Are financial constraints a determinant for long-run post-merger performance?

(1)

Are Financial Constraints a Determinant for

Long-Run Post-Merger Performance?

Master’s thesis

Author: Floris Kerkhof

Student number: 10017682

Supervisor: dr. V.N. Vladimirov

University of Amsterdam

Faculty Business Economics

Msc Finance Track

June 2015

(2)

2 Statement of Originality

This document is written by Student Floris Kerkhof who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

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

(3)

3

1. Introduction

After an IPO, mergers are seen as one of the biggest events in the life of a company. Here two companies are combined to form a new entity consisting of their combined assets. It can lead to success stories as the P&G acquisition of Gillette Talent. Gillette Talent was incorporated in a seamless manner with a high retention rate of employees (Kanter, 2009). On the other hand there are numerous examples of mergers that almost bankrupted the acquiring firm. Such as the Nextel Communications Inc. acquisition of $35 billion in 2005 by Sprint Corp., the Palm Inc. acquisition for $1.2 billion by Hewlett-Packard Co. in 2010. Both were planned to lead to new innovative products or increasing market share in new sectors, but ended up to be big financial failures for the acquiring firms (Thurm, 2013). The literature shows that the target firm usually profits the most from a merger (Healy et al., 1992). For the acquiring firm there is less consensus on the effect of mergers on their future performance. As these examples above show, even though the merger can have disastrous result for the acquiring firm. This does not seem to discourage acquiring firms, because there are still large amounts of mergers that are completed (Harford, 2005).

The performance of acquiring firms after a merger is part of the area of interest where this thesis will focus on. The literature about this subject is enormous and it would be difficult to truly make some new ground here. There are many different control variables, such as the payment of the merger and the industry that both parties operate in are found to be indicators of the long-term operating performance. They are seen as determinants that influence the performance that is found after a firm has merged. In this paper an overlooked control variable will be tested. The main focus will be on the performance of target firms that are financially constrained.

The use of this control variable is based upon a finding in a recent paper by Erel et al. (2015). The main finding of this paper is that after a successful merger the degree of financial constraints in the target firm becomes significantly lower. This means that firms have a higher liquidity and are less concerned about their capital needs. This gives these firms an opportunity to pursue more positive net present value (NPV) investments with their newfound liquidity after a merger. This should therefore lead to better performance for the new combined firm than for the two separated firms before the merger. In this thesis it will be tested if this effect can be found or if it is only applicable theoretically. This leads to the following research question:

(5)

5

Are the existence of financial constraints in a target firm a positive determinant for future long- term operating performance after a merger?

The performance of financially distressed firms in mergers has been tested by Clark and Ofek in 1994. They find that bidders that acquired distressed targets tend to decline in the long-run. This would mean that financially distressed firms have a negative effect on operating income. A reason for this finding can be that, at that time, there was no certain method yet to take the effect of financial constraints into account and this could be improved upon. Hadlock and Pierce (2010) researched the literature regarding financial constraints and found that the most important proxies are firm size and age. In this thesis the goal is to use the SA index measurement by Hadlock and Pierce and test how this affects long term operational performance and if this can confirm the finding of Erel et al. (2015).

The contribution to current research is to determine how important financial constraints in the target firm are in determining long term operating performance. The choice for using operating performance instead of stock performance is based on two reasons. The first reason is that the current research regarding operating performance is inconclusive. Some studies found a decline in operating performance (Clark and Ofek, 1994) and others an increase (Healy et al., 1992; Heron and Lie, 2002; Megginson, 2004) or no significant change at all (Ghosh, 2001; Moeller and Schlingermann, 2005).The second reason is that operating performance is less depended on market sentiments. The justification for researching financial constraints is based on the fact that according to Erel et al. (2015) this could be important in the decision making regarding potential merger targets. There is a significant amount of research on long term performance, but the effect of financial constraints has not been thoroughly tested (Clark and Ofek, 1994). The expectation is that the negative effect that financial distress causes will be low for this sample. Acquirers deemed their targets profitable enough to merge with and did proper research regarding this. The results in this thesis allow for a better understanding in determining if financial constraint in a target firm have an effect on future performance. This could lead to a higher importance on the level of liquidity of a target companies and if this could signal current untapped growth potential. An additional difference with previous literature is that a more recent time period is used.

This thesis will have the following structure. In the second section the relevant literature regarding merger performance will be discussed. The use of financial constraints as a dummy and the other control variables will be explained. The third section consists of the methodology.

(6)

6 Here the hypothesis, the selected sample and the research method will be described in more detail. In section 4 the data will be summarized and the characteristics of the sample will be described. Section 5 consists of the results and the outcome of the hypotheses. In section 6 a robustness test will be done on the sample that is used. In the last part the main findings of this paper and potential methods for subsequent research will be discussed

(7)

7

2. Literature review

The studies regarding the performance of mergers are comprehensive. There is a reasonable consensus that mergers create value for bidders and targets combined, mostly focused on the target (Healy et al., 1992; Linn and Switzer, 2001; Heron and Lie, 2002). The main difficulty is how to control for all the factors that influence performance and which benchmark group to use. The literature review will discuss the literature regarding merger performance, long-term operating performance, the use of financial constraint as a determinant and the different control variables that will be used.

2.1 Merger performance

The effect of a merger on the performance of the newly merged firms is still inconclusive. There are many different variables that can determine the success of a company. Among the literature the definition of performance and also the method of measurement that is used can differ greatly. Zollo and Meier (2008) did a comprehensive research on 88 articles about mergers from 1970 to 2006. They found that there were three broad categories of measuring that were used. The short term event study metrics, long-term event window and subjective measuring. They did not find a clearly superior method of measuring. They did find that short-term event studies do not correlate with any other method. These results are mostly based on what the market expects to happen, but the market is prone to also make mistakes. Zollo and Meier (2008) also found that the market fails to win on average. Therefore this will not be used in this paper. Subjective measures are usually used by organizational and management researchers and will therefore not be used here. In this paper the main focus is on finding the long-term operating performance after a merger and therefore accounting based measures will be used.

2.2 Long-term operating performance

The literature regarding mergers and mainly stock price performance is very extensive, but the literature about long-term operating performance is less comprehensive and shows inconclusive results. The change in performance after a merger is found to be increasing, decreasing or not significantly changing.

The most notable paper that has researched long-term operating performance was by Healy et al. (1992). They argued that previous research was flawed because they could not distinguish the merger specific gains. They made a model with the post-merger cash flow as a dependent variable and the pre-merger cash flow as an independent variable. This way the

(8)

8 intercept would capture the increase in cash flow by the acquisition. Another important addition by Healy et al. (1992) was the use of an industry median benchmark. This was done to account economy-wide or industry or industry factors that can affect the outcome. These methods lead to less measurement errors. Therefore in most of the subsequent research regarding long-term operating performance of mergers this method was used as a basic framework.

Healy et al. (1992) found an increase in operating cash flow after mergers, unlike earlier studies. He argues that operating synergies and tax savings could be reasons for the positive effect of mergers on operating performance. Positive results are also found by Megginson (2004), Heron and Lie (2002) and Linn and Switzer (2001).

Ghosh (2001) did not find any significant difference in post-merger performance. He argues that the size of the acquiring firm is systematically different than the benchmark sample and that mergers are made after a period of superior performance. Just like Healy et al. (1992) he used the same measure for operating performance and scaling measure. The benchmark was used in a different manner. It was used to adjust for size and pre-merger performance and industry performance. Therefore he argues that not taking into account the pre-merger performance and size of the merger in determining the control group leads to biased results.

Heron and Lie (2002) also tested for pre-merger performance and industry. They found that acquiring firms did outperform their industry control group. This shows that there is still a lot of uncertainty about the effect of mergers on post-merger operating performance.

Clark and Ofek (1994) were the only one to find a negative effect. This can be explained by the fact that they used a sample of financially distressed firms and their aim was to find if they were successfully restructured after merging. Another factor that could explain the negative finding is the relatively low sample size that was used.

Based on the literature regarding long-term operating performance there cannot be made a precise estimate on the findings for this paper. The Clark and Ofek (1994) paper is most similar in the goal of the research. This is because of the low sample size, older dataset and different research methods. As seen before with Megginson (2004), Heron and Lie (2002) and Linn and Switzer (2001), Ghosh (2001) there is no certainty if the expected results will be positive or not significant.

2.3 Financially constrained firms

When a firm faces financial difficulties there are several options for a firm to solve this problem. Mergers are seen as a useful tool to restructure financially distressed target firms. Clark and

(9)

9 Ofek (1994) found that bidders have a negative post-merger performance after restructuring failing firms. This means that for the acquiring firm it is not beneficial to acquire financially distressed firm. They did find that mergers were a successful tool to restructure financially distressed target firms.

Keynes (1936) gives more insight in the motivation for acquiring firms when merging with financially distressed target companies. He found that a liquid balance sheet gives firms the opportunity to undertake positive net present value (NPV) projects. Additional research by Erel et al (2015) found that acquisitions relieve financial constraints. This would mean that financially distressed firms are more liquid after mergers, thus they are more able to pursue positive NPV projects. Therefore relieving financially constrained targets of their liquidity problem should enable the target firm to achieve higher profitability. This could be a motivation for managers to undertake mergers, because of the growth potential that can be achieved by relieving firms of financial constraints.

Almeida et al (2004) found that the relation between liquidity and financial constraints determines firm behavior. The rationale behind this is that firms that are constrained are more likely to save cash than unconstrained firms who have no direct needs for cash. Almeida uses four schemes for determining financial constraints that support his hypothesis and one based on Kaplan and Zingales’s (1997) KZ index that shows contradicting results. The four schemes that were used are payout ratio of the target firm, the asset size, the availability restructure its operations, restructure the debt or by being acquired by another firm. These four methods all found that for financially constrained firms there is a significantly higher level of propensity to save cash.

Prominent research has been done by Hadlock and Pierce (2010) who researched most measures of financial constraints and their accuracy. He researched the Kaplan and Zingales (KZ) index, Whited and Wu (WW) index and firm-specific characteristics. The main finding was that the best determinants for financial constraints where firm size and age in an index with further determined factors. Corresponding to the findings of Almeida et al (2004) he also found that the widely used KZ index was a bad measure of financial constraints and advised not to use it in the future. The WW index shows better results, but has an endogeneity problem between leverage and cash flow. The SA index, consisting of a combination of firm size and age, shows the best result for measuring financial constraints and will be used in this research. Therefore in this paper the main focus will be on the SA index.

(10)

10

2.4 Additional control variables for operating performance

To determine the effect that financial constraints has on long term operational performance it is important to control for other factors that determine long-term operational performance. These variables are based on what is found in the literature as the most commonly used measures.

One of the most researched determinants for firm performance is the payment method that is used. Acquiring firms prefer to pay with stock in a merger transaction. The rationale for this is linked with the literature regarding merger waves. Mergers are likely to occur in a period of economic success. (Harford, 2005) This translates to periods with high amounts of merger frequency. In a period of financial success the firm is overvalued and therefore the stock of the firm is relatively overvalued. Cash is therefore only rationally used when the acquiring firm is undervalued or that the acquirer has significant reasons to believe that the gains of undergoing the merger are worth the extra costs that are entailed by paying with cash. This means that mergers that are paid with only cash give a signal about the subsequent performance of the firm after the merger. Loughran and Vijh (1997) and Megginson et al (2004) found that cash paid mergers outperform stock paid mergers regarding long-run stock performance. Linn and Switzer (2001) found similar results for the long-run operating performance. The main argument is that cash is used to deter competing bids in the case that the acquiring firm has private information regarding possible synergies (Fishman, 1989). Healy et al (1992) showed that there is not a significant influence of cash on performance, but that there was positive performance of mergers on the long term. The expectation is that if cash is used as the main payment method this increases the long-run operating performance.

Focused mergers are also shown to be a determinant that signals information about the performance of a merger. Theory shows that diversification can have a positive or negative effect on value. Benefits include greater operating efficiency, less incentive to forego positive NPV projects, economics of scope and scale. Potential costs are misalignment between managers, poor performing divisions draining resources of better performing divisions and resources being used for value-decreasing investments (Clark and Ofek, 1994). Clark and Ofek found that diversification reduces firm value. Heron and Lie (2002), Linn and Switzer (2001) also found that mergers in the same industry strongly outperform diversifying mergers. This gives significant reason to suspect that if a merger is focused, it can be a strong positive determinant in determining firm performance and therefore a relevant control variable.

(11)

11

2.5 Predictions and contribution to the literature

The research in table 1 shows that most literature finds an increasing post-merger performance. For this reason it is expected that in this research a positive growth of post-merger performance is found. Eres et al. (2015) expects that financial constraint is a determinant of future merger performance. The method for identifying financial constrained firms of Hadlock and Pierce (2010) will be used. This should find a positive effect of financial constraint on long-term merger performance.

The contribution the literature is that different adjustments to measuring long-term merger performance found in the literature will be combined. Furthermore the specific addition of testing the effect of financial constrained firms was only done by Clark and Ofek (1994). They used a small sample size and different methodology. For that reason this research should bring new insights and give additional perspective on the existence of financial constraints in firms as a determinant for acquiring targets.

2.6 Hypotheses

There are two objectives for this thesis. The first objective is to find if merging leads to an increase in long-run operating income for the combined firm. This is done to find out how this result corresponds to the findings of other research. Thereafter it will be tested if financial constraints are important factor in determining the long-run operation performance by adjusting for different control variables. This will be done by testing the main sample of merging firms with a benchmark of companies that operate in the same industry in the same period. The expectation is to find a positive coefficient for financial constrained firms such as the theory in the literature predicts (Erel et al., 2015). The following hypotheses will be used:

H1: The abnormal operating performance for the combined firm is higher than the abnormal operating performance of both firms before the merger separately.

H2: The existence of financial constraints is a positive determinant for a positive long-run post-merger abnormal operating performance.

(12)

12

3. Methodology

The methodology will be based around the hypothesis that were mentioned in the second part. In the literature there is a debate about the method of measuring operating performance. The method that is used in this paper will be explained. Furthermore the choice of the sample that is used and method of acquiring the data will be discussed. Thereafter the use of benchmark and pre- and post-merger performance will be described. In the last section the use of testing method and the regression will be mentioned.

3.1 Measure of operating performance

The first hypothesis is focused on the existence of a positive long-run abnormal operating performance. Therefore first it has to be determined how to define operating performance. There are different methods used throughout the literature for the long-run operating performance. Pre-tax operating cash flows is usually divided by sales or assets. Pre-tax operating cash flow can also be measured in different ways.

Operating performance is used to exclude the method of accounting or the choice of financing influencing the measurement of changes in operating income. (Healy et al., 1992; Linn and Switzer, 2001). Furthermore stock prices do not show whether mergers results in real economic gain. Stock prices also do not show what the sources of economic gains after a merger are. Therefore operating performance is used instead of stock performance. Linn and Switzer measured tax income before extraordinary items, plus depreciation or amortization, net interest expense and total income taxes. Healy et al. (1992) subtracts sales with costs of goods sold and selling and administrative expenses and adds depreciation and goodwill expenses.

There are also different methods for dividing the cash flows. Healy et al. (1992) divides pre-tax operating cash flow by the market value of assets. Market value of assets is the market value of equity plus the book value of debt. The advantage for using this is that it gives a return value. An alternative method was employed by Heron and Lie (2002) and Clark and Ofek (1994) to divide the operating income by sales. This is done to account for the fact that assets are likely to rise given the increase in operating income and are more sensitive to market expectations. Assets may understate the poor performance of the financially constrained firms (Clark and Ofek, 1994). The method of Healy (1992) will be used to measure operating performance in this research.

(13)

13

3.2 Sample and selection criteria

The data is gathered for the mergers that were concluded in the time period 1994-2012. This period is chosen to ensure that a wide sample is used and that enough data is available to test the hypotheses. No distinction is made between public and private data, the only requirement is that all relevant data is available. For every observation the operating data one year before and after the merger has to be available for the firms that are used. Furthermore an US sample will be used, because this is also done for other prominent research in this area. The year that corresponds with the effective date of the merger that is found in Thomson One will be regarded as the start of the merger. From this point the operating performance of the merged firm can be measured.

The data on mergers is acquired using the Thomson One database. In Thomson One there are 69958 mergers found that fulfill the requirements mentioned above. All observations that have a missing acquirer or target Datastream code are excluded from the sample. This leads to 5313 observations. The following requirements are used according to Moeller and Schlingermann (2005). Their requirements are used to create a usable sample. Firms with a SIC code starting with a 6 are excluded from the sample. These firms are shown to behave significantly different from regular firms and are also excluded in similar literature (Heron and Lie, 2002). Only mergers where the bidder has acquired 50% of target are used. The deal value of the transaction is divided by the market value of the acquirer to account for significant outlier firms. Only the firms with an outcome between 1% and 100% are used. Datastream is used to get data on the operational performance of the merging firms on a dataset of 2734 usable observations.

A variation on the method of Stubben (2010) is used to determine firm age. He measures firm age as the years that have passed since the first known data in Compustat for a firm. For this sample the first available data of the return index that is found in DATASTREAM is used. This measure is used because different measures such as the IPO dates are found to be incomplete in the CRSP database. This would lead to a sample that is too small to use.

Only if the information for the acquirer is available before and after the merger the merger can be used. There is specified on the payment used, the industry primary SIC-code of a company and if the firm age and the total asset value is available. Furthermore all measures of operating performance have to be available. Namely EBITDA, total debt, market value of equity and preferred stock. The same data has to be available for an industry-matched benchmark portfolio in the same period. The median of a group of firms with the same industry SIC code are used to find the abnormal difference. This leads to a sample of 490 firms.

(14)

14

3.3 Benchmark

To determine the abnormal difference in operating performance there needs to be a benchmark of companies to compare the performance against. A similar peer group as Linn and Switzer (2001) will be used. Every company is paired with a portfolio of similar companies in the same industry and time period to account for industry effects. The median of this portfolio is used to compare the benchmark to the merging company. This is done by using the SIC code found in Thomson One and CRSP. The pre-merger performance is not taken into account as was done by Moeller and Schlingermann (2005). The reason for this is that they did not find any significant effects for operating performance.

3.4 Pre- and post-merger cash flows

To determine the pre- and post-merger cash flow adjusted for industry a similar method as Linn and Switzer (2001) is used. It is based on calculating the pre-merger value for the target and bidder firm together. This is subtracted with the pre-merger industry value which is weighted by the assets of the original sample.

For the pre-merger cash flow the cash flows of the bidder and target together will be divided by a combination of the market value of assets, the book value of debt and the preferred stock values. This is the Premerger performance in equation 1. The cash flows that are used here are the EBITDA values of the corresponding companies. This will be compared to pre-merger industry cash flow value that is created with a portfolio of companies in the same industry. The industry performance of the bidder and target are weighted by the pseudo market value of assets of the original sample. This is done to normalize the industry performance to values that are in relation to the original sample. This is the premerger industry performance in equation 2. The premerger performance subtracted with the premerger industry performance lead to the industry adjusted operating performance pre-merger. This is named OCFpre in equation 3.

For the industry benchmark the median of every industry matched sample is used. This matching is done by using the SIC-code. The excess return can then be found by subtracting the post-acquisition operating cash flows with the pre-acquisition operating cash flows both adjusted for industry effects. Here data from before and after the acquisition will be used to take into account the abnormal performance of the sample in respect to market and industry behavior. This method is also employed in Linn and Switzer (2001) and similar methods are used by Healy et al. (1992). Below are the equations used to calculate the operating cash flows.

(15)

15 𝑃𝑟𝑒𝑚𝑒𝑟𝑔𝑒𝑟 𝑝𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒 = _{𝑀𝑉𝐴𝑇𝑎𝑟𝑔𝑒𝑡+𝑀𝑉𝐴𝐵𝑖𝑑}𝐶𝐹𝐵𝑖𝑑𝑑𝑒𝑟+𝐶𝐹𝑇𝑎𝑟𝑔𝑒𝑡 (1) 𝑃𝑟𝑒𝑚𝑒𝑟𝑔𝑒𝑟 𝑖𝑛𝑑𝑢𝑠𝑡𝑟𝑦 𝑝𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒 = 𝑀𝑉𝐴𝐵𝑖𝑑 𝑀𝑉𝐴𝑇𝑎𝑟𝑔𝑒𝑡 + 𝑀𝑉𝐴𝐵𝑖𝑑∗ 𝐶𝑎𝑠ℎ𝐹𝑙𝑜𝑤𝑠𝑃𝑒𝑒𝑟𝐵𝑖𝑑 𝑀𝑉𝐴𝑃𝑒𝑒𝑟𝐵𝑖𝑑 +_{𝑀𝑉𝐴𝑇𝑎𝑟𝑔𝑒𝑡+𝑀𝑉𝐴𝐵𝑖𝑑}𝑀𝑉𝐴𝑇𝑎𝑟𝑔𝑒𝑡 ∗𝐶𝑎𝑠ℎ𝑓𝑙𝑜𝑤𝑠𝑃𝑒𝑒𝑟𝑇𝑎𝑟𝑔𝑒𝑡_{𝑀𝑉𝐴𝑃𝑒𝑒𝑟𝑇𝑎𝑟𝑔𝑒𝑡} (2) 𝑂𝐶𝐹𝑝𝑟𝑒 = 𝑃𝑟𝑒𝑚𝑒𝑟𝑔𝑒𝑟 𝑝𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒 − 𝑃𝑟𝑒𝑚𝑒𝑟𝑔𝑒𝑟 𝑖𝑛𝑑𝑢𝑠𝑡𝑟𝑦 𝑝𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒 (3)

The same is done for the acquisition performance for both groups with the respective post-acquisition data. The only difference is that for the post-merger performance only the bidder values are used. This is done because the target firm is now part of the bidder firm and it has ceased to exist now individually. For the post-merger industry performance the pseudo market value of assets weights are calculated by using pre-merger values for similar reasons.

𝑃𝑜𝑠𝑡𝐶𝐹 =_{𝑀𝑉𝐴𝐶𝑜𝑚𝑏𝑖𝑛𝑒𝑑}𝐶𝐹𝐶𝑜𝑚𝑏𝑖𝑛𝑒𝑑 (4) 𝐶𝑓𝑝𝑜𝑠𝑡𝑖𝑛𝑑𝑢𝑠𝑡𝑟𝑦 = _{𝑀𝑉𝐴𝑇𝑎𝑟𝑔𝑒𝑡+𝑀𝑉𝐴𝐵𝑖𝑑}𝑀𝑉𝐴𝐵𝑖𝑑𝑑𝑒𝑟 ∗𝐶𝑎𝑠ℎ𝐹𝑙𝑜𝑤𝑠𝑃𝑒𝑒𝑟𝐵𝑖𝑑_{𝑀𝑉𝐴𝑃𝑒𝑒𝑟𝐵𝑖𝑑} + (5) 𝑀𝑉𝐴𝑇𝑎𝑟𝑔𝑒𝑡 𝑀𝑉𝐴𝑇𝑎𝑟𝑔𝑒𝑡+𝑀𝑉𝐴𝐵𝑖𝑑∗ 𝐶𝑎𝑠ℎ𝑓𝑙𝑜𝑤𝑠𝑃𝑒𝑒𝑟𝑇𝑎𝑟𝑔𝑒𝑡 𝑀𝑉𝐴𝑃𝑒𝑒𝑟𝑇𝑎𝑟𝑔𝑒𝑡 𝑂𝐶𝐹𝑝𝑜𝑠𝑡 = 𝑃𝑜𝑠𝑡𝐶𝐹 − 𝐶𝐹𝑝𝑜𝑠𝑡𝑖𝑛𝑑𝑢𝑠𝑡𝑟𝑦 (6)

The operating performance values before and after the merger are used to test our hypotheses. These industry adjusted values can be used to test a more specific effect of a merger on operating performance. Furthermore by adding relevant control variables the effect of financial constraints can be tested.

3.5 Tests and regressions

To test the first hypothesis a Wilcoxon signed rank test will be done. The post-merger cash flow return adjusted for industry is compared to the pre-merger cash flow return adjusted for industry. This is done to find if there is any abnormal difference in return found for a firm after merging.

(16)

16 For the second hypothesis the impact compared to other determinants will be tested by using an OLS regression. Here the dependent variable is the combined operating cash flow one year after a merger and the control variables are cash payment, financial constraints and if the merger is focused. This gives the following regression:

𝑃𝑜𝑠𝑡𝑂𝐶𝐹 = 𝛽₀+ 𝛽₁𝑃𝑟𝑒𝑂𝐶𝐹_𝑖 + 𝛽₂𝐶𝑎𝑠ℎ 𝑃𝑎𝑦𝑚𝑒𝑛𝑡_𝑖+𝛽₃𝐹𝑜𝑐𝑢𝑠_𝑖+ 𝛽₄𝐹𝑖𝑛𝐶𝑜𝑛_𝑖+ 𝜀_𝑖

PreOCF is a variable that consists of the operating cash flow one year before the merger. It is the pre-merger operating cash flow. This value consists of the target and bidder pre-merger operating performance adjusted for industry performance. It is matched with firms from the same industry to find the abnormal return. PostOCF are the operating cash flow for our sample one year after the merger. This variable be strongly correlated with PreOCF because it is the performance of the same companies before the merger. It is expected that companies behave similarly over time.

Cash Payment is a dummy that shows if the merger is paid with cash. If this is the case it takes on the value of 1. The literature shows that it should be expected that firms that pay with cash perform better than the rest of the sample. These results can be found in Linn and Switzer (2001) and Megginson (2004).

Focus is a dummy variable to show that the firm is merging with a company in the same industry. It takes the value of 1 if the bidder and target have the same SIC-code. Firms that merge within the same industry are shown to perform better (Heron and Lie, 2002; Linn and Switzer, 2001). Therefore this is also expected for this control variable.

FinCon is a dummy variable that measures if the firm is financially constrained. It takes the value of 1 if a combination of log firm size, log firm size squared and age are in the lowest two deciles in the industry. This measure is based on the SA index of Hadlock and Pierce (2010) and methods of Almeida et al. (2004). The expectation in this paper is that these firms will outperform the sample, because of their unfulfilled growth potential. This will be tested with the second hypothesis.

(17)

17

4. Data and descriptive statistics

In table 2 it can be observed that the amount of mergers in this sample is around a similar level of mergers compared to earlier studies over a similar time period (Linn and Switzer, 2001; Megginson, 2004). This is expected because of the growing amount of mergers in more recent times. The operating cash flow consists of several variables, namely sales subtracted with costs of goods sold and selling and administrative expenses plus depreciation and goodwill expenses. If certain parts of the operating cash flow were missing the merger is dropped from the sample. Data is also needed a year before the merger and one year after the merger. This lead to a high amount of mergers that were dropped from this sample.

Table 2

Mergers divided by date

Date Merger

Frequency

Mergers Cash Payment Focus

Financially Constrained Target 1994 6 1 0 0 1995 8 3 0 0 1996 9 2 2 1 1997 23 5 0 0 1998 39 11 5 3 1999 53 16 7 4 2000 39 10 12 9 2001 35 7 6 5 2002 23 8 7 7 2003 30 9 10 12 2004 30 9 7 8 2005 27 8 10 9 2006 24 12 8 7 2007 32 21 9 6 2008 22 9 7 9 2009 22 9 4 6 2010 31 13 7 11 2011 15 7 3 6 2012 22 13 9 8 Total 490 173 113 111

In table 2 the frequency of mergers are divided among the years that they were effective in. This table also shows how many of the mergers are paid with only cash, were part of a merger within an industry or which mergers had a financially constrained target. The mergers are

(18)

18 divided relatively equal over the time period. There is an increase in mergers in the period before the internet bubble around 2000 and a subsequent decrease thereafter. After the financial crisis in 2007 there is also a decrease in mergers. This can be explained by the market being less optimistic about future growth. These results are also in correspondence with Harford (2005). In periods of economic growth there is a high frequency of mergers and in crises there are significantly less mergers.

For cash paid mergers there is also an increase before the 2000 internet bubble (Aharon et al., 2010). At that time confidence and growth expectations were high and therefore firms were more likely to pay the merger with cash. It can be expected that this will have an effect on the cash payment dummy in the regression. These cash paid mergers are likely to perform badly as results of the subsequent crash. This could therefore lead to less favorable results for the cash dummy.

The number of mergers that are being done within the same industry remain relatively constant. This can be explained by the fact that during that period mergers with firms in the same industry were a safer options for acquiring firms. This can be seen in table 3 were the amount of mergers remain constant even in the 2007-2009 financial crisis. Heron and Lie (2002), Linn and Switzer (2001) all found that focused mergers perform better than diversifying mergers. Therefore the expectation is that this is also the case for this sample.

The frequency of financially constrained targets is divided relatively similar as the cash paid dummies. There is an increase in 2001 and more financially constrained targets are bough then. A possible explanation could be that in these periods of financial downturn there are more financially constrained firms and they would make cheaper acquisition targets. Alternatively it could be argued that after periods of financial downturn firms will go for more secure options and therefore in these periods there are less financially constrained targets bought. This can be seen in the 2007-2009 period of the financial crisis. This means that in relation to the hypothesis of the thesis no strong prediction can be made.

(19)

19 In table 3 the summary statistics of the independent and dependent variables are described. Post- and Pre-merger operating cash flow return are the respective operating cash flow values divided by the market value of assets from the firms and adjusted for industry behavior as seen in Healy (1992) among others. The difference in means between pre- and post-merger results is expected, because the post-merger results consists of the performance of the bidder and target company together. Both the post- and pre-merger cash flow have a negative mean, but it is higher for the post-merger cash flow. This means that the industry outperforms the original sample of merging firms. This would lead to hypothesis 1 to be rejected or that there is no significant difference between the merging firms and industry portfolio.

The post- and pre-merger operating cash flow both have very high minimum values and a high standard deviation. This means that there high variance in the sample. Therefore the post- and pre-merger cash flow are winsorized at 5%. This is done to normalize the top and bottom 5% of the sample and eliminate any large outliers. As can be seen in table 3 this leads to very similar means for both variables. Furthermore the standard deviations and the outliers are at more reasonable levels.

For Cash Payment, Focus and Financial Constraints the minimum and maximum variables are not given, because these are dummy variables. The mean shows how big the representation of a specific dummy is in the sample. For Cash Payment 35.3% of the observations are paid with cash. This is in correspondence with the literature were a similar amount was paid with cash. Linn and Switzer (2001) found that 51.1% of the mergers were

Table 3

Descriptive statistics dependent and independent variables Variable N Mean Standard

deviation Median Minimum Maximum Post-merger operating cash flow 490 -0.224 1912 -0.013 -39.196 0.327 Pre-merger operating cash flow 490 -0.118 0.822 -0.004 -15.569 0.711 Cash Payment 490 0.353 0.478

Focus 490 0.486 0.5 Financially constrained firm 490 0.227 0.419

Winsorized variables

Post-merger operating cash flow 490 -0.088 0.238 -0.013 -0.817 0.179

(20)

20 paid with cash and 34.5 % for Megginson (2004). 48.6% of the mergers were between firms in the same industry. According to the literature these mergers perform better, because the firms are more accustomed to each other and knowledgeable about their industry. Therefore it can be expected that these type of mergers account for a large amount of the sample. Financially constrained firms consist of the first of the five quantiles divided on firm age and log of assets. Therefore a mean of 20% would be expected. The mean is 31.1%, because there are more firms that qualify for the financial constraints measure.

In table 4 the correlations for the independent and dependent variables can be found. Only between the pre- and post-operating cash flow there is a high level of correlation. This is expected, because the performance of a company in the past influences the future performance. For the other variables the correlation is low. Therefore it can be expected that there is no multicollinearity in the sample. The positive correlations of the cash payment and focused merger dummy with the post-merger operating cash flow means that it can be expected that these variables have a positive effect on the post-merger operating performance. These correlations are low so it could also mean that there is no specific effect found. The positive correlation of financially constrained mergers with post- and pre-merger operating cash flow predict a positive effect of financially constraints on operating performance. This is in relation to our second hypothesis that financially constrained target firms have a significant effect on operating performance. This will be tested using different regressions in the next section.

Table 4

Correlation of dependent and independent variables Post-merger operating cash flow return Pre-merger operating cash flow return Cash Payment Focused merger Financially Constrained Target Post-merger operating cash flow

return 1

Pre-merger operating cash flow

return 0.557 1

Cash Payment 0.03 0.005 1

Focus 0.047 0.021 -0.052 1

(21)

21

5. Results

The first column in table 5 can be used to answer hypothesis 1. This shows the difference between post- and pre-merger operating performance. There is a negative relationship between pre- and post-merger operating cash flow return. The Wilcoxon signed rank test show that both differences are significantly different from zero. This would mean the first hypothesis will be rejected. Combined companies are found to perform worse after mergers than other firms in their industry. Similar results were found by Clark and Ofek (1994). Linn and Switzer (2001), Healy et al. (1992) and Megginson (2004) all found a positive effect for mergers. This does not mean that hypothesis 2 is also rejected, because it can still be the case that financially constrained firms are a positive determinant in predicting future performance of combined firms.

Table 5. Post-merger performance

This table looks at the median of pre- and post-merger performance. Median merger performance is described as median value of the operating performance of a merger adjusted for industry. The median difference is the z-value is found by using the Wilcoxon signed rank test on the pre- and post-merger performance. For the first column the values are winsorized at the 5% level. An *, **, and *** indicate significance at 10%, 5%, and 1%, respectively.

(1) (2)

Median pre-merger performance -0.004 -0.004

Median post-merger performance -0.013 -0.013

Median difference -0.007** -0.008***

Table 6 is used to answer hypothesis 2. As predicted from the descriptive statistics there is a large and significant effect of pre-merger operating performance on the post-merger operating performance. This should be the case, because the performance of both firms are taken together and adjusted by industry. This should be in line with the performance after the merger. The performance of a company in the past influences the future performance. For these regressions industry- and year-fixed effects are used to account for changes that vary over time or industry, but are consistent between firms.

(22)

22 Table 6. Post-Merger Operating Cash Flow Performance

This table looks at the effect of the pre-merger operating performance on the post-merger operating performance with additional control variables. Pre-tax operating cash flow is divided by the market value of assets. Market value of assets is the market value of equity plus the book value of debt. This is done before and after the merger to find the respective pre- and post-merger OCF. The OCF values are winsorized at 5% for pre- and post-post-merger operating cash flow. Cash Payment is a dummy that takes on the value of 1 if the merger is paid with cash. Focus is a dummy that takes on the value of 1 if the merger is between two firms in the same industry and have the same SIC-code. Financial Constraints Quantiles is a dummy variable that measures if the target firm is financially constrained. It takes the value of 1 if a combination of the log of firm size, the log of firm size squared and age are in the lowest two deciles in the sample. In the second column robust standard errors are used. For both regressions industry- and year-fixed effects are used. The time period for the samples that are used is from 1994-2012. *, **, and *** indicate significance at 10%, 5%, and 1%, respectively.

Dependent variable: Post-merger OCF

(1) (2) Pre-merger OCF 0.600*** 0.600*** (8.64) (5.49) Cash Payment 0.017 0.017 (0.70) (0.63) Focus 0.038 0.038 (1.62) (1.57)

Financial Constraints Quantiles 0.011 0.011

(0.43) (0.37)

Constant -0.225 -0.225

(-1.04) (-3.79)

Industry-fixed effects Yes Yes

Year-fixed effects Yes Yes

N 490 490

Adjusted R-squared 0.3625 0.5867

The negative intercept means that the abnormal post-merger performance is worse after a merger than before the merger. This is consistent with the finding in table 5. Mergers do lead to a lower post-merger operating performance than before the merger. The first hypothesis is also rejected by this result.

All control variables are shown to be not significant in column one and two. This means that the control variables are not efficient in determining the post-merger operating cash flow return. This could be caused by low amount of observations. Due to lack of data only one year before the merger and one year after the merger is used. This also leads to the high standard

(23)

23 errors in this sample. Although the control variables are shown to be non-significant, there are similar results found in other papers (Linn and Switzer, 2001; Megginson, 2004).

There is a small positive effect of cash paid mergers. The expectation was that cash payment would have the biggest effect on the post-merger operating cash flow return. This is not the case because many cash paid mergers happened just before the 2000 internet bubble (Aharon et al., 2010). These targets were therefore likely to perform worse after the subsequent crash and influence the result for this dummy variable. It is still the case that cash paid mergers lead to better post-merger operating performance. In these situations the acquirer is confident in the success of the merger and therefore willing to take on higher costs. This finding is therefore in line with other research (Linn and Switzer, 2001; Megginson, 2004). These papers did not find a significant relationship between cash and post-merger operating performance. A possible explanation is that a large amount of firms are paid with cash and this leads to it being less likely that a significant effect is found.

For focused merger there is also a positive not significant effect found on the post-merger operating performance. Even though the effect is not significant, it is almost significant. The expectation from the literature is that firms that undertake mergers within the same industry have a higher post-merger operating performance than diversifying mergers. This is because of economies of scale that there can be profited from. This finding is in line with the findings of Heron and Lie (2002). Linn and Switzer (2001) found a negative result for their complete sample and positive results for negotiated mergers. The difference in results could be explained by the different measure of operating performance that was used by Linn and Switzer for their sample. These papers also did not found a significant result for focused mergers. The explanation for this is that almost half the sample consists of focused mergers and therefore it is unlikely that a significant effect is found.

The expectation for the second hypothesis is that the financial constraint dummy will be positive. There are two measures of financial constraints in table 6. The financial constraint variable focused on the lowest two deciles ranked on log of assets, log of assets squared and size. This is in line with the measure of financial constraints by Hadlock and Pierce (2010). In both regressions the financial constraint dummy variable is positive this means that it can lead to a higher post-merger operating performance. This is in line with the expectation in this paper. Firms that are younger and smaller in size have more growth potential that can be utilized by acquiring firms. This potential growth leads to higher post-merger operating cash flow return for the new firm. There are no other papers that have researched this relationship therefore it is

(24)

24 not possible to compare these results. Clark and Ofek (1994) looked at the performance of financially distressed firms, but used highly varying methods to research this and is therefore less relevant.

Even though for the second hypothesis there is no significant effect found. There is a positive effect found that financially constrained targets have on the post-merger operating cash flow return of the new firm. Reasons for not finding a significant value can be attributed to the way that financially constrained firms are defined in the literature. The definition for financially constrained target firms is the lowest two deciles of the target firms ranked on the log of firm size and log of firm size squared and age (Hadlock and Pierce, 2010). This means that these firms are less likely to have public information available in databases regarding operating performance. Furthermore the sample is adjusted using the same method as Moeller and Schlingermann (2005). This eliminates the observations that are too small to have a significant effect on the acquiring firm. Therefore this could mean there is a selection bias in the sample that is used. This causes the data in our sample to be skewed and result in the variable not being significant.

(25)

25

6. Robustness

Two additional tests are used to verify that the earlier results hold up and are consistent with different methods. First the robustness of this research will be tested by looking at the difference in abnormal buy-and-hold return between the sample and an industry benchmark. Secondly a panel regression will be done to further test the effect of financial constraints. The difference in abnormal buy-and hold returns will be tested over a period of 36 months after the effective date of the merger. This research is done in accordance to Eckbo (2008). The motivation behind testing this is if there is a positive abnormal return for merging firms. The findings above in this paper show that this should not be the case, but large parts of literature find different results (Megginson, 2004; Healy, 1992; Linn and Switzer, 2001).

The average buy-and-hold return will be defined as a stock being bought at the effective merger date and holding the stock for three years. There will be an equal-weighted buy-and-hold return sample were the buy-and-buy-and-hold return for each firm is divided by the total number of firms. Additionally a value-weighted hold return will be created. Here the buy-and-hold return will be weighted by multiplying the buy-and-buy-and-hold return of each firm with their own market value divided by the sum of market values at the effective merger date.

To create the abnormal buy-and-hold return the buy-and-hold return of the merging firms will be matched with an industry adjusted benchmark of buy-and-hold returns. To keep this test in line with the research that was done to test the hypothesis the industry benchmark will be a median value of the industry sample finding matched on industry SIC-code and effective merger date. The industry sample consists of a population of United States companies (Linn and Switzer, 2001). There is a loss of observations compared to the initial sample used in this paper. This was caused by a lack of available stock return data on Datastream for these firms. The sample used here consists of 370 mergers.

(26)

26 Table 7. Post-Merger Average Buy-and-Hold Returns

This table looks at the average buy-and-hold return for merging firms and a matched industry benchmark. The sample consists of mergers between 1994 and 2012. After the merger there needs to be data available for 36 months after the merger. The benchmark firms are matched on industry SIC-code and the same time period as the effective merger date is used. The equally weighted sample is weighted by dividing each buy-and-hold return by the total amount of mergers in the sample used. The matched firms are matched by multiplying the buy-and-hold return of each firm with their own market value divided by the sum of market values at the effective merger date. The matched and merged values in this table are the sum of the equally-weighted and matched firms. The p-values are found by using a paired t-test on the difference between the buy-and-hold returns of the merged firms and the industry firms and if this is significantly different than zero.

Equal-weighted Average BHR Value-weighted Average BHR Date N Merged Matched Diff p(t) Merged Matched Diff p(t) 1994-2012 370 32.04% -10.94% 42.98% 0 24.78% 11.00% 13.78% 0 1994-2002 57 9.06% 3.92% 5.14% 0.27 121.35% 7.81% 113.54% 0 2002-2012 313 30.82% -19.93% 50.75% 0 13.17% 11.37% 1.80% 0.05

Table 7 shows that for equally weighted and value-weighted firms the merged firms always perform better than the industry. For the total sample period the difference in average buy-and-hold return is 42.98% for the equally-weighted sample and 13.78% for the value-weighted sample. Both differences are significant. The difference is significantly lower for the value-weighted average buy-and-hold return. This means that the industry benchmark has more large firms that perform better than the rest of the sample and the merged firms perform more uniformly.

This test shows that our initial finding for our first hypothesis could be wrong. The table shows that the merged firms should perform better than their respective industry peers for this sample. It has to be understood that this test looks at stock performance in contrast to operating performance. This means that market sentiment weighs more highly in these results and the market are prone to make mistakes (Zollo and Meier, 2008). The difference in findings can also be explained by the way that operating performance is defined in this paper. Many merging company use leveraged buyouts to finance their acquisitions. By taking on a significant amount of debt the operating cash flow decreases. Therefore in the initial tests for hypothesis two the merger is seen as value decreasing. By focusing on the stock performance not all of the debt that is taken on is directly reflected in the stock price. This gives a skewed view of the actual performance of the company and explains the difference in findings. Although a

(27)

performance-27 increasing effect of mergers is more consistent with other literature (Megginson, 2004; Healy, 1992; Linn and Switzer, 2001). Therefore it can be argued that for this sample the long-term operating performance does not adequately measure the performance of a firm after a merger. To further test if financial constraints can be used as an actual determinant of post-merger performance a panel regression will be used. The advantages of doing a panel regression are more observations and more control over omitted variables. The amount of observations used is doubled, because the operating performance is split in before and after the merger observations. By using panel data it is possible to control for changes in entities that change over time, but are similar among different states. The panel consists of observations before and after the merger. The regression takes on the following form:

𝑂𝐶𝐹𝑖𝑡 = 𝛽1𝑃𝑜𝑠𝑡𝑖𝑡 + 𝐹𝑖𝑟𝑚𝐹𝑖𝑥𝑒𝑑𝐸𝑓𝑓𝑒𝑐𝑡𝑠 + 𝑇𝑖𝑚𝑒𝐹𝑖𝑥𝑒𝑑𝐸𝑓𝑓𝑒𝑐𝑡𝑠

Here OCF is the respective industry adjusted operating performance for firms before and after the merger. The Post variable is a dummy that denotes if a firm has already merged. It is used to capture the average change in operating performance after the merger. FirmFixedEffects is a variable for the constant that controls for differences that are constant over time but differ across firms. TimeFixedEffects are different dummies for the respective year for an observation. This is used to control for differences that vary over time, but are constant between firms (Stock and Watson, 2011). The sample will be cut in two parts. The first part consists of the non-financially constrained target firms and the second part consists of the financially constrained target firms. For both samples a regression will be done. It will be tested if the coefficient of the Post dummy is higher for firms that have financially constrained targets. To further test if mergers with financially constrained target firms perform better.

Table 8. Panel Regression Post-merger Performance

This table looks at the difference in post-merger operating cash flow performance for firms before and after the merger. Pre-tax operating cash flow is divided by the market value of assets. Market value of assets is the market value of equity plus the book value of debt. Post-merger OCF is a combination of the before and after merger results that are found. Post is a dummy variable that is equal to 1 if the observation is a post-merger result. The first column includes the mergers that did not have a financially constrained target firm. The second column includes the mergers were a financially constrained target firm was acquired. Furthermore there are additional year dummies added to account for changes that vary over years, but are constant over time. For both regressions firm-fixed effects are used. The time period for the samples that are used is from 1994-2012. Robust standard errors are used. *, **, and *** indicate significance at 10%, 5%, and 1%, respectively.

(28)

28 (1) (2) Post -0.059*** -0.00 (-4.24) (0.00) Constant -0.068 -0.126 (-1.04) (-1.51)

Clustered standard errors Yes Yes

Year-fixed effects Yes Yes

N 758 222

R-Squared 0.008 0.077

As can be seen in table 8 the Post dummy is less negative for the second regression. This means that mergers were a financially constrained target firm was involved performs better than if this was not the case. Even though the mergers with financially constrained targets performed equally to the performance before the merger, no significantly better performance was found. This further verifies the second hypothesis. Important to note is that the Post dummy for the second regression is not significant. Furthermore the R-squared for both regressions is very low. The actual difference between post- and pre-merger by the Post dummy is very low. Therefore this means that the evidence for the effect of financially constrained target firms as a determinant of future performance is still not conclusive.

(29)

29

7. Conclusion

In this thesis we examined if financial constraints are a useful determinant for post-merger performance. Furthermore we tested if mergers lead to an increase in operating performance. This was done for a sample of 490 mergers in the period 1994 to 2012. This is tested by creating an operating cash flow variable for the pre- and post-merger performance. The abnormal performance is then found by subtracting the median value of the industry specific operating performance. The results that are found are that mergers lead to negative post-merger operating performance. This means that the first hypothesis in this thesis is rejected. Similar results were found for Clark and Ofek (1994). Linn and Switzer (2001), Healy et al. (1992) and Megginson (2004) all found a positive effect for mergers. Regarding the stock performance there was found a positive effect of mergers on future performance in respect to the industry. Therefore the results regarding the effect of a merger on future performance are still inconclusive.

For the second hypothesis it is found that financial constraints in the target firm are a positive determinant for post-merger performance. Our robustness also verifies this and finds that a higher post-merger performance is found for firms that merge with financially constrained targets. Firms that are younger and smaller in size have more growth potential that can be utilized by acquiring firms. This potential growth leads to higher post-merger operating cash flow return for the new firm. There are no other papers that have researched this relationship therefore it is not possible to compare these result, but it is consistent with the predictions from the literature (Erel et al., 2015).

Although a positively significant results was found for our second hypothesis there are some limitations that influence this result. The sample used is relatively equally sized in contrast to other papers (Linn and Switzer, 2001; Healy et al., 1992; Megginson, 2004), but very small in contrast to the 69958 that are reported in this time period. This is partly to the lack of available Datastream data for this sample and the restrictive requirements used following Moeller and Schlingermann (2005). For this reason it is unknown how representative this sample is for the population of mergers that were effective in this time period. Furthermore the pre-merger event performance is not taken into account into determining the benchmark that is used to find abnormal operating performance. This could lead to more accurate results. Additionally it can be argued that the current measure of financial constraints is not a very strong measure. Only looking at firm size and age to determine the financially constrained firms is still a very rudimentary. Therefore the method of measuring financially constrained firms can be improved

(30)

30 on. Hadlock and Pierce (2010) did find this to be the most accurate measure and he discredited the use of more sophisticated methods such as the Kaplan and Zingales index and the Whited and Wu index.

The findings in this paper give an insight in the possible determinants of merger performance. Financially constraints targets are not seen as a determinant of future post-merger performance in current literature. This could lead to a higher importance on the level of liquidity in a target companies and if this could signal current untapped growth potential. This would be most important for private equity firms or firms that actively undertake mergers. Of course the current findings are still not conclusive and additional research is needed to learn more of the implications that financially constraints in the target firm have on future post-merger performance. The main focus for future research should therefore be on the method of measuring financial constraint in the target firm.

(31)

31

8. References

Aharon, D. Y., Gavious, I., & Yosef, R. (2010). Stock market bubble effects on mergers and acquisitions. The Quarterly Review of Economics and Finance, 50(4), 456-470.

Almeida, H., Campello, M., & Weisbach, M. S. (2004). The cash flow sensitivity of cash. The

Journal of Finance, 59(4), 1777-1804.

Barber, B. M., & Lyon, J. D. (1997). Detecting long-run abnormal stock returns: The empirical power and specification of test statistics. Journal of financial economics, 43(3), 341-372. Clark, K., & Ofek, E. (1994). Mergers as a means of restructuring distressed firms: An empirical investigation. Journal of Financial and Quantitative Analysis, 29(04), 541-565.

Eckbo, B. E. (Ed.). (2008). Handbook of Empirical Corporate Finance SET (Vol. 1). Elsevier.

Erel, I., Jang, Y., & Weisbach, M. S. (2015). Do acquisitions relieve target firms’ financial constraints? The Journal of Finance, 70(1), 289-328.

Fishman, M., 1989. Preemptive bidding and the role of the medium of exchange in acquisitions. Journal of Finance 44, 41–58

Ghosh, A. (2001). Does operating performance really improve following corporate a

Hadlock, C. J., & Pierce, J. R. (2010). New evidence on measuring financial constraints: Moving beyond the KZ index. Review of Financial studies, 23(5), 1909-1940.

Harford, J. (2005). What drives merger waves?. Journal of financial economics, 77(3), 529-560.

Heron, R., & Lie, E. (2002). Operating performance and the method of payment in takeovers. Journal of Financial and Quantitative Analysis, 37(01), 137-155.

Kanter, R. M. (2009). Mergers that stick. Harvard Business Review, 87(10), 121-125.

Kaplan, S. N., & Zingales, L. (1997). Do investment-cash flow sensitivities provide useful measures of financing constraints? The Quarterly Journal of Economics, 169-215.

Keynes, J. M. (1937). The general theory of employment. The quarterly journal of economics, 209-223.

(32)

32

Linn, S. C., & Switzer, J. A. (2001). Are cash acquisitions associated with better postcombination operating performance than stock acquisitions? Journal of Banking &

Finance, 25(6), 1113-1138.

Loughran, T., & Vijh, A. M. (1997). Do long‐term shareholders benefit from corporate acquisitions? The Journal of Finance, 52(5), 1765-1790.

Megginson, W. L., Morgan, A., & Nail, L. (2004). The determinants of positive long-term performance in strategic mergers: Corporate focus and cash. Journal of Banking &

Finance, 28(3), 523-552.

Stock, J. H., & Watson, M. W. (2011). Introduction to Economics (3rd_{edition). Pearson}

Education.

Stubben, S. R. (2010). Discretionary revenues as a measure of earnings management. The

Accounting Review, 85(2), 695-717.

Thurm, S. (2013, September 3). History Isn't on Side of Microsoft-Nokia Tie Up. Retrieved June 15, 2015 from

http://www.wsj.com/articles/SB10001424127887324886704579053390137485358

Zollo, M., & Meier, D. (2008). What is M&A performance? The Academy of Management

Are financial constraints a determinant for long-run post-merger performance?