Malaysia - Capital controls and exit strategy

98271 Malaysia: Capital Controls and Exit Strategy1 Naoko Kojo East Asia and Pacific Region The World Bank January 1999 1. Introduction On September 1, 1998, Malaysia’s Prime Minister Mahathir surprised the international financial community by imposing selective controls on capital outflows, in effect removing the ringgit from the world capital markets. Limits were placed on ringgit- denominated onshore-offshore transactions, including the use of external ringgit accounts by non-residents, borrowing of ringgit from offshore banks, loans from domestic banks to non-resident banks and stockbrokers and trade in ringgit instruments by offshore banks. Investments in ringgit financial assets are now required to remain in the country for at least a year, foreign trade transactions must be settled only in foreign exchange, and tighter limits were placed on overseas investment by residents and on the amount of ringgit a traveler can take out (see Appendix 1). Furthermore, the ringgit has been fixed at RM3.80/US$, around 10% stronger than its levels prior to the implementation of controls. The primary objective of the capital controls was to close the off-shore ringgit market in Singapore, thereby helping reduce the short-term volatility in the exchange rates. However, another objective taken into consideration was the need to regain monetary independence so that the government could refocus on its monetary policy based on domestic conditions. By delinking monetary policy from exchange rate movements, it is asserted that the new measures will permit the government to lower interest rates without further capital outflows and a sharp decline in the ringgit. Along with the aggressive fiscal expansion, the authorities appear to be trying to boost the economy by stimulating domestic demand rather than relying on external demand. The government is well aware of the danger of the distortionary impacts that the capital controls would have. They have thus stressed that these controls are temporary, to provide some breathing space while banking and corporate sector reforms are being accelerated, and that they will be removed once stability returns to financial markets and an appropriate global regulatory framework is in place (the exact timetable has not been specified, however). Furthermore, it has been emphasized that the controls do not affect the convertibility of current account transactions or repatriation of profits, dividends, capital and capital gains arising from FDI. Despite these emphases, however, the introduction of controls has substantially undermined Malaysia’s credibility in the eyes of foreign investors. Since Dr Mahathir’s 1 A paper prepared for presentation at a workshop on currency controls, organized by ISIS Malaysia, Kuala Lumpur, Malaysia, 21-22 January 1999. announcement in September, both portfolio and FDI inflows have fallen further.2 Mounting concerns over the health of Malaysia’s local banking and corporate sectors as well as ongoing political and economic uncertainty in Indonesia are added factors that are depressing market confidence. There is also a growing concern that the easing of monetary policy through lowing interest rates, setting minimum credit growth targets and relaxing loan classification and provisioning guidelines may run the risk of throwing good money after bad, ultimately adding woes of the already battered financial sector. The purpose of this policy note is to assess Malaysia’s present strategy – the imposition of capital controls together with the fixed exchange rate and expansionary monetary and fiscal policies. Would this policy mix work to reduce Malaysia’s vulnerabilities as intended? Is there any room for Malaysia to modify these policies? If so, when and how can it do so without resulting in a massive capital outflow? What should it be prepared for in modifying the strategy? To address these issues, this note draws much on the existing literature on the cross-country experiences of capital controls. The remainder of this policy note is organized as follows. To provide some background, Section 2 reviews the instruments that have been used by various countries in dealing with large-scale capital movements in the past, and Section 3 summarizes, and draws some key policy lessons from, the recent cross-country experiences. Section 4 then turns to the specific context of Malaysia in assessing the need for, and in designing, an exit strategy in the coming months. Finally, Section 5 concludes the note. 2. Instruments of Capital Controls This section reviews the instruments that have been used by various countries in controlling cross-border capital inflows and outflows in the past. Broadly, there are two basic categories of restrictions on capital movements. The first is to use administrative measures to prohibit capital transactions directly, and the second is to use taxes or tax-like measures to indirectly regulate the volume of international capital flows. These measures are often employed in conjunction with other policy measures (such as sterilization operation and widening exchange rate bands, etc.). We argue below that tax-based measures are usually most effective when imposed in the less distorting form of a tax rather than in outright restrictions on the quantity of capital flows. Moreover, tax revenues raised can be helpful in offsetting part of the administrative costs of capital controls. (i) Administrative Controls Though widely used in the past, administrative controls are not very practical or attractive in today’s world of sophisticated trading technologies and liberalized domestic financial markets. They require an extensive bureaucracy to track, register, evaluate and approve 2 The share of trading by foreign investors in KLSE fell sharply from 42% during the first eight months of 1998 to less than 13% in September 1998. Malaysia has been dropped from Morgan Stanley International (MSCI) Indices since November 39, 1998. 2 capital transfers. The actual implementation of effective controls would be very difficult indeed. However, no matter how well administrative controls are executed, the distortions they impose on the economy are severe and tend to get worse over time. Also, the volume of bureaucratic details to administrate controls implies increased cost of business for both firms and officials, which may provide particularly strong incentives for evasion. New methods of evasion could, in turn, encourage the authorities to widen the coverage of administrative controls, compounding the initial distortions. For example, when controls are initially implemented, they typically regulate capital movements but exempt trade credits. Yet a standard technique of evasion is to use leads and lags in commercial transactions. Eventually, therefore, administrative restrictions on capital flows interfere with international trade. Examples of administrative controls (or quantitative controls) include those on the foreign exchange liabilities of banks, personal capital transfers, sale of short-term securities to foreigners, minimum maturity of foreign deposits, share of firms’ capital that can be owned by non-residents, etc. (ii) Tax or Tax-like Measures Alternatives to administrative controls are taxes and tax-like measures. An important aspect of the argument that distinguishes this type of intervention from the administrative control program is that the taxes are aimed at increasing the cost of international financial transactions, but still leaving them otherwise unrestricted. There are a wide range of instruments of this type of approach, which includes negative interest rate (Switzerland) and special reserve requirements on foreign bank deposits.3 The latter takes the form of reserve requirements for external borrowing (Chile, Colombia) and reserve requirements for banks on their net foreign currency deposits (Italy, Spain). Although such measures can provide considerable disincentive to engage in short-term speculative transactions, one advantage of implementing them is that the distortionary effects would be minimal for trade and long-term foreign direct investment (at least in the short-run). However, one should not underestimate such distortions. Given the extremely thin margins in modern financial markets, introducing even a small tax can have major effects on the allocation of resources. The second advantage is that they require lower administrative costs in that they can easily be imposed. For example, by adding a few lines of code to banks’ computerized trading programs. Compliance could be monitored by period inspection of banks’ computer records (example taken from Eichengreen and Wyplosz, 1996). Finally, tax revenues raised can be helpful in 3 Amongst these instruments, the simplest variant is known as the Tobin tax, which (at least in theory) aims at creating some friction for all foreign exchange transactions (including trade) and is therefore expected to deter speculative currency transactions, reduce the volume of short-term capital inflows and restore exchange rate stability. However, the Tobin tax requires worldwide implementation, which keeps actual implementation of such a tax impractical. See IMF (1993), Kenen (1996), Garber (1996), etc. for discussions on the feasibility of such a tax. 3 offsetting some of the costs for the implementation of controls, although, in principle, there is no reason on revenue-raising grounds to prefer a tax measure.4 If implemented properly, the tax measures would be superior to administrative controls. Still the problem of evasion remains as investors will find ways to circumvent such taxes over time. To prevent this evasion problem, for example, authorities could incur serious penalties if they have violated reserve requirement laws by setting up disguised subsidiaries. Anticipation of such penalties would limit their resort to such activities. 3. Recent Experiences of Capital Controls: Measures, Effectiveness and Exit Strategy This section first presents case studies of recent experiences with controls on capital controls with a view to assessing the impact of different instruments used. Subsequently, the section attempts to draw some policy lessons. Appendix 4 provides a table that summarizes the cross-country experiences of capital controls and their economic impacts. (i) Country Experiences Chile (June 1991 - ) The Chilean monetary authorities have used two main mechanisms to discourage foreign borrowing: (i) a stamp tax of 1.2% on all foreign loans regardless of their maturity; and (ii) unremunerated reserve requirements (URR) of 20% for a period of up to one year on all new foreign borrowing except trade credits. The URR was held at the central bank initially for a time related to the maturity of the credit (a minimum of 90 days to a maximum of one year), which meant that the impact fell mostly on short-term flows. The central bank returned the funds when the URR come due. Following the initial implementation of June 1991, the URR was extended to all existing loans (1991) and foreign currency bank deposits at commercial banks (1992). Other measures were taken concurrently (see Table 1). In January 1992, the official exchange rate of peso was revalued by 5% and the exchange rate band was widened to 10% on either side of the reference rate, allowing the exchange rate to appreciate immediately by about 3%. Meanwhile, during 1991-92 open market sterilization operations were conducted by the central bank to sterilize the monetary effects of the capital inflows, while the restructuring of the banking system was accelerated. 4 In the case of compulsory reserve requirements, implicit tax revenues accrue to the central bank in the form of interest-free deposits. The central bank is free to invest the corresponding sums in fixed-term assets and to transfer the receipts to the appropriate fiscal authorities. Only for Chile was a calculation of the revenue impact of these measures feasible, and for Chile the revenue from taxes on capital inflows or from their equivalents was small (see Agosin and Ffrench-Davis (1996)). Thus in developing countries, where international financial transactions are minuscule by international standards, these measures should be adopted essentially for their regulatory and prudential value. 4  Table 1: Chile: Capital Controls Year Unremunerated Reserve Requirements (URR) Other policy measures 1990  All capital transactions to be reported and subject to approval.  Banking sector restructuring. 1991 20%  New foreign borrowing except trade  Stamp tax of 1.2% on foreign borrowing. credits, for a time related to the maturity of  The official exchange rate of the peso was the credit, from 90 days (min) and to one devalued by 5% and the exchange rate band year (max). was widened to 10%.  Phase extension on existing borrowing.  Extended to all outstanding foreign borrowing, except short-term credits (less than 6 months) and time period for which the central bank would hold requirements was increased to one year.  Reserve requirements on existing credit maturing between July and December 92 were phased out. 1992  Extended to foreign currency bank  Limits on travel allowances were raised deposits at commercial banks. from $1,000 to $3,000 for neighboring  Reserve requirement on foreign borrowing countries and from $3,000 to $5,000 for 30% was increased to 30% from 20%. travel to other countries.  Open market sterilization operations. 1994  All URR on foreign borrowing were required to be held solely in the US dollars. 1995  Extended to secondary transactions of  Accelerated liberalization on capital ADRs. outflows.  Extended to cover investment inflows that do not constitute an increase in the capital stock but only a transaction of assets from residents to non-residents. 1998 10%  URR was lowered to 10% to attract capital inflows. 0%  URR was further lowered to 0%. In March 1992 in a further attempt to control the surge in short-term capital inflows, exporters were allowed to keep 10% of their proceeds in foreign currencies, and in May 1992 the URR on foreign liabilities of commercial banks was increased to 30% from 20%. In October 1992, the URR regulations were changed to require reserves to be maintained on deposits for a full year, regardless of the maturity of the loan. This meant that for loans with maturities of less than one year, foreign investors must maintain reserves on deposit for longer than the maturity of the loan. Capital inflows surged again in 1995 and the URR was extended further to cover all secondary transactions of ADRs and investment flows that do not constitute an increase in the capital stocks. At the same time, capital outflows were liberalized in an accelerated manner in response to a strengthening balance of payments. Chile’s capital controls remained tight until the outbreak of the Asian crisis. Though not as disruptive as the Mexican crisis was, the Asian crisis has slowed capital inflows to Chile considerably starting in November 1997 and also has lead to sharp falls in stock and bond prices in the domestic markets. In June 1998, Chile had to ease the URR from 30% to 10% but tightened the peso’s trading band, struggling with capital outflows and 5  Table 2: Chile: Selected Banking Regulations Provisions Specific to Credit Institutions Open foreign exchange position Limit of 20% of capital and reserves. Foreign financial investment ceiling of 25% of capital and reserves. Lending to domestically in foreign exchange. Only foreign-trade-related loans are allowed. Other restrictions on lending. Limits on maturity mismatches. Banks are not allowed to keep shares in their portfolios. Source: IMF (1998) as reported by Eyzaguirre and Lefort (1998) pressures for devaluation. This policy mix did not work; capital outflows continued. In September 1998, the reserve requirement was lowered to 0%, but a separate control which requires foreign investors to keep their money in the country for at least a year is being kept in place. No further control was taken on capital outflows. The Chilean experience has widely been considered as effective (at least up to 1997) in the sense that its controls managed to influence the composition of foreign capital inflows without reducing the volume of total capital inflows to the country, and has therefore invited a large number of empirical investigation. The majority of the researchers however conclude that Chile owes its stability in large part to prudential financial regulations but not solely to capital controls (see Table 2), and that if any the introduction of capital controls of itself had only temporary effects (up to 6 months) on the composition of capital inflows (see, for example, Quirk and Evans, 1995; Edwards, 1998; Cardoso and Laurens, 1998, etc. Also see IMF, 1998 and the references in it). No research result has yet to emerge on Chile’s policy response to the capital outflows that were caused by the contagion effect of the Asian crisis in 1998. Colombia (September 1993 - ) In September 1993, in response to the strong inflows of foreign capital, the central bank of Colombia imposed an implicit tax on foreign borrowing, in the form of a non-interest bearing deposit requirement of 47% of the loan amount on all loans with a maturity of 18 months or less for the duration of the loan. As in the case of Chile, this scheme intended to raise the financial cost for short-term capital flows. The reserve would be held in the central bank for one year, but could be repurchased at a discount that depended on the date of repurchase. In addition, the central bank required that import payments be made within 6 months of the due date for the purpose of accelerating payments of outflows and increasing the cost of import financing. In an effort to prevent further capital inflows, the Colombian authorities tightened the restriction on foreign borrowing twice during 1994. In March 1994, the deposit 6 requirement was extended to foreign loans with maturities of up to 3 years. The period during which the deposits had to be maintained was changed so that the magnitude of the reserve requirement was a decreasing function of the maturity. Borrowers could then choose to place a 1-year deposit for 93%, an 18-month deposit for 64%, or a 2-year deposit for 53% of the loan amount. In August 1994, the requirements were extended further to all loans of up to 5 years maturity. The deposit as a percentage of the loan was also raised, ranging from 140% for loans of 30 days to 42.8% for 5-year loans.5 Also, the maximum period for payments of imports was shortened to 4 months from 6 months. Furthermore, rules for foreign borrowing for real estate purposes were tightened in March 1994; a minimum maturity for such loans was raised from 2 years to 3 years. In August 1994, all borrowings related to real estate transactions were prohibited. Despite these controls and the other changes in policy, the volume of net capital inflows did not change, though without the measures taken, inflows might have been much larger. While there is no hard evidence on the effects of Colombia’s reserve requirement mechanism, one of the results was the changes in the composition of inflows; the share of short-term flows declined and the share of long-term flows increased after the implementation of the controls. Brazil (October 1994 - ) Brazil implemented more explicit form of taxation on capital inflows as opposed to deposit requirements as in Chile and Colombia, which can be regarded as implicit taxation. In October 1994, in line with an inflation stabilization plan, the Real Plan, several taxes on capital inflows were introduced and/or tightened. A tax of 1% was levied on foreigner’s investment in stock market; a tax on foreign purchase of domestic fixed income was increased from 5% to 9%; and the tax Brazilian firms pay to issue bonds overseas was increased from 3% to 7%. In August 1995 restrictions on capital inflow were again tightened. A 5% tax was introduced on loans in foreign currency and on interbank operations between Brazilian and foreign institutions. Foreign exchange sales on the open market were taxed at the rate of 7%. The tax rate for foreign purchases of fixed income instruments, previously 5% was raised to 7%. Furthermore, in September 1995, a capital gains tax of 15% on stock market transactions that had applied only to residents was extended to non- residents. Brazil’s capital controls do not appear to have had the intended effects of limiting the volume of flows or changing their composition. Though it is difficult to determine what capital flows to Brazil might have been in the absence of the policy measures, sterilization operations together with relatively sophisticated Brazilian financial markets are most often blamed for undermining the effectiveness of measures to discourage capital inflows. 5 See Reinhart and Smith (1997), who show that the magnitude of controls on capital inflows must be large enough to accommodate even modest adjustment in the capital account. 7  Box 1. Capital Controls in Chile (reserve requirements) and Brazil (taxes) With the aim of preventing both over-indebtedness and a large share of short-term debt to total debt, Chile controlled capital transactions in such a way that the incidence is much higher on short-term capital. The policy imposed a reserve requirement on foreign liabilities of domestic firms, which had to be deposited with the central bank, with the requirement falling as the maturity of the funds went up. Meanwhile, Brazil implemented more explicit form of taxation, with tax rates ranging depending on types of foreign exchange transaction (1% to 9%), but regardless of the holding period. Both Chile and Brazil’s instruments were imposed in the belief that that composition of capital flows does matter. Therefore, the relevant way to tax capital inflows is according to the asset holding period (Chile) or to the types of capital transaction (Brazil). Supporters of this view argue that such a tax proportional to the holding period as adopted by Chile could help prevent bunching of capital outflows, thereby reducing the likeliness of capital outflows. Brazil’s strategy of taxing capital transactions at varying rate was intended to achieve a similar result, by influencing the composition of capital inflows (see Section 3 for cross-country experiences). It is worthwhile to know, however, that there exists another argument among economists that, because of the high degrees of substitutability and fungibility, composition of capital flows does not matter (Dooley, 1990; Claessens, Dooley and Warners, 1995; etc.), and therefore it is not relevant to reduce incentives to capital inflows according to the type or maturity of flows.1 While hardly is there any empirical evidence, the fact that capital controls in Colombia and Brazil did not have intended effects could imply that this mechanism might have been at work in these cases. ________________________________ 1 In fact, on this ground Tobin (1978) proposes to tax all foreign exchange transactions, including trade. Malaysia (January – August 1994) In the early to mid 1990s Malaysia also implemented controls to curve short-term capital inflows. In comparison to the above three Latin American countries that adopted tax-based measures, Malaysia resorted to a combination of administrative and tax-based measures. Bank Negara initially attempted to offset the effects of the capital inflows on domestic liquidity by sterilization operation – increasing direct borrowing from the money market, selling Bank Negara bills, issuing long-term savings bonds, transferring government and other deposits to the central bank, and raising the statutory reserve requirement. Limited controls on capital transactions were also imposed in June 1992, when non-trade related swaps by commercial banks were subject to limits. 8 However, as liquidity continued to grow substantially and capital inflows were sustained, in January 1994 the Malaysian authorities moved to limit speculative capital inflows through more extensive capital controls, which comprised of a combination of administrative and tax-based measures. These measures include (i) a ceiling on foreign liabilities of banking institutions other than those related to trade and investment; (ii) a prohibition on residents against selling short-term monetary instruments to non-residents; (iii) a prohibition against all non-trade related swap transactions and outright transfers on the bid side with non-residents; and (iv) an obligation for commercial banks to deposit at Bank Negara the ringgit funds of foreign banking institutions (vostro accounts of non- resident banking institutions) in non-interest bearing accounts. Between February and May 1994, these ringgit funds were also included in the eligible liability base for the calculation of statutory required reserves, resulting in a negative effective interest rate on these balances. The reliance of the administrative measures was intended to be short-term and was to serve the purpose of containing price pressures by contracting liquidity. The authorities were well aware of the danger of market distortions and inefficient resource allocations arising from long-lasting capital controls. Once the objective stability in the financial markets was achieved, in August 1994 capital control measures were lifted. The domestic interest rate fell from 6.5% at the beginning of 1994 to 4.5% in September 1994, while international interest rates rose, lowering the interest rate differential from about 3% to 0.5% in favor of the US interest rate. The ringgit exchange rate appreciated against the US dollar by 5.6% during 1994. The developments described above may suggest that Malaysia’s capital controls have succeeded in stemming short-term capital inflows. However, the sharp decline in domestic interest rates, together with exchange rate appreciation during most of 1994, would also have influenced yield-sensitive short-term capital. Consequently, it is difficult to determine whether the controls had a significant impact. Spain (September – November 1992) and Portugal (September – December 1992) Spain reintroduced controls on short-term capital outflows during the ERM crisis in 1992 to prevent devaluation out of the ERM band (Spain had given up all controls by mid- 1992). Spain adopted the deposit requirement approach, intending to raise the cost of and lower the attractiveness of engaging in particular transactions, as in Chile and Colombia. However, unlike Chile and Colombia where controls took the form of deposit requirements for external borrowing, Spain imposed reserve requirements for banks on their net foreign currency deposits. Domestic banks were required (i) to deposit at the central bank for one year without interest an amount equal to the peseta value of any new long positions in foreign currencies; (ii) to deposit an amount equal to the value of new peseta denominated loans to non-residents, except loans related to commercial activities; and (iii) to hold a cash reserve equal to the full amount of peseta liabilities in branches and subsidiaries of Spanish banks abroad or in domestic branches of foreign banks. Within one month of the implementation, however, the authority discovered that the coverage of the requirements were too broad and therefore they not only affected foreign exchange rate speculation, but also seriously limited financial operations and risk 9 hedging of associated with foreign trade. To target foreign speculation more precisely, the initial measures were soon replaced by a new set of requirement for non-interest bearing deposits at the central bank for the peseta counterparts of (i) for the same-day or next-day peseta sales to non-residents and also of (ii) new forward short position in foreign currency contracted with non-residents. Portugal also implemented controls on capital outflows as part of a defense of the exchange rate arrangement during the ERM crisis, but by means of enforcing the existing quantitative restrictions on some form of short-term capital transactions: short-term escudo lending was prohibited to non-residents; and non-residents were banned from purchasing domestic money market instrument. In both Spain and Portugal, the controls did not have impressive effects. Despite controls, in Spain the domestic interest rates did not fall, giving little monetary flexibility to the authorities, while in Portugal the escudo was devalued by 6% in November (Fieleke, 1994, and Barber and Taylor, 1995). However, compared with other ERM countries which did not impose controls (Norway and Sweden), Spain and Portugal experienced much smaller fluctuations in their domestic interest rates in trying to maintain their exchange rate targets (Fieleke, 1994). It might be possible to interpret this result that Spain and Portugal did acquire some temporary insulation. Both Spain and Portugal lifted the controls by the agreed EU deadline and neither of them reintroduced controls during the subsequent episodes of exchange rate tension within the ERM. (ii) Policy Lessons The following policy lessons can be drawn from the above country-experiences:  The results of capital controls are mixed. While tax-based controls have been considered as more transparent and easier to implement, with the exception of Chile, the evidence from cross-country experiences suggests that the effectiveness of controls in reducing (the proportion of) short-term capital inflows has been inconclusive. In all the countries concerned above, capital controls were adopted as part of a (more or less) comprehensive macroeconomic package and hence it is very difficult to isolate the effects of capital controls from those of other measures taken simultaneously. Also, developments in the external environment might have had impacts on the effectiveness of capital controls.  The literature generally concludes that Chile’s success in controlling inflows is attributed to its prudential financial regulation rather than to the controls per se. Nevertheless, Chile’s experience of failing to curtail capital outflows since late 1997 leaves a question of whether a prudential regulation alone can be sufficient to combat large international capital movements under the climate of deteriorating market confidence. 10  The effectiveness of controls on inflows seems to diminish over time as the private sector invests in techniques to avoid the controls. Chile, Colombia and Brazil all had to tighten the controls subsequently to avoid evasion.  Controls on capital outflows tend to be ineffective (Chile, Spain, and Portugal). This can partly be explained by the fact that most of the controls on outflows were implemented in the midst of the crises. Once a crisis has happened and market confidence has collapsed, hardly can any measures prevent capital outflows. 4. Should Malaysia Exit from the Present Policies? This section discusses the specific context of Malaysia. What are the downside risks of Malaysia’s present policies? Is there a need for modification in the short-run? If so, how? What is the long-term strategy? These are the issues addressed in this section. (i) Downside Risks of Malaysia’s Current Strategy The Malaysian authorities’ motives of capital controls are not negative inasmuch as the controls provide a short-term breathing room by insulating the domestic economy from further international financial volatility. Nevertheless, the resulting market distortions pose short- and longer-term risks to the economy. Apart from those risks common to all capital controls (administrative costs, distortions on resource allocation, etc., as discussed in Section 2), major downside risks Malaysia is facing include the following:6  Further deterioration in market confidence Malaysia’s present controls are self-defeating in the sense that, despite the fact that these controls are placed selectively, they have worked to discourage foreign capital inflows which are needed to spur economic recovery. For foreign investors the one- year holding requirement is the most onerous among the new measures aimed at controlling the movements of funds across the border. Indeed, many are not allowed to invest in Malaysia due mainly to the restrictions on currency convertibility and repatriation; their charters often prohibit them investing in such “closed” markets. As for FDI flows, though the controls are explicitly designed to exclude these flows, the negative impact that the controls give to the investors would be substantial. Given uncertainties over repatriation of interest, profits dividends and capital, investors would think twice about investing in Malaysia, opting instead for more liberal destinations.7 As noted, portfolio capital inflows has fallen sharply and FDI inflows have weakened further after the introduction of capital controls.8 The present controls will be particularly damaging for the labor markets and, more broadly, economic development, if they continue to deter FDI which is vital for Malaysia to 6 Other risks include a possibility of a development of a parallel exchange market; a possibility of additional trade controls if the measures fail to work, etc. 7 For example, in 1993 substantial FDI flows were diverted to China, which offers preferential policies for FDI (China is not a liberal destination for FDI flows). 8 Short-term inflows in 1998 are projected to be negative, -6% of GDP, in contrast to 4% of GDP in 1997. FDI has also slowed from around 7% of GDP in 1997 to 3-4% of GDP (projection) in 1998. 11  move up the manufacturing ladder (electronics, crucially dependent on FDI, alone makes up some 51% of total exports and is worth nearly 40% of GDP in 1996).  Capital outflows in 1999 With the present policy measures unchanged, it is estimated that as much as US$10 billion of portfolio investment could leave the country when the one-year holding requirement comes due in September 1999 (or earlier if a way can be found) if market confidence continues to deteriorate. Restoring market confidence well in advance will help prevent massive capital outflows.  Capital flight With domestic confidence still uncertain, capital flight may build up over time. Although the present measure prevents under-invoicing of exports and over-invoicing of imports (a standard technique of capital flight), there are ways to evade such restrictions. In fact, the balance of payments figures for 1997 already show some sign that a substantial capital flight has taken place during 1997.9 Combating the evasion of controls could require wider coverage of controls, compounding the initial distortions.  Potential delays in structural reforms There is a risk that capital controls combined with expansionary macroeconomic policies could be used as a substitute for needed financial and corporate sector reforms. This will not only undermine investor confidence but also threaten Malaysia’s medium-term economic recovery.  Internal inconsistency The policy mix of the fixed exchange rate regime, capital controls and expansionary monetary/credit policy is potentially explosive and could make the economy more vulnerable to external shocks. (ii) Exit strategy In view of the present low market confidence in emerging markets in general, it may not be advisable that Malaysia dismantle the controls completely in the coming months, although properly sequenced capital account liberalization should be pursued in the longer-term. The short- to medium-term policy strategy proposed below is intended to provide a temporary breathing space for the Malaysian authorities while major efforts in macroeconomic stabilization and structural adjustment are being accelerated. It constitutes modifying the existing measures so that, while selective capital controls are still in place to prevent volatile capital movements, they will appear to be more market friendly, thereby helping improve investors’ perception about Malaysia. 9 In the1997 balance of payments statistics, errors and omissions turned negative for the first time in 1990s, amounting to around US$6-8 billion. Much of this reflects aimed at avoiding taxes or other official regulations. 12 Here the objective of the government is three-fold: (a) reducing vulnerabilities; (b) attracting long-term capital inflows needed for economic recovery, while preventing volatile short-term capital flows; and (c) discouraging further capital outflows. a. Reducing vulnerabilities Malaysia is still facing vulnerabilities such as the fixed exchange rate arrangement, weakness in prudential regulation and supervision of the domestic financial system, lack of well-functioning capital markets, insufficient transparency, etc. These issues need to be addressed in a comprehensive plan for macroeconomic/structural adjustment (see Malaysia CAS and SPR for detailed discussions on macroeconomic and structural policies). A credible announcement of such a plan would help boost market confidence. b. Attracting long-term capital inflows, while preventing volatile short-term transactions The development on the surge in short-term capital volatility Malaysia has gone through during 1990-95 (see Figure 1) suggests that it is essential to have in place a policy strategy that distinguishes between long-term, stable capital inflows such as FDI and those that are considerably more volatile and that could have adverse effects on long-term growth. Volatile flows include short-term financial credits to banks and large domestic firms, short-term deposits by non-residents in the domestic financial system and purchases of domestic stocks and bonds by non-residents (portfolio investment). These flows seek to arbitrage interest rate differentials or to obtain quick capital gains. Excessive inflows of this type could lead to very sharp increases in domestic asset prices and unsustainable exchange rate appreciations, which are later reversed when the effects on domestic relative prices and the current account balance become evident. Then, as happened in Malaysia as well as in other countries, there follows an overshooting of capital flows in the opposite direction with asset prices falling and the real exchange rate depreciating more than is justified by the underlying fundamentals. In replace for the one-year holding requirement on capital inflows, the government could consider the following options to influence the character of these flows (see Appendix 2 for the table that summarizes the following proposals). Graduated exit tax: In attempting to differentiate between volatile short-term investments and long-term, more stable inflows such as FDI (and medium- to long-term borrowing from multilateral institutions), the cleanest option is to impose a tax on all outward-bound transactions, with the tax rate varying inversely proportional to the maturity structure (or holding period) of the investment.10, 11 Such a tax would appear prohibitively expensive for very short-term investments but a negligible cost for long- term investments, thereby preventing a large share of short-term flows to total flows that lays the ground for financial crisis. The tax should apply to all investments, regardless of maturity and types, so that it is difficult to evade/substitute between assets. 10 Public sector transactions may be exempt. 11 Note that transaction taxes are commonly used in industrial countries. See Cambell and Froot (1994). 13 Graduated reserve requirement: Alternatively, the government could consider imposing graduated holding requirements on portfolio investments. To the Malaysian authorities, the main advantage of these measures is that it can exert an influence on the exchange rate without the need for changes in interest rates or intervention in the currency markets. It should be noted however that the tax-based measures still share the disadvantages of other forms of capital controls: no matter how modest the tax rate is set, there would be a distortionary impact on the resource allocation. In the long-run, investors might demand higher rates of return to compensate for the tax, turning their attention to more attractive locations for investment, which suggests that this measure should be limited to the temporary use. The domestic cost of capital would rise. If used, the tax-based measure needs to be designed so as to minimize these side effects. Liberalization of the FDI regime: With diminishing expectations of a recovery in Japan and a greater than anticipated slowdown in the US, attracting higher FDI inflows will remain a challenge for Malaysia. Presently, a 100% ownership is allowed for export- oriented locally incorporated firms in Malaysia (where more than 80% products exported). However, there still remains a 30% limit on foreign investment in the banking sector, which is in need of fresh capital – a need which becomes even more acute as non- performing loans rise.12 Mr. Anwar’s failure to raise the 30% ceiling during the last parliament session in March 1998 was deemed as a big disappointment by overseas investors. This is a politically sensitive issue. However, now that other East Asian countries has recently moved ahead to liberalize their FDI regimes, easing such regulations seems imperative for Malaysia so as to stimulate muted investor interest, thereby attracting foreign direct investment.13 If replacing the current measures does not lead to resumed inflows of FDI, it might be necessary to implement additional policy measures to promote inward foreign investment such as preferential tax policies, especially in priority sectors (such as electronics and services), as well as to ensure that a high proportion of profits and other investment revenues of existing FDI is reinvested within Malaysia. c. Discouraging further capital outflows Capital export tax: With the aim of stemming outflows of domestic capital, acquisition of foreign securities and overseas currency transfers by residents could also be subject to a tax, which effectively works to lower the (risk-adjusted) rates of return on their overseas investment. 5. Conclusion 12 Also, foreign M&As require a prior approval. 13 For example, in response to reduced FDI inflows Thailand lifted the foreign ownership limit of 25% for financial institutions in June 1997. In Korea the negative list on FDI was reduced from 5% to 1.5% of all industries listed in the Korean standard listed classification, while FDI via M&As, including hostile takeovers, by foreigners were also permitted. 14  It cannot be overemphasized that the above policy options need to be accompanied by a comprehensive macroeconomic policy package and structural reforms. Capital controls by themselves will by no means eliminate the economy’s fundamental problems. Until the outlook for the banking sector reform and the government’s policy of assisting ailing firms become clearer, capital and investment flows are unlikely to return in force and capital flight may continue.  Capital controls over extended periods loose effectiveness, as repeated in the literature. These measures should therefore be regarded as temporary, designed to win breathing space for an economic recovery, but not as a permanent withdrawal of Malaysia from the international capital market. The long-term goal is properly sequenced opening-up of the capital account.  It is conceivable that Malaysia will do better than its neighbors – partly because its economy was in better shape to start with, and partly because the controls might even be used to carry out needed reforms. But if it abuses the breathing space offered by capital controls, then what its officials insist was merely a narrow technical decision will rapidly be exposed as an enormous economic bungle. If Malaysia does succeed in achieving a recovery, that will be lesson enough for the rest of developing countries. References Agosin, Manuel R. and Ricardo Ffrench-Davis (1996), “Managing Capital Inflows in Latin America” in Mahbub ul Haq, et al. (eds.), The Tobin Tax: Coping with Financial Volatility, Oxford University Press, New York, 1996. Cambell, John Y. and Kenneth Froot (1994), “International Experiences with Securities Transaction Taxes”, in J. Frankel (ed.), The Internationalization of Equity Markets, University of Chicago Press, Chicago. Cardoso, Jaime and Bernard Laurens (1998), “The Effectiveness of Capital Controls on Inflows: Lessons from the Experience of Chile”, mimeo, International Monetary Fund. Cardoso, Eliana and Ilan Goldfajn (1998), “Capital Flows to Brazil: The Endogeneity of Capital Controls”, IMF Staff Papers; 45(1), March 1998, pages 161-202. Claessens, Stijn, Michael P. Dooley and Andrew Warner (1995), “Portfolio Capital Flows: Hot or Cold?” World Bank Economic Review, 9, 1: 153-74. Dooley, Michael P. (1990), “Comment (on Razin and Slemrod)” in A. Razin and J. Slemrod (eds.), Taxation in the Global Economy, University of Chicago Press for the National Bureau of Economic Research. 15 Edwards, Sebastian (1998), “The Americas: Capital Controls Are Not the Reason for Chile’s Success”, Wall Street Journal, April 3, p. A19. Eichengreen, Barry, Rose and Charles Wyplosz (1996), “Taxing International Financial Transactions”, in Mahbub ul Haq, et al. (eds.), The Tobin Tax: Coping with Financial Volatility, Oxford University Press, New York, 1996. Eichengreen, Barry, Rose and Charles Wyplosz (1994) Eichengreen, Barry, James Tobin and Charles Wyplosz (1995), “Two Cases for Sand in the Wheels of International Finance”, Economic Journal 105 (January): 162-172. Eyzaguirre, Nicolas and Fernando Lefort (1998), “Capital Markets in Chile, 1985-1997: A Case of Successful International Financial Integration”, mimeo, International Monetary Fund. Fieleke, Norman (1994), “International Capital Transactions: Should They Be Restricted?” New England Economic Review (March/April): 27-39. Haq, Mahbub ul, Inge Kaul and Isabelle Grunberg (1996), The Tobin Tax: Coping with Financial Volatility, Oxford University Press, New York. IMF (1993), International Capital Markets: Developments, Prospects and Key Policy Issues, Washington, DC., International Monetary Fund. IMF (1998), International Capital Markets: Developments, Prospects and Key Policy Issues, Washington, DC., International Monetary Fund. Kenen, Peter (1996), “The Feasibility of Taxing Foreign Exchange Transactions”, in Mahbub ul Haq, et al. (eds), The Tobin Tax: Coping with Financial Volatility, Oxford University Press, New York, 1996. Quirk, Peter and Owen Evans (1995), “Capital Account Convertibility: Review of Experience and Implications for IMF Policies”, IMF Occasional Paper No. 131, Washington DC, International Monetary Fund. Reinhart, Carmen M. and R. Todd Smith (1997), “Temporary Capital Controls”, mimeo. Tobin (1978), “A Proposal for International Monetary Reform”, Eastern Economic Journal, 4 (July-October): 153-159. 16 Appendix 1 Malaysia: Key Capital Control Measures 1998 A. External ringgit accounts (comprises vostro, non-resident accounts by embassies, individuals and corporates including offshore Labuan) 14 a) Transfers between external accounts, except purchase of ringgit assets, now require approval for any amount.15 b) Transfers to resident accounts in Malaysian require Bank Negara approval. c) Sources of funding of external accounts are limited to proceeds of sales of ringgit assets in Malaysia, salaries, wages, commissions, interest, dividend and sales of foreign currency. d) Uses of funds in external accounts are limited to purchase of ringgit assets, administrative and statutory expenses, and payment of goods and services in Malaysia as well as to granting of loans and advances to staff in Malaysia pursuant to the terms and conditions of service. B. General payments a) Residents are freely allowed to make payments to non-residents for any purpose up to RM10,000 or its equivalent in foreign currency, and in any amount for all imports of goods and services. b) Forms P&R are required for amounts greater than RM10,000 equivalent (previously RM100,000). c) Investments abroad in any form and payments under a guarantee for non-trade purposes require approval. C. Exports of goods a) Prescribed manner of payment for exports in foreign currency only, other than currencies of Israel, Serbia and Montenegro. D. Credit facilities for non-residents a) Domestic credit facilities for non-resident correspondent banks and non-resident stockbroking firms are no longer allowed. E. Investments abroad a) Payments by residents with no domestic borrowing to non-residents for purposes of investing abroad are limited to RM10,000 or its equivalent in foreign currency per transaction (previously no limit). b) All residents require prior approval to make payments to non-residents for purposes of investing abroad for amounts exceeding RM10,000 equivalent in foreign currency (previously, domestic corporates with domestic borrowing have been able to invest abroad up to RM10 million). F. Resident credit facilities from non-residents a) Residents are no longer allowed to obtain ringgit credit facilities from any non-residents (previously allowed up to RM100,000). G. Securities a) All ringgit securities must be deposited with and transacted through authorized depositories. 14 Amended rules affect mainly the ringgit transactions of non-resident corporations. For external accounts maintained by embassies, consulates, central banks and international organizations, no restriction is placed on sources and uses of funds and conversion into foreign currency. 15 Ringgit assets include properties, ringgit securities, savings accounts and FDs. 17 b) All payments by non-residents for any Malaysian securities must be made in foreign currency or in ringgit from external accounts. c) All proceeds in ringgit received by non-residents from sale of Malaysian securities must be retained in ringgit external accounts or held in another form of ringgit asset. For ringgit securities held longer than a year as of September 1, 1998, the proceeds from them can be immediately converted to foreign currency or credited to the external account. d) All payments to residents for any foreign securities from non-residents must be made in foreign currency. H. Import and export of currency notes, bills of exchange and assurance policies With effect from October 1, 1998, A resident traveler is permitted to: a) Import ringgit notes up to RM1,000; b) Import any amount of foreign currencies; c) Export ringgit notes up to RM1,000; d) Export foreign currencies up to the equivalent of RM10,000. A non-resident traveler is permitted to: a) Import ringgit notes up to RM1,000; b) Import any amount of foreign currencies; c) Export ringgit notes up to RM1,000; d) Export foreign currencies up to the amount of foreign currency brought into Malaysia. Prior BMN approval is required for import and export of ringgit notes and export of foreign currency notes other than as permitted above (approval given within one day). I. Labuan International Offshore Financial Center Licensed offshore banks in Labuan are no longer allowed to trade in ringgit instruments. 18 Appendix 2 : Summary of Capital Control Measures Objective Measures  Reduce short-term volatility in the exchange  Restriction on external account transactions rate, by curbing internationalization of ringgit. (A).16  Settlement of all trade in foreign currencies (C).  Restriction on export and import of ringgit (H).  Limits on foreign currency notes held by travelers (H).  Tightening of rules on overseas investments by residents (E).  Fixing the exchange rate.  Monitor gross flow of short-term ringgit  Transactions in securities effected through assets. authorized depositories (G-a).  Reduce volatility in short-term capital flows.  Investment in ringgit financial assets held in Malaysia for a minimum of one year (G-c). 16 Ref. table in Appendix 1. 19 Appendix 3: Proposals for Malaysia’s Exit Strategy Chile’s deposit Tobin tax Malaysia exit strategy proposal requirement on proposal Graduated exit Graduated Capital export inflows tax reserve tax requirement Motive Prevent over- Reduce volatility Reduce capital outflows indebtedness of exchange rates Reduce volatile capital movements Tax applied to Capital inflows All foreign Capital outflows Capital outflows Foreign exchange investment by transactions, residents (capital including trade outflows) Paid by Foreign investors All traders Foreign investors Foreign investors Domestic investors Paid to Central bank Tax authority Bank Negara Bank Negara Tax authority (foreign currency earning) Relationship of Rises with Invariant to Rises with Rises with Rises with tax amount to foreign interest interest rate foreign interest foreign interest foreign interest interest rate rate rate rate rate Relationship to Tax falls at 3 Fixed amount. Falls with Falls with maturity months In percent per maturity maturity year terms, falls continuously with maturity Where imposed One country Must be world- Malaysia (facing outflows) (facing inflows) wide Probable level Moderate Low (to avoid High (to discourage speculative of tax rate distortions and attacks) substitution) 20 Figure 1 Malaysia: Composition of Capital Flows (% of GDP) 20.0 15.0 10.0 Loans 5.0 FDI (net) Portfolio equity 0.0 flows Total private net 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 flows Official net -5.0 flows -10.0 21