Rauh, Joshua 2006, “Investment and Financing Constraints: Evidence from the Funding of Corporate Pension Plans,” The Journal of Finance 61 (1), 33-71.
Purpose: This paper evaluates the dependence of corporate investing on internal financial resources, using mandatory defined-benefit contributions (MCs) as an instrument for internal resources.
Findings: Capital expenditures fall by $0.60-$0.70 for $1 in mandatory pension contributions (MCs). The size of this coefficient is inversely related to credit rating, and is clearest among firms with observable financial constraints.
Motivation: The Miller-Modigliani model predicts that internal financing and external financing are perfect substitutes. However, with taxes, agency costs, and asymmetric information, internal financing may be less expensive. If so, then financial constraints should cause a drop in investment in capital or R&D (internal “financing”). Previous research observes firms’ response to exogenous funding shocks in small samples. This paper uses a much larger sample and an instrument for exogenous funding shocks. There is also a prior debate about the correlation between investment and cash flow, which this paper also addresses
- Data: 1,522 Compustat firms making DB contributions between 1990 and 1998 (8,030 firm years)
- Define cash flow as net income plus depreciation/amortization plus pension expense minus pension plan contribution from IRS forms 5500, since pension expense does not equal cash contribution
- Regress investment on cash flow and pension contribution variables
- DB pension funding may be endogenous to the firm’s investing opportunities
- Overcome endogeneity by looking at contributions to funds around the underfunded threshold
- Identifying assumption is that the relationship between unobserved investment opportunities and funding status is not the same as that between funding status and required contributions.
- Separate MCs into unexpected and “predictable”
- Use predictable, unexpected, and both contribution types as independent variables
- Divide the sample along several observable dimensions (firm age, credit rating, etc)
- Run the baseline regression (investment on MCs) on each division
- Look at whether MCs in one period affect investment in surrounding periods
- Look at whether MCs in constrained firms leads to higher investment in non-constrained firms
- Capital expenditures decrease $0.60 for $1 in MCs, which is 4-7x the magnitude of the effect of cash flows
- MCs also have marginal effects on acquisitions and dividends, but not on borrowing
- Unexpected and predictable MCs each account for equal portions of the variance in total MCs
- Firms that do not sponsor defined benefit (DB) plans take up about 12% of the capital expenditure slack that constrained DB firms leave on the table when MCs are high