-3_edited.png)
Why I Voted Against XFLT’s Proposed Change in its Sub-Adviser.
- Sean Dougherty
- Jul 27
- 11 min read
A look at the proposed Octagon-to-King Street sub-adviser change at the XAI Floating Rate & Alternative Income Trust (NYSE: XFLT), what actually drove the fund’s losses, and why the proxy fight targets the wrong manager.
I want to thank my intern, Ryan Carnicelli, a rising junior at Indiana University, for his help in pulling this information together for me. See biography on the website. LinkedIn – Ryan Carnicelli
Summary — Key Takeaways
XFLT’s poor record is real — NAV is down 54% since the 2017 IPO, and just +3.8% cumulative NAV total return since FY2022 — but it is an allocation problem, not a investment-picking problem. Portfolio allocation is the adviser’s (XA’s) job, not the sub-adviser’s (Octagon).
CLO equity pricing collapsed across the entire market — the ten-fund ($10B+ par) CLOiNVESTOR.com FV/Par index fell from 53.8% of par at YE 2024 to 38.8% at Q1 2026. XFLT’s book went 54.9% → 39.2%, indistinguishable from the Index.
EIC shows what a real fix looks like: its manager pivoted in real time — CLO equity cut, a new ABS sleeve built from ~0% to 15% of the portfolio — and its NAV held up. XA held XFLT static portfolio allocation, and its “new” mandate actually keeps combined CLO equity near 40%.
The fees are too high and the proposal doesn’t touch them: 1.70% on portfolio that is almost 50% invested in broadly syndicated loans (“BSL”) book versus a sleeve-by-sleeve appropriate fee of ~1.1%. Octagon’s counterproposal cuts the fee to 1.30. XA’s only change to the fee structure is to retain more of the fees and pay the new sub-advisor less.
King Street has its own problems — poor flagship performance, senior departures, and newly gated redemptions (per Bloomberg).
The upside is in staying the course: ~$9.18/share of unrealized losses plus a ~$5/share discount. If loan spreads widen even modestly, CLO equity re-rates across the board — recover 40% of the markdowns and halve the discount, and the stock could return ~40% over the next year or so.
I voted AGAINST the new sub-advisory agreement and support Octagon’s counter-proposal.
Background
XAI Floating Rate & Alternative Income Trust (“XFLT”) is a roughly $350M (AUM) portfolio of floating-rate credit investments. The fund launched in 2017 and invests in a combination of broadly syndicated leveraged loans (~50%) (“BSLs”), CLO debt (~15%) and CLO equity (~35%). As of June 30, 2026, XFLT’s stock price was $17.55 — a 23% discount to its $22.91 NAV — and the shares yield roughly 15% at the current $0.225 monthly dividend.
The fund has a two-tiered management structure. XA Investments LLC (“XA”) is the fund’s investment adviser: it sets the overall investment strategy, the portfolio allocation targets, and the use of leverage, and it owns financial reporting and investor relations. Octagon Credit Investors (“Octagon”) is the sub-adviser: it picks the individual loans and CLO tranches within XA’s allocation. XA now proposes to replace Octagon with Rockford Tower Asset Management, a newly formed unit of King Street Capital Management (“King Street”). Shareholders vote July 30.
Octagon was founded over 30 years ago, manages more than $30B, and was one of the first CLO managers. While I worked at S&P in the early 2000s, I rated several of their CLOs; in my humble opinion, Octagon is one of the very best credit managers in the business. They are owned by Conning, an international investment manager with almost $200B in AUM. King Street is also a significant credit manager with over $30B in AUM and a real CLO platform (Rockford Tower has managed CLOs since 2017); I have simply never worked with them.
However, according to recent Bloomberg reporting, King Street has severely limited redemptions from its flagship hedge fund, which is performing poorly, and the firm is suffering senior-employee attrition. Bloomberg – King Street’s Poor Returns, Partner Exits Spur Firmwide Revamp
The fund pays a management fee of 1.70% on average managed assets, split between XA and the sub-adviser. Octagon currently receives a bit more than half. XA’s proxy proposes no change to the 1.70% fee — King Street has simply agreed to a lower split, a concession that benefits XA, not shareholders. As I discuss later, I believe XFLT’s fees are excessive because half the assets are BSLs.
Start with the record — it is genuinely poor
In XA’s presentation supporting the change, the case is simple: performance has been bad and the stock trades at a wide discount. On the raw numbers, XA is right.
(Throughout, XFLT figures are adjusted for the March 2026 1-for-5 reverse stock split.)
Net asset value has fallen from $48.90 at the September 2017 IPO to $22.30 at March 31, 2026 — a 54.4% decline. Since inception, NAV total return annualizes to roughly 3.3%, and measured from the start of FY2022, cumulative NAV total return through 3/31/26 is just +3.8%. I will focus on NAV rather than market price deliberately: an investment manager doesn’t control the vagaries of the market, so NAV total return is the cleanest way to judge its performance.
One note on benchmarks: XA measures XFLT against the Morningstar LSTA Leveraged Loan 100 Index (“Morningstar 100 Index”), which is not a fair comparison — only half of XFLT’s portfolio is BSLs, while the other half is in far riskier, more volatile CLO debt and equity. I instead compare XFLT against the other hybrid funds that hold the same mix, ARDC and EIC, using the CLOiNVESTOR.com indexes built from the funds’ own filings (“CLOiNVESTOR.com Hybrid Index”).
XFLT’s performance has been quite poor but we need to look at what drove those losses and who is responsible.
XFLT performed exactly as XA mandated
XFLT’s losses did not come from bad credit decisions. They came from portfolio allocation — XA’s decision to put roughly 35%–40% of the portfolio in CLO equity.
Since the end of 2024, CLO equity pricing has fallen dramatically across the board. Using the CLOiNVESTOR.com CLO Equity FV/Par Index — built from the reported portfolios of the ten listed funds with CLO equity exposure (ARDC, CCIF, ECC, EIC, OCCI, OCTIX, OXLC, PDCC, SPMC and XFLT) (the index covers over $10B in par value of CLO equity) — the average fund’s CLO equity was marked at 53.8% of par at the end of 2024 and 38.8% at March 31, 2026, a 28% decline. The driver was the drop in BSL spreads: per SPMC’s quarterly presentations, the average loan spread across the CLO market fell from 3.42% at YE 2024 to 3.03% at Q1 2026 — an 11% drop that squeezed CLO equity distributions and effective yields.
Over the same period, XFLT’s CLO equity went from 54.9% of par to 39.2% — statistically indistinguishable from the CLOiNVESTOR.com Index of 38.8%. It would not have made a meaningful difference if Eagle Point, Oxford Lane, Carlyle or Sound Point had been managing the sleeve — Octagon’s marks sit mid-pack among the best CLO shops in the business.
The NAV bridge shows the same thing from the fund’s own financial statements. Since the end of FY2024, NAV fell $10.85 per share (−32.7%) — and of the $6.30 of unrealized losses, roughly 90% (~$5.50) was CLO equity; the $2.50 or 7.5% of NAV of realized losses was almost exclusively CLO equity as well. XFLT’s 7.5% average credit losses is in line with CLOiNVESTOR.com Credit Loss Index of roughly 9% for the same period. The BSL and CLO debt books were stable throughout.
What set XFLT apart was not the quality of its CLO equity but the quantity. XFLT did incur $2.50 of realized losses, whic over the last 18 months; however, its losses were in line with the CLOiNVESTOR.com credit loss indexOctagon’s individual credit decision was not bad but entire asset class performed poorly. XA was responsible for the allocation into a risky asset class and was slow to react when the market turned — unlike Eagle Point at EIC, as I will discuss below. (One silver lining flagged here and developed later: that same $9.18/share of unrealized markdowns is potentially a massive upside.)
Hybrid Funds
Comparing XFLT to the two other hybrid funds by asset class shows its books performed in line — the difference was the mix.
XFLT’s 3-year total return is just 3.4%, versus ARDC’s 7.8% and EIC’s 12.6%. The fund remains profitable — roughly 10% NII/NAV, paying a 15.4% forward dividend even though CLO equity is under pressure.
The portfolio detail makes the case plainly. XFLT’s porfolio is barbelled — ~45% in relatively safe BSLs and ~37% in volatile CLO equity — by far the riskiest mix among the hybrids (I consider BB CLO debt, with its very low default history, the safest of these asset classes). A few observations:
CLO equity and CLO debt yields have converged — EIC’s CLO debt book now out-yields its CLO equity.
The two BSL books are nearly identical: ARDC 98.9% of par at a 7.2% yield vs XFLT 95.0% at 7.1%. No selection problem there.
ARDC and XFLT held their allocations static; EIC proactively rotated out of lower-yielding CLO paper into ABS yielding 19.4%.
Allocation drove outcomes: XFLT’s NAV and NII fell 33% and 25%; ARDC’s fell just 8.9% and 11.1%; EIC’s NAV slipped 5.9% while its NII rose 2.9%.
The EIC pivot — what active advisory management actually looks like
Eagle Point Income Company (NYSE: EIC) faced the same collapsing CLO equity market — but instead of sitting pat, EIC dynamically changed its portfolio: over the last year by cutting its CLO equity from 27% to 21% and built an ABS portfolio from roughly zero to 15% of the portfolio. That is a dynamic allocation response executed in real time, without a proxy fight. Only time will tell whether the ABS bet works out, but for now it looks smart.
The uncomfortable part for XA: this is directionally what XA now proposes — yet XA was slow to react, and its revised allocation actually doubles down on the stressed asset class, keeping combined U.S.-plus-European CLO equity near 40% and adding first-loss warehouse exposure. If XA wanted dynamic allocation, it could have let Octagon invest the way EIC did.
XFLT's excessive fees
XFLT charges 1.70% on managed assets — for a portfolio that is half BSLs. Price the book by portfolio at going rates — BSL ~65bps (BKLN), CLO debt ~125bps (EIC), CLO equity ~175bps — and a fair fee for XFLT’s mix is roughly 1.1%. On ~$580M of managed assets, the difference is more than $3M a year of shareholder money.
The comps make it sharper: EIC, an all-CLO portfolio, charges 1.25% of managed assets; ECC, Eagle Point’s pure CLO equity fund, charges 1.75% — levied on its equity base, not on leveraged assets. XFLT charges nearly ECC’s pure-CLO-equity rate for a half-loan book. The King Street proposal leaves the 1.70% untouched (the fee cut in XA’s deck applies to the sister fund OCTIX, not XFLT) — the only party whose economics improve is XA. This is why I favor Octagon’s counter-proposal, which cuts the fee to a level in line with what the blended portfolio actually costs to run.
King Street’s own house
Shareholders should also know what is happening at King Street. Per Bloomberg and trade-press reporting in June and July 2026: the flagship hedge fund returned 2.4% in 2025 and lost 0.5% in the first half of 2026; more than 40 employees, including several senior partners, have departed this year; and in mid-July the firm restricted redemptions from its flagship fund, moving exiting investors into a runoff vehicle. More than 40% of King Street’s ~$30B is already in CLOs, where fees resemble mutual funds more than hedge funds — a shrinking flagship makes a listed fund’s permanent-capital fee stream strategically valuable to King Street. That doesn’t make Rockford Tower a bad manager. It does mean this transaction solves a King Street problem at least as clearly as it solves an XFLT problem.
The discount — the problem actually worth fighting over
Here is the irony of the proxy contest: the metric that most directly measures shareholder harm — the discount to NAV — is the one thing XA’s proposal doesn’t address, and the fight itself is arguably making it worse.
For most of its life XFLT traded at or above NAV — a premium on average in seven of its first eight years, peaking around +10% in 2021. The discount is new: roughly −7% on average in 2025 and −21% in 2026, coinciding with the CLO equity slide, the distribution cut, the reverse split and the proxy fight. Octagon’s proposal to cut fees and buy back shares should reduce the discount directly but CLO equity pricing improving is probably the most important factor in closing the discount.
Reasons for optimism
XFLT carries about $9.18/share of unrealized markdowns and trades roughly $5/share below NAV. As loan spreads widen — and over time they will — CLO equity distributions and pricing recover; with the CLO’s leverage, each 10bps of extra spread flows through to the equity roughly 10-fold. My guess is XFLT recovers 30–50% of the unrealized losses over the next one to two years. There are early signs of a bottom: Q2 2026 CLO fund NAVs are up just over 1% from year-end 2025 — not impressive, but finally moving in the right direction.
If NII holds, XFLT recovers ~40% of the markdowns, and the discount is cut in half over the next year or so, the stock goes from ~$17.60 to roughly $24.00 — about 30% price appreciation, or roughly 40% total return including the ~10% income yield. That would echo FY2021’s +39%, when the fund recovered its COVID-era losses. It seems silly to fire Octagon just in time to hand King Street the credit for a recovery that belongs to the asset class.
Octagon’s counter-proposal
There is a concrete alternative on the table. Octagon proposes cutting the management fee from 1.70% to 1.30%, simplifying the two-tier management structure, and prioritizing the discount — including potential buybacks or a tender offer. A 40bp fee cut is real, immediate, and accrues entirely to shareholders — roughly $2M+ a year, every year. The King Street proposal offers neither. (Activist Bulldog Investors has also urged a vote against the board’s proposal — two organized dissents against the “unanimous” recommendation.)
Bottom line — and my vote
XFLT’s record is poor because of its mandated CLO-equity weight, which was XA’s call. Octagon’s marks tracked the ten-fund average almost exactly, its loan book held near par, and the peers that fared better differ mainly in allocation — the one lever Octagon never held. EIC pivoted; XA stood still, and its “fix” keeps CLO equity near 40% while preserving an oversized fee.
Octagon is reportedly pushing for XA itself to be replaced, and candidly, I understand why. XA’s core responsibilities — strategy and allocation, fee stewardship, financial reporting and investor communication — are precisely where this fund has failed its shareholders. Octagon delivered the mandate it was given.
I voted AGAINST the new sub-advisory agreement and support Octagon’s counter-proposal. The diagnosis points at the allocation and the adviser; the prescription fires the security selector and pays the adviser the same fee.
Disclaimer
This article is for informational and educational purposes only and does not constitute investment, legal, tax or accounting advice, nor an offer to buy or sell, or a solicitation of an offer to buy or sell, any security. Nothing herein should be relied upon as the basis for an investment decision.
The author holds positions in securities discussed in this article (including XFLT) and may buy or sell such securities at any time without notice. This article reflects the author's opinions as of the date of publication, which are subject to change.
The information herein is drawn from SEC filings, fund reports and other sources believed to be reliable, but its accuracy and completeness are not guaranteed. Certain figures are derived or estimated as described in the sources note. Closed-end funds investing in CLO securities involve substantial risk, including leverage, illiquidity and the possible loss of principal; past performance is not indicative of future results. Forward-looking statements, including any recovery scenarios, are inherently uncertain and may not materialize.
Readers should conduct their own due diligence and consult a qualified financial adviser before making any investment decision.
Read the primary filings and sources
• XFLT proxy statement (Form DEF 14A): https://www.sec.gov/Archives/edgar/data/1703079/000121390026070822/ea0291793-02_def14a.htm
• XFLT soliciting materials / King Street announcement (Form DEFA14A): https://www.sec.gov/Archives/edgar/data/1703079/000121390026059094/ea0291568-01_defa14a.htm
• Octagon’s opposition — “vote AGAINST” (Form DFAN14A): https://www.sec.gov/Archives/edgar/data/1703079/000121390026073056/ea0296043-dfan14a_oct.htm
• King Street redemption restrictions: Bloomberg, “Hedge Fund Investors Face Delayed Payouts After King Street Tightens Withdrawals” (July 15, 2026) and “King Street’s Poor Returns, Partner Exits Spur Firmwide Revamp” (June 26, 2026); non-paywalled coverage at Hedgeweek and Yahoo Finance.
Sources: the XFLT proxy filings linked above; XFLT Forms N-CSR (FY2018–2025), N-CSRS (1H FY2019–2026), NPORT-EX quarterly schedules and interim financials; ARDC and EIC Forms N-CSR/N-PORT financial highlights and schedules of investments; CLOiNVESTOR.com CLO Equity Index and the uniform quarterly portfolio summaries built from each fund’s filed schedules (XFLT, ARDC and EIC, annual to FY2023 and quarterly from 2024). NAV and fiscal-year NAV total returns are fund-reported; EIC’s 2023–2025 NAV total returns are calculated from its per-share financial highlights (the method reproduces EIC’s last reported figure exactly); ARDC’s 2026 figure is estimated from its 3/31/26 N-PORT. Fee-rate comparisons are from each fund’s advisory agreements as described in its annual report. XFLT does not disclose per-position cost, so realized-loss attribution is inferred from position exits and marks. XFLT figures are post the 1-for-5 reverse split effective 3/20/2026. This is research and analysis, not investment advice.
Comments