«This version: November 2009 Abstract We present evidence of the impact of asymmetric information on ﬁrms’ likelihood to issue equity and to be ...»
Does Asymmetric Information Aﬀect SEOs and
M&A? Evidence from Corporate Pension Plans
João F. Cocco∗ Paolo F. Volpin†
This version: November 2009
We present evidence of the impact of asymmetric information on ﬁrms’ likelihood to issue
equity and to be acquired. We argue that the existence of corporate sponsored deﬁned
beneﬁt pension liabilities, and their magnitude, are a source of risk for buyers of the ﬁrm’s
shares. We show that ﬁrms that sponsor DB pension plans and have a large pension deﬁcit relative to their market capitalization are less likely: (i) to raise equity through SEOs; (ii) to be the target in an acquisition and to be taken over; and (iii) to use stock when they acquire other companies or assets.
JEL classiﬁcation: G32, G34, J32.
Keywords: asymmetric information, SEOs, M&A, corporate pensions.
Acknowledgments: We are grateful to Henri Servaes for comments and the ESRC through the Corporate Governance Centre at London Business School for ﬁnancial support.
∗ London Business School, CEPR, and Netspar, Regent’s Park, London NW1 4SA, E-mail: firstname.lastname@example.org † London Business School, CEPR, and ECGI, Regent’s Park, London NW1 4SA, E-mail: email@example.com 1 Introduction There is a vast theoretical literature emphasizing the importance of asymmetric information for corporate ﬁnancial decisions, with perhaps the most famous contribution being the pecking order theory of capital structure proposed by Myers and Majluf (1984) and Myers (1984).1 The basic idea behind these papers is Akerlof (1970)’s lemons principle: in the presence of information asymmetries between the ﬁrm and the market it is costly for ﬁrms to sell equity to outsiders. As in Akerlof’s used car market, adverse selection in the market for equity may lead to a market breakdown.
In contrast to the vast theoretical literature, evidence on the impact of asymmetric information on ﬁrms’ ﬁnancial decisions has been harder to obtain. This is to some extent due to the diﬃculty in measuring asymmetric information across a large group of ﬁrms. Our idea is to use company-sponsored deﬁned beneﬁt (DB) pension plans as a source of asymmetric information for the value of the shares of the companies that sponsor such plans.
The deﬁcit in DB pension plans (that is the diﬀerence between pension liabilities and pension assets) is a liability for the sponsoring company. The size of this deﬁcit is diﬃcult to determine since its value is very sensitive to the assumptions, which are not always made explicit, regarding discount rates, pension plan members longevity, employee mobility, among others. Naturally, uncertainty withrespect to the value of the pension deﬁcit translates into uncertainty with respect to the value of the sponsoring ﬁrm’s equity. Furthermore, the management of the companies that sponsor such plans have more information on the assumptions used and on the value of the deﬁcit than outsiders do. This may make it diﬃcult for ﬁrms that sponsor a DB pension plan to sell equity to outsiders: potential acquirers of the company shares may be worried that they are buying a lemon.
We investigate whether this is the case in two diﬀerent situations in which ﬁrms sell equity to outsiders: Seasoned Equity Oﬀerings (SEOs) and Mergers and Acquisitions (M&A). More precisely we investigate whether ﬁrms that sponsor DB pension plans, and in particular those that sponsor plans with large deﬁcits, are less likely to issue equity in a SEO, to be taken over, and to use equity as a means of payment in acquisitions.
Early contributions in this literature include Ross (1977) on the signaling role of debt, Leland and Pyle (1977) on the informational content of inside ownership, and Bhattacharya (1979) on the signaling role of dividends.
In our study we use United Kingdom (UK) data for the 2002-2008 period. The universe of ﬁrms in our sample are the Footsie 350 companies in 2002, the ﬁrst year in our dataset. Our choice of large companies is motivated by the fact that we would like a sample with variation with respect to the signiﬁcance of company sponsored DB pension plans. About two-thirds of these companies sponsor DB pension plans, and among the latter there is signiﬁcant variation in the value of the pension deﬁcit relative to the market value of the ﬁrms’ equity, ranging from 49 percent to −13 percent. We have hand-collected the value of the deﬁcits and other pension data from the footnotes to the companies’ annual reports. In addition, we use ﬁrm-level data from Worldscope, and SEOs and M&A activity data from SDC Platinum.
To assess whether ﬁrms that sponsor DB pension plans, and particularly those that sponsor plans with large deﬁcits relative to market capitalization, are less likely to issue equity, we estimate probit regressions of the decision to issue as a function of pension plan variables.
We control for several ﬁrm characteristics, including size, proﬁtability, leverage, among others.
Interestingly, we ﬁnd that ﬁrms with a large pension deﬁcit relative to the market value of their equity are less likely to issue and to raise a signiﬁcant amount of equity. It is important to note that we obtain this result controlling for a measure of ﬁrm leverage, that treats the pension deﬁcit as debt for the sponsoring company. That is: our results are not due to the fact that companies that sponsor DB pension plans tend to be more highly leveraged (Shivdasani and Stefanescu, 2009). Hence, our result is consistent with the view that DB pension plans are a source of asymmetric information for the sponsoring ﬁrm.
In addition, we ﬁnd that the sponsoring of DB pension plans has an impact on M&A activity which is consistent with the asymmetric information theory. More precisely, we ﬁnd that ﬁrms which sponsor DB pension plans are less likely to be targeted in M&A acquisitions, particularly when they have a large pension deﬁcit relative to the market value of their equity.
And conditional on being attempted, acquisitions that target ﬁrms which sponsor DB pension plans are less likely to be completed. These ﬁndings lend support to the idea that DB pension plans act as an implicit poison pill that discourages potential (and actual) bidders from acquiring the sponsoring companies.
Although our result that company sponsored pension plans act as a takeover defense resembles the results in Rauh (2006b), it is important to note that the economic mechanism at work in our paper is diﬀerent. Rauh shows that US companies that sponsor pension plans that invest in the shares of the sponsoring company are less likely to be taken over. The mechanism at work in his paper is an agency one, with the employees and the management of the sponsoring company using shares owned by the pension plan to prevent a takeover.2 In contrast, in the UK pension plans are not allowed to invest their assets in the shares of the sponsoring company, nor own assets that are related to the sponsoring company (e.g. the pension plan cannot own property that is rented out to the sponsoring company). Therefore, in our paper another mechanism is at work. Namely, the presence of a DB pension plan, particularly when running a large deﬁcit, represents a source of risk for potential acquirers of the sponsoring ﬁrm’s equity.
We also ﬁnd an eﬀect of pension plan variables on the means of payment in M&A activity. The theoretical work of Hansen (1987) and Fishman (1988) predicts that acquiring ﬁrms for which there are information asymmetries should use cash as the means of payment when acquiring other ﬁrms. This is because target ﬁrms will be reluctant to accept the shares of the acquirer as a means of payment. Consistent with this prediction, we ﬁnd that ﬁrms that sponsor DB pension plans with larger deﬁcits are more likely to use cash when they acquire other companies. This result lends support to the idea that information asymmetries related to company sponsored DB pension plans aﬀect ﬁrms ﬁnancial decisions.3 The remainder of the paper is structured as follows. In the remainder of this section we discuss related literature and we motivate our use of DB pension plans as a source of information asymmetries, and their implications for equity sales. We do so using an illustrative example that describes recent events related to a company in our sample. Section 2 describes the data that we use for our study, and it also includes a description of the methods used to determine the pension deﬁcit. Our main results are presented and discussed in Section 3. Robustness checks are in Section 4. Section 5 concludes.
1.1 An Illustrative Example
The idea that we wish to investigate in this paper is simple but, we believe, important. The value of the equity of a ﬁrm that sponsors a DB pension plan depends on the value of its corporate The employees may be against a takeover since as Pontiﬀ, Shleifer and Weisbach (1990) show DB pension plans are more likely to be terminated following hostile rather than friendly takeovers.
It also complements the early evidence on the relation between means of payment and information asymmetries in Travlos (1987), Eckbo, Giammarino, and Heinkel (1990) and Franks, Harris and Titman (1991), and the more recent evidence provided by Moeller, Schlingemann and Stulz (2004), Oﬃcer (2007), and Raman, Shivakumar, and Tamayo (2008).
pension deﬁcit (pension liabilities net of pension assets) relative to its market capitalization.
Because of the complexity in evaluating the assets and liabilities in a pension plan, corporate insiders (managers and large shareholders) may have more information than the market about the true value of the pension deﬁcit. As a consequence, investors may be reluctant to buy the ﬁrm’s shares. This will aﬀect the ﬁrm’s ability to sell its equity, either in seasoned equity oﬀerings or in M&A activity.
As an example of the relevance of corporate sponsored DB pension plans for SEOs and M&A activity we discuss recent events related to British Airways, which is one of the ﬁrms in our sample. In July 2008, British Airways (BA) and Iberia of Spain announced that they were planning to merge their operations, creating a $3.8 billion company that would beneﬁt from Iberia’s presence in Latin America and BA’s market share in North America and Asia.
Industry analysts believed the two companies to be a good “strategic ﬁt,” since there would be considerable cost savings generated by bringing their operations together.
However, in the months following the announcement of the all-share merger, discussions stalled on concerns about the size of BA’s pension deﬁcit and on stakes valuation. In March 2008, BA’s pension deﬁcit was $437 million, while BA’s market capitalization had declined to $2 billion pounds. Later that year, BA admitted that on September 18 the trustees of the pension plan had calculated the deﬁcit to be equal to $1.74 billion. Industry experts believed that the deﬁcit might be even larger.
The size of the BA pension scheme deﬁcit, and the large ﬂuctuations in its value were a concern to Iberia, which by December 2008, after months of struggling to understand it, had hired Mercer, a pension consulting ﬁrm, to review BA’s pension schemes. Fernando Conte, Iberia Chairman, said that unless Iberia could protect itself from the BA scheme, it would be “bonkers” to enter into a transaction where it was “on the hook” for it (The Daily Telegraph, 28 December 2008).
By February 2009 signs of diﬃculty in reaching a deal were apparent. Nick van den Brull, an analyst with BNP Paribas, said “I don’t see anything happening before the pension deﬁcit is known. Directors [at Iberia] would wish to know the maximum extent of the liabilities before proceeding.” (The New York Times, 8 February 2009). It was this uncertainty, and the worry that it may be buying a lemon, that led Iberia, three months later, to propose that the merger contains a post-deal adjustment to account for BA’s pension deﬁcit. More precisely, Iberia proposed an all-share merger ratio that would be adjusted if the deﬁcit widens because of such things as increased longevity assumptions. However, BA ﬁercely resisted demands for such adjustment mechanism, believing it would be impossible to sell such an open-ended deal to its shareholders (The Daily Telegraph, 22 May 2009).
At the same time, British Airways’ Chairman Martin Broughton was blaming the ﬁrm’s spiralling pension scheme deﬁcit on the processes employed by actuaries and accountants. “It’s little wonder that the Spanish have such diﬃculty understanding [our pensions]. It’s clearly time that the actuarial and the accounting world got together and recognized the folly of having mechanical processes in place that produce such divergent results, neither of which really seem in touch with reality.” At the same time he called for an overhaul of the system to help BA shareholders and Iberia understand the situation (Pensions Week, 25 June 2009).
One month later BA acknowledged that due to the crisis it needed to increase liquidity, but at the same time its Chairman Martin Broughton ruled out a rights issue, arguing that it was not a good time for the ﬁrm to carry it out. This happened amid questions about the sustainability of the ﬁrm’s business model, geared towards premium travelers, and the ﬁrm’s pension deﬁcit (Dow Jones Market Watch, 14 July 2009).
On November 12, after many months of discussions, BA and Iberia announced that they had reached a preliminary agreement for a merger expected to be completed in late 2010. Under its terms Iberia would take a 45% stake and BA a 55% stake. However, Iberia said it can pull out if BA fails to resolve its pension deﬁcit problem (BBC news, 12 November 2009).
The BA example illustrates how concerns about the value of the company sponsored pension deﬁcit, and information asymmetries regarding the value of this deﬁcit, can aﬀect M&A activity and the decision to issue equity. Although the events described suggest that there is such an eﬀect for BA, the purpose of this paper is to ﬁnd out how relevant such considerations are for a wider sample of ﬁrms. For that reason we have collected data for a large sample of UK ﬁrms.
1.2 Related Literature