Financial Fables

From a Princess to a Frog: How is AI Changing Tech Firms?

Financial Fables — Sun, 26 Jul 2026 14:31:24 GMT

Fairy tales are full of princes and princesses being transformed into animals by wicked witches. A handsome prince becomes a beast. A royal heir is reduced to an ordinary frog. The spell is eventually broken, and they reclaim their rightful place.

But financial markets, however, rarely follow fairy-tale endings and the rise of AI may be casting a very different spell.

Some of the technology companies, long regarded as the handsome princes of the digital age, are finding that the AI magic could be turning them into frogs.

The Tech Princes

When we think of technology companies, our minds often jump to Silicon Valley start-ups - small teams of coders building the next disruptive app from a garage.

Investors see something rather different. Today’s technology sector is dominated by royalty: companies like Google, Apple, Amazon and Meta. These are firms with fortress-like market positions, extraordinarily high profit margins and balance sheets overflowing with cash.

Their kingdoms are built on intangible assets rather than physical ones. Intellectual property, software, algorithms, customer relationships and vast datasets have become far more valuable than factories, warehouses or aircraft. Aside from their growing cash mountains, many of these businesses have historically owned surprisingly few tangible assets.

This asset-light, cash-generative model became the defining characteristic of the Magnificent Seven. They were the undisputed princes of the digital economy - commanding their industries, dominating investor portfolios and appearing almost impossible to dethrone.

The AI Curse?

The issue is that the rise of AI is quickly changing the nature of modern technology firms. The huge AI infrastructure buildout is fundamentally shifting the business model from an asset-light structure to one characterised by physical assets: data centres, cooling systems, networking equipment and power infrastructure.

And this buildout of hard assets is expensive. Take Alphabet, Google’s parent company, which released its second-quarter earnings this week. We can already see a marked deterioration in its free cash flow - the cash left over after funding both day-to-day operations and capital investment. As shown in the chart below, free cash flow has turned negative for the first time on record and it is expected to stay there as the firm continue to spend billions in the AI race.

Now, this shift is not in itself a problem. Investments made today that generate financial returns tomorrow are often exactly what shareholders should want to see. But it does start to put a few warts on the faces of these once-handsome tech princes.

As the chart below shows, the debt piles of these once-Magnificent Seven have begun to surge in order to fuel expensive AI buildouts. Furthermore, firms such as Meta and Google are actively contemplating equity raises - something that would have been almost unthinkable just a few years ago - in order to keep AI investment flowing.

Whether this AI spending ultimately pays off remains to be seen. But it does raise a more fundamental question for investors. If technology companies are becoming more capital-intensive, should we continue to value them as the asset-light businesses they once were?

How to Value Frogs

If asset-light princes are being turned into infrastructure-heavy frogs, then investors may need to think more carefully about how these companies should be valued, particularly firms such as Alphabet, Amazon and Microsoft, which are investing heavily in the physical infrastructure behind AI.

For years, these companies were rewarded with premium valuations because they combined rapid growth with remarkably asset-light business models. Once the software was built, it could be sold to millions of customers at relatively little additional cost, allowing profits and cash to compound with limited ongoing investment.

But data centres are not one-off investments. They require constant maintenance, consume vast amounts of power and, perhaps most importantly, have a finite economic life. Servers, networking equipment and AI chips must all be replaced as technology advances. In other words, today's capital expenditure risks becoming tomorrow's maintenance expenditure.

That doesn't immediately make these tech companies bad businesses but it does mean they start to share some of the characteristics of infrastructure providers and utilities - businesses that require continual capital investment simply to maintain their competitive position.

And that matters because investors have historically valued those businesses differently. Asset-light technology firms have commanded premium valuation multiples because they can grow with relatively little additional capital. Infrastructure-heavy businesses, by contrast, have typically traded on lower multiples because a larger share of their cash flows must be reinvested just to keep the engine running.

So it seems the real curse of AI is that it slowly turns them into infrastructure-heavy frogs - meaning investors may no longer value them like royalty.

Gone Fishing

As long-term subscribers will know, I usually take August off from writing to recharge and focus on a few other projects.

I’ll be back in September with more Financial Fables, market insights and updates. Until then, thank you for reading, and I hope you have a fantastic summer.

See you in September!

Subscribe now

Disclaimer: The information provided in this blog post is for general informational purposes only and should not be construed as financial, investment, or professional advice. The content is not intended as a solicitation, recommendation, endorsement, or offer to buy or sell any securities or financial instruments. Any reliance you place on such information is strictly at your own risk. Always consult with a qualified financial advisor or professional before making any investment decisions. The author and the website assume no responsibility for any losses or damages that may result from the use of or reliance upon the information provided in this blog post.

What is the Weakest Link in the AI Chain?

Financial Fables — Sun, 19 Jul 2026 14:31:16 GMT

There has been a lot of debate in recent weeks about what eventually ends this equity bubble. As we’ve discussed before, equity manias don’t simply die of old age. They are either murdered by the Fed, finished off by a recession, or they commit suicide through their own excesses.

The problem with this bubble is that it appears remarkably immune to the real world. The prospect of higher interest rates, stubborn inflation or even soaring energy prices has done little to cool the enthusiasm surrounding AI. That has led me to an increasingly uncomfortable conclusion: if this rally ends, it probably won’t be because of something happening outside the AI ecosystem. It will end because something inside it breaks.

So where is that weakness most likely to emerge? Which company is the weakest link in the AI chain?

The Messy AI Ecosystem

Before we identify the weakest link in the chain, it helps to understand what the AI ecosystem actually looks like. The diagram below captures a snapshot of the relationships between the major players.

The orange arrows show capital that has already changed hands, whether through equity investments or infrastructure commitments. The grey arrows show the value of future purchase commitments made in return. For example, Amazon has already invested around $13 billion into its partnership with Anthropic. In exchange, Anthropic has committed to purchasing up to $100 billion of cloud infrastructure from Amazon over time.

First things first - yes, it’s an absolute mess.

Which is precisely why the market struggles to analyse it. Capital, infrastructure commitments and commercial agreements are criss-crossing the industry in every direction. It’s also far from transparent. You’ll notice several lines on the diagram without any numbers attached. That’s not a mistake. Those represent relationships where we know a commercial arrangement exists, but nobody outside the companies has a reliable estimate of how much money has actually changed hands.

So where are the weak points?

The obvious place isn’t the chipmakers. As we’ve discussed before, AMD and Nvidia are making a fortune selling the pickaxes and shovels needed for everyone else’s AI gold rush. Likewise, infrastructure providers such as CoreWeave are largely being paid to supply computing capacity rather than taking the same strategic bets as their customers.

Instead, your attention should be drawn to the companies across the top of the diagram: the hyperscalers. Four of them look broadly familiar. One of them doesn’t.

That company is Oracle - more specifically, it’s Oracle’s relationship with OpenAI that deserves much closer scrutiny.

The Weakest Link

Why does this relationship deserve our attention?

Firstly, while there are a lot of eye-watering numbers floating around the AI ecosystem, this is by far the largest commitment. Oracle has already deployed around $7 billion building data centre capacity for OpenAI. In return, OpenAI has pledged more than $300 billion of future infrastructure spending over the coming years.

The difference is that Oracle isn’t like the other hyperscalers. Amazon, Microsoft and Google entered the AI race with fortress balance sheets and enormous cash flows. Oracle didn’t. It entered this race with an already stretched balance sheet. The chart below shows Oracle’s debt-to-equity ratio compared with many of its peers.

This leaves Oracle with very little room for error. Any delay to projects, slowdown in AI demand or disappointment in expected revenues would quickly make additional financing more expensive and place further strain on an already leveraged balance sheet.

A significant part of the Oracle AI story ultimately rests on one assumption: that OpenAI follows through on its enormous infrastructure commitments.

That assumption is becoming a serious point of concern. OpenAI lost more than $38 billion in 2025 and, at the same time, appears to be losing ground in the battle with Anthropic.

The chart below compares the annualised revenues of the two firms based on public funding disclosures. Unless OpenAI has delivered extraordinary revenue growth since then, the gap appears to be narrowing far faster than many investors appreciate.

Why does this matter?

Because Oracle has spent billions today on the expectation of hundreds of billions of dollars arriving tomorrow. If OpenAI fails to generate the cash flows the market expects - or simply scales back its infrastructure ambitions -those future commitments become far less certain. According to the credit rating agency S&P, half of the revenue Oracle expects to earn in the future is signed contracts tied to OpenAI.

So if OpenAI can’t fulfill its promises, it would leave Oracle with a massive investment built for a customer whose economics no longer stack up and a huge amount of its future revenues would evaporate.

If that were to happen it would send shockwaves through this AI bubble as investors begin to panic.

Much of today’s AI boom is built on capital being spent today against promises of revenues many years into the future. If confidence in just one of those promises begins to crack, investors won’t stop to ask whose promise it was. They’ll start questioning the entire chain.

Subscribe now

Why Markets Don't Give a Sh*t about the Economy

Financial Fables — Sun, 12 Jul 2026 14:30:36 GMT

There was a time when the stock market was widely viewed as a barometer of the economy. Strong growth meant rising profits and higher share prices; recessions meant falling markets. It wasn’t a perfect relationship, but it was intuitive.

Today, that relationship appears increasingly difficult to reconcile. Equity markets continue to scale new highs while many of the indicators we traditionally associate with economic strength tell a much less optimistic story.

Has the stock market stopped caring about the economy?

The great divergence

The equity-economy divergence is starting to show up in some odd place. The below should consumer confidence vs S&P 500. Theoretically, the equity market should care about how households feel about their financial prospects. A nervous consumer makes for a reluctant spender!

And for a long-time in the post-2008 era, this relationship held fairly well but since 2021/22 something seems to have broken.

Nor is this just contained to consumer confidence. The chart below shows total job openings vs S&P 500. Firms typically advertise and hire for positions during the good times - lowering unemployment, boosting growth and helping equities. But again since 2021/22, that relationship appears to have broken down.

Why equities don’t give a sh*t anymore?

Many commentators would acknowledge the divergence and suggest that it reflects the fact that today's equity markets simply looks different from the broader economy. Technology companies account for a much larger share of market capitalisation than they once did, while manufacturing and other labour-intensive industries have become less prominent.

Perhaps there is some truth in it, but it is far from the whole story. Indeed, in the late 70s, U.S markets were dominated by energy firm. Later on in 1999, tech dominated the index with a roughly 30% exposure. Neither of these periods triggered the economic-equity divide we are now witnessing.

So why is this happening? I think the explanation goes deeper than changes in sector composition. It reflects a more fundamental shift in the objectives of listed companies.

For much of the post-war period, corporate success was closely tied to expanding productive capacity. Firms invested heavily in factories, equipment, research and development, and growing workforces. Management could justify lower profits today if those investments generated higher profits tomorrow. As companies expanded, so too did employment, wages and output, creating a natural link between the fortunes of the stock market and the health of the wider economy.

It was this close relationship that underpinned one of Warren Buffett’s most famous valuation measures: the Buffett Indicator, which compares the total value of the stock market with the size of the economy. The intuition is straightforward. If listed companies grow alongside the economy, then over the long run the value of the stock market should broadly track GDP.

But today, the Buffett Indicator has got unhinged from the underlying economy because the objective of firms has increasingly shifted from maximising output to maximising returns on capital.

Rather than owning assets, firms lease them. Rather than employing large workforces, they outsource. Investment in physical infrastructure has given way to software, intellectual property and other intangible assets. Excess cash is increasingly returned to shareholders through dividends and buybacks rather than being reinvested into expanding capacity. A company can now create enormous value for shareholders without materially increasing employment or physical investment, weakening the relationship between equity prices and traditional measures of economic strength.

How did this come to pass? It was all due to a lack of patience. Pressure on investors to deliver returns here had a knock-on effect of pushing management teams to deliver quarterly earnings, improve margins immediately and return excess cash to investors. These decisions often make perfect sense from an equity valuation perspective, but they do not necessarily maximise long-term economic growth. As a result, the interests of financial markets and the broader economy have become increasingly separated from one another.

Why does this mean for this rally?

The current equity mania will not die of old age. Something will eventually bring it to an end.

Historically, that “something” has often come from outside the market itself. Energy shocks, aggressive Federal Reserve tightening, recessions or geopolitical crises have all been sufficient to puncture periods of market exuberance.

But today’s market feels different. Those traditional measures of economic weakness may no longer be enough on their own. Markets have become increasingly driven by expectations surrounding AI, capital allocation and future earnings rather than the current state of the economy.

If that is true, then the greatest threat to this rally is unlikely to come from disappointing GDP growth or weakening consumer confidence. Instead, it is more likely to emerge from within the AI ecosystem itself. That could take several forms: aggressive accounting practices beginning to unravel, overinvestment in AI infrastructure creating excess capacity, or AI simply failing to achieve the widespread corporate and consumer adoption needed to justify today’s extraordinary valuations.

That's why the next bear market probably won't begin with a recession - it will begin when investors stop believing in the AI story they've been telling themselves.

Subscribe now

What Will Pop This Equity Bubble - Part II?

Financial Fables — Sun, 05 Jul 2026 14:30:05 GMT

Every asset bubble is different, but what pops them is not.

Back in 2025, I put together some analysis showing that, throughout history, equity bull runs have typically been ended by one of four pinpricks: a recession, rising interest rates, speculative mania, or an exogenous shock.

As the chart below shows, before the 1990s equity bull runs were typically brought to an end by either Fed tightening or the onset of a recession. More recently, however, that dynamic appears to have shifted. Bull markets have tended to last longer and have become seemingly more impervious to real-world factors such as recessions or Fed rate hikes.

With the current bull market now ranking as the third largest in nearly 70 years, some investors are starting to wonder what will eventually pop it.

Can the Fed pop this bubble?

Historically, the Fed has been one of the most common bubble poppers, helping to bring an end to four major equity bull runs since the 1930s. Could it pop this one?

The newly appointed Fed Chair Warsh struck a notably hawkish tone in his first press conference. Market expectations have shifted from anticipating two or three rate cuts over the next year to now pricing in rate hikes over the next 12 months. It’s a significant swing in expectations, yet equity markets have largely shrugged it off.

Why? This equity bull run is taking place in what I term a “post-capitalist” financial system, whereby financial markets now dwarf the real economy and are fundamentally different in their sector composition. This transition, which has accelerated over the past decade, makes it harder for the Fed - or indeed any real-economy shock - to pierce an increasingly disconnected equity market.

Further evidence of this disconnect can be seen in the muted equity market reaction to the recent spike in oil and gas prices. In the 1970s, a surge in energy prices was enough to bring an equity bull run to an abrupt end. Today, the response has hardly triggered any reaction.

A pop from within

So, if external forces won’t be enough to pop the bubble, the catalyst is more likely to come from internal imbalances.

Exactly what that weakness will be is up for debate, but here are three things that keep me awake at night.

1. Aggressive - or even fraudulent- accounting practices

In recent weeks I’ve focused on some of the more aggressive accounting practices emerging across the AI sector. Google and Amazon, for example, have aggressively marked up the value of their stakes in Anthropic. OpenAI has also attempted to reduce reported losses by moving more than $20 billion to parent entities.

Another concern is how many tech firms are accounting for semiconductor assets. Many are treating chips as long-term assets that depreciate gradually over a decade. The technological reality, however, is that many of today’s cutting-edge chips could be obsolete within just two or three years, potentially requiring substantial write-downs.

Nothing I’ve seen so far could fairly be described as fraudulent, but it can certainly be described as aggressive accounting that leaves little room for error if AI earnings begin to disappoint.

2. AI Over-supply

In the late 1800s America experienced multiple railway bubbles. Convinced that rail transport would transform the economy, investors poured vast sums into railway companies and infrastructure, often financing routes that ultimately proved commercially unsustainable. The resulting boom gave way to a series of bankruptcies and financial crises, including the Panic of 1873. Yet the overinvestment wasn’t wasted - much of the track ultimately became the backbone of America’s industrial expansion.

As you can see from the below chart, history is littered with examples of genuinely transformative technologies that investors become overly excited about, pouring in capital only to become disappointed when adoption fails to justify the eye-watering expenditure in the near term - everything from the railway boom to

But this latest AI mania looks like a bubble on a much grander scale. The five largest hyper-scalers alone are expected to invest more than US$1 trillion in AI infrastructure between 2025 and the end of 2026, according to BIS estimates. The sheer scale of this spending dwarfs that seen during previous technology booms.

3. AI Demand Disappoints.

Perhaps the greatest risk is that AI demand simply does not come through.

Many disruptive technologies follow a J-curve, with years of heavy losses before widespread adoption allows firms to become profitable. Uber is a good example. The company generated cumulative losses of around $35bn over eight years, subsidising cheap rides to get consumers hooked on its product. As adoption took hold, Uber was eventually able to capture market share, reach sufficient scale, gain pricing power, and ultimately turn a profit.

AI firms are attempting the same strategy today - running large losses to subsidise access and get individuals and corporations hooked on their products. The difference is that the depth of the J-curve is far greater than Uber’s ever was. OpenAI lost around $35bn in just 12 months- the same amount that Uber lost over eight years.

If those earnings fail to materialise, or customers prove less willing to pay than investors currently expect, today’s extraordinary investment could quickly begin to look excessive. History suggests that bubbles are rarely popped because the underlying technology fail - they are popped because expectations run too far ahead of reality.

Subscribe now

Are Financial Markets Infected by a Tech Virus?

Financial Fables — Sun, 28 Jun 2026 14:31:02 GMT

In the closing days of 1999, the world waited nervously for midnight.

Across boardrooms, government offices and data centres, fears grew that a computer bug known as the Y2K virus would cripple the global economy. Because many computer systems stored years using just two digits, there were concerns that when the calendar rolled over from 1999 to 2000, computers would interpret the new year as 1900, triggering widespread failures in everything from banking systems to power grids.

Fortunately, some quick computer fixes and the crisis was averted.

But 26 years later, financial markets are confronting a very different kind of tech virus. This one is not infecting computers - it is infecting portfolios.

The AI revolution has turned technology from just another sector into the dominant driver of financial markets. Whether through index funds, pensions or active portfolios, investors are increasingly making the same underlying bet: that technology will continue to outperform.

Unlike Y2K, this virus cannot be patched. It spreads by concentrating more and more portfolios around the same companies, the same theme and the same source of risk.

A Tech Virus

For much of last year, I watched the AI rally with relatively little concern. Valuations were undoubtedly becoming stretched, but earnings were growing rapidly and there was at least a plausible case that artificial intelligence could justify much of the optimism.

What has changed is not simply the price of technology stocks. It is the spread of technology exposure throughout the financial system.

The next wave of mega IPOs is likely to accelerate that process. With OpenAI, Anthropic and SpaceX all expected to become among the largest companies in the market, passive investors will have little choice but to own even more technology. As these firms enter the major benchmarks, they will displace the few remaining non-tech giants at the top of the index (J.P. Morgan, Eli Lilly and Berkshire Hathaway), leaving the S&P 500 with a nearly 40% exposure to one single technological ecosystem.

This is what makes today’s concentration unusual.

Markets have always had large companies. In the 1980s, leadership was shared between oil majors, banks, industrial firms, retailers and telecoms. During the 2000s, energy and financials became more influential. Concentration itself is nothing new.

What is new is that today’s concentration revolves around a single theme. The companies may have different products and business models, but they are increasingly exposed to the same technological cycle, the same capital spending boom and the same AI narrative.

That is how a financial virus spreads - not because one company becomes larger, but because more and more portfolios become exposed to the same underlying risk.

Worldwide Tech Virus

Like any successful virus, this one has not remained confined to its original host.

What began as a concentration within U.S equity markets has spread across global financial markets. Investors looking at emerging markets or into private assets may believe they are diversifying, but many are simply gaining exposure to the same underlying AI and technology ecosystem through a different route.

The chart below illustrates the challenge. U.S equities, emerging markets and private assets are becoming increasingly tied to the same technological theme, making genuine diversification harder to achieve. The labels on the portfolio may be different, but the underlying source of risk is becoming increasingly similar.

Like any virus, prevention is easier than cure.

Completely avoiding technology exposure is becoming increasingly difficult, but investors can still install a few “firewalls” to stop it spreading unchecked through their portfolios.

1. U.S Equities: Equal Weight or Market Cap?

For passive investors, one option is to look beyond traditional market-cap weighted indices.

Market-cap weighting automatically allocates more capital to the largest companies as they grow. As technology firms continue to dominate index performance, investors may find themselves becoming increasingly concentrated in a handful of names without ever making an active decision to do so.

Equal-weighted strategies offer one way of slowing that process. While they are no silver bullet, they spread exposure more evenly across companies, reducing dependence on the largest technology firms and providing broader participation across the market.

2. Bonds, Cash and Portfolio Ballast

The same concentration is far less evident in fixed income markets.

Investment-grade and high-yield corporate bonds still have relatively limited exposure to technology, despite the surge in debt issuance from the mega-cap firms (see the earlier chart). Carefully selected bond allocations can therefore provide an additional layer of diversification.

An even less glamorous solution is cash. Holding cash has been unfashionable for much of the past decade, but maintaining liquidity can provide investors with optionality, helping reduce overall portfolio concentration while preserving capital should opportunities arise.

3. Can Gold Save Us?

Not every firewall is proving as effective as investors hoped.

At the start of the year, many institutions were discussing raising strategic allocations to gold from around 1% towards 5% or even 10%, often at the expense of government bonds. The problem with that approach is that gold itself is changing.

Historically, gold and silver were held in small allocations as portfolio insurance rather than return-generating assets. Today, the investor base is far broader, spanning central banks, institutions, momentum traders, inflation hedgers and those positioning for monetary disruption. These competing motives mean prices are increasingly driven by flows, positioning and sentiment, causing gold to behave more like a generic risk asset than a reliable hedge against inflation or geopolitical stress.

In other words, one of the market’s oldest firewalls may no longer offer quite the same protection.

Subscribe now

AI: Trying to Make the Numbers Add Up

Financial Fables — Sun, 21 Jun 2026 14:31:41 GMT

My wife is a maths teacher and she tells me that some equations have no solution.

Looking at the numbers behind SpaceX and OpenAI, I am starting to think investors may have found a few of their own. Ahead of SpaceX’s IPO last week, I struggled to make the valuation stack up. Then OpenAI’s 2025 financial accounts were leaked - and somehow the numbers look even harder to explain.

But sometimes you’ve got to sit down with a calculator and give it a go anyway, so lets try and make the numbers make sense.

SpaceX: The Moonshot Premium

Last week saw the SpaceX IPO blast off in financial markets. Shares initially soared from an IPO price of $135 to almost $220 before gravity began to take hold, with reports emerging that the stock had already attracted significant short-selling interest.

The challenge for investors is that SpaceX’s valuation appears to depend heavily on the success of several ambitious “moonshot” projects. These range from expanding Starlink’s global communications network to more speculative ventures such as space-based data centres and, ultimately, reusable missions to Mars. While these opportunities may be enormous, they remain inherently uncertain.

This creates a valuation problem. Morningstar estimated prior to the IPO that more than $1 trillion of implied company value could disappear if some of the firm’s most ambitious projects fail to materialise. In other words, a substantial proportion of today’s valuation reflects profits that investors expect SpaceX to generate many years into the future rather than cash flows being earned today.

The chart below illustrates the challenge. Even under aggressive assumptions of 50-100% annual earnings growth, SpaceX continues to trade at valuation multiples well above those of the broader technology sector. Investors are therefore not simply betting on strong growth. They are betting that SpaceX successfully delivers on some of the most ambitious commercial projects ever attempted - and, as a reminder, the Starship rocket launches (one of the most important divisions in SpaceX) currently have a 50% success rate on their launches.

OpenAI: Where Did the Losses Go?

To make matters worse, OpenAI’s 2025 financial statements were leaked this week. Technology commentator Ed Zitron highlighted that the company spent approximately $34 billion on research, development and sales initiatives (see the breakdown below) during the year while generating just $13 billion in revenue. The result was an operating loss of more than $21 billion in a single year - but there are more losses.

Zitron highlights that when accounting for interest expense and the costs associated with changing the corporate structure, OpenAI reported a consolidated net loss of more than $60 billion for 2025. That figure is so large that it exceeds the annual GDP of many sovereign nations like Paraguay or Tunisia!

Yet the final loss attributable to OpenAI was reported at a much lower $38.5 billion. More than $21 billion of losses were allocated to noncontrolling interests within the firm’s complex ownership structure.

Put differently, that means that means that OpenAI is effectively transferring a large portion of these losses to a separate controlling entity - but who that economic owner is and how they might bare those losses remains a concerning mystery.

The Pathway to Profitability

Tech firms often make losses. Airbnb lost around $5.6 billion between 2018 and 2021 before becoming consistently profitable. Uber accumulated roughly $32 billion of losses between 2014 and 2022 before eventually reaching profitability.

The issue is not that technology firms lose money. It is the scale of the losses now being tolerated by investors. OpenAI lost more money in a single year than Uber lost in eight years. Similarly, SpaceX’s high risk ventures into space and its desire to play catch-up in AI could also see the firm hemorrhage cash for a long period of time.

Yet despite these mounting losses, valuations continue to climb. Investors are being asked to assign ever-higher values to businesses generating ever-larger deficits in the hope that future profits will eventually justify today’s spending. That may prove correct. But the gap between present-day economics and future expectations is becoming increasingly difficult to ignore.

Subscribe now

Private Equity: The Second Black Hole?

Financial Fables — Sun, 14 Jun 2026 14:31:19 GMT

A few months ago, I compared private credit markets to a black hole - an asset class so opaque that it is difficult to see what is happening inside it.

Astronomers have discovered that black holes rarely exist in complete isolation. In many cases, they come in pairs, orbiting one another in what is known as a binary black hole system. And if private credit is one black hole in modern finance, there is another lurking nearby.

This second black hole is private equity (PE).

Smaller than its private credit companion but closely linked to it, PE has attracted far less scrutiny in recent years. Yet this black hole it displays many of the same characteristics: limited profitability, difficult valuations, and growing questions about whether investors will get their money back.

Private Equity: Where is the Money?

Measuring PE returns can be tricky - one of the most common approaches is to look at a metric known as Distributions to Paid-In Capital (DPI). DPI measures how much cash a PE fund has actually returned to investors relative to the amount originally invested. A DPI of 100% means investors have received back their initial capital. Anything above that represents realised profit.

The chart below tracks DPI by vintage year, with a vintage simply referring to the year in which a PE fund began investing. Unsurprisingly, older vintages have returned substantially more cash than newer ones.

The troubling feature is not that recent funds have low distributions (that is to be expected) but rather how little cash has been returned across almost all post-2017 vintages. Even 2018 vintages have yet to get back to 100% DPI despite having launched eight years ago.

From the outside, investors can still observe reported valuations and respectable performance figures. Yet the flow of cash emerging from the system has slowed to a trickle. Capital continues to disappear into the black hole, but much less is coming back out.

Less Fuel for the Black Hole

Not only is less cash escaping the PE black hole, but less cash appears to be flowing into it as well.

The chart below shows that PE fundraising has been declining for several years. While 2026 is still incomplete, fundraising remains sluggish and is on track to be one of the weakest years for asset gathering since 2018.

The slowdown is hardly surprising. PE has traditionally relied on a simple bargain: investors lock up their money for long periods in exchange for the prospect of superior returns. But that bargain becomes harder to sell when investors are waiting longer and longer to get their capital back.

This fundraising drought may also help explain why so many asset managers are now attempting to broaden access to private markets under the banner of "democratising private assets". For years, PE was largely the preserve of pension funds, endowments and sovereign wealth funds. Increasingly, however, the industry's attention has shifted towards retail investors.

But PE remains an awkward fit for many retail portfolios. Funds can lock up capital for years, distributions are unpredictable and investors often have limited visibility into the underlying assets. These characteristics may be acceptable for a pension fund with a multi-decade investment horizon, but they are less obviously suited to investors who may need access to their savings at short notice.

Cash is entering the system more slowly than before, while cash distributions to existing investors remain subdued. That does not mean the black hole is about to collapse. Many of the assets inside may ultimately justify their valuations. But when less money is flowing in and little money is flowing out, it becomes increasingly difficult to determine what is really happening beyond the event horizon.

Too Much Tech?

Another explanation for weak fundraising may be the asset class’s growing exposure to technology.

As the chart below shows, PE portfolios now have roughly 36% exposure to technology and AI-related businesses - broadly similar to both the S&P 500 and private credit markets.

This creates a potential problem. Investors often allocate to private assets seeking diversification, yet many portfolios are already heavily exposed to the same U.S. technology ecosystem through their public equity holdings.

As we discussed last week, portfolios are already full of AI and technology exposure. It is therefore possible that some investors are becoming reluctant to commit even more capital to an asset class that increasingly resembles another bet on the same theme.

How Stable are the Black Holes?

It has been a turbulent year for private assets. The challenge for investors is that it remains difficult to assess the true stability of the black hole. One way to gauge sentiment is through the share prices of listed private asset managers.

As the chart below shows, after an initial sell-off following concerns over private credit redemptions, valuations have largely stabilised.

Part of this resilience likely reflects the broader strength of financial markets. However, fresh liquidity has also emerged from newly established distressed investment funds seeking to capitalise on attractive opportunities..

PE is likely a less systemically important concern than private credit. Weak returns are easier to absorb given the underlying equity structure, and the long-term nature of investor capital reduces the liquidity pressures that have recently emerged in private credit markets. However, PE still displays many of the characteristics of an unstable black hole, with lower, and lower capital flowing back to investors.

For now, the black holes appear to have partially stabilised, although some funds continue to face redemption pressures. Yet the outlook for PE remains challenging: distributions have slowed, fundraising is weak and heavy exposure to technology leaves investors vulnerable to a reversal in a single dominant theme. The black holes are not about to collapse just yet, but in a market with limited visibility, investors would be wise to keep a close eye on the event horizon.

Subscribe now

AI & Tech IPOs: Force Feeding Investors

Financial Fables — Sun, 07 Jun 2026 14:31:41 GMT

Imagine being at your favorite restaurant having just finished its signature dish. You’re comfortably full, ready to ask for the bill and head home.

Then the waiter arrives with another round.

You explain that you’re finished. You insist that you don’t want any more food. But the waiter pays no attention. Despite your protests, they start force-feeding you another course.

Absurd? Perhaps. But over the next few months, something remarkably similar may happen to investors.

Most portfolios are already stuffed with technology and AI exposure. The S&P 500 has +35% exposure to tech and AI; the MSCI EM has 50% and the deluge of tech debt issuance means investors even have it in their bond exposure too.

Yet they may soon be forced to take on even more.

SpaceX, Anthropic and OpenAI are all reportedly preparing for public listings, with combined valuations that could comfortably exceed $3 trillion. The chart below provides a sense of the sheer scale of the potential IPO pipeline heading towards markets.

At this point, you might reasonably think: if I’ve had my fill of AI, I’ll simply choose not to participate - think again.

S&P Dow Jones is considering changes to its index inclusion rules that could allow these firms to enter the S&P 500 far sooner than would previously have been possible. And if that happens, millions of investors tracking the index may find themselves buying these companies whether they want to or not.

Force feeding Passive Investors

Every U.S firm currently IPOing strives to be included in the S&P 500. Why? Because inclusion effectively guarantees a place in the portfolios of millions of investors and trillions of dollars of passive investment funds.

But there are rules policing the inclusion of newly IPO’d companies in the S&P 500 index. These are:

At least 12 months of public trading (often called ‘seasoning’)
Positive earnings over the last four quarters
A positive most recent quarter

These requirements were designed to ensure that companies had established a track record as public businesses before passive investors were automatically exposed to them.

The problem is that many of the mega-cap technology firms currently being discussed as IPO candidates would struggle to satisfy these criteria immediately after listing and are therefore technically ineligible for inclusion within the S&P 500 - depriving them of potentially billions in capital flow from benchmark-tracking investors.

However, in April 2026, S&P Dow Jones are consulting on a rule change that would exempt “megacap IPOs” from the profitability requirements so that, as long as a firm is big enough, it can be considered to be benchmark eligible without ever having turned a profit. In addition, S&P Dow Jones are planning on shortening the seasoning period from 12 months to just 6 months in order to fast track these new IPOs into the benchmarks.

The result is that some of the largest tech IPOs in history will likely find their way into index funds far sooner than investors might expect. Meaning any benchmark-aware investors has little choice but to buy them.

Transforming the Index

As index providers start shoving large, unprofitable tech firms into the key benchmarks the index is likely to change dramatically. So what exactly is it that you will be eating if you buy a S&P 500 tracker from now on?

While valuations details are still in flux we can roughly predict what the total market value is of the three firms will be: SpaceX ($1.8 trillion); OpenAI ($1 trillion); and Anthropic ($1 trillion).

At those implied market cap weights these firms will likely enter the S&P 500 with a combined weight of roughly 5-6% of the index. By including all three of them that would mean that the entire top 10 firms in the S&P 500 would be tech/AI firms as SpaceX, Anthropic and OpenAI join the rest of the large tech firms and push out the only remaining non-tech firms left in the top 10: Berkshire Hathaway, Eli Lilly and J.P Morgan.

So if SpaceX, Open AI and Anthropic all enter at the valuations being discussed, then it is entirely plausible that:

All of the top 10 constituents of the S&P 500 are technology firms.
Nearly 40% of the entire index is concentrated in those ten names.
Close to half of the index’s performance could be driven by businesses whose fortunes are increasingly linked to tech/AI (e.g. Oracle, Palantir, AMD etc).

AI is Everywhere

We’ve seen index concentration before. Back in the 1980s the top of the index contained oil companies, industrial conglomerates, banks, retailers and telecoms. In the 2000s, energy and financials dominated. So we’ve had concentration before, but not concentration around a single technological ecosystem.

To be clear, there is nothing inherently wrong with having exposure to AI. If the technology ultimately delivers on its promise, investors will likely be rewarded for owning these businesses.

The challenge is that AI/tech exposure is increasingly seeping into every corner of financial markets. As the chart below shows, it dominates major equity indices, is becoming a growing component of corporate bond markets, and now looks set to occupy an even larger share of passive portfolios through a wave of mega-cap IPOs.

The result is that investors seeking genuine diversification may find it increasingly difficult to avoid making the same underlying bet. What appears to be a diversified portfolio on paper may, in reality, be ever more dependent on a single theme continuing to succeed.

So where can investors find genuine diversification?

One option is to move away from market-cap weighted indices such as the S&P 500 and towards equal-weighted alternatives. While still providing exposure to U.S. equities, they significantly reduce the dominance of a handful of mega-cap technology companies.

Another possibility is commodities. However, as we’ve discussed before, I remain sceptical about their long-term ability to generate returns in the same way as productive assets such as equities and bonds.

The final option is perhaps the least exciting: cash.

Yet cash may deserve more credit than it receives. If investors are being pushed towards ever greater exposure to the same technology and AI theme, then the overall risk of the equity component of a portfolio is rising. Holding a little more cash provides dry powder for future opportunities and a stable ballast when markets become turbulent.

None of this means investors should abandon AI. The technology may prove every bit as transformative as its advocates believe. But diversification only works if different parts of a portfolio respond differently when things go wrong.

And right now, investors may be discovering that the menu is becoming increasingly one-dimensional.

Subscribe now

The Accounting Alchemy of AI

Financial Fables — Sun, 31 May 2026 14:30:26 GMT

Magic spells are rarely judged on the day they are cast.

In fairy tales and folklore, the consequences often take years to reveal themselves - a spell intended to solve one problem often creates another.

Accountancy changes work in much the same way and we are living through the side-effects today. In 2016, accounting regulators tweaked the rules and a decade later tech giants like Amazon and Google are reporting huge paper profits on their private investments despite not having any hard earnings.

So how does the accounting alchemy work and what could be the unintended side-effects?

Tech Earnings Magic

On the surface, tech firms are doing well. Last quarter, both Amazon and Google reported bumper earnings. Yet hidden beneath the headlines was an interesting detail. Around half of those profits did not come from selling more products, serving more customers or generating additional advertising revenue. Instead, they came as paper profits from the mark-ups in the value of their private holdings and one in particular drove much of this - Anthropic.

Anthropic is the private AI research company behind Claude and while the firm remains privately owned, it has attracted enormous investments from Amazon, Google and, to a lesser extent, Microsoft and Nvidia.

Due to the complexities of private ownership structures, nobody is entirely sure how much these firms own, but the stakes are clearly large enough to have a meaningful impact on their financial results. As the below chart shows, over the last year over 40% of Google’s earnings have come from mark-ups in the value of their private holdings (a large chunk being Anthrophic) and over 20% in the case of Amazon.

This might sound strange. After all, how can a company report profits from an asset it has not actually sold?

The answer lies in a relatively obscure accounting change introduced less than a decade ago.

Pre-2016, if Google or Amazon owned a stake in a private company, that investment would largely sit on the balance sheet at its original purchase price. Even if the value of the company doubled or tripled, little would show up in reported earnings unless the stake was eventually sold.

Today, the rules work differently. If there is an observable change in value - for example, a new funding round that values a private company more highly than before - firms are increasingly able to mark up the value of their existing holdings and recognise those gains in their income statement.

In many ways the change made sense. Regulators wanted investors to have a clearer picture of what company assets were actually worth, rather than allowing large unrealised gains to remain hidden away on balance sheets. Yet there is a side effect. Corporate earnings have become increasingly sensitive to changes in private market valuations.

That matters because private markets have exploded in size over the last decade. Many of today’s most valuable technology companies are staying private for far longer than previous generations. At the same time, cash-rich technology giants have become some of the largest investors in those private firms.

Increasingly, companies such as Amazon and Google are not just technology businesses. They are also owners of vast portfolios of private investments. Their enormous cash piles and industry expertise allow them to acquire significant stakes in promising companies long before they ever reach public markets.

The result is that a growing share of reported profits can come not from selling cloud services, advertising or consumer products, but from revaluing holdings in private companies. In other words, future expectations are increasingly showing up in today’s earnings.

The Side-Effect: AI Circular Flow

This is where things become particularly interesting.

Amazon and Google invest billions into Anthropic. Anthropic then commits to spend enormous sums on AWS and Google Cloud infrastructure to train and run its models.

As Anthropic’s valuation rises in subsequent funding rounds, Amazon and Google are able to mark up the value of their existing stakes, generating large unrealised gains that flatter reported earnings. Those stronger earnings and growing investor enthusiasm then help sustain the broader AI trade, supporting further capital raising and even higher valuations.

Whether intentional or not, it creates a powerful feedback loop where infrastructure spending, private valuations, market sentiment and accounting gains all begin reinforcing one another.

To be clear, this does not mean the economics are fake. Anthropic may genuinely become an enormously valuable company and the funding rounds involve real third-party investors committing real capital.

But the dynamic does raise questions about how much of current profitability across parts of the AI ecosystem reflects realised cash generation versus rising paper valuations tied to private market enthusiasm. It also makes it increasingly difficult for investors to disentangle where underlying operating performance ends and AI-related mark-to-market gains begin.

How can the accountancy spell unravel?

The challenge with accounting magic is that, eventually, the audience wants to know how the trick works.

And the risk with recognizing unrealized gains is if they remain unrealized. If those gains ultimately fail to materialize, today's paper profits can quickly become tomorrow's losses.

This creates a dual risk for firms such as Amazon and Google.

The first risk is that AI simply fails to deliver on the extraordinary expectations currently being placed upon it. Many believe AI will prove transformational and perhaps, in time, it will. But given the enormous sums being invested across the technology sector, AI needs to start generating meaningful economic returns sooner rather than later.

As we spoke about last week, the evidence remains mixed. While AI is undoubtedly improving productivity in some areas, the impact on sustainable top-line revenue growth remains much harder to identify. Much of the measurable benefit currently appears to be coming through cost reductions and labour substitution rather than the creation of entirely new profit pools.

The second risk is deeper. Even if AI succeeds, who is to say Anthropic will be one of the ultimate winners?

Technology history is littered with examples of firms that backed the right trend but the wrong company. Many of today’s tech giants started life as tiny upstarts and we are still in the early innings of the AI race. Perhaps Anthropic will dominate. Perhaps OpenAI will. Perhaps a company that barely exists today will emerge as the eventual winner.

This matters because the accounting magic relies on a simple assumption: that today’s rising valuations will eventually be justified by tomorrow’s commercial success. If that assumption proves wrong, the spell can break surprisingly quickly.

Subscribe now

AI Bubble: Are We Going Too Fast?

Financial Fables — Sun, 24 May 2026 14:31:13 GMT

In Formula One if there are dangers on the track the race officials will raise yellow flags to get drivers to slow down while they manage whatever crash or obstacle is on the track.

It might be annoying for the thrill-seeking drivers but ignoring yellow flags is dangerous risk.

In many ways, investors find themselves in a similar situation - we’ve had a blistering few weeks in equity markets. US tech stocks have raced to +25% since the lows of March 2026 and global semiconductors are up 23% in the last month alone. However, whilst markets continue flying around the track, I am starting to see a number of yellow flags appearing.

So what are the warning signs and how serious are the risks of continuing with the AI race?

Yellow Flag #1: A lack of market breadth

The first yellow flag on the investment track is the narrow concentration around just a handful of names.

While bubbly periods often lift all stocks, the mania tends to concentrate on a narrow handful of names. In 1929, it was RCA. In the 1990s, it was names like Cisco and Yahoo. Today, it seems to be firms like Google and Nvidia.

To try to highlight the concentration issues in this AI bubble - take a look at the below chart which shows the performance divergence between equal weighted S&P 500 vs the market cap weighted S&P 500.

There has been a dramatic divergence from the long-run trend as market cap weighted names have significantly outperformed the equal weighted index. Put differently, the biggest names in the index keep getting bigger!

This is troubling for two core reasons.

Firstly, it’s a sign that active security selection is disappearing. The rise of passive investing and the way it allocates closely to benchmarks fuels this mindless investment style - assisting in inflating the bubble to much bigger than normal size.

But it is also troubling because for this huge AI infrastructure to be justified, the benefits of AI need to show up on the balance sheets, earnings reports and share prices of non-tech firms.

A recent Cambridge University survey suggested that 43% of firms who have introduced AI have not seen any boost to profit. A troubling statistic when combined with how detached tech firms have become from the underlying economy. From my perspective, this is where the first warning flag really begins waving.

Yellow Flag #2: Momentum and price behaviour

In Formula One, the most dangerous conditions are often the ones drivers gradually become comfortable with. Markets work in much the same way.

AI defenders are often quick to highlight to me that valuations are not excessive - proof that we aren’t in a bubble. Valuation metrics, like P/E ratios, are helpful as long-term signals but do a poor job of helping investors call the top of asset bubbles. Expensive markets can stay expensive for years. The dot-com bubble looked overvalued in 1997 and then doubled again before finally collapsing.

What matters more is often the behaviour underneath the surface of the market - because in bubbles, deteriorating track conditions rarely show up immediately in valuations. The degree to which prices become driven by momentum, crowding, and self-reinforcing flows rather than just raw P/E ratios themselves.

This is where Didier Sornette’s work becomes interesting. His research on financial bubbles focuses less on whether valuations are “too high” and more on whether markets are becoming increasingly reflexive and unstable. In simple terms, bubbles form when rising prices themselves become the reason prices keep rising.

Investors chase performance, momentum strategies attract more inflows, benchmark concentration intensifies, and volatility often remains surprisingly subdued because everyone is crowded into the same trades. The chart below captures some of this dynamic perfectly. Momentum stocks have massively outperformed minimum volatility strategies at many of the major turning points of the last 25 years - the dot-com bubble, the pre-GFC period, the post-Covid AI boom, and now again in 2026.

What makes this particularly striking is that bubbles rarely feel most dangerous when they are forming. In fact, much like a driver still setting fast lap times under yellow flags, they often feel safest right before conditions become unstable.

Volatility stays low, dips keep getting bought, and market leadership narrows into a handful of “untouchable” winners. That is why Sornette’s framework is so useful. It shifts the question away from “is the market expensive?” toward something far more uncomfortable: “has the market become dependent on price appreciation itself to keep functioning?”

Yellow Flag #3: Leverage metrics

At the tail-end of 2025, excitement around the AI bubble waned, in large part because liquidity in the system had been constrained due to a lack of excess cash in the system.

In late January, the Fed took steps to ease the pressure on the plumbing by essentially quietly pausing their balance sheet reduction and buying short-dated treasuries via a “temporary” program that will absolutely now be a staple part of their arsenal.

Regardless, the Fed’s actions have relieved pressure on the creaking leverage system allowing leverage to expand again, as seen in the chart below.

But in doing so, the AI bubble begins to become increasingly unstable - relatively high levels of speculation make the track conditions increasingly dangerous.

So what should investors do about it?

Going back to my F1 analogy, we are going fast, very fast, and the growing number of yellow flags suggests investors should at least begin easing off the accelerator.

The problem is that investors rarely have this luxury - many are assessed on short-term relative performance and sitting in safe bets whilst the market accelerates away from you is a quick way to lose assets.

In many ways, fund managers behave a bit like F1 drivers. Even when conditions are deteriorating, nobody wants to give up position while everyone else keeps pushing forward.

So for those investors unable to take a pit stop the best option is to stay off the accelerator and test the brakes.

From an accelerator standpoint, what I mean by that is now is not the time to double down on this AI bet. For example, I’ve seen a marked increase in the amount of asset managers, including my former employer, highlighting EM stocks and the AI opportunities in them. Given how much exposure investors already have to AI either via U.S stocks or private credit markets, adding exposure to EM chipmakers without reducing exposure elsewhere feels like a dangerous move to do on a hazardous track.

In terms of brakes, some exposure to short-dated bonds, cash or high quality non-tech corporate debt would help ensure that if we do hit an obstacle, your brakes might help protect you from the carnage that follows.

For now, the race continues. But the track conditions are deteriorating.

Subscribe now

Aerospace & Defence Stocks: Are They a Reliable Shield?

Financial Fables — Sun, 17 May 2026 14:31:01 GMT

We often imagine medieval battles as structured and orderly affairs. In reality, they were exercises in confusion and survival. Visibility was poor, alliances shifted unexpectedly, and knights rarely understood the battlefield they were riding into. In these environments, survival depended less on bravery and more on whether a knight’s armour could withstand the barrage of assaults long enough to endure the chaos.

Modern investors may recognise the feeling.

The world is becoming increasingly fractured and geopolitics is once again emerging as a major source of market instability. Alliances are fraying, conflicts are spreading, and the global order that underpinned financial markets for much of the past few decades appears to be shifting beneath investors’ feet.

In search of protection, many investors are turning towards aerospace & defence stocks. The logic appears straightforward: in a world of rising geopolitical tensions and growing military competition, defence spending should continue to increase - supporting the earnings of a small number of strategically important firms.

Today, aerospace & defence stocks now trade with a “conflict premium” - where investors assign higher valuations to defence stocks simply because they are perceived as beneficiaries of a more dangerous world. However, this increasingly popular portfolio shield may be far less reliable than investors assume.

To help explore the realities of the defence industry - and better understand what aerospace & defence stocks can and cannot realistically offer investors - I have partnered with Jack Wood, author of The Informed Ranger, an eight-year Army veteran who focuses on market and trade ideas connected to geopolitical developments - drawing on his extensive experience as a Ranger and Intelligence Analyst, alongside his recent work in currency and money markets.

The Myth of the Wartime Windfall

Financial markets increasingly treat aerospace and defense companies as direct beneficiaries of geopolitical instability.

Missile strikes in the Middle East, tensions in the Taiwan Strait, rising NATO spending commitments or fear surrounding the fragmentation of the global order are not often met with the same reflexive response: buy defense stocks.

As you can see from the chart below, the onset of the Iran Conflict in March 2026 has triggered a record wave into aerospace and defence ETFs as investors look for a sturdy shield to help protect their portfolio.

However, one of the biggest misconceptions investors have about the defence industry is assuming geopolitical instability quickly translates into higher revenues.

In reality, the path from conflict to corporate cashflows is slow, political and deeply bureaucratic. Modern warfare might evolves rapidly but efence procurement rarely does.

Even after governments announce higher military spending, contracts can take 12-18 months or longer to move through procurement systems, political approvals and budget negotiations. Revenues are then often recognised gradually through milestone-based payments spread across several years.

This creates an important disconnect in how investors think about the sector.

Many increasingly view aerospace & defence stocks as portfolio armour against geopolitical uncertainty. Yet these companies remain heavily dependent on the same governments and political institutions that instability can weaken.

Government shutdowns can delay procurement decisions. Fiscal pressures can slow spending commitments. Elections and political shifts can rapidly alter priorities.

In many ways, aerospace & defence firms behave less like conventional growth companies and more like highly specialized state contractors operating at the mercy of political systems.

And that makes the sector far more fragile than many investors assume.

Not Quite Silicon Valley

On the surface, aerospace & defence companies possess many of the characteristics financial markets typically reward.

They operate in highly specialised industries with enormous barriers to entry. Many possess decades of accumulated intellectual property, advanced engineering expertise and entrenched positions within oligopolistic markets that are difficult for new competitors to penetrate.

In many ways, they resemble technology companies.

But unlike Silicon Valley firms, aerospace & defence companies operate with remarkably little freedom.

Their customer base is heavily concentrated amongst a relatively small number of governments, many of which are already fiscally stretched. Revenues depend not simply on demand for military capability, but on political priorities, procurement systems and government funding cycles.

Technology firms can rapidly scale products globally, diversify revenues and allocate capital with considerable flexibility. Aerospace & defence firms cannot. Contract awards are political, export markets are tightly regulated and governments frequently exert influence over how companies deploy capital.

Another important constraint often overlooked by investors: access to financing.

Despite rising geopolitical tensions, many defence firms, particularly in Europe, still face financing frictions linked to ESG mandates and reputational concerns amongst banks and institutional investors. At the same time, governments themselves increasingly impose restrictions on shareholder distributions and capital allocation within strategically important defence contractors.

As a result, aerospace & defence firms occupy an unusual position within financial markets. They are treated like strategic growth businesses during periods of geopolitical stress, yet many remain heavily constrained by politics, regulation and state influence.

If large technology firms operated under the same restrictions - concentrated customers, political oversight, constrained capital allocation and dependence on government procurement - markets would likely assign them far lower valuations.

And this is where the idea of a permanent “conflict premium” begins to look increasingly questionable.

Politics of Protection

One of the more unusual characteristics of aerospace & defence stocks is that investors often treat them as protection against geopolitical instability, despite the sector itself being deeply dependent on political stability.

Unlike most industries, the fortunes of aerospace & defence firms are closely tied to government balance sheets, political priorities and the ability of states to consistently fund long-term procurement programmes. Rising geopolitical tensions may increase demand for military capability, but they can also intensify fiscal strains, political fragmentation and policy uncertainty within the very governments defence companies rely upon.

Furthermore, this political decision making process is highly skewed to benefit incumbents and keep challengers at a distance. When a government looks to renew a contract, as long as performance is high, the incumbent win rate is often 70-85% - so once a contract is awarded it tends to stay in the same hands. An industry challenger, who governments are less familiar with, have much lower win rates and only get success when they can demonstrate a compelling differentiation vs the wider peer group.

But despite elevated geopolitical tensions and rising global military competition, periods of government shutdown risk and fiscal instability have repeatedly created pressure within parts of the U.S aerospace & defence sector. Procurement decisions can be delayed, funding approvals become uncertain and visibility over future revenues weakens.

As the chart below highlights, European defence firms materially outperformed their U.S counterparts during periods where concerns surrounding U.S political dysfunction and fiscal instability intensified. At the same time, European governments increasingly signalled stronger long-term commitments towards defence spending and strategic rearmament.

So, while many investors increasingly view aerospace & defence stocks as straightforward beneficiaries of a more dangerous world. In reality, these companies remain heavily exposed to elections, budget negotiations, fiscal pressures and political dysfunction.

Investors may believe they are buying protection from geopolitical instability but in practice, they are often buying deeper exposure to the political systems that are generating the geopolitical instability in the first place.

The Shield Trade

Despite these structural weaknesses, capital continues to flow aggressively into aerospace & defence stocks during periods of geopolitical stress.

Interestingly, the pattern of these flows suggests investors may not be carefully repricing the sector’s underlying fundamentals. Instead, capital increasingly appears concentrated in a relatively small group of large, liquid and recognisable defence names.

The chart below compares an equal-weighted aerospace & defence basket against a market-cap weighted version of the same basket.

If investors were broadly reassessing the long-term earnings potential of the entire sector, we would expect both baskets to perform relatively similarly. Instead, the market-cap weighted basket materially outperformed, suggesting that a handful of large firms have increasingly dominated sector returns.

This matters because it hints at a growing “shield trade” developing within financial markets.

During periods of geopolitical instability, investors appear increasingly willing to crowd into the largest and most recognisable defence companies as a form of portfolio protection. In many cases, these flows are likely amplified by passive investment vehicles and ETFs, which naturally concentrate capital into the sector’s largest constituents.

The result is that aerospace & defence stocks may increasingly be trading not simply on procurement fundamentals or long-term earnings expectations, but on their perceived ability to provide psychological and financial protection during periods of geopolitical uncertainty.

And this dynamic creates its own fragilities.

When assets become widely perceived as portfolio armour, valuations can become increasingly detached from the slower-moving political, fiscal and procurement realities that ultimately determine long-term cashflows.

So, does the shield truly protect us?

Medieval knights understood an important truth about chaotic battlefields – high quality armour mattered. But history also shows that armour alone rarely guaranteed survival.

Aerospace & defence stocks may continue to benefit from a world characterised by rising geopolitical tensions, military competition and growing strategic fragmentation. Governments are likely to continue increasing defence expenditure and the sector will remain strategically important within the evolving global order.

However, investors should be careful not to confuse strategic importance with economic invulnerability.

The defence industry remains deeply dependent on political systems, government finances and slow-moving procurement structures. Revenues often materialise gradually, capital allocation is constrained and many firms operate under political and regulatory limitations that most traditional growth businesses do not face.

Yet financial markets increasingly appear willing to assign a “conflict premium” to the sector - treating aerospace & defence stocks as portfolio shields against an uncertain world.

The problem is that these shields may be more fragile than they first appear.

In periods of uncertainty, investors naturally search for protection. But as medieval knights understood all too well, survival on chaotic battlefields depended not simply on possessing armour, but on understanding its weaknesses.

Subscribe now

The Dangers of being an EM Sheep

Financial Fables — Sun, 10 May 2026 14:31:22 GMT

Next week I am speaking at the London Business School on the topic of emerging market (EM) investing – and one of my aims is to warn the audience about the dangers of becoming an EM equity sheep.

And no, before anyone asks, I am not planning to walk onto the stage and start bleating at the audience. What I am referring to is a long-running flaw in the construction of the MSCI EM benchmark which, in my view, has led many investors badly astray.

The vast majority of EM equity investing is benchmarked in some way against the MSCI EM Index, whether through passive ETFs or tightly constrained active managers. The problem is that the index has become increasingly distorted over time. By following it too closely, investors are not only missing valuable alpha opportunities, but are also sacrificing many of the diversification benefits that EM investing should provide in the first place.

In other words, a large part of the industry has stopped exploring the field and instead begun following the flock.

The Grazing Problem in Emerging Markets

EM equities have found themselves in something of a financial market sweet spot. We are in the midst of a commodity supercycle which is coinciding with what could prove to be a prolonged period of U.S dollar weakness. Together, these forces create two powerful structural tailwinds for EM equity investors and strengthen the case for larger allocations to the asset class.

But fully capturing these tailwinds cannot be achieved by simply grazing alongside the benchmark herd. EM economies are far from a homogenous group; they are a collection of smaller, highly differentiated markets, each with distinct sensitivities to commodities, currencies, and global growth.

To illustrate this point, take a look at the chart below, which plots a range of emerging and frontier markets according to their relative sensitivity to both commodities and the U.S dollar. The horizontal axis measures the share of exports tied to commodities - metals, fuel, and agriculture combined - while the vertical axis shows the correlation between each country’s MSCI index and the dollar.

The chart highlights that several Latin American economies, notably Brazil, Colombia, and Chile, alongside African markets such as Kenya and South Africa, appear particularly well positioned. These countries are significant exporters of global commodities and have historically performed well during periods of U.S dollar weakness.

By contrast, many European and Asian EM economies have relatively low exposure to commodities, limiting their ability to benefit from this tailwind. Meanwhile, the fixed exchange-rate regimes common across oil-exporting Middle Eastern economies reduce their capacity to benefit from a softer dollar, helping explain why their equity markets often move in tandem with the greenback.

The difficulty for passive investors is that the MSCI Emerging Markets benchmark remains overwhelmingly concentrated in Asia, with more than 80% of its weight allocated there. Latin America and Africa - the regions that appear best positioned to benefit from both stronger commodity prices and a weaker dollar - account for only around 7% and 4% of the index respectively.

In other words, the parts of the EM universe with the strongest cyclical tailwinds are precisely the areas where benchmark-following investors have the least exposure.

As a result, investors relying on passive or benchmark-constrained strategies risk missing some of the most attractive opportunities in EM just as the macro backdrop becomes increasingly supportive.

And this is where the dangers of behaving like an EM sheep become most apparent. By staying tightly clustered around the benchmark, investors have historically sacrificed meaningful upside and alpha opportunities from smaller EM economies. The chart below compares the performance of the MSCI EM Index against an equal-weighted basket of EM markets and highlights just how costly following the flock can become over time - with the equal-weighted index returning 2.25 times the value of the MSCI EM benchmark since 1994.

An AI wolf in EM sheep clothing

The historical evidence already highlights the dangers of behaving like an EM sheep. But looking ahead, the outlook makes me even more concerned.

A recent piece by Hakyung Kim of the Financial Times noted that developed market investors have traditionally allocated to EM equities in search of diversification. That assumption, however, is becoming increasingly questionable.

As the chart below illustrates, semiconductor manufacturers and large technology firms are now dominating EM equity indices (a combined 32% of the benchmark), steadily diluting many of the diversification benefits investors once expected from the asset class.

Companies such as TSMC and Samsung have become deeply intertwined with the same global AI and technology cycle driving U.S mega-cap equities like Nvidia and Apple. As a result, benchmark EM investing and developed market tech exposure are becoming increasingly interconnected.

For investors already heavily exposed to U.S tech, this creates an important hidden risk. What appears on the surface to be diversified EM exposure may, in reality, simply represent another expression of the same AI trade.

In other words, there may be more wolves lurking beneath the EM sheep’s clothing than investors realise. And by continuing to follow the benchmark herd, investors risk discovering that their supposedly diversified EM allocation is grazing in exactly the same field as the rest of their portfolio.

How to stop being an EM sheep

So how can investors break away from the benchmark herd? A few thoughts…

First, we need to be more intellectually honest about how we classify EM equities. Few investors who have spent time in either South Korea or Taiwan would seriously argue they resemble traditional emerging market economies. Their equity markets are dominated by globally competitive technology firms deeply embedded within developed market supply chains.

Indeed, were it not for a handful of technical and liquidity considerations, both countries would likely already be classified as developed markets by index providers such as MSCI and FTSE Russell. Some pressure should be applied so that mandates and portfolios begin to think carefully about whether to include these markets in scope, especially if the wider portfolio already has hefty U.S tech exposure.

Second, investors increasingly need to separate “EM” from “EM ex-China.” EM ex-China strategies should become far more commonplace within portfolios, with Chinese equity exposure managed through dedicated allocations rather than bundled into a broad benchmark product.

Chinese equities have simply become too large and too influential within the EM universe, often overwhelming the distinct macroeconomic and structural characteristics of smaller emerging economies.

Removing China and Taiwan from the MSCI EM benchmark and reconstituting the index materially changes regional exposures. Brazil, for example, goes from being 5% of the benchmark to closer to 10%, allowing investors greater participation in the very economies most leveraged to a weaker dollar and stronger commodity prices.

Third, investors should think more carefully about how they access EM equities. Passive ETFs may offer low fees, but cheap exposure is not necessarily good exposure. Benchmark investing has increasingly become an exercise in concentration rather than diversification.

Instead, investors should seek active managers with sufficiently wide risk bands to move meaningfully away from their benchmarks as the best EM opportunities are rarely found in the middle of the flock.

So stop behaving like an EM sheep. Lift your head up, look beyond the herd, and explore the wider field.

Subscribe now

Can Commodities Save the Day?

Financial Fables — Sun, 03 May 2026 14:30:59 GMT

Commodity funds had a record breaking month in March 2026 with $1.3bn flowing into broad-based commodity ETFs - marking the 10th consecutive month of inflows - the longest hot streak of inflows on record.

Such a surge in activity shouldn’t be a surprise. Investors are nervous that the ongoing closure of the Strait of Hormuz and the impact in energy prices will trigger another bout of inflationary pressure - commodity funds should, theoretically, provide some protection.

But while commodity funds are back in vogue, investors need to be careful - commodity funds aren’t a silver bullet and they might not save the day as investors hope.

Commodities are not born equal

One major issue with the inflows into broad commodity funds is that many investors assume that commodity markets move together. In reality, energy markets have little in common with livestock. Nor does agriculture and precious metals move together.

To prove the point, take a look at the table below which takes the five major commodity sub indices included withthe Bloomberg Commodity Index, the main benchmark for the sector.

Over the last 10 years only one commodity pairing - agriculture and industrial metals - has had a statistically significant positive correlation. Similarly on a five year basis only energy and agriculture have shown any evidence of consistently moving together.

The big takeaway from the correlations table is that it exposes the fallacy of the investment industry. Just because humans are naturally inclined to group items together for simplicity purposes does not mean that by us doing so they become magically related. Each market has their own supply and demand dynamics that influence pricing and these drivers

For investors who are allocating to broad commodity exposure, what this means is there isn’t a “commodity beta” that can be easily captured by portfolios to protect against inflation pressure. In reality it’s a series of individual sub-markets that might help or hinder performance.

So, if there is no “commodity beta” what does drive commodity fund returns?

How do commodity funds actually work?

Broad commodity funds do not function like their equity and bond peers. These traditional funds buy shares or debt and if the market price of these assets rise, the funds value rises and investors profit.

However, commodities funds are distinctively different because they do not hold the underlying asset. If you turn up at (most) commodity managers offices, there isn’t a secret room or warehouse that is physically storing barrels of oil or precious gold bars.

Instead, commodity funds get their exposure indirectly via the futures market. By buying futures contracts linked to natural gas or silver, rather than the physical asset, means that commodity funds are exposed to three unique return drivers:

Spot Return: If the market price of a commodity falls by $10, investors will feel this.
Roll Yield: This is the gain or loss you get from replacing (“rolling”) an expiring futures contract with a later one.
Collateral Yield: The return from futures collateral being invested in short dated money markets.

Of these three drivers, its roll yield that investors need to be most mindful of. Assessing the potential returns on commodity funds isn’t as simple as “prices are going up, so fund returns must be positive”. Instead, it matters what the shape of the futures curve is.

If markets are in backwardation (futures prices < spot), rolling contracts can add to returns. But in contango (futures prices > spot), investors can end up consistently “buying high and selling low” as positions are rolled forward - quietly eroding performance over time.

Right now oil and livestock futures are in backwardation, suggesting a supportive roll yield. However, gold and copper are in contago, suggesting a headwind for commodity returns.

Furthermore, investors need to be mindful of how fluid the commodity markets can be - futures curve can flip from backwardation to contango and back again quite quickly, especially if there is sudden supply changes. In doing so, it can have a big impact on the returns for commodity investors.

So can commodities save us?

As investors, we’ve been fortunate to live through a period of low, or even no, inflation. The flip side is that there are relatively few asset classes with a truly proven track record in inflationary regimes… commodities included.

2022 was the closest thing we’ve had to unanchored inflation in recent years. Commodity funds delivered strong headline returns - up 13% on the year. But as I noted a couple of weeks ago, that performance was doing a lot of heavy lifting from energy, up 36%. Strip that out, and the picture looks very different - gold and silver, for example, were flat on the year.

The honest truth is that inflationary bouts are an absolute bit*h for portfolio performance. At times like these, the goal shouldn’t necessarily be to maximise returns - it should be to limit the damage.

That’s why I’ve continued to push back against the “diversify your diversifiers” narrative coming from large asset managers. Instead, I favour a simpler approach: hold more cash.

Yes, it erodes in real terms - but it offers something that’s often in short supply during inflationary periods: predictability.

Subscribe now

Has the AI Bubble Reached a Ceiling?

Financial Fables — Sun, 26 Apr 2026 14:31:37 GMT

Asset bubbles typically have two speeds: rising fast or falling quick - there is rarely any in-between.

Back in November 2025, hype around the AI bubble had reached fever pitch but in the last six months the excitement has waned. The chart below shows the number of Google hits on “AI bubbles” has fallen sharply over the last few months as investors focus has shifted to geopolitical tensions in the Middle East.

Such a loss in interest surely should’ve led to a marked deflation in the AI bubble but in fact many AI indices have hardly fallen, with many down just 4-5% below their peak from November 2025. So what is going on, has the AI bubble hit a ceiling or have we entered a new phase of this tech mania?

Asset bubbles usually rise or fall - they rarely float sideways

Asset bubbles have long been a feature of financial markets. Since 1977, there has almost always been something inflating somewhere - gold in the late 1970s, Japanese equities in the 1980s, tech in the late 1990s, housing in the mid-2000s, and biotech more recently.

What stands out in each of these bubbly episodes is not just the scale of the move, but the shape of it. Bubbles tend to follow a familiar pattern: a steady build, a rapid acceleration, and then a violent unwind. They are dynamic, directional, and - more often than not - short-lived.

In contrast, the “disruptor” bubble, comprised primarily of tech stocks, seems to be behaving in an usual fashion - lingering in place. Prices are elevated, the narrative is still alive, but the momentum has faded.

This sideways drift is an odd place for a bubble to sit, which begs the question…

Has the AI bubble reached a ceiling?

The ‘science’ of dissecting bubbles (if we can call it that) is almost always done with the benefit of hindsight, so trying to pick apart why the AI bubble might be drifting sideways is no easy task. With that said, here are a few initial thoughts:

A lack of additional buyers: Manias rely on a constant influx of fresh capital to push prices to ever higher highs. If that buying pressure begins to fade, momentum inevitably slows. In the case of AI, it’s a fairly divisive theme - there are true believers who are already fully invested, and outright bears trying to short it at every opportunity. Perhaps we’ve reached something of a stalemate, where the marginal buyer is simply no longer there.

“Wait-and-see” mode: It’s also possible investors have shifted into a holding pattern. There has been an enormous amount of speculative hype around AI, and now the burden of proof is shifting back to companies to ‘put up or shut up’. As we’ve discussed before, for the AI story to become durable, earnings need to materialise - not just in the income statements of firms like NVIDIA, but across the broader economy. Non-tech companies need to demonstrate real, tangible benefits from this wave of innovation too - something that, so far, remains patchy at best.

Higher-for-longer rates: AI stocks are, at their core, long-duration assets, with cash flows expected far into the future. In a falling rate environment, that’s a powerful tailwind for valuations. But that tailwind is now in doubt. Rising geopolitical tensions, including the conflict involving Iran, and higher energy prices risk keeping inflation elevated and delaying central bank easing. At the start of the year, markets were pricing in 3–4 rate cuts; now, investors are hoping for just one. That shift alone may be enough to cap how far valuations can stretch.

Or maybe we are in a new phase?

There is the possibility that something is shifting within the AI bubble itself.

Earlier this year, I highlighted the different types of asset bubbles that tend to emerge in financial markets (outlined in the table below). For the initial few years, the ‘disruptor’ bubble could be safely classified as a type #2 bubble - large, but with limited risk to the broader economy as much of its done on investor capital, not borrowed funds.

However, as tech firms have embarked on a gigantic borrowing spree to fuel their AI infrastructure build-out, it feels like we are beginning to shift into a more dangerous type #3 bubble.

As part of this transition, into more of a “leveraged market mania” it’s possible that the focus of the exuberance is beginning to drift further down the market cap spectrum. In other words, the bubble isn’t necessarily bursting, it may just be rotating as we have continued to see clear signs of excess.

Last week, Allbirds, a struggling shoe company, released a vague statement indicating that it was pivoting towards “AI infrastructure” (with very little in the way of detail). A rational market would probably have punished that kind of announcement - §what do shoe executives know about AI? And yet the stock rallied nearly 700% (it has since given some of that back as a degree of sanity has returned).

Nor is Allbirds a one-off example. A small social media firm, Myseum, simply added “AI” to its name and that alone was enough to trigger a ~150% gain in the share price. Similarly, former crypto miners like Core Scientific and TeraWulf are reframing their energy-intensive operations as AI data centre infrastructure, in what look like fairly transparent attempts to tap into renewed investor enthusiasm.

These kinds of examples of AI enthusiasm (or perhaps delusion) suggest that the speculative energy around this theme hasn’t really gone away. If anything, it may just be shifting. As the largest names become more crowded and harder to push higher, investors appear to be moving further out along the risk curve in search of the next winner.

Which raises an alternative possibility - perhaps as the AI bubble continues to evolve we haven’t hit a ceiling at all. Perhaps it’s simply finding a new part of the market to inflate.

Subscribe now

The Charge of the Stag

Financial Fables — Sun, 19 Apr 2026 14:31:01 GMT

If you ever stumble across a stag in the wild, the instinct is simple: stop, stay calm, and quietly back away. The last thing you want is to startle it - uncertain animals can react quickly.

Investors today find themselves in a similarly precarious position. We’ve stumbled upon a rare and dangerous beast: stagflation.

It feels as though markets are holding their breath - watching closely, hoping to step back slowly without provoking a charge.

So as investors stand face-to-face with the stag - uncertain whether it will hold steady or charge - what might this environment mean for financial markets? And more importantly, how should investors position themselves?

A Dangerous Beast

A charging stag is a dangerous beast - and investors may find themselves caught between its antlers: lower growth on one side, higher inflation on the other.

The chart below shows the GDPNow survey - a real-time gauge of U.S. growth expectations - with 2-year U.S Treasury Inflation-Protected (TIPS) yields. In recent weeks, the two have diverged sharply: growth expectations have softened, while inflation-linked yields have moved higher.

In other words, markets are beginning to price in a more uncomfortable mix - slower growth alongside persistent inflation.

This concern is echoed in the latest Bank of America Global Fund Manager Survey, where 76% of respondents expect below-trend growth and above-trend inflation (i.e stagflation) over the next 12 months.

But why is stagflation so dangerous for financial markets?

Trampled by a Stag

The reason why stags are so dangerous is that they put policymakers in an impossible bind: faltering growth should justify lower rates but rising inflation needs to be met with interest rate increases. Given that most central bank’s target inflation, many would opt to tighten monetary policy to curb pricing pressure – placing significant strain on global financial conditions.

As policymakers try to grapple with the antlers of low growth and high inflation, investors find that their portfolios get trampled underfoot. Lower growth hurts equity market returns, meanwhile higher interest rates stamps all over bond market returns.

In short, stagflation doesn’t give any room to hide in a traditional equity-bond portfolio. As the chart below shows in 2022 - a year characterized by soaring inflation and slowing global growth - both equities and bonds got trampled.

Where to Hide?

So, if traditional equity-bond portfolios are going to let you down what should investors do to prevent themselves be skewered on the antlers of a charging stag?

Many asset managers have been calling on investors to “diversify their diversifiers” suggesting that a broad range of different assets – property, private assets, precious metals – should offer investors some protection.

Frankly, this “diversify the diversifiers” feels like marketing waffle. In reality, periods of stagflation are relatively rare and any managers suggesting they have a tried-and-test solution for withstanding the stag should be viewed with some skepticism.

I’ve become increasingly dubious of ‘alternative’ assets - they seem to be expensive and have a nasty habit of failing you just when you need them the most. Take a look at my gold post from a few weeks ago for more on that topic!

Under this alternatives banner there has also been a huge promotion of ‘commodities’ as a hedge in a stagflation environment. However, in 2022 - when the stag last stamped investors - commodities was a somewhat mixed bag. Energy prices surged as Russian energy exports ground to halt under sanctions - there is no guarantee that happens again in late 2026. Furthermore, other alternatives like precious metals, hardly rallied at all and real estate collapsed by 25%!

In my mind, if the stag does charge, the most sensible response may simply be to step aside - moving out of risk assets and into cash. It’s not a glamorous solution, but it isn’t an expensive mistake either.

Markets, for now, are offering a window to do just that. Equities have rallied on tentative optimism, but the move feels fragile. Bond and commodity markets - often more sensitive to underlying macro risks - suggest that this reprieve may prove temporary.

Some investors are already taking cover. Swiss Bank, Mirabaud, has recently increased its cash allocation to 15% - a deliberate step back from risk as uncertainty builds.

Cash may still lose value in real terms, but if the stag begins to charge - pressuring both equities and bonds - it can help preserve capital and, crucially, provide the flexibility to re-enter markets once conditions stabilise.

Subscribe now

The Private Credit Blackhole - Part II

Financial Fables — Sun, 12 Apr 2026 14:31:03 GMT

A few weeks ago I compared private credit markets to a black hole - an asset class so opaque that it’s difficult to understand what is happening inside it.

While the inner workings of black holes remain largely unobservable, scientists have developed theories to explain how they behave. It’s possible to do something similar with private credit - to piece together a view of what may be driving the growing instability within the system.

At the heart of the issue there is a simple question: is private credit facing a problem of illiquidity, or something more concerning - insolvency?

Theory #1: Private Credit is Suffering From a Lack of Liquidity

An initial theory is that the private credit black hole is facing an illiquidity crisis - where assets exceed liabilities, but cannot be converted into cash quickly enough to meet demand.

An increasing number of private credit managers have begun restricting redemptions from their funds (known as gating). The table below highlights nine separate strategies, managing over $164 billion in assets, that have gated in recent months. What stands out is not just the gating itself, but the scale of redemption requests relative to the limits imposed - particularly in the case of the Blue Owl funds.

Illiquidity-driven crises are not uncommon (the collapse of Silicon Valley Bank in 2023 being a recent example), and private credit is, by design, an illiquid asset class.

But if the instability within the private credit black hole is primarily a function of illiquidity, then it is, in principle, manageable.

Given time, assets can mature, loans can be repaid, and new capital can be attracted. Periods of stress often create opportunities - allowing those with fresh capital to step in, provide liquidity, and acquire assets at attractive valuations (cough hedge funds cough).

In other words, if private credit is simply illiquid, then with enough time and patience, the system may yet stabilise.

Theory #2: Private Credit is Insolvent

A more troubling theory is that the instability within private credit is not simply a function of liquidity - but of solvency. In this case, the issue is not whether assets can be sold quickly enough, but whether they are worth enough at all.

The lack of transparency in private credit makes this difficult to assess in real time. Instead, it tends to show up in bursts - through sudden and sometimes sharp write-downs by managers.

This was highlighted last week by bond investor Jeffrey Gundlach, who pointed to the scale of write-downs in a major private credit fund (likely BlackRock). The takeaway wasn’t just the losses themselves, but how little oversight we had to the nature and scale of those writedowns.

To me, insolvency seems to be an issue among some of the higher risk firms. Years of intense capital inflows have changed the structure of the market. Increased competition among lenders has weakened underwriting standards, with more leverage, fewer covenants, and tighter pricing. That leaves borrowers more exposed as financing conditions tighten.

At the same time, exposures within the private credit black hole are not especially diversified.

U.S private credit is heavily concentrated in a relatively narrow set of sectors - most notably technology. I borrowed the troubling chart below from my former colleagues at J.P Morgan - private credits 40% of exposure to the U.S tech ecosystem is a troubling structural feature.

If valuations in the tech sector are stretched - or if profitability fails to materialise - then the kind of write-downs Gundlach is pointing to are likely to become more frequent.

Concerns around an AI-driven valuation bubble, persistent lack of profitability, and elevated leverage levels all point in the same direction: that some assets may not simply be illiquid but entirely mispriced.

Theory #3: What if it’s both?

But what if the black hole is unstable due to a combination of both insolvency and illiquidity? This is probably the most terrifying prospect for investors.

Perhaps, the system became unstable following years of poor lending practises - but managers were reluctant to take losses so kept deferring any potential losses with fresh rounds of loans. And that the solvency concerns are being magnified by the structural design of the asset class to create a supermassive, unstable black hole.

I don’t think investors are quite at this stage of panic just yet but to try and gauge their fears I built a simple index of eight of the largest U.S private credit managers - weighted not just by size, but by how exposed each firm is to private credit.

The picture isn’t great. The index is down 27% so far in 2026 and has materially underperformed since December 2021. On an individual basis, the moves have been more severe - Blue Owl is down 44% year-to-date, while Ares is down 38%.

Ultimately, we don’t know what is driving the instability of private credit, in much the same way black holes remain a mystery to us. However, this illiquidity vs insolvency concept at least provides us with a potential framework for talking about this concept asset class and try to chart a way forward.

Subscribe now

The Fog of War is Lifting on a New Investment Landscape

Financial Fables — Sun, 05 Apr 2026 14:31:03 GMT

The fog of war has hung heavy over financial markets for the past month. This week, however, it began - if only slightly - to lift, as President Trump signalled that the United States may be preparing to wind down operations in Iran.

Before the conflict, there was a clear consensus on the market’s path: a risk-on year, shaped by moderating inflation, falling interest rates, and a weaker dollar.

That narrative now lies in tatters.

The finer details remain obscured but as the mist begins to clear, the outlines of a new landscape are starting to emerge - most notably in the shifting trajectories of interest rates, the U.S dollar, and gold.

Interest Rates: Higher For Longer?

One of the clearest shapes to emerge through the fog has been the shift in the outlook for interest rates.

The chart below shows just how sharply expectations have turned since before the Iranian war. Where markets once priced in 3–4 rate cuts from both the Federal Reserve and the Bank of England, they are now leaning toward significantly less cuts (in the case of the Fed) or significant hikes by year-end.

There is a reasonable argument that further tightening may be unnecessary. Inflation driven by higher oil prices is largely supply-led - and therefore unlikely to be meaningfully influenced by changes in short-term interest rates.

However, central bankers are unlikely to take that risk. Having spent nearly five years trying to regain control of inflation after the policy missteps of 2021, they will be wary of allowing another surge in price pressures to de-anchor expectations.

Through the haze, one conclusion is beginning to solidify: rates are likely to remain elevated for much of 2026.

In this higher-for-longer environment, the risks begin to shift elsewhere. Pressure is likely to build on more fragile parts of the system - particularly distressed borrowers in private credit, and in riskier segments of public debt markets such as emerging market debt, suggesting some de-risking in bond portfolios make sense in this emerging landscape.

U.S Dollar Dilemmas

A higher-for-longer rates environment is problematic for much of the global economy - but it remains a powerful support for the dollar.

At the start of the year, a weaker U.S dollar was a consensus forecast amongst the top banks and asset managers but that view was predicated on the assumption that the Fed would deliver 3-4 cuts this year.

But that assumption has not survived the fog of war.

As expectations for rate cuts have been pushed out - the weaker dollar narrative has faded from view, obscured by a world in which U.S rates remain higher for longer.

That said, the longer-term forces pointing toward a weaker dollar have not disappeared. The gradual unpicking of globalisation, a shifting geopolitical order, and mounting concerns around America’s fiscal trajectory all suggest that, over time, investors may begin to diversify away from dollar assets.

For now, the clearer outline emerging through the haze is this: as long as U.S rates remain elevated, the dollar is likely to stay supported through the remainder of the year.

This creates a more immediate problem elsewhere.

Emerging markets, in particular, are highly sensitive to dollar strength. A firm dollar tightens global financial conditions, raises the cost of dollar-denominated debt, and reduces capital flows into riskier markets.

That dynamic is already beginning to show. The MSCI EM index surged out of the gates this year - rising nearly 15% by the end of February, its strongest start since 2012 - but that momentum was built on the expectation of a weakening dollar.

As the chart below illustrates, EM equity outperformance is closely tied to periods of dollar weakness. When the dollar strengthens, or even stabilises, much of the potential outperformance vs DM equities fades quickly.

Gold’s Failure to Protect

One of the most shocking shapes to emerge from the fog is gold’s altered form.

Prior to the Iran war, strategic asset allocators - and major investors such as Ray Dalio - had been arguing that gold should account for 5–10% of portfolios.

Dalio logic was that holding gold provides protection against inflation, currency devaluation and economic instability, since gold tends to hold value when traditional assets struggle… that thesis seems to have faltered in March.

Gold’s failure to protect should not have come as a complete surprise. As discussed previously, gold has shown a persistently positive correlation with equities since the Russian invasion of Ukraine. Investors didn’t seem to mind this positive relationship while both assets were rising - but March provided a sharper test and failed to deliver the protection Dalio predicted.

So how should investors respond?

Some large asset managers have begun calling for investors to “diversify their diversifiers.” But many of these alternative assets - private credit in particular - are themselves showing signs of strain. Expanding further into this space risks becoming an expensive exercise in complexity, with no guarantee of protection.

Perhaps we should just keep it simple.

Swiss bank, Mirabaud, just opted to slash their EM exposure and long-duration fixed income exposure, and move the proceeds into a 15% cash weight in portfolios.

It might not be the sexiest solution but in a world still shrouded in fog, the most valuable asset may simply be the one that keeps you liquid - and ready for when the path becomes clear.

Subscribe now

Has Gold Become Cursed?

Financial Fables — Sun, 29 Mar 2026 14:31:16 GMT

In 1701, Scottish pirate William Kidd - better known as Captain Kidd - was executed in London. In the years that followed, rumors spread that he had buried vast hoards of treasure across the Caribbean.

But the stories carried a darker edge. Those who went in search of Kidd’s gold rarely prospered. Illness struck down diggers. Expeditions ended in unexplained deaths. Investors who financed the hunts were left ruined. Over time, many assumed that the treasure was cursed.

Gold investors today may find the story uncomfortably familiar.

Last week marked one of the worst periods for gold in 40 years. What makes the move more striking is the backdrop - rising geopolitical tension, the very environment in which gold has historically thrived. Instead of acting as a refuge, it faltered.

So what has gone wrong? And, in a more modern sense, has gold itself become… cursed?

Gold as an Geopolitical Hedge

If Kidd’s treasure was cursed, it was not because it lacked value - but because it failed those who sought refuge in it.

For centuries, gold has occupied a similar role in financial markets. It is the asset investors turn to when the world feels unstable - a timeless store of value during periods of geopolitical upheaval.

And, up until recently, this theory seemed to hold up in practice.

The chart below traces the performance of the Bloomberg Commodity Precious Metals Index in the 30 trading days before and after major geopolitical shocks. In episodes such as the Russia–Ukraine war, COVID-19, and even the more recent “Liberation Day,” precious metals responded as theory would predict - rallying between 3% and 15% in the immediate aftermath.

But during the Iran war, precious metal prices have collapsed by over 15% since the start of the conflict - so what has changed?

A Curse or Misaligned Expectations?

If Kidd’s treasure was cursed, it was not because it destroyed wealth - but because it betrayed the expectations of those who sought safety in it.

The simplest explanation for gold’s recent weakness is profit-taking. After a powerful rally in both gold and silver, some consolidation is natural. Investors lock in gains, positions are trimmed and prices soften.

But that feels incomplete and it doesn’t fully explain how gold has behaved.

Instead, the root of the issue is more structural: gold is increasingly trading like a risk-on asset. The chart below shows the rolling 24-month correlation of gold vs global equities. From January 2008 to just before Russia’s invasion of Ukraine in February 2022 - gold had a low and unstable correlation to global equity markets, averaging just 0.07.

But the world changed following America’s decision to freeze Russia’s dollar reserves - and effectively weaponize the dollar. Since then gold’s trades with a much higher average correlation to global equities - averaging 0.64 since the start of the war and the dollar sanctions. Even more remarkably this correlation has become increasingly high and stable in the last few years - consistently trading with a correlation of +0.90 vs. global equity markets.

Investors didn’t seem to be overly fussed about the high correlation of gold and equities when prices were rising together, helping drive large portfolio gains but in the last few weeks we’ve felt the full effect of gold’s changed status.

Rather than being a safe-haven, gold has sold off on escalations in geopolitical tension this month, only to rally when Trump’s rhetoric has softened.

The Holder Dictates the Behaviour

The behaviour of an asset class is dictated by its marginal investor - if the motive and nature of that marginal investor change then so does the underlying performance of the asset class.

And this is what may be underway in precious metals.

Historically, gold and silver sat on the fringes of portfolios. For many investors, allocations were minimal - often less than 1% - held not for return, but for protection. The marginal buyer was cautious, defensive, and largely indifferent to short-term market swings.

Since then, the composition of buyers has broadened.

What was once a relatively narrow and defensive investor base has given way to a far more diverse - and reactive - set of participants:

Central bankers trying to diversify their foreign exchange holdings
Institutional investors treating gold as a core allocation
Speculators chasing momentum
Inflation hedgers worried about pricing pressure
and those positioning for more extreme outcomes in the global monetary system

No single asset can satisfy all of these objectives at once.

Faced with competing demands, gold is increasingly defaulting into a generic risk-on financial asset - driven less by its role as a store of value and more by flows, positioning, and sentiment. The result is an asset that no longer reliably hedges geopolitical stress or inflation, but instead oscillates with broader risk.

The long-term case for gold - currency debasement, fiscal expansion, systemic risk - may still hold. But these forces play out over years, not weeks. In the meantime, price action is dictated by a far more impatient cohort of investors.

This is why gold may now appear “cursed” - so many investors are seeking its magical qualities that its very nature has changed, ultimately hurting rather than helping those hunters holding it.

Subscribe now

The Private Credit Blackhole

Financial Fables — Sun, 22 Mar 2026 15:30:56 GMT

Subscribe now

It’s not possible to directly observe a black hole. We infer its existence only through the influence it exerts on the celestial bodies around it.

I often feel the same when trying to understand private credit markets. The asset class is almost opaque - there is no continuous pricing or transparent exchange, only fragments of information. We are not even sure of its true size: some estimates place it at around $3.5 trillion today, up from roughly $250 billion in 2010.

And so, like astronomers, we are left studying the distortions of private credit: delayed exits, quiet restructurings, the occasional failure that escapes into view.

These are the only clues we have to what is happening inside the private credit black hole - and they are not encouraging.

The Gravitational Pull of Private Credit

If the structure is so difficult to observe, the obvious question is: why does so much capital continue to flow towards it?

The answer, in part, is gravitational.

As the chart below shows, private credit has offered something rare in recent years – high returns and low volatility.

While such an offering is a huge pull, it is partly an illusion. The asset class moves slowly and reports infrequently; this illiquidity disguises some of the underlying volatility and can flatter risk-return metrics.

But gravitational pull has consequences.

As capital has flooded into private credit, competition to deploy it has intensified. When lenders compete, protections tend to weaken.

One way to see this is through the steady rise of covenant-lite lending. Financial covenants are conditions written into loan agreements - such as limits on how much debt a borrower can take on or requirements to maintain a minimum level of earnings. They act as an early warning system for lenders, allowing them to intervene when a company’s finances begin to deteriorate. Remove them, and lenders lose one of their most important safeguards.

The chart below shows the proportion of direct lending deals (the largest segment of private credit) considered to be covenant-lite. By 2025, more than half of largest deals were structured with little or no covenant protection, up from just 2% in 2015.

In other words, as the private credit black hole has grown larger, the rules governing what falls into it have quietly loosened.

Signs from the Abyss

As protections weaken and visibility declines, we are left to study what escapes the system.

One such signal can be found in the growing use of payment-in-kind (PIK) interest within private credit portfolios.

PIK interest allows a borrower to pay interest not in cash, but by adding it to the loan balance - effectively borrowing more to service existing debt. Used sparingly, it can be a useful tool. But it also removes a key discipline: the requirement to make regular cash payments, which is often the clearest indication of financial health.

In earlier periods, PIK formed a relatively small share of total income, with most interest paid in cash - borrowers that could pay, did pay. But more recently, that balance has begun to shift. PIK income has risen materially as a share of total returns, suggesting that a growing number of borrowers are no longer able to service debt in cash. Rather than forcing a restructuring, this stress is deferred, rolled into the loan itself, allowing performance to appear stable even as underlying conditions deteriorate.

Other signals suggest that all is not well in the dark depths of the private credit markets, particularly with market illiquidity.

I’ve been suspicious of private credit liquidity for a while. Marketing efforts in 2024-25 aimed at ‘democratising access to private assets’ felt like a thinly veiled attempt by institutional investors and asset managers to suck in fresh retail capital in the hope of offloading bad loans to an unsuspecting new investor.

Retail investors are poorly suited to private credit markets which rely heavily on long-term patient capital. Meanwhile retail investors tend to need transparency and instant access to their capital.

Investors got a stark reminder of this in mid-March when BlackRock’s decision to restrict withdrawals from its $23 billion private credit fund - only the second such instance since 2008 - quickly highlighted that all might not be well in the private credit blackhole.

Higher for Longer Rates

The expansion of the private credit black hole has taken place largely against the backdrop of ultra-low interest rates.

For a time, the market has been able to absorb tighter policy, operating on the assumption that higher rates would prove temporary - a short-lived disruption rather than a structural shift. That assumption is now being tested.

However, as we talked about last week, surging oil prices is likely to delay further interest rate cuts from many of the major central banks. The Fed this week said that inflation had not come down as “much as they had hoped” and the possibility of further cuts in 2026 now looks unlikely.

Higher-for-longer rates expose what the system has been able to defer. Borrowers that once relied on cheap refinancing, flexible terms, or non-cash interest are increasingly confronted with the need to service debt in real terms. What was previously manageable under benign rate conditions becomes far more fragile when the cost of capital remains elevated.

A recent UBS analysis offers a useful reference point. During both the Global Financial Crisis and the COVID shock, private credit default rates peaked at around 3–5%. Today, however, estimates suggest something materially worse. Based on the deterioration in underwriting standards and the growing reliance on measures such as PIK, default rates could reach as high as 15% over the next 12–18 months as investors are finally forced to confront underperformance.

But the concern is not just the scale of potential losses - it is the way they may emerge. The private credit black hole, by its nature, is opaque, and any deterioration is unlikely to be linear or immediately visible. Instead, it will surface unevenly — through delayed exits, sudden repricings, and failures that appear isolated until they are not, potentially catching wider financial markets off guard as its true scale becomes apparent.

Unable to observe what is happening within it, the greater risk may be what eventually comes back out.

Subscribe now

Are we About to Unchain the Inflationary Wolf?

Financial Fables — Sun, 15 Mar 2026 15:30:44 GMT

In Norse mythology, the trickster god Loki’s eldest son was a wolf called Fenrir. As a pup he seemed harmless enough, but he grew at an astonishing rate. Before long the gods realised they were raising something far more dangerous than they had imagined.

Prophecy foretold that Fenrir would one day break free and help bring about the destruction of the cosmos itself. Fearing what he might become, the gods decided they had no choice but to bind him. After two attempts with ordinary chains failed, they finally restrained the wolf with Gleipnir, a magical ribbon forged by dwarves that was thin as silk but stronger than iron.

Fenrir remained chained. He was not defeated, merely contained.

Inflation behaves in much the same way. It often begins quietly, almost unnoticed. But if allowed to roam unchecked it can quickly become destructive. Containing the last bout of inflation in 2021–22 required heroic effort from central banks, which pushed interest rates to their highest levels in decades in order to restrain the inflationary wolf.

Yet the conflict in Iran is now beginning to strain those chains.

The Wolf Stirs

The surge in oil prices is understandable given the scale of the supply shock. Pipelines and production facilities have been damaged, while the Strait of Hormuz - through which roughly 20% of global oil supply normally flows - has effectively been closed by Iranian missile activity.

Energy markets have responded swiftly. Higher oil prices have fed directly into inflation expectations, raising fears that the inflationary wolf may once again slip its leash.

The chart below shows the recent move in implied inflation expectations.

Throughout much of 2025, inflation expectations had steadily eased as labour market slack increased and wage pressures cooled. The consensus view was that inflation had finally been brought under control.

The recent jump in expectations suggests that confidence may have been premature. Notably, this is not just a short-term concern - two-year and five-year inflation expectations have also ticked higher, indicating that investors fear the shock could persist.

Strain in the Chain

For financial markets, the real concern is not simply that inflation rises again - it is that central banks may be forced to tighten policy once more.

The chart below shows the remarkable shift in interest rate expectations over the past two weeks.

At the end of February - just before the first bombs were dropped on Tehran - markets were anticipating gradual interest rate cuts from several major central banks, particularly in the UK and the United States.

Since the disruption to global energy markets, expectations have shifted dramatically. Pricing in many markets has swung sharply, with derivatives now suggesting that rate hikes could return before the end of the year.

In other words, investors are beginning to worry that the wolf may require stronger chains once again.

How High & For How Long?

For the inflationary wolf to truly break free, the Middle East crisis would likely need to persist for an extended period.

Oil futures suggest that investors currently believe the disruption will prove somewhat temporary. Prices are expected to remain elevated over the next few months but then as we move into 2027, energy prices should moderate. While this still represents a material change from just a few weeks ago - there seems to be a hope that this will all be short-lived.

Betting markets tell a different story.

Polymarket currently estimates only a 3% probability that the conflict in Iran will end by the end of March. Instead, odds imply that the conflict could drag on until May (56% chance of resolution) or even June (64%).

The longer the conflict continues, the greater the risk of lasting damage to energy infrastructure - pipelines, export terminals and shipping routes. The longer oil prices remain elevated, the more likely it is that energy costs begin to seep into broader inflation.

And once inflation begins to spread, it can be very difficult to contain.

Escaping the Jaws of the Inflationary Wolf

If the wolf does slip its leash, where should investors be most concerned?

One area that worries me is private credit. Last week saw a surge of redemption requests from BlackRock’s largest private credit fund, forcing the asset manager to gate withdrawals - a step typically reserved for periods of acute stress.

The private credit market has expanded rapidly in recent years, but lending standards have often been loose and write-downs limited. The hope had been that easier monetary policy would allow struggling borrowers to refinance and stabilise. If interest rates instead remain higher for longer, that hope may prove misplaced.

Equities could also feel the strain. Inflationary environments have historically favoured value over growth, as higher interest rates compress the valuations of long-duration growth assets. Long-term readers may recall my earlier discussion of this dynamic with portfolio manager Daniel Babington - Dan may be on the right side of this debate as value is likely to outperform growth if pricing pressure rises materially.

Finally, investors should keep a close eye on emerging markets, particularly those that have accumulated large amounts of dollar-denominated debt. Countries such as Kenya, Angola, the Republic of Congo and Laos have borrowed heavily in recent years. If global interest rates remain elevated, the risk of debt distress in some of these economies rises considerably.

In Norse mythology, the gods never truly defeated Fenrir; they merely bound him, buying time but never eliminating the threat. Inflation is much the same. Central banks have spent the past few years forging stronger chains to restrain it, and for a time those restraints appeared to be working. But geopolitical shocks have a way of testing even the strongest bonds. For now the wolf remains tethered - the question for investors is whether the recent tremors are simply chains rattling, or the first sign that the wolf is beginning to stir again.

Subscribe now