Managed exchange rate regimes and monetary independence: an empirical appraisal

ABSTRACT

Under the impossible trinity, it is alleged that attempts to maintain both monetary independence and an undervalued domestic exchange rate whilst being exposed to global capital flows will culminate in macroeconomic instability. To analyse the validity of the impossible trinity in the instance of a balance of payments surplus, we first employ a number of ARDL models to investigate the potential presence of co-integration between foreign reserves and several variables on the balance sheet of the Central Bank. A VECM is then used to exemplify the fact that policymakers must respond to an increase in foreign reserves so as to steer the interbank rate to the policy rate. In total, we argue that in the case of a balance of payment surplus (i) it is problematic to validate the existence of a transmission mechanism between foreign reserve accumulation and inflation; (ii) economies with heavily managed exchange rates appear to have achieved monetary independence whilst remaining open to global capital flows; and finally, (iii) since the four Central Banks examined target an overnight rate, rather than any form of monetary aggregate, ‘exogenous’ sterilisation initiatives are involuntary.

KEYWORDS:

• Sterilisation
• Impossible trinity
• Exchange rates

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. With respect to the gold exchange standard during the interwar years, a number of involuntary constraints were imposed on policymakers, such as the requirement to maintain a system of fixed exchange rates and gold convertibility. With the breakdown in Central Bank cooperation and a series of balance of payment crises, the system collapsed (see, for example, Eichengreen 1992). Likewise, under the Bretton Woods System – in which the US Dollar acted as an anchor to a global system of fixed exchange rates and was subsequently convertible into gold – the Triffin (1960) dilemma resulted in several attempts for cooperation amongst Central Banks, with the result being the formation of the London Gold Pool 1961. Nevertheless, with US Dollar liabilities significantly exceeding US gold stocks, the Bretton Woods System was abandoned by President Nixon in 1971.

2. A balance of payment surplus – in which the Central Bank accumulates stocks of foreign reserves within a fixed or managed exchange rate regime – would arise if the current account surplus were to exceed the financial account deficit, or if both accounts were to attain a positive balance. On the other hand, if the current account deficit is smaller than the financial account surplus, the Central Bank would once again accumulate foreign reserves.

3. In contrast to various economic schools of thought (be it orthodox or heterodox), the analysis in this paper is undertaken both in its theoretical and econometric approach through a framework of de facto Institutional Practice. To this end, as opposed to examining other factors which may contribute to a nation’s position within the trilemma – such as Government ideology and the position of the respective economy within the business cycle (see Berdiev, Kim, and Chang 2012; Beckmann et al. 2017) – we have chosen to place emphasis on the Institutional Practices of the four Central Banks. Also, the paper considers de facto domestic exchange rate policies rather than de jure.

4. For an elaboration as to why we depict sterilisation as ‘exogenous’, see Section 2.

5. These include Mexico (1994), Thailand, Indonesia, Malaysia, the Philippines and South Korea (1997), Russia (1998), Argentina (2000), Turkey (2018) and Pakistan (2019).

6. Such an external balancing process is an extension of the ‘price-specie’ flow mechanism which underpinned the operation of both the classical and gold exchange standard. The price-specie flow mechanism envisaged that a trade deficit would result in gold outflows, which would in turn impose deflationary forces on the domestic economy, restoring the competitiveness of exports. On the other hand, a trade surplus would lead to inflation, thereby eroding an external surplus (Eichengreen 1992, 32).

7. Although more complex approaches have been developed since the monetary approach to the balance of payments (e.g. portfolio balance model), the assumption that a balance of payments surplus (and hence an enlarged monetary base) will lead to an increase in the domestic money supply still lingers.

8. This point requires some elaboration. According to Lavoie and Wang (2012), in the aftermath of a balance of payments surplus and subsequent intervention by the Central Bank to maintain an exchange rate parity, commercial banks which are faced with a reserve excess will endeavour to reduce their level of indebtedness vis-à-vis the Central Bank. Thus, in this instance, any excess reserves are ‘endogenously’ sterilised. We extend this analysis by distinguishing from the process of endogenous sterilisation by arguing that influxes of foreign reserves may also be ‘exogenously’ sterilised, as the removal of excess reserves is undertaken by the Central Bank.

9. For instance, the IMF (2018, 15) note that ‘ … Thailand stands out as having purchased a significant amount of reserves (and forward contracts) despite having more-than-adequate reserves and a higher-than-desirable current account balance.’

10. Fiscal costs in the context of reserve sterilisation arise when an interest rate differential exists between the rate of interest paid on instruments to withdraw excess reserves and the yield received on the asset which foreign reserves are placed in.

11. The demand-led endogeneity of money differs to what is generally portrayed in a number of textbooks. Under a fixed or managed exchange rate regime, the overall money supply is claimed to be supply-led endogenous, as the Central Bank must respond to capital flows so as to maintain a parity and prevent the money multiplier from operating.

12. For example, Becker and Sinclair (2004, 6) point out in a research paper for the Reserve Bank of Australia that to appreciate the Australian dollar, ‘ … the RBA buys Australian dollars … there is a fall in the banking system’s holdings of Australian dollars, thereby draining cash from the domestic money market … the market would be short of cash and domestic money market interest rates would tend to rise … The RBA can … act in the domestic money market to replenish the banking system’s liquidity by buying securities. This cancels, or sterilises, the liquidity effect … and leaves domestic interest rates unchanged.’ Likewise, the Bank of Thailand (2019, 261) detail that any ‘ … liquidity created by the purchase of USD needs to be fully sterilised to ensure that the prevailing money market rates are in line with the Monetary Policy Committee’s (MPC) policy interest rate.’

13. In some instances, the Central Bank may also implement its monetary policy through a floor system. In this case, the Central Bank targets the floor of the channel, with interest paid on any excess reserves. As a result, reserve draining operations are not required.

14. A repurchase agreement entails the Central Bank selling securities in exchange for reserves – with the Central Bank effectively ‘mopping up’ excess reserves. On the other hand, a reverse repurchase agreement sees the Central Bank purchasing securities so as to inject reserves into the banking system.

15. The Bank of Canada (2016, 5) summarise such a process: ‘To achieve a target level of settlement balances, the Bank … transfers net public sector payments and receipts … to and from the government.’

16. Owing to both current and financial account surpluses, the People’s Bank of China imposed higher minimum reserve requirements to mop up any excess liquidity (Glick and Hutchison 2009).

17. Mundell (1963, 475) notes that the assumption of perfect capital mobility ‘overstates the case’. For an extensive review regarding the degree of capital mobility in several economies, see Al-Jassar and Moosa (2020).

18. The notion that perfect capital mobility does not guarantee perfect asset substitutability was also noted by Keynes (1923, p. 115–132).

19. The data may be directly obtained from the author upon request.

20. Following the Asian Financial Crisis, the Bank of Thailand intervened heavily in foreign-exchange markets to curb exchange rate speculation. More recently, the Bank has been active in the foreign-exchange market to maintain external competitiveness (Bank of Thailand 2013, 327). Likewise, Malaysia moved to a managed float in July 2005 (Aziz 2013), with policymakers in South Korea also heavily active in foreign-exchange markets (Ryoo, Kwon, and Lee 2013). For a recent appraisal of exchange rate management policies in all four nations, see Bank of International Settlements (2019).

21. For example, the Bank of Thailand issue Central Bank securities in a number of different maturities, ranging from fourteen days to three years (see Bank of Thailand 2019, 263).

22. Note that the calculated F-statistic must demonstrate co-integration for the long-run coefficient to be meaningful, even if the latter is statistically significant.

23. A negative and significant $\theta$ would indicate that an adjustment process takes place following a shock. The higher the absolute value of $\theta$, the faster the adjustment process.

24. Prior to the empirical analysis, the functional form of all the variables was examined, which, in all cases, remains linear following transformation into logarithmic form.

25. To ensure robustness in the ARDL, we also employed checks to test for serial autocorrelation and heteroscedasticity, with the cumulative sum of residuals calculated to examine for any structural break in the data (see Appendix A).

26. We envisage that this result is due to the fact that there was a large spike in inflation immediately preceding the Global Financial Crisis.

27. As deterministic trends exist in the majority of the variables, we have included a constant and linear trend in the VECM.

28. The computed eigenvalues have not been included in Appendix B but are available from the author.

29. Whilst it is obvious that foreign reserves should be placed first, and the monetary base last, it is not entirely evident whether Government deposits should precede Central Bank securities or vice versa. Hence, the two were switched. However, in all four economies, the results were identical, and subsequently are not included in Appendix B.