Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Measuring corporate internationalisation

Measuring corporate internationalisation MULTINATIONAL CORPORATIONS Christoph D6rrenb&cher* A Review of Measurement Concepts and their Use Measures of corporate internationalisation have gained crucial importance in the recent debate on globalisation, since many scholars link globalisation to a quantitative increase in the international activities of firms. Opinions on the extent of this increase differ widely, however, depending on what measurement concept is used. As there is no universally applicable measurement concept, researchers face the difficult task of bringing research questions, measurement concepts and data availabifity into line. he recent debate on globalisation has generated a that ownership structure or management composition T wider interest in the transborder activities of can be internationalised too. corporations. While some authors are convinced that Given these differences in definition, this paper tries multinational corporations (MNCs) are best symbolis- to give an overview of different approaches to ed by an octopus whose tentacles try to grasp the measuring and comparing corporate internationali- whole world, 1 other authors paint a completely sation. Based on the assumption that there is more different picture. Here the MNC is seen more or less than one way to "correctly" measure internationali- as a lethargic animal, whose presence abroad is sation, this paper concentrates on the meaning- rather limited? Both metaphors can be justified when fulness of different criteria and indices. The first considering individual companies, at least if one takes section deals with individual internationalisation the following definition as a basis: according to indicators. This is followed by a discussion on how Dunning "A multinational or transnational enterprise is regional diversification in internationalisation can be an enterprise that engages in foreign direct measured and expressed. Next we look at different investment and owns and controls value adding composite indicators (or indices) and discuss issues activities in more than one country". 3 of reliability, exactness and validity of indicators and It is probably not wrong to state that today most indices. The final section deals with the difficulties of large corporations fulfil this definition, which is widely bringing research questions, indicators and data accepted in academia, business, national govern- availability into line. ments and international organisations. For instance, a Individual Internationalisation Indicators recent compilation of data on the internationalisation of the 100 largest German companies has found that As mentioned above, the term "internationalisation" it is difficult to find a strictly national company. Even has different connotations - even if restricted to some of the utility companies are to a small degree corporate internationalisation. Some might think of active in foreign markets? Thus, thinking about larger intemationalisation as a process, others see it more corporations, the question is not which companies are as a certain level of international intertwining reached internationalised, but rather how companies differ in regard to their internationalisation. This is especially true if we do not restrict internationalisation to the See R. J. Barnet, J. Cavanagh: Global Dreams. Imperial Corporations and the new world order, New York 1994; D.C. Kor- definition that value adding activities have to take t e n : When Corporations Rule the World, West Hartford 1995. place in more than one country, but also recognise 2 See P. Hirst, G. Thompson: Globalization in Question. The International Ecomomy and the Possibilities of Governance, Cambridge 1996. * Wissenschaftszentrum for Sozialforschung Berlin (WZB) and For- J. H. Dunning: Multinational Enterprises and the Global schungsgemeinschaft for Aui3enwirtschaft, Struktur- und Techno- Economy, Wokingham 1992, p. 3. Iogiepolitik (FAST) e.V., Berlin, Germany. For useful comments the author would like to thank Douglas van den Berghe (University of 4 FAST 1999, reprinted inA. Hassel, M. HSpner, A. Kurdel- Rotterdam), Dieter Plehwe (WZB), Stefano Vescovi (WZB) and busch, B. Rehder, R. Zugeh0r: Dimensionen derlnternatio- Michael Wortmann (FAST e.V., Berlin). An earlier draft of this paper nalisierung: Ergebnisse der Unternehmensdatenbank, Internationali- was presented at the workshop "Dimensionen der Internationali- sierung der 100 gr56ten Unternehmen in Deutschland, Working sierung von Unternehmen" at the Max-Planck-lnstitut f5r Gesell- Paper 1/2000, Max-Planck-lnstitut f~Jr Gesellschaftsforschung, KSIn schaftsforschung, Cologne 16th/17th December 1999. INTERECONOMICS, May/June 2000 119 MULTINATIONAL CORPORATIONS Table 1 Individual Internationalisation Indicators Structural Indicators Performance Indicators Attitudinal Indicators 1. Relating to foreign activities 1. Foreign sales 1. 'Soft' indicators 9 Number of countries a company is 9 Demand: Amount of foreign sales 9 Ethno-, poly-, regio- or geocentric management style according to: active in by customer location organisational complexity, 9 Number or proportion of foreign affiliates Exports from the home country authority, decision making, 9 Number or proportion of cases of + Revenues of foreign affiliates communication flows, non-capital involvement abroad - their revenues from exports to the recruiting, staffing, (e.g. strategic alliances, franchised home country control operations) 9 Amount or proportion of foreign assets 2. 'Hard' indicators 9 Amount or proportion of value added 9 Supply: Amount of sales of foreign 9 International experience of top abroad affiliates managers 9 Amount or proportion of sourcing abroad Sum of revenues of foreign affiliates cumulative duration of the years top 9 Number or proportion of foreign managers spent working abroad employees weighted by the total years of their working experience 2. Relating to governance structures 2. Operating income abroad 9 Number of stock markets on which a Sum of operating income of foreign company is listed affiliates 9 Amount or proportion of shares owned by foreigners 9 Number or proportion of non-nationals in the board of directors by a company. While this is only a question of different [] the number of stock markets on which a company perspectives, there are strong differences of opinion is listed, as to what indicators are suitable for measuring cor- [] the amount/proportion of shares owned by for- porate internationalisation. Following a welt establish- eigners, ed framework in international business literature, we [] the number/proportion of non-nationals on the can distinguish structural indicators, performance board of directors. indicators and attitudinal indicators2 Performance Indicators Structural Indicators Performance indicators measure the degree to Structural indicators try to give a picture of the which the success or failure of corporate activity international entanglement of a corporation at a during a certain period of time (usually one year) is certain time. Here we find several indicators that are connected to foreign countries. The two main related to the foreign activities of MNCs such as: indicators are turnover and operating income. Two [] the number of countries a company is active in, issues have to be addressed when considering turn- over: [] the number/proportion of foreign affiliates, [] The first issue is demand. The main question here [] the number/proportion of cases of non-capital in- is, to what extent turnover is generated in foreign volvement abroad (e.g. strategic alliances, franchised countries. The relevant indicator - amount of foreign operations etc.), sales by customer location - is calculated as exports [] the amount/proportion of foreign assets, from the home country plus revenues of the foreign affiliates (except those revenues that come from [] the amount/proportion of value added abroad, exports to the home country of the parent company). [] the amount/proportion of sourcing abroad, [] the number/proportion of foreign employees. 5 See H. V. Perlmutter, D.A. Heenan: Multinational Organi- zation Development, Reading 1979, p. 16; D. S u I I i v a n : Measuring the Degree of Internationalization of a Firm, in: Journal of International A second group of indicators describes the inter- Business Studies, 1994, Vol. 25, No. 2, p. 331; I. B&urle: Inter- nationalisation of the governance structures of a nationalisierung als Prozessph&nomen. Konzepte - Besonderheiten - corporation. Relevant indicators here are: Handhabung, Wiesbaden 1996, p. 9. 120 INTERECONOMICS, May/June 2000 MULTINATIONAL CORPORATIONS Table 2 The Measurement of Regional Diversification Regional Concentration Network Extension Geographical and Cultural Distance 9 Network spread index: 9 Countries are weighted according to 9 Homogeneous vs. heterogeneous distribution of foreign activities (letto-Gillies 1998) their geographic and cultural distance (Schmidt 1981) to the home country Number of foreign countries in which (Kutschker 1993) 9 Extent to which the regional distribution a company owns affiliates 9 Psychic dispersion index: of a certain indicator at a given company as a proportion of (Sullivan 1994) complies with the total distribution of this indicator in the world Number of zones with different cognitive maps relating to (Perriard 1995) total number of countries in which management principles in which a foreign direct investment has occurred minus one company is active (= home country of the company) (out of a total of 10 zones worldwide) that several attitudinal indicators, differing in how [] The second issue is supply. The main question statistically measurable they are, are in use. A rather here is, to what extent business activity is transacted "soft" qualitative indicator was developed by Perl- by subsidiaries located in foreign countries. The mutter. This indicator, which underlies or is explicitly relevant indicator - amount of sales of foreign applied in many studies 1~ distinguishes four types of subsidiaries - is calculated as the sum of turnover of headquarters management's orientation towards their foreign subsidiaries as a proportion of the sum of foreign subsidiaries. ~1 To decide whether the central turnover of the parent company and its national management of an MNC follows an ethnocentric (i.e. subsidiaries. 8 home-country oriented), a polycentric (i.e. host- A second important performance indicator is the country oriented), a regiocentric (i.e. regionally orient- amount of operating income earned abroad. This ed) or a geocentric (i.e. globally oriented) approach, indicator is usually calculated as the sum of operating Perlmutter/Heenan '2 proposed a multi-item indicator, income generated by all foreign affiliates. addressing inter alia the following aspects: complexity of organisation, authority/decision making, eva- Attitudinal Indicators luation/control, rewards/punishment, communica- Attitudinal indicators try to give a picture of how tion/information, recruiting/staffing. However, due to MNCs view foreign countries and treat their their premise that recruiting/staffing is of central subsidiaries in foreign countries. 7 They usually imply importance in the internationalisation of an MNC, an actor centred view, focusing on the management Perlmutter/Heenan 13 concentrated their own empirical or parts thereof. Although there are doubts as to study on personnel function (manpower planning, whether attitudinal characteristics can be measured manpower administration and manpower control). with sufficient reliability, 8 there is also strong support for the use of attitudinal indicators, as expressed in See H. V. P e r I m u t t e r, D.A. H e e n a n : Multinational Organiza- the following quotation: "...external and quantifiable tion Development, op. cit., p. 17. measures such as the percentage of investment ~0 See e.g.G. H e d I u n d : The Hypermodern MNC: A Heterarchy?, overseas or the distribution of equity, are useful but in: Human Resource Management, 1986, Vol. 25, Spring, pp. 9-25; C. Bartlett, S. Ghoshal: Managing across Borders. The not enough ... The more one penetrates the living transnational Solution, Boston 1989; I? Marginson, K. Sisson: reality of how decisions are made in an MNC, the The Structure of Transnational Capital in Europe: The Emerging Euro- Company, in: R. Hyman, A. Ferner (eds.): New Frontiers in more weight must be given to how executives think European Industrial Relations, Oxford 1994, pp. 15-51 ; C. D 5 r r e n - about doing business around the world". 9 b&cher: Vom Hoflieferanten zum Global Player, Unternehmens- reorganisation und nationale Politik in der Welttelekommunikations- [ndustrie, Berlin 1999. These different positrons are reflected by the fact ll See H. V. Perlmutter: The Tortuous Evolution of the Multi- national Corporation, in: Columbia Journal of World Business, 1969, Since there is no consolidation of intrafirm transfers, this indicator Vol. 4, pp. 9-18;H.V. Perlmutter, D.A. Heenan: How Multi- does net necessarily express the home/foreign split of production! national should your Top Managers be?, in: Harvard Business Review, 1974, November-December, pp. 121-132. ' Of course, in reverse this indicator also gives a picture of the role of the entities in the home country. ~2 H.V. Perlmutter, D.A. Heenan: Multinational Organization Development, op. cit., p. 18 f. See D. S u Ill v a n : Measuring the Degree of Internationalization of a Firm, op. cit., p. 325. ~3 Ibid. INTERECONOMICS, May/June 2000 121 MULTINATIONAL CORPORATIONS Table 3 Composite Indicators Used to Measure Corporate Internationalisation Transnationality Index Transnational Activities Spread Index Degree of Internationalization Scale (UNCTAD 1995) (letto-Gillies 1998) (Sullivan 1994) Ratio of foreign sales to total sales Ratio of foreign sales to total sales Ratio of foreign sales to total sales + Ratio of foreign assets to total assets + Ratio of foreign assets to total assets + Ratio of foreign assets to total assets + Ratio of foreign employment to total + Ratio of foreign employment to total + Ratio of foreign affiliates to total employment employment affiliates divided by 3 divided by 3 + "International experience of top management" (see above) multiplied by + "Psychic dispersion" of internationa~ Number of foreign countries in which a operations company owns affiliates as a proportion of total number of countries in which foreign direct investment has occurred minus one (= home country of the company) According to them, the degree of internationalisation company tends to have the same spatial distribution of a company increases from ethno-, through poly- as total world car production. and regiocentrism to geocentrism. A second way of tackling geographical variation of An attitudinal indicator which can be better corporate internationalisation is to focus on the num- measured statistically was developed by Sullivan. TM ber of countries a company is active in. Developed by According to him, the international mindset of a letto-Gillies, 17 the so called "network spread index" is company increases with the international experience calculated, ceteris paribus, by taking the number of of top managers, measured as the cumulative foreign countries in which a company has affiliates as duration of the years top managers spend working a proportion of all countries where foreign direct abroad weighted by the total years of their working investment has occurred, minus one country, i.e. the experience. home country of the company. Finally, based on the assumption that there are Internationalisation as Regional Diversification striking differences between countries, which in- In addition to the dichotomy of home versus fluence the internationalisation behaviour of com- foreign, different geographical variation indices can be panies, TM Kutschker 'g proposed that foreign countries calculated for many of the indicators given above. For should be weighted according to their geographical instance Schmidt 's uses the Herfindahl-index to place and cultural distance to the home country of the companies on a continuum between a completely company. Similarly, Sullivan 2~ introduced an indicator homogeneous and a completely heterogeneous he calls "psychic dispersion". According to this indi- distribution of their foreign activities to different cator, the world is divided into ten zones with different countries. Perriard" uses the Gini-index to measure cognitive maps relating to management principles. the extent to which the regional distribution of a The more zones a company is active in, the larger the certain indicator at a given company complies with psychic dispersion of its internationalisation. the total distribution of this indicator in the world. Take the example of car production: following Perriard's methods, one could look how far the spatial " G. letto-GiIlies: Different Conceptual Frameworks for the Assessment of the Degree of Internationalization: an Empirical distribution of the car production of a specific Analysis of Various Indices for the Top 100 Transnational Corpo- rations, in: Transnational Corporations, 1998, Vol. 7, pp. 17-39. " D. Sullivan: Measuring the Degree of Internationalization of a ~8 See J. Johanson, J. Vahlne: The Internationalization Pro- Firm, op. cit., p. 332. cess of the Firm - A Model of Knowledge Development and Increas- ing Foreign Market Commitments, in: Journal of International Busi- 1, R. S c h m i d t : Zur Messung des Internationalisierungsgrades von ness Studies, 1977, Vol. 8, pp. 23-32. Unternehmen, in: W. Wacker, H. Haussmann, B. Kumar (eds.): Internationale UnternehmensfQhrung. Managementprobleme '~ Kutschker: Dynamische Internationalisierungsstrategie, Dis- international t&tiger Unternehmen, Berlin 1981, pp. 57-70. kussionsbeitrag Nr. 41, Wirtschaftswissenschaftliche Fakult&t Ingolstadt, Katholische Universit~lt Eichst&tt, Ingolstadt 1993. '~ M. Perriard: Towards a Measure of Globalization, Institute of 2o D. S u Iliva n : Measuring the Degree of Internationalization of a Economic and Social Sciences University of Fribourg Working Paper No. 250, Fribourg 1995. Firm, op. cit. 122 INTERECONOMICS, May/June 2000 MULTINATIONAL CORPORATIONS Internationalisation Indices In order not to favour companies from one specific type of industry 27 in the ranking of companies, 28 the Indices, or composite indicators, are formed by "transnationality" index is calculated as the average of combining individual indicators. Despite the fact that the three above mentioned ratios. numerous studies are based on individual indicators, there is a broad consensus that composite indicators 2. The "transnational activity spread index". This are more suitable to measure corporate internatio- index, proposed by letto-Gillies, 29 is a combination of nalisation. Firstly, given the multidimensionality of two indices. According to letto-Gillies the index used internationalisation, 21 restricting the measurement to by UNCTAD ignores the spread of foreign activities. one single item inevitably means that only a part of the But her own network spread index, which indicates whole phenomenon of corporate internationalisation regional diversification (see above), does not take into is represented. Secondly, the multidimensionality of account the amount or the share of activities abroad. the phenomenon means that, depending on what As a result she proposes to combine the two indices indicator is used, single transactions could lead to so that the UNCTAD index is weighted by the contradictory results concerning the degree of "network spread index". In detail, the resulting new internationalisation. 22 For instance, modernising a index called the "transnational activity spread index" plant abroad might lead to an increase in the is calculated as follows: The average of the following proportion of foreign assets, but due to the labour ratios: foreign sales to total sales, foreign assets to saving effects of modernisation, the proportion of total assets, and foreign employment to total em- foreign employees might at the same time decrease. ployment, is multiplied by the number of foreign Thirdly, using individual indicators does not allow countries a company is active in as a proportion of the systematic control of measurement errors, contingent total number of countries where foreign direct invest- influences and transfer pricing manipulations (see ment has occurred, minus one (the home country). In below). addition to this comprehensive index letto-Gillies indicates that every individual indicator of the Many scholars agree that the development of UNCTAD index might also be weighted by the sophisticated indices to measure corporate inter- network spread index, thus forming three sub-indi- nationalisation is still in its infancy? 3 Only three ces: the "sales spread index", the "asset spread different composite indicators can be identified in the index" and the "employment spread index". literature on the subject: 2' 3. The "degree of internationalization scale". This 1. The "transnationality index" used by UNCTAD. index, proposed by Sullivan, 3~ was developed in a This composite indicator first appeared in the World bottom up-process, using publicly available data. Investment Report 1995.2 s The aim of the indicator is Unlike UNCTAD, which selected its three individual "... to capture fully the extent of involvement of TNCs indicators based on preliminary assumptions about [Transnational Corporations, C.D.] in the world their usefulness in expressing the phenomenon of economy"? 6 The index draws on three different ratios: internationalisation, Sullivan started with a total of Foreign sales to total sales, foreign assets to total nine indicators encompassing structural, performance assets, and foreign employment to total employment. and attitudinal items. Using item-total analysis 31 on a 2, This multidimensionality is also reflected in the wealth of different 28 Ibid., p. 24. single indicators on corporate internationalisation as described 27 Labour intensive, capital intensive or export intensive industry. above. 28 This index is annually calculated for the world's top 100 non- 22 SeeH. German, S. Raab, M Setzer: Messung derGIoba- financial MNCs, which are selected on the basis of their foreign lisierung: Ein Paradoxon, in: U. Steger (ed.): Facetten der assets. The collection and the calculation of the data is undertaken Globalisierung. C)konomische, soziale und politische Aspekte, Hei- by the Studies and Competence Centre for Organisational and Policy delberg 1999, p. 7. Research in European Business (SCOPE) at the Erasmus University of 23 See D. S u I I i v a n : Measuring the Degree of Internationalization of Rotterdam. For a recent interpretation of the data,see D. v a n d e n aFirm, op. cit.; M. Perriard, op. cit.;K. Ramaswamy, K.G. Berghe: Internationalisation Strategies, Relocation Processes and Kroeck, W. Renforth: Measuring the Degree of Internationali- Employment: Multinational Enterprises in the Age of Globalisation. zation, A Comment, in: Journal of International Business Studies, Paper Presented at the 14th Annual Employment Research Unit 1996, Vol. 27, No. 1, pp. 167-177. Conference: Globalisation, Employment and the Workplace, Cardiff Business School, 8/9. September 1999. 24 In addition to these indices, M. Perriard, op. cit., proposes a set of 13 quantitative and qualitative indicators to determine the degree ~ G. Letto-Gillies, op. cit. of the internationalisation of a corporation. However since there is no 3o D. Sullivan: Measuring the Degree of Internationalization of a integration of the indicators, the use of this instrument is very Firm, op. cit.; D. S u I l iva n : Measuring the Degree of Internatio- restricted. nalization, A Reply, in: Journal of International Business Studies, 2~ UNCTAD: World Investment Report 1995, New York. 1996, Vol. 27, No. 1, pp. 179-192. INTERECONOMICS, May/June 2000 123 MULTINATIONAL CORPORATIONS representative sample of 74 MNCs, five "good" nationalisation at the time of the headquarters indicators have been identified. These indicators, all relocation, there might be dramatic changes in some weighted by the factor one, form the degree of figures indicating the internationalisation of the internationalization scale. In detail the index is company. Some further contingent influences are composed of the following indicators: The ratio of industry specific. They occur for instance in engineer- foreign sales to total sales, foreign assets to total ing, when large projects, covering many years, are assets and foreign subsidiaries to total subsidiaries, invoiced in one year, resulting in a tremendous jump in performance indicators such as turnover or operating the international experience of top managers (measured as the cumulative duration of the years top income. managers spend working abroad weighted by the [] Transfer pricing manipulations systematically total years of their working experience) and the distort the geographic distribution of performance psychic dispersion of international operations. indicators such as turnover and especially income. Transfer pricing manipulations occur when inter- Reliability, Exactness and Validity nalised transactions between related units of an MNC are not calculated on the "arm's length principle". If one recognises that scientific progress is not Aims of transfer pricing manipulations are manifold: possible without adequate measurement 32 then it is for example, tax avoidance, lowering customs duties, important to discuss the issues of reliability, exact- repatriation of capital. However, very little is yet known ness and validity of the indicators described so far. about the significance of transfer pricing manip- As mentioned above the use of individual indicators ulations. In a recent overview of different studies, to measure corporate internationalisation is in general Plasschaert states that "... all in all, the only plausible, not very reliable since it does not include a systematic although trivial, conclusion .... is that transfer-price control of measurement errors, contingent influences manoeuvres are probably practised much more and transfer-pricing manipulations (beyond obvious frequently than TNCs [Transnational Corporations, implausibility). C.D.] are willing to admit, but much less than is alleged in some circles". 3s [] Measurement errors might occur with "soft" atti- tudinal indicators, such as the Perlmutter typology. Although composite indicators are generally more Here, biased views and cognitive dissonance, both of reliable, their use is not without problems. We have the interviewer and the interviewee, may lead to discussed three different composite indicators, each flawed results. Another source of measurement error of which claims to adequately measure corporate is the fact that the definition of a certain indicator internationalisation. However, taking the multidimen- must be clear, detailed and viable. Take the example sionality of the phenomenon as a criterion, the three of the indicator of the ratio of foreign employment to indices have different orientations. The UNCTAD total employment. In the first place, an exact definition index combines two structural and one performance of what an employee is is necessary: for example, indicator. The "transnational activity spread index", should there be a headcount, or are part time em- proposed by letto-Gillies is even more concentrated ployees converted into full time equivalents according on structural aspects of internationalisation; she uses to their working hours; 33 do employment figures the UNCTAD index as a base but weights it with represent the average of the year or is the figure taken another structural indicator. The only index covering at a common qualifying date etc. Secondly, is it viable to compare employment on a one to one basis, taking 3, Item-total analysis correlates each potential indicator at a given into account the large sectoral and national sample with the indicator-corrected scale of the total score. The aim is to get a collection of indicators which has a high average differences in the costs of creating a workplace? correlation with the total score. [] Contingent influences are influences that change 32 Korman 1974 and Schwab 1980 cited in D. Sullivan: Measuring the Degree of A Reply, op. cit., p. 338. indicators beyond real material changes. The most obvious contingent influence is exchange rate 33 This point is crucial in industries which mainly use part time workers, such as the fast food industry or the transport industry. fluctuations. Exchange rate problems occur parti- 34 One example is the relocaton of the Ikea headquarters from cularly when comparing the internationalisation of Sweden to Denmark in 1975, see C. DSrrenb&cher, H. R. companies on a historical or on a national basis. Meil3ner: IKEA -The Hollow Elk Corporation, in: Informationen 0ber multinationale Konzerne, 1991, No. 2, p. 23. Another contingent influence which occurs from time 35 S. Plasschaert: Transnational Corporations: Transfer Pricing to time is the relocation of an MNC's headquarters to and Taxation, in: UNCTAD (ed.): Transnational Corporations and another country? 4 Depending on the degree of inter- World Development, London 1996, p. 406. 1 24 May/June 2000 INTERECONOMICS, Internationalization, MULTINATIONAL CORPORATIONS all three dimensions of internationalisation is the one not the construction of such a general index is developed by Sullivan. This index combines structural feasible in the foreseeable future28 For instance, (two indicators), performance (one indicator) and Hassel et al. recently considered this question and attitudinal (two indicators) aspects. However, it is rejected the general approach. 39 questionable whether even this comprehensive Research Questions, instrument adequately covers the internationalisation Indicators and Data Availability phenomenon: The lack of a general indicator encompassing the [] Firstly, the viability of both attitudinal indicators whole phenomenon of corporate internationalisation used by Sullivan is heavily disputed. Ramaswamy et is a severe problem if the research purpose is to rank a126 maintain that the length of international experi- companies according to their overall degree of ence is not a good yardstick for the international internationalisation. However, if corporate internatio- mindset of the management, since there are several nalisation is part of a more specific research question, other factors with a stronger influence, such as the then the aim of the research might give some geographic spread of the company, its policy and its guidance on how to construct a viable index (what administrative heritage. In addition, Ramaswamy et al. indicators, how much weight). For instance, if one is criticise the plausibility of the "psychic dispersion of interested in ranking companies according to their internationalisation" concept. Using this concept production abroad, the best indicator would probably would mean, for instance, that a company operating be value added abroad. The proportion of foreign in France and the United Kingdom has a higher employees and foreign assets might be good psychic dispersion of its internationalisation than a yardsticks. In contrast, the number of stock markets company operating in Japan, India, Israel and Brazil. on which a company is listed does not seem to be a [] Secondly, there are strong concerns as to whether very plausible indicator. Sullivan's indicator can really claim to be all-encom- It might also be easier to find indices for intra- passing, because the way he chooses the indicators industry comparisons than for cross-industry studies, for the index (i.e. selecting indicators which have the since in this case indices can be optimised according highest average correlation with one another) to industry specifics. For instance the "transnational systematically excludes indicators that might be spread index" might be appropriate for industries, important to express the phenomenon. Following such as in the transport industry, in which the extent Ramaswamy et al. the index used by Sullivan "... of the geographical reach is of special importance. seems to exclude several common strategic options for internationalisation such as indirect exporting, But even if a viable index to rank companies from licensing, joint ventures ...,,37 different countries and/or industries is found, two problems remain. The first problem is data availability. A second problem is the question as to how much Apart from mail surveys, there are only a few sources weight different individual indicators should have in an from which data expressing the international index. Up to now there is no information on how to entanglement of a company is available: these are determine the importance a specific indicator should specialised company listings, company handbooks, have in the construction of an index. Because of this databases and annual reports. In general these lack of information, two of the indices mentioned here sources are very fragmented and incomplete (for treat all indicators equally. In contrast, the letto-Gillies example they lack interesting information, such as the index allots more weight to the network spread com- amount of shares owned by foreigners or the ponent than to the three other single components. proportion of value added abroad etc.). In addition, To sum up the discussion, there is neither a single indicator nor an index that satisfactorily measures the overall degree of the internationalisation of a firm. 3g Their new index, which sheds light on the previously neglected financiar dimension of intemationalisation, combines an index that Furthermore, there are differing views as to whether or measures the real dimension of corporate internationalisation (using a somewhat reworked "transnational activity spread index") with different indicators which measure the financial dimension of cor- ~6 K. Ramaswamy, K. G. Kroeck, W. Renforth, op. cit., porate internationalisation (i.e. foreign owners as percentage of total p. 173. ownership, number of listings in foreign stock exchanges and the use of international accounting standards). ,See A. Hassel, M. HOp- 37 Ibid., p. 174. ner, A. Kurdelbusch, B. Rehder, R. ZugehSr, op. cit.; 38 See D. S u Ill v a n : Measuring the Degree of Internationalization, A. Hassel, M. HSpner, A. Kurdelbusch, B. Rehder, AReply, op. cit., p. 190; K. Ramaswamy, K.G. Kroeck, W. R. Z u g e h 5 r: Two Dimensions of the Internationalization of Firms, Renforth, op. cit,, p. 176. K01n 2000. INTERECONOMICS, May/June 2000 1 25 MULTINATIONAL CORPORATIONS data availability has a high sectoral and national vari- used either to explain problems of state sovereignty ance, and is the result of different accounting rules and declining trade union power '2 or to contest these and publishing habits. For instance, in contrast to the developments by arguing that there is no, or only US General accounting principles, the German modest, growth in the international entanglement of accounting standard does not require a home/foreign MNCs. ,3 split of the assets. Different practices exist in the Regardless of the field of study or the research publication of employment data, with many German purpose, there are some strong arguments against a companies publishing a home/foreign split, some- positivistic interpretation of quantitative data on the thing that is very unfamitiar to Japanese companies. internationalisation of MNCs. Take the case of Finally, due to different modes of compilation, data is management science and its attempt to explain the often incompatible especially for companies from financial performance of an MNC by its inter- different countries. While in the long run some of nationalisation. It is a commonplace that in this kind of these problems will alleviate due to the ongoing study the measurement concept strongly influences standardisation of accounting principles, other pro- the result. However, even using the same indicator blems will persist. does not give a consistent picture." Furthermore, macroeconomic data, for example on the (probably The second and more general problem is the use of indicators and indices in wider economic and socio- growing) contribution of large MNCs to the world's political research questions. The actual use of gross domestic product, has to be interpreted quantitative data on corporate internationalisation carefully. Collecting information on the quantitative concentrates on three purposes. share of a few large MNCs in a certain market is only a first step towards finding out something about the 1. In management science, many empirical studies power of MNCs. In addition, there has to be at least a look at the correlation between the internationali- look at the development of competition and its sation of a company and its financial performance. influence on the level of prices and the structure of 2. In macroeconomics, the use of data on individual supply in this market. 's Finally, the debate on the MNCs is not very widespread. `~ However there are influence that corporate internationalisation has on some exceptions, such as concentration issues, state sovereignty and trade union power is a striking where the use of company specific data is necessary. warning against a positivistic interpretation of For instance, the list of the top 100 MNCs of the quantitative data on corporate internationalisation. world, annually published by the UNCTAD, is used to Many empirical studies have found that in the long run find out whether the share of those companies in there is no big rupture in the quantitative (and world gross domestic product is increasing or not. qualitative) dimension of corporate internationali- 3. In political science (especially in the sub-fields of sation of MNCs. '8 However this does not mean that all international political economy and industrial rela- problems concerning state sovereignty or declining tions) the internationalisation of MNCs has served trade union power are imagined. The answer is simply different purposes. In the past, the extent of foreign that other things, like general economic conditions '7 manoeuvres of MNCs was frequently used to explain or the state's need for tax income, '8 have changed underdevelopment in third world countries? 1 More dramatically over time, bringing MNCs into better recently the internationalisation of MNCs has been bargaining positions, whether their degree of internationalisation has significantly increased or not. ,0 Scholars usually use aggregate data, such as the flow or the stock of foreign direct investment. For a discussion of the problems 4~ See C. DSrrenb&cher, op. cit., p. 56. associated with these indicators see M. Wortmann, C. D6rren- b&cher: Multinationale Konzerne und der Standort Deutschland, " Seee.g.P. Hirst, G. Thompson, op. cit.; P. Doremus, W. in: W. Fricke (ed.): Jahrbuch Arbeit und Technik, 1997, pp. 28-42. Keller, L. Pauly, S. Reich: The Myth of the Global Corpo- ration, Princeton 1988; M. Wortmann: Zur Logik yon Wachstum " Seee.g.F. FrSbel, J. Heinrichs, O. Kreye: Dieneuelnter- und Restrukturierung multinationaler Konzerne, in: C. D6rren- nationale Arbeitsteilung, Reinbek 1977. b&cher, D. Plehwe (eds.): Grenzenlose Kontrolle? Organisato- rischer Wandel und politische Macht multinationaler Unternehmen, ,2 See A. Giddens: Jenseits von Links und Rechts. Die Zukunff Berlin 2000. radikaler Demokratie, Frankfurt a.M. 1996; U. Beck: Die Subpolitik der Globalisierung. Die neue Macht der Multinationaien Unter- ,7 See M. Wortmann: Globalisierung, Gewerkschaften und Euro- nehmen, in: Gewerkschaftliche Monatshefte, 1996, No. 12, pp. 673- p&ische Betriebsr&te. Lehren aus der Havanna Charta von 1947/48, 680; S. Strange: Globaloney?, in: Review of International Political in: Kurswechsel, 1999, No. 1. pp. 68-78. Economy, 1998, Vol. 5, No. 4, pp. 704-711. 48 See E W. S c h a r p f: Demokratie in der transnationalen Politik, in: ,3 Seer Hirst, G. Thompson, op. cit. Internationale Politik, 1996, No. 12, pp. 11-20; F.W. Scharpf: Konsequenzen der Globalisierung f0r die nationale Politik, in: 4, See D. S u Ill v a n : Measuring the Degree of Internationalization of a F{rm, op. cit., p. 330. Internationale Politik und Gesellschaft, 1997, No. 2, pp. 184-192. 126 INTERECONOMICS, May/June 2000 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Intereconomics Springer Journals

Measuring corporate internationalisation

Intereconomics , Volume 35 (3) – Jul 3, 2008

Loading next page...
 
/lp/springer-journals/measuring-corporate-internationalisation-XRsyrVpsZl

References (25)

Publisher
Springer Journals
Copyright
Copyright © 2000 by HWWA and Springer-Verlag
Subject
Economics / Management Science; Economic Policy; European Integration; Economics general
ISSN
0020-5346
eISSN
1613-964X
DOI
10.1007/BF02927197
Publisher site
See Article on Publisher Site

Abstract

MULTINATIONAL CORPORATIONS Christoph D6rrenb&cher* A Review of Measurement Concepts and their Use Measures of corporate internationalisation have gained crucial importance in the recent debate on globalisation, since many scholars link globalisation to a quantitative increase in the international activities of firms. Opinions on the extent of this increase differ widely, however, depending on what measurement concept is used. As there is no universally applicable measurement concept, researchers face the difficult task of bringing research questions, measurement concepts and data availabifity into line. he recent debate on globalisation has generated a that ownership structure or management composition T wider interest in the transborder activities of can be internationalised too. corporations. While some authors are convinced that Given these differences in definition, this paper tries multinational corporations (MNCs) are best symbolis- to give an overview of different approaches to ed by an octopus whose tentacles try to grasp the measuring and comparing corporate internationali- whole world, 1 other authors paint a completely sation. Based on the assumption that there is more different picture. Here the MNC is seen more or less than one way to "correctly" measure internationali- as a lethargic animal, whose presence abroad is sation, this paper concentrates on the meaning- rather limited? Both metaphors can be justified when fulness of different criteria and indices. The first considering individual companies, at least if one takes section deals with individual internationalisation the following definition as a basis: according to indicators. This is followed by a discussion on how Dunning "A multinational or transnational enterprise is regional diversification in internationalisation can be an enterprise that engages in foreign direct measured and expressed. Next we look at different investment and owns and controls value adding composite indicators (or indices) and discuss issues activities in more than one country". 3 of reliability, exactness and validity of indicators and It is probably not wrong to state that today most indices. The final section deals with the difficulties of large corporations fulfil this definition, which is widely bringing research questions, indicators and data accepted in academia, business, national govern- availability into line. ments and international organisations. For instance, a Individual Internationalisation Indicators recent compilation of data on the internationalisation of the 100 largest German companies has found that As mentioned above, the term "internationalisation" it is difficult to find a strictly national company. Even has different connotations - even if restricted to some of the utility companies are to a small degree corporate internationalisation. Some might think of active in foreign markets? Thus, thinking about larger intemationalisation as a process, others see it more corporations, the question is not which companies are as a certain level of international intertwining reached internationalised, but rather how companies differ in regard to their internationalisation. This is especially true if we do not restrict internationalisation to the See R. J. Barnet, J. Cavanagh: Global Dreams. Imperial Corporations and the new world order, New York 1994; D.C. Kor- definition that value adding activities have to take t e n : When Corporations Rule the World, West Hartford 1995. place in more than one country, but also recognise 2 See P. Hirst, G. Thompson: Globalization in Question. The International Ecomomy and the Possibilities of Governance, Cambridge 1996. * Wissenschaftszentrum for Sozialforschung Berlin (WZB) and For- J. H. Dunning: Multinational Enterprises and the Global schungsgemeinschaft for Aui3enwirtschaft, Struktur- und Techno- Economy, Wokingham 1992, p. 3. Iogiepolitik (FAST) e.V., Berlin, Germany. For useful comments the author would like to thank Douglas van den Berghe (University of 4 FAST 1999, reprinted inA. Hassel, M. HSpner, A. Kurdel- Rotterdam), Dieter Plehwe (WZB), Stefano Vescovi (WZB) and busch, B. Rehder, R. Zugeh0r: Dimensionen derlnternatio- Michael Wortmann (FAST e.V., Berlin). An earlier draft of this paper nalisierung: Ergebnisse der Unternehmensdatenbank, Internationali- was presented at the workshop "Dimensionen der Internationali- sierung der 100 gr56ten Unternehmen in Deutschland, Working sierung von Unternehmen" at the Max-Planck-lnstitut f5r Gesell- Paper 1/2000, Max-Planck-lnstitut f~Jr Gesellschaftsforschung, KSIn schaftsforschung, Cologne 16th/17th December 1999. INTERECONOMICS, May/June 2000 119 MULTINATIONAL CORPORATIONS Table 1 Individual Internationalisation Indicators Structural Indicators Performance Indicators Attitudinal Indicators 1. Relating to foreign activities 1. Foreign sales 1. 'Soft' indicators 9 Number of countries a company is 9 Demand: Amount of foreign sales 9 Ethno-, poly-, regio- or geocentric management style according to: active in by customer location organisational complexity, 9 Number or proportion of foreign affiliates Exports from the home country authority, decision making, 9 Number or proportion of cases of + Revenues of foreign affiliates communication flows, non-capital involvement abroad - their revenues from exports to the recruiting, staffing, (e.g. strategic alliances, franchised home country control operations) 9 Amount or proportion of foreign assets 2. 'Hard' indicators 9 Amount or proportion of value added 9 Supply: Amount of sales of foreign 9 International experience of top abroad affiliates managers 9 Amount or proportion of sourcing abroad Sum of revenues of foreign affiliates cumulative duration of the years top 9 Number or proportion of foreign managers spent working abroad employees weighted by the total years of their working experience 2. Relating to governance structures 2. Operating income abroad 9 Number of stock markets on which a Sum of operating income of foreign company is listed affiliates 9 Amount or proportion of shares owned by foreigners 9 Number or proportion of non-nationals in the board of directors by a company. While this is only a question of different [] the number of stock markets on which a company perspectives, there are strong differences of opinion is listed, as to what indicators are suitable for measuring cor- [] the amount/proportion of shares owned by for- porate internationalisation. Following a welt establish- eigners, ed framework in international business literature, we [] the number/proportion of non-nationals on the can distinguish structural indicators, performance board of directors. indicators and attitudinal indicators2 Performance Indicators Structural Indicators Performance indicators measure the degree to Structural indicators try to give a picture of the which the success or failure of corporate activity international entanglement of a corporation at a during a certain period of time (usually one year) is certain time. Here we find several indicators that are connected to foreign countries. The two main related to the foreign activities of MNCs such as: indicators are turnover and operating income. Two [] the number of countries a company is active in, issues have to be addressed when considering turn- over: [] the number/proportion of foreign affiliates, [] The first issue is demand. The main question here [] the number/proportion of cases of non-capital in- is, to what extent turnover is generated in foreign volvement abroad (e.g. strategic alliances, franchised countries. The relevant indicator - amount of foreign operations etc.), sales by customer location - is calculated as exports [] the amount/proportion of foreign assets, from the home country plus revenues of the foreign affiliates (except those revenues that come from [] the amount/proportion of value added abroad, exports to the home country of the parent company). [] the amount/proportion of sourcing abroad, [] the number/proportion of foreign employees. 5 See H. V. Perlmutter, D.A. Heenan: Multinational Organi- zation Development, Reading 1979, p. 16; D. S u I I i v a n : Measuring the Degree of Internationalization of a Firm, in: Journal of International A second group of indicators describes the inter- Business Studies, 1994, Vol. 25, No. 2, p. 331; I. B&urle: Inter- nationalisation of the governance structures of a nationalisierung als Prozessph&nomen. Konzepte - Besonderheiten - corporation. Relevant indicators here are: Handhabung, Wiesbaden 1996, p. 9. 120 INTERECONOMICS, May/June 2000 MULTINATIONAL CORPORATIONS Table 2 The Measurement of Regional Diversification Regional Concentration Network Extension Geographical and Cultural Distance 9 Network spread index: 9 Countries are weighted according to 9 Homogeneous vs. heterogeneous distribution of foreign activities (letto-Gillies 1998) their geographic and cultural distance (Schmidt 1981) to the home country Number of foreign countries in which (Kutschker 1993) 9 Extent to which the regional distribution a company owns affiliates 9 Psychic dispersion index: of a certain indicator at a given company as a proportion of (Sullivan 1994) complies with the total distribution of this indicator in the world Number of zones with different cognitive maps relating to (Perriard 1995) total number of countries in which management principles in which a foreign direct investment has occurred minus one company is active (= home country of the company) (out of a total of 10 zones worldwide) that several attitudinal indicators, differing in how [] The second issue is supply. The main question statistically measurable they are, are in use. A rather here is, to what extent business activity is transacted "soft" qualitative indicator was developed by Perl- by subsidiaries located in foreign countries. The mutter. This indicator, which underlies or is explicitly relevant indicator - amount of sales of foreign applied in many studies 1~ distinguishes four types of subsidiaries - is calculated as the sum of turnover of headquarters management's orientation towards their foreign subsidiaries as a proportion of the sum of foreign subsidiaries. ~1 To decide whether the central turnover of the parent company and its national management of an MNC follows an ethnocentric (i.e. subsidiaries. 8 home-country oriented), a polycentric (i.e. host- A second important performance indicator is the country oriented), a regiocentric (i.e. regionally orient- amount of operating income earned abroad. This ed) or a geocentric (i.e. globally oriented) approach, indicator is usually calculated as the sum of operating Perlmutter/Heenan '2 proposed a multi-item indicator, income generated by all foreign affiliates. addressing inter alia the following aspects: complexity of organisation, authority/decision making, eva- Attitudinal Indicators luation/control, rewards/punishment, communica- Attitudinal indicators try to give a picture of how tion/information, recruiting/staffing. However, due to MNCs view foreign countries and treat their their premise that recruiting/staffing is of central subsidiaries in foreign countries. 7 They usually imply importance in the internationalisation of an MNC, an actor centred view, focusing on the management Perlmutter/Heenan 13 concentrated their own empirical or parts thereof. Although there are doubts as to study on personnel function (manpower planning, whether attitudinal characteristics can be measured manpower administration and manpower control). with sufficient reliability, 8 there is also strong support for the use of attitudinal indicators, as expressed in See H. V. P e r I m u t t e r, D.A. H e e n a n : Multinational Organiza- the following quotation: "...external and quantifiable tion Development, op. cit., p. 17. measures such as the percentage of investment ~0 See e.g.G. H e d I u n d : The Hypermodern MNC: A Heterarchy?, overseas or the distribution of equity, are useful but in: Human Resource Management, 1986, Vol. 25, Spring, pp. 9-25; C. Bartlett, S. Ghoshal: Managing across Borders. The not enough ... The more one penetrates the living transnational Solution, Boston 1989; I? Marginson, K. Sisson: reality of how decisions are made in an MNC, the The Structure of Transnational Capital in Europe: The Emerging Euro- Company, in: R. Hyman, A. Ferner (eds.): New Frontiers in more weight must be given to how executives think European Industrial Relations, Oxford 1994, pp. 15-51 ; C. D 5 r r e n - about doing business around the world". 9 b&cher: Vom Hoflieferanten zum Global Player, Unternehmens- reorganisation und nationale Politik in der Welttelekommunikations- [ndustrie, Berlin 1999. These different positrons are reflected by the fact ll See H. V. Perlmutter: The Tortuous Evolution of the Multi- national Corporation, in: Columbia Journal of World Business, 1969, Since there is no consolidation of intrafirm transfers, this indicator Vol. 4, pp. 9-18;H.V. Perlmutter, D.A. Heenan: How Multi- does net necessarily express the home/foreign split of production! national should your Top Managers be?, in: Harvard Business Review, 1974, November-December, pp. 121-132. ' Of course, in reverse this indicator also gives a picture of the role of the entities in the home country. ~2 H.V. Perlmutter, D.A. Heenan: Multinational Organization Development, op. cit., p. 18 f. See D. S u Ill v a n : Measuring the Degree of Internationalization of a Firm, op. cit., p. 325. ~3 Ibid. INTERECONOMICS, May/June 2000 121 MULTINATIONAL CORPORATIONS Table 3 Composite Indicators Used to Measure Corporate Internationalisation Transnationality Index Transnational Activities Spread Index Degree of Internationalization Scale (UNCTAD 1995) (letto-Gillies 1998) (Sullivan 1994) Ratio of foreign sales to total sales Ratio of foreign sales to total sales Ratio of foreign sales to total sales + Ratio of foreign assets to total assets + Ratio of foreign assets to total assets + Ratio of foreign assets to total assets + Ratio of foreign employment to total + Ratio of foreign employment to total + Ratio of foreign affiliates to total employment employment affiliates divided by 3 divided by 3 + "International experience of top management" (see above) multiplied by + "Psychic dispersion" of internationa~ Number of foreign countries in which a operations company owns affiliates as a proportion of total number of countries in which foreign direct investment has occurred minus one (= home country of the company) According to them, the degree of internationalisation company tends to have the same spatial distribution of a company increases from ethno-, through poly- as total world car production. and regiocentrism to geocentrism. A second way of tackling geographical variation of An attitudinal indicator which can be better corporate internationalisation is to focus on the num- measured statistically was developed by Sullivan. TM ber of countries a company is active in. Developed by According to him, the international mindset of a letto-Gillies, 17 the so called "network spread index" is company increases with the international experience calculated, ceteris paribus, by taking the number of of top managers, measured as the cumulative foreign countries in which a company has affiliates as duration of the years top managers spend working a proportion of all countries where foreign direct abroad weighted by the total years of their working investment has occurred, minus one country, i.e. the experience. home country of the company. Finally, based on the assumption that there are Internationalisation as Regional Diversification striking differences between countries, which in- In addition to the dichotomy of home versus fluence the internationalisation behaviour of com- foreign, different geographical variation indices can be panies, TM Kutschker 'g proposed that foreign countries calculated for many of the indicators given above. For should be weighted according to their geographical instance Schmidt 's uses the Herfindahl-index to place and cultural distance to the home country of the companies on a continuum between a completely company. Similarly, Sullivan 2~ introduced an indicator homogeneous and a completely heterogeneous he calls "psychic dispersion". According to this indi- distribution of their foreign activities to different cator, the world is divided into ten zones with different countries. Perriard" uses the Gini-index to measure cognitive maps relating to management principles. the extent to which the regional distribution of a The more zones a company is active in, the larger the certain indicator at a given company complies with psychic dispersion of its internationalisation. the total distribution of this indicator in the world. Take the example of car production: following Perriard's methods, one could look how far the spatial " G. letto-GiIlies: Different Conceptual Frameworks for the Assessment of the Degree of Internationalization: an Empirical distribution of the car production of a specific Analysis of Various Indices for the Top 100 Transnational Corpo- rations, in: Transnational Corporations, 1998, Vol. 7, pp. 17-39. " D. Sullivan: Measuring the Degree of Internationalization of a ~8 See J. Johanson, J. Vahlne: The Internationalization Pro- Firm, op. cit., p. 332. cess of the Firm - A Model of Knowledge Development and Increas- ing Foreign Market Commitments, in: Journal of International Busi- 1, R. S c h m i d t : Zur Messung des Internationalisierungsgrades von ness Studies, 1977, Vol. 8, pp. 23-32. Unternehmen, in: W. Wacker, H. Haussmann, B. Kumar (eds.): Internationale UnternehmensfQhrung. Managementprobleme '~ Kutschker: Dynamische Internationalisierungsstrategie, Dis- international t&tiger Unternehmen, Berlin 1981, pp. 57-70. kussionsbeitrag Nr. 41, Wirtschaftswissenschaftliche Fakult&t Ingolstadt, Katholische Universit~lt Eichst&tt, Ingolstadt 1993. '~ M. Perriard: Towards a Measure of Globalization, Institute of 2o D. S u Iliva n : Measuring the Degree of Internationalization of a Economic and Social Sciences University of Fribourg Working Paper No. 250, Fribourg 1995. Firm, op. cit. 122 INTERECONOMICS, May/June 2000 MULTINATIONAL CORPORATIONS Internationalisation Indices In order not to favour companies from one specific type of industry 27 in the ranking of companies, 28 the Indices, or composite indicators, are formed by "transnationality" index is calculated as the average of combining individual indicators. Despite the fact that the three above mentioned ratios. numerous studies are based on individual indicators, there is a broad consensus that composite indicators 2. The "transnational activity spread index". This are more suitable to measure corporate internatio- index, proposed by letto-Gillies, 29 is a combination of nalisation. Firstly, given the multidimensionality of two indices. According to letto-Gillies the index used internationalisation, 21 restricting the measurement to by UNCTAD ignores the spread of foreign activities. one single item inevitably means that only a part of the But her own network spread index, which indicates whole phenomenon of corporate internationalisation regional diversification (see above), does not take into is represented. Secondly, the multidimensionality of account the amount or the share of activities abroad. the phenomenon means that, depending on what As a result she proposes to combine the two indices indicator is used, single transactions could lead to so that the UNCTAD index is weighted by the contradictory results concerning the degree of "network spread index". In detail, the resulting new internationalisation. 22 For instance, modernising a index called the "transnational activity spread index" plant abroad might lead to an increase in the is calculated as follows: The average of the following proportion of foreign assets, but due to the labour ratios: foreign sales to total sales, foreign assets to saving effects of modernisation, the proportion of total assets, and foreign employment to total em- foreign employees might at the same time decrease. ployment, is multiplied by the number of foreign Thirdly, using individual indicators does not allow countries a company is active in as a proportion of the systematic control of measurement errors, contingent total number of countries where foreign direct invest- influences and transfer pricing manipulations (see ment has occurred, minus one (the home country). In below). addition to this comprehensive index letto-Gillies indicates that every individual indicator of the Many scholars agree that the development of UNCTAD index might also be weighted by the sophisticated indices to measure corporate inter- network spread index, thus forming three sub-indi- nationalisation is still in its infancy? 3 Only three ces: the "sales spread index", the "asset spread different composite indicators can be identified in the index" and the "employment spread index". literature on the subject: 2' 3. The "degree of internationalization scale". This 1. The "transnationality index" used by UNCTAD. index, proposed by Sullivan, 3~ was developed in a This composite indicator first appeared in the World bottom up-process, using publicly available data. Investment Report 1995.2 s The aim of the indicator is Unlike UNCTAD, which selected its three individual "... to capture fully the extent of involvement of TNCs indicators based on preliminary assumptions about [Transnational Corporations, C.D.] in the world their usefulness in expressing the phenomenon of economy"? 6 The index draws on three different ratios: internationalisation, Sullivan started with a total of Foreign sales to total sales, foreign assets to total nine indicators encompassing structural, performance assets, and foreign employment to total employment. and attitudinal items. Using item-total analysis 31 on a 2, This multidimensionality is also reflected in the wealth of different 28 Ibid., p. 24. single indicators on corporate internationalisation as described 27 Labour intensive, capital intensive or export intensive industry. above. 28 This index is annually calculated for the world's top 100 non- 22 SeeH. German, S. Raab, M Setzer: Messung derGIoba- financial MNCs, which are selected on the basis of their foreign lisierung: Ein Paradoxon, in: U. Steger (ed.): Facetten der assets. The collection and the calculation of the data is undertaken Globalisierung. C)konomische, soziale und politische Aspekte, Hei- by the Studies and Competence Centre for Organisational and Policy delberg 1999, p. 7. Research in European Business (SCOPE) at the Erasmus University of 23 See D. S u I I i v a n : Measuring the Degree of Internationalization of Rotterdam. For a recent interpretation of the data,see D. v a n d e n aFirm, op. cit.; M. Perriard, op. cit.;K. Ramaswamy, K.G. Berghe: Internationalisation Strategies, Relocation Processes and Kroeck, W. Renforth: Measuring the Degree of Internationali- Employment: Multinational Enterprises in the Age of Globalisation. zation, A Comment, in: Journal of International Business Studies, Paper Presented at the 14th Annual Employment Research Unit 1996, Vol. 27, No. 1, pp. 167-177. Conference: Globalisation, Employment and the Workplace, Cardiff Business School, 8/9. September 1999. 24 In addition to these indices, M. Perriard, op. cit., proposes a set of 13 quantitative and qualitative indicators to determine the degree ~ G. Letto-Gillies, op. cit. of the internationalisation of a corporation. However since there is no 3o D. Sullivan: Measuring the Degree of Internationalization of a integration of the indicators, the use of this instrument is very Firm, op. cit.; D. S u I l iva n : Measuring the Degree of Internatio- restricted. nalization, A Reply, in: Journal of International Business Studies, 2~ UNCTAD: World Investment Report 1995, New York. 1996, Vol. 27, No. 1, pp. 179-192. INTERECONOMICS, May/June 2000 123 MULTINATIONAL CORPORATIONS representative sample of 74 MNCs, five "good" nationalisation at the time of the headquarters indicators have been identified. These indicators, all relocation, there might be dramatic changes in some weighted by the factor one, form the degree of figures indicating the internationalisation of the internationalization scale. In detail the index is company. Some further contingent influences are composed of the following indicators: The ratio of industry specific. They occur for instance in engineer- foreign sales to total sales, foreign assets to total ing, when large projects, covering many years, are assets and foreign subsidiaries to total subsidiaries, invoiced in one year, resulting in a tremendous jump in performance indicators such as turnover or operating the international experience of top managers (measured as the cumulative duration of the years top income. managers spend working abroad weighted by the [] Transfer pricing manipulations systematically total years of their working experience) and the distort the geographic distribution of performance psychic dispersion of international operations. indicators such as turnover and especially income. Transfer pricing manipulations occur when inter- Reliability, Exactness and Validity nalised transactions between related units of an MNC are not calculated on the "arm's length principle". If one recognises that scientific progress is not Aims of transfer pricing manipulations are manifold: possible without adequate measurement 32 then it is for example, tax avoidance, lowering customs duties, important to discuss the issues of reliability, exact- repatriation of capital. However, very little is yet known ness and validity of the indicators described so far. about the significance of transfer pricing manip- As mentioned above the use of individual indicators ulations. In a recent overview of different studies, to measure corporate internationalisation is in general Plasschaert states that "... all in all, the only plausible, not very reliable since it does not include a systematic although trivial, conclusion .... is that transfer-price control of measurement errors, contingent influences manoeuvres are probably practised much more and transfer-pricing manipulations (beyond obvious frequently than TNCs [Transnational Corporations, implausibility). C.D.] are willing to admit, but much less than is alleged in some circles". 3s [] Measurement errors might occur with "soft" atti- tudinal indicators, such as the Perlmutter typology. Although composite indicators are generally more Here, biased views and cognitive dissonance, both of reliable, their use is not without problems. We have the interviewer and the interviewee, may lead to discussed three different composite indicators, each flawed results. Another source of measurement error of which claims to adequately measure corporate is the fact that the definition of a certain indicator internationalisation. However, taking the multidimen- must be clear, detailed and viable. Take the example sionality of the phenomenon as a criterion, the three of the indicator of the ratio of foreign employment to indices have different orientations. The UNCTAD total employment. In the first place, an exact definition index combines two structural and one performance of what an employee is is necessary: for example, indicator. The "transnational activity spread index", should there be a headcount, or are part time em- proposed by letto-Gillies is even more concentrated ployees converted into full time equivalents according on structural aspects of internationalisation; she uses to their working hours; 33 do employment figures the UNCTAD index as a base but weights it with represent the average of the year or is the figure taken another structural indicator. The only index covering at a common qualifying date etc. Secondly, is it viable to compare employment on a one to one basis, taking 3, Item-total analysis correlates each potential indicator at a given into account the large sectoral and national sample with the indicator-corrected scale of the total score. The aim is to get a collection of indicators which has a high average differences in the costs of creating a workplace? correlation with the total score. [] Contingent influences are influences that change 32 Korman 1974 and Schwab 1980 cited in D. Sullivan: Measuring the Degree of A Reply, op. cit., p. 338. indicators beyond real material changes. The most obvious contingent influence is exchange rate 33 This point is crucial in industries which mainly use part time workers, such as the fast food industry or the transport industry. fluctuations. Exchange rate problems occur parti- 34 One example is the relocaton of the Ikea headquarters from cularly when comparing the internationalisation of Sweden to Denmark in 1975, see C. DSrrenb&cher, H. R. companies on a historical or on a national basis. Meil3ner: IKEA -The Hollow Elk Corporation, in: Informationen 0ber multinationale Konzerne, 1991, No. 2, p. 23. Another contingent influence which occurs from time 35 S. Plasschaert: Transnational Corporations: Transfer Pricing to time is the relocation of an MNC's headquarters to and Taxation, in: UNCTAD (ed.): Transnational Corporations and another country? 4 Depending on the degree of inter- World Development, London 1996, p. 406. 1 24 May/June 2000 INTERECONOMICS, Internationalization, MULTINATIONAL CORPORATIONS all three dimensions of internationalisation is the one not the construction of such a general index is developed by Sullivan. This index combines structural feasible in the foreseeable future28 For instance, (two indicators), performance (one indicator) and Hassel et al. recently considered this question and attitudinal (two indicators) aspects. However, it is rejected the general approach. 39 questionable whether even this comprehensive Research Questions, instrument adequately covers the internationalisation Indicators and Data Availability phenomenon: The lack of a general indicator encompassing the [] Firstly, the viability of both attitudinal indicators whole phenomenon of corporate internationalisation used by Sullivan is heavily disputed. Ramaswamy et is a severe problem if the research purpose is to rank a126 maintain that the length of international experi- companies according to their overall degree of ence is not a good yardstick for the international internationalisation. However, if corporate internatio- mindset of the management, since there are several nalisation is part of a more specific research question, other factors with a stronger influence, such as the then the aim of the research might give some geographic spread of the company, its policy and its guidance on how to construct a viable index (what administrative heritage. In addition, Ramaswamy et al. indicators, how much weight). For instance, if one is criticise the plausibility of the "psychic dispersion of interested in ranking companies according to their internationalisation" concept. Using this concept production abroad, the best indicator would probably would mean, for instance, that a company operating be value added abroad. The proportion of foreign in France and the United Kingdom has a higher employees and foreign assets might be good psychic dispersion of its internationalisation than a yardsticks. In contrast, the number of stock markets company operating in Japan, India, Israel and Brazil. on which a company is listed does not seem to be a [] Secondly, there are strong concerns as to whether very plausible indicator. Sullivan's indicator can really claim to be all-encom- It might also be easier to find indices for intra- passing, because the way he chooses the indicators industry comparisons than for cross-industry studies, for the index (i.e. selecting indicators which have the since in this case indices can be optimised according highest average correlation with one another) to industry specifics. For instance the "transnational systematically excludes indicators that might be spread index" might be appropriate for industries, important to express the phenomenon. Following such as in the transport industry, in which the extent Ramaswamy et al. the index used by Sullivan "... of the geographical reach is of special importance. seems to exclude several common strategic options for internationalisation such as indirect exporting, But even if a viable index to rank companies from licensing, joint ventures ...,,37 different countries and/or industries is found, two problems remain. The first problem is data availability. A second problem is the question as to how much Apart from mail surveys, there are only a few sources weight different individual indicators should have in an from which data expressing the international index. Up to now there is no information on how to entanglement of a company is available: these are determine the importance a specific indicator should specialised company listings, company handbooks, have in the construction of an index. Because of this databases and annual reports. In general these lack of information, two of the indices mentioned here sources are very fragmented and incomplete (for treat all indicators equally. In contrast, the letto-Gillies example they lack interesting information, such as the index allots more weight to the network spread com- amount of shares owned by foreigners or the ponent than to the three other single components. proportion of value added abroad etc.). In addition, To sum up the discussion, there is neither a single indicator nor an index that satisfactorily measures the overall degree of the internationalisation of a firm. 3g Their new index, which sheds light on the previously neglected financiar dimension of intemationalisation, combines an index that Furthermore, there are differing views as to whether or measures the real dimension of corporate internationalisation (using a somewhat reworked "transnational activity spread index") with different indicators which measure the financial dimension of cor- ~6 K. Ramaswamy, K. G. Kroeck, W. Renforth, op. cit., porate internationalisation (i.e. foreign owners as percentage of total p. 173. ownership, number of listings in foreign stock exchanges and the use of international accounting standards). ,See A. Hassel, M. HOp- 37 Ibid., p. 174. ner, A. Kurdelbusch, B. Rehder, R. ZugehSr, op. cit.; 38 See D. S u Ill v a n : Measuring the Degree of Internationalization, A. Hassel, M. HSpner, A. Kurdelbusch, B. Rehder, AReply, op. cit., p. 190; K. Ramaswamy, K.G. Kroeck, W. R. Z u g e h 5 r: Two Dimensions of the Internationalization of Firms, Renforth, op. cit,, p. 176. K01n 2000. INTERECONOMICS, May/June 2000 1 25 MULTINATIONAL CORPORATIONS data availability has a high sectoral and national vari- used either to explain problems of state sovereignty ance, and is the result of different accounting rules and declining trade union power '2 or to contest these and publishing habits. For instance, in contrast to the developments by arguing that there is no, or only US General accounting principles, the German modest, growth in the international entanglement of accounting standard does not require a home/foreign MNCs. ,3 split of the assets. Different practices exist in the Regardless of the field of study or the research publication of employment data, with many German purpose, there are some strong arguments against a companies publishing a home/foreign split, some- positivistic interpretation of quantitative data on the thing that is very unfamitiar to Japanese companies. internationalisation of MNCs. Take the case of Finally, due to different modes of compilation, data is management science and its attempt to explain the often incompatible especially for companies from financial performance of an MNC by its inter- different countries. While in the long run some of nationalisation. It is a commonplace that in this kind of these problems will alleviate due to the ongoing study the measurement concept strongly influences standardisation of accounting principles, other pro- the result. However, even using the same indicator blems will persist. does not give a consistent picture." Furthermore, macroeconomic data, for example on the (probably The second and more general problem is the use of indicators and indices in wider economic and socio- growing) contribution of large MNCs to the world's political research questions. The actual use of gross domestic product, has to be interpreted quantitative data on corporate internationalisation carefully. Collecting information on the quantitative concentrates on three purposes. share of a few large MNCs in a certain market is only a first step towards finding out something about the 1. In management science, many empirical studies power of MNCs. In addition, there has to be at least a look at the correlation between the internationali- look at the development of competition and its sation of a company and its financial performance. influence on the level of prices and the structure of 2. In macroeconomics, the use of data on individual supply in this market. 's Finally, the debate on the MNCs is not very widespread. `~ However there are influence that corporate internationalisation has on some exceptions, such as concentration issues, state sovereignty and trade union power is a striking where the use of company specific data is necessary. warning against a positivistic interpretation of For instance, the list of the top 100 MNCs of the quantitative data on corporate internationalisation. world, annually published by the UNCTAD, is used to Many empirical studies have found that in the long run find out whether the share of those companies in there is no big rupture in the quantitative (and world gross domestic product is increasing or not. qualitative) dimension of corporate internationali- 3. In political science (especially in the sub-fields of sation of MNCs. '8 However this does not mean that all international political economy and industrial rela- problems concerning state sovereignty or declining tions) the internationalisation of MNCs has served trade union power are imagined. The answer is simply different purposes. In the past, the extent of foreign that other things, like general economic conditions '7 manoeuvres of MNCs was frequently used to explain or the state's need for tax income, '8 have changed underdevelopment in third world countries? 1 More dramatically over time, bringing MNCs into better recently the internationalisation of MNCs has been bargaining positions, whether their degree of internationalisation has significantly increased or not. ,0 Scholars usually use aggregate data, such as the flow or the stock of foreign direct investment. For a discussion of the problems 4~ See C. DSrrenb&cher, op. cit., p. 56. associated with these indicators see M. Wortmann, C. D6rren- b&cher: Multinationale Konzerne und der Standort Deutschland, " Seee.g.P. Hirst, G. Thompson, op. cit.; P. Doremus, W. in: W. Fricke (ed.): Jahrbuch Arbeit und Technik, 1997, pp. 28-42. Keller, L. Pauly, S. Reich: The Myth of the Global Corpo- ration, Princeton 1988; M. Wortmann: Zur Logik yon Wachstum " Seee.g.F. FrSbel, J. Heinrichs, O. Kreye: Dieneuelnter- und Restrukturierung multinationaler Konzerne, in: C. D6rren- nationale Arbeitsteilung, Reinbek 1977. b&cher, D. Plehwe (eds.): Grenzenlose Kontrolle? Organisato- rischer Wandel und politische Macht multinationaler Unternehmen, ,2 See A. Giddens: Jenseits von Links und Rechts. Die Zukunff Berlin 2000. radikaler Demokratie, Frankfurt a.M. 1996; U. Beck: Die Subpolitik der Globalisierung. Die neue Macht der Multinationaien Unter- ,7 See M. Wortmann: Globalisierung, Gewerkschaften und Euro- nehmen, in: Gewerkschaftliche Monatshefte, 1996, No. 12, pp. 673- p&ische Betriebsr&te. Lehren aus der Havanna Charta von 1947/48, 680; S. Strange: Globaloney?, in: Review of International Political in: Kurswechsel, 1999, No. 1. pp. 68-78. Economy, 1998, Vol. 5, No. 4, pp. 704-711. 48 See E W. S c h a r p f: Demokratie in der transnationalen Politik, in: ,3 Seer Hirst, G. Thompson, op. cit. Internationale Politik, 1996, No. 12, pp. 11-20; F.W. Scharpf: Konsequenzen der Globalisierung f0r die nationale Politik, in: 4, See D. S u Ill v a n : Measuring the Degree of Internationalization of a F{rm, op. cit., p. 330. Internationale Politik und Gesellschaft, 1997, No. 2, pp. 184-192. 126 INTERECONOMICS, May/June 2000

Journal

IntereconomicsSpringer Journals

Published: Jul 3, 2008

There are no references for this article.