Third Point is the New York hedge fund Daniel Loeb founded in 1995 (now at 55 Hudson Yards), investing opportunistically across long/short equities, corporate and structured credit, CLOs and ventures; its flagship Offshore Fund has compounded at roughly 13% net since 1996. Loeb is famous for his caustic quarterly letters and activist campaigns — Yahoo, Sony, Nestlé and Disney have all been targets — and the letters are long-standing required reading on Wall Street.
This is a monthly update from Third Point, a well-known hedge fund (a firm that invests money for wealthy clients and charges high fees). The fund made 3.2% this month and 7.3% year-to-date, helped by bets on stocks like SK Hynix and Amazon. For ordinary readers, this isn't a tip to buy those stocks. Instead, it shows two things: hedge fund returns are reduced by management fees (often 2%) and performance fees (often 20%), and the numbers in such reports are partly model estimates with gray areas. It's worth a read because it gives a peek at how professional investors explain their wins and losses, and why even "transparent" disclosures have limits.
This report is a performance update for the Third Point Master Fund for May 2026. The fund returned 3.2% net in May and is up 7.3% year-to-date. Gains for the month were primarily driven by long equity positions, with SK Hynix the largest contributor, while Telephone and Data Systems and others lagg
Net return of 3.2% for the month, 7.9% QTD, and 7.3% year-to-date. The report does not disclose benchmark returns, so excess returns cannot be compared; it presents both gross P&L and net P&L, with the difference between the two reflecting fees and leverage costs.
| Basis | Month (MTD) | Quarter (QTD) | Year-to-date (YTD) |
|---|---|---|---|
| Fund — net return | 3.2% | 7.9% | 7.3% |
| Fund — gross P&L (Total) | 4.1% | — | 10.2% |
| Benchmark | Not disclosed | Not disclosed | Not disclosed |
Profits this month came primarily from equity longs (gross +3.5%), while shorts lost 0.3% in aggregate; the biggest winner was SK Hynix and the biggest loser was Telephone and Data Systems. Year to date, equity longs have contributed +6.4%, credit and ABS combined +1.5%, and Privates +0.9%.
| Position | Metric | One-line attribution |
|---|---|---|
| SK Hynix Inc | MTD winner / YTD winner | Simultaneously the largest winner for the month and the year |
| Private CRH | MTD winner / YTD winner | Contributed significant positive returns in both the month and the year (private equity holding) |
| Hut 8 Corp | MTD winner | One of the month's profit contributors |
| Quebecor Inc | MTD winner | One of the month's profit contributors |
| Rolls-Royce Holdings PLC | MTD winner | Contributor to the month's profits, and also one of the top five equity long positions |
| Telephone and Data Systems, Inc. | MTD loser | Largest source of loss for the month |
| Somnigroup International Inc | MTD loser / YTD loser | Detracted from performance in both the month and the year |
| Short | MTD loser | Short positions overall had a negative contribution in the month |
| API Group Corp | MTD loser | One of the month's sources of loss |
| MasTec Inc | YTD winner | One of the year's profit contributors |
| Amazon.com Inc. | YTD winner | Contributor to the year's profits and the largest equity long position |
| Siemens Energy AG | YTD winner | One of the year's profit contributors |
| CoStar Group Inc | YTD loser | One of the largest sources of loss for the year |
| Capital One Financial Corp | YTD loser | One of the year's sources of loss |
| Brookfield Corp | YTD loser | One of the year's sources of loss |
This monthly report is a list of existing exposures and does not disclose the detailed changes versus the prior month; this section reconstructs the capital distribution based on the disclosed holdings and exposure structure. Net equity exposure is 52.7% (long 75.1% / short 22.4%), the portfolio's core source of risk; there are 111 equity positions in total (45 long, 66 short), with the top ten positions accounting for 41% of long exposure and -9% of short exposure, indicating relatively high concentration.
This report is a performance update at the Third Point Master Fund level, with assets flowing into the master fund through Third Point Offshore Fund via a master-feeder structure; Malibu Life Reinsurance SPC was acquired on September 12, 2025, and renamed Malibu Life Holdings Limited on September 22. Fund assets are allocated 80% to Third Point Offshore Fund and 20% to Malibu Life Reinsurance SPC. Performance is stated as net returns for newly issued eligible investors in the highest-fee class (2% annual management fee, 20% incentive allocation) participating in all side-pocket private equity investments, net of operating expenses during the period; actual investor returns may vary depending on the timing of capital contributions. ASC 820 asset hierarchy: Level I 46%, Level II 36%, Level III 18%.
The follow-up's explanation of net P&L attribution reveals a dual character in this system: on the one hand it is highly granular, on the other hand it rests on a large number of human assumptions.
The modeling logic for management fees and incentive allocations contains several design choices worth noting:
Management fee allocation base. Fees are allocated proportionally to "average total exposure," rather than by capital commitment, risk contribution, or other finer metrics. This means positions with high turnover and low net exposure are subject to the same fee-base logic as core positions held over the long term, even though the actual management costs incurred (trade execution, research coverage) may differ greatly. This is a classic trade-off that prioritizes operability over precision.
The dual condition for incentive allocations. Accruing an incentive allocation requires that: i) the position's P&L is positive; and ii) there is an incentive allocation in the current MTD period. The second condition may seem redundant, but it is in fact critical—it prevents accruing incentives on individual profitable positions when the fund as a whole is in a high-water mark drawdown. This design echoes the high-water mark mechanism commonly used by hedge funds, but it also means the incentive allocation in the net P&L attribution is a modeled value, not an actual legal obligation.
The recursive nature of reversal logic. When an incentive allocation reversal occurs, the impact is based on the "incentive allocation accrued as of last month's YTD," rather than being recalculated at current market value. This is a pragmatic but potentially biased choice—if the net asset value underlying last month's accrual differs substantially from the current net asset value range (e.g., profit last month and a large loss this month), the reversal amount may deviate from the theoretical "amount that should be reversed." The text provides no further sensitivity analysis on this point.
The most critical disclosure: the document explicitly states that there is "inherent difficulty in determining and allocating the expenses on an investment or sub-portfolio group basis," and that "if expenses were to be allocated on an investment or sub-portfolio group basis, the net P&L would likely be different." This in practice concedes that net P&L attribution is not a depiction of true returns, but an approximation of true returns.
| Dimension | Gross P&L Attribution | Net P&L Attribution |
|---|---|---|
| Fee deduction | Does not deduct management fees, incentive allocations, or operating expenses | Deducts the highest-tier management fee (2% annualized) leverage multiplier, incentive allocation (20%), and proportionally allocated expenses |
| Legal effect | Reflects pre-fee sources of P&L at the portfolio level | Is only a modeled value, not a commitment of actual investor returns |
| Expense allocation method | Not applicable | Allocated proportionally to average total exposure |
| Incentive allocation accrual | Not applicable | Only for positions with positive P&L, and only during MTD periods with incentive allocations |
| Known bias | — | The arbitrariness of expense allocation can significantly affect the net attribution figures of individual positions/sub-portfolios |
Investors' actual returns are usually calculated at the aggregate fund level, whereas this document presents a process of "reverse-engineering" aggregate fees back onto individual positions. The two are mathematically unlikely to be equal, and the document dismisses this risk in a sentence.
The three-way classification of Constructivism / Activism / Post-activism — engagement that has not escalated into a campaign, campaigns currently ongoing, and "post-campaign" — has an obvious timing problem: a position's classification depends on whether the engagement has concluded, and that conclusion may itself be provisional. If a position classified as constructivism in one quarter is escalated to a proxy contest the next, must historical attribution be retrospectively adjusted? The document does not say.
More notably, the follow-up defines "Activism" as "an active campaign currently ongoing." The temporal qualifier "currently" implies that positions from concluded campaigns fall into the "post-activism" category, but the document does not develop the specific definition of post-activism at all (does it cover only residual positions still held after a campaign ends, or also related positions opened after a campaign ends?).
The numerous exclusions that appear throughout the follow-up constitute a multi-layered system of "adjusted exposure."
| Exclusion / Adjustment | Applicable Scenario | Potential Impact |
|---|---|---|
| SPAC holdings (no business combination agreement announced) | Equity positions | SPACs typically hold trust cash in the early post-listing period, with risk characteristics close to cash; excluding them better reflects true market risk exposure |
| Interest rate swaptions, interest rate and FX-related investments | Total exposure figures | Excludes the "gross" exposure of hedging strategies while retaining equity hedges — an asymmetric treatment |
| Interest rate / spread hedges | Specific scenarios | Also excluded from exposure figures |
| Options at current market value (rather than delta-adjusted) | Dollar-adjusted exposure | At high-volatility points, market-value option valuations can be far higher than on a delta basis |
The core logic behind these adjustments is consistent: when reporting "market risk exposure," the report prioritizes the exclusion of low-risk or hedging assets in order to present "purer" directional positions. But the cost of "purity" is that exposure figures use different bases across funds; an investor who directly compares Third Point's exposure with other funds could draw misleading conclusions. The document provides no general cross-fund statement on this point.
The "Other" category includes extremely heterogeneous items: currency hedges, macro investments, legacy direct real estate investments, receivables, unhedged currency gains and losses, interest income/expense, financing gains and losses, and so on.
The size of this bucket is not disclosed, but several sub-items within it are worth noting:
As long as these items are not broken out by sub-asset class, investors cannot verify the completeness of the attribution — that is, whether the sum of P&L from each sub-category equals the fund's total P&L, and whether any residual is entirely placed in Other.
"ASC levels provided are inclusive of legacy private investments and are as of 31 March, 2026." Combined with the other P&L data in this report, this may mean: the P&L and exposure data are current-period, while the ASC valuation levels are legacy data from the previous quarter end. Private investment valuations usually lag by one quarter, which is common in the PE/hedge fund space, but the document presents both together without labeling their different as-of dates, easily leading readers to assume all data have the same time consistency.
The core message conveyed by this follow-up to the Introduction is: Third Point is making an effort to improve transparency, but the ceiling of that transparency is jointly determined by the imprecision of expense allocation, the ambiguity of classification boundaries, and the lag in valuation timing.
For professional readers, there are three operational conclusions worth remembering:
1. Net P&L attribution is a model output, not actual returns. All investors assessing performance on this basis should understand the sensitivity of the expense-allocation assumptions underlying it;
2. Exposure figures are "comparably incomparable." With so many exclusions, the basis must be aligned item by item before comparing across funds;
3. The definitions related to activism are time-dependent. Classifications may drift as events evolve, calling into question the time-series stability of historical attribution.
Overall, this is a high-quality disclosure that meets institutional investors' expectations, yet its essence as a "precise approximation" remains a backdrop that cannot be ignored when reading.