Japan: monetary base vs. quantity of money

Posted by Tim Congdon in News Archive | 0 comments

Japan’s ‘Abenomics’ is reported to have three arrows,

a ‘revolution’ in monetary policy with ‘the Bank of Japan injecting huge amounts of “money” (whatever that means) into the economy’ (or something of the sort),

a short-term fiscal stimulus accompanied by long-term action to bring the public finances under control, and

– ‘a growth strategy’ (which means in practice shaking up such over-protected parts of the Japanese economy as farming and retailing).

Commentary on the last two of the three arrows has often been sceptical. Initial ‘stimulus’ (i.e., a widening of the budget deficit) is not easily reconciled with ultimate fiscal consolidation (i.e., a narrowing of the budget deficit), while Abe’s Liberal Democratic Party has drawn much of its traditional support from groups that benefit from protection and restrictive practices. By contrast, most media reporting has suggested that the Bank of Japan has definitely changed course and that a major upheaval in monetary policy is under way. This note argues  that,  although  Japanese  monetary policy  has  indeed  shifted  in  an expansionary direction, the shift is far less radical than the rhetoric that has accompanied it. Japanese policy-makers and the greater part of the commentariat seem to believe that the monetary base by itself has great macroeconomic importance. This is a mistake. National income and wealth in nominal terms are a function of the quantity of money, which must be distinguished sharply from the base. Movements in the monetary base and the quantity of money may be related, but the relationship is not necessarily all that precise or reliable. It is the quantity of money, not the monetary base by itself, that matters to macroeconomic outcomes.

Monetary base target in Japan

A key part of Abenomics in Japan is the Bank of Japan’s target to double the monetary base between April/May 2013 and April/May 2015. According to several announcements from the BOJ, the purpose of this doubling of the monetary base is to end deflation. (Japanese consumer prices have not fallen to any significant degree in recent years, but the gross domestic product deflator used in national income accounting has fallen for extended period since the early 1990s. In that sense Japan has suffered outright deflation.)

The clear implication of several of its statements is that BOJ believes that the monetary base by itself is relevant to the determination of the price level. This proposition seems to be widely believed among today’s monetary economists, although it departs radically from the standard and traditional formulations in which the quantity of money is the variable at work. A discussion of the meaning of these two terms is needed to help in clarifying the subject.Composition of the monetary base in Japan

In Japan, as elsewhere, the monetary base consists of monetary liabilities issued by the central bank. The most basic such liability is the note issue, which must generally be used to settle debts because it has legal-tender status. (The coin issue – usually a liability of the official mint – is small and a nuisance, and has no role in monetary policy. It can be ignored in the rest of this note.) The monetary base has two main components,

–     Notes and coin held by the general public (i.e., monetary base held by the public), and

–     Notes and coin held by the banking system plus banks’ cash reserves at the central bank (i.e., monetary base held by the banking system).

Banks’ cash reserves are comprised partly of so-called ‘vault cash’ (i.e., cash held ‘behind the counter’ to meet retail demand) and partly of balances at the central bank that are freely convertible into notes. Evidently, banks’ cash can be used to meet deposit withdrawals from the public and to cover debit balances in inter-bank settlement. A vital part of banking routine (and also, unfortunately, of banking crises) is to ensure that the cash balance is positive at all times, so that banks can meet their payments obligations to customers.

Historically, banks held low ratios of cash reserves to deposits, because central banks did not offer interest on cash reserves. However, nowadays central banks typically do pay some interest on reserve balances maintained with them. In Japan at present banks’ cash holdings are somewhat larger than the note issue held by the public, as the pie chart above demonstrates. Japan’s GDP is usually given as slightly under 500,000 billion yen or, translating ‘billion yen’ into ‘trillion yen’ (i.e., a trillion equals a thousand billion), between 475 and 500 trillion yen. (At an exchange rate of 100 yen to the dollar, Japan’s GDP therefore comes out as $4.75 trillion to $5 trillion, somewhat less than a third that of the USA.)

It is evident that Japan’s monetary base – at under 180 trillion yen in total – is smaller than GDP, with banks’ cash reserves alone, at a little under 94 trillion yen, being about 20% of GDP. The note issue is widely used of course in retail trade and for small local transactions, sometimes in ‘the black economy’. However, the proportion of cash payments to total payments in a modern economy is very low, being typically under 1%, according to surveys of payment patterns. Although the level of the public’s note holding is plainly related to that of retail trade, and although retail expenditure may be correlated with GDP, it is difficult to believe that the general public’s note holdings are a powerful causal influence in the determination of national income and wealth. Instead, the note holding – which is only a modest part of the total quantity of money – tends to adjust to the quantities of GDP and consumer expenditure which are set by other forces.

It follows that – if a doubling of the monetary base is to be effective in stimulating the economy – the other component of the base, banks’ cash reserves, must be the missile that really matters. What is the evidence on the power of this particular policy missile?

Banks’ cash reserves and nominal GDP in Japan: past experience

The chart below shows % annual changes in Japanese banks’ cash reserves and in Japanese nominal GDP since the early 1980s. A quick glance is sufficient to demonstrate that no relationship at all obtains between these two variables. In the 1980s, when no one was concerned about deflation in Japan, banks’ cash holdings were much more volatile than nominal GDP and on average grew somewhat more rapidly, but there was no compelling year-by-year correlation between the annual changes. In the 1990s anxiety about the solvency of Japanese banks began to emerge and, to some extent, the Japanese public may have preferred to hold money in the form of notes rather than bank deposits. It might have been expected that banks would want to hold more cash as a precautionary measure, but until the late 1990s this is not confirmed by the data. Since the turn of the century policy-makers have taken a more active interest in expanding the monetary base, partly because of advice from American economists – such as Allan Meltzer – that expansion of the base ought to revive the economy.Do changes in banks' cash holdings affect GDP in Japan?

But, quite simply, the evidence does not substantiate the case for focussing on the monetary base as, by itself, the Alpha and Omega of monetary policy. The first arrow of Abenomics envisages a doubling of the monetary base in two years. But – as the chart shows – banks’ cash reserves, the part of the base that is supposed to motivate the banking system to change behaviour and to drive the economy to a higher growth path, trebled in 2002. That trebling followed the adoption of a conscious policy of ‘quantitative easing’ in March 2001. But the trebling of banks’ cash reserves did not cause the banks to adopt more aggressive strategies of asset acquisition. On the contrary, Japanese banks remained hamstrung by past losses and worries about their capital adequacy, and potential borrowers in Japan itself were not interested in taking up the very inexpensive loan facilities that had been available before March 2001 and remained available thereafter. QE was dropped in early 2006. Wild swings in banks’ cash reserves have nevertheless continued to the present day. Movements in nominal GDP have had no visible connection whatsoever with these swings, as the table on the next page brings out very obviously. The big fall in Japanese GDP in 2009 was due, above all, to the weakness in the world economy and the large appreciation of the yen, as yen carry trades were abandoned with the crash in interest rates to virtually zero in the world’s leading economies. Changes in the monetary base and banks’ cash reserves had no obvious bearing on the severe 2009 recession or the 2010 bounceback.

The  information  presented  here  is  well-known  and  not  particularly  controversial. The  puzzle  is Japanese policy-makers’ apparent conviction that – despite the absence of a relationship between the monetary base and banks’ cash reserves on the one hand, and the behaviour of the economy on the other – they must still persevere with a target expressed in terms of the monetary base. Part of the explanation may be the strong influence of American monetary economists, many of whom see the base as the best measure of monetary policy, on Japanese officialdom.The absence of a relationship between banks' cash and nominal GDP

The quantity of money, broadly-defined, is the concept of money that matters to macroeconomic outcomes

As noted above, standard textbooks say that it is the quantity of money, not the monetary base, that is relevant in the monetary theory of the determination of national income. Over the last 20 years or so broadly-defined money measures have in fact grown a little in Japan, whereas nominal GDP has been at best flat. The numbers don’t fit together perfectly, but they are ‘in the same ballpark’. Closer examination of the data would probably find the usual ‘stylized facts’ about the role of money and banking in a modern economy, but it would require a great deal of statistical work and knowledge of institutions to make the argument compelling. It will merely be stated here – without an attempt at more rigorous demonstration – that the concept of money that matters to macroeconomic outcomes in Japan, as elsewhere, is one that is broadly-defined to include all money-like assets that can be used in transactions. (In more detail, they can be used in transactions, at little cost and after no notice period or a relatively short notice period.)

I have explained in several places – for example, in essay 4 in my 2011 book Money in a Free Society

– that, when financed from the banking system, purchases of assets by the state from non-banks directly increase non-banks’ deposits. In other words, they add – straightaway and without further ado

– to the quantity of money. The current exercise in monetary stimulus in Japan is said to include purchases of long-dated bonds, which would normally be held only by non-banks. To the extent that the BOJ bond purchases are from non-banks, that ought to mean an acceleration in the growth of broad money. (It may or may not affect the monetary base, but do I have to repeat myself by saying that the effect on the base is not the point?)

What has in fact happened to the growth rate of broad money in Japan in the last few quarters? The answer is given in the chart on the following page. The first arrow of Abenomics may be stated in terms of the monetary base, but – because asset purchases from non-banks increase the quantity of money as such – it has in fact been accompanied by an upturn in the growth of the quantity of money.Broad money developments in Japan

To that extent Japan’s monetary stimulus programme does matter. (Note that bank credit to the private sector has been sluggish in recent quarters. Another confusion in this subject is the proposition that ‘expanding the monetary base ought to cause banks to lend more, boosting the economy’. But no necessary relationship holds between new bank lending to the private sector and any other significant macroeconomic variable. Sure, if new bank lending creates extra money balances, the increase in the quantity of  money affects the equilibrium levels  of national income  and  wealth, in the manner understood by traditional theory. But, again, it is the quantity of money that matters, not bank lending to the private sector by itself.)

Conclusion: Japanese economy is being boosted by a monetary stimulus, but it is a very weak one

The chart immediately above looks quite striking in its way, with the annual growth rate of M3 reaching its highest figure in the 21st century. But the reader should take a little care: check the y axis. For much of the 2009 – 12 period the annual growth rate of M3 was about 2%. In the last few months there has been the hullabaloo about Abenomics. In the last couple of months the annual growth rate of M3 has moved up to…well, it has moved up to 3%. A 1% rise in the growth rate of the quantity of money is worth mentioning, but it is not dramatic and certainly not revolutionary. Admittedly, a 3%- a-year growth rate of broad money – if sustained – would be different from the 0.6% a year average growth rate of M3 in the eight years 2001 – 08 inclusive. The tiny minority of economists who judge M3 to be a key money aggregate might well argue that this increase in money growth ought to be sufficient to stop deflation. (I would take this line myself.) However, two conclusions may now end this piece.

First, a very weak monetary stimulus is now at work in the Japanese economy and the macroeconomic data are indeed improving, with better growth and even some glimmers of a positive change in prices instead of deflation. But the shift towards ease is small and does not justify the razzamatazz in the financial press about Abenomics. For market participants the main message is clear. The big movements in the Japanese stock market and the yen exchange rate have been disproportionate to the minor change that has in fact occurred. The trouble is that these participants seem as befuddled as the journalists  and  politicians  by  all  the  waffle  about  the  monetary  base.  Let  me  reiterate  for  the umpteenth time: the money aggregate relevant to the determination of asset prices is one that is broadly-defined to include all, or virtually all, bank deposits and money-type deposits held by genuine non-banks. Both the monetary base and narrow measure of money are by themselves irrelevant to asset prices.

Secondly, the Japanese authorities do not know what they are doing. The silly and disproportionate rhetoric on the monetary base, in the context of absurd hoopla from Abe himself and the related gang of PR advisers, is instructive in that it shows how little the top people understand of basic monetary economics. Japan’s monetary base tripled in 2002, and neither broad money nor the economy responded in any significant way whatsoever. Admittedly, the ignorance among Japanese central bankers and regulatory officialdom is not much worse than that in their counterparts in other leading nations, and this ignorance is partly to be excused by the shocking neglect of traditional monetary economics in the leading American universities. All the same, investors in Japanese assets need to notice that – since the election of the Liberal Democrats under Abe in December 2012 – the annual rate of growth of broad money has changed by a mere 1%.

Will that 1% become 3% or 5% in the next few months? Maybe, but then the important point is to monitor the figures as they emerge, and to pay attention to the actual data rather than hype from political windbags and gullible commentators.

Tags: , , , , , , , , , ,

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Time limit is exhausted. Please reload CAPTCHA.

News Archive

Article Archive