Section 2 Causes of Deflation and Challenges to Overcome It
- Japanese version
- English version
The Japanese economy has been in deflation since the mid-1990s. Deflation has been a significant fetter of the Japanese economy, and at the same time, it is a result of the economic slump in Japan, having a close relation with the current economic trends.
In this section, we will examine the causes of deflation and challenges to overcome it. First, we will address asset price deflation, analyzing it from the perspective of both land prices and stock prices. Second, we will analyze general price deflation.
1. Causes of asset deflation and challenges to overcome it
There has been asset deflation since the bubble burst in the beginning of the 1990s. In particular, land prices have been declining since 1991 and the margin of decline is still considerable. The most recent level of land prices is as low as 55% of that of the peak time.(20) On the other hand, stock prices have been on a downward trend for a long time since the end of 1989, still continuing to decline since the second half of 2001. As a result, capital losses arising from such decline in land prices and stock prices during the period from the end of 1989 to the end of 2001 reached 1,330 trillion yen (see Figure 1-2-1). These losses served as a significant downward pressure on the Japanese economy. However, among asset prices, stock prices showed signs of recovery, turning up at the end of April 2003.
We will give an outline of the present situation of land prices and stock prices and then examine causes behind it and challenges to overcome it.
Figure 1-2-1 The Capital Loss of Land and Shares
(1) Land prices continued to decline
Recent trend of land prices
Looking at the year-on-year trend of land prices after the collapse of the bubble economy, we can see that the official published land prices continued to show a sharp drop since 1992. The margin of decline reached a peak in 1993 and then temporarily reduced until 1997 (see Figure 1-2-2). However, since 1998, the margin of decline has been expanding again in places other than commercial zones in the three metropolitan regions (especially in Tokyo). In particular, land prices in local regions have recently been declining, both for commercial and residential zones, to a greater degree than those of the time immediately after the collapse of the bubble economy.
Figure 1-2-2 Officially Published Land Prices
As a result of such decline in land prices, the ratio of land assets to GDP, which rose to reach about 5 times of GDP in the bubble years, has been declining and currently dropped to about 2.5 times of GDP (see Figure 1-2-3). By region, the ratio rose significantly in the three metropolitan regions including Tokyo in the second half of the 1980s, and then dropped sharply mainly in Tokyo. Since 1996, the ratio in Tokyo has been falling below the national average ratio. Consequently, the diversion between metropolitan regions and local regions, which expanded dramatically in the second half of the 1980s, has been diminishing. The diversion between the metropolitan regions and local regions also diminished with respect to commercial zones, the ratio to GDP being about 1 to 1 in both metropolitan regions and local regions in most recent years.
What factors have been behind such trend in land prices, and what requirements should be satisfied to stop the decline in land prices? We will examine these points below.
Figure 1-2-3 Ratio of Land Assets (Private Land) to Nominal GDP
Trend of the rate of return on land
One of the important decisive factors behind land prices is the rate of return on land, which refers here to direct yields calculated by dividing land rent per unit area by land assets. We will now review the trend of the rate of return on land that is estimated for commercial zones (see Figure 1-2-4).(21)
Figure 1-2-4 Direct Yields on Land (Commercial Zones)
The rate of return on land dropped from about 8% in the middle of the 1980s to 4% after the collapse of the bubble economy, and then turned up and improved to 8% in recent years, extending far beyond the level of yields on long-term government bonds.(22)
It is necessary to take into consideration that, in this estimation, 1) land alone is regarded as a production factor and 2) the rate of return on land is considered only based on direct yields.
The first point is that land alone is regarded as a production factor in the estimation. Land may be used less effectively as in the case where it is used as parking lots or highly effectively as in the case where it is used as sites for buildings and commercial facilities. Thus, when regarding land as a production factor, it is necessary to consider it in combination with the buildings and facilities that are built on it.
According to the estimation of the rate of return on land based on the concept that land and equipment are regarded as a unit production factor in non-manufacturing industries, the rate has been rising since the middle of the 1990s, currently exceeding 6% (see Figure 1-2-5(1)).
The second point is that the rate of return on land is considered only based on direct yields. When land is regarded as an investment target, it is important to consider not only direct yields but also capital gains/losses arising from land holding.(23)
Viewing the trend of capital gains/losses from lands in commercial zones, capital gains increased significantly in the second half of the 1980s, the rate of gains nearly reaching 30% in the six metropolitan regions. However, since the beginning of the 1990s, the appraisal turned down suddenly and started to generate capital losses, the rate of gains continuing to be about -20% in the first half of the 1990s in the six metropolitan regions. At present, the rate of gains is still negative for both the six metropolitan regions and all regions in Japan. According to the estimation of the rate of return on land, with capital gains/losses from lands in commercial zones being taken into consideration, the rate rose dramatically in the second half of the 1980s and then dropped sharply in the beginning of the 1990s, remaining negative for the six metropolitan regions in the first half of the 1990s. The rate became positive for the six metropolitan regions, and it has currently been showing a modest upward trend both for all regions in Japan and the six metropolitan regions (see Figure 1-2-5(2)(3)).
Figure 1-2-5 Rate of Return on Equipment and Land
Causes of the decline in land prices
What are real causes of the decline in land prices? Here, we will conduct analysis on this issue according to the capitalization model, which is based on the idea that the value of an asset is decided by discounting potential profits from the asset to the current value.(24) In this model, the following factors are taken into consideration: 1) return currently produced by land and its future prospects; 2) return on risk-free assets; 3) effective rate of fixed asset tax; and 4) the rate of return that investors require beyond return on risk-free assets (risk premium).
There may be room for argument on whether this model applies to the actual formation of land prices. The capitalization model has not been frequently used for actual transactions, and there has been a myth of ever-rising land prices, which was evident in the bubble years in particular. However, it may also be possible to consider that such a concept as the capitalization model implicitly existed behind these actual land transactions, and the real estate market, in which land is explicitly evaluated based on its utility value, is currently expanding. For these reasons, it seems significant to use the capitalization model for analysis.
We will now take a look at what influence the four factors have had on the formation of land prices. Among the four factors mentioned above, 1) return currently produced by land corresponds to return from land estimated by regarding land alone as a production factor (based on the SNA), or return from land estimated by regarding land and equipment as a unit production factor (based on the Financial Statements Statistics of Corporation by Industry, Quarterly). Future prospects of return correspond to demand growth outlooks in non-manufacturing industries based on the Questionnaire Concerning Corporate Activities.(25) 2) Return on risk-free assets corresponds to yields on long-term government bonds. 3) Effective rate of fixed asset tax is estimated from the amount of fixed asset tax and the value of land assets. 4) Rate of return that investors require beyond return on risk-free assets is calculated backward from the other factors because it is difficult to observe premium risk itself.
Among the four factors mentioned above, return on risk-free assets has been on a downward trend since 1990 and it seems to have been supporting land prices. Other factors have been contributing to declining land prices: in particular, the expected growth rate and risk premium seems to have had a significant influence.
The average expected growth rate for the next five years in non-manufacturing industries has been declining after reaching 4% in 1990, falling to around 1% in most recent years. As future prospects of return are as cautious as such, land prices should necessarily be restrained.
Risk premium has been on an upward trend since the middle of the 1990s (see Figure 1-2-6). One of the reasons for the increase in risk premium may be the fact that, amid the collapse of the myth of ever-rising land prices, a cautious mood against risk of the decline in land prices arising from land holding has intensified due to increase in corporate restructuring and NPL disposal. Furthermore, if the rate of return expected by investors has not been achieved due to the decline in land prices, return required by investors would increase.
The effective rate of fixed asset tax has also been rising but its influence has been smaller than those of other factors.
Figure 1-2-6 Trend of Risk Premium
Characteristics of the recent trend of land prices
According to the trend of officially published land prices in 2003, the land prices rose in some high-price zones in the center of Tokyo (see Figure 1-2-7). The margin of decline in land prices decreased in some high-price zones in Osaka and Chiba Cities. On the other hand, land prices continued to decline in all areas in Yokohama City.
Figure 1-2-7 Land Prices in Central Zones, by Metropolitan Region
This may suggest that land prices are nearing levels that match profitability mainly in major cities, and they have stopped declining or turned up in some areas accordingly.
Such change is also seen when comparing the acquisition value with the term-end appraisal value with respect to some J-REIT (Japanese real estate investment trusts) that have recently been listed on the Tokyo Stock Exchange. According to the comparison, land prices are rising in the center of Tokyo and other cities. Furthermore, with respect to J-REIT, the rate of return on investment in lands in the center of Tokyo was generally 6 to 6.5%, lower than that of other cities (land prices in the center of Tokyo are higher than those of other cities). This may be largely because, in the center of Tokyo, 1) rent is less expected to decline and the prospect of successfully securing tenants is deemed to be less uncertain, and 2) future risk of the decline in land prices is deemed to be small.
Challenges to stop the decline in land prices
Signs of an end to declining land prices have just begun to emerge, but the overall downward trend of land prices will continue for some time. The tendency of forming land prices according to the capitalization model will intensify in the future. Assuming this, we can point out the following two requirements for stopping the decline in land prices.
The first requirement is to further promote effective use of land. By increasing the profitability of land through promotion of its effective use, it will be possible to increase expected return and stop the decline in land prices as assets prices.
The second is to reduce risks arising from land holding, such as price fluctuation risk and liquidity risk, thereby increasing the number people willing to purchase land. By doing so, it will be possible to reduce the risk premiums that are contained in land prices and revitalize land transactions as well as stop decline in land prices.
In order to satisfy these requirements, it is necessary to solve the NPL problem and to ease the restrictions on use of land from the perspective of promoting effective use of land.
(2) Stock price decline comes to an end
Viewed from the Nikkei Stock Average, stock prices have continued to decline since May 2002, marking a post-bubble low(26) at 7,607, on April 28, 2003, before turning upward. While the stock market has been thriving as much as it was during the bubble economy, including trading volume exceeding one billion stocks for 37 consecutive business days, stock prices have continued to rise and recovered to the 10,000 level on August 18, 2003 (see Figure 1-2-8). We will now analyze the background and characteristics of these stock price rises and consider their effects.
Figure 1-2-8 Stock Price Indexes in Japan and Other Major Countries
Background of stock price rise
Behind the rise in stock price from the end of April 2003 onward is considered to be the actions of both foreign and domestic factors.
The first factor is the rise in the U.S. stock prices. U.S. stock prices have recorded a sharp decline since 2002 due to further distrust in corporate accounting systems, as represented by Enron and World Com., and the progress of the stock price decline mainly in the communications, IT, and energy sectors triggered by this. Moreover, tensions escalated over the Iraq Crisis in the second half of 2002, which has also become a factor behind the decline in stock prices. However, U.S. stock prices turned up as it became clear that the Iraq War would be ended in a short time and expectations for U.S. economic recovery thus increased. With the sense of a fund-surplus in the world due to monetary easing, this has increased the willingness of mainly foreign investors to purchase Japanese stocks.(27)
The second factor is that an improving trend of corporate profits in Japan has become clear. Examining listed companies' settlement of accounts for FY 2002, we can see that profits are expected to increase by 15% in FY 2003, following the 70% increase in FY 2002. Many major economic indicators thereafter supported such expectation.
The third factor is increased confidence in the soundness of financial system due to the decision to inject public funds into Resona Bank in May. Bank stocks, which had led the decline, have turned up since April.
Foreign investors led the rise in stock prices
Viewed from type of investor, some noteworthy characteristics emerge in this upward phase of stock prices.
First of all, foreign investors led the rise in stock prices. In terms of the amount of trading on a net basis (difference between amount of selling and purchasing), only foreign investors have been continuing significant net buying since April. The amount of their net buying reached 1.7 trillion yen in July (see Figure 1-2-9). This is considered to have happened because a rise in U.S. stock prices spurred purchasing of Japanese stocks that were low in price in comparison with corporate profits. Examining the level of stock price as of April from price earning ratio (PER)(28) that takes interest rate differential into consideration, we can see that the ratios are generally about 1 for foreign countries while the ratio for Japan is far below that.
Figure 1-2-9 Breakdown of Stock Transactions by Investors
Domestic individual investors increased trading
The second characteristic is a sharp increase in trading by domestic individual investors. The amount of trading by domestic individual investors shows a slight net selling on a net basis (difference between amount of selling and purchasing). However, viewed from a gross basis (sum of the amount of selling and purchasing), the amount was 9.8 trillion yen in July, which is more than three times that at the beginning of the year. Seen from the number of traded stocks, it was 20 billion on a gross basis (sum of stocks sold and purchased) in July, which exceeds twice that at the beginning of the year. Increase in trading volume in the entire market also depends largely on such domestic individual investors' active trading.
Domestic individual investors' trading has increased largely because a wide variety of commissions and services for securities trading have become available. In particular, Internet trading has been rapidly expanding due to low commissions and high convenience. Consequently, stock trading commissions have decreased further since deregulation in 1999.
In relation to this, a review of the securities taxation system has also been taking place for the shift "from savings to investment." In terms of capital gains on listed stocks, etc, the system of selecting separate taxation at source or separate self-assessment taxation was abolished and unified into a separate self-assessment taxation. In addition, tax return has become unnecessary through the enabling of the use of a specific account, and the tax rate (total of income tax and inhabitants tax) was reduced to 10% for the five-year period. For dividends from listed stocks, etc., the tax rate was reduced to 10%, and tax return became unnecessary.
City, regional and trust banks continue to post net selling
The third characteristic is that domestic institutional investors continue to post net selling. The amount of net selling by trust banks is particularly large, but city banks, long-term banks and regional banks are also posting net selling. Behind this are considered to be 1) stock sales accompanied with the daiko henjo return of pension funds' assets to the state,(29) 2) stock sales for unwinding of cross shareholdings, and 3) stock sales in preparation for the enforcement of restriction of stock holdings by banks.(30) In relation to such movements, the government and the Bank of Japan have taken countermeasures.(31) In addition, examining the situation only after April, we can see that selling has proceeded considerably. Consequently, the balance of stocks held by financial institutions has been decreasing.
Effects of stock price rises on the real economy
Stock prices are one type of asset price, and they are considered to be decided based on the profits expected to be made by holding the asset. Therefore, stock prices basically reflect the trend of the real economy. However, stock prices also have an effect on the real economy.
Stock price rises generally affect the real economy in the following ways: 1) wealth effect on personal consumption, 2) improvement of the environment for corporate fund-raising, and 3) improvement in business confidence. However, 4) the improvement of balance sheets at companies and financial institutions is also possible. According to the Cabinet Office's "Questionnaire Concerning Corporate Activities," while 60% companies answered that a land price decline would have "no particular effect," 70% answered that a stock price decline would "put pressure on accounting profits through occurrence of extraordinary losses."
Looking at effects on financial institutions' balance sheets, major banks hold stocks of about 14.9 trillion yen and have unrealized losses of about 1.2 trillion yen as of the end of March 2003, on a consolidated basis. Given this, the effect of a stock price rise by 1,000 yen on unrealized losses and profits is estimated to be an improving effect of about 1.1 trillion yen.(32)
2. Current State of General Price Deflation
General price deflation has continued since the mid-1990s. Deflation restrains business investment and other activities by increasing the real debt burden and exerting pressure on corporate profits through raising the real wage or the real interest rate, and serves as a downward factor on the economy.(33) However, deflation emerges as a result of economic downturn in some respects. In the following part of this section, we will first identify the current state of the general price deflation, and then examine its causes and the challenges for overcoming deflation.
(1) Trend in price indicators
Trend in domestic corporate goods prices
Domestic corporate goods price index is the price indicators on goods that are traded between companies. This is a price index that is sensitive to the trend of import prices and the supply/demand situation, which came to be released in place of the domestic wholesale price index.
According to this, domestic corporate goods prices have been on a decline throughout 2002 (see Figure 1-2-10). The goods that made a particularly large contribution to the decline were electrical machinery & equipment among others. However, at the beginning of 2003, crude oil prices increased due to the tension of the Iraq situation and the prices of iron & steel, chemical & related products, and scrap & waste increased due to the firm export conditions. Therefore, despite the fall in prices of electrical machinery & equipment, the level was flat overall.
Although corporate goods prices became weak due to the downturn in crude oil prices that resulted from the termination of the Iraq War from the spring of 2003, they flattened out in summer since the prices of steel, and petroleum & coal products heightened to reflect the firm commodity markets.
Figure 1-2-10 Domestic Corporate Goods and Consumer Prices
Trend in consumer prices
The consumer price index comprehensively represents the prices of goods and services purchased by consumers. The consumer price index excluding fresh food, which sees considerable short-term fluctuations, has continued to fall on a year-on-year basis since October 1999.(34) However, partly because the prices of oil products such as gasoline increased with the influence of the rise in crude prices, the index has stayed flat on a month-on-month basis since the fall of 2002 (see Figure 1-2-10).
After that, oil product prices fell, but due to the revision of public utility rates at the beginning of FY 2003, prices have flattened out. The public utility rates that have been raised include: health and medical services (higher self-pay ratio of medical insurance); electricity/gas rates (adjustment of fuel cost and raw material cost); low-malt beer (higher liquor tax); and cigarettes (higher tobacco tax).
Trend in GDP deflator
The GDP deflator is the price index that comprehensively represents the goods/service activities that are traded in a nation. Since the direct impact of import prices is excluded, it shows the (home-made) price fluctuations based on domestic factors.
Based on a comparison with the same quarter of the previous year, the GDP deflator had been falling since the April-June quarter of 1998. In particular, it had been showing the drop of more than 1% since the January-March quarter of 1999 (see Figure 1-2-11). From the second half of 2002, the fall has become steeper at around minus 2%.(35)
Figure 1-2-11 Contribution to the GDP Deflator
One of the reasons that the drop of the GDP deflator became steeper when other price indices were declining less or leveling off was that external factors, such as crude oil prices, which had acted as boosting factors for other price indicators and had different impacts on the GDP deflator.(36)
The GDP deflator basically expresses the movements of profit and wages per unit of production. Accordingly, although the direct impact of the rise in crude oil prices is excluded, its indirect impact is included. Thus, if a rise in crude oil cost were not reflected or insufficiently reflected in the product prices, the profit and wage would be compressed to cover the loss, and prices would be pushed down compared to when the rise was reflected in the product prices. The result of this shows up as a fall in the GDP deflator. The decline in the GDP deflator in 2002 is considered to be the result of the deflationary tone as well as the effect of such external factors serving as downward factors.
Accordingly, is it the profit fluctuation or the wage fluctuation that is accountable for the GDP deflator fluctuation over this period? Looking at the contribution, the fall in profit made a large contribution in FY 2001, but the fall in wage contributed more in FY 2002.
Although consumer prices and domestic corporate goods prices are moving sideways as mentioned above, considering the effect of temporary factors and the negative growth from the same month of the previous year, the Japanese economy is still judged to be in a mild deflationary condition.
Column 1-2
Real GDP and Nominal GDP
There is an argument over whether to judge deflationary economic conditions by real GDP or by nominal GDP.
Gross domestic product (GDP) measures the aggregate income (= amount of added value) produced within a specific period (e.g., one year) in a nation (e.g., Japan), which is mostly distributed between wages and profits. Also, because the produced income is used for final demand, such as consumption and investment, GDP equals the annual aggregate spending.
GDP is expressed either on a nominal basis or a real basis. Nominal GDP evaluates the aggregate income based on the price level at that point of time. On the other hand, the real GDP evaluates it based on the price of a certain benchmark year (1995 at present). In other words, real GDP is calculated on the assumption that the price level does not fluctuate.
Real GDP is significant as an indicator of the real purchasing power of income. In order to gain an understanding of how much citizens' lives have been enriched through production activities, it is important to measure the actual purchasing power of the added value of those activities. For this purpose, real GDP, in which price fluctuation has been adjusted, must be used. This is the reason that real GDP is used in statistics by various nations and international organizations.
At the same time, the trend of nominal GDP is also significant, because it indicates the trend of the aggregate amount of wages and profits that we actually receive (they are given based on nominal values). In addition, as debts are fixed based on nominal values, it is possible to see the size of the debt burden by comparing the debts with nominal GDP.
By looking at both real GDP and nominal GDP, which have their respective characteristics as above, we can understand that, even when wages or profit are declining (it can be confirmed by nominal GDP), if the prices of goods/services are showing a greater decline, the actual purchasing power has not dropped so much (it can be confirmed by real GDP).
Consequently, it is necessary to look at both real GDP and nominal GDP in order to gain an understanding of the economic situation.
Unemployment rate-based Phillips curve <I>
The consumer price index has posted negative growth from the previous year for almost four years since October 1999. This expresses the seriousness of the deflation. However, the decline remained at about 0-1%. On the other hand, the employment rate surged over this period. Therefore, it would have been natural for the consumer price index to fall more dramatically in relation with the unemployment rate. If a situation where the shrinking of the real economy and the fall in prices interacted with each other in an accelerated and continuous manner were to be called a deflationary spiral, there were concerns over the emergence of such a deflationary spiral. In reality, however, such a situation did not occur. Why is that? We will examine this point based on Phillips curves.
The first Phillips curve we will consider is the one for analyzing the rate of price change in its relation with the unemployment rate. Indeed, a downward-sloping curve emerges by plotting the quarterly data of the consumer price change and the unemployment rate from 1985 to 2002 (see Figure 1-2-12). It indicates that the rate of price change declined gradually with the rise in the unemployment rate, eventually entering the deflationary region.
Usually, a Phillips curve analyzes rate of price change based on the relation between the unemployment rate and the expected rate of price change, and assumes a downward-sloping curve between the rate of price change and the unemployment rate, given that the expected rate of price change has been provided. Thus, we will estimate a Phillips curve by first obtaining the expected rate of price change. We will use the expected rate of price change that has been estimated based on the "Consumer Confidence Survey" by the Cabinet Office.(37) Figure 1-2-13 shows the results. According to this, there has been a moderate deflationary expectation since 2000.
Figure 1-2-12 Phillips Curve Based on Unemployment Rate <I>
Figure 1-2-13 Expected Rate of Price Change: Consumer Prices
The estimated Phillips curve is plotted over the earlier Figure 1-2-12. For convenience, 1985 onward is divided into three terms. By obtaining the average expected rate of price change for these respective terms, a corresponding Phillips curve has been plotted. This Phillips curve is approximated by an asymptotic hyperbola because it fits better than when approximating it by a straight line.
This reveals that the Phillips curve has become nearly flat after entering the deflationary region. It seems as if there is a lower limit to the deflation. Why is that?
Unemployment rate-based Phillips curve <II>
To answer this question, the following points need to be considered. Since the Phillips curve is based on the principle of supply and demand in the labor market: 1) the cyclical unemployment rate is more appropriate for use than the unemployment rate including structural factors; and 2) attention should be paid to the relation with the wage growth rate rather than the relation with the price change rate.
When we plot the relation between wage growth rate and the cyclical unemployment rate, we can see a downward-sloping relation as in Figure 1-2-14. We do not find a relation that becomes flat in the deflationary region. Thus, we will estimate a Phillips curve that indicates the relation between wage growth rate and cyclical unemployment rate using the earlier-obtained expected rate of price change.
The result of the estimation is shown in Figure 1-2-14. While the Phillips curve is plotted to correspond to the average expected rate of price change for the respective terms, it is better appropriated by a straight line than a hyperbola this time, and it no longer shows a relation that hints at a lower limit to the deflation. The reason that the curve became nearly flat in the deflationary region was due to the recent growth in structural unemployment. So, with that excluded, we can clearly observe a relation in which the wage growth rate declines with an increase in cyclical unemployment.(38)
Figure 1-2-14 Unemployment Rate-Based Phillips Curve <II>
GDP-based Phillips curve
What happens if we draw a Phillips curve based on GDP? We will estimate the relation between the GDP deflator and the GDP gap given that the expected rate of price change has been provided, by assuming the principle of supply and demand in the macroeconomic goods and services markets.
The GDP gap is the difference between estimated potential GDP and actual GDP. According to this, a deflationary gap has been observed since 1997, and the gap has been closing since the bottoming out of the economy in 2002 (see Figure 1-2-15). The relation between this GDP gap and GDP deflator since 1985 is plotted in Figure 1-2-16. A downward-sloping relation is again observed here.
In addition, we also need the GDP deflator-based expected rate of price change (Figure 1-2-17)(39) for the estimation.
The result of estimating a Phillips curve based on these factors is drawn over the earlier Figure 1-2-16.(40) The Phillips curve is again plotted to correspond to the average expected rate of price change for the respective terms. Approximation by a straight line fits better in this case also. It indicates that the increase rate of GDP deflator declines with the expansion of the GDP gap, and creates negative growth.
Figure 1-2-15 GDP Gap and GDP Deflator
Figure 1-2-16 GDP-Based Phillips Curve
Figure 1-2-17 Expected Rate of Price Change <II>: GDP Deflator
The demand-side factors and the supply-side factors causing the deflation
The presence of a GDP-based Phillips curve provides an important suggestion on the causes of the deflation. As indicated in the FY 2001 and FY 2002 editions of the Annual Report on the Japanese Economy and Public Finance, the causes of the deflation can be summed up into: 1) supply-side factors; 2) demand-side factors; and 3) monetary factors.
The presence of a GDP-based Phillips curve indicates that potential GDP, which is included in the "supply-side factors," and actual GDP, which is included in the "demand-side factors," both have an impact on the fluctuation rate of the GDP deflator. We can see that both "supply-side factors" and "demand-side factors" are causing the deflation.(41)
Monetary factors causing the deflation
Then, what should we think of the "monetary factors"?
The relation between money supply and prices can be summarized as below.
First of all, from a short-term perspective, if money supply increases by a monetary policy, the real GDP grows with a decline in the interest rate and raises the rate of price change.
Secondly, from a long-term perspective, growth in money supply would be equal to growth in the nominal GDP, but considering that the actual GDP and potential GDP coincide on average, growth in money supply would mainly be the factor that influences the rate of price change.
The above can also be confirmed metrically.(42) With regard to the short-term relation, it has been suggested that money supply has an impact on real GDP and real GDP has an impact on the GDP deflator. As to the long-term relation, it has been indicated that money supply and the nominal GDP has a stable relation.
Therefore, money supply also causes price fluctuations through having an impact on demand-side factors. In that sense, the present deflation reflects the lack of high growth in money supply. As discussed in Section 3, monetary policy has adopted quantitative easing and achieved a high monetary base growth, but that has not resulted in high growth in money supply. In addition, the moderate money supply growth has not resulted in an increase in nominal GDP. One of the underlying factors is that the transmission mechanism of the monetary policy has been cut off by the non-performing loans and excessive debt issues. These factors are obstructing growth in money supply and impeding the Japanese economy from overcoming the deflation. Thus, "monetary factors" also account for the deflation.
Challenges for overcoming the deflation
The causes of the deflation discussed above also indicate the requirements that must be met for overcoming the deflation. That is to say, to overcome the deflation: 1) the GDP gap that brings about deflation must be reduced; and 2) in order to deal with the "monetary factors" that cause deflation, it is necessary that effective monetary policy is implemented and the effects of the monetary policy expands aggregate demand through money supply expansion. Deflationary expectations are likely to gradually alter when these requirements come to be met.
In order for the GDP gap to decline, it is important that, with progress in disposal of non-performing loans and reduction of excessive debts, potential demand develops in growth fields and funds move smoothly from less efficient sectors to growth fields that have high efficiency and social needs. Furthermore, progress in disposal of non-performing loans and in reduction of excessive debts would also recover the transmission mechanism of monetary policy. Conversely, if aggregate demand grows with the help of the effects of monetary policy, the disposal of non-performing loans and reduction in excessive debts will make progress. In this manner, structural reform and monetary policy interact with each other.
In order to overcome the deflation, the government and the Bank of Japan need to act as one and implement strong and comprehensive measures.