HLF — Herbalife Ltd. · 2026-09-09 · Verdict: WATCH · Conviction 3
Price $12.43 (screen row dated 2026-09-08, no live price) · Mkt cap $1.30B · EV $2.97B · EV/normalised EBIT 5.7x · FCF yield ~21% · Net debt $1,669M · ADV $20.9M Sources read: 10-K 2026-02-18 (Items 1, 1A, 7), 10-Q 2026-08-05, DEF 14A 2026-03-17, 8-Ks 2026-04-14, 2026-04-16, 2026-05-01, 2026-05-06, 2026-08-05, 2026-08-31, Form 4s (12m), Q1 2026 earnings call transcript.
Desk stats - Revenue trend: FY2025 $5,037.5M, up 0.9% on FY2024; Q2 2026 $1,326.8M, up 5.4% (5.8% constant currency), a fourth consecutive quarter of growth (8-K 2026-08-05). The driver is one market: India was $270.6M in Q2, up 32.9% reported and 47.0% in local currency on 45.1% higher volume. Strip India out and worldwide Q2 net sales grew 0.1% ($1,056.2M vs $1,055.5M); ex-India first-half growth was 1.5%. EMEA volume fell 12.1%, China 28.7%, North America 1.9% (10-Q Q2 2026). - Normalised after-tax operating profit: LTM to 30 June 2026 GAAP operating income $492.3M (FY25 $481.0M less H1'25 $255.4M plus H1'26 $266.7M). Add back Technology Realignment $9.0M, Restructuring $3.0M, digital technology $3.4M, Optimization $1.3M and the September 2025 India GST transition charge $11.3M (8-K 2026-08-05, Schedule A TTM column) for normalised operating income of $520.3M. Taxed at the desk's 25% that is $390.2M; at the company's own guided ~35% adjusted effective tax rate, $338.2M. LTM GAAP net income attributable was $164.2M after a $94.6M debt extinguishment loss. - EV / normalised after-tax profit: EV $2,971.8M (104,803,380 shares at $12.43, plus principal debt $2,039.6M, less cash $370.5M). That is 8.8x at 35% tax, 7.6x at 25%. - Leverage: net debt $1,669.1M / normalised EBITDA $642.3M (normalised EBIT plus $120.8M D&A) = 2.6x. The company reports 2.2x on Credit Agreement EBITDA of $742.3M, which also adds back $42.8M of stock compensation and $23.0M of inventory write-downs. Covenants are 4.00x total leverage, 2.50x first lien net and 2.00x minimum fixed charge coverage, all met at 30 June 2026 (10-Q). - Is the growth sustainable? Organic and cash-backed but narrow: India responding to a government tax cut, plus roughly 3% of global price increases, against falling volume nearly everywhere else; the full-year adjusted EBITDA guide was cut $10M at the midpoint to $670-690M on 5 August on FX, with constant currency raised. - What the screen got wrong: the 24.6% normalised EBIT growth is a prior-year comparison artifact. FY2024 operating income of $385.9M carried $69.1M of Restructuring, $26.7M of digital technology and $13.4M of Transformation Program charges (10-K, Item 7); normalise both years and operating profit went from about $495M to about $515M, roughly 4%, not 25%. The screen also taxes at 25% against the guided 35%, omits the $12.0M current portion of debt, and its 19.4% FCF yield is stale FY2025 arithmetic.
1. What the business actually does
Herbalife sells weight management, targeted nutrition, energy/sports and outer nutrition products in 95 markets, exclusively to and through roughly 6.4 million independent Members, of whom about 750,000 had reached sales leader rank at 31 December 2025 (10-K, Item 1). Members buy at a discount of up to 50% off suggested retail and earn commissions and bonuses of up to a further 22%; the company keeps the rest, which is why gross margin is 77.7% and selling expense is 35.1% of sales (10-Q Q2 2026). In the United States a 2016 FTC Consent Order means volume credit is only awarded once product is sold to a verified customer at a profit. Weight Management is 54% of sales; the largest markets are India ($270.6M in Q2), Mexico ($161.0M) and the United States ($266.3M).
2. Why it is mispriced — the edge case
There is no identifiable edge case, and that caps this at WATCH. There is no spin, no index event, no restatement and no forced seller in the bundle. The register is the opposite of forced: Vanguard 12.08%, Nantahala 8.40%, Baupost 8.33%, Route One 7.28%, BlackRock 7.16% (DEF 14A 2026-03-17), on $20.9M of daily volume.
What is real is a dated de-risking the screen's CY2025 data cannot see. In April 2026 Herbalife replaced $800M of 12.250% secured notes due 2029 and a $365M term loan B priced at SOFR plus 6.75% with $800M of 7.750% notes due 2033 and a $650M facility priced at SOFR plus 2.50-3.25% (10-Q, Long-Term Debt). Management put the saving at "approximately $45 million in annual cash interest savings" (Q1 2026 call); quarterly interest expense fell from $53.6M to $37.4M. The 12.25% coupon was the market pricing refinancing risk, and that risk is now dated out to 2029 and 2033 with only the $277.5M of 4.25% converts due June 2028 in between. The equity has not re-rated for it: the stock is 37.7% below its 52-week high.
Against that, the market has two fresh reasons to stay away, and neither is a mistake. On 30 July the long-serving CFO told the board he will retire on 31 December 2026, and on 31 August the company announced that Stephan Gratziani will leave the CEO role effective 31 October 2026, sixteen months after taking it on 1 May 2025 (8-Ks 2026-08-05 and 2026-08-31). Buying an operationally fragile business four weeks after both principals announced their exit, with no successor named, is not a mispricing you can underwrite.
3. Unit economics and growth
Operating margin improved from 7.0% in 2023 to 7.7% in 2024 to 9.6% in 2025 (10-K, Item 7) because the 2024 Restructuring Program delivered about $80M of annual savings for $76.1M of one-time cost. That programme completed on 31 December 2025, and the improvement has stopped: Q2 2026 revenue rose 5.4% while operating income fell 3.2%, from $132.6M to $128.3M, and adjusted EBITDA margin fell 120bp to 12.6%. Gross margin slipped 34bp on mix and inventory write-downs despite 64bp of price. First-half G&A rose $58.8M, of which $26.7M was labour and $20.0M non-income tax "mainly from higher India GST expenses" (10-Q). India gives with one hand and takes with the other: the GST cut from 18% to 5% effective 22 September 2025 acted as a price cut that drove volume, but the input tax that is no longer creditable now sits permanently in G&A.
The CFO was explicit on the Q1 call: "We effectively experienced a price decrease due to the GST reduction, which created momentum... We will annualize the GST impact in September, but we do not expect growth to stop after that... growth rates will moderate." Underneath, the Member base is not growing. Sales leaders after re-qualification were 602,708 in February 2025 against 620,424 in February 2023, with North America down from 69,586 to 52,939 in two years and retention flat at 70.3% (10-K, Item 1). Adjusted EBITDA on the proxy's own definition ran $902.5M in 2021, $725.9M in 2022, $600.9M in 2023, $645.8M in 2024 and $634.9M in 2025.
4. Balance sheet and capital allocation
Debt at 30 June 2026 is $360.0M drawn on the 2026 facility (maturing April 2031, with springing dates in December 2027 and December 2028), $800M of 7.750% notes due May 2033, $600M of 4.875% notes due June 2029 and $277.5M of 4.25% converts due June 2028, against $370.5M of cash and about $245M undrawn. Shareholders' equity is negative $474.5M, the residue of $6.5 billion of lifetime buybacks (10-Q), so book value is meaningless here. Net debt barely moved in the first half, from $1,696.9M to $1,669.1M, because $49.0M of call premium and $19.7M of issuance costs consumed most of the $146.7M of operating cash flow. That same $49.0M is added back inside operating cash flow while being paid through financing, so the flattering LTM 24.7% FCF yield is nearer 21% underlying, and true capital spending exceeds the $61.5M of capex because $35-55M a year of SaaS implementation cost is capitalised outside it. No shares were repurchased in the open market in the first half of either 2026 or 2025; share count rose 1.6% year over year. Insiders sold a net $2.54M over twelve months against a single 1,200-share purchase by a director at $11.31 on 2026-06-04. Five-year total shareholder return in the proxy is $26.83 against $71.50 for the peer group.
5. Management: said versus did
Guidance discipline has been good. FY2026 adjusted EBITDA was set at $670-710M on 18 February, narrowed to $675-705M on 6 May and to $670-690M on 5 August, the last move a $10M FX cut alongside a raised constant-currency range. Q1 beat the high end and Q2 landed at the top of the sales range and the upper end of EBITDA. That is a management team hitting its numbers. It is also a management team that is leaving.
6. Valuation
Base (50%): India moderates but holds, price offsets EMEA and China decline, 2027 adjusted EBITDA $670M at 4.75x EV, net debt $1,500M, equity $16.00. Bear (30%): India volume gives back the GST-driven step after the September anniversary and ex-India volume keeps falling; 2027 adjusted EBITDA $580M at 4.0x, net debt $1,600M, equity $6.90. Bull (20%): Bioniq and Pro2col lift productivity, 2027 adjusted EBITDA $740M at 5.5x, equity $25.00. Probability-weighted $15.05, about 21% above $12.43, with a bear case 45% below. Reverse DCF: at a 10% cost of capital the $2.97B enterprise value implies about $352M of unlevered free cash flow (normalised NOPAT $338M plus $121M D&A less $61M capex less $45M capitalised SaaS) declining perpetually at roughly 2% a year, which given ex-India volume trends is not obviously wrong.
7. Catalysts and timeline
A named CEO successor before 31 October 2026. The Q3 print in early November, guided to $160-180M of adjusted EBITDA against $163.0M in Q3 2025. The Q4 print and first FY2027 guidance in February 2027, which is the first clean read on India after the GST anniversary and the first quarter with Scott Schaefer as CFO. A resumed buyback, permitted by the facility subject to leverage tests.
8. Risks and pre-registered kill criteria
- Worldwide net sales excluding India (both disclosed quarterly) decline year over year for two consecutive quarters.
- India local-currency net sales growth turns negative in any quarter from Q4 2026 onward.
- FY2027 adjusted EBITDA is first guided below $620M, that is below the FY2025 actual of $657.6M.
- Credit Agreement total leverage rises above 3.0x, or revolver drawings exceed $250M. Other risks: the FTC Consent Order's 41.75% compensation cap if eligible US sales fall below 80% of US sales; open transfer pricing and indirect tax audits in multiple jurisdictions; further Vietnam direct-selling rules effective July 2027; and a China business whose volume is down 28.7%.
9. Verdict and one-paragraph summary
WATCH, conviction 3. Herbalife at $12.43 is about 8.8x normalised after-tax operating profit, roughly 5x adjusted earnings and a 21% free cash flow yield, on 2.6x net leverage that the April 2026 refinancing has just made much cheaper and much longer, saving about $45M of annual cash interest by swapping 12.25% notes for 7.75% ones. That is genuinely inexpensive, and the screen understates the improvement in the capital structure while overstating the improvement in the business. The problem is the business. The screen's 24.6% normalised EBIT growth is almost entirely the non-repeat of 2024's $109M of restructuring and technology charges; normalise both years and operating profit grew about 4%. The "fourth consecutive quarter of topline expansion" is one market: strip out India, which grew 47% in local currency because the Indian government cut GST on its products from 18% to 5% in September 2025, and worldwide second-quarter sales grew 0.1%, with EMEA volume down 12.1%, China down 28.7% and North America down 1.9%. Operating income fell 3.2% on that 5.4% of revenue growth, partly because the same Indian tax change added $12M of non-creditable GST to quarterly G&A. Sales leaders sit below their 2023 level, no shares have been repurchased in eighteen months, insiders sold a net $2.5M, and both the CEO and the CFO announced their departures within a month of the last print with no successor named. There is no forced seller and no edge case, so wait for the February 2027 print, the first quarter that laps the India GST cut with a new CEO in place; if ex-India sales are still flat and FY2027 EBITDA is guided above $650M, this becomes a real idea.
Research for discussion, not investment advice. Positions and sizing are the reader's decision.