Good morning and welcome to the panel on the implications of private credit markets and non-bank financial intermediation for the Federal Reserve and financial stability.
Private credit. Private credit is an extraordinarily interesting and increasingly impactful asset class. Its growth since the global financial crisis has been nothing short of phenomenal. For instance, private credit loans to United States businesses have demonstrated a 12-time increase between 2012 and 2025. You can see that on the chart here.
According to a just-released paper by the Federal Reserve Board, by the way, it's an outstanding piece of research. As of the fourth quarter, let's get the next one. As of the fourth quarter of 2025, private credit accounted for 6% of total non-financial business sector debt in the United States. Okay, that is half the contribution of depository loans and twice the contribution of syndicated loans.
This growth is part of a broader mosaic with rapid shifts from banking towards non-bank financial institution activity. Benefits are clear. Private credit provided critical access to funding after bank lending shriveled in the wake of post-GFC regulation. Lending and support to economic activity continues to this day, especially for small and medium-sized companies.
Financial structures are tailored and they're tailor-made to meet the needs of borrowers, providing an attractive alternative to bank, equity, bond, and syndicated financing. And there is a shorter time to close deals and fund loans. This is truly part of the US growth and financial market outperformance story over the last decade and a half. Yet risks remain.
Loans are often unrated or marked to model. The risk of stale valuations is real. Borrowers often exhibit lower credit quality relative to syndicated lending. Some retail investors are continuing their attempts to redeem holdings in excess of permitted thresholds, albeit at a much reduced rate, significantly at a time of rising interest rates. Private credit loans are largely tied to SOFR or they float. In contrast, corporate bond loan debt is largely fixed.
The expansion of private credit has deep implications for the Fed's strategic mission to meet its 2% inflation target and safeguard the financial system. For the institution's role as a monetary policy setter, private credit and other non-bank financial institutions' activity impact the transmission mechanism through markets and then back into the real economy. For the Fed's role as a financial stability regulator, risks from leverage to interconnectedness among PE sponsors, insurance companies, banks, and pension funds present.
Sarah Breeden at the Bank of England refers to the latter as a leveraged layer cake. There is plenty for our panel and our SOMC meeting participants to address and discuss with us. We are fortunate to have spectacular panelists. I am thrilled to introduce Ken Kencel, founder and president and CEO of Churchill Asset Management; Howard Marks, co-chair and co-founder of Oaktree; and Tomas Piskorski, professor of finance and the Edward S. Gordon Professor of Real Estate at Columbia Business School.
For more, their impressive bios are available online. Each panelist will speak for 12 minutes. Questions and discussions will follow. So, let's get right into it. Howard.
I was instructed to adjust the mic. I do what I'm told, but maybe since I've been at it no longer Goodman, maybe since I've been at it longest, I appointed myself the job of providing an outline and orientation on the subject of credit. And so that's what I'll do here.
Now, credit is a relatively new term in broad use. I would date it back about maybe 20 years. In 1978, the fateful move of my life was the shift from the equity department to what was called the bond department at Citibank, where I was asked to start a high-yield bond fund. Bond department was synonymous with Siberia at that time. Then the bond department became considered an old-fashioned term and we were called fixed income, and then fixed income has mostly given way to credit today. Now, credit is non-government fixed income, but it is a big deal today and of course has become so as Larry described in the last certainly 15 or 12 years.
My history is, as I say, I was shifted from the equity department to the bond department of Citibank in '78. Before that, it was impossible for a non-investment-grade company to issue bonds if you didn't have that triple-B or better. And the fiduciary standards basically said that if you invest in a non-investment-grade bond and you lose money for your clients, you could be surcharged. So it really was, we had a mentality of good assets, bad assets, and of course you'd have to be nuts to buy the bad assets. And of course there was some sub-investment-grade debt. It was largely seen either in the banks, which did a little syndication in those days. Syndication meant calling four of their friendly banks and sharing it, or at insurance companies, what was called private placements. And my partner Sheldon Stone, who's been with me 43 years, was head of research for private placements at the Pru, but that was it for sub-investment-grade debt.
And then in August of '78, I got a call from the head of the bond department who said, 'There's some guy in California named Milken or something, and he deals in that something called high-yield bonds. Can you figure out what that means?' And I took that job. And Milken's innovation was to say that yes, they may be risky, but if they offer enough yield, on the ones that pay, it'll offset the ones that don't pay, and you'll come out ahead. And so, and this was a sea change in the investment mentality from good-bad to risk and return. Yes, it's risky, but it offers a high enough return. I'll do it. That was a major change, and we wouldn't have many of the things we have today, including especially the private equity industry, if it wasn't for risk-return thinking.
In the '90s, Wall Street developed something called, we alternatively called senior loans, leveraged loans, or broadly syndicated loans, which are kind of like bank loans, but they're sold widely, but without the ability to sell them to the public. And then of course we had the rise of the subprime. We had the global financial crisis in which bank capital was reduced and the bankers were regulated and chastened. And then we had what Larry described. We had the birth of what's called direct lending. And the direct lenders, non-banks, stepped in around 2011 to take the place of the banks that wouldn't loan enough money for the private equity industry to do all the deals it wanted to do, which was a lot of money they needed because they were getting money like mad. And so direct lending was created: loans, private loans to middle-market buyouts.
And of course today there's a lot of it. It all fits under the heading of private credit. When the newspapers write about private credit, when we think about credit, they often confuse private credit and direct lending. What they're talking about is direct lending, I think, but they say private credit. Private credit is an umbrella term that includes direct lending, mezzanine finance, asset-backed finance, lots of things under that heading. But I think that the important thing to note as a result of the history that I described, the non-investment-grade debt total in the US is between high-yield bonds, senior loans, and private credit is 5 trillion, and that's a lot of money. And you know, one of the reasons that the US is preeminent in the world economically, but that has led to nationally, is the operation of our capital markets. And that 5 trillion has facilitated a lot of transactions, some important developments, but also some unimportant shuffling of the deck chairs through buying and selling.
So, of course, the recent development was the sale of private credit to the public, direct loans in the form of BDCs. BDCs do more than just direct loans, but it's a lot of what they do. And it was done in the form of traded BDCs, and if you want to get out, you sell to somebody else like a share of stock, or non-traded BDCs, which is like a closed-end fund, and if you want to get out, you put in for withdrawal. But of course, since the collateral is illiquid, they had to limit the withdrawals to 5% a quarter. And then when there were some ripples of unhappiness about a year ago with some companies like First Brands and others, people put in to get out. They were told they couldn't get out in full. That made them want to get out more, and it cascaded for a while.
There's always the issue of whether the buyers were suitable, whether they had the knowledge to buy, whether they were informed about the illiquidity, what the sales tactics were. We have uncertainty about the marks in the portfolios, as Larry mentioned. And then we also have uncertainty with how managers of private credit will deal with worsening credits, whether they will call defaults or kick the can down the road, and which is right in their duty to their clients. I don't think there's any clarity on that.
So, the important thing about credit, as I've described it, is that, I mean, I think it was Scott Fitzgerald who said, how did he go bankrupt? Two ways: slow and then fast. So how do you go bankrupt? How do you have a really big bust? And to have it, it's not like, well, the stock market turned down. No, that's not enough. You have to have some precipice that you come off of. And usually that precipice is the result of a technological bubble. We've seen some of those. A Wall Street tactics bubble. I would put subprime mortgages under that heading. A popularity bubble, the Nifty Fifty. When I came in to work in '69, the growth companies in America, if you bought their stocks the day I got to work, five years later, you had lost over 90% of your money because they were selling at PE ratios between 60 and 90. And then finally, the other way to go bankrupt is a lot of leverage, too much leverage to get through a low point in the stream.
Systemic risk. I don't think that private credit or direct lending presents systemic risk. If you look at the global financial crisis, which was a great example of systemic risk, which had to be arrested as it was, but took down some very important institutions before it was arrested. You had massive scale in subprime mortgages. The interdependence of the participants, in the sense that each had guaranteed the other and each had sold the other derivative contracts and so forth. And then you had massive leverage, and the banks were leveraged 32 times, which means if your collateral declines 3%, you're insolvent. So that was the formula then. I don't think those elements are in place today. I don't especially, we don't have the holders of private credit are not dependent on each other and they're not highly leveraged. A highly leveraged one would be levered probably one and a half times, not 32 times.
So the point is that private credit is just like other forms of lending, is not good or bad. It can be done well or it can be done poorly. It all depends on the quality of the lending decisions. And of course, a great line on that subject is that the worst of loans are made in the best of times. We've had very good times for 17 years. But you know, it's possible to maintain discipline, and that's what gets you through the low spot. And you know, I'm pleased to say on the subject of doing it well, I think we have, and of course we have probably more experience in sub-investment-grade securities and debt than anybody else. But our portfolios are clean. We have very little PIK debt. We have very little payment-in-kind debt. Very little non-accruing debt. And we have had a few requests to withdraw, and they've all been met, so nobody's been turned away, and that's what I wanted to tell you, and I hope you'll have questions.
Thanks for the invitation. Really happy to be here, and sorry for a bit of a technical delay. So as an academic, as you know, as an economist, I want to give you a data-driven assessment of private credit and financial stability. So, one second.
Okay. So, why do we care about private credit and why the regulators are interested in it? It's a very remarkable fast growth in private credit. You know, if you look in 2016, we had about $300 billion of assets under management, and now we have about 1.5 trillion, and that includes both private equity debt funds as well as business development companies. Now, if we put it in a broader perspective of the entire credit market, it's still relatively small. If you look on the 40-plus trillion dollars of loans outstanding to households and firms, you know, private credit just accounts for 3.5% of that.
Having said that, you know, private credit is a much more important part of the landscape of lending to firms. It already accounts for 10% of business credit. But I want to stress for mid-level firms, it's much more important. They're becoming more and more reliant on that form of credit. So in that segment of the market, this really matters.
Now if we think about financial stability implications of private credit, it's usually a combination of two things. You look on the assets they hold, and you would be interested in what kind of credit risk is there, what type of exposures these assets have. But you would also look on the funding structure. We know that, for example, banking is fragile because, you know, it's thinly capitalized, it's funded with runnable liabilities, there is a duration exposure, and it's usually a combination of asset quality and funding that can create instability.
So, and we saw it, you know, during the Great Depression, savings and loan crisis, global financial crisis, and the recent regional banking crisis. So, the kind of key question is to what extent private credit replicates the type of risks that traditional banking has. So to answer this question, I'll first look on private credit funds. This is a dominant form of private credit intermediation accounting for more than 60% of private credit. So we're looking really at closed-end funds when equity investors, LPs, commit the funds for a certain period of time.
And to really get a sense of how risky these private credit funds are, we got data from Preqin Private Assets. We have data for more than 90% of funds. Detailed data on the asset structure, what type of loans and other investments they have with maturity, performance, returns, but also how they finance themselves. And there are several facts that emerge. Let me first compare private credit funds to a traditional bank. So on the left-hand side, you have a traditional bank. Typical bank is financed 10% equity, 90% debt. Vast majority of this debt are deposits. Some are uninsured, some are insured, but in principle they are all runnable.
And what makes traditional banking particularly risky is there is a fairly thin equity cushion. So in principle, you could imagine even a 10% decline in the value of bank assets. Under some conditions, it could potentially create problems to a bank. Private credit is very different. As you can see, it's very significantly equity capitalized. So a typical private credit fund has 65% of funding coming from equity, and this equity funding from LPs is in the first-loss position, and only then, if you would burn through this very thick equity cushion, the creditors, which are mainly banks that provide funding in terms of credit lines and traditional loans, would be impaired. So the first fact that sort of emerges, there's a very massive difference in capital structure. Private credit is very well equity capitalized.
Now the second thing, private credit does not have a typical fragility of traditional banking. Traditional banks finance long-term loans with short-term funding, long-term loans and investments. That creates a problem like we saw during Silicon Valley Bank and other banks. When investors show up, you know, the bank has to start liquidating losses. If we have some book value of asset that's above the market value, that can create instability and bank failure. Private credit is structured very differently. Typical maturity of a loan and other investment is 3 to 5 years, while the commitment from LPs, from the equity, is 8 to 12 years. In other words, there's plenty of time for loans to mature on average for most of them before you have to repay your creditors. And these creditors, mainly equity investors, they cannot exit before that.
Third, private credit is actually quite diversified across industries, also across geographies. As you can see, it's really a well-diversified portfolio. You know, financials are an important part of that, but there's certainly not a single sector that's very deeply, you know, overweighted there. And finally, you know, if you look on historical returns, they were pretty good. LPs that invested in private credit funds, on average, they get 10% return after fees they pay to GPs. And now in asset pricing literature, if you do some adjustments for risk and volatility, this looks like there are fair returns given the type of risk LPs get, more or less. There's not tremendous deviations.
Now there were some losses historically in the past, but given how well capitalized these private credit funds are, only about 5% of them, way less than 5%, ever impaired the bank creditors. Now it doesn't mean that the performance will be as good going forward, we might have more losses, but you know, overall the summary on the private credit funds: they are well capitalized, there is no asset-funding mismatch, and there is little runnable funding because investors are locked in.
Let me make a few comments on the business development companies. This is another sector of private credit that's also grown very rapidly. These business development companies currently account for about $600 billion of assets, so for about 38% of private credit. Importantly, you know, they are also fairly diversified, but there is a bigger exposure to software, and that's why they suffered a bit of concerns regarding, you know, AI and the implication for the software investments. But as you can see from this picture, a lot of that growth in BDCs was in private non-listed BDCs. That has important implications because if the BDC company is public, you can exit at any time by selling a share at the market price. But now we have $400 billion of private non-listed BDCs, and the mechanism there to exit is you will sell the shares at NAV, and the investors can exit at any point of time if they want, and that kind of erodes the traditional advantage of private equity funds. You start having the issue of runnable funding in principle.
Now one comment I also want to make: the BDCs are also very well capitalized, like private credit funds. Typical BDCs finance with at least 50% of equity. So because of this ability to exit these funds, this is exactly what generates a lot of maybe bad attention to the sector, because you had recently, maybe given that these withdrawals are capped, I wouldn't call them a run on these funds, but we could call them a jog, you know, a jogging on these funds. There is essentially withdrawal requests that well surpass the caps at which you can get the money out from these funds.
And I have a recent research that actually gives a little bit of economics of what's going on. And we identify one of the main problems is that net asset values, the values at which you can exit, are overstated relative to the market value of private BDCs. I created with my colleagues a synthetic price index, so we can value any private BDC using the composition of the portfolios and market prices of public BDCs. And essentially as of the middle of this year, private BDCs, the ones that are not traded, the NAVs were about 20% overvalued relative to the implied market prices. And that generated a redemption wave on them. And what's interesting, we have data on who was redeeming, it was more institutional sophisticated investors. So there's a bit of a transfer of wealth. If you're smart and you're really worried about credit quality, you might try to exit this potentially inflated asset value at the expense of those that remain.
So let me give you, let me conclude with some implications for monetary policy and regulation. So I don't see as of now private credit to be systemically concerning. It's still a relatively small share of overall credit market. There is a very good equity funding, so there is very little concern to the creditors. Banks extended about half a trillion dollars of direct funding to private credit, but they are in a very senior position, very well protected by this very hefty equity cushions both on the private equity debt part and the BDC part.
So what are the potential concerns? Well, one is the LPs and equity investors that invest in these funds. I've looked on it, it seems to be pretty dispersed, and I don't see yet any systemic problem from that, but it's some area worth watching. Maybe there's some life insurance companies or pension funds that are a little bit more loaded in certain segments. A concern maybe more not for monetary policy, but you know, if you worry about growth, employment, these mid-level firms are quite dependent on private credit, and if there is a stress in the sector, maybe partly driven by these negative events in the non-listed BDCs, you could imagine a suspension of ability to refinance your loans and obtain new loans, and that could potentially affect the ability of these firms to get funding, with more work on how substitutable this is, and more for financial regulators than for monetary policy.
If we think about semi-liquid vehicles, there are these issues of how NAVs are determined, are the sophisticated investors maybe taking advantage of some less sophisticated ones, and that it could create potential instability. In the last minute, let me just conclude by saying what about traditional banks when we kind of confront them with private credit. So if I was a monetary policy maker, I would still be more worried about the traditional banking sector. You know, they are thinly capitalized, obviously the assets are 70 times larger than that of private credit, and we know that because of a significant increase in interest rates as of 2023, the bank had a trillion-dollar loss of asset value.
And I'll just conclude: this is an independent stress test of the traditional banking sector, this is not private credit, that we did in 2023, and essentially it shows the following: as long as interest rates remain low, the banks can survive even a very severe credit distress. However, as the 10-year yield starts moving more and more, right, in 2023 the level was 5%, now it's probably closer to 6%, there is a number of banks that could get in trouble because, you know, they still have a duration exposure. And last comment I will make: what the Fed considers a best scenario, the worst test, is significant credit distress but zero interest rates. And we are advocating that we should start considering a stagflation scenario as one important source of risk for banks, but not for private credit themselves. Thank you very much.
Thank you, and thank you Mickey and panelists for joining me today. I think this is actually a fascinating discussion, and I guess my job is to give you the perspective as someone who's in the weeds literally doing private credit every single day. Churchill is the private credit manager for TIAA and over 6,000 investors globally, with over $66 billion in committed capital under management in the United States. We have another $34 billion under management in Europe. Churchill, during the first half of 2026, was named by Octus and ranked as the number one most active direct lender in the United States. So we're very busy.
At Churchill, we've spent the past 20 years providing capital to businesses in the US middle market. The middle market, interestingly, accounts for roughly one-third of private sector GDP, and if it were its own country, it would be one of the top five largest economies in the world. Within that market, we specialize in what we call the core middle market: companies with anywhere from 10 to 100 million in EBITDA.
It's been a year since Jamie Dimon warned about cockroaches in private credit, and the market's been under the spotlight ever since. That scrutiny is healthy, but it's just as important to look past the headlines and understand how private credit is structured, how risk varies across segments, and the role it plays in financing economic growth. Today, I'd like to share what we're seeing across private credit, and how we focus on the core middle market, and what 20 years in this business and nearly a hundred billion of investment has taught us about the market and the role in the broader economy.
In our view, systemic risk can be analyzed or channeled in three ways: size, leverage, and connections to the broader banking system. First, by size. Private credit has grown quickly, but it's still only about 3% of household and business debt. By comparison, in 2006, on the eve of the financial crisis, mortgages made up about 60%.
Second, by leverage. Heading into the crisis, banks were levered 30 to 1. I think you pointed out a similar statistic. So even a small drop in asset values could wipe out their equity. Direct lenders generally operate with far less leverage than banks, typically below one times leverage. That's partly because one source of private credit, business development companies or BDCs, face regulatory limits on how much they can borrow, and leveraged funds typically operate with about one times leverage on average.
The third channel is exposure to the banking system more broadly. Banks may not originate these loans directly, but they do finance direct lenders through what's called back leverage. Back leverage drew scrutiny during the first-quarter earnings calls, but the concern has faded, and for good reason. It's only about 2% of total bank assets.
Structure matters, too. As you pointed out, Tomas, which was a great point, these loans sit at the top of the capital stacks. Banks lend typically 60 to 70 cents per dollar of collateral. That's equivalent to double-A or AAA risk. So equity holders would have to take large losses before banks lost a single dollar. And I love your chart that you showed with the leverage sitting on top of significant equity within those funds. On top of that, banks hold roughly five times as much tier-one capital as back-leverage exposure. So even if stress builds, which isn't our base case, banks can absorb the losses.
Outside the banking system, exposure to private credit is highly diversified and spread across sovereign wealth funds, pensions, endowments, and insurers, with no significant concentration in any single investor group. Insurers have drawn more attention given rapid growth, but their allocations to sub-investment-grade private credit remain small. And on their first-quarter earnings calls, they reported no significant impairments across their portfolios, a trend that we expect to continue.
An underappreciated feature of private credit is it can act as a countercyclical buffer when the credit cycle turns. We've seen this play out before, most recently after Silicon Valley Bank collapsed in 2023. Financial conditions tightened sharply, particularly for smaller firms. In fact, it was the largest tightening in credit conditions we've seen outside of a formal recession. And yet, the country never tipped into one. As banks pulled back, private credit managers stepped in, helping to cushion what otherwise could have been a far more disruptive blow. Because direct lenders are funded largely with long-term committed capital from multiple funding sources, versus banks that rely heavily on deposits, they can keep lending through the cycle even when other lenders are more concentrated.
So, excuse me. So, why the focus on private credit today? Likely because of its rapid growth as institutions and individual investors have increased allocations to the asset class. Direct lending alone is about a $1.8 trillion market, and the broader asset class is expected to reach $4.6 trillion by 2030. That growth spans well beyond direct lending, from asset-based lending to distressed debt. Still, context matters. At that size, private credit is roughly on par with the leveraged loan and high-yield bond markets, which generally haven't been seen as sources of systemic risk.
The other driver: the shift from banks to direct lenders. Post-crisis regulatory reform pushed this lending off banks' balance sheets, and today direct lenders make up roughly 90% of middle-market loans. In our view, that shift has strengthened the system. Banks today are largely in the moving business. They originate loans to sell them off. Direct lenders hold what they originate, the storage business. So they live with the results and have every reason to underwrite carefully.
What also gets missed is that direct lending itself is segmented. There's the lower middle market, the core middle market where we operate, and the upper middle market. As mentioned earlier, we focus on companies with 10 to 100 million in EBITDA and credit facilities of 100 to 500 million in size. They tend to be resilient, service-oriented businesses, and many are backed by private equity sponsors. They're large enough to be regional leaders, but small enough for lenders to keep real negotiating power. This, in our view, has translated into generally lower leverage, substantial sponsor equity, and stronger covenants and attractive yields.
By contrast, a wave of new, primarily retail capital spilled into the upper middle market and broadly syndicated markets, that brought more competition with BSL loans with the banks, and generally higher leverage, lighter covenants, and more exposure to public market swings. Core middle market deals are negotiated directly, which helps insulate them from some of these dynamics. So what happened with redemptions?
So market segmentation also helps explain this year's redemptions in non-traded BDCs. Managers that raised heavily through the broader wealth channels have seen the highest redemption rates. For example, in our private BDC, our redemption rates were 1% last quarter, and they've never been remotely close to 5%. Many of the managers that took in large amounts of retail money that needed to be deployed quickly, and I think this is an important distinction: institutional funds are typically set up as drawdown vehicles. So you draw down the capital as you make the investments. Retail funds are under pressure to deploy capital immediately. So where did they go? They went to the upper middle market and they went to the syndicated loan market to deploy that capital, not exclusively into the core middle market where we operate, and they felt that pressure to deploy. This capital was invested significantly in these markets.
Institutional managers like ourselves, and we are 95% institutional, they're in a different spot. They call capital when they find the right deals, so they can be more patient about deployments. Redemptions appear to have peaked. New requests have dropped more sharply, though clearing the backlog may take a few more quarters. In our view, it's worth noting in perspective: non-traded BDCs are only about 9% of the overall private credit market.
Gates limit how much investors can withdraw in a given period. They don't eliminate liquidity risk, but they're essential for funds that hold illiquid loans. We see them as a feature, not a bug, for two reasons. First, they better match assets and liabilities. Many of these funds hold liquidity sleeves invested in cash and broadly syndicated loans. Plus, they receive steady repayments, and that's proven sufficient to meet current redemptions. Second, they protect remaining investors from forced asset sales. A recent IMF study found that these dynamics have generally allowed BDCs to meet redemptions without forced sales, even in a tougher credit environment.
What headlines miss today is that the fundamentals in private credit remain solid. We believe the credit cycle is actually normalizing, not deteriorating. That's why defaults haven't led to higher losses. Historically, middle-market direct lending has had lower losses than high yield, and covenants are a big reason why. Interest coverage, a borrower's ability to pay us interest, is roughly at its historical average, even as the Fed starts to raise rates again. At Churchill, interest coverage across our senior lending...
platform has stayed above the market even when rates were higher. Pick or payment in kind is when interest is added to the loan balance instead of paid in cash. It's often viewed as an indicator of stress across a loan portfolio. It is also peak and is returning to its longer-term average.
So here's a snapshot of what we are seeing in our portfolio. It reflects very healthy fundamentals. We have about a $40 billion US middle market senior loan portfolio with nearly 400 US middle market companies in that portfolio.
An increase in revolver utilization can be an early indicator of stress in the portfolio. Our revolver usage has not seen any material uptick and has remained in a range of roughly 20 to 25% since the fourth quarter of 2024. Our net senior and total leverage have also held steady and conservative across our senior lending platform, both point to no deterioration in fundamental credit. In addition, nearly 70% of our senior loan borrowers' EBITDA year-over-year has increased quarter-over-quarter, while our pick usage also remains well below market and barely changed in the last two years.
All of that sits within a well-diversified portfolio spread across industries, sponsors, and position sizes. Interesting point, and I think this is we're fairly typical of the largest mid-market direct lenders. Our average position size is just 31 basis points. So, hundreds of borrowers diversified across an incredibly broad portfolio. Compare that to a bank that takes the entire or used to take its entire loan onto its balance sheet. So if a JP Morgan makes a mid-market loan 15, 20 years ago when they were doing more mid-market lending, they could put that entire loan on their balance sheet, two, three, four, $500 million. Today when we make a mid-market loan, the exact number today is it's spread across 46 separate underlying investment vehicles. Much more diversification across investment vehicles and across underlying investors. So the institutionalization of private credit has had an incredibly positive impact from a diversification standpoint.
So what does this mean for returns overall? Fundamentals have delivered consistent returns for private credit over the last 20 years. The asset class has only had one negative calendar year, 2008, when it fell about 6% including unrealized losses. That year, high yield fell about 26% and leverage loans even more, about 29%. That resilience has shown up again in more recent stress periods. The 2018 trade war, COVID, and the 22 to 23 inflation and rate shock. Each time private credit held up as well or better than public credit. That consistency shows the asset class often behaves differently than the headlines suggest with a lot less underlying volatility.
And I should point out that of the loans in our portfolio, as I mentioned, nearly 400 individual loans, every single loan in our portfolio is valued by a third-party valuation firm every single quarter. I know I've heard lots about the mystery of private credit and the opaqueness of the market. I would argue strenuously that is actually not the case. Every single loan, detailed financial information provided to those valuation firms and those loans are obviously reported to our private investors as well as our public BDCs.
So given size, leverage and connections to the banking system, we believe private credit does not pose a systemic risk. But not every manager will perform equally. Only 4% of direct lending firms were around before the financial crisis. Happy to point out that both ourselves and Oaktree were. So you have two of them out of the 4% here with you today, both founders. So many managers and borrowers have never been tested through a full credit cycle. As this cycle matures, the gap between managers will widen. We will see more dispersion among managers. The real question isn't whether private credit is resilient. It's which managers are. And here's what to look for in our view. First, capital structure. Is the firm's investor base stable? Is fund level leverage modest and do will liquidity terms match the underlying loans? And Thomas, you made a great point about the maturity of the underlying loans against the backdrop of the funding sources. That's exactly what was happening in the BDCs and the private BDC redemption issues. You had funding requests coming in from investors against long-dated private market assets. But if your funding base is primarily institutions, it's sovereign wealth funds, it's insurance companies that are committing to long-term investment funds, you don't have that kind of mismatch in funding. So it's important to look at the underlying investment base, investor base of the manager and the fund investment base. Second, market segment, borrower size, sector exposure, and concentration all shape risk and return. We think the core middle market is a better spot to be and our fundamental approach is that we like larger companies but not so large to have graduated into or to compete with the syndicated loan market. And if you look at what's played out, the investors that have had to invest heavily in the syndicated loan market are exposing investors who feel that they're getting private markets exposure to much more public market volatility. So we like to stick to the core middle market where our transactions are privately negotiated and funded. Third, sourcing. Differentiated sourcing in our view is critical. Fourth, alignment. Fifth, scale of managers. Track record is important in portfolio construction. So let me close where I started. A year ago, the question was whether there were cockroaches in private credit. Today, defaults haven't led to significant losses. Leverage remains modest and bank exposure is small relative to capital. When credit tightened, private credit didn't amplify the shock. It absorbed it. Private credit is not where the next crisis starts. Done right, it helps to contain it. Thank you, and I look forward to the discussion.
Thank you for those outstanding presentations. What I'd like to do is ask one question to each of the panelists, have them answer it in succession and then open it up. Howard, you've given us a good sense for the history of credit and one of the things that you've mentioned is that the decline in interest rates of 2,000 basis points over the recent 40 years is one of the most impactful events in the financial world in the last half century. As interest rates begin to move in the opposite direction, what structural shifts do you expect for the non-bank credit industry? And Tomas, you offered a terrific framework contributing to these forces and how to think about systemic risk. When does something become a systemic problem? And you pointed at two things, namely losses on assets and fragile funding. And what would be helpful to know is in this space of fragile funding, where does interconnectedness fit in, claims and cross-fertilization among these various different holdings? Where does that fit in? And if you want to add on demand for credit from AI, Ken, 4%, that's pretty good. So congratulations to you and Howard. 4% of the lending firms were around given your perch and experience. Would you speak a little bit about where we are in the full cycle and how you see there's been an interesting transition recently, right? There's been a lot of small and medium-sized financing. There's been a transition recently from private credit engaging in that kind of financing to focusing on these mega deals with many players and a recent example is the $63 billion Broadcom Anthropic deal. So why don't we start, Howard, with you please.
I think you asked that question about 10 minutes ago. You have to remind me now.
2,000 basis points and the change.
So the point is I wrote a memo in December of 22 called Sea Change because I thought that it did represent a sea change and I mentioned that I had a personal loan outstanding from a bank in 1980 and I got a slip of paper in the mail and it said the rate on your loan is now 22 and a quarter and 40 years later I was able to borrow at two and a quarter and that 2,000 basis point decline, as I go further than Larry, I think it was the most important single event in the financial world in the last half century and declining interest rates make assets more valuable and reduce the cost of carry. So if you have a strategy of buying assets with borrowed money, you got a double bonanza. It happens that the private equity industry was invented during that period really in the 80s. And it got a huge tailwind. I describe it as like being on the moving walkway at the airport. You know, you get on, you walk at your normal speed, you make great progress over the ground, then you say, 'Boy, I'm fit.' But you know, I think the inflow of capital reached the zenith around 2021-22.
The private equity firms had a lot of investing to do. They needed a lot of capital. The private credit firms stepped up in large extent. The banks of course were limited as we have said. And a lot of that lending in particular went into software. But that's another story. But the point is that the deals that were done in 2021-22, you know, if you could buy companies that grow let's say 8, 12% a year with money you could borrow at six, you would take a lot of that money and you know the more you could borrow the more you could amp up your ROE. And so firms did borrow a lot and did do a lot of deals. The only trouble is that the deals did not anticipate an increase of five or 600 basis points in rates. And the debt that was issued in 20 or 21 or 22 was coming due in 27, 28.
And the companies are not as profitable as people thought they would be because this debt service is higher. They're not as refinancable. If they refinance, they'll have to pay 11% rather than six, which is a big difference. Some of them won't refinance because they can't carry that kind of interest rate. So, I think that you know we all describe good performance for the private credit industry and the direct lending industry so far. But as Buffett says, the tide hasn't gone out and we haven't been tested by a difficult economy, by a meltdown of companies, by a big increase in the default rates. And one day we will. So the point is, and as Ken says, the people who made better lending decisions will handle it much better and largely sail through. But there will be a moment of testing coming and we will see dislocation and we will see difficulty in refinancing. We'll see an increase in distress. We'll see restructurings. The question of when the private lender recognizes a default rather than kick the can down the road will be in play. You'll hear about that etc. But the point is we were in easy times. You talk about the credit cycle, easy times, hard times. What does it mean that the credit cycle? What it means is when the credit cycle is here, it's easy to borrow and when it's down here, it's hard to borrow. That's the whole change in the credit cycle. And we've gone through 17 and a half years from the low in the stock market, March of 09, when it was easy to borrow and the indebted company had no problems. You bet that comes to you refinance. Big deal. It's not going to be so easy in the next few years. I think so.
Let me quickly comment on two things. The interconnectedness and AI. So I don't see as much a traditional bank stability concern of interconnectedness. Their direct loans to private credit, that half a trillion, is very senior. They also provide loans to LPs that invest in private credit funds sometimes to help them manage their liquidity but it's also not very big. So you could imagine a situation when a private credit fund defaults and if they're really very severe some of the bank lenders could start seeing the pressure and also maybe some of the LPs but I generally don't see this as systemic.
Now one concern as was discussed in the panel is the mid-market firms if there was a severe credit stress and they lose funding in terms of refinancing the debts if there is not an easily available alternative and as Ken said, the private credit has played the role of stabilizer actually in terms of banking stress that could be also an issue. Let me comment also briefly on AI. As you know there's a lot of debate about the massive scale of the AI financing boom. As you may know, we have about currently 50 gigawatts of data center capacity, compute capacity. There's about 100 gigawatts being built and another maybe 200 gigawatts in plans. These are massive investment plans. If you look on hyperscalers' balance sheets, they're off-balance sheet commitments. So the leases with data centers that are being built or will be built, signed the leases on compute power, the purchases of chips and leases on energy getting from suppliers. This is already $3 trillion. This is the declared financing need based on their balance sheets. What they did initially, they funded the AI infrastructure boom with their own cash. They currently run out of cash so they went to the capital markets and where we are right now in terms of where's the credit financing, I just had a photo look on this and exposures. So currently we have about 1 trillion outstanding of which half is financed by corporate debt markets issued by hyperscalers and other up to 200 billion are construction loans done by banks but many of them were securitized and the rest is private placement debt and some CMBS, ABS, not much direct private credit exposure if I'm having private credit fund managers for the other vehicles are sometimes involved in structuring these deals in SPVs and selling things to other folks. So as of now the exposure of for example life insurance companies is $80 billion, it's relatively small relative to the $6 trillion balance sheet but for some of them is already more than 50% of the regulatory capital that's exposure to AI. Banks have still modest exposure the direct to financing. The question how the next $3 trillion will be financed and if it even happens because as we don't see much evidence on monetizability, if there is no more evidence on monetizability or profitability of these investments I would expect a lot of these things eventually won't be built which is good, you know we should essentially progress as we get on more information but it's a big issue for regulators to watch that three trillion investment plans how they will be financed and how it will evolve.
Yeah. So, I think if I recall, your question was about where we are in the credit cycle or are we in a cycle? Look, I think you know there are folks that are far more listened to than I am that predict credit cycles and when we're in a credit cycle or whether we're entering one or we've exited one. I think the reality is, you know, I'm here to report that we probably have one of the best views in terms of what we're seeing in our portfolio and what we are seeing is no evidence of fundamental credit deterioration. If you look at our non-accrual rates, if you look at our pick rates, as I alluded to, if you look at drawdowns on revolvers, which is an early sign of companies having stress and needing more capital. By the way, before the COVID crisis, we saw our revolvers, which normalized around 20 to 25% historically, went to 80% drawn down before COVID. So, you know, you can see evidence of stress in our portfolio for sure. And we're not seeing any evidence of that today. So, I'm not sure when we would see the cycle, but we certainly haven't seen it in our portfolio. I do think we are seeing a bit of a normalization of credit conditions rather than any type of broad-based deterioration as a broad matter in the market today. Non-accruals overall, and you can see this in the public BDCs' report, are up a bit but if you look at those levels, one and a half to 2% non-accruals, those are actually very much in line with historical averages so no evidence there, back to their long-term average. I think we saw a period where actually the numbers were extraordinarily low and now they're moving back to kind of where they've been historically. We do think there are some signs of stress overall, you know, just looking at other lenders in the marketplace. It looks like 2021 vintage loans and certain software-based business models are seeing some stress. There was a type of loan done during the boom years of software if you will called an ARR loan which was basically a revenue-based loan. Right? The lenders. We never did one to be honest, but they were done quite actively. And the theory was that a private equity firm would put up a lot of equity and the company was growing at such a rapid rate that they could always just turn off their growth to pay their interest expense. They could just slow growth in order to pay interest. We never believed in the fundamentals of ARRs and we never did one. And I think you're starting to see some of those loans create problems particularly where there has been a significant deterioration over a short period of time. Overall though we think we're in quite good shape. If you look at our portfolio overall, the average interest coverage ratio, meaning cash flow to interest coverage, 300 plus companies, we're looking today at around two and a half times interest coverage on average overall. Still pretty good. If you look at senior leverage, which is our attachment point, we're sitting at around four and a half times senior leverage on companies who were typically acquired for over 10 times. So you're looking at numbers in the range of 60% underlying equity in those businesses. So not seeing any direct evidence of being in a cycle, if you will, or potentially seeing long-term credit issues. I would say over a 20-year period, if you look at our average annual loss rate over 20 years, it's been five to six basis points per year on average. So I think for credit investors that manage with a highly diversified portfolio, high levels of selectivity and stay out of areas that could be potentially viewed as either more aggressively leveraged or in riskier sectors, I think you'll see direct lenders fare quite well. So we're convinced with great conviction on the core middle market. And we feel pretty good about where we sit today.
Fantastic. Questions.
John,
Thank you for the three great presentations. My question is for Ken Cassell. If PC funds rely a lot on equity funding, what particular bank regulations do you see as driving small business lending from banks to PC funds?
Oh, lending actually into the funds and how is that limited or regulated? So what you know it's very interesting if you look at, and I know Howard has had this front row seat as well, but if you look at where we've come over the last 20, 25 years, right, but banks have moved into the moving business. Why did they, why have they moved into that area? Well, the capital charges, if you make a, if you underwrite a $500 million loan or a billion dollar loan or $2 billion loan and you distribute it to investors and you hold zero of the loan, your capital charge is zero. Right? So your capital requirement is, your return is infinite. Right? So all the fees you earn against zero capital. It's a great economic model. What's limiting banks and lending to funds is that as long as they lend within the range that they are currently lending, as I mentioned earlier, banks lending typically double AAA attachment points to funds. So what does that mean? The capital requirements for banks to make those loans, fund level leverage, whether they lend one to one leverage in a BDC, whether they lend to our levered funds, typically one to two times would be the max leverage we typically would use. Banks provide that capital to us very readily, they love that business, why? It's effectively an investment grade loan for them, right? So they can lend to us, we have a diversified portfolio and they can call that loan and it is rightly an investment grade attachment point. Interestingly, if you look at CLO leverage because that's a pretty good proxy for how strong the credit quality of these loans are, there has never been a loss on a AAA rated loan to a diversified portfolio of loans. Right? So, I think we're on pretty safe ground with banks lending double AAA to funds, BDCs, private funds, but they're clearly looking to not just the equity in our fund, which as Thomas pointed out is significant, but then there's the equity in the companies, right? So, if our companies have on average 65% equity, then our funds have another 35 or 40% underlying equity, I think you're in pretty safe ground as a bank. And the model, the numbers work from a capital perspective.
