This paper examines the formation of pension plans from a corporate finance perspective. The theoretical underpinnings for selecting a definedbenefit or definedcontribution plan are discussed and used to form empirically testable hypotheses. Linear probability and logit models are used to identify corporate financial characteristics that affect the likelihood of forming a definedbenefit or definedcontribution plan. The results strongly indicate that firms with high degrees of debt and intangible assets are least likely to form definedbenefit plans in a postreversion situation, while firm size enhances the probability of forming definedbenefit plans. The growth in private retirement plans over the past quarter century has made pension fund management a critical concern for many financial managers. The total amount of assets in private pension plans amounted to approximately 150 billion in 1970, while this figure was about 2 trillion in 1989. A corresponding trend to this growth has been an acceleration in the formation of definedcontribution plans relative to definedbenefit plans. In 1975 about 29 percent of all plans were definedcontribution plans, and 71 percent were definedbenefit plans. In contrast, definedcontribution plans comprised 55 percent of all plans in 1988, while 45 percent were definedbenefit plans.1 Gustman and Steinmeier 1987 suggest that the shift to definedcontribution plans in recent years may be attributable to shifts in jobs in the economy away from the manufacturing sector and toward the service sector. Furthermore, the role of unions, firm size, and administrative costs have also been sighted as factors which partially explain the economy wide shift toward definedcontribution plans see Gustman and Steinmeier 1989, Clark and McDermed 1990, and Kruse 1991. In this paper, we address the pension choice by examining the formation of individual plans from a corporate finance perspective. Specifically, we examine the pension choice issue when firms are faced with making this decision after the termination of an overfunded definedbenefit plan. The remainder of this paper is organized as follows. Section I discusses the possible motives for selecting one plan over the other, and develops testable hypotheses. The data and methodology are discussed in section II, while section III presents the empirical results. Section IV summarizes and concludes the paper.
Managerial Finance – Emerald Publishing
Published: Aug 1, 1997