VARIANCE BOUND TESTS OF BOND MARKET-EFFICIENCY

In this paper we examine variance bound tests of the joint hypothesis that (1) bond markets are efficient and (2) the term structure is determined by the expectations hypothesis. Both the Singleton and Shiller tests are shown to be seriously biased toward rejecting the joint hypothesis in finite samples. Flavin's test is unbiased but has a very high variance leading to many false rejections of the joint hypothesis. When corrected as suggested by Flavin, Shiller's test is unbiased and has a relatively low variance. Unfortunately, it is also sensitive to measurement error.