Philippe Aghion
Collège de France
Économie des institutions, de l'innovation et de la croissance
Année 2025-2026
Conférence - Takeo Hoshi : Rapid Growth and Long Stagnation
Takeo Hoshi est invité par l'assemblée du Collège de France sur proposition du Pr Philippe Aghion.
Takeo Hoshi : Professeur d'économie et directeur du Tokyo College, université de Tokyo
Résumé
Japan's postwar economic growth is often described as a miracle, and for good reason. Having suffered immense destruction during the Second World War—with roughly one quarter of its national wealth and production facilities lost—the country emerged from devastation and rebuilt rapidly. Early stabilization policies, such as the Dodge Plan, helped curb postwar hyperinflation, and the outbreak of the Korean War in 1950 brought a surge of demand for Japanese goods. This laid the foundation for what became known as the rapid economic growth period.
Between 1950 and 1973, Japan's real GNP expanded at an average annual rate exceeding 9 percent. Growth accounting exercises show that capital accumulation was important: Japan invested heavily in physical capital, supported by high savings rates. Labor force increase also contributed, but the most significant factor was the acceleration of productivity growth. Total factor productivity accounted for more than half of Japan's overall growth in this era, highlighting the role of technological catch-up and improvements in efficiency.
The catch-up process followed the logic of diminishing marginal productivity: starting from a low capital base, the returns to investment were very high. Japan imported advanced technologies from abroad, adapted them, and combined them with its own strong base of human capital. Educational achievements of Japanese workers were already high even compared with economically advanced countries such as the U.S., so the workforce could absorb and utilize new methods quickly. In the language of growth theory, Japan had both the stock of human capital and the institutional framework for rapid technological adoption.
The first major break in this rapid expansion came with the oil shocks of the 1970s. The quadrupling of oil prices in 1973 caused inflation and forced Japanese industries to restructure. Although Japan adapted relatively well by shifting to more energy-efficient production, the overall growth rate slowed. From 1973 to 1987, Japan's average annual growth was closer to 4 percent—still impressive by international standards, but less than half of what it had been during the rapid economic growth period.
Several explanations for this slowdown have been proposed. An obvious one is the end of catch-up phase of economic growth: as Japan approached the technological frontier, it could no longer rely on simply importing and adapting foreign technologies. Productivity growth naturally decelerated as the scope for catch-up diminished.
Yet even in this period, Japan was far from stagnating. The economy continued to expand, industries remained competitive internationally, and living standards continued to improve. However, the seeds of later problems were beginning to take root—particularly in the financial sector and in the structural rigidity of industries that resisted creative destruction.
The expansion from 1986 to 1991, sometimes referred to as the "bubble economy," was one of the longest postwar upswings. Asset prices surged dramatically: the Nikkei 225 stock index tripled between 1985 and 1989, while urban land prices rose to unprecedented levels. This boom was fueled by a combination of loose monetary policy, speculative expectations, and financial liberalization that outpaced regulatory adaptation.
Many observers at the time believed that Japan had entered a new growth stage. But in retrospect, it is clear that asset values far exceeded fundamentals. It was indeed a "bubble economy." Speculative bubbles always end up collapsing. The Japanese bubble was not an exception. Japanese asset prices experienced severe declines in the early 1990s. The Nikkei 225 fell from nearly 39,000 at the end of 1989 to below 15,000 by the mid-1990s, and land prices also declined sharply. The collapse of the bubble had serious consequences on the real side of the economy: households cut back their consumption, corporations stopped borrowing money and investing in productive capitals, and banks found themselves burdened with non-performing loans.
What followed has been described as Japan's "lost decade"—and in fact, many now speak of "lost decades," as stagnation persisted well into the 2010s. Real GDP growth slowed to barely above 1 percent annually, far below the OECD average. Deflation set in, suggesting there was a serious demand shortage.
The core of the problem, however, was not simply weak demand. Japan experienced problems on the supply side. Productivity growth fell dramatically. Decompositions of total factor productivity (TFP) show that the reallocation effect—the contribution from shifting resources from low-productivity to high-productivity firms—declined significantly. Even more troubling, the exit effect turned negative: high-productivity firms exited the market while low-productivity firms remained.
The negative exit effect is closely related to the zombie problem. Banks, saddled with bad loans, kept extending credit to insolvent or unproductive firms in order to avoid recognizing losses on their balance sheets. These "zombie" firms absorbed resources but did not contribute meaningfully to productivity growth. More importantly their survival reduced profitability of otherwise healthy firms and discouraged entry by new productive firms. In other words, zombies slowed the process of creative destruction, which is essential for productivity growth in a mature economy. The result was a long stagnation in which capital and labor were trapped in inefficient uses.
Demographics also played a role in reducing the supply capacity. Japan's labor force was no longer expanding as rapidly, and the aging population began to constrain growth.
One debate during the 1990s was whether Japan's stagnation was primarily a demand-side problem (a persistent demand shortage) or a supply-side problem (lack of productivity growth). If only a prolonged demand shortage was the problem, it should have produced a deflationary spiral, but Japan instead experienced persistent but mild deflation. This suggests that both demand and supply constraints were at work. Expansionary monetary and fiscal policies occasionally provided stimulus, but without addressing the structural problems, their effects were limited.
In the 2010s, Abenomics attempted both demand stimulus and structural reforms. The first two "arrows"—monetary easing and fiscal expansion—succeeded in eliminating the demand shortage and ending deflation. But the third arrow, structural reform, was more difficult to implement. Measures such as corporate governance reform, labor market reform, and policies to increase female labor participation had some impact, but the deeper challenge of fostering productivity through creative destruction remained.
Japan's experience thus underscores a central lesson: demographics and demand matter, but productivity growth is decisive. Without mechanisms that promote restructuring and reallocation, mature economies stagnate. For Japan, the unwillingness to allow inefficient firms to exit severely depressed creative destruction. It was a key reason for the lost decades.