They Crashed Japan on Purpose
How money printing, a record Buffett exit, half-real profits, 1929-level valuations and the AI energy wall fit into one machine
Felix Nikolas Prehn · 10 August 2026 · 22 pages
Bottom line
The approved video script lays out six developments that landed in a single week. Read separately, they are six headlines. Read together, the script argues, they describe one connected machine: cheap money is being created to keep asset prices up, while the smartest money quietly heads for the exit. separate headlines the script connects into one machine
Overview
Contents
Bottom line: one machine, six warning lights
4–5
Print to hide the dump: Japan & the money printer
6–7
The smart money is leaving: Buffett's exit
8–9
Half the “record” profits are fake
10–11
The most expensive market since 1929
12–13
The AI story hits a wall of physics
14–15
Someone is holding the next Snapchat
16–17
The whole machine, at once
What it means for investors
Action checklist & free live session
Sources & disclosures
How to use this report.
Read it beside the video. The sections follow the script's information and order. The goal is not to predict the exact timing of the next market move. It is to explain the plumbing that can make apparently unrelated events move together — and why knowing when to sell matters more than what you buy.
Bottom Line
One machine, six warning lights
BOTTOM LINE:
The script's central claim is that six developments — money printing to absorb Japan's bond selling, a record Buffett exit, half-real corporate profits, 1929-level valuations, an unpowerable AI build-out, and the history of famous stocks quietly bleeding out — are not coincidences. They are connected symptoms of a market held up by cheap money and optimism.
Cheap money is created
→
It props up asset prices
→
Insiders quietly reduce risk
THE SETUP
Liquidity holds the market up. The script argues that dollars are being created to soak up the US debt Japan is selling, keeping interest rates down and stock prices up.
THE SIGNAL
The smartest money is reducing risk. A record Buffett sale of Japanese bonds is presented as a tell: watch what experienced investors do, not only what they say.
THE FRAGILITY
Prices rest on soft ground. Half of the “record” profits are described as paper gains, valuations sit at 1929-era levels, and the AI story that justifies them runs into an energy wall.
THE LESSON
Being right is not enough. The script's core message is that investors can be correct about a bubble and still lose money if they have no plan for when to sell.
Print to Hide the Dump
Money printing with a boring name
Philip Pilkington post describing the US Treasury printing dollars to buy back Treasury
bonds from Japan
Illustrative source image supplied with the video. The script describes the US absorbing the Treasury bonds Japan is selling, before a visible “dump” can alarm the market.
WHY JAPAN MATTERS
The biggest foreign lender is selling. Japan is described as the largest foreign owner of US debt. When a seller that size steps back, bond prices fall and interest rates want to rise.
WHAT RISING RATES HIT
Everything at once. The script notes higher rates touch mortgages, the government's own debt payments, and stocks
- because a safe bond that pays well competes with shares.
THE MOVE
Absorb the dump quietly. The script's claim is that new dollars are created to buy the bonds before the story reaches the front page.
The Mechanism
Repurchase agreement or photocopier?
Japan sells US debt
The US creates dollars to buy it back
The balance sheet expands; new dollars enter the system
Cheap money keeps stock prices up
THE TECHNICAL OBJECTION
“It's only a repurchase agreement.” The script acknowledges the counter-argument that this is temporary lending, not permanent printing. Its response: whatever the label, new dollars flood in and the effect is the same.
WHERE THE MONEY GOES
Cheap money finds stocks. When money is cheap and everywhere, the script argues, it flows into assets — a reason index funds kept climbing.
THE TONE
Clever, but desperate. The plan is described as keep printing, keep money cheap, keep the party going — and hope nobody looks too closely.
→
→
→
The Smart Money Is Leaving
Buffett heads for the door
Post describing Warren Buffett selling $285 billion of Japanese bonds before Japan's
new debt figures
Illustrative source image supplied with the video. The script describes about $285B of Japanese bonds sold shortly before Japan announced new government debt figures.
WHO THIS IS
The “never sell” investor. The script highlights the irony: the man whose brand is buy-and-hold-forever cleared a very large position off the table.
THE TIMING
Right before the bad numbers. The sale is described as landing in the same week Japan's new debt figures were due — the detail the script finds most interesting.
TWO READINGS
Coincidence or information. The script offers both: maybe it means nothing, or maybe the best-informed investor alive saw the same machine and chose the exit.
Watch What It Does
Follow the money, not the microphone
BOTTOM LINE:
The script's practical takeaway is simple: do not just listen to what the smart money says on television. Watch what it actually does with its own capital.
WORDS VERSUS ACTIONS
The gap is the signal. Public reassurance and private positioning can point in different directions. The script argues the positioning is the more honest of the two.
FRONT OF THE ROOM
You don't need an announcement. When the person leading the room stops clapping, the script says, you simply need to notice — not wait to be told.
THE BRIDGE
Noticing when to leave. This is where the script pivots to its central theme: the single most valuable skill an investor can own is knowing when to get out.
Not a recommendation.
This report describes the script's reasoning for educational purposes. It is not advice to buy or sell any security, currency or bond, and no rating or price target is provided.
Half the Profits Are Fake
The record profit number is only half
true
Chart showing headline S&P 500 earnings growth around 27% versus roughly 13% once
other income is removed
THE HEADLINE
~27% growth sounds spectacular. The script says this is the number the whole bull case rests on — and the number you hear on the news.
THE FOOTNOTE
Strip out “other income” and it's ~13%. The script says the ordinary company — the one actually selling products — grows about half as fast as the headline.
THE SOURCE OF THE GAP
A handful of giants marking up their own bets. Big Tech names booking paper gains on their investments, not extra sales, close the difference.
Marking Your Own Homework
A loop that works beautifully — in both
directions
Big Tech owns stakes in each other and the same AI names
Those stakes rise — booked as “profit”
Higher profits push the stocks up further
Which justifies marking the stakes up again
THE UPSIDE SPIRAL
Everyone holds everyone up. The script describes a self- reinforcing loop where paper gains become reported earnings, which lift prices, which create more paper gains.
THE REVERSE
It runs the other way too. When the paper gains reverse, the script warns, “record profits” don't just slow — they can go backwards, taking the story with them.
THE TAKEAWAY
Technically correct is not the whole truth. A market priced for record profits, sitting on profits that are half paper, has a long way to fall if the loop turns.
→
→
→
Most Expensive Since 1929
A 1929 price tag on today's market
Long-run US stock valuation chart showing today second only to the 2000 dot-com peak,
above the 1929 level
THE CLAIM
Second most expensive in history. The script says US stocks have reached the second-highest valuation ever recorded — above 1929, below only the 2000 dot-com peak.
THE REFERENCE LINE
The red line is normal. The chart's long-run average marks the sensible price paid across a century. The script's point: we are far above it.
THE COMPANY WE KEEP
1929 and 2000. The script reminds viewers how the years after both peaks played out for ordinary pensions.
Explained Simply
How many years of earnings are you paying for?
- $10
- paid for a share that earns $1 a year = ratio of 10
- $40
- paid for that same $1 = ratio of 40, badly overpaying
2nd
highest valuation in US market history today
BOTTOM LINE:
The valuation measure in the chart simply asks how many years of a company's earnings you are paying to own one share. The higher the number, the more you are overpaying for the same slice of profit.
WHAT EXPENSIVE MEANS
Little room above, long way down. The script is careful: expensive does not mean it crashes tomorrow. Bubbles usually get more expensive right up until the top.
WHAT IT DOES MEAN
Poor odds from here. When investors overpay this much, the script says, history is consistent about weak returns over the following years.
THE HONEST CAVEAT
Timing is not the claim. The report makes no prediction about the date of any move; it explains the risk of paying 1929-plus prices for the last stretch of a ride.
The AI Story Hits a Wall
You can't run a data center on a story
Post describing 474GW of interconnection requests in Texas, about 90% data centers,
versus a 91GW record peak
Illustrative source image supplied with the video. The script uses grid figures to test whether the AI build-out is physically deliverable.
THE REAL CONSTRAINT
Electricity, not enthusiasm. The script accepts that AI is real and powerful — then points out it runs on enormous, physical amounts of power.
THE REQUEST
474GW in one state. Texas alone has requests to plug in 474 gigawatts of new demand, about 90% of it data centers.
THE BRAKE
Officials hit pause. The script notes the state moved to freeze new data-center rollout — a sign the numbers do not fit the grid that exists.
The Numbers Don't Fit
The grid always wins
474GW
requested in Texas, ~90% data centers
91GW
Texas all-time record demand (normal
40–80)
- ~100
- nuclear plants' worth the US is short
(80–100GW)
THE SCALE
Roughly five times the record. The requested demand is about five times Texas's all-time peak — almost all of it to feed AI. The script calls that physically impossible with today's grid.
THE NATIONAL GAP
About 100 nuclear plants short. The script cites an estimate that the US is short 80–100 gigawatts through the end of the decade — power that does not yet exist.
SPEND NOW, REVENUE LATER
Fortunes today for promises years away. Sky-high spending and valuations rest on revenue expected far in the future, powered by electricity nobody has built. Physics, the script says, doesn't do hype.
The Next Snapchat
$10,000 turned into $682
Snap five-year chart showing a $10,000 investment worth about $682 today
THE NUMBER
A ~93% loss. The script says $10,000 put into Snap five years ago would be worth about $682 today.
NOT A SCAM
A real, famous company. The point of the example is that Snap was a household name analysts liked — not an obvious fraud — and it still deleted most of the money entrusted to it.
HOW IT HAPPENED
It bled, slowly. The script stresses it did not crash in a dramatic afternoon; it drifted lower for years while holders waited to get back to even.
“It'll Come Back”
Three words that destroy more wealth than any crash
A famous stock starts drifting lower
Didn't sell at −20% — “it'll come back”
Didn't sell at −50% — “too late to sell now”
Rode a household name almost all the way down
THE PATTERN
A slow bleed, not a bang. The script argues most hot names in a bubble do not go to a clean, dramatic zero. They quietly drift lower for years.
THE HISTORY
It rhymes every cycle. Dot-com darlings that never came back are offered as precedent — general market history, not a specific forecast.
THE DECIDER
One skill separates the outcomes. In this market, the script says, whether that holder ends up being you comes down to one thing: knowing when to get out.
→
→
→
Line Them Up
Six headlines, one connected machine
Japan dumps debt → America prints to hide it → cheap money props up stocks
Smart money leaves → half the profits are paper → priced like 1929
AI hits a wall of physics → ordinary people end up holding the next Snapchat
BOTTOM LINE:
Laid end to end, the six warning lights describe a single sequence — not six random headlines. The script is explicit that this is an explanation of the plumbing, not a prediction of the exact day anything unwinds.
Early and wrong feel the same.
The script's honest caveat: bubbles can run higher and longer than anyone believes. Panic-selling too early and holding too long can both hurt. That is precisely why a rules-based approach to selling matters — the goal is to remove guesswork, not to time a top.
→
→
What This Means for You
You can be right about the bubble and still lose
freeze as a winner turns back into a loss hold a fading name waiting to break even follow a prepared selling framework
THE BEHAVIOURAL TRAP
The gain shows up, then vanishes. The script's recurring example is the investor who watches a stock go up a great deal, freezes, tells themselves it will keep going, and hands the gain back.
KNOWLEDGE IS NOT THE GAP
It's a skill problem. Most people who owned the biggest winners still ended up poorer, the script argues, not because they picked wrong but because they had no exit plan.
THE MISSING RULEBOOK
Wall Street never handed it over. The script frames selling as a learnable process: when to take profits, when to cut losses, fitted to your own portfolio and timeline.
Turn Understanding into Process
Five questions before the next warning
shot
Free live session: when to sell
whentosell.org
Learn the rules Wall Street never gave you: when to take profits, when to cut losses, and how to never watch a 10x turn back
Do I own anything that has risen a lot with no plan for when to sell it?
✓
Am I overexposed to one crowded theme — AI, Big Tech or a single index?
✓
If a famous holding fell 20%, do I know in advance what I would do?
✓
Can I explain why each business I own actually makes money?
✓
Would a broad sell-off force me to act, or would I have a plan ready?
✓
Disclaimer
This report is for educational and informational purposes only and does not constitute financial, investment, legal, tax or accounting advice. It is not a recommendation, solicitation or offer to buy or sell any security, currency, commodity or financial product. No buy/sell rating or price target is provided. All investments involve risk, including the possible loss of principal. Market events, relationships and historical comparisons described here may not repeat. Illustrations simplify complex systems and may omit factors. Readers should conduct their own research and consult appropriately qualified professionals before making financial decisions. Past performance is not indicative of future results. The Prehn Institute makes no guarantee regarding outcomes.

