Testing the Friedman–Schwartz hypothesis using time-varying correlation analysis

ABSTRACT

This study analyses the time-varying correlation of money and output using the DCC GARCH model for the Euro, India, Poland, the UK and the US. Apart from simple sum money, this model uses Divisia monetary aggregate, which is theoretically shown as the actual measure of monetary services. The inclusion of Divisia money affirms the Friedman–Schwartz hypothesis that money is procyclical. The procyclical nature of association was not robustly observed in recent data when a simple sum money was used.

KEYWORDS:

• DCC GARCH
• Divisia
• monetary aggregates
• real output

JEL CLASSIFICATION:

• C32
• E52
• E51

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ The results are robust to the use of different types of Divisia money, different types of simple sum money and different combinations of countries.

² See Table A3 (Appendix); null is rejected at the 1% significance level.

³ Table A4 presents the conditional mean and conditional variance equations.

⁴ Table A5 presents the results for lags 10. The results are robust to the use of different lags.

⁵ With the exception of India, the Divisia data of which are available only until June 2008.

⁶ http://bruegel.org/publications/datasets/divisia-monetary-aggregates-for-the-euro-area/.