Learning from Japan: It's Hard to End a Deflation
In 1974, when the Fed faced rising inflation, the U.S. government sought to “Whip Inflation Now” by encouraging people to wear “WIN” buttons. Today, the problem is reversed. Several central banks are having trouble creating inflation. Unfortunately, we doubt that “SIN” buttons – “Support Inflation Now” – will be any more effective than the earlier variety, which served mainly as fodder for late-night comedy.
As has been the case for some time, Japan is blazing the trail into the monetary and fiscal unknown. Today’s Bank of Japan (BoJ) leadership is far more determined to promote price stability than its predecessors over the past two decades. But the deflationary hole that Japan is climbing out of is so deep that the BoJ may need some help. Unfortunately for them, the government’s fiscal position is among the worst in the OECD. Consequently, while the central bank seeks to stimulate the economy and hit a 2% inflation target, households and businesses reasonably anticipate a long period of fiscal consolidation.
From its inception in early 2013, Abenomics had three arrows: monetary, fiscal and structural. Since Haruhiko Kuroda succeeded Masaaki Shirakawa as the Governor in March of that year, the BoJ’s balance sheet has more than doubled from ¥165 trillion (33% of today’s ¥500 trillion GDP) to ¥366 trillion (73% of GDP). And, the central bank remains committed to purchasing ¥80 trillion per year (16% of GDP) of Japanese Government bonds (JGBs), in line with its “Quantitative and Qualitative Easing Policy” (QQE) begun in October 2014.
To put these numbers into perspective, Japanese central government debt is just under ¥890 trillion, and is currently increasing at a rate of something like ¥25 trillion per year. Of this, the BoJ currently holds more than ¥310 trillion, and the Japan Post Group has about ¥125 trillion. Assuming they maintain the current rate of QQE, by end-2017 the Bank of Japan will hold 50% of all Japanese sovereigns. And, by end-2020, they will hold 70%.
But there is a solid rationale for the BoJ’s Monetary Policy Committee (MPC) to expand QQE further. The reason is that inflation has not reached the 2% target, and it may not get there without a policy shift. Consider, for example, the following chart copied from the latest biannual “Outlook for Economic Activity and Prices.” It shows MPC members’ forecasts for the inflation in the CPI less fresh food, together with their risk assessments. While four of the nine members forecast that inflation will reach 2% by 2017, the majority does not. Moreover, the risk assessment reveals that five of the MPC members believe that the risk to their two-year-ahead point estimate is skewed to the down side!
Bank of Japan: Policy Board Members’ Forecasts and Risk Assessments for Year-to-year Percent Change of CPI (All Items Less Fresh Food), Fiscal Year 2010-2017
Source: Bank of Japan, Outlook for Economic Activity and Prices, October 2015.
The second figure is from the BoJ’s “Tankan” quarterly survey of enterprises. It shows that corporate expectations for output price inflation over the next one-, three-, and five-year horizons have declined. A similar retreat is evident in annual headline CPI inflation numbers, which are once again showing readings close to or even below zero. While “core-core inflation” (which excludes food and energy) was 1% over the past 12 months, our preferred indicator of the trend – the trimmed mean CPI inflation – is running at only about ½%. To put these trend indicators into perspective, since the BoJ’s aggressive monetary expansion began, the ex-food and energy core-core is up by about 1½ percentage points, but the statistical measure has risen by only 0.7 percentage points.
Household attitudes also cast doubt on an upward inflation trend. While household five-year inflation projections (measured as the median of the BoJ’s quarterly Opinion Survey) are running at 2 percent, this is a bit lower than in 2011 (when the quarterly median readings averaged 2.75%), prior to Abenomics. Moreover, as is often the case, these survey responses persistently overestimate inflation: since mid-2006, the one-year-ahead inflation forecasts anticipated a cumulative price rise of 26%, compared to the actual increase of 3%.
Japan – Business Expectations for Output Price Increases, March 2014-September 2015
Source: Bank of Japan Tankan (Quarterly Survey of Enterprises).
Finally, the BoJ knows that – on current plans – the nation’s consumption tax will rise in April 2017 by a further 2 percentage points to 10%. So far, each prior increase of that tax has been followed by a sizable drop in real GDP. With that in mind, there seems relatively little risk to price stability over the medium term if inflation in Japan were to temporarily overshoot the central bank’s 2% target.
One response to all of this is to reiterate the advice some people have given since the deflation started in the 1990s: engineer an exchange rate depreciation. While this was surely sound advice then, Japan’s effective exchange rate already has depreciated by roughly 25% over the past two years with only a small inflation pickup.
So, as the MPC meets over the coming quarters, we suspect that it will be increasingly clear to its members that current policy is inadequate to secure the 2% inflation target over the next two years, just as it was inadequate to reach the target over the past two years. Assuming, as we do, that Governor Kuroda and his colleagues are resolved to rectify this, it is natural to expect them to expand QQE in an effort to increase nominal aggregate demand (with the result that the value of the yen could decline even more). And, if they think as we do, then only a very large further expansion will have much chance of success.
As monetary economists, we often learn important lessons from examining what happens when policy hits certain limits. This prompts us to pose the following extreme question: What will happen if the BoJ continues to accelerate JGB purchases but that – even after it acquires most of the central government debt – inflation continues to fall short of the 2% target? Understanding the BoJ’s options in this case is useful because it can affect confidence in the BoJ’s current strategy – something that is critical for understanding what will happen to inflation expectations.
So, suppose the BoJ owns all the JGBs. What then? We see only two principal policy alternatives to achieve the inflation target. The BoJ could start to buy other things, or it could turn to the government to engage in a money-financed fiscal expansion.
On the first, there is a long list of non-JGB assets that the BoJ could acquire. But each item brings complications. For example, they could buy some of the ¥300 trillion in regional debt issues, effectively monetizing the debt of the prefectures and cities. But, such a policy risks subsidizing those local governments with higher debt who may be viewed as having behaved less responsibly in the past.
Another possibility would be to expand the existing program of private asset purchases already included in QQE – the “qualitative” part. That would mean purchasing instruments like commercial paper, corporate bonds, ETFs and REITs. But the sum of everything in these categories on the BoJ’s balance sheet currently only amounts to ¥13 trillion – 3½% of the total assets – and is scheduled to grow by a very modest ¥4 trillion per year. Since the capitalization of the Tokyo stock market alone exceeds ¥500 trillion, this could obviously be increased. But doing so would further move QQE toward fiscal policy: the BoJ would become more and more like a sovereign wealth fund that subsidizes corporate winners. If one of the BoJ’s goals is to maintain its independence, this looks like a move in the wrong direction.
The second alternative – no less risky – is to turn directly to the fiscal authority for cooperation. If the government wishes to support the central bank’s inflation objective, it could in theory run a sufficiently large deficit that the BoJ can monetize. A money-financed increase of government expenditures ought to lead eventually to inflation: if not, the government would be able to acquire all the goods and services in the economy without raising the price level. The other option – a money-financed government transfer to households – is a form of Milton Friedman’s famous helicopter drop. Whether it leads to inflation depends on whether people spend the transfer.
You might think that monetary-fiscal cooperation would solve a number of Japan’s problems simultaneously. Not only would it help the BoJ create inflation, but the government would no longer be constrained by the need to compensate investors who hold its debt. Inflation and debt sustainability all at once!
If this seems too good to be true, that’s because it is. To see what is happening here, we need to consolidate the balance sheet of the central bank and the fiscal authority. When we do, the fact that one part of the government is purchasing the liabilities of another drops out of the picture. What matters is the liabilities of the government as a whole. So, if the BoJ owns all the JGBs, they will have converted these fixed-rate long-term liabilities into overnight floating-rate liabilities. If, as policymakers hope, inflation rises, the largest component of the BoJ’s liabilities (the banks’ reserves) will need to be compensated at an interest rate consistent with keeping inflation close to target. The government will still have a very large debt; it’s just that it will be of extremely short maturity and held almost entirely on the balance sheets of private banks.
Would there be rollover risk? Not in the normal sense. The central bank controls the supply of reserves and cash that banks and the public have no alternative but to hold. But the situation could still be precarious: a loss of confidence could prompt a run on the currency itself. And, one way or another, the Bank of Japan’s balance sheet will deteriorate when – to keep inflation in check – it tightens monetary policy. At that stage, it will either sell a large volume of assets at a discount relative to the purchase price or fund what remains with negative carry. Regardless, the central bank faces the prospect of sizable losses.
Would this be sustainable? We don’t know, but it would surely be an unenviable position. At the same time, failing to achieve its inflation target would be no less challenging for the BoJ and the Japanese economy. The public debt managers cannot count on issuing 10-year debt with a nominal yield of only 0.3% and a real yield of roughly zero forever. Without a sustained rise in nominal GDP, even a modest deficit will lead to a rising public debt ratio.
So what are the lessons to take away from Japan’s ventures into the unknown? The first and most important is to do everything you can not to start down the deflationary road. We now see that even a resolute central bank cannot easily end a deflation that has become as deeply embedded in household and business expectations as Japan’s. Even policy actions that are difficult and inevitably risky may turn out to be insufficient. Second, there is the fact that unanticipated deflation can contribute to a country’s fiscal woes, reducing the government’s capacity to respond effectively. Put differently, even when deflation is slow and lasting (rather than sharp and acute like that of the Great Depression), it can still be debilitating.
From the narrow perspective of the central bank, there is an additional lesson about independence. Pundits typically worry that a central bank that loses independence will be induced to foment a large inflation. This is the story of fiscal dominance. It is what BoJ policymakers may have feared even after gaining legal independence in 1997, a few years after the deflation began. The irony is that an entrenched deflation also can undermine the independence of a central bank, because it may not be feasible to restore price stability without support from the fiscal authority.