The idea that money growth affects the real economic activity(e.g. real output, empolyment, etc.) only when such monetary growth is purely random or unexpected, is a key issue in modern macroeconomic theory and stabilization policy. The idea is usually called the Rational Expectation Structual Neutrality(RESN) hypothesis.
In view of its profound policy implications, the empirical validity of RESN hypothesis is perhaps an important issue in New Classical Macroeconomics. In recent years, much works have been done to test the RESN hypothesis. Unfortunately, most of these studies have limited their attention to the experience of the economy US. Also, the prior empirical studies in Korea are forcused on two-step procedure by OLS.
This study intends to investigate whether the RESN hypothesis will be fruitful in explaining the Korea economy. Also this study is intends to provide the robustness of the empirical results in Korea by employing three different method, i.e., two-step procedure by OLS, two-step procedure by ARIMA and VAR methodology with different lag lengths, respectively. The seasonal unadjusted quarterly data over the period from 1970 to 1986 have been used to test the RESN hypothesis.
This empirical evidences presented in this study on the basis of the quarterly data from 1970 to 1986 are not supportive for the RESN hypothesis. This study finds that monetary policy, irrespective of whether it is anticipated or unanticipated, affects real output in Korea economy. Therefore the results show the money non-neutrality in Korea economy. The results are robust with respect to the used different method with different lags.
However a careful reconsideration of the money growth specification is motivated. Also further studies about the method of tests of the RESN hypothesis and lag determination are required.