Author: S.C.
Download the report as a PDF
1. Introduction to the U.S. Homebuilding Sector
The U.S. homebuilding sector is dominated by single-family homebuilding: of the 1.56m aggregate housing units started in November 2023, 1.14m were single-family and 0.42 were multi-family, for a ratio of 73% and 27% respectively. The industry is fragmented except for a relatively strong position at the very top: while the two largest (D.R. Horton and Lennar) had in 2022 a combined 25% balanced market share (as measured by closings, which are a sub-set of new homes construction), the 3rd-10th largest had 19% in aggregate, with the 10th largest homebuilder (Toll Brothers) having just 1.6%[1].
There are two main categories of single-family houses built:
- manufactured homes – prefabricated, assembled in the factory and only later shipped; and
- site-built homes – as the name suggests, built on the final site.
A hybrid between these two extremes are modular homes, which after construction are equivalent to a site-built home (i.e., they are set on a permanent foundation), but are in part constructed in a factory with modules that are later assembled on-site.
Manufactured home shipments are estimated around 100k units per year and relatively stable, down significantly from a peak of almost 400k at the end of the 1990s and a spike reaching 200k in 2005 before the subprime crisis. Such homes tend to have shorter expected lives than site-built homes, but also cost significantly less: the simplest manufactured homes cost ca. 1/3 of the equivalent site-built homes[2]. The advantage is however materially reduced when including land in the calculation and when considering more complex structures. The leader in the manufactured homes space is Clayton, with a 1.8% market share of the total homebuilding market.
Figure 1: An example of U.S. single-family home
The listed space, where both site-built and manufactured homebuilders are represented, is relatively rich, with:
- “Traditional” homebuilders: D.R. Horton, Lennar, NVR Inc., Pulte Group, Meritage Homes, and several more. While single-family homebuilding is the core business of these companies, some are active in multifamily as well.
- Manufactured homes: Skyline Champion, Cavco Industries and two smaller and illiquid players.
Multi-family homebuilders are not represented in the listed space.
[1] https://eyeonhousing.org/2023/06/top-10-builder-share-jumps-in-2022/
[2] https://www.jchs.harvard.edu/blog/comparing-costs-manufactured-and-site-built-housing
2. Profitability Profile
It is well-known and obvious that the homebuilding sector is highly cyclical, and this is reflected in the profitability metrics shown in the figures below. Operating margins have been relatively homogeneous in the recent 2020-2022 phase, characterised by strong profitability with fundamental tailwinds such as the scarcity of housing inventory for sale and the pick-up in demand for homes during the pandemic[3].
Figure 2: Operating Margin and Return on Equity in different economic phases. Sources: Bloomberg. Calculations by Timber Finance.
Despite some variability within the group, most companies delivered similar operating margins during “normal” phases. Returns on capital employed have been less uniform, with Toll Brothers underperforming in the long post-subprime expansion phase, and NVR being a clear outlier on the upside. Since NVR’s extremely high ROCE is driven by the different balance sheet structure due a differentiated business model (see more below), return on equity is shown in Figure 2. Toll Brothers had one of the most conservative financial positions entering the subprime crisis (see Figure 3) and was able to robustly navigate the period – it was not able to take full advantage of the recovery phase, though (see Figure 4).
The leverage profile of homebuilders varies strongly within the sample. Manufactured homes specialists have solid net cash positions on their balance sheets (see Figure 3). While in principle we know the importance of having a strong balance sheet in a cyclical industry, not only from the time of the subprime crisis, but also of the much longer downtrend that has affected manufactured homes shipments since the end of the 1990s, the differences in free cash flows over time among homebuilders have been remarkable. Skyline (which at the time was focused on “mobile homes” – both manufactured homes and RVs) had negative free cash flow for nine years between 2007 and 2015, before becoming again cash-flow positive. Cavco did much better – in fact, it did impressively well – being free-cash-flow negative after the sub-prime peak only in 2010. If we move to on-site homebuilders, we have on one extreme NVR with a strongly positive net cash position – note that since the subprime peak, NVR had a mildly negative free-cash-flow year only in 2011. Other traditional homebuilders have (moderately) positive net debt positions, albeit much lower than they were during the subprime crisis. However, it should be noted that for these companies net debt is a consequence of their business model: owning land and holding significant unsold inventory for speculative purposes requires capital. In Figure 3 we see that most recently Toll Brothers has been displaying a relatively high debt level, compared to peers. When we look at the company’s balance sheet, we see that Toll had $8.7bn of inventory, representing 70% of its balance sheet with an asset turnover ratio of 0.8x while $1.5bn of those $8.7bn worth of inventory is land. Toll Brothers owns land, develops it, and builds residential (luxury) communities on this land. Cavco’s business model instead is based on manufacturing homes with short lead times without the need to own, control or develop any land and the company’s balance sheet displays a moderate inventory level (less than 25% of tangible assets). We understand how the difference in business models affects financial metrics and how we need to look into the specifics of each company in order to judge such metrics in the appropriate context.
Figure 3: Leverage and Return on Capital Employed in different economic phases. Sources: Bloomberg. Calculations by Timber Finance.
Figure 4: Cumulative profit evolution based on EPS, normalised to 1.00 as of 2006. Sources: Bloomberg. Calculations by Timber Finance.
When considering total shareholder returns, over the period 2007-2023, the dispersion has been significant and one should be wary of survivorship bias: among the top-performers are NVR (15% CAGR including dividends), Cavco (14%) and D.R. Horton (12%). They compare to 7% p.a. for the U.S. Homebuilders ETF. Some players who were large before the subprime crisis never fully recovered: Hovnanian Enterprises, after a series of acquisitions, was forced to raise capital and did so in three equity offerings (2008, 2011 and 2012) that were extremely dilutive for existing shareholders. It had entered the crisis with significantly higher debt (as measured by total debt to tangible book value). Beazer Homes was also forced to raise capital three times between 2010 and 2012, although Beazer did not go through an acquisition spree before the crisis and was not as indebted as Hovnanian, as measured by the same metric. In comparison, NVR only did one small acquisition in 2005, had low debt relative to tangible book, and never had to raise equity.
[3] For both primary and second homes, but stronger for second/vacation homes https://www.redfin.com/news/vacation-home-demand-drops-august-2023/
3. Business Models
There are multiple dimensions along which homebuilders differentiate in term of business model, and such differentiation has major impacts on the risk/return profile of the company.
Demographics are one dimension to consider, including buyers’ creditworthiness: Taylor Morrison’s average buyer has a 753 FICO score; D.R. Horton’s average mortgage origination FICO score was 725; Hovnanian’s 743. While not all homebuilders report their customers’ average FICO score, those who do often display scores higher than the U.S. average (718[4]), which in principle is a positive factor. That said, it should be clear that FICO scores do not, however, predict absolute earnings stability: in 2007, D.R. Horton’s average FICO was 717 (vs. a 689 U.S. average) and every homebuilder was affected by the cyclical contraction.
The type of homes built (e.g., manufactured vs. site-built) and the strategy followed for land acquisition drive asset turnover and returns on capital employed (see Figure 5). Builders of manufactured homes tend to have higher turnovers, as manufactured homes have shorter production times. Among on-site builders, NVR is an outlier: its business model is characterized by low fixed asset requirements as it does not own land and builds only pre-sold homes. NVR has been very successful long-term with this business model and other peers have started to copy this capital-light approach[5] which gives companies additional flexibility. The price for this flexibility is the premiums to be paid to landowners for the option to purchase, and the risk of not being able to obtain appropriate land for the properties to be built and sold.
Figure 5: ROCE vs. Turnover. Sources: Bloomberg. Calculations by Timber Finance.
Focusing on custom-made homes vs. holding significant inventory on a speculative basis (“spec homes”) is a major business model differentiator. “Spec” homebuilders buy lots and build homes before having a buyer – hence “speculating” that they will be able to sell the home for a good profit. This exposes homebuilders to significant real estate price risk compared to custom homebuilders who, as the name suggests, build based on a buyer’s wishes and may already have secured the sale. For completeness, there is a third type of homebuilding classification that is customary: that of “tract homebuilders”. Tract housing is a type of development characterised by a large number of (nearly) identical single-family homes built along one or more roads.
Figure 6: Tract construction.
While homebuilders in general are exposed to various types of homes and business models, some have a stronger exposure to certain dimensions. For example, Meritage Homes is strongly exposed to spec homes inventory and owns the vast majority of the land (ca. 75% owned and 25% optioned). D. R. Horton, the largest homebuilder, also has a very significant speculative exposure and owns a land (with the inverse split of Meritage – 25% owned lots and 75% optioned). On the other side of the spectrum, there are companies like NVR that avoid owning land and avoid speculative construction. As highlighted in Figure 7, NVR’s balance sheet is proportionally underexposed to inventory and land and holds a large net cash position, compared to other homebuilders like DHI whose balance sheet is dominated by inventory and land.
Figure 7: Balance sheet structure for NVR and D.R. Horton. Sources: Company annual reports (2023). Calculations by Timber Finance.
[4] https://www.fico.com/blogs/average-us-fico-score-718
[5] https://www.builderonline.com/land/builders-go-land-light-in-an-effort-to-improve-returns_o and https://www.fitchratings.com/research/corporate-finance/fitch-upgrades-toll-brothers-idr-to-bbb-outlook-stable-06-11-2023
4. Role in the Timber Value Chain
U.S. homebuilders are the primary end customers for durable wood products manufacturers. Secondary products such as Louisiana Pacific’s trims and sidings (e.g. the branded LP SmartSide) are used by the leading manufactured homes company (Clayton) as well as by on-site homebuilders[6]. In this specific case, it is interesting to note that legendary investor Warren Buffett’s holding company (Berkshire Hathaway) owns Clayton and has a large stake in Louisiana Pacific as well. Companies in the timber value chain can create significant value for homebuilders through engineered solutions that make home construction more efficient and more attractive for homebuyers. Innovative EWP solutions that are easier and faster to assemble, as in the SmartSide example above, are key drivers for timber company revenue and margins.
Digital solutions that make customisation easier to implement both in the sales process and in the supply-chain side of the business, are examples of key value drivers for the timber industry: this is Builders FirstSource’s value proposition with respect to their digital solutions for homebuilders. Builders FirstSource[7] as well as Boise Cascade, core components of the Timber Finance Carbon Capture & Storage Index, are key suppliers to the largest U.S. homebuilders such as D. R. Horton.
[6] https://lpcorp.com/blog/the-key-differences-between-tract-spec-and-custom-homebuilders and https://lpcorp.com/about-lp/media-resources/news-releases/product-news/lp-smartside-trim-siding-showcased-as-sustainable-siding-product-at-the-2023-berkshire-hathaway-shareholders-meeting
[7] https://www.bldr.com/who-we-are/in-the-news/builders-firstsource-distributor-of-the-year-dr-horton and https://investor.drhorton.com/~/media/Files/D/D-R-Horton-IR/documents/vendor-spotlight-final.pdf
5. Macro Drivers
Two big drivers of the housing market can be identified: one demographic factor – household growth, linked to population growth – and one monetary factor – interest rates, affecting both corporate cost of capital and mortgage rates for homebuyers.
5.1. Household Growth vs. Supply
The main macro driver for housing demand is household growth. Tightly connected to population growth, but at times diverging from it due to circumstances such as economic growth or pandemic-driven constraints, household growth defines the natural demand for more (or less) homes, be them single-family homes or apartments. While the downside trend in growth rates is evident, both US population and household growth rates are still positive (which is not the case in all developed economies, see the example of Japan with positive household growth but declining population, or Italy with decreasing population and stagnating household growth). The covid pandemic has played a role in slowing down both population growth and household formation.
Figure 8: Population and household growth in the U.S. Sources: U.S. Census Bureau, retrieved from FRED. Calculations by Timber Finance.
For a balanced housing market, homebuilding activity should be close to net household formation (household growth) and even slightly higher in order to account for old housing stock that needs to be replaced. When homebuilding is much higher than household formation, the market becomes oversupplied, while if homebuilding is significantly below household formation, then the housing market is undersupplied – the latter scenario provides of course for more robust prospects for the homebuilding sector.
As in the figure below – despite some challenges posed by the quality of household data – the period before the subprime housing crisis were characterised by historically very high construction activity, while household formation was relatively weak. This, combined with other factors such as issues with lending standards etc., led to a collapse in housing construction – this was needed to bring the housing market back to balance. After the Great Recession, construction levels remained relatively subdued and picked up gradually until they peaked again at the end of 2021. Since the start of the monetary tightening by the Fed in 2022, housing starts have decreased while (estimated) household formation has remained strong.
Figure 9: Construction and household growth in the U.S. Sources: U.S. Census Bureau, U.S. Department of Housing and Urban Development, retrieved from FRED. Calculations by Timber Finance.
At the beginning of 2024, the macro fundamentals for the U.S. housing market are characterised by:
5.2. Monetary Policy
Historically, in the last 50+ years, monetary tightening has always resulted in a slowdown of construction activity. This is natural, as it increases cost of capital, making homes less affordable through higher mortgage rates; additionally, the economic slowdown it causes reduces the number of potential buyers by weakening the job market. The current slowdown in construction is therefore consistent with the historical pattern.
Higher mortgage rates also make the refinancing of existing mortgages relatively unattractive – homeowners who have locked in lower mortgage rates are less willing to buy a new home if they have to refinance at much higher rates. While this reduces demand for new homes, it also reduces the inventory of existing homes for sale, making supply tighter. This, in turn, can lead to a situation, such as the current one, where new homes sales are, on a relative basis, more robust than existing homes sales: indeed, while existing home sales are currently at the lows of the subprime crisis, new home sales are below their 2020 peaks but still around 2017-2018 levels.
Figure 10: U.S. interest rates and construction. Sources: U.S. Census Bureau, U.S. Department of Housing and Urban Development, Board of Governors of the Federal Reserve System. Retrieved from FRED. Calculations by Timber Finance.
7. Conclusions
The timber value chain is directly connected to the U.S. homebuilding sector as this is dominated by single-family home construction which, in the U.S., is estimated to have timber frames in ca. 90% of cases[8]. U.S. homebuilding activity is the fundamental driver of sales by durable wood products companies.
Construction speed is extremely important for homebuilders, as it increases asset turnover thereby improving returns on capital employed – this is an area where timber companies can and do add value through engineered products as well as value-added services including optimal logistics.
The differentiation of business models across homebuilders makes an apparently uniform sub-sector more interesting than it may initially seem, as factors such as the focus on manufactured vs. site-built homes, the exposure to speculative homebuilding, and the capital intensity with respect to land ownership, make each business different from its peers and require a fundamental analysis and understanding.
[8] https://eyeonhousing.org/2022/07/the-share-of-wood-framed-homes-increased-in-2021/
Disclosures and Conflicts of Interest
Some or all the companies mentioned in this report may be included in the Timber Finance Forest-Based Construction Basket tracker and are part of the Timber Finance Carbon Capture & Storage Index. Timber Finance Management and/or the Timber Finance Initiative may have commercial relationships or be in discussions with some of the companies mentioned in this report. Specifically, Stora Enso is a member of the Timber Finance Initiative association.
Please note that this research is prepared for information purposes and targeted to institutional investors in Switzerland. They do not represent investment advice and do not take into consideration the individual requirements, risk tolerance and goals of an investor. Recipients who are not Swiss institutional investors should seek the advice of their independent financial advisor prior to taking any investment decision based on this report or for any necessary explanation of its contents.
The information presented in this report is obtained from several different public sources that we consider to be reliable. Nevertheless, we cannot guarantee the accuracy of the presented information. The information used may change quickly and we are not committed or obliged to modify the reports base on new information. The opinions and views expressed in this report reflect those of the author at the point in time of its compilation and may vary at any time. Valuation methods like DCF and any other analysis or expert judgement do not provide any guarantee that the target price or fair value will be reached, for example because of unforeseen changes in financial or economic conditions.
For more details and questions, please do not hesitate to get in touch with us at info@timberfinance.ch or:
Timber Finance
Ausstellungsstrasse 36
CH-8005 Zürich
Tel. 044 991 13 44
General disclaimer © 2023 Timber Finance Management AG (“Timber Finance”). All rights reserved. Redistribution or reproduction in whole or in part are prohibited without written permission of Timber Finance. Timber Finance Management AG makes no representation or warranty, express or implied, as to the ability of any index to accurately represent the asset class or market sector that it purports to represent and Timber Finance Management AG and its third-party licensors shall have no liability for any errors, omissions, or interruptions of any index or the data included therein. All data and information is provided by Timber Finance “as is”. Past performance is not an indication or guarantee of future results. This document does not constitute an offer of any services. It is not possible to invest directly in an index. Exposure to an asset class represented by an index may be available through investable instruments offered by third parties that are based on that index.
Copyright © 2023 Timber Finance Initiative. All rights reserved.