The government recently declared that there has been a sustained decrease in the general price level, that is, the economy continues to be in a deflationary mode, while the Bank of Japan (BOJ) reintroduced quantitative easing last week. In what follows, I would like to discuss how the ongoing decline in prices should be understood and how the government and the BOJ should respond.
* * *
Public attention is focused on whether or not the Japanese economy is in a deflation spiral. The term "deflation spiral" is not necessarily strictly defined. However, it is generally explained as: a cycle of falling prices that leads to downward pressure on corporate profits, which in turn causes wage and employment adjustments, lower household income, a slump in personal consumption, and a further drop in prices. This is nothing but a vicious circle in which falling prices bring further price falls by exerting downward pressure through wage and other channels. Indeed, it conjures up an image of a bottomless pit in which both prices and wages are in a free fall.
Whether or not deflation spirals actually exist is something that is not fully explained using the economic knowledge to date. Fortunately, at the moment very few believe that the Japanese economy is already in a deflation spiral or at risk of falling into such a state in the near future. The government's view is that although the economy is in a state of deflation, the pace of deflation is modest. BOJ Governor Masaaki Shirakawa has also said that the risk of falling into a deflation spiral is small.
Then, how should we understand the current state of prices? My view is that the situation is far from one characterized by free-falling prices and wages. On the contrary, it should be seen as one in which the response of prices has been extremely weak despite the occurrence of the Lehman shock, a massive demand shock of an unprecedented scale.
The index of industrial production (IIP) in the first half of 2009 has fallen 40% compared to levels observed prior to that period. Although the IIP has somewhat recovered in recent months, the collapse of Lehman Brothers in autumn 2008 did have an enormous impact on quantities such as those of production, orders received, shipments, and employment. In contrast, the impact on prices - i.e. prices of goods and services as well as wages or the price of labor - can be defined as mild with the rate of decline no more than 2-3%.
It can be argued that the huge quantity adjustments have been needed because the price adjustments were too small relative to the size of the negative demand shock. Had prices and wages fallen more sharply in response to the demand shock, the quantity adjustments would have been much milder.
The trade-off between price and quantity adjustments can be found in the Phillips Curve, which illustrates the relationship between the rate of inflation and the rate of unemployment. During the period 2000-09, Japan's unemployment rate significantly increased from the 3% level to above 5% but the inflation rate measured by the consumer price index (CPI) decreased only modestly. As a result, the Phillips Curve is almost flat for the period from 2000 onward ( figure ). That is, adjustments necessitated by the fluctuations in demand during that period of time have been made almost entirely in quantities with price adjustments having played only a limited role. This characteristic is particularly evident when compared to the steep downward slope for the period 1971-89, during which time the role played by price adjustments in absorbing the demand shock was certainly not small. Meanwhile, the slope for the period 1990-99 indicates that the characteristic of price adjustments playing an insignificant role did not emerge suddenly after 2000. The tendency was already there in the 1990s (for which the Phillips Curve is close to flat).
In other words, the ongoing deflation should be understood as a process in which the Lehman-triggered huge negative demand shock, which occurred amid the tendency of price inertia continuing since the early 1990s, is forcibly pushing down otherwise sticky prices. This is quite different from the free fall of prices assumed in deflation spiral arguments.
* * *
Indeed, macroeconomic indicators such as the CPI show that price adjustments have been playing only an insignificant role. However, at the microeconomic level or in the sphere of business management, price competition is said to have been intensifying lately, a situation seemingly pointing to greater elasticity in prices. Then, how do these two facts add up?
In a questionnaire survey on price-setting behavior of Japanese manufacturers conducted by Hitotsubashi University Research Center for Price Dynamics, 95% of the respondents answered "no" to the question of whether they would "immediately" change the price of a product, should they acknowledge changes in the production cost or demand for the product. The top reason for not immediately changing product prices is "the need to see how other companies within the industry respond." In other words, companies are so concerned with how competitors will react that they are apt to play a game of wait-and-see with one another, resulting in a time-lag between changes in production costs/demand and price adjustments.
In determining the prices of their products and services, companies consider two factors: the so-called "fundamentals" consisting of production costs and changes in demand, and competitors' prices. If companies assign 90% of the weight to the former factor and the remaining 10% to the latter when making decisions, they would be quick to adjust prices in response to changes in fundamentals such as a demand shock. However, in a situation where 5% of the weight is assigned to the former and the remaining 95% to the latter, prices would be slow to respond even if a demand shock occurs because companies are concerned with what others might be doing. This is exactly the above-described situation in which price adjustments play only a limited role.
This idea is called the hypothesis of real price rigidities. Price-cutting competition is most frequently observed in the online markets. In research conducted with Takayuki Mizuno and Makoto Nirei using price data observed on the kakaku.com site operated by Kakaku.com, Inc., we found that retailers registered with the site seem to be pursuing the strategy of offering prices slightly lower than their competitors.
Online retailers want to avoid setting prices higher than their rivals because doing so means losing customers. However, simply reducing prices so that they are lower than ones competitors does not mean the number of potential shoppers, or hits, to a retailer's online shop site will increase. Therefore, offering prices slightly lower than ones competitors is the best strategy.
In this way, there evolves the price-setting behavior characterized by retailers imitating one another, resulting in the lagged adjustment of prices. Recent studies in the United States have pointed out that the phenomenon in which exporting companies are slow to pass foreign exchange fluctuations on to export prices is also attributable to real rigidities.
* * *
In Outlook for Economic Activity and Prices released at the end of October 2009, the BOJ predicted that the gradual decrease in consumer prices will continue at least into fiscal 2011 (April 2011 - March 2012). This view is consistent with the hypothesis of real price rigidities, which suggests that price adjustments to a demand shock occur very slowly. Then, what are the implications of price rigidities to monetary policy management?
What is most important is that even if the decrease in prices is modest, the underlying economic losses (distortions in the economy) are not small. The very fact that prices, which are sticky due to real rigidities, are falling albeit modestly indicates that massive downward pressure is at work. We must not discount the seriousness of the situation simply because the decrease in prices is modest.
Furthermore, it is dangerous to focus only on prices. As evident from the above ( figure ), the inflation rate (CPI) changes only slightly even if a demand shock occurs. Because price statistics have their limitations in terms of accuracy or precision, it is not easy to detect slight changes in price levels. Meanwhile, according to the hypothesis of real price rigidities, the speed at which prices are adjusted is determined by the extent to which companies imitate one another. But it is difficult to predict the extent of imitation because it is subject to changes depending on the sentiment of business managers. Given these limitations, central banks should be monitoring the economy as a whole including quantities, instead of monitoring only prices. It is worth considering focusing on nominal gross domestic product (GDP) growth as an indicator for prices and quantities.
In a situation where the Phillips curve is flat, monetary authorities would face difficulty in managing monetary policy as they are left with little room to maneuver. However, the same situation works to enhance the effectiveness of monetary policy because the flatter the Phillips Curve is, the greater the impact of monetary policy in terms of changes in the money supply on quantities (such as industrial production). Monetary policy is non-neutral and central banks' role in restoring and ensuring the stability of quantities is greater in such a situation as described above than under ordinary conditions.
Japan's nominal GDP has dropped by 40 trillion yen or 8% from the level prior to the Lehman shock. In order to restore the pre-Lehman level, Japan needs to achieve, for instance, 2% growth in its nominal GDP for four consecutive years. It is necessary to implement aggressive and vigorous measures to work on the price expectations of companies and households. The BOJ, for its part, may as well announce its commitment to maintaining ultra-easy monetary policy, including quantitative easing, until Japan's nominal GDP returns to the pre-Lehman level.
* Translated by RIETI.
December 9, 2009 Nihon Keizai Shimbun
