The Problem of Discounting with an After-Tax Rate of Return in Cases of Personal Injury or Wrongful Death
Marlin, Matthew, Journal of Legal Economics
Forensic experts are often called upon to estimate the value of a lump sum payment that will replace, as closely as possible, a stream of income that has been lost as the result of a tort or some other form of negligence. Sometimes federal or state statutes require that such estimates be made on an after-tax basis using an after-tax rate of return. Three common procedures for making such estimates are discussed and compared. The conclusion is that neither calculated after-tax discount rates nor municipal bonds provide a satisfactory estimate of the appropriate rate of return. The most efficient method of dealing with after-tax requirements is essentially to ignore them, use a taxable rate of return, and then "gross up" the lump sun to cover any tax liability.
Forensic economists are often called upon to estimate the value of lost future compensation in cases of personal injury and wrongful death. Such an estimate involves projecting future lost earnings, fringe benefits, etc., and then discounting these future values back to the present with some discount rate. The result is a lump sum award that if invested today will generate an income stream that will come as close as possible to replacing the value of what has been lost due to the tort.
Two Supreme Court rulings have added to the difficulty in estimating the amount of the necessary award. In Norfolk & Western Railway Co. v. Liepelt (1980) the Court ruled that the award must be based on after-tax rather than gross income:
The amount of money that a wage earner is able to contribute to the support of his family is unquestionably affected by the amount of the tax he must pay to the Federal government. It is his aftertax income, rather than his gross income after taxes, that provide the only realistic measure of his ability to support his family.
This raises a problem in that the income stream generated by a lump-sum award also may be subject to taxation, and the income steam must therefore be adjusted in some way to ensure that the plaintiff does not pay taxes on an award that already reflects after-tax income. The Court addressed this problem in Jones & Laughlin Steel Corp. v. Pfeifer (1983) where it went on to add that the discount rate must represent the after-tax rate of return:
The discount rate should be based on the rate of interest that would be earned on "the best and safest investments"... the discount rate should not reflect the market's premium for investors who are willing to accept some risk of default. ... [T]he lost stream of income should be estimated in after-tax terms, the discount rate should also represent the after-tax rate of return to the injured worker, [emphasis added by author]
Brush and Breeden (1994) point out that not all courts, especially state courts, are bound by the above rulings and that the impact of the rulings falls primarily in personal injury and/or wrongful death tort cases tried under federal statutes. Some of the time, therefore, the issue of the after-tax rate of return may be moot since many state courts ignore the issue of taxation and economic loss estimates are calculated on a pre-tax basis. Nonetheless, the scope of the coverage is certainly broad enough that there should be a standard, replicable method for calculating damages in terms of after-tax dollars using an after-tax discount rate. Surprisingly, such a standard process is lacking.
The Basic Problem
There are two issues that must be addressed to comply with liepelt and Pfeifer. First is the calculation of after-tax wages (1 - t)Y^sub o^, and second is the calculation of an after-tax rate of return (1 - t)d. While the first issue has been adequately addressed in the literature, the second has received little attention since it was raised by Vernon (1985). In that study, Vernon noted that the procedure for estimating the after-tax award can be "cumbersome," but ignoring the effect of progressive taxation can result in significant errors. …