) That $1,280 annual difference would then represent 2.7% of the $48,000 after-tax income on a $60,000 gross income. Another 50-basis point increase in mortgage rates would lead to a 5.4% mortgage rate and equate to an increase of $168/month ($2,019/year, or 4.2% of after-tax income) relative to the mortgage at 4%.

Exhibit 5: Higher mortgage keep existing homeowners in place
Note: Analysis based on $200,000 mortgage, $60,000 gross household income, and $48,000 after-tax income.
Source: Bureau of Labor Statistics, SSR analysis

Decelerating price appreciation likely in coming years. We believe that the combination of decreasing demand and rising supply will lead to a slowing of price appreciation, which is also evidenced by the greater percentage of listings with price reductions. We expect home price appreciation to slow to 2% in 2019 and 2020, down from 5.6% in 2017 and likely 4.6% in 2018. Our expectation of modest positive pricing is based on the challenges presented from the rising listings and decreased buyer pool, but is offset by the limited new construction activity.

Exhibit 6: Price appreciation likely in low single-digits in coming years

Source: National Association of Realtors, SSR analysis

Slower home improvement spending to follow reduced turnover

Lower housing turnover to impact home center sales. We expect the reduced level of housing turnover to lead to slower growth in home center sales, as much of repair and remodeling activity takes place immediately prior to, or in the first year after the sale of a house. Existing home sales have had a 0.72 correlation with home improvement spending (Joint Center for Housing Studies). Housing turnover, measured via existing home sales, tends to be the key early indicator of home improvement spending, with trends in turnover providing a four-quarter lead ahead of residential improvement spending. As a result, we anticipate that the slowing of existing home sales will translate into slower home improvement spending in late 2018 and into 2019. Our expectation that slower home improvement activity will be seen in late 2018 and 2019 is based on the overall slowing that occurred in existing home sales since late 2017, with seasonally-adjusted sales generally closer to 5.4 million homes annualized, with months of higher readings (February and March 2018) being months of lower absolute activity, but with seasonal adjustments making these months appear more vibrant.

Risk to expectations for home centers based on impact of turnover. There’s still debate about the impact of lower turnover on sales at the home centers. Home Depot management has noted that it believes that a 15% decline in existing home sales would impact their comp sales by less than 1%. We believe the impact would be more significant, however. In addition, lower turnover levels are typically correlated with slower home price appreciation, with home price appreciation being another driver of home improvement spending.

Exhibit 7: Declining Existing Home Sales Point to Slower Home Improvement Spending Ahead
Source: Joint Center for Housing Studies, National Association of Realtors, and SSR analysis

Home centers comps sales growth likely to slip in the quarters ahead. We expect home centers to post slower sales trends in the coming quarters based on the deceleration in housing turnover and our expectation that it will lead to moderating home improvement activity. Home Depot’s recent trends have reflected the strength of the housing market through 2017 and early 2018 (home center sales activity typically reflects housing market activity with a lag), with US comparable sales growth of 8.1% in fiscal 2Q18, 3.9% in 1Q18 (weather), 7.2% in 4Q17, and 7.7% in 3Q17. This 8.1% growth in 2Q18 sales was the highest level of growth seen in 2Q since 2013, when housing enjoyed a mini-resurgence amidst recovering home prices and consumer optimism toward the sector. We would expect to see a slowing of comp sales trends in the upcoming quarters, but note that Home Depot management has indicated that it believes that a 15% decline in existing home sales would negatively impact comp sales by less than 1%. We do not believe that the valuations of the home centers reflects concern of slowing comps, given LOW’s 12% YTD increase and HD’s 1% YTD increase, relative to a 3% increase for the S&P 500.

Exhibit 8: Home Center and Home Improvement Spending to Slow in Upcoming Quarters
Source: Company reports and SSR analysis

Inventory Rising and More Supply is Coming

Lack of supply had been the theme in housing, but no longer. The housing market has generally been characterized in recent years by a severe undersupply and a lack of existing inventory, with existing home inventory reaching a multi-decade low of 3.2 months of supply (1.46 million homes) in December 2017. Since that time, there’s been an increase in inventory, with the number of homes for sale rising to 1.92 million homes in August 2018, representing 4.3 months of supply, with the statistics likely to show a further rise in September. August 2018 was the first month in which existing home inventory increased on a year/year basis since 2015, and we expect this trend of rising year/year inventory to continue and to lead to slowing price appreciation.

Inventory provides a sense of changing market conditions. Inventory trends have historically provided an early read on shifting housing dynamics, with inventory levels rising significantly in 2005 ahead of the housing downturn and inventory levels falling in 2011 as the market bottomed, along with more modest moves throughout the cycles. We believe that the recent increase in inventory (from record-low levels) may continue over the remainder of 2018 and into 2019, which would then likely lead to more modest price appreciation, as higher inventory levels are correlated with lower levels of price appreciation. We think a key to forecasting inventory trends is via trends on the supply side – the actions of existing homeowners.

Exhibit 9: Housing Inventory Rising in 2018 and Likely to Continue
Source: National Association of Realtors and SSR analysis

New listing activity provides an early indication of supply and inventory. We believe new listings provide a sense of the future direction of the market, and potentially a shift in mindset by owners of existing homes (sellers). Increasing new listing activity will likely lead to higher inventory levels in subsequent months, even assuming an unchanged level of demand. As more listings come to market, there may be two buyers for every home for sale, rather than three or four buyers. As there are still more interested buyers than homes on the market, inventory levels do not increase. However, by isolating new listing activity, we are able to get a sense of what is happening beneath the surface, even if it will not immediately lead to a change in inventory. In addition, an environment of rising new listings and slowing demand would bring about an even more rapid increase in inventory.

Recent trends show increasing new listings and likely higher inventory ahead. We analyze new listing activity across 25 key markets, and have seen an acceleration in new listings in 2018, with new listings through August 2018 running 14% above the average level of new listings seen from January-August in 2012-2017. Additionally, trends in recent months (April-August) show slight acceleration in the growth of new listings, with new listings up 3% relative to 2017. Our contacts with agents are consistent with this, with agents noting that a higher level of new listings is expected this fall relative to recent years.

Exhibit 10: New Listings Trends Point to Risk of Higher Inventory Ahead
Source: SSR analysis of MLS and Redfin data

Many key metros show high level of growth in new listings. Boston, Dallas / Fort Worth, Florida markets, Kansas City, Nashville, and San Francisco/San Jose all had an increase of 7% or more in new listings in April-August 2018 than in April-August 2017. We will watch ongoing trends in these markets and believe that they may provide an early sign of added supply coming to market, as owners recognize the price appreciation that has occurred and decide to sell their homes. Economic trends remain healthy in these markets with unemployment rates below 4% in each of these markets, and at just 2.8% in San Francisco and San Jose. We recognize that the increased new listing activity could be in response to positive economic trends, as homeowners have a desire to “trade-up”, but we nonetheless believe that these trends should be watched carefully.

Sellers more likely to reduce prices than in recent years

Price reductions often an overlooked metric that highlights imbalances in the market. We believe that one can gain a sense of pricing momentum by looking at the percentage of listings with price reductions. Existing home sales are recorded at the time of closing, with the release of that data occurring subsequent that. However, changes in asking prices precede sale transactions, with an increase in the percentage of listings with price reductions indicating both an imbalance in the market and potentially unease among sellers. We analyze listing data for 25 key metro areas across the country. Overall, we saw that 26% of listings from April-August 2018 had price reductions, well above the 18% average for those months from 2012-2017 and modestly above the 23% from April-August 2017. This current level of price reductions is the highest level of price cuts we have seen since the housing collapse a decade ago.

Exhibit 11: More listings with price cuts, indicating an imbalance and slowing of appreciation
Source: SSR analysis of MLS and Redfin data

Western markets showing greatest increases in listings with price reductions. Our analysis of the key housing markets showed significant step-ups in the percentage of listings with price reductions in many west coast markets, along with a few Texas markets. Of the western markets, we saw noticeable increases and/or high absolute levels of listings with price reductions in Denver, Orange County, Portland (a modest increase, but a high level), Riverside, San Bernardino, San Diego, Sacramento, and Seattle. Other markets with concerning increases in price reductions included Dallas/Fort Worth, Indianapolis, and San Antonio. Conversely, a few western markets (San Francisco and San Jose) still had only modest portions of listings with price reductions, despite seeing more new listing activity and slowing sales.

Exhibit 12: Previously-hot Western Markets Now Showing Signs of Cooling
Source: SSR analysis of MLS and Redfin data

Stretched Affordability and More Challenging Consumer Credit

Monitoring the financial health of the consumer to understand future demand. We track those metrics that will provide early indications of future consumer activity – housing affordability (as affordability worsens, the buyer pool shrinks, and price appreciation slows), consumer loan originations and balances outstanding (the more debt the consumer has, the more difficult it is to purchase a home), consumer loan performance, and the underwriting of mortgage insurance (easing underwriting brings in buyers initially, but has its limits).

Affordability is key long-term indicator of housing conditions, but shows stress. We believe the most relevant long-term measure of the health and stability of the housing market is affordability, whether expressed as the cost of ownership as a percentage of income, or the cost of ownership relative to the cost of renting. When looking at the cost of ownership relative to income, for-sale affordability is above its long-term average at 21.4%, higher than the 19.0% average over the past two decades. However, this affordability is not nearly as favorable as in 2011-2013, when home prices had not yet rebounded from the downturn.

Exhibit 13: Affordability No Longer Provides Support for For-Sale Housing
Source: HUD, National Association of Realtors, and SSR analysis

Own vs rent calculations may lead households to remain in the rental market. In addition to monthly mortgage payments now equating to the historical average percentage of income, the rent vs own calculation is now much more balanced than in recent years. The monthly mortgage payment is currently 9% less than the rent payment, closer than the historical average of the mortgage payment representing a 13% discount to the rent payment. We believe that this may push keep some households in the rental market, favoring both single-family rental companies (American Homes 4 Rent and Invitation Homes) and also multi-family rental companies (AvalonBay, Camden Property Trust, Equity Residential, Essex Property Trust, and UDR). We would expect to see greater pricing power in the single-family rental market, as American Homes 4 Rent and Invitation Homes achieved 96.6% and 96.0% occupancy in 2Q18, respectively.

Exhibit 14: Rental Market Now Looks Relative Affordable as Compared with Purchasing
Source: National Association of Realtors, Zillow, and SSR analysis

Affordability now most stretched in the west since 2004. We believe most western and west coast markets have returned to affordability levels that will prove to be challenging for consumers, and would indicate that there is only limited opportunity for further price appreciation, absent lower mortgage rates or significantly higher income growth. We watch trends in the west especially closely, given that many homebuilders generate a substantial portion of profitability from the region. Markets with affordability now back to levels not seen since 2004 (close to the peak before the housing downturn) include Denver, the Inland Empire (Riverside-San Bernardino), Las Vegas, Phoenix, Portland, Sacramento, San Diego, San Francisco, San Jose, and Seattle. Importantly, the tax changes in 2018 also mean that California markets, Portland, and Seattle have greater affordable concerns due to the state and local tax (SALT) burden generally exceeding the $10,000 limit on SALT deductions.

Exhibit 15: For-sale affordability appears stretched in key western markets
  
Source: HUD, National Association of Realtors and SSR analysis

Western markets are important to homebuilders and home centers, and residential REITs. The stretched affordability in the western markets is likely to inhibit the sales and profitability of homebuilders and home centers. That is, most of the large homebuilders generate between 30% and 70% of their profitability from California and the west region, We estimate that Toll Brothers is most exposed to the region, generating approximately 73% of its overall profitability from California and the West, followed by KB Home (70%), William Lyon (51%), MDC (47%), D.R. Horton (34%), Lennar (34%), and Pulte (33%). In contrast, we think rental companies with significant exposure to the west (Invitation Homes, AvalonBay, Equity Residential, and Essex Property Trust) may see some benefit from the stretched affordability of the for-sale market. Home centers also generate a significant portion of sales and profitability from the west, based on store count in the west along with greater productivity of those stores relative to stores in other regions. 12% of Home Depot’s US stores are in California, but based on the higher sales of the stores, we estimate that California may represent approximately 25% of Home Depot’s revenue.

Exhibit 16: Many Homebuilders Derive 40-70% of Profitability from CA and the West, Areas of Slowing
Source: Company reports and SSR analysis

Texas markets also more expensive than historically and have provided source of growth. The key markets in Texas – Austin, Dallas, Houston, and San Antonio all have affordability that is worse that it has been historically. Affordability is not nearly as bad as in west coast markets, but the more challenged affordability may still be an impediment to sales in these markets. The reason for focus on the Texas markets is that these four markets produced 93,530 single-family housing permits in 2017, representing 11.4% of total single-family permits in the US, with both Dallas and Houston representing more than 4% of US permits. Monthly mortgage payments on the median-priced home would now represent 20.3% of median income in Austin, above the 16.4% historical average and worse than levels seen in 2005. In Dallas, the monthly mortgage payment now represents 19.0% of median income, the highest it has been in Dallas and above the 14.3% long-term average. Houston and San Antonio have also seen affordability worsen, but not as significantly, with the monthly payment now representing 260 basis points more of income than historically in Houston (17.6% vs historical average of 15.0%) and 280 basi