HLNE › analyze
HLNE · Analyze
A portfolio-specific passage was removed from the public build.
Verdict: WATCH · conviction 5.5/10 · FV $85–128 (centre ~$100–110) · Entry $80–92 · Trim 18x fwd
A genuinely good business whose reported earnings quality is worse than the headline, at a price that is fair rather than cheap. The recurring engine is real and compounds in the mid teens. The margin expansion story — which is what a premium multiple would pay for — does not survive a like-for-like restatement. Do not buy at $103. The insiders bought at $78–90 and that is the right zone.
0. Read this first — two independent things are wrong with the public picture
Not one issue but two, from different causes, and they push in opposite directions.
0a. The vendor snapshot is corrupted by the Up-C structure
HLNE is an Up-C: Class A common (Nasdaq) sits above Hamilton Lane Advisors LLC, whose Class B/C membership units are exchangeable into Class A. Class B common stock is a non-economic voting stub. Vendor feeds mishandle this in five separate places.
| Vendor field | Shows | Truth | Cause |
|---|---|---|---|
impliedSharesOutstanding |
67.0M | 55.63M | Double-counts the exchange — adds Class B common on top of a base already containing the paired units |
marketCap |
$6.92B | $5.75B | Cascades from the above. Yahoo's own nonDilutedMarketCap ($5.78B) is the correct field |
enterpriseValue |
$5.27B | $5.68B core / $5.92B incl. TRA | Unreconcilable from any combination of Yahoo's own fields. Discard |
| "Acquisitions" −$96.7M | M&A | $0 M&A | Maps line-for-line to "Contributions to Funds" — GP commitments into HLNE's own funds. PaymentsToAcquireBusinesses is $0 in FY24, FY25 and FY26 |
| "Dividends" −$141.8M | the dividend | $87.7M | Merges Class A dividends ($87.7M) with Class B/C member distributions ($54.1M) |
totalDebt $364.8M |
— | $274.2M | Corporate debt only; the consolidated funds carry no debt |
revenuePerShare $20.62 |
— | $15.90 | Divides by Class-A basic |
Two fields that look broken are actually correct, which is the more useful finding —
trailingEps 6.57 and priceToBook 4.93 both verify. Do not discard vendor data wholesale.
The Shares(dil) series is not a vendor bug. FY25's 40.31M sits between FY24's 53.90M and
FY26's 54.47M because the FY25 10-K excluded 14,016,324 Class B/C units as antidilutive.
The if-converted test flips year to year (units excluded FY18–21 and FY25; included FY22–24
and FY26). Yahoo reported the filings faithfully.
Never CAGR an Up-C diluted share count. It is a test outcome, not a trajectory.
0b. The FRE definition was changed mid-2025, and the change is material
This is the reason the stock is down 36%, and the arithmetic supports the short seller. Full adjudication in §4. The one-line version:
On 2026-04-27 Hunterbrook Media (Bethany McLean et al., disclosed short) alleged HLNE moved fee-related performance revenue onto unrealised NAV gains and simultaneously excluded stock-based compensation from Fee-Related Earnings, claiming like-for-like FRE would have declined ~16% rather than growing 37%. Checked against the filings: −16.0% versus +36.8%. Correct to the decimal.
1. What the business actually is — and the denominator problem
Hamilton Lane is a private-markets allocator, not a GP: fund-of-funds, secondaries, co-investment, separately managed accounts, plus a data business (Cobalt) and a fast-growing evergreen retail platform.
The "$1 trillion" headline is the wrong denominator, and it is not a close call.
| Bucket | 30 Jun 2026 | Effective fee rate |
|---|---|---|
| Non-discretionary AUA | $914.1B | 0.23 bps — earned $20.5M in FY26 |
| Discretionary AUM | $146.3B | ~41 bps blended |
| — of which fee-earning AUM | $84B | 67–78 bps ← this is the business |
The FY26 10-K says it outright: "the growth in AUA from existing accounts does not have a material impact on our revenues. However, we view AUA growth as a meaningful benefit in terms of the amount of data we are able to collect." AUA is a data-acquisition subsidy run at roughly break-even — a legitimate and underrated strategic asset, but anyone valuing HLNE off the trillion is valuing the wrong number. The advisory line fell 10% in FY26 while AUA rose 10%.
Revenue map (FY26)
| Stream | $M | % | YoY | Lock | Repricing risk |
|---|---|---|---|---|---|
| Specialized funds | 374.4 | 64.1% | +18.8% | 10–14 yrs, non-redeemable; 75–85% LP vote to remove | Low (drawdown) / moderate (evergreen) |
| Customized separate accounts | 141.5 | 24.2% | +5.3% | Terminable on 30–90 days' notice | High |
| Reporting, monitoring, data (Cobalt) | 35.8 | 6.1% | +22.3% | Annual, auto-renew | Moderate — fee waived for management-fee clients |
| Advisory | 20.5 | 3.5% | −10.2% | 30–90 day termination | Very high — already repriced to ~zero |
| Fund reimbursement | 9.9 | 1.7% | +3.0% | Pass-through | n/a |
| Distribution management | 2.2 | 0.4% | −17.1% | 30–90 day termination | Very high — dying |
| Total management & advisory | 584.2 | 100% | +13.7% | ||
| Incentive fees | 174.8 | — | −14.0% | Clawback-subject | Cyclical, currently negative |
The tell is in the direction of travel. The two streams that are pure intermediation rent — advisory (−10%) and distribution management (−17%) — are the two shrinking. Specialized funds, where HLNE actually underwrites and structures, grew 19%. The market is already removing the toll booths and paying for the manufacturing.
Correcting a claim that circulates widely: the "10-year locked SMA mandates" premise is factually wrong, per HLNE's own risk factor — separate accounts "typically can be terminated by our clients generally upon 30 to 90 days' notice… or, in some cases, for any reason." The real lock is in the drawdown specialized funds. Of $84B FEAUM, roughly $43B is contractually locked for a decade and roughly $41B is on 90 days' notice. Half a moat.
Client concentration, by contrast, is genuinely excellent and deserves credit: **no client
2% of fees, top 10 = 11%, top 20 = 16%, 2,800+ institutions, 61% of fee revenue non-US.**
2. Financial health — strong, with one reservation that turns out to be the whole story
All figures reconciled to SEC primary filings (FY26 10-K 0001433642-26-000019, Q1 FY27 10-Q
0001433642-26-000042, XBRL company facts).
The scorecard
| Test (analysis_notes §1) | Result | Grade |
|---|---|---|
| FCF & 5–8yr CAGR | $419.1M FY26; 8yr CAGR 20.5%, 5yr 19.8% | A |
| Capital allocation | No M&A, 1.4% capex, 23% of FCF into GP commitments earning ~6.7% + carry rights, 34% returned | A− |
| Debt / Assets | 15.0% corporate; net cash $62.8M; 0.71x debt/FRE; 21.7x interest coverage; 70% fixed; no maturity wall to FY30 | A |
| Shares outstanding | Economic base +1.2%/yr over 8 years — and now shrinking | A− |
| Top-line growth | Total revenue 8yr 15.2%; management fees 6yr 15.6% with zero down years | A |
| Bottom-line growth | Net income attributable 6yr 26.5% | A |
| Per-share metrics | Revenue/sh 6yr 17.7%; FCF/sh 6yr 23.4% | A |
| Dividend overlay (§4) | 9yr DPS CAGR 14.7%, never cut, 34% of FCF | A |
| Earnings quality | ⚠️ See §4. This is the reservation, and it is load-bearing | C− |
The recurring engine is real
| CAGR | Value |
|---|---|
| Management & advisory fees, 6yr (FY20→FY26) | 15.6% — every single year up |
| Total revenue, 8yr | 15.2% |
| FCF, 8yr | 20.5% |
| FCF/share, 6yr (correct economic base) | 23.4% |
| Incentive fees, 6yr | 34.8% — but 157→102→199→175: no trend, only noise |
The blended fee rate has risen four straight years — 0.702% → 0.735% → 0.745% → 0.759% → 0.778% annualized. Management fees grew 14% in FY26 against 13% FEAUM growth. There is no fee compression in this business yet, which is genuinely unusual for a fund-of-funds and is the single strongest fact in HLNE's favour.
Note also AUM grew 4% while FEAUM grew 12% — growth is coming from converting the asset base into fee-earning form, not just gathering assets. Specialized funds ($43B) overtook separate accounts ($41B) for the first time in Q1 FY27.
The share count is the quiet good news
| FY | Class A basic | Fully-exchanged economic base |
|---|---|---|
| 2018 | 19.0M | 49.6M |
| 2022 | 36.5M | 53.7M |
| 2026 | 41.6M | 54.5M |
| Q1 FY27 | 41.5M | 54.0M ↓ |
Class A rises; the economic base barely moves (+1.2%/yr over 8 years). The mechanism is
visible in the FY26 financing section — Proceeds from offering 55,484 / Purchase of
membership interests (55,484). HLNE issues Class A and uses 100% of the proceeds to buy
HLA units from Class B/C holders. These are conversions, not capital raises, and the Class A
holder is not diluted by them.
New and genuinely positive: a $50.0M repurchase of Class B common in Q1 FY27 retired exchangeable units and took the economic base down — the first real return of capital. Separately, HLNE began its first-ever open-market buyback on 2026-02-20, under a program authorised in 2018 and never used, buying 559k shares at an average $89.51.
Capital allocation — clean, disciplined, unfashionable
FY26 deployment of $419.1M FCF: GP fund commitments 23.1% · Class A dividends 20.9% · Class B/C distributions 12.9% · debt repayment 3.0% · capex 1.4% · M&A 0.0%. The last real acquisition was $10.1M in FY22. The GP commitment book ($776.5M) threw off $51.9M of equity income in FY26 (~6.7%) and is the prerequisite for earning carry — a defensible reinvestment, though illiquid and marked to model.
3. The dividend — the least ambiguous part of the file
| FY27 target DPS | $2.40 (raised 11.1% Aug 2026; record 2026-09-21, payable 2026-10-06) |
| 8yr DPS CAGR (FY18→FY26) | 15.1% · 9yr incl. FY27E: 14.7% |
| Cuts or freezes since the 2017 IPO | None |
| Current yield | 2.32% (5yr average 1.73%) |
| Payout | 33.8% of FCF · 41.2% of FRE · 32.5% of FY27E EPS |
FCF payout has fallen from 61% (FY23) to 34% (FY26). FY24's 99.5% was a fees-receivable working-capital artifact that reversed immediately, not an earnings failure.
Stress test: if incentive fees went to zero, management-fee-only economics would still cover ~68% of total distributions and ~50% of the Class A dividend alone. The dividend survives a complete carry drought. This is a dividend grower, not an income holding — the case rests on the 15% CAGR compounding, not on a 2.3% yield.
⚠️ pitfall-dyt-inverts-when-price-caused-the-yield applies, and the two windows disagree
— which is itself the finding:
| Window | Price share of yield rise | Dividend share |
|---|---|---|
| Trailing 12 months (from the $161.13 high) | 76.4% | 23.6% |
| Five-year band period (Aug 2021 → today) | ~0% | ~100% |
Over five years the price is roughly flat while the payout went $1.40 → $2.40 (+71%). That is the compounder signature, not the broken-thesis signature. Over twelve months it is mostly a price move. Both are true and the report states both.
4. ⚠️ The earnings-quality problem — adjudicated against the filings
This section decides the verdict. Everything is from EDGAR primary documents.
4a. The tripwire fires
| FY26 | $M | % of revenue |
|---|---|---|
| Operating income | 324.9 | 42.8% |
| Equity in income of investees (unrealized marks) | +51.9 | |
| Net gain on investments — Consolidated Funds | +83.8 | |
| Consolidated net income | 387.7 | 51.1% ⚠️ |
Net margin exceeds operating margin — arithmetically impossible from operations. 30% of
FY26 pre-tax income came from below-the-line investment marks. Per
pitfall-unrealized-equity-marks-break-headline-pe the source is identified rather than
merely suspected, so the data is usable — but it means GAAP net income here is a mark-driven
figure and TTM EPS is a peak, not a run-rate.
4b. The Q1 FY27 "blowout" was a carry spike, not a step-change
Reported: revenue +56%, earnings +51%. Decomposed:
| Q1 FY27 | YoY | |
|---|---|---|
| Management & advisory fees | $161.4M | +21% |
| Incentive fees | $114.0M | +170% |
| Recurring share of revenue | 58.6% | vs 77.0% in FY26 |
Management attributed the incentive fees primarily to "quarterly crystallization of performance fees from our U.S. private assets Evergreen fund" — i.e. NAV-based. The durable growth rate of this business is the 21% on management fees, not the 56% headline.
4c. The definition change — located, dated, and confirmed by a restatement
Comparing the non-GAAP definitional paragraph across consecutive 8-K exhibits isolates the change to a single filing:
| Release | FRE definition |
|---|---|
| Q3 FY25 (8-K 2025-02-04) | "earnings from recurring management fees… excluding (a) incentive fees and related compensation, (b) interest, (c) tax, (d) equity in income of investees, (e) non-operating gain/loss, (f) other" — no SBC exclusion, no FRPR line |
| Q4 FY25 (8-K 2025-05-29) onward | "earnings from revenues that are measured and received on a recurring basis… excluding (a) incentive fees, net of fee related performance revenues, and related compensation, (b) equity-based compensation…" |
Both alleged changes were made simultaneously, on 2025-05-29. The wording has been stable since.
The restatement is the smoking gun for the fact of the change. The nine months to 2025-12-31 was filed twice:
| 9M ended 31 Dec 2024 | FRE |
|---|---|
| As originally filed (10-Q, 2025-02-04, old definition) | $165,789k |
| As restated comparative (10-Q, 2026-02-03, new definition) | $186,115k |
| Difference | +$20,326k (+12.3%) |
It reconciles exactly: +19,681 SBC + 1,228 FRPR − 583 incentive-comp reclass = +20,326. HLNE restated prior-period FRE upward by precisely the two items alleged.
A portfolio-specific passage was removed from the public build.
4d. Restated on one consistent, conservative definition
Management and advisory fees only, minus all fee-related expenses including SBC:
| FY24 | FY25 | FY26 | Q1 FY26 | Q1 FY27 | |
|---|---|---|---|---|---|
| Management & advisory fees | 451.9 | 513.9 | 584.7 | 133.8 | 163.2 |
| Conservative FRE ($M) | 192.2 | 185.5 | 191.1 | 41.4 | 39.0 |
| Margin | — | 36.1% | 32.7% | 30.9% | 23.9% |
| Growth | — | −3.5% | +3.1% | — | −5.7% |
Hunterbrook's exact window, using HLNE's own reported nine-month column:
| 9M Dec-2024 | 9M Dec-2025 | Δ | |
|---|---|---|---|
| Reported FRE | 186,115 | 254,575 | +36.8% |
| Conservative FRE | 165,206 | 138,713 | −16.0% |
| Conservative margin | 42.8% | 32.3% | −10.5pp |
Management fees grew 29% from FY24 to FY26. Conservative FRE went $192.2M → $191.1M — flat. There is no operating leverage in the pure fee engine; there is negative operating leverage.
This also resolves the apparent contradiction with the reported 48% → 50% → 53% margin series. The 32% and 59% figures are the same period over the same denominator (management fees only): 59.3% as reported, 32.3% like-for-like. The headline series uses total fee-related revenues including FRPR as its denominator. Different denominator, not a different trend.
4e. Adjudication
| Sub-claim | Verdict |
|---|---|
| SBC excluded from fee-related expenses, changed during CY2025 | Supported — located to 2025-05-29; prior period restated +$20.3M |
| FRPR crystallises on unrealised NAV gains | Supported on HLNE's own disclosure — "not dependent on realization events" |
| Margin 32% → 59% | Supported, but it is one period on two definitions, not a time series |
| FRE would have declined ~16% rather than growing 37% | Supported — to the decimal. +36.8% reported vs −16.0% like-for-like |
Fiscal-vs-calendar is not an escape hatch: Apr–Dec 2025 is HLNE's own "Nine Months Ended December 31" column. Hunterbrook used the company's reported period.
4f. In fairness to HLNE — three things the bear case overstates
- The conservative construction is deliberately asymmetric. It strips FRPR revenue while keeping the compensation earned on it (fee-related comp rose 44% in Q1 FY27 on the back of those revenues). −16.0% is the most punitive defensible construction, not the only one. The honest core-engine figure sits somewhere between $191M and the ~$222M you get excluding SBC. HLNE does not disclose enough to pin it — which is itself the problem.
- Excluding SBC from FRE is aggressive but not unique, and none of this touches GAAP. GAAP diluted EPS of $6.57 is verified and correct, and FY26 FCF of $419.1M is real cash. The dividend coverage in §3 is computed on cash and is unaffected.
- Management's strongest rebuttal could not be verified. The claim that the three largest evergreen products generated "over $3.6 billion of total cash realizations" appears nowhere in the filed, text-searchable record — not the FY26 10-K, not the Q1 FY27 10-Q, not any 8-K exhibit text layer. It may be on a presentation page (those are embedded images) or call-only. Unaudited and unverifiable. Realised cash would be the decisive answer to "the marks are fake," and HLNE has not put it in a filing.
What would settle this in one table: a breakout of FRPR between realised proceeds and NAV appreciation. HLNE discloses it nowhere. Until it does, an investor is taking management-defined non-GAAP metrics on trust — from a controlled company where Class B holds 73.2% of votes on 21.4% of the economics and a stockholders agreement binds ~74% of voting power. Public Class A holders have no mechanism to force the disclosure.
5. Moat — real, narrowing, and pointed at the wrong enemy
| Claimed moat | Rating |
|---|---|
| 40-year proprietary data asset | Real internally, weak as a product. Genuine underwriting edge from monitoring $914B of AUA. But Cobalt sits in a $35.8M line and the 10-K concedes the fee is waived or offset for clients who pay management fees. MSCI bought Burgiss ($697M); BlackRock bought Preqin (~$3.2B). HLNE has 785 employees and cannot win that arms race — the waivers suggest it knows |
| GP access / allocation rights | Real but cyclical. Access is only valuable when funds are oversubscribed. With ~$3.8T of unsold sponsor-backed companies and distributions stuck at 14–15% of NAV since 2022, GPs will take capital from anyone. And it is a moat against small allocators, not against BX/KKR/APO — they are the GP |
| Switching costs on SMAs | The premise is wrong — 30–90 day termination, per HLNE's own risk factor. Real lock exists only in drawdown funds |
| Brand with pensions/sovereigns | Real but losing relative share — see below |
| Efficient scale in secondaries | Narrative. Delete it. HLNE's flagship SF VI closed at $5.6B against Ardian $30B, Blackstone Strategic Partners $22B+, Lexington $20B+. Roughly one-fifth of the leaders, in a market that is de-concentrating — and efficient scale requires concentration to rise |
The comparison that should trouble a bull
| FY21 | FY26 | 5yr CAGR | AUM (3/31/26) | |
|---|---|---|---|---|
| HLNE mgmt & advisory fees | $289M | $584M | 15% | $142B (+3% YoY) |
| STEP (StepStone) | $285M | $926M | 27% | $233B (+19% YoY) |
They started within $4M of each other. StepStone is now 1.6x larger in fee revenue and growing AUM six times faster — same model, same March fiscal year. HLNE has more AUA, i.e. it is winning more of the business that pays 0.23 bps and losing the business that pays 67 bps.
The adversarial test — the attack that is already working
The mega-GPs are disintermediating the allocator in the wealth channel, and the scale gap is decisive. Blackstone's private wealth AUM is $324B, +16% YoY, with $8.6B of quarterly sales; BXPE alone did $2.4B in a quarter. HLNE's entire evergreen platform is $19B with $640M of quarterly net inflows. Blackstone's wealth book is 17x HLNE's evergreen platform and 2.3x HLNE's total discretionary AUM. One BX vehicle out-sells HLNE's whole retail franchise roughly 4:1 per quarter.
The structural asymmetry: BX has one fee layer, HLNE has two. HLNE's counter — diversification across managers rather than concentration in one brand — is real, and it is the only one it has. It is also the argument fund-of-funds has been losing for fifteen years.
And HLNE does not own its distribution. The 10-K discloses it acts as sub-advisor where "the distribution partner often aims to provide its clients with products under its own brand, which we achieve by rebranding our existing offerings." A firm whose durable retail asset is brand is voluntarily removing its brand from the product in exchange for shelf space.
The rising fee rate, correctly interpreted
Fee rate up 70→78 bps and FEAUM/AUM up 46%→58% are not the fingerprints of a fund-of-funds defending its pricing. They are the fingerprints of a company changing what it sells — from an allocator earning 30–60 bps on institutional fund selection to a product manufacturer earning 100bps+ on evergreen retail vehicles. Of the +$59.2M increase in specialized-fund revenue in FY26, +$72.1M came from evergreen alone — evergreen was more than all of the growth. Strip it and FY26 growth largely disappears: AUM +3%, SMA FEAUM +2%, advisory −10%, distribution management −17%.
Three consequences: the double-fee criticism gets worse (a retail evergreen stacks HLNE + underlying GP + distributor trail — a triple layer, sold to a buyer who cannot benchmark it); mix shift is a one-time lift, not a compounding rate; and the terminal fee rate will be set by competition with single-layer rivals holding 10–20x the distribution.
Evergreen assessment: a durable business with a fading moat. Hamilton Lane will exist in ten years, profitably, with a real dataset and a real institutional base. But the moat is narrowing while the earnings are widening — and that is the specific combination that destroys capital, because the market prices the earnings.
6. Sentiment — a multiple collapse, and the insider tape cuts the other way
Price −35.9% = earnings +21% × multiple −47%. EPS ran $3.01 → $3.69 → $5.41 → $5.92 → $6.57 straight through the drawdown. From ~29.8x trailing at the Aug-2025 peak to 15.7x today; 10.9x at the 2026-06-22 low of $71.88. Every dollar of the de-rate is multiple.
Sector leg (~two-thirds). Listed alt managers lost ~$265B of market cap Sept-2025 → Mar-2026 on semi-liquid redemption stress: Blue Owl OBDC II suspended redemptions 2026-02-19; OTIC requested 40.7% of NAV in April; Partners Group gated an $8.6B fund on 2026-06-03, paying 62 cents on the dollar. HLNE's low came three weeks later.
The cohort signal is precise: the three worst drawdowns are the three intermediated allocators — HLNE −35.9%, STEP −38.2%, RPC (ex-P10) −31.6% — while the mega-caps with balance sheets held up (APO −9.6%, BX −23.0%). The market did not de-rate "alts." It de-rated fee-on-fee, NAV-dependent alts.
Non-traded BDC context, for scale: Q2 2026 fundraising −82% YoY, net outflows $3.8B, redemption requests 12.4% of NAV — the highest ever measured, sponsors met 38%.
HLNE passed the test its peers failed. Q1 FY27: no gates on any fund, 10 of 12 evergreen funds in net inflow, $640M net. The two exceptions are the non-US credit fund (flat) and GPA, the non-US multi-strategy fund, in net outflow — and that matters because GPA ($6.73B) + US PAF ($6.50B) = $13.2B of the $19B platform.
HLNE also carries far less private-credit risk than its de-rating implies — its US credit evergreen launched April 2026 at $218M, versus APO at 86% of fee-earning assets in credit, ARES 66%, OWL 53%. It is being punished by association, not exposure.
The insider tape — the strongest bull signal in the file
⚠️ pitfall-yahoo-insider-purchases-counts-rsu-grants was applied: the May-2026 grants
(Hirsch 42,145, Delgado-Moreira 38,087 at $0.00) are awards, not purchases, and are
excluded. Only EDGAR-verified P-code transactions are counted.
| Date | Insider | Shares | Price | Value |
|---|---|---|---|---|
| 2026-05-26/27 | Hartley Rogers (Exec Co-Chairman) | 110,932 | $89.99–92.76 | $9.99M |
| 2026-06-11 | Hartley Rogers | 38,290 | $77.86–78.73 | $3.01M |
| 2026-06-11 | David Berkman (Director) | 15,000 | $76.27 | $1.14M |
| 2026-02-20 | Hirsch + Delgado-Moreira + Kramer | 20,775 | ~$107 | $2.23M |
| 2025-11-07 | Delgado-Moreira | 8,000 | ~$130 | $1.04M |
~$18.5M across 7 insiders. Zero sellers since 2025-09-04. Rogers bought $13.0M in 16 days and added lower — $90, then $78 after it fell further. The company bought alongside them.
But note precisely where they bought: $78–90. Not $103.
Analysts: 2 Strong Buy / 4 Buy / 1 Hold, mean target $133.57. No firm downgraded during the entire de-rate — targets were cut $18–59 in May–July, then UBS (+$17), Oppenheimer (+$8) and KBW (+$11) all raised within 24 hours of the print. Targets are following price, not leading it. BMO's $102 sits below spot and has not been revised since May. Short interest is 10.96% of float — the Hunterbrook thesis still on the tape, and part of the +7% earnings pop was mechanical.
Litigation: Johnson Fistel opened an investigation 2026-04-28; Robbins Geller has one open on the Private Assets Fund. No class action complaint found as of 2026-08-12 — treat as "no filing found," not "no filing exists."
7. Valuation
Models applied conditionally, and two deliberately discarded
| Model | Output | Weight | Why |
|---|---|---|---|
| Sum-of-the-parts | $85 – $128 | 40% | The only framing that separates the recurring engine from the lumpy carry and quarantines the contested number into one line |
| Relative forward P/E | $111 – $125 | 20% | Consistent basis across the cohort — but HLNE is at the median, so this is not a discount argument |
| Reverse-DCF | Implies ~2.3–4.0% perpetual earnings growth | 20% | Not a fair value — a hurdle test |
| Bogle expected return | +10.5% to +20.6%/yr | 10% | Entirely hostage to the growth assumption |
| DYT (band reset) | $89 – $114 | 10% | Survives on payout policy and coverage; the raw band is invalid |
| DDM | $56 – $94 | 0% | Values 32.5% of the earnings stream and ignores the rest. Premise fails |
| Graham IV | $55.69 | 0% | Premise fails — see below |
Why Graham is reported and then discarded. fin.py prints $55.64 and it reproduces
($55.69 on the corrected BVPS). It deserves ~zero weight, for a reason specific to this
structure: BVPS $20.98 carries a $284.2M deferred tax asset sitting against a $233.5M TRA
liability — two halves of one transaction, where the TRA is the promise to hand ~85% of that
tax shield back to the Class B/C insiders. The formula credits Class A holders a gross shield
they will receive ~15% of. Netting the pair drops Graham to $47.71 — a 14% swing from a
presentation choice, which is disqualifying for a model whose whole claim is objectivity. It
also has no discriminating power: it returns ~$56 at $161, at $103, and at $71.88.
The DYT band needed resetting. The naive calculation ($2.40 ÷ 1.73%) gives $138.73 — but that 1.73% five-year average was set during the 2024–25 melt-up at 25–30x. Reset to the 2.1–2.7% HLNE yielded when it traded at 12–16x: $89–114, which brackets the current price almost symmetrically. DYT says fairly valued, not 34% upside.
Sum-of-the-parts — the primary model
Bucket (a) uses core FRE ex-FRPR, and per §4 I widen it to span both the SBC-excluded (~$222M) and SBC-included conservative (~$191M) constructions, because the truth is between them.
| Bucket | Bear | Base | Bull | Basis |
|---|---|---|---|---|
| (a) Recurring fee engine | $3,060M | $3,895M | $4,840M | $191M×16x / $205M×19x / $220M×22x pre-tax |
| (b1) FRPR ⚠️ contested | $420M | $800M | $1,150M | ~$105M after-tax × 4x / 8x / 11x |
| (b2) Carried interest | $400M | $550M | $700M | $1.5B accrued, PV × aging × tax |
| (c) GP investment book | $621M | $699M | $776M | 0.80x / 0.90x / 1.00x of $776.5M |
| + Net cash | $63M | $63M | $63M | |
| − TRA | −$234M | −$200M | −$165M | face / PV@8% |
| Equity value | $4,330M | $5,807M | $7,364M | |
| Per share (55.63M) | $77.84 | $104.39 | $132.37 | |
| vs $103.29 | −24.6% | +1.1% | +28.2% |
Stress tests — the most valuable output in the file:
| Test | Per share |
|---|---|
| Zero FRPR entirely (accept the short thesis in full) | $90.01 |
| Zero FRPR and carry written to zero | $80.13 |
| Fee engine on the conservative $191M and zero FRPR | $83.5 |
| GP book marked at 0.70x | $101.6 |
Even if the short thesis is completely correct — FRPR worth nothing and the accrued carry never crystallizing — the recurring fee engine plus the balance sheet is worth ~$80–90/share. The downside is bounded by an identifiable, cash-generative asset, not by sentiment.
The multiple that matters
EV/FRE of 14.7x is the cheapest-looking number in the file and it is the contested one. Strip FRPR and the same enterprise value buys the pure fee engine at ~25.6x. That gap is the short thesis, expressed as a multiple.
No peer FRE table is built, deliberately. Per
pitfall-half-corrected-peer-table-invents-a-discount: HLNE's FRE is hand-reconciled to
primary filings and no peer's is. Every alt manager defines FRE differently, and the
FRPR-inclusion question contested here has different answers at BX, ARES and STEP. Building
that table would manufacture exactly the false discount the note describes. Forward P/E is the
one consistent basis available:
| HLNE | STEP | TPG | KKR | APO | OWL | BX | ARES | BN |
|---|---|---|---|---|---|---|---|---|
| 14.0x | 13.8x | 14.0x | 15.0x | 12.9x | 12.4x | 19.6x | 19.8x | 7.6x |
HLNE trades exactly at the cohort median (14.0x ex-BN). There is no relative discount. And every figure in that row is vendor consensus on Up-C issuers — HLNE's $7.378 is verified, none of the others are. Treat the row as ±10%.
Reverse-DCF: at a 10% cost of equity, $103.29 implies only ~2.3–4.0% perpetual earnings growth against a 15.6% six-year fee CAGR. On the fee-engine construction the market requires ~10.5% fee growth at a flat terminal margin — i.e. it is pricing a fade of roughly one-third off history, with no margin recovery from the depressed level. It is not pricing fee compression (which has not begun) or FEAUM outflows. If the moat fade is a decade of ~10% fee growth with stable margins, the stock is roughly fairly priced.
8. Tensions between the analysts — surfaced, not buried
1. Fundamentals graded health A−; the forensic says core earnings are flat. Both are right and they are not measuring the same thing. FCF is real and compounding at 20.5% over eight years — it includes incentive fees and fund cash flows, and it pays the dividend. The pure fee engine is not levering — conservative FRE went $192.2M (FY24) → $191.1M (FY26) while management fees rose 29%. The company generates real cash; the margin expansion story is an artifact. A premium multiple pays for the second thing.
2. Moat says the rising fee rate is evidence of a changing business; Fundamentals says it is evidence of no fee compression. Both correct, and the synthesis matters: the rate is rising because of mix, not price — and mix shifts once. The Moat reading is the more forward-looking.
3. Valuation's Bogle output (+14%/yr base case) is the most optimistic thing in the file, and the analyst distrusted it for that reason. At g=8% and 11.5x it returns 6.5%, below cost of equity. Bogle is structurally incapable of testing whether mid-teens fee growth persists — which is the entire question.
4. Unresolved, and named as a risk rather than priced: StepStone grew fee revenue 3.25x from parity while HLNE grew 2.02x, and both trade at 14.0x. Either the market has stopped paying for the growth differential — in which case both are cheap and HLNE is the safer — or HLNE is priced for STEP's growth while delivering two-thirds of it. HLNE's higher FRE margin and net cash argue the former. Nothing in this analysis resolves it.
9. Verdict
WATCH · conviction 5.5/10
FV $85–128 (centre $100–110) · Entry $80–92 · Trim 18x fwd
Under §0 of analysis_notes.md this is good company / fairly priced — "wait, watch," not
"buy." It is a notch below that on the quality axis than the headline suggests, and the price
does not compensate.
What is genuinely good, and it is a lot. A 15.6% six-year compounder on the recurring line with zero down years. A blended fee rate that has risen four straight years while the whole field discounts. Corporate net cash, 21.7x interest coverage, no maturity wall to FY30. Essentially no M&A and no empire-building. An economic share count that has moved +1.2%/yr in eight years and is now shrinking. A 14.7% nine-year dividend CAGR, never cut, at 34% of FCF — it survives a complete carry drought. Minimal private-credit exposure in a sector being sold for private-credit fear. No gates while Blue Owl and Partners Group hit theirs. And ~$18.5M of verified insider buying with the Executive Co-Chairman averaging down.
What holds it to 5.5. The FRE definition change is confirmed against the filings, dated to 2025-05-29, and evidenced by a prior-period restatement that reconciles to the dollar. Like-for-like fee-engine earnings are flat to declining — the most punitive defensible construction gives −16.0%, matching the short seller to the decimal, and even the generous construction shows a core margin falling from ~40% to ~32%. Management's best rebuttal, the $3.6B of realised cash, does not appear anywhere in the filed record. The one disclosure that would settle it — FRPR split between realised and unrealised — does not exist, and this is a controlled company where Class A holders cannot force it. The moat is narrowing against attackers 10–20x its size in the channel where all its growth now comes from. It trades at exactly the cohort median with no discount for any of this.
The decisive framing. At $103.29 you pay ~$104 of base-case value, with ~$80–90 of that resting on the uncontested fee engine and balance sheet. You are paying roughly $15–20/share for items whose accounting is disputed and undisclosed. That is not a reckless bet — the downside is genuinely bounded — but it is not a margin of safety either, and this framework demands one.
Why the entry zone is $80–92, not "buy the dip." The ceiling is set at the zero-FRPR stress value (~$90) so that entering means paying nothing for the contested item. And the cleanest argument for patience is empirical: the insiders bought at $78–90. They did not buy at $103.
Why trim at 18x rather than 20x. A mid-teens fee compounder with net cash earns a premium to the 14.0x median, and BX/ARES at ~19.6–19.8x mark where the market pays for the best of this cohort. HLNE should not clear best-in-class while its FRE definition is contested and unaddressed, governance offers Class A no recourse, and StepStone out-grows it from parity. Re-anchoring on the 29.8x of 2024–25 would be adopting the bubble as the baseline.
What would change this
| Direction | Trigger |
|---|---|
| ↑ to ACCUMULATE | HLNE discloses the FRPR realised-vs-unrealised split, or puts the $3.6B realisations figure in a filing · a clean Q2 FY27 with FRE growth on management fees alone · price into $80–92 |
| ↑↑ | Core FRE margin recovering off ~32% · GPA back to net inflow · DOL final rule with HLNE positioned as more than a sub-advised sleeve |
| ↓ to AVOID | A securities class action filing · SEC interval-fund sweep landing on HLNE · any evergreen gate or redemption cap · FEAUM growth stalling · the first sign of blended-fee-rate compression |
A portfolio-specific passage was removed from the public build.
Sources
FY2026 10-K (0001433642-26-000019) · Q1 FY27 10-Q (0001433642-26-000042) · Q1 FY27 8-K
Ex-99.1 (0001433642-26-000039) · Q4 FY26 8-K Ex-99.1 (0001433642-26-000016) · Q3 FY25 8-K
(2025-02-04) · Q4 FY25 8-K (2025-05-29) · 10-Q 2025-02-04 and 2026-02-03 (the restatement
pair) · DEF 14A 2026-07-23 · Forms 4 0001433642-26-000021/22/23/24 · SEC XBRL company facts
CIK 0001433642 · Hunterbrook Media 2026-04-27 · Q1 FY27 earnings call transcript 2026-08-04 ·
HLNE Q4/FY26 earnings presentation · StepStone FY2026 10-K · WealthManagement.com 2026-08-12 ·
Citywire Selector ~2026-08-05 · Robert A. Stanger & Co. Q1/Q2 2026 · Bain Global PE Report 2026
and Midyear 2026 · Jefferies 2025 Global Secondary Market Review · Fortune 2026-03-14 · CNBC
2026-02-19 and 2026-06-03 · Dakota 2026-07-29 · Kirkland & Ellis and Morrison Foerster on the
DOL proposed rule · Yahoo Finance MCP 2026-08-12.
Caveats. roic.ai rejects the ticker HLNE with a server-side SQL error — all long-history data came from the SEC XBRL API instead. Net inflows are not separately disclosed in extractable form (earnings-deck pages 8–23 are embedded images); FEAUM +$10B YoY is a proxy. The $3.6B realisations figure is unverified and unaudited. The class-action finding is "no filing found," not "no filing exists" — Stanford SCAC blocked automated access. Day-one markup allegations are unsizable from public filings. Peer forward P/Es are vendor consensus and unverified individually.