Comments and Questions

The authors study the relationship between inflation, inflation uncertainty and output in Tunisia employing a GARCH model and performing GRANGER-causality tests. They find that inflation does not influence inflation uncertainty and that inflation uncertainty weakly affects real inflation. General commentThe topic of the paper is promising. However, in the present form the original content of the paper is quite poor. The paper is badly written, the research hypotheses are not clearly stated, the related literature is not sufficiently explored, and the empirical results are not convincing. Detailed comments1) The English is not fluent and it should be considerably improved.2) I do not understand what is the “apperceive” hypothesis of Milton Friedman.3) The authors should better explain how empirical works have studied the relationship between price volatility and economic growth.4) The authors should better explain the relationship between expansionary monetary policy and inflation uncertainty. Moreover, to provide a clearer picture of inflation dynamics in Tunisia, the authors should add a plot of the inflation time series.5) When presenting Granger tests, the author should warn about the limits of such an analysis.6) In the last paragraph of section 2, there is a symbol which I do not understand.7) The authors do not explain what is the real inflation rate. Why is it always negative? Without having a definition of the real interest rate, it is very difficult to understand the meaning of the empirical exercises performed by the authors. 8) As Dickey-Fuller tests suggest that all the variables are I(1), do the authors estimate the GARCH model on first-differenced series?9) From the estimation of the GARCH model, only inflation uncertainty affects real inflation at the 10% significant level. The authors should try with different specifications of the econometric model to find more robust and convincing results. For instance, if inflation at time t-1 never significantly affects inflation uncertainty, the authors should try with inflation at time t-2.10) Given that the Q-statistics report that the residual of the GARCH model are serially correlated, why the authors do not try to find a different specification of the model which solve the problem?11) The evidence supporting the rejection of null hypotheses that inflation uncertainty does not Granger-cause either inflation or economic growth is not very robust. This should be mentioned by the author in the paper. In order to provide a more detailed picture, the authors should also test whether inflation or economic growth Granger-cause inflation uncertainty.12) In the Concluding Remarks section, the authors claim that “our findings conclude that recession in economy is essentially due to high inflation uncertainty enhanced by lowering the level of the interest rate of the Central Bank of Tunisia”. I do not understand how the empirical results of the authors support this statement.