Why do some firms relocate away from a country in the face of socio-political uncertainty, whilst others do not?

(1)

Sam Oxley

S3909794/B8049801

Dual Masters Award:

MSc International Business & Management

(University of Groningen)

MSc Advanced International Business Management

(Newcastle University)

Supervisors: Dr. Esha Mendiratta &

Professor Iain Munro

2

nd

_{December 2019}

Wordcount: 15,000

Why do some firms relocate away from a

country in the face of socio-political

(2)

i

Abstract:

(3)

ii

Acknowledgements:

(4)

iii

List of Figures

Figure 1: Spatial profitability boundary for satisficing behaviour Figure 2: Strategic Choice Under Conditions of Bounded Rationality Figure 3: Conceptual model of proposed relationships

List of Tables

Table 1: Location theories and factors affecting relocation decisions Table 2: Overview of key empirical studies of relocation

Table 3: Summary of variables

Table 4: Summary of data required to test each hypothesis Table 5: Descriptive statistics

Table 6: Collinearity statistics Table 7: Case processing summary Table 8: Classification table(Null model) Table 9: Omnibus tests of model coefficients Table 10: Classification table(Full model)

Table 11: Exponential coefficients of all independent variables

List of Appendices

(6)

- 1 -

1 Introduction

Investigations into factors affecting a firm’s decision to migrate have been a staple in International Business literature since the 1970s (Brouwer et al, 2004). The early studies in the 1960s were predominantly based in the United Kingdom (UK), with Keeble (1968) and Townroe (1972) being amongst the first to consider that the internal mechanisms of the firm may impact any relocation decisions. The majority of relocation studies have focused primarily on the choice of a new location as part of a growth strategy, firms often seek to explore new markets through increased internationalisation; this can be attributed in part to the increasing prevalence of globalisation, and multinational companies (MNCs) (Fosgren et al. 1995; Strauss-Kahn & Vives, 2009). However, the current political context paints a much bleaker picture for world trade, with rising protectionist measures, and nationalist feelings observable widely (Fukuyama, 2018). An example of this was the 2016 referendum, held to decide whether the UK should remain in or leave the European Union (EU), resulting in a majority of 52% voting to leave (Goodwin & Heath, 2016). On the 29th

March 2017, the British government triggered Article 50, the ‘Treaty on the European Union’, signalling the start of the UK’s exit from the EU, hitherto referred to as Brexit. At the time of writing, the process is still on going with the terms of withdrawal still unknown. Governments have long been concerned with firm relocation as firms migrating into an area bring jobs and increase the economic performance of the area, whilst firms migrating away from an area have the opposite effect (Graham & Krugman, 1995). The early work of Keeble (1968) and Townroe (1972) was revered for their contributions to spatial sciences and were used as a basis for government policies for redistributing industries around the country. Similarly, relocation literature has been central to the development of effective government policies to attract inward foreign direct investment through firm migration (Lowendahl, 2001). Today, the focus of the UK government is much more likely to be concerned with retaining firms in the face of Brexit (Tetlow & Stojanovic, 2018), a growing and highly topical issue which provides the motivation for this work.

(7)

- 2 -

Giles (2018) states that Brexit presents one of the first concrete examples of ‘deglobalisation’, as the UK attempts to break away from a well-established trading bloc. Beyond losing free-trade with EU member countries which make up the UK’s primary trading partners, such as Germany, France, and the Netherlands, the UK will also have to form new trade deals with any country it wishes to trade with, as the existing deals are made through the EU. The tariffs and quotas associated with this could be significantly less favourable for UK firms than they have currently, through higher tariff rates or more stringent quota levels. Naturally, this presents a vast amount of uncertainty for many UK firms (Thomas, 2019), especially given the complex transnational nature of many supply chains. In response to this, many firms have decided to relocate some aspects of their operations out of the UK (Laker & Roulet, 2019).

(8)

- 3 -

improve the understanding of the motivation of the decision to relocate away from the UK market.

In section 2, this paper discusses what is meant by ‘relocation’ and offers a brief overview of the three primary approaches taken in relocation literature; (1) neo-classical, (2), institutional, and (3) behavioural. In doing so, a number of the central motivations and applications for each approach will be considered, and the applicability to the current context of the UK evaluated. Following on from the theoretical overview, an introduction to the market uncertainty present in the current climate of the United Kingdom (UK) will be provided. This research aims to enrich relocation literature by providing a novel framework (section 3), highlighting how firm level factors can contribute to a firm’s decision to relocate in the face of market uncertainty. Following this, the findings of the investigation are presented and the implications for both theory and practical application are discussed.

This paper aims to answer the following research question:

“To what extent do firm level factors influence the outward relocation decisions of firms in the face of growing market uncertainty brought about by Brexit?”

In order to successfully answer the research question, the paper must first fulfil a number of objectives:

1. Consider how the firm’s dependency on the UK market affects relocation decisions 2. Examine how organisational inertia may affect firm mobility

3. Establish if the upper-echelons of a firm will influence its likelihood to relocate

(9)

- 4 -

concerning the appropriateness of the concept of organisational inertia, which dictates why certain firms may be more reluctant to change or relocation (Brouwer, 2004). The data provided found no meaningful link between the length of time the firm has been operating in the UK and its likelihood to relocate away in the Brexit period; perhaps more novel again, is the inclusion of the number of subsidiaries as an indicator of size, finding that the more subsidiaries the firm has the more likely it is to engage in some form of relocation. A key theoretical implication provided by this article is that it is amongst the first to extend the focus of the behavioural approach by explicitly linking Hambrick & Mason’s (1984) Upper-Echelon (UE) theory. The theory posits firm’s behaviour to be reflective of their top-level decision makers. Both theories rely uniformly on the existence of ‘bounded rationality’ and the ability to interpret information. There has been substantial research into the influence board characteristics can have on internationalisation and location choices (Le & Kroll, 2017)

(10)

- 5 -

2 Literature Review

2.1 A background to relocation literature:

(11)

- 6 -

distinction drawn between the corporate HQ (CHQ), regional HQ (RHQ), and core business functions from within the HQ’s must be established. The CHQ as a whole usually comprises of the executive management team, core staff functions and the corporate legal seat of the firm, the RHQ is responsible for the strategic directions of the business within a certain area, such as Europe (Desai, 2009). Partial relocations can at times be attributed to reasons fitting the rationale of all 3 primary approaches to relocation, though as much as anything the behavioural and institutional approaches may prove most useful in establishing why firms choose to relocate only elements of their operations rather than to relocate their HQ.

Throughout this investigation, although only cross boarder relocations away from the UK are being recorded- a number of forms of relocation will be included, ranging from a smaller degree of relocation beginning with a minimum level of the moving of a significant number of jobs (more than 50), a movement of a core-function (such as sales or marketing) or creation of a ‘regional business hub’, the relocation of a RHQ, and finally the greatest degree the movement of the CHQ. It was deemed appropriate to include the whole range of relocations due to the ongoing nature of Brexit which limits the scope of the experiment if only a specific form of relocations, such as HQ relocations were focussed upon

In order to best understand the necessity and scope of relocation theories, it is useful to first outline the key approaches; broadly speaking, there can be said to be three primary approaches to firm relocation; neo-classical, behavioural, and institutional (Pellenbarg et al., 2002). The premise of each approach has grounding in ‘location’ theories, though they have been manipulated to highlight not simply the favourability of one location over another, but also showcases the decision-making process of choosing between them. Firm relocation can be driven by a number of external factors which they categorised as either

pull, push, or keep factors (van Dijk & Pellenbarg, 2000). As each approach has different

(12)

- 7 -

2001). Keep factors are factors which make the current location favourable, and so there is no motivation to relocate away from there; to keep with the same example, firms who were located firmly in the centre of their main trading zone would see this as an advantage and a reason to remain in their current location.

The general principles of the neo-classical stance have roots which can be traced back to Adam Smith (Isard, 1956) and are centred around the desire of firms to profit maximise, or cost minimise. This approach considers that the location choices of a firm will be made to ensure they exist at the ‘least-cost surface’, calculating all factors including tax rates, labour, transport, and external economies (Weber, 1929). Over time, it is possible that the ‘least-cost’ location will change, and thus the neo-classical approach states that if this occurs a firm must relocate. With economic factors being central to the neo-classical approach, it follows that the primary factors which influence relocation will be linked to the potential of reducing costs or increasing profits. For example, firms could be motivated toward a relocation through either push/pull factors if they became aware of an alternative location with much lower wage rates, keep factors would include their current location being the lowest cost location. For instance, firms located within the UK could well consider the corporate tax rate as a keep factor as the UK has the fourth lowest corporation tax within the EU (Tax Foundation, 2019).

(13)

- 8 -

context of Brexit, firms would simply have to wait and see what happens before making any relocation decision.

The institutional approach is the most recent of the three to gather attention (Martin, 1999). The premise of this approach is that firm relocations occur as a result of the interactions with local institutions, both formal and informal, with whom they have to negotiate. This can include wages, taxes, subsidies amongst others. These negotiations often conclude in companies discovering the ‘least-cost’ location and so has links to the neo-classical approach. However, the majority of the work applying the institutional approach requires large corporations who have bargaining power, SMEs are likely to have limited negotiations and simply have to live within the constraints set by the institutions (Hayter, 1997). Each local context will have its own institutions and the costs and benefits presented by each location will determine how attractive it is (Meyer et al., 2011). According to Brouwer et al. (2004), trust plays a key role in these negotiations and can alter the comparable attractiveness of multiple locations. Over time, firms may have built up relationships and so trust, with institutions necessary for their markets and industries. Meyer et al. (2011) discuss the concept of ‘institutional inertia’, to highlight the fact that firms can be reluctant to move due to the largely location specific and immobile links they have created with institutions, this is a prime example of a keep factor for a location based upon the institutional approach. Push factors could be a change of government regulations which will make it more difficult for the firm to operate, such as the introduction of new licenses or increased worker’s rights. A primary pull factor could be a government subsidy to help pay for a new factory or plant, such as was the case when the UK government attracted Nissan in the early 1990’s (Lowendahl, 2001).

(14)

- 9 -

shared with the EU which will undergo significant changes (Sahr et al., 2016; Lavery et al., 2018).

Townroe (1972) was amongst the first to consider investigating a ‘behavioural’ approach to location; from this work on the distinction between the behavioural approach to movement and location was drawn, and later again relocation. The behavioural approach to relocation differs from the neo-classical view as it places greater emphasis on internal factors to the firm, such as age and size (Manjon-Antolin & Arauzo-Carod, 2009) and personal considerations (Mariotti, 2005). Beyond this, the behavioural approach acknowledges the limited access to information and therefore bounded rationality which firms use in their decision making (Brouwer et al., 2004). Whilst Brouwer et al. (2004) suggest that this leads to irrational behaviour, van Dijk & Pellenbarg (2000) argue that bounded rationality is far from synonymous with irrationality, rather it acknowledges “the limitations of the decision makers abilities to interpret information” (p.194). The behavioural approach dictates that the idea of an “optimal location” is merely a theoretical abstraction, inherently flawed by its reliance on perfect information (Simon, 1959). It should not be considered that in the behavioural approach firms do not wish to profit maximise, rather that they simply do not have the ability to do so (Boschma and Frenken, 2006). This may become an increasingly important approach when the context of Brexit is considered due to the vastly limited information available to firms and their decision makers (Wright, 2016).

(15)

- 10 -

relocate or it would be behaving sub-optimally and would ultimately fail in the long run. This requires constant evaluation of current and prospective locations profitability. Conversely, in the behavioural approach it is understood that often only when a firm moves toward the boundary of spatial profitability it will begin to seek a new location, unless a significantly cheaper alternative presents itself. This means that evaluation will only take place sporadically which will save the costs of constant evaluation (van Dijk & Pellenbarg, 2000), making the behavioural approach particularly useful in relocation decisions, rather than the initial location choice (Pellenbarg et al., 2002). When there is significant change to the environment of the firm or firm level factors, the firm is likely to evaluate the spatial profitability of their current and alternative locations (van Dijk & Pellenbarg, 2000).

Figure 1: Spatial profitability boundary for satisficing behaviour (source: van Dijk & Pellenbarg, 2000 p.196; based on Smith, 1966 p.106)

(16)

- 11 -

perception and mental mapping discussed by Pellenbarg et al. (2002) is linked with the concept of psychic distance whereby closer locations are often deemed to be more attractive as there is expected to be less disruption in the flow of information for the firm (Child et al., 2002). The fact that decisions are based on anticipated future events, of which the firm often has limited information on, means that they are by definition uncertain. A great deal of the factors motivating relocation based on the behavioural approach can be linked to knowledge. Firms are more likely to be pushed from a location as the uncertainty of their ability to meet satisficing level increases, similarly, they can be more pulled toward a location if they have good knowledge of the market/region as they are better able to reduce relocation costs and spatial bias of the decision maker. Finally, firms may well experience the keep factor of spatial bias by management. At times, it can be said that people are reluctant to move away from their current location because of the safety they find in having a great deal of knowledge and experience there.

(17)

- 12 -

Table 1: Location theories and factors affecting relocation decisions (adapted from Brouwer et al., 2004 p.338)

Approach Key concept Variables

Neo-classical Location factors •_{Market size}

•_{Country of location} •_{Terms of trade} Behavioural Internal firm-level factors • Frim size

• Firm age

• The upper-echelon of the company (unique to this research)

Institutional External factors • Role of state/EU

• Regulatory bodies

The central concepts of each approach and some of the previous variables used to test them can be found summarised in Table 1. It becomes clear that each of the approaches are based around wholly different assumptions of behaviour, both human and firm, thus cannot be easily blended together to complement one another and establish a single, more complete approach. Therefore, for the duration of this article the primary approach adopted will be behavioural, albeit with some adaptation unique to this research to improve its applicability to the context. The behavioural approach can be deemed the best fit to answer this research question due to the central role which uncertainty plays in it. In order to best ensure a satisfactory conclusion, this research will adhere to the advice of Scott (2000), who suggested that studies employing the behavioural approach have all too often focused primarily on sociological and psychological variables, and too little attention is given to the economic factors (those covered in the neo-classical approach). In order to best investigate firm relocation decisions, a combination of both approaches may provide a more comprehensive understanding.

2.2 Earlier examples of empirical investigations

(18)

- 13 -

industry through various policy tools, including both labour and capital subsidies (van Dijk & Pellenbarg, 2002). Initially these policies focused upon the redistribution of existing industry, then was later applied to targeting inwards FDI and placing it in appropriate regions, such as Nissans vast investment into Sunderland in the early 1990’s (Loewendahl, 2001). Regional policy has often been the beneficiary of the findings of relocation literature in Europe (Mariotti, 2005); in the following section, a brief overview of some of the most notable works are summarised in Table 2. Showcasing not only what is known so far, but also the broad spectrum of contexts which have influenced firm relocation. Through doing so it should become more evident that the unique external context of Brexit is likely to lead to equally unique mechanisms influencing firm’s relocation decisions.

Table 2: Overview of key empirical studies of relocation (adapted and updated from Laamanen et al., 2012 p.5)

Author, Year Setting Variables Key Findings

Keeble, 1972 UK, regional relocations (same country) 1945-65. 2756 migrant factories • Labour availability • Land availability (for

building factories) •_Government industrial location policies (ILP) • Distance of ‘peripheral area’ from ’central area’

The key finding here is that government policy can have significant impact on firm migration patterns, as can availability (both land and labour), but the distance from the central area makes a huge difference to the influence of either. More distant peripheral areas are less likely to see the benefit of the government ILP, or from an abundance of available resources, than a periphery with a short distance from the centre. They use the example of Wales benefiting much more from outward firm migration from London than Scotland.

Pellenbarg et

al., 2002 Netherlands, regional relocations, 1338 companies

• Frim internal factors (age, size, industry) •_{Location factors (size}

of market, owning/renting situation,

infrastructure quality) •_{Firm external factors}

(regional economic performance, government policy)

The industry firms operate within can have considerable impact, firms in hospitality and retail are much less likely to relocate than industrial firms. Firms with under 10 employees were more likely to move but above that there was no significant correlation with frim size and mobility.

(19)

- 14 -

relocations- though they consider this to be in part due to the general measures they tested. Brouwer, Mariotti, & Ommeren, 2004 5568 firms across 21 European countries, between 1997-1999

•_{Firm age (in years)} •_{Firm size (no of}

employees) • Market (How large) • Sector

•_{Region (Northern v} Southern Europe) • Recent growth

(increase/decrease in size, M&A activity)

This study found that firms which serve a larger market are more prone to relocation. Although firm size was not said to influence relocation decisions, change in size through internal growth factors (both positive and negative) can induce relocation activity. Firm age was said to be significant only to 10%, though any relationship is moderate at best. Birkinshaw, Braunerhjelm, Holm, & Terjesen, 2006 35 of the largest Swedish MNC’s, with a number of foreign business units •_{Amount of business} activities overseas •_{Host countries} attractiveness (pull factors) • Global capital markets •_{Ownership structure}

It was found that percentage of overseas business activities was positively correlated with the likelihood for the firm to relocate a business unit overseas. If the host country was deemed to be an attractive setting (particularly vs. the home country context) it was likely that a business unit HQ will relocate to the country. The relocation of a CHQ is more likely to relocate to a country based on the influence of global capital markets as it can influence the ease of access to capital etc.

Strauss-Kahn

& Vives, 2009 Within country HQ relocation in the USA (30,566 companies studied)

Investigated both; decision to relocate (push) & location of relocation (pull) factors. They tested on whether to relocate based upon:

•_{Firm size (revenue)} •_{Firm size (number of}

HQ) •_{Firm Age} • Foreign/Domestic ownership • Infrastructure of current location Following this they tested the lure of the following pull factors: • Wage level • Corporate tax • Infrastructure • Proximity to own production centre •_{Number of other HQ} in the area

The decision to relocate is more likely for firms which are larger, both in terms of sales and number of headquarters. Similarly, younger firms, firms with foreign ownership, and firms operating in locations with poor infrastructure were more likely to relocate.

In terms of pull factors, they concluded that; low wage and corporation tax levels provided a strong pull factor. It was also found to be important the number of other headquarters in the area and the firm’s proximity to a large airport hub.

It was very clear from this work that the push factors from one location were invariable the pull factors to another.

Laamanen et

al, 2012 Multi-country dataset (17 European

They tested the following

(20)

- 15 -

countries) to study the cross-border

relocation of HQ from 1996-2006

into country (push and pull) and firm level.

Country-level (push): •_{Wage level} •_{Taxation level} • Employment rate Country level (pull):

• Remoteness of HQ location • Proportion of revenue generated overseas Firm-level factors: • HQ size (employees) • Ownership structure • Regional HQ v Corporate HQ relocation •_{Firm performance}

alongside firm level factors it became no longer significant. The corporate tax rate however was extremely important. Finally, the employment rate was also found to be significant (to 10% confidence level). The overall conclusion is that home country factors can present a significant push factor to prompt outward firm migration.

The proportion of revenue generated overseas was highly significant (p<0.001), with export focused companies much more likely to relocate HQ.

Of the firm level factors, the only significant finding presented was that RHQ are more prone to relocation than CHQ.

2.3 The current context of Brexit and the UK:

(21)

- 16 -

cognitive capabilities, and from sizable complexities in the external environment (Dosi & Egidi, 1991; Simon, 1955). The complex and unprecedented nature of Brexit means that it will be extremely difficult for firms, no matter their cognitive capabilities, to predict the impact of Brexit on their operations, although some have already taken steps to limit the prospective impact (Sahr et al., 2016).

One of the most heavily cited papers concerning firms’ behaviour when dealing with uncertainties comes from Miller (1992), who created an integrated approach to risk management in which he outlined a number of general environmental uncertainties (Appendix 1), of which the current situation in the UK meets several. In this work, Miller suggested that “companies strategic choices determine its exposure to uncertain environmental and organisational components that impact firm performance” (p.312). As a strategic solution to exposure to uncertainties, Miller contends that ‘avoidance’ is a key option, with firms choosing to avoid entering an uncertain market, or if they are already agents within the market, divestment might well be a suitable course of action. This supports the idea presented in the behavioural approach to relocation that when there is a significant event which might impact the firm’s ability to achieve satisficing levels, firms will look to other location options. Tirtiroglu et al. (2004) were among the first to consider political risk to be a motivation for relocation, although since then there has been a number of other studies; notably research by Birkinshaw et al. (2006) which highlighted a precedent of firms relocating in response to growing political uncertainty in a country. Their study cited the cases of both S.A.B and Anglo-American, who migrated away from South Africa to London in the mid-1990’s in response to the African National Congress being voted into power and the drastic change of regime that would follow.

(22)

- 17 -

et al., 2018). These uncertainties have caused an increase in financial service relocation. By May 2017 a quarter of the largest financial firms in the UK, including some of the largest agents such as HSBC and JP Morgan were in the process of moving thousands of jobs to the EU from London (EY, 2017). Therefore, of all the uncertainties highlighted by Miller’s (1992) framework, ‘terms of trade’ may be the most poignant for this research. As a partner member of the EU, the UK was able to take advantage of the four freedoms of the single market (Sahr et al., 2016):

1. Freedom of trade in goods

2. Freedom of movement of workers

3. Freedom of rights of establishment and provision of services 4. Freedom of movement of capital

(23)

- 18 -

3 Hypotheses:

3.1 Firm level factors:

Having illustrated how the uncertainties surrounding Brexit may impact the UK business environment, this article will now focus on existing literature surrounding how firm level factors can influence relocation decisions, in an effort to establish which kind of firms are more likely to relocate under the given circumstances. Pellenbarg et al. (2002) suggest that the internal mechanisms and processes within a firm must be investigated to establish why firms want to move. There has been a vast wealth of work to date regarding how firm level factors can influence a firm’s relocation decision (see Laamanen et al., 2012 for a thorough review) ranging from the ownership structure (Strauss- Kahn & Vives, 2009) to the age of the firm (Strauss-Kahn & Vives, 2009; Brouwer et al., 2004).

One area which has received increasing attention over the years is how the dependency of a firm on a particular market may alter the likelihood of relocation. This has often been determined by the proportion of domestic versus overseas business activities, or where the company’s major customer base is (Johanson & Valhane, 2009). For the purpose of this research, the focus will be on the amount of the company’s global turnover is derived in the UK. Birkinshaw et al. (2006) focused on the proportion of business activities overseas for 40 of the largest Swedish MNC’s and found that there was a significantly increased likelihood of a firm relocating away from Sweden if it had a greater proportion of international activities. They provided a number of supporting arguments as to why this relationship exists- primarily economic in kind; there is likely to be greater efficiency in relocating towards a larger proportion of the market as there will be reduced costs, both of transport and of knowledge transfer (Mariotti & Piscitello, 1995). This reasoning becomes increasingly important in the face of Brexit where there is the potential for increased costs for import/export activities, or for compliance to regulations or legislations (Lavery et al., 2018).

(24)

- 19 -

approach through following the suggestion of Scott (2000), by placing some attention on it. As discussed, in the behavioural approach, firms are often willing to stay in their location provided they can achieve satisficing levels. However, as firms are likely to consider relocation when there is a significant change in the market conditions, Brexit is likely to represent different levels of threat to different companies. Although in the future, once the terms of trade are decided, there may well be a considerable number of firms relocating for neo-classical reasons; currently, the decision of firms to relocate is based much more closely on their ability to interpret what the prospect of Brexit could mean for them, which falls firmly in the remit of the behavioural approach.

For firms which solely operate in the domestic market of the UK, whilst they may be significantly more exposed to the socio-political uncertainty of the market, the fact that they are wholly dependent upon the UK market, significantly limits their ability to

respond to the uncertainty. Laamanen et al. (2012) suggest that through having a smaller proportion of sales generated in the home country, a firm is considerably less dependent upon the home country due to its greater exposure to other markets, enabling the firm to consider relocation as a possible response to the uncertainty. Through having

increased exposure to other markets, the firm would possess knowledge and contacts, which would enable the firm to reduce any effect of distance-decay (Pellenbarg et al., 2002), reducing the uncertainty of a new location accordingly. In doing this, the cost of relocating could fall, as less time would be needed to establish the suitability of a

potential location. Whilst the formerly presented arguments showcase how the ability of the firm to use information, it does not offer any direction on factors which may affect the willingness of the firm to use it.

(25)

- 20 -

reduce outsidership, which can lead to improved stakeholder interactions and increased likelihood of survival in the market. This can of course be a double-edged sword, as if the firm decides to relocate some of their operation away from the UK, they risk losing legitimacy in the UK, which will impact their subsequent performance in the market (Delacroix & Swaminathan, 1991). As the firm stands to lose legitimacy in one market regardless, it comes down to the perceptions of the firm and its key decision makers as to which market it prioritises. Though it can be expected that the firm will be less willing to relocate if it relies more heavily on the UK market. Therefore, through considering both the willingness and ability of the firm to relocate away from the UK in the face of Brexit, it is anticipated that:

H1- The higher the percentage of global turnover derived from the UK, the lower the likelihood of firm relocation.

(26)

- 21 -

There is evidence that even though firm mobility decreases with age, older firms are often less content with their current location than younger firms (Brouwer, 2004). This highlights that it is sometimes not a rational choice which causes a firm to remain in their current location, but rather an unwillingness, or inability to relocate due to the intangible value placed on the tradition the firm has at its current location. Brouwer et al. (2004) contend that another reason for this is that as a firm continues to exist within a specific environment it develops long-term trust-based relations, often initiated through spatial proximity (Putnam, 1993). In their work, Brouwer et al. (2004) found a moderate correlation between firm age and propensity to relocate, but only significant at the 10% level. Though they found that broadly speaking, firms which had existed in a location for over 80 years were significantly less likely to relocate than firms which had existed there for less than 30 years. This relationship has been supported in the work of Brouwer (2004), and more recently Strauss-Kahn & Vives (2009). Therefore, it is expected that:

H2- The longer the firm has operated in the UK, the lower its likelihood of relocation

(27)

- 22 -

cost of a relocation, somewhat undermining the previous work of both Kelly & Amburgey (1991) and Brouwer et al (2009). As such, there is yet to be cohesive results on whether firm size influences the decision to relocate.

As mentioned in section 2.1, there has thus far been limited scope for indicators of ‘size’, which leads to the question of what other ways the size of a firm could be measured, and indeed the effect it would have on its inertia, or willingness to relocate. It could well be argued that, in some industries, the number of employees is becoming an increasingly poor indicator of size, especially in more developed industries where there has been increased mechanisation and automation, the bigger firms can often afford to reduce staff numbers and replace them with machines.

(28)

- 23 -

company had, the more likely it was to relocate aspects of HQ, suggesting that often it was beneficial for these large firms to split some aspects of HQ functions across larger subsidiaries. Therefore, in contention to the prevailing concept of size induced organisational inertia, there is sufficient reason to expect that:

H3- The more subsidiaries a firm has, the more likely it is to relocate 3.2 Board composition factors

Following the behavioural approach to relocation, a firm’s ability to understand available information will influence how they respond to it. At this point, the key decision makers in an organisation will become increasingly important. Hambrick & Mason (1984) developed the UE theory in order to answer the question “why do organisations act as they do?” (p.193), they concluded that the strategic decisions of a firm will be the result of the collective choice of the ‘dominant coalition’. LeBlanc & Gillies (2005) further this, suggesting that the board members are the most important element to maintaining the wellbeing of the firm. The term ‘dominant coalition’ has been interpreted differently over the years, ranging from only the CEO in some instances, to all board members in others (Carpenter & Weikel, 2011). The earlier work of UE held more of a focus of top-management teams which was purely the executive, however as corporate governance improved over time, there has been a general trend for greater involvement of the board to include strategic direction and determining the firms risk position (Stiles, 2001) Therefore, for the remainder of this paper, the conceptualisation of UE theory will incorporate all board members.. According to Baysinger & Hoskisson (1990), the primary role of the board is to protect shareholders interests; and a key function within this role is to restrict the firm’s investments in ‘risky’ projects which may penalise the organisation in the competition for survival (Farma & Jensen, 1983, p.306).

(29)

- 24 -

-or in this instance collective decision makers, bring an initial cognitive base and values through which they interpret selective events as they see them. The UE cannot hope to see and understand every stimulus, and so it is from the limited field of vision that perceptions and interpretations are drawn ultimately leading to a strategic choice. In the case of this study, the situation is Brexit and the uncertainty around it, and the ‘strategic choice’ is the decision to relocate or not. Figure 2 highlights the complexity of the decision-making process for a strategic choice faced by the UE.

Figure 2: Strategic Choice Under Conditions of Bounded Rationality (Hambrick & Mason, 1984 p.195)

(30)

- 25 -

The impact of the demographic composition of a board has on company performance has long been the focus of management literature (Miller et al., 1998). Though to date there is yet to be cohesion in the findings, some suggesting that it can aid decision making (Carter, et al, 2003), and others finding that it can actually hamper the process through its divisive nature (Glick et al. 1993). Building on the UE theory, it is suggested that having a wider range of demographic and functionally (occupationally) diversity at a board level will enable the group to solve problems faster and more effectively due to the broader perspectives and experiences they can draw upon (Barsade et al., 2000; Hambrick et al., 1996).

Two of the most investigated areas of diversity at a board level are gender and nationality (Cook & Glass, 2014). The traditional homogeneous board composition in the UK is middle-class, older, white, British men, therefore the inclusion of female, non-white, or non-British directors would represent greater board diversity. Carter et al. (2002) consider that people with “different gender, ethnicity, or cultural backgrounds might ask questions that would not be posed by directors with more traditional backgrounds” (p.6). It could be assumed that one of the greatest demographic factors in this instance would be the nationality of directors; if the directors are from a foreign country, it is likely that they are less embedded in the UK (Portes & Sensenbrenner, 1993). The manifestation of embeddedness can be considered particularly important in relocations when considering the behavioural approach to relocation the idea of mental maps and perceived distance is discussed, so if the board were to be comprised wholly of British directors, it is likely that they would have a greater perception of the distance between their company and any country outside the UK.

(31)

- 26 -

of imperfect knowledge, site selection will most often be in a ‘familiar location’. Richardson (1979) attempts to explain this phenomenon stating that it maximises the “access to relevant information needed for profitable production” (p.64) A likely side effect of the rising nationalism which brought about Brexit in the first place (Gusterson, 2017), is that perception of distance between Britain and the rest of Europe is larger than previously. Through considering the available literature, it is expected that:

H4- The higher the proportion of British directors on the board, the lower the likelihood of the relocating

The proposed hypotheses lead to the creation of the following conceptual model: Figure 3: Conceptual model of proposed relationships

H1- Proportion of turnover derived

from the UK H2- How long the

(32)

- 27 -

4 Methodology

4.1 Research Philosophy:

This research falls under a positivist stance, a positivist philosophy contends that there is a single objective reality independent of the beliefs and perspective of the researcher (Carson et al., 2001). In order to adhere to this philosophy, it is necessary for the researcher to maintain rationality and logic (Hudson & Ozanne, 1988). The research will be deductive in its nature and aims to use the prevailing academia to support the hypothesis as stated. A quantitative method of analysis will be employed as this will be most appropriate to find an objective solution to the research question. Through employing a quantitative approach there is less room for researcher bias through interpretation.

4.2 Sample:

(33)

- 28 -

Although this sample is smaller than some of the seminal relocation studies, the ongoing nature of Brexit means that there is a considerably smaller pool of potential relocations than some of the studies which covered a much greater date range, up to 20 years (Keeble, 1972), or multi-country datasets covering up to 21 countries (Brouwer et al., 2004), though the study offers a larger pool of companies than Birkinshaw et al. (2006) and so there is precedent for smaller sample sizes to accommodate studies with greater constraints on time or resources. Naturally, as successfully answering the research question relies on establishing what drives firm relocation it is crucial to also include companies within the sample who have not relocated as somewhat of a ‘control group’ to better highlight any patterns (Laamanen et al., 2012). Therefore, at this stage it remained unknown how many of the sample had relocated in part since 29th_{March 2017.}

4.3 Variables:

(34)

- 29 -

Table 3: Summary of variables

4.4 Data Collection:

The first stage in data collection was to discover which companies fall within the constraints of the control variables. Following this, as there is no central database of firm relocations in the UK, relocation events will be researched following the relatively well-established method employed by Laamanen et al. (2012) amongst others, which consists of utilising LexisNexis, Reuters and other news databases. A concern highlighted in Laanmanen et al's (2012) study was the potential existence of ‘false negatives’, where a relocation has occurred which has not been accounted for; in order to significantly lessen the likelihood of such an issue, stock market notifications and search engines will be used to further the data collection and ensure its reliability. Through this technique it satisfactory that the chance of missing relocation information is limited as best as possible within the time and financial constraints of this study.

In order to gather data on the turnover generated in the UK as a proportion of the firm’s global turnover both Endole and Orbis databases will be used, supplemented with company financial reports where required. Through checking the company’s registration date in the UK with companies’ house, it is possible to establish when the company was registered, unfortunately companies can change name and re-register so it will only be the name of the company as registered in 2017 which will be considered. Orbis also provides data regarding the number of subsidiaries which are registered to a parent company, so

Dependant Variable Decision to relocate

Independent Variable UK turnover as a proportion of total turnover (H1)

How long the firm has operated in the UK (prior to 2017) (H2)

Number of registered subsidiaries (H3)

Proportion of British directors on the board (H4)

(35)

- 30 -

this data is also easy to obtain. Finally, in order to gather data on the board composition, again the listed status of all firms within the study ensures that the data is readily available. In order to gather data regarding the nationalities of directors, the BoardEx database will be used. As all of the data will be secondary data, obtained from existing databases publicly available, it follows that there are no real ethical issues to be considered through the collection of this data. This is especially true as there is no individual personal data collected of any of the board members.

Table 4: Summary of data required to test each hypothesis

H1- The higher the percentage of global turnover derived from the UK, the lower the likelihood of firm relocation.

•_{UK turnover as a proportion of global} turnover from financial year closest to 29th_{March 2017 (Orbis, Endole,}

company websites)

H2- The longer the firm has operated in the UK, the

lower its likelihood of relocation • The year the firm was registered in the UK (Companies House, Endole)

H3- The more subsidiaries a firm has, the more likely it is to relocate

•_{The total number of subsidiaries} registered to the company (Orbis)

H4- The higher the proportion of British directors on the board, the lower the likelihood of the relocating.

• Board Member nationality (BoardEx, Company websites)

4.5 Data Analysis:

(36)

- 31 -

provides the research with meaningful data regarding how a single unit increase of one of the independent variables will affect the likelihood of the firm relocating.

(37)

- 32 -

5 Findings:

Throughout this section, the results of the statistical tests for each hypothesis will be presented and interpreted before offering a decision whether or not to accept them. Table 5 offers the descriptive statistics of each of the variables, which demonstrate that each of the variables contributed to 128 cases, showing that there is no missing data. Beyond that, it provides evidence to the upper and lower values for each of the variables which can prove helpful in the later analysis.

Table 5: Descriptive statistics

N Minimum Maximum Mean Std. Deviation

Relocate (Y/N) 128 0 1 .40 .492 UK v global turnover (%) 128 .00000000000 0000 100.00000000 0000000 56.487500000 000004 40.963463651 535980 How long has the firm

operated in the UK

128 1 160 34.50 31.752

Number of subsidiaries 128 0 11360 757.33 1527.948

Proportion of directors with British nationality

128 0 100 50.15 35.049

(38)

- 33 -

Table 6: Collinearity statistics

Model Unstandardized Coefficients Standardized Coefficients t Sig. Collinearity Statistics

B Std. Error Beta Tolerance VIF

1 (Constant) .493 .095 5.167 .000

UK v global turnover (%) -.003 .001 -.284 -3.052 .003 .759 1.318

How long has the firm operated in the UK

.002 .001 .100 1.191 .236 .929 1.076

Number of subsidiaries 8.157E-5 .000 .254 3.008 .003 .921 1.085

Proportion of directors with British nationality

.000 .001 -.025 -.267 .790 .742 1.349 a. Dependent Variable: Relocate (Y/N)

(39)

- 34 -

Table 7: Case processing summary

Unweighted Casesa _N _Percent

Selected Cases Included in Analysis 128 100.0

Missing Cases 0 .0

Total 128 100.0

Unselected Cases 0 .0

Total 128 100.0

a. If weight is in effect, see classification table for the total number of cases.

Table 8: Classification table(Null model)

Observed Predicted Relocate _Percentage Correct No Yes Step 0 Relocate No 77 0 100.0 Yes 51 0 .0 Overall Percentage 60.2

a. Constant is included in the model.

b. The cut value is .500

(40)

- 35 -

Table 9: Omnibus tests of model coefficients

Chi-square df Sig.

Step 1 Step 29.486 4 .000

Block 29.486 4 .000

Model 29.486 4 .000

Table 10: Classification table(Full model)

Observed Predicted Relocate Percentage Correct No Yes Step 1 Relocate No 61 16 79.2 Yes 30 21 41.2 Overall Percentage 64.1

a. The cut value is .500

(41)

- 36 -

Table 11: Exponential coefficients of all independent variables

B S.E. Wald df Sig. Exp(B)

95% C.I.for EXP(B) Lower Upper Step 1a UK v global turnover (%) -.015 .006 7.361 1 .007 .985 .974 .996

How long has the firm operated in the UK .008 .006 1.660 1 .198 1.008 .996 1.021 Number of subsidiaries .001 .000 7.807 1 .005 1.001 1.000 1.001 Proportion of directors with British nationality -.002 .007 .120 1 .730 .998 .985 1.011 Constant -.238 .461 .266 1 .606 .788

a. Variable(s) entered on step 1: UK v global turnover (%) 1d.p, How long has the firm operated in the UK, Number of subsidiaries, Percentage of UK directors.

The co-efficient provided for each of the independent variables highlights how each variable interacts with the intercept of the direct variable, however as we want to see the effect that an increase in each of the variables in the model (Table 11) has on the decision to relocate it is more interesting to focus on the exponential of the co-efficient (Exp(B)) of each predictor. The Exp(B) effectively illustrates the change in the firm’s likelihood to relocate for every increase in each of the independent variables. The Exp(B) of the percentage of total sales derived from the UK is 0.985, which translates to that for each extra percentage of sales derived from the UK, the likelihood of the firm relocating becomes marginally lower (98.5% as likely). This result can be said to be statistically significant to 0.007, which means that there is sufficient evidence to support and accept hypothesis 1.

(42)

- 37 -

result was not statistically significant (even to the 10%) with a sig. value of .198 and so there is not sufficient statistical support to reject the hypothesis, but hypothesis 2 is certainly not supported or accepted. Once again, the 95% confidence interval for this

variable covers both above and below 1, meaning that it would be very easy to find results within the dataset which showcase the opposite relationship to the Exp(B) predicted. The number of subsidiaries provided an Exp(B) value of 1.001, which suggests that for every extra subsidiary, a firm is marginally (by 0.1%) more likely to engage in relocation behaviour. Whilst this result is statistically significant with a value of 0.005, it is so marginal that its impact should not be overstated. However, as the number of subsidiaries a firm can have is limitless, from this sample of firms alone there is a range of over 11,000 between the lowest and highest. It can be said that the effect of having an extra 100 or 1000 subsidiaries would have a considerable impact on the firm’s likelihood of relocating, with firms becoming twice as likely to engage in relocation for every 1000 subsidiaries they have. Therefore, there is statistical evidence to support and accept hypothesis 3.

The final hypothesis which can be answered from this table, regarding the percentage of British directors on a company’s board provides an Exp(B) value of .998, suggesting that for each percentage increase in British directors, the likelihood of the firm’s relocation falls slightly (99.8% as likely). Though the significance value of 0.730 is well outside the confidence level, and so although the exponential of the co-efficient would suggest that hypothesis 4 is supported, it cannot be deemed statistically significant, and so hypothesis 4 must not be accepted. The fact that within the 95% confidence interval, the upper bound

(43)

- 38 -

6 Discussion & Limitations:

6.1 Discussion:

The results presented in section 5 highlight that although some of the outcomes were as expected, firm relocation behaviour in the face of socio-political uncertainty may differ from relocation behaviour in more stable conditions. The results have enabled the objectives of the research to be fulfilled thus, provide an enrichment of relocation academia.

Redarrding the influence which the proportion of the firms’ global turnover derived from the UK, the fact that the hypothesis was supported by the empirical data adds further weight to the predicted relationship which had already been relatively well established through the work of Birkinshaw et al. (2006), and Laamanen et al. (2012) amongst others. Arguably, the predominant theoretical implication which can be drawn from this research is that the relationship still holds true in the face of socio-political uncertainty. In section 3, each of the studies discussed had placed different central assumptions as to why the relationship existed. Birkinshaw et al. (2006) suggested that firms would relocate to other countries to see gains in efficiency and control. The fact that there is significant potential for regulatory divergence between the UK and the EU could signify that there would be a considerable risk to future efficiency. Moreover, firms could be at risk of losing the legal right to provide their services in EU countries without a passport, such as the case of financial institutions.

(44)

- 39 -

aspects of their operation in anticipation for the change in terms of trade and the potential regulatory divergence. Similarly, further exploration would be needed to establish if firms placed preference on legitimacy in one market over the other. In order to establish this, qualitative research could be undertaken with the actors who chose relocate, in order to establish which factors were drivers in the relocation decision.

The inconclusive results which came from testing the length of time the firm has operated in the UK are far from surprising as there have been many unsuccessful attempts at proving the link between firm age and mobility. The results, although not statistically significant, indicated that older firms were in fact, more likely to relocate in this instance, contrary to the prevailing literature regarding organizational inertia. A potentially interesting factor to consider in this instance is the possibility that firms have become somewhat embedded in both the UK and the EU. Prior to the EU, the UK joined the EEC in 1975, meaning that all firms operating within the UK have been operating within the EEC for over 40 years, in this time the integration between the institutions of the UK has increased. As stated at the very beginning, it was felt that the institutional approach to relocation would struggle in this investigation due to the lack of any agreement (at the time of writing).

(45)

- 40 -

of convergence over the last 30 or so years with increasing globalisation. The importance of this shortcoming could well become increasingly stark in the coming years with rising nationalism observed in many countries in the EU and further afield.

The majority of research on firm relocation appears to consider firm size in terms of revenue or employees, as factors which can influence mobility. Therefore, the decision to investigate the impact that the number of subsidiaries the firm has certainly has implications for work on organisational inertia in management and spatial sciences alike, initially by offering a widening scope for the concept. To be sure, the result has implications upon the model presented in this study, allowing one of the key objectives of the research to be fulfilled as it provides evidence that the number of subsidiaries does influence firm relocation decisions. As mentioned, the increased likelihood of relocation based on a single extra subsidiary is minimal, as often is the case with size-induced inertia (Pellenbarg et al. 2002), but when extrapolated to larger amounts, paints an interesting picture. It must be noted, that the relationship evident from the data in section 5, shows that the number of subsidiaries has less baring on firm relocation than the number of HQ on HQ relocation. Strauss-Kahn & Vives (2009) concluded that for every 10% increase in the number of HQ a firm became 6.7% more likely to relocate HQ.

(46)

- 41 -

Finally, this research attempted to extend the scope of the behavioural approach by placing greater emphasis upon the role that the upper-echelon of a firm would play in its relocation decisions. Whilst there is certainly well cited and supported literature to base this extension upon, the empirical findings presented in section 5 offer little indication that the amount of the directors who were British nationals influenced the firms’ decision to relocate away. As the data was far from statistically significant, it does not rule out the fact that the upper-echelon could influence whether a firm chooses to relocate or not. However, the size of the firms which are included in the sample could well explain the lack of significance found. A recent government report of the drivers of firm relocation found that as businesses increase in size, the personal preferences of individuals within the firm become less influential (DBEIS, 2018). Similarly, these findings do not strengthen the idea of home-field spatial bias put forward by Figueiredo et al. (2002), although it must be remembered that their work had focused on smaller firms. Although it is beyond the scope of this research to support the idea that personal bias from decision makers is more prevalent in smaller firms than in very large ones, such as those included in this sample, it does serve to highlight that personal bias plays a limited role at best amongst the larger companies.. Therefore, if any future attempt is made to magnify the role of the upper-echelon in relocation decisions, it is recommended that smaller firms be used in the sample. If anything, although the results offer little empirical evidence, this research may be considered a starting block for further work by highlighting the strong theoretical overlap which had thusly been relatively underdeveloped.

6.2 Limitations

(47)

- 42 -

research initially considered extending the upper-echelon aspect of the literature by testing the influence that the political opinion (regarding Brexit) of the CEO would have on the likelihood of firm relocation, though through the data collected, it was found almost unanimously that CEO’s were against Brexit- particularly weary of the implications of a no-deal Brexit. Therefore, the data was unable to provide any meaningful results and thus, the hypothesis was dropped.

A final limitation of this work is that it groups all forms of relocations together due to the small pool of potential relocations. Although this work does not differentiate between types of relocation, it acknowledges that different levels of importance will be placed on each factor depending on the degree of relocation, for instance it is to be expected that there will be much less consideration for corporate history when relocating 50 employees than there will be for relocating a corporate headquarters. Therefore, whilst it would be wrong to proclaim that the model proposed in this article is equally appropriate for all forms of relocation, the results of this work can still be said to offer a general insight into relocation behaviour.

6.3 Future Direction

(48)

- 43 -

(49)

- 44 -

7 Conclusion

In the preceding pages, the three objectives set out in the introduction have been discussed with relevant implications highlighted. These objectives were created as a means of developing a better answer to answer the question “To what extent do firm level factors influence the outward relocation decisions of firms in the face of growing market uncertainty brought about by Brexit?”. Throughout the work, both through examining

prior theoretical and empirical research, alongside through this papers own empirical contributions it has become increasingly clear that firm level factors do indeed influence whether a firm decides to relocate away from the UK in the face of Brexit, albeit not all factors appear to impact the decision. Through the quantitative study of the relocation activities of 128 large firms registered in the UK, it can be concluded that the firms who are more dependent on the UK market for their turnover, and those with a smaller number of subsidiaries are less likely to relocate away from the UK in times of socio-political uncertainty, such as the context brought about with Brexit. The research did not find a statistically significant link between board composition and the decision to relocate, using board member nationality as a predictor. As a concluding remark, the model presented in this research can be considered satisfactory in its capability of answering the research question- though further work on the subject can only serve to better the theoretical development and understanding.

7.1 Theoretical and Practical contributions of this research

(50)

- 45 -

(51)

- 46 -

Bibliography:

Abler, R., Adams, J. and Gould, P. (1972). Spatial organization: the geographer's view of the world. Englewood Cliffs, N.J.: Prentice/Hall International.

Arauzo-Carod J.M., & Manjón- Antolín M.C., (2004) Firm size and geographical aggregation: an empirical appraisal in industrial location. Small Bus Econ 22, 299–312

Baaij, M. G., Mom, T. J., Van den Bosch, F. A., & Volberda, H. W. (2015). Why do multinational corporations relocate core parts of their corporate headquarters abroad? Long Range

Planning, 48(1), 46-58.

Barsade, S. G., Ward, A. J., Turner, J. D., & Sonnenfeld, J. A. (2000). To your heart's content: A model of affective diversity in top management teams. Administrative Science Quarterly, 45(4), 802-836. Baysinger, B., & Hoskisson, R. (1990). The Composition of Boards of Directors and Strategic Control: Effects on Corporate Strategy. The Academy of Management Review, 15(1), 72-87.

BBC, (2019). Do Boris Johnson's tax and spending plans add up? Retrieved from: https://www.bbc.co.uk/news/uk-politics-49097133

Benito, G. R. G., Lunnan, R., & Tomassen, S. (2011). Distant encounters of the third kind: Multinational companies locating divisional headquarters abroad. Journal of Management Studies, 48(2), 373–394.

Beugelsdijk, S., Kostova, T., Kunst, V., Spadafora, E., & van Essen, M. (2018). Cultural distance and firm internationalization. Journal of Management, 44(1), 89-130.

Birkinshaw, J., Braunerhjelm, P., Holm, U., & Terjesen, S. (2006). Why do some multinational corporations relocate their headquarters overseas? Strategic Management Journal, 27(7), 681-700.

Boivie, S., Bednar, M.K., Aguilera, R.V. and Andrus, J.L., (2016). Are boards designed to fail? The implausibility of effective board monitoring. Academy of Management Annals, 10(1),.319-407. Boschma, R.A. and Frenken, K., (2006). Why is economic geography not an evolutionary science? Towards an evolutionary economic geography. Journal of economic geography, 6(3), 273-302. Breinlich, H., Leromain, E., Novy, D., & Sampson, T. (2019). Voting with their Money: Brexit and Outward Investment by UK Firms. CEP Brexit Analysis, (13).

Brouwer, A.E., (2004). The inert firm; why old firms show a stickiness to their location.

Brouwer, A. E., Mariotti, I., & van Ommeren, J. N. (2004). The firm relocation decision: An empirical investigation. Annals of Regional Science, 38(2), 335–347.

Bryman, A. and Bell, E., (2015). Business Research Methods (Vol. 4). Glasgow: Bell & Bain Ltd. Carpenter, M.A., Geletkanycz, M.A. and Sanders, W.G., (2004). Upper echelons research revisited: Antecedents, elements, and consequences of top management team composition. Journal of

(52)

- 47 -

Carpenter, M.A., & Weikel, M.K. (Eds.). (2011). The handbook of research on top management

teams. Edward Elgar Publishing.

Carson, D., Gilmore, A., Perry, C., and Gronhaug, K. (2001). Qualitative Marketing Research. London: Sage.

Carter, D. A., Simkins, B. J., & Simpson, W. G. (2003). Corporate governance, board diversity, and firm value. Financial review, 38(1), 33-53.

Certo, S. T., Daily, C. M., & Dalton, D. R. (2001). Signaling firm value through board structure: An investigation of initial public offerings. Entrepreneurship Theory and Practice, 26(2), 33-50.

Child, J., Ng, S.H. and Wong, C., (2002). Psychic distance and internationalization: Evidence from Hong Kong firms. International Studies of Management & Organization, 32(1), 36-56.

Chin, M.K., Hambrick, D.C. & Treviño, L.K., (2013). Political Ideologies of CEOs: The Influence of Executives’ Values on Corporate Social Responsibility. Administrative Science Quarterly, 58(2), 197– 232.

Chin, M.K. and Semadeni, M., (2017). CEO political ideologies and pay egalitarianism within top management teams. Strategic Management Journal, 38(8), 1608-1625.

Cook, A., & Glass, C. (2014). Do diversity reputation signals increase share value? Human

DBEIS. (2018). Drivers of firm relocation: Final report, BIES Research Paper Number 2019/002, May 2018.

Delacroix, J. and Swaminathan, A., (1991). Cosmetic, speculative, and adaptive organizational change in the wine industry: A longitudinal study. Administrative Science Quarterly, 36(4).

Desai, M. A. (2009). The decentering of the global firm. World Economy, 32(9), 1271-1290.

Dosi, G. and Egidi, M. (1991). Substantive and procedural uncertainty. Journal of Evolutionary

Economics, 1(2),145-168.

Goodwin, M.J. and Heath, O., 2016. The 2016 referendum, Brexit and the left behind: An aggregate-level analysis of the result. The Political Quarterly, 87(3), 323-332.

Gupta, A. and Wowak, A.J., (2017). The elephant (or donkey) in the boardroom: How board political ideology affects CEO pay. Administrative Science Quarterly, 62(1), 1-30.

Erikson, R. S., & Tedin, K. L. (2003). American public opinion: its origins, content, and impact. New York, Longman

EY, (2017). Financial services firms starting to enact Brexit contingency plans, but those voicing intentions to move remain in the minority. Retrieved from: http://www.ey.com/uk/en/newsroom/news-releases/17-05-08—financial-services-firms-

starting-to-enact-brexit-contingency-plans-but-those-voicing-intentions-to-move- remain-in-the-minority

Fama, E.F., & Jensen, M.C. (1983). Separation of ownership and control. The journal of law and