Investigating long-run demand for broad money in the Gulf Arab countries

Abstract

We estimate the long-run demand for broad money for the six Gulf Cooperation Council countries (Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates) over the 1980–2012 period. Applying time series and panel econometric tests, we first document the existence of long-run equilibrium relationship for money demand – both nationally and regionally. The estimated income elasticities are generally lower than those reported in the literature, while the interest elasticities are more in line with the standard money demand literature. We discuss how the movements in income velocity can reconcile the varying income and interest elasticities documented across the six countries. A discussion on the homogeneity (poolability) of the long-run money demand parameters and the error correction model of money demand is also provided.

Keywords:

• money demand
• endogenous money
• Gulf monetary union
• panel cointegration
• poolabiilty

JEL Codes:

• C23
• E41
• E52

Acknowledgements

We are grateful to the two anonymous reviewers and the editor of this journal for helpful suggestions, to Elsayed Elsamadisy and Khalid Alkhater for stimulating discussions, to Megan Foster for help with proofreading, and to the participants at the 2012 Italian Statistical Society conference for comments and suggestions on earlier versions of this paper. The views expressed here are those of the authors and do not necessarily reflect the official view of the affiliated institutions. The usual disclaimer applies.

Notes

1. Rodríguez and Rowe (2007) showed that changes in the US monetary instrument (the US federal funds rate or the US money supply) have real effects in Hong Kong, where the currency is pegged to the US dollar.

2. Other exogenous factors that can lead to domestic money creation or destruction are short-term speculative flows in the stock market and the real estate sector. Furthermore, the case of money laundering as a contributing factor cannot be ruled out.

3. Nevertheless, the ability to influence money supply varies across the six Gulf countries. For example, the bulk of Saudi Arabia's oil revenues which are held as reserves with the Saudi Arabia Monetary Agency (SAMA), Saudi Arabia's central bank, allow SAMA to impact domestic money growth through buying and selling US dollars in the banking system. Further, some Gulf central banks are better equipped with instruments of monetary sterilization and other prudential tools to manage liquidity.

4. Briefly, under the accommodationist view, the money supply is demand-determined (i.e., bank loans) and the central bank provides reserves and currency on demand. The structuralist view argues that the central bank keeps some control over the supply of reserves, while the liquidity preference view implies a bidirectional causality between bank loans and money supply. See Tas and Togay (2012) for further discussion.

5. There is a burgeoning branch of literature on the feasibility and progress of the GCC monetary union – see, among many others, Buiter (2008) and AlKholifey and Alreshan (2010).

6. See Goldfeld and Sichel (1990) for a discussion of the relationships among the functional form of money demand, the existence of the liquidity trap, and the effectiveness of monetary policy.

7. According to the International Monetary Fund (2009), bank assets and equities account for 94.4% of finance in the Middle East countries.

8. National credit policies are a powerful economic/politcal tool in GCC countries. Besides stimulating investment in certain sectors, governments use the credit tool to distribute the proceeds from oil wealth to its citizens. The recent Arab uprising has prompted some GCC authorities to rely exclusively on the credit channel to provide soft long-term funding (including housing) to their nationals.

9. See Darrat (1986), Darrat and Al-Sowaidi (2009), and Sbeiti and Haddad (2011).

10. Although, in the recent decade (2001–2010), the empirical relationships between the USA and the GCC interest rates has been different compared with the findings of the long-run relationships. For instance, over the 2004–2011 period, Espinoza and Prasad (2012) reported evidence of low pass-through to GCC interest rates from the US rate. Over the same time period, Elsamadisy, Alkhater, and Basher (2014) rejected the uncovered interest rate parity between Qatar and the USA. However, since we are interested in analyzing the long-term relationships, we do not believe that these short-term results present a challenge to our analysis.

11. Time Series Regression with ARIMA Noise, Missing Observations and Outliers (TRAMO) is a program for estimating and forecasting regression models with possibly nonstationary (ARIMA) errors and any sequence of missing values. A computer code implementing the program is available as a free download at http://www.bde.es.

12. In fact, domestic liquidity was supported by the fractional withdrawal of Kuwait's external assets to financial reconstruction after the 1990–91 Gulf War. For instance, in late 1991, the government agreed to purchase US$20 billion worth of domestic debt held by Kuwaiti nationals and businesses in exchange for bonds. See Metz (1993) for further discussion.

13. The p -values for the tests with a constant have been computed using MacKinnon's (1996) urcval procedure as embedded in the econometric program Gretl. Since no routine is available for the trend case, the tests of this type have been carried out applying the critical values provided by Elliott et al. (1996).

14. Saudi Arabia's total GDP is clearly nonstationary, even allowing for a trend under the alternative hypothesis (ADF-GLS equal to –2.49, to be compared with a 10% critical value of –2.89).

15. On the connection between panel test statistics and panel null hypothesis, see Chang and Nguyen (2012), and Pedroni (2004).

16. See Basher and Elsamadisy (2012) for a discussion on the potential endogeneity between money and output in the context of GCC.

17. Strictly speaking, an income elasticity in excess of unity is generally considered as real money balances being a luxury good.

18. During the period of this study, population growth in Kuwait was lowest among the GCC countries. The Kuwaiti stock exchange is the first official bourse established in the GCC region and includes the largest number of listed companies. Furthermore, Kuwait has the largest number of investment companies proving asset/portfolio management services and, until very recently, was the only market with an active trade in derivative securities.

19. In the development literature, this transformation is often referred to as the ‘process of monetization.’ Increases in the share of national output sold in organized markets, increases in the fraction of the population working for wages, and increases in the number of intermediate payments required in production are the most frequently cited factors underlying a decline in velocity the early stage of economic growth. See Ireland (1994) for further discussion.

20. It should be remarked that this outcome is robust to the exclusion of Saudi Arabia from the panel, the country with the largest income elasticity coefficient (results not reported here, but available on request).

21. An earlier version of this paper, which included a slightly shorter sample (1980–2009), the null hypothesis of poolability could not be rejected. As a result, different pooled estimators of the regional money demand were provided. See Basher and Fachin (2012) for further discussion.

22. To the best of our knowledge, there is no available empirical analysis that concerns the short-run analysis of money demand function for the GCC countries to be able to compare our results with the existing literature.