Peter DeCaprio Insights

Where Venture Debt Fits in a Lower Middle Market Capital Stack

Peter DeCaprio — Mon, 14 Sep 2026 00:00:00 GMT

Start with the problem, because the instrument only makes sense once the problem is clear.

A company has something customers pay for. Revenue arrives every month and most of it comes back the next month. The founders have a plan that needs another year or two of spending before the business turns a profit, and they can show you where that spending goes. They take this to their bank. The bank reads the same numbers and sees losses, thin hard collateral, and a borrower whose value sits in code, contracts and people rather than in trucks and buildings. However warm the relationship, the answer is some version of not yet.

So the founders turn to their equity investors, who are willing, but at a price that resets the ownership of the company and on terms that follow every later round. The distance between what a bank will underwrite and what new equity costs is where venture debt lives. It is not a better bank loan and it is not cheaper equity. It is a different instrument with its own logic, easiest to see when the whole stack is laid out at once.

The stack, top to bottom

Layer	What it costs the company	What it demands	When it fits
Senior secured bank debt	The cheapest money available	Hard collateral, positive cash flow, tight covenants and a first claim on everything	A profitable business with assets a bank could actually sell
Venture debt	More than a bank, far less than equity, plus a small warrant	A credible equity sponsor, a clear use of proceeds, a lien on the business and real information rights	A growing company that has proved demand but not yet proved profit
Subordinated and mezzanine debt	A high coupon, often partly paid in kind	Patience on repayment, a claim behind the senior lender and a slice of the upside	A business with steady cash flow that has used up its senior capacity
Preferred equity	A share of ownership with a preference on the way out	Board rights, protective provisions and a liquidation preference ahead of common	A company raising real growth capital from investors who want downside protection
Common equity	The most expensive capital there is, because it is permanent and last in line	Belief in the plan, and control of whatever remains after everyone else is paid	Founding, and the moments when the future is genuinely open

Read down the table and the pattern shows itself.

Each layer trades price for protection.

The senior lender is cheap because it can take the assets. Common equity is dear because it gets whatever is left. Venture debt sits in the middle. It is cheaper than any form of equity precisely because it is repaid first, and dearer than a bank because it lends against a plan rather than a warehouse. The warrant is the lender admitting that some of the risk it carries is equity risk, and asking to be paid for that part in equity.

Three situations where it earns its place

Bridging to a milestone

The cleanest use is a bridge. The company is a few quarters from an event that will change how it is valued: a product release, a contract that has been signed but not yet billed, a break-even month that the model shows clearly. Raising equity before that event means selling ownership at the old price. A loan sized to reach the milestone, with a little cushion, lets the company arrive at the next round on its own terms. If the milestone slips, the loan is still there and the runway is shorter.

Extending runway without a down round

The second use is defensive. Markets turn, and a company that was worth one figure to investors last year is worth less this year through no fault of its own. Raising now would reset the price for every holder, including employees. A modest facility can carry the business through the trough so that the next equity conversation happens in a better market. This is the use I am most careful with, because it works only when the business is fundamentally sound and the trough is a market event rather than a company event. Debt does not fix a broken model. It gives a broken model more time to prove it is broken.

Funding a specific asset

The third use is the most conventional. The company needs equipment, a fleet, a build-out, the purchase of a smaller competitor, something with a definable cost and a definable return. Nobody should give up permanent ownership to buy something that will pay for itself within a few years. A term loan matched to the life of the asset, secured by the asset where possible, is the right tool.

What I ask before I lend

Who is the equity sponsor, how much have they put in, and would they put in more if the plan slipped? A lender in this layer is underwriting the sponsor's next decision.
Where exactly does the money go, month by month, and what does the business look like at the end of it?
What is the real gross margin on the revenue, after the costs the company prefers to show further down the page?
What is the downside case, and does the loan still get repaid in it? Not the model the company sends. The one I build.
If the plan stumbles, does the structure give an early signal, in the form of a covenant, a reporting requirement or a reserve, before the money is gone?
If the business were sold tomorrow at a disappointing price, where would this loan sit in the waterfall?
Do I understand the product well enough to explain, in one plain paragraph, why customers keep paying for it?

A tool, not a strategy

Venture debt is a tool with a narrow set of jobs, and it does those jobs well. It buys time, it protects ownership, and it finances things that ought to be financed rather than bought with equity. It does not make a weak company strong and it does not replace the discipline of raising the right amount of equity at the right moment. The companies that use it best treat it the way a good builder treats a power tool: chosen for one task, used with attention, and put down when the task is done. The ones that get hurt are the ones that reached for it because it was the easiest capital to get.

Reading Distressed Credit Cycles: Separating Value From a Value Trap

Peter DeCaprio — Sun, 13 Sep 2026 00:00:00 GMT

Why does a bond that has already fallen a long way keep falling?

That is the question a distressed investor has to answer before doing anything else, because the market has already priced the trouble in. What it has not priced, and cannot price, is the answer to a handful of questions about how this particular trouble gets resolved. A value trap is a distressed credit where the price looks like a discount to something, and it turns out the something was never really there. I run every situation through the same five tests. None of them lives in a spreadsheet. All of them ask what has to be true for the money to come back.

1. Is the problem the capital structure or the business?

This is the first cut and it does most of the work. Some companies are good businesses carrying the wrong balance sheet. They generate cash, customers stay, the product is fine, and the only thing wrong is that a prior owner borrowed too much against it, or the maturities landed in a bad year. That is a capital structure problem, and capital structure problems can be fixed with a new capital structure. Other companies have the opposite condition. The balance sheet is a symptom. Demand is fading, a cheaper competitor has arrived, or the cost base only worked at a scale the company no longer has. Restructuring the debt of a shrinking business buys a few more years of shrinking. The test is simple to state and hard to do honestly: if this company had no debt at all, would you want to own it? If the answer is no, the price of the debt does not matter.

2. Who controls the process?

Distressed outcomes are decided by people in rooms, and the first thing worth knowing is who will be in the room. In most structures, the class of creditors that sits right where the value runs out holds the pen. Everyone senior to them is likely to be repaid and has little reason to negotiate. Everyone junior to them is likely to be wiped out and has little ability to. If my position is the one that decides, I can shape the outcome. If my position is a passenger in someone else's process, I need to understand that person's incentives better than my own. A large holder in the controlling class may prefer to own the company rather than be repaid, and may steer toward that. A lender who is also a supplier may want continuity over recovery. The document tells you the priority. It does not tell you the behavior, and the behavior is what you are buying.

3. What is the asset worth to someone who does not need it?

Every distressed file contains a valuation prepared by someone with a reason to prefer a particular number. The company's advisers want a value high enough to keep the equity alive. The senior lenders want one low enough to keep the juniors out. I try to ignore all of them and ask a colder question: what would a buyer with no history here, no sunk cost and no emotional stake pay for these assets in a normal market? Sometimes the answer is reassuring, because the business has real customers and a real position that would attract several bidders. Sometimes the answer is that the only buyers are the people already in the structure, which means the value is whatever they decide to say it is. A recovery that depends on a friendly appraisal is not a recovery. It is a hope with a footnote.

4. How long can you be wrong?

Distressed credit punishes impatience more than it punishes error. A thesis can be entirely right and still lose money because the process took twice as long as planned, and the position had to be sold to someone else to meet a redemption or a margin call. So before I size anything I ask how long the resolution could reasonably take, then double it, and ask whether the capital I am using can sit still for that long. Interest that stops accruing, coupons that switch to payment in kind, a company that burns cash while the creditors argue, all of these change the arithmetic of waiting. The position has to be small enough, and the funding of it patient enough, that being early is merely uncomfortable rather than fatal. Most value traps I have seen were not wrong on value. They were wrong on time.

5. What does the exit look like?

The last test is the one people skip because it feels premature. It is not. A distressed position ends one of a few ways: the debt is repaid at maturity, it is refinanced, it is exchanged for new paper, it is converted into ownership, or it is sold to someone else. Each of these has a different buyer on the other side and a different set of conditions that must hold. If the plan is to be repaid in cash, where does the cash come from? If it is to own the equity, is this a business I would want to run, with a board I would want to join? If it is to sell, who buys distressed paper in the month I will need to sell it, and what will the market look like then? An exit that requires a particular kind of market to exist on a particular day is a trade, not an investment. I want the exit to work in an ordinary market, on an ordinary day.

The point of the tests

Four of these questions can come back with a good answer and the fifth can still kill the idea. That is by design.

Cheapness is the entry fee, not the thesis.

What separates value from a value trap in distressed credit is not the size of the discount but the strength of the mechanism by which the discount closes: a business worth owning, a process that can be influenced, an asset with real outside buyers, a timeline the capital can survive, and an exit that does not depend on luck. When all five hold, the fall in price was an opportunity. When they do not, it was information.

High Yield Research When Rates Stay Higher for Longer

Peter DeCaprio — Sat, 12 Sep 2026 00:00:00 GMT

You asked me what changes about high yield research when the cost of money stops falling and simply stays where it is. It is a good question, and the honest answer is that most of the research does not change at all. The issuer still has to make its coupon, the bonds still rank somewhere in a structure, and the analyst still has to decide whether the business can carry its debt through a bad year. What changes is the margin for error, and the habits that a long era of cheap money let people get away with. So rather than a rule, here are four things I believe about this work now, in the order I would want you to learn them, followed by what would make me wrong.

The first thing to accept is that the refinancing window is no longer a given. For a long stretch of the market, a maturity was a formality. A company with a bond coming due would issue a new one at a lower coupon, and the analysts would move the maturity wall out a few years and get on with their day. That habit rests on an assumption that new money will always be cheaper than old. When it is not, a maturity becomes an event. The company has to prove, in that month, that it can pay a higher coupon out of the same cash flows, and the market has to be open on the day. Your research now has to treat every maturity inside the next few years as a decision point, and ask what the company does if the window is shut when it arrives.

Second, the coupon is now doing real work, so read the cash flow statement before the income statement. When money was cheap, interest was a small line and the interesting analysis happened above it, in revenue and margins. Now interest is a large claim on operating cash, and the question of whether a business is solvent turns on the cash left after paying it. That means unwinding every adjustment management has made on the way to its favorite earnings figure. Add back nothing you would not personally sign for. Deduct the capital spending the company needs simply to stand still, and the working capital that grows with the business. What remains is the money available to service debt, and you will be surprised how often it is thinner than the headline suggests.

Third, the gap between issuers is now wider than the gap between sectors. In an easy market, credits trade on their sector. A cable operator is a cable operator, and the good and bad ones move together because the tide lifts all of them. In a market where money has a price, the tide goes out, and the difference between a business that built its balance sheet in a downturn and one that built it in a boom becomes the whole story. Two issuers in the same industry, with the same leverage, can be entirely different credits depending on when their debt was priced, how their maturities are laddered, and whether their coupons are fixed or floating. Sector views are a starting point. Issuer views are where the return is.

Fourth, the documents matter again, and most people have stopped reading them. A generation of bonds was issued with covenants loose enough to allow almost anything, and for years it did not matter because nobody needed to use them. It matters now. The question of who gets paid first when a company gets into trouble is answered in those documents, and the answers have become more creative: assets moved out of reach of existing lenders, new debt layered in ahead of old, exchanges that treat holders differently depending on whether they were in the room. Read the indenture. Find out what the company is permitted to do to you, and assume that, under enough pressure, it will.

What would change my mind

If funding markets reopened broadly and cheaply, and stayed that way long enough for issuers to refinance their whole maturity schedule at comfortable coupons, then the first thesis would fade in importance. The window would be a formality again, for a while. I would still keep the habit of treating maturities as events, because the next closure never announces itself.

If the companies in the index were paying down debt faster than I expect, either from strong cash generation or from equity raised to do it, then the coupon burden would matter less and the second thesis would matter less with it. That is a testable claim, and I check it issuer by issuer rather than trusting the aggregate.

If the economy proved less sensitive to the cost of money than the last several cycles suggest, so that businesses simply grew into their interest bills, the dispersion I expect between issuers would compress. I would be glad to be wrong about that. I would also want to know why this cycle was different before I believed it.

You will notice that none of these is about where rates go next. That is deliberate. Predicting the rate is a different job, done by different people, and usually done badly. Your job is to know what each company can survive. Read the cash flow statement first, the documents second, and the sell-side last, and write to me when you find the issuer that breaks the pattern.

Closed End Funds and the Persistence of the Discount

Peter DeCaprio — Fri, 11 Sep 2026 00:00:00 GMT

A closed end fund is a pool of investments wrapped in a company with a fixed number of shares. Unlike an open end fund, which creates and cancels shares every day at the value of what it holds, a closed end fund raised its money once, at the offering, and its shares have traded among investors ever since. That single design choice gives the fund two prices where most funds have one, and the gap between them is the subject of this note.

What a discount is and why it exists

The first price is the net asset value, which is what the portfolio inside the fund is worth. The second is the share price, which is what someone will pay you today for a claim on it. When the share price sits below the net asset value, the fund trades at a discount. When it sits above, at a premium.

The discount exists because nothing forces the two prices together. An open end fund closes the gap automatically, since anyone can hand back shares for their share of the assets. A closed end fund offers no such door. The only way out is to sell to another investor, and that investor sets the price with reference to many things besides the assets: the fees the manager charges, the leverage the fund carries, the income it pays, the difficulty of trading it, and whether anyone else wants it that day. A discount is the market's running verdict on all of those at once.

Why it persists

Discounts appear for many reasons. They persist for three.

The fee is a permanent claim on the assets. A fund that charges a management fee every year is, from the shareholder's point of view, a portfolio with a leak in it. A rational buyer will pay less than asset value for a pool that is being drained, and the larger the fee relative to what the portfolio can earn, the wider the discount will settle. Nothing about a good year changes this arithmetic, which is why the discount returns after every rally.
Nobody with the power to close it is paid to. The manager earns its fee on assets, not on the share price, and a discount does not reduce the assets. Closing the gap, whether by buying back shares, converting to an open end structure or winding the fund up, shrinks the thing the manager is paid on. The shareholders who could force the issue are usually dispersed, small and unorganized. So the one party with the means lacks the motive, and the parties with the motive lack the means.
The buyers who would arbitrage it away cannot get out. In most markets, a persistent gap between two prices for the same thing attracts capital that closes it. Here the arbitrage requires holding the shares until the discount narrows, with no way to redeem and no date by which it must happen. The people who try often need to sell before it does. The discount persists because the trade that would eliminate it demands more patience than most people who notice it have.

When a discount is an opportunity and when it is a warning

The opportunity

A discount is worth pursuing when there is a mechanism, not just a hope, by which it closes. The fund may have a term, a date on which it will liquidate or offer to buy back shares at asset value. The board may have a policy of repurchasing shares whenever the gap widens past a stated level. An activist holder may be accumulating shares and be able to force a vote. The manager may be changing, with a different view of the gap. Or the portfolio itself may be about to throw off enough cash that the distribution alone justifies the price.

The warning

A discount is a warning when it reflects something about the assets that the asset value does not. Illiquid holdings marked by the manager rather than by a market. Leverage that would force sales in a downturn. A distribution larger than what the portfolio earns, so that the fund is handing investors their own capital back and calling it income. A manager whose fee is large and whose record is short. A discount in a fund like this is not a mispricing. It is a correct pricing of the things you have not yet found.

A discount that narrows	A discount that persists
A fixed term, tender offer or liquidation date	An open ended life with no scheduled event
A board that has repurchased shares before	A board that has never acted on the gap
A known holder with enough shares to force a vote	A dispersed shareholder base with no organizer
Holdings that trade daily and are priced by the market	Holdings priced by the manager
A distribution covered by what the portfolio earns	A distribution that returns capital
A modest fee relative to what the assets can earn	A fee that consumes much of the expected return

How I approach it

I start with the assets, not the discount. If I would not want to own the portfolio at full value, a discount to full value does not interest me.

A discount is a claim on something, and the something has to be worth having.

Only then do I turn to the gap and ask whether there is a reason it should close, who would make that happen, and how long it could take.

Then I ask what I am being paid to wait. A fund with a covered distribution pays me while the mechanism works. A fund that pays nothing asks me to carry the position on faith. The first is a position. The second is a bet on other people's behavior.

The last thing I do is read the fund's own history of discounts. A fund that has traded at roughly the same discount through several markets is telling me the gap is structural, and that the market priced the fee and the governance correctly long ago. A fund whose discount has recently widened for a reason I can identify and that will pass, a forced seller, a rough quarter, a change in index membership, is a different animal. The first fund is telling you what it is. The second is telling you what just happened to it, and only the second is worth paying to wait for.

Telecommunications Sector Analysis: Cash Flow, Infrastructure, and Consolidation

Peter DeCaprio — Thu, 10 Sep 2026 00:00:00 GMT

Look at a telecommunications company through the income statement and you will see a mature, slow business with a great deal of debt. Look at it through the balance sheet and the cash flow statement and you will see something more interesting: a network that took decades and enormous capital to build, that customers pay for every month whether or not they think about it, and that would cost a competitor far more to replicate than the market is charging for it. Telecom rewards the reader who is willing to sit with the balance sheet for a long time, because that is where both the value and the risk live. Utilities are the adjacent sector I read the same way, for the same reason.

The asset is the story, and the asset is the thing most analysts skip.

I use three lenses, and I try to use all three before forming a view.

Cash flow

The first thing to understand about an operator is the difference between the profit it reports and the cash it keeps. The reported figure is flattered by the fact that the network was largely built years ago, and the depreciation of it is a charge that involves no cash. The cash figure is punished by the fact that the network never stops needing money: spectrum, fiber, towers, and equipment upgrades that arrive every few years whether the company wants them or not. The number that matters is what remains after the maintenance spending that keeps the network competitive, and companies are not always careful about separating that from the spending that grows the business.

The second thing is the shape of the revenue. Subscription income from customers who rarely switch is worth more than the same amount from customers on short contracts, and both are worth more than one-time equipment sales. Reading the mix tells you how much of the cash flow is genuinely recurring and how much has to be won again each year.

The third thing is the debt schedule. Operators carry a lot of debt because their cash flows are stable enough to support it, and that is reasonable until it is not. What matters is not the total but the timing: when the maturities land, whether the coupons are fixed, and how much of the annual cash flow is already spoken for by interest before the dividend and the capital plan get their turn.

What to watch:

The gap between the capital spending the company calls growth and the spending the network actually needs to hold its position.
The trend in customer churn, because a small change in how many people leave each month changes the value of the whole base.
The share of operating cash consumed by interest, and the direction it is moving.

Infrastructure

A network is a physical thing, and physical things have economics that do not show up on a quarterly earnings call. Fiber in the ground has a long life and a low running cost once it is laid. Wireless spectrum is a scarce right that cannot be manufactured. Towers are valuable because moving off one is expensive for every tenant on it. Each of these assets has a replacement cost, and the useful analysis is comparing that replacement cost with what the market is paying for the company that owns them.

The infrastructure lens also shows where competition is real and where it is not. In a given territory, the question is how many networks actually reach the customer. Where there is one, pricing power is durable. Where there are two, it is shared. Where a third is being built with borrowed money, the cash flows of the incumbents are about to be tested. This is knowable from public disclosures and a map, and it explains more about future margins than any forecast.

What to watch:

New network construction in the company's core territories, especially construction financed by parties who need a return quickly.
The age and technology of the plant, because an old network is a liability disguised as an asset.
Whether the company owns its infrastructure or leases it, since leased assets carry someone else's pricing power.

Consolidation

Telecom consolidates because the economics push it to. Networks have high fixed costs and low marginal costs, so two overlapping networks combined are worth more than the two apart. That logic produces a steady drumbeat of mergers, spin-offs and asset sales, and each one changes the value of everyone nearby. The acquirer takes on debt. The target's bondholders find themselves in a different structure. The competitor left out finds that its market has changed shape.

For a credit investor, consolidation is where the documents earn their keep. A change of control provision decides whether a bondholder can be repaid when the company is sold. A ratings-based coupon step decides what happens if the combined company is weaker than the parts. For an equity investor the question is simpler and harder: does the combination actually reduce the cost of serving a customer, or does it merely make a larger company with the same problems?

What to watch:

Announced transactions in adjacent territories, which tend to reprice the whole region.
Sales of towers, fiber or spectrum, which reveal what infrastructure is worth to a buyer who wants only the infrastructure.
Whether the acquirer is paying with cash, stock or debt, because that decides who carries the risk of being wrong.

Discipline

None of these lenses produces an answer quickly, and that is the point. Telecom is a sector where the fast reading, the one built on last quarter's subscriber number and next quarter's guidance, is usually wrong in the way that matters, because it prices a network as though it were a product. The slow reading, built from the balance sheet, the map and the documents, takes longer and pays better. The discipline is to keep doing the slow reading when the fast one seems to be working, and to remember that the asset is still in the ground either way.

What Disciplined Underwriting Actually Looks Like in Private Credit

Peter DeCaprio — Wed, 09 Sep 2026 00:00:00 GMT

The file arrived on a Thursday afternoon, the way they usually do: a deck, a model, a summary from the sponsor and a note asking whether I could have a view by early the following week. What follows is a composite, and every detail of it is invented. The company was a distributor of building products across the Southeast, with a founder close to retirement and a sponsor who wanted to buy the business using a mix of its own equity and a loan from someone like me. None of the people are real. The process is, and I have walked through it in this order more times than I can count.

The first read

I opened the financial statements before the deck, and the model before the summary. The deck is written to be read first, which is exactly why I read it last. The statements showed a business that had grown steadily, earned a reasonable margin on what it sold, and collected its receivables slowly. The model showed the same business growing faster, earning a wider margin and collecting faster, all beginning in the first year of new ownership. The gap between the two was the deal. Everything the sponsor needed to be true lived in that gap.

The lesson I keep relearning is that the first read is not for forming a view. It is for finding the thing the materials were arranged to keep you from noticing. There is always one. Here it was working capital: the model assumed the company could grow without tying up more cash in inventory and receivables, and the history said it never had.

The management conversation

I spent an afternoon with the founder and with the operator the sponsor intended to install after him. I asked few questions about strategy and many about mechanics. Who were the largest customers, and why did they buy here rather than from the national chain down the road? What happened the last time a major supplier raised prices? How many days did it take to collect from a builder, and which builders were slow? What broke in the last downturn, and what did the founder do about it?

The founder answered the early questions quickly and the later ones carefully, and the change of pace told me where the risk was. Builders paid slowly in bad years, and the company had carried them, because carrying them was how it kept them. That was a real competitive advantage and a real drain on cash, and the model had captured the first without the second.

Management conversations are not for hearing the strategy. The deck has already told you the strategy. They are for learning what the company actually does when something goes wrong, because that is the moment your loan will be tested.

Building the downside case

I set the sponsor's model aside and built my own from the history. I let revenue fall the way it had fallen in the worst stretch the founder described, let the margin compress as suppliers pushed through cost, and let the receivables stretch the way he told me they stretched. Then I ran the cash. In that case the business still made money, but not enough to cover the interest on the loan the sponsor had asked for and also fund the working capital it would need when volumes recovered.

That was the case I feared, and it was not exotic. It was simply the company's own history, applied to the sponsor's balance sheet.

The downside case is the underwriting. The base case tells you whether the sponsor is optimistic. The downside case tells you whether you get repaid. I spend most of my time on the second, and I build it from what the company has already survived rather than from what I can imagine.

Structuring for the case you fear

With the feared case in hand, the structure wrote itself. I offered less debt than the sponsor wanted, enough to make the purchase work but not so much that the company would have no room to breathe in the bad year. I asked for a covenant that tested cash flow against interest each quarter, set at a level that would trip early in the downside case rather than late. I asked for a portion of excess cash each year to come back to the loan, so that the good years would reduce the balance before the bad one arrived. I put limits on acquisitions and on distributions to the sponsor while the loan was outstanding. And I asked for monthly reporting on receivables, because that was where trouble would show first.

Every term in a loan agreement should answer to a specific fear.

Terms that answer to nothing are decoration, and decoration gets negotiated away. Terms that answer to a fear you can name will survive the negotiation, because you can explain why they are there.

The decision memo

I wrote the memo in plain language and put the downside case on the first page. I described the business in a paragraph a stranger could follow, named the two or three things that would have to go wrong for the loan to be impaired, and explained how the structure responded to each. I said what I did not know. I said what I would want to see in the first year that would tell me whether the thesis was holding.

If the loan cannot be explained on a page, the lender does not understand it. The memo is not a record of the decision. It is the test of whether a decision was actually made.

After the close

The loan closed. The first year went close to plan, and the cash sweep brought the balance down. The second year brought the slowdown, much as the founder had described it, and the covenant tripped in the quarter I had expected it to. That was the point. It brought the sponsor to the table early, with a smaller balance outstanding and a company that was still profitable. The conversation was uncomfortable and short, and the loan came through it.

Underwriting does not end at the close. It ends when the money is back. The structure you build on a Thursday afternoon is the only thing in the room when the bad year arrives, and disciplined underwriting is nothing more than the habit of building it as if that year were certain.

The Case for Contrarian Positioning in Public Equities

Peter DeCaprio — Tue, 08 Sep 2026 00:00:00 GMT

The strongest argument against contrarian investing comes from people who have tried it: being early looks exactly like being wrong, and for a long time it pays exactly the same. I take that seriously, because I have been on the wrong side of it. So the case for contrarian positioning in public equities is best made the way an honest case is made, by putting the objections first and answering each on its merits.

The objection: contrarian is just early and wrong

The crowd is not stupid. It is made of people who read the same filings, run the same models and talk to the same managements. When a stock has fallen for a year, the market has usually found a reason, and the contrarian who buys it is claiming to know something that a great many informed and motivated people do not. Most of the time that claim is false. The stock keeps falling, the contrarian averages down, and what was described as conviction turns out to have been stubbornness with a longer time horizon. Early and wrong feel identical from the inside, and the account statement does not distinguish between them either.

The response: early and wrong are separated by the work

The objection is right about the symptom and wrong about the cause. A contrarian position taken because a stock has fallen is not contrarian at all. It is a bet against momentum, and that bet loses more often than it wins. A contrarian position taken because you hold an independent view of what the business is worth, built from the cash flows and the balance sheet rather than from the price, is a different thing. The price tells you that other people disagree. The work tells you whether they are right.

Early is what happens when the work is right and the market has not caught up. Wrong is what happens when the work was never done.

The two look alike for a while. The difference is whether there was a thesis underneath, and whether the position was sized so that the wait could be survived.

The objection: the crowd is usually right

Markets aggregate information well. Prices move because someone learned something, and a price that has moved a long way has usually absorbed a great deal of learning. An investor who sets out to disagree with that process is setting out to disagree with the sum of everyone else's research. The record of people who make a habit of this is not good. The crowd is right far more often than it is wrong, and a strategy built on the crowd being wrong is a strategy built on the exception.

The response: usually is the whole point

Yes. The crowd is right most of the time, which is exactly why, most of the time, there is nothing to do. Contrarian positioning is not a stance to hold every day. It is a response to a specific and recognizable condition: a price that has been moved by something other than a change in value. A forced seller who has to be out by a certain date. An index change that requires funds to sell without regard to price. A narrative that has taken over from the numbers, in either direction. A whole sector marked down because of a problem at one company in it. These moments are rare because the crowd is usually right. If it were often wrong there would be no crowd worth leaning against, and no return for doing it. The rarity is what makes the return possible.

The objection: you cannot time sentiment

Suppose the contrarian is right about value. The stock is still worth what someone will pay for it, and sentiment decides that. Nobody can say when a mood will turn. A position that depends on other people changing their minds is a position with no schedule and no mechanism, and a thesis with no mechanism is a story. The market can stay wrong for longer than any investor can stay funded.

The response: you do not have to

Sentiment is not the thing being timed. What closes the gap between price and value is never a change of mood on its own. It is a cash flow that arrives, a dividend that gets paid, a buyback that shrinks the share count, an acquirer who pays for the whole business, a forced seller who finishes selling. Each of these is a mechanism, and each can be studied. The question is not when people will feel differently but what events lie ahead that will make the value visible whether or not anyone feels anything. When the mechanism is clear and the capital can wait for it, sentiment is a source of entry prices rather than a risk. When there is no mechanism, the objection is correct and the position should not be taken.

What contrarian positioning is not

It is not disagreement for its own sake. A person who reflexively takes the other side of every popular view is as predictable as the crowd and less well informed. It is not buying whatever has fallen the most. It is not shorting whatever has risen the most. I have argued short theses in public, including on Tesla, and the durable lesson of doing that is that a position stated out loud has to be defended in front of everyone, which is a useful discipline and a poor reason to hold a view. Contrarian positioning is not a personality. It is a conclusion, reached occasionally, when the price and the work disagree and there is a reason to expect the work to win.

Where that leaves it

The case comes down to this. Markets are efficient enough that most of the time the price is a fair guide to value, and inefficient enough, in specific and identifiable moments, that the price is not. A contrarian investor is someone who does the work continuously, acts rarely, sizes for the wait, and insists on a mechanism. Everything else that goes by the name is either momentum in reverse or stubbornness with a story attached.

Why Active Management Earns Its Keep in Dislocated Markets

Peter DeCaprio — Mon, 07 Sep 2026 00:00:00 GMT

The index fund is the best product the investment industry has produced for ordinary savers, and I am not going to pretend otherwise. For most investors, in most markets, in most years, a low-cost index fund is the right choice, and the argument for it is one of the strongest in finance. Every dollar invested actively is, in aggregate, the market, minus what it costs to run. An active manager who wants to be paid has to explain where the exceptions come from, and an honest one has to admit that they are exceptions. I want to make that argument narrowly, for the specific kind of market the index was never designed to handle.

Where indexing wins

The index wins on cost, and cost compounds. A fee that looks small in a single year is a large claim on the return over a working life, and no argument about skill survives the arithmetic if the skill is not there. The index wins on discipline. It never panics, never chases, never falls in love with a story and never has to explain a bad quarter to anyone. It holds what it holds by rule. And the index wins on the plain fact that the average active manager, before fees, is the market, so that after fees the average active dollar must trail it. There is no way around that. It is a matter of accounting, not opinion.

In an orderly market, one in which prices move on information and buyers and sellers meet somewhere reasonable, the index is close to unbeatable, and a manager who charges a fee to hug it is charging for nothing. I would rather concede this and be believed about what follows than argue it and be dismissed.

Where it breaks

The index is a rule, and a rule has no judgment. It owns each company in proportion to its price, so whatever has gone up becomes a larger holding and whatever has fallen becomes a smaller one. It sells what has been removed from the index and buys what has been added, on the day it is told to and at whatever price that day offers. It holds through everything, because holding through everything is the rule.

In an orderly market, that is a virtue. In a dislocated one, the rule becomes part of the problem. When a whole sector is marked down because one company in it has failed, the index cannot tell the survivors from the casualty. It owns them all, in proportion to prices that no longer say anything about the businesses. When a large fund has to sell to meet redemptions, it sells every holding, including the ones its own analysts would keep, and an index fund that is being redeemed does the same. When prices detach from value, the index keeps following the price, because the price is the only thing it can see.

None of this is a defect in the design. It is the design.

The index was built to give investors the market cheaply, and in a dislocated market the market is the thing that has stopped working.

When active management earns its fee

Market condition	What the index does	What an active manager can do
Orderly	Holds every company at its weight and collects the market return at almost no cost	Very little that justifies a fee. The honest course is to stay close to the market and keep costs down
Volatile	Rides the swings and rebalances by rule, adding to what has risen and trimming what has fallen	Hold some cash for the moments when price and value separate. Trim what has run, add to what has been marked down for no reason of its own
Dislocated	Keeps owning every constituent in proportion to price, including the ones whose price no longer reflects the business	Separate the cheap from the broken. Concentrate in the businesses whose cash flows survive. Move into the debt if that is where the value has gone
Forced selling	Becomes part of the selling, because redemptions sell every holding without regard to merit	Be the buyer of what someone must sell, at a price that pays for the risk of being the one who provides the liquidity

Read the table from top to bottom and the scorecard is clear. In the first row the index wins outright. In the second it is roughly a draw, and the manager has to be good to justify the cost. In the third and fourth rows the index cannot do the thing that needs doing, because doing it requires a judgment about value that the rule does not contain. That is where the fee is earned, and it is earned in a matter of weeks or months, not spread evenly across the years.

What earning it requires

Capital that is not itself a forced seller. A manager whose clients redeem in the dislocation becomes a seller too, and the opportunity is handed to someone else.
Research done before the dislocation, not during it. The list of what to buy, and the price at which to buy it, has to exist before the prices arrive, because there is no time to build it once they do.
The willingness to hold cash and look slow in orderly markets, which is the cost of being able to act in the other kind.
A fee that reflects what the manager does in the hard markets rather than a toll collected in the easy ones.
The humility to do very little most of the time, so that the record shows a manager who acted when acting mattered and otherwise stayed out of the way.

The keep

Active management is not a better way to own the market. In ordinary conditions it is a worse one, and the fee is the difference. It earns its keep in the markets that break the rule, when prices stop carrying information and someone with capital, research and patience has to decide what things are worth. Those markets are rare, which is why the index is right most of the time. They are also where much of the long-run return gets decided, which is why the manager who is ready for them is worth paying, and the one who is not is worth nothing at all.