Building Products: Residential Starts, R&R Spending, and the Housing Cycle

Introduction

An estimated 1,358,700 homes were started in the US in 2025, a 0.6% decline from 2024, while building permits fell 3.6% to approximately 1,425,200 units for the year. NAHB builder confidence remained below the breakeven level of 50 for every single month of 2025, ending the year in the high 30s. By any measure, new residential construction had a difficult year. Yet Masco Corporation, one of the largest building products companies in the US, maintained 16.8% operating margins on $7.6 billion in revenue, and the US remodeling market grew to a projected $509 billion. The reason: roughly 80% of Masco's revenue comes from repair and remodel (R&R) spending, not new construction. The two demand streams operate on completely different clocks, respond to completely different drivers, and require completely different analytical approaches.

Understanding this dual-demand structure is the foundation of building products coverage. A banker who models building products revenue using a single growth rate is missing the most important dynamic in the sub-sector.

New residential construction follows a chain of indicators that lead and lag each other, giving attentive analysts several months of advance warning before revenue impacts hit building products companies.

Building permits are issued before construction begins and represent the earliest hard data signal of future construction activity. Approximately 1,425,200 permits were issued in 2025, down 3.6% from 2024 and at their lowest level since 2020. Permits lead housing starts by 1-2 months because builders obtain permits before breaking ground. A decline in permits today signals weaker starts in 60-90 days.

Housing starts measure the actual commencement of construction. The 1,358,700 starts in 2025 fell just below the 2024 level of 1,367,100. For building products companies, starts are the most directly relevant indicator because the first weeks of construction are when builders order roofing, framing lumber, insulation, windows, and HVAC systems. A building products company's new-construction revenue in Q2 will closely track the starts recorded in Q1 and late Q4.

Housing completions lag starts by 6-12 months (the time required to finish a home). An estimated 1,497,800 homes were completed in 2025, down 7.9% from 2024. Completions matter for finish products (plumbing fixtures, flooring, paint, cabinetry, decorative hardware) that are installed in the final months of construction. Masco's plumbing and decorative products see demand driven more by completions than by starts.

The mortgage rate environment is the upstream driver of the entire chain. Mortgage rates averaged approximately 6.6% in 2025, and forecasts for 2026 cluster around 6.3%. The "lock-in effect," where homeowners with sub-4% mortgages are reluctant to sell because they would face significantly higher rates on a new purchase, has suppressed existing home sales by an estimated one million transactions annually and boosted home prices by roughly 5-6% above where they would otherwise be. The lock-in effect is slowly easing: the share of mortgages above 6% now exceeds the share below 3% for the first time, and inventory has risen from a 2.3-month supply in 2021 to 4.1 months in 2025 (expected to reach 4.6 months in 2026, approaching balanced market levels). This gradual normalization suggests that existing home sales will increase in 2026-2027, providing a modest tailwind for building products (each home sale tends to trigger renovation spending by both the seller pre-sale and the buyer post-purchase).

Mortgage Rate Lock-In Effect

The phenomenon where homeowners who obtained mortgages at historically low rates (sub-4% during 2020-2022) are reluctant to sell their homes and purchase new ones at current higher rates (6%+), effectively reducing housing turnover. Academic estimates suggest the lock-in effect reduced nationwide home sales by more than one million transactions annually and boosted home prices by approximately 5-6%. For building products, the lock-in effect has a nuanced impact: it suppresses existing home sales (negative for renovation activity triggered by home sales) but extends homeowner tenure in current homes (potentially positive for maintenance and upgrade spending as homeowners invest in homes they expect to stay in longer).

R&R spending follows none of the patterns that drive new construction. It does not respond meaningfully to mortgage rates. It is not sensitive to builder confidence or housing affordability. And it operates on its own set of leading indicators that are completely different from the permits-starts-completions chain.

The US has over 140 million existing housing units with a median age exceeding 40 years. Every one of those homes has roofing, HVAC, plumbing, insulation, windows, and interior finishes that deteriorate over time and eventually must be replaced. An asphalt roof lasts 15-25 years. An HVAC system lasts 12-18 years. A water heater lasts 8-12 years. These replacement cycles create a floor of non-discretionary demand that persists regardless of economic conditions.

Total homeowner remodeling spending is projected to reach $524 billion by early 2026, a record high. The Harvard Joint Center for Housing Studies (JCHS) projects R&R spending growth in the mid-single digits in 2026, accelerating to high single digits in 2027. NAHB forecasts a 5% gain in residential remodeling activity in 2025 and continued growth in 2026. These projections are underpinned by three structural forces:

Deferred renovation demand. Millions of homeowners paused planned renovation projects during 2023-2025 due to high material costs, economic uncertainty, and tariff-related hesitation. As conditions stabilize, this backlog of deferred projects is being released, creating a multi-year R&R tailwind. By mid-2026, building product spending on remodeling is forecast to grow faster than new construction spending, positioning remodeling as the primary growth driver for the sector.

Aging housing stock. Homes built during the 1970s-1990s construction boom are now reaching the age where major systems (roofing, HVAC, electrical, plumbing) require replacement. This is not discretionary spending: a 25-year-old roof that is leaking must be replaced regardless of the homeowner's economic outlook. The wave of homes aging into the replacement zone will persist for decades, providing structural demand that is independent of housing starts.

Energy efficiency mandates. Updated International Energy Conservation Code (IECC) standards require higher-performance insulation, windows, and HVAC systems on every renovation that triggers a building permit. Each code update ratchets the standard higher, meaning every renovation project generates more building products revenue per job than the prior code cycle. The heat pump transition is the most significant code-driven tailwind, as jurisdictions mandate electric heating to replace gas furnaces in both new construction and major renovations.

The leading indicators for R&R spending are different from the housing permits chain. The JCHS Leading Indicator of Remodeling Activity (LIRA) is the most widely followed predictive model, providing four quarters of forward visibility into homeowner improvement spending. The NAHB Remodeling Market Index (RMI) surveys remodelers about current and future market conditions. And home equity levels (tracked by the Federal Reserve and CoreLogic) signal homeowners' financial capacity for renovation spending: higher equity provides both the collateral for home equity loans and the psychological comfort to invest in the property.

Not all building products respond equally to new construction versus R&R. Understanding the product-level demand split sharpens the revenue model significantly.

This product-level mapping matters because a "building products" company can have vastly different exposure depending on its product mix. Builders FirstSource (80%+ new construction) and Trex (70%+ R&R) are both "building products" companies, but their revenue dynamics, cyclical sensitivity, and appropriate valuation multiples have almost nothing in common.

For sell-side timing, the R&R/new construction mix determines the optimal window. A company with heavy new-construction exposure should be sold when housing starts are at or above the 1.5-1.6 million equilibrium and leading indicators (permits trending up, NAHB confidence improving, mortgage rates declining) suggest continued momentum. Launching a sell-side when starts are at 1.36 million (as in 2025) means current earnings are below mid-cycle for the new-construction component, requiring the CIM to make a recovery argument.

For R&R-heavy companies, the timing is more flexible because earnings are less volatile. The current environment (R&R spending at record levels, deferred demand releasing, remodeling forecast to grow faster than new construction in 2026) creates a favorable narrative regardless of where housing starts sit. A banker advising Masco on a hypothetical sell-side could position the company's 80% R&R exposure as a structural advantage: "The company's revenue is anchored by the $509+ billion US remodeling market, with demand driven by 140 million aging homes and accumulating deferred renovation projects, independent of new housing start volatility."

For normalization, building products requires a dual-driver approach: normalize the new-construction revenue component to equilibrium housing starts (1.5-1.6 million), and normalize the R&R component to trend growth (mid-single digits based on the JCHS LIRA forecast). The blended normalized EBITDA will typically be closer to current earnings for R&R-heavy companies (because R&R is already near trend) and further from current earnings for new-construction-heavy companies (because starts may be significantly above or below equilibrium).