The goal of my thesis is to investigate which factors have the highest influence on American multinational companies when these companies make their headquarter location decision. We will conduct our research using a set of 35 global cities. Companies take numerous factors into account when making headquarter location decisions and we will elaborate on these later on. These factors can be influenced by the policy makers in an urban area. It is important to denote why urban areas should try to attract multinational companies’ headquarters. First, according to Holloway and Wheeler (1991) and Meijer (1993), a high agglomeration of such offices both reflects and is a causal factor in the economic power of a region. Second, Klier and Testa (2002) have found that headquarters are major consumers of high-skilled and well-paid labor. And third, technological spillovers may be caused by the presence of a high concentration of large firms’ headquarters in knowledge-intensive sectors; these spillovers have a positive influence on the neighborhood.
We will start by defining the term global city, since all cities we include in our research are of this type. Following we will explain why companies become multinational companies, by explaining this we uncover what companies need in order to locate their headquarters abroad. Then we will elaborate on the factors global cities need to possess in order to attract MNCs.
Definition of a global city.
Emergence of global cities
According to Sassen et al. global cities have become increasingly important in the last decades because of major changes in the macro-economic landscape. One of these major changes is the ascendance of information technologies and the associated increase in the mobility and liquidity of capital. Cross-border economic processes such as flows of capital, labor, goods, raw materials and tourists, have already existed for a long time. These processes, however, mainly used to take place in the inter-state system. The key actors in this system were national states and the entire international system was embedded in this inter-state system. Due to privatization, deregulation, the opening up of national economies to foreign firms, and the growing participation of national economic actors in global markets the national as a spatial unit has partially been unbundled or at least been weakened. This weakening has resulted in the emergence of other spatial units or scales. Among these are the sub-national, notably cities and regions; cross-border regions encompassing two or more sub-national entities; and supra-national entities, i.e. global digitalized markets and free trade blocs. Sassen states that the changes in the macro-economic environment, which are mentioned before, have led to the emergence of global cities.
Difference global vs world cities
Sassen points out that she knowingly has chosen the term global city, instead of the obvious alternative, world city. She has done this because the term world city refers to a type of city which has been around for centuries. Sassen states that most of today’s global cities are also world cities, but there may well be some global cities today that are not world cities in the full, rich sense of that term. We will elaborate on the world city later, when we discuss John Friedman’s work.
Sassen’s Global City Model
The goal of my thesis is to find out what global-city characteristics induce MNCs to choose a certain global city as the location for their headquarters. In order to fully be able to answer this question it is important that I start with explaining what a global city exactly is. The explanation I provide is based on Saskia Sassen’s global city model, her model consists of seven hypotheses and I will explain these briefly.
First, globalization evokes the geographical diffusion of economic activities and along with the simultaneous integration of such geographically diffused activities, this is a main factor which induces the growth and importance of central corporate functions. When a firm’s operations become more dispersed across different countries, the firm’s central functions become accordingly more complex and strategic.
Second, since these central functions become this complex, the headquarters of MNCs become more and more reliant on outsourcing: they buy a share of their central functions from highly specialized service firms – accounting, legal, public relations, programming, telecommunication and other such services. While, not long ago, the key site for the production of these central headquarter functions was the headquarter itself, nowadays a second key site has arisen. Although this is especially the case for firms involved in global markets and non-routine operations, all headquarters of large firms are buying more of such inputs rather than producing them in-house.
Third, the increasing demand for specialized services leads to a new agglomeration dynamic. The urban environment functions – due to the presence of a mix of firms, talents, and expertise from a broad range of specialized fields – as an information center. Being in a city becomes synonymous with being in an extremely and dense information loop.
Sassen’s fourth hypothesis is derived from the previous one. She states that the more headquarters outsource their most complex, unstandardized functions, the freer they are to choose any location. This is the case since less work that is done in the headquarters is subject to agglomeration economies. This indicates once again that the highly specialized and networked services sector is the key sector specifying the distinctive production advantages of global cities. While developing this hypothesis Sassen was responding to a very common notion, the fact that the amount of headquarters is what specifies a global city. In many countries it may still be the case that the country’s leading business center also possesses the largest concentration of headquarters. In some countries however, there are multiple locational options for such headquarters, since we find well-developed infrastructure outside of the leading business center.
Hypothesis five states that the specialized service firms, present in global cities, need to construct a transnational servicing network. This network consists of affiliates or other forms of partnerships, as a result of this there has been a strengthening of cross border city-to-city transactions and networks. This might be seen as the beginning of the formation of transnational urban systems. A result of this hypothesis is the fact that the gap between the economic fortune of these connected global cities and the economic fortune of their hinterland and their national economies becomes larger. Nowadays these transnational networks are of great importance for the major business centers of the world. In prospect to this, Sassen states that there is no such thing as a single global city.
Sassen’s sixth hypothesis states that there has been a raise in the degree of socio-economic inequality in global cities. This raise is caused by the growing number of high-level professionals and high profit making specialized service firms present in global cities. Since specialized services have gained significantly in importance, the value of top level professionals and their number has risen. The quality of these services depends to a large extent on the talent of the employees who perform these services, this makes proven talent an added value. This means that workers who possess of these attributes experience rapid increases in their reward structure and workers who do not possess of these attributes are likely to get caught in the opposite cycle. This is the cause of the growing level of inequality in global cities. Sassen’s sixth hypothesis has however been criticized by Chris Hamnett (1994). First, he states that Sassen’s concept of polarization is vague and undefined. She fails to make a distinction between absolute and relative social polarization and she shifts between occupational and income polarization. Second, Sassen’s paper fails to engage adequately with with existing work on social change, and the evidence of large-scale professionalization in the occupational structure of Western society and many global cities. And finally, Sassen seems to have focused only on New-York and Los Angeles when writing her thesis. These are two cities which are characterized by large-scale, continuing immigration and hence a large scale supply of low-wage workers. Hamnett argues that her attempt to extend the thesis to all global cities is problematic.
A seventh hypothesis, is that one result of the dynamics described in hypothesis six, is the growing informalization of a range of economic activities which find their effective demand in these cities, yet have profit rates that do not allow them to compete for various resources with the high-profit making firms at the top of the system. A possible solution, in order to survive under these conditions, is informalizing part of or all production and distribution activities.
The role of innovations in communication technology
Globalized economic sectors tend to be intensive users of the new telecommunications and computer technologies, on top of this, the output these sectors produce is becoming increasingly de-materialize. This raises a question as to whether they should benefit from agglomeration economies. Sassen points out the growing evidence that networks are a crucial variable that is to be distinguished from technical networks. Even before the current technologies were developed, these business networks were crucial. The business networks we have mentioned benefit significantly from agglomeration economies and hence thrive in cities even today when simultaneous global communication is possible.
Firms with large numbers of geographically dispersed factories and service outlets encounter new needs for central coordination and servicing. This implies a dynamic of simultaneous geographic dispersal and concentration and this dynamic is one of the key elements in the organizational architecture of the global economic system. Indeed, the leading financial centers inside countries concentrate a greater share of national financial activity than even ten years ago. This is an unexpected evolution for a globalized and digitized economic sector. This evolution is caused by the fact that national and global markets as well as globally integrated organizations require central places where the work of globalization gets done. In order to implement and manage global economic systems, finance and advanced corporate services are needed to be produced. The preferred sites for the production of these services are cities. Further, the firms who produce these services need a physical infrastructure containing strategic nodes with hyper-concentration of facilities. Finally, Sassen states that even the production process of the most advanced information industries is at least partly place-bound since their production process requires a combination of resources even when their outputs are hypermobile.
The changes in communication technology have not only influenced the degree of concentration of financial activity, they have also altered the role of centrality and hence of cities as economic entities. Formerly, the center was synonymous with the downtown or Central Business District. Due to new communication technologies, this is no longer the case. Nowadays what is seen as the center can assume several geographic forms, ranging from the CBD to a new global grid of cities. Sassen states that today one can observe centrality in three different forms. First, although the CBD has been altered by economic and technological change, it remains a form of centrality with great importance. Second, the center can extend into a metropolitan area in the form of a grid of nodes of intense business activity. The nodes in this grid are both connected by cyber-routes and more conventional forms of communications infrastructure, such as rail and highways connecting to airports. These conventional infrastructures are likely to maximize the economic advantages which follow from the new communication technologies. Finally, through telematics and intense economic transactions, transterritorial “centers” are being constituted, meaning that the major international financial and business centers are linked at the inter-urban level.
Why do firms become multinational firms?
This may seem like the logic consequence of a firm who grows and grows and is willing to seek other markets then its home market in order to sell more of its products and get higher profits. It is however not as easy as it sounds. Doing business in another market then your home market incurs a significant cost relative to the cost of doing business in your home market. Therefore, firms need to have large offsetting advantages in order to expand to foreign markets.
John Dunning has made a limited but nevertheless very useful framework which helps to inquire these offsetting advantages. According to this framework a there are three conditions required in order for a firm to proceed with foreign direct investments. The first condition is ‘Ownership Advantage’, this condition entails that the firm has to have a product or a production process which provides the firm a market power advantage in foreign markets. The second condition in Dunning’s framework is ‘Location Advantage’, this entails that the firm must have a reason to want to locate production abroad rather than concentrate it in the home country especially if there are scale economies at the plant level. The third and final condition of Dunning’s framework is ‘Internationalisation Advantage’. This condition tells that firms must have a reason to want to exploit its ownership advantage internally, rather than license or sell its product/process to a foreign firm.
In order to connect these theoretical ideas with practical firm and country characteristics, James R. Markusen focused on several other authors in his paper ‘Multinational Firms, Location and Trade.
Markusen calls it the ‘knowledge-capital’ model, but he notes at the same time that this is not a widely spread term.
Markusen starts with the ownership advantages in order to connect theory and practice. He notes that multinationals are related to R&D, marketing, scientific and technical workers, product newness and complexity and product differentiation. This denotes that multinational firms are firms who use an extensive amount of knowledge capital. Knowledge capital is a term which includes many different things such as the human capital of the employees; patents, blueprints, procedures and other proprietary knowledge. Finally the term also includes marketing assets such as trademarks, reputations and brand names. The reason why knowledge capital is associated with multinationals, in contrary to physical capital, is the fact that knowledge capital can easily be transported to foreign production facilities, at least relative to the services of physical capital. The second property of knowledge-capital which leads to the association of this type of capital with multinational firms is the fact that knowledge-capital often has a joint-input or ‘public-good’ property within the firm. Knowledge-capital often has a very high production cost, but once it is produced it can be transferred to foreign production facilities at a low cost and this without reducing the value or productivity of those assets in existing facilities.
The second property of knowledge capital which associates knowledge capital with multinational firms is the fact that knowledge capital often has a ‘public-good’ property within the firm. Knowledge capital generally has a large cost to be produced, but once it is produced it can be supplied at relatively low cost to foreign production facilities without reducing the value or productivity of those assets in existing facilities. In regard to this, multinationals become exporters of the services of knowledge-based assets: managerial and engineering services, financial services, reputations and trademarks.
Sources of location advantages differ between horizontal and vertical multinationals. For horizontal multinationals, which produce the same goods and services in several locations, location advantages arise from trade costs. Indeed, when trade costs were zero the multinational would produce everything in one plant, in order to maximize plant-level scale economies. The second source of location advantages for multinationals, again following form the occurrence of plant-level scale economies, is a large market in the potential host country. When the market in a possible host country is rather small, this market could be supplied by export.
For vertical MNCs, in contrary to horizontal MNCs, the sources of location advantages arise from low trade costs. Vertical MNCs, for example, exports the services of its knowledge capital and perhaps other intermediate inputs to a foreign production facility for final assembly and shipment back to the MNC’s home country. These transactions are likely to be encouraged by low rather than high trade costs. Another source of location advantages for vertical MNCs arises when the stages of production have different factor intensities and the countries have different relative factor endowments. This means that vertical MNCs will produce skilled-labor-intensive products and services in countries which have an abundant amount of skilled-labor present and less-skilled final assembly will be done in countries where low-wage unskilled labor forces are present. Fragmentation thus arises in order to exploit factor-price differences across countries.
The third and final type of advantages Dunning states in his framework are the internalization advantages. These advantages arise, like the ownership advantages, from the public-goods property of knowledge capital. The property of knowledge capital that makes it easily transferable to foreign locations makes it easily dissipated. When a MNC contracts a licensee, this licensee can absorb the MNC’s knowledge capital and then disrupt the contract with the MNC, still possessing the knowledge capital of the MNC. In regard to this risk, MNCs will prefer to internalize the transfer of knowledge capital.
Markusen states characteristics, which firms need to possess in order to become MNCs. This is not very interesting for my thesis since I am discussing firms who are already found to be MNCs.
Afterwards Markusen presents his findings on what characteristics countries should possess in order to be home and host to foreign direct investment. These findings are interesting since the goal of my thesis is to find out what characteristics of global cities can ensure MNCs to locate their headquarters in this particular city. These country characteristics are certainly not exactly the same as the global city characteristics, but they can definitely help us to form an image of what characteristics attract MNCs. We will list Markusen’s 10 country characteristics below.
1) The high-income developed countries are not only the major source of direct investment, they are also the major recipients. Most direct investment seems to be horizontal.
2) There has been a major boom of direct investment into the developing countries in the 1990s, but most of it has gone to the more advanced LDCs and to China. Little goes to the least developed countries.
3) Direct investment stocks have grown significantly faster than trade flows over the last two decades, even though trade barriers have fallen dramatically.
4) High volumes of direct investments are associated with similarities among countries in terms of relative factor endowments and per capita incomes, not differences.
5) Point 4 notwithstanding, that portion of affiliate output which is exported back to the parent country seems to depend on differences in factor endowments between the home and host country.
6) A high volume of outward direct investment is positively related to a country’s endowment of skilled labor and insignificantly or negatively related to its physical capital endowment.
7) There is weak evidence that direct investment is primarily motivated by tariff avoidance or measurable transport costs.
8) There is mixed evidence that tax avoidance and/or risk diversification are important motives for direct investment. Some evidence does suggest that political risk discourages inward investment.
9) Infrastructure, skill levels, and a minimum threshold level of per capita income seem to be very important determinants of direct investment.
10) There is evidence that agglomeration effects are important in direct investment. But it is admittedly difficult to distinguish agglomeration effects from firms being drawn to the same (unobserved) site-specific resources.
Some of these findings have can be applied to the problem I pose in my thesis, namely the headquarter location choice of multinational companies. Although my research investigates the location choice in global cities, some of Markusen’s findings are still relevant. This applies in particular on his last five findings. Markusen’s 6th finding states that a high volume of outward direct investment is positively related to a country’s endowment of skilled labor and insignificantly or negatively related to its physical capital endowment. This characteristic is also applicable on global cities since MNCs can differentiate global cities based on this characteristic. Finding seven can also be applied on global cities since MNCs can base the location choice of their headquarters on measurable transport costs, since these costs differ between different global cities. In finding eight, Markusen states that tax avoidance and/or risk diversification are important motives for direct investments. These characteristics are to be seen as practically equal for all cities which are located in the same country. It is thus not possible, based on this characteristic, to differentiate between different cities which are located in the same country but it is possible to differentiate between cities which are located in different countries. According to Markusen’s 9th finding infrastructure, skill levels and a minimum threshold level of per capita income seem to be very important determinants of direct investments. These are characteristics which can alter between different global cities, meaning that MNCs are likely to choose global cities which possess of good infrastructure, a highly skilled labor force and where a minimum threshold of per capita income exists. Markusen’s last finding states that it is difficult to analyze whether the agglomeration effects or the site-specific resources are the main reason why MNCs are drawn to a certain location. This implies that it will be difficult to check, when a MNC has chosen a certain global city, whether this choice was based on the agglomeration effects of this global city or on the site-specific resources which are present in this global city.
Why and where do headquarters move?
In my thesis, I want to investigate the headquarter location choice of American multinational companies in global cities. In order to do my research I want to find out which factors induce MNCs to choose a certain global city as the location for their headquarters. Headquarters are according to Strauss-Kahn et al. defined as a management (administration and marketing) center of a firm. This means that neither the output of headquarters nor the price of this output is observable. However, headquarters will be located in the places with the lowest costs for their sort of activities (Bel and Fageda, 2008). This statement is confirmed by Polèse and Sheamur (2004), they state that firms will indeed locate their headquarters where they can maximize their contribution to profits.
Strauss-Kahn has also found that headquarters cluster in a small number of metropolitan areas and that they are more agglomerated than economic activity, this indicates that MNCs are likely to locate their headquarters in a global city. Bel and Fageda (2008) discuss some other general features which influence headquarters location choice. First, total costs of transmitting information are determined by the proximity of other firm units and the quality of transport services. This induces firms to locate their headquarters close to their other establishments and final demand. Second, since location choice implies communication costs, firms will try to avoid other possible costs, such as congestion costs and tax payments. Third, firms try to minimize total labor costs of employees and external suppliers; they will thus prefer to locate their headquarters in places with a highly diversified pool of skilled business services providers and skilled labor. And finally, firms will benefit from the presence of headquarters of other firms, since their presence can reduce the costs of information exchanges about market conditions. Bel and Fageda (2008) mention another feature which influences the headquarter location choice which does not concern costs, namely the fact that some multinational companies prefer to locate their headquarters in the major business centers and/or the political capital of the home country, rather than in regular major cities in other countries.
Strauss-Kahn has studied the determinants of the headquarter location choice of American firms. First, we will indicate which determinants Strauss-Kahn has found to be the key determinants for these location decisions, later on we will discuss these determinants more specifically. Strauss-Kahn found that headquarters relocate to metropolitan areas wits good airport facility – with a dramatic impact, low corporate taxes, low average wages, high levels of business services, same industry specialization, and agglomeration of headquarters in the same sector of activity. She also found that younger and larger (in terms of sales) headquarters tend to relocate more. This also applies for firms that are larger (in terms of number of headquarters), are foreign, or are the outcome of a merger.
Bel and Fageda (2008) mention that information exchange between cities can be critical for the headquarters of large firms that operate on a global scale. This is the case since the role of headquarters in a corporation is to coordinate and command activities within the firm. Due to globalization, many firms have experienced a spatial separation of headquarters from product plants. Within this context, Bel and Fageda (2008) have found that passenger transportation quality is a key input for processing and transmitting information efficiently. This is the case since passenger transportation networks influence the costs and opportunities for face-to-face contacts between cities. Firms thus face a trade-off when separating headquarters from production plants, this trade-off is studied by Henderson and Ono (2005). Firms can, on the one hand, reduce communication costs by locating their headquarters close to the production facilities. On the other hand, when they locate their headquarters in a metropolitan area, they can benefit from better externalization of some services and they can obtain information from other headquarters. In addition to this, Strauss-Kahn and Vives (2005) have used airport availability in order to capture the costs of transmitting services to other firm units. In their paper they conclude that when a city has available an airport that can be considered a small or large hub, this increases the possibility of headquarters to be located in this particular city. Their research has shown that the probability of locating in a metropolitan area increases significantly with the availability of airports. Strauss-Kahn and Vives have found that, relying on odds ratio, the probability of locating in a metropolitan area increases by 40% when a city offers a small hub and increases by 90% when a city offers a large hub. This shows that the availability of airport connections has a dramatic impact on the headquarter location choice of firms and that headquarters rely intensively on airport connections in order to maintain their relation with plants and clients.
Bel and Fageda (2008) also denote that the quality of airport services offered by an urban area is closely related to the geographical scope of the destinations with direct flight, and not only to the amount of total traffic that those airports move. We thus conclude that qualitative airport facility is a key factor in the headquarter location decision of multinational companies.
Corporate taxes are expected to have a negative effect on a location’s attractiveness. Firms will be less induced to locate their headquarters in locations where corporate taxes are high. This is the logical result of firms minimizing the costs associated with certain locations. Strauss-Kahn and Vives (2009) et al. have provided us with the empirical evidence for this statement. Strauss-Kahn and Vives argue that corporate tax rate levels have a significant impact on the choice of location of headquarters. However, since corporate tax rates vary more across regions than across metropolitan areas within regions, the significance of this feature decreases when we compare metropolitan areas which are situated in the same region. Corporate taxes thus influence headquarter location choice significantly when comparing global cities which are situated in different countries or regions but corporate taxes lose their influence on headquarter location choice when comparing global cities which are situated in the same country or region.
Intuitively we would assume that higher average wages in a certain urban area would reduce the possibility of firms locating their headquarters in this location. This makes sense when you take into account the fact that firms try to minimize costs when choosing a headquarter location. Strauss-Kahn and Vives (2009) et al. state that, when using a population-nested model, a 10% increase in the average wage decreases the probability of choosing the metropolitan area by 25%.
High average wages may however also indicate a higher availability of services and qualified labor, and this location feature has a positive effect on the attractiveness of an urban area. Indeed, as stated by Markusen (1998), headquarters of multinational firms face high needs of high-skilled labor forces. This “hidden” feature of higher wages has also been observed by Strauss-Kahn and Vives (2009). They have found wages to have a positive influence on the attractiveness of urban areas in some of their specifications and they have attributed this finding to the presence of qualified labor. Notwithstanding this finding, Strauss-Kahn and Vives still assume high average wages to have a negative influence on the attractiveness of an urban area.
Levels of financial business services
According to Strauss-Kahn and Vives, both levels of relative availability of financial and business services have significant positive effects on headquarter location choice. Their research shows that a 10% increase in the measure of business services increases the probability of choosing a location by approximately 5%, this measure however does not have a significant influence when locating manufacturing headquarters. The measure of business services on the other hand remains it significant influence when regarding manufacturing headquarters. Strauss-Kahn and Vives have found that a 10% increase in the measure of business services specialization increases the probability of choosing a location by 7-13.5%.
Presence of other headquarters
Strauss-Kahn and Vives indicate that the number of headquarters and the count o