wrote the below email to a friend of mine just now and figured it was worthwhile posting.
Martin Feldman of “Harvard fame” addresses both the effectiveness of QE (it hasn’t been very effective in terms of spurring real GDP growth, asset prices is a different matter) and the low inflation rate (since 2008, for the first time in Fed history, the Fed has been paying 25bp interest on excess reserves, curbing M2 growth, which in turn has a direct corollary with inflation) in the following, accessible article.
On the latter point, consider this. In nominal terms, since 2000 the economy has increased 1.7x, but M2 has increased 2.3x. Add excess reserves and you get a 2.7x multiple. Say the Fed is unable to curtail these excess reserves from eventually finding their way into the economy – perhaps not plausible given their new “tool-kit,” but bear with me – and you’re in a situation where you’ve almost doubled the money circulating in the economy relative to the goods and services produced by the economy.
Intuition in these matters doesn’t get you far, but somehow that is a jarring number. And yet CPI is low, but how good a measure is CPI? The below shows the spread between the CPI methodology in 1980 versus today’s methodology. Point is: there is no right answer and a point-estimate won’t capture the whole truth. As with most things, there’s a distribution, and seemingly the distribution is pretty wide.
Also bear in mind that the above metrics don’t account for dollars held overseas. The status of the US dollar as a de facto reserve currency and the preferable medium of exchange for settling things like oil, iron, precious metals, etc., isn’t a given. What if the share of transactions denominated in the USD falls at a faster rate than the growth in volumes of transactions? Is it an altogether implausible scenario in which, domestically, there’s too much money chasing too few goods and, internationally, there’s less demand for dollars?