Monday, February 14, 2005

Do Foreign CBs distort the Treasury Yield?

by Calculated Risk on 2/14/2005 11:49:00 AM

Several economists have recently suggested that Foreign Central Bank purchases of treasuries have distorted bond yields by anywhere from 40 to 200 bps. As an example, from the Feb 3rd The Economist:

"By some estimates, Asian purchases of American bonds have reduced yields by between half and one percentage-point."

And from the Federal Reserve's Bernanke and Reinhart writing with Macroeconomic Advisers' Brian Sack "Monetary policy alternatives at the zero bound: an empirical assessment":
"The results ... indicate that both five-year and ten-year Treasury yields remained below the model’s predictions by an average of 50 to 100 basis points over this period. This suggests that some force not captured in the model was exerting downward pressure on yields over this period. But while the evidence is suggestive of effe[c]ts from MOF purchases, it is not conclusive."

And Roubini and Setser in "Will the Bretton Woods 2 Regime Unravel Soon? The Risk of Hard Landing in 2005-2006" reviewed recent estimates of the impact of Foreign CB purchases:
"Goldman Sachs (2004) has presented an analysis suggesting that central banks intervention is narrowing Treasury yields by only 40bps; Sack (2004) provides a similar estimate. Truman (2005) notes that sustained intervention from central banks is similar to a sustained reduction in the fiscal deficit: his ballpark estimate suggests a $300 billion in central bank intervention might have a 75 bp impact. Research from Federal Reserve suggests a 50 to 100 bps impact (see Bernanke, Reinhart and Sack (2004)); PIMCO’s Bill Gross puts it at closer to 100 bps, and Morgan Stanley’s Stephen Roach puts it at between 100 and 150 bps."

Roubini and Setser concluded that these estimates are too low:
"While estimating the effect of central banks intervention on US long rates is difficult, there is good reason to suspect that the impact of central bank purchases much larger than the 40bps static effect estimated in some studies. ... Consequently, the 40bp Goldman estimate seriously understates the effects of the Asian intervention on the market. Considering the size of recent central bank purchases, the indirect impact of central bank intervention on private demand for Treasuries, the interaction between central bank reserve accumulation and Treasury debt management policy and the effects of Asian reserve accumulation on inflation and growth (general equilibrium effects), we would bet the overall impact would be closer to 200bps." emphasis Added

Over the weekend, the IHT had an article (by Daniel Altman) titled: "U.S. debt: Watch out for the domino effect". Although the article didn't mention lowering of American bond yields by foreign CBs, it touched on the impacts of Foreign CBs diversifying away from dollar denominated assets.

With regards to foreign CBs distorting the treasury yield, I remain skeptical but intrigued. If Foreign CBs are depressing American bond yields, this has already boosted any RE bubble by lowering borrowing costs for homebuyers. Any unwinding of these positions could potentially lead to a slowdown in the US economy with rising interest rates. That would not be a good combination.