EXCESS MARKET VALUE, THE MULTINATIONAL ... - SSRN

3 downloads 0 Views 214KB Size Report
Mar 29, 1985 - Hofstra University. Abstract. The results of this study show that there is a positive and significant relationship between excess market value of ...
EXCESS MARKET VALUE, THE MULTINATIONAL CORPORATION,AND TOBIN'S q-RATIO Wi Saeng Kim * Baruch College-CUNY EsmeraldaO. Lyn** Hofstra University Abstract. The results of this study show that there is a positive and significant relationship between excess market value of multinational corporations and the degree of international involvement as measured by foreign sales percentage. However, the excess market value is not determined by the number of foreign subsidiaries, nor the interaction between foreign sales and the number of foreign subsidiaries. The relationship between excess market value and the two frequently used measures of monopoly power-concentration ratio and the Lerner Index-was also examined. This study found that the Lerner Index contributes significantly in explaining excess market value but concentration ratio does not. Furthermore, the positive and significant coefficients of advertising and R&D intensity serve as evidence that product market imperfections appear to play a bigger role in explaining the excess market value experienced by MNCs during our sample period. INTRODUCTION

A number of motivations have been assigned to the corporate international operations of multinational corporations.' Many extant theories on foreign direct investment recognize the existence of imperfections in the product or factor market. It is argued by some that multinational firms invest abroad because they obtain monopolistic advantages and hence can generate higher profits than those achieved by local competitors. Others argue that real asset diversification is the major motivation for investing in foreign countries.2 The empirical evidence regardingthe benefits to shareholders of multinational corporations (MNCs) arising from international diversification is, however, inconclusive. Agmon and Lessard (1977) found that the higher the degree of an MNC's international involvement, the * Wi SaengKim is AssistantProfessorof Economicsand Finance, Baruch College,The City University of New York. His researchhas centered on the agency problems in corporatedecisions,and also include FDI theories. ** EsmeraldaO. Lyn receiveda Ph.D. in Businessfrom BaruchCollege, The City University of New York, 1982, and is now an Assistant Professor at Hofstra University. Date Received: April 2, 1984; Revised: March 29, 1985; Accepted: June 25, 1985.

119

Palgrave Macmillan Journals is collaborating with JSTOR to digitize, preserve, and extend access to Journal of International Business Studies ® www.jstor.org

120

JOURNALOF INTERNATIONALBUSINESSSTUDIES, SPRING 1986

lower its market-assignedmeasure of systematic risk. Jacquillat and Solnik (1978), on the other hand, concluded that investing in multinationals is a poor substitute to international portfolio diversification. Recently, Errunza and Senbet (1981, 1984) investigated the existence of monopoly rents associated with international operations in a market-value theoretic framework. They argued that advantages arising from international operations, if any, would be "priced-out" in a rational and efficient market, and therefore an investigation of the rates of return realized by shareholders of MNCs will not capture the monopolistic rents for the MNCs. In the spirit of Errunzaand Senbet, we investigate the determinants of the excess market value of multinational corporations. To the extent that multinational corporations possess the monopolistic power to compete with local competitors, it is expected that the excess market value of MNCs is determined by the intensity of R&D expenditures and of advertising efforts, which are believed to be the fountainhead of the creation and maintenance of product market imperfections. EXCESSMARKETVALUE OF MULTINATIONAL CORPORATIONS

This note empirically assesses the effect of international operations in the market-value theoretic framework. As in Errunza and Senbet (1981), we make no a priori specification of the structure of the international capital market, nor do we postulate the degree of international capital market integration. We only rely on the functioning of the U.S. capital market and focus on U.S.-based multinationals' excess market valuation. We relate the excess market valuation to the degree of the MNCs' international involvement. The basic generalized relationship which serves as an underpinning for the model is VMNC = VD +VF

(1)

That is, the value of a multinational firm, VMNC,has two components: VD which represents the present value of cash inflows from domestic operations and VF which represents the capitalized value of the international operations. The hypothesis we want to test is whether there is a positive relationship between market value (VMNC) and degree of international involvement (DI), holding other variables constant. Since it is not such an easy task to hold constant other variableswhich influence market value, we deflate the market value of firms by a common yardstick. The common yardstick we employ in this study is the replacement costs of assets. As we divide equation (1) through by R, the replacement costs of the assets in place, we have the Tobin's q expression.3 VMNC= VD+VF

R

q

(2)

R

If securities of MNCs enjoy a premium because they take advantage of imperfections in the international market and U.S. investors recognize it,

EXCESSMARKETVALUE

121

then the current market value of MNCs will be a positive function of the degree of international involvement. Furthermore, extensive expenditure on advertising and R&D may create imperfections in the product market, such as entry barriers, thus enhancing the ability of firms to earn monopoly rents. Mueller and Rogers (1980), for instance, found that advertising creates and maintains product differentiation, which may lead to entry barriers. Grabowski and Mueller (1978) and Branch (1974) found that firms in research-intensive industries earn significantly above-average returns. We also include measures of future growth opportunities and of monopoly power. Thus, we express the functional relationship between Tobin's q and its explanatory variables as follows.4 qMNC = qMNC(DI,A,RD,G,MP)

where DI = A = RD= G = MP =

(3)

degree of international involvement for an MNC intensity of advertising expenditures intensity of research and development expenditure expected growth rate as measured by past growth proxy for monopoly power METHODOLOGY

Data Sources

Four major publications that report foreign involvement measures are used. The Standard and Poor's "Outlook" and Forbes report foreign sales figures. Moody's Directory of Corporate Affiliation and Dun and Bradstreet's Corporate Families and International Affiliates list the foreign affiliates of U.S.-based MNCs. The 1972 Census of Manufactures is used for data on monopoly power. The Standard and Poor's Compustat Industrial Tape is utilized for all the other firm-specific variables in the model specification. This study uses three different measures of foreign involvement, namely, (1) Foreign Sales as a percentage of total sales (FS), (2) Number of foreign affiliates (NOFA), and (3) FS x NOFA, an interactive term.5 The study also employs the two most commonly used measures of monopoly power, namely, (1) The four-firm concentration ratio (CON) which is based on the value of shipments of the four largest firms in its industry, and (2) the Lerner Index (LI) which is the percentage difference between price and marginal cost.6 The empirical work covers the sample period from 1974-1978. There are 154 firms in the sample with complete data for any of the years in the sample period. There are 458 observations in all. Testsand Results

A stepwise regression procedure called the maximum R2 improvement technique (MAXR) is used to test the significance of the variables in the

122

JOURNAL OF INTERNATIONAL BUSINESS STUDIES, SPRING 1986

model. MAXR calculates the R2 improvement and the F-statistics reflecting each variable'scontribution to the model if it were to be included. The MAXR method begins by finding the one-variable model producing the highest R2. It then adds another variable that would yield the greatest increase in R2 by comparing each variable with all other variables and doing all possible switches until no switch could increase R2. The two-variable model achieved is considered the "best" two variable model the technique can find. The comparing and switching process is repeated to find the "best" three variable model, and so forth. For testing purposes, the model specification is given as a linear form V/R = q =ao + a DI + a2A+ a3RD+ a4G+a MP+e

(4)

As in Thomadakis (1977) and Errunza and Senbet (1981, 1984), a proxy for q is employed in this note. Excess valuation (EV) of MNCsis calculated as the difference between the market value of common equity and its book value, normalized by sales, i.e., EV =

marketvalueof commonequity- net worth sales

Thus, equation (4) becomes EV = ao + a1

international involvement

[R&D expenditure

+ a3

a2 sales 1

+ a4 sales

sales -A sales t10

+ as [ degreeof monopolypower] + errorterm

1

(5)

The second DI proxy; NOFA, will be represented by the dummy variable, D, whereby it takes a value of one when the NOFA > 10, zero value otherwise. The interactive DI proxy will then become the product of FS and D (FSD). The proxy for monopoly power, LI, is measured by the percentage difference between sales and operating expenses.7 The test results for equation (4), using CON as a proxy for monopoly power, are presented in Table 1. The MAXR procedure entered RD first as the best one-variable model with R2 equal to 26.4%. As shown in the first panel, the predicted power of the model increased to 33.4% when G was included, to 34.7% for A, and to 35.6% for FS. The results indicate that the excess market value of MNCs is more strongly determined by imperfections in the product market arising from heavy advertising and R&D expenditures rather than from financial imperfection.

123

EXCESSMARKETVALUE

TABLE1 RegressionResultsUsingCONas a Measureof Monopoly Power Coefficients of Dl Proxy

Intercept

FS

-.486

R2 improvement

D

-.406

R improvement

FSD

R2

-.419

DI

A

RD

CON

G

(4) .003 5.76 .017

(3) 1.742 8.12 .005

(1) 10.237 119.51 .0001

(5) .001 .10 .751

(2) 2.580 45.66 .0001

.356

.347

(5) .007 .03 .857

(3) 1.790 8.33 .004

.346

.347

(5) .001 .70 .404

(3) 1.838 8.91 .003

.348

.347

.264 (1) 10.915 145.96 .0001 .264 (1) 10.765 138.65 .0001 .264

.356 (4) .001 .19 .660 .348 (4) .001 .180 .6701 .348

R2 for the 5-variable model

.356

.334 (2) 2.655 47.69 .0001

.348

.334 (2) 2.620 46.44 .0001

.349

.334

Note: The numbers in parentheses denote the sequence each variable entered the model. The second row in each panel represents calculated F-values, and the third row represents F-significance level.

When the Lerner Index is used as a proxy for monopoly power, Table 2 shows that the LI enters the model first with R2 equal to 49.7% for the best one-variablemodel. Thus, LI becomes the most important contributor to the explanation of the excess market value of MNCs. The predictive power of the model improved significantly when LI is used instead of CON. The behaviorof the other variablesis consistent with those in Table 1. With respect to the proxies for the degree of international involvement, the signs of the coefficients are consistent with those predicted by the model except for the dummy variable, D, for NOFA, but only FS is significant. As shown in the two tables, when the DI proxy is introduced to the model, R2 improved only for FS, and not for D or FSD. CONCLUSION

This note extended earlier work on the benefits of corporate multinationality by empirically examining whether the excess market value of multinational corporations is determined by the degree of international

124

JOURNAL OF INTERNATIONAL BUSINESS STUDIES, SPRING 1986

TABLE 2 Regression Results Using LI as a Measure of Monopoly Power Coefficients of Dl Proxy

Intercept

G

A

RD

(5) .002 3.05 .0812

(4) 1.099 4.57 .0330

(2) 4.356 23,71 .0001

(1) 8.931 187.00 .0001

(3) 1.138 11.91 .0006

DI

LI

R2

FS

-.744

D

-.697

.005 .02 .8931

1.144 4.84 .0283

4.686 23.38 .0001

9.027 190.93 .0001

1.155 12.09 .0006

.542

FSD

-.710

.001 1.68 .1951

1.175 5.20 .0231

4.508 25.86 .0001

9.403 192.33 .0001

1.120 11.42 .0008

.544

R2 improvement

.542

.525

.497

.545

.537

Note: The numbers in parentheses denote the sequence each variable entered the model. The second row in each panel represents calculated F-values, and the third row represents F significance level.

involvement and by the intensity of advertising and R&D effects. We employed market value in excess of book value normalized by sales (EV) as a proxy for the premium U.S. investors may pay for MNCs. The results of this study show that there is a positive and significant relationship between EV and the degree of international involvement as measured by foreign sales percentage. However, the EV is not determined by the number of foreign subsidiaries, nor the interaction between foreign sales and the number of foreign subsidiaries. The relationship between EV and the two frequently used measures of monopoly power-concentration ratio and the Lerner Index-was also examined. This study found that the Lerner Index contributes significantly in explaining EV but concentration ratio does not. Furthermore, the positive and significant coefficients of advertising and R&D intensity serve as evidence that product market imperfections appear to play a bigger role in explaining the excess market value experienced by MNCs during our sample period. NOTES 1. See Hymer (1966), Kindleberger (1969), Aggarwal (1977), and Ragazzi (1973) for a comprehensive review of the theories on the foreign direct investments. 2. See Rugman (1976). 3. See Tobin and Brainard (1977) and Smirlock (1984) for a discussion on the concept of q-ratio. 4. The measurement of q incorporates the risk factors, so we chose not to include it in our model. See Smirlock, et al (1984).

125

EXCESSMARKETVALUE

5. We thank an anonymous referee for suggesting that we should employ alternative measure of international involvement. There are other measures of international involvement such as foreign assets and foreign earnings. We employ the number of foreign subsidiaries because earlier studies indicated that foreign earnings measure may not be comparable due to non-uniformity of accounting methods. Furthermore, Errunza and Senbet (1981) found that these measures were not significant in explaining MNCs' excess market value. See Errunza and Senbet (1981) footnote 19. 6. See Scherer (1971) and Lindenberg and Ross (1981). 7. As in Lindenberg and Ross (1981), we assumed constant returns to scale and thus, average cost equals marginal cost.

REFERENCES Aggarwal, Raj (1977). "Theories of Foreign Direct Investment: A Summary of Recent Research and a Proposed Unifying Paradigm," Economic Affairs (September). Agmon, Tamir and Lessard, Donald R. (1977). "Investment Recognition of Corporation International Diversification," Journal of Finance (September): 1049-1055. Branch, Ben (1974). "Research and Development Activity and Profitability: A Distributed Log Analysis," Journal of Political Economy 82 (September/October): 999-1011. Errunza, Vihang R. and Senbet, Lemma W. (1981). "The Effects of International Operations on the Market Value of the Firm: Theory and Evidence," Journal of Finance (May): 401-417. Errunza, Vihang R. and Senbet, Lemma W. (1984). "International Corporate Diversification, Market Valuation, and Size-Adjusted Evidence," Journal of Finance 39 (May): 727-742. Grabowski, H. G. and Mueller, D. C. (1978). "Industrial Research and Development Intangible Capital Stocks, and Firm Profit Rates," The Bell Journal of Economics 9, No. 2; 32843. Hymer, S. H. (1966). "The International Operations of National Firms: A Study of Direct Foreign Investment" (Doctoral Dissertation, M.I.T.). Jacquillat, B. and Solnik, B. H. (1978). "Multinationals are Poor Tools for Diversification." Journal of Portfolio Management 4 (Winter): 8-12.

Kindleberger,C. P. (1969). AmericanBusinessAbroad:Six Lectureson Direct Investment,New Haven and London: Yale University Press. Lindenberg, E. and Ross, S. (1981). "Tobin's 'Q' Ratio and Industrial Organization," Journal of Business 54:1-32. Mueller, Willard F. and Rogers, Richard, T. (1980). "The Role of Advertising in Changing Concentration and Manufacturing Industries," The Review of Economics and Statistics (February): 7-13. Ragazzi, G. (1973). "Theories of the Determinants of Direct Foreign Investment," IMF Staff Papers 20 (July): 471-98. Rugman, A. (1976). "Risk Reduction by International Diversification," Journal of International Business Studies, 7 (Fall/Winter): 75-80. Scherer, F. M. (1971). Industrial Market Structure and Economic Performance, New York: Rand McNally & Co. Smirlock, M., T. Gilligan, and W. Marshall (1984). "Tobin's q and the Structure-Performance Relationship," American Economic Review (December): 1051-1060. Thomadakis, Stavros (1977). "A Value-Based Test of Profitability and Market Structure," Review

of Economicsand Statistics(May): 179-85. Tobin, J. and Brainard, W. (1977). Asset Markets and the Cost of Capital. In B. Belassa and R. Nelson (eds.) Economic Progress, Private Values and Public Policies, North Holland, Amsterdam.