This complex and interesting paper explores the relationships among the ability of a government to ‘precommit’ and the optimal degree of debt indexation and maturity structure. The paper is essentially a series of examples using a government budget constraint and a loss function, with no specified model of how the economy works, so it is not clear how general the results are. But the results are intuitively clear. Precommitment and indexation are good, but not perfect, substitutes; indexation seems to me to be a form of precommitment. With precommitment, long debt is optimal, because it allows maximum flexibility in using the inflation tax to smooth the path of the tax burden once uncertainty about shocks is resolved. Without precommitment, the optimal maturity structure becomes shorter, to reduce the temptation on the government to use the inflation tax unexpectedly.

The paper is sufficiently complex that I had to spend quite a bit of time working through the models to see the results. Part of the difficulty comes from the authors' tendency to discuss results before deriving them. But part also is in the complexity of the problems being studied. So before I come to some critical questions at the end of this comment, I will provide a brief reader's guide to the paper, going over the model, the structure of the paper, and the results.