Economic policy uncertainty and demand for money in Australia

ABSTRACT

The demand for money has received a great deal of attention in the empirical literature. This literature, however, has emphasized factors such as interest rate, income, inflation rate and exchange rate as the primary determinants of money demand. Although an emerging strand of literature examines uncertainty as a potential determinant of money demand, findings have been mixed. Using a news-based Economic Policy Uncertainty (EPU) index and Australian quarterly data from 1998 to 2017, we study the impact of policy uncertainty on demand for money. Autoregressive distributed lag (ARDL) results show that the economic policy uncertainty measure has a negative short-run effect on the demand for money, suggesting the wider public hedge against future expected inflation, and positive long-run effect, whereby the broader public hold more cash to stay liquid during times of economic uncertainty. Also, introducing nonlinearity into the money demand equation, we find an asymmetric effect, more in favour of currency appreciations, supporting the expectations effect of further appreciations in exchange rate movements.

KEYWORDS:

• Money demand
• policy uncertainty
• demand for money
• Australia

JEL CLASSIFICATION:

• E41
• E40

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Please see http://www.policyuncertainty.com/australia_monthly.html for details on index construction.

² See Appendix A for data definition and sources.

³ Arango and Nadiri (1981) argued that a wealth effect leads to an increase in consumption and thus an increase in money demand since a deprecation of domestic currency raises the domestic currency value of foreign assets. Alternatively, Bahmani-Oskooee and Pourheydarian (1990) and Bahmani-Oskooee (1996) argued that if domestic currency depreciations lead to further expected depreciations, economic agents may choose to hold more foreign currency and less domestic currency.

⁴ The advantage of using the ARDL procedure is that it can be applied regardless of whether the variables are stationary or non-stationary, thus avoiding the pre-testing problems associated with conventional cointegration methods such as the Engle and Granger (1987) and Johansen and Juselius (1990). In addition, any endogeneity concerns are avoided with the appropriate modification of lag lengths in the ARDL model (Pesaran, Shin, and Smith 2001).

⁵ A similar estimate for income elasticity in the Australian case in found in Bahmani-Oskooee and Xi (2011) who report a value of 2.01.

⁶ For example, if domestic depreciations (or foreign appreciations) induce a stronger wealth effect, then domestic economic agents will demand more Australian dollars. However, foreign appreciations may lead to further and more severe appreciations leading to intervention in the foreign exchange market by the central bank. Thus, as expectations become relatively stronger over time, Australians may decide to hold more foreign currency and less Australian dollars, producing an asymmetric effect. See Bahmani-Oskooee, Halicioglu, and Bahmani (2017) and Bahmani-Oskooee, Xi, and Bahmani (2016) who also study the impact of asymmetric effect of exchange rates in money demand equations.