I was hiking in the Colorado Rockies with my family a month ago when the May payrolls report landed. Strong across the board — 172,000 jobs added, more than double the consensus forecast. My first thought: great news, puts a real dent in those lingering stagflation fears. I wondered if this perverse market would sell off on a bit of profit-taking. What I did not expect was the largest selloff in months while I was out on a trail somewhere.

My son asked me that night if he should buy the dip. I didn’t have a good answer — I’d been away from screens all day and couldn’t piece together what had actually occurred. The headlines weren’t much help: Market sells off on fears of Fed hikes behind strong jobs data.

The following week, the same sources informed me that the market was rallying on anticipation of a geopolitical de-escalation that would reduce oil prices, thus easing inflation, thus increasing the chances of a Fed cut. Then it sold off again when the Fed struck a hawkish tone at its June meeting, coinciding with strong retail sales data — and on and on. Future Fed actions were cited as the primary, or even sole, reason for every single market wiggle.

Last week, chips sold off sharply — Micron, coming off its best quarter ever, led the way down — because a South Korean politician mentioned a possible AI tax in a Facebook post. Google lost two senior engineers to OpenAI and Anthropic and promptly shed $250 billion in market cap. The valuations of both high-multiple companies are certainly up for debate, yet what is not debatable is that either of those moves had anything to do with monetary policy. Yet the financial press helpfully informed me, both times, that markets were selling off on fears of Fed hikes.

I swear that every Financial Journalism 101 course includes ‘always mention the Fed’ somewhere in the syllabus. This piece is my attempt to untangle why that bothers me so much — and why it should bother you too, even if you’ve never traded a single share of stock. The Fed is important, yes, but we confuse multiple Fed functions and overattribute all sorts of things to them. In doing so, we obscure what’s actually happening, make worse decisions because of it, and give ourselves false comfort at exactly the moment we should be asking harder questions.

The Inversion Problem

One of the worst legacies of the post-2008 zero interest rate era is the way valuations became so dependent on central bank support that market logic inverted. Good news became bad news because it might make the Fed less accommodative — and vice versa. We’ve left ZIRP behind, but we haven’t shaken the distorted logic that came with it.

Benjamin Graham famously described the market as a short-term voting machine but a long-term weighing machine. Right now, the voters are obsessing over a single variable — future Fed actions — while the weighing machine is trying to price AI disruption, fiscal fragility, geopolitical realignment, a productivity revolution, and the actual structural health of the economy.

Strong jobs data and a reduced likelihood of stagflation is not bad news simply because it reduces the probability of an immediate rate cut. If the economy is performing well and the risk of a hard landing is reduced, that is a structural positive. Period. Valuations are ultimately driven by earnings and earnings expectations, not by the tortuous idea that corporate earnings are so good they might someday be less good because the Fed tweaked the overnight rate by 25 basis points. Obsessing over minor policy adjustments while ignoring fundamental economic strength is the financial equivalent of hyper-focusing on the wind resistance of the hood ornament while ignoring the performance of the V8 engine under it.

Similarly, if oil prices ease, that’s genuinely good for corporate planning and input costs — not because it changes the Fed’s calculus, but because stable energy prices make the real economy function better. Moreover, the idea that elevated commodity prices from temporary supply shocks should derail the Fed’s long-term reaction function misses how monetary policy actually works. There’s a reason the Fed focuses on core inflation: a spike in energy prices doesn’t structurally inflate the entire economy; it acts as a tax that diverts discretionary spending away from other areas. The Fed knows this. The headlines don’t seem to.

This hyper-focus on a single scapegoat muddies our understanding of everything else. Recently, both my wife and my mother complained about high airfares and attributed them to elevated oil prices — an explanation that has been plastered all over their news feeds. Meanwhile, crude prices had already retraced nearly to their pre-conflict levels.

The real drivers of elevated airfare are structural capacity constraints: Boeing production delays, chronic pilot shortages, and a consolidated industry exercising massive pricing power with sustained passenger demand. But those aren’t easily fixable problems, nor do they fit neatly into a headline. People crave tidy explanations they can see, and easy villains. Geopolitical oil shocks dominate headlines; disrupted supply chains and onerous regulatory policy do not make for good clickbait. Policymakers much prefer to point fingers at convenient bogeymen rather than address the messy underlying realities.

As H.L. Mencken put it: “For every human problem, there is a solution that is neat, plausible, and wrong.”

Why This Actually Matters

So who really cares if the headlines are lazy? Markets eventually sort things out and people misunderstand complex issues all the time.

I fully admit that I can be relentless (annoying?) when I see a false narrative taking root. To my former colleagues who had to endure my soapbox rants about why an inverted curve did not mean an imminent recession, or how rates near zero could no longer hedge equity risk — I am sorry. To my family, exhausted from rolling their eyes every time I rail against the invasive plant species taking over my backyard — yeah, also sorry. As I write this, Progressive is probably scripting a commercial about me turning into my father, lecturing unsuspecting people about things they simply don’t care about.

And yet.

Invasive species do real harm to an ecosystem. So do false narratives in financial media. They crowd out the stories and structural insights we actually need to survive the next cycle.

Retail investors are paying more attention to their portfolios than ever before. Fed a steady diet of bogus narratives, they misunderstand the actual drivers of risk, leading to poor decisions at the worst possible moments — selling on “Fed hike fears” when the real move was driven by shifts in AI competitive dynamics, or buying on geopolitical optimism when the energy-to-inflation-to-Fed chain of reasoning barely holds together.

Policymakers start to believe these narratives too. We have a structural housing shortage in this country that won’t be fixed by Fed rate cuts; it requires addressing local zoning laws, permitting timelines, and construction capacity. But when enough politicians come to believe the problem begins and ends with the central bank, they fail to address the actual bottleneck. Instead, they create new problems by politicizing the institution, pressuring it to cut for the wrong reasons, and undermining the very independence that makes monetary policy credible. Ironically, as I have written here and here, the US economy has rarely been less sensitive to the Fed Funds rate than it is today — even as local market moves overreact to every single central bank utterance.

Two Feds, One Narrative

This brings us to the Fed’s own institutional dilemma.

Kevin Warsh, the newly sworn-in Fed Chair, is deeply frustrated by the central bank’s outsized footprint in asset markets — and rightly so. His stated plan is to shrink the balance sheet and reduce what he views as market-moving overcommunication. I am broadly sympathetic. The post-GFC balance sheet expansion distorted risk assets and enabled fifteen years of fiscal recklessness.

But the communication strategy misdiagnoses the problem.

For many decades, the Fed was famously opaque — and that opacity had a perverse effect. Investors parsed every stray scrap of information for clues, and a single word change could move markets violently. Greenspan’s eventual shift toward transparency was deliberate: flood the zone with words, dilute the impact of any single one. Verbal inflation. It worked, up to a point.

Warsh inverts that logic. Less communication, applied to a market that already believes the Fed is the dominant variable, doesn’t reduce sensitivity — it concentrates it. You’d be recreating the exact dynamic that made Fed-watching such a hair-trigger exercise in the first place. Warsh is a smart guy and none of this is lost on him. But he is operating with an audience that has been conditioned by something much harder to undo than press conference frequency.

Over fifteen years of crisis interventions, balance sheet expansions, and reliable emergency backstops, the rumored Fed Put became entirely real. Investors learned, rationally, that the Fed would show up when things broke. That conditioning created a powerful spotlight on every Fed action — and it won’t be dimmed by talking less. The market will simply read more into whatever signals remain, and when the next crisis hits, the game of chicken between markets and the Fed will settle the question of how high the new hurdle is to trigger it.

But here’s where the conflation causes the most damage. The Fed Put — exercised through emergency facilities, lending channels, and regulatory interventions — is a genuine systemic shock absorber. The Fed Funds rate is simply the rate at which commercial banks lend excess reserves to one another overnight. These are different mechanisms with different implications, and the market treats them as just “The Fed”.

The Fed Funds rate matters far less to the real economy than it used to — I’ve written about why at length (link). The idea that a quarter-point decision months away carries relevance that trumps actual corporate earnings or technological disruption is simply wrong. Yet it is the idea that headlines cheerfully reinforce every afternoon.

The Fed didn’t become the dominant narrative variable because it talked too much. It became dominant because it kept showing up with a fire hose every time there was smoke in the market. Warsh can talk less. He cannot put that genie back in the bottle.

The Risk We’re Not Pricing

Here is the irony that ties this all together.

Equity valuations are theoretically long-duration instruments — decades of discounted cash flows priced off long-term growth and productivity assumptions. A quarter-point Fed decision eight weeks out should be mathematically immaterial to that calculation. It slightly tweaks the discount rate (the denominator), but we are entirely ignoring the actual earnings generation (the numerator).

And yet, we obsess over the short-term policy path while largely ignoring the tectonic shifts that actually matter over the next decade.

Those shifts are genuinely uncertain right now. AI is introducing non-linear change to productivity, labor markets, and corporate competitive moats in ways that completely resist traditional modeling. The global geopolitical order is being actively renegotiated. Our domestic fiscal trajectory is mathematically unsustainable, and nobody knows what kind of crisis or adjustment mechanism will eventually force the resolution. The range of potential macro outcomes is incredibly wide.

The rational response to this level of structural complexity is to demand a higher equity risk premium — more compensation for bearing uncertainty, wider distributions of outcomes, and far more humility regarding valuations.

Instead, the opposite has happened. The Fed’s ZIRP era guidance and balance sheet expansion contributed to a perception of stability and risk mitigation that endured. Meanwhile, the media’s hyper-focus on the Fed creates a false illusion of short-term clarity, as if the next dot plot tells us something meaningful about the next decade. It substitutes a simple, trackable variable for the harder, more uncomfortable work of sitting with genuine structural uncertainty.

This has become a vicious cycle. Financial media covers the Fed because that’s what audiences want; audiences want it because that’s what they’ve been trained to focus on. Fifteen years of central bank interventions quietly embedded the belief that equity risk had somehow been socialized away — that buying every dip was riskless because the Fed would always be there to bail them out.

That muscle memory has persisted long after the conditions that produced it have vanished. Maybe it ends badly. Maybe AI justifies everything. The point isn’t the outcome — it’s that we’re not being paid to find out.

What This Means

The investors who navigate this regime shift well won’t be the ones who most accurately predict the next Fed dot plot. They’ll be the ones who resist the seduction of simple explanations long enough to investigate what’s actually happening — who can hold genuine complexity rather than retreating to the nearest convenient variable.

That requires a kind of intellectual stubbornness: the willingness to say, “I don’t know exactly what drove that market move today, but I am fairly confident it wasn’t primarily about the Fed,” and sit with that uncertainty rather than reaching for an easy headline.

I’m not going to pretend I have a simple alternative framework to offer — doing so would make me exactly what I’m criticizing here. But I’ll keep pointing out the invasive species in the yard. Someone has to.

While I laugh — the image captures something true about how this asset class has been marketed — the fact that private credit is generating headlines at a time when geopolitical tensions are at a fever pitch tells you it has officially gone mainstream.

The headlines tend to fall into two categories: warnings about liquidity gates trapping investors, and warnings about deteriorating lending standards producing waves of defaults. Both risks are real. Neither is the disaster the headlines imply.

Liquidity mismatches create friction — and the headlines about them are doing something useful, forcing investors to ask whether their investment actually matches their time horizon and risk tolerance. That reassessment is healthy. The investors who discover they were mismatched are better off repositioning now than during a genuine crisis. Credit losses from poor underwriting are disappointing, but contained — they hurt the funds that made the bad loans, not the broader financial system. Both are symptoms of what always happens when capital floods into a market that has only known one direction: standards slip, mismatches accumulate, and a correction eventually forces the reckoning. That reckoning is uncomfortable. It isn’t a crash.

What causes crashes is leverage. Not redemption gates, not bad loans — leverage, acting at multiple levels simultaneously, transforming contained problems into contagion. That’s the deeper risk in private credit — not because a crash is inevitable, or even likely, but because leverage is the mechanism that would turn a contained problem into a systemic one. It has been the common denominator in every major crash. It’s worth understanding before it becomes relevant.

This isn’t a call to panic — it’s a reason to pay attention. The current mini-panic feels like a healthy corrective. The investors reassessing their positions and reducing where appropriate are making the system more stable, not less. But underneath that growth and that healthy reassessment, leverage has been accumulating across the private credit ecosystem in ways that deserve more scrutiny than the headlines are giving it.

The Hidden Migration of Leverage

I’m a bond guy at heart. I have no objection to leverage. But I’ve had uncomfortably close views of what happens when it builds unchecked, migrates to less visible corners, and then gets discovered all at once.

Lloyd Blankfein recently put it well: “I wonder where there’s hidden secret leverage? Everyone says the world’s not leveraged… that’s exactly what everybody said in the mortgage crisis until you suddenly discovered there was mortgage risk in Iceland.”

2008 wasn’t primarily a story about bad loans; it was about leverage acting at multiple levels simultaneously — households, investment vehicles, and leveraged entities adding more debt on top. The problem wasn’t any single point of failure, but that all the layers deleveraged together. The pieces looked manageable. The aggregate was not.

Problem leverage behaves like water — it migrates to the cracks that are less visible and less regulated. Each point of leverage may look justifiable in isolation. In aggregate, the picture looks different.

The Leverage Stack

Think of the private credit ecosystem as a stack: borrower → fund → sponsor → vehicle.

At the borrower level, middle-market companies typically carry 5–6x debt-to-EBITDA — the market’s preferred metric, though it flatters the picture by stripping out the very interest expense that’s under pressure. Cash flow coverage tells a more honest story, and those numbers are under increasing strain. Earnings disappointments, already obscured by EBITDA metrics, can result in rapid repricings when debt service costs start rising as a percent of cash flow. At the fund level, managers use subscription credit lines to bridge capital calls and improve IRRs. At the sponsor level, private equity firms facing historically low distribution rates (DPI) are turning to NAV loans — borrowing against portfolio value to manufacture liquidity. The growth of NAV lending alongside record low DPI tells a consistent story: cash in the PE ecosystem is becoming scarce, and sponsors are using leverage to cover the gap.

Perhaps most pressingly, the retail vehicle layer is where the risks all converge. BDCs operate with permitted 1:1 leverage on top of portfolios that are themselves already levered. BDC marks are model-based, meaning cash flow impairment at the borrower level may not yet be visible to the retail investors funding these vehicles. That opacity cuts both ways. In normal markets it creates smooth reported returns. In stress it creates suspicion — and suspicion can be more dangerous than actual impairment. BDCs are already trading at meaningful discounts to NAV, suggesting some investors don’t trust the marks regardless of what the models say. We saw this dynamic in 2008: when confidence in marks evaporates, investors stop asking whether specific assets are impaired and start asking whether any marks can be trusted at all. The answer, for many investors, was to convert everything to cash — and that impulse, multiplied across a retail investor base, is precisely how a markdown becomes a spiral.

There’s another dimension to the leverage stack that doesn’t show up in any single balance sheet: crowding. When capital floods into a one-way market, managers across the industry tend to converge on the same opportunities. In private credit, the most visible example is the concentration in PE-backed technology and SaaS companies — businesses acquired at peak multiples during the zero-rate era, financed with private credit, and now facing a fundamentally different environment. The concentration isn’t visible the way it would be in public markets. A portfolio of 50 loans looks diversified until you notice that 30 of them share the same vintage, the same sponsor profile, and the same sensitivity to software valuations in a higher-rate world. When that sector comes under pressure, it doesn’t just affect individual loans — it affects the marks, the DPI, and the NAV loan collateral simultaneously across the industry. When that concentration is broad enough, writedown and redemption pressures can spill over to other sectors, undermining even well-diversified portfolios.

None of these layers is disqualifying on its own. But they don’t just add — they multiply. When underlying cash flows disappoint — as they eventually do when underwriting standards erode — the pressure doesn’t stay at the borrower level. It travels upward through a stack that assumed those cash flows would be there, and each level gets weaker at the same time. The 2008 lesson isn’t that we’re facing the same crisis — I was there, and this feels nothing like it. But I also learned some critical and painful lessons, and the relevant one today is that leveraged systems don’t unwind linearly. They tip.

When Fundamentals Become Noise

Leverage accumulates quietly. Credit deterioration tends to hide until it can’t. The dangerous moment is when both arrive together — when assumed cash flows fail to materialize, and excess leverage transforms the system from shock absorber into propagator. There are early signs of both in the current private credit market.

Private equity’s cash flow cycle has broken down. DPI — the rate at which investors actually receive distributions — has hit historic lows. Liquidity is being manufactured via NAV loans to bridge the gap. At the borrower level, stressed companies unable to pay cash interest shift to PIK (Payment-in-Kind) structures — adding interest to principal. It’s labeled “flexibility.” PIK structures are not inherently bad, and can be an important tool for creative underwriting. But when a cash-flowing loan converts to PIK because the borrower can’t make the payment, that’s a shadow default — and that’s a trend worth watching.

Both are early warnings. If the current corrective halts or reverses these trends, they remain manageable signals. If it doesn’t, the leverage underneath has a way of turning gradual deterioration into something that moves much faster than anyone expects. A potential private credit spiral runs like this: PIK ratios rise, cash distributions disappoint retail investors seeking income, and redemption pressure builds before marks move. Forced selling then creates new marks that reprice the broader industry. NAV loans that seemed conservative against estimated valuations suddenly aren’t. Margin calls trigger more selling. The good loans get sold alongside the bad; when cash is all that matters, quality is secondary.

Professional investors love to talk about “fundamentals” — right up until the moment when the fundamentals stop mattering. When a markdown spiral begins — selling to meet redemptions pushing prices lower and triggering more selling — only cash safety matters. Relative quality offers no protection when both portfolios are being marked down by the same forced-selling dynamic.

We’ve seen this before. In the cash grab of March 2020, the best loans — the ones able to be sold — were marked down to catastrophe levels simply because investors needed liquidity and the alternative was no cash at all. 2008 was worse. Valuations were irrelevant, and the only thing that mattered was who had available balance sheet and who didn’t: investors who were right about the relative quality of their assets still watched them get written to zero.

The Quality of the Liabilities

In true stress, what matters more than the quality of the asset? The stability of the funding behind it — the ability to stay patient when others cannot. Many platforms have invested heavily in democratizing access to private credit, bringing in a wave of retail capital. But retail investors behave differently than institutional investors — and that difference has systemic implications.

Identical portfolios will perform differently depending on who funds them. A portfolio backed by long-term insurance capital focuses only on loan creditworthiness and cash flow. Insurance CIOs don’t react to headlines — they may even seek opportunities in times of stress to deploy excess cash. Retail investors, on the other hand, famously redeem simultaneously and at the worst time. Headlines stoke fear, and fear — especially among investors with limited understanding of what they own — has a way of building on itself.

Gates and redemption limits can handle normal outflow patterns. They were not designed for the moment when every retail investor in every vehicle reaches for the exit simultaneously. Even managers who anticipated behavioral risk and built conservative structures can find themselves overwhelmed — not by the size of any single redemption, but by the simultaneity of all of them.

The point isn’t that retail investors are unsophisticated. It’s that the system wasn’t designed for their simultaneity. Same assets, different liabilities, completely different crisis profile.

What This Means

The headlines aren’t wrong about the risks; they’re wrong about what those risks mean. Liquidity friction and credit losses are real and worthy of attention, but they are the ordinary symptoms of a rapidly growing asset class finding its equilibrium. They don’t cause crashes. Leverage does. And the leverage accumulating across this ecosystem isn’t a glaring problem in any single balance sheet, any single fund, or any single headline. It just adds up when you look at the stack as a whole. My hope is that the current stress does what stress is supposed to do: shake out the mismatched and sharpen the rest. My experience says it does — just not completely, and never cleanly.

Most managers focus relentlessly on loan quality — and most are probably right that their underwriting is solid. But we don’t invest in Lake Wobegon, where every loan is above average. And even if yours are, loan quality is the wrong variable when the risk is in the system rather than in your specific credits. Good underwriting matters — it determines how quickly you recover and how much you lose before you do. But in a system-level event, it doesn’t determine whether you’re affected. This isn’t the regional mortgage monoculture of 2006, where ‘dispersion’ meant the difference between Texas and Arizona — there is real dispersion in private credit across sectors and vintages. But SaaS concentration in some portfolios is worth watching specifically, not just as a credit story but as a potential spiral trigger, where mark pressure in one sector creates liquidity demands that ripple across otherwise healthy portfolios.

There’s no precise dashboard for when this becomes critical. DPI and PIK ratios are the closest things to early warning signals about flawed cash flow assumptions. Vintage and covenant quality tell you how much margin for error was built in when the loans were made. But the most important variable isn’t in the loans at all — it’s in who’s funding them.

The investors best positioned through a genuine stress event won’t be those with more vehicle liquidity or even those with the best loans. In private credit, your liabilities are your ultimate risk management. If the fire is coming, look up the hill and ask who built the firebreaks before the season started — and who’s just hoping the wind changes.

Most market writing optimizes for speed and simplicity. Headlines need to be clean. Narratives need to fit in a paragraph. The problem is that markets don’t cooperate — they’re complex, contradictory, and the dynamics that drove outcomes last cycle often don’t apply to this one.

This publication exists for readers who already know that. You won’t find daily calls or data release commentary here. You will find reasons to question prevailing narratives — whether they rest on selectively chosen history, missed regime shifts, fundamental misunderstandings of how things actually work, or data that’s been asked to carry more weight than it can bear.

Sometimes the conclusion is obvious, sometimes counterintuitive, sometimes it raises more questions than it answers. The goal is always the same: cutting through noise that doesn’t need to be there.

Less than 30 minutes before a payrolls report is not the time you want to question your risk models.

Markets had been choppy and I didn’t have a strong view, so I had worked to take my risk down to what I thought was a manageable level into the number. I thought I was lightly positioned. My largest trade was just offsetting positions in different months. Then the head of trading tapped me on the shoulder. Why, he asked, was I carrying more than $1 million per basis point of rate exposure into a major data release?

Thanks for reading The Paulson Perspective! Subscribe for free to receive new posts and support my work.

Subscribe

That is not the question you want right before a market-moving number hits the screen. It turned out a newly formed risk team—brilliant quants applying a regression overlay to firmwide positions—had flagged my position as dangerously exposed.

Their math may have been right, but their conclusions were wrong. As I figured out later that afternoon, they had relied on a narrow historical sample distorted by idiosyncratic moves. The correlation they quantified was significant, but they had mistaken temporary statistical noise for a structural relationship. (It also served as one of many early contributors to some well-earned gray hair.)

It was my first encounter with a lesson I’ve seen repeatedly since: Risk models don’t measure risk. They measure recent behavior—and present it as structure.

We see this play out in the finance industry in several ways: in the way modern finance defines risk, in the statistical tools used to summarize it, and in the assumptions about diversification that quietly sit underneath most portfolios.

The Volatility Trap

The problem starts with how we define “risk.” In modern finance, risk is almost entirely synonymous with volatility. We owe this to Modern Portfolio Theory, and it provided vital insights for an era. Volatility also has a practical advantage: it is easy to observe and quantify, which makes it a convenient foundation for models and risk reports. But for a long-term investor, this definition is often misleading.

If you are saving into an IRA for thirty years, does a 10% decline next Tuesday matter? Not much. The real risk is the permanent loss of purchasing power over decades.

When we define risk solely as short-term volatility—measuring it on a single axis—we create distortions. Private credit funds, for example, are often marketed as “lower risk” than liquid bonds because they are marked to market less frequently. They look smoother on paper. But the underlying credit risk hasn’t disappeared; it’s just less visible.

Optimizing for volatility encourages portfolios that feel comfortable in the short run while quietly accumulating other risks: inflation, liquidity stress, or structural credit deterioration. We focus on the smoothness of the path and ignore the fragility of the destination.

The VaR Mirage

The industry’s favorite security blanket is Value at Risk (VaR). It compresses market complexity into a single number: “95% of the time, losses should not exceed X.”

But VaR is built on history. It assumes tomorrow’s distribution resembles yesterday’s. By focusing on a 95% confidence interval, it treats the rare, extreme days where real wealth is destroyed as statistical noise. George Box famously wrote, “All models are wrong, but some are useful.” It’s a helpful reminder that models simplify by filtering out noise—but in finance, that “noise” is sometimes the only part of the signal that matters.

In the case of VaR, we reduce and offset thousands of different risks into a single estimate. To make that estimate appear stable, the models rely heavily on a second assumption: correlation.

The Correlation Illusion

Correlation, the core concept behind the value of diversification, is the idea that if two assets move differently, combining them reduces risk. But correlations are not set in stone; they are descriptions of past behavior, and they can change quite suddenly.

During calm periods, correlations appear stable. During stress, however, they tend to converge to “1.” Assets that had moved independently begin moving together—usually down. Models reward portfolios with low historical correlation, but low correlation is not structural protection; it is merely recent independence. Confusing the two is the difference between a hedge and a potentially crowded trap.

The “Safe” Duration Trap

For forty years, bonds and equities moved in opposite directions. This relationship became embedded in models as historic fact, not recent anomaly. Investors were told that interest rate risk was “good” risk because it diversified stocks.

That worked until 2021. When inflation returned, bonds and stocks fell together. The result was 2022, the worst year for the “safe” 60/40 portfolio since 1937. Too many modelers failed to recognize that low inflation was a necessary condition for their models to work. When regimes change, the variables the models choose to ignore are often the only ones that matter. In each case—volatility, VaR, correlation—the limitation is the same: the model assumes that yesterday’s relationships will remain intact tomorrow.

Authority Laundering

If models fail when we need them most, why do we rely on them so heavily?

Because they solve a psychological problem: they convert ambiguity into numbers. A risk report is formatted, mathematical, and definitive. Many decision-makers will not act until they see a quantity—even if that quantity is little more than a model assumption expressed with precision.

This is Authority Laundering: the process of taking a subjective assumption, running it through a complex algorithm, and having it emerge on the other side as an objective “fact.”

The model appears to possess authority, but it only reflects the assumptions embedded in it. When the output says a portfolio is “safe,” responsibility shifts from judgment to the model itself. To be clear, models are not useless. In stable conditions, they are indispensable. When I was trading, I relied on them constantly to estimate exposures and keep day-to-day risk within limits. You cannot manage what you cannot measure.

The problem begins when tools designed for normal market conditions are treated as universal truths. Models built to navigate the 99% of ordinary days are suddenly asked to guide decisions during the 1% of days when the regime itself is changing. Authority Laundering doesn’t eliminate uncertainty; it simply hides it. This leads to the Precision Trap: the more sophisticated the model, the more confident the output feels—even when the underlying regime is shifting beneath your feet.

The Hidden Fragility

The most dangerous portfolios are the ones that look safe under a narrow definition of risk. When an entire industry relies on similar volatility and correlation assumptions, positioning converges. Everyone owns the same “optimized” assets.

This False Homogeneity creates systemic fragility. We saw this in 2008, when “safe” AAA-rated tranches were held by everyone because the models said the risk was non-existent. When everyone uses the same map, they all walk into the same hole at the same time.

Models are tools, not maps of the future. In the final piece of this series, I’ll move from critique to construction: how to shift from prediction to resilience, and from point estimates to scenarios.

One of my go-to jokes when I’m asked to make predictions is of three economists who go duck hunting. The first misses five feet to the left, the second misses five feet to the right, and the third yells, “We got him!”

I thought of that joke every December when, like clockwork, the forecast requests began to trickle in. A spreadsheet here, a survey there. By mid-month, the trickle became a deluge. I was expected to fill hundreds of fields with point estimates: Where will the S&P 500 be in twelve months? The 10-year Treasury? CPI? And, for the truly ambitious, where will those variables be each year for the next ten.

Thanks for reading! Subscribe for free to receive new posts and support my work.

Subscribe

I’d love to claim that my forecasts were the result of a proprietary model that hummed in the background of my office, constantly updated and presciently accurate. The truth was closer to throwing them in the air and seeing how they landed. Either method would have had a similar shot at being correct.

Why? Because even the most highly valued forecasts—the gold standards of analytical rigor—are fatally flawed. Even the modelers know it. The future is better described as a probability distribution, yet forecasters are forced to compress a range of possibilities and potential outcomes into a single data point. Frankly, being asked to decide between 2.83% vs 2.82% for Core PCE three years from now is a sharp reminder of the fundamental absurdity of the undertaking—false precision covering wild speculation using infinitely complex variables, shifting relationships, and imperfect data inputs.

The Imprimatur of Authority

It pained me greatly when the head of a sales team proudly placed an aggregated forecast from a client—an average of many forecasts they had gathered—as if it were a holy relic. “This represents the best and brightest!” I was told. “We should be acting on this!”

My own guesses, of course, were buried in that average. What disturbed me was how much stock he, and many others, put in this aggregation of guesswork—not to mention how many billions of dollars were being allocated based on these survey results.

Of course, nobody got it “right.” But accuracy wasn’t really the point. The number carried institutional authority. Once averaged, formatted, and circulated, it stopped looking like a guess and started looking like knowledge. At those moments, I felt great kinship with John Kenneth Galbraith, who famously quipped, “The only function of economic forecasting is to make astrology look respectable.”

The Fortress of False Security

Humans crave concrete answers. Studies in behavioral psychology consistently show that people prefer a wrong answer to no answer at all. Ambiguity is uncomfortable; closure feels safe.

Malcolm Gladwell once noted that drivers preferred SUVs because the high-walled, “cocoon” design made them feel safer, even though that same design often made the vehicles less stable, and with less visibility and maneuverability. We treat forecasts the same way. Forty pages of model outputs filled with decimals feel safer than admitting we are navigating open terrain. We would rather feel “closed in” by a model than have a 360-degree view of the actual risks.

The Incentive to Be Common

There is also a powerful professional incentive to stay behind those walls. For financial consultants, bank treasurers, and many who manage money for others, there is a massive penalty for being “uniquely wrong” and almost no penalty for being “commonly wrong.”

This creates a Consensus Trap. We don’t produce the most accurate forecasts; we produce the most defensible ones. In a world where the real danger is tail risks hurting those who are overly complacent in crowded trades, we should be putting our effort into considering why these “averages” might be wrong, not fine-tuning the average. When a structural shift finally occurs, everyone tries to exit through the same narrow door at once.

The Lamppost Problem

We still need frameworks. Planning provides guidance. Good planning clarifies assumptions and exposes internal contradictions. That exercise has value. But most forecasts are, at their core, structured extrapolations. They begin with historical data, assume that underlying relationships are stable, and then adjust the recent trend at the margin.

This is the “Lamppost Effect”—searching for our keys under the light because the light is better there, even if we dropped them in the dark. We model what is easy to measure, rather than what is important to know. When policy, demographics, inflation dynamics, or capital flows shift structurally, the past becomes a poor map of the future. We are trying to drive a car while making steering adjustments based solely on the rearview mirror. It works fine as long as the road is straight.

Resilience Over Precision

The problem, ultimately, is not the forecast—it’s the faith we place in it.

Relationships will shift. Regimes will change. Tail events will dominate averages. What matters is not whether the number is correct, but what breaks if it isn’t.

A more useful question than “What will CPI be in 12 months?” is: If CPI moves in a direction the forecast never imagined, does my portfolio—or my strategy—survive? If everybody shares my forecast, are there additional risks created by the consensus itself being wrong? That is where real preparation lives, and where the true investment work begins.

In the next piece on modeling, I’ll examine how this same precision trap shows up in financial risk models—how VAR, correlation assumptions, and option-adjusted spread frameworks create the illusion of control while quietly embedding fragile assumptions. The math may be sophisticated. The certainty is not.

Many people still think of the Fed’s primary role as managing the business cycle. In fact, the Fed has shifted back to focusing on its original mission - ensuring the stability of the financial system. With only two recessions (not counting the brief COVID shutdown) in the last 35 years, the risks have shifted away from those of suboptimal growth and towards the systemic. The Fed has responded accordingly.

We have established that the Fed Funds rate doesn’t impact the economy the way it used to, and the economy itself has changed such that the old “cyclical manager” role of the Fed has become largely obsolete. But as we learned in 2008, systemic financial risks pose far greater economic dangers than traditional inventory or capex cycles ever did.

Thanks for reading! Subscribe for free to receive new posts and support my work.

In response, the Fed has undergone a quiet but radical evolution. They have moved from being a “Manager of Growth” to a “Stabilizer of the System.” Their primary focus is no longer fine-tuning GDP through the Fed Funds Rate; it is maintaining the stability of a massive, complex, and increasingly interdependent financial system.

For all of the focus on the Fed’s impact on monetary policy, the Fed was created as a Banker to the Banks, and tasked with ensuring the stability of the system, along with setting monetary policy. That role evolved along with the economy, particularly as recurrent boom/bust cycles and the high inflation of the 1970s required active monetary policy, but as the risks have shifted back, the Fed is being recast in its most fundamental role: keeping the financial machinery working smoothly so the real economy doesn’t seize up.

The New Toolkit: Targeted Stabilization

Working hand-in-hand with the Treasury Department in 2008, the Fed built a host of new tools to respond to the unique challenges of the day. Since that time, they have grown their balance sheet dramatically while responding to a variety of flare-ups, including dollar liquidity issues with other central banks, markets freezing around COVID, and the 2023 bank failures.

None of these issues would have been resolved with a lower Fed Funds rate. Too many people still assume that cutting the rate is always the answer, but the 2023 failures of Silicon Valley Bank and First Republic proved that the problem was a liquidity leak, not a monetary one. While pundits called for rate cuts, the Fed instead utilized the Bank Term Funding Program (BTFP)—a targeted tool designed to stabilize bank liquidity without having to flood the entire economy with unnecessary monetary stimulus.

This toolkit is now expansive, including the Standing Repo Facility (SRF) to keep overnight markets liquid, and various facilities used in 2020 to stabilize money markets and corporate credit. These tools aren’t about driving growth; they are about preventing a systemic heart attack by targeting specific vulnerabilities.

The “Systemic Put”: A Necessary Evil?

People have speculated about the existence of a “Fed Put” for decades—the belief that the Fed will always step in to save the market. But in today’s economy, where asset prices impact the financial conditions the Fed monitors so closely, the Put has likely become much more of a reality.

It’s not about “saving billionaires.” It’s about the fact that in a hyper-financialized system, asset prices are the collateral that fuels everything from global repo markets to small-business credit lines. If that collateral loses too much value too quickly, we risk a spiral of de-levering that could be incredibly damaging. (The de-levering death spiral was the true issue in the much-misunderstood 2008 crisis—I’ll have more on that in other posts).

Now in 2026, the relevant conversation for investors is no longer whether the Put exists, but how “out of the money” the strike price is. How much market pain will the Fed tolerate before they decide the system itself is at risk?

Note that the Fed does not want the market believing that such a Put exists, as they believe it might encourage unhealthily risky behavior (Moral Hazard). However, the market now firmly believes in it, and so far, the Fed has blinked first—at least with regard to sending calming messages whenever markets begin to panic.

The Warsh Transition: A Pressure Test

For all the talk about geopolitical risks and disruptive AI, the most imminent wildcard is the potential regime change at the Fed. With Kevin Warsh nominated to succeed Jerome Powell, the era of reliable excess liquidity in the financial system is likely to come to an end.

Warsh has long been a vocal critic of what he calls “balance sheet bloat” and has suggested working with the Treasury to limit the Fed’s role and shrink its footprint. The Fed has already made strides, shrinking the balance sheet from a peak of nearly $9 trillion down to roughly $6.6 trillion as of early 2026. While the Fed has signaled that it believes the “optimal” size is still considerably larger than pre-2008 levels, Warsh will undoubtedly work to reduce that number further.

But the market has been skittish about every modest reduction thus far, including several liquidity panics since 2018. Funding markets are akin to breathing—you take it for granted until it’s a struggle, at which point you move immediately to panic. Fortunately, the Fed has more tools to address liquidity than just the balance sheet, and I can’t imagine Warsh will suddenly decide he wants to be the Chair who collapses the system over an ideological dislike of balance sheet size. However, I do anticipate some tests of just how much the market can take. This likely means headlines and volatility from a market that has proved quite prone to overreactions.

The New Frontier: Risks Outside the Map

Agree or disagree with what the Fed has done (usually a combination of both), we remain in deeply uncharted territory. Markets will panic any time something new happens, and it’s very likely the existing tools will eventually be inadequate.

What concerns me is that many of the new risks are structural and likely immune to the blunt tool of interest rates. Addressing them will require a combination of existing surgical tools and, likely, the development of new ones on the fly:

· The Private Credit Black Box: With over $2 trillion in unregulated debt, we have a massive pool of credit that doesn’t answer to the same rules as banks. Is it a shock absorber or a “hidden time bomb” of deferred defaults that current tools can’t reach?

· The AI/Tech Concentration: Five companies now drive more capex spending than entire sectors of the old economy. A sudden shift in the AI cycle would be a systemic shock that a lower Fed Funds rate cannot “fix.”

· The Demographic Wall: We are in a “low-hire, low-fire” environment where old unemployment models don’t apply. What if we hit a wave of white-collar AI job replacement in a market with exceptionally low job turnover?

Conclusion: The Limits of the Map

We have to accept that the Fed’s evolution from a growth manager to a systemic stabilizer is a permanent shift. They have built an impressive array of tools to keep the system functioning, but those tools are only as good as the information the Fed receives.

As we move into a new era of leadership and potential balance sheet reduction, the new plumbing and maintenance tools the Fed has built will be tested. If we want to stay grounded, we need to stop hyperfocusing on the Fed Funds rate, or the latest historical scare chart of the day, and start paying attention to the actual health of the financial system—focusing on liquidity and credit spreads rather than some dots on a chart.

Ultimately, no amount of intervention can permanently suspend the laws of human nature. Speculative cycles will still happen, and things will still break. At the same time, we are NOT heading back to a 2008 collapse. Whatever crash happens next, it will look very different, with different causes, different problems, different outcomes. The future belongs to those who recognize that the old maps are no longer accurate and are willing to navigate by what they actually see, rather than what the models tell them they should see.

The Series Wrap-Up: This three-part series has explored why the old maps of the 20th-century economy are leading us astray. The Fed has adapted by becoming a stabilizer of the financial plumbing, but as we move into a new era of leadership and potential balance sheet reduction, that plumbing is about to be tested.

Coming Soon: Why I Hate Forecasting — and why our obsession with point estimates may be doing more harm than good.

For almost a decade now, the consensus has been waiting for a crash that hasn’t arrived. Since the Fed first began to hike out of their zero-rate policy in 2018, we have been treated to a non-stop parade of recession calls from analysts, TV talking heads, and all sorts of recession-predicting models. Invariably, when the data fails to comply, the goalposts move.

I’ve been wrong plenty of times, but I always try to figure out what I missed. Sure, “this time is different” is often considered the most dangerous phrase in the world of investing, but at what point in a series of false recession signals do we start to question the models? They are still looking for the ghost of 1976 in the 2026 data. It feels like listening to a sports announcer trying to call today’s game with yesterday’s playbook.

Why have they been so wrong? The world has changed in critical ways from the world their models are based on. The economic cycle has lengthened, if not decoupled entirely from the traditional “boom/bust” manufacturing cycle of the 20th century. More notably, the role interest rates play in the economy has shifted fundamentally.

1. From Capex to IP: The Shift in Sensitivity

The textbook boom/bust cycle describes a Capex-heavy world. Factories and heavy machinery tied up capital and required substantial, rate-sensitive planning. Capacity tended to be overbuilt during boom times, leading to painful downturns while excess capacity was shuttered. You cut spending, laid off workers, and triggered a recession.

But the modern US economy is driven by Intellectual Property, software, and services. These sectors are far less capital-intensive. Tech titans don’t need the bond market to fund their next innovation—they fund it out of massive, existing cash flows. For these companies, 5% rates aren’t a headwind; they are a tailwind for their massive cash piles. Unlike banks, companies with high profit margins and exponential growth models don’t make decisions about deploying cash based on marginal changes in interest rates. We have moved from an economy that borrows for business investment to one that scales through code and high-margin services.

2. Fixed-Rate Households and the Labor Moat

On the household side, the impact of rates on monthly spending and credit expansion is greatly diminished. In the 50 years leading up to the 2008 crisis, household debt climbed from roughly 25% of GDP to nearly 100%. Credit expansion boosted growth for decades, with lower rates fueling demand during every dip. But eventually, the US consumer simply ran out of capacity to borrow more. Household debt has actually decreased to around 71% of GDP since that 2008 peak, despite historically low rates for much of that time.

Furthermore, the US consumer is largely protected by the 30-year fixed-rate mortgage. Roughly 90% of US mortgages are fixed. While the Fed was hiking to “cool the consumer,” the average homeowner’s largest monthly expense didn’t budge. With most outstanding mortgages well below prevailing rates, even significant declines in rates won’t trigger the massive refinancing activity that used to inject cash into the economy.

Then there is the Phillips Curve, which links rates to unemployment. It is increasingly seen as a historical artifact. A tight labor market has immunized employment against rate hikes, leading to an era of “labor hoarding.” Whether the Phillips Curve should be discarded or if this is an anomaly of pandemic disruption is up for debate. But it’s a strong reminder: stop looking backwards and saying, “This is how it always was,” when every key measure of the system has changed.

3. The Private Credit Black Box

While traditional bank lending has been hampered by regulations, private credit has stepped in to fill the gap. Private credit is sensitive to the Fed Funds rate—most loans are floating-rate—but it behaves differently than the banking sector. Private lenders don’t answer to the same regulators and don’t face bank-run risks.

Crucially, private lenders have shown a flexibility for working with borrowers during stress that banks simply cannot match. A surge in loan modifications, including PIK (Payment-in-kind) structures, has likely suppressed the actual default cycle. Whether these modifications are merely deferring pain or representing a new, more durable credit model remains to be seen. But expecting the economy to mirror prior cycles while ignoring this shift in credit delivery is a mistake.

4. The K-Shaped Heat (Asset Prices as the New Dashboard)

Interest rates still matter, but they matter less, and the way they move through the system has changed. Because asset prices now drive so much of the spending in our economy (the “Wealth Effect”), the Fed has been forced to broaden its focus. They now use Financial Conditions—a metric heavily influenced by the stock market and credit spreads, in addition to interest rates—as their primary dashboard to see how healthy the system is.

But while asset prices impact both spending and access to capital, they don’t impact everyone equally—just as with the lone boiler trying to heat a 19th-century building. A tech billionaire may out-earn a substantial portion of the population on a good day in the market. The resulting imbalance is the “K-shaped economy”—an inevitable consequence of the fact that asset prices now move the needle on the economy more directly than the Fed Funds rate.

The Takeaway

The people waiting for the “inevitable” crash are misreading the temperature because they are using an old thermometer. The insulation of fixed-rate debt, the shift to IP-heavy margins, and the rise of private credit have created an economy that is far more resistant to the Fed’s traditional tools than the consensus appreciates.

However, we have entered uncharted territory. The system has changed, the “breaks” are in different places, and models should be in “low confidence” mode. In fact, the Fed has implicitly shifted its focus away from the “thermostat”, and toward a much more critical priority: Financial System Stability. If the boiler can’t heat the building efficiently anymore, the Fed’s real job is making sure the pipes don’t burst.

What’s Next: If interest rates no longer move the needle for the average family or CEO, has the Fed become irrelevant? Hardly. It has simply changed its job description.

In the final installment, Part 3: The Systemic Pivot, we’ll explore the Fed’s evolution into the “Chief Janitor” of the financial system—and why the most important tools in their kit are no longer the ones you see on the news.

The financial media feeds its audience a steady diet of narrative junk food, and all too often, the Federal Reserve is the primary ingredient. We’ve been conditioned to believe the Fed has an “easy button” thermostat that can crank the US economy up and down at will. While that fits a primal urge for simple explanations, it is increasingly detached from reality.

It is a lazy narrative that leads to expensive conclusions.

To be clear: there has never been a direct, mechanical link between the cost of overnight money and the decisions made by a family in Ohio or a CEO in Silicon Valley. It affects bank decision-making and the cost of overnight money—things that matter, but don’t impact the economy the way they used to.

For evidence, look no farther than the post-pandemic cycle. The Fed hiked rates by more than 500 basis points—accompanied by dire predictions of imminent recession—yet unemployment stayed at historic lows and the S&P 500 hit new highs. The old understanding of how things worked said the economy should have frozen over. It didn’t. The “plumbing” of how monetary policy moves through the economy has changed, yet the obsession with the Fed’s thermostat dial persists.

No “Right Rate” for a Diverse Economy

A closer analogy for our current economy isn’t a modern Nest thermostat; it’s my first New York apartment. It was a 19th-century building with one central boiler in the basement. When the tenants on the top floor complained they were freezing, the super would crank the boiler. The result? The lower floors were sweltering with their windows wide open in January, the top floors were still cold, we wasted energy, and nobody was happy.

The Fed is currently that super. The US economy is a sprawling complex of different structures, each with its own insulation. There is no single “right rate.” A 5% rate might be a death sentence for a small manufacturer relying on a floating-rate bank line—the freezing tenants. Meanwhile, a Big Tech titan sitting on $50 billion in cash gets a nice tailwind from higher interest income.

The Short Lever

The biggest flaw in the “Fed Obsession” is the belief that the Fed controls the rates that actually matter. Roughly 70% of U.S. household debt is tied to 30-year fixed mortgages. Corporate borrowing and auto loans are largely tied to 5- or 10-year rates—none of which are tethered to the Fed in the way people assume.

We saw this throughout 2024 and 2025: the Fed began cutting the overnight rate, but mortgage rates actually rose. The Fed’s lever is simply too short to reach the parts of the economy that move the needle for most Americans.

The Affordability Trap

Nowhere is this “lazy narrative” more obvious than in housing. The common cry is that lower rates will solve the affordability crisis. But housing is a structural problem—we simply have not built enough homes.

You cannot fix a supply shortage with a monetary knob. Lowering rates in a supply-starved market doesn’t build houses; it just reprices a scarce asset. Lower rates can actually make it harder for first-time homeowners by inflating prices and requiring even larger down payments.

The Political Game and Fiscal Dominance

If lower rates won’t fix housing or trigger a growth boom, why is the pressure on Jerome Powell so relentless? The answer lies in the math of the federal deficit.

We have entered an era of Fiscal Dominance, where the sheer magnitude of U.S. national debt dictates policy. We now spend more each year servicing existing debt than we do on defense. Almost one-third of all outstanding marketable Treasury debt—roughly $10 trillion—is set to mature in 2026.

The government isn’t asking for a growth miracle; they are asking for a cheaper credit card payment to maintain their deficit spending.

The Bottom Line: SO WHAT?!

The Fed can try to set the temperature, but the markets determine where the heat actually goes.

Stop obsessing over the timing of a 25bp move. I can’t count the number of times I’ve sat in meetings where people report some small change in Fed path expectations, as if this matters more than major economic developments. No. A 25bp Fed Funds rate change will impact a levered bond trader, even a private equity borrower, but it makes almost no difference to the health of the broader economy.

The health of the economy depends far more on navigating tariffs, the regulatory environment, and the massive fiscal deficit than on the rate at which banks lend to each other overnight. Focusing on the 25bp move while ignoring the $10 trillion debt wall is the definition of missing the forest for the trees.

What’s Next: Identifying that the “thermostat” is broken is only half the battle. If the Fed’s primary tool no longer moves the needle, it’s because the economy has developed a thick layer of insulation.