In the aftermath of the financial crisis of 2007-08 the blame for much of what went wrong was laid at the door of the global financial industry. Bankers, regulators and economists lost much of their public standing as hundreds of billions of taxpayers’ money was used to bail out troubled financial institutions, leaving already precarious government budgets in dire straits, nearly blowing up the Eurozone and forcing austerity measures upon innocent citizens, that will hamper economic growth for years to come. The role of finance in these affairs as well as its historical association with the rich and influential propagated the views that, first, sophisticated financial products and services are being used by those at the very top of the income distribution to further increase their share of world’s assets, and second, the finance industry is creating wider segregation of income in the society through generous pay packages to the highly-skilled finance professionals. The Occupy movements with their slogan “We are the 99%” launched a storm of public outrage against the finance-driven capitalist appropriation and the so-called self-enriching ‘financial elite’.
The apparent discontent stemming from the class divide between the capital owners and the laborers is nothing new. However, the political attack on the finance profession highlights an interesting shift that is occurring in today’s society: a social divide within the labor force, namely between working-rich and working-poor. Thomas Piketty and Emmanuel Saez (2004) used the term “working-rich” to distinguish the top income earners in the contemporary society from earlier elites who generated most of their income from the ownership of assets, also known as the “rentier class” (Keynes) or “leisure class” (Veblen). In the past 30 years, the working-rich have been working their way up to the top layers of the income distribution; they now outnumber heirs to existing wealth in the lists of world’s rich published by Forbes, Sunday Times and the like. When the income inequalities started to increase globally, after a prolonged period between the 1920s and late 1970s of decreasing inequalities, it happened largely as the result of developments at the top end of the income distribution. As Hacker and Pierson (2010) put it quite succinctly, “…the bottom of the distribution went nowhere, the middle saw a modest gain, and the top ran away with the grand prize”. The working-rich, among which there are many finance professionals, are at the center of these developments.
There is no clear consensus regarding how these developments came about. In the economic literature four broad types of explanation can be found. The first one is “globalization” (Milanovic, 2002a). Due to the increasing international integration of production and consumption markets, the world has for all economic purposes in fact become a single labor market. Since labor costs differ substantially between countries and since new (organizational) technologies allow firms to by and large overcome differences in labor productivity, in the high-wage West in particular, low skilled workers increasingly experience downward wage pressures.
The second one is what is called “skill-biased technological change” (Jaumotte, et al., 2013). In this storyline, technological changes, especially the widespread introduction of increasingly powerful and ever cheaper Information and Communication Technologies (ICT) have put a growing premium on workers who possess the skills to develop, implement and update new hardware and software. As a result, the income of the workers with these skills has pulled away from the incomes of those without, resulting in an ever widening income gap between workers with only secondary education and workers with tertiary education, especially in the fields of computing, finance, economics, mathematics and the natural sciences.
The third one is the increasing size and number of so-called “winner-take-all” markets. (Frank & Cook, 1996) . This explanation combines elements of the former two to argue that, for instance, in the fields of sports, culture and entertainment the rewards for top performers increased exponentially due to the ever-expanding global audiences reached through new electronic distribution channels. A similar trend is occurring outside these industries, as chief executives of multinational businesses, top corporate lawyers and ‘Wall Street’ financiers are earning superstar-worthy remunerations1 (Bakija, et al., 2008; Kaplan & Rauh, 2010).
The fourth is “financialization” (Zalewski & Whalen, 2010). The claim here is that a longer term transformation of contemporary capitalism has resulted in the increasing size, share and importance of financial transactions for firms, households and national economies (Pike & Pollard, 2010; Freeman, 2010; Lapavitsas, 2011). The ensuing growth of global capital flows has created plenty of opportunities for well-positioned insiders to skim off substantial slices of wealth, causing huge increases in income inequality, predominantly at the high end tail of the income distribution.
It is easy to see that these explanations are complementary rather than mutually exclusive. The globalization-thesis primarily addresses worsening remuneration conditions at the low end of the income distribution, while the other three provide (partial) explanations for the developments at the other tail of the distribution, resulting in a rapid global increase in the number of High Net Worth Individuals (HNWIs)2. The extent of this development is staggering: in 2010 alone, that is two years after the bankruptcy of Lehman Brothers, the number of HNWIs increased by 8.3 percent to 10.9 million, while the aggregate value of their assets rose by 9.7 percent to US$42.7 trillion (Capgemini/Merrill Lynch, 2011). More recently, Oxfam reported that 85 richest individuals in the world now own as much assets as the bottom 50% of the global population combined (Oxfam, 2014).
Adjudicating which one of these explanations provides most explanatory leverage is not the aim of this paper. Nor is a political or economic evaluation of these inequalities, their effects and possible remedying. Rather, using one of the causal theories for the global increase in income inequalities – financialization – this paper aims to explore empirically the linkages between different modes of financialization (Engelen & Konings, 2010) and different levels of income inequality, through a focus on the urban nodes where finance has historically tended to concentrate, i.e. the International Financial Centers (IFCs).
This paper puts forward two assumptions. First, that financialization is indeed a driver of growing inequalities, primarily through its effects on rising income at the high end of the income distribution. And second, that functional differences that exist within the finance industry influence the development of income inequalities in the IFCs. In other words, IFCs that tend to specialize in higher rewarding financial activities will also demonstrate a higher level of income inequality and vice versa.
From these assumptions follow two key objectives. First, to examine the extent to which financialization as a defining feature of the contemporary capitalism can be linked to increasing levels of urban income inequalities. We know from a growing body of financial geographical literature as well as industry reports that financial activities are not spread evenly across the globe but instead follow distinct spatial articulations. For instance, bank-based financial systems have a tendency to be dominated by IFCs which are composed predominantly of national banking holding corporations, Paris, Frankfurt, Brussels, Tokyo and Amsterdam being the prime example. The reverse is true for IFCs situated in market-based financial systems3. These tend to be dominated by financial market operators and their prime customers. Here London and New York may serve as examples. In other words, IFCs tend to occupy distinct niches in the global financial division of labor.
Second, to investigate to what extent these spatially distinct distributions of financial activities are related to different levels of income inequalities in the cities that house different types of IFCs. From industry reports we know that the market-based activities of investment banks are associated with much higher bonuses and other modes of remuneration than retail banking activities. Conversely, the head offices of large bank holding corporations tend to employ tens of thousands of employees, many of which fulfill lowly remunerated back office functions. Hence, we expect to observe distinct inequality patterns in different types of IFCs – namely, when comparing the lower and higher concentration levels of financial activities, we expect to find limited surge in inequalities in bank-dominated IFCs versus steep rise of inequalities in market-dominated IFCs. By using a unique, self-constructed database on functional differentiation of financial activities across 53 major financial centers, we aim to trace how these niches are linked to different degrees of income inequalities in different International Financial Centers.
The paper is organized as follows. Section 2 defines the main concepts and reviews briefly the underlying literature. Section 3 describes the data sources and methodology employed in the paper. Section 4 outlines the results of the analysis and discusses the evidence in support of our assumptions. Section 5 concludes and sketches an agenda for further research.
Key concepts and literature
This paper takes its cue from both old and new work on the linkages between IFCs – the urban nodes where traditionally most of the wealth of ‘financial elites’ accrues – and the financialized phase which capitalism as a regime of accumulation has putatively entered into since the late 1970s, early 1980s (Arrighi, 1994; Krippner, 2012). The Swiss economic historian Youssef Cassis paired this with a history of Western finance and its urban articulations more broadly, including the spatial ramifications of the Great Financial Crisis of 2008 (Cassis, 2010).
Similarly, Allen Scott recently wrote about “cognitive-cultural capitalism” to denote the socio-economic transformations that our contemporary capitalist societies are experiencing. Scott built his thesis on the idea that the new stage of capitalist development is characterized by increasing dependence of labor processes on intellectual and affective human assets and decreasing importance of “bluntly routinized mental or manual forms of work” (Scott, 2007: 1466) – statements which clearly reflect the ‘skill-biased technological change’ theory adumbrated above. Like Cassis, Scott emphasizes the fundamental urban nature of this transformation. It is (specific) cities much more than national economies that serve as the principal stages and crucibles for these developments. By doing so, Scott combines modern economic theory with an older strand of geographical work that emphasized the spatial concentration of financial and corporate activities, namely the body of work that has become known as ‘World Cities’, ‘Global Cities’ and other deflections of similar ideas (Friedmann, 1986; Sassen, 1991; Beaverstock, et al., 1999; Faulconbridge, et al., 2007).
This paper similarly zooms in on the urban field, but uses a lens that is simultaneously more restrictive and more general. As was stated in the introduction, our take on urban inequalities as the dependent variable of this study is strongly informed by the intuition that financialization as one of the independent variables should be taken more seriously and deserves self-standing research interest. Hence, financialization is the first key concept we will discuss in this section. The second key concept is income inequality.
There is now a growing body of multidisciplinary literature on financialization. In broad terms, it denotes an array of economic developments such as the increasing importance of financial markets, financial metric and financial motives in different societal spheres and at different levels of aggregation, which add up to a structural transformation of contemporary capitalism that can be read off from metrics such as share of GDP, value added, employment, profits and revenues, size of bank balance sheets, nominal value of financial assets, etc. (Epstein, 2005).
Broadly speaking there are three strands of research identifiable in this corpus of literature (Froud, et al., 2006; Engelen, et al., 2010; Van der Zwan, 2014). The first is the firm-based approach, which views financialization as a process that primarily affects non-financial corporations through the increased use of finance-based performance metrics, stock market driven corporate strategies and through a regulatory environment that increasingly promotes the pursuit of shareholder value (Boyer, 2000; Cutler & Waine, 2001; Stockhammer, 2004; Milberg, 2008). This approach reverberates with a wider academic debate on the politics of neoliberalism, that sees the institutional transformations conducted during the neoliberal revolutions of Thatcher and Reagan – privatization, liberalization, deregulation – as the prime drivers of the process of financialization (Crotty, 2005; Harvey, 2006) .
The second approach combines insights from the French regulation school and the so-called Varieties of Capitalism literature to describe financialization as a new, encompassing ‘accumulation regime’ or ‘growth model’ (Duménil & Lévy, 2004; Foster, 2008; Orhangazi, 2008; Engelen, et al., 2010; French, et al., 2011; Krippner, 2012). Here financialization is conceptualized as the collective answer to the profitability of Fordism in the mid-1970s. The main metrics used in this approach are the standard macroeconomic ones such as GDP growth, inflation and employment figures as well as shifts in sectoral GDP contributions. The accumulation perspectives come in more Weberian (Hall & Soskice, 2001) or more Marxian (Arrighi, 1994; Boyer, 2000) inflections.
The third approach is more sociological in nature and focuses on the changes in identity and subjectivity caused by the increasing penetration of financial metrics in daily life. The topics of study, typical for this approach, range from the effects of financial literacy programs (Martin, 2002) , the performative effects of the increasing emphasis on the exchange value of real estate over its use value (Langley, 2007) and the manipulation of credit scores to keep access to consumer credit (Montgomerie, 2006) to the self manipulations of traders in order to withstand the emotional stresses of risk taking (Zailoom, 2006). In all these cases, micro-sociological data are used to describe the birth and maintenance of so-called ‘financialized subjects’.
What is missing from these three strands is a feel for the spatial unevenness of the process of financialization. Both the firm and the subjectivity-oriented strands present descriptions and explanations of their objects that are in essence universal and decontextualized in nature. The accumulation regime-approach, on the other hand, given its comparative origins, in principle contains the promise of a more place sensitive analysis, but tends to sweep intra-national differences under the carpet of national political economies conceptualized as self contained wholes, without internal (sectoral or geographical) differentiation (Agnew, 1994). As such, this paper tries to infuse the wider financialization literature with a pinch of spatial sensitivity by investigating the causal linkages between the global division of financial activities and their effects on urban inequalities, through a closer analysis of 53 IFCs.
Income inequality, our second key concept, is defined here as the uneven distribution of gains from production and labor in society and as such must be kept distinct from wealth inequality and the rentier interests derived from the ownership of property. Our focus on income inequality as the dependent variable must thus be seen as a reflection of the observation coming out of the inequality research (see above) that the rich nowadays are by and large ‘working-rich’ instead of the rentiers of an earlier phase (1850-1914) of globalization.
In global terms, the income inequality between nations, especially between developed and developing nations, has been a topic hotly debated among academics and practitioners over the past 30 years, resulting in the recognition that at the global level income inequalities have actually declined. Nevertheless, there is now a growing concern about income inequality being on the rise within nation states, especially those that belong to global North, in sharp contrast to the decreasing inequalities associated with the Fordist settlement (Kuznets, 1955) , raising pertinent questions about the causal drivers behind it (Sheehey, 1996; Atkinson, 1999; Milanovic, 2002; Piketty & Saez, 2003).
This goes together with a more disaggregated view on income inequality, with a stronger emphasis on the economic fates of different social groups. Citigroup (2005) went so far as to suggest that the world is divided into two camps: Plutonomies, whose economies are powered by and largely consumed by the wealthy few (such as the US and UK) and the rest. Plutonomy allegedly ensures an exclusive circle of wealth generation for High Net Worth Individuals, who are becoming increasingly detached from their local environments and are instead sharing more and more similar lifestyles, preferences and social and economic networks with other HNWIs of the world. Hacker and Pierson (2010) have found some evidence for such level of polarization for the US, while Useem (1983) and Moran (2008) presented evidence from the UK.
Similarly an increasing number of scholars has zoomed in on trends within top-end tail of the income distribution in different countries, i.e. the share of total income captured by the top 1 percent, the 0.1 percent or even the 0.01 percent (Piketty & Saez, 2003; Duménil & Lévy, 2004). The World Top Incomes Database, set up by Atkinson, Saez and Piketty, among others, is a case in point. Here an increasing amount of data on the income of the ‘working rich’ is made publicly available in an attempt to spawn new research on the drivers of increasing global intra-national income inequalities.
While this is a very welcome development, allowing us to shed stronger lights on the formal and informal mechanisms of distribution of economic rewards, it fails to address the spatially highly uneven articulation of these increasing income inequalities. Statistical categories (the top 10 percent) or sociological categories (classes) are not evenly distributed over space, but tend to cluster in specific places (Pow, 2011; Hay & Muller, 2012). Taking its cue from a seminal paper by Kaplan and Rauh (2010), which explicitly traces increasing income inequality in New York to the skill composition of ‘Wall Street’, this paper too aims to examine the linkages between the different skill compositions of IFCs to urban-level income inequalities.
Data and methodology
The analysis in this paper requires specific data on financial sector activities detailed down to functional specializations within the industry. Since city-level financial sector data spanning multiple regions of the world is simply not available from traditional data sources (e.g. OECD, Eurostat, IMF), we constructed our own database using data from an online social networking platform, Linkedin. Social networks are a fairly new phenomenon and their appearance in academic literature has been limited to a handful of publications of social media studies (e.g. Papacharissi, 2009). We chose Linkedin because it contains a wealth of (financial) industry knowledge that cannot be obtained from elsewhere. Since this paper is the first in its field to use a social media-sourced statistical data, our database is accordingly unique.
Linkedin is an online professional network with over 3 million company pages and 250 million individual members from 200 countries (Linkedin, 2014a). Launched in 2003, Linkedin grew into the largest and the most popular professional network on the Internet. The company and individual-specific data are proprietary in Linkedin, but we used publicly available, non-company specific aggregate data from Linkedin to create our database4. More specifically, we collected one-time, snapshot data (January 2014) on the number of finance companies registered in Linkedin, per city and differentiated by the finance sectors they belong.
Linkedin differentiates 147 industry sectors, among which we identified 6 financial sector categories: Financial services, Banking, Investment banking, Investment management, Venture capital & Private equity and Capital markets (Linkedin, 2014b). Since Linkedin does not give definitions to the industry categories it uses, we will provide our own description for the six finance sectors.
Figure 1: Linkedin-based, self-constructed data set, final composition
As we described above, conceptually there is a degree of overlap between different finance categories in Linkedin, and indeed we observe strong correlations between the sectors in our data set (Appendix Table A.4). Practically, however, the companies sorted into different sectors in Linkedin do not overlap, because every company can choose only one industry code upon registering to the network. There are companies that have different industry codes for different subsidiaries they own (such as a banking group that has commercial as well as investment banking arms). However, this does not affect the quality of the data, as these businesses run independently of each other for all administrative and legislative purposes, in fact it adds depth to it. The finance industry data we gathered are hence sectorally differentiated and mutually exclusive, which are essential conditions for our analysis.
When constructing the database, we focused on the major International Financial Centers of the world, because the size and variety of financial activities in these cities are unmatched by any others, and also this is where the sophisticated, high-risk and high-reward financial services that give rise to the name ‘high-finance’ are located. As reference for the IFCs we use “Global Financial Centres Index 14” (GFCI) report by London-based commercial think-tank Z/Yen Group (2013) which ranks the world’s 80 most prominent financial centers.
We collected city-level finance data from Linkedin for 53 out of the 80 centers. Out of the 27 cities not included in our dataset, 15 are identified as offshore financial centers by the Z/Yen Group (e.g. Channel Islands, Cayman Islands, Malta, Cyprus) and the IMF (Monaco, Dubai, Dublin, Panama). The remaining 12 financial centers are either small (e.g. Vienna, Tel Aviv, Ryadh) or with recent history of financial meltdown (e.g. Athens, Rejkyavik). The absence of these centers does not affect our analysis much, because their contribution would have been either minimal (off-shore centers) or greatly distorting (troubled economies).
There are several advantages of using Linkedin as a data source: it is up-to-date and linked to the primary source (finance industry). The sectorally differentiated industry statistics from Linkedin are fairly reliable, because companies normally would not misrepresent themselves in a professional network such as Linkedin. Our data set confirms this assertion: New York and London are leading in every finance sector category, Chicago is second only to New York in Banking and San Francisco beats every major IFC in Venture Capital & Private Equity category.
Next we collected income inequality data, having chosen GINI index as the metric for inequality. The GINI index measures inequality on the scale of 0-1, where 0 denotes perfect equality and 1 denotes perfect inequality. It was not possible to obtain city-level GINI coefficients for every city in our data set. We gathered 29 city-level GINI coefficients from various sources, ranging from National Bureaus of Statistics, UN Habitat, UNDP, World Bank reports (UN Habitat, 2009; 2011; 2013; UNDP, 2011; World Bank, 2001; 2012; Census and Statistics Department of Hong Kong, 2006; US Census Bureau, 2011; Department of Statistics Singapore, 2012) to books and academic publications (Rode & Kandt, 2011; Bolton & Breau, 2012; Miranti, et al., 2013; Wegren, 2013).
For the 24 cities for which city-level GINI data was not available, we employ country-level GINI coefficients. Out of these cities 18 belong to European Union countries, 4 are from high-income OECD countries (Norway, Japan, South Korea and New Zealand) and 2 are high-income Middle Eastern countries (Bahrain and Qatar)6. Since these countries have relatively even income distribution, the use of country-level GINI coefficients does not disadvantage our study.
Our database hence contains sectorally differentiated finance industry data from Linkedin for 53 financial centers in 35 countries and the corresponding GINI coefficients (see Appendix Table A.1 for the full list). There are, however, limitations to Linkedin data inherent in the structure of the network: geographical and industry biases. The network is popular in some regions of the world (North America, Western Europe, South East Asia and Oceania) and not so much in others (South America excluding Brazil, Africa, Eastern Europe, Central and East Asia) (Linkedin, 2014a). The popularity of Linkedin in a region appears to be determined by a number of factors: the economic success of the country/region (US, India, Brazil, UK have the largest membership rates in Linkedin), the Internet usage rate (low membership rate for China) and the region’s strength of economic and cultural ties (e.g. English language) to the global North (or more specifically to the US and the UK). Not surprisingly, therefore, Australia and New Zealand have high Linkedin memberships, whereas Japan has very low membership rate, which can only be attributed to the nature of Japanese culture both at corporate and individual levels. Our data set is undoubtedly affected by the geographical bias, but it is not clear how strong that bias is, given the fact that finance industry is also highly concentrated in those regions where Linkedin is popular. Linkedin’s other bias – the industry bias – however clearly plays to our advantage, because in terms of the number of companies registered, finance industry (the six finance sectors combined) is the third largest industry in Linkedin (4% of all companies), after Information Technology & Services (7%) and Marketing & Advertising (5%).
Before discussing the method used in this paper, let us describe how we preprocessed the data.
First we checked our data for outliers. In finance data, outliers are in fact expected and they contribute vital information to the analysis. In income inequality data, however, an outlier can distort the outcome of the analysis. In our GINI data we found one outlier – Johannesburg, South Africa – with urban income GINI coefficient of 0.75, which we removed from our analysis. In finance data, New York and London are the two biggest outliers, even though London has nearly 50% less companies than New York in every sector. New York leads in almost every category, but San Francisco beats the Big Apple in Venture Capital & Private Equity, albeit by a narrow margin.
Second, we divided our data into 7 subsets (6 finance sectors plus a new category, All Finance, which is the sum of the six). We then paired each set with the GINI index and ranked the cities in every subset according to the size of its finance sector. In effect, we created 7 hierarchical indexes for the financial centers, each in different financial sector category. Here are the top 10 cities in each sector (see Appendix Table A.1 for overall finance ranking):
Table 1: Hierarchy of financial centers based on specific finance sectors
Next we performed a simple test of relation between finance and inequality. We ran a series of simple linear regressions, one for every subset, using the GINI index as dependent variable and finance sector data as independent variable. The regression results show that there is a statistically significant (p-value≤0.05) positive linear relation between finance and income inequality for every finance sector (Appendix Table A.4). The outcome of this exercise therefore indicates that we are on the right track with our assumptions. Our next analysis however goes further and deeper by looking into the differences in inequality between various types of IFCs.
We use in this paper a statistical method known as clustering. The clustering method is a simple, yet effective tool that is particularly suited for the purpose. Clustering generally involves dividing a non-homogeneous group of observations into smaller, more homogeneous subgroups (Lattin, et al., 2003). In our case, we wanted to divide each of our financial sector data into three subgroups: LOW, MEDIUM and HIGH concentration of financial activities. As the names suggest, the low concentration subgroup would include cities with a small number of financial firms, the medium concentration subgroup would include the up-and-coming financial centers that have well above the average number of financial firms, and the high concentration subgroup would only take cities that are at the very top of the financial center hierarchy.
Our aim is to create mutually exclusive and collectively exhaustive subgroups within the sector subsets, a method that is sometimes referred to as ‘partitioning’ (Lattin, et al., 2003). Also, it is critically important for our analysis that we create more or less naturally occurring clusters, and hence we did not use statistical algorithms and software to divide the data. Instead we looked at our data, its distribution, range and variance and created clusters manually (see Appendix Table A.2 for descriptive statistics). It is a rudimentary but often highly reliable way of clustering small size, univariate data like ours. For instance, if we glance at the distribution of All Finance activities in a simple line chart (Figure 2), we can already tell how the subgroups should roughly look like.
Figure 2: The distribution of all financial activities in the IFCs
As we see, the distribution of financial activities is highly skewed, with a large right-tail and a sharp peak at the top. We can observe that it would describe the data fairly well if we divide the data around about city ranks of 4 and 15 (Figure 2). We examined the distribution of finance activities in each of the six finance sectors in a similar way and found that the same partitioning method would work for every sector. The real issue here is to develop a rule that determines the cut-off point for every subgroup exactly, a rule that can be applied to each sector equally well.
After some experimentation with the data, we determined that the arithmetic mean of the group (average number of finance companies in the sector) and the value that is approximately two standard deviations away from the mean are the most natural cut-off points. We also looked at the value that is one standard deviation away from the group mean, but that creates unnaturally crowded HIGH concentration subgroup, where less important finance centers are clustered together with the top ones. When we apply our partitioning rule (mean value plus two standard deviations) for the All Finance group shown in Figure 2, we get cut-off points corresponding to the cities ranked 3 and 14 respectively. The subgroups are then divided as follows: HIGH (cities 1-2), MEDIUM (cities 3-13) and LOW (cities 14-52) (bottom-up division). This is very close to the estimate we made above by looking at the distribution of the values, which means that our partitioning rule works particularly well. For the other six finance sectors the method is applied with the same level of success. The uniform application of the method across all sectors ensures that the results of the analysis are comparable across sectors (see Appendix Table A.3 for cluster compositions).
Next we calculated the average GINI coefficient for each cluster of cities (i.e. LOW, MEDIUM, HIGH subgroups), in every finance sector. Although GINI index is measured on ordinal scale, average values for regions or selected groups of countries are commonly reported by national and international economic agencies and are widely used in economic literature. Therefore, we base our subsequent analysis on the implicit assumption that the average GINI values are representative of the groups they are derived from. In the following section we will show the results of our analysis, followed by a discussion.
Results and discussion
In this paper we set out to analyze the effect of financialization on urban income inequality. We developed two main assumptions regarding this relation, which are: first, financialization is a driver of income inequalities, and second, different finance specializations affect income inequality differently.
Using the clustering method discussed above we obtained the following results. For the finance industry as a whole, we note that the higher the size of finance industry in the cities the higher the observed income inequality becomes. Figure 3 shows that income inequality increases as the level of finance concentration in the cities changes from LOW to MEDIUM and from MEDIUM to HIGH. This implies that the urban centers where finance industry has a strong foothold tend to have higher urban socio-economic polarizations than others, likely due to the disproportionate increases at the top-end of the income distribution.
Figure 3: Finance industry as a whole and the urban inequality
The concentration of financial activities is measured by the absolute size of the industry in this paper. If we however look at the relative sizes, that is, at the values adjusted by urban population size, we see that Amsterdam tops the list for Financial Services, New York for Investment Banking and Capital Markets, Geneva for Banking and Investment Management, and San Francisco for Venture Capital & Private Equity. We also find that, based on the relative measure, New York, Geneva, Zurich and Amsterdam are the only cities that are in the top 10 of every finance sector and hence are the ‘true’ financial centers (i.e. where finance flourishes in all of its variety, not just in a certain niche). London, Luxembourg and San Francisco come close to this, only London is low in Banking and the other two lack Capital Markets. Table 2 below compares the top 10 financial centers according to absolute and relative size of finance industry.
Table 2: Resulting financial center hierarchies using absolute and relative measures
The relative size of the finance industry is a useful measure for gauging the importance of finance in the urban economies, but not so useful for comparing cities based on it, because a city that has more financial activities than the size of its local population justifies is not always a financial center of global importance (e.g. San Francisco). But relative size can also indicate a level of financialization that does not translate into local industry, a tax haven, for instance, in cases such as Amsterdam. We will, however, proceed our analysis using the absolute sizes, because we are interested in the heavyweights in the global finance industry and their effects on income inequality.
Let us now turn our attention to the individual sectors in finance industry. We proposed that these sectors would have different effects on income inequality. In particular, we predicted that the high-risk and high-reward activities which are collectively known as ‘high-finance’ (Investment Banking, Investment Management, Capital Markets and Venture Capital & Private Equity sectors included here), would produce higher surge in income inequalities than the traditional finance activities that are Financial Services and Banking. This appears to be the case, as shown in Figure 4. Although we observe across all sectors an upward movement of GINI as financial activities intensify, the patterns of increase in inequality are different for every finance sector. We believe that these patterns tell volumes about the nature of the finance industry’s involvement in growing urban inequalities.
Figure 4: Financial sectors and the urban inequality
For instance, we see that for Financial Services the rise of GINI is steady and evenly paced. The increase in GINI in this case is likely to be produced not by Financial Services sector itself, but by the growth of industries whose activities it supports, in particular the other highly-specialized finance sectors. Banking, on the other hand, appears to have almost no impact on income inequalities. The surge of GINI observed at HIGH level of Banking activities can reasonably be attributed to the same causes as in Financial Services, because the top 2 banking centers in our data set (New York and Chicago) have equally high concentrations of other (higher level) financial activities.
If we look at the other four sectors, which we identified as belonging to ‘high-finance’, we see marked difference from the previous two sectors in terms of their pattern of GINI increase. For these sectors, GINI increases by 0.06-0.09 points as financial activities increase from LOW to MEDIUM intensity, compared to only 0.03 point increase in the case of Financial Services. Investment Management and Venture Capital & Private Equity sectors showed the highest increase of GINI at the MEDIUM level of concentration. It is in line with our expectations, because these sectors include wealthy hedge fund owners, venture capitalists and mutual fund managers, all of whom are in the top 1% of the income distribution, or even at the top 0.1% and higher. This result thus corroborates our early claim that finance industry contributes to urban inequalities primarily through its involvement in the unrelenting and extraordinary increases of top incomes in the world.
We place particular importance on the GINI differences between LOW and MEDIUM concentration subgroups, because essentially what we are comparing here is the cities with little or no amount of high-level financial sector presence and the cities that have abundance in high-level financial activities. In other words, have’s and have-nots of the financial world. Therefore, a significant difference in average GINI between ‘have’ and ‘have-not’ cities is a telltale sign that the surge of high-finance activities in the cities brings with it a rapid development at the top end of income distribution.
The HIGH concentration of the sophisticated financial activities, however, does not seem to contribute to urban inequalities by much more than the MEDIUM concentration of the same activities. For Investment Banking and Capital Markets, the effect of finance on inequality tapers off after the sectors reach MEDIUM concentration, and for Investment Banking, the inequality even reduces slightly between the MEDIUM and HIGH subgroups. In our opinion, it implies in no uncertain terms that ‘the damage is already done’ when the high-finance sectors establish themselves in the urban centers en mass (MEDIUM group is that level, at which cities have greater than global average number of highly specialized finance companies). For Venture Capital & Private Equity the rise in GINI is maintained at the HIGH level of concentration, but as there are only two cities at this level the observation is not that informative.
Finally, the results of our analysis have provided evidence in support of both of our assumptions. First, finance industry as a whole is found to be linked to income inequalities in the cities, although without further information we are unable to say to what extent this happens. Second, the so-called ‘high-finance’ activities such as Investment Management are indeed shown to have greater impact on urban income inequalities than the traditional finance activities such as Banking. The evidence also hints at the possibility that the more sophisticated a finance activity becomes, the higher effect it has on income inequality. This goes parallel to our earlier observation that the deeper the specialization of a sector, the more geographically concentrated its activities become. A good example for this is hedge fund management, where only two main players exist: New York, which manages over 41% of all global hedge fund assets and London, which is in the second place with over 18% global hedge fund assets under its management (85% of the European hedge fund assets) (TheCityUK, 2013).
In the next section we summarize our findings and outline a direction for further research.
Since early 1980s income inequality has been on the rise globally, the world’s rich have become progressively richer, while the circumstances of the poor have improved only marginally. Economic literature identifies a number of contributing factors to this development, including globalization, technological change and financialization. In this paper we aim to examine one of these explanations, namely financialization, and systematically and empirically analyze its relation to rising income inequality. This paper offers a distinctly urban narrative on the subject, as we believe that financialization is a spatially highly concentrated phenomenon, especially at the top end of its calibration.
Using a unique, self-constructed database, we explored the link between finance industry and urban income inequalities, guided by two assumptions. First, we argued that financialization is a driver of income inequalities, and we found evidence supporting this premise. Second, we posited that functionally diverse finance specializations have uneven effect on the development of inequalities in the urban centers. The results of our analysis also corroborated this claim. Especially for the higher order financial activities (i.e. ‘high-finance’) the effect on income distribution was found to be more pronounced than in the case of more mainstream financial activities.
This study hence contributes to the economic geography literature by providing the first detailed study into the differences between the global financial centers based on their specializations, and a varied effect their functional differences produce on the local socio-economic order. However, due to the limited scope of our analysis and data constraints, we were not able to establish beyond reasonable doubt the causal nature of relation between financialization and urban income inequality. Nor could we offer insight into the channels through which finance affects income inequality, namely the direct (through finance sector remunerations) and indirect (through ensuring the growth of income of the world’s rich) ones. A further study with more in-depth view of the causes of (urban) income inequalities, aided with detailed statistics on finance sector remunerations, we believe, will shed more light on these subjects.
Table A.1: The data set
Legend: GFCI14 rank – “Global Financial Centres Index 14” rank (Z/Yen Group, 2013); GINI – Gini coefficients; All Fin. – All Finance; FS – Financial Services; Bank – Banking, IB – Investment Banking; IM – Investment Management; CM – Capital Markets; VCPE – Venture Capital and Private Equity.
Table A.2: Descriptive statistics of the data set
Table A.3: Number of cities in each cluster
Table A.4: Regression results and correlations
Agnew, J. 1994. The Territorial Trap: The Geographical Assumptions of International Relations Theory. Review of International Political Economy, 1(1): 53-80.
Arrighi, G. 1994. The Long Twentieth Century: Money, Power and the Origins of Our Times. New York: Verso.
Atkinson, A. B. 1999. The Distribution of Income in the UK and OECD Countries in the Twentieth Century. Oxford Review of Economic Policy, 15(4): 56-75.
Bakija, J., Cole, A. & Heim, B. T. 2010. Jobs and Income Growth of Top Earners and the Causes of Changing Income Inequality: Evidence from U.S. Tax Return Data. Williams College Department of Economics Working Papers, Williamstown, MA.
Beaverstock, J. V., Smith, R. G. & Taylor, P. J. 1999. A Roster of World Cities. Cities, 16(6): 445-458.
Bolton, K. & Breau, S. 2012. Growing Unequal? Changes in the Distribution of Earnings across Canadian Cities. Urban Studies, 49(6): 1377-1396.
Boyer, R. 2000. Is a Finance-led Growth Regime a Viable Alternative to Fordism? A Preliminary Analysis. Economy and Society, 29(1): 111-145.
Capgemini/Merrill Lynch. 2011. World Wealth Report 2011. Available at: <http://www.ml.com/>
Cassis, Y. 2010. Capitals of Capital : A History of International Financial Centres, 1780-2009. Cambridge: Cambridge University Press.
Census and Statistics Department of Hong Kong. 2006. Population By-census. Available at: <http://www.bycensus2006.gov.hk/>
Citigroup. 2005. Plutonomy: Buying Luxury, Explaining Global Imbalances. Available at: <http://www.slideshare.net/tlyon415/>
Crotty, J. 2005. The Neoliberal Paradox: The Impact of Destructive Product Market Competition and ‘Modern’ Financial Markets on Nonfinancial Corporation Performance in the Neoliberal Era. In: G. Epstein, ed. Financialization and the World Economy. Northampton, MA: Edward Elgar, pp. 77-110.
Cutler, T. & Waine, B. 2001. Social Insecurity and the Retreat from Social Democracy: Occupational Welfare in the Long Boom and Financialization. Review of International Political Economy, 8(1): 96-118.
Department of Statistics Singapore. 2012. Household Income and Expenditure Survey. Available at: <http://www.singstat.gov.sg/>
Duménil, G. & Lévy, D. 2004. Capital Resurgent. Cambridge, MA: Harvard University Press.
Engelen, E., Erturk, I., Froud, J., Leaver, A. & Williams, K. 2010. Reconceptualizing Financial Innovation: Frame, Conjuncture and Bricolage. Economy and Society, 39(1): 33-63.
Engelen, E. & Konings, M. 2010. Financial Capitalism Resurgent: Comparative Institutionalism and the Challenges of Financialization. In: G. Morgan, J. Campbell, C. Crouch, O.K. Pedersen & R. Whitley, eds. The Oxford Handbook of Comparative Institutional Analysis The Oxford Handbook of Comparative Institutional Analysis [online]. Available at: <http://www.oxfordhandbooks.com/>
Engelen, E., Konings, M. & Fernandez, R. 2010. Geographies of Financialization in Disarray: The Dutch Case in Comparative Perspective. Economic Geography, 86(1): 53-73.
Epstein, G. A. ed. 2005. Financialization And The World Economy. Northampton, MA: Edward Elgar.
Faulconbridge, J., Engelen, E., Hoyler, M. & Beaverstock, J. 2007. Analysing the Changing Landscape of European Financial Centres: The Role of Financial Products and the Case of Amsterdam. Growth and Change, 38(2): 279-303.
Foster, J. B. 2008. The Financialization of Capital and the Crisis. Monthly Review: An Independent Socialist Magazine, 59(11): 1-19.
Frank, R. H., & Cook, P. J. 1996. The Winner-Take-All Society: Why the Few at the Top Get So Much More Than the Rest of Us. New York: Penguin Books.
Freeman, R. B. 2010. It's financialization! International Labour Review, 149(2): 163-183.
French, S., Leyshon, A. & Wainwright, T. 2011. Financializing space, spacing financialization. Progress in Human Geography, 35(6): 798-819.
Friedmann, J. 1986. The world city hypothesis. Development and Change, 17(1): 69-83.
Froud, J., Johal, S., Leaver, A. & Williams, K. 2006. Financialization and Strategy. London: Routledge.
FuturesIndustry. 2013. FIA Annual Volume Survey. Available at: <http://www.futuresindustry.org/>
Hacker, J. S. & Pierson, P. 2010. Winner-Take-All Politics: Public Policy, Political Organization, and the Precipitous Rise of Top Incomes in the United States. Politics & Society, 38(2): 152-204.
Hall, P. A. & Soskice, D. eds. 2001. Varieties of Capitalism: The Institutional Foundations of Comparative Advantage. Oxford: Oxford University Press.
Harvey, D. 2006. Neo-Liberalism as Creative Destruction. Geografiska Annaler: Series B, Human Geography, 88(2): 145-158.
Hay, I. & Muller, S. 2012. That Tiny, Stratospheric Apex That Owns Most of the World: Exploring Geographies of the Super-Rich. Geographical Research, 50(1): 75-88.
Jaumotte, F., Lall, S. & Papageorgiou, C. 2013. Rising Income Inequality: Technology, or Trade and Financial Globalization? IMF Economic Review, 61(2): 271-309. Available at: <http://www.imf.org/>
Kaplan, S. N. & Rauh, J. 2010. Wall Street and Main Street: What Contributes to the Rise in the Highest Incomes? The Review of Financial Studies, 23(3): 1004-1050.
Keynes, J. M. 1923. A Tract on Monetary Reform. London: MacMillan.
Krippner, G. 2012. Capitalizing on Crisis: The Political Origins of the Rise of Finance. Harvard: Harvard University Press.
Kuznets, S.1955. Economic Growth and Income Inequality. The American Economic Review, 45(1): 1-28.
Langley, P. 2007. The Uncertain Subjects of Anglo-American Financialization. Cultural Critique, 65: 66-91.
Lapavitsas, C. 2011. Theorizing Financialization. Work, Employment & Society 25(4): 611-626.
Lattin, J., Carroll, J. D. & Green, P. E. 2003. Analyzing multivariate data. Belmont, CA: Brooks/Cole.
Linkedin. 2014a. Company Information. Available at: <http://press.linkedin.com/>
Linkedin. 2014b. Industry Codes. Available at: <http://developer.linkedin.com/>
Martin, R. 2002. Financialization of Daily Life (Labor in Crisis series). Philadelphia, PA: Temple University Press.
Milanovic, B. 2002a. Can We Discern the Effect of Globalization on Income Distribution? Evidence From Household Budget Surveys. World Bank Policy Research Working Paper, no. 2876. Available at: < http://elibrary.worldbank.org/>
Milanovic, B. 2002b. True World Income Distribution 1988 and 1993: First Calculation Based on Household Surveys Alone. Economic Journal, 112(476): 51-92.
Milberg, W. 2008. Shifting Sources and Uses of Profits: Sustaining US Financialization with Global Value Chains. Economy and Society, 37(3): 420-451.
Miranti, R., Cassells, R., Vidyattama, Y. & McNamara, J. 2013. Measuring Inequality for Small Areas Using Spatial Microsimulation: An Australian Case Study. Conference proceedings of the 42nd Australian Conference of Economists, 7-10 July 2013. Perth, Western Australia: Murdoch University. Available at: <http://www.murdoch.edu.au/>
Montgomerie, J. 2006. The Financialization of the American Credit Card Industry. Competition & Change, 10(3): 301-319.
Moran, M. 2008. Representing the Corporate Elite in Britain: Capitalist Solidarity and Capitalist Legitimacy. The Sociological Review, 56: 64-79.
Orhangazi, O. 2008. Financialisation and Capital Accumulation in the Non-Financial Corporate Sector: A Theoretical and Empirical Investigation on the US Economy: 1973-2003. Cambridge Journal of Economics, 32(6): 863-886.
Oxfam. 2014. Working for the Few. Available at: <http://www.oxfam.org/>
Papacharissi, Z. 2009. The Virtual Geographies of Social Networks: A Comparative Analysis of Facebook, Linkedin and AsmallWorld. New Media & Society, 11(1-2): 199-220.
Pike, A. & Pollard, J. 2010. Economic Geographies of Financialization. Economic Geography, 86(1): 29-51.
Piketty, T. & Saez, E. 2004. Income Inequality in the United States, 1913-2002. UC Berkeley Working Paper.
Piketty, T. & Saez, E. 2003. Income Inequality in the United States, 1913-1998. Quarterly Journal of Economics, 118(1): 1-39.
Pow, C. 2011. Living It Up: Super-Rich Enclave and Transnational Elite Urbanism in Singapore. Geoforum, 42(3): 382-393.
Private Equity International. 2014. The 2013 PEI 300 (Top 50) [online]. Available at: <http://www.peimedia.com/>
Rode, P. & Kandt, J. (2011) Inequality in Transport Accessibility: Sao Paulo, Istanbul and Mumbai. LSE Cities/Urban Age Programme Presentation Material. Transport Studies Unit, Oxford University. Available at: <http://www.tsu.ox.ac.uk/>
Sassen, S. (1991) The Global City: New York, London, Tokyo, 1st ed. Princeton, NJ: Princeton University Press.
Scott, A. J. 2007. Capitalism and Urbanization in a New Key? The Cognitive-Cultural Dimension. Social Forces, 85(4): 1465-1482.
Sheehey, E. J. 1996. The Growing Gap Between Rich and Poor Countries: A Proposed Explanation. World Development, 24(8): 1379-1384.
Stockhammer, E. 2004. Financialisation and the Slowdown of Accumulation. Cambridge Journal of Economics, 28(5): 719-741.
UN Habitat. 2009. State of the world's cities 2008/9. Available at: <http://www.unhabitat.org/>
UN Habitat. 2011. Global Urban Indicators Database. Available at: <http://www.devinfo.info/>
UN Habitat. 2013. State of the world's cities 2012/13. Available at: <http://www.unhabitat.org/>
UNDP. 2011. UN Human Development Report 2011. Available at: <http://www.undp.org/>
US Census Bureau. 2011. U.S. Neighborhood Income Inequality in the 2005–2009 Period. Available at: <http://www.census.gov/>
US Census Bureau. 2014. North American Industry Classification System. Available at: <https://www.census.gov/>
Useem, M. 1983. Business and Politics in the United States and United Kingdom: The Origins of Heightened Political Activity of Large Corporations during the 1970s and Early 1980s. Theory and Society, 12(3): 281-308.
Van der Zwan, N. 2014. Making Sense of Financialization. Socio-Economic Review, 12(1): 99-129.
Veblen, T. 1899. The Theory of the Leisure Class: An Economic Study of Institutions. New York: MacMillan.
Wegren, S. K. 2013. Rural Inequality in Divided Russia. New York: Routledge.
World Bank. 2001. Dimensions of Urban Poverty in the Europe and Central Asia Region. Available at: <http://elibrary.worldbank.org/>
World Bank. 2012. World Development Indicators 2012. Available at: <http://data.worldbank.org/>
Z/Yen Group. 2013. Global Financial Centres Index 14 [September 2013]. Available at: <http://www.longfinance.net/>
Zailoom, C. 2006. Out of the Pits: Traders and Technology from Chicago to London. Chicago: University of Chicago Press.
Zalewski, D. A. & Whalen, C. J. 2010. Financialization and Income Inequality: A Post Keynesian Institutionalist Analysis. Journal of Economic Issues, 44(3): 757-777.
1. Wall Street banks have handed out US$20.8 billion bonuses in 2010 to their employees. The top paid City banker in the UK was awarded £14 million in 2010, which is over 1200 times higher than the lowest paid City staff (High Pay Commission UK, 2012).
2. HNWIs are defined as those having investable assets of US$1 million or more, excluding primary residence, collectibles, consumables, and consumer durables (Capgemini/Merrill Lynch, 2011).
3.The distinction between bank-based and market-based is of course derived from Gershenkrons 1965 classic. See for a more recent reworking Hall & Soskice (2001), where the same distinction is rephrased as Liberal Market Economies (LMEs) versus Coordinated Market Economies (CMEs).
4. The data we gathered is neither uniquely identifiable nor is retraceable to any individual company. The resulting data set is used exclusively for the purposes of academic research.
5. Although Chicago is not only big in Banking - the city is also home to world’s number one Derivatives Exchange (Chicago Mercantile Exchange) (FuturesIndustry, 2013) - its banking sector is singled out here because it is second only to New York in our database.
6. The Global Financial Centres Index of Z/Yen Group lists a number of countries among the IFCs, such as Bahrain, Qatar, Malta, Cyprus, Mauritius. Unlike the rest, Bahrain and Qatar are not offshore financial centers, but the level of financial activities is very low in these centers; there are mainly some investment management firms for managing the oil-wealth of the region and some banking and financial services auxiliary to the former.