An ACU for denominating financial transactions

A. The Store of Value Function of an Asian Regional Unit

It is possible for the ACU to perform at least some of the functions of an international currency basket. It might be assumed that the ACU would be a low risk currency, well suited to risk-averse investors. By backtracking a synthetic ACU, it is possible to explore the extent to which the risk properties of such an ACU would have met these priors. We examine the risk properties of short-term investments in ACU as compared to investments in national instruments. The unavailability of data for a large number of East Asian countries over a long enough sample precludes us from examining the more complex issue linked to investments in long maturity instruments in ACU. Therefore, we study the latter issue in Appendix 2 in the light of the European experience with the ECU.

1. Short Maturities

The risk properties of short-term investments in ACU should be compared with the risk properties of short-term investments in each national currency. We consider such risk properties from the point of view of returns on ACU holdings in each national habitat. For the 2000:I–2006:I period, we examined twelve East Asian countries.⁶ Quarterly data were used in order to match the maturity of the investments with the frequency of the data. We present the mean return, risk (as measured by the standard deviation), and the ratio between the two (the Sharpe ratio).

The formula in definition (5) was used for Table 6 to compute the return in a national habitat of an investment in an ACU short-term instrument:

(5) Racuj=[(Et/Et-1)*((iacu,t-1)/4)+( (E t- E t-1)/E t-1)*100)]*4

Et = number of units of domestic currency per ACU

iacu= interest rate on 3-month investment in ACU instrument at annual rate

In national habitats, mean returns were higher for investments in ACU instruments than in the instruments of Cambodia, Japan, Singapore, and Thailand (Table 6 [ PDF 88.8KB | 1 page ]). However, in national habitats, volatility was always much higher for an investment in an ACU instrument. Therefore, Sharpe ratios are considerably smaller for investment in ACU instruments than for investments in national instruments.⁷

2. Long Maturities: the ACU as a Special Currency

In principle the interest rate on ACU instruments should be a weighted average of component interest rates. But because of its basket nature, the ACU will be a rather complex and special currency. First, the interest rate elasticity of the ACU with respect to interest rates on component currencies will be a function of maturity. The longer the maturity of an ACU-denominated bond, the higher will be its elasticity with respect to changes in low yield currencies and vice versa. The reason is that market participants need to project exchange rate movements until maturity, on the basis of existing interest rate differentials among component currencies to remain consistent with interest rate parity constraints.

Second, the above result implies that the ACU yield curve will be flatter than the curve obtained by aggregation of individual component currency yield curves based on actual weights in the ACU. To illustrate this point, it is useful to consider a case where yield curves of all component currencies are flat, as we did in Example 4. The ACU yield curve would then be negatively sloped; this is because, at any point in time, current interest rate differentials suggest (on an interest rate parity basis) that, over time and up to maturity, lowyield currencies will appreciate against high-yield currencies. Hence, the weight (i.e., the product of the amounts of a currency in the ACU definition times the ACU exchange rate of this currency) of low-yield currencies will increase over time and the weight of highyield currencies will decline (see Girardin and Steinherr 2006). The implication is that the weighted average of interest rates on component currencies is declining over time.

Third, variations of exchange rates between component currencies will affect ACU interest rates only very marginally. In fact, the sign of the exchange rate effect is not clear a priori: a devaluation of a component currency in terms of all others may result in a decrease or an increase in the ACU interest rate. Devaluation of a currency with interest rates above (below) ACU interest rates result in a reduction (increase) of ACU interest rates. Because devaluation lowers the weight of the devalued currency in the ACU basket, the variation in ACU interest rates depends on whether the devalued currency's interest rates are above or below ACU rates. This effect is obviously small because all other component currencies are, by definition, revalued, some of them with interest rates above ACU rates. Therefore, the reduction in ACU interest rates will be dampened.

The case of long-maturity investments is examined in detail in Appendix 2. We cannot do this for East Asia since small countries among ASEAN+3 which do not have an active bond market would have to be excluded. We analyzed a variety of risks that are specific to the ECU as a basket currency and those that are shared with any other foreign currency. We are able to identify two sources of risk that are analytically identical but differ from a policy perspective. One derives from the uncertainty associated with exchange rate changes on component currencies. The premium attached to this type of risk is factored in the spread between theoretical and market interest rates for ECU-denominated assets.

The second type of risk is due to the fact that interest rate parity must be used to compute a time path for future weights of basket components. Forward rates are considered (though by a shrinking set of economists) as best-linear-unbiased estimators of future exchange rates, but their performance is very poor. Hence, forecast errors will be made and the prices of ECU-denominated bonds factored in a risk premium. Clearly, both types of risk were due to the variable weights nature of an ECU basket, as opposed to a fixed-weight one.

3. Risk in the ACU Habitat

Table 7 [ PDF 79.2KB | 1 page ] lists "beta" coefficients for the sake of comparing the attractiveness of the ACU in the East Asian habitat with that of domestic currencies. We set the beta of the ACU returns in the East Asian habitat equal to unity (equivalent to the relevant "market" risk) and for all other currencies we ran the following regression:

(6) Yi = α + β X_E

where Yi is the return on a comparable domestic money-market instrument (three month returns) converted into ACUs (same as computed in column 2 of Table 6), and XE the return on ACU-denominated money-market instruments. If the estimated beta for currency i is equal to one, then, on average, it would be as risky as the ACU. A beta larger than one suggests that the currency is riskier; a beta of under one, but positive, then the currency is less risky than the ACU.

The results in Table 7 suggest that in the East Asian habitat, for most currencies, the beta is insignificant. One exception is Indonesia, whose currency with a beta significantly positive is riskier than the ACU. An important result is that no currency has a significantly negative beta. A significantly negative beta would imply that a currency would be very risky, but also, and at the same time, that it would be a useful hedge for the ACU. In other words, an important implication of such results is that none of the currencies examined is such that its non-participation would make the ACU more attractive than otherwise.

B. Launching and Nurturing the ACU

1. Cost-Benefit Analysis of ACU-Bond Issuance

The existing experience with the issuance of basket bonds is only recent. Multi-currency bonds issued before the First World War or dual currency bonds issued after the collapse of the Bretton-Woods system were not real precedents since they simply embedded currency options protecting the investor against the depreciation of any of the currencies (Dammers and McCauley 2005). Basket currencies such as special drawing rights or the initial ECU were virtual currencies since settlement had to be made in other “real” currencies, but subsequently this did not apply to the ECU any more since settlement could be made in that currency.

a. Benefits: Is Diversification Sufficient?

In principle even the benefit of diversification may not seem to be a necessary condition. In principle, an investor could by herself buy a portfolio of bonds denominated in all the currencies in the basket. However, in practice, this may not be feasible because of the unavailability of the denominations for small amounts, and it may not even be available since bonds may simply not exist in the required currency or maturity, and international issuance is simply prohibited for residents of some countries.

Even if diversification benefits are present for the issuance of ACU-denominated instruments, would this be a sufficient condition for the success of such issuance? Some observers of the European experience with the ECU bond markets in the 1980s and 1990s argue that the success of issues on such a market had more to do with either regulatory arbitrage or convergence trades, than with diversification benefits. On the former, German regulation enforced the anchoring principle, implying that foreign issuance of German mark bonds had to be underwritten by a German bank. As a response to such regulations, non- German banks attempted to create an alternative to the US dollar which could mimic the mark. As a result, the underwriters of ECU bond issues were exclusively non-German over the 1980–1987 period. The convergence trade interpretation is based on the subsequent period when the stability of the European currencies within the European exchange rate mechanism (ERM) led investors to benefit from excess returns over German mark yields. Issuance collapsed when such stability vanished after the 1992 ERM crisis, and only took off again when a clear commitment to EMU was made in the second half of the 1990s.

In East Asia, regulatory arbitrage may work also but in a different way: there is currently no possibility for international or regional investors to access the issuance of bonds denominated in many East Asian currencies because of capital controls. However, this leads to a dilemma (Eichengreen and Luengnaruemitchai 2004) because it is wise to liberalize only after financial markets have become robust enough, i.e., sufficiently liquid and deep. The development of ACU bond markets could help sidestep this dilemma. By providing special treatment, within domestic foreign exchange regulations, to domestic issuance of bonds denominated in ACU, foreign investors would be given access to securities issued by domestic residents, and financial market development and deepening would be initiated, with immediate access to the international financial environment.

If the argument about the predominance of convergence trade is right, then exchange rate stability before the East Asian crisis was a missed opportunity. Indeed, as was clearly explained in the moral hazard theory of the crisis, investors were convinced that the dollar peg was going to last. However, such a discussion centered purely on the European experience may miss three other important dimensions specific to East Asia (Park and Park 2005). First on efficiency grounds, a regional ACU-denominated bond market may provide issuers, public or private, with a greater availability of funds at a lower cost, with deeper, more liquid markets on which a larger diversity of instruments is traded. Second, even though some observers may argue that there is no place for a regional market on top of existing national and global bond markets, there may be a case for a regional bond market based on “missing markets.” Indeed, issuing corporate bonds with a low investment grade (B to BBB) is not always available to Asian firms on global markets. One of the reasons for such a missing market is that these firms inherit from the rating of their sovereign, which may not be representative of their own financial performance. Regional rating agencies would not be victims of such an asymmetry of information. Securitization on a regional market would be a partial answer if complemented by credit guarantees helping to dispose of subordinated bonds. Third, the current pattern of financial flows between East Asia and the rest of the world still involves investing official reserves in short-term dollar denominated assets, and borrowing short-term from banks in US dollars. Regional bond markets would contribute to reducing the exclusive reliance on such short-term recycling.

Some regional initiatives have been started, such as the Asian Bond Market Initiative and the Asian Bond Fund. However, since the latter should invest in bonds denominated in US dollars by East Asian issuers, it is not clear that it will help to develop the East Asian bond market. It would be better for such a fund to invest in ACU-denominated bonds.

b. Costs: a Zero-Sum Game?

A major worry attached to the development of sovereign issues of ACU-denominated bonds is that if a given government were to choose the ACU-bond market instead of the domestic currency-bond market, the latter would shrink. Its turnover and liquidity would diminish and its long-term development would be impaired. Such a concern seems overplayed in the East Asian case. Indeed, first, this may actually leave more room for private issuers in the domestic market; second, the funding needs of many governments in East Asia are huge, for example for infrastructure development or pension funds. Even for intermediate-sized countries, such worries may not materialize. More fundamentally, such concerns ignore what was one of the main tenets of the “original sin” theory of the East Asian crisis, i.e., the inability of many countries in the region to issue bonds denominated in their own currencies. For such countries, the issuance of ACU-denominated bonds would be a golden opportunity, enabling them to avoid putting all their eggs in the same basket, i.e., to be able not to issue bonds denominated in one single international currency, but in a basket of regional currencies.

Another concern would relate to the difficulty of interpretation and associated risks inherent in a basket or virtual currency. Supposedly, it would be difficult to market basket bonds to issuers and investors given their complexity. Such an argument seems rather far-fetched at a time when the sophistication of financial markets has reached such high levels, and the understanding of financial products has become general knowledge.

2. The Regulation and Supervision of the ACU

a. How to Treat Banking Activities in ACU?

The experience of Singapore, with its offshore foreign currency business, enjoying different regulations, rules, and supervisory regime (Box 1 [ PDF 70KB | 1 page ]), since the late 1960s provides some guidance on ways to treat loans and deposits in foreign currencies (of which the potential ACU would be an example). Strictly speaking, the differential treatment of foreign currency business in Singapore has not required the continuation of de jure capital controls, in as much as the latter were discontinued in the late 1970s. However, the framework shows that countries whose currency would be included in the ACU could allow a different treatment, in the activities of their banks, for deposits and loans in domestic currency and deposits or loans in ACU. This could prove important given the widely different roles and structures of national banking systems in the region. The differential treatment of domestic currency and ACU business would allow regulations for the domestic currency activity of banks to gradually evolve while business in ACU would be treated as offshore activities. Such differential treatment would amount to accounting distinctions for financial legal entities established within the same financial institutions and registered to operate according to separate and specific guidelines decided by national banking regulators.

Foreign banks could be offered incentives to participate in offshore ACU activities as well as deterrents against operating in the onshore domestic currency market, whenever national regulators wish so. These deterrents could include: a ceiling on domestic currency loans, the relative inaccessibility of local deposits, as well as high reserve costs. Incentives for offshore ACU activities would comprehend the lifting of any withholding tax on the interest income of non-residents, the waiving of statutory reserve requirements, as well as a large number of fiscal incentives with respect to foreign securities trading, fund management, and syndicated loans.

The dichotomization of activities of domestic banks would be a way for domestic banks to learn more quickly how to deal with a liquid and active environment. In particular, this would allow the involvement of such bank in the development of an active ACU inter-bank market while, in many small ASEAN countries, the inter-bank market is still at a very early stage. It is possible that such a scheme may result in the predominance of the ACU business over the domestic currency one. The latter possibility, of course, would be welcome from the point of view of the development of the ACU.

Reference can also be made to the European experience with dual foreign exchange markets in Belgium, tax exemption on ECU transactions in Italy, and special treatment under capital controls in Ireland (Gros and Thygesen 1998).

b. The ACU as a Parallel Currency

For the ACU, a number of potential choices can be derived from the ECU experience (Box 2 [ PDF 69.8KB | 1 page ]). One of those concerns is the possible treatment of the ACU as a parallel currency. In countries with weak and volatile currencies, economic agents tend to hold cash positions in foreign currencies (dollar) anyhow. In such countries, the ACU would be rapidly accepted and would possibly crowd out national currencies. All the more so if national authorities accept ACU balance sheets and tax payments. In hard currency countries, this process would be much slower and possibly only begin when the ACU is already a major currency in some countries. Over time, the market would decide; a process that European countries had not accepted. Second, it would be advisable not to repeat the European experience in creating an official and a private ACU (Girardin and Steinherr 2006).

As far as consumers are concerned, it is only in partly US dollarized economies that replacing dollar holdings with holdings of ACU could be considered a positive step. However, as a general principle, the experience of most developing economies shows that currency substitution for consumers is a source of monetary instability. Therefore, any scheme intended to introduce such substitution, where it is absent, should be seen with great skepticism, and used only when there is full certainty that monetary instability will not be the first and foremost by-product.

Download this Discussion Paper [ PDF 274.5KB| 38 pages ].

Comment(s)

There are [0] comment(s) for this entry. Post a comment.