The long-term view

Can the conclusion that a stronger CNY does not exert upward pressure on reported inflation in the U.S. be confirmed over a longer period of time and across a wider spectrum of currencies? After all, the revaluation of the CNY should not be confused with the movements of a freely floating currency, and the process has been underway only since July 2005.

If we map the year-over-year changes in the dollar index on an inverse scale against the year over year changes for the CPI and PPI, we find only a weak leading relationship. A weaker dollar leads changes in the CPI by eight months and changes in the PPI by nine months, on average. However, the respective r2 values of 0.003 and 0.035 are statistically insignificant. No link between the dollar index and reported inflation can be asserted over the long-term. However, this statement would not have been made prior to the turning point in the dollar’s early-1980s strong period.

The dollar index’s rally started to reverse in February 1985, as marked with a magenta vertical line. Prior to this reversal, the r2 values for the rates of change of the PPI and CPI, respectively, against the rate of change for the dollar index were 0.226 in both cases. After February 1985, the respective r2 values fell to 0.0002 and 0.0368. An F-test of these regressions to determine whether they were statistically different before and after February 1985 confirmed they were at near 100 percent confidence.

What changed? This was the beginning of central bank coordination of monetary policy and efforts to drive the dollar both lower (the September 1985 Plaza Accord) and higher (the February 1987 Louvre Accord). Once central banks realized independent monetary policies could not affect short-term interest rates and currency rates simultaneously, they turned away from direct currency management and toward national monetary policies. To the extent these policies matched — and they often did currency rates could stay relatively static while inflation rates diverged. The opposite could be obtained as well; consider the Federal Reserve’s willingness to accept greater inflation in 2003-2004 while the dollar fell.

Restated, a central bank can fix its short-term interest rates or it can manage its currency, but it cannot do both simultaneously.

Those who continue to believe in the inflationary consequences of currency changes are viewing the world through the pre-1985 prism of non coordinated central bank policies. Until and unless the era of central bank coordination ends, we should expect the disconnection between currencies and inflation — and between the CNY and American inflation, both expected and reported — to continue.