One question raised by our numerical results concerns the absence of conclusive evidence
from existing empirical studies, which, for the most part, have failed to find a sizable and
robust effect of long-run in
ation on a country's GDP. Interestingly, we show that if the
baseline calibrated model were true, then there would still be a 15 percent chance that a
cross-country regression of log GDP on in
ation would reveal a statistically insignificant
effect. The reason is that our model implies an inherent endogeneity issue, which arises from
a positive Phillips-type relationship between output and in
ation that tends to mask the
underlying negative long-run causal effect of trend in
ation on output.
The paper is organized as follows. In Section 2, we brie
y survey some of the relevant
literature on the real effects of in
ation. Section 3 lays out the basic structure of our
agent-based model and discusses its workings. Section 4 presents our method for calibrating
the model's parameters. Section 5 reveals our main results and those from our sensitivity
analysis. In Section 6, we discuss our technique for dealing with the Lucas critique. Section
7 shows the implications of the model for cross-country regressions of output on in
flation,
and Section 8 concludes.