The FHLB-Private Credit Nexus
Congress created the Federal Home Loan Banks (FHLBs) in 1932 to support housing finance. At March 31, 2026, the third-largest borrower in the system was Apollo Global Management, with $28.2 billion of advances outstanding—behind Wells Fargo and Truist, and ahead of every other bank in the country.[1] Apollo is a private equity firm, not a housing lender. It also happens to own Athene, the largest seller of annuities in the United States.[2]
This note is about why a private equity firm is borrowing $28 billion from the housing bank system, and who absorbs the loss if the assets bought with that money go bad.
TL;DR—
PE-owned life insurers, which are regulated at the state level, are making use of two mechanics to add significant risk to their balance sheets, the losses on which will be borne by policyholders and state guaranty associations if we have a large-scale credit event. First, the National Association of Insurance Commissioners (NAIC) have established a rule that makes low-quality private credit cheap for a life insurer to hold: their capital charges key off of a credit rating, not risk. Second, FHLB advances make these low-quality private credit assets cheap to fund, as advances have a weighted-average cost of ~3.85% against pledged collateral.
Insurance policyholders, or state insurance support funds, will ultimately absorb the credit losses due to their subordination on the life insurance balance sheet. Each dollar of advance encumbers about $1.86 of the insurance company's assets, which are pledged to FHLBs.[3] RBC does not charge for this. When the private credit loans deteriorate, the FHLBs are paid out of the pledged collateral, and the lower-quality assets are left to the policyholders.
The Federal Home Loan Banks
The 11 FHLBs are government-sponsored enterprises (GSEs), established by Congress in 1932 with the aim of supporting housing finance and community lending. Like the better-known GSEs Fannie Mae and Freddie Mac, they are privately-owned but regulated by the Federal Housing Finance Agency (FHFA). The FHLBs are structured as cooperatives, and are owned by their members, which include banks, credit unions, savings and loan associations, and insurance companies.[4]
The FHLBs' collective balance sheet (the collective balance sheet matters because the FHLBs are jointly and severally liable for all debts) amounts to roughly $1.3 trillion.[5] About $735 billion of FHLB assets consist of "advances," which are essentially collateralized short-term loans (about 60% have a maturity of 1 year or less) to their members. With a weighted-average interest rate of 3.85% as of March 31, 2026, advances are a cheaper source of financing than what FHLB members would be able to access from capital markets directly.[6]
The FHLBs are able to offer cheap funding for two reasons: 1) capital market participants believe they operate with an implicit government guarantee similar to Fannie Mae and Freddie Mac, and 2) the advances are collateralized. Importantly, FHLB advances enjoy large haircuts—averaging about 28% on single-family mortgage collateral and about 34% on commercial real estate loans—and they are super-senior in the waterfall of claims. This means, if a bank fails, the FHLB gets paid before the Federal Deposit Insurance Corporation (FDIC) does. According to the Congressional Budget Office (CBO), the FHLB system has never suffered losses on an advance.[7]
Insurance Companies and Private Credit
Insurance companies can be members of the FHLB system, and this is where the linkages to private credit come in.
The presence is significant. Although insurance companies only make up 7.7% of FHLB member borrowers,[8] they are drawing 27% of advances.[9] Furthermore, Apollo Global Management, owner of Athene Annuity and Life Company, is the only non-bank borrower in the Top 5, with about $28 billion of advances outstanding.[1:1] The Federal Home Loan Bank of Des Moines reported that Athene's Iowa subsidiary was its largest borrower and accounted for 21% of its total outstanding advances, as of December 31, 2025.[10]
Since the 2008 crisis and the subsequent period of ultra-low interest rates (thanks Fed!), private equity firms have become increasingly large players in the life insurance industry. According to the NAIC, 139 insurance companies were private equity-owned as of June 2025.[11] Annuity and life insurance are attractive to PE firms because they provide "permanent capital" or "float," (best described by Warren Buffett in his 1995 investor letter[12]) which can be deployed into private equity and private credit investments.
Risk-Based Capital Classification at PE-Owned Life Insurers
U.S. insurance companies are regulated at the state level, and state insurance commissions set investment rules that govern what locally-domiciled insurance companies can hold on their balance sheets. Generally speaking, insurance companies have risk-based capital (RBC) requirements similar to banks, that are meant to limit insurance companies' exposure to risky assets.
Insurance companies invest in a mix of assets, ranging from Treasuries and Agency MBS to corporate bonds, ABS and private credit loans. RBC and liquidity requirements push insurers to hold part of their balance sheet in Treasuries and MBS, which earn near the risk-free rate (adjusted for duration). In order to pay higher yields to annuity customers, and to produce higher returns for shareholders, insurers blend higher-yielding corporates, ABS and private credit into the balance sheet.
Since ABS and private credit should theoretically have significantly higher RBC requirements, which would depress ROE, risky assets can only boost ROE if the capital requirements can be arbitraged somehow. The central game I believe private equity-owned insurance companies are playing is that they are classifying high-risk assets as low-risk ones. The amount of risk-based capital an insurance company has to hold against an asset is heavily determined by its NAIC classification, a range from 1-6. Since year-end 2021 the six classes have been split into twenty designation categories, and the ramp-up in capital charges is steep:
| NAIC category | Moody's | S&P | Life C-1 factor (pre-tax) |
|---|---|---|---|
| Exempt | — | — | 0.000% |
| 1.A | Aaa | AAA | 0.158% |
| 1.B | Aa1 | AA+ | 0.271% |
| 1.C | Aa2 | AA | 0.419% |
| 1.D | Aa3 | AA− | 0.523% |
| 1.E | A1 | A+ | 0.657% |
| 1.F | A2 | A | 0.816% |
| 1.G | A3 | A− | 1.016% |
| 2.A | Baa1 | BBB+ | 1.261% |
| 2.B | Baa2 | BBB | 1.523% |
| 2.C | Baa3 | BBB− | 2.168% |
| 3.A | Ba1 | BB+ | 3.151% |
| 3.B | Ba2 | BB | 4.537% |
| 3.C | Ba3 | BB− | 6.017% |
| 4.A | B1 | B+ | 7.386% |
| 4.B | B2 | B | 9.535% |
| 4.C | B3 | B− | 12.428% |
| 5.A | Caa1 | CCC+ | 16.942% |
| 5.B | Caa2 | CCC | 23.798% |
| 5.C | Caa3 | CCC− | 30.000% |
| 6 | Ca and below | CC+ and below | 30.000% |
Source: NAIC, 2024 Life/Fraternal Risk-Based Capital Forecasting and Instructions, page LR002 (Bonds).[13]
NAIC does not have the staff capacity to do its own credit ratings, given that the universe of securities that life insurers could hold numbers in the millions. Instead, they allow standard rating agency credit ratings to be used as the primary input for determining the NAIC rating. So, a corporate bond rated A+ could receive an NAIC 1 designation.
This is where ratings-shopping comes in. Journalists and market commentators have observed a growing trend of issuers seeking ratings from small credit ratings agencies like Egan-Jones,[14] or issuing "private" credit ratings. (A private credit rating is a rating issued by a ratings agency, but only disclosed to the issuer and select investors, rather than to the general public.) As a result, there is "ratings inflation" going on in these credit books, where the securities would be rated lower (and carry higher capital charges) if all of the ratings were being done by larger agencies, with all information fully public.
A working paper from a team at Columbia says it best:[15]
We show that private ratings systematically understate credit risk and depress required capital. Focusing on the U.S. life insurance sector, we document a ten-fold increase in the use of private ratings since 2018, predominantly in opaque securities and concentrated among large and PE-affiliated insurers. Within the same rating, privately rated bonds are twice as likely to impair than publicly rated bonds, yet are downgraded less often, not more.
As you can see, a bond designated 1.A instead of 3.A carries one-twentieth the capital charge on identical dollars of exposure. The researchers at Columbia find that the worse performance of privately-rated securities corresponds to ratings inflation of 2.5-3.2 notches. Adjusting NAIC classifications downwards by even two notches would have raised required capital by $4.5 billion per year from 2021-2025.[16]
In my view, the Columbia paper is conservative—in the event of a downward credit cycle, I expect the gap between publicly- and privately-rated securities to widen, and the "notch inflation" to be worse than 2-3 notches.
How Insurance Companies Use FHLB Advances
FHLB advances represent extra leverage on top of this ratings manipulation structure. In order to acquire advances, borrowers pledge collateral—Treasuries, agency securities, mortgages and other real estate-related assets.[17] In exchange, they receive cash, at around 3.85% in today's rates.
Viewed in a vacuum, this looks a lot like a repo trade, and not at all profitable with the haircuts FHLBs require. But there are two differences. First, most repo lending is much shorter in term, ranging from overnight to three months, than the balance of advances, which are closer to one year in maturity. Second, repo borrowing funds the asset it is being used to purchase, in essentially a maturity transformation trade. FHLB advances, however, have no restriction on what they can be invested in and ownership of the pledged collateral does not change hands. They are fungible with other sources of funds, and are probably being used to purchase more private credit assets.
Purchasing medium duration 3-5 year private credit assets with an inflated NAIC classification of 1 or 2, at yields that could be north of 9%, using FHLB advances with 1 year maturity at 3.85% becomes both a maturity transformation and a credit transformation trade. With net investment spread in the 5-6% range, and little additional capital required, it is easy to see why this "collateral arbitrage" is attractive to firms like Athene. (The alternative to funding private credit with FHLB advances would be funding private credit with insurance premiums, which cost the yield on the annuity plus sales commissions of 4-7% plus however long the sales cycle is for those annuities.)
When Things Go South
The problem arises when there are capital losses in the private credit book.
Life insurance policyholders and annuity investors pay their premiums with the expectation that they are buying a safe asset. Normally, other safe assets—Treasuries, publicly-rated high-grade corporate bonds and Agency MBS—provide part of the cushion that ensures policyholders will get paid.
In the event of significant private credit losses, however, the FHLB has first claim on the assets used as collateral for its advances. This is slightly different for an FDIC-insured bank, where the FHLB has a super-senior claim. In insurance land, funding agreements (how FHLB advances are structured) are pari passu to policyholder claims. However, state law allows[18] the FHLB to take possession of collateral outside the receivership estate after a seven-day stay, which is effectively the same thing as a super-senior claim.
If cash flows are tight, advances from the FHLB will almost certainly be covered by the pledged collateral. However, since these assets are now encumbered by the FHLB's claim, they are no longer available to make policyholders whole. In Athene's case, this appears to be $1.86 of pledged collateral for every $1 of advance. So whatever losses policyholders already would have taken through the insurance company's NAIC rating manipulation are magnified by the additional leverage drawn from the FHLB.
Since state guaranty associations protect the value of insurance liabilities (similar to the way the FDIC protects deposit liabilities), smaller policyholders won't see direct reductions in their promised payments. But this situation echoes 2008, where privatization of gains (Apollo's juiced ROE from a government-supported leveraged spread trade) and socialization of losses (state guaranty associations rebuilding their rescue funds from assessments on other insurance firms, which will be passed on to policyholders) sparked widespread public anger.
Note: I am not making an accusation of culpability towards federal or state governments. The NAIC is a group of state regulators, who came up with the ratings designation that allows PE-owned life insurers to book dodgy assets with low capital charges. The federal government subsidizes the FHLBs, who add leverage to the state-enabled trade. If and when the trade breaks, absorbing the losses returns to the states.
So What?
Hyman Minsky said it best:
In particular, over a protracted period of good times, capitalist economies tend to move from a financial structure dominated by hedge finance units to a structure in which there is large weight to units engaged in speculative and Ponzi finance. Furthermore, if an economy with a sizeable body of speculative financial units is in an inflationary state, and the authorities attempt to exorcise inflation by monetary constraint, then speculative units will become Ponzi units and the net worth of previously Ponzi units will quickly evaporate. Consequently, units with cash flow shortfalls will be forced to try to make position by selling out position. This is likely to lead to a collapse of asset values.[19]
and
An easier filter for financing ruled after securitization was developed than before. Furthermore more money was chasing financing deals than hitherto. As the thrifts were released from financing single family homes, their funds became available for financing new activities: land development, construction financing and commercial mortgages. This funds availability was combined with a pricing structure by which developers made money from construction quite independently of the success of their projects. The combination of perverse incentives guaranteed that both over and wrong type of building would take place.[20]
The FHLBs can't stop the exuberance in credit markets right now, but regulatory changes aimed at tying their operations to their original mission—supporting housing finance, rather than Ponzi finance (how many private credit loans have now moved to pay-in-kind?[21])—would reduce the pain that policyholders and state guaranty associations will ultimately feel when all this goes bust. Unfortunately, I think we are well past the hedge finance phase and are well into speculative or Ponzi finance.
Although the NAIC, journalists and market analysts have begun to scrutinize the life insurance-private credit nexus and the FHLB's role in it, any regulatory changes that have teeth are going to be too little, too late, as they always are in financial bubbles. Get ready for Occupy Wall Street and Tea Party 2.0.
FHLBanks Office of Finance, Combined Financial Report, Q1 2026 (May 14, 2026), Table 5, Top 10 Advance Holding Borrowers by Holding Company, p. 11. https://cdn.fhlb-of.com/files/2026/2026Q1CFR.pdf ↩︎ ↩︎
Athene Holding Ltd., "Athene Tops Annual Annuity Sales for Third Consecutive Year," West Des Moines, Iowa, March 24, 2026. https://ir.athene.com/news-events/press-releases/detail/202/athene-tops-annual-annuity-sales-for-third-consecutive-year ↩︎
Athene Annuity and Life Company, Annual Statement for the Year 2025, Note 11.B, FHLB Agreements, item (3)(a): total general and separate account collateral pledged to the Federal Home Loan Bank of Des Moines of $43,386,185,339 at fair value against aggregate total borrowing of $23,271,200,000 at December 31, 2025. https://d1io3yog0oux5.cloudfront.net/_92d9911b7eec681c8b848a515199ec98/athene/db/2370/22531/pdf/AAIA_4Q_2025_Statement_-_FINAL.pdf ↩︎
Combined Financial Report, Q1 2026, Notes to Combined Financial Statements, "Background Information," p. F-8. ↩︎
Combined Financial Report, Q1 2026: Combined Statements of Condition, p. F-1; Selected Financial Data, p. 1. ↩︎
Combined Financial Report, Q1 2026: Table 3, p. 10 and p. F-1 (advances principal amount); Figure 3, p. 9 (advances due in one year or less); Table 4.1, p. F-14 (weighted-average interest rate). ↩︎
Congressional Budget Office, The Role of Federal Home Loan Banks in the Financial System (March 2024), p. 10. https://www.cbo.gov/system/files/2024-03/59712-FHLB.pdf ↩︎
Combined Financial Report, Q1 2026, Table 4, Member Borrowers by Type of Member, p. 10. ↩︎
Combined Financial Report, Q1 2026, Table 3, Principal Amount of Advances by Type of Borrower, p. 10. ↩︎
FHLBank Des Moines, Form 10-K for fiscal year 2025, Item 7 (MD&A), p. 38. https://www.sec.gov/Archives/edgar/data/1325814/000162828026016325/fhlbdm-20251231.htm ↩︎
NAIC, "Private Equity" topic page (last updated October 24, 2025). https://content.naic.org/insurance-topics/private-equity ↩︎
Berkshire Hathaway, 1995 Chairman's Letter, "Insurance Operations": "Float is money we hold but don't own." https://www.berkshirehathaway.com/letters/1995.html ↩︎
NAIC, 2024 Life/Fraternal Risk-Based Capital Forecasting and Instructions, page LR002 (Bonds), PDF p. 61. https://in.gov/idoi/files/RBCL24-INpdf.pdf ↩︎
Financial Times, "Does private credit have a ratings problem?" 11 November 2025 (paywalled). https://www.ft.com/content/045232ac-e5a3-43e8-9291-368b44581e79 ↩︎
Xuelin Li, Sangmin Oh, and Giacomo Ricciardi, Rating Without Market Discipline, Columbia Business School Research Paper (May 31, 2026). https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6859158 ↩︎
Li, Oh, and Ricciardi, Rating Without Market Discipline, p. 4. ↩︎
NAIC Capital Markets Bureau (Jennifer Johnson), "Federal Home Loan Banks," pp. 4–5. https://content.naic.org/sites/default/files/capital-markets-primer-federal-home-loan-banks.pdf ↩︎
Iowa Code § 507C.5(3)(a) (2026): "Notwithstanding any other provision to the contrary, after the seventh day following the filing of a delinquency proceeding a federal home loan bank shall not be stayed or prohibited from exercising its rights regarding collateral pledged by an insurer-member." https://www.legis.iowa.gov/docs/code/2026/507C.pdf ↩︎
Hyman P. Minsky, "The Financial Instability Hypothesis," Levy Economics Institute Working Paper No. 74 (May 1992), p. 7. https://www.levyinstitute.org/pubs/wp74.pdf ↩︎
Hyman P. Minsky, "The Capital Development of the Economy and the Structure of Financial Institutions," Levy Economics Institute Working Paper No. 72 (January 1992), p. 24. https://www.levyinstitute.org/pubs/wp72.pdf ↩︎
Rod Dubitsky, "The Vanishing: How BDCs Disappear Bad Loans," April 14, 2026. https://roddubitsky.substack.com/p/the-vanishing-how-bdcs-disappear ↩︎