Phase Space AI

Valuation

Turning Point Brands [TPB]

Turning Point Brands [TPB] — Valuation

Spot $75.71 · 19.475m shares · market cap $1,474.5m · net debt $118.0m · EV $1,592.5m · TTM revenue $480.9m (to 2026-03-31) · EV/Sales 3.31x · EV/EBIT 18.7x on TTM EBIT of $85.0m

EV is stated LEASE-EXCLUSIVE. Operating lease liabilities are immaterial on this filer (<$5m). Net debt = cash $192m − total debt $310m. Economic net debt is $132m once the 50%-owned half of the ALP VIE's $28.1m of consolidated cash is removed; that basis is stated but the headline uses the GAAP consolidated figure.


Company state: A (mature and structurally stable) — with the mix shift flagged

Test applied: profitable; operating margin low-variance across ≥5 years; no structural regime change.

2020 2021 2022 2023 2024 2025 TTM Q1'26
Operating margin 15.9% 20.3% 24.7% 24.7% 24.0% 20.6% 17.7% 10.0%
Gross margin 46.9% 48.9% 55.4% 56.3% 55.9% 57.1% 56.8% 55.0%

Declared A — profitable in every year of a nineteen-year AV record, no acquisition of scale, consumable nicotine demand with no exogenous cycle to date. It is not STATE B: nothing here is driven by fab utilisation, memory pricing or freight.

But two regime breaks are on the record and must not be smoothed over:

  1. The Creative Distribution Solutions wind-down. FY2021 revenue $445.0m → FY2022 $321.2m. The screen's cagr_base_revenue of $321.2m is FY2022 — the post-divestiture trough — so its 13.0% "demonstrated CAGR" measures recovery off a divested base. Arithmetically correct, economically a different question.
  2. The mix shift now underway. Modern oral went from a rounding error to 105.2% of consolidated growth in FY2025 and the margin structure of the two businesses is materially different (see below). A company whose growth engine has a negative incremental operating margin in the most recent quarter is not, in that quarter, structurally stable.

evidence_grade: B−. Segment disclosure is genuinely good — two segments, revenue, cost of sales, gross profit, "other segment items" and operating income for each, three years and quarterly. Absent: any A&P line, any volume/price split, any named PMTA product line, any FDA decision date.


Terminal margin: 18.0%, below the trailing FY2025 actual, with the causal bridge required

valuation.md rule 2: "What a below-current terminal margin requires is an explicit causal bridge, not prohibition." Here is the bridge.

Opex bridge reconciled from the 10-K segment note, FY2025 (% of net sales of $463.1m):

line FY2025 $m % of sales
Gross profit 264.3 57.1%
Other segment items (primarily selling and marketing) 96.3 20.8%
Corporate unallocated (incl. $4.8m PMTA) 72.7 15.7%
Consolidated operating income 95.3 20.6%

Terminal bridge (year 5) — and the causal argument for each line:

terminal FY2025 Δ cause
Gross margin 56.0% 57.1% −1.1pp Mix. Modern oral carries a lower gross margin than legacy MST — Stoker's segment gross margin fell 57.5% → 54.0% in Q1'26 as modern oral scaled, and Zig-Zag's fell 55.4% → 53.7% on tariffs. Modern oral is contract-manufactured, not owned-plant. As modern oral goes from ~29% to a majority of revenue, the blended gross margin follows it down.
Selling & marketing (other segment items) 23.0% 20.8% +2.2pp This is the A&P assumption and it is the opposite of leverage. Stoker's S&M intensity has gone 13.6% → 15.9% → 20.8% → 31.5% in Q1'26. Nicotine pouches are a share war against Zyn, Velo and On!; the growth is the trial and shelf spend. I model it declining from Q1'26's 31.5% (a launch peak) but settling above the FY2025 level, not below it.
Corporate unallocated 15.0% 15.7% −0.7pp The only genuine leverage assumed. FY2025's +34.3% jump was driven by joint-venture expenses; PMTA spend ($4.8m) caps out because the company states it will file no further applications.
m_EBIT,T 18.0% 20.6% −2.6pp

Constraints checked: - m_EBIT,T (18.0%) ≤ m_gross,T (56.0%) ✔ — 38pp of headroom. - Terminal is 2.6pp below the FY2025 actual and 0.3pp above the TTM actual of 17.68%. The DATA_DEFECTS.md warning — "if your terminal margin is below the company's own trailing actual, that is almost certainly an error — say why it is not" — is answered directly: it is below FY2025 because the revenue mix in year 5 is a different mix, and the incoming mix is lower-gross-margin and higher-S&M. It is above TTM because Q1'26's 10.0% is a launch quarter, not a run rate. Both directions are argued rather than asserted. - It is not monotonicity-violating: valuation.md rule 2 requires V_bull > V_base > V_bear, not m_bull > m_base > m_bear. A bull case here pairs a lower terminal margin with much higher revenue — modern oral winning national share — and that is a legitimate, higher-value outcome.

What this terminal margin assumes about A&P, stated as the brief requires: it assumes A&P intensity RISES 2.2pp and never levers. That is the honest reading of a business whose growth segment tripled its selling intensity in nine quarters while its operating income fell.


Implied-path test (primary output — terminal value ~80% of EV)

reverse_dcf.py --spot 75.71 --shares 19.475 --net-cash -118 --revenue 480.9
               --years 5 --wacc 0.095 --terminal-margin 0.18 --exit-multiple 15.0
               --fcf-margin 0.008 --hist-cagr 0.130

--fcf-margin passed as +0.008 (TTM FCF $4.0m / revenue $480.9m = +0.83%). The sign is positive but the magnitude is near zero, so the interim-FCF correction is worth well under 1pp here — the smallest correction of the three names in this cluster. The sign matters and it is positive, so omission would have overstated required CAGR and suppressed a PASS, not manufactured one. Basis: trailing TTM. Forward basis would be worse, not better — Q1'26 CFO was −$22m and the company is building manufacturing capacity, so if anything the forward FCF margin is negative and the correction would invert. Flagged as the GOOGL pattern in miniature: a company crossing toward negative free cash flow has a trailing sign that misleads about the direction of its own correction. I have used trailing, stated it, and flagged that forward would move the other way.

held fixed value
terminal margin 18.0%
exit multiple 15.0x EV/EBIT
WACC 9.5% (small cap, 54.2% realised vol, $310m of debt)
horizon 5 years
FCF margin +0.83% (trailing TTM; forward likely negative — flagged)

Required revenue CAGR: 13.8% Demonstrated: 13.0% (FY2022 $321.2m → FY2025 $463.1m; base year is the post-divestiture trough) Margin: demonstrated − required = −0.8pp

Verdict: FAIL — but marginally, and the correct label is FAIL rather than PASS WITH ARGUMENT.

The framework allows PASS WITH ARGUMENT where "the price requires more than demonstrated and there is a specific, evidenced reason (a named product cycle, mix shift, pricing action)." TPB has a named product cycle — modern oral, $107.7m of FY2025 growth, sized by the company. So the question is genuinely live.

I am scoring it FAIL, for a reason that is evidenced rather than narrative: the named product cycle is real but its unit economics at the operating line are currently negative on the increment. Stoker's added $28.4m of revenue in Q1'26 and lost $4.4m of operating income. A PASS WITH ARGUMENT requires the mix shift to be accretive; this one is dilutive as measured. If the Q2'26 print shows Stoker's operating income growing on growing revenue, this flips.

Implied compression: TPB trades at 18.7x TTM EV/EBIT. Solving at 15.0x assumes −19.8% compression.

Sensitivity over the exit multiple

exit EV/EBIT required CAGR vs demonstrated 13.0%
8.2x (warranted at steady state) ~24% −11pp FAIL
12.0x 17.0% −4.0pp FAIL
15.0x (base) 13.8% −0.8pp FAIL
17.0x 9.5% +3.5pp PASS
18.7x (no compression) ~7.5% +5.5pp PASS
20.0x 6.2% +6.8pp PASS

Flip point sits between 15x and 17x — i.e. the name passes only if it does NOT de-rate. For a business whose consolidated operating income fell 46.2% year over year in the most recent quarter, underwriting zero multiple compression is not conservative.

What steady_state_check.py flagged

steady_state_check.py --ticker TPB --terminal-margin 0.18 --exit-multiple 15.0

My response: this flag has more force on TPB than on the other two, because TPB's 10.9% ROIC is barely above its ~9.5% cost of capital. A business earning ~1.4pp above WACC does not deserve an 18.7x EV/EBIT multiple on any steady-state logic; it deserves something near book-value economics. The 15.0x base is the traded record's compromise, not a warranted number, and at the warranted 8.2x the name fails by roughly 11pp. Recorded as the single strongest objection to owning this.


12-month target: $66.85, −11.7% to spot (54th percentile of its own post-divestiture history)

Multiple history (own, EV/Sales, as-known TTM revenue lagged 45 days):

window n days min p25 p50 p75 max current percentile
Full, 2020-07-27 → 2026-07-29 1,506 1.17 1.45 2.15 3.45 6.25 3.31x 73rd
Post-divestiture, 2023-01-01 → 895 1.20 1.42 2.52 3.99 6.25 3.31x 54th

The post-divestiture window is the one I use, and the reason is a defect the full series contains: the full window spans the Creative Distribution Solutions wind-down, which cut TTM revenue from $445m to $321m mechanically. An EV/Sales series across a divestiture prints a spike that is a denominator artifact, not a re-rating. Using the full window's 2.15x median would import that artifact into the anchor. This is the "regime change" case valuation.md names, and the response is to sub-window rather than to declare UNIDENTIFIED — because a 895-day post-break window with 1.20x–6.25x dispersion is a genuinely informative sample, unlike SN's.

Build: - NTM revenue = 5/12 × FY2026 consensus $526.5m + 7/12 × FY2027 consensus $590.6m = $563.6m (AV EARNINGS_ESTIMATES, 4–5 analysts — thin coverage, stated). - Anchor multiple 2.52x = the 54th percentile, i.e. the median of the post-divestiture window, which is where the stock currently sits. - EV = 2.52 × $563.6m = $1,420.3m; less net debt $118.0m = $1,302.3m equity; ÷ 19.475m shares. - Target $66.85, −11.7% to spot.

At the p75 of 3.99x the target is $109.4 (+44.5%); at the p25 of 1.42x it is $35.0 (−53.8%). The dispersion is enormous and that is the honest picture of a $1.5bn small cap at 54.2% realised vol with a binary regulatory overhang.

Named product events inside 12 months (see TPB_Catalyst_Calendar.md): ALP 12mg strength introduction (2026, month not disclosed); the option to acquire the SBB Canadian distribution business becomes exercisable March 2027 at ~$20.0m. No FDA determination date exists.

Estimate-revision direction (AV, 30-day trailing): FY2026 EPS 0 up / 3 down; FY2027 0 up / 3 down; Q3'26 0 up / 1 down. Uniformly negative across every horizon — the only name of the three with no upward revisions anywhere.


Downside case with a named cause (Type: MEASURED — does not reject the name)

Named cause: an FDA Marketing Denial Order on either pending PMTA product line.

This is a genuinely existential, genuinely undated risk. TPB has two product lines under PMTA review, does not name them, and there is no FDA decision timeline. An MDO on the modern-oral line removes $107.7m of FY2025 growth and 105.2% of consolidated growth at a stroke, leaving a Zig-Zag business whose operating income is already down 13.7% over two years plus a legacy MST business.

Quantified: ex-modern-oral, FY2025 revenue is roughly $355m (Zig-Zag $178.5m + Stoker's ex-modern oral ~$177m) at a blended operating margin nearer FY2023's 24.7% on the legacy mix — EBIT ~$88m — but a declining, terminal-value-impaired asset trades at 8–10x, not 15x. At 9x: EV $792m, less $118m of debt, $34.6/share, −54.3%.

A second, non-regulatory path to a similar number: modern oral fails to reach operating breakeven at scale and the S&M spend proves permanent. Stoker's Q1'26 already ran at a negative incremental operating margin; if that is structural rather than a launch phase, the whole growth engine is a value-destroying reinvestment of a 10.9%-ROIC business's cash flow.