Private equity bids in Australia: an exploratory study
We would like to thank Gerry Gallery and workshop participants at the School of Accountancy, Queensland University of Technology for helpful comments. We also thank Stephen Lindsay and Matt Davies for initial help and advice regarding the sample selection.
Abstract
In this study, we provide an insight into how private equity players choose their targets and the bid arrangements they prefer. We test our expectations of the unique features of private equity targets using a sample of 23 listed private equity target firms during 2001–2007. We find, relative to a benchmark sample of 81 corporate targets matched by year and industry, the private equity target firms to be larger, more profitable, use their assets more efficiently, more highly levered and have greater cash flow. Multivariate testing indicates that private equity targets have relatively greater financial slack, greater financial stability, greater free cash flow and lower measurable growth prospects. All conclusions are found to be robust to a control sample of 502 takeover bids during 2001–2007.
1.Introduction
The private equity investment sector in the Australian economy has achieved unusual growth over recent years. For example, in 2006, the value of private equity transactions announced and endorsed by target firm boards surged to $26 billion, up from an average of $2 billion over the previous 5 years.1 This new wave of highly publicised private equity takeovers has become a topic of significant interest to policy makers, professionals and academics. Nonetheless, there is no current theory and no extant literature encompassing the preferences of private equity bidders in Australia.
Private equity in the context of this study refers to the acquisition of a public company with the intention of taking it private through a delisting. The private equity funds are typically organised through limited partnerships managed by the general partner (fund managers) who receive a management fee based on the size of the fund and also share in the capital gains delivered. The primary investors, the limited partners, do not participate in the management of the partnership but benefit from limited liability.
Given that these deals can feature a high level of insider knowledge and/or a significant degree of leverage employed in their financing mix, the role of private equity in the market has attracted recent attention. Regulators in Australia and the UK, e.g. have held recent enquiries into private equity investment and its effects on capital markets.2
Considering the significance of private equity, arguably a formal study into the features of private equity bids in Australia is warranted, in particular, an investigation into the attributes of target firms and bid provisions by private equity bidders. Prior literature identifies some generally accepted theories about the motivations of participants in traditional merger and acquisition (M&A) deals, including disciplinary (hostile) or synergistic (friendly) motivations (Powell, 1997). However, private equity bidders occupy an unusual place in the market with their role appearing to be as disciplinary, friendly acquirers. To avoid being dubbed ‘Barbarians at the gate’,3 private equity firms consistently endeavour to negotiate friendly deals with target management (Cheffins and Armour, 2007). Thus, arguably it is difficult to classify private equity bids in the traditional manner, suggesting that private equity offers a new form of corporate control. This view is held by Michael Jensen who argues that private equity is a new model of general management, applicable to many, if not most firms and organisations. He argues that private equity enables the capture of value destroyed by agency problems, as evidenced from the growth and success of the private equity sector, particularly in the mature segment of the market with high free cash flows (Jensen, 2007).
Using a framework developed from academic and industry literature, we propose that relative to other corporate target firms, private equity target firms should have greater financial slack (both debt capacity and free cash flow), greater business stability and lower measurable growth prospects. We also propose that private equity bids will be predominantly of a friendly nature, more so than other corporate bids. The private equity sample we consider consists of 23 listed target firms subject to a takeover bid involving a private equity bidder during the period 2001–2007. For purposes of comparison, we benchmark this sample against a sample of 81 corporate bids matched by year and industry. For sensitivity, we also consider an alternative benchmark consisting of 502 corporate takeover bids during the study period.
The descriptive statistics for our sample suggest that private equity target firms are larger, more profitable, use their assets more efficiently, more highly levered, and have greater cash flow than other corporate targets. They also suggest that the bid premium offered in private equity deals is lower and a slightly smaller proportion of bids are successful. Private equity bids also involve an independent expert more often, due predominantly to the structure of the transaction (scheme of arrangement) as opposed to mandated due to bidder conflicts. Given the debt funding of private equity investment, the preferred mode of payment for private equity bidders is cash, in contrast with other offers that can and do involve scrip. Finally, the industry composition of the private equity targets suggests that private equity bidders are reticent to bid in either the Financial or Metals and Mining sectors.
Our univariate and multivariate test results provide relatively consistent support for the assertions that private equity targets have greater financial slack, greater financial stability, greater free cash flow and lower measurable growth prospects than other corporate targets. They also provide some mixed support regarding the friendliness of private equity bids.
This study is of necessity exploratory in nature given the lack of prior research on private equity in Australia. Its innovation lies in the untested relationship between private equity firms and their takeover targets. In this regard, it provides insight into how private equity players set about choosing their targets and the bid arrangements that private equity bidders prefer. It examines target characteristics and bid attributes, and compares these data across a sample comprising private equity bids and other corporate targets. This study also potentially benefits regulators and policy makers who have been closely monitoring the impact of the increased private equity presence in the Australian market.
The remainder of this document is structured as follows. Section 2 provides a background to the private equity market. Section 3 reviews the extant literature concerning M&A and establishes a foundation for the propositions. The theory and proposition development is delineated in Section 4. Section 5 describes the sample data. The results of the study are presented in Section 6 and Section 7 presents a summary and conclusions.
2. Industry background
According to the Australian Private Equity & Venture Capital Association Limited (AVCAL), the following five factors, said to be the hallmarks of private equity investment, allow private equity to add value to businesses:4 (i) an alignment of interest between the owners and the management; (ii) long-term success is not sacrificed in the pursuit of short-term growth; (iii) detailed due diligence performed by private equity fund managers; (iv) the private equity fund ensures that there is appropriate planning for the growth of the investment; and (5) active stewardship is a significant factor in adding value to investments.
Austin and Tuch (2008) recognise that there is a number of distinctive characteristics of private equity, include relating to takeover transactions (vis-à-vis venture capital/start ups):
- •
the targeting of poorly managed businesses
- •
high levels of debt funding from third party lenders
- •
high gearing post acquisition
- •
medium- to long-term investment strategy (say 3–7 years) for resale
- •
control of the acquired businesses and the injection of management disciplines to achieve aggressive business plans.
These observations by industry commentators are highly suggestive of the disciplinary motivation for takeovers as identified from the classic literature.
3. Literature review
3.1. Disciplinary and synergistic motives
A common premise of the existing literature is that traditional acquisitions are either disciplinary or synergistic (Powell, 1997; Weir and Wright, 2006), with the disciplinary motive the more common explanation. Disciplinary acquisitions are regarded as hostile and are traditionally viewed as a key mechanism for disciplining under-performing managers (Manne, 1965). Such inefficiency includes managerial incentive problems stemming from information asymmetry (Scharfstein, 1988), entrenched strategies no longer beneficial to firm value (Jensen, 1988) and suboptimal governance structures (Shivdasani, 1993).
According to the Reserve Bank of Australia, private equity transactions in the Australian market support the inefficient management hypothesis, and play an important role in ensuring an efficient and dynamic business sector.5 However, in a review of the literature on Australian M&A since the 1980s, da Silva Rosa and Walter (2004) conclude that overall, the evidence to date is supportive of M&A activity being initiated by differentially efficient firms. In their opinion, the disciplinary motive for takeovers is arguably oversold.
In contrast, synergistic acquisitions are traditionally classified as friendly, with both parties recognising the potential benefits, such as increased market power, marketing economies, improved technical expertise and better research and development (Morck et al., 1988). Friendly acquisitions are therefore less likely to be disciplinary in nature with targets selected because they possess characteristics that the bidding firm specifically desires.
3.2. Financial incentives
Both motivations (disciplinary or synergistic) pre-suppose some compelling evidence that draws the bidder’s attention to a potential target. Here, the evidence consistently supports the notion that target firms underperform in the pre-bid period. For example, using US data, Morck et al. (1989) find that hostile targets have poor performance prior to the bid as measured by both Tobin’s Q and stock market performance. In Australia, both Brown and da Silva Rosa (1997) and Eddey and Taylor (1999) find that target firms underperform in the pre-bid period. Brown and da Silva Rosa then argue that takeovers add value in these cases by realigning the interests of management and shareholders.
Adopting the bidder’s financial perspective, a bidder can only act if they have the resources to do so. Myers and Majluf (1984) model a setting wherein a firm with too little financial slack can increase its value by acquiring more. Their model considers the scenario where a financial slack-rich firm acquires a financial slack-poor target with growth opportunities. Bruner (1988) tests the proposition finding that pre-merger, bidders have more financial slack and lower leverage than do the control firms, and that bidding firms build a debt capacity in advance of acquiring. Bruner also finds that target firms have significantly higher financial leverage at the time of merger than non-targets. Moreover, his results suggest that bidders are less levered than target firms ex ante, also consistent with the information-asymmetry hypothesis of the use of excess cash and unused debt capacity through merger.
Smith and Kim (1994) examine the extent to which takeovers mitigate the underinvestment problem (Myers and Majluf, 1984) and the free cash flow overinvestment problem (Jensen, 1986). Consistent with Myers and Majluf (1984), they find more positive total returns for bids combining slack-poor firms and firms with excess cash than for bids that perpetuate financial slack and free cash flow problems. Overall, their results suggest that merger gains arise from the resolution of over- and underinvestment problems. Smith and Kim also contend that the Myers-Majluf theory of partnering slack-rich and slack-poor firms applies equally to slack-poor bidders and slack-poor targets, positing that the asymmetry problem can be avoided if an undervalued slack-poor firm can acquire target shares for cash.
In sum, the literature indicates that the evidence is weighed towards the view that a firm is more likely to become a target if it is performing poorly or if it is financial slack-poor, but has good growth opportunities. The evidence on governance structures is inconsistent.
3.3. Public-to-private studies
In his widely cited work concerning free cash flow theory and the agency problem, Jensen (1986) posits that takeovers, and in particular leveraged buyouts, mitigate the agency problem in firms with substantial free cash flow. He suggests that debt plays an important role in motivating organisational efficiency. Debt creation, without retention of the proceeds, enables managers to effectively bond their promise to pay out future cash flows. The debt reduces agency costs by reducing the cash flow available for discretionary spending by managers. The debt created in the takeover of a firm experiencing high agency costs of free cash flow is not usually permanent. The process results in a review of the organisation’s strategy and structure, leading to a more efficient and competitive organisation.
Following Jensen’s (1986) free cash flow hypothesis, Lehn and Poulsen (1989) investigate the source of stockholder gains in going private transactions. While their primary focus in on stockholder gains, by comparing firms that went private with a control group of non-targets, they also find that the likelihood of going private is directly related to the ratio of undistributed cash flow to equity value and inversely related to the growth rate in sales.
Weir and Wright (2006) examine whether public-to-private transactions are different from traditional acquisitions of listed firms in the UK. They posit that firms going private are expected to be in mature, low-growth sectors with substantial free cash flow, with the transaction enabling the return of some of the free cash flow to shareholders as a result of improved governance and incentive realignment post-buyout. They find support for the disciplinary/substitutability perspective in the sense that public-to-private transactions have lower growth prospects. Conversely, they find support for the non-disciplinary/complementary perspective as public-to-private transactions do not have greater free cash flows than traditional acquisitions of listed firms by existing corporate groups.
4. Empirical propositions and proxy measurement
4.1. Target firm characteristics
The central concern behind the Senate inquiry into private equity takeovers is their debt profile. Capacity for debt financing is the maximum amount of interest payments that the firm can support without incurring financial distress. Extending debt capacity is the notion of financial slack, defined as having cash (or near-cash) and/or spare debt capacity available to exploit potential growth opportunities. The theoretical work suggests the plausibility of combining slack-poor and slack-rich firms, particularly if the slack-poor bidder acquires target shares for cash. Financing can be secured by the assets of the target and its free cash flows can be used to service the acquisition debt (Myers and Majluf, 1984; Bruner, 1988; Smith and Kim, 1994; Bugeja and Walter, 1995). Given the significant use of debt by private equity and the literature cited before, we propose as our first proposition:
Proposition 1: Private equity targets are financial slack-rich relative to other targets.
Continuing, Jensen (1986) argues that many of the benefits of going private are due to the control function of debt. These transactions create a new organisational form with the advantage of controlling the agency costs associated with the excess free cash flow. While the evidence regarding the role of free cash flow in the decision to go private is mixed (Lehn and Poulsen, 1989; Opler and Titman, 1993; Weir and Wright, 2006), stable cash flows are found to be a sought-after attribute in the stated investment criteria of many private equity firms.6 We therefore propose as our second proposition (in two parts):
Proposition 2a: Private equity targets have stable business histories relative to other targets.
Proposition 2b: Private equity targets have substantial free cash flow relative to other targets.
Finally, the growth prospects of takeover targets have also been widely discussed in the literature but with again inconsistent findings (Jensen, 1986; Lehn and Poulsen, 1989; Opler and Titman, 1993). Notwithstanding, positive growth prospects are frequently identified as a desirable target attribute by private equity firms.7 While value and growth firms are typically seen as operating at different life stages, this apparent contradiction may diminish if the role of private equity is to identify growth prospects overlooked by the capital markets. Since stability is the result of an established business in a mature industry, leading to substantial free cash flow as recognised growth prospects have been exploited, it is plausible for private equity bidders to seek stable firms with unrecognised growth prospects.8 Accordingly, we propose as our third proposition:
Proposition 3: Private equity targets will have lower measureable prospects for growth relative to other targets.
Taken together, the arguments presented above suggest that private equity target firms are likely to have greater financial slack (debt capacity and free cash flow) and more stable business histories, but lower measurable growth prospects. To test these propositions, we employ the following proxies. First, for debt capacity, we use the debt-to-equity and interest coverage ratios. Based on proposition 1, we expect the debt-to-equity ratio to be lower and the interest coverage to be higher for private equity target firms than for other targets.
Second, we proxy for the stability of the target’s business history using as our primary proxy, the standard deviation of per share cash flow from operations measured over the 4-year period immediately preceding the bid. Based on proposition 2a, we expect private equity targets to exhibit a more stable cash flow pattern. We also consider the target firm’s dividend payout and price-earnings ratios. Since firms that pay higher dividends tend to be in mature industries with limited growth prospects, we also expect private equity target firms to have higher dividend payout ratios. Finally, since low price-earnings ratios are typically associated with value stocks, we expect the price-earnings ratio of private equity target firms to be lower.
Third, we measure free cash flow both in total and scaled by the target’s total assets. In each instance, the measure is expected to be larger for the private equity target firms. Finally, we proxy for the target firm’s growth prospects using both its market-to-book ratio and its capital expenditures. The market-to-book ratio has been widely used to proxy for a firm’s investment opportunities or growth potential (e.g. Comment and Schwert, 1995), with a high market-to-book ratio seen to imply that the capital markets regard the firm as having good growth prospects. We expect private equity target firms to have lower observable measures for growth prospects and hence, a relatively lower market-to-book ratio. Since private equity target firms are argued to be those for which management perceives limited growth opportunities, we expect the private equity target firms’ capital expenditures to be relatively lower.9 In measuring a target firm’s capital expenditure, we scale by total assets.
4.2. Bid attributes
Private equity firms have frequently been dubbed ‘Barbarians at the gate’ or the ‘purest form of capitalism’ and hence are often associated with hostility. This would not be consistent, however, with the recent experience in Australia, where private equity firms prefer to proceed with target board support (Austin and Tuch, 2008).10 Based on the arguments in Austin and Tuch (2008) and Cheffins and Armour (2007), and the expressed attitude of private equity towards a friendly takeover, we propose:
Proposition 4: Private equity takeover bids will be friendly relative to other takeover bids.
To test this proposition, we consider five measures identified in the literature as being aligned with the notion of friendliness and/or with bid success (board recommendation, bid revision, bidder toehold, break fees and blockholdings). Hostility is most often defined in terms of the recommendation made by the target board of directors (Morck et al., 1988). In Australia, under s. 638(3)(a) Corporations Act, each target director must state, and provide reasons for, their recommendation that the bid be accepted or not accepted. Recent evidence suggests, however, that the percentage of favourable recommendations has increased significantly in recent years (Chapple and Treepongkaruna 2006). We therefore appeal to additional measures in hopes of further discriminating the samples.
A second measure of hostility is that of a bid revision (Henry, 2004). Henry finds that the outcome of a takeover is predominantly determined by takeover-specific factors, such as the recommendation made by the target board, the magnitude of the bidding firm’s toehold in the target and the existence of offer price revisions during the bidding process. A bid revision can be considered as a measure of hostility because it is usually undertaken subsequent to unreceptive attempts at negotiation or a reject recommendation by the target board. Here, private equity bidders are expected to make fewer bid revisions than other bidders. Third, in relation to toeholds (defined as pre-bid shareholdings of the bidder), Betton et al. (2005) conclude that friendly bidders tend to refrain from toeholds primarily to avoid ‘toehold-induced target resistance’. They posit that proposing a bid with a toehold may be viewed as aggressive and consequently endanger friendly merger negotiations. We expect acquisitions involving private equity firms to have lower toeholds than for other takeover bids.
Break fees, our fourth measure, have been found to deter takeover competition and increase the rate of bid success (Bates and Lemmon, 2003; Officer, 2003). In Australia, based on the anti-competitive effect of break fees, the Australian Takeovers Panel has fixed break fee usage to one per cent of deal value. Chapple et al. (2007) find this intervention is effective in diluting the otherwise anti-competitive effects of break fees, but that the probability of bid success is lower in the presence of a break fee. Since a bidder must be in a certain position to be able to negotiate a break fee, we expect break fees to be relatively more prevalent in private equity bids. Our fifth and final measure of friendliness is then blockholdings. The literature suggests that large external shareholders have an increased incentive and ability to monitor and discipline management (Shleifer and Vishny, 1986). Further, external blockholders are in a better position to negotiate directly with the bidding firm. Eddey (1991) posits that it makes little sense to proceed with a formal bid unless substantial target shareholders have been consulted and have agreed to the terms of the bid. We thus expect the blockholdings to be greater in private equity targets. The proxies we adopt are the number of blockholders and ownership concentration.
5. Sample data
5.1. Data
Our private equity sample consists of all listed target firms identified in the Connect 4 ‘Takeovers’ database over the period 2001–2007 where the bidder was a private equity firm or the bidding consortium involved a private equity firm.11 Since our research interests are the target firm characteristics and the bid attributes, we consider all bids irrespective of outcome (successful, current, withdrawn and unsuccessful). In addition, we consider both schemes of arrangement (mergers) and takeover bids.
The final private equity sample consists of the 23 sample unique target firms private equity bids for which the required target firm characteristic and bid attribute data could be obtained from the Connect 4 ‘Takeovers’, AspectHuntley FinAnalysis, and/or the AspectHuntley DatAnalysis databases. Panel A of Table 1 presents a frequency distribution by year for the sample bids. The majority occurred in 2006 and 2007 (13/23). Panel A also presents a frequency distribution for the 502 corporate (non-private equity) bids by year over the study period for which the required data could be obtained from the same databases, and in addition, private equity bids expressed as a proportion of the total bids. Here, the private equity bids are again revealed play a relatively more significant role in 2006 and 2007 when they represent approximately 6 per cent of all bids.
| Panel A: Frequency distribution by year | |||
|---|---|---|---|
| Year | Private equity bids | Non-private equity bids | Private equity proportion |
| 2001 | 1 | 68 | 0.014 |
| 2002 | 1 | 47 | 0.021 |
| 2003 | 3 | 59 | 0.048 |
| 2004 | 3 | 63 | 0.045 |
| 2005 | 2 | 59 | 0.033 |
| 2006 | 7 | 108 | 0.061 |
| 2007 | 6 | 98 | 0.058 |
| Total | 23 | 502 | 0.044 |
| Panel B: Frequency distribution by industry | ||||
|---|---|---|---|---|
| GICS sector | GICS code | Private equity | Corporate takeovers | Year and industry match |
| Energy | 10 | 1 (0.043) | 34 (0.068) | 1 (0.012) |
| Materials | 15 | 1 (0.043) | 117 (0.233) | 7 (0.086) |
| Industrials | 20 | 2 (0.087) | 53 (0.106) | 8 (0.099) |
| Consumer discretionary | 25 | 12 (0.522) | 85 (0.169) | 27 (0.333) |
| Consumer staples | 30 | 1 (0.043) | 33 (0.066) | 5 (0.062) |
| Health care | 35 | 1 (0.043) | 35 (0.067) | 12 (0.148) |
| Financials | 40 | 1 (0.043) | 80 (0.159) | 13 (0.160) |
| Information technology | 45 | 2 (0.087) | 33 (0.066) | 6 (0.074) |
| Telecommunications | 50 | 0 (0.000) | 14 (0.028) | — |
| Utilities | 55 | 2 (0.087) | 18 (0.036) | 2 (0.025) |
| Total | 23 (1.000) | 502 (1.000) | 81 (1.000) | |
- The samples consist of 23 private equity bids, 502 corporate bids, and 81 corporate bids matched with the private equity bids by year and industry. Panel A present the frequency distribution of private equity and corporate bids by year, and the number of private equity bids as a proportion of total bids (private equity plus corporate). Panel B present the frequency distribution of private equity, corporate and matched corporate bids by two-digit GICS Industry Sector.
Panel B of Table 1 presents a frequency distribution by GICS industry sector. As revealed, while all GICS sectors except Telecommunications Services are represented within the private equity bids, they are heavily concentrated in the Consumer Discretionary sector (52.2 per cent). In contrast, the corporate sample bids are more evenly spread across the sectors, with Materials being the most frequently represented (23.3 per cent).
These profiles reveal three notable differences. First, while 106 of the 117 corporate sample bids within the Materials sector occur in the Metals and Mining group, none of the sample private equity bids do (the one bid is in the Chemicals group). Second, while 15.9 per cent of corporate sample bids occur within the Financial sector, only one private equity bid does (in the Real Estate group). Finally, the private equity sample bids are heavily weighed into the Consumer Discretionary sector, whereas this sector plays a much more modest role in the corporate sample. Together, these contrasts suggest that private equity bidders are reticent to bid in either the Financial or the Metals and Mining sectors. While conjectural, possible reasons as to why private equity may shy away from the Metals and Mining group include the often unpredictable outcomes of these ventures and their dependence on equity funding. For the Financial sector, one possible reason is its strict regulatory environment. Thus, the apparent private equity ‘overweighting’ of the Consumer Discretionary sector may simply reflect an avoidance of targets in Financial and/or Metals and Mining sectors.12
5.2. Benchmark strategy
As revealed in Table 1 and discussed before, the distributions of the 23 private equity and 502 corporate bids differ both over time and across industry. Given these differences, the strategy we adopt to benchmark the private equity target firm characteristics and bid attributes is against corporate takeover targets matched by year and industry sector (two-digit GICS code). Typically, there is more than one eligible corporate match. Thus, to eliminate the possibility for selection bias, we pool the private equity and corporate takeover firms within industry and year and rank each measure in ascending order. We then convert the ranks to percentile ranks given the differing sample sizes within each industry/year group and base our analyses on these percentile ranks.13 In total, there are 81 matched corporate targets. As revealed in Panel B of Table 1, while the industry distribution of matched bids is closer to that of the private equity bids, the overlay is imperfect given the differing number of corporate bids within each industry/year group. While for completeness we also present results based on raw data, we focus on the results using the percentile rank data since our analyses are conducted using the pooled sample of private equity and matched corporate bids drawn from different years and industries, with target firm characteristics and bid attributes likely differ across both.
5.3. Target firm characteristics
Mean and median values of pre-bid target firm characteristics for the reporting period immediately preceding the takeover offer are presented in Panel A of Table 2. The first set of columns present values based on the raw data while the second set present values based on the percentile rank data. In each instance, the figures presented are those for the private equity sample, those for the matched corporate sample, and p-values for tests of differences in mean and median values between the two samples. For the raw data, we emphasise median tests given the distributional properties of the measures.
| Measure | Raw data | Percentile rank data | |||||
|---|---|---|---|---|---|---|---|
| PE | Match | p-value | PE | Match | p-value | ||
| Panel A: Target firm characteristics | |||||||
| Deal market value ($ millions) | Mean | $1 332.727 | $544.154 | 0.049 | 0.755 | 0.428 | <0.001 |
| Median | $409.338 | $96.208 | <0.001 | 0.833 | 0.400 | <0.001 | |
| Total assets ($ millions) | Mean | $2 141.716 | $1 019.905 | 0.096 | 0.735 | 0.433 | <0.001 |
| Median | $443.547 | $86.017 | <0.001 | 0.857 | 0.417 | <0.001 | |
| Total revenue ($ millions) | Mean | $1 115.149 | $281.604 | 0.021 | 0.769 | 0.424 | <0.001 |
| Median | $425.560 | $63.148 | <0.001 | 0.800 | 0.400 | <0.001 | |
| Return on equity | Mean | 15.90% | −1.85% | 0.042 | 0.651 | 0.457 | 0.046 |
| Median | 12.19% | 7.78% | 0.019 | 0.714 | 0.444 | 0.050 | |
| Return on assets | Mean | 7.18% | −1.39% | 0.042 | 0.719 | 0.438 | 0.003 |
| Median | 6.97% | 3.89% | 0.003 | 0.667 | 0.444 | 0.003 | |
| Operating cash flow | Mean | $160.337 | $22.994 | 0.006 | 0.787 | 0.419 | <0.001 |
| Median | $48.240 | $3.4500 | <0.001 | 0.857 | 0.400 | <0.001 | |
| Earnings after tax | Mean | $71.134 | $30.192 | 0.127 | 0.792 | 0.417 | <0.001 |
| Median | $34.544 | $1.972 | 0.001 | 0.875 | 0.400 | <0.001 | |
| Asset turnover ratio | Mean | 1.072 | 0.883 | 0.144 | 0.649 | 0.458 | 0.070 |
| Median | 0.946 | 0.477 | 0.051 | 0.667 | 0.400 | 0.051 | |
| Current ratio | Mean | 1.430 | 4.484 | 0.116 | 0.364 | 0.539 | 0.054 |
| Median | 1.310 | 1.630 | 0.005 | 0.400 | 0.583 | 0.075 | |
| Debt-to-assets ratio | Mean | 0.291 | 0.233 | 0.396 | 0.573 | 0.479 | 0.085 |
| Median | 0.267 | 0.141 | 0.079 | 0.571 | 0.444 | 0.087 | |
| Panel B: Bid attributes | ||||||
| Bid Premium, 20 days prior | ||||||
| Mean | 0.144 | 0.272 | 0.059 | 0.377 | 0.555 | 0.031 |
| Median | 0.154 | 0.220 | 0.038 | 0.400 | 0.586 | 0.042 |
| Independent expert reports | ||||||
| No report | 2 | 36 | 0.002 | |||
| Report | 21 | 45 | ||||
| Best interests | 14 | 17 | 0.071 | |||
| Fair and reasonable | 3 | 18 | ||||
| Not fair but reasonable | — | 3 | ||||
| Neither fair nor reasonable | 4 | 7 | ||||
| Foreign bid | ||||||
| Foreign bid | 6 | 34 | 0.167 | |||
| Domestic bid | 17 | 47 | ||||
| Offer type | ||||||
| Cash | 23 | 56 | 0.009 | |||
| Scrip | — | 14 | ||||
| Cash/scrip | — | 11 | ||||
| Bid success | ||||||
| Successful | 14 | 68 | 0.042 | |||
| Unsuccessful | 5 | 9 | ||||
| Withdrawn | 4 | 4 | ||||
- The samples are 23 private equity bids (PE) and 81 corporate bids matched by year and industry (Match). The measures presented are mean and median values for both raw and percentile rank data, and two-tailed p-value for tests of differences. Percentile ranks were determined by pooling the private and matched corporate bids, ranking each measure within industry and year, and then converting the ranks to percentiles. For the categorical variables presented in Panel B, the p-values are for the Chi-square test of independence.
The reported statistics suggest that private equity bidders seek larger, more profitable and more efficient firms with higher operating cash flows. Curiously, the private equity targets also appear to be more highly levered and less liquid. In detail, the raw median value and respective mean and median percentile rank values of each of our three size proxies (deal implied market value, total assets, and revenues) are all significantly greater for the private equity target sample than for the applicable benchmark sample at the 1 per cent level.
A similar pattern presents for our three profitability measures (ROE, ROA, and earnings after tax) and for operating cash flow, with raw median and percentile rank mean and median values all significantly higher for the private equity target sample at the 5 per cent level for the profitability measures and at less than the 1 per cent level for the cash flow measure. Interestingly, a focus on better performing firms is inconsistent with Manne’s (1965) inefficient management hypothesis and thereby suggestive of a non-disciplinary function for the acquisition. Of note, the pre-bid private equity target ROEs ranged from 0.044 to 0.517. whereas 138 of the 502 corporate target firms registered a negative ROE. Also of note, 182 of the 502 corporate target firms experienced a negative operating cash flow whereas only four of the 23 private equity target firms did.
Finally, in terms of efficiency, liquidity, and leverage, the asset turnover and debt-to-asset ratios are greater for the private equity sample, while the median current ratio value is lower. In each instance, the difference is statistically significant at the 10 per cent level.
5.4. Bid Attributes
Bid attribute descriptive statistics are presented in Panel B of Table 2. To begin, there is consistent evidence that private equity bids involve a significantly lower bid premium. For example, based on the target share price 20 trading days prior to the bid, the median bid premium for the private equity sample is 15.4 per cent whereas it is 22.0 per cent for the corporate sample. The p-value on the difference is 0.038. An identical conclusion follows from the percentile rank statistics. Thus, it appears that, on average, private equity bidders either do not find it necessary to offer as high a premium or are unwilling to do so.14
The evidence also indicates that significantly more private equity bids involve an independent expert. Such a finding is, however, not unexpected. Specifically, the Corporations Act mandates expert reports in three circumstances: (i) where the bid is a takeover by scheme of arrangement (under s. 411); (ii) where the bidder already has a stake in the target of 30 per cent (under s. 640(1)(a)); or (iii) where the bidder and target have common directors (under s. 640(1)(b)).15 Against this backdrop, of greater interest then is the expert’s recommendation. Here, for the private equity bids, 81.0 per cent were deemed to be either in the ‘best interests’ of target shareholders or ‘fair and reasonable’ compared with 77.8 per cent of the matched bids. On the other side, independent experts deemed 19.0 per cent of private equity bids and 22.2 per cent of the matched bids to be either ‘not fair but reasonable’ or ‘neither fair nor reasonable’. The p-value for the Chi-square test of independence between recommendation and type of bid is 0.071. Thus, there appears to be a modest difference in the recommendations, although the evidence is at best weak.
The evidence also suggests differences in the source of the bid, foreign or domestic, with 26.1 per cent of the private equity bids and 42.0 per cent of the matched bids involving a foreign bidder. The Chi-square tests, however, fail to reject independence. While not unexpected given the nature of private equity firms and their use of debt to fund bids, the profile for the method of payment also potentially presents an interesting contrast. The only form of payment offered in the private equity bids was cash. In contrast, 69.1 per cent of the matched bids were cash only, with the remainder involving scrip only or a mixture of cash and scrip. Finally, there is evidence that private equity bids are ultimately less successful. Here, 60.9 per cent of the private equity bids are successful, 21.7 per cent unsuccessful, and 17.4 per cent withdrawn. In contrast, 84.0 per cent of the matched bids are successful, 11.1 per cent unsuccessful, and 4.9 per cent withdrawn. The p-value on the Chi-square test is 0.042.
6. Empirical results
6.1. Univariate target firm characteristic results
Univariate target firm characteristic results are presented in Panel A of Table 3. Again, the first set of columns present values based on the raw data and the second set based on the percentile rank data. In each instance, the figures presented are those for the private equity sample, the matched corporate sample, and p-values for tests of differences in values between the two samples. In brief, the results provide strong support for the free cash flow and stability propositions (Propositions 2a and b), modest support for the growth prospects proposition (Proposition 3), but only mixed support Proposition 1 regarding financial slack.
| Measure | Sign | Raw data | Percentile rank data | |||||
|---|---|---|---|---|---|---|---|---|
| PE | Match | p-value | PE | Match | p-value | |||
| Panel A: Target firm characteristics | ||||||||
| Financial slack | ||||||||
| Debt-to-equity ratio | (−) | Mean | 0.426 | 0.158 | 0.385 | 0.584 | 0.476 | 0.152 |
| Median | 0.496 | 0.173 | 0.198 | 0.600 | 0.472 | 0.173 | ||
| Interest coverage ratio | (+) | Mean | 51.89 | 5.363 | 0.383 | 0.676 | 0.450 | 0.025 |
| Median | 6.55 | 3.160 | 0.049 | 0.655 | 0.400 | 0.026 | ||
| Stability | ||||||||
| SD (cash flow from operations) | (−) | Mean | 12.389 | 16.652 | 0.069 | 0.411 | 0.525 | 0.032 |
| Median | 10.166 | 14.017 | 0.038 | 0.400 | 0.543 | 0.029 | ||
| Dividend payout ratio | (+) | Mean | 0.568 | 0.650 | 0.643 | 0.537 | 0.489 | 0.230 |
| Median | 0.550 | 0.468 | 0.378 | 0.615 | 0.500 | 0.511 | ||
| Price to earnings ratio | (−) | Mean | 17.45 | 38.447 | 0.103 | 0.345 | 0.544 | 0.019 |
| Median | 17.30 | 20.050 | 0.236 | 0.279 | 0.600 | 0.002 | ||
| Free cash flow | ||||||||
| Free cash flow | (+) | Mean | $128.356 | −$83.642 | 0.068 | 0.799 | 0.415 | <0.001 |
| Median | $16.930 | −$2.290 | <0.001 | 0.875 | 0.400 | <0.001 | ||
| Free cash flow/total assets | (+) | Mean | 0.057 | −0.073 | 0.002 | 0.699 | 0.443 | 0.002 |
| Median | 0.050 | −0.060 | 0.001 | 0.700 | 0.400 | 0.001 | ||
| Growth Prospects | ||||||||
| Market-to-book ratio | (−) | Mean | 2.07 | 3.25 | 0.091 | 0.395 | 0.530 | 0.060 |
| Median | 1.90 | 2.50 | 0.083 | 0.333 | 0.556 | 0.063 | ||
| Capital expenditure/total assets | (−) | Mean | 0.063 | 0.077 | 0.685 | 0.539 | 0.489 | 0.534 |
| Median | 0.048 | 0.065 | 0.073 | 0.500 | 0.500 | 0.950 | ||
| Panel B: Bid attributes | |||||||
| Board recommendation | Accept | 16 | 56 | 0.250 | |||
| Reject | 5 | 9 | |||||
| No act | 2 | 16 | |||||
| Bid revision | Yes | 7 | 17 | 0.343 | |||
| No | 16 | 64 | |||||
| Toehold | Freq | 8 | 41 | 0.809 | |||
| Mean | 0.309 | 0.240 | 0.403 | 0.440 | 0.516 | 0.628 | |
| Median | 0.184 | 0.198 | 0.892 | 0.333 | 0.500 | 0.620 | |
| Break fee | Freq | 14 | 18 | <0.001 | |||
| Mean | 0.012 | 0.012 | 0.704 | 0.511 | 0.497 | 0.587 | |
| Median | 0.011 | 0.010 | 0.841 | 0.500 | 0.500 | 0.722 | |
| Number of blockholders | Mean | 3.830 | 2.910 | 0.028 | 0.647 | 0.458 | 0.031 |
| Median | 4.000 | 3.000 | 0.040 | 0.700 | 0.400 | 0.027 | |
| Ownership concentration | Mean | 0.709 | 0.682 | 0.230 | 0.596 | 0.455 | 0.074 |
| Median | 0.752 | 0.686 | 0.066 | 0.583 | 0.444 | 0.075 | |
- The samples are 23 private equity bids (PE) and 81 corporate bids matched by year and industry (Match). The measures presented are mean and median values for both raw and percentile rank data, and two-tailed p-value for tests of differences. Percentile ranks were determined by pooling the private and matched corporate bids, ranking each measure within industry and year, and then converting the ranks to percentiles. For the categorical variables presented in Panel B, the p-values are for the Chi-square test of independence.
In detail, for the debt-to-equity ratio, none of the tests for differences are significant at conventional levels. If anything, directionally the percentile data hint at private equity firms having more debt, not less as predicted. Conversely, for the interest coverage ratio, tests for differences in the median values and percentile rank mean and median values are all significant at the 5 per cent level in the predicted direction. Thus, the univariate results provide mixed support for Proposition 1, revealing that private equity targets do not have the predicted lower levels of debt but that they do nevertheless have superior coverage.
For stability, we find as predicted, the median value and mean and median percentile rank values of our primary proxy, the standard deviation of per share cash flows from operations, to be significantly lower for private equity targets at the 5 per cent level.16 For the price-earnings ratio, all differences are in the predicted direction, with differences in mean and median percentile ranks significant at the 5 per cent level. Conversely, for the dividend payout ratio, while differences are in the predicted direction, none of the differences are statistically significant at conventional levels. Taken together, we interpret these univariate results as favouring Proposition 2a, especially given the significance on the primary proxy.
Our results provide strong support for Proposition 2b, with the median value and mean and median percentile rank values for both free cash flow and FCF scaled by total assets are all significantly greater for the private equity target sample than for the benchmark sample at the 1 per cent level. Thus, private equity targets have substantially higher free cash flow. Finally, the results also provide some support for the prediction that the private equity target firms have lower measurable growth prospects (Proposition 3). For our primary measure, the market-to-book ratio, the median value and mean and median percentile rank values are lower for the private equity target firms with the differences weakly significant at the 10 per cent level. The results are less clear for the capital expenditure measure, with the median value for the private equity sample lower, as predicted, at the 10 per cent level but tests based on the percentile rank statistics insignificant.
6.2. Univariate bid attributes results
Univariate bid attribute results are presented in Panel B of Table 3. The results provide at best mixed support for Proposition 4, with only arguments relating to break fee usage and the importance of blockholders supported. There is no direct support for arguments relating to differences in the friendliness of the bid based on the most commonly used proxies, the board’s recommendation, the frequency of bid revision or existence of a bidder toehold.
In detail, 69.6 per cent of the private equity bids elicited an ‘accept’ recommendation from the target board whereas 21.8 per cent, a ‘reject’ recommendation. The comparable figures for the matched bids are 69.1 and 11.1 per cent. Similarly, 30.4 per cent of the private equity bids involved a bid revision compared with 21.0 per cent of the matched bids. In both instances, the Chi-square tests fail to reject independence. Thus, we find no evidence of a difference in the target board recommendation or the incidence of a bid revision.
There is also no evidence of a relation between the source of bid (private equity or corporate) and either the existence or magnitude of a bidder toehold. Here, 34.8 per cent of the private equity bidders had a toehold compared with 32.1 per cent of matched bidders. The Chi-square test again fails to reject independence. Further, tests for differences in both raw data and percentile ranks mean and median values are insignificant.17
Alternatively, as predicted, private equity bids appear far more likely to involve a break fee. For the private equity bids, 60.9 per cent included a break fee whereas only 22.2 per cent of the matched bids did. The Chi-square test rejects independence at less than the 1 per cent level. On the other hand, a comparison of the mean and median break figures reveals no difference. Such a result is not, however, unexpected given the cap imposed on break fees in the Australian setting. Finally, as predicted, private equity targets have both more blockholders and a higher ownership concentration, with differences in median values and mean and median percentile rank values significant at the 5 per cent level for the blockholder measure and the 10 per cent level for the shareholder concentration measure.
6.3. Multivariate results
Results for variants of the model are presented in Table 4. The first set of columns is run based on the raw data, while the second set of columns is run based on percentile rank measures. Within each, the first model includes only the primary measures, while the second model additionally includes the four control variables. Pair-wise correlations (not tabulated) suggest that there is unlikely to be a threat of multicollinearity. The largest pair-wise correlation, between ROE and FCF/TA, is 0.497.19
| Variable | Sign | Raw data | Percentile rank data | ||
|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | ||
| Intercept | −0.354 | −20.366 | −1.756 | −4.637 | |
| (0.072) | (0.001) | (0.034) | (0.067) | ||
| D/E | − | −0.308 | −0.877 | −1.836 | −2.745 |
| (0.593) | (0.078) | (0.647) | (0.034) | ||
| σ(CFO) | − | −0.547 | −0.222 | −0.621 | −0.774 |
| (0.079) | (0.037) | (0.012) | (0.029) | ||
| FCF/TA | + | 5.048 | 3.966 | 12.832 | 36.099 |
| (0.015) | (0.008) | (0.004) | (0.002) | ||
| M/B | − | −0.554 | −0.653 | −6.962 | −5.210 |
| (0.053) | (0.043) | (0.005) | (0.049) | ||
| BLOCK | + | 0.206 | 0.243 | 5.835 | 7.394 |
| (0.062) | (0.019) | (0.040) | (0.011) | ||
| ln TA | 0.862 | 25.439 | |||
| (0.005) | (0.032) | ||||
| ROE | 5.065 | −3.925 | |||
| (0.135) | (0.306) | ||||
| TURN | 0.791 | −7.135 | |||
| (0.128) | (0.402) | ||||
| CR | −0.051 | −8.174 | |||
| (0.822) | (0.540) | ||||
| Percentage correct | 0.704 | 0.796 | 0.891 | 0.954 | |
| Nagelkerke, R2 | 0.208 | 0.450 | 0.736 | 0.910 | |
- PRIV, dichotomous variable set to 1 for private equity deals and 0 otherwise; D/E, the firm’s debt-to-equity ratio; σ (CFO), the standard deviation of the firm’s per share cash flow from operations; FCF/TA, the firm’s free cash flow scaled by total assets; M/B, the firm’s market-to-book ratio; BLOCK, the number of blockholders; TA, the firm’s total assets; ROE, the firm’s return on equity; TURN, the firm’s asset turnover; CR, the firm’s current ratio. All accounting measures are for the reporting period immediately preceding the takeover offer. The analysis is conducted on a pooled sample of 23 private equity bids (PRIV = 1) and 81 corporate bids matched by year and industry (PRIV = 0). Two-tailed p-values presented in parentheses.
As can be seen, the results are, in the main, consistent with the univariate results presented in Table 3. They reveal, as predicted, significant and positive coefficients on the cash flow (FCF/TA) and blockholder (BLOCK) measures, and negative and significant coefficients on the stability (σ(CFO)) and growth (M/B) measures. Further and in contrast with the univariate results, they also reveal, as predicted, a negative and significant coefficient on the financial slack measure, D/E, in the complete models. For example, for the complete model based on the percentile rank data, their coefficients (p-values) are: D/E, −2.745 (p = 0.034); σ(CFO), −0.774 (p = 0.029); FCF/TA, 36.099 (p = 0.002); M/B, −5.210 (p = 0.049); and BLOCK, 7.394 (p = 0.011). This model has a prediction accuracy of 95.4 per cent and a Nagelkerke R2 of 0.910. For the control variables, we find a positive and significant coefficient on size while remaining measures are insignificant. Thus, taken together, the results for the multivariate analyses provide direct support for all propositions.
6.4. Sensitivity to choice of benchmark sample
For the analysis reported above, we benchmark the 23 private equity bids against a sample of 81 corporate bids matched by year and industry. To consider the sensitivity of our results and conclusions to the choice of benchmark strategy, in this section we alternatively benchmark the private equity bids against the full sample of 502 corporate bids over the study period 2001–2007 with available data. Notwithstanding the significant differences across time and industry between this sample and the private equity sample revealed by Table 1,20 the results (not tabulated) are qualitatively similar to those reported in all tables and thereby indicate that conclusions are indeed robust to the choice of benchmark strategy.
Specifically, focusing on median values, against this alternative benchmark we again find our sample private equity target firms to have a lower standard deviation of per share cash flow (p = 0.046), greater free cash flow (p = 0.004), a lower market-to-book ratio (p = 0.079), and a higher interest coverage ratio (p = 0.022). Differences in remaining measures are also qualitatively similar to those reported in Table 3. Thus, relative to the sample of 502 corporate targets, the private equity targets again have greater and more stable cash flows (Propositions 2a and b), modestly greater growth prospects (Proposition 3), and greater financial slack as measured by the coverage ratio but not the leverage ratio (Proposition 1).
For the bid results, we again find evidence of greater break fee usage (p = 0.002), more blockholdings (p = 0.014), and greater ownership concentration (p = 0.079) with the private equity bids (Proposition 4), but no evidence of differences in board recommendation (p = 0.436), the incidence of bid revision (p = 0.353), or either the existence (p = 0.244) or magnitude (p = 0.889) of a toehold. Finally, the results for the logistic regression model are also qualitatively similar to those reported in Table 4. Specifically, for the complete model (based on the raw data), the coefficients (p-values) are now: D/E, −1.368 (p = 0.061); σ(CFO), −0.167 (p = 0.036); FCF/TA, 4.210 (p = 0.010); M/B, −0.802 (p = 0.031); and BLOCK, 0.462 (p = 0.009). Thus, the results for the multivariate analyses using the alternative benchmark sample again provide direct support for all propositions.
7. Summary and conclusions
In recent years, the growth in the private equity sector has provoked an increased curiosity surrounding private equity bids from academics, practitioners and regulators alike. Although there is no extant literature encompassing the determinants of private equity bids, current theory and existing literature on the motivations for M&A is used to inform this study into the determinants of private equity bids. Hence, the current study provides the first formal albeit preliminary investigation into the determinants of private equity bids in Australia.
More specifically, the focus of this study is on whether there are unique aspects of both the target firms that private equity pursues and the nature of the bids that ensue. The framework for the investigation derives from the Senate inquiry into private equity, most notably a submission made to the inquiry by the Australian Private Equity and Venture Capital Association Limited, which identified the value-adding attributes of private equity investment. Based on this framework, the related academic literature, and anecdotal evidence from the investment criteria proclaimed by private equity firms, we propose that relative to other corporate target firms, private equity target firms should have greater financial slack, both debt capacity and free cash flow, greater business stability and lower growth prospects. In addition, we also propose that private equity bids will be predominantly of a friendly nature, more so than other corporate bids.
The private equity sample we consider consists of 23 target firms subject to a takeover bid involving a private equity bidder over the period 2001–2007. The benchmark sample we use consists of 81 corporate bids matched by year and industry. For sensitivity, we also consider an alternative benchmark consisting of all corporate bids during the study period. Results based on this alternative benchmark sample reveal our conclusions to be robust to the choice of benchmark strategy.
In detail, the descriptive statistics provide an overview of a typical private equity bid. They suggest that private equity target firms are larger, more profitable, use their assets more efficiently, more highly levered, and have greater cash flow than other corporate targets. Further, the bid descriptive statistics suggest that the bid premium offered in private equity deals is lower and a slightly smaller proportion of bids are successful. Private equity bids also involve an independent expert more often, but this is because the sample consists predominantly of mergers by scheme of arrangement. Perhaps as expected given the debt funding of private equity investment, the preferred mode of payment for private equity bidders is cash, in contrast with other offers that can and do involve scrip. Finally, the industry composition of the private equity targets suggests that private equity bidders are reticent to bid in either the Financial or the Metals and Mining sectors. Possible reasons include the strict regulatory environment in which the Financial sector operates, and the unpredictable outcomes of Metals and Mining ventures and their dependence on equity funding.
Turning to our primary propositions, we interpret the results from our multivariate analyses as providing consistent and direct for all propositions, support which is in the main echoed by our univariate analyses. Specifically, they indicate that private equity targets have greater financial slack, greater financial stability, greater free cash flow and lower measurable growth prospects than other corporate targets. Further, they provide some support regarding the friendliness of private equity bids, with arguments relating to break fee usage and the importance of blockholders supported.
The study potentially informs regulators and policy makers, such as the Senate inquiry in Australia and the Financial Services Authority Inquiry in the UK. We find that private equity bidders appear to occupy an unusual place in the market as disciplinary, friendly acquirers. However, consistent with da Silva Rosa and Walter (2004), the disciplinary motive for takeover activity in Australia appears to be oversold. Moreover, as private equity bidders appear not to focus on particular industries, the synergistic motive is obviously less plausible. Consequently, private equity bidders appear to play an ‘opportunistic’ role not readily explained by either the disciplinary or synergistic hypothesis.
In sum, while clearly exploratory in nature, arguably this research provides the beginnings of a scientific foundation for further investigation into the determinants of private equity bids in Australia. Although an expanded sample of private equity bids would be preferable, the results in general are surprisingly consistent.
