Japan's Bubble and Lost Decades

Why this matters

On December 29, 1989, Japan's main stock index, the Nikkei 225, closed at 38,915.87. It didn't close higher again until February 22, 2024, more than 34 years later.¹ Japan's bubble and the slump that followed are the clearest modern counterexample to the idea that a major stock market always recovers within a few years. They're also the main case study of an economy stuck with falling prices and interest rates near zero, which Federal Reserve economists studied closely when US rates fell in the early 2000s.²

The bubble

After the Plaza Agreement in September 1985, the yen rose rapidly and Japan slipped into recession. To counter it, the Bank of Japan cut its official discount rate five times between January 1986 and February 1987, to 2.5%, then the lowest on record. The rate stayed there for about two years and three months, until May 1989.³

Asset prices took off. The Nikkei peaked at the end of 1989 at 3.1 times its level at the time of the Plaza Agreement. Land prices followed with a lag, spreading from Tokyo to Osaka, Nagoya, and other cities, and commercial land in the six largest cities peaked in September 1990 at almost four times its 1985 level. Over 1986 to 1989, gains on stocks and land added up to 452% of a year's nominal GDP. The economy boomed as well: during the bubble years real GDP grew 5.5% a year, and business investment ran at almost 20% of GDP.³

Expectations did much of the work. In April 1990 the Bank of Japan described a widespread "myth of ever-rising land prices": an assumption that land would keep rising in price, or at least never fall.³ During the bubble, real estate and stocks were often accepted as collateral, so rising prices let firms borrow more.³ The same link worked in reverse once prices fell.

Why the central bank waited

With hindsight, rates stayed low too long. At the time, the case for raising them looked weaker. Consumer prices rose only about 0.1% in 1987 and 0.7% in 1988.⁴ Under the Louvre Accord of February 1987, the major economies had agreed to foster stable exchange rates around their current levels. When the Bank of Japan began nudging market rates up later that year, the October crash described in The 1987 Crash led it to stop, and some argued that higher Japanese rates could set off a plunge in US stock prices and the dollar.³ Three of the Bank's own economists later concluded that the start of tightening was significantly delayed.³

The bust

The Bank of Japan raised the discount rate five times between May 1989 and August 1990, from 2.5% to 6.0%. About a month after the last increase, stock prices had fallen to half their peak.³

Daily closes of the Nikkei 225.¹ Land fell more slowly but further. By 1999, commercial land in the six largest cities was about 80% below its 1990 peak and about 20% below its 1985 level, and losses on stocks and land from 1990 to 1993 came to 159% of a year's GDP.³

Why the slump lasted

At first the downturn looked like an ordinary one, and most forecasters expected a return to strong growth within a couple of years.² But falling stock and land prices had cut the value of the collateral behind loans and the net worth of many firms, especially in construction and real estate, and the banking system was left weakened by a growing pile of bad loans.²

Rather than recognize those losses, regulators and banks put them off. Most large banks could meet capital standards only because inspections were lax, and banks often kept lending to insolvent borrowers through sham restructurings, betting that the borrowers would recover or that the government would bail them out. The economists Ricardo Caballero, Takeo Hoshi, and Anil Kashyap call those borrowers zombies. By their measure, zombies were 5% to 15% of firms through 1993, and more than 25% in every year from 1995 to the end of their data in 2002. As late as 1997, at least five years after the bad-loan problem was recognized, the Ministry of Finance was still insisting that no public money would be needed to help the banks.⁵

	Recognize the loss	Roll the loan over
Loan to a property developer	10	10
Value of the land behind it	3	3
Loss shown this year	7	0
Bank's capital, starting at 20	13, below its required minimum	20 on paper

Rolling the loan over, and lending the developer enough to keep paying interest, keeps the loss off the books for now. The bank hopes land prices recover. If they don't, the loss is still there, and meanwhile money that could have gone to healthy firms keeps a failing one in business.

Zombies did more damage than tying up bank loans. Kept alive, they held on to workers and market share, cutting profits for healthy competitors and discouraging new firms from entering. The researchers found that industries with more zombies had less job creation and lower productivity, and that healthy firms in those industries invested and hired less.⁵

The rest of this lesson is for members

Still to come in this lesson: Deflation, The long road back, What it means for investors, the key takeaways, and 2 quiz questions with explanations.

Membership opens every lesson in the curriculum, the capstone projects, the interactive tools, and the members-only channels in the Hyperion Entities Discord. The first lesson of every module stays free.

Become a member Sign in