The impact of the leveraged buyouts on the target firm's cash holdings

(1)

THE IMPACT OF THE LEVERAGED

BUYOUTS ON THE TARGET FIRM’S

CASH HOLDINGS

MSC Finance: Corporate Finance track Master Thesis

Dmitrijs Borscevskis (11403004)

Supervisor: dhr. dr. Vladimir Vladimirov

(2)

1

Statement of Originality

This document is written by Dmitrijs Borscevskis, who takes full responsibility for the contents of this document.

I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Finance is responsible solely for the supervision of completion of the work, not for the contents.

(3)

2

Abstract

This paper investigates the impact of the Leveraged Buyouts on the target firm’s cash holdings. In addition, I examined the effect of LBO’s by Private Equity firms on the cash flow sensitivity to cash and investment. Moreover, I checked the LBO’s effect on the target firm’s investment. Additionally, I ran “placebo” test using the sample of industry-size-country matched companies. After investigating European LBO’s deals from 2004 to 2015, it can be concluded target firm’s cash holdings decline following the acquisition. The drop in cash holdings is about 2.9% and statistically significant different from the zero. The cash flow sensitivity to cash and investment increased prior to the acquisition and declined after the LBO transaction and the target firm’s investment increased following the acquisition. These results confirm that acquisitions relieve financial constraints of the target firms. Finally, “Placebo” test revealed that the matched firms do not encounter changes in financial policies related to financial constraints reduction.

(4)

3

1. Introduction

For a long time, there are debates about firm’s optimal level of cash holdings. As debates continue, American corporations go on piling up cash. One of acknowledgement of these abnormal cash holdings is that the average cash-to-assets ratio for the US corporations more than doubled in the period from 1980 to 2006 (Bates, et al., 2009) and according to the

“Financial Times” news portal, 1.7 trillion dollars are held in cash by the US corporations.1_This

tendency could be explained by the academic literature on corporate cash holdings which specify two primary motives why firms hold cash – precautionary and agency motive (Opler, et al., 1999). The first motive relates to company’s upcoming financial constraints and the second one is about imperfections in the dispersed form of ownership of the companies (Bates, et al., 2009).

However, the modern world is full of imperfect capital markets with firms which are facing financial constraints. Financial constraints facilitate firms to pass up profitable investment opportunities. In that kind of situations, an acquisition could potentially ease financial constraints through post-acquisition reallocation of capital between large organization divisions or better access to capital markets (Erel, et al., 2015). Although a colossal number of papers focusing on mergers and acquisitions, there is still no clear answer whether acquisitions mitigate financial constraints of the target firms.

A potential empirical examination of this unanswered question requires financial data of the target firms before and after the acquisition in order to evaluate financial constraints of the target firms. Moreover, target firms are becoming part of the acquirers in the post-acquisition process, in that case, these companies are no longer independent firms, so it is impossible to evaluate changes following the acquisition unless the target is becoming an independent

(6)

5

subsidiary. However, if the acquirer is Private Equity firm (PE), targets remain independent firms after the acquisition, board members and top executives could be changed, but firms do not lose its autonomy. Given that, it is possible to collect financial data of target firms before and after acquisition process. Additionally, most of the targets are small privately held companies which do not disclose financial information. In the United States, financial information on private firms is not publicly available. Nonetheless, in Europe, most small privately held companies disclose financial data, so it is possible to collect financial information on target firms.

There are many ways of measuring whether the firm is financially constrained. One of the methods is to measure companies cash holdings. In times of uncertainty, it is harder to access capital markets to finance some valuable projects, so the only option managers have is to use internal resources such as cash holdings. Consequently, managers are willing to hold more cash on the balance sheet during uncertain times. The acquisition would facilitate a decline in targets cash holdings and therefore financial constraints will be also reduced.

Traditionally, there is the difference between strategic buyers and financial investors, such as PE firms. Both parties are engaged in the acquisition process while acquisition strategy and post-acquisition goals are different. Strategic buyers aim to fully integrate target firms into their business and benefit from the synergies. Strategic buyers pursue “buy and hold” strategy and are willing to realise synergies in the long-term perspective. On the other hand, PE firms do not have long-term incentives and are guided by “buy, improve and sell” strategy. Target firms are usually heavily leveraged and management is under constant pressure to operate in an efficient manner in the post-acquisition period. Financial investors could change the board of directors and managers in order to improve results of target firms. Ultimately, PE firms usually exit the investment in 3 to 6-year period, selling the company to another investor, strategic buyer or make company public through Initial Public Offering.

(7)

6

Previous research by Erel et al. (2015), indeed, showed significant results that target firms cash holdings decreased after the acquisition process, thereby mitigating financial constraints. However, these results could be biased due to acquirer’s incentives. As a strategic buyer, the acquirer has long-term goals concerning the target firm. An acquirer could have incentives of tunneling resources such as assets and cash holdings from the acquired firms. For instance, annual dividend payout to shareholders could lead to decrease in target firm’s cash holdings (Baek, et al., 2006). PE firms do not have incentives in tunneling resources from targets because of “buy, improve and sell” strategy, therefore estimating the change in cash holding of target firms before and after LBO transaction could better assess whether financial constraints are mitigated.

First, I measured the impact of LBO’s on target firm's cash holdings. The results suggest that target firms are less constrained after the acquisition by PE firms. Descriptive statistics show that target firm’s mean of cash and cash equivalents normalized by total assets declined by 2% after the acquisition. Running the first difference regression showed that the cash ratio, indeed, decreased approximately by 2.9% after the LBO transaction and this result is statistically significant at 1% level. Second, I used a measure of the cash sensitivity to cash flow proposed by Almeida et al. (2004) in order to additionally verify if cash ratio is a good measurement of financial constraints. The cash sensitivity to cash flow also decreases from 0.08% to -0.07%, however, the result is not statistically significant at any level. Nevertheless, the investment sensitivity to cash flow shows a statistically significant drop from 4.38% to -3.84%. These results suggesting that target firms became less financially constrained after being acquired by PE firms. Third, previous results showed that acquisition could mitigate constraints, therefore targets facing an opportunity to revise their investing strategy. Accordingly, targets should increase the amount of investment due to relieved constraints. Summary statistics shows that the mean of gross investment (as a fraction of total assets) declined by 1.4%. Nonetheless,

(8)

7

additional first different regressions showed that gross investment normalized by total assets dropped by up to 1.6%, controlling for different factors. Nevertheless, these results are in line with theory but are not statistically significant different from zero. Overall, results suggest that institutional investor improves financial situation in the target firms and these companies become less financially constraint after being acquired by PE firms. Targets cash ratio declines, a fraction of cash savings from incremental cash flows also drops and investments increase following the acquisition. Hence, institutional investor facilitates target firms easier and cheaper to use external sources of financing in order to invest in the new projects. Even though acquisition through LBO heavily leverages targets, results suggest that financial constraints are relieved by acquisition.

My thesis contributes to the existing merger and acquisition literature and theory of financial constraints. First of all, I used a different sample of acquisitions (by PE firms using LBO’s as a method of payment) compare to previous research. Acquisition by an institutional investor could better explain post-acquisition decline of target firm’s cash holdings compare to the same acquisition by a strategic buyer. As it is mentioned before, PE firms are less eager to engage in tunneling resources from their targets compare to a strategic buyer. Moreover, I used more than one year “before” and “after” acquisition observations. PE firms tend to exit investment up to a 6-year period of time, so I used on average 3 years “before” and 5 years “after” acquisition observation in order to better explore the impact of the LBO transaction on cash holdings and financial constraints. In general, my results add additional evidence that acquisitions facilitate a decline in target firm’s cash holdings and therefore these companies face relaxation of financial constraints.

This paper is organized as follows. Section 2 presents related literature review. In Section 3, I provide a description of collected data sample. In Section 4, I describe the methodology for analyzing the impact of LBO’s on cash holdings. Section 5 provides results of

(9)

8

regression analysis. In Section 6 and 7, I provide robustness check and potential issues connected to the research. Section 8 concludes.

2. Literature Review

2.1 Cash holdings

Cash holdings are one of the most valuable resources of each company because it provides the firms with a liquidity. During the bad times, a company can use cash holdings as an emergency fund in order to pay out all the short-term obligations. There are different theories why firms holding cash and cash equivalents on their balance sheets. Two main theories are – precautionary and agency motives.

Bates et al. (2009) found that firms tend to hold more cash to cover unexpected expenses. Moreover, cash reserves are used to finance different investment in the situations when markets are volatile and external methods of payments are too costly. The firms which are engaged in cash hoarding tend to have risky cash flows and poor access to capital markets. Additionally, Bates et al. (2009) documented that firms which have good investment opportunities, still tend to hold more cash on balance sheets because financial distress costs and adverse shocks outweigh benefits of these investments. Almeida et al. (2004) used a model of the cash flow sensitivities and found that firms which are financially constrained, save a cash as a fraction of cash flows.

Managers of firms have power over cash holdings and decide whether the firm would hold cash or cash holdings would be transferred to shareholders through the dividend payout. Consequently, Jensen (1986) documented that entrenched managers are likely to hold cash in the company rather than the payout to shareholders if firm facing poor investment

(10)

9

opportunities. Agency motives play a big role in cash holdings determination. Pinkowitz et al. (2006) found that firms tend to increase cash holding in countries with greater agency problems and showed that with greater agency problems between inside and outside shareholders, cash holdings would lose its worth.

All in all, Bates et al. (2009) did not just find precautionary and agency motives to hold cash but also they looked at the changes in firm characteristics and how idiosyncratic risk could affect cash holdings. Moreover, Opler et al. (1999) established determinants and implications of corporate cash holdings and provided evidence for precautionary cash savings and as well introduced the agency motives of cash holdings.

2.2 Financial constraints

Financial constraints of the firms could be explained by high costs of external finance and if firms are financially constrained, they rely on internal financial resources such as cash holdings and cash flows. Fazzari et al. (1988) found evidence that the sensitivity of investment to cash flow increases if external financing is costlier than internal financing, meaning that firm is financially constrained. However, Kaplan and Zingales (1997) criticized findings that the cash flow sensitivity to investment is a valid measure of financial constraints and found opposite effect. Furthermore, Almeida et al. (2004) used another approach and focused on the cash flow sensitivities to cash. They found that the cash flow sensitivity to cash increasing for financially constrained firms and decreasing for unconstrained. Moreover, Kim et al. (1998) discovered that cash holdings are positively correlated with industry cash flow volatility, meaning that firms are facing high costs of external finance in the industries with high cash volatility. Therefore, financially constrained firms hold more cash than unconstrained. Lastly, Erel et al. (2015) used a large sample of acquisitions in Europe and found that acquisitions potentially relieve financial

(11)

10

constraints and target firm’s cash holdings decline following the acquisition, especially for the target firm’s which are relatively small in size.

2.3 Private Equity firms

PE firm is an investment management company which makes investments in the private equity using different approaches such as leveraged buyouts (LBO’s), venture capital and growth capital. PE aims to acquire a controlling or considerable minority stake of the target in order to maximize the value of the investment with a subsequent exit through the initial public offering, another merger or acquisition or recapitalization (Kaplan & Stromberg, 2009). Furthermore, PE represents an improvement in the capital market with a potential to mitigate financial constraints (Amess, et al., 2015). In that case, it is reasonable to predict that companies will decrease their precautionary cash reserves after being acquired by PE firm. Moreover, improved level of corporate governance would facilitate drop of cash holdings due to agency motives (Jensen & Meckling, 1976). Nonetheless, PE firms backed by LBO’s have been often criticized but only for the creation of financial inflexibility (Alfred, 1990).

According to Kaplan (2009), PE firms create funds with the goal to raise capital that is then put towards the acquisition of a portfolio of firms. Gilligan and Wright (2014) stated that one of the strategies of acquisition of the portfolio firms is leveraged buyout strategy when private equity firms use debt finance, secured against target firms assets and/or obligations to future cash flows in order to make a deal.

However, there is controversy about the economic consequences for target firms subject to the leveraged buyouts. Boucly et al. (2011) as proponents suggest that after the leveraged buyouts, corporate governance induces incentives for managers to create value by reducing unprofitable expenditures following profit growth opportunities. On the other hand, Alfred

(12)

11

(1990) as critic argued that PE firms pursue short-term investment strategy in order to receive short-term profit and high level of leverage diverts cash away from long-term investment opportunities towards servicing the debt.

The expectations of the listed corporations regarding financial constraints are

ambiguous. Generally, listed companies experience financial constraints due to investor asymmetric information. However, some of the listed companies might face financial constraints due to unwillingness of the short-term investors to allow use resources and time in order to achieve long-term profitable goals. According to Lerner et al. (2011) turning to private through leveraged buyout may relax these financial constraints as private equity firms’ managers have the long-term expectations compare to stock market investors. In contrast, Carpenter and Petersen (2002) found that private firms mostly rely on internal finance. Therefore, they might suffer from underinvestment in profitable projects and may have problems with access to financial support from banks. Furthermore, owners of private firms may also constrain access to the external source of financing in order to fund some profitable projects due to their behavior and goals. For example, owners would like to remain in control of the company and that is why they do not want to dilute ownership through giving up their share of capital to external investors. This means that companies that are financially constrained are motivated to have cash holdings for precautionary reasons.

Boucly et al. (2011) found that PE firms use mechanisms associated with corporate governance and finance expertise in order to relax financial constraints and therefore have an access to finance. First of all, PE firms are active investors who monitor management and their strategic decisions by being represented on the board of directors. This improvement of corporate governance could mitigate moral hazard issue due to creditors confidence that company’s resources are used productively. Second of all, the financial expertise of PE firms is

(13)

12

the type of reassurance to creditors. Based on that, creditors would be more likely to provide funds for the investment.

3. Data

To evaluate the impact of LBO’s on cash holdings, it is necessary to have access to a sample of LBO’s transactions with financial measures of target companies before and after the transaction in order to detect changes. Most of the PE firm’s targets are privately held companies (Isil, et al., 2012). Given that there are different disclosure requirements around the world, it is almost impossible to collect data on the US companies due to low financial information disclosure of privately held companies. However, it is required to disclose financial data in the most European countries for private companies. Therefore, it is possible to collect target firms financials before and after the LBO transaction. Data about companies financials are available at the Amadeus database.

I started with collecting data about European LBO transactions from the Zephyr database. I used Zephyr database instead of Thomson One due to common data vendor “Bureau Van Dijck” of Amadeus and Zephyr and they both have common firm identifiers. Having the same identifier, it is possible to match LBO’s transactions from Zephyr with firms financials from Amadeus more precisely compare to using Thomson One or any other database.

According to Zephyr database, there were 595 European LBO transactions by PE firms from 2000 to 2014. Merging transactions from Zephyr with firm financials from Amadeus, I obtained only 141 transactions which have financial data before and after the transaction. Unfortunately, it is impossible to obtain more transaction due to limited data availability. My final dataset is consisting of 141 LBO transaction from 2005 to 2014 with acquirers as PE firms from all over the world and targets from the European countries. All variables are winsorized at 1% level in order to eliminate the effect of outliers. Only “Leverage” estimator is not winsorized,

(14)

13

however, all variables which are less than 0 and higher than 1 is dropped from the data set. Moreover, all the financials of target firms are converted from local to USD currency.

Table I presents statistics of target firms on European LBO deals from 2005 to 2014. Most of the deals are relatively small with the median targets total assets size of USD 46.2 million. Nevertheless, the size of distribution is slightly skewed. Mean of the targets total assets is almost USD 203.74 million which is around four times higher than the median. Moreover, it is clearly seen that deal numbers are increasing till 2007 and consequently declines till 2011. It can be explained by the financial crisis of 2007-2008.

(15)

14

Table II provides summary statistics for the financial variables of the target in the sample before and after the acquisition. Data used from unconsolidated statements provided by Amadeus database. It is clearly seen that mean total assets of the targets was USD 118 million before the LBO transaction took place and after the acquisition, mean total assets of target became USD 235 million. As it was expected, cash holdings declined after the LBO transaction, both mean and median is lower compare to pre-LBO cash holdings. Moreover, using LBO as a method of payment, target firms became more leveraged. However, it is not possible to draw a conclusion from the results from this table. To evaluate the impact of LBO’s on cash holdings and financial constraints it is needed to run first difference regressions with firm and country specific control variables. Additionally, there is an uneven distribution of accounting variables before and after LBO transaction due to limited availability of data. In this table, I presented all the available data before and after the acquisition. However, below in

(16)

15

regression analysis, I used on average 3 years before and 5 years after transaction because I am interested in the post-acquisition effect and PE firms try to exit the initial investment in up to a 6-year period.

4. Methodology

The empirical strategy of the thesis is aimed to identify the causal effect of the LBO’s on target firm's cash holdings. Therefore, I evaluate how cash holdings of target firms are changing after the acquisition of PE firms. Precautionary theory of cash holdings predicts that firms will hold less cash if the firm is not facing financial constraints. Given that financial constraints are eased by the LBO transaction, cash holdings of the target firm should decline after the acquisition. Consequently, my first hypothesis is:

H1: “Target firm's cash holdings decrease following the LBO”

This hypothesis can be tested using the first difference regression with dummy variable which shows post-acquisition effect on the dependent variable (cash ratio). I used the regression equation from the Erel et al. (2015) paper:

/ = + + +

Cash Holdings

Cash holdings is the dependent variable which measures cash and cash equivalents of the target firms. According to the traditional financial literature, there are two cash holdings measurements: cash and cash equivalents to total assets by Bates et al. (2009) and cash and cash equivalents to net assets by Opler et al. (1999). Cash to net assets measures how much firm has assets in cash. This ratio is used with the aim to see changes in daily cash holdings

(17)

16

due to changes in cash, cash flows, accounts receivable etc. Therefore, cash to total assets is a more convenient ratio to use in my research because I am interested in exploring changes in cash holdings in the end of the year.

After

After is a dummy variable which equals to one if acquisition took place and zero otherwise. All the variables are excluded of the same year when LBO transaction happened. Controls

Firm-specific and country-level control variables were used as explanatory variables. In order to control for variation in external financing opportunity, domestic credit to the private sector by banks normalized by gross domestic product (GDP), the market capitalization of listed domestic companies to GDP and annual growth of GDP per capita were used as country-level control variables. The sample consists of target firms which are part of the European Union (EU) that is why country level controls were used as EU variables.

Additionally, firm-specific controls were included in the regression. First of all, the natural logarithm of total assets was included. Second of all, I added target firms cash flow normalized by total assets with return on total assets (ROA). Furthermore, leverage, sales growth and the number of employees were added. The last three firm-level control variables could describe growth opportunities of the target firms. Moreover, all the firm related variables were excluded at the deal completion year. Furthermore, fixed effects for the target firms were included in order to control for time-invariant firm characteristics omitted in regression analysis.

(18)

17

After testing the impact of LBO’s on cash holdings, I proceeded with additional tests to verify whether the decline in cash holdings following the acquisition is explained by a reduction in financial constraints. To test that, I tested the cash flow sensitivity of cash and investment and target firm’s investment value itself using the same model changing the dependent variable.

5. Results

5.1 Cash Holdings

The results of the first difference regressions with cash ratio as dependent variable are presented in Table III. In column (1), the natural logarithm of total assets is only included as a firm-level control. In column (2), only cash flow to total assets is included as a firm-level control. Column (3) includes ROA instead of cash flow to total assets because these two estimates are highly correlated and could lead to bias in results. In column (4), the number of employees, leverage and sales growth were included to control for potential firm’s growth opportunities.

The estimates of cash holdings in Table III are in line with the hypothesis that target firm’s cash holdings decline after the LBO. Dummy variable’s “AFTER” coefficients documents that estimators after the acquisition are between -0.0254 and -0.0372. This means a decrease between 2.54% and 3.72% in the cash holdings following the acquisitions. Moreover, all specifications are showing that this fall in cash to assets ratio is statistically significant different from the zero. Additionally, these results confirm the trend presented in descriptive statistics in Table II, that there is a drop in cash holdings after the target is acquired.

(19)

18

The decrease in target firm’s cash holdings could be potentially explained by financial constraints reduction following the LBO’s by PE firms. Therefore, management of target firms now has lower precautionary incentives for holding cash.

Table III

The effect of the LBO’s on target firms Cash Holdings

This table presents results of the first difference regression model with cash holdings to total assets used as dependent variable. Dummy variable “AFTER” is equal to one for the years before the acquisition and zero otherwise. The accounting data of target firms is used from unconsolidated financial statements provided by Amadeus database. Definitions and formulas of estimates are provided in the Appendix. Firm and time fixed effect are included in all equations. Standard errors are presented in parentheses, *** p<0.01, ** p<0.05, * p<0.01.

(1) (2) (3) (4)

Cash Holdings/Total Assets

AFTER -0.0254** -0.0372*** -0.0281** -0.0281*

(-2.84) (-3.83) (-3.03) (-1.98)

Ln(Total Assets) -0.0164* 0.0137 -0.0116 0.00920

(-2.11) (1.54) (-1.39) (0.66)

Cash Flow/Total Assets 0.0818* 0.0587

(2.15) (1.05) ROA 0.000539 (1.70) Ln(Number of Employees) -0.00351 (-0.48) Sales Growth -0.00180 (-0.84) Leverage -0.0394 (-1.23) Credit of Banks/GDP 0.0123 0.00704 -0.00453 0.0194 (0.20) (0.11) (-0.07) (0.66) Market Cap/GDP 0.0328 0.00194 0.0162 -0.0751* (1.06) (0.06) (0.51) (-2.12)

GPD per Capita growth -0.00220 -0.00290 -0.00302 -0.00172

(-1.42) (-1.86) (-1.91) (-0.96)

Constant 0.388* -0.118 0.328* 0.00652

(2.39) (-0.66) (1.92) (0.02)

Observations 853 742 803 362

(20)

19

5.2 The Cash Flow Sensitivity of Cash and Investment

According to Almeida et al. (2004), firms might save cash from incremental cash flows because of financial constraints. In this paper, authors showed that financially constrained firms try to save a positive fraction of incremental cash flows in order to finance future projects with internal resources. On the other hand, firms which are not constrained already have their investments at the optimal level, so these firms do not have incentives to adjust their saving strategy. Moreover, financial literature provides evidence that there is a high correlation between estimates of the cash sensitivity to cash flow and other measures of financial constraints. Consequently, changes in financial constraints could be explained by the changes in the cash flow sensitivity of cash around the LBO transaction.

I used the same first difference regression model as presented above to estimate the change in the cash flow sensitivity to cash around the acquisition time. I used the change in cash to assets ratio as the dependent variable. Moreover, I added interaction variable (cash flow normalized by total assets multiplied by dummy variable “AFTER”). This interaction term indicates whether the firm year is after the LBO transaction. In this model, cash flow to assets estimator shows the cash flow sensitivity before target firms were acquired and the sum of cash flow to assets coefficient and interaction variable show the cash flow sensitivity after the acquisition.

The estimates for the cash flow sensitivity to cash holdings are presented in first two columns of Table IV. The coefficient on Cash Flow normalized by Total Assets shows if target firms were financially constrained prior the LBO. Column (1) presents a negative coefficient of Cash Flow to Total Assets including firm size, country-level controls and firm and time fixed effects. This result is not statistically significant and not in line with the theory.

(21)

20

Table IV

The effect of the LBO’s on the target firm’s Cash Flow Sensitivities to Cash

and Investment

This table presents results of the first difference regression model. In columns (1) and (2), the change of the cash holdings to assets is used as dependent variable. In columns (3) and (4), the gross investment normalized by total assets used as dependent variable. Dummy variable “AFTER” is equal to one for the years before the acquisition and zero otherwise. Interaction variable (AFTER x Cash Flow/TA) is included to observe changes in the sensitivities following the LBO. The accounting data of target firms is used from unconsolidated financial statements provided by Amadeus database. Definitions and formulas of estimates are provided in the Appendix. Firm and time fixed effect are included in all equations. Standard errors are presented in parentheses, *** p<0.01, ** p<0.05, * p<0.01.

(1) (2) (3) (4)

∆(Cash/Total Assets) Gross Investment/Total Assets

AFTER -0.0464 -0.00570 0.0515* 0.0346

(-1.91) (-0.20) (2.38) (1.17)

Cash Flow/Total Assets -0.113 0.0827 0.0438*** 0.0436**

(-1.06) (0.61) (4.65) (3.16)

AFTER x Cash Flow/TA 0.187 -0.0753 -0.0310** -0.0278

(1.67) (-0.55) (-3.03) (-1.82) Ln(Total Assets) 0.0389** 0.0424* 0.0384** 0.0634** (2.82) (2.26) (2.69) (2.74) Ln(Number of Employees) 0.0147 -0.0158 (1.52) (-1.29) Sales Growth 0.00219 0.0113** (0.82) (3.15) Leverage 0.114* -0.0608 (2.57) (-1.12) Credit of Banks/GDP -0.181 -0.0427 0.161 0.279 (-1.54) (-0.31) (1.38) (1.78) Market Cap/GDP -0.115* -0.0732 -0.0199 -0.00233 (-2.58) (-1.60) (-0.44) (-0.04)

GDP per Capita growth -0.00536* -0.00207 -0.000865 0.000427

(-2.54) (-0.88) (-0.39) (0.14)

Constant -0.396 -0.808* -0.857** -1.341**

(-1.33) (-2.20) (-2.80) (-2.91)

R-squared 0.056 0.102 0.063 0.126

However, in column (2), when I added more firm-level controls, the coefficient of Cash Flow to Totals Assets became positive and equal to 0.0827. Almeida et al. (2004) used the same model in their research using different sample consisting of large and publicly traded U.S. companies. In the constrained subsamples, they found statistically significant coefficients of

(22)

21

0.05 and 0.06. In the subsample of unconstrained firms, estimates were close to zero. Given that there are differences in the samples, the coefficient from column (2) is roughly the same to the estimates found by Almeida et al. (2004). Unfortunately, I cannot draw the conclusion that target firms are financially constrained before the acquisition.

The interaction variable with dummy variable “AFTER” is only negative and equals to -0.0753 in the Column (2) and this means that the cash flow sensitivity of cash declined after the acquisition. Moreover, this coefficient has almost opposite value to the cash flow to assets estimate. In column (2), the sum of both coefficients is very close to zero, suggesting that financial constraints of target firms relieved after the LBO transactions. The result in column (2) documents that prior LBO target firms are financially constrained and there is a reduction in constraints after the LBO. However, these results are not statistically significant from the zero.

Another approach of measuring financial constraints is to estimate the sensitivity of investment to cash flow. This approach is suggested by (Fazzari, et al., 1988). The main idea behind the sensitivity of investment to cash flow is that unconstrained firms should invest in all value-enhancing projects disregarding firm’s current financial conditions. On the other hand, managers of constrained firms should more rationally choose investment projects and if firms cash flows are increasing this would enable to undertake more projects. Therefore, financially constrained firms should observe a relation between firm’s cash flow and its investment (Fazzari, et al., 1988).

I used the same model for estimating the cash flow sensitivity of investment, changing the only dependent variable to gross investment normalized by total assets. The estimates are reported in columns (3) and (4) of Table IV. In both columns, the cash flow to assets coefficients are positive and statistically significant different from zero. This finding suggests

(23)

22

that target firms were financially constrained prior to the LBO. Nevertheless, the interaction coefficients with “AFTER” dummy are negative and one of them is statistically significant at 5% level, suggesting that the cash flow sensitivity of investment decreased after target firms were acquired.

The results of the cash flow sensitivity to investment regression could be interpreted as financial constraints are declined following the acquisition. However, this explanation of the results could be criticized. The cash flow and investment opportunities are probably correlated between each other. If the cash flow sensitivity to investment is positive, then it would reflect investment opportunities instead of financial constraints (Kaplan & Zingales, 1997). Nevertheless, cash holding decreased following the acquisition and one of the cash flow sensitivity to cash result is in line with the theory that the LBO’s reduce financial constraints. Consequently, I could consider the cash flow sensitivity to investment results as additional confirmation that LBO’s lead to financial constraints relaxation.

5.3 Target firm’s investment

Finally, additionally to cash holdings and cash flow sensitivity findings, I decided to check another method if financial constraints are eased following the LBO. Financial constraints force firms to have a higher cost of capital compare to the situation then firms operate on stable markets. Given that the previous results showed that the acquisitions relieve financial constraints, target firm’s managers could change their investment strategies and consequently undertake more investment at a lower cost of capital.

(24)

23

Table V

The effect of the LBO’s on target firm’s Investment

This table presents results of the first difference regression model with the gross investment normalized by total assets used as dependent variable. Dummy variable “AFTER” is equal to one for the years before the acquisition and zero otherwise. The accounting data of target firms is used from unconsolidated financial statements provided by Amadeus database. Definitions and formulas of estimates are provided in the Appendix. Firm and time fixed effect are included in all equations. Standard errors are presented in parentheses, *** p<0.01, ** p<0.05, * p<0.01.

(1) (2) (3) (4)

Gross Investment/Total Assets

AFTER 0.00370 0.0101 0.0162 0.00281

(0.0158) (0.0170) (0.0159) (0.0238)

Ln(Total Assets) 0.0342** 0.0349** 0.0307** 0.0639***

(0.0136) (0.0143) (0.0139) (0.0233)

Cash Flow/Total Assets 0.211*** 0.252***

(0.0574) (0.0941) ROA 0.00121** (0.000498) Ln(Number of Employees) -0.0151 (0.0123) Sales Growth 0.0111*** (0.00360) Leverage -0.0441 (0.0537) Credit of Bank/GDP 0.216* 0.190 0.102 0.298* (0.112) (0.117) (0.112) (0.158) Market Cap/GDP -0.00842 -0.0230 -0.0206 0.000149 (0.0443) (0.0456) (0.0438) (0.0593)

GDP per capita Growth 0.000121 -0.000644 -0.00112 0.000807

(0.00213) (0.00221) (0.00211) (0.00299)

Constant -0.788*** -0.793** -0.618** -1.359***

(0.294) (0.308) (0.299) (0.463)

R-squared 0.019 0.046 0.029 0.115

I used the same regression model with a change of dependent variable to investment normalized by total assets. Table V presents estimates of gross investment to total assets with the same controls used in Table III for cash holdings. All “AFTER” dummy variables are having positive coefficients, indicating an increase in investment following the acquisition. However, these coefficients are not statistically significant at any level. In contrast, descriptive

(25)

24

statistics documented decrease of the mean gross investment to assets after the LBO’s of target firms. Therefore, the results of my research show that target firm’s cash holdings decreased following the LBO and these firms became less financially constrained.

6. Robustness check

I explored changes in the cash holdings and financial constraints of target firms before and after the acquisition by PE firms. However, it is impossible to predict that these changes would have been opposite to findings if those target firms did not experience LBO. On the other hand, it is possible to observe the changes in policies of firms which are similar in the size and from the same country/industry to the targets of PE firms and which have not been acquired. Apparently, it is impossible to control for every observed or unobserved factor which could potentially influence the results. Apart from the LBO, any unobservable factor could have influence financial policies of target firms. Consequently, these factors would have also affected financial policies of comparable firms. To test these unobservable factors which could coincide with the LBO and affect target firm’s financial policies, I ran “placebo test” on the sample of matched firms.

I constructed a sample of similar firms to the target firms which were acquired. I matched target firms with the firms from Amadeus database by the same country and the same national industry code. Additionally, firms were matched based on the value of total assets one year before the LBO with a possible difference of 20%. Finally, I was able to collect 141 comparable firms to the target firms which were acquired by PE firms.

In the placebo test, I ran the similar regressions to those in Tables III – V on the sample of matched firms. The results of those regressions are presented in Table VI. Obtained results

(26)

25

are in sharp contrast to the previous results. The coefficient on cash holdings in the first column, also predicts a decrease in cash holdings for the matching firms.

Table VI

“Placebo test” on the matched firms sample

This table presents the results of the “Placebo” test using the sample industry, size, and country matched firms. Each target from the sample were matched with firms similar in size (total assets) and from the same country and industry. In column (1), cash holdings to total assets were used as dependent variable. In column (2), the change in cash to total assets was used as dependent variable. In columns (3) and (4), the gross investment to total assets was used as dependent variable. Dummy variable “AFTER” is equal to one for the years before the acquisition and zero otherwise. Interaction variable (AFTER x Cash Flow/TA) is included to observe changes in the sensitivities following the LBO. The accounting data of target firms is used from unconsolidated financial statements provided by Amadeus database. Definitions and formulas of estimates are provided in the Appendix. Firm and time fixed effect are included in all equations. Standard errors are presented in parentheses, *** p<0.01, ** p<0.05, * p<0.01.

(1) (2) (3) (4)

Cash/Total

Assets ∆(Cash/Total Assets) Inv/Total Assets Inv/Total Assets

AFTER -0.00665 -0.0119 -0.00543 -0.0349

(-0.44) (-0.42) (-0.21) (-1.60)

Cash Flow/Total Assets 0.226*** 0.337* 0.311* 0.0513

(5.38) (2.36) (2.41) (0.82)

AFTER x Cash Flow/TA -0.229 -0.310*

(-1.53) (-2.30) Ln(Total Assets) -0.0329*** 0.0582** 0.118*** 0.111*** (-3.58) (3.19) (7.35) (7.01) Credit of Banks/GDP -0.162 1.480 -0.915 -0.746 (-0.19) (0.45) (-0.31) (-0.25) Market Cap/GDP 0.0148 0.121 -0.147 -0.121 (0.06) (0.26) (-0.35) (-0.29)

GDP per capita Growth -0.00416 0.0891 -0.0283 -0.0215

(-0.10) (0.68) (-0.24) (-0.18)

Constant 0.840 -2.649 -0.960 -1.005

(0.82) (-0.72) (-0.29) (-0.31)

R-squared 0.083 0.073 0.128 0.119

However, this decline is much smaller (0.7% vs 2.5%) and the coefficient is not statistically significantly different from the zero. While I found an increase in gross investments normalized by total assets for the target firms, matched firms experienced opposite effect on investment. Both cash flow sensitivity of cash and investment are significantly positive for

(27)

26

matched firms before the “LBO”. Additionally, both cash flow sensitivity of cash and investment declined after the “LBO” and only one coefficient is statistically significant at 10%. Overall, the results of the placebo test suggest that the matched firms do not encounter changes in financial policies related to financial constraints reduction.

7. Potential limitations

7.1 Cross-country difference.

The first potential issue in my research could be that sample of LBO deals includes cross-border deals. Targets firms are from the European Union, but PE firms are from around the world. Given that, it is impossible to completely control for institutional differences in those deals. For instance, there are restrictions in cash and cash flow movement from parent companies to their subsidiaries in some countries. Nevertheless, target firms remain independent companies in case of acquisition by PE firms and PE firms could only change management and organize monitoring of existing management actions, so these cross-border restrictions could not bias the results.

7.2 Managerial risk aversion.

Another potential issue which could bias results that target top management could be replaced by less risk-averse management after the acquisition. On the other hand, if top management stays the same under PE firm control, a fraction of their ownership declines to zero because of acquisition. Consequently, management is going to be less risk averse having zero ownership of the company, so they would apply lower discount rates in the valuation of potential projects, thereby showing that firm is not financially constrained. In this case, if acquisition leads to lower risk aversion among management then it would be in line with our

(28)

27

findings that acquisition leads to a reduction in financial constraints. However, the sample consists of LBO’s deals with 100% ownership acquisition. All the ownership is transferred to the PE firms and managerial risk aversion, in that case, cannot be viewed as a potential concern. 7.3 Selection issues.

The main issue of this research is selection issue because of Amadeus database and data availability. Amadeus has a maximum data of 10 years per firm. Given the sample of LBO deals from 2000 to 2015, data is not available for deals happened from 2000 to 2004. I collected all available data for target firms that firms at least had accounting data one year before and after the LBO. However, the data sample is not balanced. Deals which happened earlier in the sample have more post-LBO observations and deals which happened later, have more pre-LBO observations. Since the sample is very small and consists only of 141 deals with firm financials, I decided to use more than one before and after LBO’s years in order to have a larger number of observations. Moreover, I could not use an equal number of years pre and post-LBO due to small sample, so I used on average 3 years before and 5 years after the LBO’s. I used more year observations after transaction because of focus on post-LBO financial situation of target firms and because PE firms on average try to exit the investment in 3-6 years. Additionally, I included time fixed-effect in my regression model to adjust for any macroeconomic shocks to ensure that unbalanced data sample will not significantly bias results.

(29)

28

8. Conclusion

Last decade showed a tendency that firms tend to hoard cash mostly due to

precautionary reasons. Financially constrained firms are forced to hold more cash in order to finance some future profitable projects because external sources of financing are too expensive or even sometimes impossible to obtain. Existing literature justifies that acquisition could potentially lead to a reduction of targets’ financially constraints and this could be observed through a decrease in cash holdings. Previous research focused on acquisitions made by strategic buyers and found that target firms cash ratio drops after the acquisition. However, this decline could be biased because of acquirer incentives – parent firms could potentially exploit targets and engage in tunneling resources from them. Given that, I decided to explore the LBO’s by PE firms and how LBO’s affect target firm’s cash holdings. In this case, target firms remain autonomous and PE firms pursue a strategy of improving target firms’ business and consequently sell it to another investor or exit investment through IPO.

It is difficult to test empirically impact of LBO’s on target cash holdings because most

of the targets are privately held and financial information is needed before and after the acquisition of target firms. Disclosure requirements in the European Union enable to construct a sample of LBO’s across Europe. Final data sample allows testing the hypothesis that target firms cash holdings will decline after LBO transaction.

Using the first difference regression model, I document that there is a decline in target

firm's cash holdings after the LBO and this result is in line with my hypothesis. To confirm that cash holdings declined due to a reduction in financial constraints, I decided additionally to check the LBO’s impact on the cash flow sensitivity to cash and investment and investment of target firms. Theory predicts that the cash flow sensitivity to cash and investment would increase and firm’s investment amount would decrease if the firm is financially constraint. These effects should be reversed if financial constraints are mitigated following the LBO. Running additional

(30)

29

regressions, I found that the cash flow sensitivity to cash and investment is negative and investment amount increased after target firms were acquired by PE firms. These results confirm that target firms were financially constrained before the LBO’s and after the LBO’s, these targets became less financially constrained. Lastly, I run placebo test for robustness check. I matched acquired targets to other firms from the same country, same industry and with approximately the same value of total assets which have not been acquired at that time. Placebo test revealed that matched firms which were not subject to the LBO’s did not experience any changes in financial policies indicating that financial constraints have not been relieved.

Overall, every research faces some potential limitations. Cross-border differences and

managerial risk aversion did not influence results of my research. However, selection issues could potentially bias the results. To deal with potential selection issues, there should be more accounting data for acquired targets available. This would enable to have a larger sample and select more evenly before and after LBO observations.

(31)

30

Appendix

This table presents description and formulas of accounting variables used in the research.

Firm-level variables (from Amadeus database)

Ln (Total Assets) Natural logarithm of total assets (in USD dollars)

Cash/Total Assets Cash and cash equivalents divided by total assets

Gross Investment/Total Assets

((Fixed assets – lagged fixed assets + Depreciation)/Total assets)

Cash Flow/Total Assets Cash flow divided by total assets

∆ (Cash/Total Assets) Cash/Totals assets – lagged(Cash/Total assets)

Sales growth (Sales – lagged sales)/Total assets

Leverage (Long-term debt + Current liabilities)/Total assets

Country-level variables

Credit of Banks/GDP Domestic credit to the private sector by banks (% of GDP)

of the European Union. (Source: the IMF and The World Bank)

Market Cap/GDP Market capitalization of listed domestic companies (% of

GDP) of the European Union. (Source: The World Bank)

GDP per capita Growth Annual percentage change in GDP per capita (Source: The

(32)

31

References:

Alfred, R., 1990. "The staying power of the public corporation". Harvard business review, pp. 96-104.

Almeida, H., Campello, M. & Weisbach , M. S., 2004. "The Cash Flow Sensitivity of Cash". The Journal of Finance, 59(4), pp. 1777-1804.

Amess, K., Stiebale, J. & Wright, M., 2015. "The impact of private equity on firms' patenting activity". European Economic Review, pp. 421-433.

Baek, J.-S., Kang, J.-K. & Lee, I., 2006. "Business Groups and Tunneling: Evidence from Private Securities Offerings by Korean Chaebols". "The Journal of Finance", 61(5), pp. 2415-2449.

Bates, T. W., Kahle, K. M. & Stulz, R. M., 2009. "Why Do U.S. Firms Hold So Much More Cash than They Used To?". The Journal of Finance, p. 1985–2021.

Boucly, Q., Sraer, D. & Thesmar, D., 2011. "Growth LBOs". Journal of Financial Economics, pp. 432-453.

Carpenter, R. E. & Petersen, B. C., 2002. "Is the Growth of Small Firms Constrained by Internal Finance?". The Review of Economics and Statistics, 84(2), pp. 298-309.

Erel, I., Jang, Y. & Weisbach, M. S., 2015. "Do Acquisitions Relieve Target Firms’ Financial Constraints?". The Journal of Finance, 70(1), pp. 289-239.

Fazzari, S. M., Hubbard, G. R. & Petersen, B. C., 1988. "Financing Constraints and Corporate Investment". Brooking Papers on Economic Activity, 1988(1), pp. 141-195.

Gilligan , J. & Wright, M., 2014. "Private equity demystified". ICAEW, pp. 1-44.

Isil, E., Rose, L. C. & Weisbach, M. S., 2012. "Determinants of Cross-Border Mergers and Acquisitios". The Journal of Finance, 67(2), pp. 1-38.

Jensen, M. C., 1986. "Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers". "The American Economic Review", 76(2), pp. 323-329.

Jensen, M. C. & Meckling, W. H., 1976. "Theory of the firm: Managerial behavior, agency costs and ownership structure". Journal of Financial Economics, pp. 305-360.

Kaplan, S. N., 2009. "Leveraged Buyouts and Private Equity". Journal of Economic Perspectives, pp. 121-146.

Kaplan, S. N. & Stromberg, P., 2009. "Leveraged Buyouts and Private Equity". Journal of Economic Perspectives, 23(1), pp. 121-146.

(33)

32

Kaplan, S. N. & Zingales, L., 1997. "Do Investment-Cash Flow Sensitivities Provide Useful Measures of Financing Constraints?". The Quarterly Journal of Economics, 112(1), pp. 169-215.

Kim, C.-S., Mauer, D. C. & Sherman, A. E., 1998. "The Determinants of Corporate Liquidity: Theory and Evidence". The Journal of Financial and Quantitative Analysis, 33(3), pp. 335-359.

Lerner, J., Sorensen, M. & Stromberg, P., 2011. "Private Equity and Long-Run Investment: The Case of Innovation". The Journal of Finance, 66(2), pp. 445-477.

Opler, T., Pinkowitz, L., Stulz, R. & Williamson, R., 1999. "The determinants and implications of corporate cash holdings". Journal of Financial Economics, Issue 52, pp. 3-46.

Pinkowitz, L., Stulz, R. & Williamson, R., 2006. "Does the Contribution of Corporate Cash Holdings and Dividends to Firm Value Depend on Governance? A Cross-country Analysis". The Journal of Finance, LXI(6), pp. 2725-2751.

The impact of the leveraged buyouts on the target firm's cash holdings