1. Asking "that" question

  Hong Kong's currency board system, the link, was adopted in October 1983 as a rescue measure to stop the Hong Kong dollar (HK$) from collapsing in the midst of a political row between China and Britain on the future of the territory. There was no time for debating whether a fixed exchange rate regime was "optimal" for the Hong Kong economy. Nor was the rate of HK$7.80/US$ scientifically determined.

  Then came the question of the 1997 political transition, when Hong Kong ceased to be a British colony and returned to become a special administrative region (SAR) under Chinese sovereignty. In the run-up to 1997, few brave souls dared to advocate abandoning the link for economic idealism. The perceived political risk was simply too huge.

Now after the transition, the still unfolding East Asian financial turmoil (which has seen many currencies in the region plunging, in contrast to the survival of the Hong Kong dollar peg) adds a frustrating dimension to the equation. With the impending recession in the territory, partly as a result of the loss in competitiveness, it is perhaps justifiable to seriously ponder "that" question. Is the link optimal for Hong Kong? Of course, this is not be the time to float the HK dollar or adjust the peg. Such actions are bordering on the suicidal. That is why I have been advocating as a near-term response the adoption of the AEL model to strengthen the link (Tsang, 1996b; 1998b), and to reduce the "efficiency risk" in the system so that the "interest rate pain" can be alleviated (if not eliminated). However, what if the crisis finally subsides and the situation returns to normal, should Hong Kong then choose an opportune time and change the system?

Williamson (1995) gave a catalogue of advantages and disadvantages of currency boards, largely from the viewpoint of optimality. Tsang (1996b) used those criteria to evaluate the case of Hong Kong, and came out with an overall affirmation of the link. Ghosh, Gulde and Wolf (1998) also had some nice things to say about the general performance of economies adopting currency board arrangements (CBAs), which contrasted sharply with the critique of Roubini (1998). Lithuania, on the other hand, has decided that it has had enough of its currency board (Bank of Lithuania, 1997). The country is quitting the system in order to prepare to join the European economic and monetary union. In any case, the East Asian turbulence raises serious new questions and concerns about fixed exchange rates in general, and Hong Kong's own experience also calls for a review of the long-term desirability of the link as a CBA.

2. In favour of fixed exchange rate and CBA: flexibility, stability, anchorage and banking strength

  A CBA represents one way of fixing the exchange rate, albeit a very strong one, and theoretically market driven. So the debate about the fixed versus the floating exchange rate regime is relevant.

In favour of the link, there are several arguments. First, in principle, the more flexible the real sector of an economy is, the more suitable is a fixed exchange rate to it, because there is no need to change nominal relative prices through exchange rate movements in the event of a shock. The real sector will execute the job by quickly adjusting. Hong Kong seems to qualify as such an economy. Or does it not?

Second, for a small open economy like Hong Kong, a floating exchange rate may not bring much monetary "autonomy". Local interest rates cannot deviate from foreign ones by too much, lest there might be exchange rate gyrations. In both the floating rate era (late 1974 to September 1983) and the fixed rate regime up to the end of 1997 (October 1983 to December 1997), the differentials between three-month HK$ HIBOR and US$ LIBOR were relatively small. A quick calculation shows that the standard deviation was 1.89% in the former period, and 1.44 % in the latter (on the basis of average monthly figures).

Third, a floating exchange rate system for a financial centre like Hong Kong carries considerable instability risk, particularly as the territory is going through a very sensitive political period. Speculative capital movements and attack on the currency are difficult to contain when the exchange rate floats. A fixed rate regime does have the advantage of diverting pressure away from the exchange rate to other aspects of the economy, which may be less vulnerable to financial fluctuations and "hit and run" activities. It thus serves an important anchorage function.

Fourth, a fixed exchange rate regime, and in particular a CBA, keeps the banking system under tight rein. Under a CBA, banks are presumably forced to quote the fixed exchange rate because of effective cash arbitrage (Tsang, 1984) or electronic interbank arbitrage via the central bank (Tsang, 1998b), irrespective of whether the rate chosen is "optimal" in a microeconomic sense. Banks would have little room, if at all, to manoeuvre. Also, the monetary authority is seriously constrained in providing liquidity to the banking system, even as the latter is under stress because of a lack of confidence. The reason is that the injection of liquidity may contradict the supreme goal of defending the exchange rate (Caprio, Jr. et al., 1996; Bank of Lithuania, 1997; Santiprabhob, 1997). These have been cited as arguments against implementing a CBA in Indonesia (Roubini, 1998), at least if it is done too hastily, without proper regard for the "systemic risk" (Tsang, 1998a). However, the problem concerning "bank soundness" is much less serious in Hong Kong as a financial centre, because the local banking sector is strong by international standards (FSB, 1998, paras. 3.70-3.109).

3. Against fixed exchange rate and CBA: sea change, poor barometer, internal versus external crises, and exit

  There are of course counter-arguments. First, if the shocks to an economy are very dramatic, to the tune of a "sea change", it may be too much to ask the real economy to shoulder all the adjustments. The exchange rate should at least share part of the burden.

Second, even for a financial centre, a floating exchange rate would give useful signals about unfolding imbalances in the economy. New York, London, Frankfurt, Zurich, Tokyo and, to a lesser extent, Singapore are all operating under floating exchange rates. A fixed exchange rate is a much less sensitive barometer to ever changing market conditions.

Third, in the longer run, the problem is whether the fixed rate needs to be modified at all, should there be an irreversible structural divergence between the home economy and the foreign economy to whose currency the domestic one is pegged. As a very strong form of fixed exchange rate regime, a CBA was in a number of cases used as a rescue measure for an internal crisis e.g., Hong Kong in 1983 to save the HK dollar (as pointed out at the beginning); Argentina in 1991 to stop hyperinflation; Estonia and Lithuania in 1992 and 1994 respectively to prop up their fragile new national currency. On the whole, it has been quite successful in containing home grown disasters.

However, a CBA may find it difficult to deal with major external crises, structural divergence, or deep-seated economic imbalances that do not show up easily. The reason is surprisingly simple: a CBA has to peg its currency to a foreign counterpart at a certain rate on a certain day, without "perfect" foresight about the future. How can such a decision on the "anchor" (particularly if it is made in a desperate situation) automatically claim economic immortality in a world of rough oceans? Have we forgotten how and why the old gold standard was deserted?

The difference between internal crisis and external crisis is crucial in the assessment of the economic optimality of a CBA. This I am afraid is a neglected topic in the debate of fixed versus floating exchange rate, as well as in the discourse about currency board economics.

A key issue for a CBA, which was not discussed by Williamson (1995), is that of exit cost, i.e., the cost of quitting the peg. If more is invested in building confidence in the CBA or even in "guaranteeing" the fixed exchange rate, like the insurance schemes proposed by some academics to strengthen the Hong Kong link (FSB, 1998, paras. 3.46 - 3.59), more has to be foregone when it is finally deemed necessary to change track, because of factors that are outside domestic control.

Having said all these, though, one has to refrain from over-reacting to the mood of the day. Is the world economy into a rough ride (with an impending meltdown on Wall Street)? Even if it is, giving up the anchor in a stormy sea at the wrong moment is not necessarily less suicidal. When to jump the ship is a "life or death" question. Lithuania's case is "unique" (Estonia and Bulgaria following?) in the sense that the country actually wants to "throw" itself into a powerful union.

4. Empirics of a CBA: growth and inflation in Hong Kong

Hong Kong is caught in a dilemma. The link survives, but the aftermath of its survival has led to a harsh reality: the unemployment rate shot up to the highest level in more than a decade. While only a minority of economists is forecasting negative real growth for 1998, most find the government's prediction of 3.5% hopelessly optimistic.

But let us go beyond the immediate and look at some empirical evidence on the performance of the Hong Kong economy since the adoption of the link in late 1983.

While the nominal exchange rate of the HK dollar against the US dollar has remained around the 7.80 level, cumulative inflation differential between the two economies was about 90% in the period of 1984-1997. In other words, the HK dollar has appreciated substantially in real terms against the US dollar. Even if we consider that the HK dollar was undervalued at the time of the inception of the link, the discounted real appreciation would still be notable. This might have been one of the important reasons why real GDP growth has declined under the link rate (see Table 1).

The problem seems to lie with the 1990s when low GDP growth has combined with high inflation. The fall in the real growth rate of per capita GDP has been particularly disappointing, as the Hong Kong economy should not have "matured" so quickly. Common perception has it that the "China factor" boosted the Hong Kong economy, particularly after the "Deng Whirlwind" of 1992, when the paramount leader Deng Xiaoping made a tour to Southern China and enjoined everyone to speed up economic growth and reform. In Hong Kong, however, the whirlwind has apparently resulted in more inflation than growth.

It is however necessary to look at the inflation issue more closely. We can investigate it from two angles: (1) inflation transmitted through trade; and (2) asset inflation.

As far as inflation transmitted through trade is concerned, there is little evidence that the link has been the culprit, as Table 2 testifies. The unit value index (UVI) of import is the deflator for imports, and it has not shown any serious upward pressure since the inception of the linked rate system. Of course, the deflator is determined by two factors: (1) the exchange rate and (2) foreign inflation. While foreign inflation has remained low, one could always argue that a stronger exchange rate would have kept the UVI under even tighter rein.

¡@

5. Asset price bubble in Hong Kong

Another aspect is asset inflation. In so far as the linked rate under-reflected the "fundamental strength" of the Hong Kong economy, in the light of its rapid "integration" with the Chinese economy, local assets might look exceptionally attractive to foreign investors, and inflow of outside money would cause asset inflation and potential instability. Sheng (1995) tackled the hypothesis of asset bubble and found it wanting. He pointed to the fact that a flexible exchange rate might not avoid asset inflation as "recent evidence of asset price bubbles occurred mainly in countries with flexible exchange rates, such as Japan, Australia, Scandinavian countries and the United Kingdom. All these countries were free to use their interest rate tools, and yet the bubble occurred." (p.56)

Nevertheless, the East Asian financial crisis does focus attention on the necessity of keeping strict economic discipline under a fixed or pseudo-fixed exchange rate regime. Over-investment in the real estate and financial sectors had fuelled bubbles in a number of economies in the region (in particular Thailand, Malaysia and Indonesia), which unwittingly also ran considerable current account deficits financed by capital inflows (quite a large share of which was short-term hot money). Since these economies adopted implicit nominal or real peg to the US dollar (Corsetti, Pesenti and Roubini, 1998), the underlying contradictions could have built up over a period of years without showing explicit strains. Ironically, that is the key weakness of a successful fixed exchange rate regime: short-term problems are "hidden" under the superficial calm of the peg. A floating exchange rate system would be a better barometer of economic imbalances.

Hong Kong has also shown signs of an economic bubble in the 1990s. Table 3 tracks the relative performance of the annual averages of per capita GDP, private residential property prices, and the Hang Seng Index of the local stock market since 1984 (when all the indexes were normalized to 100). Per capita GDP is chosen as a proxy for average income and purchasing power. Prices of real estate and stock shares should not overshoot it by too much, even granted considerations of changes in saving behaviour and portfolio selection. From Table 3, it can be observed that while the three indexes were within each other's neighbourhood in the 1980s, wide gaps had opened up since 1992, but without any ramifications on the exchange rate, which has stayed on the strong side of 7.80.

A more deep-seated factor behind these developments has been the "Manhattanization" or the "de-industrialization" of Hong Kong (Tsang, 1994; 1996a). With the Chinese economy opening up since 1979, Hong Kong entrepreneurs have relocated in massive scale their plants and factories into the Mainland, making use of the much cheaper labour and material resources. Manufacturing industries dwindled dramatically within Hong Kong, but repeated themselves at lamentably stagnant technological levels further north. Surplus capital, human and other resources in Hong Kong then moved rapidly into the service sectors. Unfortunately, with the profitability of many of those sectors eroded by extremely high property prices and rentals, an increasingly number of investors found it more rewarding to desert "productive" endeavours. Instead, they pushed onto the frontiers of speculative activities. Acute structural imbalances in the local economy were formed (Tsang, 1997).

The whole phenomenon resembled that of the "Dutch disease", when sudden fortunes derailed a balanced growth trajectory. What is amazing is that the currency board system in Hong Kong has not prevented, nor given any clear warning signals about, this structural deterioration until the East Asian financial turmoil.

The "China factor" has apparently also contributed to a financial bubble in Hong Kong. In the run-up to 1 July 1997, Chinese capital funds reportedly poured into the property and stock markets in the territory, perhaps in an effort to demonstrate support to the incoming SAR, or simply as a result of uncontrollable speculative activities in an era of decentralization and liberalization. In any case, international fund managers followed suit. A bubble rapidly emerged and then became self-destructive.

Property prices were on a dramatic upswing in the fourth quarter of 1996, when rumours about Chinese capital abounded. In the twelve months that followed, the average price index of residential property units rose by about 50%, while that of large units surged by almost 60%. Equally breathtaking was the rise in China-related stock shares listed in Hong Kong. There are two types of such shares, for which price indexes are compiled: (1) the China Enterprises Index, or the so-called "H-shares" index; and (2) the China-Affiliated Corporations Index, or the "red-chips" index. Both exhibited huge volatility around the 1997 transition, reflecting highly speculative activities.

All these point to one important conclusion: an economic bubble can emerge from a fixed exchange rate regime, including one that professedly follows the strong principles of the currency board system. It is crucial that the authorities closely monitor the longer-term viability of the fixed exchange rate.

6. In sum ........

In recent years, CBAs have been used by a number of economies as crisis response measures. Their record in defusing internal crises has been remarkable, as the experiences of Hong Kong, Argentina, Estonia and Lithuania have demonstrated. A key to success is the very strong internal commitment to the fixed exchange rate, shown in a "transparent" manner. But not all economies could save their exchange rates via this route: "bank soundness" (Santiprabhob, 1997; Roubini, 1998; Tsang, 1998) and the containment of "efficiency" and "systemic" risks are legitimate concerns (Tsang, 1998a; 1998b).

In any case, even if a CBA can solve short-term internal crises, its long-term desirability as an economic institution remains an open question.

In theory, the more flexible the real sector of an economy is, the more suitable is a fixed exchange rate to it, as the real economy will adjust quickly, without having to go through nominal and relative price changes via exchange rate movements. For a small open economy with a huge financial sector, a floating exchange rate system may bring instability, as it will be difficult to contain speculative capital movements and attack on the currency. A fixed exchange rate regime, especially a CBA that commands credibility, does have the advantage of diverting pressure away from the exchange rate to other aspects of the economy and thus providing an anchorage.

The problem is that a CBA in the modern context often provides an anchorage for a desperate economic "ship" in a rough sea, and the arrangement may not have been well planned for uncertainties beyond the immediate horizon. While a CBA requires "strict" fiscal and monetary discipline, it does not serve as a very good barometer of imbalances that might be building up in the economy. Bubbles could emerge, as in the case of Hong Kong, and the consequences would be very serious if the fixed exchange rate collapses under pressure, as testified by the experience of several East Asian economies.

As a very strong form of fixed exchange rate regime, a currency board system may find it hard to deal with major external crises, structural divergence and deep-seated economic imbalances that do not show up easily. A key issue is that of the exit cost of quitting the peg. If more is invested in building confidence in the system, more has to be foregone when it is finally deemed necessary to abandon the ship. A balance needs to be kept over the conflicting considerations of ensuring short-term stability and of maintaining long-term flexibility.

¡@

References

Bank of Lithuania (1997) Monetary Policy Programme of the Bank of Lithuania for 1997- 1999, Resolution No.14 of the Board of the Bank of Lithuania, 16 January.

Caprio Jr., G., Dooley, M., Leipziger, D. and Walsh, C. (1996) "The Lender of Last Resort Function Under a Currency Board", The World Bank, Policy Research Working Paper, No. 1648.

Corsetti, G., Pesenti, P. and Roubini, N. (1998) What caused the Asian currency and financial crisis? at www.stern.nyu.edu/~nroubini/asia/AsiaHomepage.html.

Financial Services Bureau (FSB) (1998) Report on Financial Market Review, Hong Kong Government.

Ghosh, Atish R., Gulde, Anne-Marie, and Wolf, Holger C. (1998) "Currency Boards: The Ultimate Fix?" IMF Working Paper WP/98/8.

Roubini, Nouriel (1998) The Case Against Currency Boards: Debunking 10 Myths about the Benefits of Currency Boards, at www.stern.nyu.edu/~nroubini/asia/CurrencyBoards roubini.html.

Santiprabhob, Veerathai (1997) "Bank Soundness and Currency Board Arrangements: Issues and Experience", IMF Papers on Policy Analysis and Assessment, PPAA/97/11.

Sheng, Andrew (1995) "The Linked Exchange Rate System: Review and Prospects", Quarterly Bulletin, Hong Kong Monetary Authority, May, pp.54-61.

Tsang, Shu-ki (1984) "On the Cash-based Fixed Exchange Rate System", in Tsang Shu-ki, The Pearl in the Mouth of the Dragon: Collected Essays (in Chinese), Hong Kong: Wide Angle Press, pp.179-201.

Tsang, Shu-ki (1994) "The Economy", in The Other Hong Kong Report 1994, Hong Kong: The Chinese University Press, pp.125-148.

Tsang Shu-ki (1996a) "The Political Economy of Greater China", Asian Pacific Business Review, vol.2 no.3, pp.23-43.

Tsang, Shu-ki (1996b) A Study of the Linked Exchange Rate System and Policy Options for Hong Kong, a report commissioned by the Hong Kong Policy Research Institute, October.

Tsang, Shu-ki (1997) "Structural imbalances must be rectified" (in Chinese) Ming Pao, Hong Kong, 19 June.

Tsang, Shu-ki (1998a) Indonesian Currency Board? Two Questions to Answer, at www. hkbu.edu. hk/~econ/web984.html.

Tsang, Shu-ki (1998b) "The Case for Adopting the Convertible Reserves System in Hong Kong", Pacific Economic Review, forthcoming.

Williamson, John (1995) What Role for Currency Boards? US: Institute for International Economics.