The wonderful thing about statistics is that you don’t have to make up numbers to mislead. You just have to choose carefully. Pick the best starting date. Use averages instead of the median. Quote the percentage but forget the denominator. Show the winners and quietly bury the losers. Turn correlation into causation. And when all else fails, add a decimal point.

Wall Street has elevated this to something approaching an art form. A fund can outperform depending on when to start measuring. A strategist can have a spectacular record if he cherry-picks the winners and ignores the losers. A stock can look cheap if an analyst uses earnings that haven’t yet happened. And a forecast becomes strangely more convincing when someone changes “about 10%” to “10.2%.”

Read more

“There can be few fields of human endeavor in which history counts for so little as in the world of finance.” John Kenneth Galbraith

A constant challenge for long-term investors is the flood of confusing and contradictory predictions about financial markets and emerging technologies. Many Wall Street analysts seem to believe that they can read tomorrow’s headlines today. However, if Wall Street could reliably predict the future, we would never be surprised by recessions, technologies that fizzle, high-risk financial bubbles and bear markets. Instead, history tells a different story.

The biggest market booms are often accompanied by the greatest confidence, and the worst downturns usually begin just after experts explain why this time is different. From economists and Federal Reserve chairmen to Nobel Prize winners and technology visionaries, some of history’s smartest people have made spectacularly wrong forecasts. The lesson isn’t that experts are foolish – it’s that the future is far less predictable than our confidence would lead us believe.

In a recent article (Cutting Through The Noise), I explained why investors should ignore commentary about the Federal Reserve and the direction of interest rates, daily announcements of monthly economic data, predictions for the future level of the stock market, advice to jump into hot new technologies, and the promotion of exciting short-term trends. This article shows how far those efforts have missed actual outcomes.

Read more

Major market tops rarely announce themselves with a single, obvious event. The following observations individually are not meant to predict an imminent market decline, but collectively they’re signs of a record level of risk in an already overpriced market. Speculation has run rampant. In honor of Leo’s time-honored concept, we have identified eighteen thin reeds in today’s financial markets.

Thin reeds in market structure: market internals and prices become increasingly irrational

Read more

Individual investors can benefit from a similar review. Like that for professionals, the objective for high-net-worth investors is to avoid risks that lead to a permanent impairment of capital, while at the same time preserving enough growth to meet long-term goals. The following stress test 101 follows a series of logical steps.

Step 1: Create a set of likely possible adverse events

Using past historical episodes as templates, create possible scenarios such as a mild recession, an average bear market, a financial meltdown or a liquidity crisis. My consulting service provides a range of detailed outcomes for these scenarios, but for the following analysis I use an average bear market decline, which is more likely to occur than a single worst historical episode. Even so, typical bear markets reduce stock prices by about 35%.

In addition to asset price scenarios, professional investors also place a significant emphasis on the ability to easily access funds. As much as declining prices matter, it’s equally important that a portfolio contains sufficient liquidity in the event an investor needs funds to meet upcoming spending needs.

During a bear market, spreads widen, market depth falls, trade sizes shrink and funding disappears. When this happens, liquidity usually deteriorates 30% to 70% in public markets and 80% to 100% in private or illiquid markets. If an original $10 million portfolio has $7 million of assets at bear market prices, the usable liquidity may be closer to $4 million to $5 million without accepting fire-sale prices. The following analysis includes not only the risk of falling prices but the impact of reduced liquidity.

Step 2: Identify asset classes

With peak stock valuations, current financial markets are especially vulnerable if AI spending slows, interest rates stay elevated, credit spreads tighten, the economy enters a recession or sentiment turns bearish.

To evaluate the impact of these and other risks, a granular identification of specific asset classes produces better insight. For example, small cap companies typically have more leverage, less diversified businesses and depend more on economic growth than larger businesses. So, they generally decline five to ten percent more than large caps in stressful environments. While all stocks fall during bear markets, the following sectors respond differently.

• Technology experiences the widest range of outcomes and poses the greatest risk. From 2000 to 2002 the NASDAQ Composite fell roughly 78%, in the 2008 Financial Crisis by 55%, and in the rate hike bear market of 2022 by 37%. High-priced stocks fell even more. NVIDIA fell 89% in 2008 and 66% in 2022, Meta fell 77% in 2022 as did Tesla by 74%.

• Tech stocks fall more because their valuations depend on projections far into the future and because ETFs and mutual funds often own the same large growth names creating concentrated selling pressure. These stocks normally exhibit greater volatility than the broader market, and investors abandon the fastest-rising stocks first when market sentiment changes.

• Financial stocks are effectively leveraged businesses, and when credit deteriorates loan losses rise, funding costs increase and capital becomes scarce. As a result, financials often underperform during stock price declines.

• Historically, Healthcare has been one of the more defensive sectors because demand is less tied to economic cycles. So, these assets fall but usually less than the market.

• Private equity often appears less volatile because valuations are updated quarterly or less frequently and withdrawals are limited (see my reports, Cracks Appear in Private Assets and All That Glitters Isn’t Gold). If private assets were marked continuously like public equities, many analysts believe their declines would resemble those for leveraged small-cap stocks.

• Investment grade bonds and municipals normally provide diversification against declining stock prices except in the rare case when interest rates rise rapidly, such as 2022 when the Fed abandoned its zero-rate policy.

• High yield bonds behave more like equities than investment grade bonds, because in recessions defaults rise, credit spreads widen and prices decline substantially.

Step 3: Stress Test Your Portfolio Before the Market Does

The following back-of-the envelope stress test is an example of the type of analysis a long-term investor needs to perform to measure portfolio risk. I’m using the impact of a typical bear market, but it’s also possible to test for a financial crisis, an interest rate shock or a more severe market decline. Investors should understand and correct their potential for loss as well as their prospective liquidity to reduce their risk before the inevitable down market.

In general, keep in mind the following possibilities.

Look for exposure to overpriced securities. Valuation risk includes concentration in AI investments that now dominate the market as well as overreliance on index ETFs that promise diversification but hide sector risk (see my report, The Tail That Wags the Market).

Look out for single-point failures. Make sure that no more than 10% of the portfolio ties to a single theme, which substantially increases concentration risk.

Limit illiquid investments. Investments such as private equity and real estate can’t convert to cash for years, even at much lower prices.

Test cash-flow resilience. For investors relying on portfolio withdrawals, estimate whether spending can continue if stock market prices fall, dividends decline and private investments stop distributing cash. The portfolio should be able to fund spending for at least several years without forcing distressed sales.

While there is no such thing as an average portfolio, I have constructed a simple stress test for commonly held assets in a sample $10 million portfolio during an average bear market.

* Amounts in millions of dollars

While these returns look unsettling at first, keep in mind that this incorporates average – not extreme – losses in a bear market. Bear markets are not outliers but occur with regularity in the stock market, and this is the typical impact.

This illustrates why capital allocation and risk allocation are not the same. For the average declines in both market prices and liquidity, the table uses historical averages for at least the last ten bear markets. In this test, the portfolio valuation falls from $10.0 million to $6.4 million. Unfortunately, in bear markets liquidity falls as well, so the readily available funds drop to $4.0 million, a 60% decline from the starting level.

This shortfall occurs because the portfolio is overexposed to high-risk assets, such as equities (66%) and illiquid private asset and real estate investments (25%). While many investors believe they own independent asset classes, these investments move together during stress. Losses could be even greater if either tranche has a major but hidden position in an overvalued sector such as AI. Given the current record valuation for stocks, a reversion to the long-term mean could create even steeper losses in these categories.

In this case, a larger investment in bonds and cash – now only 9% of assets – would provide much more resilience. Reallocation of these assets to 40% to 50% of a portfolio substantially reduces risk, provides meaningful income and produces liquidity for spending needs or even funds to buy stocks later at fire-sale prices.

How long can this go on? Unfortunately, a bear market usually lasts at least half the time of the prior bull market. If the current bull market began in the spring of 2020, then it’s about four years old, which is the average of post-World War II bull markets. Given that length, then an ensuing bear market now could last for at least two years. On the other hand, if this is an extended bull market that began in 2009, which has reached the highest valuation ever, then a market that reverts to the mean would last a lot longer. Either way, an upcoming bear market could try the patience and liquidity of an investor who wants to wait it out to recover losses. Better to avoid major losses in the first place with an annual stress test.

This is just an overview assessment. The table is too general to assess other hidden risks, such as exposure to the boom in AI valuations, but those appear with a more in-depth analysis.

Why do we care?

• No one wants to buy insurance until the house is on fire, but intelligent planning provides for reasonable contingencies. Take on as much or as little risk to enable a sound night’s sleep.

• A rebalance from higher risk assets to those that are less volatile and more liquid can still provide sufficient returns for possible growth. If you own a complex portfolio, with positions across a spectrum of investment vehicles and durations, you may not understand the hidden risks in each investment. My consulting service can analyze and stress test your portfolio for risk factors such as concentration, illiquidity, and private valuation and give you guidance on how to make your portfolio safer and more resilient. Click on this link to set up an introductory call to discuss your needs. Click on this link to learn more about me.

• If considering ETFs and mutual funds, I also recommend a service that ranks these investments against every other ETF or mutual fund with similar characteristics. The database runs a program on more than 14,000 funds with selected and weighted blends of up to 48 different metrics to evaluate any ETF for past performance and underlying risk. As far as we know, it’s the only program that can do this. To test the professional tool – the ProRRT^sm – visit https://prorrt.com/. To access a slimmed-down retail version, go to Rita℠ - The Retail Investment Tracking Application℠

I have written before on these hidden risks (see my report, The Big Passive Bet: How ETFs Quietly Concentrate Market Risk). The forced inclusion of SpaceX into major indexes will further increase concentration in already top-heavy indexes dominated by ten or fewer AI stocks.

Read more

Richard M. Nixon, White House Tapes 1972

Every day, Wall Street produces an endless stream of forecasts, warnings, upgrades, downgrades, “urgent” market alerts and economic predictions. This output produces more forecasts than a psychic convention, each presented with the certainty of the laws of physics. The overall effect, however, is to increase the anxiety that investors have somehow missed a key element of their financial future.

To state the obvious, if Wall Street could reliably predict the future, economists would own tropical islands instead of appearing on cable television explaining why their last forecast was unexpectedly affected by recent developments. Keep in mind that markets are the result of billions of daily global transactions, so the likelihood is next to nil that one or another of these analyses can predict the future consistently.

This flurry of investment advice exists because the financial industry survives on activity, not patience. Investors are constantly encouraged to react to interest-rate forecasts, economic projections, political headlines and year-end market targets that are often revised repeatedly and forgotten shortly afterward. The sheer volume of commentary creates the impression of precision. In reality, much of it is sophisticated noise.

To save time and help investors sleep better, this report cuts through some of the generic noise and offers a few tried and tested pieces of advice for long-term investors. I’ve highlighted a few common themes that get repeated attention.

Ignore commentary on the Federal Reserve and the direction of interest rates

Up until the war in Iran, the primary focus of Wall Street strategists and the financial media was the machinations at the Federal Reserve and the upcoming direction of interest rates. As I have written before, this day-to-day gaming is largely noise for long-term investors (see my report Fed Watching: The Most Popular Waste of Time on Wall Street).

The output for this exercise produces opinions on irrelevancies such as quarter-point rate moves, press conference wording, “dot plots,” the yield curve and inflation surprises. But over the long run, business strategy, earnings growth, productivity, and valuation matter far more than whether rates moved 25 basis points in June versus July.

Ignore the daily announcements and revisions to monthly economic data

Along the same lines, there’s a daily stream of economic data that Wall Street and the media compare to “expectations,” and a “hit” or “miss” often leads to short-term blips in the market. This data ranges from changes in consumer behavior to oil price fluctuations to developments in the housing and labor markets.

Wall Street analysts typically react to this data stream by first revising their estimates for the direction of the economy (almost always up), then using that to predict the direction of the stock market (going higher because of the revised economic outlook). The first problem is the sequence is wrong: financial markets usually lead the economy by three months to more than a year. It’s the reason why the S&P 500 is included as a component of the index of leading economic indicators.

The second problem is it’s impossible to predict something as complex as the economy for any length of time (see my report Why Markets are Unpredictable). The proof for this statement is no one has done it well consistently.

The following chart of Goldman Sachs’ projections for global economic growth in 2026 is a case in point.

The blue line shows the changes in the firm’s estimate for global growth as economic data rolled in during 2025 and early 2026. It shows no fewer than 20 changes in 15 months, that ranged from a low of 2.2% to a high of 3.0%, with no guarantee that any of these will be correct. Anyone investing based on these projections should be anxious.

Wall Street economic projections are also contradictory, adding to the confusion. Here is a sample of recent Wall Street pronouncements that cannot be true simultaneously.

• “AI is a transformational productivity revolution” as opposed to “AI is the largest bubble since the dot-com era.”

• “The economy is headed for a soft landing” as opposed to “Recession risks are elevated.”

• “Tariffs won’t matter” as opposed to “Tariffs will reignite inflation.”

• “The consumer is resilient” as opposed to “The consumer is cracking.”

None of these can help predict the level of the economy or financial markets ten years from now.

Ignore the constant Wall Street predictions for the future level of the stock market

“Forecasts may tell you a great deal about the forecaster; they tell you nothing about the future.” — Warren Buffett

As with economic projections, most annual stock market forecasts are entertainment disguised as analysis. Wall Street analysts routinely predict, especially in January, year-end S&P 500 targets based on such unknowable factors as recession timing, election impacts and sector rotations. Not only do these efforts miss major market distress but few consistently outperform random guesses for any length of time.

On the other hand, long-term investment success is driven by starting valuations, avoiding major mistakes, investment discipline and tax efficiency more than forecasting precision.

Ignore Wall Street’s advice to jump into hot new technologies and ignore cash as a tactical asset

These opportunities are excellent advice for Wall Street fee generation, but it exposes investors to the risk of losses from overvalued investments and unwanted drawdowns. Sometimes holding cash is prudent, reducing leverage matters and preserving capital remains a priority. Avoiding catastrophic loss often matters more than squeezing out the final 2% of upside.

Wall Street fans the fear of missing out with blue sky predictions of exciting new technologies (see my report Wall Street Loves Hype). It urges investors to jump on the bandwagon and ignore valuations, yet major investment mistakes occurred with the adoption of disruptive technologies, such as railroads, radio, widespread computer adoption, the internet and the dot.coms, and now AI (see my report The AI Gold Rush: Chasing the Rainbow).

Every major bubble has its intellectual justification. These novel technologies were profound and real, but the investment conclusions drawn from them were wildly exaggerated, leading to major losses.

Ignore Wall Street’s promotion of huge short-term gains

The financial media breathlessly report on huge gains from various investments, sectors or themes. As a result, a huge flood of Wall Street marketing relies on recent returns as the concerns of months ago are forgotten (see my report Recency Can Cloud Long-Term Investment Decisions).

What those headlines don’t reveal are the effects of this short-term outperformance from leverage, concentration, momentum exposure, favorable market regimes and, most importantly, luck. The key question should not be, “How much did this make?” but “What risks were taken to achieve those returns?” A recent ad on Bloomberg promotes examples of short-term investment returns of more than 100% using the advertiser’s option strategies, which the viewer can access for free and only $29.95 for shipping and handling. What the ad doesn’t mention is the strategy’s level of risk and combined losses over time that it took to produce that outcome.

For long-term investors, the greater risk is large drawdowns, illiquidity and sequence-of-return risk. Remember, a 50% loss requires a 100% gain just to recover. That arithmetic never changes.

Wall Street advice rarely mentions risks that truly affect long-term investors

I have repeatedly warned about the risks in this market from overvaluation and the hidden risks in a variety of popular investments, from rising index concentration to hidden risks in private assets that should be of key concern for long-term investors (see my reports on Valuation: The Anchor Wall Street Wants You to Ignore, the two part series Beyond Forecasts: Three Real Risks Facing Markets and The Big Passive Bet: How ETFs Quietly Concentrate Market Risk). Apart from a few scattered newsletters, these issues are rarely addressed in Wall Street research.

Ignore often contradictory Wall Street slogans that are excuses to limit rational analysis

Wall Street has a saying for almost everything. Wall Street brokers are famous for using well-worn phrases to keep investors constantly trading their accounts. Here a few examples of common slogans that contradict each other that a long-term investor should ignore.

• “When the ducks are quacking, feed them” as opposed to “Trees don’t grow to the sky.”

• “Sell in May then go away” as opposed to “This time is different.”

• “Don’t fight the Fed” as opposed to “Markets bottom before the economy.”

• “Buy low, sell high” as opposed to “The trend is your friend.”

• “Nobody ever went broke taking a profit” as opposed to “Let your winners run.”

• “Don’t fight the tape” as opposed to “By the time you read it on the front page, it’s already in the price.”

On the other hand, there are timeless truisms that you hear rarely.

• “In the short run, the market is a voting machine; in the long run, it’s a weighing machine.” — Benjamin Graham

• “Rule No. 1: Never lose money. Rule No. 2: Never forget Rule No. 1.” – Warren Buffett

• “What feels safest is often riskiest.”

• “Leverage is a way of turning a small mistake into a life-changing event.”

• “The four most dangerous words in investing are: ‘This time it’s different.’” – John Templeton

• “The five most dangerous words in finance are: ‘It works in theory.’”

Returns are set before you invest—through valuation. That’s why experienced investors focus less on slogans and more on valuation, risk management, time horizon, liquidity and investment discipline.

Why do we care?

• I hope this critique of common Wall Street advice reduces your anxiety about trying to follow the fire hose of information from investment reports and the financial media. More data doesn’t automatically mean more predictive power.

• Investment success depends on avoiding potential losses, which are often hidden in rising markets. Bull markets don’t eliminate risk. They anesthetize it.

If you’re concerned about the hidden risks lurking in your portfolio, I provide professional, independent consulting advice to investment advisors and independent investors. You can learn more about my services here: Consulting – Apex Equity Research.

Years of rising asset prices create the impression that most portfolios are stable, diversified, and resilient. The opposite is often true. Leverage builds quietly, liquidity assumptions go untested, and correlations that once seemed independent begin to move together. The system looks strongest just as it becomes most fragile. But portfolios are not transparent. By the time risk becomes obvious, it is no longer manageable.

Most investment reports focus on returns but very few explain the underlying risk that drives those returns. Financial cycles tend to last around 15 to 20 years, and it’s been nearly two decades since the last downturn, so there is scant analysis of investment risk in Wall Street reports and headline stories. Few are aware of hidden risks: a recent report found that only 8.8 percent of allocators feel like they know the true investment risk in their portfolio (Institutional Investor). However, the amount of hidden risk in a portfolio has a dramatic impact on long-term investment performance when markets suddenly turn.

Wall Street strategists usually define risk as volatility — not capital impairment. Volatility measures how much the price of an investment moves up and down, but real risk is the chance that you’ll lose money. So, using past volatility to measure risk is like reading yesterday’s weather report to plan an outing.

A real risk assessment requires a structured analysis, and investors rarely model drawdowns in dollar terms compared to their ongoing and prospective cash needs. If you’ve never seen your portfolio analyzed this way, you’re not alone.

This review identifies major areas of hidden risk and discusses investors’ risk tolerance when considering a portfolio’s underlying liquidity, investment concentration, hidden leverage, potential correlation breakdowns, and absolute drawdowns.

Read more

A friend of mine who used to manage the bond portfolio at U.S. Trust would often remind me that, “financial events cast long shadows.” What he meant was that major changes in markets don’t occur because of a recent event or two; they’re the result of years of accumulated events that lead to unsustainable positions.

Using a metaphor from complex systems research, markets fail the way avalanches develop – they build with a long, invisible accumulation followed by a moment that looks trivial in isolation that starts the decline. By the time the final snowflake falls, the outcome was already determined. The only uncertainty is timing.

What matters most, then, isn’t predicting the trigger – it’s identifying where the snow has been quietly piling up. Today, one of the largest accumulations sits in private equity and credit, where years of easy money, opaque pricing, and rising leverage have created a structure that appeared stable – until the first cracks are now forcing investors to question the wisdom of financing illiquid assets with “semi-liquid” claims. The recent surprises with Blue Owl Capital and the growing redemption requests at other private asset managers look like the beginning signs of an avalanche.

Read more

Wars, terrorist attacks, political crises, and other dramatic events that dominate headlines seem like obvious triggers for market turmoil. The outbreak of the war in Iran two weeks ago raised an already high level of investor uncertainty even higher due to the war’s potential impact on the energy markets as well as pressure on interest rates from a heightened risk for inflation.

Surprisingly, history repeatedly shows that markets usually absorb the shock from wars with surprising resilience. After the outbreak of hostilities, markets adjust to the new reality, then usually rebound to their pre-war pattern, even rising in value during the course of the war. This report outlines the oft-repeated reaction of financial markets to war and explains the reason for investors’ resilience. It’s unlikely that the current war will be the trigger that causes the next major downturn.

Keep in mind, this is one of the most overvalued public and private markets in history, so it remains a high-risk environment. To put the current situation in perspective, we explained in our last note (see The Limits of AI in Investment Analysis) the real reasons for major market failures:

“After the financial crisis in 2008, the physicist Didier Sornette performed a detailed autopsy of major market downturns in his book, Why Markets Crash (2009). He found that the danger wasn’t any obvious shock to financial markets but a long period of reinforcing optimism that quietly built system fragility. Periods of accelerating prices lead to speculative bubbles where investors are increasingly influenced by listening to each other rather than the underlying valuation reality.”

As a result, “… major downturns did not result from major events, such as wars or bank failures. They arose from the internal dynamics of the market itself as rising prices built self-reinforcing feedback until the system became unstable …” then fell from exhaustion.

A good analogy is the formation of avalanches, which like financial markets are also complex systems. As snow builds up on the mountain, it only takes a few final snowflakes to bring the pile cascading down. For example, historians still debate the proximate causes of the crash in October 1929. The current cracks in the private asset market are more likely to have a destabilizing effect than war.

An examination of financial markets during wars in the last century – even worldwide conflagrations – shows very little impact on the long-term direction of markets. To be more specific, the pattern upon the outbreak of war takes three general phases.

The Immediate Shock

• Immediately before the outbreak of war, uncertainty intensifies and stocks decline slightly leading to short-term volatility. They decline further once war begins.

• Stephen McBride of Mauldin Economics examined more than 50 geopolitical conflicts since the 1950s and found on average, US stocks drop around 7% in the days before the start of a conflict.

• The effect of the Iran War on the current market is even more muted than average – as of today the S&P 500 is down by only 2.6%.

The subsequent rally

• As war progresses, markets usually rise as investors adapt, and the extent of the war becomes clearer.

• McBride found that one year after the beginning of war, 85% of stocks are at or above their prior levels with a median gain of 7%.

Long-term effects (months or years)

• If the conflict is limited, markets return to their pre-war levels quickly.

• If the conflict becomes global, prices for commodities and other supply chain materials rise, and government spending on the war increases, which also fuel higher market levels.

• Currently, the market anticipates a short, limited war based on the belief that Trump doesn’t have the motivation for an extended conflict.

Remarkably, the market’s reaction to war was similar for global conflagrations as well as more limited outbreaks. Here’s a brief summary of wars in the 20^th century leading up to the 2003 Gulf War.

Global wars:

• 1914 – Outbreak of World War I. When war broke out in July 1914, financial markets experienced one of the most dramatic disruptions in history. The New York Stock Exchange closed for over four months (July–December 1914). Stock prices initially fell, then rallied strongly through 1915, which became one of the strongest markets in the 20th century.

• 1941 – U.S. Enters World War II. After Pearl Harbor in December 1941, stocks initially fell, but the market bottomed in April 1942. Due to massive government spending, the S&P 500 rose roughly 50% during the rest of the war.

• 1950 – Korean War. At the outbreak of the war in June 1950, stocks dropped about 5–7% in the first weeks, but the market recovered quickly and gained roughly 20% within a year due to massive U.S. defense spending.

Limited or regional wars:

• 1991 – Gulf War. Markets were weak during the buildup to war in 1990 because of Iraq’s invasion of Kuwait and oil price spikes, but once the air campaign began in January 1991, uncertainty collapsed and the S&P 500 rose about 15% in the following two months.

• 2003 – Iraq War. Leading into the war, markets were already depressed by the Dot-com Bubble collapse, but once the invasion began in March 2003, investor uncertainty dropped and the economic recovery accelerated. As a result, the S&P 500 rose roughly 26% in 2003.

The outbreak of the Iran bombing has produced a more muted initial response even though it’s difficult to tell how long the war will last. No matter the length of the conflict, the above examples imply that the markets are unlikely to decline further solely due to the war.

Why do we care?

• It’s a waste of time for long-term investors to significantly rearrange their portfolios due to the Iran War.

• This is still a high-risk market due to other reasons, so remain cautious. See our reports, Valuation: The Anchor Wall Street Wants You to Ignore, Beyond Forecasts: Three Real Risks Facing Markets (Part 1), and Beyond Forecasts: Three Real Risks Facing Markets (Part 2).

New technologies emerge during these periods and seem to confirm the market narrative. Each computing innovation appears to solve the flaws of the last era — until it doesn’t. Didier Sornette, in Why Markets Crash (2009), found that major market collapses are often preceded not by obvious shocks, but by long stretches of reinforcing optimism that increased computing capability can’t capture.

The latest supercomputing claim is the belief that AI can now replicate institutional-grade investment analysis — replacing Bloomberg terminals and six-figure analysts with a few clever prompts.

Read more

Likewise, the most dangerous risks in financial markets are rarely the ones investors argue about on television. They are the risks few people even realize exist. Today, beneath record stock prices and calm headlines, a vast shadow system of debt has grown larger than the regulated banking sector itself. It is complex, opaque, and deeply interconnected—and in a crisis, it has the potential to move markets faster than anyone expects.

Understanding that hidden architecture is the purpose of this report. The first part of this report (see Beyond Forecasts: Three Real Risks Facing Markets) discussed the risks of record high equity valuations and the huge rise in private investing. This part will explain a third risk – the hidden debt in the global financial system that exists outside the purview of regulators. It is possibly the most dangerous risk, yet given its complexity, it is virtually absent from any Wall Street investment advice.

Read more

As predictions for 2026 proliferate, we focus instead on three structural risks that history suggests matter far more than near-term forecasts. The first two risks – record high equity valuations and the huge rise in private investing – seem hidden to Wall Street strategists and the financial media. They’re rarely discussed and widely ignored.

Read more

Now a more complicated force is entering the picture: an administration openly intent on reshaping the Federal Reserve. What happens when political pressure meets persistent inflation, soaring deficits, and a bond market that isn’t easily fooled?

What we can see clearly is that the Federal Reserve’s independence is now squarely in the political crosshairs. Unfortunately, politicians pushing for greater outside control are even worse at anticipating the unintended consequences of monetary policy than economists.

The Fed has faced political pressure before – sometimes it ended with a rate cut; sometimes it ended with inflation locked into the national psyche for a decade, which means the next chapter in interest-rate policy will not be straightforward. And, even if Fed policy becomes clear, the outcomes from that policy are not and may include sudden drops, unexpected twists, and a few moments where you wonder who’s steering the ship.

This article explores how politics is reshaping the Fed, what drives long-term interest rates and why long-term returns in the stock market depend more on valuation than on the next move in Washington.

To begin, we can make one safe prediction: Trump will reduce the Fed’s independence in May when he appoints a new chair who, along with other Trump appointees, will move quickly to push for substantially lower interest rates.

• Already Trump’s call for the resignation of Fed Chair Jay Powell, the attempted firing of Lisa Cook and his repeated demand that the Fed lower interest rates to zero have shaken the market’s perception of the central bank’s independence. Trump already appointed two members to the Federal Open Market Committee (FOMC) – Christopher Waller and Michelle Bowman – and his recent appointment of Stephen Miran to the Fed, who will also retain his job at the White House, has further eroded the perception of independence.

• Trump’s most important appointment will be the replacement in May of the current Fed Chair, Jerome Powell. The Fed chair has a significant influence on FOMC decisions.

• The FOMC consists of twelve members – seven nominated by the President and confirmed by the Senate and five appointed by regional Federal Reserve Banks. That means Trump has already appointed three out of twelve seats and will have appointed four out of twelve (25%) after he appoints a new chair in May. The Board of Governors also sets another short-term rate, which we discuss below.

• With the new Trump appointees, the votes on the FOMC are now showing discord, unlike the usual unanimous votes taken during the last decades. A vote of the FOMC in July saw two dissents for the first time since 1993. The October meeting also saw two dissents from different members than those who dissented in July.

Since 2008, the process for controlling money, and therefore short-term interest rates, has changed, perhaps giving Trump even greater control over policy.

• The Fed has for decades used the “federal funds rate” as the headline target for its key policy rate, which gets all the media attention. This rate – a target set by the FOMC – is the rate that banks use to lend overnight to each other. However, in 2008 the Fed flooded the banks with excess reserves using quantitative easing, or QE. As a result, banks no longer need to borrow from each other, making this rate obsolete.

• Now the Fed maintains control over short rates using the interest rate it pays on bank reserve balances. That sets a floor for short-term rates because banks won’t lend their balances at a lower rate than they can earn from the Fed. This rate is not set by the FOMC, but rather by the Board of Governors. By May, Trump will have appointed three out of the seven members of the Board (43%), a greater percentage than his control over the FOMC.

Given this situation, investment advisors get asked, “Should I be concerned if the Fed loses its independence to make monetary policy?” Hard to say, but keep in mind that political efforts to sway the Fed are nothing new.

• Every US president wants lower interest rates to stimulate the economy and boost their economic performance (and popularity). The President and Congress, as well as Wall Street, almost always jawbone for lower rates. A well-known confrontation happened in 1965 when Lyndon Johnson invited Fed chair William McChesney Martin to the Texas White House, and LBJ picked him up and threw him against the wall to make his point. Martin caved and subsequently lowered short-term rates, which contributed to crippling stagflation.

• As another cautionary tale, around 2022, Recep Tayyip Erdogan, President of Turkey, beat the same relentless tune as Trump for sharply lower interest rates. He finally took control of the country’s central bank, which promptly cut interest rates dramatically even though the country had persistent inflation. As a result, the inflation rate in Turkey soared to over 72%. To tame this wild inflation, the central bank was then forced to raise interest rates, which reached 50% in 2024.

• Asset inflation. Setting interest rate policy is hard enough without political meddling. Many recognize that the Fed’s easy money policies in the early 2000s kept mortgage rates too low for too long and led to today’s bubble in housing prices.

Even with Trump’s influence, his effort to pack the Fed does not guarantee lower long-term rates.

• If the Fed lowers short-term rates dramatically, it risks losing credibility for fighting inflation. In this scenario, if the Fed slashes short-term rates and the market then fears inflation from such a large drop, long-term interest rates will likely rise. Long-term rates – especially the yield on the 10-year Treasury note – matter more to the economy than the Fed’s short-term rates because they serve as the benchmark for most lending, including commercial loans and mortgages.

• The Fed’s decisions have much less of an impact on long-term rates, which are instead set by millions of participants in the market for other reasons, such as the path of the economy and individual hedging and portfolio allocations. For perspective on the impact of interest on long-term investment decisions, see our report All Eyes on Interest Rates.

• Even now, the Trump economic team is promoting monetary policies that could produce other unintended results. Several on the short list for Fed Chair, as well as Treasury Secretary Scott Bessent, argue for a resumption of quantitative tightening (taking money out of the economy) through reduced bank reserves. That would tighten the supply of money for loans, which ironically puts upward pressure on long-term rates, the opposite of their intentions.

Maintaining price stability (i.e., managing inflation through monetary policy) is one of two mandates that Congress has assigned to the Fed.

• All this said, the risk of inflation is the counterbalance to big rate cuts. Since the end of COVID-era supply disruptions, inflation has been persistently sticky and remains one hundred basis points above the Fed’s stated target of 2%. The lesson of the runaway inflation in the 1970s is that short-term inflation not addressed gets fixed into the minds of consumers, making it much harder to fight as time passes.

• Trump’s policies on tariffs, immigration and the federal deficit are all highly inflationary, and lowering rates in the face of these actions could add gasoline to the fire. To date, American importers have used creative short-term strategies to soften the impact of tariffs on prices, such as overordering inventory before the tariffs took effect, but as companies work down those stockpiles, that cushion will evaporate, and they’ll be forced to raise prices.

• The central bank’s ability to manage long-term rates is further compromised by soaring federal deficits, which the President and Congress seem unable to contain (see our post Should you care about the Federal Deficit?). To fund expanding deficits, the Treasury Department needs to issue even more Treasury bills, notes and bonds, which increases supply. All things being equal, the larger supply lowers the price for these investments, which means higher rates.

• As the recent fall elections revealed, affordability – or the impact of inflation on everyday purchases – remains the top concern of Americans. Consumers are also facing higher prices in a range of services that aren’t included in the consumer price index, such as record high home prices, higher property taxes and soaring homeowners’ insurance. If inflation persists, political pressure could switch from an insistence on lower rates to a focus on managing affordability, which means higher rates.

• In summary, it’s just as likely that a Fed cut could lead to higher long-term interest rates if the market perceives it’s failing to control inflation. Even though Trump’s influence over the Fed will grow, it’s not clear how this will play out. Ultimately, the market – not the Fed – determines long-term interest rates.

Given this uncertainty, how will upcoming Fed policy to lower short-term rates affect the stock market and my portfolio?

• Even knowing the direction of Fed policy, it’s difficult to know how the stock market will respond to short-term rate cuts. When the Fed lowered interest rates in the past decades, it produced mixed results on stock prices (see our report Fed Watching: The Most Popular Waste of Time on Wall Street.) As one recent example, when the Fed started lowering the Fed Funds rate last December, the rates on the 10-year Treasury note, which determines the rate for most commercial credit, actually rose and had little impact on equities.

• Over the past decades, the impact of Fed cutting cycles on the stock market has been “extremely mixed” (Enterprising Investor, CFA Institute). While one could expect a positive impact, all things being equal, the path of markets is largely influenced by the global economic environment.

• The best predictor of long-term returns is current stock market valuations, not Fed timing (see our reports Valuation and Risk Management and Valuation: The Anchor Wall Street Wants You to Ignore). The impact of Fed cuts on the stock market depends on the level of stock prices when it begins to lower rates, as well as many other economic factors, and stock prices are now at record highs. Since 1965, there have been 12 distinct rate-cutting cycles, which failed to consistently produce higher markets.

The most popular options to hedge rising inflation have their own issues.

• At current high asset prices, there are few low-risk options to hedge against uncertainty. The two investments most often mentioned as hedges both have sizable issues. Gold has traditionally served as a hedge against inflation and political turmoil, but the price has more than doubled in the last year, so it may already reflect investor fears of inflation and political turmoil. Gold has risen with central bank and new gold ETFs buying bars and coins and has become the world’s second highest reserve asset. A modest position can serve to diversify away from stocks and bonds, but at these levels it carries some risk.

• Wall Street also touts cryptocurrency as a counterbalance to economic distress. The bull case is that it’s a hedge against currency debasement by central banks (i.e., long-term inflation), but it trades nothing like a counterbalance or a store of value. Instead until recently, Bitcoin has tracked closely with price changes in the Magnificent Seven high-tech stocks that dominate the market, which speaks to its investment appeal as speculation rather than investment. Bitcoin has no intrinsic value, so it’s impossible to tell what it’s worth, and we need to know that to calculate a reasonable floor for the price. Its price fell by more than 33% since October, so basically, it’s a measure of animal spirits.

Why do we care? All this gets back to our original comment: no one knows what long-term rates will do, especially over the long run, which is the period we care about. For long-term investors, ignoring the Fed guessing game isn’t negligence; it’s discipline.

Focus on fundamentals—valuations, earnings power, and long-term opportunity. It saves time, reduces anxiety, and leads to better decisions. Upcoming reports will explore other possible hedges to inflation.

Yet behind the spotlight, the economics tell a far more complicated story—one of soaring costs, elusive revenues, fragile demand, and unbridled speculation. This note looks past the stagecraft to examine the machinery underneath and find the signals through the noise.

The results from generative AI (GenAI) models are impressive, leading to enthusiastic, but as yet distant and imaginary, benefits. These models read all possible information from the Internet and uses large language models to string words together to produce encyclopedic answers. Wall Street believes that this ability will unlock a massive productivity supercycle and ultimately cure diseases, revolutionize education, streamline the law, solve climate change and usher in a new age of prosperity.

Investing in GenAI is not as easy as these possibilities suggest. It’s important to distinguish between the prospects for the technology, and the price investors are paying for a piece of that future. Every mania begins with the kernel of a new and exciting technology – manias generally don’t emerge from bad ideas. John Kenneth Galbraith called this the indispensable element of truth. Railroads, national electrification and the internet were all major technological advancements that provided profound benefits, yet as we show below, the initial overinvestment in these themes led to overcapacity and crippling investment losses.

The returns from investing in GenAI capacity now are fraught with all sorts of risks. The current GenAI investment boom – centered on building massive data centers and associated infrastructure – has become a speculative bubble, driven more by the fear of missing out than by sound economics.

Rapid GenAI adoption has so far produced mixed results.

• Great features with sizable drawbacks. It’s easy to salivate over the potential to add a natural language interface to every human interaction, but GenAI often produces inaccurate results.

  • Garbage in …. While GenAI results are impressive to read, they’re only as good as the underlying data, which is a flood of sometimes unreliable and conflicting information from the Internet. AI does not create new ideas; it just scans the Internet to group words and images that appear together.

  • As a result, GenAI answers can be misleading, and so-called “AI slop” is growing exponentially (Institutional Investor). As these models run out of new Internet data and increasingly train on other AI-generated content, their results have a greater tendency to produce “hallucinations,” and they sometimes invent data to complete unfinished answers. To correct these errors requires a comprehensive human review of essentially black boxes.

  • Partial answers. GenAI doesn’t “read” entire data collections but stops once it has enough data to generate a credible response. GenAI can also misinterpret text out of context. A troubling report found that a program that created medical advice for cancer patients produced many misleading and inaccurate results, and reports from law firms found numerous references to legal cases that didn’t exist.

  • GenAI has already vacuumed up all the information on the Internet, so it has turned using output generated from other large language models to train its algorithms, which progressively degrades output further.

  • Bias. When AI is trained on human generated data, such as the posts on chat rooms, it adopts all the human biases. For example

  • According to researchers at the Sloan School at MIT, GenAI will also face challenges with intellectual property restrictions as it downloads all sorts of proprietary reports and images from conventional news sources and research institutions.

• The unlimited potential for GenAI may actually be limited. As companies rush to incorporate AI into their operations – such as customer service – the benefits have remained elusive.

  • GenAI adoption has occurred at the fastest pace ever for a disruptive technology. In the three years since the launch of ChatGPT, more than 54% of Americans now use the technology, which is 2.8 times faster than the rate of adoption for personal computers and 1.8 times faster than the adoption of the Internet. However, the rate of adoption has begun to slow for larger firms that have begun to use the technology. In the three months since June, the adoption rate at firms with more than 250 employees has declined by 14% (US Census Bureau, Macrobond, Apollo).

  • Part of the problem is that GenAI adoption has produced mixed results. Massachusetts Institute of Technology (MIT) published research showing that despite U.S. companies already spending upwards of $40 billion on Al adoption, 95% of these organizations have seen little if any real value or monetary return.

  • A five-year Harvard Business School study of AI generated work schedules for more than 300,000 employees at large retail chains found that managers needed to make manual overrides in 84% of the cases. The study also called into question GenAI’s ability to manage inventory levels or pricing as well as its role in hiring decisions.

  • Nick Rubright, CEO of Rank Media, sums it up. “There’s a lot of talk about Al being a human replacement, and lots of Al startups claim their tools can replace workers, but I’ve never found this to be true because these new tools still need management” (Epoch Times).

So far, revenues from GenAI services are minimal or nonexistent.

• Low or no revenues per search. Many GenAI searches on platforms such as ChatGPT are free or at most charge a minimum fee for large scale users, and the most popular GenAI searches are for those on which Google does not make money. Even the modest subscription fees are too low to cover high variable costs.

• Low barriers to entry will keep prices and profits low. There are already scores of GenAI models, and all use the same data source: the Internet. This speaks to low barriers-to-entry, so GenAI models risk becoming a commodity. Ultimately, high fixed costs models with low barriers will see revenues fall to the level of the variable cost, which could keep prices low.

  • The market is flooded with GenAI competition. The major, highly rated platforms include ChatGPT (open AI), Claude (Anthropic), Azure (Microsoft), Watsonx (IBM), Firefly (Adobe), Lumio AI, ImagineArt (Vyro.ai), Vertex AI (Google), Gemini (Google), Bedrock (Amazon Web Services), Jasper, Copy.ai, GitHub Copilot, Synthesia, Grammerly, and Grok (xAI).

  • International competition. DeepSeek, with made-in-China GenAI models released earlier this year, is achieving results comparable to the best GenAI models using less-advanced chips and innovative model training at one-fourth the cost. It is already among the top U.S. GenAI downloads.

The cost to provide GenAI is ridiculously expensive. In addition to monumental fixed costs to create the infrastructure, the operating costs for each search are also expensive and will likely rise as usage scales. In a Working Knowledge report from Harvard Business School, Professor Andy Wu says, “while GenAI’s promise is clear, the potential return on these massive investments remains murky.”

• Huge Capital Investment. The data centers to process GenAI are enormously expensive to build.

  • Record CapEx. This year alone, Google, Meta, Amazon and Microsoft will spend more than $350 billion on GenAI data centers or double the cost in one year (inflation adjusted) of the entire Apollo moon program.

  • Looking ahead, Goldman Sachs Research projects that the five highest-spending US enterprises building AI data centers will make a combined $736 billion of capital expenditures in 2025 and 2026.

  • Expect a big drain on capital markets. Until now, developers have paid for their GenAI investment with funds generated from operating cash flows, but there’s no way that this source can fund future ambitions which will rise to $500 billion per year. This means that these companies will have to raise huge amounts of funds from equity and credit markets, which will further dilute the return on GenAI investments (Zero Hedge).

• Huge capital investment will need even larger revenues. Spending on GenAI is at epic levels and magnitudes above levels that can provide an adequate return.

  • Morgan Stanley forecasts the global investment in data centers will cost $2.9 trillion by 2028. This is what’s driving chip sales at companies like Nvidia.

  • However, it’s not clear if this investment can produce an adequate return. A report from Bain & Company estimates that the current level of investment will ultimately need $2 trillion in revenues to produce even a modest return.

• GenAI search is costly. In addition to the massive investment required to train models and build capacity to scale – the technology suffers from high operating costs per search with as yet no or minimal revenues per search.

  • Huge power hog. There are about seven thousand data centers worldwide, with the largest concentration in the United States. In 2022, American data centers consumed roughly 4 percent of all U.S. demand but given that ChatGT uses 17,000 times the electricity of the average American home, Goldman Sachs predicts that global data center power demand will increase by 165% by 2030.

  • Surging utility bills for consumers. Current GenAI queries use so much electricity that the DOE warns that within the next year several US cities could suffer rolling blackouts, and communities near data centers have already seen electric bills more than double. Sixty percent of Santa Clara’s electricity now goes to GenAI data centers, and in New Jersey, Pennsylvania and Ohio annual electric utility bills in data center areas increased this summer by as much as $300. The Department of Energy warns that blackouts will rise by 100 times in the next five years.

  • Power demands for each search will get worse. Unfortunately, the newest GenAI models use more energy with each successive generation. Academics estimate that the newest version of ChatGPT – ChatGT-5 – uses up to twenty times more power than the original version, and the newer chips, such as Nvidia’s next generation, draw more kilowatts per server with each successive version.

  • US already near electric capacity. PacifiCorp has already failed to deliver more than a minimum amount of power to four data-center campuses in Oregon. In the longer term, data centers will have to build their own expensive, dedicated power plants to supply expected power demands.

  • With high variable costs and low or nonexistent revenues, profitability remains a distant dream, and those companies that are producing GenAI services are losing boatloads of money on each search. So far, market enthusiasm has overlooked the current lack of any potential profitability, but at some point investors could become impatient with insufficient returns on these massive investments.

• If that is not enough, the current boom in GenAI investment comes from circular payments, a major red flag that has produced much corporate distress in the past. The practice begins when a GenAI chip company lends money to the company building the data center, who then turns around to buy the company’s chips. When a company lends to its customers to buy its products, the discipline to keep lending standards pales with the temptation to show immediate sales – the company is essentially trading short-term gains for longer-term risks.

As cautionary tales, the country faced similar technological revolutions in the past with the construction of the railroads, the push for electrification and the buildout of the Internet, yet rampant speculation in these transformations led to severe overbuilding and massive losses for investors. Ultimately, these technologies produced enormous benefits, but the process matters for investors: huge infrastructure spending using leverage led to massive overcapacity, which led to bankruptcies and restructuring, which ultimately led to much lower costs that encouraged adoption. You want to be at the end of that sequence, not the beginning.

Railroads radically transformed the nation’s transportation network.

• Between 1866 and 1890, the U.S. quintupled its track mileage, far outpacing both population growth and the expansion of freight demand. Economists generally agree that perhaps 25% to 40% of the track built in the 1870s–1880s was economically unnecessary at the time.

• By 1890, over 25% of U.S. rail mileage went into receivership.

The process of electrification at the beginning of the 20^th century let to profound benefits for the U.S. economy but investing in that process led to a great deal of overbuilding.

• Between 1919 and 1929, the U.S. more than tripled its electric generating capacity. That’s an average increase of about 12% per year, far above the growth rate of electricity consumption of 7–8% per year.

• By the late 1920s, the U.S. electric power industry had 30% to 50% more generating capacity than demand required, and as a result, roughly one-third of capacity sat idle most of the time.

• By 1932, dozens of utility holding companies collapsed under heavy debt burdens, the Insull collapse of 1932 being the most infamous. That led to major government intervention in the 1930s.

Between 1995 and 2001, U.S. (and global) telecom and internet firms invested approximated 1.5 trillion of dollars in U.S. physical infrastructure, including fiber optic networks, switches, routers, and data centers.

• Between 1998–2001, network capacity grew by about 1000%, but internet traffic only grew by about 400%. By 2001, only about 5 to 10% of fiber optic capacity (“lit fiber”) was actually activated, so more than 90% remained “dark fiber” — installed but unused — for much of the 2000s.

• As a result, more than half of all competitive local exchange carriers (CLECs) and most long-haul fiber startups went bankrupt by 2002, and global telecom equipment sales fell by about 70% from their 2000 peak to 2002. Major companies such as WorldCom, Global Crossing, 360Networks, Level3 and Qwest lost billions of dollars.

• The good news is that the unused fiber enabled cheap broadband expansion in the 2010s, cloud computing and streaming as well as the 5G and global internet backbones. Initial investors, however, lost their shirts.

Why do we care? Given the huge expense to build, train and operate GenAI platforms that produce minimal revenue, and the subsequent issues with reliability, data degradation and so far uncertain value, investment in AI is highly speculative.

Furthermore, most investors in the current stock market fail to realize that they’re already heavily invested in the GenAI mania. Only ten stocks – including those making the largest AI investments such as Amazon, Apple, Google, Microsoft, Meta and Nvidia – now account for 38% of the S&P 500 and more than 70% of the index’s returns. We can’t predict when or how these investments will disappoint the imbedded expectations, but this is a time of historically high risk so a prudent investor should act with caution.

AI is a high cost, high risk technology with great promise, but without any clear path to profitability. There will be some victors but far more casualties.

Read more

The stock market isn’t so different. When enthusiasm runs high, investors convince themselves it’s different this time. Valuation, they say, is just an outdated relic, like rotary phones or balanced budgets. However, history has proved this wrong – every time. Long-term investing isn’t about riding waves of euphoria; it’s about returns, and returns are anchored to what you paid in the first place. Markets can stay irrational longer than expected, but they never escape the anchor of fundamentals.

With stock valuation at all-time high multiples, it’s currently popular for Wall Street strategists to point out that valuation is a poor predictor for stock prices in the next year or so. That’s true, however, the animal spirits that float the market to new heights can’t persist indefinitely because ultimately companies are valued for their ability to earn profits over the long haul.

• Markets go through cycles. When investors buy a stock, they’re buying a share of a company, and ultimately the value of a company depends on its underlying cash flow. Unless you’re a day trader, repeated research has shown that market valuation is the best – actually the only – influence on long-term returns.

• Let’s take a step back and see if it makes sense, in a general way, if what you pay for something has an impact on its return. Keep in mind that real investment risk is the likelihood that you’ll lose money. If you underpay for an asset (i.e., pick up a bargain) you’re going to get a better return than if you overpaid for the same asset (i.e., got fleeced). That’s just math.

• This simple concept is completely overlooked by Wall Street, who urges you to buy stocks under any circumstance. Take the example of a house that’s worth $100,000. Your investment risk depends on the future price, but if you paid only $50,000 you’d have a margin of safety. On the other hand, if you paid $200,000 your future return depends on finding someone to pay an even higher price. (By the way, here’s a piece of free advice: If at all possible, sell your house to an architect. They have such wonderful imaginations that they can envision any number of fantastic possibilities that could lead to an even higher price.)

• This inclination to overpay is also called the greater fool theory, which speaks for itself. In a market with roaring optimism, it’s possible in the short term to find that buyer. But, in the long term, investors return to reality. In the case of the stock market, periods of optimism are always followed by periods of skepticism

• With respect to stocks, valuation matters because if you pay too high multiple to the company’s earnings or cash flow, it will take decades for that company to return your capital, even assuming all goes well. If you pay more than 35x or 40x earnings – the average multiple for the seven stocks that currently dominate the S&P 500* – it will take decades to break even. And that assumes that earnings continue to grow. No one ever predicts a future down cycle for earnings, which happens regularly and goes through long periods of weak performance.

• For perspective, the following chart – produced by Nobel Prize winner Robert Shiller – uses a ten-year moving average to smooth out highly volatile earnings for the S&P 500 and shows the degree of the current overvaluation.

This is the same data that measures the market deviation from its long-term average PE.

• With the S&P 500 currently trading at 31x trailing 12-month earnings, valuation hovers at record levels (the market is also at record levels for just about every other measure, such as price-to-sales and dividend payouts). To figure out if that’s a reasonable level, we need to compare stock prices with the discounted value of their underlying future earnings.

• Assumption #1: To make that calculation, we make the following assumptions. Wall Street analysts expect earnings to grow at 11% per year for the next several years. The Street’s estimates are almost always overly optimistic, and the likelihood that this growth will persist for many years into the future is fanciful given an economy growing at 3% a year (and even that assumes no recessions). However, let’s go with a rosy prediction of 11% earnings growth into the hereafter.

• Assumption #2: To discount this earnings growth, we’ll use the risk-free rate of Treasury bills, which is currently around 4%. Again, let’s be generous and assume that the Fed will make more cuts to lower the rate to 3.5%.

• The result: Applying an 11% growth rate for earnings and a 3.5% risk free rate produces a fair multiple for stocks of approximately 15x earnings, which is a neutral, real value for the market. This also happens to be the average multiple for the S&P 500 for the last eight decades after World War II. Unfortunately, it’s half the level of current prices.

• Wall Street strategists dismiss valuation as a poor predictor of short-term performance – and for the last several years, Wall Street has been right. However, as the above charts show, the market always returns to its long-term value (called “reversion to the mean”). Valuation is the tether between price and reality—and when it stretches too far, the snapback is rarely gentle.

• How did valuation get so out of touch with reality? The primary stimulus for these levels came from years of zero interest rates and, most recently, the government’s response to COVID. In 2020 to 2021, the Fed printed a huge amount of new money, and at the same time the Federal government ran record deficits. In a short period of less than two years, these two actions nearly doubled the US monetary base, far outpacing economic growth of around 3%.

• These monetary and fiscal propellants can’t continue indefinitely. The stimulus provided by the Fed has ended, and quantitative tightening has reduced reserves in the banking system by 26% to the same level as 2019. The Fed also needs to control an inflation rate that persistently remains above targets, which limits its ability to use interest rate cuts to further fuel the market. In addition, government efforts to reduce the federal deficit will dampen greater fiscal stimulus.

• The hangover from this money surge has left a euphoric market that is complacent about long-term risk. It produced a generation of investors conditioned to believe that current market levels are normal (see our post, “Recency Can Cloud Long-Term Investment Decisions” at https://marketrisks.substack.com/p/recency-can-cloud-long-term-investment). It also led to new highs in the market this year due to the huge surge of new retail investors, who are famous for buying heavily at market peaks.

• Part of the overvaluation is also due to quantitative models that drive most investment activity today and fail to measure risk accurately. These models use price volatility – the degree that a stock price move up and down – to measure risk because it’s the only way to apply math to something a computer can model. Unfortunately, volatility turns out to be a poor measure of real risk because future returns depend on all sorts of other intangible and unanticipated factors that can’t be quantified, and thus the models miss.

• At the same time, successful long-term investors who rely on fundamental analysis have been selling. Warren Buffett’s Berkshire Hathaway has accumulated record levels of cash (see our post, “Buffett Sells Stocks and Raises Cash” at https://marketrisks.substack.com/p/buffett-sells-stocks-and-raises-cash).

Our notes on thin reed indicators and other red flags have mentioned many examples of recent euphoric investment behavior that typify market peaks.

• The highest level of institutional investors’ exposure to stocks since the market peak in 2007.

• The surge in plans for corporate buybacks, which occur at times of high complacency.

• The proliferation of ETFs, which now outnumber the number of publicly traded stocks and defy fundamental analysis.

• The record level of exposure of retail investors to the stock market.

• The current dominance of index funds that automatically buy more of the few concentrated stocks that produce most of the market’s return. This process hides underlying risk by ignoring valuation.

• The recent behavior of IPOs, that after careful pricing by investment bankers rise as much as 30% to 100% in the first day of trading.

• The record low spread between the yields on risky junk bonds and safe corporate debt demonstrating investor complacency.

• The rise in long-term interest rates at the times when the Fed cuts short-term rates.

• And, many other signs, which we will continue to point out in our notes section.

When will valuation return to reality? We don’t know, but stock market history shows that periods of exuberance are always followed by retrenchment to long-term averages.

• It’s impossible to predict what will lead to a reversal in market sentiment. Since markets are complex systems (see our report “Why Markets are Unpredictable” at https://marketrisks.substack.com/p/why-markets-are-unpredictable), they can turn down merely due to exhaustion rather than any major market catalyst. In the study of avalanches, this occurrence is called “self-organized criticality.” Imagine adding a snowflake to pile of snow at the top of a mountain. As the snow stacks up it reaches a point of criticality when an avalanche occurs. Unfortunately, it’s impossible to tell when this will occur – we can’t tell which snowflake will finally topple the pile – we only know that the pile has reached extreme levels. As an investment example, economists still debate the precipitating causes of the crash of 1929, nearly 100 years after the event.

• Given the extreme concentration in the market, it may not take much to turn investor perceptions, especially if a few key stocks fail to meet estimates. To justify stratospheric multiples, Wall Street has issued projections for the few stocks that dominate the market – the Magnificent Seven stocks* – that assume huge earnings growth. When a few of these companies miss the consensus for revenue or earnings or offer lower guidance moving forward, that could cause a sharp reevaluation of future expectations.

* The Magnificent Seven stocks are Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla

Why do we care?

• For long-term investors, valuation matters. Eventually, market valuations return to reality, so the current record valuation level creates a high degree of risk for stocks. If a market decline for months or possibly years would cause you sleepless nights, or worse, you’ll want to adopt a cautious approach and take steps to preserve capital. A cautious approach also keeps dry powder to buy stocks when they reach significantly lower levels.

But what happens when too many investment boats set out on the same course, pushed forward not by skilled navigation but by the sheer weight of money flowing in the same direction? The waters begin to shift. Prices rise not because companies are thriving, but because liquidity forces indexes to buy. Diversification starts to blur as the same stocks dominate portfolio after portfolio. And when the tide inevitably goes

Read more

Unless you’re an investment professional, avoid the new craze for private assets.

Wall Street is marketing broader access to private equity with several benefits.

• Rather than invest in individual equities, private equity (PE) firms buy an entire company privately. Ideally, they then restructure the company, cut costs and add debt to later take the company public or flip the investment to another PE firm at a profit.

• Private assets are perceived as less volatile than publicly traded stocks because they’re not marked-to-market daily, rather their asset prices are set by the firms that own them. Because these assets don’t sink with every decline in the market, they have an obvious appeal to long-term investors such as pension funds and university endowments.

• The new public vehicles, along with the Trump administration’s green light for placement in retirement accounts, will allow smaller investors to participate in what was previously available to only large institutional investors. The problem is, these opportunities are arriving just as the smart money is bailing. With the pressure to produce realized gains, the waiting list to exit private positions has now grown to eight years (Wall Street Journal).

Investments in private assets produced stellar returns in the last two decades.

• Mesmerized by the exceptional returns of the Yale University Endowment and David Swenson’s Pioneering Portfolio Management published in 2000, trustees for endowments and public pension funds soon followed suit by investing substantial amounts of capital into illiquid alternative markets, including venture capital, buyouts, real estate, hedge funds, and private credit. By the 2020s, almost every institutional and individual investor believed that private markets offered a foolproof way to enhance returns and reduce portfolio risk.

• This led to a flood of capital moving into the sector during the last decade. With low financing costs and rising demand, prices for private assets skyrocketed. Deal volume exploded to $3.1 trillion in 2021 and prices more than doubled with buyout multiples soaring from 6.5x EBITDA to 11.9x EBITDA in ten years (Zero Hedge, McKinsey).

• After years of rising investment in private assets, the popularity of private investments has waned, and most institutions now want to sell their positions to realize gains. For example, universities need to raise cash to deal with the Trump administration’s new charges. As a result, private markets have entered what may be the most precarious phase of a decades-long speculative cycle, defined by high valuations, opaque pricing, low liquidity and a high exit backlog.

Beware of the new ETFs that will buy these private assets. They contain many obvious and not so obvious risks for unsophisticated retail investors.

Risk #1: The current roster of private equity investments occurred during a period of huge demand from PE firms. As a result, these investments suffer from higher buyout prices and lower asset quality.

• As money poured into the PE market in the last decade, deal volume exploded leading to peak prices for deals around 2020 and 2021 and inevitably leading to lower quality investments. In June, State Street Global Markets found that the S&P 500 outperformed its private market index for all time horizons. This year PE distributions are running 50% below normal (Morgan Stanley).

• Given the high initial prices paid for private companies, the future returns for new ETF portfolios that buy these assets will likely underperform as well. PE firms still have record amounts of unallocated capital, and pressure to deploy those funds means that ETFs looking to buy new private deals could face more competition and even higher deal pricing leading to even less attractive future returns.

• In addition, investments in private assets contain normal operating risks. More than 65% of private equity investments either fail or fail to return their initial investment (Zero Hedge). PE investments made in the peak fundraising years, such as 2007 and 2021 produced the worst returns, with distributions for the 2021 vintage running 80% below normal.

• Rising interest rates also lower future returns. PE returns depend not only on valuation but on borrowing costs to fund these deals. As the low-interest loans that financed private deals in 2021 and 2022 now mature, the major PE firms are under even greater pressure to unload underperforming assets before incurring much higher interest rates.

• As a result, the waiting list to exit private equity positions has surged to eight years. According to Pitchbook, a backlog of approximately 30,000 companies now sits on the balance sheets of private equity firms. With the surge in the current exit backlog, it could take longer to realize any profits, which could mean delayed distributions and sales at significant discounts.

• With their normal exit channels stuffed, the large PE firms are now forming ETFs to sell these assets to less skeptical and unsophisticated retail investors. The firms are touting these new investments as democratizing access to private markets, but essentially, the smart money is selling.

Risk #2: Because they don’t trade publicly, the underlying private equity Investments are Illiquid.

• To perform their restructuring magic on their target companies, PE firms need to lock up capital for as long as seven to ten years. A retail investor is out of luck if he changes his mind or needs cash for unexpected expenses.

• Most funds have strict limits on redemptions, so investors may find themselves stuck in poor performing funds for a long time.

Risk #3: Private equity funds make a few large investments, so they face concentration risk.

• PE funds often focus on a narrow set of companies or industries. As a result, A single failed investment can have an outsized impact compared to diversified public ETF.

• In addition, the steps that PE firms take to improve corporate performance – such as turnaround strategies, cost cutting, and growth initiatives – may not realize the anticipated benefits. Current PE investments have likely implemented many of these efforts already, leaving little room for further operating improvement.

Risk #4: Private equity investments require continuing capital infusions.

• When they make an investment, PE funds don’t commit all the necessary capital at once. In fact, they invest capital over time as required by the target companies. It’s unclear how a publicly traded ETF will make cash calls, but investors may have to supply cash when required, even during downturns.

Risk #5: PE firms value their investments using opaque, and sometime dodgy, methods.

• One of the touted benefits of private equity is more stable valuation compared to publicly traded assets, but these firms operate with limited regulatory oversight. Rather than price assets based on prices determined in the market, PE firms value their assets infrequently using opaque valuation models. During periods of falling stock prices, this can delay recognizing losses which can hide the real the underlying risk.

• Many PE funds have reported large gains within a day of purchasing assets. One pricing technique is known as NAV squeezing, which creates one-time bookkeeping gains even though the value of the underlying asset is unchanged. PE firms can buy companies in secondary deals at a discount to their net asset value (NAV). They can then mark up the value of their investment to the higher NAV, which has led to immediate gains of more than 1000% in a day giving the illusion of exceptional returns for what is just a bookkeeping gain (Wall Street Journal). The Journal further reports that secondary deals in 2024 changed hands at an average discount of 11% creating ample room to mark up assets.

• This process has another more subtle risk. To show further gains, PE funds need to continually make more purchases to take advantage of NAV markups. This could increase the pressure to make less discriminate acquisitions.

Risk #6: PE investments are highly leveraged, which will accelerate potential losses.

• In the zero-interest rate world after the 2008 financial crisis, ultra-low interest rates allowed PE firms to use greater leverage and less equity and pay up for acquisitions. According to JP Morgan, PE firms have financed more than half their acquisition costs with debt.

• While leverage can increase investment returns, it also works in the other direction, accelerating losses when prices decline.

• Many of the firms that bought companies in the past several years using cheap debt now face much higher interest rates as the old low-interest loans roll over.

Risk #7: PE firms charge enormous fees and often charge additional management compensation on top.

• To work their magic as active managers, PE firms charge hedge fund type fees. A typical PE fund charges a 2.0% annual management fee as well as 20% of the gains, not to mention consulting fees they charge to companies they oversee. This compares with publicly traded ETFs that charge an annual fee of anywhere from 0.03% to 0.10%, a huge difference over the seven to ten year life of a private investment.

Why do we care?

To sum up, here’s a quote from Ludovic Phalippou, a professor of economics at the University of Oxford, “Giving individual investors exposure to a high-fee, opaque, conflict-ridden asset class — marketed using gameable metrics and de facto fictional track records — is not governance for the people. You can't call it democratization if the system is designed to benefit the insiders while leaving the public exposed to risks that they may not understand.”

Avoid!

Dan, what are the three most important investment issues that keep you up at night?

To answer that, I need to explain that our clients generally fall into two different buckets, both with similar but distinct investment demands. The first bucket is clients that are near or in retirement. This group wants to maintain their lifestyle for the next 25 years or so but generally owns assets of less than $5 million. For them, it’s important that we continue to grow their assets but balance that with capital preservation.

The second group is typically high-earning clients in their 40s and 50s, who have a variety of spending requirements but also need to save and grow their assets in planning for future expenses and eventual retirement. This group can take a more patient, longer-term investment approach.

First concern: Stock Market Levels

We track stock market valuations generally and wonder if the market can sustain and even rise from such lofty levels. What happens to current investment decisions if the market produces substandard returns for the next 10 to 20 years? With greater than 20% annual appreciation for the last several years, our clients have come to expect high returns, so lower returns could lead to frustration.

Second concern: Interest Rates

This is of particular concern to our clients in their 40s, who either own a home or are thinking about buying a home. A house is often one of our clients’ largest assets. For those who are current homeowners, many are locked into low mortgage rates which they’d have to relinquish to trade up to another house, which raises the concern of absorbing a much higher expense. For those that are looking to buy for the first time, should they wait for lower mortgage rates or will those rates remain high? Will rates ever return to the low levels of the 2010s?

Third concern: The Impact of Artificial Intelligence

Much of the rise in the stock market this year has been due to the enthusiasm for AI stocks. How much of this is due to real long-term benefits or temporary investor excitement? How much will AI impact the economy in the next decade, and how will it affect jobs, job creation and the economy?

Dan, thank you for taking the time to inform Market Risks readers. Your thoughts clearly bring into focus concerns that are on the minds of most advisors.

Why do we care?

• The current high level of uncertainty and rapid change in financial markets makes investment decisions even more difficult. No one can predict the outcome for these issues, however, Market Risks can explain the market context and provide a long-term perspective for thinking about their impact. We recently provided background on two of these concerns and plan to keep an eye on all three moving forward.

Yes! But the reason may not be what you think. Many economists and investment strategists have worried about the geometrically expanding federal debt for decades (see the following chart), but until now it’s had a minimal impact on the US economy and financial markets. The usual concerns are the level of federal debt compared to GDP, which is a measure of the country’s ability to service the debt, and the impact of higher interest payments on the federal budget. Both have risen to levels that are concerning – the debt to GDP is at a level that matches the prior peak during World War II, and interest payments now amount to 13% of the federal budget, which is the second highest line item behind Social Security and higher than outlays for Medicare, Medicaid and defense spending among other critical functions.

Read more