Japan experienced a boom-and-bust cycle in the real estate and stock markets almost 20 years earlier than Europe. Since the bursting of the Japanese bubble economy, the country has fallen into a deep recession and has experimented with crisis therapies in the form of unconventional monetary expansion, Keynesian fiscal stimulus, and recapitalization of financial institutions. Japan reached a low interest rate environment in the mid 1990s and has accumulated an exceptionally high level of public debt during more than two decades of economic stagnation. This paper compares the boom-and-bust cycles in Japan and Europe with respect to the reasons for excessive booms, the characteristics of the crises, and the (potential) effects of the crisis therapies. It is argued that in both Japan and Europe the consequences of expansionary monetary and fiscal policies include the hysteresis of a low-interest rate and high government debt environment, the erosion of the allocation and signaling functions of the interest rate, the gradual quasi-nationalization of financial institutions, as well as gradual real income losses. The economic policy implication for Europe and Japan is the timely exit from crisis therapies in the form of excessively expansionary monetary and fiscal policies.
While a growing number of European countries are struggling for the consolidation of their general government deficits, the crisis seems to have become the normal course of life. Whereas deep crises in the past were followed by buoyant recoveries, the European financial, debt, and economic crisis seems to have become a chronic disease. Reiterated rescue activities by governments, central banks, and supranational institutions were able to prevent major meltdowns, but could not trigger a sustained European recovery, despite buoyant German economic activity. Will the persistent instability in a rising number of European crisis countries (now including economic giants such as Italy and France) lead to uncontrolled contagion and financial meltdown? Or do we stand at the beginning of a persistent lingering crisis as has been observed in Japan for more than two decades?
Japan not only moved through a boom-and-bust cycle—the so-called bubble economy—almost 20 years earlier than Europe, but also offers an important experience with crisis management in the form of monetary expansion, unconventional monetary policymaking, fiscal expansion, and the recapitalization of banks. Although Japan has reached (close to) a zero interest rate environment more than a decade earlier than Europe and gross general government debt (in terms of GDP) has gone far beyond the levels that are currently prevalent in Europe, growth continues to stagger.
A growing number of papers have analyzed the origins of the current crisis and crisis therapies with a focus on the United States, the UK, China, or Northern Europe (e.g., Bosworth and Flaaen 2009; Ostrup, Oxelheim, and Wihlborg 2009; Woo and Zhang 2011). Other papers have scrutinized Japan's crisis therapies and the consequences for the Japanese financial sector and growth (e.g., Mikitani and Posen 2000; Bayoumi and Collyns 2000; Sekine, Kobayashi, and Saita 2003; Koo 2003; Caballero, Hoshi, und Kashyap 2008; Ueda 2012). The origins of the European crisis have been associated with the introduction of the euro (Berger and Nitsch 2010, diverging competitiveness (Zemanek 2010), speculation inherent to human action (De Grauwe 2011), and uncoordinated fiscal policies (Schnabl and Wollmershäuser 2013). Although Ueda (2012) compares the Japanese post-bubble crisis with the sub-prime crisis in the United States, a comprehensive comparison between the boom-and-bust cycles in Japan and Europe, including an analysis of the effects of Japan's crisis management, has not yet been made.
The comparison between the boom-and-bust cycles in Japan and Europe with respect to the origins of exuberant booms, crisis patterns and therapies, and the (possible) effects of those therapies shows that despite significant differences, important similarities exist. With the growing socialization of risk Europe appears to be following Japanese economic policy, with—possibly—a similar outcome for European growth and welfare. The gradual decline in real income in Japan should be incentive enough for a turnaround in economic policymaking in both Europe and Japan.
2. Monetary cum fiscal expansion as the origin of the crisis
The exuberance on the Japanese stock and real estate markets in the second half of the 1980s originated, inter alia,1 in a combination of expansionary monetary and fiscal policies (Schnabl and Hoffmann 2008). The starting point of the Japanese bubble economy was a political conflict about the U.S.–Japanese trade imbalance, which had emerged after the liberalization of Japanese international capital flows during the first half of the 1980s. The United States pressured Japan to cure the trade imbalance by a strong revaluation of the yen against the dollar (McKinnon and Ohno 1997). With the Plaza Agreement of September 1985, a substantial appreciation of the Japanese yen against the U.S. dollar was announced and underpinned by monetary tightening in Japan (Funabashi 1989). This opened the door for one-way bets on yen appreciation (McKinnon and Ohno 1997), which led to a yen appreciation far beyond the targeted range. The sharp decline of Japanese exports and a high-yen–induced recession triggered attempts by the Bank of Japan to soften the appreciation pressure by cutting the short-term interest rate from roughly 8 percent in 1985 to 3.5 percent in 1987 (Figure 1).
The resulting easing of credit conditions not only faciliated investment in Japanese industry to regain international competitiveness, but also contributed to unsustainable credit growth and an unprecedented stock market and real estate boom. Revankar and Yoshino (2008) provide empirical evidence that the low interest policy of the Bank of Japan during the bubble economy influenced the lending behavior of banks, which provided loans beyond the threshold that would have been justied by profit maximization. They identify competition for market share and rising real estate prices as transmission channels from interest rate cuts to accelerating credit growth. In particular, as real estate was widely used in the banking sector as collateral for credit, rising real estate prices provided an incentive for growing supply and demand in the credit market.
The boom on stock and real estate markets was further fuelled by an expansionary fiscal policy following the Louvre Accord in February 1987. After the yen appreciation had failed to reduce Japan's current account surplus,2 Japan was urged to stimulate the domestic economy by fiscal expansion to reduce the current account surplus via the import channel (Funabashi 1989). Monetary expansion thus became paired with fiscal expansion. Figure 2 shows the Japanese bubble economy from a stock market perspective. The Nikkei 225 hiked from about 15,000 points in January 1985 to just below 40,000 points in December 1989. Additionally, real estate prices, in particular in the metropolitan areas, increased rapidly. In densely populated Japan, real estate became an important form of collateral for speculative stock purchases.
In the European (Monetary) Union the origin of the excessive boom events in the later crisis countries can also be traced, among other things,3 to a policy mix of expansionary fiscal and monetary policies (Schnabl and Wollmershäuser 2013). In contrast to Japan, the trigger of an expansionary monetary policy by the European Cental Bank was not a trade imbalance. The intra–euro area current account positions in Europe were widely contained through to the turn of the millennium (Figure 3). Before, during, and after the speculation boom in Europe the current account position of the euro area as a whole remained roughly balanced. Rising intra-European current account imbalances during the boom period can be seen more as a symptom of the speculation booms, rather than a cause (Schnabl and Wollmershäuser 2013).
Moreover, two years after the introduction of the euro, the European Central Bank sharply cut interest rates in response to the bursting dotcom bubble, from about 5 percent in early 2001 to 2 percent in 2003 (i.e., to a historical low in Europe). Figure 1 provides a comparison of the central bank response to the respective exogeneous shocks, namely, the Plaza Agreement in September 1985 for Japan (and the resulting yen appreciation) (lower x-axis) and the bursting of the dotcom bubble during 2000 for Europe (and the resulting decline in stock prices) (upper x-axis). The interest rate cuts of the European Central Bank started from a lower level and lasted longer than the interest rate cuts of the Bank of Japan, but were less pronounced.
Monetary expansion starting in 2001 was paired with asymmetric fiscal policies stances in different parts of Europe (Schnabl and Wollmershäuser 2013). In Germany—which was facing the long-term consequences of the German unification in the form of low international competitiveness, unprecedented high unemployment, and historically high government debt—sustained efforts were made to consolidate government finance (Schnabl and Zemanek 2011). This public consolidation process was accompanied by a thorough restructuring process in the enterprise sector, which aimed at restoring its international competitiveness and had declined substantially during the unification boom. The German policy mix of expansionary monetary, restrictive fiscal, and restrictive wage policies4 favored capital outflows to the periphery of the European Monetary Union (EMU), other member states of the European Union (Abad et al. 2013), and beyond.
At the European periphery, buoyant capital inflows and historically low interest rates provided an incentive for rising government spending and proliferating wage policies in the public and private sectors (e.g., Greece and Portugal). In other countries buoyant capital inflows from northern European countries with high saving rates, which assumed the de facto role of an expansionary monetary policy, triggered real estate and financial market booms (e.g., Spain, Ireland, and Cyprus). Similar to Japan during the second half of the 1980s, rising government expenditure and the wealth effects of rising asset prices contributed to rising consumption (Abad et al. 2013). Depending on whether the speculation and consumption booms generated government revenues beyond rising government expenditure the debt levels of the later crisis countries increased (e.g., Greece and Portugal) or decreased (e.g., Ireland and Spain).
Figure 2 compares the development of stock prices in response to interest cuts in Japan and Europe. In Japan the interest rate cuts starting in 1986 were associated with sharply increasing stock prices without major lag. In Europe, the stock market response to interest rate cuts starting in 2001 was delayed. After the dotcom shock, stock prices continued to decline until 2003 and rose sharply thereafter. The stock market booms were more pronounced in the later crisis countries (e.g., Greece and Ireland) compared with Germany (Figure 2). German stock prices were dragged down by the consolidation efforts of German enterprises and austerity in public spending, but stimulated by the improving performance of German export enterprises and financial institutions. Country-specific real estate booms emerged in different parts of Europe, being particularly pronounced in Spain and Ireland (but also Central and Eastern Europe), while real estate prices in Germany remained widely flat.5
The current account surplus of Japan declined during the speculation boom on stock and real estate markets in the second half of the 1980s as domestic demand and imports were stimulated by rising government expenditure and the positive wealth effects of rising asset prices, although Japan's current account position returned to its pre-1985 level after the bubble burst in December 1989 (Figure 3). In Europe, different monetary and fiscal policy mixes led to sharply diverging intra-European current account balances starting from 2001 (Figure 3). In the north of the European (Monetary) Union (in particular in Germany) current account balances improved on the back of public austerity, depressed investment, and sluggish growth. In the south, the current account balances of the booming countries of the EMU as well as of many European countries outside EMU (e.g., Iceland, Bulgaria, and the Baltics) turned dramatically negative on the back of overconsumption and/or speculation booms. Despite strongly diverging intra–euro area current account balances, the current account balance of the euro area as a whole remained widely balanced.
3. Trigger of the crisis and crisis therapies
In Japan, the Bank of Japan recognized that the credit-driven stock and real estate boom, which was transmitted with a lag via wealth effects into a consumption boom and rising inflation, was not sustainable. Starting from 1988 the Bank of Japan lifted interest rates to deflate the bubble (Figure 1). Attempts were made to reduce real estate transactions through tighter regulations and higher taxes.6 In December 1989 the stock market bubble burst (Figure 2), and in 1991 the upward trend on real estate markets was reversed.7 The Bank of Japan kept monetary conditions tight until 1991 (Figure 1) to ensure a consolidation of credit to the private sector in the balance sheets of private commercial banks. Both the restrictive monetary policy in response to the bubble as well as the restrictive monetary policy stance during the first phase of the post-bubble recession were ex post seen as monetary policy mistakes (Posen 2000). Bernanke (2000) dubbed the Japanese post-bubble stagnation of the 1990s as “self-induced paralysis”, which could have been avoided by decisive policy action and “unconventional measures” such as outright purchases of government bonds and/or discretionary currency depreciation.
In Europe, the financial, government debt and economic crisis was initiated by monetary tightening starting from end of 2005 (Figure 1), which followed the monetary tightening by the U.S. Federal Reserve. The outbreak of the U.S. sub-prime crisis in August 2007 can be seen as the final trigger to the change in the risk assessment of global financial markets, in particular concerning international lending. The possible emergence of financial market bubbles was not regarded as a motivation for monetary tightening by the European Central Bank—rather, the concern about rising inflation had grown. As in Japan, inflationary pressure had not emerged for a long time (on average) in the EMU. Fiscal and wage austerity in Germany had kept average EMU inflation low, although inflation rates in the later crisis countries had grown substantially beyond the 2 percent benchmark. In contrast to Japan, the European Central Bank did not lift the interest rate up to the pre-boom level (Figure 1).
Also in contrast to Japan, the characteristics of the crisis in Europe are—as the pre-crisis boom—heterogeneous. The countries at the periphery of the EMU (which, similar to Japan in the 1980s, have experienced a pronounced boom-and-bust cycle since the turn of the millennium) are going through severe recessions. Germany, which experienced a severe phase of austerity before the crisis, is blessed by a robust economic performance and a continuing decline in unemployment. Pre-crisis reforms, which have resulted in low unit labor costs and comparatively sound government budget balances, are the basis for Germany's role as the growth engine in Europe.8
3.1 Interest rate cuts and unconventional monetary policy
In the early years of the crisis the Bank of Japan cut interest rates hesitantly. When the crisis persisted, interst rates were cut more decisively and the overnight money market rate reached a level of 0.5 percent in the mid 1990s (Figure 1) during a period of painful appreciation pressure on the Japanese yen. Money market interest rates converged toward zero during the Japanese financial crisis in 1998–99, about ten years after the bursting of the bubble on the stock market. Because the zero interest rate policy could not reanimate the Japanese economy, unconventional monetary policy measures supported fragile financial institutions, weak small and medium enterprises, as well as large export enterprises by preventing further yen appreciation.
In the late 1990s, the Bank of Japan pathed the way into “unconventional monetary policy,” which went far beyond the textbook monetary policy pattern (Posen 2000; Iwata and Takenaka 2011; Ueda 2012). The Bank of Japan balance sheet was gradually inflated by purchases of stocks, bonds, and government bonds (Bernanke 2000). Figure 4 shows as a proxy for the scope of unconventional monetary policy measures the Bank of Japan's monetary base as a share of GDP, which increased from about 10 percent in 1989 to about 25 percent in 2012. In periods of economic recovery the exit from this expansionary monetary policy was occasionally attempted, but interrupted because of re-emerging financial fragility and re-emerging appreciation pressure on the Japanese yen. Figure 5 shows how in 2000 and 2006–08 short-term interest rates were temporarily lifted and then lowered again. The temporary increases in short-term interest rates are reflected in respective declines in the monetary base.
In Europe, despite objections from Germany (where high private savings of the household sector underpin a widespread concern about inflation), the European Central Bank cut interest rates faster toward zero and then commenced with unconventional monetary policy measures much earlier than the Bank of Japan had in response to the bursting bubble. Similar to the period prior to the outbreak of the crisis, one-size monetary policy clearly does not fit all. For the European crisis countries the harsh monetary expansion seems appropriate as deeper financial and economic meltdowns are prevented. For Germany, as in Japan during the second half of the 1980s, the current monetary policy stance seems too loose, triggering a real estate and stock market bubble, with rapidly rising real estate prices in metropolitan areas and stock prices that have already reached historical hights (Figure 2).
Despite rising government debt to unprecedented high levels, risk premiums on Japanese government bonds never became visible. If short-term interest rates are depressed close to zero indefinitely, this pulls down long-term interest rates, because long-term interest rates are equivalent to expected future short-term interest rates plus a liquidity premium. The clearly intertwined downward trend of short- and long-term rates in Japan toward zero is shown in Figure 5.9 This implies that, linked to the monetary expansion in the form of interest rate cuts and unconventional monetary policy measures (including government bond purchases), the Bank of Japan provided an implicit insurance mechanism against outright government default.
In contrast to Japan, at the beginning of the crisis the European crisis countries could not hope that the monetary policy of the European Central Bank would be fitted to their country-specific needs. This led to dramatic hikes in the risk premiums on the crisis countries’ government bonds.10 Only the gradual implementation of rescue mechanisms by the EU and IMF as well as outright purchases of crisis countries’ government bonds in the secondary sovereign bond market by the European Central Bank—all of which can be seen as a kind of country-specific monetary policy stance—initiated the reduction of interest rate spreads versus German government bonds.
3.2 Recapitalization of banks
Ueda (2012) sees the retarded recapitalization of Japanese banks as the main reason for the delayed recovery from the slump. The recapitalization of Japanese banks was postponed until the late 1990s because of strong opposition from the manufacturing sector and unfavorable public opinion against the injection of public money into speculative banks. Recapitalization was only initiated after the Japanese financial crisis had substantially worsened the bad loan problem.11 Before, the government had urged stronger banks to take over non-profitable banks in what was dubbed the “convoy approach” (Shimizu 2000). In 1998, the Financial Reconstruction Law was put in place to oblige banks to publish detailed information on their bad loans. The new law also allowed for the nationalization of banks. In the same year, the Early Strengthening Law, which allowed bank recapitalization, came into force. In March 1999 all big banks were recapitalized with an amount equivalent to 80 billion euros. Through to March 2001, 169 credit institutions were nationalized with a sum equivalent to 170 billion euros.
The notion that the delayed recapitalization of commercial banks led to an unnecessary prolangation of the Japanese crisis contributed to the faster recapitalization of European banks in distress to prevent a credit crunch resulting from falling asset prices and shrinking equity ratios. In Germany, at the beginning of the crisis, takeovers took place in the form of a Japanese-type convoy system (IKB, SachsenLB). Then, commercial banks were recapitalized (Commerzbank), nationalized (Hypo Real Estate), or closed down (WestLB). In contrast to Japan, where bancruptcies of financial institutions such as Sanyo Securities, Yamaichi Shoken, and Long-Term Credit Bank were tolerated, this was not the case in Europe. In particular, the Emergency Liqudity Assistance of the national central banks provided liquidity for eased collateral conditions to commercial banks in distress in the European crisis countries.
3.3 Zombie lending and evergreening
Caballero, Hoshi, and Kashyap (2008) pioneered the notion of “zombie institutions” that depend on public subsidies (i.e., the steady provision of low-cost liquidity by the central bank). The profitability of banks becomes contingent on the perpetuation of zero interest rate policies (zombie banks). Caballero, Hoshi, and Kashyab (2008, 1) stress the detrimental impact of zombie lending on the enterprise sector: “We confirm that zombie dominated industries exhibit more depressed job creation and destruction, and lower productivity.” With the cost of money being (close to) zero, banks became tempted to roll over loans to enterprises to avoid bad loans from becoming visible, although the enterprises were no longer profitable.
Sekina, Kobayashi, and Saita (2003) model “forbearence lending”: Credit provision to the private sector becomes delinked from the profitability of investment projects, rather becoming path-dependent on previous credit provision. This implies that the low profitability of the banking sector is rolled over to the enterprise sector, with the marginal efficiency of investment declining as a result. Peek and Rosengren (2005) identify a misallocation of capital in the credit sector, which keeps enterprises with gloomy business perspectives alive (“evergreening”).12 Since 2012, a Law on Comprehensive Measures to Facilitate the Financing for Small and Medium-Sized Enterprises13 forces banks to continue credit provision to small and medium enterprises at low rates of interest.
3.4 Keynesian spending programs
After the bursting of the Japanese bubble economy, enterprises reduced investment due to the negative wealth effect of declining asset prices. The public sector smoothed out the resulting negative business cycle effect by providing stimulus via credit-financed public spending. The Japanese government implemented several Keynesian spending programs to stabilize the Japanese economy, and in particular outside the economic centers of Tokyo and Aichi. As the public spending programs were focused on public infrastructure projects such as highways, bridges, bullet-train tracks, and public buildings, the construction industry experienced an extraordinary stimulus (Yoshino and Mizoguchi 2010). Given the continuous growth of public spending, the marginal efficiency of public investment can be assumed to have decreased over time. Driven by the costs of public spending programs, losses of tax revenues due to the sustained recession and the costs of commercial bank recapitalization, the gross public debt of Japan climbed to a historical high of 235 percent of GDP in 2012 (Figure 6).14
While the credit of the banking sector to enterprises and households remained sluggish—linked to Keynesian public economic stimulus programs and declining tax revenues—the public demand for credit increased (Yoshino 2013). As a result, in the aggregated balance sheet of the private banking sector credit to the private sector continues to be stepwise-substituted by credit to the public sector. As shown in Figure 7, between 2001 and 2012 claims on the private sector as a share of overall assets declined from about 77 percent to 65 percent. During the same time period claims on the public sector increased from about 23 percent to about 34 percent.
In Europe, there are restrictions to Keynesian spending programs, as govenrment debt in most EU member states has climbed to historically high levels, far above the limits of the tightened stability and growth pact. This shifts the responsibility for macroeconomic stabilization back to structural reforms or—alternatively—to the European Central Bank, which is urged into easier monetary conditions to keep government debt under control. In the euro area the monetary base as a share of GDP increased more rapidly than in Japan, although the degree of base money expansion is still lower (Figure 4). Figure 6 shows a substantial increase in the public debt of European countries since the start of the crisis. In Germany, where the economic performance is sound and where (linked to the low interest rate policy of the European Central Bank and the function of Germany as a safe haven in Europe) interest payments on government debt have further declined, public debt remains contained. Yet the rising liabilities for the rescue measures of a rising number of crisis countries have increased the implicit liabilities for the German government.
Thus, in the EMU the effectiveness of any attempts to limit government debt might be undermined as large-scale rescue packages and risks accumulated in the balance sheet of the European Central Bank have inflated the implicit government debt of the rescuing countries far beyond the institutional limits of the Maastricht Treaty. In addition, the low interest rates guaranteed by the European Central Bank provide an incentive to further increase government debt. Nevertheless, the framework of the EMU to control government debt based on institutionalized rules serves as a role model for Japan in its attempts to keep public debt under control (Yoshino 2011).
4. Economic consequences of Japan's crisis therapies
The economic consequences of Japan's crisis therapies are linked to the unintended side effects of rescue measures beyond financial market stabilization, namely, the collapse of financial intermediation, hysteresis of the low-interest rate trap, the erosion of the allocation and signaling functions of the interst rate, as well as the creeping quasi-nationalization of aggregate demand.
4.1 Unintended side effects of rescue measures
The gradual interest cuts of the Bank of Japan since the early 1990s (Figure 1) were linked to continuous net capital outflows (including carry trades).15 In the first half of the 1990s Japanese banks could compensate for their losses from the bursting Japanese bubble economy by refinancing at cheap rates in Japan and expanding credit to the fast-growing Southeast Asian economies. Japanese export enterprises participated in the Southeast Asian boom via rising demand from these countries and by building up subsidiaries with low labor costs. When the Southeast Asian boom turned sour in 1997, both Japanese banks and export enterprises suffered new losses, which became reflected in a significant decline of prices on the Tokyo stock exchange, the Japanese financial crisis, and the collapse of large financial institutions. Since then, foreign financial market shocks have had negative feedback effects on Japans’ economic recovery, contributing to the continuous decline of stock prices as shown in Figure 2.
Similar negative feedback effects on the private sector are less likely in Europe, as the assets of German and other European banks were to a substantial degree nationalized in the course of the rescue measures. On the other hand, while the current account deficits of the crisis countries are stepwise reduced toward zero on the back of extensive austerity programs, the German current account surplus persists (Figure 3). This implies that German net capital exports are redirected to third countries (in particular France and countries outside the euro area) where new risks might be accumulated, similar to Japanese banks during the Southeast Asian boom.
The high and (based on sustained current account surpluses) ever-growing net international assets of the Japanese private sector made enterprises, households, and financial institutions vulnerable for the appreication of the Japanese yen. As Japan's international assets are denominated in foreign currency (mostly U.S. dollars), an appreciation of the Japanese yen against the U.S. dollar not only slows down Japanese exports, which remain the most important pillar of growth, but also reduces net international assets in terms of domestic currency (McKinnon and Schnabl 2004). The outcome is a negative balance sheet effect, to which financial institutions may respond by reducing credit to the private sector. For this reason exchange rate stability against the dollar became a pivotal pillar for financial and macroeconomic stability in Japan (Goyal and McKinnon 2003). This contributes to the persistence of a zero interest rate policy and quantitative easing, with both monetary policy measures aiming to dampen the appreciation pressure on the Japanse currency.
For Europe, the net international asset position is, in contrast to Japan, of minor concern for two reasons. First, the current account position and therefore the net international investment position of the EMU as a whole is widely balanced. From this perspective the European crisis is more a question of intra-European redistribution. Second, the foreign assets and/or liabilities of euro area countries are to a larger extent denominated in euros (i.e., in domestic currency). This makes the financial sector less vulnerable to euro appreciation in the face of monetary expansion in the United States. Nevertheless, as the appreciation of the euro against the dollar is a major concern for the export sectors of the crisis countries, their competitiveness can be seen as a major determinant of the European Central Banks’ interest rate decisions.
4.2 The collapse of financial intermediation
Both in Japan and the EMU low interest rate policies provoked the collapse of money markets with trading volumes declining dramatically. With the rising perceived risk of interbank lending, liquidity provisions among commercial banks were substituted for the liquidity provision of central banks to commercial banks. McKinnon (2012) explains how zero interest rate policies undermine financial intermediation on money markets. With yields on money markets having declined to low levels, potential lenders are disinclined to lend. Finanical institutions with excess cash prefer to deposit their excess cash at the central bank. If any financial institution would bid for funds at a substantially higher rate than the prevailing money market rate this would signal high risk, impeding it even more from raising liquidity. To sustain credit to the private sector, the central bank has to provide a substitute for private money market lending.
Furthermore, the zero interest rate policy contributed to a decline of bank-based lending to the private sector (credit crunch) since the Japanese financial crisis (Figure 8).16 It has been argued that the Japanese credit crunch has been driven either by the necessary consolidation of commercial bank balance sheets or by the shrinking credit demand of the enterprise sector (Ishikawa und Tsutsui 2005). Proponents of a supply-driven credit crunch hypothesis see asset price deflation as the origin of shrinking credit to the private sector (Posen 2000). As asset prices fall, the equity of financial institutions falls with them, forcing financial institutions not to renew outstanding credit to enterprises. Bankruptcies in the enterprise sector cause new bad loans for financial institutions, which have to further tighten their credit conditions.17 Given this process, monetary expansion and recapitalization of banks was seen as an effetive remedy against this supply driven credit crunch (Posen 2000; Bernanke 2000).18
The demand-side credit crunch hypothesis stresses the negative wealth effect of asset price deflation. As enterprises suffer from revaluation losses on their assets they reduce investment to consolidate their balance sheets (Koo 2003). The credit opportunities of financial institutions deteriorate. This forces them to further tighten credit conditions and/or to sell stocks and real estate, which leads to a further decline of asset prices. Furthermore, the substantial decline in interest rates improved the financing conditions for large enterprises, because better financing conditions for enterprises increased aggregated enterprise saving.
Figure 9 shows how the decline in interest rates after the bursting of the bubble economy went along with the transformation of the non-financial corporations sector from a net debtor into a net saver. The financial surplus (net saving) of non-financial corporations increased from about –10 percent of GDP in 1989 to +5 percent in 2011.19 This allowed enterprises to close credit lines with commercial banks.20 In addition, large non-financial corporations tended to finance directly via issuing commercial bills to savers. This left (in particular smaller) banks with a riskier portfolio of loans to small and medium enterprises (which could not enter the capital market) facing a higher capital constraint. Banks have to hold more capital for a rising average default risk and/or to reduce the outstanding amount of credit to small and medium enterprises.21
The zero interest rate policy also created a drag on the reanimation of the banking sector as it compressed the deposit-lending spread. Commercial banks usually cover their costs and generate profits by accepting deposits (at short maturities) at low rates and providing credit (at longer maturities) at higher rates. The deposit-lending spread is the basis for a profitable banking business engaging in maturity transformation. After the bursting of the Japanese bubble economy high deposit–lending spreads would have been the basis for the solution of the non-performing loan problem (Goyal and McKinnon 2003). The low-interest rate policy of the Bank of Japan, however, compressed the lending–deposit spread, thereby impeding the Japanese banking sector from cleaning up their balance sheets. As shown in Figure 10, the non-risk–adjusted deposit–lending spread was on average above three percentage points throughout the 1980s. With the advent of the zero interest rate policy and quantitative easing since the early 1990s, the lending–deposit spread declined toward 0.5 percent, thereby marginalizing the traditional banking business.
The credit crunch that has been observed in Japan is in Europe limited to the crisis countries suffering from capital flight. It became virulent as banks from non-crisis countries started to close their credit lines. The deflationary effects of tightening credit conditions became attenuated by credit provision at eased collateral conditions by the national central banks in the form of Emergency Liquidity Assistance. The national central banks in crisis countries were refinanced by the European System of Central Banks via the TARGET2 payment system, which took over a form of unlimited supranational liquidity/credit provision (Sinn and Wollmershäuser 2012). In Germany, sufficient liquidity is available in the commerical bank sector, which continues to reduce credit lines to the crisis countries. Given rising deposits of German commercial banks at the central bank, the Deutsche Bundesbank is being transformed from a central bank that supplies credit to the domestic banking system into one that absorbs liquidity from the domestic banking system.
European banks were recapitalized faster than Japanese banks to prevent a balance sheet recession. Polleit (2011) provides an explanation as to why the recapitalization of banks is not a sufficient condition for a sustained economic recovery, however. Because the governments have no liquidity buffers available, they have to raise the funds needed for recapitalization by issuing more government debt. The recapitalization of commercial banks allows banks to expand their credit supply, which will, however, be absorbed by the additional credit demand of the government sector. As shown in Figure 7, the recapitalization of Japanese banks at the turn of the millennium was accompanied by a decrease in credit to the private sector.22
Furthermore, zero interest rates per se constitute an impediment for a positive effect of the recapitalization of banks on the credit provision to private enterprises. With interest rates at zero the banks are missing an appropriate benchmark at which an investment project can be assessed to be profitable. From this point of view the recapitalization of banks is a necessary but not sufficient condition for the reanimation of credit to the private sector. The recapitalization of banks would have to be paired with a reconstitution of the allocation function of the interest rate (see the following). Therefore, the Japanese credit crunch has to be seen as a crowding out of credit to the private sector by credit to the public sector, which is further enhanced by the Basel requirements on equity ratios and the implicit guarantee on government debt provided by zero interest rate policies.
As in Japan, assessments concerning bad assets in the European banking system (so-called stress tests) are sugarcoated by the expansionary monetary policy in the form of historically low interest rate levels, direct emergency liquidity assistance, and outright purchases of crisis countries’ government bonds. Therefore, as in Japan, in most European countries the number of zombie banks, which have to rely on low-cost liquidity injections of the central bank, is likely to have significantly increased. In addition, the notion of “zombie states” emerges—that is, countries that have to rely on supra-national liquidity provisions (European Financial Stability Facility, European Stability Mechanism, European System of Central Banks, etc.)—to safeguard the stability of national financial sectors and government finance. Would the European Central Bank lift the interest rate to a level of, say, 3 percent (which is still low from a historical perspective), to countersteer a real estate and stock market bubble in Germany, the stress in the European banking systems and crisis countries (as well as in current non-crisis countries) would grow substantially.
4.3 Hysteresis of the low interest rate trap
A second stylized fact of Japanese macroeconomic crisis management is the interdependence of declining interest rates and rising government debt in the form of lock-in effects of interest rate cuts (i.e. the hysteresis of the low-interest rate trap [Schnabl 2012]). Figure 5 shows the gradual decline of 10-year government bonds yields, which provided an incentive for the level of government debt to be increased without a higher burden of debt service (e.g., measured as a share of government expenditure).
Once a low interest rate level and a high government debt level are reached, short-term interest rates have to stay (close to) zero, and unconventional monetary policy has to further depress the interest rate at the long end of the yield curve. Otherwise, public finance would be destabilized. Figure 11 shows a simulation of interest rate payments as a share of the Japanese central government budget dependent on varying interest rates on Japanese government bonds and varying debt levels. Currently, gross Japanese government debt is about 235 percent of GDP and the interest rate on outstanding 10-year government bonds is assumed to be on average at around 2 percent. Interest rate payments as a share of central government expenditure are about 20 percent. If the average yield on Japanese government bonds increased to 4 percent, the interest rate payments as a share of the central government budget would increase to 40 percent and the Japanese government would become financially restricted. At an interest rate level on 10-year government bonds of 7 percent, about 70 percent of the central government budget would have to be spent on debt service. The higher the debt level, the more this effect is reinforced.
The simulation presented in Figure 11 does not include second-round effects. For instance, interest rate increases by the Bank of Japan would be followed by negative business-cycle effects, and therefore declining tax revenues and rising instability of the financial sector. If financial institutions were to default, the costs of recapitalization would additionally contribute to rising government debt. Furthermore, if interest rates were lifted, the Bank of Japan itself would suffer from substantial sterilization costs of liquidity in the domestic banking system and from significant revaluation losses on bond holdings in its balance sheet. As the Bank of Japan would become dependent on recapitalization by the government, it seems to have lost its independence.23 For this reason the political pressure on the Bank of Japan to further depress interest rates is strong. This limited degree of monetary policy independence has become most visible under Prime Minister Abe and the new Bank of Japan President Kuroda.
4.4 Erosion of interest rate functions and the nationalization of demand
Hayek (1929, 1944) has stressed the pivotal role of the interest rate for stable economic development. If the prime interest rate is kept at or close to zero for too long, not only are new boom-and-bust cycles in financial markets (domestic and/or foreign) encouraged, but also the allocation function of the interest rate, which separates between investment projects with high and low expected returns, is destroyed. Instead, the credit allocation works through zombie banks with path dependence seeming to be the most important determinant of credit allocation. Structural distortions, which have emerged during the boom, are preserved. Furthermore, the interest rate has ceded to signal default risk, as the central bank implicitly gurantees government debt as well as the debt of private financial institutions and enterprises.
Although the expansionary monetary and fiscal policy in response to the lasting Japanese crisis has prevented an even deeper crisis, it has come along with the gradual decline of private investment both in absolute terms and as a share of GDP. Private investment in Japan has been gradually substituted by public spending. Figure 12 shows Japanese GDP by expenditure. The share of gross capital formation out of GDP was 32 percent in the 1990, and has declined to 20 percent in the year 2011. During the same time period the share of government spending increased from 13 percent to 21 percent of GDP.
The projection of Bernanke (2000) that interest rate cuts and Keynesian economic stimulus would revive Japanese growth has not been fulfilled. As shown in Figure 8, the real growth rate in Japan has fallen together with the interest rate. This can be interpreted as a gradual decline of the marginal efficiency of investment as the interest rate benchmark for investment projects has gradually declined. Whereas real growth rates since 1990 still add up to a yearly average of 1 percent, the stock price index as indicator for the performance of the Japanese economy has declined to 14,000 points24 even below the level before the start of the Japanese bubble economy in 1985 (Figure 2). The growth perspective remains gloomy because the United States and Europe as important target regions of Japanese exports have gone through similar boom-and-bust-cycles and are facing declining growth momentum. And Chinese growth, another important source of demand for Japanese products, is slowing down as well.
4.5 The wealth effects of quasi-nationalization
Although Abe has launched a comprehensive program of new fiscal and monetary stimulus in combination with real wage austerity and structural reforms, that program's long-term success is uncertain. The reason is that at the core of persistent stagnation the quasi-nationalized finanical sector allocates capital—similarly to former planning economies—based on political criteria rather than on the principles of allocation efficiency. Kornai (1993) coined the concept of “soft budget constraints” for enterprises in the Central and Eastern European planning economies. In Japan, financial institutions and enterprises have become reliant on soft budget constraints in the form of quasi-unlimited liquidity injections via unconventional monetary policy. As in planned economies, the borders between the central bank, commercial banks, enterprises, and government have become blurred. If the Bank of Japan were to lift the interest rate to a level that would eliminate the structurcal distortions originating in the low interest policies, the financial sector in Japan would be financially destabilized. The government would have to further increase debt, because substantial parts of the financial sector would have to be recapitalized or nationalized. The already advanced quasi-nationalization of the Japanese financial sector, which extends to the whole economy, would proceed even further.
The price of the creeping quasi-nationalization of the Japanese finanical and enterprise sectors has become obvious not only in the form of a stagnating real GDP (Figure 8), but in particular in form of shrinking real personal incomes. Whereas during the bubble economy the incomes from wages and property increased, the expansionary macroeconomic policies during the post-bubble crisis could not prevent the decline of the real incomes of Japanese households (Figure 13). Real wages have been gradually declining since the Japanese financial crisis in 1998, on average by 1 percent per year. The incomes from property (interest, dividends, insurances, rents) have fallen since the bursting of the Japanese bubble even more dramatically, mainly driven by the zero-interest rate policy of the Bank of Japan. Property incomes are about 70 percent lower than in December 1989 (when the bubble burst) and more than 40 percent lower than in 1985 (when the bubble started).
5. Conclusions: Hayek's lessons for Japan and Europe
It seems that Europe has taken a similar path to Japan in economic policy patterns in response to the crisis. Yet, as Hayek (1944, 86) put it: “Any attempt to control prices or quantities of particular commoditites deprives competition of its power of bringing about an effective coordination of individual efforts, because price changes then cease to register all relevant changes in circumstances and no longer provide a reliable guide for the individual's action.” Hayek (1944, 103) further stressed: “Although the state controls directly the use of only a large part of the available resources, the effects of its decisions on the remaining part of the economic system become so great that indirectly it controls almost everything.” What Hayek fitted back in the 1940s for the means of industrial productions in Germany and the Central and Eastern European planning economies seems to apply today to the Japanese and (increasingly) the European financial sectors. The control of interest rates and credit allocation via unconventional monetary policy has triggered in Japan a quasi-nationalization of the financial and enterprise sectors and therefore the creeping advent of a quasi-planned economy in form of pervasive soft budget constraints. The outcome seems to be a persistent recession. In Europe, in response to the European debt, financial, and economic crisis, a similar path is being taken.
The longer the crises smolder, the more monetary policy and fiscal policy are used as tools to preserve financial and economic stability. Expansionary monetary policy in the face of any future crisis can be regarded as an indispensable tool to stabilize finanical markets and employment. Yet this type of economic stimulus can only help to stabilize employment in the long term if it works symmetrically. Fiscal and monetary expansions during recessions have to be matched by symmetric fiscal and monetary tightening during the recovery after the slump. Otherwise, given the asymmetric realization of macroeconomic policies, the gradual decline of interest rates and the gradual increase of government debt lead into a quasi-nationalization of the financial sector, the evergreening of bad loans, a systematic substitution of private investment by public demand, a declining marginal efficiency of investment, the cementation of structural distortions, and the erosion of real income and wealth. The negative impact of this process on political stability will be amplified by rising income inequality, which is a globally visible unintended side effect of excessive monetary expansion.
If Keynesian macroeconomic stimulus was not able to reanimate the Japanese growth during the last two decades, in the spirit of Hayek (1929, 1944), the inversed process (i.e., gradual macroeconomic consolidation) may well be the appropriate strategy for a sustained economic recovery. If in the long term the extensive macroeconomic expansion of the past two decades has led to declining growth, a gradual monetary tightening should lead to the revival of Japan's and Europe's economic dynamics, despite deflationary effects in the short term. The marginal efficiency of investment would increase and rising shares of economic activity would be reallocated to the private sector and open markets.
As gradual asymmetric interest rate cuts have in the past favored speculation and crisis, a gradual increase of interest rates would reduce speculative activities and contribute to a higher marginal efficiency of investment and steadier economic devolopment. The consolidation of government debt would free resources for private investment and would allow the return to an independent, stability oriented monetary policy both in Japan and Europe to pave the way towards a sustained recovery.
I acknowledge Osamu Ishikawa, Raphael Fischer, and Stefan Angrick for valuable support. I thank Naoyoki Yoshino, Zhichao Zhang, the participants of the Asian Economic Panel in Dublin, the participants of the Workshop on Financial Research of Hitotsubashi University, as well as the participants of the Global Economy Round Table of the European Central Bank for useful comments.
The literature further identifies the liberalization of capital markets and the resulting erosion of bank-based lending as an important determinant of the bubble economy. (See also footnote 3.)
Further possible reasons for the speculation booms and rising intra-European current account imbalances were provided in Section 1. As in Japan, the liberalization of financial markets is assumed to have played an important role. In this context, however, we will scrutinize to what extent the liberalization process of financial markets has been driven by monetary expansion, as rising money supply increases the need for investment opportunities, which constitutes a push factor for lobbying pressure in favor of financial market liberalization.
Wage austerity in the public sector was accompanied by wage austerity in the private sector, thereby implying an overall restrictive wage policy for Germany.
Germany had experienced a real estate boom in Eastern Germany during the unification boom. The bursting of the Eastern German real estate bubble led to a significant decline of Eastern German real estate prices.
Currently, the Chinese government tries to cool down the real estate boom in China by similar measures.
Since then real estate prices have continued to fall below even the pre-1985 level.
The extensive restructuring processes in the public and the private sectors during the second half of the 1990s and after the turn of the millennium can be also seen as the main driving force of fast-rising German net capital exports and current account surpluses after 2001 (Abad et al. 2013).
Note that this downward trend of short-term and long-term interest rates is also clearly correlated with the downward trend of U.S. short-term and long-term interest rates.
Furthermore, Japanese government debt is held primarily by domestic agents, whereas the government debt of the European crisis countries is to a large extend held by foreign or supranational agents.
Up to the 1997–98 Asian financial crisis Japanese banks could (partially) compensate their losses from domestic asset price deflation by expanding international credit, in particular to the booming Southeast Asian countries such as Indonesia, Malaysia, the Philippines, South Korea, and Thailand. When the Southeast Asian economic miracle came to an end in 1997 with the outbreak of the Asian financial crisis, Japanese financial institutions realized new losses, which triggered the Japanese financial crisis.
In practice, the insolvency law did not work, as large banks were considered to be too big to fail. The de facto missing legal framework became an impediment for market-based solutions of the financial crisis. The reforms of the Koizumi government of the insolvency and labor laws were only partially successful and were reversed in 2009 when the Democratic Party of Japan took over the government. The reform of the insolvency and labor laws are an important challenge for the new Abe government.
The law is seen as the result of the lost elections of the Prime Minister Koizumi in 2009, which can be interpreted as the voters’ response to Koizumi's reform efforts. As the law did not lead to the expected results, it is to be abolished under the Abe government, even though small and medium enterprises provide political support for the Liberal Democratic Party.
Note that net public debt, which accounts for the large foreign assets held by the Japanese Ministry of Finance and is linked to massive foreign exchange intervention against yen appreciation is substantially lower, at about 140 percent of GDP.
Capital could be raised at low interest rates in Japan and could be invested in countries with higher interest rate levels (often emerging market economies such as New Zealand or Russia). The returns from positive interest rate spreads were threatened to turn negative by potential yen appreciation versus the investment currencies. This risk was mediated by foreign exchange intervention against yen apprecation as well as Bank of Japan interest rate cuts and quantitative easing, which softened yen appreciation pressure. This made carry trades a lucrative source of income.
Woo and Zhang (2011) find a different impact of very expansionary monetary policies on credit growth in the United States and the UK in contrast to China.
Fisher (1933) formed a debt deflation theory for the Great Depression with respect to the impact of mild deflation on real debt. In contrast to the Great Depression, however, a sustained process of consumer price deflation did not set in in Japan.
In effect the money multiplier defined as the ratio between a broader measure of money supply (for instance M2) and the monetary base declined (Ueda 2012, 193–196).
During the same time period household saving declined from about 10 percent of GDP in 1990 to about 4.5 percent of GDP in 2011.
Some large enterprises in Germany such as Siemens and BMW have direct access to low-cost central bank financing via own financial subsidiaries. This is a significant competitive advantage versus smaller enterprises, which have to rely on financing by the banking sector.
The credit provision of commercial banks to small and medium enterprises declined from 347 trillion yen in 1993 to 245 trillion yen in 2011.
If, alternatively, the bonds issued by the government to recapitalize commercial banks are bought by the non-bank private sector, the money supply shrinks. In the aggregated balance sheet of the commercial banks private deposits on the liability side of the balance sheet are substituted by rising capital. This is followed by recession and declining prices, which is likely to trigger an additional monetary expansion by the central bank. Government bond purchases of the central bank will put the money supply back to the original level.
Carpenter et al. (2012) provide a detailed insight into the impact of an exit from low interest rate policy and quantitative easing on the equity of the Federal Reserve Bank.
Note that Japanese stock prices have been substantially inflated after the recent announcement of excessive monetary expansion by the Bank of Japan in the context of the so-called Abenomics.