Are the Minutes of the Bank of Korea Informative on the Policy Rate Decisions?

23 Pages Posted: 18 Nov 2018

Date Written: October 27, 2018


We estimate the policy stance on the monetary policy rate decision, which stands for the intensity of committee's consensus to adjust the policy rate, with the minutes of the Bank of Korea from 1999 to May 2018. We try to estimate the policy stance more directly without relying on external dictionary information by using a model that combines the text of the minutes and the actual policy rate decisions, both of which are the observations generated from the policy stance. We also propose an algorithm which generates additional N-grams to improve the quality of Korean text data. We see the estimated policy stance as an information variable summarizing large information set on the economic situation and prospects used in the policy rate decision. By including the policy stance in VAR model, the identification of the policy rate shock is effectively improved, without losing much degree of freedom by adding only one variable. The results suggest that using the policy stance estimated from the text data can be a good way of including information on the policy decision factors such as inflation and output forecasts, as an alternative to other approaches which use economic indicators, even though no direct information on these factors are used for the estimation. This can be thought of as one example shows that the text data work as alternatives to the traditional economic data, possibly even more parsimoniously and efficiently.

Keywords: Monetary Policy, Policy Stance, Text Mining, Korean Text, Minutes, Bank of Korea, VAR, Price Puzzle

JEL Classification: E52

Suggested Citation

Jung, Dong Jae, Are the Minutes of the Bank of Korea Informative on the Policy Rate Decisions? (October 27, 2018). Available at SSRN: or

Dong Jae Jung (Contact Author)

Seoul National University ( email )

Seoul, 151-742
Korea, Republic of (South Korea)

Register to save articles to
your library


Paper statistics

Abstract Views
PlumX Metrics