Financialisation and its effects on fixed investment at non-financial businesses in the Netherlands

(1)

“Financialisation and its effects on fixed investment

at non-financial businesses in the Netherlands”

By Fransje Puts, 10658076

Bachelor Thesis Finance & Organisation

Supervisor: Mr. R.C Sperna Weiland

(2)

2

Statement of Originality

This document is written by Student Fransje Puts who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3

I. Introduction

According to the WRR Report of 2016, the Dutch society and economy are getting more dependent on the financial system. Financial products and instruments are becoming highly important for citizens, companies and governments. Households and their increasing mortgage debt, the privatization of public institutions and their

financial self- reliance, the high dependence of debt finance at small businesses, and the confrontation of a more volatile and technical financial surrounding of large companies. This issue is related to the relatively new concept 'financialisation'. A common used definition for financialisation is the “increasing role of financial

motives, financial actors and financial institutions in the operation of the domestic and international economies” (Epstein 2005, p. 3). This growing size and importance of the financial sector started around the 1980’s in the United States, along with the deregulations of the financial sector under president Reagan. These developments were also visible in other wealthy countries, like the Netherlands.

Well- functioning financial markets could lead to economic growth. The financial sector serve various functions like facilitating trade, mobilizing and allocating money from people who want to save to people who want to spend,

spreading risk and generating information of investment opportunities (Levine, 1997). The problem is that these financial markets are not always functioning well.

An increasing number of studies point out a non-linear relationship between financial development and economic growth. These empirical studies show that countries with an underdeveloped financial system benefit from an increase in credit facilities, but after a certain point, credit expansion could lead to a slowdown of activities in the real economy (Cecchetti en Kharroubi, 2015). Instead of fulfilling its original character of supporting the real economy, the financial sector started acting more like an independent industry (Boot, 2011). The financial sector is “outgrowing” and sometimes even withdrawing money from the real economy. It uses (human) resources that could otherwise be utilized, maybe more productively, in another sector. In a cross-country analysis of 17 OECD countries, it shows that the

Netherlands is ranked at the fifth place of highest gross financial income (Karwowski, Shabani & Stockhammer, 2016).

(5)

5 It is also questioned whether the expansion in private debt, partially caused by the growing financial sector, results in socially optimal outcomes (Turner, 2012). Higher leverage could result in rigorousness and financial instability. The last twenty years, private debt increased significantly in the Netherlands. Although households are the main player in this occurrence, non- financial corporations hold, relative to other developed countries, also a large amount of debt. According to the previous named cross-country analysis, Dutch non-financial businesses have the highest debt to income ratio (Karwowski, Shabani & Stockhammer, 2016).

At the same time, somewhat contradictory, the Dutch gross savings of non-financial businesses are substantially higher relative to other EU-countries (Jansen & Ligthart, 2014). The growth in gross savings is accompanied by detained real

investments. This movement is also apparent in other developed countries but not as significant as in the Netherlands. The investment quote, calculated as gross

investments relative to value added, at non- financial companies decreased

remarkably. Whereas this quote was 28 per cent in 1970, it is now approximately 16 per cent (WRR, 2016).

A growing amount of empirical literature emphasizes the linkage between a slowdown in capital accumulation and the growing size and importance of the financial sector (Lazonick & O’Sullivan 2000; Stockhammer, 2004; Orhangazi, 2008). These studies were country-specific and mainly focused on the United States, United Kingdom, France and Germany. The advice report of the WRR 2016 states that too little attention is given yet to the financialisation of the Dutch society. More empirical research should be done whether this growing financial sector is really a problem, and how it affects the real economy. Productive investments are an important determinant of growth, employment and income, that is, how well an economy is doing and therefore meaningful to study. Because of the relatively high financial income and strong decreasing investment quote in the Netherlands, it is a preeminent country to investigate.

To examine the relationship of financialisation and the acquisition of fixed investments, a time-series analysis will be done over the years 1995-2015. The research is based on the sector 'non- financial companies' in the Netherlands. A negative relationship is found in most of the regressions. However, the results should be taken with caution due to the relatively low amount of observations.

(6)

6 First, in the theoretical part some explanations of the decreasing investment quote and the increasing importance of financial assets at non- financial companies in the Netherlands will be given. Also previous studies will be discussed and analyzed and linked to the special case of the Netherla nds. Thereafter, a regression equation based on the theory will be specified, followed by the presentation of the econometric results. After that, the limitations and short-comings of the study are discussed and a final conclusion is made.

(7)

7

II. Literature Review

The ministry of Economical Affairs, Agriculture and Innovation (ministerie van Economische Zaken, Landbouw en Innovatie, EL&I) and other advisory institutions like the Bureau for Economic Policy Analysis (Centraal Plan Bureau, CPB) and the Scientific Council for Government Policy (Wetenschappelijke Raad voor

Regeringsbeleid, WRR) took notice of the strongly decreasing investment quote at non- financial businesses over the last twenty years. The Netherlands currently has the lowest investment quote among all European countries.

Source CBS Gross investments relative to value added of non-financial corporations (%).

A possible determinant of the decreasing investment quote is the increasing service sector in the wealthy countries. More Dutch companies outsource or offshore the manufacturing part to other countries to lower the costs or to focus on their core business (Jansen & Ligthart, 2014). As a result, the investments done abroad are not part of the Dutch investment quote, while the revenues from this product are obtained in the Netherlands. And thus value added will increase more significantly than real investments. However, this argument could be weakened by the fact that in

comparable European countries the service sector is also rising, while their investment quote doesn't show such a strong decline.

15 16 17 18 19 20 21 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 Investment Quote in %

(8)

8 Another possible determinant of the decreasing investment quote could be the growing productivity; fewer resources are needed to produce the same amount of goods (Verhoeven et al., 2012). Besides, relative to other European countries, fixed capital in the Netherlands has a longer longevity, and thus reduces the purchase of new fixed capital.

The above named causes are not necessarily an issue; it doesn’t affect the amount of products or services that could be produced. A more problematic

hypothesis states that the financialisation of the business sector hampers productive investments. This cause would lower the amount of goods and services in the real economy. An indicator of financialisation that will also be used in the empirical part, and is used in previous studies (Stockhammer 2004, 2016), is the possibly rising rentier income share of non- financial companies, that is, the income out of interest and dividend.

Source: OECD. Interest and dividend income relative to value added (%) in the Netherlands

The chart shows that from 1995 until around 2013, there is indeed a significant increase in dividend and interest income, with a short downfall around the crisis. By the use of this proxy, we could say financialisation took place at non-financial businesses during this period. Remarkably is the downward trend from around 2012 onwards. 5,00% 7,00% 9,00% 11,00% 13,00% 15,00% 17,00% 19,00% 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

(9)

9 The potential negative relationship of financial development and capital accumulation could be explained via two channels (Orhangazi, 2008). First, when non- financial companies increase their financial investments, fewer funds are available for real investments. This is also called the ‘crowding out’ effect. Non-financial companies are getting more engaged in Non-financial assets, even though this is not part of their core competence. Because of technological changes and financial globalisation, transaction costs are lower and the investment possibilities for

stockholders are significantly increased (Boot, 2014). Non- financial corporations are therefore confronted with more competition and have to keep up with the pace of a higher dynamic financial surrounding. The crowding-out effect also appears at the level of human capital. The most ‘able’ or skilled workers are attracted for fulfilling these complex tasks, while these high-ability workers could have worked in other technical industries (Cecchetti & Kharroubi, 2015). In this way the growing

importance of financial positions is subtracting talented workers from positions that are important for the growth in the real economy (Kneer, 2013). The relatively high salaries in the financial sector are an important determinant of the popularity among skilled workers to work in this sector (Phillipon & Reshef, 2012).

The second channel through which financialisation could negatively affect capital accumulation is the increased shareholder pressure. Managers are forced to increase dividends and stock buy-backs to higher the value of their stocks.According to the instigators of the dividend theory, Modigliani and Miller (1961), dividend policy is irrelevant when all the excess financial means of a company will be paid out eventually. This statement is true under perfect market conditio ns and the assumption of rational behaviour of stockholders. Though, these conditions do not hold in the real world and when companies never pay out the excess financial means dividend policy do play a role in the shareholder value. In the Netherlands the pay-out of dividends at non- financial companies is low relative to their obtained profits and compared to other European countries (Van Ees et al., 2008). On the other hand, the buy-backs of own stock, which could be seen as an indirect way of shareholder compensation, are relatively high in the Netherlands. When a firm buys back its own shares, less stocks are available, accordingly the price of the stocks and thus shareholder-value will increase. Consequently, fewer funds are available for fixed investments.

(10)

10 A general change in corporate governance structure is the main cause of the above described shareholder pressure. This structure transformed from a managerial-based firm to a rentier-dominated firm. Under the managerial firm, fixed dividends were paid and excessive profits were used for productive investments (Klundert, 2016). But around the 1980’s a shift took place that increased the power of the shareholders. The main aim of a rentier-dominated firm was maximizing shareholder value, which resulted in higher dividend pay-outs, and in the Netherlands particularly, in the increase of buybacks of own stock. The core strategy changed from ‘retain and reinvest’ towards ‘downsize and distribute’ (Lazonick & O’Sullivan, 2000). The first was focused on retaining profits and employees, and reinvesting this in fixed as well as human capital, whereas the second, and now prominent principle was all about raising stock prices and increasing the value of stock options. This focus on

maximizing shareholder value could partly be explained by “short-term performance obsession” (Rappaport, 2005). He states it is simply easier to make decisions based on short-term investments that result in directly visible results, than on long-run projects. Besides, also supported by the liquidity preference theory, investors prefer liquid assets to illiquid assets (Turberville, 2015). The process of liquidating fixed

investments is hard and reduces a company’s flexibility. Also because of the higher instability of financial markets the variance in expected returns of fixed investments has increased over time, which makes these kind of investments more risky and volatile (Onaran et. al, 2009). Therefore managers are tempted to invest in financial assets that could result in quick returns, and long-term (fixed) investments got neglected

The change in governance structure is a result of the objective to align the incentives of the shareholders with the incentives of the managers. Because of information asymmetry managers were able to act in their own interest and were thus focused on the size of the firm to enhance their own income. Increasing firm size doesn’t necessarily result in higher profitability, while high profits are the main aim of the shareholders of the firm. To solve this classical principal-agent problem, the goal was to align the objectives of the managers with the objectives of the shareholders (Grossman & Hart, 1982). Alignment could be achieved by the possibility of firing managers in the form of hostile takeovers, as well as the implementation of

(11)

11 performance related pay schemes (Stockhammer, 2004). Salaries of the managers were linked to the changes in stock prices of the company and resulted in short-term actions to influence the stock value positively (Boot, 2010). These short-term actions are primarily a reflection of the trade in financial assets to generate quicker and higher profits.

Whether the increasing usage of financial assets and the growing importance of shareholder-value indeed decreased the real investments at non- financial

businesses, is tested empirically by a couple of researchers. Stockhammer (2004) performed a cross-country time series analysis for the United States, United Kingdom, France and Germany. For France he found that the entire slowdown of capital

accumulation could be explained by financialisation, and for the United States one-third of the slowdown could be explained by financialisation. Some as well as none support are found for the United Kingdom and Germany respectively.

Orhangazi (2008) tested the relation of financialisation and decreasing

investments at firm- level basis in the United States. He categorised the firms by size and sector. His findings indicate a negative relationship, especially for the large businesses.

Kliman and Williams (2014) oppose these studies by their findings that the slowdown in accumulation is mainly a result of the lowered profit share of non-financial corporations. They also find that non-financial payments and acquisitions in the United States are mainly financed by debt, and thus the claim that profits have been diverted from fixed investment to financial purchases is not true. But even if these financial assets are paid by debt, it might be still at the expense of fixed investments. As companies are restricted to a certain debt-equity structure, less debt could be issued to finance for example new machines. Besides, in the Netherlands the profit share merely increased over time, and the debt level relative to financial assets of non-financial businesses decreased significantly. Therefore the findings of Kliman and Williams do not hold per se for the Netherlands.

(12)

12 Source: CBS, Profits relative to value Source: OECD, Debt relative to financial

added at non-financial businesses assets at non-financial businesses

To test whether financialisation at non-financial companies in the Netherlands deduced capital accumulation, an investment equation will be specified and tested. Both of the channels through which financialisation might have resulted in lower fixed investments are incorporated in this function. The income out of interest and dividend covers the increased shareholder value, whereas the amount of financial assets a company holds reflect the crowding-out effect. To control for these

financialisation measures standard investment variables, like productivity and profits, are added. By testing these variables empirically, we will hopefully gain some

information whether the theoretical part will be supported by the numbers. 36 37 38 39 40 41 42 43 44 45 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 Profit Share in % 0,5 0,7 0,9 1,1 1,3 1,5 1,7 1,9 1990 1995 2000 2005 2010 2015 2020 Debt to financial assets ratio in

(13)

13

III. Methodology

Regression specification

To test the relation of real investments and financialisation of non- financial corporations empirically, a time series analysis of the investment quote in the Netherlands will be carried out. As explained in the litera ture part, the hampering of capital accumulation caused by financialisation appeared via two channels; via the so-called crowding-out effect, and via the growing importance of shareholder value. The aim is to test both of these causes empirically. Therefore, two separate regressions will be studied. Each using the same control variables but a different measure of financialisation. The general investment equation is specified as follows:

where , denote investment quote, capacity utilization, profit share, and financialisation, respectively.

is the investment quote and is calculated as gross fixed investments relative to the value added of non- financial businesses; is capacity utilization and is measured by dividing value added of non-financial businesses by fixed assets in stock; is the profit rate and is the gross operating surplus relative to value added of non- financial businesses. The financialisation measure that will be used in the first regressions is RSNF. This is the rentier share of non-financial businesses and is the sum of interest- and dividend income divided by the value added. The second

financialisation measure represents financial assets relative to total assets. All the variables are in current prices.

According to Keynesian as well as neoclassical models, investment equations depend on output. This output is based on both the standard accelerator effect and the cost of capital (Chirinko, 1993). The accelerator effect is represented by the productivity measure capacity utilization and proved to be significant in previous studies (De la Croix & Licandro, 1993; Stockhammer, 2004). Higher capacity utilization should influence business investments positively ( ); a company is expected to invest more the closer it gets to full capacity. To the contrary, the cost of capital measure in the research of Stockhammer (2004) failed to be significant. Also, in most of the

(14)

14 regressions the cost of capital did not have the expected sign. For this reason, I

decided to leave out this variable.

Subsequently, real investments are expected to increase when the profit share rises. The higher the profits, the more available intern funds to finance new

investments ( ). Non-financial businesses seem to rely growingly on internal funds for fixed investments (OECD, 2007). Though, in previous empirical studies, profits play a moderate role in explaining aggregate investment (Ford and Poret, 1991). Inverse causation, the case that investments might result in higher profits as well as the other way around, might be an explanation (Kalecki, 1954). Still, profits play an important role and are also backed up by the theoretical part.

The measure of financialisation that is used in previous studies explains the growing shareholder importance, namely the income out of interest and dividends (Stockhammer 2004; Onaran et al., 2009; Stockhammer, 2016). This proxy proved to be significant and had the expected sign in different time periods ranging from 1963 to 1997, particularly in France and the United States. The relationship of

financialisation and capital accumulation is expected to be negative ( ). This might be in contrast with the reasoning of firms being finance constraint, rentier income is still income, and thus more funds to invest in fixed assets. Nevertheless, I will argue that this income is not used for fixed investments and thus have a negative effect on the desired level of real investment.

The crowding-out effect will be tested via another proxy of financialisation in a separate regression. This measure has not been supported by empirical studies yet, but relating to the theory it might be an important factor. It is an indicator of the amount of financial investment of non- financial businesses and is calculated by financial assets relative to total assets. When financial assets increase relative to the total obtained assets, it shows the growing importance of financial assets compared to fixed assets. The higher the investment in financial assets, the fewer funds available for fixed investment, and thus a negative relationship is expected ( ).

In time-series analysis it is common practice to use lagged variables to account for the time delay between the investment decision and the purchase of fixed

investment (Fazzari & Mott, 1986). The dependent variable is lagged and is used as an independent variable. Besides, all explanatory variables are lagged. This will prevent problems related to dynamic effects.

(15)

15 Data

For this research a combination of data sources is used, namely the Dutch Central Bureau of Statistics (CBS) database, the OECD database, and the database of the European Union called Eurostat. They all make use of the same accounting methods and report their values in current prices, and thus combining these variables is possible. Because of data availability of all variables needed, the research covers the time-period from 1995 until 2015, and is on yearly basis. As this study is about the financialisation of the non-financial businesses, only data for non- financial companies are used. This sector contains all the private-, public-, and quasi limited liability companies, except for financial institutions.

The gross fixed investments and value added are obtained via the 'kerngetallen' of the national accounts of the CBS database Statline. Besides, the operating surplus is found on the Eurostat database. Furthermore fixed assets are from the balance sheet of non-financial assets of the national accounts of the OECD

database, whereas the financial assets are obtained via the financial accounts tables of the national accounts. The interest- and dividend income and payments are from the transactions account of the national accounts, also from the OECD database. (see Appendix Table III.)

(16)

16

IV. Results

In the first regression, a first order autoregressive model will be tested. That is, a regression of the investment quotes onto its own lag. This shows the relationship of the dependent variable with its own past. Second, the financialisation measure will be added to check the direct relationship of the dependent variable and the explanatory variable we wish to investigate. Thirdly, following the literature (Bhaskar & Glyn, 1995) a partial adjustment model (PAM) is in this case an appropriate estimation method. Thereafter, a reparameterised autoregressive distributed lag (ADL) model in levels as well as in difference form is estimated to make sure the estimates in the PAM specification are free of spurious effects. This estimation strategy is deducted from the study of Stockhammer (2004). All the regressions are carried out twice. First the financialisation measure is included. By the use of this indicator, an extra regression is added, to control for the rentier payments of non- financial businesses ( ). Secondly, the financialisation measure is used in all the regressions. As the lagged value of the dependent variable is added as an explanatory variable, each regression is tested for serial correlation with the Godfrey-Breusch test.

In the first regression of Table I., an OLS first order autoregression is displayed. That is, a regression of the series onto its own lag. The positive sign indicates that a positive value of the investment quote in the past is associated with a positive value of the investment quote in the future; the variables move in the same direction. The significant t-value indicates that the lagged investment quote is an important determinant of the current investment quote.

The second regression in Table I. shows the relationship of the investment quote, lagged investment quote as independent variable and the lagged value of the rentier income share. In Table II. this regression is shown, but now the financial assets are used as the financialisation measure. Although in both regressions the

financialisation indicator is not significant, the coeffic ients have the expected negative signs. The low t- values could be explained by missing explanatory variables. Again the autoregressive terms are highly significant.

(17)

17 Regression 3 in Table I. and regression 2 in Table II. display a PAM

specification. The PAM is a special case of the more general ADL model, and comprises a static as well as a dynamic part. OLS estimation is applicable whenever the error term is not autocorrelated, the dependent and explanatory variables are jointly stationary and weakly dependent, no outliers and no multicollinearity exists (Stock & Watson, 2007). The lag values of the dependent as well as the explanatory variables entered this regression. To avoid endogeneity problems, the

contemporaneous values of the explanatory variables are excluded. In both regressions, the autoregressive term and the profit rate appear to be significant, whereas capacity utilization and the financialisation measures are not. In Table I. capacity utilization has a negative sign, while we expect a positive relationship with the investment quote. Both, the profit rate and the financialisation measure RSNF have the expected signs. In Table II. the financialisation measure FA has a positive sign while a negative sign was expected. In this regression the profit rate and capacity utilization also have positive signs, which was expected. The opposite signs of

capacity utilization and the financialisation measure FA respectively, might be caused by their high correlation. Also, this might inflate their standard e rrors, and lowers their significance.

Regression 4 in Table I. and regression 3 in Table II. summarises the results of a reparameterised ADL model. An ADL model may contain lags of the dependent variable and distributed lags of one or more other explanatory variables (Stock & Watson, 2007). Common practice with ADL models is to include many lags of each explanatory variable and gradually reduce this number by dropping the most

insignificant lags, step-by-step. Including a higher order of lags also reduces the amount of observation points. Due to the already low number of observation points it is chosen to include only first lags. Because the independent variables might be integrated of order one, all variables, except for the lagged dependent, entered the regression in level and in difference form to check whether the PAM is a reasonable special case of the more conventional ADL model. This is done to ensure that the PAM results are not 'spurious'. This strategy is replicated from the study of Stockhammer (2004). This form of the ADL model is the starting point to narrow down the number of variables. The general to specific method is used to choose

(18)

18 whether the variables should be included in levels or differences. In table I., based on the most significant t-values, the profit share as well as the financialisation measure suggested to be in levels, whereas for capacity utilization it might be more appropriate to use the difference form. The variables have its expected signs, and the measure of financialisation appears to be significant at the 10-percent level. Using the same method in Table II. suggest a somewhat different approach; all variables should be in difference form. The results are remarkable, as the profit sign as well as the

financialisation measure have the opposite sign of what was expected. Though, they are not significant. Compared with the PAM specification, both of the financialisation measures keep their sign and stay in the same order of magnitude. This indicates that the estimates of the PAM specification were not because of spurious results.

Again, the autoregressive term is highly significant in both regressions.

As the rentier payments increased along with the rentier income, an extra regression in table I. is included to control for the case that the financialisation measure is picking up some effects of the rentier payments. Although the financialisation measure looses its significance, it keeps its negative sign. A reason for the insignificance of both variables might be their high correlation.

In both tables, the Godfrey-Breusch test for autocorrelation is showed in the last column. None of the tests appeared to be significant, so the null- hypothesis of 'no serial correlation' could not be rejected, and thus autocorrelation do not appear to be a problem.

(19)

19

Table I. Regression Results

Dependent Variable: Investment Quote (IQ) Financialisation Measure RSNF Variables ₁ ₂ ₃ ₄ ₅ 0,828 *** 0,772 ** 0,849 *** 0,834 *** 0,817 *** (,119) (,143) (,146) (,143) (,144) 6,98 5,41 5,81 5,83 5,67 0,457 ** 0,300 0,325 * (,184) (,187) (,189) 2,49 1,61 1,72 -0,107 (,121) -0,88 0,174 0,210 * (,120) (,126) 1,45 1,67 -0,045 -0,153 -0,152 * -0,102 (,063) (,095) (,091) (,105) -0,72 -1,60 -1,67 -0,98 -0,109 (,113) -0,97 0,029 -0,090 -0,124

-,0727 (,021) (,086) (,076) (.0836) 1,41 -1,05 -1,62 -0,87 R Square 0,7303 0,7382 0,8147 0,8051 0,8245 Adj. R Square 0,7153 0,7074 0,7653 0,7685 0,7570 GB 1,059 2,589 0 2,227 0,517

Standard errors are in parentheses and in italic, t-values are in italic

*, **, *** denote significance at 10, 5 and 1percent, respectively

GB is the Godfrey-Breusch test for

(20)

20

Table II. Regression Results

Dependent Variable: Investment Quote (IQ) Financialisation Measure FA Variables ₁ ₂ ₃ 0,821 *** 0,819 ** 0,652 *** (,123) (,159) (,130) 6,69 5,17 5,04 0,324 * (,178) 1,82 -0,450 (,276) -1,63 -0,206 (,178) -1,62 0,500 ** (,201) 2,49 -0,030 0,016 (,079) (,085) -0,38 0,18 0,043 (,056) 0,78 0,033 0,000 0,058 ** (,023) (,072) (,022) 1,41 0,00 2,59 0,7325 0,7834 0,8166 Adj. R Square 0,7011 0,7256 0,7642 GB 0,879 0,103 1,706

Standard errors are in parentheses and in italic, t-values are in

italic

*, **, *** denote significance at 10, 5 and

1 percent respectively

(21)

21

V. Discussion

The regression results show some support for the occurrence of fina ncialisation at non- financial companies in the Netherlands; the RSNF financialisation measure is consistently negatively correlated with fixed investments and appeared to be

significant or nearly significant in some regressions. Though, the outcomes should be taken with caution. Unfortunately, due to data availability and data revisions, only the time-period from 1995 to 2015 could be empirically studied. Some of the variables were available on quarterly basis but various crucial variable s were only on annual basis. Therefore yearly data was used. Because of the low amount of observations and the relatively high number of parameters, multicollinearity is a problem and thus the estimates might be biased. In the closely followed research of Stockhammer (2004), high correlations among variables are also pointed out as a problem, however a correlation matrix is left out. Therefore it is not clear how severe his correlations exactly are. Leaving out highly correlated variables is an option, as those variables are reflecting the same occurrence. However, leaving out important explanatory variables could result in omitted variable bias, and co uld affect the remaining explanatory variables.

It should also be noted that the observed time period 1995-2015 covered several instable events. First the dot.com bubble around 2000, shortly after the 9/11 terrorist attacks in 2001, and most recently the 2008 credit crisis. These occurrences could influence the investment quote negatively. Because of the low amount of data points, a Chow test for a structural break was not applicable.Desirably, empirical research should be done for the time period 1965 until now based on quarterly data, as this will also cover the years preceding the financialisation got started. Also

fundamental changes in economic policy or financial crises should be taken into account, either via a Chow test for a structural break, or by adding dummies for these special occurrences. It is not clear whether the data is not publicly available, or whether the data does not exist at all.

Additionally, the variables used covered the whole sector of non-financial businesses. Thus, no distinction has been made between small or large firms, or the type of industry the firms were doing business in. Several Dutch economic advice

(22)

22 reports show there are differences in the usage of financial assets among large

(multinationals) and smaller firms. Besides, tt would be interesting to distinguish the investment behaviour of manufacturing firms compared to service-driven firms. This would reveal whether companies where investing in fixed assets is of high

importance, also show a declining capital accumulation trend. Further research sho uld be done on firm- level basis.

Furthermore, as is already stated by previous studies, investment eq uations are hard to formulate. This might be the reason the standard variables in this research have

problems in explaining the investment quote. The variables capacity utilization as well as the profit share are not consistently significant. It looks like the lagged investment quote itself explains most of the current investment quote. The lack of a precise measure for capacity utilization and the usage of a somewhat unconventional proxy might be a determinant of these kinds of problems.

Also the two measures of financialisation, the rentier income share and the amount of financial assets non-financial companies own, have its limitations. These financialisation indicators do not cover the whole theoretical story. By covering the crowding-out effect, I tried to broaden the existing literature that mainly focuses on dividend and income as financialisation indicator. Unfortunately, this measure have problems in explaining the investment quote, and show different relationships.

Source OECD, financial assets relative to total assets in % 4,00% 6,00% 8,00% 10,00% 12,00% 14,00% 16,00% 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 Finacial Assets Rate

(23)

23 This inconclusive relationship might be explained by the above graph. The downward slope of financial assets also contradicts the explained theory.

Additionally other significant aspects are excluded. For example the

repurchases of own stock is particularly in the Netherlands an important determinant of the growing importance of shareholder value and indirectly an explanation of the financialisation at non- financial businesses. Unfortunately, due to the lack of data this effect is not taken into account.

Moreover, the empirical part doesn’t capture the finance structure of those companies. It is not clear whether these increased financial assets are financed by debt or equity, and in how far we can speak of a crowding-out effect. But the graph in the literature part show that financial assets are increasing stronger than the issued debt so it shows some support, must be also financed out of internal funds.

Finally, the measure of fixed investment itself is not free of problems. Different studies use other sorts of indicators to show the movement of fixed investments over time. Other papers looked at fixed investments relative to capital stock (Orhangazi, 2008), or to the growth rate of gross capital stock.

Source: CBS, author’s own calculations. Gross investments relative to capital stock & the growth rate of gross investments at non-financial corporations (NFC’S) in millions of euro’s

These charts show another picture compared to the plot of the investment quote. Instead of a straight decreasing line, the upper charts do not show a

straightforward direction. The first two dips of the investment growth rate might be an 0,07 0,075 0,08 0,085 0,09 0,095 0,1 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015

Investment relative to capital stock -0,1 -0,05 0 0,05 0,1 0,15 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015

(24)

24 explanation of the burst of the dot.com bubble and the credit crisis respectively. The cause of the third dip is not clear. In the regressions performed in this study, we experimented with different measures of fixed investments. Although the values of the coefficients differed somewhat, most of the variables kept its signs. As the focus in the theoretical part was merely on the investment quote, we decided to use this measure also in the empirical part. By showing these charts it must be pointed out that it depends on how the fixed investments are measured, what direction these charts are showing, despite the fact some Dutch advisory reports only emphasise the strongly decreasing investment quote.

(25)

25

VI. Conclusion

This study analysed the potential (negative) effects of the growing size and importance of the financial sector on the investment behaviour of non- financial companies. While the investment quote of the Netherlands decreased significantly over the past twenty years, at the same time the financial sector increased remarkably. In this thesis the linkage between them is studied.

According to the literature, financialisation at non-financial businesses decreased the acquisition of fixed investments through two channels. First,

investments in financial assets crowds out the funds available for the investment in fixed assets. Instead of investing in goods that enhance their core business, firms tend to use their funds for financing activities. Second, the increased importance of

shareholder value tempted managers to increase dividends and repurchase shares to higher the value of stock. Especially the buybacks of stocks appear to be significantly higher in the Netherlands compared to other European countries. This also means that less funds are available for capital accumulation.

These phenomena were tested empirically mainly based on a model employed by Stockhammer (2004). The results show some support for the hypotheses that decreasing investment quote is a cause of the financialisation of non-financial businesses in the Netherlands. Most of the times the financialisation variables had a negative sign, though only once the measure appeared to be significant. Lack of data availability might have influenced the outcomes.

More research should be done in this field. A study on firm- level basis would enable to distinguish the financialisation character and its influence on investment behaviour among small and large firms as well as among different sorts of non-financial industries.

(26)

26

References

Boot, A.W.A. (2010) De ontwortelde onderneming. Ondernemingen overgeleverd aan financiers? 2e druk, Assen: Van Gorcum.

Boot, A. W. A. (2014). Financial Sector in Flux. Journal of Money, Credit, and Banking, 46(1), 129-135.

Cecchetti, S.G. en E. Kharroubi (2015) ‘Why does financial sector growth crowd out real economic growth’, bis Working papers 490, Bazel: Bank for International settlements

Clark, J.M. (1917) - Business Acceleration and the Law of Demand: a Technical Factor in Economic Cycles, Journal of Political Economy, 25, (3), 217- 235.

Croix, de la, D. & O. Licandro (1993) - the q-Theory of Investment under Unit Root Test, Cahiers Economiques de Bruxelles, 139, (3), 329-339:

Brussel.

CPB Discussion Paper | 266. Banks, Financial Markets and Growth in Developed Countries A survey of the empirical literature

Fazzari, S. and Mott, T. 1986. The investment theories of Kalecki and Keynes: an empirical study of firm data, 1970–1982, Journal of Post Keynesian Economics, vol. IX, 171–87

Grossman, S. J., & Hart, O. D. (1982). Corporate Financial Structure and Managerial Incentives. National Bureau of Economic Research.

Hebbink, G. & Amerongen. Van., E. (2002) - een Q-Investeringsmodel voor Nederland, De Nederlandsche Bank, Meb 2002-10: Amsterdam. Jansen, C., & Ligthart. M. (2014). Spaaroverschot niet- financiele bedrijven:

(27)

27 Karwowski, E., Shabani, M., & Stockhammer, E. (2016). Financialisati Dimensions and Determinants. A Cross-country Study. Post Keynesian Economics Study Group Working Paper 1619

Klundert, T. (2016). Financialisering en economische groei. TPE Digitaal, 10(3), 14-26.

Kopcke, R. and Brauman, R. 2001. The performance of traditional macroeconomic models of businesses’ investment spending, New England Economic Review, vol. 2, 3–39

Krippner, G. 2005. The financialization of the American economy, Socio-Economic Review, vol. 3, no. 2, 173–208

Lazonick, W. & O’Sullivan, M. (2000). Maximizing shareholder value: a new ideology for corporate governance. Economy and Society Volume 29 Number 1: 13–35.

Levine, R. (1997). Financial Development and Economic Growth: Views and

Agenda. Journal of Economic Literature Vol. XXXV (June 1997), pp. 688–726 Kneer, Ch. (2013) ‘Finance as a magnet for the best and brightest: Implications for

the real economy’, dnb Working Paper 392.

Orhangazi, Ö., (2008). Financialisation and capital accumula tion in the non- financial corporate sector: A theoretical and empirical investigation on the US

economy: 1973-2003. Cambridge Journal of Economics

Philippon, T. en A. Reshef (2012) ‘Wages and human capital in the us Finance industry: 1909-2006’, The Quarterly Journal of Economics, 127, 4: 1551-1609.

(28)

28 Power, D., Epstein, G., & Abrena, M. (2003).Trends In The Rentier Income Share In

OECD Countries, 1960-2000. Political Economy Research Institute, Workingpaper Series.

Rappaport, A., 2005, The Economics of Short-Term Performance Obsession, Financial Analyst Journal, vol. 61:.65-79.

Ruis, A., Span, T., & Verhoeven, W. H. J. (2012). Dynamiek van Investeringen in Nederland. EIM: onderdeel van Panteia.

Stockhammer E., 2004. Financialization and the slowdown of accumulation. Cambridge Journal of Economics 28, 5: 719-74

WRR Rapport 2016: Samenleving en financiele sector in evenwicht

Tinbergen, J. (1938) - Statistical Evidence on the Acceleration Principle, Economica New Series, 5, (18), 164-176.

Turner, A. (2012). Credit creation and social optimality. International Review of Financial Analysis 26z; 142-15

Zwan, N. van der (2014) ‘Making sense of financialization’, Socio-economic Review 12, 1: 99-129.

(29)

29

Appendix

Table III.

Variable Definition Calculation Source

Investment Quote Gross investments

relative to value added of non- financial corporations (NFC)

CBS

National Accounts Statline Database Capacity Utilization Value added of NFC's

relative to fixed assets

OECD,

National Accounts Database

Profit Share Gross operating surplus

NFC's relative to value added of NFC's

Eurostat & CBS database

respectively Rentier Share NFC Interest + Dividend

income relative to value added NFC

OECD National Accounts & CBS Statline Financial Assets Rate Financial assets relative

to total assets of NFC's

OECD,

National Accounts Database

Rentier Payments NFC Interest + Dividend payments of NFC's

OECD National Accounts & CBS Statline

Table IV. Correlation Matrix

1.0 0.8 1.0 -0.4 -0.6 1.0 -0.1 -0.3 -0.3 1.0 -0.5 -0.6 0.8 0.2 1.0 0.3 0.1 0.1 0.7 0.4 1.0 -0.4 -0.5 0.8 0.2 0.8 -0.02 1.0 0.1 0.1 0.1 0.6 0.2 0.4 0.3 1.0 -0.4 -0.5 0.7 0.5 0.7 0.2 0.8 0.5 1.0 -0.1 -0.1 0.2 0.3 0.3 0.2 0.1 -0.002 0.5 1.0 0.2 0.1 -0.2 -0.00 -0.2 0.1 -0.002 0.5 0.06 0.6 1.0

Financialisation and its effects on fixed investment at non-financial businesses in the Netherlands