Initiated 2026-09-24 · Price $43.62 (as of 2026-09-23, stockanalysis.com) · Mkt cap $4.44bn · Building products / outdoor living · Model: verified

Rating: Hold — Conviction: Medium

Trex is a genuinely good business — 38% gross margin, 13% ROIC, 0.8x leverage, three quarters of its market still unconverted — priced at a level that needs most of that story to arrive on schedule. The base driver set, which grants the guided year and five more of steady conversion, values the equity below today's price; the upside case is real and worth about 30%, but the market already pays roughly halfway to it.

Model value range vs price
Bear $19.25Base $36.83Bull $57.12Price $43.62

Business overview

Trex makes wood-alternative composite decking and railing from ~95% recycled material (reclaimed polyethylene film and wood flour), sold through a two-step distributor and dealer channel plus big-box retail into US residential repair-and-remodel. It reports as a single segment since exiting Trex Commercial; decking is the large majority of FY2025's $1.174bn of revenue, railing the faster-growing minority. Three variables drive earnings: the rate at which homeowners convert from wood (still ~75% of the decking market, each point of share ~$80mm of sales); utilization at the new Little Rock, Arkansas plant, which turns today's depreciation drag into tomorrow's cost advantage; and a household's willingness to commit to a $10k–$30k discretionary project, which is rate- and confidence-sensitive.

Bull case

  1. The conversion runway is long and needs no boom — wood is still ~75% of decking and Trex is the default premium alternative; management also flags ~$100mm of revenue held by tertiary brands it can take through distribution. Plays out if housing turnover and R&R activity stay merely adequate. Model: rev_growth
  2. Little Rock flips from drag to engine in 2027 — the plant ran six months ahead of schedule, half its decking lines live by end-2026, described as the lowest-cost base from 2027 and sized for $1.8–2.0bn of revenue; add a targeted 500bp of railing gross-margin gain over three years and ~100bp of margin per $100mm of incremental revenue. Plays out if utilization rises without a demand air pocket. Model: ebitda_margin
  3. Adjacency optionality is cheap and starting — entry into PVC decking opens a category Trex has never competed in, and Q2 2026 brought the first meaningful entry-level Enhanced increase in years. Plays out if the budget-conscious consumer keeps returning. Model: rev_growth
  4. The capex cliff is the near-term cash event — capex was $223.6mm in FY2025 (19.0% of revenue) and $232.3mm in FY2024; FY2026 guidance is $100–120mm and steps down after. That swing already funds ~$300mm of 2026 repurchases ($153.6mm in H1, $150mm more authorized). Model: capex_pct_rev

Bear case

  1. This is the most deferrable purchase in housing — revenue fell 8.5% from FY2021's $1.197bn to FY2023's $1.095bn, and FY2025 is still below that peak five years on. Plays out if unemployment rises or home-equity borrowing costs stay where they are. Model: rev_growth
  2. The fixed-cost base just got much bigger — Q2 2026 gross margin fell to 37.9% from 40.8% on Little Rock depreciation, mix and ramp inefficiency (~100bp of it), and FY2025 EBITDA fell to $320.9mm from $360.3mm while revenue grew. Capacity sized for $1.8–2.0bn against $1.2bn of revenue cuts both ways. Model: ebitda_margin, da_pct_rev
  3. Pricing power is thinner than the brand implies — management's stated plan is modest pricing to offset inflation, not annual increases, prioritizing volume and operating leverage. Freight, diesel and aluminium inflation therefore lands in margin, not price. Model: ebitda_margin
  4. Working capital has quietly absorbed the growth — inventory went from $107.1mm at FY2023 to $238.7mm at FY2025, operating NWC to 21.5% of revenue against an 18.1% five-year average. Plays out if the build proves permanent rather than pre-ramp. Model: nwc_pct_rev

Valuation & balance sheet

Metric (definition) Current Own history (range or 5y avg) Peers Source, as-of
Forward P/E (consensus NTM EPS) 22.7x ~40.5x 5y avg trailing; 74.6x peak Q4 2021 n/a (unverified) stockanalysis.com / fullratio.com, 2026-09-24
EV/EBITDA (trailing: EV $4.74bn / TTM EBITDA $312.8mm) 15.2x ~21.0x 10y median; range ~10.1x–49.4x James Hardie ~17x fwd (absorbed AZEK, the only pure-play comp) stockanalysis.com / gurufocus.com, 2026-09-24
FCF yield (TTM FCF $208.1mm / mkt cap) 4.7% n/a (unverified) n/a (unverified) stockanalysis.com, 2026-09-24
Net leverage (net debt / FY2025 EBITDA) 0.77x n/a (unverified) n/a (unverified) model-summary.json; 10-Q 2026-06-30

Model-implied value range (from model-summary.json; generic module, Gordon-growth DCF and exit-EV/EBITDA DCF, midpoints): Bear $19.25 · Base $36.83 · Bull $57.12 per share, i.e. implied returns of −55.9% / −15.6% / +30.9% vs $43.62. These ranges show how the bull and bear drivers translate into value; they are not price targets. Nothing floors at zero — the company is barely levered — so the whole range and every scenario row are informative. The price sits between Base and Bull, roughly 45% of the way up that span: the market pays for a good part of the conversion-plus-Little-Rock case, not for the base case. The method spread is wider than most names in this rotation and is itself a finding: base Gordon $28.34 against exit-multiple $45.31. Gordon penalises near-term free cash flow still absorbing 21% working capital and 8.9% capex; the multiple method looks past both to Y5 EBITDA of $478mm at 14x. Which is right is largely the question of whether the capex and inventory build normalise.

Why Hold and not Sell. The Base implied return of −15.6% is more bearish than the call. I stop short of Sell because the base already embeds a de-rating (14x exit against 15.2x trailing and a 21x ten-year median) and a 10.5% WACC on a raw 1.47 beta, so part of the shortfall is discount-rate conservatism rather than business deterioration; because 0.8x leverage with nothing due before 2031 means no path is forced; and because the capex step-down is guided, not hoped for. It is not a Buy either: the base branch loses money and the bear branch is a real −56%, because the fixed-cost base is now sized for a company half again this size.

Balance sheet: net leverage 0.77x FY2025 EBITDA, gross 0.79x, coverage 27.6x. Total debt is $253.0mm, all drawn on a $700mm revolver at Term SOFR plus spread (4.6% weighted average at 2026-06-30) running to 2031-03-26 — no maturity before then. Liquidity is $443.9mm undrawn (after $3.1mm of letters of credit) plus $6.5mm of cash; the thin cash balance is a choice, with free cash flow sweeping the revolver. Ratings: n/a (unverified) — no rated public debt identified this session.

Model note: tier full, status built, verification verified (all 3,188 formula cells recalculated in LibreOffice and matched). No unverified inputs, no assumptions without basis, all 14 scenarios consistency OK. Two conventions: historicals.ebitda is GAAP operating income plus D&A from two EDGAR endpoints and runs modestly below management's "adjusted EBITDA", so the FY2026 guidance midpoint was restated onto the GAAP basis for the Y1 margin; and operating NWC is receivables plus inventories less payables, because revolver draws in current liabilities in FY2022 and FY2024 make current assets less current liabilities unusable as a series. Generator note (2026-09-24): in this module share_change, capital_return_pct and cost_of_debt move EPS and the net-cash path only, and peer_pe feeds a cross-check only; none changes the value range, so bull/bear tags no longer cite them. Scenario deltas were rebuilt after multiple_pct shocks were extended to the Gordon method (previously they moved only the exit-multiple half), and EPS now uses average diluted shares.

Scenario stress test

Reasoned from the bull/bear drivers above. The model column comes from the scenario overlays (Base case + shock).

Scenario Effect Mechanism Magnitude Model Δ value vs Base ($/sh)
S1 Fast equity crash − 1.47-beta discretionary name de-rates with the tape; no fundamental change in weeks High −$5.62
S2 Slow bear / recession − The deck project is deferred first; FY2021–FY2023 already showed an 8.5% revenue decline High −$9.48
S3 Rapid rate shock − Home-equity and HELOC rates set the project's financing cost; demand and discount rate hit together High −$9.50
S4 Slow rate grind − Same financing channel, grinding; the floating revolver reprices continuously Med −$4.94
S5 Soft-landing cuts + Cheaper home-equity borrowing releases deferred projects with the labour market intact — the cleanest positive case here High +$7.93
S6 Recession-driven cuts − Rate relief cannot offset a breaking labour market when the purchase is deferrable Med −$3.56
S7a Credit liquidity shock − Little direct exposure at 0.8x with a committed 2031 revolver; the hit is multiple, not funding Low −$3.12
S7b Slow default cycle − Distributor and contractor channel stress plus tighter consumer credit slows sell-through Low −$1.98
S8 Stagflation − Freight, diesel and aluminium inflation against a stated policy of only modest pricing, volume already slowing Med −$4.47
S9a Dollar spike 0 No material effect, not modeled — revenue essentially all domestic, 95% of input recycled US polyethylene — $0.00
S9b Dollar slide 0 No material effect, not modeled — mirror of S9a — $0.00
S10 Melt-up + High-beta name with a visible 2030 growth target re-rates fastest when risk appetite outruns fundamentals Med +$4.68
S11 Energy supply shock − Diesel and freight costs (flagged on the Q2 call) plus a gasoline squeeze on the discretionary budget Low −$1.26
S12 Mega-cap/AI derating 0 No material effect, not modeled — no mega-cap or AI linkage; rotation benefit is relative performance the model does not price — $0.00

Magnitude labels re-ranked 2026-09-25 to match the model's dollar deltas (skill 3a-v-b item 7; the 2026-09-24 generator fix enlarged multiple-shock rows): S11 Med→Low, S8 High→Med.

Three of the four scenarios currently live or on watch are negative channels here and the one clean positive is not among them — part of why this is a Hold. Currently active/on watch per the playbook: S3 partially active (price leg met, pace legs short); S8, S10 and S11 on watch, S11 escalated after the 2026-09-19/20 Saudi/Houthi events.

Model value change vs Base, by scenario
S3 Rapid rate shock−$9.50S2 Slow bear / recession−$9.48S1 Fast equity crash−$5.62S4 Slow rate grind−$4.94S8 Stagflation−$4.47S6 Recession-driven cuts−$3.56S7a Credit liquidity shock−$3.12S7b Slow default cycle−$1.98S11 Energy supply shock−$1.26S9a Dollar spike$0.00S9b Dollar slide$0.00S12 Mega-cap/AI derating$0.00S10 Melt-up+$4.68S5 Soft-landing cuts+$7.93

What would change the call

Upgrades if: the price moves materially below the Base midpoint with FY2026 guidance intact; or gross margin holds at or above 39% for two consecutive quarters, putting the bull margin path in play; or FY2026 capex lands at the $100mm low end with revenue at the top of the guided range. Downgrades if: FY2026 revenue guidance is cut; or gross margin falls below ~36.5% on ramp costs that prove structural rather than transitional; or operating NWC exceeds 22% of revenue at a second consecutive year-end.

Watch items

Sources