Global cities and the geographical transfer of value

Introduction

The aim of this article is to deepen an analysis, which in my view is central to the global city concept, ¹ but which has not been elaborated satisfactorily since the publication of Sassen’s The Global City in 1991. I argue that, although the title of Sassen’s book suggests otherwise, the global city model is primarily not about urban theory. Rather, authors who have developed the concept such as Cohen (1981), Friedmann (1986), Friedmann and Wolff (1982), Hymer (1972) and Sassen (1988, 1991) were concerned with the role of specific cities in the organisation of economic globalisation, and not (primarily) with cities themselves. As Sassen (2016: 100) recaps, ‘When I first began my research on the global economy I was not thinking cities at all’. It is, thus, fair to say that the global city model has been developed as an economic geography concept to comprehend the role of specialised economic actors in certain cities: ‘The key indicator of global city status is whether a city contains the capabilities for servicing, managing, and financing the global operations of firms and markets’ (Sassen, 2001: 359). More in particular, global cities are conceptualised on the basis of their capabilities for the ‘command and control’ of the world economy. ² The concept thus aimed to understand the role of producer service firms (henceforth PSFs) clustered in global cities in the organisation of the uneven relations underlying economic globalisation. The global city concept has therefore been, as Sassen (2001: 351) reiterates in the second edition of The Global City, developed as a theory ‘attuned to questions of power and inequality’.

The call to understand the novel geographies of economic power and the functions which global cities assume in it was, however, more programmatic claim than lived research practice. Already in 1986, in a comment on the then emerging global city literature, Roberts (1986: 459; emphasis in original) criticised that ‘there is not enough emphasis […] on how cities and the classes within them achieve control over other regions’. Thirty years and thousands of publications on the subject of global cities later, one must admit that this criticism is still valid. Though studies substantiating that global cities are organisational and power nodes in globalisation exist (Hanssens et al., 2012; Jacobs, 2014; Jacobs et al., 2011; Parnreiter, 2010, 2015; Rossi et al., 2007), we have not advanced satisfactorily in understanding the role of global cities in the exercise of economic power, because (too) much of the literature has either dealt with measuring the connectivity of world cities or focused on urban phenomena related to globalisation (Acuto, 2011; Derudder and Parnreiter, 2014; Parnreiter, 2013).

Against this backdrop, this article seeks to make two contributions. The first goal is to deepen the examination of the role of PSFs in global cities in the organisation of economic globalisation and, in particular, in the making and reproduction of uneven development. To specify this analysis, I make three moves. Firstly, I use the related concepts of global commodity chains, global value chains and global production networks ³ (henceforth GCCs, GVCs and GPNs) to denote what is meant by ‘world economy’ or ‘globalisation’. This is appropriate not only because such networks form the world economy’s backbone, but also because the respective literatures’ interest in uneven development (Bair, 2014; Coe et al., 2008; Dicken, 2011; Gereffi and Korzeniewicz, 1994; Henderson et al., 2002; Werner, 2016; Yeung and Coe, 2015) overlaps with key concerns of global city researchers. Secondly, I borrow the governance concept of the GCC literature (Gereffi, 1994) because its actor-focused perspective on asymmetrical power relationships is apt to substantiate the notion of ‘command and control’, which is only vaguely defined in the global city concept. Thirdly, I refer to Hadjimichalis’ (1984, 1987) concept of the ‘geographical transfer of value’ (henceforth GTV) to analyse uneven development in GPNs. Conceptualising global cities as governance nodes in GCCs (Brown et al., 2010; Parnreiter et al., 2007), I seek to scrutinise the role of PSFs in global cities in such centripetal wealth transfers.

The second contribution I seek to make is to show that GPN governance through PSFs is not a matter of a very limited number of ‘Global Elite’ cities (AT Kearney, 2015). In the context of the persistent criticism that the global city concept has a Northern/Western bias on the basis of which ‘millions of people and hundreds of cities are dropped off the map’ of urban studies (Robinson, 2002: 535), I will argue that PSFs in ‘minor’ Northern/Western cities and in cities in the Global South are also busy in developing strategies that contribute to reconfiguring relations in GPNs in a way that wealth flows are diverted towards their clients (and themselves) in the world economy’s centres. This contention is illustrated by the example of the role of PSFs in two not so prominent global cities – Hamburg and Mexico City – in two crucial fields of GCC governance (entry barriers and labour relations). In exemplifying the conceptual arguments with cases from cities beyond ‘the usual suspects’, this article contradicts postcolonial urban studies’ critique (e.g. Robinson, 2002, 2006; Roy, 2009) of a Northern/Western bias of the global city concept: PSFs in ‘minor’ global cities are strategic agents, too. They are strongly involved in shaping the relations along GCCs in a way that wealth can be transferred towards the centres of the world economy.

Global cities and the governance of GCCs

As stated above, there is a paucity of studies detailing the economic control functions of global cities. A related, but even more serious critique is that the global city concept lacks an adequate theorisation of economic power, its subject matter. Jones (2002: 340) has criticised that the model is ‘unnecessarily focused on urban places and spaces at the expense of transnational firms and most especially key social actors’. Imbuing places with agency, the concept fails ‘to get to grips with theorising power’. In a similar vein, Allen (2003: 154) contended that ‘Sassen conflates resources and power’ because she seeks to read power off from the number of PSFs in a city. Yet, because power is ‘a relational effect of social interaction’ (Allen, 2003: 2), it cannot be stored nor radiate out from an authoritative centre like a global city.

This criticism of an insufficient theorisation of economic power has to be taken very seriously, because inter-firm and inter-city control relationships are the raison d’être of the global city concept. To overcome this flaw, I suggest turning to the concept of governance which is central to the GCCs, GVCs and GPNs literatures. In particular, I consider Gereffi’s (1994: 97) original conceptualisation of governance as the ‘authority and power relationships that determine how financial, material and human resources are allocated and flow within a chain’ as particularly useful because it draws attention to power relations in uneven development. ⁴ Moreover, Gereffi’s concept implies a multi-actor and multi-scalar analysis, what is important because it invites scrutiny of the role of actors which are put centre stage by the global city concepts: PSFs. Their involvement in economic governance has been neglected for a long time, although in the early phase of GCC research it has been acknowledged that PSFs are ‘economic agents exerting a dominant influence over the governance of GCCs’ (Rabach and Kim, 1994: 125). Recently, however, the significant influence that PSFs in general, and financial firms in particular, exercise on corporate strategies has been put centre stage in GPN research (Coe et al., 2014). The consequences of lead firms’ financialisation have been discussed as regards the shape and extension of GCCs (Gibbon, 2002; Milberg, 2008) and the distribution of value amongst different actors (Palpacuer, 2008). Wójcik (2013) and Bassens and Van Meeteren (2015) point to the critical role of PSFs in enabling their clients to make superprofits and to then keep them through tax evasion. ⁵ As to non-financial PSFs, Coe (2014) drew attention to logistic firms, while the IGLP Law and Global Production Working Group (2016) is launching an initiative to focus more on the role of law firms in GVC governance. These studies echo the findings of the broader literatures on financialisation and transnational private governance which confirm that PSFs supply key inputs for economic governance (Baud and Durand, 2012; Djelic and Sahlin-Anderson, 2006; Flood and Sosa, 2008; Froud et al., 2006; Graz and Nölke, 2008; Milberg and Winkler, 2010).

All these studies’ dealing with PSFs’ role in economic governance rest on the assumption that producer services ⁶ (PSs) are more than just ‘technical’ support. Rather, they are guidance. In the perception of their clients, PSFs enjoy authority in Arendt’s (1961) or Sennett’s (1980) sense, because they master knowledge which is seen to satisfy the desire and need for leadership in very complex economic settings. Since PSFs are (believed to be) the providers of the know-how necessary for economic success (cf. Bassens and Van Meeteren, 2015), their counsels are usually followed. Because an advice which is known to be obeyed becomes to be more than an advice, PSFs do, though they have no formal autonomy in decision making, pre-structure corporate decisions in the sense that they shape a decision in advance. Sassen (2010: 158) calls this influence by means of the services provided ‘embedded governance – embedded in the lawyering, the accounting and the investment choices of the firm’. PSFs have thus become strategic partners of their clients, with their relationships being best understood as horizontal (Bryson and Daniels, 2007). Accordingly, PSFs’ power is not as exercised over clients, but together with them. It is a resource to empower clients vis-a-vis business partners, unions or governments. Therefore, the even and mutual relationships between PSFs and their clients are embedded into and serve the hierarchical relations which PSFs’ clients maintain in ‘their’ GPNs.

PSFs, global cities and the geographical transfer of value

To operationalise an analysis of uneven development in GPNs, it is worthwhile to return to Hadjimichalis’ (1984, 1987) concept of the ‘geographical transfer of value’. In general terms, this GTV refers to a process through which value produced at one location is transferred to and realised in another, with the result that the latter location’s accumulation base grows at the expense of the first. GTV thus leads to and intensifies differences in localised accumulation processes, which in turn increases and accelerates inter-regional inequality. This connection between the geographical reallocation of wealth, regionally differentiated growth rates and uneven development becomes most obvious in the case of the ‘urbanization of capital’ (Harvey, 1985).

According to Hadjimichalis (1984: 340), value is transferred geographically either indirectly through the market, or directly through violent methods and ‘more sophisticated policies such as state taxation policies, price policy, public transfer payments, multinational profit repatriation, and transfer-pricing’. This listing suggests that PSFs are strategic agents facilitating the GTV.

As to the market-mediated shifting of value, GCC, GVC and GPN analysis consistently points to entry barriers as a key instrument for achieving rents and hence an uneven distribution of surplus: ‘The GCCs approach explains the distribution of wealth within a chain as an outcome of the relative intensity of competition within different nodes’ (Gereffi et al., 1994: 4). For the purpose of this article, two things are important. Firstly, erecting entry barriers is a function of governance: ‘It is this role […] of identifying dynamic rent opportunities and apportioning roles to key players which reflects the act of governance’ (Kaplinsky, 2000: 124). Secondly, for creating and managing entry barriers, lead firms have to rely on PSFs. These are particularly important for erecting ‘strategic’ barriers, which the OECD (2005) defines as being intentionally created to deter entry of competitors. Prime examples are the creation and subsequent protection of intellectual property or the establishing of brands, firm reputation and customer loyalty, which would be impossible without the involvement of patent lawyers or advertisers. Yet, PSFs are also important as regards some structural entry barriers (which are related to industry conditions) such as financing, which obviously requires the engagement of banks and other financial institutions.

PSFs are also key agents in many direct mechanisms of value transfer. It is hard to imagine that policies related to taxation, pricing, foreign direct investment or the repatriation of profits could be put into practice without the involvement of bankers, lawyers, accountants and other consultants (e.g. Bassens and Van Meeteren, 2015; Clark and Monk, 2013; Flood and Sosa, 2008; Sikkaa and Willmottb, 2010; Wainwright, 2011; Wójcik, 2013). Another field in which PSFs are involved in value transfer are labour relations. These are critical to GCC governance not least because under competitive conditions, companies must reduce costs without impairing performance (Yeung and Coe, 2015). Labour costs and regulations are therefore essential criteria as to whether and how companies in a particular region become integrated into GPNs (Henderson et al., 2002; Selwyn, 2012; Smith et al., 2002). Finally, PSFs are key agents in the GTV for yet another reason. As ‘gatekeeper(s) of an enormous stock of sought-after knowledge’ (Bassens and Van Meeteren, 2015: 763), they draw massive fees from their clients, appropriating thereby an increasing part of the wealth produced along GVCs.

Considering PSFs as strategic actors in the GTV carries consequences for the geographies of the wealth transfer, in particular as regards its organisation. Contrary to Hadjimichalis (who developed his arguments more than 30 years ago, when the new geographical contours of the world economy were just emerging), I contend that the social agents active in transferring value are not located ‘usually …outside the donor region’ (Hadjimichalis, 1984: 340; emphasis mine). Current urban theorising stresses that knowledge intensive activities require constant face-to-face interactions and, thus, co-presence (e.g. Storper, 2013). This certainly applies to high-end bankers, lawyers or accountants, too: As distance-sensitive knowledge workers, professionals of PSFs had to follow their clients in globalisation processes (Bryson and Daniels, 2007). Accordingly, the global cities where these services cluster are necessarily spread over the world. Remember that Sassen (1991) contended that New York, London and Tokyo are leading examples of a much broader process of global city formation which cuts across the traditional North–South divide. Sassen’s (2001: 348) wording that global cities form ‘the operational scaffolding of … the global economy’ is quite unambiguous – scaffoldings typically cover the whole area ‘under construction’, and not just key intersections. Accordingly, Taylor and Derudder (2016) identify 525 cities creating the world city network – almost half of them in poorer countries (including classical mega-cities such as Mumbai, São Paulo, Mexico City or Jakarta). While the resulting geography of economic governance looks far more decentralised than the one deduced from a focus on lead firms only, the theoretical insight drawn from these spatialities of PSFs’ influence on GCC governance is that wealth transfer is organised from outside the donor regions as well as from within. Centre-periphery relations unfold thus as successive inter-city connections on various scales – an idea that Frank (1969) suggested with his spatialised account of the ‘Development of Underdevelopment’, in which Latin American capital cities were conceived of as bridgeheads for the interests of the dominant centres.

Conceptualising global cities in this way as governance nodes in GCCs (Brown et al., 2010; Parnreiter et al., 2007) draws attention to cities beyond the world city network. This is built upon inter-city ramifications which link global cities to the countless non-global, but yet globalised cities where production for the world market is carried out and from where wealth is being drained off (Parnreiter, 2014). In fact, the very idea of a world/global city as introduced by Friedmann and Sassen only makes sense if these cities are seen in a relational perspective, tied to each other, but also to all the globalised ‘ordinary cities’ (Robinson, 2006) whose work has to be articulated with the world market. There is, thus, no contradiction between Sassen’s (2016: 103) ‘analytic tactic … to focus on the extreme ends’ such as the financial sector in New York, and Robinson’s (2002: 547) assertion that export processing zones in poor countries constitute ‘quite the other end of the command and control continuum of global city functions’. It is precisely such a relational approach to centre-periphery interactions in GVCs and in urban networks that justifies the selection of PSs in global cities as key analytical lenses. The often disapproved ‘narrowing down’ of the focus on PSs (e.g. Robinson, 2002) is thus a means of a ‘critical exploration of exceptional power’ (Peck, 2015: 164), conceptually intended to draw attention to the ugly truth that most people in most cities (Southern and Northern alike) are off the map of economic decision-making. Because the world economy is so ‘spiky’, it is neither analytically accurate nor politically progressive to speak homogeneously of ‘globalising cities’ (Marcuse and van Kempen, 2000).

Seen in this light, the critique that global city researchers were treating non-global cities as ‘out of the game’, as ‘excluded from global capitalism’ and therefore as irrelevant to their theoretical reflections’ (Robinson, 2002: 538) is left without substance. Though the lack of empirical work on global cities’ role in the inter-city GTV must be recognised self-critically, it is not true that for global city scholars only the core matters. That such allegations arise from straw man rhetoric (Van Meeteren et al., 2016) can be illustrated with the travelling of the following quote. Referring to Robinson (2002), Roy (2009: 824; emphasis mine) states: ‘global/world cities mapping drops all other cities from the map, arguing that they are structurally irrelevant to the functioning of economic globalization’. Yet, in Robinson’s (2002) article there is no ‘arguing’ of a global/world cities’ representative in that respect, but a specification of what Robinson conceives to be a conclusion from global city research: ‘Many populations are excluded from the space of global capitalism, and thus from the field of world cities: they are “economically irrelevant” (Knox, 1995: 41)’ (Robinson, 2002: 534). Besides the fact that the quote is from Friedmann and not from Knox, the more substantial critique is that Robinson misinterprets Friedmann. Speaking of a polarised world, Friedmann (1995: 41) suggested that ‘São Paulo and the affluent metropolitan classes of Brazil do not require the country’s disempowered poor as either producers or consumers. In that sense … more than 50 per cent of Brazil’s population is economically irrelevant’. While it should be subject to empirical analyses whether this is correct, ⁷ it is evident that Friedmann never suggested deleting the poor from the map of global cities. He did the opposite: ‘We must understand global cities in relation to their respective peripheries, to both their external and internal proletariat’ (Friedmann, 1995: 42; emphasis mine). Since the notions of periphery and proletariat are conceptually based upon uneven and exploitative relations, what Friedmann was really calling for was an analysis of the intra- and inter-city structure of wealth transfers.

Hamburg and Mexico City as governance nodes: Exemplifying the role of PSFs in erecting entry barriers and in shaping labour relations

So far I have developed two ideas: Firstly, PSFs are important governance actors shaping ‘how financial, material and human resources are allocated and flow within a [commodity] chain’ (Gereffi, 1994: 97), which is why global cities are governance nodes. Secondly, the specialised knowledge PSFs supply helps their clients to improve their position in GPNs – often at the expense of other actors. PSFs are thus involved in the organisation of centripetal wealth flows, what makes global cities to specific nodes of governance, namely of the GTV. A third notion touched upon is that this governing of GCCs and their inherent value transfers is not accomplished from a small number of Northern or Western cities. Rather, I have claimed that it is conceptually cogent (Sassen, 1991) and empirically established (Taylor and Derudder, 2016) that the world city network is truly global.

In order to exemplify the role of PSFs in the GTV, in this section I will give some examples of how professionals describe their role in shaping their clients’ governance processes. Since my intention is to show that this involvement in governance processes goes beyond ‘Global Elite’ cities (AT Kearney, 2015), the cases are taken from two ‘minor’ global cities – Hamburg and Mexico City. While Hamburg, though being a crucial hub in many GPNs due to its port (Hesse, 2006), is integrated into the world city network only relatively weakly (Taylor and Derudder, 2016), Mexico City is, as a classical ‘Third World megacity’, a case of a city which the postcolonial urban studies’ critique sees dropped off the global city map (Robinson, 2002, 2006). ⁸

While many insights from my interviews tie in with the results of studies on the role and practices of PSFs in GPNs and in the world city network (e.g. Bassens and Van Meeteren, 2015; Beaverstock et al., 2015; Coe et al., 2014; Faulconbridge, 2007; Wójcik, 2013), here I pay particular attention to two aspects of governance, which are frequently denoted as being strategic: entry barriers (represented by cases of PSFs in Hamburg) and labour relations (cases from Mexico City) (Coe and Hess, 2013; Gereffi, 1994; Kaplinsky and Morris, 2002; Selwyn, 2012; Smith et al., 2002). The examples given draw from semi-structured, open-ended interviews with representatives of global PSFs drawn from the GaWC ⁹ list (Taylor and Derudder, 2016) in the two cities (19 in Hamburg in 2013, 18 in Mexico City between 2007 and 2013). Respondents were mainly partners of the interviewed firm. Interviews lasted from 30 to 65 minutes, were recorded and completely transcribed without using transcription software. The analysis was carried out on the basis of the structuring content analysis (a variant of qualitative content analysis). For the development of the coding structure, I used a mixed deductive-inductive approach (Gläser and Laudel, 2010; Mayring, 2010). No software was used in the analysis of the transcript. Quotes have been translated by the author.

According to the insight that patents represent one of the clearest cases of an entry barrier (Kaplinsky and Morris, 2002), two interviewed business lawyers in Hamburg identify intellectual property as a key area both of GCC governance and of their influence on it. They also make clear that their objective in erecting such barriers is to alter the distribution of wealth created in a GPN in favour of their clients. The office managing partner of a law firm in Hamburg says that his service (patenting):

of course, quite clearly [has the purpose] to prevent others from entering the market … Because intellectual property, very, very clearly leads to monopoly rights. And monopoly rights of course always serve to crowd out a competitor. Thus, to have a monopoly … would always be suitable.

Yet, altering the distribution of value in a GPN requires more than just legal advice. In full agreement with Sassen’s (2010: 158) notion of governance as being ‘embedded’ into the services provided, the partner of another global law firm in Hamburg asserts that:

we go beyond the [legal] issue, considering how we can best serve the client, to help to achieve his economic goal as efficiently and profitably as possible … That’s precisely our job, to go beyond [the legal] and to develop things. Yet this development is part of the legal advice.

The example he gives refers to the recommendation to geographically restructure a client’s GVC because in a particular country the entrance barrier ‘intellectual property’ could not be guaranteed because of a ‘loose’ handling of laws:

So if I, if in some place I don’t have patent protection or not sufficient [protection], then the competition comes and takes [my product] away or modifies it or makes it cheaper … And that is certainly very significant … So, depending on whether you have them [patents], and where [a patent is registered], that has immediate impact on the added value, on value distribution.

Financing is another powerful entry barrier, in particular for suppliers in poorer countries (Gibbon, 2008). The head of the business centre of the Hamburg office of a global bank gives an example of how he can take advantage of the dearth of financial resources in an Asian country where his client’s supplier is located:

Well, in essence, it’s all about the question: When will I get my money? … That’s the most elementary payment condition, [which] also determines the strength vis-a-vis my suppliers … For example, you enable your supplier to get his money earlier … You can either put him in the position that he can no longer [hold out] because the payment terms are changed so that he is virtually no longer able to cover the pre-financing requirements, or you can make life easier for him, so to speak.

Offering the Asian supplier ‘mechanisms of liquidity’ brings him competitive advantages over other Asian suppliers. Yet, what the Hamburg financial service provider has ultimately in mind is, of course, not the supplier’s economic fate, but his client’s profit. The price the supplier has to pay for being empowered against local competitors is to be weakened against the German lead firm. As the banker says:

if you are paying him [the supplier] so early, that his position is clearly, significantly improved, then that in turn has implications on, for example, prices, yes, that is the other side, so to speak, the sooner I tell him: hey, I will pay you earlier, but you then have to make concessions, too, regarding the prices.

The example reveals, thus, two dimensions of how PSFs influence GCC governance. In organising the entry barrier ‘liquidity’, the Hamburg office of the PSF controls which companies can integrate themselves into the Asian segment of the client’s GVC. This powerful position is in turn used to siphon off wealth from the ‘empowered’ Asian firm and shift it to the Hamburg client (and to itself).

The second example of how PSFs are involved in GCC governance and in the organisation of the GTV refers to labour relations. According to the interviews in Mexico City, lawyers see themselves as strategic actors in their shaping. One issue recurrently addressed relates to the payment of social benefits. For example, the level of insurance benefits for employees depends on the risk class in which a company is classified by the Mexican Institute for Social Security. In order to reduce payments for the insurance protection of the employees of one client, an interviewed global law firm has successfully claimed in the courts the client’s re-classification in a less risky business field. Another issue mentioned repeatedly by labour lawyers concerns the configuration of employment contracts. If workers, for example, belong directly to the hiring company, they normally receive higher wages and more social benefits than if they were employed by a subsidiary founded for this purpose.

Another frequent topic is the ‘restructuring’ of labour relations, e.g. through the switching to a different union with which the collective bargaining agreement is signed. This is a critical issue, due to the closed-shop principle according to which companies in Mexico have to accept only one union as the workers’ legitimate representation. Because unions differ strongly in strength and political orientation, the closed-shop principle boosts high selectivity as regards with whom to sign the collective bargaining agreement. Lawyers therefore recommend their clients to ‘actively hunt for a trade union’, as the German Association for Foreign Trade and Location Marketing informs (Buerstedde and Päffgen, 2015). This ‘chasing’ of convenient unions is particularly important because of the geographical dimension of unions’ strength. Since unions are more employer-friendly in some federal states than in others, the proactive selection of workers’ representation involves localisation decisions: Because of unions’ impacts on the working environment, wages, the payment of social benefits and hence on profits, companies decide where to establish a plant also with regard to the geography of union power.

However, these spatialities of labour relations are difficult to grasp for foreign companies and especially for newcomers to the Mexican market. Lawyers from global law firms describe it therefore as one of their tasks to guide the client in this respect. As one lawyer underscores, the partners of his firm:

have conferences with the clients to plan a deal, to structure a deal … The one who takes the decision how things should be done, rather would be here, from the [law] firm … I do believe that the one who makes the strategy, it’s the partners of the law firm.

A partner of another law firm corroborates that the client:

contracts you precisely for that. That you tell him which union is the most convenient one … I think there’s much influence one has. A good example, as to labour issues, is … the firm x. X has closed plants, and has put other plants, only and exclusively because of the advice we have given … [They faced] problems with the union, because the union is very aggressive … Because [in that federal state] a contract was common, they had to pay certain, hm, benefits that were above market standards …

And the decision to relocate the plants, it was a joint decision, the client and you?

Well, … the client decides because it is his money, (but) based on the advice you give him … If you tell him the best place to establish a plant, then it is unlikely that the client will decide for anything else.

While these examples of the strategic involvement of PFSs in GCC governance certainly require a critical assessment through triangulation, it is also true that the interviews fit both the literature on the importance of barriers to entry and labour relations in shaping the structure, geography and governance of GCCs (Gereffi, 1994; Kaplinsky and Morris, 2002; Mahutga, 2012; Selwyn, 2012; Smith et al., 2002) and the literature on the role of financial and legal firms in GPN governance processes (Coe et al., 2014; IGLP Law and Global Production Working Group, 2016). To make it clear: PSFs are not only enablers of GCCs through ‘greasing’ assembly lines, they are also drivers, shaping how and where value is produced and accumulated. That is valid also for ‘minor’ global cities such as Hamburg and Mexico City, which is in line with the quantitative GaWC research that proves that the world city network is formed of more than 500 cities around the world (Taylor and Derudder, 2016).

Without claiming that the cases presented draw a complete picture, I think it is fair to say that they do support Sassen’s (2010: 158) contention that governance is being ‘embedded’ into the services provided. The cases show that services such as patenting, financing or making labour contracts ‘entail command functions that are distributed across those operations’ (Sassen, 2010: 158). The Mexico City lawyer, for example, who in his search for ‘convenient’ unions pre-structures clients’ strategies for plant relocation, is an active agent in GCC governance in two ways. Suggesting to clients a geographical reorganisation of their GPNs implies that the lawyer functions as a ‘disarticulator’ (cf. Bair and Werner, 2011) of plants, cities or federal states with combative trade unions. Because such exclusions are as ‘constitutive’ (Werner, 2016: 458) for the formation and restructuring of GVCs as their opposite, namely the integration through ‘strategic coupling’ (Yeung, 2015), lawyers can be identified as important network shapers. Secondly, plant relocation with the purpose of reducing wages increases a local GTV from workers’ households to the firm: Weaker unions imply less direct and indirect payments and therefore more value to be siphoned off. Yet, workers compensate the loss of income amongst other things through externalisation of costs to their households, which feed values into GPNs because (mainly female) household members shoulder more unpaid work for producing and reproducing the labour force (Dunaway, 2014).

As to the scales of this involvement of PFSs in GCC governance, the interviews with PSFs in Hamburg reveal that their influence clearly extends beyond national borders. When a lawyer advocates the withdrawal of a client from India because protection of intellectual property for pharmaceuticals there is ‘weak’, he/she defends ‘monopoly rights’ (and the corresponding profits) against a foreign government that seeks to provide affordable medicines. Suggesting to his client a geographical restructuring of the medical GCC with the purpose of bypassing the Indian market, the Hamburg lawyer once again can be identified as a ‘disarticulator’, and that on a global scale. In a similar vein, the Hamburg banker who provides financing to a client’s supplier in Asia is influencing GCC governance worldwide. His service can be decisive for which Asian firm will be integrated into the Hamburg client’s GCC, and he contributes to a global transfer of value because his financial service to the Asian supplier allows the Hamburg firm to beat down prices in backward linkages. On the other hand, PSFs in Hamburg also function as a pivot to other global cities, facilitating Asian or US companies’ entry into the European and/or German market (Parnreiter, 2015). While this role of a gateway city (Rossi and Taylor, 2006) for foreign firms to the national market is also assumed by Mexico City, and while Mexico City’s PSFs do service (the not so many) Mexican firms going global, the geographical reach of their shaping capacities seems to be more limited: None of the respondents brought up a case of modelling labour relations in, say, Central America (Parnreiter, 2010).

In any case, the consequences of the wealth transfers co-organised by PSFs in Hamburg and Mexico City are serious. In Latin America, for example, about 45% of total profits or 5% of total FDI were repatriated between 2000 and 2010 (Neme Castillo and Perezyera, 2013). Because a city’s or a region’s development depends critically on the capacity to ‘hold down’ the value created there (Coe and Hess, 2013), such policies of disinvestment and the GTV are part of the ongoing shaping of the landscapes of uneven development. Now, it is my contention that such relocations of resources are inconceivable without the daily collaboration of PSFs. And it is equally inconceivable that this GTV could be accomplished from a few places in the ‘North’, because high-end servicing is knowledge intensive and therefore distance sensitive. Mexico City is, thus, a place for the making of Mexico’s misery.

Conclusions

The starting point of this article was the disappointing acknowledgement that 30 years after the publication of Sassen’s The Global City (1991) our understanding of ‘how cities and the classes within them achieve control over other regions’ (Roberts, 1986: 459) has not improved substantially. In order to find answers to this question and to substantiate the notion of global cities’ ‘command and control’ functions, I have drawn on conceptualisations of global cities as nodes in GCCs (Brown et al., 2010). More precisely, and relying on the governance concept as originally defined by Gereffi (1994), I have scrutinised the role of PSFs in uneven development and, in particular, in the GTV (Hadjimichalis, 1984, 1987). My key contention is that because the direct as well as the indirect mechanisms through which value is transferred geographically require the intervention of PSFs, global cities can be theorised as governance nodes for centripetal wealth transfers along GCCs.

To exemplify this notion, I turned to PSFs’ activities in two less ‘eye-catching’ global cities. In the context of the persisting criticism that the global city concept has a bias towards ‘elite cities’, it has been my intention to substantiate the claim that global cities are critical places for the organisation of uneven development with reference to cities beyond ‘the usual suspects’. Referring to cases of how PSFs in Hamburg and Mexico City erect entry barriers to protect their clients from competition and of how they shape labour relations at the expense of employees, I have maintained that governance is, as Sassen (2010: 158) has argued, indeed ‘embedded’ into the services provided. From that follows that even ‘minor’ global cities are strategic governance places from where the transfer of wealth towards the centres of the world economy is organised.

Yet, this form of governance does not directly correspond to the notion of ‘command and control’, nor does it fit into the dominant approach in GCC, GVC and GPN studies, which until very recently has focused on bipolar governance relations between lead firms and key suppliers. I think that economic governance is best understood as a multi-actor, multi-scale and multi-local process. Paraphrasing Jane Jacobs (1970: 121), one could probably say that PSFs’ agency resembles the ‘little movements’ that turn ‘the great wheels’ of the GTV. Hamburg and Mexico City are two of the many global cities where these wheels are turned, and that is why the global city concept is a meaningful approach for studying the role of cities all over the world in uneven development. Thus, the challenge for urban studies is not so much to theorise the city ‘from the Global South’ (Parnell and Robinson, 2012), but to comprehend the role of global cities all over the world in the shaping of the uneven relations between cities that guarantee the centripetal GTV.