Keystone National Group / Keystone Private Income Fund (KPIF): Diligence Note (First Pass)
Keystone National Group / Keystone Private Income Fund (KPIF): Diligence Note (First Pass)
Prepared 2026-10-20 from the files in sources/ (93 PDFs, four of them exact duplicates). The newest evidence is the August 2026 DDQ, the Q2 2026 investor update and the 30 June 2026 quarterly report, so the note runs about four months behind. No Q3 2026 data was available. This is a first-pass note. It is not full investment diligence, not ODD, and not a recommendation.
Format note: the hedge-fund-diligence template has been adapted. KPIF is a 1940 Act registered, non-listed closed-end fund that offers quarterly tender offers at Board discretion. It is privately placed to accredited investors and qualified clients. It is not an interval fund: repurchases are discretionary, not mandatory [PPM25; TO7/26].
Source key: [DDQ26] Keystone DDQ, Aug 2026 · [Upd2Q26] Q2 2026 Investor Update deck, Aug 2026 · [QR2Q26] Quarterly Investor Report, 30 Jun 2026 (includes unaudited FS) · [SAR26] N-CSRS semi-annual report, 31 Mar 2026 · [AR25] Annual report FY to 30 Sep 2025 (audited, Grant Thornton) · [AR21–AR24] prior annual reports · [Proxy26] Special-meeting proxy, Jan 2026 · [ADV2A] Form ADV Part 2A, 31 Mar 2026 · [ADV1A] Form ADV Part 1A, filed 14 Jul 2026 · [PPM25] Private Placement Memorandum, Jun 2025 · [TS3/26] Tear sheet as of 31 Mar 2026 (printed May 2026) · [Snap1/26], [Snap3/26] Monthly Snapshots · [DbD] "Different by Design" letter, 26 Mar 2026 · [Web4Q25] Q4 2025 webinar deck · [Deck1/26] Fund presentation dated Jan 2026 (file labelled February 2026) · [TO7/26] Tender offer dated 30 Jul 2026
1. First Impression
What it is. KPIF is an asset-backed private credit fund of about $2.06B in total assets and $1.77B NAV as of 30 Jun 2026 [QR2Q26 p.158]. It makes senior secured, mostly first-lien, short-dated loans and leases in four sleeves: Financial Assets, Specialty Real Estate Finance, Equipment Leasing and Corporate Finance [TS3/26 p.3]. Weighted average maturity is 23 months and "over 60%" of positions amortise [QR2Q26 p.3]. Keystone National Group (Salt Lake City, founded 2006, about 2.0B of $2.5B) [DDQ26 §II.1].
Why it may be worth a meeting.
- The record since July 2020 is steady. Class Y has returned 10.12% a year net since inception, and 9.02% over the year to Jun 2026 [DDQ26 p.24]. Every monthly return shown is positive [TS3/26 p.3].
- The style is different from mainstream direct lending: asset collateral, amortising, short tenor, mostly fixed rate, and no software concentration. That difference is real in the schedules, not only in the marketing.
- During the 1H 2026 private-credit redemption wave, when peers gated, KPIF kept running its ~5% quarterly tenders. Whether requests were met in full is not disclosed. Q2 2026 return was still 2.06% for Class Z after writedowns [Upd2Q26 pp.8, 17].
- Service providers for the registered fund are institutional: UMB (administrator and custodian), Ultimus (transfer agent), Grant Thornton (auditor) and Faegre Drinker (fund counsel) [DDQ26 §I.9; AR25 p.48].
What is unclear, and the biggest gaps:
- Performance was restated on 22 May 2026. Every current document says the figures were "recalculated using a revised methodology, which supersedes previously reported figures" [TS3/26; DDQ26; Upd2Q26]. No reconciliation was found (§4).
- The default book is larger than the "<1% loss ratio" headline suggests. By my tally, positions footnoted "In default" in the SEC-filed schedules were about $226M at cost and $190M at fair value on 31 Mar 2026. That is roughly 10.8% of NAV (§6).
- Very smooth marks. All holdings are Level 3, and most sit at or near cost. Keystone itself is the Rule 2a-5 valuation designee. Scalar independently values only positions above 0.9% of the portfolio [AR25 p.13; PPM25; DDQ26 §IV.2].
- Fees and alignment moved against investors. The incentive fee became income-based in Feb 2026, so capital losses no longer reduce it. It is capped at the old level for only two years [Proxy26]. Total expenses are about 3.9–4.9% of NAV depending on share class. Trustees and officers own less than 1% of the fund [Proxy26].
- Liquidity demand has tripled. Redemptions were $210M in the nine months to Jun 2026, against $68M in all of FY2025. Tenders payable at 30 Jun 2026 were $77M, about 4.4% of NAV, close to the ~5% offer size [QR2Q26 pp.158–161; AR25].
- The ownership and distribution model is changing. The Virtus stake, a pending 1933 Act registration that would remove the accredited-investor limit, and Virtus becoming distributor together point toward a wider retail base [Web4Q25 p.8; ADV2A Item 12].
2. Team
Leadership. The co-founders and Managing Partners are John Earl (ex-Lehman Brothers, Shattuck Hammond, Dillon Read and Credit Suisse; leads non-real-estate credit) and Brandon Nielson (ex-Partners Group VP and KPMG CPA; leads real estate). Both have been there since 2006. Brad Allen (since 2010; ex-Wilson Sonsini, CPA/JD) runs finance, legal and operations. Taylor Jackson (since 2017; 14 years at Nearon Enterprises) covers real estate. Grant Calder (since 2018) is head of origination [DDQ26 §I.6]. These five are the named portfolio managers [PPM25, "Portfolio Managers"]. Earl is fund President, Allen is Treasurer, and Nielson is the one interested trustee [AR25 pp.46–47].
Depth. The firm has 42 employees, 40 of them in advisory roles [ADV1A Item 5]. Senior credit hires include Chris Gilker and Suman Mallick (both ex-Revere Capital and Fortress), Raymond Chan (ex-Citi principal investments), Joel Hawkins (ex-CarVal, Aristeia and Ensign Peak), Cody Solomon (ex-Park Square) and Ryan Ruth (ex-Ares middle-market CLO) [DDQ26]. Fourteen of the 49 people listed joined in 2026, and most are analysts and associates.
Turnover. Six people left in 2025 and three in 2026, all described as voluntary [DDQ26 §I.7]. The 2026 leavers include J. Troy Beatty, General Counsel and CCO, and Connor Hoisington, a private credit analyst [Deck1/26 p.9]. The ADV Part 1A filed 14 Jul 2026 still names Beatty as CCO [ADV1A Item 1.J]. Who holds the adviser CCO role today needs confirming. The fund's own CCO is outsourced to PINE Advisor Solutions (Randi Jean Roessler) [AR25 p.47].
Key-person read (interpretation). The process is IC-driven. Earl and Nielson sit on every relevant committee, and the founders sign every investor letter. The bench is broader than a two-person shop. Even so, sourcing relationships, IC authority all sit with four or five Managing Partners, who signed long-term employment agreements with Virtus. The terms of those agreements are not disclosed [Proxy26 p.3]. The edge looks partly institutionalised (three "centers of excellence", an asset-management team) and partly founder-dependent. There is a related-party point too: Keystone may invest alongside "related parties or family members of Brandon R. Nielson", and Nielson recuses from those IC votes [ADV2A Item 10.C].
Governance. The entire Board was replaced in March 2025. The prior trustees resigned and Kent Misener (Chair), Daren Shaw, Herbert Scruggs and Nielson were elected [AR25 Note 1]. The reason for the change was not found.
3. Terms, Liquidity and Alignment
| Term | Detail | Source |
|---|---|---|
| Structure | Delaware statutory trust; 1940 Act closed-end; PPM says non-diversified | PPM25 cover |
| Investors | Accredited investors and qualified clients; $50k minimum; 1933 Act registration pending | TS3/26; Web4Q25 |
| Share classes | A/D (up to 3.50% load), Y, I, Z | ADV2A Item 5.B |
| Management fee | 1.50% of month-end net assets | ADV2A; Proxy26 |
| Incentive fee | 15% of pre-incentive-fee net investment income over a 7% hurdle, full catch-up to 8.24%, paid monthly; excludes realized and unrealized gains and losses (approved 25 Feb 2026) | ADV2A Item 5.B |
| Prior incentive fee | 15% of net profits including realized and unrealized P&L, with a loss recovery account | Proxy26 pp.12–13 |
| Distribution and servicing | 1.00% (A), 0.90% (D), 0.25% (Y), 0.15% (I), none (Z) | Proxy26 |
| Total annual expenses (FY25 actual) | 4.93% A · 4.83% D · 4.18% Y · 4.08% I · 3.93% Z; pro forma under the new fee 4.38% Y / 4.13% Z, before the cap | Proxy26 pp.11–12 |
| Expense limit | 3.00% cap excluding incentive fee, interest, distribution fees and others; no waivers in FY25 | AR25 p.6 |
| Liquidity | Quarterly tender offers of ~5% of net assets at Board discretion; 2% early repurchase fee within 12 months; no lock-up | PPM25; TO7/26 |
| Leverage | $400M revolver maturing 2 Aug 2027; whole portfolio pledged; 300% asset-coverage limit | SAR26; AR25 |
Fee load. The manager earned $50.0M in FY2025: $24.0M of management fees and $26.0M of incentive fees [Proxy26 p.10]. A weighted average gross yield of 12.7% [Upd2Q26 p.3] against net returns of about 9–10% puts fee and expense drag at roughly 3 points or more. With a 12–13% gross coupon the 7% hurdle is cleared every month, so in practice the incentive fee works as a near-permanent 15% share of income. That is my reading.
The incentive-fee change is the main alignment issue. Under the new basis, Keystone is paid on income even when NAV falls from markdowns or realized losses. The proxy shows that, had the new method applied in FY2025, incentive fees would have been $31.0M rather than $26.0M, a 19% increase, because FY2025 losses would no longer have reduced the fee [Proxy26 p.10]. Keystone agreed to cap the fee at the old method for "not less than two years" [Proxy26 pp.10–11]. The stated reason is to let non-qualified-client investors buy in, which is a distribution reason rather than an investor-protection one. Keystone also argues the old basis could reward chasing appreciation [Proxy26 p.4].
GP commitment. Not disclosed. Trustees and officers, who include the Managing Partners, held less than 1% of shares as of 30 Nov 2025 [Proxy26 p.42]. At that NAV, 1% would be up to about $17M, so the actual dollar figure is a primary gap. The founders sold a majority of the management company to Virtus. Remaining insider stakes: Nielson 10–25%, Earls Gate Investments LLC 10–25%, and Hallstatt Holdings LLC 5–10% [ADV1A Sched. A]. Who owns those two vehicles is not stated.
Other conflicts disclosed:
- An affiliated third party in which Keystone principals have economic interests holds a right of first refusal over exclusively floating-rate CRE first-mortgage loans [PPM25, floating-rate loan risk].
- Cross trades and principal transactions are permitted, the latter with CCO approval [ADV2A Item 11.B].
- A co-investment exemptive application filed 19 Dec 2025 is pending [ADV2A Item 12.B].
- Some placement agents share in management and incentive fees [ADV2A Item 5].
4. Performance and Drawdowns
Reported (post-restatement) returns, net, as of 30 Jun 2026 [DDQ26 p.24; QR2Q26 p.9]:
| Class | 1-yr | 3-yr | 5-yr | ITD (Jul 2020) |
|---|---|---|---|---|
| Y | 9.02% | 10.48% | 10.44% | 10.12% |
| Z (largest class) | 9.29% | 10.75% | 10.71% | 10.44% |
| D (no load) | 8.32% | 9.78% | 9.74% | 9.52% |
Calendar-year Class Y total net return [TS3/26 p.3]: 2021 9.78% · 2022 10.76% · 2023 11.19% · 2024 10.72% · 2025 9.85% · 2026 Q1 ~2.0%. Audited Class Y returns for fiscal years to September: FY2025 10.03%, ITD 10.37% at 30 Sep 2025 [AR25 p.6].
Restatement flag. Before May 2026, marketing showed an all-class "Net IRR" and monthly "returns" defined as aggregate net income divided by NAV [Snap1/26; Snap3/26; Web4Q25 p.1]. On that basis the numbers were higher in most years:
| Year | Old basis: "Annual Net IRR", all classes [Snap3/26] | New basis: Class Y total net return [TS3/26] |
|---|---|---|
| 2021 | 10.41% | 9.78% |
| 2022 | 10.86% | 10.76% |
| 2023 | 11.54% | 11.19% |
| 2024 | 10.81% | 10.72% |
| 2025 | 9.99% | 9.85% |
| ITD | 10.5–10.6% | 10.20% (3/26) |
Individual months move by more than the annual totals. For example, Dec 2023 was 1.14% on the old basis and 1.57% on the new, Jan 2024 0.95% vs 1.28%, and Mar 2024 0.85% vs 0.48% [Snap3/26 vs TS3/26]. This shows the old monthly series was not NAV-based. Marketing claims built on the old series, such as "66 consecutive months of positive returns", "0.4% standard deviation" and the KPIF vs CCLFX/BCRED comparison in [DbD], should be re-checked against the restated series. The reason for the change, who required it, and whether past distributions or fees were affected are not stated.
Trend. The return is drifting down: 1-year Class Y went from 10.03% (Sep 2025) to 9.34% (Mar 2026) to 9.02% (Jun 2026). Management guides to about 9% for 2026 [DbD p.2] and flags "some softness expected/possible in 2H 2026" because of a non-accruing hotel-to-apartment book and a $23M corporate loan with a negative credit development [Upd2Q26 pp.9, 35].
Drawdowns. No negative month has been reported since 2020. That should be read as a property of Level 3 marks and short history, not as proof that drawdowns do not occur. Keystone itself notes its volatility figures "may understate the volatility that would result if the Fund's assets were priced in a liquid trading market" [Upd2Q26 p.32]. Loss recognition so far has been gradual:
- Net unrealized depreciation was −2.1M (FY24), −13.1M (9M FY26) [AR23–AR25; QR2Q26 p.159].
- The first realized loss line appears in 9M FY26 at −$3.1M [QR2Q26 p.159].
- Several positions are carried at $0 against about $12M of cost without being realised [AR25 SOI].
Cycle coverage. The fund has lived through the 2022–23 rate shock and the 2026 private-credit stress, but not a full credit-default cycle. Keystone's earlier private credit funds go back to 2008 and give longer context [DDQ26 §III.5]. Net IRRs fall by vintage: Fund II 10.5%, III 11.2%, IV 8.3%, V 8.1%, VI 7.8%, VII 7.4%, VIII 7.4%, IX (2020 close) 5.8%. KPIF's ~10% net over the same 2020–26 window as Fund IX's 5.8% needs explaining: fees, leverage, mark methodology, PIK and mix are the candidates.
5. Strategy and Edge
Claim. Keystone says banks have retreated from smaller, asset-heavy borrowers, and that it earns a premium by lending first-out, asset-collateralised, amortising, covenant-heavy, 2–4 year money in $10–150M tickets where large direct lenders do not compete [TS3/26 p.2; DbD; Web4Q25 p.10]. Targets are a 9–11% net yield and a 12–13% gross yield on new deals [TS3/26; DbD p.2].
Evidence that supports it:
- In the schedules, collateral is mostly hard assets or receivables: equipment leases, construction and bridge loans, receivables facilities. Software is essentially absent [QR2Q26 pp.10–13].
- Turnover is real. There were 65 full exits in the four quarters to Q2 2026, returning $793M on $649M invested (1.22x gross deal-level MOIC) [Upd2Q26 p.11]. Payoffs and amortisation in Q2 2026 were over $201M [QR2Q26 p.3].
- Active investment summaries, totalling about 140 pages per quarter, give unusual position-level transparency [QR2Q26].
Evidence that complicates it:
- "Asset-backed" covers a wide range. The Financial Assets sleeve includes lender finance to specialty and merchant-finance lenders (for example BizFund LLC, $108.5M fair value at 31 Mar 2026) and NAV loans to private-equity funds (for example Kensington Private Equity Fund, $40.1M; "NAV Portfolio Loans I/II/V", $156M combined at 30 Jun 2026) [SAR26 SOI; QR2Q26 p.12]. These rely on underlying pools and fund valuations rather than directly held hard assets.
- PIK is significant. PIK income was 20% of net investment income in FY21, 14% in FY22, 30% in FY23, 33% in FY24 and 16% in FY25 [AR21–AR25]. The Q2 2026 letter criticises direct lenders for "allowing PIK loans" [QR2Q26 p.5].
- The real estate sleeve has been the weak spot. That includes hotel-to-apartment conversions (about $105M outstanding [Upd2Q26 p.9]), three REO properties and a restructured hotel loan [QR2Q26 p.13].
- Exit tables show only completed exits, and each quarter's heading says "all of them with attractive results" [Upd2Q26 pp.39–42]. Positions written down, foreclosed or still open are excluded by construction.
Still to diligence. The loss-given-default history by sleeve. Realised recoveries on foreclosed and REO assets compared with their marks. How much of the 12.7% gross yield is cash versus PIK. Whether the edge (sourcing smaller, asset-heavy deals) scales alongside planned retail growth: inflows have averaged $94M a quarter [Web4Q25 p.3], and the manager says it is "difficult to build a $20 billion+ fund with deals like" these [QR2Q26 p.6].
6. Portfolio, Risk and Capacity
Portfolio at 30 Jun 2026 [Upd2Q26 p.3; QR2Q26]: 136 active investments; 96.5% first lien; $17.8M average invested capital; 23.2 months weighted average to maturity. Allocation is Financial Assets ~28%, Specialty Real Estate ~27%, Equipment Leasing ~23%, Corporate ~22% (Mar 2026 figures) [TS3/26].
Concentration (my computation from QR2Q26 pp.10–12 fair values). The average ticket understates exposure. The seven largest positions total about $704M, which is 35% of investments and 40% of NAV:
- Equipment Leasing Credit Facility V, $145M (cut from $175M per Upd2Q26 p.9)
- Commercial Finance Credit Facility II, $124M
- Corporate Lending Facility IV, $118M
- Corporate Lending Facility V, $93M
- High Voltage Equipment Lease, $78M
- NAV Portfolio Loan V, $76M
- Medical Assets Credit Facility, $70M
The top two are about 13.5% of investments. Keystone upsized one corporate loan to $220M and syndicated $120M of it [Upd2Q26 p.9]. Concentration limits are not stated beyond the RIC tax tests [PPM25].
Credit quality (my tally of SOI positions footnoted "In default"):
| Date | Cost | Fair value | % of NAV | Source |
|---|---|---|---|---|
| 30 Sep 2025 | ~$216M | ~$184M | ~10.7% | AR25 |
| 31 Mar 2026 | ~$226M | ~$190M | ~10.8% | SAR26 |
Several large defaulted positions are carried at or near par, for example Carnaby FA (17M). The cluster of "MC Oslo / MC Rye" loans is marked at 37–99% of cost [SAR26]. At 30 Jun 2026 there were also $51M of REO, foreclosed and restructured positions at fair value [QR2Q26 p.13]. Total cost-to-fair-value markdown across the whole book was about $47M (2.6% of NAV) at 31 Mar 2026 [SAR26].
- Interpretation: "In default" includes maturity and technical defaults that may still be covered by collateral. That is consistent with Keystone's loss ratio, defined as realised plus unrealised losses over cumulative capital invested since 2006 (about $6B), staying under 1% [TS3/26 p.4]. Even so, about a tenth of NAV in defaulted names, mostly marked near par, is the central valuation question for this fund. The tally was computed from extracted PDF text and should be checked against the filings.
Leverage. The line of credit was $130M at 30 Sep 2025 (7.6% of NAV), $125M at 31 Mar 2026 and $167M at 30 Jun 2026 (9.4% of NAV; 8.1% of total assets) [AR25; SAR26; QR2Q26 p.158]. The weighted average rate was 6.82% [SAR26]. Marketing variously says "unlevered" [Snap3/26; DbD], "low-to-no leverage" [TS3/26] and "~7% fund leverage" [Upd2Q26 p.14]. The line has first claim on all fund assets [Upd2Q26 p.19]. The lender is not identified in the reviewed text, and the facility matures August 2027.
Liquidity.
- Repurchases were $28.9M (FY22), $23.5M (FY23), $84.4M (FY24), $67.8M (FY25) and $210.3M (9M FY26). Redemptions payable at 30 Jun 2026 were $77.1M [AR22–AR25; QR2Q26].
- Contributions were $281.8M in 9M FY26, so net flows were still positive.
- The sources say nothing on how much was requested versus accepted in each tender, or whether any tender was prorated.
- The quarterly asset runoff ($150–200M from payoffs and amortisation) is a genuine liquidity buffer.
Valuation process. Every holding is Level 3 and restricted [AR25 SOI fn.1–2]. Keystone is the valuation designee and acknowledges a conflict [PPM25]. Scalar values positions above 0.9% of the portfolio each quarter; Keystone values the rest using the same DCF method [DDQ26 §IV.2]. The Q2 2026 deck describes the marks as "third-party, asset-based valuations" [Upd2Q26 p.14], which overstates the independent coverage the DDQ describes.
7. Key Counterparties and ODD Snapshot
Disclosed facts only. This is not an ODD assessment.
| Role | Provider | Source |
|---|---|---|
| Administrator / fund accountant | UMB Fund Services | AR25 p.48 |
| Custodian | UMB Bank, n.a. | AR25 p.48 |
| Transfer agent / tender administrator | Ultimus Fund Solutions | AR25; TO7/26 |
| Auditor | Grant Thornton (from FY2023; Cohen & Co. for FY2020–22; no disagreements reported) | AR23 |
| Fund counsel | Faegre Drinker Biddle & Reath | DDQ26 |
| Valuation agent | Scalar (positions >0.9% only) | DDQ26 |
| Placement agent | PINE Distributors; VP Distributors (Virtus) approved once 1933 Act registration is effective | ADV2A Item 12 |
| Fund CCO | PINE Advisor Solutions (outsourced) | AR25 p.47 |
| Adviser CCO | Listed as J. Troy Beatty in Jul 2026 ADV; DDQ says he left in 2026; current holder unknown | ADV1A; DDQ26 |
| Credit facility lender | Not identified | — |
| Insurance | D&O, E&O and EPL under Virtus programmes (2025–26) | Virtus coverage files |
Internal inconsistencies to resolve:
- The DDQ says Virtus acquired "an economic interest" and Keystone retains "autonomous control" [DDQ26 §I.5c]. The ADV says Virtus took a "controlling" and majority interest, ownership code D (50–75%) [ADV2A Item 4; ADV1A].
- The DDQ's provider table names WSRP as "Private Credit Fund Auditor", while its text says KPMG audits the private credit funds [DDQ26 pp.15, 26]. This does not affect KPIF but is sloppy.
- The PPM calls KPIF "non-diversified" [PPM25 cover]. The July 2026 tender offer calls it "diversified" [TO7/26 p.2].
- Leverage wording varies across documents (§6), and so does the definition of the performance series (§4).
Unreadable sources: the Keystone certificate of formation and the PINE compliance services agreement are image-only scans with no text layer. They were not reviewed.
8. Known / Unknown / Ask Next
| Area | What we know | What we don't know | Ask next |
|---|---|---|---|
| Team | 5 named PMs, founders since 2006; 42 staff; 9 departures 2025–26 including GC/CCO | Current adviser CCO; Virtus employment and retention terms; IC voting rules | Org chart, IC charter, CCO identity, Virtus agreement summary |
| Terms/Liquidity | 1.5% + 15% income incentive over 7%; ~3.9–4.9% TER; ~5% quarterly tenders | Tender requests vs fills; behaviour after the 2-year fee cap | Tender history (requested/accepted) since 2020; fee-cap letter |
| Performance | Class Y 10.12% ITD; restated May 2026 | Why it was restated; full reconciliation; month-level impact | Old vs new monthly series with explanation; auditor sign-off on method |
| Strategy/Edge | Asset-backed, short, amortising; 65 exits at 1.22x | Loss-given-default by sleeve; PIK share of gross yield; full realised history including losers | All-in realised and unrealised deal ledger since inception |
| Risk/Exposures | ~10.8% NAV in default (my tally); top 7 = 35% of investments; LOC 8–9% | Non-accrual list; collateral coverage on defaulted names; concentration limits | Watch list, non-accrual schedule, LTVs on the top 20 and all defaults |
| Capacity | ~$94M/quarter inflows; manager says $20B is not feasible | Target size after 1933 Act registration; Virtus channel plans | Capacity plan and pipeline conversion rates |
| ODD/Counterparties | UMB, Ultimus, GT, Faegre Drinker, Scalar | Lender identity; Scalar scope beyond the 0.9% threshold; reason for the March 2025 Board turnover | Credit agreement summary; valuation policy; Board minutes summary |
| Alignment | Trustees and officers <1% of fund; founders sold majority to Virtus | Dollar value of insider holdings in KPIF; owners of Earls Gate and Hallstatt | Partner and employee capital in KPIF, history and trend |
9. Diligence Questions
Incentive alignment
- How many dollars do the Managing Partners and employees hold in KPIF today, and how has that changed since the Virtus sale?
- After the two-year cap ends, what stops the income-based fee from rewarding yield at the expense of NAV? Would you accept a total-return lookback?
- Who owns Earls Gate Investments and Hallstatt Holdings, and what are the Virtus earn-out and retention terms?
- Which placement agents share in the incentive fee, and on what terms?
Edge and repeatability
- What share of the 12.7% gross yield is PIK or capitalised interest, by sleeve?
- How do lender-finance and NAV loans fit the "asset-heavy" thesis, and what is their target share?
- Why has KPIF returned about 10% net while Fund IX, with an overlapping 2020 vintage, returned 5.8%?
Performance and drawdowns
- What exactly changed in the 22 May 2026 recalculation? Who initiated it (SEC exam, auditor, Virtus)? Were any fees, distributions or past marketing affected?
- For each defaulted position, especially those carried at par (Carnaby, Trico, Vivo Portfolio), what are the collateral coverage, recovery path and expected loss?
- What were realised recoveries on REO, foreclosed and restructured assets compared with their prior marks?
Team stability
- Who is the CCO now, and why did the GC/CCO leave?
- Why did the entire Board turn over in March 2025?
- How are IC decisions made (votes, vetoes), and what does Virtus approve?
Risk, financing and capacity
- For each 2026 tender: amount requested and amount accepted? Has the LOC been used to fund repurchases?
- Who is the revolver lender, what are the covenants and borrowing-base terms, and what is the refinancing plan before Aug 2027?
- Is there a single-borrower concentration limit? What drove the $175M position, and what are the paydown terms?
- How will the 1933 Act registration and Virtus retail distribution change the investor base and redemption behaviour?
10. Reference Check Prep
- Current investors and RIAs: How did Keystone explain the May 2026 restatement and the default book? Were 2026 tenders filled in full? Were defaults and REO disclosed before investors asked?
- Former investors: Why did you redeem in 2025–26? Did tender timing or proration ever surprise you?
- Former employees (e.g., the 2025–26 leavers): How independent is valuation from the investment team? How are defaulted loans marked? What is the IC culture?
- Service providers (Grant Thornton, UMB, Scalar): What is Scalar's actual coverage? Any valuation disagreements or late adjustments? Were there recurring audit findings?
11. Suggested Materials Request List
- A reconciliation of the old and new performance methods, with monthly series for both.
- Tender offer history: requested, accepted and proration, Q4 2020 to Q3 2026.
- Non-accrual and watch list with collateral coverage and marks; a default and recovery ledger since inception.
- The full deal ledger including losses (not only completed exits), with gross and net IRR by sleeve.
- Valuation policy, Scalar engagement scope and sample reports; minutes of the Board valuation oversight committee.
- Credit agreement summary (lender, covenants, borrowing base, maturity).
- Partner and employee holdings in KPIF over time.
- Virtus transaction summary: governance rights, employment and retention terms, distribution plans.
- Current Form ADV with the updated CCO; compliance testing summary; any SEC exam correspondence.
- Reason for the March 2025 Board change.
- Concentration and limits policy; top-20 exposures with LTV.
- Q3 2026 (30 Sep) quarterly report and the FY2026 audited financial statements once available.
12. Sources Reviewed
Read in full or in the relevant sections:
- Keystone DDQ (Aug 2026)
- Q2 2026 Investor Update (Aug 2026)
- Quarterly Investor Report Q2 2026 (letter, schedules and FS)
- N-CSRS semi-annual report (31 Mar 2026)
- Annual reports FY2020–FY2025 (key statements)
- Proxy statement (Jan 2026)
- Form ADV Part 2A (31 Mar 2026; two identical copies)
- Form ADV Part 1A (filed 14 Jul 2026; Items 1, 5 and Schedule A)
- PPM (Jun 2025; targeted sections)
- Tear sheet (31 Mar 2026)
- Monthly Snapshots (Jan and Mar 2026)
- "Different by Design" letter (26 Mar 2026)
- Q4 2025 webinar deck
- Fund presentation (Jan 2026)
- Tender offer (30 Jul 2026)
Scanned for specific items only:
- Quarterly reports Q4 2025 and Q1 2026 (letters)
- Presentation6 (Jun 2026 figures)
- Repurchase notices 2020–2026 (offer size)
- Virtus D&O, E&O and EPL coverage
- 2022–23 KNG commercial package summary
Not reviewed in depth:
- Quarterly reports Q3 2020 to Q3 2025
- Semi-annual reports 2021–2024
- Declaration of Trust (2019), IMA (2020), Expense Limitation Agreement (2020)
- Subscription booklet
- Privacy notices
- "The Keystone Private Credit Difference" (Jan 2024)
Unreadable (image-only): Certificate of Formation; PINE compliance services agreement.
Duplicates (identical hashes): Q1 2026 quarterly report; 2024 audited FS = "09.30.24_FINAL"; ADV 2A (two copies); 31 Mar 2026 repurchase notice.
No Coverage File existed for this fund, and nothing outside sources/ was used as evidence.
