To the extent that business cycles tend to behave independently across different countries and that there exist imperfections in international markets for products, capital, and other factors, foreign operations by firms should provide their stockholders risk-return opportunities superior to those available to the stockholders of purely domestic firms. In this paper, we examine the empirical evidence with regard to these advantages of investing in multinational firms as a method of diversifying internationally, investigating the issue of investor recognition of the advantages enjoyed by a multinational firm.
This issue is important not only for investors but also for corporate management, for example, when formulating corporate strategy (see, for example, Hisey and Caves ). It is hypothesized that, consistent with the capital asset pricing model, international diversification of real assets by firms is rewarded in their domestic capital markets by a reduction in their systematic risk (beta) and by an increase in their price/earnings ratios. Further, it is hypothesized that the strength of these relationships is related to the degree of international involvement.
An early application of portfolio theory in an international context was by Grubel . He demonstrated that for the individual portfolio investor, risk reduction is facilitated by holding a diversified portfolio of international securities. These results have been subsequently confirmed and extended by Levy and Sarnat , Grubel and Fadner , Solnik , Agmon and Lessard , and Solnik and Noetzlin . These empirical studies have confirmed benefits from international diversification at the shareholder level in the form of risk-adjusted returns that are superior to those achievable in a single national market.
However, there are many restrictions that may prevent investors from achieving an efficient, internationally diversified portfolio. Among the barriers to investments in individual foreign markets are higher information processiong and transactions costs in the form, for example, of a general lack of information about foreign securities, the peculiar nature of trading and brokerage arrangements in the foreign capital market, lack of liquidyt, foreign exchange and other controls on the movement of capital, and the fear of expropriation and other political risks. To avoid these limitations of international portfolio diversification, it has been suggested that investors could achieve the advantages of international diversification by investing in domestically traded multinational firms.
A large number of researchers (Severn , Hughes, Logue, and Sweeney , Rugman , Agmon and Lessard , Miller and Pras , Aggarwal [1,2], Mikhail and Shawky , Barone , Kumar , Yoshihara , Grant , Aggarwal and Soenen , Daniels and Bracher , and Brewer  have provided empirical evidence of a positive relationship between international involvement and firm performance. These studies reported that multinational firms tend to be more profitable) measured as return on assets, return on sales, return on equity, and the price-earnings ratio) and/or showed lower risk (in terms of diversification of the unsystematic component of risk, beta coefficients, probability of insolvency, and equity variability) than domestically-based firms.
However, some authors )Severn , Rugman , Barone , and Aggarwal and Soenen ) have indicated that benefits of international diversification via the multinational firm seem to be deteriorating over time and will eventually fade away as the integration of economies proceeds.
A smaller number of researchers (Jacquillat and Solnik , Brewer [6, 7], Mathur and Hanagan , Fatemi , Shaked , and Michel and Shaked ) have provided empirical evidence that when both risk and return were considered, the risk-adjusted returns of multinational firms do not (always) outperform those provided by purely domestic firms. Therefore, according to them, multinational firms cannot be relied upon to provide the complete benefits available from international portfolio diversification.
Thus, there seems to be some controversy regarding the continuing diversification advantages of multinational firms. This paper attempts to provide further empirical evidence regarding the stock market valuation of multinationality using more recent data (1978-1986) than previous studies did. A clear distinction is made in the type of international involvement as measured by the percentage of foreign assets, the percentage of foreign sales, and the atio of foreign to total profits. In addition to the level of international involvement, this study also considers the variability of the three measures of multinationality. Stock market recognition of multinationality is further examined for two subsamples corresponding to firms with a relatively high and a low degree of overseas investment. The relationship between risk/return performace of MNFs' common stock and the different measures of multinationality are also examined for a strong dollar cycle (1979-1984) and a weak dollar cycle (1985-1986). Finally, the productive use of foreign assets and foreign sales is used as a variable to investigate stock market appreciation of corporate multinationility.
DATA SOURCES AND RESEARCH DESIGN
The sample of multinational firms examined in this study is the 46 manufacturing MNFs for which complete data are available for the period 1978-1986 taken from the annual issues of Forbes that list the largest MNFs and their percentage of foreign assets, foreign sales, and foreign profits. Oil and service companies were excluded to obtain a more uniform sample of manufacturing MNFs. In this study, the overseas involvement of the U.S.-based MNFs is measured not only by the percentage of foreign assets (% FA) but also by their percentage of foreign sales (% FS) and percentage of foreign profits (% FP). The relationship between multinationality and stock market valuation is examined by regressing the market's measure of systematic risk (beta) and the price-earings ratio (as a surrogate for the quality of the expected return) against the three measures of multinationality (mean values and standard deviations) for each company over the period 1978-1986. In equation form, the six regressions estimated are:
BETA = [a.sub.0] = [a.sub.1] %FA + [a.sub.2] SD (%FA) (1)
BETA = [b.sub.0] + [b.sub.1] %FS + [b.sub.2] SD (%FS) (2)
BETA = [c.sub.0] + [c.sub.1] %FP + [c.sub.2] SD (%FP) (3)
P/E = [u.sub.0] + [u.sub.1] %FA + [u.sub.2] SD (%FA) (4)
P/E = [v.sub.0] + [v.sub.1] %FS + [v.sub.2] SD (%FS) (5)
P/E = [w.sub.0] + [w.sub.1] %FP + [w.sub.2] SD (%FP) (6)
For each year, the beta coefficients used were taken from the Value Line issue of the end of that year, while the P/E ratios were collected from that year's Standard & Poors.
To examine the impact of the relative strength of the U.S. dollar on the relation between multinationality and stock market valuation, it is hypothesized that in a weak dollar cycle, international involvement via foreign sales is more favorable than having foreign assets and vice versa in a strong dollar cycle. Finally, the productive use of foreign assets or foreign sales is investigated as a variable to explain stock market reaction to multinationality. In this regard, the average beta- and P/E-values of the 46 MNFs are regressed against the ratio of the percentage foreign profits to the percentage foreign assets and the ratio of the percentage foreign profits to the percentage foreign sales. The estimated regression equations are:
BETA = [p.sub.0] + [p.sub.1] %FP/%FA (7)
BETA = [q.sub.0 + [q.sub.1] %FP/%FS (8)
P/E = [r.sub.0] + [r.sub.1] %FP/%FA (9)
P/E = [s.sub.0] + [s.sub.1] %FP/%FS (10)
Multiple regression analysis is used to test the hypothesized positive relation between the average P/E ratio over the nine year period (1978-1986) and the degree of overseas involvement and the assumed negative relation between the average beta-value and different measures of multinationality (i.e., that [u.sub.1], [v.sub.1], [w.sub.1] > 0 and [a.sub.1], [b.sub.1], [c.sub.1] < 0 in equation (1) to (6)). An opposite relationship is assumed with regard to the standard deviation of the measures of multinationality. This hypothesis implies that variability in the degree of international involvement, as measured by the standard deviation of % FA, % FS, and % FP, shows a negative relation with the P/E ratio and a positive relation with systematic risk (i.e., that [u.sub.2], [v.sub.2], [w.sub.2] < 0 and [a.sub.2], [b.sub.2], [c.sub.2] > 0 in equations (1) to (6)). Because all three measures of involvement are highly correlated (i.e., CORR (% FS, % FA) = 0.86, CORR (% FS, % FP) = 0.68), the regression lines are estimated using only one measure at a time to avoid the problem of multicollinearity. The regression results are summarized in Table 1.
[TABULAR DATE OMITTED]
As Table 1 shows, no significant relationship is observed between multinationality as measured by the percentage of foreign assets, foreign sales, and foreign profits and the market measure of systematic risk. None of the regression coefficients is significant at the 0.10 level (see t-values between parentheses) except for the regression coefficient of the variability in the percentage foreign profits. The lack of any significant relationship between beta and the percentage of foreign assets (foreign sales and foreign profits) indicates that the U.S. capital market does not value an increasing proportion of international activities by reducting proportionally the systematic risk associated with such firms. As the adjusted [R.sup.2.-values] show, the proportion of changes in the systematic risk of U.S. MNFs explained by changes in the percentage of foreign assets or foreign sales is close to zero. This finding is confirmed when analyzing the regression results for the relationship between the P/E ratio and the three measures of multinationality. As can be seen from Table 1, no significant relationships exist, and the stock market does not seem to be paying any premium for multinationality in the firms' operations.
To further examine stock market recognition of overseas investment, the 46 sample MNFs are classified into two groups according to the degree of international involvement as measured by the ratio of foreign assets to total assets. First, there is a subsample of 16 firms with at least 40 percent of their assets invested abroad. The second group consists of 18 firms whose foreign assets represent less than 25 percent of total assets. The cut-off points for the ratio foreign assets to total assets are selected so that two sufficiently large samples result to perform a regression analysis. However, we would like to emphasize that our findings must be accepted guardedly as the sample size is rather small. Regression results using equations (1) to (6) for both groups of companies are summarized in Table 2.
As Table 2(a) shows, no significant relationship (at the 0.10 or better level of significance) is observed between multinationality and beta for the group of firms with an extensive overseas involvement. As a consequence, the resulting [R.sup.2.-values] and regression coefficients, although mostly showing the expected sign, are all close to zero. These results provide further evidence that the percentage
[TABULAR DATA OMITTED] of FA, FS, or FP of U.S.-based firms has no significant impact on lowering the systematic risk for investors. A similar conclusion follows concerning the relationship between the P/E and the different measures of multinationality. However, a significant negative impact on the P/E ratio is found for the percentage FA (at the 0.10 level) and for the standard deviation of the percentage FP (at the 0.05 level).
The results are quite different for the firms with relatively low international involvement (see Table 2(b)). There is a very significant relationship (at the 0.01 level) between the P/E ratio and the percentage of foreign assets. The hypothesized relationship between P/E and the percentage foreign sales is also supported but at a better level of significance (0.05). These results support the hypothesis that invesyors pay a premium for the benefits of international diversification enjoyed by the multinational firm. However, no significant relation is found with regard to the percentage foreign profits. On the other hand, the beta value of the MNF's stock shows a significant association with the percentage foreign assets but not with the two other measures of multinationality.
The results of Table 2 suggest that the stock market seems to recognize multinationality in the firms' operations when international involvement is low (less than 25 percent) but not when companies realize a major portion of their business (more than 40 percent) overseas.
Next we examine if a strengthening or weakening of the U.S. dollar has any bearing on the hypothesized relationship between the stock market risk/return parameters and the degree of multinational operations. The regressions (1) to (6) are estimated for a 6-year strong dollar cycle (1979-1984) and a 2-year weak dollar cycle (1985-1986). Table 3 summarizes the result of these regressions. In addition, it is assumed that in a weak dollar cycle, international involvement via foreign sales is more favorable than having foreign assets. A depreciation of the dollar is expected to stimulate foreign sales, while foreign assets would only result in translation gains that have no cash flow impact. It is also assumed that a strong dollar would hurt foreign sales and lead to translation losses. Overseas involvement because of foreign sales is, therefore, expected to be affected more unfavorably by a dollar appreciation than cases of foreign direct investment. Cash flow losses outweigh translation losses in an efficient market setting.
The result do not support any of the assumptions made with regard to the value of the U.S. dollar and reaction of the U.S. stock market to multinationality. They provide more evidence of the general lack of a significant market recognition of the firms' international involvement.
Previous findings deal with capital market reaction to the level of multinationality as measured by the percentage of foreign assets, sales, and profits to their total values. However, market participants might react to the degree of profitability of foreign involvement rather than the level of overseas operations. Table 4 shows, in accordance with equations (7) to (10), the regression results for the relationship between the market measures of risk/return performance (beta and P/E) and the ratios of the percentage of foreign profits to the percentage of foreign assets and the percentage of foreign profits to the percentage of foreign sales for the period 1978-1986.
[TABULAR DATA OMITTED]
The regression results reported in Table 4 show that no significant relation is found between the P/E ratio and the relative profitability of foreign assets and foreign sales. Again, the results provide evidence that investors are not paying a premium for the relative profitability of MNF's foreign operations. However, a significant positive relationship is demonstrated between beta and the ratio of percentage foreign profits to the percentage of foreign assets (at the 0.01 level) and the percentage of foreign sales (at the 0.05 level). This result implies that systematic risk of MNF's stock show a positive association with the profitability of their international involvement as measured by foreign assets and foreign sales. The positive sign of the regression coefficient indicated that foreign profits may be viewed by the investor as more risky than profits generated domestically because they are influenced by exchange risk and political risk. The higher beta-values would then also explain the negative sign of the coefficients in the regressions with the P/E ratio as the dependent variable.
[TABLUAR DATA OMITTED]
In general, the results of this study show that for the 1978-1986 period, there seems to be no significant indication that U.S. investors reward U.S. firms for going overseas. The results do not support the traditional assumption that multinationality is rewarded with a higher P/E ratio or a reduction in the systematic risk. This finding is confirmed when comparing the results for a strong dollar cycle versus a weak dollar cycle. The general results, therefore, are consistent with the conclusions arrived at by Jacquillat and Solnik , Brewer [6, 7], Mathur and Hanagan , Fatemi , Shaked  and Michael and Shaked  that MNFs provide no discernable benefits over purely domestic firms with regard to the risk/return combinations desired by investors. Since we use a similar research design but a more recent data base compared with prior studies that found positive effects of multinationality, one explanation is that the extent of multinationality in a firm's operations no longer seems to be rewarded in U.S. capital markets. This recent decline in the strength of the relationships between degree of multinationality and systematic risk (beta) and P/E ratios is consistent with the assertion that international capital markets are becoming increasingly integrated with an apparent decrease in the benefits of international diversification undertaken by investing in MNFs.
Nevertheless, a significantly higher P/E ratio and a less significant reduction in systematic risk were found to be associated with the level of international operations for those MNFs with less than 25 percent of their total assets held abroad. This result suggests that multinationality is recognized by U.S. stock markets for companies whose overseas investments are relatively low but not when they are high (i.e., at least 40 percent of the firm's assets invested abroad).
In addition, this study provides evidence of a positive relationship between a MNF's beta and the relative profitability of its international business. Although there is no stock market appreciation for the level of corporate multinationality (except for companies with a low degree of foreign investment), the productive use of these foreign assets or the profitability of foreign sales has a substantial positive impact on the level of systematic risk.
 Aggarwal,Raj. "Investment Performance of U.S.-based Multinational Companies: Comments and a Perspective on International Diversification of Real Assets." Journal of International Business Studies, 11, 1 (Spring/Summer 1980), 98-104.
 Aggarwal, Raj. "Multinationality and Stock Market Valuation: An Empirical Study of U.S. Markets and Companies." Management International Review, 19, 2 (May, 1979), 5-21.
 Aggarwal, Raj, and Luc Soenen. "Changing Benefits of International Diversification of Real Assets." Rivista Internationale Di Science Economicsche & Commerciali, 34, 11-12 (Novermber/December, 1987), 1103-11.
 Agmon, Tamir, and Donald R. Lessard. "Investor Recognition of Corporate International Diversification." Journal of Finance, 32, 4 (September, 1977), 1049-55.
 Barone, Robert N. "Risk and International Diversification, Another Look." The Financial Review, 18, 2 (May, 1983), 184-95.
 Brewer, H. L. "Components of Investment Risk and Return: The Effects on Common Shareholders from Firm Level International Involvement." Management International Review, 29, 1 (January, 1989), 17-28.
 Brewer, Robert N. "Investor Benefits from Corporate International Diversification." Journal of Financial and Quantitative Analysis, 16, 1 (March, 1981), 113-26.
 Daniels, J. D., and J. Bracker. "Profit Performance: Do Foreign Operations Make a Difference?" Management International Review, 29, 1 (January, 1989), 46-56.
 Fatemi, Ali M. "Shareholder Benefits from Corporate International Diversification." Journal of Finance, 39, 5 (December, 1984), 1325-44.
 Grant, R. M. "Multinationality and Performance Among British Manufacturing Companies." Journal of International Business Studies, 18, 3 (Fall, 1987), 79-89.
 Grubel, Herbert C. "Internationally Diversified Portfolios: Welfare Gains and Capital Flows." American Economic Review, 58, 5 (December, 1968), 1299-314.
 Grubel, Herbert C, and K. Fadner. "The Interdependence of International Equity Markets." Journal of Finance, 26, 4 (March, 1971), 89-96.
 Hisey, Karen B., and Richard E. Caves. "Diversification Strategy and Choice of Country: Diversifying Acquisitions Abroad by U.S. Multinationals (1978-1880)." Journal of International Business Studies, 16, 2 (Summer, 1985), 51-64.
 Hughes, John S., Dennis E. Logue, and Richard T. Sweeney. "Corporate International Diversification and Market-Assigned Measures of Risk and Diversification." Journal of Financial and Quantitative Analysis, 10, 4 (November, 1975) 627-37.
 Jacquillat, Bertrand, and Bruno H. Solnik. "Multinationals are Poor Tools for Diversification." Journal of Portfolio Management, 4, 2 (Winter, 1978), 8-12.
 Kumar, M. S. Growth Acquisition and Investment. Cambridge, England: Cambridge University Press, 1984.
 Levy, Haim, and Marshall Samet. "International Diversification of Investment Portfolios." American Economic Review, 10, 4 (September, 1970), 668-75.
 Mathur, Ike, and Hyran Hanagan. "Are Multinational Corporations Superior Investment Vehicles for Achieving International Diversifications." Journal of International Business Studies, 14, 3 (Winter, 1983), 135-46.
 Michel, Allen, and Israel Shaked. "Multinational Corporations vs. Domestic Corporations: Financial Performance and Characteristics." Journal of International Business Studies, 16, 3 (Fall, 1986), 89-106.
 Mikhail, Armi D., and Hany I. Shawky. "Investment Performance of U.S.-Based Multinational Companies: Comments and a Perspective on International Diversification of Real Assets: A Reply." Journal of International Business Studies, 11, 1 (Spring/Summer, 1980), 53-66.
 Miller, J. C., and B. Pras. "The Effect of Multinational and Export Diversification on the Profit Stability of U.S. Corporations." Southern Economic Journal, 46, 3 (January, 1980), 792-805.
 Rugman, Alan M. "Risk Reduction by International Diversification." Journal of International Business Studies, 7, 2 (Fall/Winter 1976), 75-80.
 Severn, Alan K. "Investor Evaluation of Foreign and Domestic Risk." Journal of Finance, 29, 2 (May, 1974), 545-50.
 Shaked, I. "Are Multinational Corporations Safer? Journal of International Business Studies, 17, 1 (Spring, 1986), 83-106.
 Solnik, Bruno. "International Pricing of Risk: An Empirical Investigation of the World Capital Market Structure." Journal of Finance, 29, 2 (May, 1974), 365-78.
 Solnik, Bruno, and B. Noetzlin. "Optimal International Asset Allocation." Journal of Portfolio Management, 9, 1 (Fall, 1982), 11-21.
 Yang, H. C., James W. Wansley, and William R. Lane. "Stock Market Recognition of Multinationality of a firm and International Events." Journal of Business Finance and Accounting, 12, 2 (Summer, 1985), 263-74.
 Yoshihara, H. "Multinational Growth of Japanese Manufacturing Enterprises and the Postwar Period." Proceedings of the Fuji International Conference on Business History, Tokyo, University of Tokyo Press, 1985.
LUC A. SOENEN is Professor of Finance at California Polytenic State University…