The pdf is here, and the core result is that during the recent global financial crisis, fixed exchange rate regimes fared as well as floating rates with inflation targeting. (Krugman comments here, Antonio Fatas comments here.) That finding should not come as a surprise, for a few reasons:
1. Floating rates can be tough on countries which import a lot of oil, or which import critical inputs for their production or exports.
2. A lot of countries in 2008-2009 did not experience a traditional AD-only shock, rather they experienced trade shocks, which for the purposes of macro are like real shocks. Floating rates offer less protection against real shocks than against nominal shocks.
3. One of the benefits of floating rates is that a country can crank up the rate of price inflation when it needs to, to offset negative demand shocks. If you are targeting an inflation rate, you are not doing this and that means you lose out on a potential benefit of floating rates.
4. Fixed rates (or rather a single currency) have been a definite disaster for the eurozone. Because of the nature of the EU, the absence of capital controls (except for Cyprus), and proximity, it is easy to switch your funds out of the failing countries, thereby whacking their banking systems. I would rather have a euro in a German bank than a Greek bank, and so on. This problem with “fixed rates” is especially strong in common currency areas with insolvent banks, and excess sovereign-bank connections, and it is less difficult in other fixed rate settings.