Houlihan Lokey, Inc. (NYSE: HLI) — The Best House in Independent Advisory, Marked Down for Being Cyclical

Independent equity research · Report date: 2026-07-10 · Sector: Independent Investment-Banking Advisory · Fiscal year ends March 31 (“FY2026” = year ended 2026-03-31, reported May 2026). Figures reconcile to the FY2026 Form 10-K (filed 2026-05-22), the Q3/Q4-FY2026 earnings calls, the FY2025 proxy, and the 5-year SEC corpus unless noted.

⚡ Claude’s Take

This is the author’s own subjective opinion and general information, not investment advice. The analysis that follows carries no recommendation and no price target; only this opening block takes a view.

Verdict: HOLD — constructive / accumulate on weakness ~$120–135 (already at the low end of the range). Quality-compounder-on-sale, not a fat pitch. Not a short. Conviction: medium. Tag: “A countercyclical compounder, de-rated for being cyclical — the market took away the premium, not the franchise.”

Houlihan Lokey is, to my eye, the single highest-quality business in the independent-advisory group and one of the better franchises in all of financials: #1 in global M&A by deal count, #1 in restructuring, the largest US fairness-opinion practice, a genuine countercyclical hedge (restructuring earns more when M&A seizes), ~28% ROE, a net-cash fortress with no funded debt, and a disciplined, culture-first, tuck-in acquisition machine that has grown revenue in 9 of its 10 public years. It just printed a record FY2026 (revenue $2.62B, +9.5%; adjusted EPS $7.56, +20%). And the stock is down ~34% from its September-2025 all-time high (~$207) to ~$136 — because a Q4 revenue miss (−4.6% YoY, the first of the recovery) punctured the “straight-line M&A re-acceleration” narrative and unwound the ~30x growth-extrapolation multiple back toward ~22x GAAP / ~18x adjusted / ~13.4x EV/EBITDA.

That de-rate is the opportunity and the caution in the same sentence. The opportunity: you rarely get to buy this franchise cheap, and the market has re-rated the price without impairing the business — the moat (independence, restructuring depth, mid-market scale, the data-scale flywheel in valuation) is fully intact, the FR hedge is firming even as M&A recovers, and the balance sheet is bulletproof. The caution: ~22x GAAP on earnings that are recovering toward a cyclical high, not sitting at a trough is not statistically cheap, and advisory is the most cyclical fee business there is — if FY2026 proves near a cyclical peak and the recovery stalls, forward earnings and the multiple can both compress. So I can’t call this a fat pitch at $136; I can call it a high-quality compounder that has handed you the best entry it’s offered in a couple of years, with a favorable skew (I see meaningfully more upside toward the prior highs on a real M&A upcycle than downside in a stall). Bullish trigger: two consecutive quarters of CF revenue re-acceleration confirming the M&A cycle resumed. Bearish trigger: a second YoY revenue decline / a stalling deal calendar that reveals FY2026 as a cyclical top. The insiders offer no help either way — they control ~75% of the votes through the dual-class trust and are structural net-sellers; there’s no conviction open-market buying to lean on.

📈 Stock Price Action — Five-Year Event Map

Factual price history — no recommendation, no price target. Prices are nominal closes from the AZI 5-year CSV (IPO Aug-2015 → Jul-2026); the attributed cause of each move is Interpretation, the move itself is Fact.

Arc. Over five years HLI ran from a COVID-era low of ~$46 (Mar 2020) to an all-time high of ~$207–209 (Sep 23, 2025) — a ~4.5x move — then de-rated to ~$136 (Jul 9, 2026), sitting ~34% below the ATH and down ~27% over the trailing twelve months, near the low end of a 52-week range of ~$133 → ~$207. The round trip is the advisory cycle in miniature: COVID crash, 2021 M&A boom, 2022–23 downturn, a powerful 2024–25 recovery to record highs, and a 2026 de-rate.

(Current ~$136 on Jul 9, 2026, a small bounce off the Jun-29 low ~$133.) The chart is a full advisory cycle capped by a 2025 blow-off to ~30x and a 2026 de-rate back to a cyclically-sober multiple — a re-pricing of the premium, not a break in the franchise.

1. Executive Summary

Houlihan Lokey (founded 1972, public since 2015) is a leading global independent, advice-only investment bank — no lending, no trading, no research, and therefore none of the conflicts a bulge-bracket carries. It advises corporations, private-equity sponsors, and creditors through three segments: Corporate Finance (CF — M&A + a fast-growing Capital Solutions/private-capital business; ~67% of revenue), Financial Restructuring (FR — #1 globally; ~20%), and Financial & Valuation Advisory (FVA — the largest US fairness-opinion/valuation practice; ~13%). The defining feature is the CF-vs-FR seesaw: CF booms in up-cycles, FR earns more in downturns, and FVA grinds recurring fee-event volume through both — a structure that produced revenue growth in 9 of 10 public years and far lower earnings volatility than any pure-play advisory peer.

FY2026 was a record: revenue $2.62B (+9.5%), GAAP diluted EPS $6.22, adjusted EPS $7.56 (+20%), ROE ~28%, ROIC ~17%, on a net-cash balance sheet with no funded debt (~$1.36B liquidity). The economics are elite-but-capped: a ~64% GAAP compensation ratio (~61.5% adjusted) is stable through the cycle — the source of margin stability and a structural ceiling, since the bankers capture most of the value. Returns look extraordinary on tangible equity (~57%) because the “assets” (bankers) are expensed, not capitalized; reported ROE (28%) is lower because the acquisition roll-up loads ~$1.4B of goodwill.

The moat is real but moderate and multi-part, not a wide fortress: genuine (reputational-intangible + niche-scale, with an emerging data-scale flywheel) in FVA; moderate (reputational, repeat-mandate) in restructuring; weak-to-moderate (brand + sub-sector density + sponsor relationships) in mid-market M&A. The durable edge is the combination — a diversified, cycle-smoothing platform plus a culture-first talent-retention and serial-acquisition machine that compounds bankers better than pure-plays, riding a genuine secular tailwind (independent-boutique US M&A fee share rose from <15% in 2018 to >27% by 2024).

The setup is a quality compounder caught mid-cycle: the business is stronger than two years ago, but the stock de-rated ~34% from its ATH after a Q4 revenue decline (the first of the recovery) unwound a ~30x multiple. At ~$136 it trades ~22x GAAP / ~18x adjusted / ~13.4x EV/EBITDA / ~2% yield — reasonable for the quality, but not statistically cheap on earnings that are recovering toward a cyclical high, not sitting at a trough. The core debate is whether FY2026 is near a cyclical peak (making ~22x full) or the recovery merely paused (making an out-of-favor 28%-ROE franchise a quality name the market stopped loving). No recommendation and no price target appear below; valuation is discussed only as embedded expectations.

2. Business Overview

Houlihan Lokey sells advice, not balance sheet — no lending, no sales & trading, no research — so it carries none of the conflicts a bulge-bracket bank does when it lends to, trades against, and researches the same companies it advises. At 2026-03-31 it employed >1,900 financial professionals (354 Managing Directors) across 30+ offices, ~2,800 total staff, serving >2,000 clients annually — corporations, financial sponsors and their portfolio companies, and creditor/bondholder groups. It markets across product groups, ~200 industry sub-sector groups, and a dedicated Financial Sponsors group, and reports three segments:

Corporate Finance (~⅔ of revenue) — the growth engine. M&A advisory plus a fast-growing Capital Solutions business (sponsor financing, private placements, secondaries/GP-stakes) now >20% of CF revenue. CF is the most M&A-cyclical segment — it exploded in FY2022, fell hard in FY2023, and re-accelerated +14% in FY2026. HLI is deliberately a mid-market franchise: its edge is winning the ~$25M–$500M deal-count game, not mega-deal value league tables.

Financial Restructuring (~20%) — the countercyclical hedge. The #1 global restructuring/liability-management franchise (59 MDs, one of the deepest benches), advising debtors and creditors in and out of court. It works when CF doesn’t — FR revenue fell only −3% to $529M in FY2026 even as M&A recovered, still “one of the strongest years on record,” illustrating the smoothing dynamic.

Financial & Valuation Advisory (~13%) — the recurring ballast. The largest, most respected US fairness-opinion/valuation practice: 2,519 fee events (vs a few hundred episodic transactions elsewhere) — a high-volume, low-average-fee, quasi-recurring business, and the one most exposed to (and, per management, most defended by) AI/technology. Portfolio-valuation demand is a secular grower tied to private-credit expansion and the LP/regulatory push for more-frequent marks.

Fee model. Revenue is earned under individually negotiated engagement letters dominated by success-based Completion Fees (paid on closing) plus Progress Fees and modest retainers. Two consequences: (1) revenue is episodic, not recurring — a quarter depends on the timing of closings (the Q4-FY26 “miss” was timeline slippage, not lost mandates); and (2) the flex-comp model absorbs the volatility — pay flexes with revenue, so downturns compress the bonus pool rather than break the P&L.

Verdict. A capital-light, advice-only, globally diversified investment bank whose revenue is ~⅔ M&A-cyclical (CF), ~20% counter-cyclical (FR), and ~13% volume-recurring (FVA). Revenue is episodic and closing-timing-dependent, but the three-legged structure plus flex-comp produces earnings stability no pure-play advisory peer matches.

3. Industry Dynamics

The structural growth story: independents taking the bulge bracket’s wallet. A firm that only advises — no lending relationship, no trading desk, no research to compromise — can give conflict-free advice, and boards/creditors increasingly demand exactly that on the decisions that matter most. The data confirm the migration: the five public elite boutiques (Lazard, Evercore, PJT, Moelis, Houlihan Lokey) plus Centerview captured under 15% of US M&A advisory fees in 2018, rising to over 27% by 2024. That is a supply-side migration of the fee pool, not merely a cyclical bounce — the tailwind under HLI’s entire franchise.

Where we are in the cycle (mid-2026). M&A recovery is early-innings (“roughly third inning”); global PE dry powder is ~$1.3–2.6 trillion, and ~34% of PE portfolio companies have been held >5 years (up from 28%), with the current inventory taking a near-record ~9 years to clear — a large, pent-up, sponsor-driven backlog that is the sweet spot of HLI’s mid-market, sponsor-heavy CF business. The recovery is choppy: geopolitical shocks and a software-sector dislocation extended deal timelines and dented Q4-FY26 CF/FVA growth. Restructuring is elevated and, unusually, improving — HLI raised its FR outlook for FY2027 citing widening credit spreads, private-credit dislocation, and software/energy stress: a rare setup where the counter-cyclical hedge firms even as the pro-cyclical engine recovers. FVA/portfolio valuation is a secular grower on private-credit marking demand.

Competitive intensity. The space is crowded and intensifying — elite boutiques on the large-cap fringe (Evercore, Lazard, Moelis, PJT, Centerview, Perella Weinberg, Rothschild) and mid-market specialists head-on (Lincoln International, William Blair, Harris Williams, Baird, Piper Sandler, Jefferies). FY2025 advisory scale of the public set: Evercore ~$3.9B, HLI ~$2.4B, Centerview ~$2.1B, Lazard FA ~$1.8B, PJT ~$1.7B, Moelis ~$1.5B. In restructuring, the tier-1 triad is PJT, Houlihan Lokey, and Evercore. The binding constraint is talent, not capital — growth is bought by hiring/retaining MDs, and ~60–62% comp ratios are the “tax” bankers extract; the economics accrue heavily to labor. This is both the industry’s attraction (capital-light, ~28% ROE) and its structural vulnerability (the productive assets can, and do, walk).

The AI question. Management frames AI as a double-edged sword that favors scale: FVA valuation work could face pricing pressure, but (a) HLI has out-grown that pressure “for a decade” by expanding the TAM faster than price falls; (b) only firms of HLI’s size can afford the multi-year tech spend, so AI should consolidate the valuation industry toward large players and kill sub-scale boutiques; and © the proprietary dataset from doing an enormous volume of marks makes its models better — an emerging data-scale flywheel. Directionally plausible, consistent with FVA’s volume economics; unproven.

Verdict — structurally attractive, with a talent-cost caveat. Independent advisory enjoys a genuine, multi-year secular share gain from the bulge bracket, capital-light economics, high returns, and — for a diversified player — a natural cyclical hedge. Offsets: intense competition, high M&A-cycle sensitivity, and a labor-cost structure capturing ~60%+ of revenue. A good industry, better for the diversified, scaled incumbents than for sub-scale pure-plays.

4. Competitive Position & Moat

HLI’s positioning rests on three "#1"s and one structural feature: #1 global M&A by deal count (LSEG 2025; ~552 transactions in FY2025 per PitchBook — outside the top-five on value, by design); #1 global restructuring/liability-management; the largest US fairness-opinion/valuation practice (2,519 FY2026 fee events); and the diversified, cycle-smoothing model itself.

Naming the moat in Greenwald’s taxonomy — and pressure-testing it. Advisory is a people business, so the classic objection is that the “assets” walk out the door every night. Applying the framework honestly:

• FVA is closest to a real Greenwald moat — reputational intangible + niche scale economies + a nascent data advantage. Fairness/solvency opinions and portfolio valuation reward the largest, most credible, most independent provider: boards hire the name that withstands litigation and regulatory scrutiny, and that name is self-reinforcing (more opinions → more precedent/data/credibility → more opinions). The high fixed cost of the required AI/technology build spreads over 2,500+ fee events — a scale-over-volume cost advantage sub-scale boutiques cannot match. Independence is a differentiator a bulge-bracket structurally cannot replicate. Grade: narrow but genuine.
• Restructuring is a reputational/relationship moat with a self-reinforcing flywheel, but thinner. A small pool of distress situations flows repeatedly to the same top-tier names; being #1 begets the next mandate. Real reputational intangible with mild network characteristics — but still people-anchored, against an elite set (PJT, EVR). Grade: moderate.
• Mid-market M&A is the thinnest — brand + sub-sector density + sponsor relationships, closest to “a collection of bankers.” #1 by deal count reflects a dense platform (~200 sub-sector groups, deep sponsor coverage, global reach), and the platform’s data/deal-flow/brand is hard for any single departing MD to replicate — but head-to-head competition is fierce and client switching costs are near zero. Grade: weak-to-moderate.

The real, defensible edge: the diversified model + the retention/acquisition platform. The durable advantage is not any single league-table crown but the combination: (1) a product-breadth/diversification advantage that smooths the cycle better than any pure-play (revenue growth in 9 of 10 public years, far lower earnings volatility than EVR/MC/PJT); and (2) an institutional platform (brand, deal flow, sub-sector density, restructuring bench, global footprint, proprietary valuation data, a disciplined comp culture) that retains and compounds talent, and serially absorbs boutiques, better than peers. Bankers can walk — but away from a data/deal-flow/brand engine a single MD cannot carry; steady MD growth (339→354 YoY, +33 hired/acquired in FY26) suggests the platform, not any individual, is the asset. Greenwald share test: HLI passes in the strongest sense that matters — durable and rising share (#1 deal count year after year while the cohort’s US fee share climbed <15%→>27%). Returns confirm quality (ROE ~28%, ROIC ~17%), though ROIC is depressed by acquisition goodwill and the comp ratio caps the equity holder’s take.

Verdict — a real but moderate, multi-part moat: durable, not wide. The genuine, financially-visible advantages are a narrow reputational-plus-scale moat in FVA (and an emerging data-scale flywheel), a moderate reputational moat in restructuring, and a structural diversification/retention-platform advantage no pure-play matches. Mid-market M&A alone is weak and contestable. Not an unassailable fortress — a people business where the “franchise” is a brand, platform, and culture that retain and compound talent better than competitors. Bankers can walk; the platform is harder to replicate than any one of them. Durable advantage: yes, moderate — stronger and far more cycle-resilient than EVR/MC/PJT, but not a monopoly.

5. Growth History and Forward Opportunities

The record — compounding through a full cycle, FR cushioning the trough:

An ~11% revenue CAGR FY21→FY26 across a peak-to-trough-to-record cycle — the smoothing model, not a straight line.

Decomposing growth — organic vs acquired vs cycle (the critical distinction):

1. Organic share gains (highest quality): rising boutique fee share, #1 deal count sustained, ~200 sub-sector groups built out globally, and Capital Solutions growing from a standing start to >20% of CF — recurring, self-funded.
2. Acquisition-fueled (medium quality — the roll-up): HLI is a serial acquirer of boutiques. Landmark GCA Corporation (2021, ~$591M, ~500 people) established the Japan/Europe tech platform; steady tuck-ins since (7 Mile, Triago, Waller Helms, Prytania, Audere/France, Mellum), 33 MDs hired-or-acquired in FY2026 alone, and the June-2026 Intrepid Financial Partners energy deal (34 professionals). Inorganic MD adds inflate the growth rate and load goodwill that depresses ROIC (17% vs ~28% ROE) — franchise-accretive but return-dilutive; integration/retention/earn-out risk is real.
3. Cycle-driven (lowest quality — mean-reverting): CF’s +14% FY26 and FY25 +25% owe heavily to the M&A recovery; the FY22-peak-then-FY23 −20% is the reminder.

A volume story, not a productivity story. Revenue-per-MD is roughly flat-to-modestly-up; growth is driven by more MDs (promotes + hires + acquisitions), not dramatically rising output per banker. FY26 MD count 339→354 (+4.4%). Fine — it is how advisory scales — but investors should not extrapolate boom-year revenue-per-MD.

Forward drivers. The M&A recovery (largest lever; pent-up sponsor exits, aging portfolios, dry-powder pressure — choppy but up); Capital Solutions (highest-conviction internal grower); restructuring optionality (a free countercyclical call option, guided up for FY27); international build-out (Europe/France, Asia post-GCA growing faster than the US); new verticals via M&A (Intrepid energy the template); and FVA/portfolio-valuation + data monetization.

Verdict — high-quality growth in aggregate, but decompose it. The durable core — recurring share gains, capital-light economics, Capital Solutions, the countercyclical FR hedge — is genuinely high quality. But a meaningful slice of reported growth is acquisition-fueled (goodwill drag on ROIC) and cycle-timed (mean-reverting), and the story is fundamentally more bankers, not rising productivity per banker. A well-run, disciplined compounder whose ~11% blended CAGR is real and repeatable on average through a cycle, provided the roll-up keeps clearing its integration bar and the M&A cycle cooperates.

6. Financial Quality

The revenue engine. FY2026 revenue $2,617.5M (+9.5%; +37% over FY2024) is pure fee income — no COGS, no inventory, no capital equipment. The year’s story is the counter-cyclical hedge working in reverse: as M&A recovered, CF revenue jumped +14% and CF segment profit +23% (operating leverage), while FR revenue fell −3% and FR profit −14% (negative leverage as its comp ratio rose on softer revenue). FR is the shock absorber (record $544M in FY25, giving back only 3%); FVA is the small, stable, high-recurrence annuity. Caution: FY26 is a CF-led (M&A-cycle-led) year — 67% of revenue now rides the deal cycle, and FR’s cushion is thinner than in FY24–25. Earnings are not at a cyclical trough; they are recovering toward a cyclical mid-to-high.

The compensation ratio — the number that governs this business. For an advisory firm the key ratio is comp & benefits / revenue. HLI’s disclosed GAAP comp ratio was 64% in both FY2026 and FY2025 (~61.5% on an adjusted basis excluding acquisition-related comp):

~64 cents of every revenue dollar goes to the bankers, and that ratio is remarkably stable through up- and down-cycles because management flexes the bonus pool with revenue. The residual ~15–17% non-comp is disciplined and largely fixed, so incremental margins on a good year are high (27% FY26, 33% FY25). The crucial QoE point: HLI’s $201.3M of stock-based comp (7.7% of revenue) is fully expensed inside the 64% comp ratio and is NOT added back to adjusted EPS — unlike a software company. Its non-GAAP adjustments are narrow (acquisition amortization, $73.6M acquisition-related retention comp, a $17.9M contingent-consideration mark), each separately disclosed. The comp ratio is honest, and GAAP EPS ($6.22) and adjusted EPS (~low-$7 / $7.56) are close enough that the choice does not move the thesis — a rare virtue.

Returns — very high, understand why. ROE 28% (FY26), ROIC 16.9%, EBITDA margin 25.3%. A 28% ROE with no funded debt is genuinely elite — a capital-light model where the “assets” (bankers) are expensed, not capitalized. Two nuances: (1) equity of $2,342M carries $1,396M goodwill + $204M intangibles, so tangible equity is only ~$742M and net income of $426M is a ~57% tangible ROE — the organic advisory business is a cash spigot; the gap between 28% and 57% is capital tied up in acquisitions (why ROIC 17% sits below ROE — the roll-up is return-dilutive but franchise-additive). (2) ROE is flattered by the cycle and by buybacks. The comp ratio is a structural ceiling on margin expansion — this will never be a 40%-operating-margin business; the talent captures the economics.

Balance sheet — a net-cash fortress. Cash & equivalents $1,189.5M + investments $170.3M = $1,359.7M liquidity; no funded debt (the “$492M long-term borrowing” in aggregator feeds is operating-lease liabilities, confirmed on the balance-sheet face); the only facility is a $150M undrawn revolver. Treat HLI as net cash ~$1.36B (EV ~$8.3–8.8B vs ~$9.6B market cap). Caveat: a material portion is spoken for — accrued salaries & bonuses of $1,076.6M sit at year-end, paid the following May/November; true structural excess is smaller than the headline, but the firm is still unambiguously over-capitalized and bulletproof through any cycle.

Free cash flow — high quality, genuinely lumpy. OCF swung from $136M (FY23) to $849M (FY25); FCF/share from $1.35 to $12.31 — a 9x range while net income was far steadier. This is a working-capital timing artifact, not an earnings-quality problem: HLI accrues bonuses monthly and pays the bulk in fiscal Q1/Q3. Normalize by averaging — five-year average FCF ≈ $520M/yr, cumulative FCF closely tracks cumulative net income. Any single-year FCF yield must be read against the bonus-cycle position, not annualized naively.

Verdict — financial quality is HIGH. Honest comp accounting (SBC fully expensed), disciplined non-comp, elite tangible returns, no debt, ~$1.36B liquidity, and cash flow that converts fully across the cycle. Two things to keep front-of-mind: the 64% comp ratio caps margins, and FCF must be read on a multi-year average. Neither is a knock on quality; both are features of the model.

7. Capital Allocation

Framework. HLI throws off far more cash than it needs (capital-light, no debt service). Revealed priorities: (1) fund the acquisition roll-up, (2) pay a growing dividend, (3) repurchase stock to offset dilution, (4) let the rest accrete on the balance sheet. It has deliberately hoarded net cash rather than lever up or pay it all out.

The M&A roll-up — the growth strategy and main capital use. HLI is a serial tuck-in acquirer, entering sectors/geographies faster than it could hire: GCA Corporation (2021, ~$361M cash, transformational for international CF), then 7 Mile, Triago, Waller Helms (Dec 2024), Prytania, Audere/France (Feb 2026, controlling interest + $110.6M redeemable NCI), and Intrepid (June 2026, energy, post-FY26). The discipline is in the structure: acquired bankers are tied in with contingent consideration (earnouts) and deferred/retention comp (the $73.6M line) — a seller-alignment mechanism. Cash deployed post-GCA is modest (FY26 only $2.5M net; FY25 $69.2M), so the roll-up is funded from cash flow, not debt, and is not bet-the-company sized; there have been no goodwill impairments across the five-year corpus. Bear caveat: ROIC (17%) sits well below tangible ROE precisely because acquisitions add goodwill — franchise-accretive, return-dilutive — and a serial acquirer always carries integration/overpayment risk (sharpened by Intrepid’s undisclosed terms).

Dividend — steadily growing, moderate payout. FY2026 dividends paid $174.0M (~$2.62/share); the quarterly dividend was raised to $0.70 (→ ~$2.80 FY27 run-rate). Five-year growth ~$1.40→$2.62 = ~13%/yr, raised every year including the FY23 trough. Payout ~42% of GAAP EPS (lower on adjusted) — safe, well-covered, room to run.

Buybacks — mostly dilution-offset, with a FY26 step-up. Two streams: open-market buyback ($175.4M FY26, 3x FY25, at ~$179) plus tax-withholding repurchase ($142.7M). Net effect: period-end share count essentially flat at ~69M over five years — buybacks + withholding roughly offset the ~1.5–2M shares/year issued via equity comp. FY26 is the first year the open-market buyback meaningfully shrank the float beyond neutralizing SBC — a signal management is beginning to treat the over-capitalized balance sheet as returnable ($230M authorization remaining).

The over-capitalization question. With ~$1.36B liquidity (net of the bonus accrual, still several hundred million true excess), no debt, and ~$500M/yr normalized FCF against ~$175M dividends + ~$175M buybacks, HLI retains a large, growing cash pile. Charitable read: dry powder for counter-cyclical boutique acquisitions and a stability signal to clients/regulators. Critical read: a 28%-ROE business letting cash earn ~4% T-bill yields is lazy-balance-sheet capital allocation; the FY26 buyback step-up is the first evidence management may be addressing it.

Governance — dual-class control. Class A (1 vote) trades publicly; Class B (10 votes) is held by employees/MDs through the HL Voting Trust, controlling ~74.6% of voting power — HLI is a “controlled company” and (post-2025 AGM) no longer has a majority of independent directors; a Dec-2025 amended Voting Trust Agreement refreshed the structure. Public Class A holders have almost no say over capital allocation — the insider partnership controls it. For a founder-partnership advisory firm this cuts both ways: it enables the long-term, retention-first culture that is the franchise, but it means the over-capitalized balance sheet and buyback pace are entirely at insider discretion. CEO Adelson’s FY25 comp was $11.0M (trivial $500K salary + $8.25M cash incentive + $2.25M Class B stock); say-on-pay ~97%; clawback in place. The equity-heavy, forfeitable Class B comp is the economic engine of the moat (it locks in senior bankers) — governance and business quality are inseparable here.

Verdict — good, with one reservation. Positives: disciplined, cash-funded tuck-in roll-up with strong integration and zero impairments; a safe ~13%/yr-growing dividend at a conservative payout; no debt; a FY26 buyback step-up. Reservation: the balance sheet is over-capitalized and under-worked — years of dilution-offset-only buybacks and a swelling cash pile earning T-bill yields is sub-optimal for a 28%-ROE business. Management allocates capital prudently but not aggressively; the FY26 open-market buyback is the first sign of a shift.

8. Changes and Headwinds — Last Two Years

The two years to FY2026 tell a clean story: a record operating recovery colliding with a sharp multiple de-rate. The business got materially bigger and more profitable; the stock still fell ~27% over the trailing twelve months.

The record-revenue recovery. HLI came out of the 2022–23 downturn faster than almost any peer: revenue $1.91B (FY24) → $2.39B (FY25, +25%) → $2.62B (FY26, record, +9.5%), surpassing the FY2022 boom peak by ~15%; adjusted EPS $7.56 (+20%). ROE ~28%, net cash. The margin tell: GAAP operating margin 23.6% and profit margin 16.3% remain below the FY2022 peak (27.0% / 19.3%) even at higher revenue — the comp ratio held ~stable, so the gap is below the comp line (non-comp/amortization on a larger, more-acquisitive firm). The record top line is not translating to record GAAP margins — a real, if modest, headwind.

The de-rate and the Q4-FY26 miss. HLI peaked at ~$207–209 (Sep 23, 2025), then de-rated to ~$136 — down ~34% from the ATH — as the P/E compressed from ~30x toward ~22x. The accelerant was the Q4-FY26 print (May 6, 2026): adjusted EPS $1.63 vs ~$1.79 (~9% miss); revenue $635.6M vs ~$687M (~7.5% short, −4.6% YoY — the first YoY quarterly revenue decline of the recovery), management citing geopolitical/policy uncertainty and software-sector pressure delaying CF closings. The market had extrapolated the FY25–26 recovery into a full re-acceleration and priced HLI near a record multiple; the Q4 revenue decline punctured the “straight-up” narrative and forced a re-rate toward a cyclically-sober multiple.

The Intrepid acquisition (June 2026). HLI announced (6/30) the acquisition of Intrepid Financial Partners, a premier independent energy bank (34 professionals; O&G team to 70+; founder “Skip” McGee, ex-Barclays/Lehman, becomes Global Chairman of Oil & Gas). Terms undisclosed; closing before 9/30/2026. Textbook HLI capital allocation — a capability/coverage tuck-in (acqui-hire of a senior team) in a sector where it was underweight; the risk is generic to the model (integration, retention, and — terms undisclosed — unknowable price discipline).

Governance/leadership. CEO Scott Adelson (succeeded Beiser June-2024, a planned internal handoff); founders Gold/Beiser remain co-chairmen; CFO J. Lindsey Alley. Board refresh (Mund, Oct-2025); the Dec-2025 Voting Trust refresh perpetuates insider control. An orderly, low-drama filing history — no litigation bombshells, no restatements, KPMG throughout with unqualified opinions.

Verdict — the operating changes strengthen the franchise; the price change created the setup. The business is bigger, more diversified (Intrepid adds energy depth), record-revenue, high-ROE and net-cash — genuinely stronger than two years ago. What weakened is the valuation cushion and near-term momentum: the Q4 revenue decline confirms advisory revenue is still cyclical and lumpy, GAAP margins sit below the last peak, and the market re-rated from a growth-extrapolation multiple to a more cyclically-honest one. A quality-compounder-caught-mid-cycle situation, not a broken thesis — the changes reset the price, not the moat.

9. Risk Analysis

HLI is an asset-light, net-cash, people-based advisory franchise; its risks are overwhelmingly cyclical and human-capital, not balance-sheet or solvency.

Verdict — a high-quality, low-solvency-risk franchise whose principal risk is the one it cannot escape: the M&A cycle, amplified today by an elevated earnings base and a still-above-average multiple. The people-and-comp risks are real but chronic and historically well-managed. No risk rises to catastrophic; the honest framing is cyclical downside, not permanent-loss downside.

10. Valuation Discussion (Embedded Expectations)

No price target and no recommendation. This frames what the current ~$136 price embeds and the scenario band around it.

The anchors. At $136.44 (2026-07-09): P/E ~21.9x GAAP ($6.22) / ~18.0x adjusted ($7.56); EV/EBITDA ~13.4x (EV ~$8.84B / EBITDA $661M); P/B ~4.0x (but book is largely goodwill — a poor lens for a talent business); dividend yield ~1.9% (~2.0% on the raised run-rate). AZI’s own-history percentiles read composite ~65th, P/E ~67th — i.e., cheaper than its 2025 peak (~30x) but still above its own median, because earnings and the multiple both fell.

What the multiple should be. Advisory earnings are cyclical, so a single-year P/E overstates or understates depending on cycle position. HLI has historically commanded a premium to pure-play advisory peers (EVR, MC, PJT, LAZ) precisely for its countercyclical stability, net-cash balance sheet, and consistency — often mid-to-high-teens EV/EBITDA and low-to-mid-20s P/E in normal conditions, compressing in downturns. At ~13.4x EV/EBITDA and ~18x adjusted EPS, the stock is toward the lower end of its normal band but not at a crisis trough. The decisive question is cycle position: the fin workstream reads FY2026 as “recovering toward a cyclical mid-to-high,” not a trough — so ~22x GAAP is reasonable if the recovery has legs and full if FY2026 is near a cyclical peak.

What the price embeds (the market is underwriting): (i) the M&A recovery resumes after the 2025–26 pause and CF re-accelerates; (ii) FR stays elevated (the hedge holds); (iii) the comp ratio and margins hold; and (iv) the roll-up keeps clearing its integration bar. If those hold, an out-of-favor ~18x adjusted on a 28%-ROE, net-cash, share-gaining franchise is undemanding. If the deal calendar stalls and FY2026 proves a cyclical top, forward earnings and the multiple can both compress.

Scenario band (blends forward adjusted P/E and EV/EBITDA):

At $136 the stock sits at the low end of Base — roughly ~12–25% downside to Bear versus ~35–55% upside to Bull toward the prior highs. For a franchise of this quality that skew is favorable, and better than most de-rated cyclicals offer — the caveat being that “cyclical peak vs paused recovery” is genuinely uncertain and the bear leg is a real M&A-cycle risk, not a tail. Comp context: HLI moves with the independent-advisory cohort (EVR, LAZ, SF, JEF, PIPR, PJT, MC are its nearest factor neighbors), so its de-rate is best judged relative to that group — where HLI’s diversification and net-cash balance sheet justify a premium the market has partly withdrawn.

11. Variant Perception

Consensus. After the Q4 miss, the sell-side turned cautious/mixed — price targets trimmed (UBS Neutral ~$161, Morgan Stanley OW cut to ~$187, KBW Outperform to ~$172) — pricing a slower, choppier M&A recovery and removing the growth-extrapolation premium.

The bull case. The highest-quality, most-diversified independent advisor, de-rated ~34% on a timing-driven Q4 miss, not a franchise break; a genuine secular tailwind (boutique fee share <15%→>27%); a countercyclical FR hedge guided up for FY27; a net-cash fortress; a disciplined roll-up with zero impairments; ~28% ROE (57% tangible); and a multiple back to a reasonable level with room to re-rate toward the prior highs on a real M&A upcycle.

The bear case. ~22x GAAP on earnings recovering toward a cyclical high, not a trough — not cheap; advisory is the most cyclical fee business there is, and the Q4 revenue decline shows the CF engine can stall; GAAP margins sit below the last peak (comp ratio caps them permanently); “growth” is part acquisition (goodwill-diluted ROIC) and part cycle (mean-reverting); the balance sheet is over-capitalized and under-worked; and the dual-class trust means outside holders have no lever to force better capital allocation.

The 3–5 assumptions that matter most: (1) the M&A recovery resumes (CF re-accelerates) — the single most important variable; (2) FR stays elevated through the recovery (the hedge holds — currently guided up); (3) the comp ratio holds ~61–64% despite the talent war; (4) the roll-up keeps integrating without overpayment (Intrepid and future deals); (5) FY2026 is not a cyclical peak. Falsifiers: two consecutive quarters of CF revenue re-acceleration (bull-confirming); a second YoY revenue decline / stalling deal calendar (bear-confirming); a comp-ratio breakout above ~65% (bear); a goodwill impairment (bear).

Factor-positioning read (evidence, not a price call). HLI is a market-and-dividend-sensitive financial/broker-dealer (loadings: Market ~0.90, DividendYield ~0.71, Broker-Dealers ~0.43, Financials ~0.36; R² ~0.52 — much less idiosyncratic than GL) — with Momentum zeroed and no meaningful Value or Quality load. The 10-year track record is a genuine compounder (+22.4%/yr, Sharpe 0.76), but every short window is deeply negative (y1 −27%, m6 −43% annualized) and the stock is ~34% off its ATH near its 52-week low. The synthesis: an out-of-favor quality compounder in a real, cyclically-driven de-rate — a controlled falling knife, not a bubble and not yet a screaming statistical bargain. Where consensus is most exposed: it has extrapolated the Q4 stumble into cyclical-peak fear; if the recovery merely paused (management’s framing) and re-accelerates, an out-of-favor ~18x adjusted on this franchise is a quality name the market has stopped loving. Regime caveat: the DividendYield/Financials tilt means HLI will trade with the financials complex and deal-cycle sentiment regardless of execution — a factor headwind if financials fall out of favor, a tailwind if the M&A cycle re-rates.

12. Fact vs. Interpretation

13. Open Questions

1. Is FY2026 near a cyclical peak or a paused recovery? The single most important valuation unknown — it determines whether ~22x GAAP is full or cheap.
2. Does CF re-accelerate in FY2027? The Q4 revenue decline must prove transitory (timing) rather than a cycle roll-over.
3. What did HLI pay for Intrepid? Terms undisclosed — price discipline on the roll-up is unverifiable for this deal.
4. Will management work the over-capitalized balance sheet harder? The FY26 buyback step-up is one data point; the dual-class trust removes external pressure.
5. Can the comp ratio hold ~61–64% through an intensifying talent war, or does banker pay inflation compress margins?
6. How real is the FVA data-scale/AI flywheel — TAM-expanding moat, or eventual price commoditization?

14. What Must Be True

Bull case — for HLI to compound from here, all of the following must hold:

• The M&A recovery resumes and CF re-accelerates within a couple of quarters, confirming the Q4 stumble was timing, not a cycle top. Falsification test: a second consecutive YoY quarterly revenue decline / a stalling deal calendar.
• FR stays elevated (the countercyclical hedge holds — currently guided up for FY27), so a soft CF patch doesn’t open a revenue air pocket. Falsification: FR revenue falls sharply while CF has not yet re-accelerated.
• The comp ratio holds ~61–64% and the roll-up keeps integrating acquired teams without overpayment, so per-share value compounds. Falsification: a comp-ratio breakout above ~65%, or a goodwill impairment.

Bear case — for HLI to de-rate further, any of the following is sufficient:

• FY2026 proves near a cyclical peak and forward earnings decline. Falsification of the bear: two consecutive quarters of CF revenue growth.
• The talent war forces the comp ratio up and compresses margins structurally. Falsification: margins hold at higher revenue.
• A large or dilutive acquisition (or a goodwill impairment) reveals the roll-up over-reaching. Falsification: disciplined, accretive tuck-ins continue.

The pivotal, monitorable variable that discriminates bull from bear is Corporate Finance revenue trajectory over the next two to three quarters.

15. Source Appendix

Primary sources: HLI FY2026 Form 10-K (filed 2026-05-22) and FY2022–FY2025 10-Ks; HLI Q3-FY26 (2026-01-28) and Q4-FY26 (2026-05-06) earnings calls/releases via ROIC.ai; FY2025 DEF 14A (filed 2025-07-25); the 5-year SEC corpus (CIK 0001302215; 39× 8-K, 167× Form 4, proxies); HL/BusinessWire Intrepid release (2026-06-30). Third-party/data: ROIC.ai (statements, ratios, EV); AZI (price CSV, news feed, valuation percentiles); FactorsToday (factor loadings, leaderboard, related-stocks); LSEG/PitchBook 2025 league tables; ION Analytics / Bain / PwC 2026 industry outlooks. Facts reconcile to primary filings; third-party data is labeled and used as cross-check only.

APPENDIX A — Standard Diligence Questionnaire — Houlihan Lokey, Inc. (NYSE: HLI)

Report date 2026-07-10. Fact/Interpretation/Assumption labeled where it matters. Where a question does not map to an asset-light advisory firm, the correct sector analog is given.

General

What thoughtful questions have other investors asked? (1) Is FY2026 near a cyclical peak or a paused recovery? — The central valuation debate; the fin read says “recovering toward a cyclical mid-to-high,” not a trough (Interpretation). (2) Was the Q4 miss timing or a cycle roll-over? — Management frames it as deal-closing slippage from geopolitical/policy shocks (management hypothesis). (3) Is the buyback real capital return or just dilution-offset? — Mostly dilution-offset historically; FY26 open-market buyback stepped up 3x — a genuine but early shift (Fact). (4) Does the moat survive if bankers walk? — The platform (brand, data, deal flow, culture) is harder to replicate than any one MD (Interpretation).

Cyclicality & Earnings Nature

Earnings are cyclical and recovering toward a mid-to-high, not at a trough — advisory revenue is deal-close-dependent and lumpy (FY22 $2.27B → FY23 $1.81B, −20%). Driven by both the external M&A cycle (CF, ~⅔ of revenue) and internal actions (MD hiring, share gains, the roll-up); the countercyclical Financial Restructuring segment (~20%) is the deliberate hedge. Revenue stability is high for advisory (revenue grew in 9 of 10 public years) but low in absolute terms vs a subscription business. Market outlook: a large, growing, global independent-advisory fee pool taking secular share from bulge brackets (<15%→>27% US M&A fee share 2018→2024); international and domestic.

Business Quality & Competitive Moat

More or less competitive? Intensifying (talent war among boutiques), but HLI is gaining share. How profitable (ROIC/ROE)? ROE ~28%, ROIC ~17%, tangible ROE ~57% — elite, capital-light. How profitable is the industry? Very (capital-light, high-ROE) but ~60%+ of revenue accrues to labor via the comp ratio. Understandable? Yes. Undermined by foreign low-cost labor? No — it is the high-cost senior talent; AI is the disruption/consolidation vector (management argues it favors scale). Do brands matter? Yes — reputation is the moat in FVA/restructuring. Switching costs? Near zero in mid-market M&A (you hire the banker); higher in FVA (litigation-defensible name). Moat: real but moderate/multi-part — narrow-genuine in FVA (reputational + niche scale + data flywheel), moderate in restructuring, weak in mid-market M&A; the durable edge is the diversified platform + talent-retention machine.

Financial Condition & Balance Sheet

Unrecognized assets? The banker franchise/brand/data are the real assets and are largely un-capitalized (expensed via comp). Off-balance-sheet liabilities? Minimal — operating leases (~$492M) are on-balance-sheet; earnout/contingent-consideration liabilities are disclosed; a $110.6M redeemable NCI from Audere. Accounting conservatism: high — SBC ($201M) is fully expensed inside the 64% comp ratio and NOT added back to adjusted EPS; narrow, disclosed non-GAAP adjustments; KPMG unqualified opinions, no restatements. CapEx-hungry? No — de minimis capex (~$22M); the only “investment” is acquisitions and MD hiring.

Capital Allocation & Management

FCF generation & use? ~$500M/yr normalized FCF (lumpy year-to-year on bonus timing); used for tuck-in acquisitions, a growing dividend (~$2.62/sh, ~13%/yr, ~42% payout), and buybacks (mostly dilution-offset, FY26 stepped up 3x). Recent acquisitions? Serial tuck-in roll-up (GCA 2021, Waller Helms, Audere/France, Intrepid/energy June 2026); cash-funded, earnout-aligned, zero goodwill impairments. Buying back shares? Yes, but net share count is ~flat (~69M) — buybacks + tax-withholding offset SBC issuance; FY26 open-market step-up is the first real float shrink. Issuing shares to insiders? Yes — equity-heavy Class B comp (a forfeitable retention handcuff). Compensation policy: CEO $11.0M FY25 (trivial salary + large discretionary incentive, part in vesting Class B); say-on-pay ~97%. Management motivations: insiders control ~75% of votes via the dual-class HL Voting Trust (“controlled company”) — long-term, retention-first, but no external check on capital allocation.

Valuation & Market Data

ADR/MLP/K-1? No — a US C-corp; dual-class (Class A public, Class B insider 10-vote). Dividend policy: growing ~13%/yr, ~42% payout, ~1.9–2.0% yield. How profitable? ~28% ROE, ~$426M net income on $2.62B revenue. Net income vs cash from operations? OCF is lumpy (bonus timing) but cumulative OCF/FCF tracks cumulative net income across the cycle — read on a multi-year average, not a single year.

Risks & Downside

What would cause the stock to decline? An M&A-cycle stall / second YoY revenue decline confirming a cyclical peak; a comp-ratio breakout; a goodwill impairment / integration failure; multiple compression from ~22x on cyclical earnings; a financials-sector de-rating (factor headwind). Catastrophic loss risk? Very low — net-cash, asset-light, no trading/lending/inventory, no debt; the FR engine earns more in downturns. Total loss? Not a credible balance-sheet scenario; the realistic downside is a cyclical earnings + multiple contraction (a drawdown), not permanent capital impairment.

Recent News & Events

Environment changed recently? Yes: record FY2026 but a Q4 miss (May 2026) and ~34% de-rate from the Sep-2025 ATH; analyst PT trims; the June-2026 Intrepid energy acquisition; a Dec-2025 Voting Trust refresh; a dividend raise to $0.70/quarter. Significant acquisitions? Intrepid (energy, June 2026, terms undisclosed) — the latest in a steady tuck-in cadence. Accounting-policy changes? None material. Recent changes — CEO Adelson (since June 2024); board refresh; ongoing international build-out (Europe/Asia).

APPENDIX B — Source Appendix

Report date 2026-07-10. Primary (public) sources first; third-party/aggregated data labeled and used as cross-check only.

Primary — SEC filings (EDGAR, CIK 0001302215)

• FY2026 Form 10-K — filed 2026-05-22 (hli-20260331.htm). Business, three-segment MD&A, comp ratio, balance sheet, risk factors, KPMG unqualified opinion. https://www.sec.gov/Archives/edgar/data/1302215/000130221526000053/hli-20260331.htm
• FY2022–FY2025 Form 10-Ks — filed 2022-05-27 / 2023-05-25 / 2024-05-21 / 2025-05-15 — multi-year revenue/margin/segment trends.
• FY2025 DEF 14A (proxy) — filed 2025-07-25 — executive comp, dual-class/voting-trust control, NEO incentive design, insider ownership.
• 5-year SEC corpus (mirrored locally): 5× 10-K, 15× 10-Q, 39× 8-K (+2 8-K/A), 5× DEF 14A, 167× Form 4/3, plus S-3ASR/S-8/ARS. Used for the 8-K timeline and the insider (Form 4) read.
• Key 8-Ks: Q4-FY26 earnings (2026-05-07, adjusted figures in Ex-99.1); Amended & Restated Voting Trust Agreement (2025-12-30); board changes (2025-09-19, 2025-10-06); CEO transition (2024-06-10).

Primary — company disclosures & calls

• Q4/FY2026 earnings release & call — 2026-05-06/07 (via ROIC.ai) — record revenue, Q4 miss, comp ratio, buyback step-up, FR FY27 outlook raised.
• Q3-FY2026 earnings call — 2026-01-28 (via ROIC.ai) — M&A “third inning,” Capital Solutions, mid-market framing.
• HL newsroom / BusinessWire — Intrepid Financial Partners acquisition announcement, 2026-06-30.

Industry / third-party

• LSEG / PitchBook — 2025 global M&A league tables (HLI #1 by deal count; #1 restructuring).
• ION Analytics / WSJ / Dealogic — independent-boutique US M&A fee share (<15% 2018 → >27% 2024).
• Bain & Company / PwC — 2026 PE/M&A outlooks (dry powder ~$1.3–2.6T; ~34% of PE holds >5 years).
• Analyst notes (UBS, Morgan Stanley, KBW) via Yahoo/MarketBeat, May–June 2026 — sentiment context only.

Quantitative data feeds (cross-check; reconciled to filings)

• ROIC.ai MCP — income statement, balance sheet, cash flow, profitability ratios, per-share data, enterprise value, valuation multiples, earnings-call transcripts.
• AZI — 5-year price CSV (OHLCV, beta ~0.94), news feed, valuation_index own-history percentiles (P/E ~67th, P/B ~60th, P/S ~67th; composite ~65th).
• FactorsToday — factor loadings (Market ~0.90, DividendYield ~0.71, Broker-Dealers ~0.43; R² ~0.52), leaderboard (10-yr +22.4%/yr, y1 −27%), related-stocks (EVR/LAZ/SF/JEF/PIPR/PJT/MC comp cross-check).

Note on authority

For US-filer facts, EDGAR and the 10-K/10-Q/DEF 14A are primary; ROIC.ai, AZI, and FactorsToday are third-party aggregated data used to accelerate and cross-check, not to replace the filing. The “$492M long-term borrowing” reported by aggregator feeds is confirmed on the 10-K balance-sheet face to be operating-lease liabilities, not funded debt — the memo treats HLI as net cash accordingly.