WATCHconviction 3published 2026-09-07

MGPI — MGP Ingredients, Inc. · 2026-09-07 · Verdict: WATCH · Conviction 3

Price $16.40 (screen row close 2026-09-06; no live quotes available) · Mkt cap $351M · EV ~$705M · EV/2026E adj. EBITDA 7.5x (EV/EBIT n/m, FY25 GAAP operating loss $94.6M) · FCF yield ~6% on 2026 guidance math, negative in H1 2026 actual · Net debt ~$353M (3.5x) · ADV $3.4M Sources read: 10-K 2026-02-25 (Items 1, 1A, 7), 10-Q 2026-07-29, DEF 14A 2026-04-09, 8-Ks of 2026-04-07, 04-29, 05-14, 07-29, 08-07, Form 4s (12 of 29 in trailing 12 months). No transcript or prepared remarks were filed as an exhibit, so section 5 rests on written guidance versus results.

1. What the business actually does

Three segments (10-K 2025, Item 1). Branded Spirits is an owned portfolio sold to distributors and control states, spanning premium plus (Penelope, Yellowstone, Rebel, Remus, El Mayor tequila), mid (Pearl vodka), value (Arrow, Canada House) and private label. Distilling Solutions sells bourbon, rye and grain neutral spirits to other brand owners and rents barrel warehouse space. Ingredient Solutions makes specialty wheat starches and proteins (Fibersym, Arise, Proterra) for bakers and food processors. Distilleries are in Indiana and Kentucky, with a tequila JV in Mexico and bottling in Missouri, Ohio and Northern Ireland.

2. Why it is mispriced — the edge case

GAAP has gone useless while the cash business has not, and the segment doing the damage has shrunk too far to do much more. H1 2026 shows a $155.5M operating loss and $(5.74) basic EPS, almost entirely from $179.5M of non-cash writedowns: $115.7M of goodwill, $37.0M of trade names and a $26.9M fixed asset impairment at Lux Row (Q1 release 2026-04-29; 10-Q Q2 2026). Goodwill is now zero. Any screen keying on GAAP EBIT, ROIC or EPS reads a company losing $100M a year; the same six months produced $42.6M of adjusted EBITDA and $12.0M of Q2 net income.

Second, the melting part is nearly melted. Brown goods went from $265.9M of sales in 2024 to $128.5M in 2025 to $29.2M in H1 2026 (10-K Item 7; 10-Q). The line that erased $137M of sales last year has roughly $58M annualized left to lose. Distilling Solutions is now 24% of quarterly gross profit ($11.3M of $46.5M), while Branded Spirits is 68% at a 53.0% gross margin, up 20bp, with sales ex private label up 3%, the best in two years, and Penelope up 13% (Q2 release).

Who is selling, and are they wrong? Not obviously, and that is the problem. This is no forced-seller story. The two largest informed holders sold: 10% owners Caroline Lux Kaplan and Paul S. Lux disposed of 93,000 shares for $1.67M at $17.76 to $18.18 between May and August 2026, with no open-market insider purchases in the trailing twelve months (Form 4s 2026-05-28, 2026-08-24).

3. Unit economics and growth

FY2025: sales $536.4M (-24%), gross margin 37.2%, adjusted EBITDA $116.0M (sum of the quarterly reconciliations in the Q1 and Q2 2026 releases), with segment gross profit of $115.3M at 49.5% in Branded Spirits, $68.6M at 37.8% in Distilling Solutions and $15.5M at 12.7% in Ingredient Solutions (10-K Item 7). Q2 2026: sales $124.4M (-15%), adjusted EBITDA $27.6M (-23%), adjusted basic EPS $0.72. Pricing power shows where it counts: brown goods net price/mix was positive 9% in Q2 even as volume fell 68%, and segment gross margin rose to 38.7% (10-Q). Ingredient Solutions is the sore spot at 10.1% margin against 21.7%, on waste starch disposal costs, despite 2% sales growth. One Branded Spirits customer was 16% of consolidated 2025 sales, one Ingredient customer 14% (Item 1).

4. Balance sheet and capital allocation

This is the problem. Total debt went from $252.3M at 12/31/25 to $369.6M at 6/30/26 because MGP paid the full $110.8M Penelope earnout in April with borrowings (10-Q), taking net debt leverage from 1.8x to 3.5x year over year (Q2 release). Inventory is $408.4M and still growing, up $25.6M in H1 on more barreled distillate, in a market management itself calls "structurally oversupplied" (8-K 2026-04-07). H1 operating cash flow was negative $40.7M; excluding the $48.7M earnout piece run through operating it was positive $8.0M, against $10.2M of capex paid. On August 6 the company amended both the credit agreement and the PGIM notes to add back up to $20M of receivable losses from named customers through 2027, and elected an Elevated Ratio Period lifting the net leverage covenant from 4.00x to 4.50x for Q2 2026 and three quarters after (8-K 2026-08-07). Only $2.1M is provisioned for the customer that filed Chapter 11 on July 26 (10-Q); the gap between $2.1M booked and $20M of headroom negotiated is a tell. About $196.3M of convertible notes are expected to be put in Q4 2026 and refinanced on the revolver ($338M available) or the PGIM shelf ($236.4M); the pack does not disclose the convert coupon, so that interest swap cannot be sized here. Buybacks are paused with $53.4M authorized and nothing repurchased since 2024. The $0.12 quarterly dividend continues.

Governance is closed. Five of nine directors are elected by 297 voting preferred shares, 68% held by Karen Seaberg, who owns 9.7% of the common (8-K 2026-05-14; DEF 14A); the Lux Family Group holds 31.3%. A sale or activist campaign is structurally hard. Executive PSUs use a one-year performance period vesting over three years (DEF 14A), weak alignment for a multi-year down-cycle; the CEO's 138,310 options strike at $28.54.

5. Management: said versus did

Guidance has held all year: $480-500M sales, $90-98M adjusted EBITDA, $1.50-1.80 adjusted basic EPS, reaffirmed April 7 while announcing the Kentucky idling, again April 29 and July 29. CEO Julie Francis said Q2 "adjusted EBITDA and adjusted basic EPS came in ahead of our expectations" (Q2 release), consistent with a path needing about $25M a quarter in H2. Management called Q3 2026 its "peak leverage, after which it expects leverage to decline" (8-K 2026-08-07). That claim is what the thesis rests on and it is not yet demonstrated.

6. Valuation

Assume 21.4M shares and $353M net debt throughout.

Bear (35%): the glut runs into 2028, brown goods halves again, Ingredient Solutions stays near 10% margin, adjusted EBITDA falls to $75M in 2027, the convert refinances at higher cash cost, receivable losses reach the $20M the amendment contemplates, and leverage sits above 4x into the step-down back to 4.00x. 6.5x on $75M is $488M EV, about $6 a share.

Base (45%): 2026 lands at the $94M midpoint; Distilling Solutions holds its H1 2026 run rate, Branded Spirits compounds low single digits at 51-53% gross margin, Ingredient Solutions recovers to mid-teens, adjusted EBITDA reaches $110M by 2028 with debt paid to $300M. 8x is $880M EV, about $27 in 2028, roughly $20 discounted at 12%.

Bull (20%): the cycle turns in 2028, brown goods recovers half its decline, premium plus compounds high single digits, adjusted EBITDA hits $150M by 2029 and leverage falls under 2x. 9x is about $51, roughly $38 discounted.

Probability-weighted: about $18.35 against $16.40, a 12% margin. Too thin to underwrite.

Reverse DCF: at $16.40 the roughly $705M enterprise value, against about $50M of steady-state free cash flow (guided $94M adjusted EBITDA less roughly $20M capex, interest and cash tax), implies the market expects MGP to hold today's trough cash generation forever and grow it about 2% a year at a 9% discount rate, neither recovering with the whiskey cycle nor deteriorating further.

7. Catalysts and timeline

Q3 2026 results in late October: whether leverage actually peaked, and the terms on which the converts refinance. Q4 2026: settlement of the convertible put. February 2027: first FY2027 guidance, the first year that could show EBITDA growth. A restart of the buyback, or resumption of Kentucky distilling, which management said could come as early as May 2027 (8-K 2026-04-07), would each mark the turn.

8. Risks and pre-registered kill criteria

Structural risks: a whiskey glut with no dated end; a bank group amended twice in six months; a $408M inventory book whose value depends on demand two to six years out (Item 1A); family and preferred-share control that removes the takeover floor. Kill criteria, all observable in filings: 1. Reported net debt leverage above 4.0x in any quarter after Q3 2026, or a further amendment or extension of the Elevated Ratio Period. 2. FY2026 adjusted EBITDA guidance cut below $90M, or FY2027 guided below $85M. 3. Branded Spirits gross margin below 50% for two consecutive quarters, or premium plus sales down year over year in any quarter. 4. Cumulative provision for credit loss on the bankrupt customer above $10M.

9. Verdict and one-paragraph summary

WATCH, conviction 3. MGP's headline numbers are wrecked by $179.5M of non-cash impairments that took goodwill to zero, while the cash business still guides to $90-98M of adjusted EBITDA and Branded Spirits, now 68% of gross profit at a 53% margin, grew 3% ex private label with Penelope up 13%. The genuinely interesting fact is that brown goods, which erased $137M of sales in 2025, is down to a $58M annualized run rate and cannot do much more damage. What stops this being a buy is the balance sheet at exactly the wrong moment: the $110.8M Penelope earnout paid in cash took leverage from 1.8x to 3.5x, $196M of converts get put in Q4 2026 at a refinancing cost the filings do not disclose, the covenant was loosened to 4.50x in August only through Q1 2027, and management booked $2.1M for a bankrupt customer while negotiating $20M of covenant headroom for exactly that. Add no insider buying against $1.67M of family selling above today's price, and a probability-weighted value near $18 against $16.40, and the honest answer is to wait for the Q3 print that proves leverage peaked and prices the convert refinancing.

Research for discussion, not investment advice. Positions and sizing are the reader's decision.

Source markdown: 2026-09-07_MGPI.md · how these notes are built · every verdict tracked since publication.

Research and education only. Nothing here is investment advice or a recommendation to buy or sell any security. No price targets are recommendations; positions and sizing are the reader's decision. Past performance does not predict future results.