The Pfeifer decision, risk and damage awards: An extended response to Albrecht
and WoodJournal of Legal Economics; Florence; Winter 1997/1998; Thomas R Ireland;
Volume:
7
Issue:
3
Start Page:
23-34
ISSN:
10543023
Subject Terms:
Supreme Court decisions
Settlements & damages
Inflation
Economic theory
Treasury bonds
Discount rates
Uncertainty
Classification Codes:
9190: US
4330: Litigation
1130: Economic theory
3400: Investment analysis
Geographic Names:
USAbstract:
Albright and Wood (1997) have argued that the Jones & Laughlin Corp. v. Pfeifer (1983)
decision of the US Supreme Court creates an entitlement that a damage award should not be
subject to inflation risk. They argued that discount rates should be based on short-term
government securities, rather than long-term securities. This paper suggests that: 1.
Albrecht and Wood confuse the meaning of inflation risk with inflation uncertainty, 2. nominal
interest rates already fully discount inflation risk but cannot fully eliminate loss due to
unexpected inflation, 3. the new Treasury Inflation-Indexed Securities bonds provide a
mechanism that is free of inflation uncertainty, 4. the Jones & Laughlin Corp. v. Pfeifer
decision does not advocate a reward that is free of inflation uncertainty, and 5. short-term
bonds carry liquidity premiums that make them unsuitable for discount rates in damage
awards even though the Pfeifer decision does permit such usage.Full Text:
Copyright American Academy of Economic and Financial Experts Winter 1997/1998
Gary R. Albrecht and John H. Wood (1997) argue in "Risk and damage Awards: Short-Term Bonds Vs.
Long-Term Bonds" that the Jones & Laughlin Corp. v. Pfeifer (1983) decision of the United State
Supreme Court creates an entitlement that a damage award should not be subject to "inflation risk." On
the basis of this presumed entitlement, they argue that discount rates should be based on short-term
government securities, rather than long-term securities. This comment suggests that (1) Albrecht and
Wood confuse the meaning of inflation risk with inflation uncertainty, (2) nominal interest rates already fully
discount inflation risk but cannot fully eliminate loss due to unexpected inflation, (3) the new Treasury
Inflation-Indexed Securities (TIIS) bonds provide a mechanism that is free of inflation uncertainty, (4) the
Jones & Laughlin Corp. v. Pfeifer decision does not advocate a reward that is free of inflation uncertainty,
and (5) short-term bonds carry liquidity premiums that make them unsuitable for discount rates in damage
awards even though the Pfeifer decision does permit such usage.lInflation Risk vs. Inflation Uncertainty
Albrecht and Wood do not correctly understand the meaning of inflation risk, which they confuse with
inflation uncertainty.l Inflation Risk is the known risk that inflation may occur in the future. At any given
time, borrowers and lenders anticipate that inflation (or deflation) will occur in the future. What is known is
a probability distribution of possible future rates of inflation (or deflation), with a central tendency that can
be expressed as the expected rate of inflation. The actual frequency distribution may vary by market
participant, but the market as a whole develops an inflation premium that is incorporated in existing
nominal interest rates. This is the meaning of the Fisher equation, and is a general economic theory.2 Thus,
nominal interest rates already fully compensate for the risk of inflation itself.What is not included in the Fisher equation or in nominal interest rates is the premium for inflation
uncertainty. Risk is the appropriate term for all known information about future possibilities. Uncertainty
refers to what is not known that might affect what is known about future possibilities. This is Frank
Knight's distinction between risk and uncertainty applied to interest rates on financial securities. At any
given time, there is a known frequency distribution of possible future rates of inflation. That frequency
distribution will be summarized into an expected rate of inflation and compensated by a risk premium that
is equal to the expected rate of inflation plus some compensation for the variance in the rate. To illustrate
this point, suppose that investors generally presumed that the expected rate of inflation over a particular
period to maturity is 2%. Suppose further that the real rate of interest is 3% over that period. Finally,
suppose that the investors rate the probabilities of outcomes as a 25% chance that there will be 1%
inflation, a 50% chance that there will be 2% inflation, and a 25% chance that the rate of inflation will be
3%. With this simplified frequency distribution of expected outcomes, the expected rate of inflation is 2%.
The nominal interest rate will include 3% for the real interest rate and 2% as a premium for the expected
rate of inflation. It will also include a small premium for the risk caused by the known variance in the
expected rate of inflation. This premium has some positive value. Assume that the premium for this
variance is one fourth of a percent.2 Thus, the interest rate at this point should be 5.25%.But even this is not the whole story because there is also risk that the existing frequency distribution will
turn out not to be the correct frequency distribution of risk. This is what is meant by inflation uncertainty.
There is some risk that the known risks will be rendered inaccurate by unknown factors. The fact that they
cannot be known in the present means that no specific premium will fully eliminate this risk. Risk refers to
known variance. Uncertainty refers to unknown sources of possible variance. It is important that unknown
factors can confer both benefits and costs, with almost equal probability. It is almost equally likely that
new information will result in increasing a damage award as that it will reduce the value of that award. Still,
uncertainty reduces the ability to plan effectively and is typically also compensated by a premium. Thus,
we have one premium that is equal to the expected rate of inflation, another premium that compensates for
the known variance in the expected rate of inflation, and still another premium that compensates for the
unexpected variance in the expected rate of inflation. Nominal interest rates include all three types of
premia. If another quarter of a percent is added, the interest rate is now 5.5%. But because no premium
can preclude results that cannot be known, all of these premia taken together cannot preclude the remote
possibility that the damage award fund may not hypothetically be able to make all scheduled payments.2That is the kernel of truth at the core of the Albrecht and Wood argument. With a portfolio of three-month
Treasury bills that were reinvested each three months, the inflation risk would be limited to losses during
three month intervals. If sources of inflation uncertainty shifted the frequency distribution implicit in the
expected rate of inflation, this would be captured after a period of no longer than the three months
between reinvestment periods. Inflation losses could not be more than one fourth of the annual rate of
inflation during any one quarter. Nor, of course, could inflation gains be any greater than one fourth of the
annual rate of inflation during any one quarter. However, if the worker chose to invest his award in any
other type of portfolio, this presumed advantage of minimum risk from inflation uncertainty would
disappear. It is generally a valid argument that what a worker does with an award after it is received is not
an issue. But in this case, the only advantage of three-month Treasury bills is that they can be used to
eliminate a special, and quite small, risk of inflation uncertainty. If the worker chooses to invest in a
portfolio of higher yield-higher risk assets, that advantage is given up. Thus, this argument directly depends
on the actual use of threemonth Treasury bills.In summary, inflation risk is already compensated by a premium that exists in all nominal interest rates,
both short-term and long-term. The risk implied by variance in the known frequency distribution of
possible inflationary outcomes is also compensated by a premium that exists in all nominal interest rates,
both short-term and long-term. And the market offers a premium for the uncertainty that the known
variance will be correct. But the market cannot provide absolute certainty that a given fund invested in any
portfolio of existing financial securities will turn out to be adequate to make all scheduled payments,
defined in terms of real purchasing power. Only government can do that, and the federal government has
taken steps in that direction. The instrument involved is TIIS bonds, which is the subject of the next
section. Other than TIIS bonds, however, three-month Treasury bills do reduce potential losses due to
inflation uncertainty to the smallest possible level. But they only do that if the loss award is actually
invested in three-month Treasury bills. If the award recipient invests a loss award in any other type of
asset with longer terms to maturity, even the safest possible longer term bonds, this supposed advantage
of three-month Treasury bills is lost.TIIS Bonds Compensate Inflation Uncertainty More Effectively Than T-Bills
It turns out that the correct premia for inflation uncertainty are extremely small and can be determined from
the commercial marketplace in a quite easy manner. Beginning in January 1996, the U.S. Treasury began
issuing what were then called TIPS (Treasury Inflation Protected Securities), but are now known as TIIS
(Treasury Inflation Indexed Securities). Since these rates are designed to eliminate the impact of inflation
uncertainty by being payable on a CPI adjusted basis, TIIS securities can be compared with other U.S.
Treasury securities of similar maturities, adjusted for the current expected rate of inflation. For example, a
10-year TIIS bond can be compared with a regular 10-year U.S. Treasury bond. Since the TIIS security
is defined in real CPI adjusted terms, the rate will be smaller than the 10-year Treasury bond that is not
inflation indexed. In general, the rates should differ by the existing expected rate of inflation over the next
10 years, so that amount should be subtracted from the rate offered on the nonindexed 10-year bond. The
differences between the two types of 10-year Treasury bonds left at that point should be based on special
tax complications with TIIS bonds and inflation uncertainty. The tax consequences have been estimated as
requiring a market premium of perhaps a fourth of a percent for the rate on the TIIS bonds. After this
adjustment is also made in the comparative rates, the remaining differences are very small (in the range of
no more than a half of a percent and probably less than one fourth of a percent). Thus, it is fair to
conclude that inflation-uncertainty premia are quite small, regardless of measurement difficulties created by
special tax features of TIIS securities (Ireland 1997).TIIS bonds provide more complete protection against inflation uncertainty than do three-month Treasury
bills, which have some risk during the three-month intervals between reinvestments. After the fact, any
losses in purchasing power due to inflation are fully compensated by additions to the eventual principle
value of the bonds. To be sure, TIIS bonds have undesirable tax features, but they are unlikely to be used
by damage award recipients for reasons similar to why three-month Treasury bills are not likely to be
used. However, in the hypothetical frame of reference one must use to even talk about eliminating inflation
uncertainty as a factor, TIIS bonds will do the same job three-month Treasury bills will do, only more
completely because inflation losses during three-month intervals-will be eliminated.The Pfeifer Decision Does Not Require Compensation for Inflation Uncertainty
Albrecht and Wood argue that the 1983 U. S. Supreme Court decision in Jones & Laughlin v. Pfeifer
mandates elimination of the risk imposed by inflation uncertainty. This is simply not true. However, it is a
decision that is often cited but little read by forensic economists. The purpose of the Pfeifer decision was
to provide a general framework within which an economic damage analysis for a personal injury should be
conducted. In most respects, economists who have been worried about the mandates of the Daubert
decision can take a great deal of comfort from the fact that the United States Supreme Court spoke so
cogently about the nature of damage estimates by economists. Note especially the commentary of the
Court about the predictability of inflation (at 2555):First, by its very nature the calculation of an award for damages must be a rough approximation. Because
the lost stream can never be predicted with complete confidence, any lump sum represents on a "rough
and ready" effort to put the plaintiff in the position he would have been in had he not been injured. Second,
sustained price inflation can make the award substantially less precise. Inflation's current magnitude and
unpredictability create a substantial risk that the damages award will have little relation to the lost wages it
purports to replace. Third, the question of lost earnings can arise in many different contexts. In some
sectors of the economy, it is far easier to assemble evidence of an individual's most likely career path than
in others.While the impact of inflationary pressures in 1983 was much greater than it is today, it seems clear that the
Supreme Court did not believe that inflation variance could be eliminated. Further, one cannot find any
statement in the Pfeifer decision that requires that an award must provide certainty in the purchasing power
of monies provided in a lost earnings stream. Abrecht and Wood cite Bryan and Linke (1988) as authority
for their claim that Pfeifer implies that inflation "risk" should be minimized. Bryan and Linke were wrong as
well. Pfeifer carefully avoided such a pronouncement, making it clear that either short-term or mixed-term
portfolios could be used to develop discount rate estimates.Pfeifer was very firm on only two issues. First, it explicitly reaffirmed Norfolk & Western R. Co. v. Piepelt
(1980), which requires that taxes be removed from estimates of lost earnings of injured workers. Second,
Pfeifer required that the discount rate or rates that are used must be free of (default) risk. The claim made
by Bryan and Linke that Pfeifer meant that inflation risk should be minimized is based on a presumption
that the Court included inflation risk when it made the following statement (at 2550): "Once it is assumed
that the injured worker would definitely have worked for a specific term of years, he is entitled to a
risk-free stream of future income to replace his lost wages; therefore the discount rate should not reflect
the market's premium for investors who are willing to accept some risk of default. " (Italics added for
emphasis.)The Court clearly does refer to a discount rate that is free of default-risk in this passage, which qualifies
what is meant by a "riskfree stream of future income." Read in its entirety, there is no implication in this
passage that the injured worker is entitled to certainty in purchasing power, but only that the discount rate
should be free of risk of default. The need for a discount rate that is free of default-risk is the point that
Albrecht (1997) makes in his separate comment in this journal and is correct. The Pfeifer Court does not
ignore inflation, but discusses the impact of inflation in the various methods for analyzing discount rates that
it specifically allows, saying (at 2551) that: "Price inflation - or more precisely anticipated price inflation -
certainly affects market rates of return. If a lender knows that his loan is to be repaid a year later with
dollars that are less valuable than those he has advanced, he will charge an interest rate that is high enough
both to compensate him for the temporary use of the loan proceeds and also to make up for their
shrinkage in value" [fn 23].Footnote 23 elaborates the Pfeifer Court's analysis of inflation premia for the expected rate of inflation, the
variance in the frequency distribution in that rate and inflation uncertainty, as follows (also at 2551):The effect of price inflation on the discount rate may be less speculative that its effect on the lost stream of
future income. The latter effect always requires a prediction of the future, for the existence of a contractual
cost-ofliving adjustment gives no guidance about how big that adjustment will be in some future year.
However, whether the discount rate actually turns on predictions of the future depends on how it is
assumed that the worker will invest his award.On the other hand, it might be assumed that at the time of the award the worker will invest in a mixture of
safe short-term, medium-term, and longterm bonds, with one scheduled to mature each year of his
expected worklife. In that event, by purchasing bonds immediately after judgement, the worker can be
ensured whatever stream of nominal income is predicted. Since all relevant effects of inflation on the
market interest rate will have occurred at that time, future changes in the rate of price inflation will have no
effect on the stream of income he receives...On the other hand, it might be assumed that the worker will invest exclusively in safe short-term notes,
reinvesting them at the new market rate whenever they mature. Future market rates would be quite
important to such a worker. Predictions of what they will be would therefore also be relevant to the choice
of an appropriate discount rate...We perceive no intrinsic reason to prefer one assumption over the other...
In light of the foregoing passages, it seems clear that the Court was aware of the impact of inflation but did
not intend to imply a requirement that inflation-risk must be minimized, as assumed by Bryan and Linke
and cited by Albrecht and Wood. The Court fully understood the issues posed by inflation and
intentionally avoided making any prescriptions with respect to how this issue should be treated. The Pfeifer
Court allows the short-term bonds approach recommended by Albrecht and Wood but carefully avoids
recommending this approach over a mixed term approach. Thus, Pfeifer is not an authority for or against
the approach recommended by Albrecht and Wood.Liquidity Premia Make Short-Term Bond Rates Unsuitable as Discount Rates
The final point of this comment is that the short-term bond approach recommended by Albrecht and
Wood, although permissible under Pfeifer, is very inaccurate for an entirely different reason. On the
surface, it would appear that portfolios of TIIS bonds and threemonth U.S. Treasury bills would both
reduce the degree of inflation uncertainty to about the same degree, though with a slight superiority for the
TIIS bonds. A portfolio that was turned over every three months could be reinvested at the higher or
lower rates that would prevail at those very short maturity dates. This would prevent significant losses of
purchasing power due to inflation. However, if this were all that was going on, one would expect the rates
on threemonth Treasury bills to reasonably match the rates on TIIS bonds, perhaps with small differentials
(one fourth of a point or less) because of the special tax features of the TIIS bonds. Because of the
unattractive tax features (Ireland 1997), we would expect the rate on TIIS bonds to be slightly higher than
the rate on three-month Treasury bills after reduction for expected inflation.In fact, however, TIIS bonds have rates in the range of 3.5% in real terms, while the real rate of the return
on three-month U.S. treasury Bills is typically more than a full percentage point below that. The reason for
this disparity is that the rate of return on threemonth U.S. Treasury bills contains a significant negative
liquidity premium. Three-month Treasury bills are not used in long-term portfolios at all, except as
temporary stores of value. The rates are very low because the purchasing power will become available
within three months without the kinds of risks that exist with prematurity market sales. They are only held
in financial portfolios when portfolio managers believe that current rates will change favorably in the future.
And in those instances, fund managers are willing to accept lower than market rates of interest because of
their liquidity advantages.Although both TIIS bonds have even slightly less inflation uncertainty than three-month Treasury bills, they
do not have the liquidity advantages of three-month Treasury bills. As with all other longer term securities,
their prices are free to fluctuate more freely on the basis of long-term changes in real interest rates. Thus,
TIIS bonds must offer substantially higher rates of return than three-month Treasury bills. This poses no
disadvantage for TIIS bonds in terms of replacing lost earnings. Since the future earnings of workers that
are being replaced in damage awards would have had even worse liquidity characteristics than those in
TIIS bonds, there is no reason to try to maximize liquidity in a damage award. The lack of liquidity for
future returns on human capital investments are a profound feature of both labor markets and financial
markets. In other words, the absence of price fluctuations for three-month Treasury bills is an important
advantage of three-month Treasury bills over TIIS bonds. The higher returns on TIIS bonds are
substantially related to the lack of these price advantages by TIIS Bonds. But because these advantages
did not exist with the lost earnings that are being replaced, there is no basis in either legal requirements or
in economic theory for arguing that the loss replacement fund should have the liquidity advantages
available in three-month Treasury bills.Conclusion
On the basis of the Pfeifer decision, there is no requirement that a loss award should be invested in such a
way that there is no risk from inflation uncertainty. The odds are relatively even that the loss award would
be benefited or hurt by unexpected changes in the rate of inflation. This means that there is no requirement
that even TIIS bonds should be used to eliminate the risks posed by unnecessary inflation. However,
market evidence suggests that the costs of eliminating these risks by use of TIIS bonds are extremely
small, certainly no more than half of a percent in yield. As such, returns on TIIS bonds in the range of
3.5% would imply real rates of interest in the range of 2.75 to 3.25%. (A 2.75% rate is calculated by
subtracting 0.25% for tax consequences and 0.5% for inflation uncertainty from the starting TIIS rate of
3.5%, whereas a 3.25% rate is calculated by subtracting 0.25% for both tax consequences and inflation
uncertainty from the TIIS rate.) This is not what one finds with rates of interest on short-term government
securities. While the courts may allow use of such rates, economic analysis indicates that this is very
shortsighted.[Footnote]
Endnotes[Footnote]
1. As used here, the terms risk and uncertainty have the meanings ascribed to them by Frank Knight in 1921 in
Risk, Uncertainty and Profit.[Footnote]
2. For this to be relevant, one must implicitly assume that the damage award recipient would invest the damage
award in the manner implied by an economist's discount rate. Thus, if the discount rate is based on shortterm
securities, the entire damage award must be invested in short-term securities, which are continuously reinvested
as short-term securities mature. The award recipient would extract scheduled payments equal in purchasing power
to amounts projected by the economist and the fund would be said to be adequate if the damage award fund was
sufficient to make all necessary payments to replace the lost purchasing power projected by the economist.[Reference]
References[Reference]
Albrecht, Gary R. 1997. "The Need to Use Risk Free Discount Rates: Comment" Journal of Legal Economics. 7(1):
92-95.[Reference]
, and John H. Wood. 1997. "Risk and Damage Awards: ShortTerm Bonds Vs. Long-Term Bonds." Journal of Legal
Economics. 7(1): 48-58.[Reference]
Bryan, William R., and Charles M. Linke. 1988. "Estimating Present Value of Future Earnings: Experience with
Dedicated Porfolios." Journal of Risk and Insurance. 55:273-86.[Reference]
Ireland, Thomas R. 1997. "Forensic Implications of Inflation-Adjusted Bonds." Litigation Economics Digest. 1(2):
92-102[Reference]
Jones & Laughlin Steel Corp. v. Pfeifer, 103 S.Ct. 2541 (1983)[Reference]
Knight, Frank H. 1965. Risk, Uncertainty and Profit. New York: Harper Torchbacks, Harper & Row. Original
publication in 1921 by Hart, Shaffner & Marx.[Reference]
Norfolk & Western R. Co. v. Liepelt, 444 U.S. 490, 100 S.Ct. 755, 62 L.Ed. 689 (1980)[Author note]
*Thomas R. Ireland, Professor of Economics, University of Missouri, St. Louis, Missouri