Did Direct Access to the Treasury Bid System Affect Interest Rates?

In June 2011 the U.S. Treasury Department, for the first time, allowed direct access to its auction system to an entity outside the investment banking community.  The entity: the People’s Bank of China.  The deal involved giving the People’s Bank of China a direct computer link to the Treasury’s bidding system. Overall, this historical change didn’t save the People’s Bank any money, since primary dealers are prohibited from charging clients for bidding on their behalf at Treasury auctions. Since the change doesn’t save the People’s Bank any money, why did they want it? The answer comes down to simple economics: asymmetric information.  Essentially, by limiting the amount of information representatives of the People’s Bank share with primary dealing investment banks, the People’s Bank may be able to get a better price on their bond purchases.

With this background in mind, what has happened since China has had direct access?  Has it affected interest rates?  Here is some empirical evidence:

In June 2011, the time when the People’s Bank was set up with a direct computer link to the Treasury’s auction system, the 1-year T-Bill was trading at 0.18% per annum, the 10-year T-Note was trading at 3.01% per annum, and the 30-year T-Bond was trading at 4.24% per annum.

Since then, as of the first week of June 2013, the 1-year T-Bill traded at 0.14% per annum, the 10-year T-Note traded at 2.13% per annum, and the 30-year T-Bond traded at 3.28% per annum.

Treasury Rate Change

Given these facts, let’s look at the People’s Bank’s conflicting interests.

On the one hand, higher yields increase interest income to its own balance sheet.  Most in the investment banking community would likely find this as the sole goal of individuals running the People’s Bank. In contrast to the return maximization interest, the People’s Bank also has an interest to maximize Chinese exports.  Since higher interest rates (yields) put downward pressure on economic growth, the People’s Bank may, in a sort of hidden agenda type of view, want lower Treasury rates because lower interest rates increase demand for Chinese exports.

With the conflicting interests in mind, here’s some regression evidence when controlling for other factors that influence the market yield on Treasury debt securities.  These factors include equity market performance, the U.S. dollar exchange rate against other competing currencies, money supply (specifically quantitative easing), the effective federal funds target rate, the labor market, inflation, and other actual and expected economic growth factors.


The results shown above are for the 1-Year, 10-Year, and 30-Year Treasury bond yields.   Overall, all three parameter estimates for the “China Direct Access Effect” are negative (10-Year Note=-0.3273, 1-Year T-Bill=-0.0574, and 30-Year T-Bond=-.2044), although only one, 10-Year Note, is close to 95 percent statistical significance (92% level for T-Note). (A negative coefficient means that when China was given direct access (i.e. change in dummy variable from 0 to 1 since June 2011) it reduced yields.  For instance, if China hadn’t been given direct access, the model would imply 10-Year yields of 2.46 today instead of 2.13 with 92% confidence that the effect is a reduction in the yield.) As a note of reference, assuming every year the federal government needed to finance $2 trillion in short-term and long-term expenditures, then every 1% increase in yield associated with a mixed basket of short-term and long-term financing would equate to about $20 billion in increased interest expense.

Overall, since China was provided a direct computer link to the Treasury’s auction system in June 2011, rates have generally floated lower.  Regression analysis shows negative correction, meaning the People’s Bank’s direct access has been correlated with decreased Treasury yields.  Interesting, although not conclusive by any stretch of the imagination.

Bookmark and Share

Comments on this entry are closed.

Previous post:

Next post:

Real Time Web Analytics