Why do independent central banks (ICBs) generally produce low inflation? Alex Tabarrok says it's because bankers tend to get appointed to run ICBs and bankers have a bias toward low inflation. Megan McArdle says no, it's because nobody holds Congress accountable for the performance of ICBs, which gives them the freedom to appoint people who will do things they don't have the guts to do themselves. Matt Yglesias isn't sure that the Fed is really an ICB in the first place: it was pretty clearly politicized in the 70s and produced high inflation, and it was quite likely politicized in the oughts and helped produce an economic meltdown.
These all sound like reasonable points to me, and I don't think they're mutually exclusive. Maybe they're all right. But I'm curious about something else: what does the academic literature say about the performance of ICBs in advanced economies in the first place? There must be a considerable amount of research on this point. If we take some measure of "independence" and compare that to various measures of medium-term economic performance (inflation, wage growth, GDP growth, unemployment, etc.), what do we find? Does independence really matter very much? Is there some specific aspect of independence that matters more than others? Can some friendly econoblogger summarize the literature for us?