Faculty of Economics and Business
Bachelor’s thesis
The performance of Dutch private
equity-backed companies during the current
economic crisis (2007-2012)
Written by Anne-Lotte Meier (10190546)
Supervised by Marijn Kool
Bachelor:
Economics and Business
Specialization: Economics and Finance
Field:
Finance
Abstract
Since the eighties, the private equity industry has experienced enormous growth, with its highest peak in 2007. However, 2007 also happens to be the same year in which the economic crisis has started. The future of many businesses is becoming severely challenged. Several studies state that private equity-backed companies are better able to manage financial distress and are more likely to restructure out of court. Therefore, this study investigates whether private equity-backed companies performed better than publicly traded companies during the current economic crisis, by examining the difference in return on assets and employment growth. It is found that, the average decrease in return on assets is lower for listed companies (-3,55%) than for private equity-backed companies (-7,47%). Further, it can be concluded that the increase in employment growth is approximately five times higher for listed companies (43,96%) than for private equity-backed companies (8,73%). And that private equity-equity-backed companies are better able to manage financial risk than listed companies.
Table of contents
Table of contents……… 3
1. Introduction……… 4
2. Literature review……… 6
2.1. A private equity firm and a publicly traded company………
6
2.2. The governance framework of a private equity firm………..
7
2.3. The private equity developments since the eighties……… 8
2.4. Private equity in The Netherlands………..
8
2.5. The performance of private equity-backed buyouts………...
10
2.5.1. Profitability………..
10
2.5.2. Growth……….
10
2.5.3. Financial distress risk and bankruptcy rates………
11
3. Methodology……….
12
3.1 Data selection………..
12
3.2 Method………. 13
4. Results……… 15
4.1 Summary statistics……… 15
4.1.1 The return on assets………... 15
4.1.2 Employment growth……….. 17
4.2 Correlations……….. 18
4.3 Regression results return on assets……… 21
4.4 Regression results employment growth……… 23
5. Limitations……….. 26
6. Conclusion……….. 28
1. Introduction
Several studies (Cressy, Munari & Malipiero, 2007, p. 648; Kaplan & Strömberg, 2008, p. 2; Palepu, 1990, pp. 247-248) state that the eighties was the beginning of an enormous growth in the American and European private equity (PE) industry. In particular, the leveraged buy-out (LBO, the financing of acquisitions or restructuring with debt by a group of private investors.) enjoyed a real boom. On the other hand, in the nineties there were rumors of an economic downturn, however, according to Kaplan and Strömberg (2008) “the PE-firms continued to purchase private companies and divisions, while the leveraged buyouts of public companies were relative scarce”. Since then, the industry has continued to grow during the last two decades (p. 2). In 2007, the PE-industry was at its highest peak, called the second wave (Wilson et al., 2012, p. 193).
A private equity firm is considered to be a new and efficient form of organization that, for example though a LBO, buys a majority stake in an existing or mature firm, called the portfolio company or PE-backed company (Cressy et al., 2007, p. 648; Kaplan & Strömberg, 2008, p. 2). The PE-firm adds value to their investees, create a superior governance framework and use their skills and expertise to give advice on how to improve operational efficiency (Cressy et al., 2007, p. 648). Subsequently, after a couple of years of successful managing and
restructuring the PE-backed company, the company can hopefully be sold for more than was previously invested (Kaplan & Strömberg, 2008, p. 11).
The enormous growth of the PE industry in 2007 attracted considerableattention of governments and media. Since then, the economic consequence of PE transactions has been a popular topic of discussion. However, the theoretical and empirical research concerning the performance of PE-backed companies in the current economic crisis, which according to Bernstein, Lerner, Sørensen and Strömberg (2010) become noticeable in 2007, is limited. In this period of economic downturn,the future of many businesses is becoming severely challenged.
Until now, the existing literature is about the first and second wave in particular. How do PE-backed companies perform several years after a LBO and how do they perform in comparison with companies that are not backed by PE firms? Only one article by Wilson et al. (2012) has already examined the current global recession. In addition, in the existing literature, research was mainly focused on the US or the UK. For that reason, this thesis analyzes the following research question: Based on companies in The Netherlands, have private equity-backed companies performed better than publicly traded companies during the current economic crisis?
To assess the performance of PE-backed companies, this thesis will look at two dimensions, the financial health and the size of the company, specifically, the return on assets
(ROA) and the employment growth. Furthermore, the period of economic crisis that will be analyzed concerns the period 2007 and including 2012.
Based on the previous literature of Cressy et al. (2007), Tykvová and Borell (2012) and Wilson et al. (2012), it will be hypothesized that in this period of economic crisis, PE-backed companies have performed better than publicly traded companies in the field of financial health. However, there is also evidence that the PE industry follows the bust and boom cycles (Kaplan & Strömberg, 2008, p. 28). Furthermore, Kaplan and Strömberg (2008, pp. 17-18) and Wilson et al. (2012, p. 195) stated that employment growth of a PE-backed company will increase after the LBO, but will be equal or less than other comparable firms. Following their findings, it is hypothesized that employment growth will increase through PE transactions in the current economic crisis, but not more than comparable publicly traded companies.
This thesis is structured as follows. The next chapter provides an overview of existing literature on the performance of PE-backed companies. The third chapter described the data and method used to answer the research question. Thereafter, in chapter four, the results are presented and analyzed, followed by chapter five, which discusses the limitations of this thesis. Finally, chapter six contains the conclusion.
2. Literature review
This chapter gives an overview of the existing literature on private equity. The first paragraph explains what a private equity firm is and in short what publicly traded companies are. Thereafter, the governance framework of a private equity-backed firm is discussed. The next paragraph describes the developments of private equity since 2008. Followed by an explanation of the developments in The Netherlands. Finally, the performance of private equity firms is discussed.
2.1 A private equity firm and a publicly traded company
A private equity (PE) firm is an organization that buys a majority stake, mainly in an existing or mature firm, to restructure the governance framework and try to create better and healthier companies (Kaplan & Strömberg, 2008). The firm in which the PE firm invests is called a private equity-backed company or the portfolio company. The PE firm becomes part of the board of directors and makes contractual rules for the management’s behavior. Therefore it becomes an active investor in the portfolio company (Wilson et al., 2012, p. 194). This is different from venture capital (VC), which invest in young and new developing companies (Kaplan & Strömberg, 2008).
A private equity firm’s equity capital is financed by a private equity fund. According to Kaplan and Strömberg (2008), a private equity fund is organized as a limited liability partnership. They state that the limited partners are institutional investors and wealthy individuals. The general partner consists of the private equity firm, which manage the fund (p. 4). Thus, a private equity firm is completely funded with private equity and is therefore not vulnerable for financial instability (Nederlandse Vereniging voor Participatiemaatschappijen & PricewaterhouseCoopers, 2008, p. 5).
The private equity firm earns its money in several ways. Firstly, itis compensated through a share of the profits of the fund, called ‘carried interest’. Secondly, it earns a percentage of capital employed, the annual management fee. Furthermore, the private equity firm can charge monitoring and deal fees to the invested firms (Kaplan & Strömberg, 2008).
According to Scellato and Ughetto (2013), the most common form of a private equity transaction is a leveraged buyout (LBO). They state that in 2007 82% of the funds raised consisted of buyouts (p. 2642). A leveraged buyout means that the private equity firm buys the portfolio company with a large fraction of outside debt, approximately 60 to 90 percent, and the rest is covered by equity (Kaplan & Strömberg, 2008, p. 6).
In a private equity transaction, there are stand-alone investors or syndicates. A syndicate is a group of private equity investors who each have a partial ownership stake in the portfolio firm. The reason behind this is that they can combine their complementary skills and financial resources and therefore are better able to monitor and support the portfolio firm during the investment phase (Tykvová & Borell, 2012, p. 140). Tykvovà and Borell also state (2012) that syndication can result in agency problems and conflicting interests, which lead to inefficiencies. But when the benefits outweigh the costs, syndication can be preferable over investments from stand-alone investors (p. 140).
A publicly traded company or a public company is a company that issues securities and of which value will be determined through daily trading on the stock exchange.
2.2 The governance framework of a private equity firm
Private equity firms implement a superior governance framework, which add value to their portfolio companies and aim to improve efficiencies (Cressy et al., 2007). The PE firm adds value by means of giving advice and monitoring, which is possible because of the investors skills, experiences and resources. When portfolio companies experience trading or financial difficulties, close monitoring enables investors to intervene promptly, which reduces the likelihood of
financial distress (Cressy et al., 2007; Scellato & Ughetto, 2013; Tykvová & Borell, 2012). Furthermore, according to Tykvová and Borell (2012), experienced investors are better able to manage financial distress risk and bankruptcy rates compared to inexperienced investors because of three reasons. First, experienced investors need to protect their reputation and therefore are more ambitious to avoid bankruptcy. Second, they can obtain loans with better conditions. And lastly, they have better selection and value-adding abilities. However, because of their superior negotiation skills, they can obtain more debt, which can in turn increase financial risk (pp. 140-141).
The amount of debt also plays an important role. High leverage in case of an LBO puts pressure on the managers, so they are less inclined to waste money and are better able to service the debt used by the acquisition (Wilson et al., 2012).
Besides, Kaplan and Strömberg (2008) suggest that the equity of a PE-backed company is illiquid, because they are private. Managers are therefore not able to sell equity until an exit transaction is made and this reduces wasteful spending.
Agency problems are problems between the owners of the company and its management, because they are separated and have other interests. According to Scellato and Ughetto (2013), private equity firms mitigate the agency problem because of the managerial ownership in an
LBO, which in turn leads to better oversight and control of the target company, because the interests are more closely aligned. This in contrary to companies with dispersed external shareholders (p. 2643)
Finally, Kaplan and Strömberg (2008) argue that PE-backed firms are superior compared to public companies, which have less leverage, dispersed shareholders and a weak governance framework (p. 2). They imply that this is because they are more exposed to agency problems as consequence of their separation of ownership and management. Furthermore, they have a larger board of directors, which leads to more difficult negotiation (p. 14).
2.3 The private equity developments since the eighties
As mentioned before, Kaplan and Strömberg (2008) suggest there was a substantial increase in the US private equity transactions between 1980 and 2007. They state that the funds in the US have increased from 0.2 billion dollars in 1980 to approximately 200 billion dollars in 2007 (pp. 2-3). In addition, the buyout market has also grown in Europe. Between 2000 and 2004, the US had only 43.7% of the total world value, while West Europe had 48.9% (Kaplan & Strömberg, 2008, p. 10). Meanwhile, in 1990 and the early 2000, there was an interruption in this enormous growth, “but the LBO market had not died - it was only in hiding” (Kaplan & Strömberg, 2008, p. 2).
However, Kaplan and Strömberg (2008) indicate that in 2008 private equity transactions had also declined because of the start of the economic crisis. But they also state that, because of the less fragile capital structures and more refined private equity firms, the magnitude of possible defaults was less severe than in the nineties (p. 3).Therefore they suspected that the activity would continue (p. 28).According to Scellato and Ughetto (2013), the European private equity investments decreased from 74 billion euro in 2007 to 54 billion euros in 2008 and to 24 billion euros in 2009 (p. 2642). However, the evidence from Europe in general is diverse and usually limited to the United Kingdom (UK) (Cressy et al., 2007, p. 649).
2.4 Private equity in The Netherlands
According to the Nederlandse Vereniging van Participatiemaatschappijen (NVP) and
PricewaterhouseCoopers (2008), the Dutch private equity firms reached a record in 2007. The amount of investments in the Dutch companies had increased to 3,04 billion euros. The liquidity crisis, which began mid 2007, was not yet noticeable (pp. 5-6). They announce that the average investment in a company had increased from 7 million to 9,9 million euros by 2007 and that the overall amount invested in Dutch companies from Dutch and foreign private equity firms
combined is 5,4 billion euros (2008, p. 8). In addition, there was an amount of 2,5 billion euros invested in buyouts, which means a 50% expansion compared to 2006 (NVP &
PricewaterhouseCoopers, 2008, p. 8).
The NVP and PricewaterhouseCoopers (2009) suggest that mid 2008 was the year in which the negative effects of the economic crisis became noticeable (p. 5). The investments in private equity decreased to 1,8 billion euros and the average amount of investments for each company declined to 6,3 million euros. A total of 2,7 billion euros private equity was invested in Dutch companies by the domestic and foreign private equity firms in 2008, compared to 5,4 billion euros in 2007 (NVP and PricewaterhouseCoopers, 2009, p. 8). The Nvp and
PricewaterhouseCoopers (2009) point out that, because of the economic crisis, private equity firms concentrated more on existing portfolios and the banking sector became severe with respect to crediting. It was difficult to determine the value of companies because of the uncertain
expectations, which consequence in sellers and buyers not reaching an agreement (p. 11). As a result, in 2008, large buyouts do not take place (NVP and PricewaterhouseCoopers, 2009, p. 11).
2010 was a year in which private equity transactions increased again on all areas. According to NVP and PricewaterhouseCoopers (2011), the investments of Dutch private equity firms increased to 1.4 billion euros compared to 0,78 billion euros in 2009. On the other hand, they suggest that, in 2012, big transactions do not take place and therefore the investments of the Dutch private equity firms declined again to 1.3 billion euros (2013, p. 6). Only three firms received an investment of more than 50 million euros, compared to nine firms in 2011 (NVP and PricewaterhouseCoopers, 2013, p. 10).
Table 1: Summary of important investments
2007 2008 2009 2010 2011 2012
Investments by Dutch PE
companies €3,04 billion €1,8 billion €0,78 billion €1,4 billion. €2,1 billion €1,3 billion
investments in Dutch companies by Dutch and
foreign PE firms €5,4 billion €2,7 billion €0,82 billion €2,0 billion €2,9 billion €1,3 billion
Investment in buyouts €2,5 billion €1,1 billion €0,21 billion €0,78 billion €1,4 billion €0,84 billion
Average investment in a
company (in million) €9,9 million €6,3 million €2,5 million €4,4 million €5,9 million €3,9 million
Note: All amounts are given in billion euros, except for the average amount of investments, which is given in million euros.
2.5 The performance of private equity-backed buyouts
The existing research is mainly focused on the performance of target companies several years after the buyout. Except for one research, it is all based on the years before 2008, the period before the economic crisis.
2.5.1 Profitability
According to Cressy et al. (2007), based on firms operating between 1995 and 2002, they found that private equity-backed companies’ operating profitability was 4.5% greater compared to companies not backed by private equity three years after the deal was made. Thereby, they argue that the skills and experience of an investor in that specific industry of the target company increases the profitability of a typical PE investment by an extra 8.5% (pp. 665-666). This increase in profitability is also confirmed by Kaplan (as cited by Bernstein et al., 2010) and by Hotchkiss, Smith and Strömberg (2012, p. 21).
On the other hand, based on firms operating between 1997 until and including 2004, Scellato and Ughetto (2013) show that the operating profitability for target companies is less than that of the control group over the first three post-buyout years (p. 2642). They did their research for profitability based on the mean of the variable ‘EBITDA/total assets’ (p. 2646).
To add, Scellato and Ughetto (2013) state that investors’ specialization of the turnaround of a company is positively related to operating profitability. This is investigated by means of a dummy for a group of turnaround specialists compared to generalists, only based on the buyout companies (p. 2647).
When considering the profitability of buyout companies during the current economic crisis, Wilson et al (2012) looked at the return on assets (ROA) in the pre-recession (2003-2006) and recession period (2007-2010) (p. 199).They found that the profitability of the buyout companies was 2.7% to 4.7% higher than for the profitability of non-buyout firms (p. 201). Thereby they point out that, concerning buyout companies, the average ROA and profit margin in the recession period has a positive differential compared to the pre-recession period (p. 197).
2.5.2 Growth
When considering growth, Wilson et al. (2012) found that prior to the crisis, employment growth, turnover growth and the growth in value added of PE target companies was lower compared to companies without such transactions. This went on until 2009 (p. 197). Unlike the period before the crisis, they argue that in the crisis period, the growth in employment, value added and turnover of portfolio companies was higher than for public companies.
Consistent with Wilson et al. (2012), Scellato and Ughetto (2013) also analyze
employment dynamics. They point out that, in the period from 1997 until 2004, the employment growth of PE-backed companies was greater compared to counterparts, at least till three years after the deal was made. To add, they indicate a higher total asset growth for target companies as opposed to non-PE-backed companies (p. 2646).
Cressy et al. (2007) analyzed the relation between PE firms and company growth by means of the variable ‘sales growth’ (p. 653). They found that independent PE firms add approximately 15%-22.5% to the annual growth of the company (p. 666). Cressy et al. (2007) describes an independent PE firms as a private equity firm not affiliated to a bank, which means that they are not funded by banks, but have to raise their funds by third parties, like wealthy individuals. For this reason they have more pressure to let the firm grow (p. 655).
Critics, which are against acquisitions by private equity firms, suggest that these
investments are harmful to employees (Kaplan & Strömberg, 2008, p. 17). Several studies (Davis et al.; Kaplan; Lichtenberg and Siegel, as cited in Kaplan & Strömberg, 2008) point out that, in the period after the buyout, employment growth is lower than for other companies in the industry (pp. 17-18).
2.5.3 Financial distress risk and bankruptcy rates
According to Tykvovà and Borell (2012), European companies backed by private equity firms experience higher financial distress risk until three years after the buyout, but, in these years, the risk is not higher than the financial risk which non-private equity-backed firms experience(p. 143). In addition, stand-alone investors have more difficulties with handling companies in financial distress compared to syndicates (Tykvovà & Borell, 2012, p. 143).
With respect to bankruptcy rates, target companies do not have higher rates contrary to non-buyout companies after the buyout (Tykvovà & Borell, 2012, p. 145). To add, Tykvovà and Borell (2012) state that experienced investors are even able to bring down these bankruptcy rates to below those of non-buyout firms, despite the fact that the financial distress risk is higher (p. 146). They conclude that experienced investors are clearly better in managing this risk (p. 147).
3. Methodology
3.1 Data selection
To answer the research question, two databases of ‘Bureau van Dijk’ have been used. According to the Bureau van Dijk website (http://www.bvdinfo.com/nl-nl/home), the company Bureau van Dijk was established in 1991 and is now one of the world’s biggest and fastest growing providers of business information. The databases contain information about private firms, publicly traded companies, mergers and acquisitions for individual countries, regions and the world.
Firstly,the ‘Zephyr’ database was used to gather the firms that are backed by private equity firms. Secondly, the ‘Orbis’ database was accessed to obtain data concerning the
performance of these specific private equity-backed firms. The second database was also used to find data for the listed companies. Because this thesis focuses on companies in The Netherlands, only Dutch PE-backed firms and Dutch listed companies were selected.
With respect to finding PE-backed companies, which are used in this thesis, the Zephyr database found 60.000 private equity deals worldwide and 1500 private equity deals for Dutch companies only. To be certain know that the firms used are private equity-backed during the current economic crisis, only firms that made deals in 2007 and before were considered. This determines that these firms are private equity-backed until at least 2008. To be sure that these firms are still PE-backed in 2012, this paper also looked at their status and legal form in 2012. Otherwise it is possible, for example, that a firm that made a deal in 2009 was not yet PE-backed in 2007, or not PE-backed anymore in 2012. This is important in order to answer the research question, because then it is possible to compare listed companies with companies that are completely PE-backed during the whole crisis period from 2007 until and including 2012. After this selection, 204 private equity-backed firms remain. With regard to listed companies, ‘Orbis’ found approximately 63.000 companies worldwide. After selecting only Dutch companies, 158 firms remain.
Subsequently, the variables to measure the performance of PE-backed and listed companies are obtained in the Orbis database. As previously mentioned, this paper assesses performance with respect to size and financial health.
Size is a measure of quantitative growth; it measures the overall scale growth of a
company. In the existing literature different measures are used to assess size, namely fixed assets, sales and the number of employees. The last two are the most common measures (Delmar, 2003; Scellato & Ughetto, 2013; Wilson et al, 2012). This paper uses the number of employees as its measure. Employees are the labor resources of a company and contribute to the firm’s size.
Financial health or profitability is in the previous literature evaluated through sales, total assets, the return on assets (ROA) or the earnings before interest, tax, depreciation and
amortization (EBITDA) margin (Cressy, 2007; Wilson, 2012). In this thesis the ROA is used. This reflects the earnings with respect to size. It measures the ability to generate profits, specifically, how efficient a firm is in using its assets to generate revenue (Stolowy, Lebas & Yuan Ding, 2013, p. 77).
After deciding which measures to use, they are acquired from the database for the specific 204 PE-backed companies and 158 listed companies. Seeing as not every company provides all its data for the years 2007 until and including 2012, firms with incomplete data were deleted. This means that only 61 of the 204 PE-backed companies and 67 of the 158 listed companies remain when looking at the employment. Concerning the ROA, only 51 of the 204 PE-backed companies and 99 of the 158 listed companies remain.
Alongside the possibledifference between listed and PE-backed companies, there are also other variables that can have influence on the ROA and employment growth, but which are not the variables of interest. These are called control variables. For employment growth, the variables to be considered are the ROA and the age of the company. If the profitability of a company increases, the ability to grow in size increases as well. According to Cressy et al. (2007), “younger investee companies tend to grow faster, but are more likely to fail” (p. 655). Concerning the ROA, age can also have an impact here. Younger firms are riskier and more vulnerable to macroeconomic shocks, which can result in higher fluctuation in the returns compared to mature firms (Bodie, Kane & Marcus, 2011). Furthermore, both fixed assets and the solvency ratio, is used as a control variable. The solvency ratio is used to control for
macroeconomic shocks (Wilson et al., 2010, p. 9). Stoffels (2012) suggest that the solvency ratio measures the firm’s ability to meet its obligations. They state that the lower the solvency ratio, the more likely the firm will end up in financial distress (p. 85). The variable ‘Fixed assets’ is used as an indicator of firm size.
3.2 Method
To test the performance of private equity-backed firms compared to public companies during the current economic crisis, regression analyses have been used. The model used is a simple multiple regression model. Seeing as more than one regressor is implemented, no special model is needed. The equations for ROA is as follows:
ROA = α + β1 * dummy private equity + β2 * Fixed assets + β3 * Age +
β4* Solvency ratio + ε (1)
Where the ROA is the dependent variable. The coefficient β1 is the relation between the dummy
variable private equity and the dependent variable ROA. A dummy or binary variable is used to indicate a binary outcome (Stock & Watson, 2012, p. 805). In this model it equals one (d=1) if it concerns private equity-backed firms and zero (d=0) if it concerns listed firms. The coefficients β2, β3and β4 give the relationship between the control variables and the ROA. According to Stock
and Watson (2012) control variables are not the variables of interest in this model. But, because they could have an influence on the dependent variable, they need to be included in the model. Otherwise this could lead to ‘omitted variable bias’ (p. 272). The second equation, the equation for employment growth looks like this:
Employment growth = α + β1 * dummy private equity + β2 * Age + β3 *
Solvency ratio + ε (2)
Where employment growth is the dependent variable. Like in the first equation, β1 is the
coefficient for the dummy variable. Furthermore β2 and β3 contain the relation between the
control variables and employment growth.
Like already told, the period that will be analyzed is the crisis period from 2007 until and including 2012. Due to the fact that not all years were available for the dependent variables, this paper will only look at the years 2007 and 2012. For each dependent variable and for each firm, the percentage change is calculated between these years. For example, the employment growth of firm1 will look like this:
𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 𝑜𝑜𝑜𝑜 𝑁𝑁𝑁𝑁𝑒𝑒𝑒𝑒𝑜𝑜𝑒𝑒𝑁𝑁𝑁𝑁𝑒𝑒 𝑜𝑜𝑓𝑓𝑁𝑁𝑁𝑁 12012− 𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 𝑜𝑜𝑜𝑜 𝑁𝑁𝑁𝑁𝑒𝑒𝑒𝑒𝑜𝑜𝑒𝑒𝑁𝑁𝑁𝑁𝑒𝑒 𝑜𝑜𝑓𝑓𝑁𝑁𝑁𝑁 12007
𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 𝑜𝑜𝑜𝑜 𝑁𝑁𝑁𝑁𝑒𝑒𝑒𝑒𝑜𝑜𝑒𝑒𝑁𝑁𝑁𝑁𝑒𝑒 𝑜𝑜𝑓𝑓𝑁𝑁𝑁𝑁 12007 ∗ 100 %
The percentage change is also calculated for the control variables, except for the control variable ‘Age’. It shows the development of each variable of each firm from 2007 until 2012.
After doing the regressions, the Chow test was used to test whether a slope coefficient in the two linear regressions is equal (Gujarati, 1970, p. 50). In other words, are the separated models for the two groups, PE-backed companies and listed companies, different (dr. J.C.M van Ophem, personal communication, June 12, 2014)? Gujarati (1970) states that the formula of the chow test is as follows:
𝐹𝐹 =
𝑆𝑆𝑆𝑆𝑆𝑆𝑐𝑐− (𝑆𝑆𝑆𝑆𝑆𝑆1+ 𝑆𝑆𝑆𝑆𝑆𝑆2)
𝑘𝑘 𝑆𝑆𝑆𝑆𝑆𝑆1+ 𝑆𝑆𝑆𝑆𝑆𝑆2
𝑁𝑁1+ 𝑁𝑁2− 2 ∗ 𝑘𝑘
The SSR1 and SSR2 are the sum of squared residuals from the separate regression of each type of
company. The SSRc is de sum of squared residuals for the pooled regression, k is the number of
parameters in the model and N1 and N2 are the number of observations of the two groups. The test
statistic is: F(k, N1 + N2 – 2 * k) (p. 50). As a result, it is possible to determine if there is a
difference in performance between private equity-backed companies and listed companies.
4. Results
This chapter gives an overview of the results, generated from the regressions that are discussed in the method part of this thesis. The results are achieved using STATA. Firstly, the summary statistics for both models are presented and examined. Thereafter, the correlation tables are given, followed by the results with respect to the return on assets and employment growth.
4.1 Summary statistics 4.1.1 The return on assets
In table 2A, the summary statistics for the whole sample are given, which are private equity-backed firms and listed companies combined. Firstly, the table shows the number of observations (N) for each variable. This is the total number of firms for each variable. It reveals that the amount of firms for fixed assets and solvency ratio is less than for the other variables. This is due to missing data. For example, the variable ‘fixed assets’ is missing data for three firms.
Another thing that has to be mentioned is that the ROA (the dependent variable), fixed assets and solvency ratio (the control variables), are presented in percentages. To be specific, these are the percentage changes of 2012 compared to 2007, the period of economic crisis. When considering the first variable, the table shows that the return on assets has a mean of
-4,88%. This means that, on average, in the period of economic crisis, the return on assets is decreased by 4,88%. The maximum decrease of the return on assets over this period is 47,73%.
Furthermore, it can be noticed that there is a huge spread in the amount of fixed assets of 1.249,41%. This is logical, because the investments or disinvestments in fixed assets can be
extremely high, so it is possible that the amount has substantially increased or decreased over this period. The maximum increase is 11.119,62%.
Table 2A: Summary statistics – Private equity-backed companies and listed companies
N Mean Std. Dev Min Max
ROA (dependent) 150 -4,88% 14,66% -47,73% 66,23%
Fixed Assets 147 210,35% 1.249,41% -95,42% 11.199,62%
Solvency Ratio 149 -1,68% 18,05% -70,82% 59,21%
Age (years) 150 45,61 48,80 6 329
The following tables reveal the summary statistics separately for the private equity-backed companies (table 2B) and the listed companies (table 2C). Like table 2A, the percentages are percentage changes over the period 2007-2012. Here too there are missing values, but only for the listed companies.
When comparing these statistic tables, it shows that the average decrease in the return on asset is lower for the listed companies 3,55%) than for the private equity-backed companies (-7,47%). Thus the profitability for private equity-backed firms has, on average, decreased more during the current economic crisis. Further, the solvency ratio of listed companies has decreased by 4%, while the solvency ratio of the private equity-backed companies has increased by 2,77%. When the solvency ratio decreased, this means that these firms are less able to meet its
obligations and have a higher probability to default on its debt (Stoffels, 2012, p. 85). Therefore, contrary to private equity-backed firms, the probability that listed companies would default on its debt increased during the crisis.
But it must be mentioned that the number of observations for the private equity firms is approximately twice as low, which makes these statistics less reliable.
Table 2B: Summary statistics – Private equity-backed companies (PE=1)
N Mean Std. Dev Min Max
ROA (dependent) 51 -7,47% 13,21% -47,73% 25,38%
Fixed Assets 51 293,97% 1432,54% -90,15% 9.975,74%
Solvency Ratio 51 2,77% 23,18% -70,82% 59,21%
Table 2C: summary statistics – Listed companies (PE=0)
N Mean Std. Dev Min Max
ROA (dependent) 99 -3,55% 15,25% -38,23% 66,23%
Fixed Assets 96 165,93% 1145,95% -95,42% 11199,62%
Solvency Ratio 98 -4,00% 14,29% -55,65% 28,48%
Age (years) 99 53,04 55,64 6 329
4.1.2 Employment growth
There are also three tables of summary statistics for the dependent variable employment growth. Table 3A consist of the statistics for the whole sample, table 3B for the private equity-backed firms and 3C for the listed companies. Both ‘Employment Growth’ and ‘Solvency Ratio’ are given in percentage changes over the period 2007-2012.
From table 3A it is evident that the employment growth for the whole sample, on average, increased by 27,17%. This means that the size of the companies, on average, increased during the economic crisis. This is surprising, because, in times of recession, a lot of firms have difficulties to keep their head above water, which can result in a substantial rise in unemployment rates (Mankiw, 2010, p. 332). The table also reveals that the maximum employment growth is 748,28%, which would imply that the employment of one of these companies became
approximately seven times higher in five years. This is very unlikely, especially in this period of economic downturn. Therefore, this value can be regarded as an outlier, which in turn could have affected the mean.
Table 3A: summary statistics – Private equity-backed firms and listed companies
N Mean Std. Dev Min Max
Employment Growth (Y) 128 27,17% 105,42% -79,40% 748,28%
Solvency Ratio 124 -1,71% 29,31% -107,12% 157,89%
Age (years) 128 44,23 48,96 7 329
When table 3B and 3C are compared, it is clear that the increase in employment growth for listed companies is, on average, approximately five timeshigher than the increase for private equity-backed companies.Despite the fact that this difference in growth could be realistic, it is possible that a part of this deviation is due to the outlier in the sample for listed companies. Moreover, the solvency ratio of listed companies decreased by 3,96% over the period, which is
still contrary to the private equity-backed firms, of which the probability to default on their debt decreased, because the solvency ratio increased (0,85%).
Contrary to the summary statistics of the dependent variable ROA, these statistics are more reliable, because the number of observations is approximately equal.
Table 3B: summary statistics – Private equity-backed firms (PE=1)
N Mean Std. Dev Min Max
Employment growth 61 8,73% 52,64% -77,71% 223,19%
Solvency Ratio 58 0,85% 39,22% -107,12% 157,89%
Age (years) 61 25,61 19,36 7 91
Table 3C: summary statistics – Listed companies (PE=0)
N Mean Std. Dev Min Max
Employment growth 67 43,96 135,14 -79,40 748,28
Solvency Ratio 66 -3,96 16,31 -75,77 28,48
Age (years) 67 61,19 60,49 7 329
4.2 Correlations
Correlations are values between -1 and 1, which determine to what extent the variables depend on each other (Stock and Watson, 2012, p. 74). The first three tables (table 4A, 4B & 4C) give the correlations with respect to the dependent variable ROA. The second three tables (table 5A, 5B & 5C) present the correlations concerning the dependent variable employment growth.
In all of the three tables below, the solvency ratio is positive, and highly significant, correlated with the dependent variable ROA (α=1%). This is also the highest correlation coefficient in the three tables (0,3626, 0,5558 and 0,3134 respectively), but the coefficients for the two types of firms combined (0,3626) and for the listed companies alone (0,3134) are far from perfect, because these values are closer to 0. There is a moderate linear relation between the solvency ratio and ROA. But for the private equity-backed firms there is a stronger positive linear relation between these variables (0,5558). Because these correlations are highly significant, it is unlikely that these relations are occurred by chance. These positive and significant correlations say that these variables co-moving in the same direction.
Further, in table 4A, PE and age are also highly significant correlated with each other, but have a weak negative correlation (-0,2128). In addition, table 4A shows that both PE and
solvency ratio and fixed asset and solvency ratio are significant correlated as well (for α=0,05 and α=0,1 respectively), but the relations are weak. Moreover, the table indicates that the dummy PE
is low and not significant correlated with ROA, which are the variables of interest. Therefore it could have given a problem with respect to accuracy and reliability of this thesis. However the values of the dummy only indicate if a firm is PE-backed or listed, so the values have in itself no meaning and therefore will not affect the results.
In general, the correlations between the independent variables in table 4A are close to zero, so they are low, which is a good, because the independent variables are not allowed being influenced by each other. Otherwise this could lead to perfect multicollinearity (Stock & Watson, 2012, p. 239).
To add, in table 4C, solvency ratio and fixed assets are negative and highly significant correlated, but they have a weak relation. This weak relation will not be a problem, because they are just the control variables.
Table 4A: Correlations – private equity-backed companies and listed companies
ROA PE Fixed Assets Solvency Ratio Age
ROA 1
PE -0,127 1
Fixed Assets -0,0026 0,0489 1
Solvency ratio 0,3626 *** 0,1785 ** -0,1462 * 1
Age -0,1177 -0,2128 *** -0,0949 -0,0791 1
Note: *** significance level of 1%, ** significance value of 5%, *significance value of 10%
Table 4B: Correlations – Private equity-backed companies (PE=1)
ROA Fixed Assets Solvency Ratio Age
ROA 1
Fixed Assets 0,0702 1
Solvency ratio 0,5558 *** -0,0585 1
Age -0,1462 -0,1245 -0,0812 1
Note: *** significance level of 1%, ** significance value of 5%, *significance value of 10%
Table 4C: Correlations – Listed companies (PE=0)
ROA Fixed Assets Solvency Ratio Age
ROA 1
Fixed Assets -0,0344 1
Solvency ratio 0,3134 *** -0,2648 *** 1
Age -0,1546 -0,0846 -0,0362 1
The first correlation table (5A) for employment growth indicates that the dummy PE and the dependent variable employment growth are negative and significant correlated with each other (α=10%). This indicates that this relation is unlikely occurred by chance, which is good, because something can be said about this variable. But, on the other hand, because the linear relation is not that strong (-0,1676), it could have affected the reliability. However, again, nothing can be said about this correlation, because the values of dummy variable have in itself no
meaning.
The same applies for the significant correlation between ‘Age’ and ‘PE’.
Table 5A: Correlations – Private equity-backed firms and listed companies
Employment growth PE Solvency Ratio Age
Employment growth 1
PE -0,1676 * 1
Solvency Ratio -0,0782 0,0823 1
Age -0,0295 -0,3645 *** -0,0459 1
Note: *** significance level of 1%, ** significance value of 5%, *significance value of 10%
Table 5B: Correlations – Private equity-backed companies (PE=1)
Employment growth Solvency Ratio Age
Employment growth 1
Solvency Ratio 0,0067 1
Age -0,0925 -0,0328 1
Note: *** significance level of 1%, ** significance value of 5%, *significance value of 10%
Table 5C: Correlations – Listed companies (PE=0)
Employment growth Solvency Ratio Age
Employment growth 1
Solvency Ratio -0,1790 1
Age -0,0995 -0,0224 1
4.3 Regression results ROA
Table 6 gives the regression results for the ROA model over the period 2007-2012. Number (1) present the regression results for PE-backed firms and listed companies together, so for all observations. Number (2) gives the regression results when the dummy variable PE takes on value one. The regression results where the dummy PE takes on value zero are given in number (3). At last, number (4) reveals the results for the whole sample, but exclusive the dummy variable PE.
In order to test if the ROA with respect to private equity-backed companies differ from the ROA concerning listed companies, the Chow test can be done (dr. J.C.M van Ophem, personal communication, June 12, 2014). Therefore model (2) and (3) will be tested against model (4), with 4, 138 degrees of freedom, whereby the critical value is 1,99 when α=10%, 2,44 when α=5% and 3,46 when α=1% (dr. K.J van Garderen, personal communication, June 29, 2014; Keller, 2009, pp. B-12-B-16)). The result of the Chow test is 2,11960797. Because 2,1196 is larger than 1,99 (when α=10%), it can be concluded that the models differ from each other with at least one coefficient. That the two models differ can also be concluded from the p-value of the dummy variable PE. Because the dummy PE is included in the first model (1), the intercepts of the groups are different. And because this variable is highly significant, it is clear that these models for private equity-backed firms (regression 2) and listed companies (regression 3) differ and can be compared.
When regression (2) and (3) are compared, it can be concluded that the adjusted R2 is higher for model (2) than for model (3). This means that the model for ROA with private equity firms is better explained by its variables than those of listed companies and therefore more accurate and reliable (Stock and Watson, 2012, p. 236). However, it must be taken into account that the number of observations is twice as low as the observations of the model with listed companies, 51 compared to 95.
In these models, the solvency ratio coefficient for private equity-backed companies (0,3154) shows that when it increases, the return on assets increases as well and this is highly significant (α=1%). Concerning the current economic crisis, this implies that when the probability that private equity-backed companies defaulted on their debt decreases (thus the solvency ratio increases), the profitability will also increase, so they move in the same direction. This is a logical causation, because if a firm is able to meet its debt, it will not end up in financial distress, hence it will stay alive and probably earn profits. However, when considering the coefficient of the solvency ratio in model (2), the model for the listed companies, a contrary effect is revealed. It implies that when the solvency ratio increases, the profitability will decrease, which suggests that
when a firm not defaults on their debt, the profitability decreases, so they move in the opposite direction. This is kind of strange causation and therefore, it can be argued that there must be another factor that causes the ROA to decrease or increase during this economic crisis. The adjusted R2 in model (3) is also extremely low, which indicate that these variables poorly explain the variation in the ROA for the listed companies.
Furthermore, the results disclose that the p-value of fixed assets in each regression is higher than 0,4, which is even higher than 0,1 and thus, according to Keller (2009), not
significant. In addition, he argues that if the p-value lies between 0,05 and 0,1, it is not significant either, which is the case for the variable ‘Age’ (p. 355). Therefore these variables do not make sense in determining the difference in ROA between these two types of firms. They are just control variables, thus give no problem.
From regression (1) can be achieved that the coefficient of the dummy ‘PE’, the variable of interest, is a negative number and thus has a negative influence on the dependent variable ROA. Because the p-value of this dummy is 0,004, which according to Keller (2009) is highly significant (p. 355), it can be concluded that Dutch private equity-backed companies (when PE=1) generate, on average, lower ROA during this crisis compared to Dutch listed companies (when PE-0). It confirms what Scellato and Ughetto (2013) have suggested, who imply that the operating profitability for target companies is worse. But, this is contrary to most of the existing literature (Bernstein et al., 2010; Cressy et al., 2007; Hotchkiss et al., 2012), which imply that private equity-backed firms have a higher operating profitability than comparable companies. However, this existing literature has studied these two different types of firms in the years before the crisis and is based on other countries than The Netherlands, which could imply other results than this thesis present. Wilson et al. (2012) suggests that the PE-backed companies did better in the current economic crisis, but his research is only based on the first three years of the crisis compared to the first five years that his thesis cover, which better represent these years of current economic crisis. Wilson et al. (2012) also based his research on firms in the UK, which is according to him the largest PE market in Europe. Because the PE market in The Netherlands is growing and not yet that big, it is possible that the Dutch listed companies will do better in terms of profitability compared to the Dutch PE-backed companies.
Table 6: Regression results ROA
(1) (2)
Coefficient T-Statistic P-value Coefficient T-Statistic P-value
Constant -0,1928 -2,93 0,919 -7,1963 -2,86 0,006** PE -7,1107 0,69 0,004** Fixed Assets 0,0062 5,17 0,494 0,0008 0,76 0,453 Solvency Ratio 0,3261 -1,72 0,000** 0,3154 4,6 0,000** Age -0,0401 -0,1 0,088 -0,0447 -0,74 0,462 R2 0,1919 0,3273 Adjusted R2 0,169 0,2844 SSR 25450,8377 5865,30569 k 5 4 N 146 51 (3) (4)
Coefficient T-Statistic P-value Coefficient T-Statistic P-value
Constant -0,1430 -0,07 -0,07 -3,3499 -2,10 0,037* Fixed Assets 0,0005 0,34 0,735 0,0005 0,51 0,608 Solvency Ratio -0,3372 3,12 0,002** 0,2941 4,61 0,000** Age -0,0396 -1,47 0,145 -0,0260 -1,11 0,269 R2 0,1215 0,1429 Adjusted R2 0,0926 0,1248 SSR 19568,6431 26996,5575 k 4 4 N 95 146
Note1: (1) Whole sample, inclusive the dummy variable PE. (2) Private equity-backed firms, when PE=1. (3) Listed companies, when PE=0 (4) Whole sample, exclusive the dummy variable.
Note2: ** significance value of 1%, *significance value of 5%.
4.4 Regression results employment growth
Table 7 gives the results of the regressions with respect to employment growth over the period 2007-2012.
The Chow test, which is done for the ROA regression, will also be used for this
employment model. The result of the test is 2,37291727. According to Keller (2009) and dr K.J van Garderen (personal communication, June 29, 2014), the critical value, with respect to (3, 118) degrees of freedom and a significance level of 10% is 2,12999. When alpha is 5%, the critical value is 2,68 (pp. B-12-B-16). 2,37291 is bigger than 2,12999, meaning that the model for listed companies differs from the model of PE-backed companies and they can be compared.
When comparing these two models, only the solvency ratio for listed companies, in regression (3), is highly significant (α=1%). Like the results for ROA, the coefficient is negative, which indicate that when the solvency ratio increases, thus the ability of the firm to meet its debt increases, the employment growth will decrease. However, this variable is only implemented to control for the financial health of the company and is therefore not important in determining the difference in employment growth between these two types of firms.
From regression number (1) it is clear that the coefficient of PE is negative, which should mean that, over the period 2007-2012, the Dutch private equity-backed firms (when PE=1) score on average lower, with respect to employment growth, compared to the Dutch listed companies (when PE=0). The p-value of this variable is 0,045, which means that the variable is significant (α=5%). This corresponds to the results from Scellato and Ughetto (2013) and the critics, which imply that the employment growth was greater for PE-backed companies compared to their counterpart, but their research was based the years before the crisis. Wilson et al. (2012) also confirmed the lower employment growth for PE-backed companies before the crisis. However, he argues that the employment growth is greater for PE-backed companies during the recession years 2008 until and including 2010. As previous mentioned, his research is based on firms in the UK, which market is already better developed, so these PE-backed firms may already be more evolved with respect to their corporate governance framework. This may be the reason that Dutch PE-backed companies have lower employment growth during the crisis than the Dutch listed companies.
Contrary to the dummy variable, the p-value of ‘Solvency ratio’ and ‘Age’ is even higher than α=10%, which imply that they are not significant. However, these variables are just control variables, so it makes no sense.
When considering the R2 and adjusted R2 of model (1), (2) en (3), it is clear that these measures of fit are extremely low. That means that these models are weakly explained by the variables and less reliable than the models of ROA.
Table 7: Regression results employment growth
(1) (2)
Coefficient T-Statistic Significance Coefficient T-Statistic Significance
Constant 57,96115 3,17 0,002** 17,67618 1,48 0,143 PE -41,44643 -2,02 0,045* Solvency Ratio -0,2448783 -0,75 0,456 0,004421 0,02 0,981 Age -0,2360483 -1,14 0,257 -0,2886021 -0,79 0,432 R2 0,0404 0,0113 Adjusted R2 0,0164 -0,0247 SSR 1347361,14 161298,37 k 4 3 N 124 58 (3) (4)
Coefficient T-Statistic Significance Coefficient T-Statistic Significance
Constant -0,1430343 -0,07 -0,07 31,71199 2,43 0,017* Solvency Ratio -0,337211 3,12 0,002** -0,291514 -0,88 0,379 Age -0,0396437 -1,47 0,145 -0,0857654 -0,44 0,662 R2 0,0432 0,0077 Adjusted R2 0,0129 -0,0087 SSR 1152678,35 1393246,84 k 3 3 N 66 124
Note1: (1) Whole sample, inclusive the dummy variable PE. (2) Private equity-backed firms, when PE=1. (3) Listed companies, when PE=0 (4) Whole sample, exclusive the dummy variable.
5. Limitations
There are some limitations in this paper. Firstly, it is difficult to obtain data for private equity-backed companies, because they are not publicly traded and therefore are not forced to publication (Wilson, Wright & Cressy, 2010, p. 8).This resulted, for the ROA, in a sample of private equity-backed firms smaller than the sample of listed companies, so they are not equally distributed. As the results in table 6 show, the sample of PE-backed companies is twice as low as the sample of the listed companies. Furthermore, the fact that this research is based only on Dutch companies can also result in a smaller sample. The researchers in previous literature were focused on bigger areas, like Europe, UK and US (Cressy et al., 2007; Scellato & Ughetto, 2013; Wilson et al., 2012).
The problem that only few firms have published their data has resulted in selection bias. In the regression model for ROA, only firms that completely reveal their ROA in the years 2007 and 2012 are included in the sample. This also applies to employment growth.
Another complication is the outliers. The data for employment growth contain a couple of outliers, which, as already mentioned, is revealed in the summary statistics. This could have an impact on regression results for employment growth. Nevertheless, these outliers are included in the analysis, because, if neglected, the sample became even smaller than it already is.This could also have a negative effect on the regression results.
Once the firms were selected, the other (control) variables were added to these specific firms. However, not all variables were completely available for these years. As consequence, the missing data poses a threat to internal validity (Stork and Watson, 2012, p. 364).
Because the R2 and adjusted R2 are extremely low in the regression models for
employment growth and ROA, it may be valuable to add more and better explanatory variables. A variable like ‘wage’ can have influence on employment growth. On one hand, high wage increases the cost of a company, which can decrease the demand for employees. On the other hand, high wage may attract more employees, which can increase the employment growth. A dummy for the industries may be relevant to add to the models, because employees are, for example, less important in capitalized industries, but in the service sector they have a leading role.
The biggest problem of this paper is that the regression model lacks explanatory
variables. The regression model only consists of a dummy variable and some control variables, so the focus is on the dummy variable ‘PE’. A regression model with just one dummy variable is like performing a difference of means analysis (Stock and Watson, 2012, p. 195), The control
variables ensure that it is not just a difference of means analysis. However, this raises the question whether these regression results really contribute to answer the question if private equity firms, compared to listed companies, do better with respect to ROA and employment growth.
6. Conclusion
This thesis compares the performance of Dutch private equity-backed firms with Dutch listed companies during the current economic crisis (2007-2012). To access this performance, the difference in ROA and employment growth between these two groups is studied.
From the summary statistics is it clear that, on average, in the period of economic crisis, the ROA is decreased by 4,88%, with a maximum decrease of 47,73%. When comparing the statistics of listed companies with those of private equity-backed companies, it shows that the average decrease in return on assets is lower for Dutch listed companies (-3,55%) than for Dutch private equity-backed companies (-7,47%). This is confirmed by the regression results, because the coefficient of the dummy variable ‘PE’ is a negative number (-7,11), which means that Dutch PE-backed companies (when PE=1) generate on average lower ROA during this crisis compared to Dutch listed companies (when PE=0). From these regression results it is also clear that the dummy variable is highly significant, because the p-value (0,004) is lower than an alpha of 1%.
When looking at the coefficient for ‘Solvency ratio’ in regression (1) of the ROA model, which is highly significant (p-value=0,000), it is clear that this ratio moves in the same direction as the ROA, which also applies to the PE-backed companies. This result combined with the summary statistics implies that PE-backed firms are better able to manage their debt and thus are less likely to default on its debt, which is a good characteristic in times of economic crisis. This is because the solvency ratio of PE-backed companies has increased and that of listed companies has decreased during this period.
The summary statistics concerning employment growth reveal that the employment growth is, on average, increased by 27,17%, with which can be concluded that the average size of both type of companies is increased during the economic crisis. But when comparing the
separated statistics of both types, it is clear that this growth is approximately five times higher for listed companies than for PE-backed companies. This is again confirmed by the regression results, which show a negative coefficient (-41,45) for the dummy variable. Also in this model, the dummy variable is significant (α=5%). This high growth is surprising during the recession years and therefore it is possible that the outliers in this model affected the result.
To access whether the models of both type of firms are reliable, the measures of fit (R2 and adjusted R2) of these models are compared. This measure reveals that the model for
employment growth is less explained by its variables (approximately 1%) compared to the model of ROA (approximately 10%).
To summarize, is it clear that, according to this research, Dutch publicly traded companies perform better with respect to ROA and employment growth compared to Dutch private equity-backed companies during the current economic crisis. This is contrary to what is hypothesized about the return on asset, because it was expected that Dutch PE-backed firms would perform better during the recession period. However, the higher decrease in ROA for PE-backed companies is possible, because the Dutch private equity market is growing, but not yet developed like the market in the UK, that was considered in the inquiry of Wilson et al. (2012). The hypothesis about employment growth is confirmed, because, like Wilson et al. (2012) implies, the employment growth for both type of firms this is increased during the crisis, but less for Dutch PE-backed companies than for Dutch listed companies.
7. References
Berger, A. N., & Udell, G. F. (1998). The economics of small business finance: The roles of private equity and debt markets in the financial growth cycle. Journal of Banking &
Finance, 22, 613-673.
Bernile, G., Cumming, D., & Lyandres, E. (2007). The size of venture capital and private equity fund portfolios. Journal of Corporate Finance, 13, 564-590.
Bernstein, S., Lerner, J., Sørensen, M., & Strömberg, P. (2010). Private equity and industry
performance (NBER Working Paper No. 15632).
Bodie, Z., Kane, A., & Marcus, A. J. (2011). Investments and Portfolio Management (9th ed.). New York. USA: McGraw-Hill.
Cressy, R., Munari, F., & Malipiero, A. (2007). Playing to their strengths? Evidence that specialization in the private equity industry confers competitive advantage. Journal of
Corporate Finance, 13, 647-669.
Hotchkiss, E., Smith, D.C., & Strömberg P., (2012). Private Equity and the Resolution of
Financial Distress.
Kaplan, S. N., & Strömberg, P. (2008). Leveraged buyouts and private equity (NBER Working Paper No. 14207).
Keller. (2009). Managerial statistics (8th ed.). Mason, USA: South-West Cengage Learning. Mankiw, N. G. (2010). Macroeconomics (7th ed.). New York, USA: Worth Palgrave Macmillan. Nederlandse Vereniging van Participatiemaatschappijen (NVP) and PricewaterhouseCoopers.
(2013). Ondernemend vermogen: De Nederlandse private equity-markt in 2012. Nederlandse Vereniging van Participatiemaatschappijen (NVP) and PricewaterhouseCoopers.
(2012). Ondernemend vermogen: De Nederlandse private equity-markt in 2011. Nederlandse Vereniging van Participatiemaatschappijen (NVP) and PricewaterhouseCoopers.
(2011). Ondernemend vermogen: De Nederlandse private equity-markt in 2010. Nederlandse Vereniging van Participatiemaatschappijen (NVP) and PricewaterhouseCoopers.
(2010). Ondernemend vermogen: De Nederlandse private equity-markt in 2009. Nederlandse Vereniging van Participatiemaatschappijen (NVP) and PricewaterhouseCoopers.
(2009). Ondernemend vermogen: De Nederlandse private equity-markt in 2008. Nederlandse Vereniging van Participatiemaatschappijen (NVP) and PricewaterhouseCoopers.
(2008). Ondernemend vermogen: De Nederlandse private equity-markt in 2007. Palepu, K. G., (1990). Consequences of leveraged buyouts. Journal of financial economics,
Scellato, G., & Ughetto, E. (2013). Real effects of private equity investments: Evidence from European buyouts. Journal of Business Research, 66, 2642-2649.
Stoffels, H. (2010). De naamloze vennootschap (4th ed.). Maastricht, The Netherlands: Stoffels.
Stolowy, H., Lebas, M. J., & Ding, Y. (2013). Financial accounting and reporting: A global
perspective (4th ed.). Andover, England: Cengage Learning EMEA.
Tykvová, T., & Borell, M. (2012). Do private equity owners increase risk of financial distress and bankruptcy? Journal of Corporate Finance, 18, 138-150.
Wilson, N., Wright, M., & Cressy, R. (2010). Private equity and insolvency.
Wilson, N., Wright, M., Siegel D. S., & Scholes, L. (2012). Private equity portfolio company performance during the global recession. Journal of Corporate Finance, 18, 193-205.
