Private equity has a returns problem that is not showing up in IRRs. It is showing up in distributions.
DPI (distributions to paid-in capital, the measure of actual cash returned to investors relative to capital called) has collapsed for recent vintage years. For the 2021 and 2022 vintages, DPI is running below 0.05x, more than 90% below the 15-year average for comparable vintage years. (Source: Pitchbook) This is not a data anomaly. It reflects a structural problem: sponsors are sitting on portfolios they cannot sell at prices they can accept, cannot refinance at rates that work, and in many cases cannot simply hold indefinitely without LP pressure reaching a breaking point.
That pressure does not stay contained within the equity. It flows directly into the debt. Understanding the transmission mechanism from PE distress to leveraged loan market dynamics is, in our view, one of the more important and underappreciated analytical frameworks for credit investors operating in the current environment.
The IRR illusion in private equity today is the product of a simple dynamic: denominator suppression. Funds that have not returned capital show unrealized marks, not losses. Continuation vehicles and GP-led secondaries delay the distribution clock without resolving the underlying economic question. The question being: what is this portfolio worth to a buyer who is not the original sponsor?
The answer, for most of the 2020 to 2022 vintage, is considerably less than the carrying value implies.
Companies acquired at peak multiples (the average LBO purchase price during the 2020 to 2022 period was approximately 12x EBITDA, compared to roughly 8x in 2007), with peak leverage levels at historically low interest rates are now being evaluated by potential buyers using higher discount rates, lower exit multiples, and earnings that in many cases are below the COVID-era highs that justified the original underwriting. (Source: Pitchbook) The buyer pool has not disappeared. But the valuation gap between what sellers need and what buyers will pay has been wide enough, for long enough, to produce the lowest sponsor exit activity in over a decade.
The consequence is a massive inventory of unrealized positions, a growing cohort of portfolio companies that have been sponsor-held for five years or more, and LPs with capital committed to funds that are not distributing. That last point is where the pressure builds.
For PE-backed companies that cannot be sold, the remaining path is typically to refinance and extend the hold period. The problem is that the refinancing math has fundamentally changed.
A company bought in 2021 at 12x EBITDA with 6x leverage, financed at SOFR plus 350 basis points when SOFR was near zero, was underwritten to all-in interest costs of roughly 3.5 to 4%. That same company, attempting to refinance today at SOFR of approximately 4.3% plus current spread, faces all-in costs of 8 to 9% or higher. For a business with 6x leverage, the difference between 4% and 9% in borrowing costs is the difference between serviceable debt and a balance sheet under fundamental stress.
The sponsors who bought at peak multiples with peak leverage at trough rates are, in aggregate, now managing a large portfolio of companies that were never underwritten to the current financing environment. Many of these companies can service their debt. Many of them cannot do so while also investing for growth, maintaining adequate liquidity, and preserving the operational flexibility required to improve the business. And for a meaningful subset, the debt service itself is the primary impediment to the operational recovery that would justify the original valuation.
This is the refinancing trap. The exit is unavailable at acceptable prices. The refinancing is available, but at terms that compound the original stress rather than relieve it.
The pressure described above has a direct and visible transmission mechanism into leveraged loan markets, and it operates through three channels.
The first is LME activity. Sponsors facing an inability to exit at acceptable valuations and an inability to refinance on workable terms are the primary drivers of the LME volume we documented in our prior piece in this series. The LME is the sponsor's preferred tool for extending the hold period without triggering a formal default. As we described there, these transactions rarely resolve the fundamental balance sheet problem. They create a more complex capital structure, displace incumbent CLO holders, and generate the post-transaction opportunity set that we find increasingly actionable.
The second channel is loan amendments and covenant modifications. Before a company reaches LME threshold, it typically cycles through multiple rounds of documentation relief, EBITDA definition expansion, and covenant holiday requests. Each amendment adds complexity, weakens lender protections, and increases the probability that when the eventual restructuring occurs, the recovery analysis is more complicated than a clean read of the original credit agreement would suggest.
The third channel is defaults, both hard defaults and the LME-driven "synthetic defaults" that now constitute a meaningful portion of the overall default rate. The dual-track default rate, which includes both traditional hard defaults and LMEs, has reached approximately 8%, up from 1% at year-end 2021. (Source: BofA Global Research) At the same time, the maturity wall through 2028 represents over $400 billion in loans that require resolution. (Source: LCD) With PE DPI at historic lows and exit markets still functionally impaired for the 2021 to 2022 vintages, a significant portion of that maturity wall will arrive without the refinancing optionality that sponsors would prefer.
The maturity wall is, in a precise sense, the forcing function that converts latent PE overhang stress into actionable credit market events.
Forward indicators of LME activity suggest the pipeline remains full. Octus data on lender cooperation agreements executed, financial advisors hired by stressed borrowers, and the continued historically low level of DPI all point toward LME activity running at or above its recent pace through at least 2026. The companies that entered LMEs in 2022 and 2023 to address the initial rate shock are now approaching the maturities those transactions were designed to extend. Many will need to act again.
For institutional investors in credit, the implications are specific. The supply of complex, post-LME and pre-restructuring situations in the sub-$1 billion segment of the leveraged loan market is not declining. The structural retreat of experienced distressed capital, which we detailed in our earlier piece in this series, means that supply continues to clear into a thinner bid environment than the historical record would suggest is normal.
The combination of these dynamics (PE overhang driving sustained credit event activity, CLO structural pressure amplifying selling behavior, and reduced competition limiting price discovery) describes a credit market where the dislocations are not cyclical noise. In our view, they are structural features of a market that was built during a decade of conditions that have now reversed, and they may persist for as long as the inventory of 2021 to 2022 vintage PE investments requires resolution.
That process likely has years, not quarters, left to run.
For the full analysis of how these dynamics interact across the credit cycle, download Credit Unbound at axarcapital.com.
This content is provided for informational and educational purposes only and does not constitute investment advice or a solicitation to invest. All investments involve risk, including potential loss of principal. The views expressed reflect the current opinions of Axar Capital Management LP, which may change without notice. Past performance is not indicative of future results. Axar Capital Management LP | 402 W 13th Street, 5th Floor | New York, NY 10014
What is DPI and why is it a useful measure of private equity performance?
DPI, or distributions to paid-in capital, measures the actual cash returned to limited partners relative to the capital they contributed. It is a realized return metric, in contrast to IRR or MOIC, which can reflect unrealized marks that have not been tested in a transaction. For current vintage years, DPI is a particularly important lens because the gap between fund-reported unrealized values and the prices that buyers in the current environment would actually pay is wide. A fund with a strong IRR and minimal DPI has generated returns that exist, for now, only on paper.
Why are sponsor exit volumes so depressed despite strong public equity markets?
PE-backed company valuations were set during the 2020 to 2022 period using purchase multiples, leverage levels, and discount rates that no longer prevail. Strategic and financial buyers evaluating the same businesses today are using higher discount rates and, in many cases, lower earnings estimates, as COVID-era EBITDA highs normalize. The resulting valuation gap between seller expectations and buyer willingness has been sufficient to suppress transaction volumes even as public equity markets have recovered.
How does the PE DPI problem translate into leveraged loan market activity?
Sponsors unable to exit their investments and unable to refinance at workable rates have three options: hold and service the debt, execute an LME to extend the maturity runway, or allow a formal default to occur. The first option is viable for companies with adequate debt service coverage. The second has driven record LME volumes in 2023 and 2024. The third option, while least preferred, is increasingly present in the dual-track default rate data. Each of these paths has direct implications for leveraged loan pricing, CLO portfolio composition, and the supply of distressed and stressed credit in the secondary market.
What is the maturity wall and why does it matter?
The maturity wall refers to the concentration of leveraged loan maturities scheduled to occur over a defined forward period. Over $400 billion in US leveraged loans mature through 2028. (Source: LCD) For companies that were financed at historically low rates in 2020 to 2022 and have not meaningfully deleveraged since, reaching those maturities without a refinancing solution or a change of control transaction means a credit event, either a restructuring or a default. The maturity wall is, in this sense, a forcing function that converts latent balance sheet stress into time-bound credit events.
Is the PE overhang a cyclical problem or something more structural?
It is both, in different respects. The specific vintage concentration (2020 to 2022 deals underwritten at peak multiples and trough rates) is cyclical in origin. But the structural features that prevent a rapid resolution are more durable: the size of the overhang, the mismatch between carrying values and current transaction prices, LP pressure to realize returns, and the absence of the easy refinancing environment that would otherwise allow sponsors to extend their holds indefinitely. In our view, working through the 2020 to 2022 vintage inventory is a multi-year process, and the credit market implications will persist well beyond the current cycle.
Axar Capital Management LP is a $3.3 billion opportunistic investment manager focused on U.S. middle market companies with $200 million to $800 million in debt outstanding. Founded in 2015 and 100% employee-owned, Axar specializes in complex situations where traditional institutional capital operates inefficiently. The firm is headquartered at 402 W 13th Street, New York, NY.
Disclaimer: Axar Capital Management LP ("Axar") has prepared this content for informational purposes only. This content is not intended to constitute legal, tax, financial, or investment advice. While all information contained herein is believed to be accurate, no guarantee, representation or warranty is made as to its accuracy, completeness or fairness. Certain information reflects the current opinions of Axar which may prove to be incorrect and are subject to change. Past performance is not indicative of future results. Investing involves a material risk of loss. Certain statements herein reflect forward-looking views, which are inherently uncertain and subject to change. Actual results may differ materially.