Lumber Builds the Future

Projected Housing Demand and Lumber Consumption in the U.S.

A third of all U.S. lumber is used to build new homes or apartment buildings. Photo by byrev, CC 2.0.

“Lumber and economic growth are tightly connected,” says USDA Forest Service scientist Jeffrey Prestemon. “At a certain rate of GDP growth, you get a certain path of lumber consumption.”

Prestemon recently led a modeling study that projects softwood lumber demand under different economic growth scenarios, represented by real gross domestic product. Real GDP is adjusted for inflation and reflects the value of all the goods and services an economy produces.

SRS scientists David Wear and Karen Abt contributed to the study, along with NCSU professor Robert Abt. The study was published in the journal Forest Science.

Housing demand reflects prices, income, taxes, credit conditions, interest rates, and demographic factors. Fully specified models use many such variables to predict the future. Prestemon and his colleagues developed a reduced-form model, which is simpler.

“Simplified models can be just as precise and accurate as the more complex ones,” says Prestemon. “Our model lets us make the projections with fewer data. And testing shows that it’s as unbiased as models that are fully specified.”

Prestemon has already begun using the simplified model for his work under the Resource Planning Act and with the U.N. Economic Commission for Europe. The Commission is a postwar institution that keeps Europe, the U.S., and other partner countries talking about a number of issues – including housing – and their policy solutions.

“We’re always looking forward,” says Prestemon. “What could the future of the forest sector look like?”

Housing demand is a key factor. A third of all U.S. lumber is used to build new homes or apartment buildings.

How many houses will be built in 2050? How about 2070? The model uses three variables to predict the number of housing starts: the rate of economic growth, the number of houses built in the past, and mortgage delinquency rates. Mortgage delinquencies, when a homeowner falls behind on monthly payments to a lender, spike during recessions but otherwise occur at fairly stable rates.

Softwood lumber – much of it imported – is used for new homes and apartment buildings. Photo by Bart Everson, CC 2.0.

“Those three variables can predict 90 percent of the quarter to quarter variation in the national total number of housing starts,” says Prestemon.

The scientists tested several future economic growth scenarios, represented by real GDP, through the year 2070. They used Monte Carlo simulations to identify confidence bands in the projected number of total housing starts, given the GDP growth assumptions.

For the past 15 years, GDP has grown by an average of 2.4 percent each year.

In the future, if real GDP grows by two percent, the model suggests 1.1 million new homes would be started. In that scenario, lumber demand would stay roughly the same as today, neither increasing or decreasing substantially.

If GDP grows faster than two percent, lumber demand will increase. “In a vigorous economy, some houses built before the 1970s – little houses of 1,000 square feet with eight-foot ceilings – will likely be torn down, even if population growth is slow,” says Prestemon.

But if GDP grows more slowly than two percent, demand for lumber would likely fall. “People would be more likely to repair old houses, use less new wood, and become more efficient in how much wood they use in new houses,” says Prestemon.

Since 1950, the inflation-adjusted rate of economic growth in the U.S. has slowed by an average of 0.04 to 0.05 percent of GDP each year. The U.S. population is also growing more slowly, down from 1.8 percent per year in the 1950s to less than 0.9 percent per year in the last ten years, which could explain some of this slowing economic growth.

“We didn’t set out to model a world of shrinking population,” says Prestemon. “But we modeled forest sectors under different economic scenarios. One of those was low economic growth, which could be caused by low population growth.”

If GDP grows more slowly than 2 percet, people would be more likely to repair old homes. Photo by Joanna M. Kresge, U.S. Air Force.

Low or no population growth is a reality for Japan and some European countries. Countries with shrinking populations can still build wealth, but their demand for lumber is expected to decline.

Currently, the U.S. is the biggest producer and consumer of forest products in the world, though dependent on wood products imports to meet demand for construction.

Prestemon and his colleagues aim to understand how the evolution of the U.S. and world economies will affect the U.S. forest sector. As Prestemon says, “My interest is to study how the world’s forest sector evolves over time. The U.S. housing sector is a key driver of demand for wood products, and it’s also important for the broader economy.”

Read the full text of the article.

For more information, email Jeffrey Prestemon at

Access the latest publications by SRS scientists.

Subscribe to our newsletter!

Receive weekly updates