Flexible, fixed, or a middle ground?

Regionally integrated economies need to balance trade-offs in the choice of their exchange rate regime vis-à-vis one another.¹⁸

• On the one hand, predictable external values of national currencies within the region facilitate intraregional trade, simplify economic decisions, and promote cross-border holdings of local securities. A system of stable intraregional exchange rates could also ease an eventual transition to a common currency.

• On the other, international capital mobility makes defending intraregional currency pegs difficult, absent close convergence of policies and strong market confidence in the system. In fact, in the “impossible trinity” view, an economy can have only two of the following: an independent monetary policy, a fixed exchange rate, and capital account openness. In the textbook version, a monetary loosening to support GDP growth, for example, would trigger incipient capital outflows that would put downward pressure on the exchange rate peg and lead to an unsustainable drawdown of official reserves.¹⁹ Something has got to give.

Capital controls do not offer a durable way out of the dilemma. A margin for policy maneuver can perhaps be reconstituted by recognizing that capital account openness is not an all-or-nothing proposition: capital flows can be managed through capital controls. While attractive as a tactical solution, this approach has limitations in practice. Capital controls may provide temporary “breathing space” for the pursuit of domestic policy objectives but their long-term effectiveness is questionable in sophisticated global financial systems. More importantly, barriers to capital mobility entail costs in terms of a less efficient allocation of international savings and the foregone benefits of the diversification provided by unencumbered trade in assets. Recognition of these costs motivates, in fact, the wide-ranging initiatives underway to liberalize financial markets in Asia.

Intraregional exchange rate flexibility is a more promising strategic alternative. Exchange rate systems that do not present one-way bets on the external value of a currency eliminate the rationale for speculative attacks. As a result, they are consistent with steps to encourage intraregional capital flows while safeguarding the ability of policy to pursue domestic objectives. There is a downside, though: excessive exchange rate volatility could hamper trade and complicate investment decisions by heightening uncertainty. A clear commitment to sound macroeconomic policies and the development of forward markets to hedge exposure to exchange rate risk would go a long way toward limiting these undesirable side effects. So would two-way intervention in the foreign exchange market that dampens exchange rate volatility at high frequencies but does not undermine the role of exchange rate movements in the longer-term adjustment to shocks or structural changes.²⁰

In sum, while trade considerations may favor greater fixity in exchange rates to lower transaction costs, steps to promote capital mobility point in the other direction. Exchange rate systems that allow exchange rates to operate as shock absorbers offer the best hope to resolve this tension. Rare exceptions notwithstanding (Hong Kong SAR, for one), greater exchange rate variability seems unavoidable when the capital account becomes more open.²¹ So, greater flexibility in exchange rate management is a key element in a strategy to foster regional integration: it allows economies to share regional savings and expand trade, without giving up the freedom to pursue domestic objectives at the same time. This is indeed the direction in which emerging Asia has been moving: exchange rate flexibility vis-à-vis both regional and G-3 currencies has been increasing in most cases over time.

Are currency basket pegs an alternative?

It has been suggested that currency basket pegs could also be an option for regionally integrated economies in emerging Asia. There are two versions to this proposal—one involves each country pegging its currency to a common basket of G-3 currencies, the other would have each country choosing an appropriately weighted basket of currencies including those of intraregional partners. In the first approach, the main focus is to ensure exchange rate stability vis-à-vis the rest of the world and intraregionally. In the second, the main goal is to better capture country-specific features and limit intraregional exchange rate variability to promote trade and investment. Both approaches have been also put forth as possible intermediate regimes on the road to monetary unification.

The rationale for a collective basket peg against G-3 currencies lies in the view that predictable exchange rates against the dollar, the euro, and the yen would facilitate export growth, a pillar of the Asian development model. Common weights in the national currency baskets would also limit intraregional fluctuations in effective exchange rates. In one proposal reminiscent of the (pre-euro) European Monetary System (EMS), a country’s exchange rate would be allowed to float within a band around a predetermined nominal value of a basket of currencies. Intervention—perhaps financed by a common pool of reserves—would prevent the exchange rate from straying further. Gradual realignments of the central parity would be possible in case of a significant drift in the equilibrium exchange rate.²²

Critics of this “basket-band-crawl” arrangement question the premise that export-led growth should continue to play a central role in Asia’s development.²³ They argue that productivity growth (which can only indirectly be steered by monetary arrangements), and not the level of nominal wages, holds the key to rising living standards. On this view, exchange rate stability may be less important for growth, going forward. Furthermore, critics note credibility issues related to the adoption of “soft margins” in the target band as well as the political difficulty of re-aligning central parities when the underlying determinants of the equilibrium exchange rate change.²⁴ More broadly, the precedent of the EMS—which was supported by pan-regional political, economic, and financial commitments—is generally seen of limited relevance for emerging Asia where these pre-conditions are still absent. The robustness of a regional exchange rate system depends on shared economic structures, policies, and shocks among participating countries. For this reason, a regional grid of fixed exchange rates would seem at odds with emerging Asia’s economic and institutional diversity. A case in point is that trade patterns vary substantially within the region and some countries would be disadvantaged under a common peg.²⁵

A variant of the G-3 currency basket proposal focuses on the advantages of a basket—tailored to each country—that includes both external and regional currencies. Such a basket arises, for example, if trade weights are used in choosing its composition. By construction, it could not be a common basket.²⁶ In principle, a system of country-specific basket pegs would insulate the trade relations of the region from outside disturbances (as the common-basket proposal does) but not fully eliminate intraregional instability in bilateral rates. Whether customized baskets would be superior in stabilizing real exchange rates depends on the relative variability of the external values of the constituent currencies.²⁷ A further advantage of country-specific baskets pegs is that they may be better suited to a context of evolving intraregional trade patterns and could represent an interim step toward the adoption of a common basket as real convergence among Asian economies proceeds.²⁸

Currency basket pegs tailored to individual members of an integrated region suffer from many of the shortcomings of a common-basket scheme. In addition, there may be heightened difficulties in deciding the appropriate weighting scheme for the component currencies. Should weights be based on import, exports, or total trade? Should they reflect the direction of trade or its currency composition? Should financial flows be considered in the determination of the appropriate weights? Beyond these technical issues, there remains the concern that limited exchange rate flexibility as implied by either sort of basket pegs carries the risk of fundamental inconsistencies between internal or external policy objectives. The “impossibility trinity” casts a shadow on all these proposals.

However implemented, currency pegs would in the end constrain the scope for an independent monetary policy—the more so the more mobile capital is internationally. Preserving the viability of a multilateral grid of pegged exchange rates potentially comes at the expense of domestic price, output, or financial stability. This was what happened during the 1992–93 turmoil in European exchange rate markets that shook the EMS’ Exchange Rate Mechanism (ERM)—the system of intraregional currency basket pegs of the time. Insufficient convergence of inflation rates and cyclical divergences between Germany and many other European countries fanned massive speculative attacks on the weakest ERM currencies. In the often-futile attempt to defend existing parities, interest rates increased in the countries under attack, in most cases in conflict with the domestic objective of supporting activity in an environment on increasing economic weakness.²⁹ Regional integration may in the end be held back if countries are forced to trade off domestic stability for deeper trade linkages. More broadly, the adoption of currency basket proposals would involve a change from the current systems to which economic agents have already adapted. Barring indisputable evidence of their superiority, a policy change could undermine hard-won credibility. Overall, limiting intraregional flexibility through various forms of basket pegs seems a sensible option only in the very last stages of a transition to monetary unification.

Is there a role for an Asian Currency Unit?

The view that a currency basket peg could strengthen intraregional integration—and ultimately pave the way for a regional currency—is related to proposals for establishing a basket of regional currencies as a unit of account.³⁰ In the Asian context, an Asia Currency Unit (ACU) is in the works. Some observers have noted that the ACU could play the role of the European Currency Unit (ECU) as a stepping stone to monetary unification.

Like the ECU, the ACU would be defined as fixed number of units of the constituent (possibly ASEAN+3) currencies. Thus, the contribution of each individual currency to the value of the ACU would fall or rise with its exchange rate. In principle, official steps such as the denomination of portion of official reserves at participating central banks in ACU or the commitment of participating governments to issue ACU bonds, could make it more attractive for financial and non-financial firms to issue and accept ACU-denominated liabilities and assets.³¹ ACU-bonds could offer the benefit of currency diversification: if one constituent currency lost value against, say, the investor’s home currency, the impact would be limited by its weight in the basket. In time, the ACU could become an invoice currency in trade, gaining the status of a medium of exchange as well as that of a store of value and a unit of account.

Be that as it may, history shows that the ECU played a limited part in facilitating the launch of the euro (see for example Eichengreen (2005)):

• Although the ECU was established in 1974 as a unit of account for European Community institutions and in 1979 as the settlement currency for interventions in the EMS, it never acquired a significant role in either capacity.³²

• The bulk of European transactions continued to be conducted in national currencies until the euro was introduced. “Network externalities” weighed heavily against the widespread usage of the ECU: the benefits of adopting a new money depend on other agents doing the same, but there is no incentive to be the first-mover. With no easy way to resolve this collective action problem, the status quo tends to get locked in.³³

• With a strong predominance of the U.S. dollar in intra- and extra-European commercial and financial transactions, the need for a new regional currency for invoicing and settlement purposes was weak all along.

• As for the ECU’s modest success as a unit of account for bonds, it owed much to currency-plays in the run-up to unification and regulatory barriers to the internationalization of the Deutsche mark. Once these barriers fell, so did the demand for ECU-claims.³⁴

Overall, the European experience seems to provide a cautionary tale on the role that the ACU could play in promoting regional integration in Asia. Network externalities and the incumbency of the U.S. dollar as the main invoice currency in intraregional trade are likely to militate strongly against a rapid acceptance of an Asian parallel currency. Further headwinds may arise from the risks to financial stability as banks, firms and households take on ACU-denominated claims. For example, currency mismatches may reappear on banks’ books if the growth of ACU-liabilities and assets is not kept in step—the more so, if lack of exchange rate flexibility within the region weakens incentives to hedge open positions.³⁵ Tighter supervision and reserve requirements may mitigate these dangers, but there may be a price to pay in terms of forgone intermediation.

The ACU could play a part in the development of regional bond markets, but the benefits may be limited. Dammers and McCauley (2006) underscore the attendant trade-offs. In particular:

• While retail investors may benefit from the pre-packaged diversification provided by basket bonds, institutional investors are less likely to find in the basket weights a close match for the currency distribution of their liabilities or the preferred mix of foreign exchange risk exposure.

• For issuers, the availability of currency swaps to transform liabilities in one currency into liabilities in another may also weaken interest in basket bonds.

• The need for underlying contracts to stipulate contingencies in the event of changes in the official composition of the ACU adds a layer of legal complexity that may deter widespread acceptance of ACU-denominated bonds.

• Finally, government issuance of ACU-bonds may take a toll on local bond market liquidity to the extend that it diverts issuance away from national markets or undermines the benefits of large benchmarks. Liquidity may also be adversely affected by the fact that basket bonds tend to attract buy-and-hold investors who want to lock in the diversification benefits they provide.

What monetary framework?

Unlike a commitment to a fixed exchange rate, freer floating of exchange rates is not an operating strategy for monetary policy. The presumption that greater flexibility in exchange rate systems will support and facilitate regional integration raises the question of what is then an appropriate supporting monetary framework. A popular strategy to accompany (more) flexible exchange rates is a central bank’s commitment to the pursuit of price stability.

Price stability, defined qualitatively or operationalized as a target for the inflation rate, has gained broad acceptance as a key goal for monetary policy. A formal inflation-targeting regime requires an institutional commitment to price stability (appropriately defined), mechanisms to ensure accountability of the monetary authority, announced targets for inflation as well as a possible horizon for achieving these targets, and regular reporting to the public and the markets of the rationale for monetary policy decisions. Taken together, these elements define a regime of constrained discretion—balancing the need for flexibility in implementation with the need for a credible commitment to domestic price stability. In other expressions of this commitment, the balance between flexibility and credibility may be made operational in less formal ways.

Country-specific factors will determine the case for adopting inflation targeting, broadly defined. A history of weak central bank independence, reliance on commodity exports, and relative closeness to international trade, all increase the attractiveness of a formal inflation-targeting regime.³⁶

• A formal arrangement in which the government sets a clear mandate for the central bank (the inflation objective) but leaves up to the bank the choice of the policy instruments lends credibility to monetary policy makers who might have lacked before operational independence.

• Commodity exporting countries are vulnerable to wide swings in the terms of trade, which make it difficult for monetary policy to maintain a fixed exchange rate. Inflation targeting would be for them a more attractive monetary framework.³⁷

• Finally, openness to international trade imposes by itself a measure of discipline on the conduct of monetary policy, since any attempt to boost the economy by laxer monetary settings would trigger an exchange rate depreciation and a surge in inflation. On this account, more open economies would have less of an incentive to embrace inflation targeting. Other considerations, however, such as the need to limit the domestic impact of external shocks by adopting a flexible exchange rate regime, strengthen the case for inflation targeting.

In line with these stylized facts, Asia has been moving away from monetary frameworks that favored an exchange rate target toward those that focus on the pursuit of price stability. Several countries in the region are explicit inflation targeters. Most other countries have price stability as a key goal for monetary policy. The few economies that still manage their exchange rates heavily outside of a currency board arrangement seem to be in a transitional stage and steps are being taken to enhance the flexibility of their exchange rate systems. Overall, the transition to inflation-centered monetary frameworks underscores the close connections between deepening trade and financial linkages in the region, on the one hand, and the need for a monetary framework that anchors price and wage expectations, on the other. Flexible exchange rates and monetary decisions geared to the achievement of price stability can—together—contribute strongly to intraregional cohesion.

What is the role of the exchange rate in inflation targeting regimes?

A commitment to price stability or the adoption of an inflation target does not imply “benign neglect” of the exchange rate. Exchange rates convey information about future movements in prices and output, particularly so in highly open economies. A central bank that, like many in East Asia, tries in practice to minimize deviations of inflation and output from their targets will in general respond to movements in the exchange rates. Its response should of course reflect the anticipated impact of the exchange rate movement on output and prices. It may also be guided by considerations such as the nature of the shock affecting the exchange rate or the desire to limit interest rate volatility. The important point is that inflation targeting does not mean ignoring the exchange rate but rather that the entire operating strategy for monetary policy is not organized around a target level (or range) for that rate.

In fact, a floating exchanger rate takes on a new and useful role in monetary frameworks centered on price stability. For instance, a rapid depreciation may indicate the risk of an inflationary surge, signaling the need for monetary tightening. More generally, sustained exchange rate movements can provide early warnings of underlying economic or financial imbalances that warrant a policy response.

Does regional integration facilitate the conduct of monetary policy? And can monetary policy coordination facilitate regional integration?

As intraregional trade and financial ties grow, the business cycles in individual economies become better synchronized. Country-specific shocks are dissipated throughout the region; and output and inflation tend to co-vary more across national borders. In this setting, the monetary policy requirements of different economies also become similar: the likelihood of contemporaneous shifts in the monetary stance in the same directions across regional economies (e.g., toward tightening) increases. Policy becomes better harmonized—and its intraregional spillovers potentially less disruptive.³⁸ For example, with interest rates moving together, volatility in intraregional exchange rates will diminish, limiting the impact of the policy moves on intraregional trade.

Monetary policy coordination may in turn aid regional integration. For example, the choice of a common inflation target would speed up the convergence of national inflation rates to a common level. As nominal and real convergence advances, greater policy coordination becomes critical for an eventual monetary unification.

How does regional integration affect fiscal decision making?

A first issue is whether economic cohesion among open and financially interconnected economies constrains the scope for fiscal stabilization at the national level. The standard view holds that intraregional exchange rate flexibility and capital mobility reduce the effectiveness of fiscal policy: other things equal, a fiscal stimulus will put upward pressure on the interest rates and exchange rates, partially crowding out exports and interest-sensitive demand. If monetary policy targets price stability, an induced tightening may further mute the fiscal impact. The point is however a bit academic. There is a growing recognition that the best course for fiscal policy makers is to take on a longer-term orientation and avoid attempts at short-run stabilization—which often run afoul of long decision and implementation lags.

From this perspective, a relevant question is whether regional integration has implications for budgetary financing and broadly defined tax structures.

• Deficit financing may benefit from deeper regional capital markets as well as from any convergence to lower interest rates. This has been the experience in Europe, for example, in the run-up to the introduction of the euro.³⁹ Much of that gain, however, has been traced back to a progressive elimination of exchange and inflation risk premia as prospects for a single currency strengthened and it is unlikely to be relevant in the Asian context for some time to come.⁴⁰

• Intraregional financial integration and the expansion of vertical trade (i.e., in goods for further processing) will likely require a degree of tax harmonization.⁴¹ Capital and labor may become less nationally located and harder to tax, setting the stage for tax competition that, in a race to the bottom, may harm the provision of public goods. There may be a case for shifting the tax base to consumption, for example through more widespread reliance on a value added tax, with the additional benefit to limit trade distortions.⁴²

What are the policy requirements for a monetary union?

There is a broad recognition that a regional monetary union in Asia is a distant goal, at best. Asia’s growing interconnectedness through trade in goods and capital is suggestive of the road taken by countries in continental Europe on the way to the Economic and Monetary Union. However, a realistic view needs to take into account the difficulties posed by Asia’s cultural and political diversity, as well as lack of supra-national institutions and incomplete intraregional convergence.

• The Maastricht criteria provide reference points for assessing the region’s degree of nominal convergence (Table). It is telling that, in 2005 no group of countries met all the Maastricht thresholds and most countries missed at least two (or more, if the debt ceilings were recalibrated to the Asian context).

• Convergence of production and economic structures is even more elusive, suggesting a vulnerability to country specific shocks that a monetary policy cast from a region-wide perspective could not counter.⁴³

Table III.1.

Convergence in Asia: Maastricht Criteria

Note: Numbers in bold type exceed the reference values set out in the Maastricht Treaty.

^a/The criterion for compliance is obtained by taking the average for the three countries with the lowest inflation rates and adding 1.5.

^b/The criterion for compliance is obtained by taking the average for the three countries with the lowest inflation rates and adding 2 percent margin allowed by the treaty.

^c/The reference value is -3 percent of GDP.

^d/The reference value is 60 percent of GDP.

Table III.1.

Convergence in Asia: Maastricht Criteria

	Inflation Ratea	Interest Rateb	Fiscal Balancec	Gross Debtd
Australia	2.7	5.3	0.8	10.7
Cambodia	5.8	…	-3.1	41.4
China	1.8	2.8	-1.3	19.3
Hong Kong SAR	1.1	3.6	0.3	1.9
India	4.3	6.6	-7.5	83.8
Indonesia	10.5	13.0	0.4	47.7
Japan	-0.3	1.4	-5.8	175.5
Korea, Rep. Of	2.7	3.5	-0.8	32.0
Malaysia	3.0	3.6	-3.6	45.4
New Zealand	3.0	6.0	3.1	22.2
Philippines	7.6	10.9	-1.9	66.9
Singapore	0.5	3.4	6.0	…
Thailand	4.5	5.0	0.1	47.4
Vietnam	8.7	8.0	-6.4	43.7

Note: Numbers in bold type exceed the reference values set out in the Maastricht Treaty.

^a/The criterion for compliance is obtained by taking the average for the three countries with the lowest inflation rates and adding 1.5.

^b/The criterion for compliance is obtained by taking the average for the three countries with the lowest inflation rates and adding 2 percent margin allowed by the treaty.

^c/The reference value is -3 percent of GDP.

^d/The reference value is 60 percent of GDP.

View Table

Table III.1.

Convergence in Asia: Maastricht Criteria

	Inflation Ratea	Interest Rateb	Fiscal Balancec	Gross Debtd
Australia	2.7	5.3	0.8	10.7
Cambodia	5.8	…	-3.1	41.4
China	1.8	2.8	-1.3	19.3
Hong Kong SAR	1.1	3.6	0.3	1.9
India	4.3	6.6	-7.5	83.8
Indonesia	10.5	13.0	0.4	47.7
Japan	-0.3	1.4	-5.8	175.5
Korea, Rep. Of	2.7	3.5	-0.8	32.0
Malaysia	3.0	3.6	-3.6	45.4
New Zealand	3.0	6.0	3.1	22.2
Philippines	7.6	10.9	-1.9	66.9
Singapore	0.5	3.4	6.0	…
Thailand	4.5	5.0	0.1	47.4
Vietnam	8.7	8.0	-6.4	43.7

Note: Numbers in bold type exceed the reference values set out in the Maastricht Treaty.

^a/The criterion for compliance is obtained by taking the average for the three countries with the lowest inflation rates and adding 1.5.

^b/The criterion for compliance is obtained by taking the average for the three countries with the lowest inflation rates and adding 2 percent margin allowed by the treaty.

^c/The reference value is -3 percent of GDP.

^d/The reference value is 60 percent of GDP.

Policies can foster the convergence needed for a workable monetary union.⁴⁴ At the most basic level, there is the need to get the macroeconomic framework right. Domestic stability and growth must be secured through flexible, transparent, market-based economic systems supported by responsible monetary, fiscal, and financial frameworks. In particular:

• Sound fiscal policies make for stronger economic performance which will ease the transition to—and the operation of—a monetary union. Fiscal prudence would also build the margin for maneuver to address country-specific shocks.

• A monetary policy firmly geared to price stability can lock in a history of low and stable inflation needed to back the eventual adoption of (irrevocably) fixed exchange rates among participating countries. Exchange rate flexibility in the transition to monetary unification will continue to provide the supporting exchange rate regime.

• Regulatory harmonization and shared governance practices can foster convergence of financial structures. Financial market reforms may in many instances help a move to greater exchange rate flexibility. Strengthened supervisory and prudential systems will also pave the way for greater capital account openness.

• Finally, an improved intraregional policy dialog and more effective mechanisms of regional surveillance will facilitate policy coordination and a greater focus on common goals. Further integration of the markets for good and services, and in capital and labor, will result.