Chapter

IV. Policy Responses

Author(s):
Robert Kahn, Adam Bennett, María Carkovic S., and Susan Schadler
Published Date:
September 1993
Share
  • ShareShare
Show Summary Details

In each of the six countries, the choice between allowing the inflows to stimulate demand and growth, on the one hand, and curbing their possibly destabilizing effects, on the other, came quickly to the fore. There were a wide range of responses (Box 1). In Egypt and Mexico, policy action was relatively subdued for various reasons: a stronger desire to stick to a predetermined course for policies; in Mexico, a somewhat greater willingness to accept the Lawson doctrine than in the other countries; and in Egypt, the still relatively short experience with large inflows. Actions were relatively aggressive in Chile and Colombia, where credit policies were fully dedicated to reducing inflation. Spain and Thailand occupied a middle ground: inflows were perceived as a response to fundamental changes in the attractiveness to foreign investors, but the threat of overheating was still a major concern.

The absorptive capacity of the economy was also an important issue, particularly in two areas. The first was the ability of the banking system to intermediate a large increase in credit. The efficiency of the banking system, always a concern, took on greater urgency as flows through the banking system and the scope for systemic risk rose. A second concern was the adequacy of physical infrastructure. In Thailand, where the inflows produced exceptionally high growth, this was acute and conflicted with the desire to accommodate inflows through a wider fiscal surplus.

This section provides a review of the policies with which countries attempted to avert destabilizing effects and improve the absorptive capacity of the economy. An assessment of the effectiveness of these policies is given in the following two sections.

Sterilization

Sterilization was the first line of defense against the surge in inflows, regardless of its cause, in each of the countries. Sterilizing was attractive because it could be implemented quickly and bought time to consider the likely causes and persistence of inflows and to formulate a longer-term response. It sought to curb the monetary effect of the inflows and tended to lock in a cushion of reserves against a possible reversal. As inflows persisted most governments eased the degree of sterilization.

Narrowly defined, sterilization is the exchange of bonds rather than money for foreign exchange. In fact, open-market operations were only one of the mechanisms that central banks employed to limit the influence of capital inflows on money. Increases in reserve requirements on all or selected parts of bank deposits, often designed to lengthen the maturity structure of deposits and discourage foreign inflows to the banking system, were also important. Other sterilization-type policies included various forms of central bank borrowing from commercial banks; the shifting of government deposits from commercial banks to the central bank; raising interest rates on central bank assets and liabilities; curtailing access to rediscount facilities; and, in Spain, direct credit controls. Sterilization was also accomplished through sales of government liabilities: this was important in Egypt, where the Government sold bonds to reduce money creation from deficit financing. To capture the bulk of these mechanisms, a broad measure of sterilization is required—namely, changes in the domestic assets of the central bank net of liabilities to the public sector and banks—that is, the domestic counterpart of currency issue.

Another difficulty in identifying sterilization is determining causality—whether the reduction in net domestic assets caused subsequent capital inflows or offset previous inflows. In fact, for most of the countries both lines of causality were undoubtedly at play. Except in extreme circumstances—when interest rates are invariant to the supply of bonds or to banks’ free reserves, or inflows are insensitive to interest rate differentials—sterilization will, at least to some degree, increase inflows. Nevertheless, estimates, reported in Appendix I, of the offset coefficient, that is, the degree to which sterilization induces offsetting capital inflows, suggest that, at least historically, the countries under review, with the possible exception of Thailand, had scope for effective sterilization. It must be remembered, however, that for all of these countries domestic financial markets have become increasingly integrated with foreign markets and estimates of offset coefficients using historical data probably underestimate the mobility of capital.

Box 1.Timing and Sequencing of Policy Responses to Capital Inflows1

PolicyTiming
Chile (1990)
Sterilization1990–91
Lifting of restrictions on capital outflowsFebruary, April, May, October 1991
Nominal appreciationApril, May, June 1991
Trade liberalizationJune 1991
Capital controlsJune, July 1991
Nominal appreciation and widening of exchange rate bandJanuary 1992
Capital controlsJanuary, May 1992
Lifting of restrictions on capital outflowsMarch 1992
Colombia (1991)
Sterilization1991–92
Introduction of foreign exchange certificatesJune 1991
Taxes on capital inflowsJune 1991
Trade liberalizationJune, August 1991, and January 1992
Lengthening of maturity of exchange certificatesNovember 1991
Lifting of restrictions on capital outflowsJanuary–February 1992
Taxes on capital inflowsFebruary 1992
Egypt (1992)1991/92
Fiscal adjustment1991/92
Sterilization
Mexico (1989)
Reduced rate of crawlMay, November 1990
Trade liberalization1990
Lifting of restrictions on capital outflowsNovember 1991
Introduction of widening exchange rate bandNovember 1991
Enhanced bank supervision and capital controlsApril 1992
Further widening of exchange rate bandOctober 1992
Spain (19B7)
Trade liberalizationYearly during episode
Sterilization1987
Capital controlsApril, July 1987
Lifting of restrictions on capital outflowsMay 1987
Capital controlsJune 1988, January 1989
Joining the exchange rate mechanism (ERM) of the European Monetary SystemJune 1989
Lifting of restrictions on capital outflowsDecember 1989 March, December 1990 March, April 1991
Elimination of capital controlsFebruary 1992
Exchange rate devaluationSeptember 1992
Thailand (1988)1988–91
Fiscal adjustment1988–90
SterilizationOctober 1990
Trade liberalizationApril 1991
Lifting of restrictions on capital outflows
Source: IMF staff reports.1 First year of episode noted in parentheses after country name.

Ignoring questions of causality, sterilization—broadly defined as a deceleration or drop in the net domestic assets offsetting the monetary effects of capital inflows—occurred in each country in the first year of the capital inflow episode (Table 3). This deceleration reflected restraint in net credit to the private sector, except in Mexico and Thailand. After the first year, the decline in net domestic assets slackened in most countries owing partly to lower growth in reserves, but also reduced reliance on sterilization. The exceptions were Mexico and Thailand, where restraint of credit to the private sector strengthened (although only moderately) in the second year of the inflow. Sterilization occurred on the largest scale in Chile and Colombia—countries where the initial tightening of credit probably also played a large role in attracting the increase in inflows. In these countries, sterilization took the form of increases in reserve requirements and open market operations. In Colombia, about one year into the inflow episode, the central bank shifted to heavy reliance on the sale of foreign exchange certificates—forward foreign exchange assets bearing a market interest rate in exchange for current foreign exchange receipts.

Table 3.Net Domestic Assets of the Central Banks1(Percent contribution to increase in currency)
Average of

Three Years

Prior to

Episode
Year

Prior to

Episode
Inflow Episode
Year 1Year 2Year 3Year 4
Chile (1990)
Net domestic assets–73.2–223.2–414.3–178.0–230.5
Public sector2310.1–127.1115.2–0.9–95.1
(Public sector borrowing requirement)(–7.7)(–129.6)(–62.2)(–45.2)(–76.5)
Private sector3–383.3–96.1–529.5–177.1–135.4
Colombia (1991)
Net domestic assets–67.4–82.0–186.6–115.5
Public sector6.716.613.414.5
(Public sector borrowing requirement)(45.7)(21.4)(32.5)(30.5)
Private sector3–74.1–98.6–200.0–130.1
Egypt (1991/92, July–June)
Net domestic assets41.5–10.0–57.6
Public sector43.112.5–24.1
(Public sector borrowing requirement)(118.2)(139.7)(45.0)
Private sector3–1.6–22.5–33.5
Mexico (1989)
Net domestic assets–31.9352.542.4–53.6–104.9–2.1
Public sector184.5369.535.430.7–75.8–16.7
(Public sector borrowing requirement)(804.1)(700.1)(219.0)(127.7)(9.2)(–14.2)
Private sector3–216.4–17.07.0–84.3–29.114.6
Spain (1987)
Net domestic assets–1.23.1–55.2–18.48.37.9
Public sector–13.2–20.113.7–1.2–13.3–6.7
(Public sector borrowing requirement)(100.2)(70.3)(58.4)(52.6)(49.6)(48.6)
Private sector312.023.1–68.9–17.321.614.6
Thailand (1988)
Net domestic assets–13.6–24.0–43.6–69.6–57.9–55.7
Public sector–1.1–9.2–47.0–9.916.3–2.0
(Public sector borrowing requirement)4(13.0)(–36.9)(–78.9)(–89.2)(–52.5)(–4.7)
Private sector3–12.5–14.83.5–59.7–74.2–53.7
Source: IMF staff reports and estimates.

Domestic assets of the central bank net of liabilities to the government and commercial banks. First year of episode noted in parentheses after country name.

Public sector assets in Chile are calculated net of medium- and long-term foreign liabilities of the central bank, most of which were acquired in the context of 1983–85 rescheduling agreements with foreign commercial banks.

Including the banking sector.

Fiscal accounts (fiscal year October–September) converted to calendar-year basis.

Source: IMF staff reports and estimates.

Domestic assets of the central bank net of liabilities to the government and commercial banks. First year of episode noted in parentheses after country name.

Public sector assets in Chile are calculated net of medium- and long-term foreign liabilities of the central bank, most of which were acquired in the context of 1983–85 rescheduling agreements with foreign commercial banks.

Including the banking sector.

Fiscal accounts (fiscal year October–September) converted to calendar-year basis.

Countries reduced their reliance on aggressive sterilization for several reasons. First, it carried large quasi-fiscal costs. These were most clear when sterilization took the form of open market operations. Then, the net cost to the central bank per bond issued was the difference between the interest on domestic bonds and the return on foreign reserves—interest earned, for example on U.S. Treasury bills, plus the actual depreciation of domestic currency. This difference was largely a reflection of the domestic risk premium and ranged in 1990–91 from virtually zero in Thailand to over 10 percent in Chile, Colombia, and Spain. For Colombia—where bond sales were greatest among the countries reviewed—the direct quasi-fiscal cost in 1991 is estimated at 0.8 percent of GDP. Such calculations establish the lower limit of actual costs. A broader measure would also capture the effect of higher domestic interest rates on the government’s debt-service payments. The costs of other forms of sterilization are varied. For example, an increase in unremunerated reserve requirements would be profitable for the central bank.

Second, aggressive sterilization through open market operations risked intensifying the conditions that attracted large inflows: by augmenting the supply of bonds, it put upward pressure on domestic interest rates, sustaining an important attraction to additional inflows. Again the experience of Colombia is instructive. A marked reduction in sterilization in mid-1992 resulted in a large drop in short-term interest rates, a slackening of the inflows, and a deceleration of the monetary aggregates.

Third, by forcing foreign inflows into purchases of government bonds and sustaining high domestic interest rates, sterilization deprived the economy of the benefits of capital inflows—higher domestic investment and growth. This consideration carried particular weight in Thailand as the durability of the inflows and their contribution to investment and growth became clear.

Exchange Rate Policy

At the outset of the surge in inflows, each of the countries, except Spain, maintained a fixed, or broadly fixed, rule for the exchange rate. Initially, the authorities in each country took the position that the sustainable real exchange rate over the medium term was not significantly higher than its level prior to the surge in inflows. However, unchecked, inflows fed demand for nontraded goods, pushing up prices and the real exchange rate. Thus, whether inflationary pressures should be pre-empted through a nominal appreciation became a policy issue.

Four countries, Chile, Colombia, Mexico, and Spain, resorted to nominal appreciation or a modification in the announced path for the exchange rate (Chart 4 and Box 2). Each of these countries, except Mexico, had initially responded to the surge in inflows with relatively aggressive sterilization. When it became apparent that this could not be done on a scale sufficient to reduce money growth and inflation to targeted levels, exchange rate policy was adjusted, often in conjunction with the imposition of capital controls.

Chart 4.Nominal Exchange Rate Indices1

(Index of SDR per unit of national currency; period t—1 = 100)

Source: Information Notice System.

1 Period t is the first year of the surge in capital inflows, the timing of which is indicated in parentheses after country name.

Box 2.Exchange Rate Policy Changes During Episodes1

Chile (1990)

  • At the outset of the inflow episode, the government’s exchange rate rule was to adjust the reference rate for the peso against the U.S. dollar by the difference between domestic and U.S. inflation. The peso was allowed to fluctuate within a band around the reference rate. In response to the capital inflows, the Government undertook a series of small revaluations.
  • April 1991: Revalued reference exchange rate 0.7 percent.
  • May 1991: Revalued reference exchange rate 0.7 percent.
  • June 1991: Revalued reference exchange rate 2 percent.
  • January 1992: Revalued the reference exchange rate by 5 percent and widened the band around the reference rate within which the market rate is allowed to fluctuate to 10 percent in either direction (previously 5 percent). Within a short time, the peso moved to the top of the band. Thus, January 1992 measures led to a nominal appreciation of about 10 percent.

Colombia (1991)

  • June 1991: Nominal revaluation of 2.6 percent in foreign currency terms. Previously, foreign exchange operations were conducted at an official exchange rate that crawled against the U.S. dollar. Also, a system was introduced to exchange foreign exchange receipts for non-interest-bearing exchange certificates denominated in U.S. dollars with a 90-day maturity issued by Banco de la Republica. At maturity, the certificates are exchanged for domestic currency at a “reference” exchange rate set by Banco de la Republica. The certificates are redeemable before maturity at a discount. The market price of foreign exchange certificates became the exchange rate.
  • November 1991: The maturity of the certificates was extended to one year. This amounted to another nominal revaluation of 1.9 percent in foreign currency terms.

Mexico (1989)

  • 1990: Preannounced depreciation of peso against U.S. dollar reduced from around 15 percent a year (1.3 percent a month) to 10 percent a year (May) and 5 percent a year (November).
  • November 1991: Reduced preannounced depreciation of the peso against U.S. dollar from annual rate of 5 percent to 2½ percent a year and simultaneously created and gradually widened the band within which the peso could fluctuate. The aim was to move gradually toward a margin of about 4 percentage points between the maximum and minimum points of the band.
  • October 1992: Increased the rate of depreciation of the maximum point of the exchange rate band so as to widen the band by some 9 percentage points by the end of 1993.

Spain (1987)

  • June 1989: Entry into the exchange rate mechanism (ERM) of the European Monetary System.
  • September 1992: Devaluation of the peseta by 5 percent from the reference rate within the ERM.
  • November 1992: Devaluation of the peseta by 6 percent from the reference rate within the ERM.
Source: IMF staff reports.1 First year of episode noted in parentheses after country name.

Exchange rate action unleashed questions about what role market forces should have in determining the change. The case against giving freer rein to market forces rested on several considerations: a view—prevalent in Spain—that a market-determined appreciation encouraged speculative behavior; the presumption that restraining reliance on nominal appreciation would limit the risk of overshooting and force more of the adjustment onto fiscal policy, which would limit inflation and the real appreciation; and the desirability of preserving a credible nominal anchor should the inflows reverse. For Spain, these arguments supplemented the underlying objective of harmonizing with the EC and led to the government’s decision to join the ERM two years into the inflow episode.

By contrast, Chile and Mexico moved to greater, albeit limited, flexibility by enlarging the band within which their exchange rates moved. Three considerations were important: first, the position of the exchange rate in the band provided an indicator of the appropriateness of the rate, which helped assuage a public perception that the domestic currency was overvalued; second, permitting some flexibility increased the exchange rate risk for foreign investors; and third, a reversal in inflows would elicit an immediate offsetting policy response through a change of the exchange rate in the band.

Paradoxically, the increase in current account deficits that frequently accompanied surges in inflows often raised questions of whether a devaluation would be desirable. In general, the answer was no; by attracting larger capital inflows, a devaluation would exacerbate the signs of overheating, inflation and widening current account deficits, and ultimately be self-defeating. Indeed, it is reasonable to ask whether even the small nominal crawls in the Latin American countries did not simply hold up inflation and domestic interest rates.

Fiscal Policy

In the face of large and persistent inflows, a tightening of fiscal policy is generally the only means of containing inflation and avoiding a real appreciation. To the extent that the inflows stem from an unsustainable financial policy mix—relatively tight credit with an easier fiscal policy—reducing the fiscal deficit eliminates the problem at its source. When other causes for inflows are at play, fiscal adjustment—additional to the adjustment that might already have occurred—restrains domestic demand and inflation.

The form of fiscal adjustment influences its restraining effect. First, an adjustment that moderates demand for nontraded goods and services, either directly by cutting government spending on them or indirectly by raising taxes, reduces domestic inflationary pressures; however, one that falls on traded goods tends to strengthen the external current account, which could even add to pressures for appreciation. Second, discretionary measures, to the extent that they are perceived as sustainable, have a stronger effect than cyclical influences on inflation expectations and interest rates, although both reduce demand. Third, unless private savings behavior fully offsets shifts in government saving, containing expenditure should have a stronger effect on domestic demand than revenue increases: expenditure cuts affect demand directly, but revenue increases absorb resources that might have been saved.

In fact, despite the strong cyclical positions of the economies, fiscal adjustment did not figure significantly in most countries’ responses to the inflows (Table 4). Although fiscal positions generally improved in the first year of the inflow episode, this was less a response to the inflows than a continuation of the fiscal consolidation that in part attracted the inflows in the first place: little further change occurred thereafter.5 Thailand stands out as an exception. There, the position of the central government relative to GDP swung by over 7 percentage points from the immediate pre-inflow year to the peak fiscal surplus four years later. The improvement reflected fiscal measures—particularly to contain current expenditure—and a strong cyclical improvement in revenues. The financial position of the overall public sector also improved but by less. In Egypt, where fiscal consolidation began only in the first year of the inflow, efforts to raise revenues and contain expenditure continue, but the inflow period is not long enough to establish a meaningful record.

Table 4.Fiscal Policy During Inflow Episode1(In percent of GDP, unless otherwise indicated)
Average of

Three Years

Prior to

Episode
Year

Prior to

Episode
Inflow Episode
Year 1Year 2Year 3Year 4
Chile (1990)
Nonfinancial public sector balance2.85.33.82.33.2
Government revenue221.621.118.320.821.6
Government current expenditure24.522.021.121.420.5
Operating surplus of public enterprises312.011.510.97.67.0
Capital expenditure6.35.44.44.64.9
Real GDP growth7.710.02.16.010.4
Colombia (1991)
Nonfinancial public sector balance–1.5–0.6–0.2–0.4
Revenue22.723.725.026.2
Current expenditure16.616.818.119.1
Capital expenditure7.77.57.17.6
Real GDP growth3.94.32.13.5
Egypt (1991/92, July–June)
Overall government balance–16.0–17.2–5.0
Revenue26.528.934.8
Current expenditure28.930.130.6
Capital expenditure13.516.09.2
Real GDP growth2.52.10.3
Mexico (1989)
Operational balance4–1.4–4.5–1.72.62.33.3
Primary balance4, 54.46.08.17.85.55.6
Revenue30.230.429.629.927.027.0
Current expenditure520.720.618.018.217.918.2
Capital expenditure5.13.93.53.93.63.2
Real GDP growth–0.31.23.34.43.62.6
Spain (1987)
General government balance–6.2–6.0–3.1–3.3–2.8–4.0
Revenue35.436.237.837.839.839.3
Current expenditure38.138.537.537.338.238.4
Capital expenditure3.53.73.43.84.44.9
Real GDP growth2.63.85.65.24.83.6
Thailand (1987/88, October–September)
Consolidated nonfinancial public sector balance–4.2–1.61.34.14.73.9
Central government balance–4.1–2.20.73.34.84.9
Revenue16.316.517.618.620.120.3
Current expenditure16.315.313.612.912.311.9
Capital expenditure4.13.53.22.43.03.5
Real GDP growth6.89.513.212.010.08.2
Source: IMF staff reports and estimates.

First year of the episode noted in parentheses after country name.

Includes net capital revenue.

Before taxes and transfers.

Includes public sector earnings (primarily oil revenues).

Excluding interest.

Source: IMF staff reports and estimates.

First year of the episode noted in parentheses after country name.

Includes net capital revenue.

Before taxes and transfers.

Includes public sector earnings (primarily oil revenues).

Excluding interest.

In general, the minimal use of fiscal adjustment as a response to the inflows reflected the difficulty of garnering public support for a tightening of fiscal policy, particularly in the wake of already significant adjustments. Added to this was the need in some countries for investment in infrastructure to accommodate higher investment stemming from the inflows. Yet, while this was a central factor in Spain, where capital outlays rose significantly, in Thailand even a vigorous public investment program just kept pace with rapidly rising GDP.

Were there other considerations that would have militated against fiscal action in the face of inflows? In principle, the initial fiscal position and stock of government debt could be important. Of course, for Spain, and especially Egypt, where deficits were still large, the inflow episode offered an ideal opportunity for vigorous adjustment. For the other countries, where fiscal positions were almost in balance or in surplus by the first year of the surge in inflows, fiscal adjustment needed to be viewed against the desired fiscal balance and debtor/creditor position of the government over the medium term. In fact, even for these countries, the debtor position of the government still provided substantial scope for further fiscal consolidation.

Microeconomic Policies

Although the thrust of the response to surges in inflows was focused on macroeconomic policy, microeconomic policies also played a role in two ways: in some countries, impediments to inflows or trade liberalization appear to have slowed down the inflows for short periods; and financial sector reform and trade liberalization increased the efficiency with which inflows were absorbed.

Impediments to Inflows

Like sterilization, impediments to inflows were quick and easy to impose. Unlike sterilization, however, by locking out inflows they precluded even future benefits of incipient inflows for investment and growth. Resort to them, therefore, followed efforts to sterilize and reflected several considerations: a judgment that the inflows were prone to reversal and would not provide lasting benefits to the economy; concerns about the effect of sterilization on interest rates; and fears about the efficiency with which the economy could absorb inflows—because of implicit government guarantees, inadequate infrastructure, or deficiencies in the banking system. Against these considerations was the threat that prolonged restrictions or taxes on inflows to the banking system would weaken bank profits and encourage disintermediation.

Each of the countries, except Egypt, attempted to slow the inflows through controls, taxes, or other impediments (Box 3). Most efforts to limit inflows relied on some form of taxation: adjustments, extensions, or increases in reserve requirements, often unremunerated, on bank deposits or other credits from abroad; changes in commissions on foreign exchange transactions with the central bank; or withholding taxes on interest payments abroad. Direct controls were imposed in Spain (1988), where prior authorization for certain foreign borrowing was required, and Mexico (1992), where banks’ acceptance of foreign deposits was limited to a fraction of domestic deposits. Another approach was to eliminate existing market imperfections or implicit subsidies that were attracting inflows: Chile eliminated special debt and equity and other swap facilities for nonresidents; and Mexico tightened prudential standards on banking activities funded with external borrowing.

Box 3.Controls, Taxes, and Other Impediments to Capital Inflows During Episodes1

Chile (1990)

  • June 1991: The central bank introduced a non-interest-bearing reserve requirement of 20 percent on all new credits from abroad (excluding trade credits) and increased commissions on swap operations.
  • July 1991: Reserve requirement extended to all outstanding external credits (rescheduled, government-guaranteed bank loans exempted); implemented over a six-month period.
  • January 1992: Reserve requirement extended to cover foreign loans to foreign currency denominated deposits held by commercial banks.
  • May 1992: Government broadened the base of the 20 percent reserve requirement on foreign currency bank deposits and loans by introducing a marginal reserve requirement on interbank deposits of 30 percent; it also raised the reserve requirement on new credits to 30 percent. To encourage borrowing with longer maturities, the requirement was designed to make the tax fall as the maturity of the foreign loan increased.

Colombia (1991)

  • June 1991: Imposition of 3 percent withholding tax on foreign exchange receipts from personal services rendered abroad and other transfers, which could be claimed as credit against income tax liability.
  • February 1992: Banco de la Republica increased commission on its cash purchases of foreign exchange from 1.5 percent to 5 percent.

Mexico (1989)

  • April 1992: Government issued new regulations limiting most foreign currency liabilities of Mexican banks (e.g., Euro CD issues) to 10 percent of their total loan portfolio; in addition, 15 percent of that amount had to be placed in highly liquid instruments, such as top quality commercial paper.

Spain (1987)

  • April 1987: Imposed reserve requirements on Spanish bank accounts in convertible pesetas held by nonresidents. Banks prohibited from paying interest on convertible peseta balances exceeding Ptas 10 billion.
  • July 1987: Short-term repurchase agreements by nonresidents on domestic financial assets were forbidden.
  • June 1988: Reintroduced prior authorization for foreign financial loans over Ptas 1,500 million to resident borrowers, with average maturity less than three years.
  • January 1989: Imposed 30 percent nonremunerated reserve requirement on new foreign loans by residents and 20 percent nonremunerated reserve requirement on banks’ net foreign currency borrowing.

Thailand (1988)

  • March 1990: Eliminated exemption from with-holding tax on foreign borrowing.
Source: IMF staff reports.1 First year of episode noted in parentheses after country name.

Without counterfactuals, it is impossible to verify the period over which impediments affected inflows; however, the experiences of Chile and Spain, where they were implemented most aggressively, illustrate two points.

First, capital inflows slackened temporarily when the impediments were implemented (the second year of the inflow episode). Although nominal exchange rates were adjusted and sterilization was eased at this stage, it is likely that the introduction of impediments contributed to this slowdown.

Second, within one year impediments needed to be broadened several times. In Chile, reserve requirements were initially placed on foreign credits but, following the course of evasion, were extended over the course of one and a half years, first to all foreign deposits and then to interbank deposits. A similar pattern occurred in Spain. There, evasion also occurred by channeling inflows through subsidiaries of foreign companies, making them direct investment rather than loans or deposits.

Financial Sector Reform

In several of the countries, there were longstanding distortions in the financial sector and weaknesses in banking supervision and prudential standards.6 Phased programs for addressing some of these problems were planned or in progress, but the surge in inflows added to the urgency of the reforms for three reasons: first, because the inflows tended to increase the flow of funds through the banking system they could encourage risky lending behavior; second, to the extent that inflows increased banks’ open foreign exchange positions, they increased their exposure to exchange risks; and, third, the inflows raised the potential for pressures on the government to support institutions in the event of any reversal in inflows. To address both of these types of risks, regulations to ensure the adequacy of capital bases and limit banks’ exposure to foreign exchange risk were essential. Although explicit deposit insurance existed only in Chile, Colombia, and Spain, other governments had stepped in to avoid bank failures in previous banking crises, and the risk of implicit government guarantees was substantial.

Generally, programs of financial reform were being implemented at the time the surge in inflows began and remaining steps took place as planned—not accelerated or broadened in response to the inflows. Most countries focused on improving supervision and prudential standards and adopting capital adequacy standards in accordance with guidelines under the Basle accord. Other notable steps were reducing direct controls on interest rates and credit in Egypt, Spain, and Thailand; downsizing and privatizing government-owned banks in Colombia and Mexico; and easing access of foreigners to the public debt market in Spain. By and large, these changes were not a response to the surges in inflows and may even have increased the attractiveness of domestic financial markets. One area where steps were taken in response to the inflows, however, was the limiting of bank’s open foreign exchange positions. Such limits, which were in place in most of the countries prior to the surge in inflows, were introduced or tightened in Egypt and Mexico during the inflow episode.

Easing Restrictions on Capital Outflows

There were two motives for easing restrictions on capital flows: first, to enhance efficiency, by subjecting domestic financial markets to greater international competition and by enabling residents to reduce risk by diversifying their portfolios; and second, to encourage outflows. Whether such measures actually result in net outflows is unclear. Where capital flight had been sizable, the volume of pent up capital would not be large. Also, simplifying the process for repatriating profits and capital might attract inflows.

Most of the countries under review relaxed controls on private capital outflows (Box 4). Generally, this step was taken after others had been tried. This reflected the recognition that the relaxation of controls on outflows should be part of a move to capital account convertibility insofar as reimposing capital controls in previously integrated capital markets is usually not effective. Changes fell into three broad categories: domestic residents and institutions were allowed to make portfolio investments abroad; surrender requirements on foreign exchange earnings were reduced; and restrictions on profit and capital repatriation by foreign firms were eased. The most significant changes were in Chile, where pension funds were allowed to invest a portion of their portfolios in foreign assets; in Colombia, where limits on exporters’ retention of proceeds abroad were extended; and in Spain, where controls on outflows were eliminated. The latter was done to comply with EC provisions rather than as a response to the surge in inflows. Another way of promoting outflows, prepayment of public sector foreign debt, occurred on a small scale in Thailand. It does not, however, have a monetary effect beyond any corresponding reduction in the fiscal deficit.

Box 4.Elimination of Restrictions on Capital Outflows During Episodes1

Chile (1990)

  • 1991: In a number of steps (February, April, May, and October), commercial banks were permitted to increase external trade financing and use up to 25 percent of foreign exchange time deposits for foreign trade financing. Joint venture rules were simplified, and the waiting period for remitting capital invested in Chile under the debt conversion program was shortened. Procedures for enterprises to directly invest abroad were modified and made easier. (These types of transactions were already done through the legal informal market.)
  • March 1992: Pension funds allowed to hold a portion of their portfolio in foreign assets (government bonds, certificates of deposit, and bankers’ acceptances). Limit on these investments increased gradually to 10 percent of investment portfolio.
  • March 1992: Limit on net foreign exchange holdings of commercial banks was doubled. Share of export receipts exempt from surrender requirements increased. Allocations of foreign exchange for a variety of payments abroad (including travel) raised. Period for advance purchase of foreign exchange for debt service extended.

Colombia

  • January 1992: Export surrender requirement proceeds eased: all exporters allowed to retain part of export proceeds abroad. Previously, this was granted only to coffee growers and to state enterprises exporting oil and minerals.
  • January 1992: Residents allowed to hold foreign stocks and other foreign portfolio investments abroad up to $500,000. Higher amounts require approval of the National Planning Department.
  • February 1992: Minimum maturity on foreign loans reduced from five years with two years’ grace, to one year. Such loans permitted only to finance working capital or fixed investment. Limit on contractual interest rate (London interbank offered rate (LIBOR) + 2.5 percent) eliminated for the private sector.

Egypt (1991/92)

  • February 1991: Foreign exchange made freely transferable abroad for visible and invisible transactions, including capital transfers. Authorized banks and nonbank dealers free to sell foreign exchange to individuals and public and private sector entities.

Mexico (1989)

  • November 1991: Abolished foreign exchange surrender requirements and related exchange control measures permitting unification of controlled and free market exchange rates.

Spain (1987)

(Many of the changes were made to comply with EC provisions)

  • May 1987: Spanish investment abroad liberalized; real estate investments partially liberalized.
  • July 1989: Complete liberalization of forward operations in the foreign exchange market.
  • December 1989: Direct investments abroad in companies engaged in portfolio and real estate investment activities permitted. Ceilings on real estate investments abroad abolished.
  • March 1990: 20 percent nonremunerated deposit requirement applied to banks’ net foreign currency borrowing abolished.
  • December 1990: Most regulations limiting portfolio investments abroad abolished.
  • March 1991: 30 percent nonremunerated deposit requirement applied to all loans contracted by Spanish residents abolished.
  • April 1991: All remaining restrictions on purchases of marketable foreign monetary instruments by Spanish residents removed; residents allowed to open bank accounts in foreign currencies.
  • February 1992: All remaining controls on capital flows eliminated.

Thailand

  • April 1991: Foreign exchange earners allowed to open foreign exchange accounts with commercial banks in Thailand up to $500,000 for individuals and $2 million for corporations. Thai investors could freely transfer up to $5 million abroad for direct investment. Bank of Thailand approval requirement of repatriation of investment funds eliminated.
Source: IMF staff reports.1 First year of episode noted in parentheses after country name.

Trade Reform

For most of the countries, trade liberalization, and particularly tariff reform and reduction, was in train before the surge in inflows began. The initial motivation for trade reform—to improve economic efficiency—became all the more important when large inflows needed to be allocated throughout the economy. Moreover, the surge in inflows eased any perceived external constraint on the speed of trade liberalization and tariff reduction. There is a question, however, of whether an acceleration of trade reform helped contain increases in reserves. Initially, the main effects of lowering tariffs would be to increase import demand, weaken the current account, and slow reserve accumulation. Over time, however, efficiency gains, particularly if tariff reductions affected mainly intermediate inputs, would improve competitiveness and narrow the current account deficit.7

Most of the countries reduced tariffs during the period of large inflows, although only three explicitly accelerated or changed the process in response to the inflows. Colombia made the most significant adaptation of trade reform by accelerating a planned reduction in tariffs; Mexico modified its ongoing trade liberalization program in a more piecemeal fashion; and Thailand implemented selective tariff cuts and liberalization measures.

    Other Resources Citing This Publication