The influence of heterogeneous institutional investors on company strategies in emerging countries: Evidence from Brazil

Abstract

The objective of this study is to evaluate the impact of institutional investor shareholding on strategic corporate management. Focusing on Brazilian non-financial corporations, we consider the heterogeneity of institutional investors – based on their main characteristics and investment profile –, in the context of Brazilian institutional framework, to explore the dynamics between institutional ownership and corporate strategies. Applying principal component and cluster analysis, we identified four distinct groups of institutional investors. The results from our dynamic panel data model indicate that institutional investors do not impact dividend policy but do have a positive influence over financial investments. While neither short-term investors nor those with a longer term view were found to have any influence on the financialization of corporate strategies, we found that presence of specialized investors characterized by intermediate shareholder positions – primarily independent and national private asset managers – are associated with greater financial investments, but lower dividend payments.

Keywords

Financialization institutional investors corporate strategies emerging markets

Introduction

The structural correlation between productive and financial wealth has given rise to a new model of corporate governance (CG) with direct consequences on the regime of accumulation (Kliman and Williams, 2015). In this context, the rentier class, formed by agents who acquire capital gains, interest and dividend payments, becomes increasingly powerful from both a political and economic perspective (Stockhammer, 2004). Due to their high concentration of capital, these agents – particularly institutional investors – have been exerting considerable influence on economies. In particular, institutional investors have acquired an increasing role in the corporate decision-making process due to their presence in company ownership structure (Aglietta, 2000; Crotty, 2002; Guttmann, 2008; Orhangazi, 2008; Stockhammer, 2004).

Understanding the effects of institutional ownership on corporate strategies is of great relevance. Financialized dynamics, possibly promoted by these agents, affects the productive investment, competitiveness and stability not only of corporations but also of overall economies (Bortz and Kaltenbrunner, 2018; Davis, 2017; Kaltenbrunner and Painceira, 2015, 2018; Tori and Onaran, 2018a, 2018b). Previous studies have found evidence that institutional investors have a relevant effect on the strategies of non-financial corporations (NFC). In addition, studies have indicated that this influence varies according to the characteristics of these agents and the domestic institutional environment (Becker et al., 2010; Brossard et al., 2013; Bushee, 2004; Çelik and Isaksson, 2013; Chen et al., 2007; Crane et al., 2016; David et al., 2001; Ferreira and Matos, 2008; Karwowski and Stockhammer, 2017; Katan and Mat Nor, 2015).

In addition to focusing on U.S. and European markets, studies regarding the heterogeneity of institutional investors have considered general characteristics to form investor groups and evaluate the relationship between institutional ownership and corporate strategies. In this context, the influence of institutional investors on corporate decisions in emerging market economies (EME) remains to be further investigated, and the question of how a particular institutional framework and the heterogeneity of these agents could affect this dynamic must be further explored.

The economic environment in Brazil, in particular, has been characterized since the 1980s by high uncertainty. Consequently, the public debt market has presented investment options based on safety, liquidity and high real interest rates, motivating portfolio allocation in Treasury bonds and negatively impacting the development of Brazilian capital markets (Bruno and Caffé, 2017; Miranda et al., 2017). Specifically, the presence of high risk-free rates of return can reduce the traditional active role of institutional investors in a company’s decision-making process and the presence of such rates therefore favours financial investments over productive investments. In addition, the influence of international investors on corporate decisions may be limited, considering the underdeveloped nature of Brazilian capital markets and the small portion of Brazilian stocks in the portfolios of international investors, since international investors likely follow a portfolio diversification strategy.

In this context, the objective of this article is to evaluate the impact of institutional investor ownership on the strategic management of Brazilian NFC between 2010 and 2016. The analysis focuses on a specific country and considers domestic institutional aspects to verify whether the above-mentioned impact leads to the financialization of NFC. Moreover, using a more sophisticated method of classifying and clustering institutional investors, we define groups of institutions based on their main characteristics and investment profiles, to better incorporate investor heterogeneity in our empirical analysis. We hypothesize that the influence exerted by institutional investors on corporate strategies is not homogeneous, depending both on the investment profile, characteristics and motivations of each investor, and on the institutional framework of the country in which the investor operates.

This study provides a comprehensive analysis of how institutional investors have influenced the decision-making of NFC. The results offer useful insights for policymakers, executives and stakeholders, given the growing importance of institutional investors in financial markets. The findings also provide meaningful points for academic discussion, with respect to the impact of institutional investor portfolio allocation on corporate finance in EME. In particular, this study sheds more light on the discussion of the financialization process, providing new empirical evidence of the potential heterogeneous influence of distinct institutional investor groups on company strategies in an important emerging market economy.

Literature review

The concept of financialization has been widely used to refer to a finance-dominated accumulation regime, which is characterized by a structural intertwining of productive and finance wealth, with the latter overriding the former (Krippner, 2005; Stockhammer, 2008; Tori and Onaran, 2018a). The reduction of the role of the state in economies, the increased international capital flows, and the rise of financial operations all reflect the financial hegemony that followed a period of intense regulation during the Golden Age (Clark, 2000; Crotty, 2002; Duménil and Lévy, 2011; Epstein, 2005).

In this historical context, corporate management has adopted the ‘portfolio view of the firm’. Greater financial investments – in conjunction with increasing transfers in the form of interest, dividends and stock repurchases – have indicated a shift from a ‘retain-and-reinvest’ resource-allocation criterion to a ‘downsize-and-distribute’ criterion (Barradas and Lagoa, 2017; Davis, 2009; Tori and Onaran, 2018b). The ‘maximizing shareholder value’ principle was then reaffirmed as the main instrument of CG, resulting in an increasing importance of shareholders over the stakeholders of other companies (Lazonick and O’Sullivan, 2000; Rabinovich, 2019; Tori and Onaran, 2018a).

Institutional investors play a prominent role in this corporate dynamic. Large shareholdings provide them with influence over the decision-making process. In addition, since institutional investors hold large positions in company ownership structure, which creates illiquidity exposure, the high costs of exiting their positions can provide incentives for shareholder activism (Fichtner, 2013). Moreover, institutional investors may exert external control regardless of the magnitude of their ownership, given the high amount of capital they manage in capital markets.

Several empirical studies have recently investigated the influence exerted by institutional investors on corporate management, and suggest a greater investor role in managerial control. On one hand, the presence of institutional investors in company ownership structure could be associated with a proactive role, with investors outlining strategies to ensure the company’s sustainability and growth (Crane et al., 2016; David et al., 2001; Ferreira and Matos, 2008). On the other hand, these agents could exert ‘myopic’ pressures on management priorities to achieve short-term returns (Aglietta, 2000; Brossard et al., 2013; Bushee, 1998, 2001; Chen et al., 2007; Crotty, 2002; Fichtner, 2013; Guttmann, 2008; Orhangazi, 2008; Stockhammer, 2004).

Recent studies have also explored the heterogeneity of institutional investors and their distinct impact on corporate strategic management. For example, Chen et al. (2007) examined the behaviour of institutional holdings, using a sample of 1815 completed acquisitions announced in the U.S. between 1984 and 2011. The authors found that independent and long-term-oriented institutional investors focused on monitoring activities, rather than short-term trading. Using various categories and characteristics of institutional investors, Ferreira and Matos (2008) investigated the role of institutional investors in 27 countries. Their findings suggested that the presence of foreign and independent institutions in the ownership structure of the companies enhanced shareholder value. Brossard et al. (2013) also contributed to this debate, finding a positive effect of institutional investors on innovative activity in European companies. Focusing their analysis on ‘impatient’ institutional investors, their results showed a negative influence of this group on R&D spending.

Furthermore, differentiating institutional investors into two groups, transient and dedicated owners, Katan and Mat Nor (2015) verified that, in general, the presence of institutional investors in the ownership structure of Malaysian companies had no effect on firm performance.¹ However, when the authors specifically analysed the transient ownership group, the results showed a significant influence on firm performance. Finally, Muniandy et al. (2016) evaluated the relationship between company performance and the presence of institutional investors in Australia. The authors found that the presence of ‘pressure-resistant’ investors improved the short-term performance of companies, while the influence of ‘pressure-sensitive’ institutional investors was less clear. In turn, nominee and trustee investors played an important direct or indirect monitoring role and, in consequence, created long-term firm value.²

In general, the hypothesis of the aforementioned studies maintains that since institutional investors do not form a homogenous group, their influence may also vary. Certain studies have used legal type classifications³ to group these agents (Bushee, 2004; Katan and Mat Nor, 2015), while other studies have considered other factors, including the degree of investor independence,⁴ investor country of origin and whether the investor is private or public (Brossard et al., 2013; Chen et al., 2007; Ferreira and Matos, 2008). However, several study approaches do not necessarily incorporate background information into the behaviour of institutional investors and disregard significant intra-group variation associated with the size and stability of ownership, sensitivity to current returns, and other factors associated with investment profile (Bushee, 1998, 2001; Crane et al., 2016). Moreover, most studies do not consider various organizational forms that include either independent institutions, large corporation subsidiaries or financial conglomerates (Çelik and Isaksson, 2013). All of these factors have a direct impact on the investment profile of an institutional investor and change the direction and degree of investor influence on corporate strategies (Bushee, 2004).

The influence of institutional investors on corporate management also varies according to the institutional framework in which the firm operates. To explore this issue in EME, more structural aspects related to the subordinated position of EME within the international financial system should be considered (Bortz and Kaltenbrunner, 2018; Kaltenbrunner and Painceira, 2015, 2018). For instance, when addressing the case of emerging markets, it is crucial to consider not only the role of interest and exchange rates but also the role of the government bond market and its relationship with capital markets and with the corporate and financial pattern of NFC (Becker et al., 2010; Bin, 2016; Black et al., 2008; Bonizzi, 2017; Demir, 2007, 2009).

In Brazil specifically, on one hand, stock markets are poorly used as a source of financing, which restricts shareholder influence on corporate decisions. On the other hand, the availability of Treasury bonds with high risk-free rates of return could reduce the interest of institutional investors in exerting their expected influence. In short, the underdeveloped nature of capital markets in Brazil – dominated by a well-developed public debt market – may alter the pattern of corporate funding and governance in the country (Bruno and Caffé, 2017; Miranda et al., 2015, 2017). This differentiated pattern thus calls for a revision of the hypothesis in the literature (which is almost exclusively based on developed economies, and particularly on that of the U.S.), in which institutional investors tend to have a short-term view and pressure companies for short-term returns.

Moreover, despite having the capacity to influence the management of investee companies, institutional investors may allocate their own resources to financial investments. Therefore, when investing in Brazilian companies, foreign investors may be almost exclusively interested in diversifying their portfolios and demonstrate a reduced interest in playing an active role in the company’s decision-making process (Kaltenbrunner and Painceira, 2015). Even in cases where the investor – foreign or national – is inclined to exert a close monitoring effort, the influence may be directed to financial investments, not necessarily due to a search for short-term returns, but instead to the investor’s preference for investments with a beneficial risk/return relationship for the company itself (Demir, 2007, 2009; Katan and Mat Nor, 2015).

Karwowski and Stockhammer (2017) highlighted that despite an increasing number of studies focusing on the financialization of emerging countries, no systematic comparison across studies existed. The authors found significant variation in the financialization experience of each country and underlined the importance of more detailed case studies. For instance, using firm-level data, Demir (2007, 2009) found evidence that for Argentine, Mexican and Turkish firms, the differential return rate between financial and non-financial investment significantly affected the investment strategy of the NFC. The increasing rate of return gap that favours financial investments, in conjunction with the high risks involved in long-term investments, justified a shift of NFC away from real investments toward more short-term financial investments, which thereby reduced capital accumulation.

Bonizzi (2017) developed an empirical study to identify the determinants of capital allocation by institutional investors in EME. The author considers the role played by foreign exchange reserves and the balance sheet of the investors themselves. The results of the study indicated that the portfolio allocation by institutional investors had a procyclical characteristic, regardless of whether it assumed a longer term profile. This result was a function of decision-making being subject to factors that are sensitive to the economic cycle (e.g. exchange reserves, which have a positive impact on portfolio allocation), and a function of the need for higher returns to strengthen the balance sheets of institutions. However, the results did not find the returns on safe assets to be a determinant of the demand for EME assets. Nevertheless, according to the author, as the returns on safe assets do affect the balance sheets of institutions, they may have an indirect negative effect on demand.

Overall, several studies have found evidence that the investor influence on corporate strategies in EME varies according to the institutional ownership by distinct groups of investors. This study contributes to the debate by further exploring the role of domestic institutional framework and institutional investor heterogeneity on an important emerging economy, Brazil.

Methods

Econometric model

The allocation of corporate resources to financial assets, in addition to their distribution in the form of dividend payments, is considered evidence of the financialization of NFC (Lazonick and O’Sullivan, 2000). As a consequence, it has established a reciprocal relationship between companies and financial markets – NFC would allocate increasing amounts of their revenues to financial investments, and the financial market would exert a greater influence over company management. A greater share of financial investment in the total assets of corporations would increase the revenue from financial sources, resulting in increasing transfers of income flows to shareholders (Orhangazi, 2008; Stockhammer, 2004).

In this context, we defined the following dependent variables: dividend payments and financial investments, both as a proportion of total assets. We used the generalized method of moments (GMM) – developed by Arellano and Bond (1991) – to estimate the models. GMM is considered the most adequate method for treating endogeneity when strictly exogenous instruments are not available for all pre-determined or endogenous regressors (Roodman, 2009). Equations (1) and (2) specify the two models

DIVI D_{it} = α DIVI D_{it - 1} + β_{0} + β_{1}^{T} x_{1 i t} + β_{2}^{T} x_{2 i t} + a_{i} + μ_{it}

(1)

F I_{it} = αA F_{it - 1} + β_{0} + β_{1}^{T} x_{1 i t} + β_{2}^{T} x_{2 i t} + a_{i} + μ_{it}

(2)

where i represents the firm and t the time period. The dependent variables

DIVI D_{it}

and FI_it represent, respectively, the share of dividend payments and the share of financial investments in total assets. The intercept is represented by

β_{0}

, and

β_{1}^{T}

and

β_{2}^{T}

correspond to the transposed vectors of parameters. The vector of regressors, represented by x_it, is divided into two subvectors:

x_{1 i t}

is composed of endogenous or predetermined regressors and

x_{2 i t}

is composed of exogenous regressors. Finally,

a_{i}

represents the random effects and

μ_{it}

is the error term, with

E [μ_{it}] = E [a_{i}] = 0

According to Arellano and Bover (1995) and Blundell and Bond (1998), the difference in GMM, while asymptotically valid, may generate inaccurate and biased estimates (even if consistent) in finite samples. This would occur whenever the endogenous explanatory variables are too persistent in time and therefore have little correlation with the first differences – the latter becoming weak instruments for the former. To mitigate this problem, Blundell and Bond (1998) developed the important methodological extension, the System GMM, which we used in this article. The System GMM simultaneously estimates the equations in difference and in level, using the lags of the endogenous variables in difference as instruments.

We have also applied Windmeijer’s finite sample correction to ensure the efficiency of the two-step estimation by preventing downward-biased standard errors. In addition, to assess whether autocorrelation exists between the instruments and the endogenous regressors, but not between the instruments and the error terms, we applied the Hansen test of overidentifying restrictions and the test for autocorrelation suggested by Arellano and Bond (1991).

Data and explanatory variables

The study sample was based on 261 Brazilian publicly traded companies, using the period of 2010 to 2016. We excluded companies with negative net equity and/or without five consecutive observations for the dependent variables.⁵ Therefore, the sample for the dividend model was composed of a total of 216 companies, while the financial investment model consisted of 223 companies.

To capture the influence of institutional investors on investment and payout policies, we considered the following control variables: company size, return on asset (ROA), growth opportunity (GO), productive investment, leverage (LEV), liquidity, CG and sector of activity. Data on the companies were obtained from the Economatica database.

Next, we present both our reasoning for using the selected control variables and their respective hypotheses:

Size (SIZE): Larger companies are more capable of overcoming capital market information asymmetries (Fazzari et al., 1988). In addition, larger companies are more likely to have a more diversified and complex organizational and financial structure (Brossard et al., 2013; Orhangazi, 2008). Thus, we expect larger companies to have a higher share of financial investments in total assets, and an increased free float and liquidity in their equities. Consequently, they would be subject to larger market control, which in turn would lead to larger transfers in the form of dividend payments.

ROA: Managers decide whether to reinvest or distribute profits based on the net income of the company. More profitable companies are expected to either allocate more resources to financial investments or to pay more dividends (Gill et al., 2010; Mehta, 2012; Martins and Famá, 2012). As financial investments can also impact interest expense and therefore net income, we define the ROA variable as non-exogenous in the FI model.

GO: The agency theory predicts that companies in the maturity stage are subject to managerial discretion and therefore to overinvestment (Jensen, 1986). As a result, we expect that firms with reduced GOs (i.e. with lower sales growth rates) have more incentive to distribute dividends to signal that the strategies of the firm are consistent with the expectation of its investors. However, firms with greater GOs generally commit less of their capital to financial investments in order to increase production, and they therefore distribute fewer dividends in order to reinvest (Fazzari et al., 1988).

Capital expenditure (CAPEX): According to the financialization literature, the allocation of resources in financial assets and the payment of dividends limit the amount of resources available for productive investment and vice versa (Duménil and Lévy, 2011; Orhangazi, 2008; Stockhammer, 2004). Thus, we expect that the greater the share of CAPEX in total assets, the lower the financial investments and dividend payments. Considering the expected feedback relationship between this variable and the dependent variables, the former is defined as non-exogenous in both models.

LEV: According to Kliman and Williams (2015), a firm can borrow to make new investments and/or pay more dividends. Therefore, we also define this variable as non-exogenous in both models. With respect to dividend payments, we must consider both the possibility of a constraint in future cash flow and the signalling effect that the literature on finance associates with the level of LEV (Martins and Famá, 2012). The signalling effect refers to the fact that a rise in LEV of the company may signal a larger future cash flow and motivate managers to pay more dividends in the next period (Jensen et al., 1992). As a result, the level of LEV is lagged by one period in the DIVID model. We expect a positive relationship between LEV and financial investment. The relationship between LEV and dividend payments, however, is undefined.

Current ratio (CR): This ratio indicates the level of liquidity of the company. As financial investments compose a portion of current assets, this variable is not used in the FI model. On one hand, in our DIVID model, we expect more liquid firms to have better conditions for preserving or increasing dividend distribution (Acharya and Viswanathan, 2011). On the other hand, firms can choose to distribute less dividends to ensure higher liquidity. Therefore, this variable is defined as non-exogenous in the DIVID model.

CG: Firms with better CG are committed to a series of practices, such as issuing only common shares, ensuring more transparency, and maintaining a minimum free float.⁶ These firms are consequently expected to pay more dividends due to their commitment to maximizing shareholder value, since according to the agency theory, the greater the cash flow under a manager’s discretion, the greater the chances of expropriation (Jensen, 1986).

We additionally used binary variables related to the activity sector of the company to isolate idiosyncrasies of the various sectors that are not captured by the other regressors. In addition, we added dummies for certain years to isolate macroeconomic shocks that are dissociated with the individual heterogeneity of the companies. Table 1 summarizes the description, definition and expected sign of each variable.

Table 1.

Explanatory variables with their definitions and expected relations.

Explanatory variable	Description	Definition	Expected signal
SIZE	Size	Natural log of total asset	(+)
ROA	Return on asset	Net income to total asset	(+)
GO	Growth opportunity	Annual rate of sales growth	(–)
CAPEX	Capital expenditure	Capital expenditure to total assets. If the variable was null, we considered the investment on fixed assets minus the sale of fixed assets	(–)
LEV	Leverage	Gross debt to total assets	(+/undefined)
CR	Current ratio	Current assets to current liabilities	(+)
GC	Corporate governance	Binary variable was set to one for the firms that presented higher levels of corporate governance (listed on ‘New Market’ and ‘Level 2’ segments of the Brazilian Stock Exchange – B3) in a given year. Firms listed in other segments were the reference category	(+)
II	Institutional investors	Binary variable was set to one for the firms that have institutional investors with a share of more than 5% in a given year. Firms without institutional investors among their shareholders were the reference category	(+)
II₁, II₂, II₃ and II₄	Clusters of institutional investors	Categorical variables that represent the cluster of institutional investors with the largest joint share in the firm ownership structure. Firms without institutional investors among their shareholders were the reference category
SECTOR	Sector of activity	Binary variables for each sector (19 sectors in total)
YEAR	Year	Binary variables for each year of the sample period (seven years in total)

We estimated two model specifications for each dependent variable. Within the first, we considered the presence of institutional investors in the aggregate (variable II), where II equals 1 when the company has an institutional investor in its ownership structure, and 0 otherwise. In this respect, we must note that Brazilian publicly traded companies are only required to disclose shareholders with an ownership of more than 5%.

Within the second model specification, we took into account the heterogeneity of institutional investors (variables II₁ to II₄). To classify institutional investors, we used factor and cluster analysis. First, according to Brossard et al. (2013), Bushee (1998, 2001, 2004), Chen et al. (2007) and Crane et al. (2016), we applied factor analysis (specifically, principal component analysis – PCA), considering 17 indicators related to institutional investor investment profiles in the Brazilian stock market.⁷ Using PCA with Promax rotation, we extracted four factors: (i) ownership size; (ii) portfolio diversification; (iii) ownership stability and (iv) marginal influence, which collectively accounted for 68% of the total variance of the indicators.

Second, we applied cluster analysis by adding further individual characteristics to the four aforementioned factors to group similar agents into clusters, without disregarding investor heterogeneity.⁸ The individual characteristics were based on the following factors: (i) whether the institutional investors were asset managers, pension funds or insurance companies; (ii) whether they were national or international; (iii) whether they were public or private and (iv) whether they were dependent or independent, where we considered dependent investors those who were part of a financial conglomerate or managed by the government.⁹

Table 2 summarizes the main characteristics of the four clusters obtained by the classification and cluster methodologies applied in this study. While Clusters 1 and 2 were composed of international investors, Clusters 3 and 4 were both formed by Brazilian (national) investors. Meanwhile, government (public) and dependent investors were grouped into Cluster 1 and, in particular, into Cluster 4, while private and independent institutional investors were grouped into Clusters 2 and 3.

Table 2.

Classification of institutional investors.

Cluster	Variable	Institutional investors’ characteristics	Investment profile
1	II₁	International13% Public and 87% PrivateDependent	– Position: small– Portfolio: diversified– Stability: low– Marginal influence: high
2	II₂	InternationalPrivateIndependent	– Position: intermediate– Portfolio: less diversified– Stability: low– Marginal influence: high
3	II₃	NationalPrivateIndependent	– Position: intermediate– Portfolio: not very diversified– Stability: low– Marginal influence: low
4	II₄	National21% Public and 79% PrivateDependent	– Position: large– Portfolio: diversified– Stability: high– Marginal influence: low

Clusters 1 and 2 included institutional investors with lower stability and greater marginal influence.¹⁰ Given such characteristics, we hypothesized that these agents tend to exert ‘myopic’ pressures on company strategies to achieve short-term returns (Brossard et al., 2013; Bushee, 2004) and favour financial investments and dividend payments. Although Clusters 2 and 3 were similar in terms of position size, stability and degree of portfolio diversification, the institutional investors grouped in Cluster 3 – in addition to all being national – assumed relatively larger positions and presented less marginal influence. Consequently, we hypothesized that the investors in Group 3 tend to focus on profit reinvestment and present a long-term profile. In addition, the group of institutional investors in Cluster 4, which includes Brazil’s largest pension funds (those of public companies), had the greatest stability, the highest positions and the lowest marginal influence of all clusters. Observing these ‘conservative’ characteristics, we also assumed that these investors tend to have a long-term view rather than a short-term profile (Chen et al., 2007; Crane et al., 2016; Ferreira and Matos, 2008).

Finally, considering the explanatory variables of companies, three variables – ROA, GO and CR – demonstrated both wide dispersion and outlier values. Therefore, we winsorized the observations of these variables and eliminated the observations from the lowest 1% and the highest 1% of the distribution in the case of ROA and CR. In the case of GO, we eliminated only the observations from the highest 1% of the distribution, as the inferior limit was adequate. Tables 3 and 4 present the descriptive statistics.¹¹

Table 3.

Descriptive statistics of variables – FI model.

Variables	Obs.	Mean	Median	Standard deviation	Minimum	Maximum
IF	1508	5.55	0.81	10.75	0.00	89.24
SIZE	1508	14.78	14.93	1.75	4.91	20.62
ROA	1508	3.47	3.54	9.04	–51.78	36.16
GO	1439	12.62	9.53	38.23	–100.00	368.84
CAPEX	1434	5.51	4.31	8.56	–84.38	109.39
LEV	1508	27.19	28.02	17.68	0.00	80.88

FI: financial investments; SIZE: size; ROA: return on asset; GO: growth opportunity; CAPEX: capital expenditure; LEV: leverage.

Table 4.

Descriptive statistics of variables – DIVID model.

Variables	Obs.	Mean	Median	Standard deviation	Minimum	Maximum
DIVID	1440	2.97	1.17	5.45	0.00	65.75
SIZE	1440	14.89	15.00	1.69	4.91	20.62
ROA	1440	3.61	3.67	9.08	–65.85	36.16
GO	1379	12.87	9.72	38.04	–100.00	368.84
CAPEX	1426	5.53	4.32	8.56	–84.38	109.39
LEV	1440	27.48	28.41	17.76	0.00	80.88
CR	1440	2.31	1.61	3.42	0.06	44.73

DIVID: dividend payments; SIZE: size; ROA: return on asset; GO: growth opportunity; CAPEX: capital expenditure; LEV: leverage; CR: current ratio.

Results

Table 5 shows the results of the estimated models. For each model, we first considered the effect of institutional investor ownership in the aggregate (greater than 5%). Next, we analysed the influence of investor heterogeneity.

Table 5.

System GMM estimation results.

	Model DIVID				Model FI
	Coef.	SD	Coef.	SD	Coef.	SD	Coef.	SD
Y _t–1	0.2297**	(0.103)	0.2550***	(0.095)	0.6053***	(0.099)	0.6109***	(0.096)
SIZE	–0.0666	(0.133)	–0.1048	(0.151)	–0.2897	(0.204)	–0.2876	(0.244)
ROA	0.1557***	(0.033)	0.1589***	(0.034)	0.0798*	(0.048)	0.0956*	(0.053)
GO	–0.0075**	(0.003)	–0.0085***	(0.003)	–0.0056	(0.005)	–0.0037	(0.005)
CAPEX	–0.0280	(0.017)	–0.0317	(0.021)	–0.1288***	(0.041)	–0.1368***	(0.046)
LEV	–0.0324**	(0.015)	–0.0295**	(0.014)	–0.0157	(0.028)	–0.0106	(0.027)
CR	0.0433	(0.061)	0.0444	(0.077)	–	–	–	–
CG (= 1)	0.0573	(0.323)	0.3095	(0.359)	–	–	–	–
II (= 1)	–0.4354	(0.300)	–	–	1.5063**	(0.800)	–	–
II ₁	–	–	–0.6622	(0.814)	–	–	0.6051	(1.148)
II ₂	–	–	–0.1673	(0.594)	–	–	1.6468	(1.229)
II ₃	–	–	–1.1124**	(0.482)	–	–	1.8064**	(0.935)
II ₄	–	–	–0.8136	(0.544)	–	–	1.4213	(1.352)
Sector	YES		YES		YES		YES
Year	YES		YES		YES		YES
N	1750		1178		1193		1193
Hansen	141.81	(0.285)	148.250	(0.260)	103.62	(0.122)	150.10	(0.369)
AR(1)	–6.54	(0.000)	–2.52	(0.012)	–3.63	(0.000)	–3.71	(0.000)
AR(2)	0.74	(0.459)	–0.66	(0.509)	–1.87	(0.061)	–1.83	(0.067)

Note: (i) ***, ** and * indicate statistical significance at the 0.01, 0.05 and .1 levels, respectively. (ii) The ‘YES’ indicates that sector and year dummies were incorporated into the model, but their coefficients were not presented due to a space constraint. (iii) For the Hansen, AR(1) and AR(2) tests, the statistic and the descriptive levels (p-value) are reported. (iv) SIZE: size; ROA: return on asset; GO: growth opportunity; CAPEX: capital expenditure; LEV: leverage; LIQ: liquidity; CG: corporate governance; II: institutional investors – II₁, II₂, II₃ and II₄: clusters to which institutional investors belong.

The results indicated a good fit of the models to the data. For all specifications, the null hypothesis was not rejected in the Hansen test of overidentifying restrictions, suggesting the instruments used were valid and not correlated with the error term. The autocorrelation tests indicated a negative first-order serial correlation in the first-difference residuals, but no negative second-order serial correlation, confirming that errors were uncorrelated in both models, regardless of the specification.¹²

In addition, the findings suggested the presence of a mean-reverting tendency and a high persistence of financial investments and dividend payments, which justify the use of a dynamic panel data model. The lagged dependent variable was statistically significant and positively correlated with its contemporary value for all specifications. The same applies to the profitability variable (ROA), which demonstrated a positive and statistically significant effect. Previous studies have shown that ROA has always been considered an important indicator of investment and payout policies (Gill et al., 2010; Martins and Famá, 2012; Mehta, 2012). Similarly, we found that the higher the net profit, and consequently, the ROA, the greater the availability of resources for reinvestment and the greater the investor expectation for dividend payments.

Results from the DIVID model indicated that firms with greater GOs pay less dividends, as do firms with higher levels of LEV. These results, as well as those concerning profitability, are in line with studies on the determinants of dividend policy in Brazil (Martins and Famá, 2012). Firms with few GOs incur higher agency costs, thus making dividend payments an important instrument of CG (Jensen, 1986). The negative relationship between LEV and dividend payments implied that LEV constrains future cash flows and, consequently, dividend payout. However, the SIZE and the CAPEX variables, as well as the CR and the pattern of GC variables, proved irrelevant in explaining dividend distribution.

With respect to the FI model, the variables GO and LEV were not statistically significant. In this model, CAPEX was the variable that best explained the firm’s decision to direct resources to financial investments. The CAPEX variable demonstrated a negative relationship, implying that firms with larger productive investments had a smaller share of total assets allocated to financial investments. The cycle of increasing financial investments over productive investments is extensively debated in the literature on financialization, given the significant fall in the rate of capital accumulation and economic growth observed in advanced economies (Duménil and Lévy, 2011; Orhangazi, 2008; Stockhammer, 2004).

Our results also indicated that the presence of institutional investors in company ownership structure, regardless of investor heterogeneity, had no influence on dividend policy, contrary to the findings of Crane et al. (2016). Previous studies exploring the determinants of dividend payments in Brazil found evidence of an ex ante payout policy decision among NFC. Specifically, payout is less subject to the immediate interests of shareholders, and instead aligned with a company’s statute and Brazilian corporate law (Martins and Famá, 2012). With respect to the FI model, the findings demonstrated that institutional investors influenced company investment strategy and favoured financial investment.

However, Table 5 also presents evidence that a particular group of institutional investors (II₃) is determinant of higher financial investment, but lower dividend payments, confirming the importance of incorporating the heterogeneity of institutional investors in analysing their influence on NFC (Brossard et al., 2013; Bushee, 1998, 2001, 2004; Çelik and Isaksson, 2013; Chen et al., 2007; Crane et al., 2016). This group of institutional investors was characterized by intermediate shareholder positions in a limited number of companies and sectors (Table 2). While the investors did not necessarily maintain their positions for long periods – greater than two years – they did not have high marginal influence. Specifically, these investors had holdings large enough to exert influence, but not to incur liquidity problems if they wanted to liquidate their position. Private and independent asset managers – mostly national – comprised the majority of the investors in this group (Cluster 3).

Conversely, neither institutional investors characterized by lower ownership stability and high marginal influence nor those characterized by larger positions and stability demonstrated any influence over company strategies. Specifically, neither short-term (Clusters 1 and 2) nor long-term (Cluster 4) investors influenced the proxy variables for the financialization of Brazilian NFC. These results contradicted those found in the literature (Brossard et al., 2013; Bushee, 1998, 2001). Katan and Mat Nor (2015), however, found evidence that neither institutional ownership (in the aggregate), nor long-term investor ownership in particular has a significant effect on company performance.

The findings from this analysis raise several questions. First, why did volatile investors (Clusters 1 and 2, and mostly international asset managers) show no positive influence on either payout or investment policy? Second, why did conservative investors (Cluster 4, including pension funds of Brazilian public companies, and with more stable ownership and holding larger positions) demonstrate no influence over financial investment and dividend payouts? Third, why was the influence exerted by investors with no diversified portfolio, no large marginal influence, and who did not hold the largest positions (Cluster 3)? Finally, why was influence exerted in favour of financial investments, yet contrary to dividend payments?

In summary, the results suggested that a significant number of institutional investors did not influence the financialization of Brazilian NFC. To understand this behaviour, Brazilian institutional framework should be taken into consideration. Specifically, the interest rate gap, exchange rate gap and return gap between productive and financial investment in Brazil (and in other EME) are possibly among the main determinants of the portfolio allocation of international institutional investors (Bortz and Kaltenbrunner, 2018; Kaltenbrunner and Painceira, 2015). Moreover, while Brazilian capital markets are poorly developed, the government debt market – which the investor can access directly – is large, liquid, riskless and profitable, as is the case in other EME (Becker et al., 2010; Bin, 2016; Bruno and Caffé, 2017).

In this context, foreign asset manager ownership in Brazilian companies may likely follow a portfolio diversification policy or a conjunctural portfolio allocation decision. The fact that the Brazilian stock market likely represents a very small portion of the portfolios of foreign asset managers, in addition to the investment and CG patterns of Brazilian NFC, could also explain the lack of foreign institutional investor interest in influencing corporate management. That is, foreign institutional investor ownership in Brazilian companies indicates a passive profile associated with the portfolio’s diversification.

This explanation, however, does not apply to all institutional investors. Conservative investors (Cluster 4) presented diverse characteristics and consequently, diverse motivations. They were shareholders of several companies from various sectors, had a high level of stability and had large positions. If their motivation were portfolio diversification, they would have no reason to either assume large holdings or keep their holdings for a long period.

Two main factors may explain the lack of influence of these investors. First, the conservative investor cluster contains national pension funds of public companies that adopted a specific portfolio strategy during the period under analysis (2010–2016). They reduced equity investments due to both performance problems and increasing volatility in Brazilian stock markets. The investment strategy of these agents, focused on fixed income, may have affected their capacity and interest in interfering in company management (De Conti, 2016a, 2016b). Second, these agents tended to increase their indirect ownership through investment funds. Even larger pension funds, which generally have a complex organizational structure, outsourced the management of their portfolio (De Conti, 2016a).¹³

The results confirmed the hypothesis that institutional investor influence is mostly motivated by the particularities of each group of investors. However, this begs the questions of which Cluster 3 characteristics explain the group’s influence, and moreover, why Cluster 3 is determinant of less dividend payments and more financial investments.

The results suggested that investors exerting significant influence in both models held a less diversified investment strategy. Specifically, such agents invested in a limited number of firms and sectors. These investors held intermediate positions and were also less volatile, meaning they did not oscillate around the minimum position required to exert influence. They tended to exert influence in favour of smaller dividend payments, which indicated their support for reinvesting these profits back into the company. In sum, these investors assumed a longer term view consistent with the portfolio specialization associated with relatively larger positions.

Furthermore, unlike Clusters 1 and 2, which were mainly composed of international investors whose participation in Brazilian companies was likely marginal in their portfolios, Cluster 3 was composed entirely of national institutions. Therefore, we would expect the institutional investors in this cluster to be more dependent and consequently more interested in the results of the investee companies, which in turn explains the larger investor monitoring capability and their influence on the strategic management of companies.

However, findings from the FI model, in principle, contradict the conclusion that such shareholders do not hold a short-term view. This is so because their ownership is associated with larger financial investments, which indicates a search for short-term returns that are detrimental to real investment and consequently, to the sustainability of the company in the long term. However, their positive influence in favour of larger financial investments may be explained by the opportunity cost of capital in Brazil. The existence of a developed government debt market, in a context of economic recession and slow growth, may justify the reinvestment of profit in financial investments, which is accepted as the most appropriate investment strategy for the company itself (Demir, 2007, 2009).

In addition, the lack of liquidity in capital markets and higher Treasury bond yields may discourage long-term productive investment in Brazil even in a different economic context. The tendency to invest in fixed income bonds (particularly in government bonds) is a characteristic of EME, at least in Latin America and in Turkey (Demir, 2009). In Brazil, since the 1980s, the country’s high inflation rate and consequently the greater importance attributed to the safety and liquidity of assets – both of which are associated with the external debt crisis – motivated investors to invest in fixed income instruments indexed to the overnight interest rate. Despite the stabilization of prices observed during the 1990s, macroeconomic vulnerability and high real interest rates have limited the development of capital markets. The profitability of productive investment continued to be supplanted by fixed income securities, sustaining the transfer of wealth to the financial sector (Belluzzo and Almeida, 2002; Bruno and Caffé, 2017).

Conclusions

Several recent studies have examined the potential effects of institutional ownership on management priorities. However, the influence of institutional investor heterogeneity on NFC strategies is an issue not yet well explored by the literature on EME. Our contribution lies in further understanding the impact of institutional investor shareholding on the strategic management of NFC, taking into account institutional investor heterogeneity and aspects of the Brazilian institutional framework.

Our results indicated that institutional ownership in Brazilian NFC had no influence on dividend policy, but had a positive influence on the share of total assets allocated to financial investments. With respect to the control variables, while variables that affect the availability of resources for distribution did impact the dividend policy, financial investment was positively related to profitability and negatively related to productive investment, which is consistent with a crowding-out effect widely discussed in the literature on financialization.

Thus, we found evidence that a particular type of investor influenced management of the investee companies, corroborating our claim that institutional investor heterogeneity should be considered in the analysis. Institutional investors with intermediate positions in a limited number of firms and sectors – primarily independent and national private asset managers – were found to be determinant of higher financial investments, but of lower dividend payments.

Profit reinvestment support is aligned with a long-term view compatible with portfolio specialization and relatively larger positions. On one hand, since Cluster 3 mostly consists of national asset managers, the potential dependence of the investee companies could explain the investor activism. On the other hand, our findings suggest that this particular group of investors was determinant of higher financial investments. Factors such as high interest rates, high uncertainty and the high opportunity cost of productive capital in Brazil may explain institutional investor influence in favour of financial investments.

Furthermore, our results indicated that investors characterized by short-term preferences or with a potential long-term strategy did not appear to influence the financialization of Brazilian non-financial corporation strategies. This may be a result of the characteristics of Brazilian institutional framework. The marginal allocation of funds by international investors in domestic companies is an additional aspect to be considered, as it would reinforce the assumption of a passive profile associated with portfolio diversification. In the case of conservative investors, we highlight a tendency of increasing indirect ownership in companies through investment funds.

Future research should analyse the entire portfolio of institutional investors, including international investments and investments in other bonds, rather than shares. Filling this gap would allow us to confirm the hypotheses that were proposed – but not tested – in this article. For example, whether the passivity of volatile investors can be effectively associated with the marginality of their investments in EME companies could be verified in further studies, as these investors appear to have a short-term profile essentially driven by the return gap. Additionally, future research could test whether more conservative investors, instead of exerting a direct influence over investee companies, have indirect involvement in the decision-making process.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Appendix 1. Distributions and descriptive statistics of the variables used in the FI model

References

Acharya

Viswanathan

(2011) Leverage, moral hazard, and liquidity. The Journal of Finance 66(1): 99–138.

Aglietta

(2000) Shareholder value and corporate governance: some tricky questions. Economy and Society 29(1): 146–159.

Arellano

Bond

(1991) Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. The Review of Economic Studies 58(2): 277–297.

Arellano

Bover

(1995) Another look at the instrumental variable estimation of error-components models. Journal of Econometrics 68(1): 29–51.

Barradas

Lagoa

(2017) Financialization and Portuguese real investment: A supportive or disruptive relationship? Journal of Post Keynesian Economics 40(3): 413–439.

Becker

Jäger

Leubolt

et al. (2010) Peripheral financialization and vulnerability to crisis: a regulationist perspective. Competition and Competition & Change 14(3–4): 225–247.

Belluzzo

Almeida

(2002) Depois da Queda [after the Fall]. Rio de Janeiro: Editora Civilização Brasileira.

Bin

(2016) The politics of financialization in Brazil. World Review of Political Economy 7(1): 106–126.

Black

Carvalho

Gorga

(2008) An overview of Brazilian corporate governance. Cornell Law School Research Paper No. 08-014, Cornell University Law School, Ithaca.

10.

Blundell

Bond

(1998) Initial conditions and moment restrictions in dynamic panel data models. Journal of Econometrics 87: 115–143.

11.

Bonizzi

(2017) Institutional investors allocation to emerging markets: A panel approach to asset demand. Journal of International Financial Markets, Institutions and Money 47: 47–64.

12.

Bortz

Kaltenbrunner

(2018) The international dimension of financialization in developing and emerging economies. Development and Change 49(2): 375–393.

13.

Brickley

Lease

Smith

(1988) Ownership structure and voting on anti-takeover amendments. Journal of Financial Economics 20: 267–291.

14.

Brossard

Lavigne

Sakinç

(2013) Ownership structures and R&D in Europe: The good institutional investors, the bad and ugly impatient shareholders. Industrial and Corporate Change 22(4): 1031–1068.

15.

Bruno

Caffé

(2017) Estado e financeirização no Brasil: interdependências macroeconômicas e limites estruturais ao desenvolvimento [State and financialization in Brazil: macroeconomic interdependencies and structural limits to development]. Economia e Sociedade 26: 1025–1062.

16.

Bushee

(1998) The influence of institutional investors on myopic R&D investment behavior. The Accounting Review 73(3): 305–333.

17.

Bushee

(2001) Do institutional investors prefer near-term earnings over long-run value? Contemporary Accounting Research 18(2): 207–246.

18.

Bushee

(2004) Identifying and attracting the “right” investors: Evidence on the behavior of institutional investors. Journal of Applied Corporate Finance 16(4): 28–35.

19.

Çelik

Isaksson

(2013) Institutional investors as owners: Who are they and what do they do? OECD Corporate Governance Working Paper 11, OECD, Paris.

20.

Chen

Harford

(2007) Monitoring, which institutions matter? Journal of Financial Economics 86(2): 279–305.

21.

Clark

(2000) Pension Fund Capitalism. New York: Oxford University Press.

22.

Crane

Michenaud

Weston

(2016) The effect of institutional ownership on payout policy: Evidence from index thresholds. Review of Financial Studies 29(6): 1377–1408.

23.

Crotty

(2002) The effects of increased product market competition and changes in financial markets on the performance of nonfinancial corporations in the neoliberal era. PERI Working Paper 44, Political Economy Research Institute, University of Massachusetts at Amherst, Amherst.

24.

David

Hitt

Jimeno

(2001) The influence of activism by institutional investors on R&D. Academy of Management Journal 44(1): 144–157.

25.

Davis

(2009) Managed by the Markets. Oxford: Oxford University Press.

26.

Davis

(2017) Financialization and investment: A survey of the empirical literature. Journal of Economic Surveys 31: 1332–1358.

27.

De Conti

(2016a) Os fundos brasileiros de previdência complementar, segmentações analíticas e estudos preliminares sobre a alocação de seus recursos [Brazilian complementary welfare funds, analytical segmentations and preliminary studies on the allocation of its resources]. In: Dos santos C (ed.) Características Estruturais Do Sistema Financeiro Brasileiro, um Registro da Reflexão Do Ipea No Biênio 2014–2015 [Structural Characteristics of the Brazilian Financial System, a Record of Ipeas Reflection in the Biennium 2014–2015]. Brasília: IPEA, pp.343–374.

28.

De Conti

(2016b) Previ, Petros e Funcef: uma análise da alocação das carteiras das três maiores entidades brasileiras de previdência complementar (2003–2013) [Previ, Petros and Funcef: An analysis of asset allocation of the three largest Brazilian private pension entities (2003–2013)]. In: Dos Santos C (ed.) Características Estruturais Do Sistema Financeiro Brasileiro, um Registro da Reflexão Do Ipea No Biênio 2014–2015 [Structural Characteristics of the Brazilian Financial System, a Record of Ipeas Reflection in the Biennium 2014–2015]. Brasília: IPEA, pp.375–425.

29.

Demir

(2007) The rise of rentier capitalism and the financialization of real sectors in developing countries. Review of Radical Political Economics 39: 351–359.

30.

Demir

(2009) Financial liberalization, private investment and portfolio choice: Financialization of real sectors in emerging markets. Journal of Development Economics 88: 314–324.

31.

Duménil

Lévy

(2011) The Crisis of Neoliberalism. Cambridge: Harvard University Press.

32.

Epstein

(2005) Introduction, financialization and the world economy. In: Epstein

(ed.) Financialization and the World Economy. Cheltenha: Edward Elgar, pp.3–16.

33.

Fazzari

Hubbard

Petersen

(1988) Financing constraints and corporate investment. Brookings Papers on Economic Activity 19: 141–195.

34.

Ferreira

Matos

(2008) The colors of investors’ money: The role of institutional investors around the world. Journal of Financial Economics 88(3): 499–533.

35.

Fichtner

(2013) Hedge funds: agents of change for financialization. Critical Perspectives on International Business 9(4): 358–376.

36.

Gill

Biger

Tibrewala

(2010) Determinants of dividend payout ratios: Evidence from United States. The Open Business Journal 3: 8–14.

37.

Guttmann

(2008) Uma introdução ao capitalismo dirigido pelas finanças [An Introduction to Finance-Driven Capitalism]. Novos Estudos - CEBRAP 82: 11–33.

38.

Hutchinson

Seamer

and Chapple

(2015) Institutional investors, risk/performance and corporate governance. The International Journal of Accounting 50(1): 31–52.

39.

Jensen

Solberg

Zorn

(1992) Simultaneous determination of insider ownership, debt, and dividend policies. The Journal of Financial and Quantitative Analysis 27(1): 247–263.

40.

Jensen

(1986) Agency costs of free cash flow: Corporate finance, and takeovers. American Economic Review 76(2): 323–329.

41.

Kaltenbrunner

Painceira

(2015) Developing countries changing nature of financial integration and new forms of external vulnerability: The Brazilian experience. Cambridge Journal of Economics 39(5): 1281–1306.

42.

Kaltenbrunner

Painceira

(2018) Subordinated financial integration and financialisation in emerging capitalist economies: The Brazilian experience. New Political Economy 23(3): 290–313.

43.

Karwowski

Stockhammer

(2017) Financialisation in emerging economies: A systematic overview and comparison with Anglo-Saxon economies. Economic and Political Studies 5(1): 60–86.

44.

Katan

Mat Nor

(2015) Institutional ownership heterogeneity and firm performance: Evidence from Malaysia. International Journal of Economics and Finance 7(12): 176–188.

45.

Kliman

Williams

(2015) Why financialization hasn’t depressed US productive investment. Cambridge Journal of Economics 39: 67–92.

46.

Krippner

(2005) The financialization of the American economy. Socio-Economic Review 3(2): 173–208.

47.

Lazonick

O’Sullivan

(2000) Maximizing shareholder value: A new ideology for corporate governance. Economy and Society 29(1): 13–35.

48.

Martins

Famá

(2012) O que revelam os estudos realizados no Brasil sobre política de dividendos? [What do the studies carried out in Brazil on dividend policy reveal?]. Revista de Administração de Empresas 52(1): 24–39.

49.

Mehta

(2012) An empirical analysis of determinants of dividend policy: Evidence from the UAE companies. Global Review of Accounting and Finance 3(1): 18–31.

50.

Miranda

BPJ

Crocco

Santos

(2015) Impactos da financeirização sobre o padrão de financiamento e a governança das empresas não-financeiras brasileiras de capital aberto: período de 1995-2008 [Financialization impacts on the financing and governance pattern of Brazilian publicly-held non-financial corporations: 1995–2008 period]. Revista de Economia 41(2): 7–30.

51.

Miranda

BPJ

Crocco

Santos

(2017) Financeirização e governança corporativa: um estudo sobre a estrutura de controle das empresas não-financeiras do Novo Mercado da BM&FBOVESPA. [Financialization and corporate governance: a study on the control structure of the non-financial companies of the New Market of BM&FBOVESPA]. Brazilian Keynesian Review 3(1): 75–94.

52.

Muniandy

Tanewski

Johl

(2016) Institutional investors in Australia: Do they play a homogenous monitoring role? Pacific-Basin Finance Journal 40: 266–288.

53.

Orhangazi

(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.

54.

Rabinovich

(2019) The financialization of the nonfinancial corporation: A critique to the financial turn of accumulation hypothesis. Metroeconomica 1–38. DOI: 10.1111/meca.12251.

55.

Roodman

(2009) How to do xtabond2: An introduction to diﬀerence and system GMM in Stata. The Stata Journal: Promoting Communications on Statistics and Stata 9(1): 86–136.

56.

Stockhammer

(2004) Financialisation and the slowdown of accumulation. Cambridge Journal of Economics 28(5): 719–741.

57.

Stockhammer

(2008) Some stylized facts on the finance-dominated accumulation regime. Competition and Change 12(2): 184–202.

58.

Tori

Onaran

(2018a) The effects of financialization on investment: Evidence from firm-level data for the UK. Cambridge Journal of Economics 42(5): 1393–1416.

59.

Tori

Onaran

, (2018b) Financialization, financial development and investment: Evidence from European non-financial corporations. Socio-Economic Review 1–43. DOI: 10.1093/ser/mwy044.