In 1987 the G6 countries (Canada, France, Germany, Italy, Japan, and the UK) accounted for 55% of U.S. goods imports. That same year, China, Mexico and Brazil only accounted for 8% of imports.
In 2010 the U.S. reached a milestone–for the first time, imports from China/Mexico/Brazil exceeded imports from the G6 countries. In the year ending March 2011, imports from China/Mexico/Brazil equaled 32% of goods imports, compared to 31% for the G^ countries. Here’s another way of seeing the same thing. Please note that OPEC’s share, and the share of “all other countries,” don’t change very much. It’s really the G6 versus a handful of low-cost importers.
One final note. The shift in sourcing is most likely happening because the goods made in China/Mexico/Brazil are less expensive than the same goods made in France/Germany/UK. Unfortunately, the BLS import price statistics are not able to pick up the price drops from shifts in country sourcing.
Suppose for example that goods made in China are sold for one-third less than the same goods made in Japan. Then for the same physical quantity of imports, that shift in sourcing will cause the nominal value of imports to be one-third lower. This imparts a significant downward bias to the import penetration ratio.