"The rebalancing debate has sparked renewed interest in Japan’s experience since the 1980s. Some argue that this is a cautionary tale, exemplifying the dangers of reorienting economies through currency appreciation (People’s Daily, 2010). They claim that the appreciation of the yen after the Plaza Accord forced the authorities to introduce an offsetting macroeconomic stimulus, which then led to an extraordinary asset price boom followed by an extraordinarily painful bust. Japan was one of the world’s fastest-growing economies for three decades but has averaged only 1.1 percent real GDP growth since 1990, while prices have steadily declined. Consequently, the size of Japan’s economy today is about the same as in the early 1990s. The sequence of events is clear and striking. But there are reasons to doubt that it was truly inevitable, whether the Plaza Accord was really the direct cause of Japan’s “Lost Decades.”
What Happened?
The events began in September 1985, when delegates from the G5 countries met at the Plaza Hotel in New York, declared the U.S. dollar overvalued, and announced a plan to correct the situation.1 The essence of the plan was that the main current account surplus countries (Japan and Germany) would boost domestic demand and appreciate their currencies. In effect, this agreement marked a major change in policy regime: the Federal Reserve was signaling that after a long and successful fight against inflation, it was now prepared to ease policies, allow the dollar to decline, and focus more on growth. This signal was backed by coordinated currency market intervention and a steady reduction in U.S. short-term rates. Accordingly, it triggered an exceptionally large appreciation of the yen, amounting to 46 percent against the dollar and 30 percent in real effective terms by the end of 1986. (The deutsche mark appreciated similarly.) As a result, Japan’s export and GDP growth essentially halted in the first half of 1986. With the economy in recession and the exchange rate appreciating rapidly, the authorities were under considerable pressure to respond. They did so by introducing a sizable macroeconomic stimulus. Policy interest rates were reduced by about 3 percentage points, a stance that was sustained until 1989. A large fiscal package was introduced in 1987, even though a vigorous recovery had already started in the second half of 1986. By 1987, Japan’s output was booming, but so were credit growth and asset prices, with stock and urban land prices tripling from 1985 to 1989. Then, in January 1990, the stock price bubble burst. Share prices lost a third of their value within a year, and two decades of dismal economic performance followed (Figure 1.4.1). Today, nominal stock and land prices are back at their early 1980s levels, one- quarter to one-third of their previous peaks.
The critical question is whether this sequence was inevitable. In other words, did the appreciation force Japan to introduce a powerful stimulus to sustain growth, which then triggered a bubble, which caused the Lost Decades when it collapsed? Let’s consider each step in turn.
Was Such a Large Stimulus Needed?
Studies suggest that, in fact, the monetary policy easing may have been excessive. Estimates by Jinushi, Kuroki, and Miyao (2000) and Leigh (2010), among others, suggest that the policy rate was up to 4 percentage points too low during 1986–88 relative to an implicit Taylor rule based on the output and inflation outlook. Why, then, did the central bank sustain such a policy? A key reason is that current inflation remained reasonably well behaved, which led some economists to argue that soaring growth rates did not represent a cyclical boom but rather a “new era” of higher potential growth. This growth was particularly gratifying because it was led by domestic demand, a key commitment under Plaza.
But IMF reports at the time suggest another factor was also at work. The authorities worried that higher interest rates would further strengthen the yen and feared that appreciation would eventually have serious effects on the economy. In the end, external demand did indeed diminish. But it did not collapse. Real exports continued to grow in the five years after Plaza, by an average of 2½ percent a year (half the rate of the previous five years), while the current account surplus diminished by a moderate 2 percentage points of GDP. (Similarly, Germany’s currency appreciation failed to derail its export or GDP expansion, even with a smaller monetary response.) Put another way, excessive stimulus was adopted in part because there was excessive concern about the impact of appreciation.
Did the Stimulus Cause the Bubble?
Although the monetary easing was certainly large, it is far from clear that it alone was responsible for the asset price bubble. Chapter 3 of the October 2009 World Economic Outlook and Posen (2003) have examined the link between monetary policy and asset price booms in advanced economies over the past 25 years. They conclude that policy easing is neither necessary nor sufficient to generate asset price booms and busts. In Japan’s case, two other elements seem to have played a large role. As Hoshi and Kashyap (2000) explain, financial deregulation in the 1970s and early 1980s allowed large firms to access capital markets instead of depending on bank financing, leading banks to lend instead to real estate developers and households seeking mortgages. As a result, bank credit to these two sectors grew by about 150 percent during 1985–90, roughly twice as fast as the 77 percent increase in overall bank credit to the private sector. Finally, because the dangers of real estate bubbles were not well understood in those years, the Japanese government did not deploy countervailing regulatory and fiscal policies until 1990.
Did the Bubble’s Collapse Cause the Lost Decades?
The aftermath of the bubble proved extraordinarily painful for Japan. But the collapse of a bubble does not inevitably have such powerful and long-lasting effects. What was special about Japan’s case? A key factor was the buildup of considerable leverage in the financial system, similar to what occurred in the United States before 2008. Tier 1 capital of Japanese banks in the 1980s was very low, much lower than elsewhere, as global standards (the Basel I accord) had not yet gone into effect. Moreover, much of the collateral for loans was in the form of real estate, whereas under the keiretsu system a significant portion of bank assets consisted of shares in other firms from the same group. So, when real estate and share prices collapsed, the banking system was badly damaged.
This underlying vulnerability was exacerbated by a slow policy response. The authorities delayed forcing banks to recognize the losses on their balance sheets and allowed them to continue lending to firms that had themselves become insolvent, a process Caballero, Hoshi, and Kashyap (2008) call “zombie lending.” This process continued into the early 2000s, stifling productivity growth and prolonging Japan’s slump. Why did the authorities not force faster restructuring? Possibly because restructuring would have required additional bank capital, which they were not in a position to provide in light of the strong political backlash after an initial injection of public capital in 1995. Consequently, the authorities exercised forbearance instead.
The postbubble slump may also have been exacerbated by the macroeconomic policy response and adverse external shocks. Some argue that premature monetary tightening and the lack of a clear commitment to raising inflation led to unduly high real interest rates (Ito and Mishkin, 2006; Leigh, 2010). In addition, the tightening of fiscal policy in 1997 may have undercut the nascent 1995–96 recovery (Posen, 2003; Corbett and Ito, 2010). Finally, adverse external shocks played a role, including the 1997–98 Asian financial crisis.
In sum, Japan’s experience shows that currency appreciation does not, in fact, inevitably lead to “lost decades.” The appreciation did not inevitably require such a large macroeconomic stimulus. The stimulus did not inevitably lead to the bubble. Nor did the bubble’s collapse inevitably lead to the Lost Decades. Instead, it was the particular combination of circumstances and choices that led to that result.
Lessons for Rebalancing Today
Calibrating a policy response to exceptionally large appreciations and movements in asset prices remains an extraordinarily difficult task. But some pointers can be gleaned from Japan’s experience. The keys are to
• avoid an excessive macroeconomic response to currency appreciations;• use prudential policies to prevent vulnerabilities from building up, especially in the form of leverage;• address banking problems quickly if they do materialize; and• provide significant macroeconomic support when banking systems and economies come under stress.An even broader lesson is that bubbles can prove dangerous. Accordingly, Japan has introduced a two- perspective framework for monetary policy, with one pillar focusing on price stability and the other looking out for financial imbalances such as asset price bubbles.
But even as Japan’s experience offers lessons to countries considering rebalancing today, the direct parallels are limited. Most notably, circumstances in China today differ from those in Japan in the 1980s in ways that should help it avoid Japan’s disappointing outcomes (Figure 1.4.2). First, the leverage of households, corporations, and the government in China is lower now than it was in Japan before the bubble (N’Diaye, 2010), and the risk of excessive borrowing may thus be smaller. Second, as Chapter 4 of the April 2010 World Economic Outlook and Igan, Fabrizio, and Mody (2007) find, climbing the quality ladder helps offset the impact on growth of currency appreciation, and China has more room to climb the export quality ladder than Japan did. (At the same time, the impact on labor-intensive industries may be greater.) Third, Japan had a floating exchange rate regime in the 1980s, but China has a managed exchange rate supported by vast foreign currency reserves and strong restrictions on capital inflows. This difference in currency regimes should help China avoid the sharp appreciation observed in Japan. Most important, China should be able to reap the benefits of learning from Japan’s experience."
