0%
Chat
Research note

Canada Softwood and US Housing

Lumber prices move with housing demand and interest rates, but the wood behind them takes decades to grow. This note puts the forest, mill, housing and rate figures side by side.

Data as of 29 September 2026
$539

Lumber futures, per 1,000 board feet

29 Sep 2026, about 12% below a year ago

20 bn

Canadian softwood output in board feet

2024, down from 35 bn in 2004

1.275 M

US housing starts, annual pace

Aug 2026, lowest since Oct 2025

7.03%

US 30-year fixed mortgage rate

24 Sep 2026, 6.30% a year earlier

Summary

Calm for buyers, tight for producers

Buyers

Futures sit near a nine-month low. US mills and imports have covered most of the Canadian shortfall, and total North American shipments have been close to flat.

Producers

Canadian output is down about 43% since 2004. Duties, weak demand and fewer accessible trees have closed mills and left others part-idle.

Forest

Fire and insects affect far more area each year than harvesting does. Permanent forest loss is small, but the wood supply that can be reached is shrinking in some regions.

What would change it

A housing recovery that lifts demand faster than supply can respond. The 2021 surge is the example of how quickly that can happen.

Trade

US and Canada goods trade

The United States runs a goods deficit with Canada, but it is small next to its deficits with Mexico, Vietnam, Taiwan and China. A trade balance is an accounting figure. Bilateral balances reflect supply chains and capital flows, and they say little on their own about whether trade is fair.

US goods trade with CanadaExportsImportsBalance
2025, full year$333.6 bn$381.9 bn−$48.3 bn
Jan to Jul 2026$205.5 bn$233.7 bn−$28.2 bn
Jul 2026, seasonally adjusted$29.3 bn$32.5 bn−$3.2 bn

US Census Bureau and Bureau of Economic Analysis. Not seasonally adjusted except the July line.

Agreements

Free trade agreements have been in force since 1 January 1989, about 37 years. The Canada–US agreement ran from 1989, NAFTA from January 1994 to June 2020, and the USMCA since July 2020. On 1 July 2026 the US declined to extend the USMCA for a new 16-year term. The agreement stays in force, with joint reviews every year through 2036.

Forest

Harvest, regrowth and disturbance

  • Canada has about 369 million hectares of forest (2023). Less than 0.5% has been deforested since 1990, and about 94% of it is public land.
  • Harvest covered about 669,000 hectares in 2023, roughly 0.2% of the forest. Regeneration is a licence condition on public land, through planting or natural regrowth.
  • Deforestation means permanent conversion to farms, mines, energy or development. Harvest, fire and insect damage do not count, because the land is expected to regrow.

Area burned compared with area harvested

Fire 202317.2 M ha
Fire 20258.78 M ha
Fire 20245.4 M ha
Fire 2026, to 31 Aug4.49 M ha
Harvest 20230.67 M ha

Fire totals for 2024 are to October and for 2025 to mid-September. Much of the burned area is remote northern forest outside the commercial base. In 2023, 7.9 M of the 17.2 M ha burned was in managed forest. Insects affected 11.6 M ha in 2023 and 13.1 M ha in 2022, and that count includes light defoliation.

The open questions are about standards. Burned or beetle-killed land often has no harvester responsible for replanting, natural regrowth can return a different species mix, and old-growth forest does not come back on a planting timetable.

Supply

Mills and lumber prices

01020304035≈27.520.919.8≈2020042017202220232024Billion board feet
Canadian softwood lumber production. The 2017 value is derived from a reported 28% decline to 20 bn in 2024, and 2004 and 2024 come from a trade-dispute response, so treat them as approximate. The 2022 and 2023 values are reported shipments.
  • Total North American shipments were about 57.2 bn board feet in 2023 and 57.0 bn in 2024. US mills produced about 37 bn.
  • One US Lumber Coalition paper puts Canadian mill utilization at about 70% in 2024. That group is a party to the trade dispute, so the figure is an advocate's estimate.
  • British Columbia's production is about half its 2017 level (RBC Economics).

Lumber futures, USD per 1,000 board feet

2021 peak≈1,711
2026 high, 23 Jul664
2025 close576
2026 low, 18 Sep541
29 Sep 2026539

CME lumber futures are thinly traded, so prices can move sharply on small volumes. Physical framing lumber (NAHB index) was $506.76 on 25 Sep, up 4.5% year over year while futures were down about 10%.

Land

How much forest one mill needsEstimate

Under a steady annual cycle, land needed equals annual volume divided by growth per hectare per year. Equivalently, it is the acres cut each year multiplied by the rotation length. The example below is a mill making 200 million board feet a year, using about 3.6 m³ of logs per 1,000 board feet. These are my own assumptions and arithmetic, and real mills vary widely.

Forest typeRotationGrowth, m³/ha/yrLand neededCut per year
Southern-style pine plantation30 yr10≈180,000 ac≈6,000 ac
Intensive fast-growing plantation25 yr15≈120,000 ac≈4,800 ac
Canadian boreal forest80 yr2≈890,000 ac≈11,000 ac

Gross area will be larger once roads, streams, buffers and reserves are removed. A slow growth rate is the reason boreal supply needs five to seven times the land of a plantation.

What a short rotation is worth

A US South pine stand at 26+ years can bring about $2,110 to $2,275 per acre at clearcut, against $200 to $400 per acre to establish. Taking $2,200 at year 28 and $300 at planting, the value of the bare land depends heavily on the discount rate.

Discount rateBare-land value per acreEstimate
4%≈ $650
6%≈ $160
8%about zero or negative
Demand

US housing

  • Starts were 1.275 M at an annual pace in August 2026: 918,000 single-family and 344,000 multifamily. Completions were 1.128 M, down 27.1% from August 2025.
  • In 2025, starts were about 1.36 M. Goldman Sachs research, cited in industry coverage, says annual starts near 2 M would be needed to close the gap.
  • Homeownership was 65.0% in Q2 2026, unchanged from a year earlier. For under-35s it fell from 36.4% to 35.2%.
  • There were 133.8 M households, up from 132.6 M a year earlier, so about 1.2 M net new households against roughly 1.3 M starts.

Estimates of the shortage, in millions of homes

John Burns≈1.0 M
NAHB≈1.2 M
Realtor.com4 M+
White House CEA10 M+

The methods differ. The White House figure counts missing single-family homes against a historical building trend. The others use vacancy, total units or population trends.

Who owns the stock

  • Investors made 27% of single-family purchases in Q2 2026 (Cotality), about 273,000 homes and 40,000 fewer than a year earlier. Their share peaked near 30% in late 2025 and was under 20% for much of the 2010s.
  • About 17 M of 86 M single-family homes and townhouses are not owner-occupied, and 87% of those belong to owners with one to five properties (BatchData). Large institutions made about 1% of purchases (Realtor.com).
Rates

Interest rates and oil

IndicatorLevelDate and note
Fed funds target3.75% to 4.00%Raised 25 bp on 16 Sep 2026
30-year mortgage7.03%24 Sep 2026, first reading above 7% since Jan 2025
15-year mortgage6.42%24 Sep 2026
10-year Treasury≈4.95%10 Sep 2026
Brent crude≈$102 to $10625 to 28 Sep 2026, peak near $118 in late March
WTI crude≈$9328 Sep 2026; record is $147.27 in July 2008

The energy shock comes from the 2026 Iran war and disruption around the Strait of Hormuz. It keeps inflation up and rates high, which weighs on housing starts and so on lumber demand. My own reasoning, not from the sources, is that wood also gains against steel and concrete when energy costs rise, because those take more energy to make.

For scale, from general knowledge and not from the sources below: the Fed funds rate reached about 20% and mortgage rates about 18% around 1981.

Outlook

How the pieces connect

Forest growth sets how fast supply can respond. Housing starts and mortgage rates set demand. Oil and inflation feed into rates, and fire and insects reduce the wood base underneath all of it. Price is where they meet, which is why it can stay calm for years and then move sharply.

Prices are soft now because housing is subdued, and not because supply is abundant. If borrowing costs fall and starts climb toward the level analysts say is needed, demand could outrun Canadian output, which is smaller than it was and cannot be expanded quickly. Replanting is already a built-in cost, so it does not move prices. Losses to fire and insects, trade duties and interest rates do.

Sources

Where the figures come from

Trade and policy

Forest

Lumber

Housing

Rates and oil

Figures were current on 29 and 30 September 2026. Items marked Estimate are my own arithmetic from the stated assumptions, and the one general-knowledge comparison is labelled as such. This is general information and not financial advice.