(Reprinted from HKCER Letters, Vol. 75 Sept-Oct 2003)
From Relink to Parallel Currencies to Monetary Union
Kam Hon Chu
The lackluster performance of the Hong Kong economy following the Asian financial crisis in 1997 has been attributed by some commentators to Hong Kong's linked exchange rate system, which has pegged the local currency against the greenback at an official rate of US$1 = HK$7.8 since October 1983. To some extent the link has eroded the international competitiveness of Hong Kong's exports, as most East Asian currencies depreciated considerably against the US dollar after the outbreak of the Asian flu. More important, Hong Kong¡¦s currency board system bears much blame for amplifying and prolonging the ongoing deflationary trend after the burst of the bubble economy about four years ago. Under a fixed exchange rate regime, the domestic economy has had to adjust through changes in prices and wages, as evidenced by their downward trends over the last few years. Most notably, property prices have on average plunged by more than half from their 1997 peaks. Had Hong Kong been on a floating exchange rate regime, domestic prices would have fallen less, and the painful adjustment process would have been ameliorated as well.
There are many potentially feasible economic policies that could mitigate the current economic malaise and resurrect Hong Kong's competitiveness. As far as exchange rate policy is concerned, a theoretically possible (although not necessarily practical or optimal) solution is to abolish the linked exchange rate system and allow market forces to determine the external value of the Hong Kong dollar. If such a step were taken, the local currency would in all likelihood depreciate against its US counterpart. One would hope that the depreciation would help the Hong Kong economy extricate itself from its current doldrums. Alternatively, the Hong Kong Monetary Authority (HKMA) could repeg the Hong Kong dollar to the greenback at a higher official exchange rate so as to virtually devalue the local currency without abandoning the existing currency board system. Unfortunately, the decision of whether to delink or repeg the Hong Kong dollar is problematic. For example, it may be difficult to choose the right time to enforce the policy or to determine the "optimal" new official exchange rate if the Hong Kong dollar were to be repegged. More critically, both exchange rate policies could trigger a confidence or credibility problem, as they might be misinterpreted as evidence of a lack of commitment on the part of the HKMA to defend the local currency. As Kasa (1999) correctly points out in explaining why a currency board is still subject to speculative attacks, willingness to defend a peg should be distinguished from ability to defend a peg. In other words, the large quantity of foreign currency reserves held by the Hong Kong government is no guarantee of exchange rate stability.1 The credibility problem, if it arose, would induce massive capital outflows and hence sharp rises in domestic interest rates in the case of repegging; these problems would be compounded by overshooting the Hong Kong dollar in the case of delink. In either case, there would be undesirable, if not disastrous, economic and social consequences.
On the surface, such unfavorable scenarios tend to suggest that the HKMA should follow the admonition of an old adage ¡X if it ain't broke, don't fix it ¡X against changing the status quo. To be fair, the linked exchange rate system has served Hong Kong quite well in maintaining exchange rate stability since its inception in 1983. Although there have been occasional speculative attacks, notably the so-called double plays by speculators in October 1998, the linked exchange rate system has so far stood up fairly well against such attacks and has maintained the stability of the Hong Kong dollar.
Despite its respectable track record, however, it is unlikely that the linked exchange rate system can remain the optimal exchange rate regime for Hong Kong in the long run. Two decades ago, when the system was introduced, the United States was Hong Kong's largest export market. Given this close trade tie and the international status of the US dollar as a reserve currency, it was appropriate for Hong Kong to choose the greenback as the anchor or reserve currency for its currency board. On balance, the choice has so far proved to be the right one, as the link has not only maintained the public's confidence in the Hong Kong dollar but has also to some extent facilitated Hong Kong's exports to the United States as a result of reduced exchange rate volatility.
Nevertheless, there has never been a "one-size-fits-all" currency regime that is right for all countries at all times. From the point of view of the theory of optimal currency areas (Mundell 1961), Hong Kong's linked exchange rate system may have outlived its effectiveness. To date China has already overtaken the United States as Hong Kong's largest trading partner. The "one country, two systems" arrangement notwithstanding, developments in both China and Hong Kong point to continuous integration of the two economies in terms of bilateral flows of not only goods and services but also labor and capital. As its relationship with China is becoming much closer than that with the United States, Hong Kong stands to benefit more from forming a monetary union with the former than with the latter. Therefore, it is time for the linked exchange rate system to be modified to meet changing business and economic environments.2
The Proposed System
It is premature, however, for the current link to completely switch from the US dollar to the renminbi as the anchor currency, because the latter is not yet a freely convertible currency in international foreign exchange markets. This inconvertibility also rules out the possibility of adopting a dual currency board system such as the one put forward by Oppers (2000), under which currency issued by the currency board can be redeemed into one of two reserve currencies (e.g., the US dollar, the Euro).3 In addition to raising potential economic difficulties, replacing the Hong Kong dollar with the renminbi may not be politically feasible, because the Hong Kong dollar has to continue to circulate and be fully convertible in accordance with Article 112 of the Basic Law. Nonetheless, the renminbi can still play a role in modifying the current linked exchange rate system to allow it to accommodate changing needs in the years to come and to facilitate the formation of a long-term monetary union between Hong Kong and China.
Theoretically, a currency board can link the currency of an economy to a basket of foreign currencies. Since currency movements are not perfectly positively correlated, such a link has an advantage in that the external value of the domestic currency is less volatile than it would be if the currency were linked to a single anchor currency. But in practice a single anchor currency is often chosen on grounds of transparency and simplicity in operation. For the proposed monetary reform, the Hong Kong dollar would be linked to a currency basket consisting of the US dollar and the renminbi only.4 Its theoretical foundation is analogous to symmetallism, which is advocated by Marshall (1887), according to whom the currency unit would consist of a specified weight of gold and a specified weight of silver.5 The price of this composite gold-silver currency would be kept fixed, but the individual prices of gold and silver could vary freely. Under Marshall's proposal, the composite currency would mitigate the impact on the general price level that could be caused by supply shocks of one metal alone.
Similar to Marshall's proposal, the current proposal would define the Hong Kong dollar in terms of a currency basket consisting of a specified quantity of US dollars and a specified quantity of renminbi. The official price of the currency basket would be fixed in terms of the Hong Kong dollar, but the individual prices of the US dollar and of the renminbi in terms of the Hong Kong dollar (i.e., the exchange rates) could vary freely. However, the proposal differs from Marshall's in that the weights or the specified quantities of the two currencies would be revised over time, preferably following a preannounced schedule, and would shift gradually from the US dollar to the renminbi in response to economic integration between China and Hong Kong as well as to the emergence of the renminbi as an international hard currency.6
To begin with, the existing linked exchange rate regime with an official exchange rate of US$1 = HK$7.8 can be considered a special case of the proposed system in which the currency basket comprises 0.1282 units of the US dollar only. The official price of one unit of the currency basket is fixed at HK$1. The fixed exchange rate together with full convertibility implies that the Hong Kong dollar serves as a store of value, allowing its holders stable purchasing power in terms of the US dollar.
To transform the existing system, the HKMA should ideally preannounce that on a specified date the renminbi will replace the US dollar such that the purchasing power of the Hong Kong dollar (or the currency basket) will be maintained at a value equivalent to US$0.1282.7 As an illustrative example, suppose the current spot-market exchange rates were US$1 = HK$7.8 and US$1= RMB 8.268 yuan. Further suppose that the HKMA would like to define a currency basket such that 90% of its value was derived from its US dollar component. The given data imply that one currency basket should consist of 0.1154 (i.e., 0.1282 x 0.9) units of the US dollar and 0.1060 (i.e., 0.1282 x 0.1 x 8.268) units of the renminbi. By definition the official price of one currency basket would always be HK$1, regardless of the weight of the US dollar in the basket. Needless to say, given the above quantities and exchange rates, the market price of the currency basket in nominal terms would coincide with its official price. As is shown below and in the appendix, the market and official prices of the currency basket would not always be the same; and arbitrage would arise whenever there were discrepancies.
Although it will not affect the official price of the currency basket, the weight of the US dollar in the basket is not entirely arbitrary. It depends on factors such as the amount of renminbi at the HKMA's disposal, China's schedule for opening its capital account, and the degree of integration between Hong Kong and China, to name just a few factors. Assuming the above-mentioned exchange rates remain unchanged, the currency basket could be revised to consist of, say, a 0.8 weight in the US dollar later on, when it has become more usual to use the renminbi in international transactions. In this case, the currency basket would consist of 0.1026 units of the US dollar and 0.212 units of the renminbi, while its official price would remain intact at HK$1.8 The ultimate currency basket would be made up entirely of the renminbi when the two economies formed a monetary union.
Like the quantities in the currency basket, the roles of the currencies would change over time with respect to increasing circulation and convertibility of the renminbi in international markets. For the purpose of analysis, it is useful to distinguish between the medium of exchange, the unit of account, the medium of account, and the medium of redemption (see, for example, McCallum 1985). Under the proposed exchange rate system, Hong Kong dollar banknotes and coins (or more generally Hong Kong dollar funds, such as demand deposits) would remain the medium of exchange, routinely offered and taken in trade for goods and services. The Hong Kong dollar would also remain the unit of account (i.e., the unit used for pricing, bargaining, and bookkeeping). These two roles would remain the same as they are today, at least until the Hong Kong dollar and the renminbi became perfect substitutes for each other. By definition, the medium of account is the commodity (or a bundle of commodities), some specified quantity of which serves as the unit of account. Consider the United States under the pre-1914 gold standard as a simple illustrative example. Gold was the medium of account, and 0.04838 troy ounce of gold was defined as "the US dollar," which was the unit of account. Gold was also the medium of redemption, as economic agents could redeem their holdings of US dollars into gold at the rate specified above.
For the proposed exchange rate system, the medium of account would be the chosen currency basket comprising the US dollar and the renminbi in different specified quantities, and one unit of the currency basket would be officially defined as one Hong Kong dollar. As already mentioned, the composition of the medium of account would vary over time.
In contrast to the current linked exchange rate system, in which the sole medium of redemption is the US dollar, the HKMA would be obligated to redeem Hong Kong dollar currency into both US dollars and renminbi in quantities as legally specified by the currency basket under the reformed system. Conversely, one currency basket would have to be surrendered by note-issuing banks to the HKMA as reserves whenever they issued Hong Kong dollar banknotes worth HK$1. The existing currency board arrangements would need only minor modifications, and the transmission mechanism would remain largely intact under the proposed system. In principle, changes in the monetary base would still be determined by the balance of payments, and aggregate economic activity would adjust in a fashion similar to that in which it adjusts under the automatic adjustment mechanism of a classical currency board system.
As in the current linked exchange rate system, arbitrage would continue to play a role in maintaining exchange rate stability. Deviations of the market price of the currency basket from the official price would set in motion profit opportunities from arbitrage. The arbitrage process would restore equality of the market price with the official price, and hence equilibrium in the foreign exchange markets. To illustrate, let us suppose that the currency basket was composed of 0.1154 units of the US dollar and 0.1060 units of the renminbi, as previously described. Further suppose that for some reason the market exchange rate of the US dollar rose to HK$7.9, while the other exchange rates remained unchanged. It follows that the market price of the currency basket would be HK$1.0117, higher than its official price. Hence, arbitrageurs would step in to buy currency baskets from the HKMA at the official price of HK$1 per unit; they would break up the baskets and sell the individual currencies on the foreign exchange markets to obtain HK$1.0117, thereby making a profit of HK$0.0117 per basket. The arbitrage would bring down the exchange rates of the US dollar and the renminbi by increasing their supplies, and the process would continue until the market exchange rates had fallen sufficiently to restore the market price of the basket to its official price.9 The appendix provides a more detailed analysis of how the arbitrage process would restore equilibrium.
As a medium of redemption, the renminbi would initially play a less prominent role than the US dollar, simply because the former has not yet established its status as an international hard reserve currency, as has the latter. The amount of renminbi circulating outside the Mainland is estimated to be about 30 billion yuan.10 Even if we assume, rather unrealistically, that this amount of renminbi would all be held by the HKMA as reserves and that the current ratio of foreign currency reserve assets to currency in circulation would remain unchanged, the renminbi would account for less than 5% of the currency basket.
This seemingly small percentage, however, could be raised to enhance the renminbi's role. Theoretically, the reformed scheme would operate in the same way as a currency board, and it presumes that the HKMA has a sufficient stock of currency basket at its disposal to support arbitrage activities. This assumption is reasonable, in spite of the HKMA's limited reserves of renminbi in the very short run. With its considerable foreign currency reserves, the HKMA could raise the weight of the renminbi in the currency basket if it were allowed to use US dollars in lieu of the renminbi for the redemption of Hong Kong dollars when its renminbi reserves were insufficient. Alternatively, the same objective could be achieved as long as the HKMA was able to obtain renminbi loans from the People's Bank of China to support the redemption of Hong Kong dollars (see also Note 10). Neither measure would undermine the credibility of the HKMA to defend the Hong Kong dollar, because convertibility would be guaranteed.
These institutional arrangements, if needed at all, would probably be short-term, transitory measures. In all likelihood, they would be phased out in the longer run because of several factors leading to broader currency of the renminbi. First, the forthcoming qualified domestic institutional investors and qualified foreign institutional investors schemes would enhance portfolio capital flows and increase the openness of China's financial and capital accounts. As a regional financial center, Hong Kong could serve as an offshore financial and clearing center for the renminbi. In particular, the HKMA could have more renminbi at its disposal if it administered clearings of renminbi deposits in the local economy.11 The development of a deep and broad foreign exchange market for the renminbi would also enhance the international acceptability of the currency. The result would be beneficial for both China and Hong Kong.
Second, regional cooperation and vibrant economic developments in the greater Pearl River Delta ¡X encompassing Guangdong, Hong Kong, and Macau ¡X are additional favorable factors. With closer ties in both the financial and real sectors of these economies, it is not difficult to imagine the existence of parallel currencies in Hong Kong and South China in the foreseeable future as a result of higher substitutability between the Hong Kong dollar and the renminbi.12 The increasing regional economic integration is an important reason explaining why the current link of the Hong Kong dollar to the greenback will not be an optimal monetary arrangement in the long run. A common currency for Guangdong, Hong Kong, and Macau should encourage trade and growth within the region by reducing transaction costs and exchange risks.
Third, China's stable exchange rate policy over the years is a positive factor reinforcing the international role of the renminbi. During the Asian currency crisis of 1997-98, China was able to resist mounting pressure and maintain its currency's stable external value against the US dollar, thus averting a war of competitive currency devaluation among East Asian countries. As China has registered a trade surplus for nine years in a row, and as it is one of the largest capital-importing countries in the world, the renminbi is expected to continue its stable long-term trend of appreciation that has been in affect since 1994. These track records undoubtedly encourage the demand for the renminbi, particularly from politically unstable neighboring countries.13
Last but not least, with its accession to the World Trade Organization China is further transforming itself into not only a world factory but also a world market for consumer goods and services. This increasing economic power is decidedly conducive to the development of the renminbi into an international currency in the long run, as evidenced by the experiences of industrialized countries like the United States, Japan, Germany, and the United Kingdom. In sum, the factors discussed above point to a trend of internationalization of the renminbi. Thus, the renminbi is expected to play an increasingly important role as an anchor currency in Hong Kong's currency board system over time.
In the long run, the Hong Kong dollar could be entirely linked to the renminbi as the sole anchor currency when the latter becomes a freely convertible currency like the US dollar or the Japanese yen. And when Hong Kong's economy is fully integrated with China's, the two parallel currencies are expected to be perfect substitutes for each other and to circulate at par in Pearl River Delta. In addition to its role as a medium of redemption, the renminbi could function as a medium of exchange and a unit of account in Hong Kong. Under these circumstances, Hong Kong would have virtually formed a monetary union with China even though the circulation of Hong Kong dollars might continue as stipulated by the Basic Law.
The new system proposed in this paper has several potential advantages over the current linked exchange rate system. As already mentioned, the individual exchange rates of the US dollar and the renminbi vis-à-vis the Hong Kong dollar are flexible under the proposed system. Theoretically, the Hong Kong dollar could therefore depreciate against the US dollar to improve Hong Kong's competitiveness in the global market.14 The depreciation of the Hong Kong dollar could also be attainable through the monetary policy of the People's Bank of China if the Hong Kong dollar were fully backed by the renminbi or if a monetary union were formed.15 This also suggests another potential advantage of the new system ¡X the possibility of tackling macroeconomic problems through monetary policy, which is totally absent under a classical currency board system.
The benefits described above pale, however, when compared with the fact that the proposed scheme is more consistent than the current linked system with trends in economic developments in Hong Kong and China. There is no need to explain in detail why parallel currencies or a monetary union is better than the current monetary arrangements for promoting trade and growth in the region. One can simply imagine what would happen if the US dollar circulated everywhere in the United States except in New York City, which issued its own New York dollar or used the Japanese yen as its currency.
A less obvious but closely related potential benefit of the proposed system is that it might reduce the probability that the Hong Kong dollar will suffer a currency crisis at some point in the future. As historical evidence indicates, currency crises are largely due to inconsistency between exchange rate regimes and economic fundamentals (Bordo and Schwartz 1996). The longer the current linked exchange rate regime remains, the more likely it is to become inconsistent with current underlying economic development trends. Maintaining the current link is often perceived as a signal of the HKMA's commitment to defending the Hong Kong dollar. But it could also be interpreted, or perhaps misinterpreted, as the HKMA's failure to overcome the drawbacks of the existing system. Reform of the exchange rate regime in the "right" direction could reinforce the public's faith in the HKMA's commitment to maintaining monetary stability.16 Financial markets are difficult, if not impossible, to predict. Hence there is a risk, no matter how small, that any proposal for, or even discussion of, reforming the current exchange rate regime could trigger rumors and speculative attacks on the Hong Kong dollar. Unfortunately, the absence of such discussion or proposals does not eliminate the risk of speculative attacks either, because rogue speculators can always test the credibility of the HKMA by initiating rumor-based speculative attacks.
Much ink has been spilled over Hong Kong's linked exchange rate system since the Asian currency crisis. Many studies, however, focus on proposals to protect the Hong Kong dollar from speculative attacks. Recently, the discourse on monetary reform seems to have narrowed its focus to two polar cases only (i.e., to maintain the link or to float the Hong Kong dollar). Many other possible solutions may have been overlooked.17 This article offers a proposal to reform the current linked exchange rate system to meet Hong Kong's changing needs. It covers mainly monetary reform. Many operational details, such as the proportion of the two reserve currencies in the currency basket and the timetable for revising this proportion, remain to be worked out. The proposal may turn out to be infeasible. Nonetheless, this article's main objective has been achieved if it has drawn the attention of economists and policymakers toward exploring other possible exchange rate arrangements for Hong Kong.
Bordo, Michael, and Anna J. Schwartz (1996) "Why Currency Clashes Between Internal and External Stability Goals End in Currency Crises, 1797¡V1994," Open Economy Review 7: 437: 468.
Chan, Kenneth S. (2002) "Currency Substitution Between the Hong Kong Dollar and the Renminbi in South China," Pacific Economic Review 7: 37¡V30.
Fisher, Irving (1920) Stabilizing the Dollar: A Plan to Stabilize the General Price Level without Fixing Individual Prices, New York: Macmillan Company.
Friedman, Milton (1951) "Commodity Reserve Currency," Journal of Political Economy 59: 203¡V32.
Goodhart, Charles (1988) The Evolution of Central Banks, MIT Press.
Greenfield, R., and L. Yeager (1983) "Laissez-Faire Approach to Monetary Stability," Journal of Money, Credit and Banking 15: 302¡V15.
Jao, Y.C. (1998) "The Working of the Currency Board: The Experience of Hong Kong 1935¡V1997," Pacific Economic Review 3: 219¡V41.
Kasa, Kenneth (1999) "Why Attack a Currency Board?" FRBSF Economic Letter 99¡V36, Federal Reserve Bank of San Francisco.
K ves, Andr<s (1992) Central and East European Economies in Transition: The International Dimension, Boulder: Westview Press.
Lawrence, Susan (2003) "The Renminbi Zone," Far Eastern Economic Review (May 29, 2003): 24¡V28.
Marshall, Alfred (1887) "Remedies for Fluctuations of General Prices," reprinted in A.C. Pigou (ed.) Memorials of Alfred Marshall, London: Macmillan and Company.
McCallum, Bennet T. (1985) "Bank Deregulation, Accounting Systems of Exchange and the Unit of Account: A Critical Review," Carnegie-Rochester Conference Series on Public Policy 23: 13¡V46.
Melvin, Michael (2000) International Money and Finance, 6th edition, Addison-Wesley Longman.
Mundell, Robert (1961) "A Theory of Optimum Currency Areas," American Economic Review 51: 657¡V665.
Oppers, Stefan Eric (2000) "Dual Currency Boards: A Proposal for Currency Stability," IMF Working Paper WP/00/199, Washington, D.C.: International Monetary Fund.
1 As of the end of 2002, foreign currency reserve assets held by the Hong Kong government totaled US$111.9 billion, or seven times the number of Hong Kong dollars in circulation.
2 A few years ago, Jao (1998) had already pointed out the suboptimality of the current peg to the US dollar because of the divergence in the cyclical movements of the two economies.
3 In the dual currency board system, if the payee's choice of foreign currency, say, the US dollar, in return for his domestic currency is not available, he is nevertheless guaranteed payment in the other reserve currency from the currency board. See Oppers (2000) for more details. As far as Hong Kong's case is concerned, a dual currency board system is inapplicable as long as the renminbi (as one of the reserve currencies) is not convertible and seems unnecessary if the renminbi is fully convertible. Nevertheless, as is shown below, our proposal also makes use of the notion of convertibility subject to the availability of the renminbi.
4 Alternatively, a trade-weighted currency basket consisting of the US dollar, the renminbi, the yen, and so forth can be chosen. Such a currency board system has the advantage of minimizing external shocks from a particular trading economy, but it does so at higher transaction costs and with less simplicity and lower transparency in operation when compared with the case of a single reserve currency.
5 Subsequent extensions of Marshall's symmetallism to the case of more than two commodities include the "compensated-dollar" scheme (Fisher 1920), the commodity reserve standard (Friedman 1951) and a modern commodity basket proposal commonly known as the Black-Fama-Hall system (Greenfield and Yeager 1983), among many others. These schemes are not discussed here because of space limitations.
6 It should be emphasized that allowing the weights of the currency basket to change over time does not imply a switch from a rule-based regime (i.e., the currency board) to discretionary monetary policymaking. The proposed system is still a rule-based system. The operations of the system are rule based, as the system is essentially a modified currency board system. Changes in the weights of the currency basket can still be rule based so long as the HKMA changes the weights in a period (say, a year) by following a rule or formula design that is applicable within the planning horizon (i.e., a large number of periods or years). The pre-announced rule and schedule for changes do not necessarily mean risk-free profit opportunities for speculators. See explanations below, especially Notes 7 and 17.
7 Preannouncement seems to be more desirable than a sudden uninformed change, because the latter might undermine the HKMA's credibility and hence induce currency speculation in future. See also the discussion on this point in Note 17. By the same token, subsequent changes in the weights of the currency basket should preferably follow a preannounced schedule. As long as the HKMA's commitment to maintaining a stable Hong Kong dollar remains credible, such preannounced changes by themselves should not trigger currency speculation, because the weights are calculated in such a way that the value of the Hong Kong dollar in terms of the US dollar remains intact regardless of the change in weighting. Undeniably, it is difficult, if not impossible, to predict how economic agents will change their expectations in response to regime changes. However, the experience of the European Monetary Union suggests the feasibility of a smooth and gradual regime change following a preannounced schedule.
8 The actual quantities of the two currencies in the basket would also depend on their exchange rates in terms of Hong Kong dollars at the time of revision.
9 This process presumes that the HKMA, the de facto central bank in Hong Kong, has enough currency baskets at its disposal to support the arbitrage activity. This condition can be fulfilled with the support and cooperation of the People's Bank of China, the central bank of China. On the one hand, the combined foreign currency reserve assets, mostly in US dollars, held by the two central banks amounted to over US$400 billion as of the end of 2002. On the other hand, the People's Bank of China could provide the HKMA with the amount of renminbi required for supporting the arbitrage activity.
10 This estimate is based on an article in a local newspaper, Ming Pao, dated March 7, 2003. It may be an underestimate, as Lawrence (2003) reports that renminbi worth $3.8-5.1 billion is in circulation in Hong Kong.
11 There are several advantages in having the clearing function taken up by the HKMA instead of by a commercial bank. First, if the clearing bank was a commercial bank, there could be potential or perceived conflicts of interest between the clearing bank and other commercial banks, but this could be avoided in the case of the HKMA as the clearing bank simply because it is not a commercial bank (see for example Goodhart 1988). Second, the centralization of renminbi reserves at the HKMA means that the HKMA could have more renminbi at its disposal to support arbitrage activities in the currency baskets. Third, with its renminbi reserves the HKMA could maintain a renminbi balance with the People's Bank of China. On the one hand, the reserves could be used as collateral to obtain renminbi loans from the People's Bank to support the arbitrage activities. On the other hand, the clearing bank would offer renminbi deposits collected in Hong Kong a temporary outlet, at least before China opens up its domestic loan markets. Finally, it would be "natural" and least costly if all the central banking functions were held by the HKMA, which would one would expect to be converted into a regional branch of the People's Bank of China (like the U.S. Federal Reserves System) later on, when a monetary union was formed.
12 There is already some basis for the formation of parallel currencies within the region. For instance, the Hong Kong dollar has been generally accepted as a medium of exchange for a long time in Macau. This acceptance was extended to Guangdong following the setup of the Shenzhen Special Economic Zone in the early 1980s. In a recent econometric study, Chan (2002) estimates that about 3.2% and 7.4% of the total Hong Kong dollar currency issued was circulating in Macau and Guangdong, respectively. Reciprocally, the renminbi has been increasingly accepted for transaction purposes in Hong Kong as a result of increasing numbers of tourists from China coming to Hong Kong in recent years.
13 See also the recent article by Lawrence (2003).
14 Although theoretically possible, this potential benefit should not be exaggerated. First, when the Hong Kong dollar starts to depreciate against the US dollar, all other things equal, the arbitrage process will tend to offset the extent of depreciation. Second, and more important, Hong Kong's international competitiveness depends on the value added and quality of its products more than on their prices.
15 The current exchange rates of US$1 = HK$7.8 and US$1 = RMB 8.268 yuan suggest that there is room for the Hong Kong dollar to depreciate while the renminbi appreciates, both against the US dollar, at the same time before the two exchange rates converge to, say, US$1 = RMB 8 yuan and HK$1 = RMB 1 yuan. After a currency union had formed and the Hong Kong dollar was fully backed by the renminbi, the Hong Kong dollar would follow the renminbi to depreciate against the US dollar if China deemed depreciation a desirable exchange rate policy.
16 Viewing the evolution of economic institutions as a spontaneous discovery process, this author would definitely be pretentious to claim ex ante that this proposal is the "right" reform. It can also be assumed that this proposal would also not be "right" if merely discussing it prematurely triggered rumor-based speculative attacks on Hong Kong and led to a currency crisis.
17 Another possible solution is simply for the HKMA to suddenly announce a regime change by repegging the Hong Kong dollar to the renminbi when the latter has become fully convertible. This solution may be preferred to the proposal presented here because it is simpler to implement. However, this strategy has at least a couple of potential problems. First, the HKMA could use it only once because after doing so it would lose much if not all of its credibility. This might not be a problem if there was no other repegging in the future, say, if the Hong Kong dollar and the renminbi became perfect substitutes for each other. Second, and more important, it is unclear how long it will take for the renminbi to become a fully convertible international reserve currency. The waiting cost for Hong Kong could be high, especially if there were speculative attacks on the Hong Kong dollar during the interim. Empirically, the lessons from many emerging economies indicate that gradualism or a gradual approach to systemic transformation is more likely to be successful than a sudden change or shock therapy. An example of some relevance to our case here is German unification in 1990, in which East Germany adopted the deutsche mark as its currency overnight, and its economy suffered as a consequence of a revaluation shock. While West German financial support made the shock therapy manageable, as K ves (1992) argues, the short-term shock to East Germany could have been lessened if a gradual approach to economic and monetary union had been adopted instead.
Acknowledgments: This article is a revised version of a paper written when the author visited the Hong Kong Institute of Economic and Business Strategies, The University of Hong Kong, in early 2003. The author is grateful for the hospitality of the staff of the Institute and would like to thank Alex Chan, Y.C. Jao and Alan Siu for their comments on earlier versions of this paper. The views in this paper do not necessarily reflect theirs or those of the Institute. As usual, the author is solely responsible for all errors.
Chu Kam Hon is Associate Professor at the Department of Economics, Memorial University of Newfoundland, St. John's, Newfoundland & Labrador, Canada A1C 5S7.
Appendix: Analysis of Arbitrage under the Proposed Exchange Rate Regime
This appendix discusses the arbitrage process and equilibrium determination under the proposed exchange rate regime. It begins with a brief theoretical explanation of how equilibrium in foreign exchange markets is characterized under the proposed exchange rate regime, which is followed by a simple hypothetical numerical example to illustrate how the arbitrage process would work to restore equilibrium.
Algebraically, an equilibrium in foreign exchange markets under the proposed regime has to satisfy the following two equations simultaneously:
(1) QUS$ x SHK$/US$ + QRMB x SHK$/RMB = 1, and
(2) SUS$/HK$ x SHK$/RMB x SRMB/US$ = 1,
where QUS$ is the quantity of the US dollar in the currency basket,
QRMB is the quantity of the renminbi in the currency basket,
and Sij denotes the spot exchange rate of the ith currency defined in terms of the number of that currency per unit of the jth currency.
For example, SUS$/HK$ is the spot exchange rate of the US dollar per unit of the Hong Kong dollar, say, US$0.1282 /HK$. Other exchange rates are defined in the same fashion.
The first equation is based on the definition of the currency basket and the commitment of the HKMA to converting Hong Kong dollars into the specified quantities of US dollars and renminbi. In equilibrium the nominal value of the currency basket in terms of the Hong Kong dollar would be equal to one Hong Kong dollar; otherwise an arbitrage opportunity would exist. The second equation is the familiar triangular (three-point) arbitrage condition presented in the international finance literature. Similarly, if Equation (2) did not hold with equality, in theory, an arbitrage opportunity would exist (see any standard textbook in international finance; for example, Melvin 2000). In both equations, we have assumed no transaction costs involved in arbitrage.
Suppose the quantities of the US dollar and the renminbi in the currency basket were pre-specified. This parsimonious model by itself cannot determine a unique equilibrium because there are three unknown variables (the exchange rates) but only two equations (the arbitrage conditions). There are multiple equilibria, and any combination of the three exchange rates that satisfies the two equations simultaneously can possibly be an equilibrium.
Under the proposed exchange rate regime, the HKMA would have to stand by and convert any amount of Hong Kong dollars into an equivalent quantity of the currency basket, or vice versa. However, the HKMA would not be obligated to commit to convertibility between the US dollar and the renminbi. The renminbi-US dollar exchange rate could continue to be determined by China's exchange rate policy as it stands under the current regime, or by market forces beyond the influence of the HKMA if the renminbi were finally fully convertible. From this perspective, the spot exchange rate of the renminbi in terms of the US dollar can be viewed as an exogenous variable, whereas the equilibrium exchange rates of the US dollar and the renminbi, both in terms of the Hong Kong dollar, can be regarded as endogenously determined by the institutional settings.
Hence solving the above simultaneous-equation system yields the unique spot exchange rates of the Hong Kong dollar in terms of the other two currencies in equilibrium:
(3) SHK$/US$ = 1/ (QUS$ + QRMB / SRMB/US$ ), and
(4) SHK$/RMB = 1/ (QRMB + QUS$ x SRMB/US$).
In words, the values of the spot exchange rates of the Hong Kong dollar in equilibrium depends on the quantities of the US dollar and the renminbi in the currency basket as well as on the renminbi-US dollar exchange rate.
Let us now illustrate how the arbitrage process and China's exchange rate policy would determine the equilibrium exchange rates of the US dollar and the renminbi vis-à-vis the Hong Kong dollar with a simple hypothetical numerical example.
To begin with, assume initially the equilibrium spot exchange rates to be US$1 = HK$7.8, RMB1 yuan = HK$0.9434, and US$1 = RMB 8.268 yuan. It can be verified that these rates satisfy Equation (2) and that no profit opportunities arise from buying and selling these three currencies at the same time in the foreign exchange markets. Further suppose that the HKMA defined the currency basket such that the quantity of US dollars accounted for 90% of the value of the basket in terms of the Hong Kong dollar. Given the above exchange rates, one currency basket thus consists of 0.1154 units of the US dollar and 0.1060 units of the renminbi. From Equation (1) it can be easily shown that there would be no profit opportunities from converting Hong Kong dollars into the currency baskets, or vice versa, because the market value of the currency basket is HK$1, the same as its official price.
Now suppose, for one reason or another, that the Hong Kong dollar depreciated in the open market against its US counterpart, and as a result the spot exchange rate was US$1 = HK$7.9, while the other exchange rates are assumed to remain intact. As is shown below, such a change in the exchange rates would create arbitrage profit opportunities. In practice, arbitrageurs would convert Hong Kong dollars into the currency baskets and simultaneously trade the currencies in the spot foreign exchange markets to make profits. For purposes of analytical tractability and illustration, let us assume that these transactions can be separated.
We consider first how the foreign exchange markets would respond to such a disturbance without taking into consideration the convertibility of the Hong Kong dollar. It can be easily verified that there are profit opportunities from arbitrage in the foreign exchange markets, as Equation (2) no longer holds with equality. More specifically, an arbitrageur could first buy RMB 1.06 yuan with HK$1 at the exchange rate of RMB1 yuan = HK$0.9434, then sell this amount of renminbi for US$0.1279 at the exchange rate of US$1 = RMB 8.268 yuan, and finally convert the greenback at the new exchange rate of US$1 = HK$7.9 back into HK$1.0106 to net a profit of HK$0.0106. The first transaction would cause the Hong Kong dollar to depreciate against the renminbi, the second would put pressure on the renminbi to weaken against the US dollar, and the last would induce the Hong Kong dollar-US dollar exchange rate to fall, partly offsetting its initial rise. These effects are summarized in Table 1. The process of triangular arbitrage would continue until no more arbitrage profits could be further exploited.
On the other hand, the initial increase in the Hong Kong dollar-US dollar exchange rate to US$1 = HK$7.9 would cause the market value of the currency basket to increase to HK$1.0117 (i.e., from Equation (1), we have QUS$ x SHK$/US$ + QRMB x SHK$/RMB = 0.1154 x 7.9 + 0.106 x 0.9434 = 1.0117). This implies arbitrage profit opportunities by converting Hong Kong dollars into currency baskets. Arbitrageurs would sell Hong Kong dollars to the HKMA for currency baskets, break up the baskets, and sell the individual currencies in the foreign exchange markets to obtain a profit margin of HK$0.0117 per Hong Kong dollar. The increase in supplies of the US dollar and of the renminbi would cause their exchange values in terms of the Hong Kong dollar to fall. The process would come to an end when no arbitrage profits could be made as a result of exchange rate adjustments. The impacts on the exchange rates are also summarized in Table 1.
As can be seen from Table 1, in this hypothetical example the combined net effect of triangular arbitrage and the conversion of Hong Kong dollars would be a fall in the Hong Kong dollar-US dollar exchange rate, offsetting its initial increase. Meanwhile, the net impact on the renminbi would be a depreciation against the US dollar, whereas the net effect on the renminbi-Hong Kong dollar exchange rate would be ambiguous.
The analysis so far provides some information about the directions of changes in the exchange rates but does not tell us exactly what the final equilibrium exchange rates would be in this example. In other words, the Hong Kong dollar-US dollar exchange rate would fall from 7.9, but its final equilibrium value could be 7.85, 7.8, or any other figure. The precise value cannot be determined a priori unless further information is gleaned or further assumptions are made. The final equilibrium depends on how the arbitrageurs readjust their portfolios.
However, if China maintained a stable exchange rate policy such that it restored the initial exchange rate of US$1 = RMB 8.268 yuan, then the exchange rates of the US dollar and the renminbi in terms of the Hong Kong dollar would return to their initial levels of US$1 = HK$7.8 and RMB 1 yuan = 0.9434, according to our theory's prediction as from Equations (3) and (4). Consequently, the proposed scheme and China's stable exchange rate policy would work together to stabilize the external value of the Hong Kong dollar.
Table 1: Effects of Arbitrage on the Exchange Rates
Effect Due to
HK$ per US$
HK$ per RMB
RMB per US$
Conversion of HK$
Note: Since the exchange rates are defined in terms of the number of local currency (say, HK$) per unit of foreign currency (say, US$), a decrease (increase) in the exchange rate means a stronger (weaker) local currency.