Chapter 3 Cost of capital and Investment

Transcription

1 Chapter 3 and Investment Masashi Manabe Satoko Maekawa 1. Introduction As one of the economic stimulus measures, reduction of corporate tax burden has been discussed in order to cope with globalization and international severe competition. In such discussions it should be necessary to make clear the effective level of corporate tax burden and the effects of reduction of corporate taxation on investment in Japan. In Chapter 1, we estimated and compared the effective rates of corporate tax in Japan as well as in other major developed countries. In Chapter 2, we calculated tax-adjusted costs of and the contribution of tax factor to cost of. The results of those analyses indicate that it is still possible to lower the corporate tax burden in Japan and to reduce the costs of by such a reduction. Could the cost of really affect the corporate investment? If so, how much could the reduction of corporate tax raise the investment? In order to answer these questions we analyze the relationship between the tax-adjusted cost of and investment in Japan. Theoretically Jorgenson (1963) formulated the companies investment function with an intellection of cost of, which is one of the indicators made by integrated funding cost, investment-goods prices, rate of depreciation allowance and corporate tax rate. As a matter of course, the corporate value maximization model derives this connection. Based on the theoretical development, a Jorgenson-type investment function was often used for estimation in Japan. Representative studies are Takenaka, Takabayashi, Tsukakoshi, Kuwana and Yoshida (1986) and Tajika and Yui (1988). 1 1 Other studies that analyzed the connection between investment and taxation system are Tajika and Yui (1984, ), Honma, Hayashi, Atoda and Hata (1984), Honma, Iwamoto, Asada, 44

2 As globalization has been progressing, the factors that should be considered in investment decision are becoming more diverse than ever before. Global companies certainly have to consider the political conditions, economic situation and taxation system in a potential investment destination. Focusing oneself solely on the investment inflow to Japan, traditional factors such as interest rates for funding, goods prices and land prices weigh heavily for the decision-makers. However, studies about corporate investment have been concentrated on certain financial aspects. Issues like bad-loan problems have been the main focus and have neglected overall investment determinants analysis including the taxation system. This is mainly due to Japan s recent macroeconomic situation. In the period of the economic bubble we saw rising stock prices and land values steeply and after that experienced the substantial decline in them as the bubble burst. A long economic slowdown followed along with policy failure. Corporate investment could have been influenced in the past two decades by the changing economic climate; economic bubble period, period of fallout from the bursting bubble, deflationary period and minor recovery period. This paper will examine how the relation between cost of and investment changed in this turbulent period and will quantitatively measure those changes. Beyond any doubt, corporate investment is influenced not only by the cost of but also other economic variables and environmental variables. Hence these factors will be considered as well. Moreover, this paper will investigate the effects of reduction of corporate tax rate on promotion of corporate investment. In particular, investment functions will be estimated using the cost of, which was calculated in the previous chapter. We will be look at three categories of industry that have been stipulated by the Japan Standardized Industrial Classification Code. The first category is large groups by industry, which is made up of nine major industries such as agriculture, forestry and fisheries, mining, manufacturing, etc. The second one is 15 manufacturing industries of medium groups by industry, which is mainly made up of food and tobacco, textile, chemicals, etc. And the third category is size. Based on the results of investment functions we will estimate how much investment could be stimulated by reduction of corporate tax rate. The following is the contexture of this chapter: In Section 2, we will explain the investment Sunagawa and Sano (1989), Iwata, Suzuki and Yoshida (1987). In the 199s, Hayashi and Inoue (1991) conducted the empirical analysis about Q and business growth and Asako, Kuninori, Inoue and Murase (1997) estimated the investment function which includes the land investment. Uemura and Maekawa (2) is a recent study. 45

3 function model and estimation method. In section 3, estimation results will be examined in order of the three industrial categories. As a policy implication, in section 4, we will derive the impact of a 1% corporation tax rate reduction to investment based on the previous section results. Then we will conclude by mentioning the remaining future issues. 2. Model and Data 2-1 Estimation Model Neoclassical-type investment function defined by Jorgenson (1963) can also be derived by a dynamic optimized approach that maximizes present discounted value of net cash flow. The simplest investment function determining the production only by and labor is: It = It ( Ct, K t 1) (1) Where It denotes the investment in the period t, K t 1 denotes the stock in the period t-1 and Ct denotes the cost of in the period t. In actual estimation, data including the so-called bubble period is used. Hence supposing the production function which includes the land as variable, the following estimation model is used: I K t t 1 = α + βc + γp t L t + u t (2) L P t t Where indicates the land price in the period t and u indicates the disturbance term in the t. Now, let us explain the sign condition. Degradation of the cost of stimulates the accumulation, then β <. Regarding the land price, the rise in land prices facilitates the fund procurement pledging the land and stimulates investment, hence γ>. During a bubble period, 46

4 corporations put up the land as collateral for fund-raising Data and Estimation Method We will estimate the investment functions by industry according to the categories set by the Japan Standardized Industrial Classification Code and we also estimate by size. The estimation period is from to 25. Table3-1 summarizes the data used for each variable in the model. Basically, other tangible fixed asset in "Houjin, Kigyou, Toukei, and Kihou,( Financial Statement Statistics of Corporations )" (Ministry of Finance) is used as stock (K) and the amount of the investment (I) is measured by ordinary formula of accumulation. Depending on the industries type and scale, investment is not always done every year. Some companies chose large investment once a several year. As represented by the construction industry, fluctuation of investment for each industry is considerably large. Hence tertiary moving average derivation is used. data calculated by the model built in the statutory tax rate in the Chapter 2 is used as cost of. Land price is commercial land price in six major metropolitan areas of Urban Land Price Index (Japan Real Estate Institute). We estimated equation (2) using Ordinary Least Square (OLS). The data ranges are from the bubble period in the late 198s, the depression period after bubble period in the middle of the 199s, the deflation period over the late 199s to 2. Hence dummy coefficients of cost of are included in and after 98 years, because structural change of the relationship between investment rate and cost of might have occurred. 3. Estimation Results 3-1 Results in Large Groups by industry Table3-2 shows the estimation results of the all industries and the nine individual industries. None of the coefficient of cost of is significantly negative through the whole period. In only 2 Ogawa and Suzuki (2) estimated the investment function building the land price into the independent variables and derived the significant result. 47

5 the manufacturing and electricity, gas & water supply industries, adjusted R square are over.9. Then there is a possibility that significant determinants cannot be considered. On the other hand, coefficient of cost of after 1992 is significantly negative in many industries, including manufacturing, construction, wholesale and retail trade and information & telecommunications. In particular, by dividing the dummy variables from 1992 to 1997 and after 1998, better results are estimated. Coefficient of cost of through the whole period is not significant hence it can be considered as. Therefore, in the case of all industries coefficient of cost of from 1992 to 1997 is and that after 1998 is -3.. In the case of manufacturing, coefficient of cost of until 1997 is and that after 1998 is Until 1997, construction is and after 1998 is Regarding other industries, coefficient of real estate and information & telecommunications is relatively high; and each after The coefficient is significant for land price in manufacturing, construction, wholesale and retail trade, and the service activities industries. This would be caused by the abnormal investment (behavior) pledging of the land in the bubble period when the coefficient of the stock is not significant. In Table3-2, showing estimation results, some industries coefficient of cost of through the whole period is significantly positive against the theory. These are agriculture, forestry, the fisheries industry and electricity, gas & water supply industry. Also, at no point did mining show any significance. The relationship between investment and cost of might be disturbed within these four industries due to other factors. For example, electricity, gas & water supply industry is greatly influenced by the price of oil. The other three industries are labor-intensive industries and scale is fairly small. Hence assumption of corporate value maximization activity of the theoretical model deriving the investment function cannot be necessarily applied to these industries and the result was not significant. 3-2 Results in 15 Manufacturing in the Medium Groups by industry Next, we will present Table3-3 to see the estimation results for the 15 industries in manufacturing within the medium category. Regrettably, coefficient of cost of after 1992 is significantly negative only in the three industries; general machinery, electrical machinery and other manufacturing. Coefficient of general machinery after 1992 is -1.21, electrical machines are 48

6 -1.43 and other manufacturing is Those results are consistent with the results of total manufacturing in Table3-2, which was from 1992 to 1997 and after However, a number of industries coefficients are significantly positive even after 1992 and some industries are not significant. Furthermore, nine industries coefficients are significantly positive over whole period. (textiles, pulp, paper and paper products, chemicals, petroleum and coal products, iron and steel, non-ferrous metals, fabricated metals, transport equipment, precision instruments) On the other hand, eight industries land price variables are significantly positive. (foods and tobacco, chemicals, petroleum and coal products, ceramic, stone and clay products, non-ferrous metals, fabricated metals, other manufacturing) Also six industries land price variables are significantly positive and coefficients of cost of for all period are significantly positive as well. These results can be seen as showing how much Japanese manufacturing can be influenced by a bubble. More industries in this category did not meet the sign condition compared to the large category. One of the reasons for these results would be considered that data of Houjin, Kigyou, Toukei, Kihou Financial Statement Statistics of Corporation- (Ministry of Finance) is a sample survey, hence more impacts for each company are reflected in focusing on the detail. However, we can hardly avoid the conclusion that only general machinery, electrical machinery and other manufacturing invested following the corporate value maximization model, which is assumed in the theory, and other industries investment is influenced by the bubble economy and specific factors of each. 3-3 Results by Capital Size Finally, we will focus on the estimation results by size in Table3-4, which classified companies into three categories: big, medium and small. Small size companies (less then 1 million yen) coefficient is not significant in any period. This could because most of the companies size is small and they do not go on the market. So, they do not choose to follow the corporate value maximization model. In medium size companies (more than 1 million yen, but less than 1 million yen), the coefficient is significant negative through every period. Coefficient of the whole period is and that of after 1992 is fairly high: (= ). Coefficient of land price is also 49

7 significantly positive. It can be considered that medium size companies used land-collateral loans during the bubble period but also achieved sound management using the cost of as an objective indicator. On the other hand, big-size companies (more than 1 million yen) coefficients of cost of during the whole period and from 1992 to 1997 are not significant either. This could be down to sloppily management during the bubble period and after the collapse. However, the coefficient after 1998 is -1.75, which is significant. Big-size companies coefficient is smaller than that of middle- size companies. It could be thought that larger effects of the burst bubble remain in big-size companies and they are unable to respond in situations to change of the cost of. 4. Conclusion We calculate the impact of a 1% corporation tax rate reduction on investment with the previous chapter results about the relationship between the corporation tax rate and cost of. The independent variable in the equation is investment rate, hence the amount of investment increase caused the by a 1% corporation tax rate reduction can be derived by multiplying the variation of cost of, in the results of the previous chapter, by the coefficient in the estimation results from Table3-2 to Table3-4 and then by the amount of the stock. Table3-5 shows the calculation results of the stimulated amount of investment only for the industries in which coefficients of cost of were significant: all industries, five industries within the large category (manufacturing, construction, wholesale & retail trade, real estate, information and telecommunications), and big and medium sized companies. By industry, the largest stimulated amount is within information and telecommunications, which is less than 23 billion yen. The second is less than 17 billion yen for manufacturing. The third is over 1 billion yen for construction. Stimulated investment amount across all industries can be expected over 78 billion yen. By size, stimulated amount within big-size companies can be estimated at less than 41 billion yen and that of medium-size companies can be expected to be over 58 billion yen. If anything, a reduction of the corporation tax rate would possibly stimulate the investment of 5

8 medium-size companies. This chapter has examined quantitatively the relationship between investment and cost of built in corporation tax. As a result, a negative relationship between cost of and investment was observed. Obviously, using the land price variable also develops aberrant investment for some industries in the bubble period. Despite a drop in financial revenue for the government due to a 1% reduction in corporation tax the national economy is greatly boosted by the inflow of further corporate investment. This investment inflow outstrips any tax loss. The remaining issue is the fact that estimation results fail to offer strong explanation of the results. More accurate and robust estimations with additional variables are necessary. Also, some coefficients of cost of in manufacturing are not significant. That might be led by the small-size industries but cause investigation is required. Trying data collaboration could do this. This is an issue for the future. 51

9 Reference Asako, K. M. Kuninori, Inoue, T. and H. Murase (1997), Setsubi Toushi to Tochi Toushi: , [Investment and Land Investment] in Asako. K and M. Otaki (Ed.) Gendai Makuro Keizai Dougaku [Modern Macro Economic Dynamics], Tokyo Daigaku Syuppan Kai, in Japanese. Hayashi, F and T, Inoue (1991), The Relation Between Firm Growth and Q with Multiple Capital Goods; Theory and Evidence from Panel Data on Japanese Firms, Econometrica, 59, pp Honma, M. Y. Iwamoto, Asada, T. K. Sunagawa and H. Sano (1989), Setsubi Toushi no Zissyo Bunseki, [Empirical Analysis on Investment], Financial Review, 9, pp.56-95, in Japanese. Honma, M. H. Hayahi, Atoda. N and K. Hata (1984), Setsubi Toushi to Kigyo Zeisei, [Investment and Corporate Taxation System], Keizai Kikaku Cho Keizai Kenkyu Zyo Kenkyu Series, 41, in Japanese. Iwata, K. I. Suzuki and A. Yoshida (1987), Setsubi Toushi no Sihon Kosuto to Zeisei, [ Capital and Taxation System of Investment], Keizai Bunseki [Economic Analysis] 17, pp.1-72, in Japanese. Jorgenson, D.W. (1963), Capital Theory and Investment Behavior, American Economic Review 53, pp Ogawa, K. and K. Suzuki (2), Demand for Bank Loans and Investment under Borrowing Constraints: A Panel Study Japanese Firm Data, Journal of the Japanese and International Economies 14, 2, pp Takenaka, H. K. Takabayashi, Tsukakoshi, Y. Y. Kuwana and Y. Yoshida (1986), Shihon Kosuto no Kokusai Hikaku, [International Comparison of Capital], Financial Review, first issue, pp.29-59, in Japanese. Tajika, E and Y. Yui (1984), Sengo Nihon no Houzin Zeisei to Setsubi Toushi-Houzin Zei Keigen Ritsu no Gyousyu Betsu Keisoku wo Chuushin to site. [Corporation Taxation System and Investment in the Post War Era-With a Central Focus on Measuring Method by sector for Corporation Reduced Tax Rate], Kikan Gendai Keizai, [Modern Economics], 59, pp.29-49, in Japanese Tajika, E. and Y. Yui (), Sengo Nihon no Houzin Zeisei to Setsubi Toushi, [Corporate Taxation 52

10 System and Investment in the Post War Era] in Ito, K. A. Kousaka and Tajika, E (Ed.) Keizai Hatten to Zaisei Kinyu, [Economic Development and Finance], pp , Institute of Developing Economics, in Japanese. Tajika, E. and Y. Yui (1988), Shihon Kosuto to Houzin Zikkou Zeiritsu, [ and Effective Corporation Tax Rate], Keizai Kenkyu,[Economic Research], 39.2, pp , Hitotsubashi University, in Japanese. Uemura, T. and S. Maekawa (2), Sangyo Betsu no Toushi Koudou to Houjin Shotokuzei-Kigyo Zaimu Data wo Riyou sita Tax-Adjusted Q ni yoru Zissyou Bunseki, [Investment and Corporate Income Tax by industry- Empirical Analysis of Tax-Adjusted Q with Corporate Financial Data], Nihon Keizai Kenkyu, [Japanese Economic Research], pp.45-7, Japan Center for Economic Research, in Japanese. 53

11 Table3-1 Data Used for Estimation of Investment Function K Capital Stock Variable Statistics Calculation Methods "Other Tangible Fixed Assets" in Financial Statements Statistics of Corporations by Industry, Quarterly, Ministry of Finance P L Land Price Urban Land Price Index, Japan Real Estate Institute C Capital C estimated in Chapter2 π Capital Goods Inflation Rates π estimated in Chapter2 I Investment I(t)=K(t)-{1+π(t)}K(t-1) 54

12 Table3-2 The Estimation Results of Investment Function : All Industry and the Major Groups by industry Industy Dependent variable Test statistic All industies constant Land price adjr 2 DW (investment rate) all period 1992~ ~25 coefficients t-value p-value Agriculture, forestry & fishing constant adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Mining constant adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Manufacturing constant Land price adjr 2 DW (investment rate) all period 1992~ ~25 coefficients t-value p-value Construction constant Land price adjr 2 DW (moving average) all period 1992~ ~24 coefficients t-value p-value Electricity, gas & water supply constant adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value.2..4 Wholesale & retail trade constant Land price adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Real estate constant adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Information & telecommunications constant adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Service activities constant Land price adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value

13 Table3-3 The Estimation Results of Investment Function: 15 Manufacturing in the Medium Groups by Industry: No1 Industy Dependent variable Test statistic Food and tobacco constant Land price adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Textile constant adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Pulp, paper and paper products constant adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Chemicals constant Land price adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Petroleum and coal products constant Land price adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Ceramic, stone and clay products constant Land price adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Iron and steel constant Land price adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Non-ferrous metals constant Land price adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value

14 Table3-3 The Estimation Results of Investment Function: 15 Manufacturing in the Medium Groups by Industry: No2 Industy Dependent variable Test statistic Fabricated metals constant Land price adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value General machinery constant adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Electrical machinery constant adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Transport equipment constant adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Precision instruments constant Land price adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Newspaper publishing and publishing constant adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Other manufacturing constant Land price adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value

15 Table3-4 The Estimation Results of Investment Function : by Asset Size Size of Total Assets Dependent variable Test statistic Big business constant adjr 2 DW (investment rate) all period 1992~ ~25 coefficients t-value p-value Middle-scale business constant Land price adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value....1 Small business constant adjr 2 DW (moving average) all period 1992~24 coefficients t-value p-value Table3-5 The Effects of a 1% Corporation Tax Rate Reduction on Investment Amount change for cost of Coefficients of investment function Amount of stimulated investment Stock of (A) (B) (C) (A) (B) (C) unit - - million yen million yen industy All industies Agriculture, forestry & fishing Mining Manufacturing Construction Electricity, gas & water supply Wholesale & retail trade Real estate Information and telecommunications Service activities Size of Total Assets Big business Middle-size business Small business

16 Figure A3-1 Capital (C) and investment rate (I/K) in All industries 5 5 Figure A3-2 Capital (C) and investment rate (I/K) in Agriculture, Forestry & Fishing

17 Figure A3-3 Capital (C) and investment rate (I/K) in Mining 5 5 Figure A3-4 Capital (C) and investment rate (I/K) in Manufacturing 5 5 6

18 Figure A3-5 Capital (C) and investment rate (I/K) in Construction 5 5 Figure A3-6 Capital (C) and investment rate (I/K) in Electricity, Gas & Water supply

19 Figure A3-7 Capital (C) and investment rate (I/K) in Wholesale & Retail trade 5 5 Figure A3-8 Capital (C) and investment rate (I/K) in Real estate

20 Figure A3-9 Capital (C) and investment rate (I/K) in Information & Telecommunication 5 5 Figure A3-1 Capital (C) and investment rate (I/K) in Service activities

21 Figure A3-11 Capital (C) and investment rate (I/K) in Food & Tobacco 5 5 Figure A3-12 Capital (C) and investment rate (I/K) in Textile

22 Figure A3-13 Capital (C) and investment rate (I/K) in Pulp, Paper & Paper products 5 5 Figure A3-14 Capital (C) and investment rate (I/K) in Chemicals

23 Figure A3-15 Capital (C) and investment rate (I/K) in Petroleum & Coal products 5 5 Figure A3-16 Capital (C) and investment rate (I/K) in Ceramic, Stone & Clay products

24 Figure A3-17 Capital (C) and investment rate (I/K) in Iron & Steal 5 5 Figure A3-18 Capital (C) and investment rate (I/K) in Non-ferrous metals

25 Figure A3-19 Capital (C) and investment rate (I/K) in Fabricated metals 5 5 Figure A3-2 Capital (C) and investment rate (I/K) in General machinery

26 Figure A3-21 Capital (C) and investment rate (I/K) in Electrical machinery 5 5 Figure A3-22 Capital (C) and investment rate (I/K) in Transport equipment

27 Figure A3-23 Capital (C) and investment rate (I/K) in Precision instruments 5 5 Figure A3-24 Capital (C) and investment rate (I/K) in Newspaper publishing & Publishing 5 5 7

28 Figure A3-25 Capital (C) and investment rate (I/K) in Other Manufacturing 5 5 Figure A3-26 Capital (C) and investment rate (I/K) in Big business

29 Figure A3-27 Capital (C) and investment rate (I/K) in Middle-scale business 5 5 Figure A3-28 Capital (C) and investment rate (I/K) in Small business