Japan suffered a massive banking and financial crisis starting in 1989, which was followed by a long period of slow growth and deflation. Even after the Japanese economy appeared to recover, growth remained low, prompting various observers to claim that the financial crisis had not only irretrievably reduced the level of Japanese output but also altered Japan’s long-term growth potential. Given the similarities between Japan’s crisis and that in the United States—a bursting housing bubble, a collapse in equity prices, and failing financial institutions—influential observers have speculated that growth in the U.S. economy will also remain depressed even after a sustained recovery is achieved, the so-called “new normal.”
An examination of the data calls into question the hypothesis that Japan’s crisis caused its secular growth deceleration, and also exposes the limitations of drawing a simple comparison between Japan and the United States. Well before the outbreak of its financial crisis, economic growth in Japan was slowing. A bubble that emerged around 1985 concealed this decline, and the crisis that followed exaggerated it, but the fundamentals of the Japanese economy predicted an era of much slower growth well before 1990. The data shows no such secular deceleration in the United States, suggesting that Japan’s experience gives little insight into long-term growth prospects in the United States.
Following World War II, Japan experienced a period of extraordinary growth—dubbed the “Japanese Miracle.” From 1955 to 1990, growth averaged 6.8 percent per year and GDP multiplied eight times. Growth fell below 3 percent only one year, during the first oil shock in 1974. In 1979, Harvard economist Ezra Vogel predicted that Japan would surpass the United States as the world’s leading economy, while others maintained this prediction as late as 1995.
At the end of 1990, however, Japan’s bubble economy imploded. Output growth slowed to 3.3 percent in 1991 and 0.8 percent in 1992. The financial crisis reached its most acute phase in 1997 when failing banks had to be merged. From 1991 to 2002, real GDP growth averaged 1 percent per year; GDP per capita grew by an average of 0.8 percent annually.
Had the bubble and crisis not occurred, Japan’s economy would be 10 percent larger than it is today, but it would be growing no faster.
Though hardly disastrous, this rate of growth was interpreted as a “Lost Decade” compared to Japan’s glory years. Furthermore, progress following the sustained recovery that began much later, in 2003, also appeared slow. Growth averaged “only” 2.1 percent per year (2 percent per capita) from 2003 through 2007.
But the claim that the credit crunch engendered by the crisis could depress demand for 20 years, even after Japan’s banking system was effectively reorganized and returned to health in 1999, is not plausible. The logarithmic trend line of GDP growth from 1960 to 1983 (approximately when the Nikkei 225 began to surge) illustrates that GDP growth in Japan was already trending down, well before the real estate bubble and subsequent banking crisis. Since the bubble and the crisis, Japan’s GDP level has fallen 10 percent below the trend line, reflecting the permanent output loss caused by the crisis. The GDP growth trend, however, predicts an average growth rate of 2.1 percent from 2003 to 2007—the actual growth rate over that period—suggesting that had the bubble and crisis not occurred, Japan’s economy would be 10 percent larger than it is today, but it would be growing no faster.
What accounted for Japan’s growth slowdown, if not the banking crisis? A standard decomposition of Japan’s GDP growth suggests that three factors were at work: labor force growth falling, and investment and productivity growth rates converging to levels broadly consistent with those of other advanced economies.
From 1955 to 1970, Japan’s labor force grew at an average of 1.8 percent per year, nearly twice as fast as recent growth in the United States. By 1985, growth had fallen to nearly a third of this rate; by 1996, the labor force was shrinking.
Investment rates also peaked around 1970, at 35 percent of GDP—again, nearly twice the recent average in the United States. By 1985, investment had fallen to 27 percent. After surging over 30 percent at the height of the real estate bubble, it continued its decline, dropping to 23 percent in 2007.
Finally, Japanese total factor productivity (TFP) growth slowed from 6.5 percent per year (compared to a recent average of 1.3 percent in the United States) in the 1960s to 1.4 percent in the 1970s, long before the crisis began.1
As the graph shows, the major drivers of economic growth were weakening by 1975, and GDP growth decelerated. Additionally, projecting the trends of these factors, rather than simply GDP growth itself, forward can provide a more articulated estimate of the long term growth potential of the Japanese economy. Based on this calculation, growth in Japan was driven significantly above potential during the bubble—and below it during the crisis—but it returned to a level in line with economic fundamentals once the recovery took hold.
Fully accounting for this slowdown in TFP and investment goes beyond the scope of this note, but the data is consistent with the idea that Japan had finished “catching up,” as economic fundamentals converged to those of other developed countries. In 1975, Japanese GDP per capita reached over 70 percent of that of the United States in purchasing power parity terms. Historically, advanced economies that have achieved this level of income relative to the United States have averaged annual productivity growth of 1.2 percent and an investment rate of 20 percent. From 1975 to 1980, Japanese productivity was growing at 1.4 percent annually, slowing from 4.5 percent annually during the previous decade. Investment remained high, averaging 31 percent per year from 1975 to 1980, and while it adjusted quickly following the crisis, it remained above the average of other advanced countries. Japan’s shrinking labor force completes the explanation.
Unlike Japan, the U.S. economy was on only a modestly slowing secular growth path before the crisis. The U.S. economy experienced five recessions between 1970 and 2007, but it has maintained a steady trend growth rate of 3 percent per year over that period, slowing to 2.7 percent in the years preceding the crisis. Unlike in Japan, labor force growth, investment rates, and productivity growth showed little sign of slowing.
Thus, the most that can be said about the United States from Japan’s experience is that output lost during the crisis may be lost forever. Furthermore, once a sustained recovery has taken hold, the growth rate in the United States should return to its trend, as it did in Japan, implying average growth of around 2.5 to 3 percent per year.
Sources: BEA, The World Order in 2050, author's calculation.
Though this is encouraging news for the United States, it also carries an important policy message: the severe and permanent loss of output during the Japanese crisis was clearly associated with the twelve-year duration of the crisis, reflecting in part the delay and modesty of Japan’s policy response, including the time taken to reorganize and repair the banks. Though the U.S. macroeconomic policy response has been more vigorous than Japan’s during its financial crisis, there is more work to be done. The quicker the United States completes restructuring its banks, insurance companies, and government-sponsored enterprises, the smaller the permanent output loss will be.
Bennett Stancil is a junior fellow in Carnegie’s International Economics Program.
1 This calculation, based on a simple Cobb-Douglas production function, is roughly in line with academic estimates.
You are leaving the Carnegie–Tsinghua Center for Global Policy's website and entering another Carnegie global site.