School of Banking and Finance Working Paper 2002 University of New South Wales Multinational Financing Strategies in High Political Risk Countries Abstract This paper investigates the importance of various issues in relation to the financing of U.K. and U.S. multinationals subsidiaries in high political risk countries. It is discovered that multinationals with operations in high risk countries adopt strategies which are nationally responsive to the host nation. The matching of assets with liabilities is considered to be an effective method for reducing exchange rate risk and political risk. Local debt is viewed as crucial to the management of political risk by harnessing local participants in the ownership structure of the firm. In addition, it is found that firms with operations in high risk countries tend to be relatively less centralised. Although, expropriation of multinationals overseas assets has become a past occurrence, multinationals need to recognise the time-varying nature of political risks that they encounter. Vince Hooper Senior Lecturer School of Banking and Finance University of New South Wales Kensington, Sydney AUSTRALIA Tel +61 2 9385 5984 Fax +61 2 9385 6347 Email V.Hooper@unsw.edu.au 1
INTRODUCTION The purpose of this paper is to investigate the rationale underlying financial policies adopted by multinationals in high political risk countries. A survey was sent to a sample of U.K. and U.S. multinationals in order to discover the financial policies implemented for overseas subsidiaries. The findings of this study reveal that debt finance has multiple applications within an international financing context. The research has practical implications for the multinational manager who should use local debt finance in order to reduce political and foreign exchange rate risks. The financing decisions of the multinational are distorted by barriers to financial flows such as exchange controls, different national taxation systems, political risks, international capital market segmentation and foreign exchange risks, in addition to the internal organizational structure of the enterprise. While there have been many theoretical papers focussing upon financial policy, few have taken a comprehensive approach to understanding the rationale underlying the financing of multinational enterprises (MNE). The strands in the research literature encompass mainly signaling approaches, taxation, foreign exchange exposure but tend to ignore political risk perspectives. Multinational enterprises have three major choices relating to their financing of overseas subsidiaries or affiliates. They can finance from local sources such as host country debt and equity markets; international equity and bond markets; or fund from internally generated sources such as the parent's reserves. The latter can be used in order to offset any additional bankruptcy risks because multinationals would not need to raise additional external finance. Nowhere in the literature are there any studies that assess the relative importance of each of the three classes of financing overseas subsidiaries and affiliates, especially for those located in a high political risk country. Political risk is essentially an integral component of country risk, which involves the possibility of losses due to country-specific, economic, political and social events. The major political risks that impact upon the multinational are wide ranging. These could involve the hostile government blocking the repatriation of funds from the host country to the home country, expropriating or confiscating the multinational's assets or imposing wider economic constraints perhaps in the form of exchange controls [Buckley, 1992]. There are two major approaches to managing political risk. One is for the multinational to negotiate with the host country government and the second is for the multinational to alter its operating and financing strategies. With respect to the first approach, Heenan 2
(1988) supports a rapprochement between multinational companies and host governments as a means of reducing the extent of political risk faced by the MNE. A relationship may be extended from short-term to long-term. Strategic partnerships in the form of an equity joint venture can also be viewed as a means of reducing conflict between private and public enterprises and the host country government. However, the relationship between the multinational and host country is asymmetrical and the multinational is unable to prevent the host country s government from changing the contractual base on which the initial investment was first made [Sachs, 1983]. In effect, whenever a multinational undertakes an investment in an overseas country it has written a call option which a hostile government could exercise if it so wishes [Hooper and Pointon, 1996]. A decision matrix is offered by Kennedy (1988) that integrates political risk management concepts with portfolio planning. The Boston Consulting Group's market growth and relative market share matrix is used as a foundation. In general, five political risk strategies are available to the multinational enterprise: 1. adapt, by conforming to government policies. 2. politic, by acting informally behind the scenes, 3. withdraw from or avoid the country, 4. restructure with a serve strategy, involving the sale of equity and its replacement by management-service contracts, and 5. restructure as a joint venture Therefore the level of political risk that multinational enterprises often encounter can dictate the types of strategies implemented such as joint ventures [Beamish and Banks, 1987] which in turn influences the MNEs financing arrangements. Joint ventures should be viewed as a flexible method of control, but a reduction in the locus of control of the parent. In addition, the multinational enterprise can mitigate political risks to a great extent by raising local debt from the bond markets of the host country or from host country financial institutions. This has the result of reducing the net assets of the subsidiary and makes any possible expropriation of assets less likely. In 1988, the US and the U.K. ratified a convention establishing the Multilateral Investment Guarantee Agency (MIGA), a wholly owned subsidiary of the World Bank. MIGA's main objective is to encourage the flow of foreign direct investment to developing countries. MIGA hopes to achieve a co-operative synergism between developing and developed countries. It has two primary functions, which are (1) providing advisory and technical services for the improvement of investment conditions and (2) guaranteeing foreign 3
investments against non-commercial risks. There is now an array of private sector insurance companies entering the political risk underwriting business. However, in recent times the risk of expropriation of assets has been replaced by the threat of greater legislative controls upon businesses, in the form of increased regulation or tax penalties. The multinational s strategy for overcoming high political risk is inherently related to organizational considerations. In this regard, Martinez and Jarillo (1989) uncover that multinationals are sometimes forced to adopt decentralized decision-making structures by a host country government. Doz and Prahalad (1987) suggest a management philosophy that multinationals need to pursue in order to overcome the complexities of operating within an international arena. They suggest that multinationals should couple national responsiveness and cultural awareness with that of maintaining a global vision. This is a recent corporate ideology because multinationals have been criticized due to their unwillingness to conform to the requirements of the host country. This implies a degree of decentralization in decision-making. There have been a few papers in the literature which have provided guidance on debt denomination under exchange rate risk. By operating within an international setting, the multinational has the potential to lower its weighted average cost of capital by having access to a larger number of financing sources. The relationships that exist between different currencies, i.e. the risk diversification associated with debt portfolios, may benefit the multinational [Madura and Nosari, 1984 and Soenen, 1988]. Shapiro (1984) shows that in the absence of taxation, multinationals are indifferent between issuing debt denominated in one currency or another. However, with differential corporate taxes, a firm should borrow in the country with the weaker currency in order to minimize the expected financing costs. Rhee, Chang and Koveos (1985) support this. More recently, Madura and Fosberg (1990) demonstrate that assuming no corporate taxes and that if the International Fisher Effect holds then the expected net present value of a multinational project is invariant to debt denomination. Madura and Fosberg extend their analysis to incorporate market imperfections such as taxation and conclude that taxation considerations cause the multinational to have a debt denomination preference. The limited literature on multinational debt denomination decisions tends to support the idea of value enhancement to the multinational when it raises debt finance from countries with high rates of corporation tax. The literature review has attempted to appraise the broad issues involved with a multinational's financing strategy in relation to political risk. The next sections outline the empirical part of the paper. 4
HYPOTHESIS The null hypothesis is H0: There are no significant differences between the financial policy of multinationals that have operations in high political risk countries and those that do not have operations in high political risk countries. METHODOLOGY A questionnaire was used to collect the data necessary to test the above hypothesis. The survey questions were based around the issues that emerged from the literature review. Using the Dunn and Bradstreet stock market information system, Datastream, four criteria were used to identify the sample of multinationals. First, the market capitalization of the multinational needed to be in excess of 100 million pounds ($US160 million). Second, it had to have overseas production capabilities in the form of production facilities. Thirdly, the company had to be paying tax to an overseas government, and fourth the company was not a subsidiary of any other company. The questionnaire was mailed to the finance directors (Vice-President of finance) of United Kingdom (U.K.) and United States (U.S.) multinationals. Questionnaires were received from thirty-two U.K. multinationals and forty-two U.S. multinationals. A non-response bias was conducted by comparing market value and overseas tax paid by responding and non-responding firms. Responding firms had significantly higher market values and paid relatively greater overseas tax than non-responding firms, which implied that responding firms were more "multinational" than non-responding firms. Multinationals were identified as having operations in high political risk countries if they had business ventures in countries identified by Political Risk Services Inc. as being of high political risk. The finance director was asked to indicate the importance of financial policies [see Table 1 to Table 8]. The survey was designed around accepting or rejecting hypothesis H0. In particular, the financial policy questions related to capital structure decisions; the importance of debt finance for overseas subsidiaries; financing source; country-specific issues in relation to overseas financing; project-specific issues involved with raising 5
finance from overseas; and the degree of centralization of decision-making. Responses were measured on a Likert scale from 1 to 5 (1 indicated of no importance to 5, which indicated of highest importance). RESULTS The entire sample of U.K. and U.S. multinationals was split according to whether the multinational group had operations in a high political risk country or not. Group A represented those companies that did not have operations a high political risk country and Group B represented companies that did have operations in a high political risk country. Group A consisted of 38 companies, whilst Group comprised 36 companies. A chi-squared test revealed no association between the nationality of the multinational and whether or not the multinational had operations in a high political risk country. t-tests were conducted on the two groups in relation to the survey items in order to test the hypothesis. Significant differences are reported at the 5% level. The hypothesis H 0 was rejected on the basis of the t-tests. 6
In relation to the issues involved in the capital structure decision [Table 1], Group B companies (that had operations in countries with high levels of political risk) tended to stress greater importance on conforming to the industry and cultural norms of the host nation [Issue 1.5] than Group A companies (that did not have operations in countries with high political risk). This reflects Doz and Prahalad's (1987) argument that multinationals need to be nationally responsive whilst maintaining a global vision. This is probably why Group A companies placed more emphasis upon minimizing the cost of capital of the parent multinational [Issue 1.3]. Hence Group B companies show more evidence of adaptability than Group A companies, in the presence of high political risk. Table 1 The Importance of Issues Involved with the Capital Structure Decision Issue Grp A St. dev Grp B St. dev 1.1 Minimize the global cost of capital of the multinational group 1.2 Maximize the value of the tax shield on debt 1.3 Minimize cost of capital of the parent multinational 1.4 Achieve the target currency configuration of debt 1.5 Conform to the industry and cultural norms of the host nation 1.6 Minimize cost of capital of the subsidiaries 3.97 1.29 4.04 1.30 3.63 1.32 3.82 1.20 4.09 1.19 3.70 1.54 3.00 1.20 3.08 1.44 2.38 1.17 3.04 1.15 2.93 1.30 2.91 1.31 1.7 Diversify the investor base 2.27 1.00 2.39 1.03 Scale: 1=of no importance, 5=of greatest importance, Grp A=Companies that do not have operations in countries with high political risk, Grp B=Companies that do have operations in countries with high political risk. s significantly different at the 5% level. 7
With regard to the importance of strategies adopted in relation to the financing of overseas subsidiaries [Table 2], the avoidance of high political risk countries applied more strongly to Group A companies, that did not have operations in countries with high political risk than it did for Group B multinationals [Issue 2.2]. For all companies, there was a preference for structuring finances in the form of an equity joint venture [Issue 2.3] and moderate support for insuring a project with a political risk insurer [Issue 2.3]. For those companies with operations in high political risk countries, there was weak preference exhibited for the monitoring of the company's operations by host country institutions [Issue 2.6]. Table 2 The Importance of Strategies Adopted in Relation to Financing of Overseas Subsidiaries Strategy Grp A Grp B 2.1 Adapt by conforming to the host government's directives 3.40 1.03 3.79 0.88 2.2 Avoid a high political risk country 3.75 1.22 2.91 1.08 2.3 Structure finances in the form of an equity joint venture 2.4 Insure the project with a political risk insurer 2.60 1.02 2.81 0.96 1.78 1.14 2.09 1.15 2.5 Lobby other groups and institutions 1.95 1.17 1.90 0.83 2.6 Allow host institutions to monitor the company's operations 2.23 1.10 1.76 0.83 2.7 Politic with the World Bank 1.33 0.65 1.59 0.80 Scale: 1=of no importance, 5=of greatest importance, Grp A=Companies that do not have operations in countries with high political risk, Grp B=Companies that do have operations in countries with high political risk. s significantly different at the 5% level. 8
In relation to the importance of resource allocation policies [Table 3] there was overwhelming support for using portfolio theory to allocate assets and liabilities in an overall risk minimizing configuration [Issue 3.1], as well as upon the policy of matching [Issue 3.3]. There were no significant differences between Group A and Group B companies across these issues. Table 3 The Importance of Resource Allocation Policies Resource Allocation Policies Grp A Grp B 3.1 Allocate assets and liabilities in an overall risk minimizing configuration 3.2 Allocate assets and liabilities in an overall tax minimizing configuration 3.3 Match values of assets and liabilities in each respective currency 3.4 Allocate debt and equity in a risk minimizing configuration 3.5 Allocate liabilities in proportion to net project cash flows in each currency 3.6 Allocate assets and liabilities in a portfolio to maximize expected currency returns 3.84 1.05 3.95 1.23 3.66 0.86 3.58 1.10 3.84 1.09 3.54 1.56 3.47 1.05 3.47 1.23 2.77 1.03 2.90 1.15 2.50 1.00 2.35 1.19 Scale: 1=of no importance, 5=of greatest importance, Grp A=Companies that do not have operations in countries with high political risk, Grp B=Companies that do have operations in countries with high political risk. s significantly different at the 5% level. 9
Group B companies (that did have operations in countries with high political risk) stressed greater importance than Group A companies on borrowing in a weak currency [Table 4, Issue 4.1], in relation to the importance of debt finance. This finding is reinforced by the fact that companies that raised finance from countries with high levels of political risk (Group B) indicated greater importance of local debt markets of host country [Issue 5.3] as a means of financing than Group A companies. This also applied to the raising of finance from host country banks and governments [Issues 5.1 & 5.6]. For Group B companies, debt is seen more as a mechanism for reducing political and foreign exchange rate risks than taking advantage of tax shields on debt. Group B companies placed less importance upon achieving the correct portfolio configuration of debt [Issue 4.7] than Group A companies, which reflects the other priorities placed upon the utilization of debt. Table 4 Rationale for Raising Debt Finance from High Political Risk Countries Rationale Grp A Grp B 4.1 To lessen exchange rate risk by borrowing in a weak currency 4.2 To match assets against liabilities for subsidiary 3.00 1.41 4.08 1.18 3.61 1.26 3.91 1.16 4.3 To reduce the impact of exchange controls 3.78 1.20 3.83 1.03 4.4 To decrease the risk that assets may be expropriated 4.5 To take advantage of generally higher tax shields on debt 3.89 1.10 3.74 1.25 3.29 1.21 3.65 1.07 4.6 To obtain cheap government financing 3.06 1.39 3.33 1.13 4.7 To achieve the correct portfolio configuration of debt 3.78 1.31 2.62 1.20 Scale: 1=of no importance, 5=of greatest importance, Grp A=Companies that do not have operations in countries with high political risk, Grp B=Companies that do have operations in countries with high political risk. s significantly different at the 5% level. 10
With regard to sources of finance there was greater preference by Group B companies for local sources of finance [Issues 5.1, 5.3 & 5.6] than international sources which reaffirms the usage of local debt as a hedge against political risks [Table 5]. Table 5 The Importance of Sources of Finance Sources of Finance Grp A St. dev Grp B St. dev 5.1 Host country banks 3.60 0.92 4.04 0.97 5.2 Internally generated funds from the subsidiary's reserves 3.49 1.01 3.74 1.05 5.3 Local debt markets of the host country 3.18 1.15 3.71 0.96 5.4 Host country financial institutions 2.64 1.09 2.95 1.21 5.5 Internally generated funds from the parent's reserves 3.09 1.24 2.78 1.08 5.6 Host country governments 1.93 1.09 2.74 1.10 5.7 International bond markets 2.22 1.30 2.00 1.11 5.8 Local equity markets of the host country 1.60 1.03 1.91 0.95 5.9 International equity markets 1.76 1.11 1.69 1.06 5.10 Co-financing with the World Bank 1.35 0.84 1.52 0.85 Scale: 1=of no importance, 5=of greatest importance, Grp A=Companies that do not have operations in countries with high political risk, Grp B=Companies that do have operations in countries with high political risk. s significantly different at the 5% level. 11
Group B placed greater emphasis upon the importance of the host country inflation rate [Issue 6.4] than those that did not have operations in countries with high political risk [Table 6]. Both Group A and Group B companies shared the same level of concern for country-specific issues in relation to financing. Of greatest concern was the level and variability of the underlying economic fundamentals. Table 6 Country Specific Issues in Relation to Financing Country Specific Issues Grp A Grp B 6.1 Level of real interest rates of host country 3.83 1.10 4.20 1.02 6.2 Variability of exchange rate between the home and host country 6.3 Level of money interest rates of host country 3.65 0.84 3.82 0.94 3.72 0.98 3.79 1.02 6.4 Host country inflation rate 3.23 1.08 3.79 1.06 6.5 Variability of host country interest rates 3.46 0.94 3.58 1.14 6.6 Level of political risk of host country 3.42 1.24 3.45 0.93 6.7 Exchange controls 3.65 0.95 3.42 0.92 6.8 Exchange rate between home and host country 3.25 1.14 3.30 1.14 6.9 Transaction costs 2.91 0.97 2.74 1.13 Scale: 1=of no importance, 5=of greatest importance, Grp A=Companies that do not have operations in countries with high political risk, Grp B=Companies that do have operations in countries with high political risk. s significantly different at the 5% level. 12
In relation to project-specific considerations Group B companies were less concerned about the costs of monitoring the project [Issue 7.4], "bail out" options [Issue 7.6] and the costs of insolvency [Issue 7.7] than Group A companies which again supports the notion of local debt being used as a hedge against political risk [Table 7] as it would be local debt stakeholders who would be most affected if the multinational was tampered with by a hostile host country government. Table 7 Project-Specific Considerations in Relation to Financing Project Specific Issues Grp A Grp B 7.1 Variability of project cash flows denominated in foreign currency 3.81 0.91 3.42 1.26 7.2 Time horizon of project cash flows 3.52 0.83 3.73 1.09 7.3 Variability of project cash flows denominated in home currency 3.04 0.95 3.26 1.09 7.4 Costs of monitoring the project 2.52 1.01 2.00 0.97 7.5 Life of project 3.38 0.88 3.10 1.10 7.6 "Bail out" options 3.19 0.94 2.63 1.01 7.7 Costs of insolvency of the project 2.76 1.08 1.95 1.08 Scale: 1=of no importance, 5=of greatest importance, Grp A=Companies that do not have operations in countries with high political risk, Grp B=Companies that do have operations in countries with high political risk. s significantly different at the 5% level. 13
With respect to the degree of centralization of the finance functions [Table 8], companies that raise debt finance in countries with high political risk indicated slightly lower levels of centralization [Issue 8.4]. This may be explained by the fact that some host governments in high political risk countries force multinationals to have decentralized decision-making structures [Martinez and Jarillo, 1989]. Although benefits accrue to the multinational in the form of lowering political risk exposure by issuing more debt locally, there is a cost in the form of a loss in the locus of control, which can be interpreted as governance costs. Table 8 Degree of Centralization of Finance Functions Centralization Grp A Grp B 8.1 Centralization of debt financing 4.46 0.79 4.48 0.79 8.2 Centralization of tax planning 4.53 0.73 4.21 0.93 8.3 Centralization of interest rate risk hedging of subsidiaries 4.50 0.96 4.18 1.00 8.4 Centralization of hedging 4.56 0.67 4.17 0.87 8.5 Centralization of translation risk subsidiaries 4.41 1.00 4.14 1.28 8.6 Centralization of economic exposure risk 4.14 1.30 4.04 1.24 8.7 Centralization of cash management 4.02 1.10 3.54 1.25 8.8 Centralization of transaction risk subsidiaries 3.64 1.44 3.43 1.27 8.9 Centralization of capital budgeting 3.62 1.23 3.09 1.00 Scale: 1=completely decentralized, 5=completely centralized. s significantly different at the 5% level. 14
CONCLUSION The purpose of this paper was to compare the financial policies of multinationals with operations in high political risk countries with those in low political risk countries. The null hypothesis that there were no significant differences between the financial policies of multinationals with operations in high and low risk countries was rejected. Multinationals with operations in high political risk countries tended to conform more to the norms of the host country. This reflects the phenomenon that multinationals need to be responsive and adaptable to the demands of the host nation especially in the presence of political risk. This is reflected in the discovery that companies with operations in high political risk countries tended to be less centralized. Often, companies with operations in high political risk countries are forced to have greater involvement in decision-making by host country governments, which supports the notion of national responsiveness. Political risk by multinationals is mitigated to a greater extent through the manipulation of financial strategy and participation in joint ventures than insurance with a political risk insurer. This reflects a change in the business climate since the 1970s because few multinational enterprises have been expropriated in the past decade, hence the reduced need for projects to be insured. Expropriations have been replaced by more legislative controls upon businesses such as greater regulation and the imposition of windfall taxes on profits. In general, multinationals are reducing foreign exchange and political risks through the operation of a zero-net exposure policy, i.e. matching and utilization of portfolio theory. The rationale for raising local debt finance for companies with operations in high political risk countries was primarily targeted at taking advantage of depreciating currencies and reducing political risks through matching. There was a trade-off between companies that placed a high emphasis upon local sources of finance and the importance placed upon project-specific considerations such as bail out options, costs of insolvency of the project and costs of monitoring the project. Those companies that tended to raise local debt finance tended to place less emphasis upon these project-specific factors because the costs would be born by the local bondholders, should the project suffer distress. Therefore in practice, companies with operations in high political risk countries should attempt to raise as much local debt possible from host country capital markets and local financial institutions. 15
REFERENCES Beamish, P. W. and Banks, J. C. (1987). "Equity Joint Ventures and the Theory of the Multinational Enterprise", Journal of International Business Studies, vol. 18, no. 2, pp. 1-16. Buckley, A. (1992). Multinational Finance, second edition, Prentice-Hall International. Doz, Y. L. and Prahalad, C. K. (1987). The Multinational Mission: Balancing Local Demands and a Global Vision, Free Press. Heenan, D. (1988). "A Different Outlook for Multinational Companies", Journal of Business Strategy, vol. 9, no. 4, pp. 51-54. Hooper, V. J. and Pointon, J. (1996). "The Valuation of the Option to Expropriate a Multinational's Assets", Australian National University Working Paper Series in Economics and Econometrics, No. 305. Kennedy, C. R. (1988). "Political Risk Management: A Portfolio Planning Model", Business Horizons, vol. 31, no. 6, pp. 26-33. Madura, J. and Fosberg, R. H. (1990). "The Impact of Financing Sources on Multinational Projects". Journal of Financial Research, vol 13, no. 1, pp. 61-69. Madura, J. and Nosari, E. J. (1984). `Utilizing Currency Portfolios to Mitigate Exchange Rate Risk', Columbia Journal of World Business, vol. 29, no. 1, pp. 96-99. Martinez, J. and Jarillo, J. (1989). "The Evolution of Research on Co-ordination Mechanisms in Multinational Enterprises", Journal of International Business Studies, vol. 20, no. 3, pp. 489-514. Rhee, G., Chang, R., Koveos, R. (1985). "The Currency of Denomination Decisions for Debt Financing", Journal of International Business Studies, vol. 16, no. 4, pp. 143-50. Sachs, J. D., 1983, Theoretical Issues in International Borrowing, NBER Working Paper 1189, Cambridge, MA August. Shapiro, A. C. (1984). "The Impact of Taxation on the Currency of Denomination for Long-term Foreign Borrowing and Lending", Journal of International Business Studies, vol. 15, no. 1, pp. 15-25. Soenen, L. A. (1988). "Risk Diversification Characteristics of Currency Cocktails", Journal of Economics and Business, vol. 40, no. 2, pp. 177-189. 16