Jim O'C on expansiveness
Doug, the nation income accounts give the perverse result that an export-led expansion accompanied by growing surplus (two features of the US economy much of the 1990s) show up in national income accounting as a negative factor in growth (because M exceeds X).
If you assume different compositions of X and M in terms of C vs. I goods, you also get results that don't truly describe what's going on. For example, if all X=I, then a country is more productive in I than appears in the income accounts, but since a portion of I is exported, less productive, potentially, in the future. If, cet. oar., all M=I, then the reverse is true.
A limiting case is when a country exports all of its I output and/or imports all of its C needs. This comes close to the old model of super-exploitation by an imperialist country of its colony, where I ouput=say, oil. Where all C needs are supplied by imports. And where wage income is remitted to the home country and all profits are repatriated to the imperialist country.
I believe that the US exported relatively more I (including primary products) than C goods in the 1990s, hence produced more than it appears from the national income tables. And imported relatively more C than I goods. Trading away future production capability for present consumption gratification, as it were.
As for the story of Tobin and JFK and the 1950s, for him to be right that I was negative over the period while C+G was growing c. 115 percent, either the statistics have been changed, presumably improved, and he was wrong. Or net I was very negative in 53Q3/4-54Q1 and other recession of the decade. Because the figures you give don't support Tobin at all.