What is the effect of European banks’ R&D expenditures on their sales and employment?

(1)

What is the effect of European

banks’ R&D expenditures on

their sales and employment?

MSc International Economics

and Business thesis

(2)

Master of Science International Economics and Business Thesis

What effect do R&D investments of European banks

have on their sales and employment?

University: Rijksuniversity Groningen

Postbus 72

9700 AB Groningen

Company: KPMG N.V. Advisory

(3)

3 Preface

This master thesis is written as final paper of the study International Economics and Business

given at the Rijks university of Groningen, in the year 2014 – 2015. After starting in 2010 with

my bachelor study International Business and Management, this master thesis is the end of 5

years of knowledge gathering in the Business and Economic field. ...

(4)

4 ...

1. Introduction ...6 1.1 Financial innovation ...6 1.2 Research question ...7 1.3 Paper structure ...8 2. Theoretical framework ...9 2.1 Innovation ...9 2.2 Growth ...10 2.3 R&D ...11 2.4 Performance ...12 2.5 Hypotheses ...14 3. Methodology ...16 3.1 Performance measurements ...16 3.2 Dependent variable ...17 3.3 Independent variable ...18 3.4 Control variables ...18 3.5 Models designed ...20 4. Results ...23

4.1 Total R&D expenditure banks ...23

4.2 Hofstede’s PDI and summary statistics ...24

4.3 Regression sales/employees to R&D ...25

4.4 Regression employees/R&D ...26

5. Conclusion ...27

6. Discussion and Limitations ...29

7. References ...31

(5)

5 Abstract

Innovation seems to be the holy grail to success, a lot is written and said, however, is our current banking system entirely prepared for this innovation push? Although innovation is hard to measure, R&D investments provide us with an indicator of banks initiatives and desire to innovate. This thesis searches for the correlation between this desire to innovate, which might lead to actual innovation, and the sales/employment of European banks. What kind of innovation do we see in the financial sector and how is the commercial bank going to develop?

(6)

6 Introduction

“Everybody talks about financial innovation, but (almost) nobody empirically tests hypotheses about it”

Frame and White (2004)

It’s hard to ignore the concept of ‘crowdfunding’ after reviewing the growth of U.S.-based; Lending Club platform. This startup grew by 1 billion dollar in the first quarter of 2014 alone and has a loan portfolio of 4 billion dollar. Lending club is rapidly growing and a prime example of a financial technology (FinTech) initiative that is transforming the financial sector. The idea behind Lending club is to attract private investors to lend directly to individuals and small business, called peer to peer lending. By using technology, that helps it better assess credit risks and service loans more efficiently than traditional banks, the company offers lower rates and higher returns for investors. Additional to Lending Club, there are innovative companies active in the field of transactions, crypto currencies, peer to peer payments and fully integrated banking solutions. All are typical examples of financial innovation (KPMG, 2014).

(7)

7

account balances, pay bills and transfer funds through mobile devices instead of visiting banks on the corner of the street (Gu, Lee, Suh, 2009).

This thesis provides more insight into the relationship between the innovation initiatives of European bank as measured by R&D investments and their sales/employees ratio. While the issues in measuring innovation and productivity in the banking sector remains, the thesis makes a first attempt to underscore the possible relationship between the two, using available indicators. The main research question I pose is ‘What

effect do R&D investments of European banks have on their sales divided by employment ratio?’. To

answer this question, I measure the initiative to innovate by R&D investments and compare it with the sales/employees ratio per bank.

In addition, the thesis reviews the relationship between employment and R&D expenditures, to see if the decrease in jobs in the financial sector is explained by the increase in innovative initiatives. And so the sub-research question answered in this thesis is ‘Is employment in European banks negatively related to innovation initiatives measured as R&D expenses?’. Even though, general evidence concerning the impact of innovation on employment varies across sectors, it has been often argued that the effect of innovation on employment in the financial sector is negative (Evangelista and Savona, 2002). Apparently, in the financial sector the widespread use of information and communication technology has been resulting in displacement of workers at the lower segment of skill level. After the financial crisis of 2008, the banking world experiences large re-structuring and lay-offs, though these banks are still aiming for faster growth. Innovation might help banks reduce jobs, still keeping their competitive position in the market by implementing ICT (capital) (Evangelista, 2002).

There is not one standard form or definition of innovation. Innovation can be radical, disruptive, incremental or on the job (World Bank, 2005). Innovation could as well be product focused, process focused or price reducing (Worldbank, 2005). Even though the measurement of innovation is widely debated, this thesis follows the most common practice, and use R&D investments to represent de initiative to innovate. As mentioned, R&D is only an input, more on this can be found in the literature review. This thesis fills a gap in the literature, as far as I know, previous quantitate research on the R&D expenses of European banks compared to their sales/employment ratio has never been done, connecting R&D expenditures to its effects over time. Innovation and performance is measured using several datasets combined. The findings of this thesis provides banks, clients, companies, and researchers with valuable insights into the importance of the ongoing (technological) movement in the financial world. The motivation for choosing R&D as innovation initiative and the influences of innovation on the financial sector asks for the highest attention of all banks, financial institutions but also technology companies. This is because our society is changing rapidly, people are getting used to technology and expect businesses and banks to be part of this digital movement (KPMG, 2014).

(8)

8

use in our innovative world. This because banks might face disruption, startups, technology companies and customers which are all taking over tasks, priory performed by banks (Liebenau et al., 2014). The managerial relevance of this thesis can be found in the importance of the outcome for financial managements. Banks managers may benefit from this paper as it provide the facts, numbers and literature available on financial innovation and its effects over time. This thesis explains that financial innovation is related to financial technology and that research and development are a necessity, in the financial service sector to all stakeholders involved.

Paper structure

(9)

9 Theoretical framework

“Innovation is the central issue in economic prosperity”

Michael Porter

In this section, I provide a review of existing literature concerning R&D investments, financial innovation, financial technology and its impact on sales and employment. In the literature, papers on financial innovation, its importance, and impact are hard to find. Therefore, this review focusses in general on the available scattered literature on the impact of innovation in the financial sector and on various aspects of economic performance.

Tufano (2003) and Subramenian (2012) both take R&D as a proxy for innovation initiatives. As Tufano states, R&D can be a motive for innovation and so explain the urge to aim for innovation practices. Innovation is any new or substantially improved good or service which has been commercialized or any new or substantially improved process used for the commercial production of services and goods. ‘New’ means new to the type of business (Rogers, 1998). Innovation is hard to define because when is an innovation new to a certain business? Technologies are ‘rules and ideas that direct the way goods and services are produced’ (Kemeny, 2010). Technological inventions are new rules and ideas about what to produce and how to do it. This results in technological innovations, when new rules and ideas find practical use through being applied and commercialized by entrepreneurs (Naudé and Szirmai, 2013). Technological innovation contributes to higher levels of economic output and can deliver new goods and services that change human lives and capabilities (Lipsey et al., 2005).

(10)

10

Several papers examine the relationship between finance and growth over time (Greenwood and Jovanovic, 1990; Bencivenga and Smith, 1991; Levine, 1991; King and Levine, 1993a). They all mainly insist on the influence of finance allocating capital and so hence long term economic growth. The level of financial development influences growth (Aghion and Howitt, 2009). Michalopoulos et al., suggest that regulations that hinder financial innovation might have a long-lasting negative effect on economic growth. Thus the literature on financial innovation and its effect on growth emphasizes a positive role for innovation, whereas the regulatory environment, if hampers the financial innovation, can have a negative impact on growth. On the other hand, several people claim that not all financial innovations promote growth. Regulators have legitimate concerns about the impact of some financial innovations on financial stability in particular, and the allocation of credit more generally (Beck et al., 2012).

Beck, Chen, Lin and Song (2012), using OECD data, attempt to relate financial innovation in the banking sector to economic growth and bank fragility. They found positive and negative effects of financial innovation, however, a higher level of financial innovation is associated with a stronger relationship between a country’s growth opportunities and capital. The former chairman of the Federal Reserve, Paul Volcker, claims he can find very little evidence that the financial innovations in recent years have done anything to boost the economy (Beck et al., 2012). Chava, Oettl, Subramanian and Subramanian (2012) document empirical support for a key micro level channel-innovation by young private firms, through which the financial sector deregulation affects economic growth. In their paper they compare U.S.A intra-state and inter-state banking deregulations and observe that innovation has been fountainhead of long-term economic growth. There is evidence that finance can affect long-term economic growth by influencing innovative activity. Banks should try to relax financial constraints and boost innovation (Subramanian et al., 2012). Furthermore, financial innovation is positively related to GDP per capita growth and to higher growth rates in industries that rely more on external financing and depend more on innovation (Beck, Chen, Lin and Song, 2012). For instance, Beck et al., find that a specific financial innovation -the introduction of private credit bureaus - results in faster convergence of countries towards the growth path of the most advanced country. They analyzed two views on the impact of innovation: the innovation-growth view and the innovation-fragility view. The innovation-growth view posits a positive perspective on financial innovation that financial innovation helps improve the quality and variety of banking services (Merton 1992; Berger, 2003). In addition, within this perspective, earlier studies have observed, that innovation facilitates risk sharing (Allen and Gale, 1991) and improves allocative efficiency (Ross, 1976; Housten et al., 2010). Innovation has played a key role in reducing the volatility of economic activity in the early parts of the 21st

(11)

11

high income countries. In the paper of Leaven, Levine and Michalopoulus (2009), this was different as they focused on specific financial innovation and private credit bureaus. The paper of Beck et al., finds supportive evidences for both innovation-growth view and innovation-fragility view, both deriving from financial innovation. Banks located in countries with higher pre-crisis levels of financial innovation experienced larger drops in ROA (return on assets) and ROE (return on equity) between 2006 and 2008 (Beck, Chen, Lin and Song, 2012). Financial innovation encourages banks to take on more risk, but also increases volatility. As a conclusion of this first part of the literature review, it is priory written that R&D reflects innovation initiatives, financial innovation is related to technological innovation and both are positively related to (firm) growth.

Research and Development spending (R&D) is considered to be related to innovation initiatives of banks. However, R&D is an input, not an output of innovation. R&D is by the OECD defined as ‘the comprise of creative work undertaken on a systematic basis in order to increase the stock of knowledge and the use of this stock of knowledge to devise new applications’ (OECD, 1993). In the academic literature, papers have been written upon R&D expenditures and its effect on innovation and firm performance, like Garcia-Manjón and Romero-Merino (2012), Morbey (1988), Sher and Yang (2005). As mentioned before, by calculating R&D investments in the financial sector, I want to reflect upon past and current investments of European banks and review what effects do R&D investment have on the banks sales / employees ratio. As also mentioned before, the literature on innovation in manufacturing has focused mainly on patents as a measure of innovation (Beck, Chen, Lin and Song, 2012). Patents are mainly available in the manufacturing sector, but rarely exist in the financial sector and not at all in the European union (Frame and White, 2004) therefore this thesis ignores patents and focuses on R&D expenditures, since I focus on the financial sector, not on the manufacturing sector. Although, patents and research and development expenditures are both highly limited as complete measurements of innovation, it provides us with some quantitative data. R&D, the way it is used in this thesis is described as the “cash investment funded by the companies themselves in research and development” (European scoreboard, 2014). This is further explained in the methodology part of this thesis, however, for now this explains the relation between literature on innovation, R&D and the numerical value in this thesis representing the initiative to innovate.

Morbey (1988) was one of the first to identify that the “decisions made by management today on investment in research and development can influence the viability and growth of a corporation into the 21st_century”.

(12)

12

expenditures on a company’s performance. The time-lag could exist, how do we measure the effects of R&D expenditures over time? However, to compare a company’s or industry’s absolute growth rate over a number of years with the R&D expenditures for an earlier period might be valid and useful. If R&D expenditures, for large established companies, are measured as a percentage of their sales revenue; the year-to-year variations in R&D expenditures are modest (Morbey, 1988). Therefore, sales growth over a time period can be compared with R&D expenditures of an earlier period without detailed knowledge of the exact lag time of R&D effects. As long as a broad time spectrum is considered the results will be sound (Morbey, 1988). Sustained R&D investments often lead a company to long-term growth, whereas growth is measured as sales (Morbey, 1988). Only with sustained R&D investment it is likely that an enterprise will continue to grow. Maybe some companies today are paying for the cut-back support of R&D in the past.

Holak, Parry and Song (1991) find that firm and industry characteristics have an important influence on the research and development – performance relationship. Griliches (1998) reflected on the empirical patterns of firm growth and R&D investment using a quality ladder model interpretation. He presents a specified model of endogenous firm growth and concludes that R&D and innovations are the engines of growth. OECD published a supportive paper written by Sheehan and Wyckoff (2003) in which the most popular indicator is the R&D intensity of a country as measured by the amount of R&D it performs divided by GDP. However, the ability of firms to boost their expenditures on R&D can be influenced by rates of GDP growth as well. R&D is mainly funded from retained earnings especially within larger established firms, implying that that it might be difficult for firms to invest in R&D during periods of slow economic growth (OECD, 2003).

(13)

13

(14)

14

Hofstede’s cultural dimensions are used to represent the cultural component of an economy and the results of their paper indicate that the adoption decisions to ICT are influenced by the cultural setting of the economy considered. Hofstede’s power distance is a significant cultural factor that explains some of the differences in ICT adoption rates between countries (Erumban, Jong, 2006). The value power distance, measured by Hofstede, is used in this thesis to control for the potential cultural effects on productivity.

It is quite clear from the review above that existing evidence suggests that R&D spending has an effect on the sales and total employees of firms. However, the specific impact of innovation in financial sector on the performance of financial sector, in terms of productivity, is less understood. This paper, which aims to address this question in the context of European banks, postulate the following hypothesis.

H1: European banks investing more in R&D will face a higher sales/employees ratio than those which do not.

This hypothesis is also been supported by findings in Garcia-Manjon and Romero-Merino (2012) who observe a positive effect of R&D intensity on the sales growth for a sample of 754 European firms.

The literature is quite clear on the positive effect of innovation on growth and productivity. However, there exist another side to the impact of technological change, which is its potential negative effect on employment. Banks and insurance companies experience a negative impact and a strong skill bias because of innovation (Evangelista and Savona, 2002). In banks the labor saving effects linked to the introduction of ICTs have been more severe. In all these industries the creation of high skilled jobs linked to the introduction of new services has not been strong enough to offset the destruction of low qualified jobs linked to the introduction of labor saving technologies (Evangelista and Savona, 2002). It has been the case that the number of job loss in the financial sector in the recent years has been quite substantial. While part of these might be an effect of the global financial crisis, it is important to see the relationship between employment and innovation. As observed by Garcia and Romero (2012) innovation remains highly significant nowadays, since firms want to reduce their costs and inputs and increase their outputs and time efficiency. This leads us to an important question, which is the unemployment in the financial sector and the effect of technology on disruption of the banking system. I examine this issue, with the following hypothesis:

H2: There exist a negative relationship between R&D expenditure and headcount (employees) in European banks.

(15)

15

(16)

16 Methodology

To answer the research question ‘What effect do R&D investments of European banks have on their sales

and employment?’ we need to understand the relationship between R&D investments and the

sales/employment ration of European banks. This in a quantitative way for which an econometric regression approach is used. In this approach the indicators of performance are sales per employees. I am conscious that this sales/employees ratio does not measure any financial performance, neither a productivity. However, it provides us with a measure of how sales, which could also be an outcome of something else than labor, for example capital, increases or decreases. And if the amount of employees, as an input, increases or decreases, likewise, employees do not have to be the only input, this can also be capital. For the ease, this ratio does provide us with a reflection of the banks performance, keep in mind these previously named limitations though; sales and employees are limited output/input measurements. The outcome of the sales/employees ratio, is offset against the measure of the initiative to innovate; research and development expenditures, and a set of control variables. For now, sales/employees is used as an (limited) indicator of performance.

Several different measurements of performance exist. General performance indicators as sales, profitability or labor productivity. As well as ratio’s, like return on investment (ROI), return on assets (ROA), return on equity (ROE) can be assessed to measure a bank’s performance (Bureau of Labor Statistics). Sales is considered to be one of the original commercial banking measures (Royster, 2012). The US Bureau of Labor Statistics (BLS) confirms this, and states that output measures are mainly based on annual sales or revenue data (Royster, 2012). Business performance measurements generally used to evaluate businesses are sales, sales margin and average number of hours employed per customer (Wiele et al., 2002). Sales is pointed out to be the most suitable indicator to measure growth and productivity (Hoy et al, 1992). Sales is yet further explained in the literature section.

(17)

17

productivity effect and an activity effect (Miller, 1987; Kurosawa, 1975). From this, it is possible to conclude that profitability is seen as another performance indicator.

Labor productivity and economic growth are two key factors in any economy. Labor productivity depends on the availability and quality of labor resources and makes that labor productivity heavily influences the production process and production costs (Emsina, 2014). The main definition used for labor productivity is the real output per hour worked. The Bureau of Labor Statistics constructs two basic measures; output and total labor hours. Output measures are seen as real value added in the industry based on national accounts (Calcagnini, 2013). Labor productivity can be seen as a partial productivity index, typically describing the relation of the output of the process to the used capacity given in time unites, or the number of persons involved. Labor productivity is relying on the output per hour, per worker (Czumanski, 2012). Another measurement of performance is the ROA, the return on assets ratio. Measures such as the ROA, ROE and other financial ratios have been accepted by most practitioners as possible indicators of performance (Doyle, 1994). ROA is an indicator how profitable a company is relative to its total assets. Profitability is often measured as net income. Assets are a measure of firm seize, net income a measure of return (Andrés et al., 2008).

Sales is discussed in a previous paragraph, market capitalization and added value are not used in this paper and thus undiscussed. However, employment growth is an interesting variable representing growth and performance (Manjón et al., 2012). Policy makers often view employment growth as an indicator of an healthy economy (Birch, 1979). Managers of emerging firms often perceive this differently and find employment growth to be a particular dilemma for independent business owner-managers. Employment growth is namely associated with desirable and undesirable consequences (Wiklund et al., 2003). Hiring new employees and expanding employment may be a response to opportunities to expand, provide employment and gain market share. However, it could also be asking for responsibility, administrative hassle and high cost (Davidsson and Wiklund, 2013). Resulting in not pursuing growth or alternatively peruse growth in revenues only when it can be done without hiring additional permanent employees (Davidsson and Wiklund, 2013).

(18)

18

The fact that both variables represent, as stated in previous literature, a kind of performance, however, are not heavily correlated, made me decide to use both. Recall that growth is a performance indicator and often multifaceted, the ‘performance indicator’ used in this thesis, sales/employment is also multifaceted and provides us with sales per employee. As mentioned, sales is the most common used variable when measuring firms performance (Hoy et al., 1992) and represents an output. While to me, employment is a very interesting variable in the financial service sector, because of all the recent news on job losses and employment diminishing. As mentioned before, sales/employees are a limited measure of performance and do not at all measure any financial performance or productivity.

The other variables, profitability, labor productivity per hour, return on asset and total value added are all not considered to be the dependent variable in this thesis. Profitability has a pricing element focus and no information was available on the costs of each transaction, each year, made by European banks. This information was too detailed and impossible to collect. The exact time schedules of all employees of each bank could not be found. Labor productivity focusses on the output per employee, per hour, divided by total revenue. This detailed information of working hours per employee, per bank, was impossible to find over the 10 years considered in the thesis. ROA consists often of assets divided by total income. Neither assets, nor total income are as strongly related to performance as sales and employment.

Sales and employment data used in this thesis is collected from the EU R&D investment scoreboard. The European commission published surveys and reports upon research and innovation, at the Economics of Industrial Research and Innovation website (IRI). This source is used for the three main variables used in this thesis, sales, employment and R&D investments. This thesis uses this information based upon data available from the EU R&D investment scoreboards. The IRI database provides time series data for a range of industries from aerospace and defense to chemicals, automobiles, pharmaceutics and software. Banks are one of the industries included and the database provide annual information on their R&D expenditures, employees, sales and profit. The scoreboards range from 2004 until 2013, leaving us with up to date information. The sales term used in the European scoreboard and as well in this paper, when concerning the sales of banks, is defined as the ‘Total (operating) income’ plus any insurance income. For insurance companies, sales are defined as “Gross premiums written” plus any banking income. In this thesis I only use the sales of banks as financial institutions. The number of employees is the total consolidated average employees or year-end employees if average is not stated per country.

(19)

19

their innovation investments (Frame and White, 2004). However, the general IRI database provides time series data for a range of industries from aerospace and defense to chemicals, automobiles, pharmaceutics and software applications. Out of these data sets I collected the data focusing on banks and made one final database covering 10 years of sales, employment and R&D investments. The term R&D investment used in this thesis refers to corporate investment in R&D, funded by companies themselves and their subsidiaries. R&D financed by governments, or other third parties are not included. Besides, it excludes a given company’s share of any associated company or minority joint venture R&D investment. The definitions of R&D used by companies, which are following accepted international accounting standards, are in line with definitions used in official statistics (OECD, Frascati Manual).

In the European scoreboard ‘research’ is defined as the original and planned investigation undertaken with the prospect of gaining new scientific or technical knowledge and understanding. Expenditure on the research is recognized as an expense when it is incurred (European scoreboard, 2014). Development is the application of research findings or other knowledge to a plan or design for the production of new or substantially improved materials, devices, products, processes, systems or services before the start of commercial production or use. Development costs are capitalized when they meet certain criteria and when it can be demonstrated that the asset will generate probable future economic benefits (European scoreboard, 2014).

Furthermore, the IRI database helps to identify the parent entity of firms. The terms ‘Finnish company, German company, Dutch company’ are used in the source report to refer to a company whose parent entity has chosen to locate its registered office in that country stated. The term real R&D is used to refer to the companies R&D/employees ratio, to measure the ratio of company’s R&D investments to its employees. The years which are mentioned throughout this data, refer to the company’s published account for the given financial year. Many companies do have discretion in the choice of accounting period end and therefore the current year set can include accounts ending on a range of dates from the middle of one year to the early stage of next year.

(20)

20

Another control variable added, nearly impossible to influence by a company, is the power distance defined by Geert Hofstede (1988). The variable power distance explains the extent to which the less powerful members of society accept that power is distributed unequally (Hofstede, 1988). In organizations this distribution of power is reflected in the hierarchy. Authority, centralized decision structures and the use of formal rules are characteristics of organizations in countries with high power distance. Organizations located in these high power distant countries, have been associated with lower rates of innovation and adaption (Zmud, 1982). Cultures with high power distance often are expected to be less open-minded, as it involves decision making on issues that did not happen before and little information is about (Lee and Peterson, 2000). In this thesis, power distance covers a cultural country aspect, while GDP per capita focusses on economic welfare and population. Power distance might have an effect on the sales/employees ratio which I use in this thesis. Countries with a high power distance score will face lower initiatives of ICT adaption (Erumban, 2006). As we saw, ICT The data on power distance is collected from the original Geert Hofstede website, for all countries used in our dataset.

Two other control variables, which are possible to influence by the bank itself, are its total assets and total equity. Both might influence the sales/employees ratio in different manners. Total assets reviews the amount of capital possession by the bank and is often referred to as the banks’ size (Pervan and Visic, 2012). Pervan and Visic, focus on firm size and evaluate its influence on firm performance, they consider total assets, asset turnover and employment as indicators of company size. Their conclusion drawn is a weak positive effect between a firm’s assets (size) and performance. This influence of total assets on firm performance, made me add this as the third control variable in my regression. The total equity reveals the power distribution and explains the financial structure of banks. Equity as derived in this thesis are the total assets minus all liabilities and debts paid. The data on total equity and total assets are both derived from Bankscope. Following the above discussion, I formulate a regression equation, relating banking performance to innovation and the other control variables as:

lnProd = α + β0R&D + β1Tot.ass + β2Tot.eq + β3GDPpc + β4PDI (1)

Where lnProd is the log of labor productivity measured as sales/employment, R&D is the R&D/employment ratio, and the control variables, Tot.ass is total assets, Eq is equity, Inc is income and PDI is the cultural dimension power distance, α is the intercept. The second equitation is designed to review the correlation between employees and innovation. The banks innovation incentive is again captured in the R&D expense, while the control variables are kept the same, as in the previous regression. Total assets, total equity, GDP per capita and power distance.

(21)

21

The regression might give us more insight into the effects of innovation on the amount of employees. Employment can be influenced as well by the GDP per capita or power distance.

The data collection of R&D expenditures, sales, employees, total assets and GDP, I extended with the numbers on the total of a countries bank assets on its GDP in percentages. To measure the size of the banking system, a country’s banking assets divided by the country’s gross domestic product (GDP) is a yardstick often applied (Schoenmaker, 2012). Banking assets are everything that a bank owns, e.g. loans, physical assets and securities and can be found on the left side of the balance sheet. Banking assets usually account for over 100% GDP in developed economies. A summary of the banking assets/GDP of the countries used in the result section of this thesis is stated below.

Table 1: Banking assets per country/GDP%

Country 2007 2011 2013 Belgium 386% 316% 245% Cz. Republic 99% 111% 120% Denmark 246% 244% 279% Finland 152% 280% 220% France 298% 368% 381% Germany 300% 311% 280% Ireland 688% 600% 422% Italy 206% 208% 210% Luxembourg 2570% 1872% 1600% Poland 69% 83% 91% Portugal 251% 291% 273% Spain 0% 295% 260% Sweden 186% 209% 220% Netherlands 548% 441% 379% UK 423% 442% 375%

Source: European central bank, 2013

(22)

22

The dataset derived from the IRI, the European scoreboard consists out of 10 years, 47 different banks responded their R&D investments, these banks were divided over 15 European countries (Belgium, Czech Republic, Denmark, Finland, France, Germany, Ireland, Italy, Luxembourg, Poland, Portugal, Spain, Sweden, the Netherlands and the United Kingdom). The data is measured over 10 years, starting from 2004, ending with the last variable from 2013. Since the bank sets over the year changes, I can only find general effects over time concerning R&D investments and the banks differences in employment and sales. The total amount of data accounts for 235 observations, which leaves us with an average of 23 observations per year. These observations consist of a different set of banks per year, one year it might be possible Banco Santander is included in the dataset since they reported their investments, other years they did not and are left out. All banks included in the data set can be found in the appendix attached to this thesis.

I am conscious this data set is limited and does not include all European banks. However, the banks included in the data set appeared to be the largest banks of their country. In the figure below the largest banking groups of the European Union are displayed and divided by their consolidates assets. The GDP of its home country (domestic GDP) and the GDP of the European union as a whole are used to depict the banking assets. This figure is designed based on numbers collected in 2011, I use it here as a benchmark to show the including of large banking groups in my dataset. The comparison to domestic GDP and EU GDP provides a yardstick to measure the size of a countries banking sector (Schoenmaker, Werkhoven, 2011).

Figure 1: Large EU banking groups consolidated assets against GDP%

(23)

23

(24)

24 Results

This chapter provides us with the outcomes I found while using Stata and running the two regressions based on the dataset explained in the methodology. As mentioned in the methodology the amount of R&D investment is representing the initiative to innovate. The data is gathered from the European investment scoreboard, the World bank and Eurostat, over a time span of 10 years, 2004 – 2013. I take into account this dataset is not the complete R&D investment done by all banks the past 10 years, since some banks might not want to provide competitors with their R&D investment data or simply do not track the value of their innovations.

Table 1 illustrates the R&D investment divided by country over the last 10 years projected by the investment scoreboard. I keep in mind the previously recalled division of banking sectors, in which Luxembourg, Ireland, the Netherlands and Belgium represented a large share of banking assets. This makes Luxembourg and Ireland compared to their banking sector score relatively low on R&D investments reported in the IRI scoreboard over the years.

Table 1: Total R&D expenditure of banks by country (2004-2013)

Source: IRI international investment scoreboard

Germany and the United Kingdom are reporting high R&D expenditures. Belgium is the third country reporting high R&D expenditure and Italy the fourth. Italy was impacted fairly by the global crisis of 2008 (Di Quirico, 2010), the high investment in R&D of banks is therefore surprising to me. Finland, the Czech Republic and Spain report low R&D investments. Referring to the power distance dimension, as designed by Hofstede, and used while running the regressions as a control variable for the cultural effect on R&D

0 5 10 15 20 25 30 35 40 45 50

(25)

25

spending. Germany and the United Kingdom do belong to the group of countries with a lower score on power distance. Belgian banks however, score high on R&D investment and on power distance, this would be contradicting, following the literature, but could be an exception. The Czech Republic does score high on power distance and low on R&D expenditure. On average the hypothesis holds as well for European banks and innovation; there exists a negative correlation between power distance and innovation.

Table 2: Hofstede’s Power distance indicators by country

Source: Website Geert Hofstede, cultural differences indicators

The first regression ran, reflects the correlation between banks performance (sales/employees) and innovation (R&D). The regression was completed by adding the control variables explained, GDP per capita, power distance, total assets and total equity. Over 10 years, 47 different banks responded their R&D investments, these banks were divided over 15 European countries (Belgium, Czech Republic, Denmark, Finland, France, Germany, Ireland, Italy, Luxembourg, Poland, Portugal, Spain, Sweden, the Netherlands and the United Kingdom). The data is measured over 10 years, starting from 2004, ending with the last variable from 2013. Additional summarizing information is displayed in this table below.

Table 3. General Summary of dataset used

Variable Obs. Mean Std. Dev. Min. Max.

(26)

26

Total Equity 235 31042 30535 -320 145657

PDI 235 42.35 14.45 18 68

A total of 235 observations were collected. The independent variable, innovation is strongly correlated and significantly related to the dependent variable; performance (sales/employees). Table 4 presents the results from the first regression ran.

Table 4: The effect of R&D expenditures on (sales/employment) Variable Coefficient t statistic P value R&D 9.40 *** 3.98 (0.000) GDP per capita 1.07e-06 *** 2.90 (0.001) Total Assets 7.00e-08 *** 3.52 (0.006) Total Equity -1.22e-06 *** -2.80 (0.004)

PDI -0.0016 ** -2.46 (0.015)

Constant 0.21 *** 4.77 (0.000)

R square 0.3429

Adjusted R square 0.3279

Number of observations 229

*** indicates significance at 1%, ** at 5 % and * at 10%

(27)

27

expect because of the papers written by Morbey, 1988 (sales positively related to R&D investment) and the paper written by Evangelista and Savona, 2002 (employment negatively related to innovation in financial sector).

Since the sales/employees ratio might be related to total assets, as explained in the control variables, an additional question might arise: what happens to the regression model when we divide the R&D expenditures by the total assets of each bank? The size of a bank might influence the R&D expenditures. To do this, R&D is divided by Total Assets, both in million dollars and control for the size of the bank.

Table 5: The effect of (R&D expenditures/total assets) on (sales/employees) Variable Coefficient t statistic P value R&D / Assets 3.505 ** - 1.99 (0.058) Total Assets 0.002 0.83 (0.407) Total Equity 0.115 1.57 (0.119) GDP per capita - 0.12 *** - 3.03 (0.003) PDI 48.51 0.85 (0.394) Constant 1092 ** 2.48 (0.014) R square 0.144 Number of observations 232

Now the results of the regression ran are very different. When we control for the size of the bank and divide their R&D spending by their assets, we find a negative coefficient R&D/Assets compared to Sales/Employees. The R&D/Assets variable coefficient is now even half of the number it was. The significance decreases, only the constant and the GDP per capita is showing a significant outcome. An explanation for this difference might be the large banks spending a very high amount of money on R&D and so the general average is lower than when reviewing the relation between simple R&D spending without any relationship to assets or firm size.

(28)

28

Table 6: The effect of R&D expenditures on employment

Variable Coefficient t statistic P value R&D - 11.23 ** - 1.98 (0.049) Net sales 4.63 *** 41.64 (0.000) Total Assets - 0.01 *** - 2.95 (0.004) Total Equity 0.29 ** 2.46 (0.015) GDP per capita 0.05 0.45 (0.651) PDI 237.91 * 1.77 (0.078) Constant - 5617 - 0.54 (0.587) R square 0.9038 Adjusted R square 0.9012 Number of observations 229

From this second regression we see the negative influence of innovation (R&D) on employment in European banks. The employment decreases when the R&D expense increases. The coefficient reflects a large value, since the amount of employees used in our data set is in thousands. This regression does also fit in the line of reasoning of Evangelista and Savona (2002).

(29)

29

Figure 2: Internet users (by 100 people)

Source: World Development Indicators

Conclusion

The results presented in this thesis imply a strong correlation between the sales/employees ratio in European banks and R&D investments (initiative to innovate). This is supporting the literature written on the subject, previously explained in the theoretical framework. Performance, often measured by sales, shows a strong correlation between innovation and sales (Morbey, 1988). Whereas employment in the financial sector experiences a negative effect of innovation (Evangelista and Savona, 2002). The research question ‘What effects do European banks that spend a lot on R&D experience in their sales/employees ratio?’ Is therefore as follows; banks who invest high on R&D experience a decrease in employees while still facing the same or a higher amount of sales.

(30)

30

employment instead of on sales to understand the change in employment as an effect of R&D investments done.

The conclusion from this paper is as follows. The financial innovation sector (European banks) are facing a reformation. This because ICT entered into the banking world and banks appear to be highly dependent on ICT and innovation technology (Evangelista, 2002). The effect of this technology focused development is a decrease in employment caused by an increase in innovation investments. Since mobile banking, peer to peer lending and fin-tech startups start to becoming popular in our modern world, banks feel the need to aim for higher productivity and therefore we see an increase in R&D spending over the past 10 years. Sales and R&D are strongly correlated (Morbey, 1988), which is another incentive for banks to invest in R&D and technology. The two variables, financial innovation and financial technology are strongly correlated and therefore one will probably encourage the other. As long as technology continues to grow, investments are continuing to be done and so the financial sector again experience growth. Growth, R&D expenditures and sales are positively correlated. Financial innovation and financial technology as well. R&D spending and employment are negatively correlated in the financial sector. This because of ICT taking over jobs formerly performed by the banks employees.

(31)

31 Limitations and Discussion

The research into innovation initiatives of European banks is very immature, as Frame and White stated, more hypothesis should be tested, while using quantitative data, on financial innovation.

Limitations of this research should be taken into account regarding the results. First of all, specific literature on innovation in European banks has not been supported by any quantitative evidence so far. Therefore it was hard to construct a theoretical framework applicable to European banks. A lot has been written on innovation, R&D, sales and employment, but not especially on R&D investments of European banks. The paper written by Beck et al., was the closest attempt to explain innovation in European banks Secondly, the scoreboard data is very limited, over 10 years of time, only 235 banks reported their investment in R&D. This data should be extended with more banks and an increase in transparency of data on innovation. Several hurdles are faced, while aiming for this transparency; all innovations continue to be hard to measure and evaluate. What is the value of a brilliant idea? Or of the implementation of new technology in the production process? Besides the difficulty of measuring innovation, it is yet hard to define innovation. Therefore innovation has been given a definition in this paper which is rather broad, neither R&D has been specified in this thesis. The R&D investments reviewed in this paper could be technological, product focused, process focused, incremental, disruptive or simply hiring more research personal. The exact distribution of the R&D investments inside the European banks was not evident from the data collected and therefore it was decided to continue a wide spreading investigation. More specification concerning the kind of innovation, invested by banks would help researchers to understand possible effects of innovation in the financial service sector. However, I understand this might give away sensitive information to its competition about the innovation initiatives of banks.

(32)

32

After writing this thesis a lot of questions remain unanswered, in which further research should be done. One of these questions is addressing the startup companies entering the field of financial innovation and financial technology. At the beginning of this thesis, a young firm called ‘Lending club’ was mentioned and the immense investment done in this startup facilitating peer to peer lending. An interesting phenomena, young companies taking over tasks formally performed by banks. Since a quantitative database of these companies does not exist, neither there is much data about startup companies, a quantitative analyses in this thesis could not be done.

The question remains, what is the position banks are going to take in our society and how can this transformation be supported? How can banks become modern and attractive to new customers, giving them the feeling they want to be a customer of that specific bank. Is research and development the ultimate manner? Or is simply attraction of new technological focused R&D personal the key to success?

(33)

33 References

Aghion, P. and Howitt, P. (2009) The Economics of Growth, MIT Press.

Allen, F. and Gale, D. (1994) A Welfare Comparison of the German and U.S. Financial Systems, Center for Financial Institutions Working Papers 94-12, Wharton School Center for Financial Institutions.

Andres, F., Corradi, V., Swanson N. (2008) Information in the revision process of real-time datasets,

Working Papers 08-27, Federal Reserve Bank of Philadelphia.

Beck, T., Demirgüc-Kunt, A., and Levine, R. (2000) A new database on financial development and structure.

World Bank Economic Review 14, 597-605.

Beck, T. and Levine, R. (2002) Industry growth and capital allocation: does having a market or bank-based system matter? Journal of Finance Economics 64, 147-180.

Beck, T., Lundberg, M. and Majnoni G. (2006) Financial intermediary development and growth volatility: Do intermediaries dampen or magnify shocks. Journal of International Money and Finance 25, 1146-67.

Beck, T., Tao, C., Chen, L., Song, F. (2012) Financial Innovation: The Bright and the Dark sides.

Working Papers 052012, Hong Kong Institute for Monetary Research.

Bencivenga, V. and Smith, B. (1991) Financial intermediation and endogenous growth, Review of

Economics Studies 58(2), 195–209.

Berger, A. (2003) The economic effects of technological progress: evidence from the banking industry.

Journal of Money, Credit and Banking 35, 141-176.

Bohm-Bawerk, E. (1891) Positive Theory of Capital, Mcmillan and Company [translated by William A. Smart].

(34)

34

Brunnermeier, M. (2009) Deciphering the Liquidity and Credit Crunch 2007-2008. Journal of Economic

Perspectives 23(1): 77-100.

Calcagnini, G., Ferrando, A., Giombini, G. (2013) Multiple Market Imperfections, Firm Profitability and Investment, Working Papers 1305, University of Urbino Carlo Bo, Department of Economics, Society & Politics.

Chava, S., Oettl, A., Subramanian, A., Subramanian, K. (2012) Banking deregulation, bargaining power, and innovation, Unpublished Working Paper, Georgia Institute of Technology.

Czumanski, T. and Lšdding, H. (2012) Integral Analysis of Labor Productivity. Procedia CIRP 3, 55-60.

Diewert, W. (1973) Functional forms for profit and transformation functions, Journal of Economic Theory, Elsevier, 6(3), 284-316.

Doyle, P. (1994). Setting Business Objectives and Measuring Performance. European Management

Journal 12, 123-132.

Dynan, K., Douglas W., Sichel, D. (2006) Can financial innovation help to explain the reduced volatility of economic activity? Journal of Monetary Economics, Elsevier 53(1), 123-150.

Emsina, A. (2014) Labor productivity, economic growth and global competitiveness in post-crisis period,

Procedia - Social and Behavioral Sciences 156, 317 – 321.

Erumban, A. and Jong, S. (2006) Cross Country differences in ICT adoption: A consequence of culture?

Journal of World Business, 41(4), 302-314.

Evangelista, R. and Savona, M. (2002) The Impact of Innovation on Employment in Services: Evidence from Italy, International Review of Applied Economics, 16, 309-18.

Farrell, M. (1957) The measurement of productive efficiency. J. R. Statistic Soc. A 120, 253-281.

Frame, W. and White, L. (2004) Empirical Studies of Financial Innovation: Mostly Talk and Not Much Action? Journal of Economic Literature 42, 116-44.

(35)

35

Omega 20:5/6, 553-568.

Gennaioli, N., Vishny A., Robert W. (2012) A Model of Shadow Banking, Journal of Finance, forthcoming.

Geroski, P. (1995) Innovation and Competitive Advantage, OECD Economics Department Working Papers 159, OECD Publishing.

Goetzmann, N. (2009) Financing Civilization, mimeo, Yale University.

Greenwood, J. and Jovanovic, B. (1990) The Journal of Political Economy 98, 1076-1107.

Grifell-Tatjé, E. and Lovell, C. (1996) Profits and Productivity, Working Papers 97-18, Wharton School Center for Financial Institutions, University of Pennsylvania.

Gu, J., Lee, S., Suh, Y. (2009) Determinants of behavioral intention to mobile banking, Expert Systems with

Applications, 36(9), 11605-11616.

Holak, S., Song, X., Parry, M. (1991) The Relationship of R&D/Sales to Firm Performance: An Investigation of Marketing Contingencies, The Journal of Product Innovation Management 8(4), 267-282.

Hoy, F., McDougall, P., D'Souza, D. (1992) Strategies and environments of high growth firms. In The State

of the Art of Entrepreneurship, Sexton, D. and Kasarda, J., (Eds.), Boston, MA: PWS-KENT Publishing

Company.

Henderson, B. and N. Pearson (2011) The Dark Side of Financial Innovation: A Case Study of the Pricing of a Retail Financial Product, Journal of Financial Economics 100, 227–247.

Houston, J., Lin C., Lin P., Ma, Y. (2010) Creditor rights, information sharing, and bank risk taking. Journal

of Financial Economics 96, 485-512.

Kemeny, T. (2010) Does Foreign Direct Investment Stimulate Technological Upgrading? World

Development, 38(11).

Kendrick, J. (1961) Productivity Trends in the United States, Princeton: Princeton University Press.

(36)

36

intermediation in the construction of Europe, London: Centre for Economic Policy Research, 156-89.

Klette, T. and Griliches, Z. (1998) Empirical Patterns of Firm Growth and R&D Investment: a Quality Ladder Model Interpretation, Memorandum 23/1998, Oslo University, Department of Economics.

KPMG, Leclercq, R. (2014) Banks look to co-creation with FinTech start-ups, [online] available from:

http://www.kpmg.com/global/en/issuesandinsights/articlespublications/perspectives/pages/banks-co-creation-fintech-start-ups-fs.aspx

Kurosawa, K. (1975) An Aggregate Index for the Analysis of Productivity and Profitability, Omega

3:2, 157-68.

Levine, R. (2005) Finance and growth: Theory and evidence. In Handbook of Economic Growth. Aghion, P. and Durlauf, S., (Eds.), Amsterdam: North-Holland Elsevier Publishers.

Levine, R. (1991) Stock Markets, Growth, and Tax Policy, Journal of Finance 46(4), 1445-65.

Lipsey, R., Carlaw, K., Bekar, C. (2005) Economic Transformations: General Purpose Technologies and Long-Term Economic Growth, Oxford University Press (12)

Manjon, J. and Merino E. (2012) Innovation systems and policy design: The European experience.

Innovation : Management, Policy & Practice 14(1): 33-42.

Manjon, J. and Merino E. (2012) Research, development, and firm growth. Empirical evidence from European top R&D spending firms. Research Policy, 41(6): 1084–1092.

Merton, R. (1992) Financial Innovation and Economic Performance. Journal of Applied Corporate

Finance 4, 12-22.

Michalopoulos, S., Laeven, L., Levine, R. (2009) Financial innovation and endogenous growth. Unpublished

working paper, Brown University.

(37)

37

Miller, D. (1986) Financial Innovation: The Last Twenty Years and the Next. Journal of Financial and

Quantitative Analysis, 21, 459-471.

Pagano, P. and Schivardi, F. (2003) Firm Size Distribution and Growth, The Scandinavian Journal of

Economics 25, 555-589.

Pervan, M. and Visic, J. (2012) Influence of firm size on its business success, Croatian Operational

Research Review (CRORR), Vol. 3, 2012

Petit, P. and Soete, L. (1996) Technical change and employment growth in services: analytical and policy challenges, paper presented at the conference Technology, Employment and Labor Markets, Athens.

Rogers, M. (1998) The definition and measurement of innovation, Working Paper 10/98, Melbourne institute.

Royster, S. (2012) Improved measures of commercial banking output and productivity, Monthly Labor

Review 7, 3–17.

Sheehan, J. and A. Wyckoff (2003) Targeting R&D: Economic and Policy Implications of Increasing R&D Spending, OECD Science, Technology and Industry Working Papers, 2013/08.

Sher, P. and Yang, P. (2005) The effects of innovative capabilities and R&D clustering on firm performance: the evidence of Taiwan’s semiconductor industry. Technovation, 25 (1), 33‐43.

Smith, A. (1776) An inquiry into the nature and causes of the Wealth of Nations, Methuen & Co., London.

Syverson, C. (2011) What Determines Productivity? Journal of Economic Literature 49(2): 326-65.

Szirmai, A., Naudé, W. and Alcorta, L. (2003) Pathways to Industrialization in the 21st Century, New Challenges and Emerging Paradigms, Oxford University Press (5).

Tangen, S. (2002) Understanding the concept of productivity. Proceedings of the 7th_{APIEMS Conference,}

1-4.

(38)

38

Finance. Elsevier, 307–336.

Varadi, K., Mavaluri, K., Boppana, N. (2006) Measurement of Efficiency of Banks in India, MPRA Paper 17350.

Wiele, A., Hesselink, M., Boselie, J. (2002) The Importance Of Customer Satisfaction In Organizational Transformation, ERIM Report Series Research in Management ERS-2002-60-ORG, Erasmus Research Institute of Management (ERIM).

(39)

39 Appendix

Banks considered in the dataset:

Aareal Bank ABN AMRO Alliance & Leicester Allied Irish Banks Banca Intesa

Banca Popolare di Milano Banco Popular

Banco Santander

Bank Ochrony Srodowiska Bank of Ireland

Barclays BHW BRE Bank

Caixa General de Depositos Commerzbank Credito Agricola Danske Bank DekaBank Deutsche Bank Dexia

Espirito Santo Financial Fortis HSBC HSH Nordbank HVB ING Intesa Sanpaolo KBC Komercni banka Landesbank Berlin Lansforsakringar Lloyds Banking Norddeutsche Landesbank Nordea Bank Nykredit Realkredit OP Bank OP-Pohjola Rabobank

(40)

40

Table 1: Total R&D expenditure of banks by country (2004-2013)

Table 2: Hofstede’s Power distance indicators by country

0 5 10 15 20 25 30 35 40 45 50

R&D per Country

(41)

41

Table 3. General Summary of dataset used

Variable Obs. Mean Std. Dev. Min. Max.

Productivity 231 0.23 0.11 -0.31 0.67 RD inv. 235 166.429 259.17 5 1420 Net sales 234 12541 14830 -4383 63720 Employees 232 60104 72613 1341 331458 GDP per capita 235 44150 15955 9001 113738 Total Assets 235 733934 769250 1096 3.543.974 Total Equity 235 31042 30535 -320 145657 PDI 235 42.35 14.45 18 68

Table 4: The correlation between bank performance and innovation Variable Coefficient t statistic P value R&D 9.40 *** 3.98 (0.000) GDP per capita 1.07e-06 *** 2.90 (0.001) Total Assets 7.00e-08 *** 3.52 (0.006) Total Equity -1.22e-06 *** -2.80 (0.004)