The four senses, briefly, so a reader who arrived from the wrong one can leave. The present-value sense is the subject of this page: an assumed annual rate applied to future dollars to state them in today's money. The Federal Reserve sense is a bank borrowing rate, the primary credit rate charged at a Reserve Bank's discount window; 12 U.S.C. 357 is its statutory authority and speaks of "rates of discount to be charged by the Federal reserve bank for each class of paper," established by each Reserve Bank "subject to review and determination of the Board of Governors of the Federal Reserve System." It has nothing to do with converting future dollars into present ones, and the federal funds rate page carries it. The Treasury bill sense is a price convention, and it too is codified, in the Treasury regulation governing its own auctions: 31 CFR 356.2 provides that "discount rate means a rate of return, on an annual basis, on bills held until they mature," that it "is expressed in percentage terms and based on a 360-day year," and that it "is also referred to as the 'bank discount rate.'" The regulation's own price formula applies that rate to face value, so the figure is a pricing basis rather than a present-value rate and is not directly comparable to one; the treasury bill page carries the arithmetic. The actuarial sense is a present-value rate with an institutional consequence, described below.
A real rate belongs with real amounts and a nominal rate with nominal amounts, and the pairing is not optional. Circular A-94 states the rule directly at section 8.a: "A real discount rate that has been adjusted to eliminate the effect of expected inflation should be used to discount constant-dollar or real benefits and costs," while "a nominal discount rate that reflects expected inflation should be used to discount nominal benefits and costs. Market interest rates are nominal interest rates in this sense." That last sentence is the trap: quoted market rates already contain expected inflation, so applying one to a figure stated in today's purchasing power removes inflation twice and understates the answer. The circular adds the practical approximation: "a real discount rate can be approximated by subtracting expected inflation from a nominal interest rate."
Nobody publishes the correct rate, and the choice moves the answer more than most people expect. A Federal Reserve Board economist made exactly this point in a 2014 FEDS Note on equity valuations, building a simple constant-growth present-value framework and concluding that "even a fairly modest decline in discount rates of 1 percentage point can have a marked effect on valuation ratios," so an elevated valuation "may not necessarily be an indicator of an overvalued market, but instead a reflection of a change in discount rates." The general reading holds outside equities: over long horizons the rate, not the cash flows, usually dominates the result. Which is why a present value or a net present value quoted as a single number, without the rate beside it, cannot be evaluated.
The actuarial sense is where the choice has the largest institutional consequences. A pension plan's promised benefits are future payments, so stating what they cost today requires a discount rate, and the same promise looks smaller at a higher rate and larger at a lower one. Because the reported size of the obligation drives contributions and funded-status disclosures, the rate is prescribed rather than left open in the places it matters most: the Pension Benefit Guaranty Corporation publishes the interest assumptions used to value annuities in terminating single-employer plans, alongside the mortality tables that go with them, and pension liability figures computed at different rates are not comparable even for the same plan in the same year. The relevant instinct for a reader meeting a funded-status headline is to ask what rate produced it before asking whether the number is alarming.
What actually goes into a household's choice. Two considerations, and both are judgments. The first is opportunity cost: what the money would realistically earn in its next-best use, which sets a floor. The second is certainty: a guaranteed payment deserves a rate near a safe interest rate, while an uncertain one deserves a higher rate, which is the arithmetic way of saying a shaky promise is worth less. The discipline that follows is to run the calculation at more than one rate and see whether the conclusion survives, rather than to hunt for the one right rate, which does not exist.
