Strong Money Demand in Financing War and Peace: The Cases of Wartime and Contemporary Japan

By Makoto Saito

This book theoretically and empirically investigates the emergence of strong money demand in wartime Japan (1937–1945), its disappearance after the end of the war (1945–1949), and the reemergence of strong money demand in contemporary Japan (from 1995 to the present) in terms of the effects on fiscal activities and the price level. An augmented fiscal/monetary theory of the price level is constructed from a close examination of the strong money demand present in these periods. Then, profoundly puzzling phenomena such as mild deflation despite monetary expansion, low long-term interest rates despite fiscal unsustainability, and weak aggregate demand despite near-zero rates of interest, all of which are actually being observed in contemporary Japan, can now be interpreted in line with the above augmented theory. In the present, strong money demand at near-zero rates endows the Japanese government with maximum fiscal flexibility. However, if it disappeared for some reason, prices wouldsurge to the quantity theory of money level, and fiscal sustainability would have to be restored. In the future, alternative currency units issued by private banks might carry out a purge of such strong demand for the yen.

• Language: English
• Publisher: Springer
• Release date: Jun 17, 2021
• ISBN: 9789811624469

Book preview

© Springer Nature Singapore Pte Ltd. 2021

M. SaitoStrong Money Demand in Financing War and PeaceAdvances in Japanese Business and Economics28https://doi.org/10.1007/978-981-16-2446-9_1

Introduction: Toward a Monetary and Fiscal Theory of the Price Level

Makoto Saito¹  

(1)

Graduate School of Economics, Nagoya University, Nagoya, Aichi, Japan

Makoto Saito

Email: [e-mail]

Abstract

This chapter briefly provides empirical and theoretical motivations for research papers that are collected in this book. In the modern Japanese economy, strong money demand has appeared twice. It first appeared during World War II and then, more recently, after the mid-1990s. Such strong money demand enabled the Japanese government to finance large-scale military spending in the first occasion, while it has supported massive fiscal operations developed by the government in the second. For the former part, the economy was forced to implement strict fiscal reforms to prevent hyperinflations from happening once the strong money demand disappeared immediately after the war ended. For the latter, on the other hand, near-zero rates of interest and stable prices continue to coexist with affluent money and poor fiscal surpluses, while strong money demand survives.

The observations associated with the emergence and disappearance of strong money demand are hard to explain by either the quantity theory of money or the fiscal theory of the price level. Here, a monetary and fiscal theory of the price level is proposed as their alternative. In particular, it can successfully identify additional sources of money demand. More concretely, the strong money demand appearing during the war was attributed to immense demand for BOJ notes from black markets, while the current one has been driven by extremely low interest rates, starting from 1995. Employing rich implications from this alternative theory, I fully elucidate the extent to which the current monetary and fiscal situation of the Japanese economy is sustainable, and how it will break down in the near or far future.

1 How Is Strong Money Demand Characterized?

1.1 Two Occasions When Money Demand Was Strong in Modern Japan

Let me first define ‘strong money demand’, which is the subject of the analysis in this book. What is implied by ‘money demand is strong’ is the situation where the scale of money demand exceeds that of money demand driven by normal transaction motives. It is well known that the ordinary transaction demand for money is approximately proportional to nominal gross domestic product (GDP) at a fixed coefficient. Thus, if the balance of money in circulation is well above such a fixed share of nominal GDP, then money demand is strong. A typical phenomenon, which appears when money demand is strong, is a very weak relationship between the price level and money quantity, such that the price level does not increase as much as the quantity of money.

In the modern Japanese economy,¹ ordinary transaction motives have mostly driven money demand. Figure 1 plots the time series of the ratio of Bank of Japan (BOJ) notes in circulation to nominal gross national expenditure (GNE) up to 1954 and to nominal GDP from 1955 through 2019.² This ratio is known as the Marshallian k by convention. Founded in February 1882, the BOJ began to issue its own notes in May 1885, and these were well established about five years after the first issue. In the prewar period from the 1890s to 1937, k stayed at around 10% of nominal GNE. On the other hand, in the postwar period from 1952 to 1995, k remained at around 8% of nominal GDP (GNE).³ The long-run average of Marshallian k differs slightly between the prewar period (10%) and the postwar period (8%) because, in the latter period, high volume payments are made not by BOJ notes, but through the BOJ reserve requirement system, which was formally established in 1957.

../images/499648_1_En_1_Chapter/499648_1_En_1_Fig1_HTML.png

Fig. 1

BOJ notes/nominal GNE (GDP) and the official discount rate, 1885–2019. Notes (1) In terms of BOJ notes in circulation (measured at the end of each calendar year), Statistics Department, BOJ (1966) for the period 1885–1969, and Money and Deposits → Currency in Circulation in BOJ (2020) for the period 1970–2019. (2) In terms of nominal GNE (GDP), Ohkawa et al. (1974) for the period 1885–1929, EPA (1964) for the period 1930–1954, except for 1945, ESRI (1998) for the period 1955–1979, ESRI (2009) for the period 1980–1993, and ESRI (2019) for the period 1994–2019. Section 4.​2 in chapter Central Banknotes and Black Markets:​ The Case of the Japanese Economy During and Immediately After World War II describes the approximation of the 1945 value of nominal GNE. (3) GNE (1885–1944, and 1952–1954) and GDP (1955–2019) are for calendar years, while GNE (1946–1951) are for fiscal years. (4) For the official discount rate (the calendar-year average of daily data), Interest Rates on Deposits and Loans → the Basic Discount Rates and Basic Loan Rates in BOJ (2020)

However, there were two occasions in which k deviated upward from its long-run trend. Thus, money demand was especially strong on these two occasions. First, k started to deviate upward in 1937 when the Second Sino-Japanese War began and reached just below 50% in 1945 when World War II ended.⁴ It then declined quickly toward 10%, its prewar long-run level, in 1949 just before the Allied occupation of Japan ended. The period in which the Marshallian k displayed such a large upward swing almost overlaps with the period in which the Japanese economy was subject to strict economic controls in terms of prices and quantities. More precisely, control of the Japanese economy started in July 1937 with the outbreak of the Second Sino-Japanese War and ended in April 1952 when the Allied occupation closed, but with most economic controls effectively lifted by the late 1940s.

Second, the Marshallian k started to deviate upward in 1995 when the short-term interest rate was historically low, as shown in Fig. 1. The BOJ then began to guide its official discount rate and overnight call rates (interbank rates) were below 0.5% in September 1995. The k then increased gradually from 8% in 1994 to above 15% in 2004 and exceeded 20% in 2018. As of late 2020, this upward trend in k from its long-run level continued, with the above two occasions examined in more detail below.

1.2 Money Demand Expansion and Contraction During the Controlled-Economy Period

Let me first make remarks on the wartime and immediate postwar price indexes, which are both employed in this analysis. As discussed in Sect. 2 in chapter "Central Banknotes and Black Markets:​ The Case of the Japanese Economy During and Immediately After World War II", the GNE deflator , as adopted for the national accounts by the Economic Planning Agency (EPA, 1964), reflected both heavily restricted official prices and excessive black market prices . Thus, the EPA did not underestimate nominal GNE but calculated it as properly as reasonably achievable. However, as the wholesale and retail price indexes were only based on official prices, both were heavily underestimated during the war, with a gradual correction as official prices approached market prices up until the late 1940s. If we were to calculate the real balance of BOJ notes in circulation using the wholesale/retail price indexes for the wartime period, then the real balance would be heavily overestimated.

As mentioned, the Marshallian k displays an upward curve from 1937 to 1945. For this period, nominal GNE (the denominator of k) increased by a factor of 4.9, but outstanding BOJ notes (its numerator) expanded by a factor of 24.1. Accordingly, k rose from 10.1 to 48.4%. In contrast, k declined quickly from 1945 to 1949. Over that period, nominal GNE expanded by a factor of 29.5, but outstanding BOJ notes increased by a factor of 6.4. Consequently, k fell heavily from 48.4 to 10.5%.

Behind the sharp upswing in the Marshallian k, the price level did not increase as much as outstanding BOJ notes did during the war, but surged despite relatively moderate monetary expansion after the war. By comparison, the GNE deflator increased by a factor of 6.8 from 1937 to 1945 and by a factor of 31.1 from 1945 to 1949. According to the wholesale and retail price indexes, which were based on official prices only, the immediate postwar inflation in Japan was even more remarkable. For the period 1945–1949, the wholesale price index increased by a factor of 59.6 (2.8 for the period 1937–1949), and the retail price index increased by a factor of 78.9 (2.7 for the period 1937–1949).

The prevailing economic history of Japan, including Ito (2012), often affirms that the immediate postwar inflation was hyperinflation. Such a firm judgement is based on the behavior of the wholesale/retail price indexes. Including the years 1950 and 1951, the wholesale (retail) price index soared by a factor of 97.8 (100.4) for the period between 1945 and 1951. A substantial decline in the real balance of outstanding BOJ notes was also clear, being a typical phenomenon of hyperinflation. Because outstanding BOJ notes in circulation increased by just a factor of 9.1, the real balance reduced to less than one-tenth of its 1945 level by 1951.

But it is difficult to judge that the immediate postwar inflation was hyperinflation for the following reasons. First, inflation based on either the wholesale or retail price index, in reflecting only official prices, underestimates the wartime price level. As discussed, the GNE deflator, reflecting both official and market prices, only recorded lower inflation, with a factor of 39.4 in the period 1945–1951. Second, money quantity and nominal GNE moved together for the entire period of the controlled economy. For the period between 1937 and 1949, the nominal balance of BOJ notes in circulation and nominal GNE increased by a factor of 154.1 and 144.1, respectively. That is, the value of the Marshallian k in 1937 and the value in 1949 were almost equal, at just above 10%. In the first half of the controlled-economy period, k changed from normal to extremely high and, in the second half, it changed from exceedingly high to normal. Third, price surges suddenly ceased in late 1949 when k returned to its prewar level, although the price level resumed increasing temporarily between mid-1950 and early 1951. During 1951–1960, the wholesale (retail) price index increased slightly by 0.3% (0.1%) per year. A more precise statement of immediate postwar inflation is that price surges came to a sudden halt before it fell into genuine hyperinflation, in which the Marshallian k would have declined from normal to extremely low.

We can confirm this using monthly data of the real balance of BOJ notes in circulation calculated by the wholesale price index, even though the index was heavily underestimated during the war. As shown in Fig. 2, the real banknote balance began to increase in July 1937, when the so-called Marco Polo Bridge Incident (a trigger for the Second Sino-Japanese War) took place and peaked in August 1945 with Japan’s defeat. During the war, it increased by a factor of more than eight. Immediately after the war, the real banknote balance contracted quickly, and it reached its prewar level by the late 1940s. As discussed in Sect. 2.4.2 in chapter Central Banknotes and Black Markets:​ The Case of the Japanese Economy During and Immediately After World War II, the real banknote balance had a temporary but large fall in February 1946 because the BOJ collected old bills under the Emergency Financial Measures (EFM). According to Fig. 3, which plots the monthly time series of the wholesale and retail price indexes in common logarithms, price surges suddenly stopped after the real banknote balance returned to its prewar level in late 1949. The price level resumed increasing in June 1950 but was almost stable after April 1951.

../images/499648_1_En_1_Chapter/499648_1_En_1_Fig2_HTML.png

Fig. 2

Real balance of BOJ notes in circulation, January 1930–December 1952 (January 1930 = 100). Notes (1) In terms of BOJ notes in circulation (measured at the end of each month), see Asakura and Nishiyama (1974) for the period 1930–1945, and Editorial Office of History of Public Finance in Showa Era (EPHPF), MOF (1976) for the period 1946–1952. (2) For the wholesale price index (monthly data), see Research and Statistics Department, BOJ (1987)

../images/499648_1_En_1_Chapter/499648_1_En_1_Fig3_HTML.png

Fig. 3

Wholesale and retail price indexes, January 1930–December 1960. Notes (1) For the wholesale price index, Research and Statistics Department, BOJ (1987). (2) For the Tokyo retail price indexes, Statistics Department, BOJ (1968)

Given the above observations, it may not be proper to consider that the immediate postwar inflation in Japan was indeed hyperinflation. Instead, the following questions are much more legitimate. Why did strong money demand appear during the war? Why did money demand shrink after the war? Why did the postwar price surge stop suddenly before sharp inflation fell into genuine hyperinflation? As discussed in chapter "Central Banknotes and Black Markets:​ The Case of the Japanese Economy During and Immediately After World War II, we can hold the behavior of black marketeers responsible for the rapid expansion and quick contraction of money demand during the controlled economy in Japan. During the war, black marketeers held massive amounts of BOJ notes to mask their illicit income given their uninscribed nature. But these black marketeers then shifted their asset portfolios from BOJ notes to real assets after the war for the reasons described in chapter Central Banknotes and Black Markets:​ The Case of the Japanese Economy During and Immediately After World War II". Through strict tax reforms immediately after the war ended, the new Japanese government changed its fiscal surpluses from largely negative to nearly always positive from 1947 onward. Once money demand returned to its normal level, and sufficient fiscal surpluses fully backed the public debt, the upward pressure on the price level disappeared. Sharp inflation stopped abruptly in the late 1940s before it fell into genuine hyperinflation.

1.3 Strong Money Demand at Near-Zero Interest Rates

While it may be difficult to identify why strong money demand emerged during the war, it is much easier to understand why strong money demand appeared after late 1995. As mentioned, the BOJ began to control the official discount rate and overnight call rates below 0.5% in September 1995. As shown in Fig. 1, these near-zero interest rates were unparalleled in the modern Japanese economy. While the BOJ and the Ministry of Finance jointly developed a low interest rate policy during the war, the official discount rate was 3.3%. With the low interest rate policy of the second half of the 1980s, which might have been responsible for the asset price bubble at that time, the official discount rate was 2.5% between February 1987 and May 1989. Near-zero costs of money holding because of this historically low interest policy have been pointed at as a primary factor behind the recent appearance of strong money demand.

We can confirm this argument using the quarterly data. Figure 4 plots the quarterly time series of the real balance of BOJ notes, calculated using the GDP deflator, together with those of real GDP. Up until the first quarter (Q1) of 1995, the real banknote balance moved in parallel with real GDP, which drove the transaction motive of money demand. Up until then, money demand moved above ordinary transaction demand. However, while real GDP was stagnant from 1995, the real banknote balance started to expand. With a value of 100 at Q1 1980, the real banknote balance surpassed 200 in Q1 1996, 300 in Q2 2001, 400 in Q1 2006, 500 in Q2 2015, and 600 in Q1 2020. These figures show that money demand exceeded transaction motives from late 1995 onward.

../images/499648_1_En_1_Chapter/499648_1_En_1_Fig4_HTML.png

Fig. 4

Real balance of BOJ notes in circulation and real GDP, Q1 1980–Q3 2020. Notes (1) For BOJ notes in circulation (measured at the end of each quarter), Money and Deposits → Currency in Circulation in BOJ (2020). (2) For the corporate goods price index (seasonally adjusted quarterly data), Prices → Corporate Goods Price Index in BOJ (2020). (3) For real GDP (seasonally adjusted quarterly data), ESRI (2018, 2020)

../images/499648_1_En_1_Chapter/499648_1_En_1_Fig5_HTML.png

Fig. 5

Inflation rates according to three price indexes, 1981–2019. Notes (1) For GDP deflator, ESRI (2018, 2020). (2) For the corporate goods price index (the fiscal year average of monthly data), Prices → Corporate Goods Price Index in BOJ (2020). (3) For the consumer price index, Statistics Bureau of Japan