Timber

The question

Are timber REITs, wood products makers, and containerboard producers priced for a permanently lower housing plateau and a price-taker packaging industry, or for a housing cycle that mean-reverts off a demographic deficit and a packaging industry that just consolidated from five serious North American players to three?

Base rates

The reference class is prior housing down-cycles in the United States: 1990-1993, 2006-2011, and 2022-present. In each prior episode, starts troughed near 1.0-1.1 million units and recovered to 1.5-1.7 million over a three-to-five year cycle. Timber REIT total returns from trough averaged +14-19% annualized over the subsequent three years, with dividend support providing a floor. Wood products equity returns were more dispersed -- +25% annualized in the LPX-TREX archetypes and closer to index returns in the commodity-lumber archetypes -- because product mix at the mill determines how much of a price move reaches the equity.

The cohort has trailed SPY trailing-1Y meaningfully. The index compounded roughly 28.6% while timber and wood products were flat and paper-and-packaging lagged mid-single digits. That underperformance is the thesis. Multiples on IP, SW, PKG, WY, and LPX have compressed to levels that embed either a permanent housing reset or a failure of post-merger containerboard discipline. Neither is the modal outcome.

The packaging sub-base-rate is more specific. After the IP-DS Smith and Smurfit-WestRock combinations, roughly 70% of North American containerboard capacity sits with three operators. In prior consolidation episodes -- cement in the late 1990s, railroads in the early 2000s, US airlines post-2013 -- the two-to-three-year path has been margin expansion of 300-500 basis points driven by capacity discipline rather than price collusion. The reference class is mechanical, not speculative.

For timber REITs, the binding variable is NAV per acre. Current NAV discounts on WY, RYN, and PCH are roughly 15-25%, comparable to prior cycle troughs in 1995 and 2011, both of which closed within 24 months of the cyclical inflection in starts.

Why consensus is wrong

Consensus treats the current housing cycle as demand-impaired. The binding constraint is supply. The US housing deficit accumulated since 2008 is roughly 3.5-5.5 million units depending on the definition. Household formation is still running ahead of completions. The mortgage-rate sensitivity that dominated 2022-2024 has already shown up in starts; the deferred demand has not. When mortgage rates move through the low-6s, the elasticity is stepwise, not linear, because affordability constraints clear in thresholds rather than smoothly.

The second miss is containerboard pricing. Consensus reads the three-player industry as still price-taking because inventory-to-shipment ratios have not yet compressed. The reference class says the compression lags the consolidation by two to four quarters -- operators take time to realign capacity, close sub-scale mills, and re-sign contracts on the new footprint. Looking at the spread rather than the inventory print is the correct signal, and the spread has already inflected.

Third, the timber REIT segment is being priced as an interest-rate asset because of the REIT wrapper. The underlying business is a biological compound machine -- trees grow roughly 4-6% a year in merchantable volume regardless of what the Fed does. The NAV floor is structural. The equity is trading through that floor, which has historically been a reliable three-year signal.

Position construction

The book has one 20% anchor, three cluster sub-books, and a specialty consumer tail.

Anchor. IP at 20% is the North American containerboard leader post-DS Smith integration, with the broadest capacity footprint and the clearest synergy path. Incremental margins at normalized utilization are the single largest driver in the book.

Paper and packaging cluster (~42.2%). SW at ~18.6% is the Smurfit-WestRock combined entity, the second leg of the consolidation thesis with European and Americas exposure. PKG at ~15.8% is the highest-margin containerboard pure-play with a disciplined operating history that pre-dates the current consolidation wave. GPK at ~4.4% adds consumer paperboard with structural growth from plastic-to-fiber substitution. SON at ~3.4% is the specialty converter franchise with a distinct end-market mix.

Timber REIT cluster (~20.8%). WY at ~14.7% is the largest US timberland owner with a diversified product mix across lumber, engineered wood, and real estate. RYN at ~3.3% adds Pacific Northwest and New Zealand timberland with a cleaner REIT structure. PCH at ~2.9% rounds out the Southern yellow pine exposure.

Wood products cluster (~16.9%). LPX at ~5.4% is the OSB and SmartSide franchise -- the siding sub-segment is taking share from fiber cement and vinyl on a structural curve. BCC at ~2.3% is the distribution and engineered wood products play with counter-cyclical margin characteristics. UFPI at ~4.6% spans industrial, construction, and retail wood products with a notably high return on invested capital. TREX at ~4.7% is the composite decking franchise where incremental growth is substitution-driven rather than housing-cycle-dependent.

Asymmetric payoff

If US housing starts recover toward 1.5 million over the next two years, containerboard operating rates tighten into the consolidated footprint, and timber REIT NAV discounts close to historical mid-cycle levels, the weighted book returns roughly 17-24% annualized over three years. If starts flat-line at 1.1-1.2 million and containerboard discipline breaks, the book returns roughly -3% to +4% with dividend support from WY, IP, SW, and PKG providing a floor. If mortgage rates reach the mid-5s and consolidation margin capture exceeds 500 basis points, the right tail is 32-42% with multiple expansion concentrated in the three containerboard anchors.

At 55% base, 25% bear, and 20% bull, expected value is roughly +13 to +19% annualized against an SPY base rate near +8%. The payoff is asymmetric because the NAV floor on timber, the dividend support across the cohort, and the mechanical nature of post-consolidation margin expansion truncate the bear case while housing mean reversion remains the base case rather than the bull case.

Three things that would change our mind

1. US housing starts failing to cross 1.3 million on a trailing four-quarter basis through the end of 2026, with Census data indicating household formation is slowing rather than accumulating against a supply deficit.
2. Containerboard operator capacity announcements showing a net addition rather than a net reduction in North American tonnes through 2026, with management commentary framing the additions as competitive rather than committed customer demand -- this would falsify the consolidation-discipline thesis directly.
3. Timber REIT NAV discounts widening beyond 30% for two consecutive quarters, with private timberland transaction data showing implied acre values compressing rather than holding, which would indicate a structural rather than cyclical repricing of the underlying asset.

What we're explicitly NOT betting on

We are not betting on a specific Fed rate path. We are not betting on a particular lumber price print. We are not betting on a single homebuilder -- the book holds the land and the inputs, not the finished product. We are not betting on the timing of any FTC second review or antitrust challenge to post-merger integration. We are not betting on a plastic-packaging ban. The thesis requires only that housing starts mean-revert toward trend, that three-player containerboard behaves like every prior consolidated commodity industry, and that timber NAV stays anchored to private-market transactions. All three are weaker claims than betting on rates or housing timing, and the book is sized for them.