Japan's 1989 bubble vs the AI bubble: a lesson for Asian investors
The essentials in 30 seconds
- Japan’s stock index peaked in December 1989, and recovery took decades.
- The case teaches that the horizon of your savings matters as much as the size of a fall.
- Japan’s bubble ran on domestic credit, while the AI bubble runs on global capital and company cash.
- Four checks suit Asian investors: horizon, home concentration, credit link, and currency.
- The EQOS Asia AI study records AI valuations up ×4 in 18 months and a market score of 0.337.
Japan’s 1989 bubble vs the AI bubble is the comparison that speaks most directly to Asian investors, because it happened in Asia and its aftermath lasted for decades. As of October 2026, this page sets the two cases side by side on credit, concentration, and horizon, and gives 4 checks you can apply to your own savings. It sits under the guide to bubble history, Tulips, railways, dot-com. This page is general information, not investment advice.
On this page
Compare the EQOS Asia offers and request a 30-minute pre-diagnostic.
What happened in Japan’s 1989 bubble, and why compare it with the AI bubble?
Japan’s stock index reached a record on 29 December 1989, after a boom in shares and in land. Prices then fell, and the years after 1991 became known as the Lost Decade. Urban land prices fell from their peak, and nationwide land prices rose year on year for the first time only in 2018. The case set a reference for slow recovery.
It is cited for AI because it shows an outcome that headlines about sudden crashes leave out. A bubble can deflate over many years, and recovery of the old peak can take far longer than a typical investment horizon. That possibility deserves a place in any AI bubble plan, next to the quick-fall scenario.
How did credit shape Japan’s 1989 bubble compared with the AI bubble?
Japan’s bubble is commonly linked to domestic bank lending against rising land and share prices. When prices fell, the collateral lost value, banks held bad loans, and lending slowed for years. Credit tied the stock market, the property market, and the banks into one loop, which made the aftermath long.
Today the AI boom draws on company cash, bonds, private credit, and global investors. The EQOS Asia AI study records about $200B a year invested in AI (EQOS Asia AI study, April 2026). Ask how much of that money is borrowed against AI assets. The more borrowing is tied to AI prices, the more a fall can reach lenders and slow lending.
Why does your time horizon matter most in the Japan 1989 lesson?
An investor who bought at the peak in December 1989 faced a wait of many years, and recovery needs far more time than most plans allow. The same fall hurts a saver who needs money in 5 years much more than one who needs it in 30. The horizon decides the damage, more than the size of the drop.
List each pot of savings with the date you need it. Match the AI exposure of each pot to its date, and keep the shortest pots away from concentrated positions. This exercise needs no forecast. It protects you whether the AI bubble ends in a quick fall, a slow decline, or a soft landing.
How does home concentration raise the AI bubble risk for Asian investors?
A saver who holds mostly home assets sees a fall in domestic shares and land hit their wealth twice. Asian investors today can show a similar pattern: a home market, a home currency, and a technology theme that overlaps with local chip makers. Singapore, ASEAN, and Hong Kong portfolios often hold Taiwan and Korea chip names alongside US AI leaders.
Count exposure by theme, not by country. A portfolio that looks diverse across 5 markets can still rest on one AI supply chain. Add up the share of chips, cloud, and data centre names across all your funds. If one theme exceeds the share you can carry through a long decline, rebalance toward the horizon rule above.
Where does Japan’s 1989 bubble differ from the AI bubble?
Japan’s bubble was national, driven by land and bank credit, while the AI boom is global and centred on technology companies with large cash flows. These differences make a repeat of the exact Japanese path less likely.
A less likely repeat is not a zero risk. The useful response is to prepare for several paths, with a plan for each, and to review the plan on a fixed date every quarter. The shared feature is a theme that investors came to treat as a one-way bet. Valuations that assume uninterrupted growth leave little room for a pause. Keep the Japan case as a stress scenario in your plan, and give it a weight that reflects the differences.
Which four checks apply the Japan 1989 lesson to the AI bubble?
Check horizon: match each pot of money to the date you need it. Check home concentration: add up the AI theme across funds and home holdings. Check the credit link: how much borrowing sits against AI assets. Check currency: does a fall in AI shares coincide with a move in the currency you spend in?
Write the 4 answers with a date, and compare them with the previous quarter so that you see the direction of travel, which says more than a single snapshot. The mother guide, Tulips, railways, dot-com: where the AI bubble fits in bubble history, places Japan beside the other cases, and it shows which risk each case highlights. For Japan the risk is duration, which is the one a plan can address in advance.
How can EQOS Asia help Asian investors apply the Japan 1989 lesson?
The EQOS Asia AI study scores the enterprise AI market at φ(S) = 0.337, in the fragile zone where a shock can tip the system (EQOS Asia AI study, April 2026). The EQOS Asia pricing page lists the diagnostic that applies this scoring to your case.
A 30-minute pre-diagnostic is the first step. You bring your horizon, your holdings, and the decision in front of you, and the session returns a score and a ranked set of options for a soft landing, a correction, and a long decline.
See also in this guide
Frequently asked questions
What was Japan’s 1989 bubble?
Japan’s stock index reached a record on 29 December 1989 after a boom in shares and land. Prices then fell, and the years after 1991 became known as the Lost Decade. Nationwide land prices rose year on year only in 2018.
What does Japan’s 1989 bubble teach about the AI bubble?
It teaches that an aftermath can last many years, so the horizon of your savings matters as much as the size of a fall. It also shows how credit and home concentration lengthen a decline. As of October 2026 these lessons apply to AI holdings.
Is the AI bubble the same as Japan’s bubble?
No single case repeats exactly. Japan’s bubble was national and ran on bank credit against land, while the AI boom is global and draws on company cash, bonds, and private credit. The shared feature is a theme treated as a one-way bet. This is general information, not investment advice.
What should Asian investors check first?
Start with horizon: match each pot of savings to the date you need it. Then add up the AI theme across funds and home holdings, check how much borrowing sits against AI assets, and note the currency you spend in.
Test your knowledge
1. When did Japan’s stock index reach its record?
- March 2000
- December 1989
- February 1637
Show the answer
December 1989.
2. What lengthened Japan’s aftermath?
- A credit loop through banks, property, and shares
- A shortage of chips
- A new tax on savings
Show the answer
A credit loop through banks, property, and shares.
3. What decides the damage of a long decline?
- The logo of the fund
- The country of the broker
- The horizon of your savings
Show the answer
The horizon of your savings.
4. Which count should Asian investors use for exposure?
- By theme
- By country
- By currency symbol
Show the answer
By theme.
5. How many checks apply the Japan 1989 lesson?
- 2
- 4
- 9
Show the answer
4.
About the author
Stéphane Jambu
President, EQOS Forecast Asia
Stéphane Jambu leads EQOS Forecast Asia and works as a consultant in e-commerce, search visibility, and AI. LinkedIn profile
Sources: EQOS Asia AI impact study (eqos-asia.com/case-studies/ia-2026.html), EQOS Asia pricing page (eqos-asia.com/pricing.html) and contact page (eqos-asia.com/contact.html); Japan index peak, Lost Decade and land price dates checked on en.wikipedia.org/wiki/Japanese_asset_price_bubble, consulted on 22 October 2026. Page updated on 22 October 2026.