Lessons from Japan’s Economic Experience
Click here for a printer-friendly version of this article
Last week, the Wall Street Journal featured a long article about how the Federal Reserve is attempting to apply the lessons learned from Japan’s ongoing struggle to recover from its real estate and stock bubble collapse, which dates back to the 1990s. The reporter writes: “Tokyo's economic problems are more than academic for the Federal Reserve chairman. They are a window into his own situation as he stares at what could be a long period of slow growth, high unemployment and declining inflation in the U.S.” Donald Kohn, who stepped down recently as the Vice Chairman of the Fed, has commented repeatedly that he took away two lessons from Japan: 1) be very aggressive about stimulating the economy early and 2) don’t remove or wind down those stimulus efforts too soon when things might appear to be stabilizing.
In reviewing the minutes from the Fed’s last meeting (in September), it is now clear that many Fed officials agree that if growth doesn’t start accelerating (and inflation remains modest to non-existent) then “it would be appropriate to provide additional monetary policy accommodation." The leading option or lever for the Fed, appears to be restarting a program that would have the Fed buy more of its own debt to try and drive down long-term interest rates. The goal would be to boost growth, but there are clear downsides, such as indirectly calling for more inflation and growing the Fed’s balance sheet beyond its already record level of $2.3 trillion. This is the backdrop to policy discussions today about Japan’s history and the lessons for the U.S. Yet, it’s wise to dive a bit deeper into the fundamentals of the two economies – both as they entered their respective recessions and as they plotted their recovery strategies – before jumping to conclusions.
Over the past 15 years, the U.S. economy has absorbed the rise and fall of two huge asset price bubbles: the Internet or “dot-com” stock market bubble of 1997-2000 and the housing bubble of 2003-2007. In both cases, some expressed fear in the aftermath of the bubbles’ collapse that the U.S. would enter a lengthy period of malaise similar to that experienced by Japan after its real estate and stock bubbles burst at the start of the 1990s. Again, in both cases, those voicing concerns tended to overlook key differences in U.S. circumstances that made this outcome improbable. While the collapse of the more recent U.S. housing bubble was far closer to the Japanese experience for reasons discussed below, the fundamental health of the U.S. corporate sector and the large U.S. capital account surplus undercuts any direct comparison.
More consequentially, the worldview that suggests the Japanese policy response to its crisis was right directionally but insufficient in magnitude could badly impair the nascent recovery if applied to policymaking in the U.S. Large scale fiscal consolidation and policies that promote investment in the tradable goods sector are more likely to aid U.S. economic growth going forward. If the supposed lessons from Japan are that the authorities need to run larger fiscal deficits or funnel more societal resources into low value-added infrastructure projects, they would be best to go unlearned.
The most popular explanation for Japan’s nearly 20 year slump is that the country experienced a “balance sheet recession” where the collapse of an asset price bubble leaves a large number of balance sheets with more liabilities than assets. In Japan, the sharp decline in land and stock prices left businesses technically insolvent – the market value of assets fell while nominally fixed liabilities remained unchanged, causing net worth to turn negative. While Japan could have resolved the problem by pushing insolvent businesses into bankruptcy, forcing banks to recognize losses on loans made to these businesses would have left the entire banking system insolvent. Fearing this would have deepened the recession and ultimate resolution costs, the banks were allowed to hide losses on bad loans. The companies that owed the banks money continued in operation without being forced to restructure obligations. The result was a combination of zombie borrowers and zombie lenders that, in the telling of Richard Koo, pursued debt minimization instead of profit maximization.
When the private sector seeks to reduce debt at all costs, normal monetary policy fails to function. The goal of businesses is to reduce the unpaid principal balance on outstanding debt below the market value of assets, not refinance the unsustainable balance at a lower rate. Thus, the Bank of Japan was unable to stimulate lending even at zero interest rates. Some argue this was because of the capital conservation policies of the insolvent banks; others contend it was because of the low demand for loans due to a dearth of unencumbered collateral to be pledged for new loans (insolvent borrowers). Either way, Japan’s banks and businesses entered a prolonged period of deflation brought about by “negative borrowing” – cumulative declines in outstanding debt levels caused by debt repayments.
There is obviously something to this notion of a “balance sheet recession” that makes the resulting downturn more debilitating than orthodox macroeconomics would suggest. The recent financial crisis made it clear that the collapse of a bubble in assets financed with debt is far more consequential than the collapse of a bubble in assets financed with cash. According to the Federal Reserve (B.102), the value of corporate equities fell by $7.8 trillion from the first quarter of 2000 to the third quarter of 2002 after the dot-com bubble burst. By comparison, from 2006 to the first quarter of 2009 the Fed (newly) estimates that the collapse of the residential real estate bubble shaved $6.2 trillion (B.101) from household net worth. After adjusting for inflation, the destruction of household wealth from the collapse of the dot-com bubble was actually 50% larger than the associated losses from the housing bubble. Yet the dot-com bubble resulted in a relatively mild recession, few bank failures, and a peak unemployment rate barely above 6%, while the housing bubble’s collapse contributed directly to the deepest and most protracted recession in 70 years.
The difference in the magnitude of the two recessions was directly attributable to the balance sheet effect. At the end of 2007, nearly 100% of U.S. home mortgage debt was held by leveraged financial institutions like Fannie Mae and Freddie Mac, banks, and asset-backed security issuers. By contrast, leveraged financial institutions held only 2.3% of corporate equities at the end of 1999. When short-term creditors fear that a leveraged institution’s assets are insufficient to cover liabilities, they tend to demand repayment, triggering a “run on the bank.” This amplifies the economic impact of the initial decline in wealth. Since equities were held predominately by unleveraged investors like pension and mutual funds, the steep stock market decline did not trigger solvency concerns. Banks were not forced to stop lending or shed assets because they were not directly affected by the collapse in stock prices (they did suffer losses on loans to technology companies, but these were small by comparison). The government did not have to intervene directly to keep insolvent technology companies afloat because these businesses’ failures had little to no implication for the functioning of the payments system.
The slowdown following the U.S. housing bubble was a clear case of a balance sheet recession, with some obvious parallels to the Japanese experience. The first parallel is the decline in wealth from the collapse of the bubble, though U.S. peak-to-trough wealth losses were about one-third as large as Japan’s (as a share of GDP).
Second, the U.S., like Japan, took steps to allow financial institutions to hide the size of their losses. Under immense political pressure from Congress, the Financial Accounting Standards Board (FASB) provided banks greater flexibility with respect to valuing certain distressed assets. However, one must again put this parallel in context. As explained in an e21 editorial, the rule change came too late to cause much damage. Banks had already reported $680 billion in fair value losses at that point and the political pendulum seems to have swung back the other way, suggesting forbearance is unlikely going forward. Perhaps most significant, the capital injections through TARP occurred because of the pressure created by fair value. In Japan, the first capital injections into technically-insolvent banks occurred many years after the bubble burst, by which time deflation had already taken hold
Third, the U.S., like Japan, seems to think that refinancing unsustainable debts into lower rates is an effective strategy. Just as zero percent interest rates failed to stimulate loan demand in Japan, so too would new special refinancing rates fail to erase the asset-liability imbalance on U.S. homeowners’ balance sheets. As argued in a recent e21 reaction piece, when a borrower owes $300,000 on a house that’s only worth $150,000, the problem is unlikely to be solved by getting the borrower a better rate on that $300,000 mortgage. Unfortunately, on this point the U.S. experience seems not at all divergent from Japan. The Administration continues to push out more refinancing programs and other policy analysts embrace even more ambitious “mass refi” solutions. Unsustainable debt levels are still the fundamental problem. It this can only be solved through massive debt forgiveness or natural declines in debt levels through foreclosures or short sales. Rather than pushing mortgage rates to zero and instituting new housing stimulus measures, the U.S. should heed the Japanese experience and support a reduction in mortgage balances. Government-assisted restructurings delay this day of reckoning.
To close the GDP shortfall caused by the overly indebted private sector’s decrease in spending, the Japanese government instituted huge stimulus programs. Between 1991 and 2000, Japanese public expenditures increased by 23% as a share of the economy. Much of the increase was used to fund infrastructure spending allocated towards key political constituencies. Though the stimulus never succeeded to generate sustainable growth, it caused Japan to become far and away the most indebted government in the developed world. Ironically, Japan has spent so much on failed stimulus that it would have cost less – and been far more efficacious – to simply pay down business and banking debts.
More significant than the similarities are the differences. Unlike Japan, the balance sheet recession affects U.S. households and banks, not U.S. businesses. Borrowing by non-financial businesses in Japan ranged from between 8% and 18% of GDP from 1985 to 1990. And from 1985 to 1993, Japanese business debt averaged more than 4-times earnings (see chart). This means that the Japanese borrowed $6 for every $1 of corporate profits, compared to about $1.25 dollar of borrowing for every $1 of corporate profits in the U.S.
Corporate Debt as a Multiple of Earnings
By comparison, U.S. non-financial businesses’ borrowing exceeded 8% of GDP in only two of the past 60 quarters, while these businesses’ debt levels are barely greater than 1.5-times earnings. Today, non-financial U.S. businesses have a combined $2.4 trillion in cash holdings (deposits and cash equivalent accounts of corporations and non-corporate, non-financial businesses) – an all-time record. With real capital spending currently below 2004 levels, the cash-rich and lightly-leveraged U.S. corporate sector is poised to make large contributions to growth going forward as soon as uncertainty concerning the direction of taxes and regulatory policy is removed.
Beyond the health of the corporate sector, the U.S. is also well-positioned relative to Japan because much of the past excess came from global imbalances recycled inefficiently into excess household expenditure. As the chart below shows, from 2003-2007 households ran a financial deficit of an astonishing 7.8% of GDP. Most of that deficit was financed by foreigners (the external sector), who ran a financial surplus of 5.4% of GDP during that period (the mirror of the U.S. current account deficit). In essence, foreigners provided households with a line of credit that averaged $720 billion (relative to 2010 GDP) to spend on houses, goods, and services. The inefficiency of this arrangement is obvious, but it came from an inability to intermediate efficiently foreign excess savings, not domestic excess savings like Japan.
Since the financial crisis began in earnest in the 3rd quarter of 2007, the trends have reversed sharply. Households have seen an 8% of GDP adjustment, going from a 7.8% deficit to a small surplus. This has been offset by a 6% increase in public sector indebtedness and a 2.8% decline in external sector surpluses. Rather than run unaffordable fiscal deficits like Japan, the U.S. can offset the decline in household spending (in excess of income) by further reducing the external sector’s surplus with the aid of additional business investment. Were the external sector’s surplus at its 1951-1992 average of 0.3% of GDP and the non-household private sector deficit at its 1998-2000 level of 1.7%, the government could run 2% of GDP deficits with an economy growing at potential.
Emerging from slow growth in this manner would require sizeable increases in capital expenditures and net exports, but the point is that this outcome is attainable. The current U.S. circumstances in no way necessitate massive government borrowing to sustain output as some believe was required in Japan. Given the white elephant projects brought about by misallocated government stimulus and resulting two-decade-long malaise, an export- and investment-led expansion is not only the preferred outcome for the U.S., but the only one that seems capable of generating a reliable expansion. Simply put, the path charted by Japan is a dead end and should not be pursued.
U.S. household balance sheets are damaged. The sector is likely to be running a financial surplus for the next several years to rebuild balance sheets, but if U.S. fiscal and regulatory policies are rationalized, a strong corporate sector will ensure the outcome need not be a prolonged Japanese-style “balance sheet recession.” The U.S. is not Japan, nor are the policies embraced by the Japanese fiscal or monetary authorities the right ones for America at this time.
Christopher Papagianis is the Managing Director of e21 and was Special Assistant for Domestic Policy to President George W. Bush.