Building Costs and House Prices
Introduction
Building Costs and House Prices have historically been viewed as inextricably linked, yet new research suggests this relationship is far more complex and weakening over time. In a June 2025 working paper titled "Building Costs and House Prices," economists Brian Potter and Chad Syverson provide a comprehensive analysis of the United States housing market spanning nearly three-quarters of a century.
Their findings challenge the conventional wisdom that rising construction expenses are the primary driver of escalating home values. By examining long-term trends and cross-city data, the authors demonstrate that while building costs remain a significant component of new home production, their explanatory power over final house prices has diminished significantly in recent decades. This summary explores the methodology, key statistical findings, and policy implications detailed in the National Bureau of Economic Research (NBER) working paper.The Weakening Link Between Construction Expenses and Market Values
The central thesis of the paper is that Building Costs and House Prices no longer move in lockstep. While it is intuitive to assume that higher costs to construct a home lead directly to higher sale prices, the data reveal a growing divergence. The authors note that building costs have never had overwhelming explanatory power over US housing prices, but even the imperfect correlations of the past have weakened further in recent years. This decoupling is evident across multiple dimensions, including aggregate national trends and specific metropolitan areas.
For much of the early post-1950 period, cumulative growth in construction costs actually outpaced housing prices. It was not until the late 1990s that cumulative price growth finally caught up to and surpassed cost growth.
Since then, two distinct periods have emerged where house prices rose notably faster than building costs: the mid-2000s leading up to the financial crisis, and the period from the mid-2010s to the present. This recent divergence occurs despite poor productivity growth in the construction sector, suggesting that factors other than pure construction inefficiency are driving price surges.
Methodology: Measuring Building Costs and House Prices Accurately
To analyze the relationship between Building Costs and House Prices, the authors rely on robust, industry-standard datasets. For construction costs, they utilize the Housing Cost database from RSMeans, which has provided cost estimates to the construction industry since the mid-20th century.
These estimates capture the costs and margins associated with physically constructing a building, excluding land acquisition and ancillary soft costs. The validity of RSMeans data is confirmed by its widespread use in both industry and academic research.
The authors verify the accuracy of RSMeans data by comparing it to two others widely known indexes. First, they compare it to the Census Single-Family Houses under Construction price index, finding that the two track each other closely nationwide, with minor departures during the volatile late 2000s.
Second, they compare city-level RSMeans data with the Turner and Townsend International Construction Market Survey. The cross-city correlations between these datasets range from 0.70 to 0.97, indicating that RSMeans data reliably reflects geographic differences in building costs.
For house prices, the study primarily uses the Federal Housing Finance Agency (FHFA) index, which is constructed from repeat sales or refinancings of the same house, thereby adjusting for quality changes in the housing stock. They also reference the Shiller home price index for longer historical trends. By normalizing these indexes against the GDP deflator, the authors isolate real trends in Building Costs and House Prices relative to the broader economy.
Historical Trends in Building Costs and House Prices
A long-run view of the data reveals several critical patterns regarding Building Costs and House Prices. Since 1950, construction-related prices have grown significantly faster than general consumer prices.
While the Consumer Price Index (CPI) saw cumulative growth of about 20 percent relative to the GDP deflator, housing construction costs and prices both more than doubled. This outsized growth is consistent with the well-documented productivity underperformance of the construction sector relative to the rest of the economy since 1970.
However, the relationship between the two-construction metrics has shifted. In the early decades of the sample, building costs grew faster than house prices. From 1950 to 1975, cumulative growth in the Shiller house price index was 30 percent less than the change in the RSMeans building cost index. The dynamic flipped in the late 1970s, with house prices beginning to rise faster than construction costs.
Over the entire sweep of the data, house prices cumulatively rose only modestly more than building costs. However, focusing on the period since 1975 reveals a starker divergence, with cumulative FHFA house price growth outpacing building costs by over 60 percent.
City-Level Divergence in Building Costs and House Prices
When drilling down to the city level, the disconnect between Building Costs and House Prices becomes even more apparent. The authors match city-specific building cost indexes from RSMeans with FHFA housing price indexes for metropolitan statistical areas. Summary statistics show that for seven of the ten five-year periods analyzed, average city-level house price growth rates exceeded average city-level building cost growth rates.
The volatility of house prices is substantially higher than that of building costs. The standard deviation across cities for house price growth rates ranges from 2 to 4 percentage points per year, while for building costs, it ranges from only 0.5 to 1 percentage point.
Cross-sectional regressions reveal a declining ability of building cost growth to predict price growth. In the 1975-80 period, building costs explained 64 percent of the variation in house prices (R-squared of 0.641). By the 1995-2000 period, this explanatory power dropped to virtually zero, and it has remained low in subsequent decades.
Certain cities experience massive gaps between Building Costs and House Prices. Coastal metros such as San Francisco, Los Angeles, Seattle, New York, and Boston show extreme rises in housing prices compared to costs. In contrast, cities that have historically built housing in large volumes, such as Dallas, Houston, Atlanta, and Phoenix, show a smaller difference. However, even in these high-volume markets, house prices have risen dramatically faster than costs over the last decade.
Material Versus Installation Costs in Housing Production
An important nuance in understanding Building Costs and House Prices is the breakdown of construction expenses into materials and installation. Since 2007, RSMeans has provided separate indexes for these components. The material index covers items like concrete, lumber, and fixtures, while the installation index reflects labor-centered services and assembly costs.
In the early 2010s, installation cost growth outpaced material cost growth. This pattern reversed in the latter half of the decade, with materials cost growth accelerating due to supply chain frictions and goods inflation. During the post-Covid period, inflation in materials far outpaced installation costs.
However, at the city level, installation cost growth varies much more across geographies than material cost growth. This suggests that labor markets are more local and fragmented, while material prices are more uniform due to tradeability. Consequently, any remaining correlation between Building Costs and House Prices is likely driven more by local labor costs than by material prices.
Price Levels and Residuals in Major Metropolitan Areas
The authors also examine cross-sectional relationships in levels rather than just growth rates. Using 2024 data, they compare Zillow median house prices with RSMeans construction cost estimates for a standardized 2,000-square-foot home. While there is a positive correlation, building costs explain only a fraction of the variation in house prices. When adjusting for new construction specifically and measuring price per square foot, the explanatory power improves, but nearly half of the variation remains unexplained.
Cities with the highest residuals—where prices are much higher than predicted by Building Costs and House Prices models—include Miami, San Jose, Los Angeles, and San Francisco. Miami, for instance, has prices per square foot that are 98 percent above what would be predicted by local construction costs.
Interestingly, many of these high-residual cities also have high absolute building costs. San Francisco and San Jose have the second and third highest estimated building costs among all cities studied. This indicates that their high price residuals exist despite their high costs, not because of low costs. Conversely, cities like Chicago and Minneapolis have prices significantly lower than predicted by their relatively high building costs.
Policy Implications for Building Costs and House Prices
The findings have significant implications for housing policy. If Building Costs and House Prices were tightly linked, policies aimed at reducing construction costs would be the most effective way to lower housing prices. However, the weakening correlation suggests that other factors are increasingly dominant. The authors point to increasingly binding restrictions that prevent housing supply expansion in high-demand areas as a primary driver of the decoupling.
While stagnation in construction productivity and high building costs certainly contribute to high housing prices, they are overshadowed by land-use regulations and supply constraints. In cities with severe zoning restrictions, the price of land and the scarcity of permitted housing units drive prices far above the replacement cost of the structure itself. Therefore, addressing the housing affordability crisis requires looking beyond construction efficiency to broader regulatory reforms.
Conclusion
The relationship between Building Costs and House Prices is evolving, with construction expenses playing a diminishing role in explaining final market values. As demonstrated by Potter and Syverson, while building costs remain a fundamental input, they no longer dictate housing prices in the way they once did.
The growing divergence, particularly in high-demand coastal markets, highlights the critical importance of land-use policy and supply-side constraints. For researchers, policymakers, and housing professionals, understanding this decoupling is essential for developing effective strategies to address housing affordability.
Future interventions must look beyond the construction site to the regulatory environment that governs land use and development rights. The ongoing value of this research lies in its clear empirical demonstration that solving the housing crisis requires a holistic approach that addresses both Building Costs and House Prices within the broader context of urban economics and policy.