BY Camille Cornand and Cheick Kader M’Baye
http://d.repec.org/n?u=RePEc:hal:wpaper:halshs-01313095&r=dge
We conduct laboratory experiments with human subjects to test the rationale of adopting a band versus point inflation targeting regime. Within the standard New Keynesian model, we evaluate the macroeconomic performances of both regimes according to the strength of shocks affecting the economy. We find that when the economy faces small shocks, the average level of inflation as well as its volatility are significantly lower in a band targeting regime, while the output gap and interest rate levels and volatility are significantly lower in a point targeting regime with tolerance bands. However, when the economy faces large shocks, choosing the suitable inflation targeting regime is irrelevant because both regimes lead to comparable performances. These findings stand in contrast to those of the literature and question the relevance of clarifying a mid-point target within the bands, especially in emerging market economies more inclined to large and frequent shocks.
Interesting approach to use an experiment with live actors to test a model. I am not sure about its validity though, as outcomes are provided by the model, not endogenously through the interaction of the participants. And the model is extremely crude.
NEP-DGE Blog 
Thanks for the comments.
This experimental approach in macro follows the learning to forecast literature, notably developed by the team of Cars Hommes in Amsterdam.
The model is of course very simple, to keep things as general as possible. And the outcomes are provided by the model, conditional on the heterogeneous expectations formed by participants in the lab.
The main contribution of the experimental approach consists in designing comparable treatments by varying only a single feature at a time. In our case, we have two comparable treatments, namely whether a point or a band target is disclosed to participants. We can thus evaluate to what extend communication about band or point target matters for economic outcomes.
The replicability is also of interest.
Obviously, in the real world, it is extremely difficult to have so pure data and to bench test alternative policies.