A note on Japan’s experiment with quantitative easing
Marshall Auerback here. Just yesterday, Ed wrote a good post on quantitative easing. As much as I liked the post, frankly I think Ed omitted some crucial factors about Japan, which I will detail here for you.
Japan’s policy makers generally procrastinated considerably in terms of implementing any kind of stimulative measures, as well as prematurely reversing the benign impact of policies which had some earlier success. In terms of monetary policy, the BOJ did not actually embrace quantitative monetary easing until 2001, eleven years after their bubble had burst. Furthermore, the authorities waited until 1998, a full 8 years after the real estate crash, before embarking on a program of banking recapitalization, the upshot being that the traditional mechanisms through which credit was to be intermediated were still severely impaired. The benign impact of fiscal reflation was also undermined by a wrong-headed increase in the consumption tax in 1995, which was followed on by the Asian financial crisis, creating an additional exogenous shock to an economy still recovering from the throes of a major credit bubble collapse.
Japan’s lost decade can be divided into three sections: the 1991-93 recession, the temporary recovery of 1994-96, and the deep recession of 1997-99. Each episode offers important lessons.
The post-1989 collapse of Japan’s equity bubble was, at first, welcomed by the Bank of Japan. Even though stock prices were falling rapidly, land prices continued to rise for another year at least, and the Bank of Japan remained concerned about inflation pressures. The Bank of Japan actually boosted the discount rate from 4.75 percent to 6 percent in August 1990 and held it at that level until June 1991. As the collapse of equity prices intensified and land prices began to fall, the Bank of Japan reversed course and cut its discount rate rapidly by 275 basis points in the year following June 1991. Subsequent additional cuts of 150 basis points in 1993 brought the discount rate down to 1.75 percent by September 1993. Simultaneously, Japan pursued three major fiscal stimulus packages totalling 6 percent of GDP between August 1992 and September 1993.
The effects on the economy of these extraordinary stimulative measures were limited. Growth recovered only slightly–to just over 1 percent in 1994–for several reasons. First, the Bank of Japan was late to initiate its easing. By the time of the first rate cut in June 1991, Japan’s nominal GDP growth, which in the long run should not be allowed to fall below policy interest rates, had dropped to about 2.5 percent, headed down to a negative level by 1994. That level was far below the 6 percent discount rate in place. Simultaneously, falling land and equity prices were erasing household wealth and resulting in a sharp curtailment of credit because of the heavy exposure of Japan’s banks to the commercial land bubble. This was especially damaging because Japanese companies are far more reliant on banks for financing than American companies, which have access to broader credit markets.
More broadly, deflation began to emerge, which boosted real growth numbers artificially. (Real growth is calculated by subtracting inflation from nominal growth or adding deflation to nominal growth.) Real growth of 2 percent, generated by zero nominal growth and 2 percent deflation, is too weak.
Japan’s large fiscal stimulus packages, which became legendary during the 1990s, were ineffective for several reasons. First of all, the packages were not as large as advertised, often inflated by double counting as stimulus government programs that were already slated to be undertaken. More importantly, the packages were poorly directed–largely toward unproductive public works projects and credits to small businesses that were no longer economically viable. Also, the need to move production facilities abroad grew as the economic environment in Japan deteriorated and deflation strengthened the yen, making goods produced in Japan too expensive in world markets, which should have induced greater moves to induce aggregate demand, but instead was deployed in inefficient “bridges to nowhere”, given the continued dominance of the construction companies and the like on the Japanese political system.
Japan’s sharp interest rate reductions and sizeable public works programs did help to boost the economy mid-decade. The real growth rate reached nearly 4 per-cent in 1996 and early 1997. Yet an ominous period of deflation that emerged mid-decade and dragged down nominal GDP growth was largely ignored by Japan’s policymakers. The deflation, coupled with a loss of public confidence tied to the Kobe earthquake in January 1995, resulted in an elevation of liquidity preference by Japanese households and firms.
The sharp rise in the demand for money was signaled by a rapid appreciation of the yen, which reached 79.8 per dollar on April 19, 1995. As Japanese households sought to raise cash balances, Japan’s external surplus created a strong demand for yen that was not offset by capital outflows. The sharp appreciation of Japan’s yen early in 1995 signaled a severe liquidity shortage, which resulted in a sharp drop in demand, even though market interest rates were low. In 1995, Japanese policymakers responded with another 125 basis points of rate cuts that brought the discount rate to 0.5 percent in September of that year. Another two major fiscal stimulus packages in 1994 and 1995, again totalling 6 percent of GDP, helped to sustain a recovery of growth until 1997.
Japan’s biggest policy mistake came in 1997 when the government raised its consumption tax from 3 to 5 percent. The aim was to help compensate for the large run-up in Japanese debt that resulted from the series of unproductive fiscal stimulus packages expended largely on wasteful public works projects. The combination of higher consumption taxes, the continued fall in land prices that persisted in preventing Japan’s banks from operating as financial intermediaries because of their heavy exposure to real estate losses, and a rapid return to deflation in 1998 resulted in a virtual collapse of the Japanese economy.
Japan’s poorly timed attempt to redress its large budget deficit and rapidly rising public debt provided a compelling reminder of the lessons that John Maynard Keynes taught in the General Theory of Employment, Interest, and Money. Fiscal stringency in the form of a tax on consumption in an economy weakened by massive wealth losses and an erosion of confidence that results in a virtual liquidity trap is an extraordinarily harmful policy. Japan’s nominal GDP growth rate was below zero for most of the five years after 1997, with most of its positive real growth resulting from the technical application of GDP deflators averaging about -1.5 percent. In a deflationary economy, it is important to watch nominal, not real, GDP growth.
Japan persisted in believing that monetary policy was extremely accommodative because of low nominal interest rates and the Bank of Japan discount rate at 0.5 percent. But with deflation at 1.5-2 percent and negative nominal GDP growth, a 0.5 percent interest rate was actually restrictive.
After five years in a deflationary economic wilderness, the Bank of Japan switched during the spring of 2001 to a policy of quantitative easing–targeting the growth of the money supply instead of nominal interest rates–in order to engineer a rebound in demand growth. Simultaneously, the energetic and innovative Junichiro Koizumi was elected prime minister in April 2001, and he undertook aggressive measures to recapitalize Japan’s banks, which were still heavily burdened by non-performing loans.
The need to recapitalize Japan’s financial sector had been evident since late 1997 when Yamaichi Securities–a midsized Japanese securities house–collapsed, followed by failures of some regional banks in 1998. The persistent problem was non-performing loans whose total value rose as high as 20 to 25 percent of Japanese GDP. While substantial public funds had been made available to Japanese banks in 1999, balance sheet restructuring was undertaken only slowly and reluctantly, partly because of the stigma attached to revealing long-concealed losses and fears of losing control to foreign investors. A lack of transparency in the balance sheets of Japanese banks and a passive approach by the Japanese government to restructuring those balance sheets contributed substantially to the prolonged period of economic stagnation. Deflation exacerbated the problem by further depressing the land prices to which Japan’s balance sheets were closely tied and by elevating real interest rates. When cash–a riskless asset–was earning the deflation rate of 2 percent or higher, few wanted to leave their deposits in banks whose solvency was in question because of heavy exposure to the collapse in land prices and that offered interest rates to depositors that did not compensate for those perceived risks.
The move by the Bank of Japan to quantitative easing and the large increase in liquidity that followed helped to stabilize land prices by 2003. The Japanese economy then enjoyed modest growth averaging around 2 percent per year. The Bank of Japan held interest rates at zero until early 2007, when it boosted its discount rate back to 0.5 percent in two steps by midyear. Since then, however, the slowdown in global economic activity and persistent deflation stayed the Bank of Japan’s hand from further rate increases and raised the possibility of even returning to rate cuts in 2008 as the outlook for the Japanese economy worsened.
By contrast, everything that Japan did in terms of banking recapitalisation and quantitative monetary easing has been effected by the Bernanke Fed within a year of the credit crisis enveloping the U.S. economy. And President-elect Obama has already promised major fiscal stimulus early next year. The one aspect of the Japan analogy which does suggest ominous problems for the U.S. is the latter’s exceptionally high debt to GDP ratio of 350 per cent which contrasts negatively with Japan’s huge accumulated savings surplus at the time their bubble burst. Of course, one could easily turn that argument around and suggest that the savings surplus gave Japan’s monetary and financial authorities a significant margin of error in policy making, a margin which they comfortably exceeded through a combination of political inertia and economic incompetence.
It is possible that, for the time being, the existence of record leverage caused the U.S. stock market to undershoot its current quite disastrous fundamentals. Of course, these fundamentals could get worse over time, warranting a yet deeper stock market decline. The economic contraction could be more severe and sustained than anything in the post war period. The price level could fall for the first time in the post war period. The financial crisis could become far more severe. And the market cap to GDP ratio is still well above the post war lows. All this bad stuff has not yet happened, so I’m not trying to say it can’t happen. But there has never been so immense a policy response in history.
History says that, for the time being, this stock market probably overshot to the downside relative to current economic conditions and investor psychology when it hit 750 on the S&P about 10 days ago. The explanation probably lies in record leverage. I should add that the largest group of leveraged investors – the hedge funds – are probably now, in aggregate, net short. Let’s wait a bit before we start rendering definitive judgements about what we should do.
Your comments are welcomed.
That’s valuable detail on the history of Japan’s crisis, thanks.
“The one aspect of the Japan analogy which does suggest ominous problems for the US is the latter’s exceptionally high debt to GDP ratio of 350 per cent which contrasts negatively with Japan’s huge accumulated savings surplus at the time their bubble burst.”
To quantify this a little more, according to economist Steve Keen Japan’s debt-to-GDP ratio at the peak of their crisis (162%) was half that of the US earlier this year (~350%) (though another estimate has Japan’s ratio at 250% in 1990 so one of the two is off).
My feeling is that while it SEEMS like the Western govts have done ‘the right things’ in response to the crisis, it by no means follows that we will have a traditional V or U shaped recession, followed by a recovery to historical growth levels within 2-3 years. I think what has been done has just prevented the total collapse of the financial system (and even that is open to question). The economy will now have to bump along with weak or zero growth for a number of years (5+, possibly up to a decade) while everyone, govts and individuals, work at paying down the levels of debt that have reached such epic proportions. Only when all the usual ratios have returned to, and possibly undershot, their long term averages can we resume some sort of meaningful increases in national and private wealth.
Of course all the above analysis is pointless if govts decide that the best way to pay off all the debt they have taken on to their books is to create an inflationary spiral by whatever means necessary. As that is the ‘easiest’ method for debtors to escape their creditors, I fully expect govts to take that route. After all who votes for a politician who promises a decade of high taxes and low growth to pay off all the old debt from past consumption?
Sobers: Good points. I would simply argue that either scenario is VASTLY superior to the Great Depression where the economy halves in 4 years and unemployed people comprise one-third of the population. That’s why I hate the Austrian economists so much. Their policies will bring us there.
Marshall “I would simply argue that either scenario is VASTLY superior to the Great Depression where the economy halves in 4 years and unemployed people comprise one-third of the population.”
But if only to play the devil’s advocate, that leaves 66% of the population who would still not be unemployed. Or put another way, the easy, simple way out of this mess appears to be to potentially screw ALL of us instead of 33% of the population (or whatever the percentage might be if we include families). Even if those employed but with debts got bailed-out so that some of the 66% were net beneficiaries of inflation, the margin between winners and losers would not be “vastly” clear-cut, & would it be morally right for savers and pensioners to be thrown to the wolves whilst the profligate are rescued?
Re: Austrian economists – I agree that their medecine COULD return us to the 1930s. But had we listened to them in the first place we would not be in this mess now. To use the Keynesian/Inflationary escape route now is a bit like the old prayer “Lord make me good (but not just yet)!”.
It is always a ‘one last hit’ situation with debt and Keynsian stimuli. When will the discipline be reintroduced into the system? There always seems to be a reason for ‘not yet’. If there was a proposal to introduce some new system of economic governance that took the decisions away from politicians, and defined the maximum size of the state, and maximum tax levels etc etc, I could agree to the ‘one last hit’. Govts must be made in the future to have the same rules apply to them as apply to the rest of us – live within your means, and if spending rises faster than income, cut spending, not raise taxes.
Plus as mentioned above the hardship under an inflationary spiral is probably just as great, just not as visible. People on fixed incomes, and with savings are impoverished at the expense of the profligate. What sort of message is that to send? Don’t work hard, be thrifty and save, just get yourself into massive debt and consume, as the govt will inflate your debt away? Moral hazard to the nth degree I think!
“in the long run we’re all dead” means “screw the next generations, they’ll pay our debt. i’m old, bitter and childless, so i don’t give a shit.” typical baby boomer thinking. it’s just bad morality and here’s something else for you. japanese managers have a huge problem now because young’uns don’t want any responsibility. only want to put 9-5 and go home. why? because they saw what happened to their parents who got screwed while banks got “recapitalized”. the future is bright ahead.
Thanks for this great post. One thing that I’ve wondered about is how the carry trade affected Japan’s attempts to stimulate internal demand with easy money. I’m not sure about the timing, but it seems like it would be hard to boost demand with much of the newly created money flowing out of the country to be invested elsewhere. Any thoughts on this?
Thanks, Jim
Thanks Jim. You raise a good point. I used to always say, “liquidity abhors a vacuum” and definitely one adverse side effect of the BOJ’s move to quantitative easing was the rise of the Japanese retail carry trade. In a February 2005 piece for FinanceAsia , economist Richard Duncan made the startling observation that in 2003 and the first quarter of 2004, monetary authorities in Japan created 35 trillion yen:
“To put that in perspective, ¥35 trillion is approximately 1% of the world’s annual economic output. It is roughly the size of Japan’s annual tax revenue base or nearly as large as the loan of $2500 for every person in Japan and, in fact, would amount to $50 per person if distributed equally among the entire population of the planet. In short, it was money creation on a scale never before attempted during peacetime.”
Not all of that surplus liquidity has found its way back into the Japanese economy. Rather, it has simply flowed into the U.S., helping to sustain the latter’s ongoing credit bubble. If “liquidity does indeed abhor a vacuum,” then the actions of Japan’s monetary and financial authorities demonstrated this conclusively.
Today, however, we have major risk aversion so I think there would be less “leakage” particularly as all major G7 economies are now reflating and undertaking significant fiscal stimulus. If anything, the “leakage” ultimately is likely to flow the other way, given the trillions of dollars being created by the Fed. But a large chunk will impact on the US economy as well, which is why I do think the measures in place will ultimately “take” and give us a horrendous inflation problem down the road. No good policy choices left, I’m afraid, just least bad policy options.
I hope this helps.
@Marshall Auerback: Marshall those are very good points especially as regards leakage. I would agree that leakage is less of a problem than it was when Japan was conducting its experiment with quantitative easing.
In normal circumstances, you would expect leakage as funds flow to the path of least resistance, inflating whatever asset sector is in favour globally i.e. commodities and oil in the past year. But, as almost all asset prices are falling simultaneously, there will be less leakage.
David Rosenberg, Merrill Lynch’s Chief North American Economist has pointed out that Treasuries may be unsustainably high — in bubble territory. The yield on the 3-month U.S. Treasury is effectively zero and the yields on the 2-year and 10-year are at all time lows. Price moves in the opposite direction to yield so prices of U.S. Treasuries have been skyrocketing. One might argue that excess liquidity is flowing there.
On the whole, in a gloablised financial system without significant capital controls, there is always a danger that there will be some asset class globally which will receive excess demand, creating a bubble.
For anyone interested, Richard Koo, chief economist at Nomura Research Institute, has written two illuminating books about Japan’s Great Recession (as he calls it). The most recent is The Holy Grail of Macroeconomics: Lessons from Japan’s Great Recession. The other (circa ~2003) is called Balance Sheet Recession. In a nutshell, he classifies Japan’s lost decade as a \balance sheet recession\ caused by the corporate sector shifting into balance-sheet-repair mode after the asset bubble burst. He says in such an environment monetary policy is largely impotent because demand for funds is feeble. The only effective policy prescription is increased fiscal spending to close the output gap resulting from retrenchment in corporate sector demand (capex). He says that Japan’s oft-maligned fiscal stimulus programs successfully averted depression.
“The one aspect of the Japan analogy which does suggest ominous problems for the US is the latter’s exceptionally high debt to GDP ratio of 350 per cent which contrasts negatively with Japan’s huge accumulated savings surplus at the time their bubble burst.”
Ed, this is something I’ve often wondered about. If it’s the case that the Japanese people (and government) had a savings surplus and high rate of savings at the time of their economic woes and deflationary outlook, then wouldn’t it have been very difficult politically to put into place policies that might cause large amounts of inflation? If deflation “helps” savers, then perhaps that was the reason for their inertia and delay regarding quantitative easing?
In the US, we’re the opposite. Most people are in debt. Most people don’t have savings. Most people would probably welcome a little dose of inflation when compared to a little dose of deflation. Politically, it’s much easier to inflate our way out of this problem in the US. The only people that would complain are those with money, and if you have money right now, you’re going to be considered fortunate and should have no reason to complain when others have nothing and are unemployed.
Back in the summer I would read articles regarding whether the fed would ever go under 2% interest rates. The predominant theory was that this would “punish savers” and that was supposed to be one of the reasons why it wouldn’t happen. But, who exactly are these savers? (umm, me?) Not the majority, that’s for sure.
Your thoughts are much appreciated.
Sorry, my last post should have been addressed to Marshall who wrote the original post.
@Knute K Knutson:
Yes, I’ve already read parts of Koo’s book and he is absolutely right. And my understanding is that Summers has read the book as well and agrees with it.
@Knute K Knutson: Knute, I haven’t read Richard’s books but I would argue that we are seeing something similar here today on a global scale. The difference is the it is the supply of lending due to a financial sector balance sheet sector that is causing the problem today. I am less optimistic about the benefits of low interest rates and quantitative easing (although I prefer the latter to the former as policy tool).
That said, I agree with Marshall that inflation would be a real concern if this easing does succeed.
As for the restrictive supply of credit, it seems that the only way the government can get banks to lend again is to credibly threaten to nationalize them. Otherwise, lending is unlikely to resume in earnest due to either low rates or a flood of high-powered money. We are likely to see a massive build up in excess reserves and a contraction of the money multiplier.
@Matthew:
That’s a great argument, Matthew, one that makes a lot of sense, and explains largely why the Japanese people were so adamantly opposed to taxpayers’ funds for the banks. Of course, it ended up being a case of cutting their collective noses to spite their faces, so while politically your argument explains a lot, it was an economic disaster.
You could also argue that Japan’s huge bulk of savings induced more political inertia, because the resultant social strains were not as great. Japan’s policy makers had a huge margin of error which they still managed to exceed!
You could also argue that Japan’s huge bulk of savings induced more political inertia, because the resultant social strains were not as great. Japan’s policy makers had a huge margin of error which they still managed to exceed!
Marshall,
Yes I can absolutely see that. If everyone (including the government) had the 6-12 months of emergency savings and socialized health care, we wouldn’t *have* an emergency to deal with yet, and the economy might be left to collapse further before everyone ran out of savings. By then you might be left with a depression rather than a normal recession.
This is echoed by my anecdotal sampling of friends. Those that just sold their house, have no debt, but have a few hundred thousand in the bank are all for NO BAILOUTS. If the economy collapses they’ll be just fine, and can buy a house back for less money. Deflation and mass unemployment for a few years wouldn’t be a big issue for them, or so they think. Other friends that are more reliant on the paycheck to meet their bills are also angry at [insert entity] for this problem, but support spending (AKA borrowing) to support the economy. Inflation for them means less vacations/wine and not starvation, but unemployment would be a killer.
The very mindset of America is about bringing forward consumption to today, at the expense of future consumption. Yes, you hear the occasional economist or commentator talk about the immorality of leaving our children with debt, but that’s like telling an alcoholic not to drink that last bottle of vodka today as another future alcoholic might get a headache tomorrow. It is considered patriotic to spend, and normal to be in silly amounts of debt. Our government does it, we do it, and it seemed to be working. Until it didn’t.
“That’s why I hate the Austrian economists so much. Their policies will bring us there.”
Their policies? Who for god’s sake is “they”? Ron Paul?
It is the bubble that will bring us there: that is what bubbles do: burst and create messes.
In your analysis, you left out one important piece of Japan’s response to their crisis. Beginning in 1992 and continuing until 1996 the government embarked on what jokingly was called the PKO, or price keeping operation (a take on the Middle East peace process). In order to attempt to reflate the balance sheets of Japanese financial institutions, the government handed out Postal Savings (Yucho) and Postal Insurance (Kampo) money to foreign and domestic trust banks and told them to “buy” stocks. Just buy.
Altogether upwards of $350 billion was distributed in this manner. The average cost was around 19000 Nikkei. The reflation of the balance sheets of the banks (who held massive amounts of shares) approved the appearance of bank capital ratios (similar to the current $350 billion in handouts to the major banks and remainng IB’s in the US), but it did nothing to promote renewed lending.
The real downside to that PKO has yet to come home to roost. With the Nikkei below 9000, those funds are deep in the red. But it is worse than it appears because of peculiarities in the accounting method used in managing those funds (as well as Japanese pension funds). I’ll briefly explain: at the time of the PKO, payout rates for Kampo and Yucho were approximately 6%. Payouts could only be made from recognized gains. The Japanese government forced trust banks to supply the needed payout from the funds under management, so what occured was that “winning” equity positions were sold in order to recognize the gain. Losers were not sold, as recognized losses had to be used to offset gains. All the trust banks built up substantial pools of dead funds. The stocks held in these pools could not be sold, because then the required payout could not be met. What it means is that today, more than a decade later—and with Japan’s rapidly ageing population in need of its pension/savings/insurance—the effective average cost of these funds is probably equivalent to 25000-30000 Nikkei.
Of course some of these losses were offset by gains on JGB’s, but not enough.
Tokyo Joe, I’m sure Marshall would want to reply but he’s off in California. That is a great pice of analysis and I appreciate your providing it. My understanding is that the Japanese are back at this again now:
“Japan’s government will buy as much as 20 trillion yen ($223 billion) of shares held by banks to boost their capital and support a sagging stock market.
The amount is part of a stimulus package totaling 75 trillion yen that also includes 10 trillion yen for capital injections into banks, the Cabinet Office said in a statement. The buyback plan is subject to approval by Japan’s parliament.
The government is encouraging banks to lend more to small companies struggling to secure funds amid the worst global financial crisis since the Great Depression. Investment losses on stocks have eroded capital at banks including Mitsubishi UFJ Financial Group Inc., crimping their capacity to lend.
“The government’s purchase will prevent banks from selling their stockholdings in the market and creating oversupply,” said Hideo Arimura, who oversees about $1.9 billion at Mizuho Asset Management Co. in Tokyo. “However, there is no guarantee banks will sell their holdings.”
Mitsubishi UFJ, the nation’s biggest bank, rose 3.3 percent to 560 yen as of the 3 p.m. trading close in Tokyo, while Mizuho Financial Group Inc., the second-largest by revenue, climbed 1.2 percent. The Topix Banks Index, tracking 84 lenders, gained 0.6 percent.
The Nikkei 225 Stock Average fell 44 percent this year and is set for a record drop. It’s decline has been compounded by mark-to-market accounting rules, forcing companies to book losses on illiquid holdings of securities.
The Bank of Japan cut its benchmark interest rate to 0.1 percent from 0.3 percent today and said it would buy more corporate debt as a deepening recession chokes off funding for businesses. ”
https://www.bloomberg.com/apps/news?pid=20601087&sid=a.Xq_Or5JUyk&refer=home
This certainly seems like they are about to re-attempt what you have said they already did and failed with. Am I missing something here or is this not what is happening?