Lin Research 050709

Japan’s ‘Lost Decade’: Analysis of the Effects of Interest Rate Manipulation

Albert Lin Rutgers University [email protected]

Submitted to Professor Hiroki Tsurumi 5/07/09 Abstract: Horioka (Japan and World Economy 18(4):378-400, 2006) argues that demand side factors offer a better explanation of Japan’s economic stagnation (misguided government

policies) while others including Hayashi and Prescott (Review of Economic Dynamics 5(1):206235, 2002) suggest that supply side factors (low productivity growth) are more poignant in explaining Japan’s economic stagnation. I hypothesized that theoretically interest rate manipulation should increase private investment but in this case overall have no effect on GDP growth due to supply-side problems such as low productivity growth but find that interest rate changes have weak correlation to private investment. However, a simple regression showed moderatecorrelation of interest rate change to real GDP growth. This, I conclude, creates a disturbing impression that interest rate adjustments have helped real GDP growth. Also, I find that private and public investment has little correlationto real GDP growth. Keywords: Bank of Japan, Japanese Economy, Lost Decade, Monetary Policy, Ordinary Least Squared Regression Model

Introduction: Since 1990, Japan’s economic stagnation continues to persist. Although the credit boom in the year 2000 created the illusion to declare Japan out its “Lost Decade”, the years did not publish any higher levels of GDP growth higher than 2.90 percent according to Fig. 1. Hailing as the second largest economy in the world, Japan has experienced significantly lower rates of GDP growth (average of 1.17 percent from 1996 to 2005) when compared to that of the United

States (3.26 percent), Asia excluding the Middle East (3.24 percent) and the World (3.04 percent). Fig. 1

Source: http://earthtrends.wri.org/text/economics-business/variable-227.html from EarthTrends Org.

Fig. 1 displays the trend in Japan’s GDP growth percentile from 1990 to the year 2005. There is a decline in GDP growth starting at 1990, with the burst of the real estate bubble until there seems to be modest recovery starting from 1993 (strong-armed fiscal policy in 1995) until the shocks of the 97’ Asian financial crisis adversely affected the revival. In the year 2000, a credit boom creates the illusion of a robust recovery of the Japanese economy (sixth column of Table 1); however, GDP growth rates slump back to 0.20 percent in 2001. Chart 1 The GDP growth in percentiles of these respective regions, 1996-2003 2

2

2

2

2

2

1

1

1

1

World

005 3.5

004 4.1

003 2.7

002 1.9

001 1.5

000 4.1

999 3.2

998 2.3

997 3.7

996 3.4

Asia

4.7

5.1

3.9

2.8

2.2

4.6

2.4

-1

3.2

4.5

2

2.7

1.5

1.3

2.1

4.0

3.2

2.7

2.7

1.7

United States

3.2

3.9

2.5

1.6

0.8

3.7

4.5

4.2

4.5

3.7

Japan

1.9

2.7

1.4

0.3

0.2

2.9

-0.1

-2

1.6

2.7

Europe

Source: same as Fig. 1

Fig. 2

Source: http://www.e-stat.go.jp/SG1/estat/ListE.do?lid=000000730075 Official Statistics of Japan.

Fig. 2 shows the trend over time of Japanese CPI. CPI, a common measure of inflation, reveals that inflation rose steadily beginning from 1990 and peaking at 1998 even as GDP growth dwindled (Fig. 1). The sharp turn upward in trend in year 1996 may be attributed to the strong fiscal spending policies adopted by the government in 1995 as there had been an increase in money supply. What had been the government’s role during the so-called “post bubble”period? In this paper I specifically investigate the government’s monetary policy of interest rate manipulation. I assume that the government’s goals are to encourage the steady growth of GDP, maintain a low rate of inflation, creating a positive trend of productivity and real GDP (per worker), and making the most of resources/labor consistent with stable prices. Because relevant data on total factor productivity had been processed and limited to 3 sets of data http://www.kojin.org/papers/06_Japan_TFP.pdf, I chose data on the growth percentage of real GDP (seasonally adjusted) to use in demand-side factors calculations as the determining factor of policy effectiveness. I assumed that an effective change (that can be numerically measured) in government policy will yield a positive change in real GDP growth. Also, since maintaining a low rate of inflation and making use of resource/labor are inherently contradictory goals, I choose the growth of GDP as an indicator of the government’s policy success. But to briefly provide background on government policy tools; demand-side policies would include fiscal and monetary stimulus, as well as interest rate manipulation. Fiscal policies including taxation (changes in marginal tax-rate) and subsidizing of industries would all be considered supply-side. However, other fiscal methods such as issuing and buying up bonds affect aggregate demand. The demand-side policy effect I choose investigate is primarily Bank of Japan’s interest ratebecause interest rate has a negative correlation to investment. Because Japan has a cultural

tendency to save, increasing investment becomes increasingly important during economic downturns so that aggregate demand will be sustained by raising investment: Yd = C + I + G + (X – M) where C is consumption, I is investment, G is government spending, and (X-M) is the net export. I develop a comparative regression analysis using this data, interest rate adjustments, and that of private and public investment in order to discover if there are strong relationships between these factors.

Background Discussion: I chose to use the ordinary least squares regression model through the eviews program to see if there is a strong relationship between interest rate change and private/public investment. The OLS model is especially useful and accurate in predicting expected values of the dependent variable. Hayashi and Prescott (Review of Economic Dynamics 5(1):206-235, 2002) run simpler regression models in which they produce a cross section regression. However, they develop a more complex equation of the aggregate output function where they combined the results of their regression. But their main focus is to discover the relationship between aggregate production and the shortened work week implications i.e. hours worked per employee and aggregate employment.

Ceteris paribus, I view the relationship of investment to be positively correlated to aggregate demand: Yd r Ipub + Ipriv r Ratebank Therefore, I mimic the same regressions methods used by Hayashi and Prescott, OLS;

however, because my data is in a time series, I ignore the stationarity hypothesis test and proceed with the OLS regression.

Regression Model: OLS model, which is mathematically explained by the following equation:

S xy := ∑ xi2 −

1 ( ∑ yi ) 2 n

In my case:

where r = real gdp growth and i = investment, interest rate S ri := ∑ ri ii −

1 ( ∑ ri )( ∑ ii ) n

I determine the real GDP growth to be the dependent variable because of the framework: Yd = C + I + G + (X – M) where I = investment, keeping all other factors constant.

Data: Sources and Construction I use the information from the Japanese statistics bureau, a division of the Ministry of Internal Affairs and Communications as my primary data source. Ultimately I take GDP growth change to be the dependent variable, as well as the ‘measurement’ for the effectiveness of each government policy alteration. A few adjustments were made to the GDP growth rate so that it is realGDP and also seasonally adjusted. Interest rate data from the Bank of Japan was initially downloaded to contain monthly data; I refitted the data to quarterly, taking the lowest interest value of each quarter to be

significant because in the economic downturn, the government aims to increase credit flow by loweringinterest rate as much as possible. I took data available from the official Bank of Japan website. The private and public investment data had been left as-is. The statistics were taken from the Cabinet office of Japan website official statistics page that can be found here: http://www.esri.cao.go.jp/en/sna/qe084-2/gdemenuea.html.

Empirical Results: Dependent Variable: REALGDP Method: Least Squares Date: 03/29/09 Time: 10:45 Sample: 1994Q1 2000Q4 Included observations: 28 Variable C BANKRATE R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Durbin-Watson stat

Coefficie nt

Std. Error

t-Statistic

Prob.

519235.1 -15903.9 3

3935.536

131.9351

0.0000

2019.718

-7.874335

0.0000

0.704563 0.693200 6303.679 1.03E+09 -283.661 6 0.466860

Mean dependent var S.D. dependent var Akaike info criterion Schwarz criterion

489699.2 11380.63 20.40440 20.49956

F-statistic Prob(F-statistic)

62.00515 0.000000

Dependent Variable: REALGDP Method: Least Squares Date: 03/29/09 Time: 10:58 Sample: 1994Q1 2000Q4 Included observations: 28 Variable

Coefficie nt

Std. Error

t-Statistic

Prob.

C PRIV_INVEST PUB_INVEST R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Durbin-Watson stat

541478.6 1.480922 -1.40630 1

31590.57 0.963323

17.14051 1.537306

0.0000 0.1368

0.831727

-1.690821

0.1033

0.178335 0.112602 10720.76 2.87E+09 -297.981 9 0.214283

Mean dependent var S.D. dependent var Akaike info criterion Schwarz criterion

489699.2 11380.63 21.49871 21.64144

F-statistic Prob(F-statistic)

2.713021 0.085838

Dependent Variable: PRIV_INVEST Method: Least Squares Date: 03/29/09 Time: 11:01 Sample: 1994Q1 2000Q4 Included observations: 28 Variable C BANKRATE R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Durbin-Watson stat

Coefficie nt

Std. Error

t-Statistic

Prob.

1388.426 -223.550 3

1360.918

1.020213

0.3170

698.4232

-0.320079

0.7515

0.003925 -0.03438 6 2179.827 1.24E+08 -253.928 8 0.739380

Mean dependent var

973.2607

S.D. dependent var Akaike info criterion Schwarz criterion

2143.289 18.28063 18.37579

F-statistic Prob(F-statistic)

0.102450 0.751467

Conclusion: Empirical results show that there is weak relationship between bank interest rate adjustment and private investment. This result seems anomalous because theoretically and logically there should exista negative relationship between lowered interest rates and increased

aptness to invest. My hypothesis becomes disproved by further multiple regression testing between the relationship of private and public investment to real GDP. According to the R2values, and ignoring the t-statistic because the sample size is smaller than 30, the probabilities of bank rate having a relationship with GDP growth is insignificant at the 5% level. However, at first glance there seems to be moderate correlation between interest rate adjustment and real GDP growth. I believe this creates the false impression that these factors may be directly related. In theory, interest rates affect investment, which in turn contribute to GDP. However, there proves to be barely any relationship between the government’s interest rate policy and private investment. This leads to other possible studies about the efficacy of interest rate adjustment. Also, it suggests that there is indeed a supply-side economic problem of low productivity.

REFERENCES K. Assenmacher-Wesche, Stefan Gerlach, Toshitaka Sekine, Monetary factors and inflation in Japan, Journal of the Japanese and International Economies, Volume 22, Issue 3, September 2008, Pages 343-363, ISSN 0889-1583, DOI: 10.1016/j.jjie.2007.09.001. http://www.sciencedirect.com/science/article/B6WMC-4RJK125-1/2/fd4cbe575fd0d918840d30e691150ed1 T. Bayoumi, The morning after: Explaining the slowdown in Japanese growth in the 1990s, Journal of International Economics 53 (2001) (2), pp. 241–259. http://www.nber.org/papers/w7350.pdf

BRAUN, R. ANTON; WAKI, YUICHIRO. Japanese Economic Review, Jun2006, Vol. 57 Issue 2, p324-344, 21p, 3 charts, 2 graphs. http://web.ebscohost.com/ehost/pdf?vid=3&hid=6&sid=bdd0c0d8-160e-44bc-a8da-243d3da0ed0d%40sessionmgr3 Fukao, Mitsuhiro. "Japan's Lost Decade and Its Financial System." World Economy 26.3 (Mar. 2003): 365384. EconLit. EBSCO. 15 Apr 2009 https://login.proxy.libraries.rutgers.edu/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=ecn&AN =0737221&site=ehost-live. F. Hayashi and E.C. Prescott, The 1990s in Japan: A lost decade, Review of Economic Dynamics 5 (2002) (1), pp. 206–235 http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6VF1-4JRVF1V2&_user=526750&_coverDate=12%2F31%2F2006&_alid=914194105&_rdoc=4&_fmt=high&_orig=search&_cdi= 5997&_sort=d&_docanchor=&view=c&_ct=10&_acct=C000023759&_version=1&_urlVersion=0&_userid=52675 0&md5=c3be8806dd1000306182b70604c6bdae#bbib14 Horioka, Charles Yuji. Japan & the World Economy, Dec2006, Vol. 18 Issue 4, p378-400, 23p. http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6VF1-4JRVF1V2&_user=526750&_coverDate=12%2F31%2F2006&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_acct=C00 0023759&_version=1&_urlVersion=0&_userid=526750&md5=0c02917d27925f9afedca2e266b89771 Koo, Richard C. International Economy, Fall 2008, Vol. 22 Issue 4, p69-73, 5p. http://web.ebscohost.com/ehost/pdf?vid=3&hid=6&sid=bdd0c0d8-160e-44bc-a8da-243d3da0ed0d%40sessionmgr3 Portier, Franck. "Comment 'A 'News' View of Japan's Lost Decade': Monetary Policy during Japan's Lost Decade." Japanese Economic Review 57.2 (June 2006): 345-357. EconLit. EBSCO. 1 May 2009 https://login.proxy.libraries.rutgers.edu/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=ecn&AN =0879108&site=ehost-live. Yoshikawa, Hiroshi. "Japan's Lost Decade: What Have We Learned and Where Are We Heading?." Asian Economic Policy Review 2.2 (Dec. 2007): 186-203. EconLit. EBSCO. 15 Apr 2009 https://login.proxy.libraries.rutgers.edu/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=ecn&AN =0991412&site=ehost-live.

NEWS SOURCES "Asia 'must cut export dependency'" BBC NEWS | News Front Page. 01 May 2009 http://news.bbc.co.uk/2/hi/business/8031456.stm "Fiscal Policy: The Concise Encyclopedia of Economics |." Library of Economics and Liberty. 02 May 2009 . "Japan cuts rates to zero." BBC NEWS | News Front Page. 01 May 2009 . "Japan's economy in quarterly dive." BBC NEWS | News Front Page. 12 Apr 2009 .

"Supply-Side Economics: The Concise Encyclopedia of Economics |." Library of Economics and Liberty. 12 Apr 2009 .