Fama's evidence that the term premium on Treasury securities is not monotonically increasing is found to depend entirely on the behavior of bid-asked mean returns on 9- and 10-month bills, and only during the subperiod 8/64–12/72. When transactions costs, as reflected in the bid-asked spread, are taken into account, there is found to be no way to exploit this non-monotonicity. The anomalous behavior of the quotations is attributed to the Treasury's auctions of 9-month bills during the period 9/66–10/72. The hypothesis that the term premium is a monotonically increasing function of maturity remains unrefuted.

Journal Article Operational Aspects of the Siegel Paradox Get access J. Huston McCulloch J. Huston McCulloch Boston College Search for other works by this author on: Oxford Academic Google Scholar The Quarterly Journal of Economics, Volume 89, Issue 1, February 1975, Pages 170–172, https://doi.org/10.2307/1881720 Published: 01 February 1975

a recent article in this Review, Nissan Liviatan and David Levhari attempt to explain the phenomenon that In spite of recent marked inflationary trends the important capital markets have not developed a significant (private enterprise) market for [cost-of-living] linked (p. 366). They note that the familiar theoretical complete dominance of indexed bonds over nonindexed bonds holds unambiguously only in an economy in which there are no nominal assets. They demonstrate that borrowers may in theory insist on hedging the inflation risk on their nonindexed cash balances by issuing some nonindexed bonds at a discount that makes them attractive in spite of the risk they impose on lenders. Their model leads them to conclude that indexed bonds might not completely dominate nonindexed bonds. However, they do not claim that their model explains the phenomenon they set out to explain: Why is it that the free market, to use their expression, has come up with virtually no indexed bonds at all? Fortunately, a simple explanation for the nonexistence of indexed bonds in the United States was provided, two years

An example of a continuous time economy is given whose general equilibrium term structure of interest rates obeys the Expectations Hypothesis for continuously compounded interest rates and returns, contradicting the 1981 claim by Cox, Ingersoll, and Ross that such an economy is mathematically impossible. This example does not generate exploitable arbitrage opportunities of the type Cox, Ingersoll, and Ross claim must arise. The “Logarithmic Expectations Hypothesis,” as we call it, it therefore an acceptable benchmark from which to measure term premia in continuous time term structure modeling.