Leonard N. Stern School of Business Administration
New York University
Corresponding Author: Dov Fried, Stern School of Business Administration, New York University,
40 West 4th St., NY, NY 10012, (212) 998-0005, email@example.com
Pension Plan Contributions, Free Cash Flows and Financial Slack Abstract The agency theory literature suggests that conflicts of interest between management and shareholders and/or information asymmetries can lead managers to hoard excess free cash flows. This study examines whether pension plans are used by managers to build financial slack.
The study finds a significant and positive relationship between levels of free cash flows and pension plan contributions, after controlling for various variables that may affect pension contributions. Because this relationship is driven by firms with positive free cash flows and not by firms with negative free cash flows, the phenomenon is suggestive of more than just an “ability to pay”. The results are consistent with the Myers and Majluf theory that management stores excess cash flows to build financial slack, which can be used for future (profitable) investment projects.
The study further examines the relationship of pension plans and financial slack by testing whether declines in pension plan contributions coincide with increases in profitable investment opportunities. The results confirm this to be the case; firms with declining contributions after a period of increasing contributions are found to invest more in capital expenditures, to have greater future accounting profits and future stock returns than firms that continue to increase pension plan contributions.
The study also provides evidence that, in addition to storing financial slack, pension plans may be viewed by management as one alternative in its portfolio of investment choices. That is, the returns earned on pension plan assets may be more attractive than other investment opportunities leading to more investment in (i.e. contributions to) pension plans.
Pension Plan Contributions, Free Cash Flows and Financial Slack
Pension plans have increasingly become an essential part of a firm’s overall financial management plan. This study examines the corporate pension funding decision and the (managerial) incentives that can effect that decision. Specifically, the paper examines the relationship between firms’ pension plan contributions, the level of their (excess) free cash flows and characteristics of subsequent investment decisions taken by the firms. The current study differs from prior studies in this area, because it focuses on annual cash contributions to the pension plan, instead of the funding status of the plan, i.e., the excess of plan assets over pension obligations. The current study also uses the pension plan contributions to distinguish Myers and Majluf (1984) firms from Jensen (1986) firms, by their motives for increasing pension plan contributions.
Firms, which have cash flows in excess of current reinvestment needs, should, theoretically, distribute these free cash flows to their shareholders by way of dividends and/or stock repurchases. In practice, however, firms often retain a portion or all of these free cash flows. The agency theory literature suggests that conflicts of interest between management and shareholders and/or information asymmetries can explain management’s motivation to hoard free cash flows and lead to differences in payout policies.
Background and Literature Review
Jensen's Free Cash Flow Hypothesis: Jensen (1986) argues that managers, in their own self-interest, seek to accumulate perquisites and as a firm becomes larger, more opportunities exist for managers to indulge their needs for pecuniary and non-pecuniary (power and prestige) benefits. Unless properly controlled, such behavior can lead to managers making inefficient expenditures by taking on less than optimal (i.e. below cost of capital) investments as they attempt to “grow” the firm. In Jensen's world, "good" managers are those who commit to dispose of excess cash flows by increasing dividends and instituting share repurchase programs. Bad managers, interested in increasing their perquisites, will retain cash in order to "grow" the firm by engaging in (unprofitable) takeover and/or expansion behavior1.
Myers and Majluf's Information Asymmetry Approach: Myers and Majluf (1984), on the other hand, suggest a more benign reason for management's reluctance to distribute excess free cash flows. They argue that management is motivated by their desire to "protect" current shareholders' interests relative to new shareholders. Due to information asymmetries, management possesses superior information relative to investors and/or creditors as to the desirability of (new) investment projects. As such, they are reluctant to finance these projects with the use of external funding for fear that any new securities issued would be underpriced. Therefore, in order to have internal funds available for investment use and avoid issuing underpriced securities (and cutting future dividends), managers build up “financial slack” by storing excess funds until they are needed.
Unlike the Jensen scenario, in the Myers-Majluf world, firms have future (positive NPV) investment projects. Thus, cash flows will be retained in a form where they are readily accessible; i.e. as cash or short-term financial assets2. Management would be willing to store the excess cash in these assets despite their low returns because of their ease of recoverability.
From an empirical perspective, both the Jensen and Myers and Majluf approaches suggest a similar behavior pattern; i.e. managers do not distribute free cash flows but rather invest/hoard them in alternative venues. Given the identical outcome, distinguishing the underlying motivation from the behavior pattern itself is not always feasible3. However, although managers of both Jensen-type and Myers-Majluf firms engage in the hoarding of future cash flows, the nature of the hoarding behavior as well as the characteristics of the two type of firms may be discernible (at least on an ex post basis). Thus, it would be possible to distinguish between Jensen and Myers-Majluf firms by
comparing the means used to hoard excess cash flows and
examining, ex post, the nature of the investments undertaken.
In this study, we examine pension plan contributions and argue that storing excess cash flows in pension plans is consistent with the Myers-Majluf view but not the Jensen approach. Pension plans can serve as an alternative and (with the additional benefit of their favorable tax treatment) perhaps superior place to store free cash flows and build financial slack.4 The firm can build financial slack by increasing contributions to the pension fund. Subsequently, as new projects are undertaken, the firm can draw upon these cash flows by either reducing future contributions directly (having previously “over” contributed), or altering actuarial assumptions5, thus reducing future required contributions. Additionally, in certain situations, the firm can access the funds by outright termination (or restructuring)6of the plan. Thus, pension plans, with their higher (after-tax) returns and relative ease of accessibility may be a superior place for Myers-Majluf type managers to store free cash flows.
For a Jensen type firm, however, contributions to pension plans do not grow the firm in a manner that would allow for greater pecuniary and non-pecuniary benefits for managers. In fact, by contributing to pension plans, managers, in a sense, are giving up control of the cash flows to the plan trustees, at least, insofar as enjoying the perquisites which could emanate from alternative uses (takeovers) of the funds.
Thus, our study examines the relationship between pension plan contributions and financial slack in two steps. First, we examine whether firms use pension plans to store excess cash flows. Finding such behavior would be consistent with Myers and Majluf's financial slack theory. Second, our study examines the implications of the Myers and Majluf's financial slack theory with respect to the expected behavior once hoarding stops. In the Myers-Majluf world, management has knowledge of future positive NPV projects. Thus, in time (assuming managers' information is superior), the firm should "switch" its behavior pattern by utilizing the financial slack previously accumulated as they take on (profitable) investments. As such, our study also examines the characteristics of firms which initially increase pension plan contributions and subsequently reverse their pattern of pension plan contributions to see whether these activities coincide (as predicted by Myers and Majluf) with increased (profitable) investment activity.
The use of pension plans to maintain financial slack was recently studied by Datta et al (1996). They postulated (and found) that managers with no ownership stake have more incentives than those with an ownership stake to maintain financial slack in the form of excess pension funding. Their study used the pension plan funding status; i.e. the difference between the plan assets and liabilities, to evaluate the degree to which managers build up financial slack. While funding status does indeed measure financial slack, it is a cumulative product of many factors (e.g. benefit obligations, actuarial assumptions, return earned on pension plan assets) that are not under management’s control. As such, it only indirectly measures currentactions taken by management to build up financial slack.
A more appropriate and direct measure of management behavior is the actual level of contributions made by the firm to the pension plan. This latter number is not currently7 provided explicitly in the financial statements and perhaps explains why Datta et al did not use that metric in their study. In this study, we use the actual level of contributions to measure management behavior. As demonstrated by Stickney (1995), White et al (1997), and Ballester et al (1998) (and described in the appendix to the paper) contribution levels are derivable from current accounting disclosures. Additionally, this paper examines the relationship of free cash flows and pension contributions using a longer and more recent time frame (1991-1996) as opposed to just the two years (1984 and 1987) studied by Datta et al.