The End of Easy Money: Why Index Investing Stops Working

In the second and third weeks of June 2026, four events broke in the US and Korea with no connection to one another, and all of them happened to point in the same direction. Kevin Warsh's first FOMC as the new Fed chair, Bloomberg flagging that excess liquidity had turned negative, Goldman Sachs publishing its "Post-Modern Cycle" report, and Bank of Korea Governor Shin Hyun-song signaling rate hikes.

Taken separately, each is just a headline. Stack the four on top of one another, though, and a larger picture appears. The common sense that ruled the last 40 years, and especially the 15 years since 2008, that rates stay low, money stays cheap, and risk assets all rise in the end, is breaking down at a structural level.

This piece is a macro structural analysis of what that shift is and what it means for stocks, real estate, companies, and investors. Think of it as a map of how the board is changing, not a recommendation to buy any particular name. I have tried to separate verified facts from interpretation and debate as I go.

The One-Line Thesis: Money Isn't Drying Up, It's Getting More Expensive

Money isn't drying up. It is getting more expensive and more selective.

The absolute quantity of liquidity keeps growing. Governments keep spending, central banks ultimately backstop, and private debt expands. But the "excess liquidity," the surplus that the real economy absorbs and then leaves over to flow into risk assets, is shrinking.

As a result money no longer goes everywhere. It flows selectively to where the real money is made. Markets, real estate, and companies all move from an era where everything rose together to a K-shaped era where the winners and losers split sharply. Goldman Sachs calls this regime the Post-Modern Cycle.

Four Signals Pointing the Same Way

Signal 1. Kevin Warsh's Fed: "You're On Your Own"

Kevin Warsh succeeded Jerome Powell as Fed chair on May 22, 2026. The Senate confirmation vote was 54 to 45, the most divided in the Fed's history. June 17 was his first FOMC.

The decision itself was a hold (3.50 to 3.75 percent, the fourth in a row). Yet the market read it as a hawkish shock.

The real message was not the rate number. Warsh refused to submit his own dot (forecast), explicitly scrapped forward guidance ("forward guidance isn't the business we should be in"), and stripped the easing bias from the statement. He also launched five task forces meant to overhaul how the Fed operates (communication, balance sheet, data, productivity and labor, inflation measurement).

The point is not that "Warsh is a hawk." He has barely shown whether he is a hawk or a dove. The point is a declaration that he will end the era of the Fed Put and forward guidance with his own hands. The 20-year belief that the Fed will ultimately catch the market when stocks wobble, and the habit of the market front-running the Fed's pre-announced direction, are both being withdrawn. Warsh's logic runs like this: because the market front-ran the Fed by watching its mouth, the last leg of 2023 to 2024 inflation never got caught. So the message is "stop watching the Fed, look at the data yourself, you're on your own."

The reaction from bond king Jeffrey Gundlach (DoubleLine) was telling. He said "there will not be as much easy-money policy as everyone thought," that "a new sheriff is in town and the case for holding long-dated Treasuries is stronger," and judged that the Fed had entered an unpredictable "new era."

Two twists are worth noting here. First, the market read it as hawkish because of the other members' dots, not Warsh's. Warsh did not plot his own dot, and what was actually hawkish was the other members' projections (a majority leaning toward a hike this year, with six members penciling in two). Powell would have soothed this in the press conference, but Warsh did not, stressing principles and task forces instead. So the hawkish projections landed on the market untempered, while Warsh's own policy hand stays hidden. Second, there is a dovish opening inside the hawkish shell. Warsh said he cares more about "the integer to the left of the 2.0 percent target," which Gundlach read as meaning "if 2.99 percent counts as price stability, you can cut." He also hinted, in Greenspan fashion, that "the AI productivity revolution works as a disinflationary force, so I won't rush to hike and throw cold water on it."

Signal 2. Excess Liquidity Turns Negative for the First Time Since 2021

On June 18, Bloomberg columnist Simon White noted that global excess liquidity had turned negative for the first time since 2021. The column was titled, fittingly, "The Squeeze on Liquidity Is Just Getting Started."

Excess liquidity is real money growth minus economic growth. It means money is now growing slower than the speed at which inflation and real growth absorb it. In White's words, both banks and central banks create liquidity, but asset prices only rise when that liquidity grows faster than the real economy can absorb, and that surplus is now disappearing.

He sees this gauge leading stocks and commodities by three to six months. After it turns negative, stock returns usually worsen and demand shifts toward safe assets after that lag. One caveat: excess liquidity is a model-based indicator that different analysts compute differently (Simon Ward and others argue it is "closer to measuring monetary acceleration than the growth rate"). It is best read as a strong warning light, not absolute truth.

Signal 3. The Whole Rate Regime Has Stepped Up

The point is not the numbers themselves but the fact that they "do not come back." The market printed 0.3 percent, hit 5 percent, and eased to 4.5 percent, but rather than returning to the pre-pandemic zero, it is locking in at a 4 percent handle. The decisive part is the real yield moving from 0 to +2 percent. That is what it means for the true price of money to get expensive.

Signal 4. The Bank of Korea Pivots to Hikes Too

On April 21, 2026, Shin Hyun-song, former head of the BIS Monetary and Economic Department, was sworn in as the 28th Governor of the Bank of Korea (succeeding Rhee Chang-yong at the end of his four-year term). May CPI came in at +3.1 percent, back above 3 percent for the first time in 26 months, with core at +2.5 percent. The BOK projects roughly 3.0 percent headline and 2.6 percent core for the second half.

At the bank's 76th anniversary ceremony on June 12, Governor Shin effectively formalized a hike, saying there was "a need to raise rates without delay" and "a need to raise the base rate at an appropriate time." The market is watching the July 16 rate meeting (though a "hike next month" is an observation, not a certainty). Some forecasts see a cumulative +100bp to a terminal 3.50 percent. On a big step (0.5 percentage point), he drew a line on June 17, calling it "an exceptional measure for when bond yields and the won are excessively high." The stance is gradual.

Why Korea's implications are especially heavy: if the US story is "high rates locked in," Korea is, on top of that, reversing direction from cutting back to hiking. The cutting cycle is over. For Korean equities and real estate too, this signals that easy money is winding down.

This is not a coincidence stacking up. It is one large regime change.

Why Warsh Is Doing This: The Project 2025 Blueprint

Warsh's debut moves are not improvisation. There is a blueprint. They overlap almost exactly with the Federal Reserve chapter of the Heritage Foundation's Project 2025 (authored by Paul Winfree), and Warsh hired that same Winfree as his first adviser right after taking office on June 2 (along with Daniel Heil of the Hoover Institution).

The blueprint has three pillars.

1. Abolish the dual mandate. Drop "maximum employment" and focus only on price stability. This fits the debut statement, which erased the easing bias and left only "price stability." That said, Warsh said at his swearing-in that he "respects both sides of the dual mandate," a surface contradiction. The direction leans entirely toward prices, but a formal abolition has not happened yet.
2. From discretion to rules. Instead of signaling the market case by case while raising and cutting, move mechanically by rules (a Taylor rule, money-supply targets, and so on). This matches the press-conference line, "react to data, not to the Fed."
3. Shrink the lender of last resort and asset purchases (QE). Treat buying bonds to inject money as market manipulation and moral hazard, and hold back on rescues even if banks break. Classic sound-money thinking.

Pushed to the extreme it becomes this: take Warsh at face value and you get "don't watch the Fed or what the Fed says. Just watch the 2-year Treasury yield." The market reacts to data, and the Fed merely follows that market.

But there are two big holes (the fuller counterargument comes later in "The Counterarguments and Limits"). One, can you trust the data? When shutdowns leave figures missing or mixed and confidence in statistical agencies wobbles, removing the Fed that would smooth that volatility only makes the bond market more volatile. Two, what happens when a crisis comes? It may work in calm times, but in a recession or pandemic the Fed cannot "just watch the market and sit still." It ends up having to step in again. The moment ideals collide with reality arrives during a crisis.

Goldman's "Post-Modern Cycle"

A report by Goldman Sachs chief global equity strategist Peter Oppenheimer, published in mid-June and titled "The Post-Modern Cycle: Navigating the Capex Boom," ties these signals into a single frame (it extends his 2024 book Any Happy Returns). He divides market history into three eras.

The reason global tech traded at nearly 3x the PER of non-tech from 2009 to 2021 was not just better fundamentals. Zero rates held up the multiple. That prop is now being pulled away.

Oppenheimer names seven structural drivers.

1. Rising cost of capital. The pandemic supply shock produced the century's first inflation shock and lifted real rates.
2. Surging government debt. Public debt to GDP rose from 55 to 124 percent in the US, 37 to 95 percent in the UK, and 22 to 102 percent in China. The CBO projects debt to GDP at 120 percent and net interest above $1 trillion a year by 2036.
3. Tariffs and deglobalization. The US effective tariff rate is the highest since the 1930s. Trade barriers push up costs.
4. A reshaping geopolitical order. Cracks in the rules-based international order, with policy uncertainty at multi-year highs.
5. Energy and commodity security. The pursuit of supply-chain and energy independence raises related capex.
6. A rebound in defense spending. Wars in Ukraine and the Middle East, with major military expansion in Germany and Japan.
7. The AI capex revolution. The biggest driver. S&P 500 capex jumped +38 percent year over year in Q1 2026, while buybacks grew just +1 percent over the same period. Capital has turned away from wages and buybacks toward data centers, chip fabs, power grids, and fiber.

All seven converge on one conclusion. Because demand for capital rises structurally (governments borrow and companies invest), the price of money (rates) stays high, and the room for multiples to inflate further shrinks.

The Core Mechanism: Excess Liquidity and the Cost of Capital

To really understand the shift, watch two gears.

Absolute Liquidity vs Excess Liquidity

Money is not vanishing. Governments loosen fiscal policy further and central banks ultimately backstop, so the absolute amount grows. But when inflation rises and real growth (AI infrastructure investment and the like) speeds up, the real economy soaks all of it up. The surplus left for risk assets shrinks.

Picture a paycheck (money supply) coming in while rent, food, and interest (inflation, cost of capital, real investment) eat all of it, leaving nothing spare to buy stocks. So the era where money made everyone happy ends, and money flows only to where the real money is made. A K-shaped market.

One common misconception needs correcting here. The intuition that "less money goes to risk assets" is only half right. The reduced pool does not leave all risk assets. It piles even more into those that prove earnings and cash flow (semis, AI, infrastructure), while pure speculation with no cash flow (speculative coins, loss-making growth stocks) loses its fuel first. The K splits once more inside risk assets.

From the Age of Multiples to the Age of Earnings

A "multiple" is the umbrella term for the ratio at which a stock trades to some measure of the business. The most common, PER, is market cap divided by net income ("times earnings"), while early high-growth names that are loss-making or barely profitable use PS, market cap divided by revenue ("times sales"). There are others like EV/EBITDA and PBR.

When rates are low, the market grants a generous multiple even on the same earnings. A company earning 10 billion won a year might get 30x (300 billion) in a low-rate world and only 15x (150 billion) when rates rise. Earnings are unchanged, yet the price is halved.

Why? A stock's value is future earnings translated into present value, and you discount by rates when you translate. One million won received in ten years is worth about 910,000 today at 1 percent, but shrinks to about 610,000 at 5 percent. Higher rates discount future earnings more, so multiples come down. Growth stocks, whose earnings sit far in the future, take the hardest hit. The 2022 Big Tech sell-off ran on exactly this principle.

Now the engine pushing prices up is earnings growth and cash flow, not multiple expansion. Companies that prove it survive, and those that cannot watch their multiples get cut without mercy. Not "a good company," but "a company making money and generating cash right now" earns the premium.

Goldman's conclusion compresses into one line. Alpha beats Beta. This is not an era where simply owning the index rises. The value of being able to select names and industries is greater than ever. At the same time, index-level returns are lower and volatility higher, a "fatter and flatter" market.

It Doesn't Mean the Index Won't Rise

That conclusion actually packs two different phenomena into one sentence. Separating them sharpens the picture.

"The index won't rise" is wrong. The main engine that lifted the index from 2009 to 2021 was multiple expansion (low rates inflating PER). That engine simply turns off, and the index now rises only as much as earnings grow, with more chop along the way. A market with a wide range but only gentle trend progress.

"Liquidity supply gets cut off" is also wrong. The absolute amount grows. What disappears is the "free lift" that raised the whole index, as the surplus left over after the real economy absorbs its share shrinks. Put it together: the index loses its free lift (time-series, flatter and choppier), and inside it money moves with earnings (cross-section, alpha and K-shaped). The two are two sides of the same coin rooted in the same cause (rising cost of capital, falling excess liquidity).

How the Stock Market Changes: Seven Axes

(1) From Beta to Alpha

The era where simply owning the S&P 500 let multiple expansion lift it for you is fading. As dispersion across sectors and stocks widens, what you buy matters more than whether you buy. The weight shifts from all-passive toward active selection.

One misconception to flag. The money here does not go to traditional blue chips (large, stable, high-dividend). The criteria are different: does it prove earnings and cash flow in numbers, does it own a bottleneck (a moat), does it sit in a structural tailwind. So an old large-cap software "blue chip" can de-rate while an expensive 26x-PER semiconductor name holds its premium. The yardstick is not "safe and blue chip" but "structural tailwind plus proven earnings."

(2) Leadership Rotation: From Software to Hard Infrastructure

The leadership of asset-light software, which led the market for more than a decade after the financial crisis, is ending, and the center of gravity shifts to those who make physical things.

The new leaders are semiconductors and data-center hardware, power and energy (nuclear, grids, generation equipment), defense, industrials, materials (copper, metal smelting), and infrastructure. Companies with the capacity to smelt metal, build chips, put up buildings, and make weapons earn a premium again. What actually got expensive in the AI era is not "AI software" but the electricity, copper, steel plate, cooling, and land needed to run it. For reference, Goldman's example capex-beneficiary weights run roughly industrials 30 percent, materials 20 percent, tech 15 percent, utilities 10 percent.

(3) The AI Capex Supercycle

The scale comes into focus through numbers.

Add sovereign AI, cloud builders (like the Neocloud trio), and enterprise investment on top of the hyperscalers and the scale grows further. Token consumption is projected to rise 24x by 2030.

Even so, the overheating warning comes from Goldman itself. AI infrastructure names trade at a 26x PER, the highest since ChatGPT launched, yet only 1 to 2 percent of companies quantified AI's earnings impact on Q1 calls. The spending is certain, the earning is not yet proven. How the market reacted to this "ROI fear" is covered separately in the AI fear trade and SaaSpocalypse.

Where Are We in the AI Cycle? Three Stages

Unfold the leadership rotation in time order and you see the same three stages every great infrastructure revolution has passed through. First, infrastructure is scarce and the picks and shovels (this time GPUs, HBM, power, copper, cooling) command fat margins. Then supply overtakes demand, prices collapse, and the excess gets cleared (the bust). Finally, applications (a handful of moated models and apps) absorb the value on top of the now-cheap infrastructure.

The past five infrastructure cycles all followed the same script (boom to bust to applications).

Diagnosing honestly where we are, the transition signals are switching on one by one along the historical path.

Putting it together, we are somewhere between late Stage 1 (picks) and early Stage 2. Capex is still strong, but growth is showing signs of slowing, prices have already collapsed (commoditization), and Stage 3 applications like Anthropic and Cursor have already sprouted. The stages do not split cleanly with a knife, they overlap. The lesson from history is clear. The picks suppliers see margins cut by commoditization, and the true long-run winners are likely to be the handful of applications and models that turn cheap intelligence into money on an exclusive basis. And the payoff arrives later than you expect (electricity took fully 40 years).

That said, some of the figures above (2027 capex, the timing of oversupply) mix in forecasts and are not settled. "Inference cost down 280x" is cost per fixed performance, not the list price of the latest frontier model. Treat the signals as observation gauges, not prophecy.

(4) Companies Become "National Strategic Assets"

In the US-China contest for supremacy, AI and semiconductors are treated not as mere technology but as national strategic assets. As the trend of governments backing companies, "even if you don't make money now, we will subsidize you" (the Intel case), spreads, "how important is this company to national security and strategy" becomes a new variable in valuing it. Subsidies, tariff protection, and government orders matter as much as the fundamentals.

(5) The Return of Valuation Discipline

Not all software dies. Some shines brighter in the AI era. But for software that worries AI coding might replace infinitely copyable software, and for software that has not yet proven its business model in earnings, the recommendation is to reduce weight. It is not "going out of style" so much as the bar for proof rising.

Conversely, some sectors like semis and power carry valuations that look "absurd" by old-cycle yardsticks, yet a different reading is possible once you account for the paradigm shift. Looking only through the old lens can itself be irrational. But this is a double-edged sword. When the supercycle breaks, that premium breaks with it.

(6) Gold and Bitcoin: A Split in Hedging

As of April 2026, gold was around $4,800/oz, up +46 percent year over year, while bitcoin was around $74,000, actually down from $93,000 at the start of the year. Same "hedge" assets, opposite directions.

Take coins first. When excess liquidity drains, the assets with zero cash flow and the highest beta lose their fuel first. Coins, especially alts and memes, are the front line. They have no earnings to prove and eat the full +2.2 percent real-rate opportunity cost head-on. This is where the intuition that "less money goes to risk assets" fits most cleanly.

So why did gold rise? The textbook says "rates up, gold down," because high rates are a headwind for the yieldless metal. But this round of high rates is not about healthy growth, it is about government debt, inflation, and geopolitics, and those very causes create gold demand: a hedge against debasement, safe-haven demand amid war and fragmentation, and central banks buying gold to de-dollarize. The force that lifted rates lifted gold at the same time. This time, that tailwind beat the high-rate headwind.

Why gold and bitcoin split has a hypothesis (a market debate, not settled): bitcoin is a hedge against monetary inflation (over-issuance), gold a hedge against price and scarcity (supply, energy, war). This time gold won. But there is no agreed conclusion, and if real rates climb more steeply the yieldless metal could correct too.

(7) The Paradox of Crowding

Money does pile into the top edge of the K (the winners). But that crowding itself creates a new risk. When funds concentrate in a few names (AI, semis), the index's dependence on a few stocks goes to an extreme. KOSPI leans on Samsung and Hynix, the US on a few AI names. When one stock wobbles, the whole index wobbles. AI infrastructure trades at 26x while only 1 to 2 percent of companies quantified earnings impact, so a crowd stands together in a narrow, expensive spot. The place everyone crowded into because it looks safe can swing the hardest. Momentum and crowding are two sides of one coin, so you ride the crowding but diversify rather than putting everything in one basket.

Bonds, FX, and Oil

Bonds cut both ways. Gundlach's line that "the case for holding long Treasuries is stronger" means that the stronger Warsh's resolve to crush inflation, the more the inflation premium in long rates can settle. But surging government debt (more supply) pulls the term premium the other way. The two forces tug at each other around the 4 percent handle. Watch for this: when the stock-bond correlation flips from negative to positive, the classic 60/40 diversification breaks, and bonds stop cushioning equity drawdowns more often. On the dollar and emerging markets, US high rates and a strong dollar weigh on EM and high-FX countries. Even so, Goldman recommends diversifying outside the US, noting that gold, emerging markets, and Japanese stocks led 2025 to 2026 returns. Step away from an all-US-tech tilt.

Oil is turning back toward structural oversupply. As the Israel-Iran war moves toward a ceasefire, Brent fell to the mid-$70s. Three supply factors overlap: crude released as Iran sanctions lift, OPEC+ increases actually reaching the market as Hormuz normalizes, and the UAE leaving OPEC and expanding bypass pipelines. Add Brazil and Venezuela increases and a market already oversupplied before the war returns to oversupply. Over time even the $50 to $60 range gets mentioned (the timing is heavily debated). The implication cuts both ways. Lower oil pulls down headline inflation and helps Warsh and the BOK in their inflation fight. But a late-normalization reverse shock (strategic reserve refills, a China demand recovery) could shake the comfort that "ex-energy, core is fine." And Wall Street's oil forecasts missed badly again ($150 warnings followed by a plunge). Oil is best treated as a risk to manage, not a thing to forecast.

Korea: The End of Cuts, and Real Estate as the Fuse

The path by which the US shift arrives in Korea is a chain where rates, inflation, real estate, and FX are all linked.

Why Inflation Won't Cool

The BOK's read is that prices stay high even as the ceasefire pulls oil down. Costs raised by high oil and a high won are spreading to non-oil items with a lag (the pass-through of accumulated high inflation). What is interesting is the BOK's point about "bonus-driven demand pull." Its analysis finds that wage hikes at top-decile high-bonus firms lift CPI by +0.05 percentage point with a five-month lag. Of the +3.4 percent nominal wage growth in Q1 2026, IT bonuses contributed +1.3 percentage points, in the historical 97th percentile. The second-order effects spread through three channels: labor moving to high-wage IT firms and pushing wages up, more dining and department-store spending near bonus recipients, and other industries using IT pay as a reference wage in their own bargaining. Samsung and SK bonuses became one axis of broad demand pressure.

Real Estate: The Weakest Link Entering a Hiking Cycle

Household credit stood at roughly 1,993 trillion won at the end of Q1 2026, just shy of 2,000 trillion. Household debt at 88.6 percent of GDP is a six-year low, but not because debt shrank, rather because nominal GDP surged (a denominator effect). Capital-region home prices rising about 10 percent a year are the core burden on the BOK's rate decision. The math of real estate that ran on easy money (jeonse leverage, all-in borrowing, presale premiums) is being recalculated by the rising cost of capital.

The problem is that a rate hike is not even across everyone. The size and order of the shock differ by leverage structure.

Real estate is K-shaped too. Prime locations with good rental cash flow hold, while outer, oversupplied, and highly leveraged assets take the first and biggest hit. Not "everything rises" but "only the right locations work." The real trigger is not rates alone but the "combination" of a hike with an income shock and a jeonse-deposit or loan-rollover date. The marginal borrowers break first. Because the BOK drew a line on big steps and chose gradual hikes, though, it is closer to a slow squeeze than a crash trigger.

The BOK's dilemma sits here. To cap home prices it must raise, but raising deepens the risk in 1,993 trillion won of household debt. Hence gradual. Real estate is both the target and the constraint, and that is why Korea has less room to maneuver than the US.

KOSPI and the Won

In the first half of 2026, KOSPI surged on its two semiconductor giants, touching near the 8,900 level intraday and prompting talk of "8,000-pi." That is the Korean version of the K (the same "paradox of crowding" playing out live in Korea). Foreigners turned large net sellers around quarter-end in June (a semiconductor rebalancing) and named "global rate hikes" as the biggest second-half variable. The won held in the 1,500s, where a high won feeds back into import prices and CPI, a vicious loop that ties the BOK's hands.

The Grammar of Corporate Management Changes

When capital is cheap and plentiful, growth itself is a virtue. When capital is expensive and selective, return on invested capital (ROIC) and cash flow become the virtues. The changes companies will face come in five strands.

Discipline in capital allocation. The era of deficit-funded expansion and "share first" strategy fades. Propping up the stock with buybacks and dividends hits a limit too (Q1 buybacks +1 percent vs capex +38 percent). The CFO's focus shifts from "how fast do we grow" to "can we earn more than the borrowed money costs (return on invested capital above the cost of capital)."

The types of winners and losers sharpen. Winners are firms that own a bottleneck (power, HBM, advanced packaging, cooling, power semis), essential goods with pricing power, cash generators, and companies tied to national strategy. Losers are growth stocks holding on by multiples without earnings, highly indebted firms with heavy refinancing burdens, and undifferentiated commodity software.

Zombies and the refinancing cliff. Marginal firms (zombies) that survived on cheap debt in the low-rate years face rates 2 to 3x higher at rollover. The rising cost of capital is a selective clean-out of weakness, accelerating the culling of weak companies. Credit spreads and the maturity wall decide each firm's fate.

Software's survival strategy. A simple subscription-growth story is not enough. Firms must prove a moat AI cannot replace (proprietary data, workflow lock-in, regulatory barriers). At the same time, every company must prove in numbers that "we raised productivity and margins by adopting AI" to hold its multiple (only 1 to 2 percent have quantified it, so this is the battleground of separating wheat from chaff).

The renaissance of physical producers. Manufacturing, materials, energy, and defense, long undervalued as "old economy," get re-rated as beneficiaries of scarcity. Expansion capacity, long-term contracts, and a government-order pipeline become new premium factors.

What Investors Should Check and Build

The Questions Change

Portfolio Frame

Goldman's four capex-beneficiary themes are AI and semiconductor hardware (with bottleneck, cash generation, and quantified earnings first), power and energy infrastructure (AI's real bottleneck), defense (a direct beneficiary of geopolitical fragmentation), and infrastructure, materials, and industrials (physical production capacity). Add discipline: cut weight on high-multiple, high-debt names that have not proven earnings, select by alpha over beta, and run diversification that does not put everything in one basket given crowding risk.

Is Buy-and-Hold Dead?

When the shape of the chart changes, so does not only what you buy but how you hold it. Translate the earlier "fatter and flatter" into a trading strategy and it looks like this.

History is the evidence. In the high-inflation, high-rate years of 1966 to 1982, US indices went sideways for 16 years with big swings, and index buy-and-hold was a waste. The dot-com and financial-crisis stretch of 2000 to 2013 was likewise a trader's market, with the S&P 500 barely back to break-even after 13 years. The "buy-and-hold paradise" of the 2009 to 2021 zero-rate era we are used to was actually the exception. The post-modern cycle is closer to those first two eras.

But do not mistake "favorable" for "easy." High volatility eats clumsy swing traders even faster. Range-bound markets are full of false signals (whipsaws) that trap "buy high, stop out low." Volatility rewards skill and punishes the lack of it. Both the one who just holds and the clumsy trader get cut, which is the other face of "alpha beats beta." And what died is not "buy-and-hold" but "index buy-and-hold." The real winners at the top of the K (the semis, AI, and infrastructure that own a bottleneck and generate cash) still trend strongly upward, so selective buy-and-hold, picking the right winners, works and may be the strongest. When you swing, wider stops and smaller positions are essential given the chop, the positive stock-bond correlation weakens the 60/40 cushion, and the fee and tax drag of frequent trading accumulates.

Scenarios

The base case (most likely) is high rates locked in with the K deepening. Bottleneck and physical assets stay strong, while weak growth stocks and high debt de-rate. Diversification and name selection decide returns. The optimistic case is the AI productivity revolution producing real disinflation (the Warsh scenario) and earnings justifying capex, so the supercycle runs long. The risk case is risk assets correcting three to six months after excess liquidity turns negative (White's warning), the 26x multiple de-rating sharply if AI capex fails to prove earnings, 60/40 falling together, and Korea facing financial stress from real estate and household debt.

The Counterarguments and Limits

To avoid blind faith in this thesis, here is the other side, stated plainly.

• "Excess liquidity" is a model-dependent gauge. The conclusion changes with the method (Ward's critique). It is a warning light, not a prophecy.
• "Warsh equals hawk" is not a given. He welcomes AI productivity and has a dovish side, "cut rates to help small business and the real economy." Some (Infrastructure Capital's Hatfield) forecast three cuts over the next twelve months on falling oil. "The end of easy money" is the dominant narrative, but not a unanimous consensus.
• Global 10-year yields are still below 4 percent (Germany ~2.9 percent, Japan ~2.6 percent). The "about 4 percent" refers to Japan's ultra-long (40-year) bond, so do not mix maturities.
• The AI capex supercycle is a double-edged sword. If earnings proof lags (1 to 2 percent quantifying now), the 26x multiple can collapse fast. "This time is different" may be the most expensive phrase.
• Korea's "hike next month" is an observation. The governor formalized the direction, but it is not confirmed by a rate-meeting result.
• If a real crisis comes, the Fed and government open the taps again. "The end of easy money" is a peacetime structural change, not the disappearance of crisis intervention. Shrinking the balance sheet is not as easy as it sounds either.

The "BOJ cautionary tale" matters in particular. The real enemy may not be the Fed but fiscal dominance. The Bank of Japan raised its policy rate to 1 percent and still got trapped (40-year above 4 percent, 20-year above 3.5 percent) into propping up the JGB market every time long rates spiked. The government floods the market with bonds figuring "the BOJ will buy them anyway," and the BOJ, knowing it must fight inflation, is left with its hands tied. The core point is that the root cause is not a Fed or BOJ mistake but a fiscal deficit pouring out faster than the bond market can absorb. Unless the US cuts its deficit too, whatever Warsh does, it is hard to escape ending up like the BOJ.

Even so, there is value in "showing resolve." Even if the ideal is hard in reality (as in the SVB episode, where intervention came fast in the end), the very act of imprinting "we will defend 2 percent no matter what" anchors inflation expectations and slows the fall into the "BOJ trap." It matters as expectations management, separate from whether it is realized. Finally, the Fed is a committee. The chair cannot run solo like a president. Reforms like scrapping the dots or trimming the statement need board and FOMC votes, and pushback from regional Fed presidents is a variable.

What History Says: The 1970s and the 1940s

There have been two eras in history that resemble today's thesis (high rates locked in, inflation resurgent, fiscal dominance, hard assets favored, K-shaped): 1970s stagflation and 1940s postwar financial repression.

1970s Stagflation (1966-1982): Inflation Flips the Asset Hierarchy

On the macro picture, CPI soared from 2.9 percent ('66) to 13.5 percent ('80), the US 10-year rose from about 4 percent to 15.8 percent ('81), there were two oil shocks, and Bretton Woods collapsed. Volcker raised the funds rate to about 19 to 20 percent and only subdued it in 1982.

The real-return hierarchy of assets flipped.

Stocks were stuck under Dow 1,000 for 16 years and fell -48 percent in 1973-74. The Nifty Fifty (high-growth names at PER 40 to 90) crashed deepest (Polaroid -91 percent, Avon -86 percent). But real total return including dividends (3 to 6 percent then) was roughly break-even, so do not confuse it with "price down two-thirds."

The real reason real estate was "the best hedge" was not the price itself but leverage and the erosion of fixed-rate debt in real terms. Negative real rates melted the debt away with inflation. But highly leveraged mortgage REITs saw about two-thirds go bankrupt in 1973-74. The light and shade of "owning the asset versus leverage." And when Volcker pushed mortgage rates to 18.5 percent, the same houses were crushed, with transactions and starts down 55 percent. The 1979 BusinessWeek "Death of Equities" cover happened, fittingly, to be a contrarian bottom signal three years before the 1982 bull market.

1940s Postwar Fiscal Dominance and Financial Repression (1946-1951): Melting Debt with Inflation

After WWII, US debt to GDP peaked at about 108 percent public and 122 percent total, similar to today. The Fed pegged Treasury yields (short 0.375 percent, long 2.5 percent) to underwrite cheap government funding, the textbook of fiscal dominance. Postwar inflation surged (14 percent in '47), driving real rates to -8 to -14 percent.

The 1951 Treasury-Fed Accord was the event where the Fed broke the peg and regained independence amid Korean War inflation. It is also the template for the "new Accord" Warsh invokes (though Warsh blames the Fed's bloated balance sheet more than fiscal dominance and wants to solve it with QT).

How did the debt come down? Not by surplus repayment but by a mix of growth, inflation, and financial repression. Public debt to GDP fell from 108 percent to about 25 percent by the 1970s. On the asset side, depositors and bondholders were the biggest losers (real -8 to -14 percent), while stocks enjoyed the 1949-66 bull market (~19 percent nominal in the '50s). Real estate was the biggest beneficiary of financial repression. The GI Bill and suburbanization, cheap fixed mortgages (~4 to 5 percent) plus inflation melted the debt, and homeownership rose from 44 to 62 percent.

The Iron Law of Real Estate: The Real Rate Is the Switch

The reason the same real estate moved in opposite directions across eras is one thing: the real rate (nominal rate minus inflation).

Korea today sits exactly on the last row of this table. Household debt about 1,993 trillion won, Seoul apartments +11.26 percent ('25, the most in 19 years) versus outer-area declines (Pyeongtaek -6.9 percent), a K. About 40 percent of first-half transactions were gap investments, plus the June 27 lending rules. Because debt is high, it carries both Japan-style risk and K-shaped polarization at once.

Then vs Now

The same parts are high debt, inflation, fiscal dominance, high rates, hard-asset advantage, and high-PER growth de-rating. The differences matter. First, there is the AI productivity disinflation card the 1970s lacked (Warsh's bet). Second, without 1940s-style capital controls, "melting debt with inflation" is harder, and bond vigilantes punish first. Third, Korea, Japan, and Europe face Japan-style trap risk from aging. Fourth, the real rate now is in the +2 percent range, not the 1940s-style deep negative. So real estate is closer to the 1980-82 or Japan-style "squeeze" than the 1970s "hedge." Only if the government presses real rates back down with inflation does it return to the 1970s form.

Regimes are not forever, worth remembering. As "Death of Equities" (1979) sat three years before the bottom, extreme pessimism is a contrarian signal. In 1982 inflation broke and an 18-year bull market began. Today's pessimism, and optimism, will flip someday too.

Korea Outlook: Scenarios from Today's Coordinates (Forecast)

What follows is a forecast that maps the macro and historical frame onto Korea's current data. It is a scenario, not a certainty, and the biggest external variables are the Fed (Warsh), global rates, and the AI capex cycle.

Core Diagnosis: Korea Is the Intersection of "1970s x Japan"

Korea carries the 1970s form (inflation plus hikes) and the Japan form (high debt plus aging), and a K on top, all in one body. The switch that decides its fate is, again, the real rate. The policy real rate is still negative so it looks favorable for real estate on the surface, but the lending real rate borrowers feel (a 4 to 5 percent mortgage minus 3.1 percent CPI) is already positive. That is why frozen transactions and the collapse of gap-investment math have been underway since 2022. If the BOK raises to about 3.5 percent and inflation holds at 3 percent, the policy real rate flips positive, and that timing is the inflection point for asset markets over the next 6 to 18 months.

The Most Likely Path: Hikes Pushed by FX

The official rationale is prices and financial stability, but the decisive variable that pulls the trigger is likely FX, because FX and import prices are the same coin. A weak won (the 1,500s) feeds CPI through import prices, so "fight inflation" and "defend the won" are effectively the same action. With Warsh's Fed holding hawkishly at 3.50 to 3.75 percent while the BOK sits at 2.5 percent, the gap invites capital outflows and won weakness, so a hike becomes a defensive tool. In short, domestic prices and home prices supply the rationale, and FX pushes from behind.

The expected path is two 25bp hikes (to 3.0 percent) as the near-term base, and up to 3.5 percent (the KDI forecast) if FX and Fed pressure persist. A big step is off the table. In real-rate terms, two hikes "step on" the threshold (policy real rate from -0.6 percent to about 0), while four (3.5 percent) "cross" it (real rate clearly positive). The latter is the true switch for punishing real estate. The fallout from two hikes (to 3.0 percent):

There is a Korea-style mini-BOJ trap here. If won weakness is structural (low growth, strong dollar, the semiconductor cycle), two hikes will not fix it. Then the BOK is trapped: raising does not catch the won, so it must raise more, and household debt turns bad. If Japan's weak spot is JGBs and fiscal, Korea's weak spot is household debt, and that is the ceiling on hikes. Korea is at least not as deep a trap as Japan, thanks to a current-account surplus and FX reserves.

Sector Forecast (6 to 18 Months)

On rates and prices, the cutting cycle is effectively over, with gradual hikes (no big step, a cumulative ~100bp to 3.5 percent) possible from July into the second half. Prices face 3 percent-locked pressure from bonus-driven demand, second-order effects, and high FX, against the cushion of falling oil. The real signal is the flip of the real rate to positive, more than the hike itself.

On real estate, the near term sees frozen transactions and collapsing gap-investment math from the early hikes and the June 27 rules, with the K deepening (Seoul prime holds while outer, provincial, and gap-invested areas fall). When the real rate flips positive, all-in, gap, and multi-home borrowers get punished first. But Seoul's structural undersupply and the jeonse system defend the downside in prime areas, so it is more polarization and frozen transactions than a "crash." Non-prime and provincial areas face Japan-style long stagnation risk from aging and population decline.

On stocks (KOSPI), semiconductor concentration ("8,000-pi") is the Korean version of the K and a crowding risk (extreme dependence on Samsung and Hynix). Hikes are a headwind for high-PER, high-debt names and a tailwind for physical, value, and dividend names (shipbuilding, defense, nuclear, power equipment). Korean defense, nuclear, and shipbuilding are direct post-modern (security and infrastructure) beneficiary themes. Second-half variables are global rates, the AI capex cycle, and foreign flows. On FX, the 1,500s high-FX pressure persists, and the vicious loop from high FX to import prices to CPI ties the BOK's hands.

KOSPI's New Axis: Value-Up and Dividends (FX Is the Premise)

Separate from semiconductor concentration, a fast-growing second axis for foreign money is value-up, dividends, and shareholder returns. In a market where multiple expansion is blocked, most of the return comes from dividends (about three-quarters of US stock total return in the 1970s). Rotating from growth into dividends and value is the natural conclusion of the regime.

Korea happens to have a policy accelerator. The Corporate Value-up Program launched in February 2024 (modeled on Japan's TSE), with the first and second commercial-law amendments legislating shareholder protection and a third amendment (mandatory treasury-share cancellation) under way. The effect already shows: KOSPI broke 5,000 on January 22 and 6,000 on February 25, with 177 value-up disclosures. Breaking the Korea discount (low dividends, chaebol governance) by policy is a structural change that appeals precisely to foreign value and dividend investors.

The precondition, though, is FX. A foreigner's total return is dividends plus capital gains minus FX losses, so if the won stays weak, even a 5 percent dividend is eaten by FX losses and the theme collapses. Conversely, if the BOK's defensive hike stabilizes the won, dividend carry becomes attractive to foreigners. The defensive hike, paradoxically, helps foreign dividend inflows.

Three Scenarios and Checkpoints

The base case (most likely) is gradual hikes then high rates locked in. K-shaped real estate and frozen transactions, KOSPI semiconductor concentration persisting, the won in the 1,500s, a "slow squeeze." The optimistic case has falling oil and AI disinflation calming prices, minimizing hikes and a soft landing. The risk case has the real rate flipping positive and household debt going bad, leading to financial stress from real estate and PF.

Four checkpoints for investors in Korea: the timing when the real rate (base rate minus CPI) turns positive (the inflection for real estate and leverage), selecting real estate by location and rental cash flow plus a stress test on rollover timing, diversifying away from semiconductor concentration into physical, defense, nuclear, power, and dividend value names, and at the household level a stress test on loan rates and a look at switching to fixed.

Key Numbers at a Glance

Conclusion: Resetting the Conventional Wisdom

The investing common sense of the last 15 years was "cheap money floats everyone." The common sense of the next era is "expensive money sorts the wheat from the chaff."

In stocks, the weight shifts from beta to alpha, from software to physical infrastructure, from multiples to earnings and cash flow. Money piles into the real moated companies that own a bottleneck, not the index, but when that crowding gets too dense it becomes a risk in itself. Real estate moves from "everything rises" to "only the right location, the right cash flow works," and Korea, with real estate and household debt as its weakest link, sees variable-rate, all-in, and gap investors pressured first. Companies move from the virtue of growth to the virtue of return on capital, and those that cannot clear their cost of capital are culled at the refinancing wall. In investing, the keys are diversification and selection, and the ability to price things on your own without a safety net (the Fed Put).

One question to ask yourself at the end. Is my investing common sense still stuck in the easy money of the 2010s? The larger the turning point, the more dangerous the assumption that what worked before will keep working. At the same time, the euphoria that everything has changed is dangerous too. Holding together what changed (the price of money, the selectivity of money) and what did not (in the end, earnings, cash flow, and a moat create value) is, I think, the real way to survive the post-modern cycle.

Disclaimer

This post is a macro structural analysis based on public information, written for educational and informational purposes. It is not investment advice to buy or sell any particular stock or asset. The figures reflect the time of writing (June 19, 2026), and markets and conditions change constantly. Phrases like "a hike in July" are market observations, not confirmed facts, and some figures such as the 1,993 trillion won household-debt number are provisional. I have tried to separate "verified facts" from "interpretation and forecast," but model-based gauges and scenarios are inherently uncertain. All investment decisions and their outcomes are the responsibility of the investor.

References

• Bloomberg: The Squeeze on Liquidity Is Just Getting Started (Simon White, 2026.6.18)
• Fortune: Kevin Warsh's first Fed meeting, forward guidance dropped (2026.6.17)
• Goldman Sachs Global Strategy Paper No.76: The Post-Modern Cycle
• Fortune: Data centers, buybacks, and the capex shift (Goldman, Oppenheimer, 2026.6.17)
• CNBC: Fed Chair Warsh's first hires include Project 2025 author (2026.6.2)
• Yahoo Finance: Goldman says consensus 2027 hyperscaler capex is too conservative
• Federal Reserve: The Treasury Tantrum of 2023
• CBO: The Budget and Economic Outlook, 2026-2036
• Money Moves Markets: Assessing excess liquidity, reasons for caution
• Reinhart & Sbrancia: The Liquidation of Government Debt (NBER 16893)