The Private Credit Screen

Alex Rivera, Wealth Architect at The Wealth Grid

Deep Dive | August 14, 2026 | The Wealth Grid

For about thirty years, private credit was a country club.

Direct loans to mid sized companies, real estate debt, specialty finance. Minimums in the millions, ten year lockups, and a guest list of pensions and endowments. If you were a normal person with a brokerage account you were not invited, and the industry was comfortable calling that a feature.

That is over. In 2026 you can get in with minimums as low as a few thousand dollars. Blackstone, Apollo, KKR, Ares and Blue Owl have all built retail distribution machines. Capital Group and KKR launched public private credit funds in April that pulled in over a hundred million dollars in three months. Hamilton Lane cleared the SEC with its first interval fund in March. Six more filed draft registrations in a single month this spring.

And the prize they are really circling is the roughly fourteen trillion dollars sitting in American 401k plans, which regulators have been moving to open up. So it is coming to a menu near you, probably inside eighteen months, with a deck showing something like nine point four percent annualized through multiple credit cycles.

Here is the part the deck will not lead with. Earlier this year, retail investors in these structures tried to leave and could not. In June, Partners Group gated a fund. Redemption requests across the space blew through the quarterly limits. Blackstone's flagship raised its repurchase ceiling to seven percent in the first quarter just to work through the backlog.

Two stress tests, one year, and we are about to hand the product to a hundred million retirement savers.

I am not here to tell you private credit is bad. Some of it is genuinely good. I am also not your advisor, and nothing here is a recommendation on any particular fund. What I am handing you is a screen. Seven questions. If a fund cannot answer all seven cleanly, you have learned something important for free.

Let's build it.

Three structures dominate retail private credit, and they behave completely differently under stress. People conflate them constantly.

Listed BDCs trade on an exchange like a stock. You can sell any day the market is open. The catch is that you sell at whatever price the market offers, which in a panic can be well below the stated net asset value. Your liquidity is real. Your price is not guaranteed.

Non traded BDCs and interval funds are the ones flooding retail right now. An interval fund is a closed end fund registered under the Investment Company Act of 1940. It does not trade on an exchange. Instead it makes formal repurchase offers at set intervals, almost always quarterly, and typically offers to buy back up to five percent of net asset value.

Write this next sentence on something. That five percent is not a floor. It is a ceiling set by SEC rules.

The fund is not promising to buy back five percent of your money every quarter. It is promising to consider offers up to that amount. If everyone wants out at once, the offers get prorated. You request your full position, you get a fraction, and you go back in line next quarter. That is the gate, and it is not a malfunction. It is the design working exactly as written. Tender offer funds work similarly, with even more discretion over whether to hold a window at all.

The industry sells all of this as "semi liquid," which is a marketing term, not a financial one. The honest translation is: liquid when nobody needs liquidity.

One upside worth naming, because I want this to be fair. Interval funds are registered vehicles. They issue a 1099 instead of a K-1, they are open to non accredited investors, and they carry disclosure obligations private partnerships do not. That is a real improvement over the old structures. It is just not the same thing as being able to get your money.

Do not accept "quarterly liquidity." Push until you get the mechanics.

When is the window, and what is the notice deadline? What percentage of NAV has the board actually approved in each of the last eight quarters, not what is permitted? Has any window been prorated, and by how much? Do I get the NAV on the request date or the fulfillment date?

That last one matters more than people expect. If the fund marks down in between, you eat the markdown.

The tell is a manager who has never had a prorated window and treats the question as hypothetical. In a market where multiple large funds have gated in the last fourteen months, that is either very new or not straight with you.

Public bonds have a price because strangers trade them all day. Private loans do not. They are marked to model, using a valuation process the manager runs and a board reviews.

This produces the most seductive property in the asset class: the returns look smooth. Beautifully, impossibly smooth, month after month, while public credit chops around.

That smoothness is not stability. It is a measurement artifact. The volatility did not disappear, it just did not get printed. People confuse the two, size as if they own something safe, and find out in a real credit cycle that what they owned was never low risk, only low resolution.

So ask: who values it, how often, is a third party involved, and how many loans are on non accrual, meaning the borrower stopped paying and the fund stopped booking income. That non accrual number is the closest thing to an honest signal you will get, and its direction over four quarters tells you more than any performance chart.

Not the management fee. The stack. These structures are layered in a way that makes the headline number nearly meaningless. Add up:

Run a real example. Say the loan book yields eleven percent gross. A management fee of one and a quarter on gross assets with modest leverage is closer to one and three quarters against your net. Add servicing. Add an incentive fee above the hurdle. Add the fund's borrowing cost.

You can start at eleven and hand the investor the high sevens without anything going wrong. Nothing defaulted. That is just the toll booth.

Now compare. As I write this the thirty year Treasury pays above five point two percent for zero credit risk and daily liquidity. So the question is not "is eight percent good." It is "am I paid enough over a risk free five to accept credit risk, leverage, model based pricing and a gate." Sometimes yes. Often not. You cannot answer it without the fee stack in dollars.

You are not just buying loans. You are buying loans bought with borrowed money.

BDCs can generally run up to about two to one debt to equity. Leverage is what turns senior loans yielding around nine into a distribution yield that prints double digits and looks irresistible on a fact sheet.

It works in reverse just as fast. A portfolio at two to one leverage that takes a ten percent hit on the loan book does not lose ten percent of your equity. It loses roughly thirty.

Ask for current debt to equity, the covenants on the fund's own credit facilities, and what happens if asset values fall enough to trip them. A fund forced to sell illiquid loans to satisfy its own lenders is exactly when your gate slams shut, because its creditors get paid before you get your redemption.

Almost no retail investor knows to ask this, and it is the best early warning signal in the asset class.

PIK means payment in kind. Instead of sending cash interest, the borrower adds it to the principal. The loan grows. The fund books the income. Everyone reports a great quarter.

But no cash moved. And you, the investor, may owe tax on income you never received.

A small amount of PIK is normal and can be a legitimate structure. A rising PIK percentage across a portfolio usually means something else: borrowers are struggling, and rather than mark loans as troubled, the manager is letting them defer. It is the accounting equivalent of a company paying its rent with an IOU while telling shareholders revenue is up.

Ask for PIK income as a share of total investment income, for the last eight quarters, as a series. If that line is climbing steadily, you have found the story the performance chart is hiding.

There is a real thing called an illiquidity premium. If you genuinely cannot touch your money for years, you should earn more than someone who can touch theirs daily. That is a fair trade and it is the intellectual foundation of the entire asset class.

The trouble is that as trillions piled in chasing that premium, competition compressed spreads. More lenders, same borrowers, looser terms, thinner covenants. In some corners the premium has been bid down to almost nothing, which means investors absorb all of the illiquidity and get paid very little for it.

So compare against a liquid alternative. A broadly syndicated loan fund or a public high yield fund holds similar credit risk and trades every day. If the private fund's net yield after the full fee stack is only a point above the liquid version, you are accepting a gate, model based pricing and leverage for one point. You can see that in ten minutes with a calculator.

Everything above is documents. This one is behavior, and behavior is the only real test.

This market just got its first genuine retail stress events. Ask what this manager did in them. Honor full requests, prorate, or suspend? Raise the ceiling to clear a backlog the way Blackstone's flagship did in the first quarter, or hold at five and let the queue build? And how did they communicate while it was happening?

A manager who gated and explained it clearly and early is not disqualified. Gates exist to prevent forced selling that would hurt everyone who stayed. What should disqualify a manager is having gated and being vague about it now.

Track records in a bull market are marketing. Behavior in a queue is information.

The documents that answer these seven questions exist. They are in the prospectus, the N-2 registration statement, and the quarterly and annual reports. They are also written to be technically complete and practically unreadable, which is not entirely an accident.

I do not read them cover to cover anymore. I run them through Galaxy.ai and ask for exactly the seven things above, one at a time, with a page citation for each so I can verify the ones that matter. Putting the same extraction question to two or three models is a real check here, because if they disagree about what the repurchase terms say, that usually means the language is genuinely ambiguous, and ambiguity in a liquidity provision is itself the finding. The model is doing retrieval, not judgment. I still make the call.

Then I wire the monitoring, because a screen you run once at purchase is worth little in an asset class where risk builds slowly. I keep a Make.com scenario watching the filings feed for anything I hold. Every new quarterly report or repurchase notice, it pulls the non accrual rate, the PIK share and the leverage ratio into a running sheet and pings me if any of the three moves past a threshold. Three numbers, tracked forever. That is a two hour build that replaces a quarterly chore nobody actually does.

Even if a fund passes all seven questions, there is one more discipline, and it is the one that actually protects you.

Size the position as though the gate is closed the entire time you own it.

Not as a worst case. As the base case. Assume you cannot get this money for at least a year, and specifically that you cannot get it during the exact period you most want it, because gates close when credit is stressed, which is the same moment your job or the rest of your portfolio is under pressure. Illiquid assets correlate with your own bad luck. That is the whole risk in one sentence.

Which means: emergency fund untouched and fully liquid. Nothing you need for a known expense in the next three to five years. And a total allocation small enough that a gate is an annoyance rather than an emergency. For most people that is a single digit slice of the portfolio, not a core holding, no matter how good the deck looks.

I built the seven questions into a one page scorecard you can take into any meeting or run against any prospectus. You get all seven questions with the follow up that stops a vague answer, the fee stack worksheet that shows your real net yield against a liquid comparison, the four quarter tracking table for non accrual, PIK share and leverage, and the sizing rule as a simple test you can apply before committing a dollar.

Want it? Reply with the word SCREEN and I will send you The Private Credit Screen, free. Just reply SCREEN and it lands in your inbox.

This is what the paid community is for. When a product like this lands in someone's plan menu, the useful thing is not an opinion. It is a room where people paste in the actual fund documents and work the screen together.

Inside you get my live positioning notes ahead of events like the September meeting, the build sessions where we wire up Make scenarios together, and a group of people who read the prospectus rather than the fact sheet.

Membership runs 29 dollars a month. The first 100 founding members can lock a lifetime seat for a one time 499 dollars, which means every system we ever build, forever, with no monthly bill again. Those seats do not come back once they are gone. Reply with the word GRID and I will send you the founding member door.

Private credit is not a scam. It is a real asset class doing real economic work, and some of these managers are excellent.

But an asset class built for institutions with thirty year horizons and dedicated analysts is being repackaged for people who will click it on a retirement menu next to an S and P index fund, with no real understanding that one of those can be sold on Tuesday and the other cannot.

The seven questions are your defense. They take about an hour per fund, and most products fail in the first ten minutes.

Ask the questions. Make them show you the queue.

Sunday we are stepping back from the mechanics to talk about something that has been sitting underneath all three of this week's issues, which is the specific place where smart people lose the most money. It is not the stuff they know nothing about.

Alex Rivera, Wealth Architect at The Wealth Grid

Wealth is a system, not a guess.