Mark Thoma suggests that people need to pay more attention to the macroeconomic foundations of microeconomics. We talk about the microeconomic foundations of macroeconomics all the time (this is one of those areas where Keynesians have been working and publishing seminal stuff for decades, but critics ignorantly insist they never even tak about). As a result, we have good instincts to guard against committing ecological fallacies. But we never consider the reverse, and so we often very easily slip into fallacies of composition. Things that ought to be intuitive, like the "paradox of thrift", instead become... well... paradoxical.
The other major problem that comes of this micro-obsession is the confusion of accounting identities with behavioral laws. When you think that all you need to explain is individual behavior, and then you just have to add up all that individual behavior, then you can put an awful lot of faith in accounting identities. Ceteris paribus (the most dangerous two words in economics) allows all sorts of false inferences based on intuition about individual behavior.
The really interesting question is why we're always so interested in microfoundations in the first place and not macrofoundations. I honestly never considered this until I saw this by Thoma today. Why?