A dollar-versus-commodity extreme as a regime case

Read the extreme as a regime

Editorial. A dollar-versus-commodity extreme is a case for classifying an inflation-stress regime. The follow-on work is to put a moving-average baseline and a buy-weakness, sell-strength momentum procedure next to buy-and-hold, so the inflation trade is a testable sequence rather than a narrative.

In late 2004 the US Dollar Index reached 80.53, a nine-year low and a degree of dollar weakness seen only twice in the prior 30 years. By September 2005 the Continuous Commodity Index reached 336.56, just below its 20 November 1980 peak of 337.60, marking a 25-year high for the basket.

The dollar and the commodity basket sat at simultaneous opposite extremes in November 1980 and again around September 2005, but the federal funds rate was about 3.75 percent in the later episode versus near 20 percent in 1980.

What followed the 1980 high-rate squeeze

Gold peaked near 875 dollars in January 1980, oil exceeded 40 dollars a barrel, the federal funds rate reached about 19 percent in mid-1981, and the prime rate rose above 21 percent.

After that high-rate squeeze, Mexico devalued the peso by 30 percent in early 1982, the Dow Jones Industrial Average fell 22 percent by July 1982, some US and Canadian home prices dropped by as much as 50 percent, and by early 1983 the US prime rate was back to 10.5 percent with gold near 350 dollars and oil below 30 dollars a barrel.

Same November date, different policy-rate worlds

On 1 November 1980 gold was near 650 dollars an ounce and falling from above 800. On 1 November 2005 it was roughly 465 dollars while the federal funds rate was about 4 percent versus the 1980-81 peaks near 19 percent.

A rising-rate setting can pull foreign buying into dollar assets even as commodity prices stay elevated, creating a regime in which the dollar and gold can advance together rather than only moving in opposition.

Core consumer-price measures can understate household inflation pressure because they do not fully include housing, food, and fuel, two of the largest consumer expenses.

A moving-average and momentum sequence versus buy-and-hold

A 2005 gold case using oscillators, moving averages, and an intermarket predictive moving average reversed at defined points from a 20 June long at 445.90 through later shorts and longs, producing 43.80 dollars of net gain including one 15.60-dollar loss, versus 21 dollars from holding the first long to 31 October.

The illustrated momentum procedure of buying weakness and selling strength at plateaus is presented as outperforming buy-and-hold except in a sustained long-term uptrend, which is described as occurring less than 20 percent of the time.