Lancaster University: LUMS study simplifies the Treasury security issuance process
The Treasury security issuance process can be greatly simplified, as the choice of auction rule creates essentially no revenue difference, new research involving Lancaster University Management School suggests.
The new study, forthcoming in the American Economic Journal: Microeconomics, investigates whether Treasury securities – which are sold by nations to cover their debts – should be sold through discriminatory-price auctions (where traders pay as they bid) or uniform-price auctions (where all winners obtain the same interest rate). Researchers say the findings indicate that Treasuries across the world could shift their focus away from auction rules and towards other important aspects of the financial market.
One of the authors, Professor Dakshina De Silva, from Lancaster University Management School, said: “There are two commonly used auction rules across Treasuries worldwide – discriminatory and uniform pricing. The Nobel laureate Milton Friedman (1912-2006) predicted that uniform pricing could save a nation’s interest rate by 0.5% – a considerable interest rate difference, which could drastically reduce the taxpayers’ burden. However, Friedman passed away before providing a concrete answer to this literal million-dollar question. So, the main question is, ‘what type of auctions to use?’ This question has occupied Treasuries and researchers for about fifty years.”
The study is a collaboration between academics from Lancaster University Management School, SKEMA Business School, Nanjing Audit University, and the University of Glasgow. It examines the auction outcomes of more than 300 Treasury securities from two Chinese Government security issuers, the Chinese Development Bank and the Export-Import Bank.
Surprisingly, the researchers found no significant statistical or economic difference in either yield or revenue between the two popular auction formats. Looking at the yield rates of securities sold through discriminatory and uniform auctions during a three-year market experiment by the two Chinese Government banks, the researchers found just a 0.001 to 0.008 percent difference between discriminatory- and uniform-price auctions. The difference in revenue was also marginal. If the banks had issued their bonds in an alternative auction format to the one used, the researchers estimated just a 0.00041 to 0.00054 percent difference in Chinese Government expenditure.
Professor Dakshina De Silva continued: “For decades, researchers have tried to understand which auction format generates the most attractive yield and revenue.
“We discovered that bidders do not appear to favour one auction rule over another. Thus, our finding eliminates a headache that Treasuries worldwide have had for half a century and guides Treasuries towards more sophisticated Government debt market design. Rather than focusing on auction formats, Treasuries should concentrate on other crucial aspects of the financial market, such as market stability and the primary dealer system.”
The study is the first to examine Treasury auction revenue using real market data from a large-scale market-based experiment. The studied auctions involved 1.95 trillion Chinese Yuan (approximately 291 billion US Dollars) and took place between 2012-2015.