Government Securities--the Scandal of '91
-by Kate Reynolds
One in a Series of Articles from Agorics, Inc.
Each year the U.S. Treasury Department offers several trillion dollars of debt. Thirteen and twenty-six week maturities are auctioned weekly on Mondays; longer maturities are offered several times each year. Each week the Treasury Department announces the amount of debt it will auction off the next week. The market can be thought of as a "forward" market that serves both to allocate and to evaluate. Auction results are considered an indication of the way institutions regard the fiscal policies of Congress and the President and their own beliefs about the marketability of the securities.It should be mentioned that treasuries are sold in terms of yield, but for simplification the word "price" will be used here. In general, two types of bids are allowed. Competitive bids are for large markets such as money market banks, dealers, institutions. Non-competitive bids are the small scale bidders open to the public.
There are thirty-nine "primary dealers", institutions allowed to participate in the competitive bidding. They submit bids specifying the price they will pay and the number of securities desired. Up until the recent past the U.S. Treasury used the discriminatory sealed-bid auction (first price) which means that winners paid different prices with some people paying a great deal less than others. The came the great Salomon Brothers disaster.
In August of 1991, Salomon Brothers, Inc. admitted that it had seriously violated U.S. Treasury auction rules by submitting fraudulent bids a year earlier. (They ended up paying a fine of $290 million.) Also in 1991 there were two other successful efforts to manipulate the market. A huge uproar resulted and hearings were held to determine whether or not to change the type of auction used to sell securities. This attracted considerable attention as one might expect when billions are at stake.
As early as 1959 Milton Friedman spoke before the Joint Economic Committee urging a switch to the uniform-price auction system (Vickrey--all new issues awarded at the same amount which is equal to the highest losing bid). In fact, the Treasury did experiment briefly in the 1970's but changed back when results proved inconclusive. Many believe the experiment was abandoned too quickly, and the Salomon Brothers disaster prompted a reopening of the matter. Milton Friedman spoke and wrote again on the same subject in 1991.[Milton Friedman]
Merton Miller (winner of the 1990 Nobel Prize in economics) also argued in 1992 in favor of experimenting again with the uniform, second-price auction. Other people appeared before Congress urging a switch to the Dutch auction (in financial markets the term "Dutch" means the Vickrey auction.) Chari and Weber [Chari] argued that a switch to a uniform-price auction would be a good change because participants would be less inclined to acquire information about other bidders because there would be less market manipulation. The discriminatory format strongly encourages bidders to learn about their competitor's valuations.
A number of arguments make the Vickrey auction more attractive. A Vickrey auction tends to mute "winner's curse" and therefore increases revenue to the Treasury because bidders are more aggressive in that format. Also, the Vickrey method is considered less prone to collusion (one group cornering the market such as in the Salomon Brothers case). In general, the Vickrey auction is a simpler auction type, requires less bid preparation time, is less costly, and allows for greater bidder participation.
An argument against the experiment is that it is believed that interest rates have less to do with how securities are sold than with demand. Another compelling argument against the Vickrey auction is that a market manipulator could place bids for a substantial portion of some issue at a price considerably higher than the expected market consensus. This would ensure significant awards but the winner would pay only the second-highest price. Of course if more than one bidder attempts this, someone would pay dearly.
It was agreed to experiment for a year with the new technique. The government hoped that pension funds and other large investors would bid. As of this writing the results have been inconclusive and the experiment continues.
Thomas Vogel [Vogel] reported that the government possibly saved money on two-year notes but may have lost money on the five-year notes. More people did participate as hoped.
[Previous] [Tech Library] [Home] [Next]