Comment on "Collusion in the Auction Market for Treasury Bills"

Abstract

M ICIIhAEL RIEIBER s thorough examination of the auction market for Treasury bills makes it clear that the casual empiricism imbedded in my remarks to the Joint Economic Committee was, as such empiricism so often is, wide of the mark. Clearly, I misinterpreted the Treasury's published data on auction prices, overestimated the extent to which the market is limited to dealers, and gave insufficient attention to the information provided by the secondary market for bills as a basis for rational bidding in the auction. I am grateful to Rieber for setting m-e straight on these matters. At the same time, Rieber's comments (like those of Brimmer and Goldstein) leave entirely untouched my main theoretical point and my policy proposal: that the pricediscrimination method of bidding used by the Treasury reduces participation, changes the demand schedule as it appears to the Treasury, probably raises the costs for the Treasury, and is inferior to an alternative method of bidding under which all purchasers would pay the same price. These defects have always seemed to me far more important for long-term securities than for shortterm securities, and my major reason for urging a change in method of bidding has been not to save money on bills but rather to facilitate the adoption of auctioning as the exclusive method of selling all Treasury securities, regardless of maturity. I have argued that alleged "failures" of the few experiments that have been made in auctioning long-term securities have reflected the use of the method that is now used to auction bills rather than any inherent difficulty in using auctioning to sell long-term Treasury securities. The offering method now used to sell long-term securities, involving as it does a bias toward unduly generous terms in order to assure "success" of an issue and hence generally a "free ride" for initial subscribers, very likely makes the Treasury's cost for long-term securities appreciably higher than it would be under either alternative method of bidding. Equally important, it introduces unnecessary discontinuity and uncertainty into the market. Rieber shows that the method of auctioning has an even more trivial effect for bills than I had supposed. Rieber calculates in his final paragraph that, if the demand curve as viewed by the Treasury is assumed to be the same under either method of bidding, what he calls "the Treasury's returns from discrimination" equals a sum that averages 0.01 per cent of the face value of the amounts auctioned; or, in perhaps more meaningful terms, less than 1 per cent of total interest payments. But surely, on the matter of principle, Rieber is wrong in asserting that his arithmetic calculation gives "the Treasury's returns from discrimination" and that "had the Friedman proposal been in effect these profits would not have been realized"-as Rieber himself implicitly recognizes in the last half of the final sentence of his paper. The method of bidding may not limit the volume of bidding much, but it does limit it some; it may not alter the bids submitted by those who enter bids much, but it does alter them some; it may not add much to the returns of dealers who have superior ability to bid and who buy to resell, but it does add some. In the circumstances of the bill market, all of these effects are clearly much smaller than I initially thought them to be, but the direction is what I asserted it to be. The demand curve faced by the