Intro to Capital Controls

8/6/2019 Intro to Capital Controls

http://slidepdf.com/reader/full/intro-to-capital-controls 1/18

FED ERA L R ESERV E B A N K O F S T. LO U I S

13

N O V EM B ER / D ECEM BER 1 9 9 9

1 “ Article VI. Section 3. Controls

of capital transfers: Members

may exercise such controls as

are necessary to regulate inter-national capital movements,

but no member may exercise

these controls in a manner

which will restrict payments for

current transactions or which

will unduly delay transfers of

funds in settlement of commit-

ments, except as provided in

Article VII, Section 3(b) and in

Article XIV, Section 2.”

An In t roduc t iont o Capi t a lControls

Christ opher J. Neely

Moreover, it may well be asked whether

we can take it for granted that a return

to freedom of exchanges is really a ques-

tion of time. Even if the reply were in the

affirmative, it is safe to assume that after a period of freedom the regime of control

will be restored as a result of the next

economic crisis.

—Paul Einzig, Exchange Control,MacMillan and Company, 1934.

Currency controls are a risky, stopgap

measure, but some gaps desperately

need to be stopped.

—Paul Krugman, “Free Advice:A Letter to Malaysia’s Prime Minister,”Fortune

, September 28, 1998.

Unlike many topics in international eco-nomics, capital controls—taxes orrestrictions on international transac-

tions in assets like stocks or bonds—havereceived cursory treatment in textbooks andscant attention from researchers. The con-sensus among economists has been that cap-ital controls—like tariffs on goods—areobviously detrimental to economic efficiencybecause they prevent productive resources

from being used where they are most need-ed. As a result, capital controls graduallyhad been phased out in developed countriesduring the 1970s and 1980s, and by the1990s there was substantial pressure on less-developed countries to remove their restric-tions, too ( New York Times, 1999). The topicalmost had been relegated to a curiosity.

Several recent developments, however,have rekindled interest in the use and

study of capital controls. First, the resump-tion of large capital flows—trade in assets—to developing countries during the late1980s and early 1990s created new prob-lems for policymakers. Second, a string of exchange rate/financial crises during the1990s—the European Monetary Systemcrises of 1992-93, the Mexican crisis of 1994 and the Asian financial crisis of 1997-98—focused attention on the assettransactions that precipitated them. Inparticular, Malaysia’s adoption of capitalcontrols on September 1, 1998, has

prompted increased media attention andhas renewed debate on the topic.

Modern capital controls were devel-oped by the belligerents in World War Ito maintain a tax base to finance wartimeexpenditures. Controls began to disappearafter the war, only to return during theGreat Depression of the 1930s. At thattime, their purpose was to permit coun-tries greater ability to reflate their econo-mies without the danger of capital flight.In fact, the International Monetary Fund

(IMF) Articles of Agreement (Article VI,section 3) signed at the Bretton-Woodsconference in 1944 explicitly permittedcapital controls.1 One of the architects of those articles, John Maynard Keynes, wasa strong proponent of capital controls andthe IMF often was seen as such during itsearly years. During the Bretton-Woods eraof fixed-exchange rates, many countrieslimited asset transactions to cope with bal-ance-of-payments difficulties. But, recog-nition of the costs and distortions createdby these restrictions led to their gradual

removal in developed countries over thelast 30 years. The United States, for exam-ple, removed its most prominent capitalcontrols in 1974 (Congressional QuarterlyService, 1977) . During the last 10 yearseven less-developed countries began toliberalize trade in assets.

The purpose of this article is to intro-duce Review readers to the debate on capi-tal controls, to explain the purposes

Christopher J. Neely i s a senior economist at the Federal Reserve Bank of St. Louis. Kent Koch provided r esearch assistance.



N O V EM BE R / D ECEM BER 1 9 9 9

F ED ERA L R ESERV E B A N K O F S T. LO U I S

14

2 The U.S. Department of

Commerce does not recognize

“ real” assets as a separate

class. The purchase of assets

such as foreign production facili-

ties is recorded under financial

assets in their accounts(Department of Commerce,

1990).

3 The capital account records

both loans and asset purchases

because both involve buying a

claim on future income. A

bank making a car loan obtains

a legal claim on the borrower’s

future income.

4 Equity investment is considered

portfolio investment in national

accounts until it exceeds 10

percent of the market capital-

ization of the firm, then it is

considered direct investment.

5 The current account records

trade in goods, services, and

unilateral transfers. A nation’s

capital account balance must

be equal to and opposite in

sign from its current account

balance because a nation that

imports more goods and ser-

vices than it exports must pay

for those extra imports by sell-

ing assets or borrowing money.The sum of the current account

balance and the capital account

balance is the balance of pay-

ments.

6 The composition as well as the

magnitude of capital flows also

may influence the sustainability

of policies, as will be discussed

in section 3.

and costs of controls and why some advo-cate their reintroduction. To lay the ground-work for understanding restrictions oncapital flows, the next section of the article

describes capital flows and their benefits.The third section characterizes the mostcommon objectives of capital controls withan emphasis on the recent debate aboutusing controls to foster macroeconomicstability. Then the many types of capitalcontrols are distinguished from each otherand their effectiveness and costs are con-sidered. In addition, accompanying shad-ed inserts outline specific case studies incapital controls: the U.S. Interest Equali-zation Tax of 1963, the Chilean encaje of the 1990s, and the restrictions imposed

by Malaysia in September 1998.

CAPITAL FLOWS

To understand what capital controls do,it is useful to examine capital flows—tradein real and financial assets. Internationalpurchases and sales of existing real andfinancial assets are recorded in the capital

account of the balance of payments.2 Realassets include production facilities and realestate while financial assets include stocks,bonds, loans, and claims to bank deposits.3

Capital account transactions often are classi-fied into portfolio investment and directinvestment. Portfolio investment encom-passes trade in securities like stocks, bonds,bank loans, derivatives, and various formsof credit (commercial, financial, guaran-tees). Direct investment involves the pur-chase of real estate, production facilities,or substantial equity investment.4 When aGerman corporation, BMW, for example,builds an automobile factory in SouthCarolina, that is direct investment. On the

other hand, when U.S. investors buyMexican government bonds, that is portfo-lio investment.

A country is said to have a deficit inthe capital account if it is accumulating netclaims on the rest of the world by purchas-ing more assets and/or making more loansto the rest of the world than it is receiving.A country, like Japan, with a capital accountdeficit is also said to experience a capital

outflow. Accumulating claims on the restof the world is a form of national saving.Conversely, a country is said to have a sur-

plus in the capital account—or a capital

inflow—if the rest of the world is accumu-lating net claims on it, as is the case withthe United States.5 Just as individualsmust avoid borrowing excessively, policy-makers must make sure that the rest of theworld does not accumulate too many netclaims on their countries—in other words,that their countries do not sell assets/bor-row at an unsustainable rate.6

Benefits of Capital Flows

Economists have long argued that

trade in assets (capital flows) providessubstantial economic benefits by enablingresidents of different countries to capitalizeon their differences. Fundamentally, capi-tal flows permit nations to trade consump-tion today for consumption in the future—to engage in intertemporal trade (Eichen-green, et al. 1999) . Because Japan has apopulation that is aging more rapidly thanthat of the United States, it makes sensefor Japanese residents to purchase moreU.S. assets than they sell to us. This allowsthe Japanese to save for their retirement bybuilding up claims on future income in theUnited States while permitting residents of the United States to borrow at lower inter-est rates than they could otherwise pay.

A closely related concept is that capitalflows permit countries to avoid large fallsin national consumption from economicdownturn or natural disaster by sellingassets to and/or borrowing from the rest of the world. For example, after an earth-quake devastated southern Italy on Novem-ber 23, 1980, leaving 4,800 people dead,

Italians borrowed from abroad (ran a capi-tal account surplus) to help repair thedamage. Figure 1 illustrates the timeseries of the Italian capital account from1975 through 1985.

A third benefit is that capital flowspermit countries as a whole to borrow inorder to improve their ability to producegoods and services in the future—likeindividuals borrowing to finance an educa-





15

7 Alesina, Grilli, and Milesi-

Ferretti (1994) and Grilli and

Milesi-Ferretti (1995) empiri-

cally examine factors associat-

ed with capital controls.

tion. To cite just one example, between1960 and 1980 Koreans borrowed funds fromthe rest of the world equal to about 4.3percent of gross domestic product (GDP)

annually to finance investment during Korea’speriod of very strong growth (see Figure 2).

These arguments for free capital mobili-ty are similar to those that are used to sup-port free trade. Countries with different agestructures, saving rates, opportunities forinvestment, or risk profiles can benefit fromtrade in assets. More recently, economistshave emphasized other benefits of capitalflows such as the technology transfer thatoften accompanies foreign investment, orthe greater competition in domestic mar-kets that results from permitting foreign

firms to invest locally (Eichengreen, et al.1999). The benefits of capital flows do notcome without a price, however. Becausecapital flows can complicate economic poli-cy or even be a source of instability them-selves, governments have used capitalcontrols to limit their effects (Johnstonand Tamirisa, 1998).

PURPOSES OF

CAPITAL CONTROLSA capital control is any policy

designed to limit or redirect capitalaccount transactions. This broad defini-tion suggests that it will be difficult to gen-eralize about capital controls because theycan take many forms and may be appliedfor various purposes (Bakker, 1996).Controls may take the form of taxes, priceor quantity controls, or outright prohibi-tions on international trade in assets.7

Revenue Generation and

Credit Allocation The first widespread capital controlswere adopted in WWI as a method tofinance the war effort. At the start of thewar, all the major powers suspended theirparticipation in the gold standard for theduration of the conflict but maintainedfixed-exchange rates. All the belligerentsrestricted capital outflows, the purchase of foreign assets or loans abroad. These

restrictions raised revenues in two ways.First, by keeping capital in the domesticeconomy, it facilitated the taxation of wealth and interest income (Bakker, 1996).Second, it permitted a higher inflation rate,

which generated more revenue. Capitalcontrols also reduced interest rates andtherefore the government’s borrowing costson its own debt (Johnston and Tamirisa,1998). Since WWI, controls on capitaloutflows have been used similarly in other—mostly developing—economies to gen-erate revenue for governments or to permitthem to allocate credit domestically with-out risking capital flight (Johnston and

Figure 1

2

1.5

1

0.5

0

–0.5

–1

–1.51975 1977 1979

SOURCE: Inter national Financial Stati stics

1981 1983 1985

I ta l ian Capi ta l Account Surplusas a Percent age of GDP

Percent

November 23, 1980Earthquake

Figure 2

20

15

10

5

0

– 5

–101960 1964 1968 1972 1976 1980 1984 1988 1992 1996

SOURCE: Inter national Financial Statistics and Mitchell ( 1998)

South K orean Growt h and Capi ta l Surplus

Percent

Real GDP Growth

Capita l Account Balanceas a percentage of GDP





16

Purposes of Capi t al Controls

Table 1

Purpose ofControl

Generate Revenue/Finance War Eff ort

Financial Repression/Credit Allocation

Correct a Balance ofPayments Deficit

Correct a Balance ofPayments Surplus

Prevent Potentiall yVolatile Inflows

Prevent FinancialDestabilization

Prevent RealAppreciation

Restrict ForeignOwnership ofDomestic Assets

Preserve Savingsfor Domestic Use

Protect DomesticFinancial Firms

Method

Controls on capital outflows permit a country to runhigher inflation with a given fixed-exchange rate andalso hold down domestic interest rates.

Governments that use the financial system to rewardfavored industries or to raise revenue, may use capitalcontrols to prevent capital f rom going abroad to seekhigher returns.

Controls on outflows reduce demand for foreign assetswithout contractionary monetary policy or devalua-tion. This allows a higher rate of inflation than other-wise would be possible.

Controls on inflows reduce foreign demand for domes-tic assets without expansionary monetary policy orrevaluation. This allows a lower rate of inflation thanwould otherwise be possible.

Restricting inflows enhances macroeconomic stabilityby reducing the pool of capital that can leave a coun-try during a crisis.

Capital controls can restrict or change the compositionof international capital flows that can exacerbate dis-torted incentives in the domestic financial system.

Restricting inflows prevents the necessity of monetaryexpansion and greater domestic infl ation that wouldcause a real appreciation of the currency.

Foreign ownership of certain domestic assets— espe-cially natural r esources— can generate resentment.

The benefits of investing in the domestic economymay not fully accrue to savers so the economy, asa whole, can be made better off by restricting theoutflow of capital.

Controls that temporarily segregate domestic financialsectors from the rest of the world may permit domesticfi rms to attai n economies of scale to compete in worldmarkets.

Directionof Control

Outflows

Outflows

Outflows

Inflows

Inflows

Inflows

Inflows

Inflows

Outflows

Infl ows andOutflows

Example

Most belligerentsduring WWIand WWII

Common indevelopingcountries

U.S. i nterestequalization tax,

1963-74

German Bardepotscheme, 1972-74

Chilean encaje ,

1991-98

Chilean encaje ,

1991-98

Chilean encaje ,

1991-98

Article 27 ofthe Mexicanconstitution





17

Tamirisa). Table 1 summarizes the pur-poses of capital controls.

Balance of Payments Crises During the Great Depression, controls

simultaneously were used to achieve greaterfreedom for monetary policy and exchangerate stability—goals that have remainedpopular. To understand why controls havebeen used in this way, it is necessary tounderstand balance of payments problemsand their solutions (Johnston and Tamirisa,1998). At a given exchange rate, a countryoften will want to collectively purchase moregoods, services and assets than the rest of the world will buy from it. Such an imbal-

ance is called a balance of payments deficit and may come about for any one of a numberof reasons: 1) The domestic business cyclemay be out of sync with that of the rest of the world; 2) There may have been a rapidchange in the world price of key com-modities like oil; 3) Expansionary domes-tic policy may have increased demand forthe rest of the world’s goods; 4) Large for-eign debt interest obligations may surpassthe value of the domestic economy’s exports;5) Or, a perception of deteriorating eco-nomic policy may have reduced interna-tional demand for domestic assets.8 In theabsence of some combination of exchangerate and monetary policy by the deficitcountry, excess demand for foreign goodsand assets would bid up their prices—typi-cally through a fall in the foreign exchangevalue (a devaluation or depreciation) of the domestic currency—until the deficitwas eliminated.9

There are four policy alternatives tocorrect an imbalance in international pay-ments: 1) Permit the exchange rate to

change, as described above; 2) Use mone-tary policy—unsterilized foreign exchangeintervention—to correct the imbalancethrough domestic demand; 3) Attempt tosterilize the monetary changes to isolatethe domestic economy from the capitalflows; and 4) Restrict capital flows.10

Each alternative has disadvantages.When domestic residents purchase

more goods and assets from foreigners

than foreigners purchase from domesticresidents, the exchange rate (the price of foreign currency) tends to rise. If theexchange rate is flexible, the foreign cur-

rency tends to appreciate and the domesticcurrency tends to depreciate. The depreci-ation of the domestic currency raises pricesof imported goods and assets to domesticresidents and lowers the prices of domesticgoods and assets on world markets, reduc-ing the relative demand for foreign goodsand assets until the imbalance in the bal-ance of payments is eliminated. A countrywith a fixed exchange rate similarly maycorrect a balance of payments deficit bychanging the exchange rate peg—devalu-ing the currency—but this option foregoes

the benefits of exchange rate stability forinternational trade and policy discipline.In addition, it may reduce the public’s con-fidence in the monetary authorities’ anti-inflation program.

If a government is committed to main-taining a particular fixed exchange rate, onthe other hand, its central bank can pre-vent the depreciation of its currency withcontractionary monetary policy—by sell-ing domestic bonds.11 Alternatively, thecentral bank might sell foreign exchangeto affect the monetary base, in which casethe action is known as unsterilized foreign

exchange intervention. In either case, sucha sale lowers the domestic money supplyand raises domestic interest rates—lower-ing domestic demand for imports—whilereducing the prices of domestic goods, ser-vices, and assets relative to their foreigncounterparts. The reduced demand andhigher prices for foreign goods, services,and assets would eliminate a balance of payments deficit. However, this defense of the exchange rate requires that monetary

policy be devoted solely to maintaining theexchange rate; it cannot be used to achieveindependent domestic inflation or employ-ment goals. In this case, for example, thecontraction temporarily will reducedomestic demand and employment, whichmay be undesirable. A country that usesmonetary policy to defend the exchangerate in the face of imbalances in interna-tional payments is said to subordinate

8 Often, the term “ balance of

payments deficit” describes an

imbalance in the current

account (goods, services, fac-

tor payments, and unilateral

transfers). Here, it describes

the sum of the current account

and the capital account.

Countries also may demand

fewer goods and assets from

the rest of the world— balance

of payments surpluses— butthis article concentrates on bal-

ance of payments deficits

because most countries find

balance of payments surpluses

easier to manage.

9 When a flexible exchange rate

currency gains or loses value, it

is said to appreciate or depreci-

ate , respectively. Fixed-rate

currencies are said to be reval-

ued or devalued when their

price rises or falls.

10 There are other policies—fiscal and regulatory— that

may be used to manage the

effects of capital flows but they

will be ignored to simplify the

discussion.

11 Foreign exchange operations

that do not affect the domestic

monetary base are called “ ster-

ilized,” while those that do

affect the monetary base are

called “ unsterilized.” Sales of

any asset would tend to lower

the domestic money supply andraise interest rates because

when the monetary authority

receives payment for the asset,

the payment ceases to be part

of the money supply. Fiscal

policy also may have an effect

on exchange rates but taxing

and spending decisions usually

are more constrained than

monetary decisions.



domestic monetary policy to exchangerate concerns.

Rather than subordinate monetary pol-icy to maintaining the exchange rate, some

central banks have attempted to recapturesome monetary independence by steriliz-

ing—or reversing—the effect of foreignexchange operations on the domesticmoney supply. Sterilization of sales of for-eign exchange (foreign bonds), for exam-ple, would require the central bank to buyan equal amount of domestic bonds, leav-ing domestic interest rates unchanged afterthe inflow. It generally is believed thatsterilized intervention does not affect theexchange rate and so it is not very effectivein recapturing monetary independence

(Edwards, 1998b).If international investors don’t believe

that the monetary authorities will defendthe exchange rate with tighter monetarypolicy, they will expect devaluation—a fallin the relative price of domestic goods andassets—and will sell domestic assets toavoid a loss. Such a sale increases relativedemand for foreign assets, exacerbatingthe balance of payments deficit, and speedsthe devaluation.12

Capital flows play a crucial role in bal-ance of payments crises in two ways.Swings in international capital flows cancreate both a balance-of-payments problemand—if the exchange rate is not defend-ed—expedite devaluation under fixed-exchange rates. Thus, in the presence of free capital flows, a country wishing tomaintain a fixed-exchange rate must usemonetary policy solely for that purpose.As McKinnon and Oates (1966) argued, nogovernment can maintain fixed-exchangerates, free capital mobility, and have anindependent monetary policy; one of the

three options must give. This is known asthe “incompatible trinity” or the trilemma(Obstfeld and Taylor, 1998) . Policymakerswishing to avoid exchange-rate fluctuationand retain scope for independent monetarypolicy must choose the fourth option,restrict capital flows.

By directly reducing demand for for-eign assets and the potential for specula-tion against the fixed-exchange rate,

controls on capital outflows allow a coun-try to maintain fixed-exchange rates andan independent domestic monetary policywhile alleviating a balance-of-payments

deficit.13 The monetary authorities canmeet both their internal goals (employ-ment and inflation) and their externalgoals (balance of payments).14 Thus, capi-tal controls are sometimes described interms of the choices they avoid: to prevent

capital outflows that, through their effecton the balance of payments, might eitherendanger fixed-exchange rates or indepen-

dence of monetary policy.

Real Appreciation of the

Exchange Rate While capital outflows can create bal-

ance-of-payments deficits, capital inflowscan cause real appreciation of the exchangerate.15 During the 1980s and 1990s anumber of developing countries completedimportant policy reforms that made themmuch more attractive investment environ-ments. Eichengreen, et al. (1999) reportthat net capital flows to developing coun-tries tripled from $50 billion in 1987-89 tomore than $150 billion in 1995-97. Theselarge capital inflows to the reformingcountries tended to drive up the prices of domestic assets. For countr ies with flexi-ble exchange rates, the exchange ratesappreciated, raising the relative prices of the domestic countries’ goods. For coun-tries with fixed-exchange rates, theincreased demand for domestic assets ledthe monetary authorities to buy foreignexchange (sell domestic currency), increas-ing the domestic money supply and ulti-mately the prices of domestic goods andassets. In either case, the prices of domes-

tic goods and assets rose relative to thosein the rest of the world—a real apprecia-tion—making domestic exported goodsless competitive on world markets andhurting exporting and import-competingindustries.16 Because of these effects, theproblem of real exchange-rate appreciationfrom capital inflows is described variouslyas real exchange-rate instability, real appre-ciation, or loss of competitiveness.

12 Of course, capital outflows are

not necessary to force devalua-tion. If the domestic economy

still demands more goods and

services than it supplies to the

rest of the world, the exchange

rate cannot be maintained with-

out a monetary contraction.

13 Countries that face balance of

payments surpluses — the

desire to purchase fewer goods

and services from the rest of

the world at the fixed-exchange

rate— would restrict capital

inflows, rather than outflows,to reduce demand for their own

assets.

14 If there is still excess demand

for foreign goods, the fixed-

exchange rate will still be only

temporarily sustainable.

15 A nominal appreciation is a rise

in the foreign exchange value

of a country’s currency. A real

appreciation is a rise in the rela-

tive price of domestic goods

and services compared to for-

eign goods and services. Thismay result from a nominal

appreciation, domestic inflation

that is higher than foreign infla-

tion, or some combination of

the two.

16 Empirically, Edwards (1998b)

finds a consistent, but limited,

tendency toward real apprecia-

tion from capital inflows.



18



Countries have a number of policyoptions to prevent real appreciation in theface of capital inflows (Goldstein, 1995;Corbo and Hernandez, 1996). Permitting

the exchange rate to change still results innominal and real appreciation but avoidsdomestic inflation. A very common tacticfor fixed exchange-rate regimes is to steril-ize the monetary effects of the inflows, pre-venting an expansion of the money supplyby reversing the effect on the domesticmoney market (Edwards, 1998b). It gener-ally is believed that sterilization is not veryeffective in recapturing monetary indepen-dence as it keeps domestic real interestrates high and leads to continued inflows.Sterilization of inflows also is a potentially

expensive strategy for the government asthe domestic bonds that the central bank sells may pay higher interest than the for-eign bonds the central bank buys. Fiscalcontraction is an effective way to preventreal appreciation because it lowers domes-tic interest rates, and likewise, the demandfor domestic assets; but raising taxes and/orreducing government spending may bepolitically unpalatable. Because of theproblems associated with these first threepolicies, countries like Brazil, Chile, andColumbia chose to use capital controls—restricting purchase of domestic assets(inflows)—to try to prevent real apprecia-tion and substitute for fiscal policy flexibil-ity in the face of heavy inflows.

Theories of the Second Best

More recently, economists have con-sidered other circumstances—other thanbalance-of-payments needs or real appreci-ation—under which capital controls mightbe a useful policy. As a rule, economists

emphasize that restrictions on trade andinvestment impose costs on the economy.There are exceptions to that rule, however.Taxes and quantitative restrictions may begood for the economy—welfare improving,in technical jargon—if they are used tocorrect some other, pre-existing distortionto free markets that cannot be correctedotherwise. The idea that a tax or quantita-tive restriction can improve economic wel-

fare in this way is called a “theory of thesecond best.”17

Capital controls preserve domesticsavings for domestic use. From a national

point of view, there might be benefits froma greater rate of domestic investment thatdo not fully accrue to the investors. Forexample, domestic savers might invest dis-proportionately overseas because of politi-cal risk of expropriation or a desire toescape taxation. In either case, the nationas a whole could be made better off by lim-iting or taxing domestic investment abroad(Harberger, 1986).

The infant industry argument—an oldidea often used to justify tariffs in goodsmarkets —has been resurrected to ratio-

nalize the use of capital controls on bothinflows and outflows. This idea starts withthe premise that small, domestic firms areless efficient than larger, foreign firms andso will be unable to compete on an equalbasis. To permit small domestic firms togrow to the efficient scale that they need toenjoy to compete in world markets, theymust be protected temporarily from inter-national competition by trade barriers. Asapplied to capital markets, the argumenturges that capital controls be used to protectunderdeveloped financial markets fromforeign competition. The problem withthis argument, as in goods markets, is thatprotected industries often never grow upand end up seeking perpetual protection.

Financial Sector Distortions

In reality, capital controls rarely havebeen imposed in a well-thought-out way tocorrect clearly defined pre-existing distor-tions. Instead, capital controls most oftenhave been used as a tool to postpone diffi-

cult decisions on monetary and fiscal poli-cies. Recently, however, the case has beenmade that capital controls may be the leastdisadvantageous solution to the destabiliz-ing effects of capital flows on inadequatelyregulated financial systems.

Recall that when a country with afixed-exchange rate has a net capital out-flow, the increase in relative demand forforeign assets means that there is insuffi-

17 The classic example of a tax

that improves welfare is the

one imposed on a polluting

industry. Because a polluting

industry imposes non-market

costs on others, for which it

does not compensate them, a

government could improve

everyone’s well being if it were

to tax pollution. The factory

would produce less pollutionand people would be happier.

The desire to meet employmen

goals with an independent

monetary policy is really a ver-

sion of a “second best” story in

which the pre-existing distortion

is price inertia (or a similar fric-

tion) in the real economy that

causes monetary policy to have

real effects (Dooley, 1996).



19



cient demand for domestic goods andassets at the fixed-exchange rate. Thedomestic monetary authorities may con-duct contractionary monetary policy—

raise domestic interest rates to make theirassets more attractive—or lower the pricesof their goods and assets (devalue the cur-rency). This is a special case of a balance-of-payments deficit and presents the samechoice—to raise interest rates or devalue—but in the case of a sudden capital outflow,the crisis manifests itself in large suddencapital outflows rather than more gradualbalance-of-payments pressures from othercauses. Governments must choosebetween high interest rates coupled withsome capital outflows or an exchange rate

devaluation that provokes fear of inflationand policy instability leading to greatercapital outflows. In either case, a seriousrecession seems unavoidable.

The recent case for capital controlsrecognizes that a monetary contraction notonly slows economic activity through thenormal interest-rate channels, but also canthreaten the health of the economy throughthe banking system (Kaminsky and Reinhart,1999). If the monetary authorities raiseinterest rates, they increase the costs of fundsfor banks and—by slowing economicgrowth—reduce the demand for loans andincrease the number of nonperforming loans.Choosing to devalue the currency rather thanraising interest rates does not necessarilyhelp banks either, as they may have bor-rowed in foreign currency. A devaluationwould increase the banks’ obligations totheir foreign creditors. Thus, capital out-flows from the banking system pose spe-cial problems for the monetary authorities,as banks’ liabilities are usually implicitly orexplicitly guaranteed by the government.

Indeed, the very nature of the financialsystem creates perverse incentives (distor-tions) that international capital flows oftenexacerbate (Mishkin, 1998). For example,in a purely domestic context, banks haveincentives to make risky loans, as theirlosses are limited to the owners’ equitycapital, but their potential profits areunlimited. The existence of deposit insur-ance worsens this problem by reducing

depositors’ incentive to monitor their banks’loan portfolio for excessive risk. Depositinsurance, in turn, exists precisely becausedepositors can not easily monitor the riskiness

of their banks. In the absence of depositinsurance, depositors would find it diffi-cult to tell good banks from bad banks andwould withdraw their money at any sign of danger to the bank. Once some depositorsbegan to withdraw their money from thebank, all depositors would try to do so,forcing the bank to close, even if its under-lying assets were productive (Diamond andDybvig, 1983). This puts the whole bank-ing system at risk.

To avoid this problem, most developedcountries combine implicit or explicit insur-

ance of bank deposits with government reg-ulation of depository institutions, especiallytheir asset portfolios (loans). In emergingmarkets, however, banking regulation ismuch more difficult as the examiners areless experienced, have fewer resources andless strict accounting standards by which tooperate. Thus, banking problems are moreserious in emerging markets.

Large international capital inflows,especially short-term foreign borrowing,can exacerbate these perverse incentives

and pose a real danger to banking systems.Domestic banks often view borrowing fromabroad in foreign currency sources as alow-cost source of funds—as long as thedomestic currency is not devalued. Withthis additional funding, banks expand intounfamiliar areas, generating risky loansthat potentially create systemic risk to thebanking system (Eichengreen, 1999;Garber, 1998; Goldstein, 1995; Dorn-busch, 1998). If capital outflows force adevaluation, the foreign-currency denomi-nated debts of the banking system increase

when measured in the domestic currency,possibly leading to bank failures. Thebanking system is a particularly vulnerableconduit by which capital flows can desta-bilize an economy because widespreadbank failures impose large costs on taxpay-ers and can disrupt the payments systemand the relationships between banks andfirms who borrow from them (Friedmanand Schwartz, 1963; Bernanke, 1983). The



20



difficulty of effective banking regulation cre-ates an argument for capital controls as asecond-best solution to the existence of thedistorted incentives in the banking system.18

There are two ways in which capitalcontrols might be imposed to limit capitalflow fluctuations and achieve economicstability. First, capital controls may beused to discourage capital outflows in theevent of a crisis—as Malaysia did inSeptember 1998—permitting looserdomestic monetary policy. Controls onoutflows are ideally taken as a transitionalmeasure to buy time to achieve goals, as anaid to reform rather than as a substitute(Krugman, 1998). Second, controls canprevent destabilizing outflows by discourag-

ing or changing the composition of capitalinflows, as Chile did for most of the 1990s.

The second method—to discourage orchange the composition of capital inflowswith controls—requires some explanation.A prime fear of those who seek to limitcapital flows is that sudden outflows mayendanger economic stability becauseinvestors are subject to panics, fads, andbubbles (Kindleberger, 1978; Krugman,1998; Wade and Veneroso, 1998).Investors may panic because they, as indi-viduals, have limited information aboutthe true value of the assets they are buyingor selling. They can, however, infer infor-mation from the actions of others. Forexample, one might assume that a crowd-ed restaurant serves good food, even if onehas never eaten there. In financial mar-kets, participants learn about other partici-pants’ information by watching pricemovements. An increase in the price of anasset might be interpreted as new informa-tion that the asset had been underpriced,for example. Such a process might lead to

“herding” behavior, in which asset pricechanges tend to cause further changes inthe same direction, creating a boom-bustcycle and instability in financial markets,potentially just ifying capital controls. Bydiscouraging inflows of foreign capital,governments can limit the pool of volatilecapital that may leave on short notice.

Instead of limiting the total quantity of capital inflows, some would argue that

changing the composition of that inflow is just as important . For example, it often isclaimed that direct investment is likely tobe more stable than portfolio investment

because stocks or bonds can be sold moreeasily than real assets (like productionfacilities) can be liquidated (Dixit andPindyck, 1994; Frankel and Rose, 1996;Dornbusch, 1998). In contrast, Garber(1998) argues that tracking portfolio anddirect investment data may be misleading;derivatives (options, futures, swaps, etc.)can disguise the source of a crisis, makingit look like the source is excessive short-term debt. Goldstein (1995) says there islittle evidence that direct investment is less“reversible” than portfolio investment. For

example, foreign firms with domestic pro-duction facilities abroad can use th osefacilities as collateral for bank loans thatthen can be converted to assets in anothercurrency, effectively moving the capitalback out of the country.

TYPES OF CAPITAL

CON TROLS

To meet the many possible objectivesdescribed for them, there are many typesof capital controls, distinguished by thetype of asset transaction they affect andwhether they tax the transaction, limit it,or prohibit it outright. This section distin-guishes the many types of capital controlsby this taxonomy.

Capital controls are not, strictly speak-ing, the same as exchange controls, therestriction of trade in currencies, although thetwo are closely related (Bakker, 1996).Although currency and bank deposits are onetype of asset—money—exchange controlsmay be used to control the current account

rather than the capital account. For example,by requiring importers to buy foreignexchange from the government for a statedpurpose, exchange controls may be used toprohibit the legal importation of “luxury”goods, thereby rationing “scarce” foreignexchange for more politically desirablepurposes. So, while exchange controls areinherently a type of limited capital control,they are neither necessary to restrict capital

18 The capital adequacy standards

of the Basle accords penalize

long-term international inter-

bank lending relative to short-

term lending, exacerbating the

problem (Corsetti, Pesenti, andRoubini, 1998b). Although

banks are important and heavi-

ly regulated almost every-

where, banks play an especially

important role in developing

countries because information

problems tend to be more

important in the developing

world than they are in the

developed world.



21



movement nor are they necessarily intendedto control capital account transactions.

Controls on Inflows vs. Outflows

Capital controls on some long-term(more than a year) inflows—direct invest-

ment and equity—often are imposed fordifferent reasons than those on short-terminflows—bank deposits and money marketinstruments. While the recent trend hasbeen to limit short-term capital flowsbecause of their allegedly greater volatilityand potential to destabilize the economy,



22

MALAYSIA’S CAPITAL CONTROLS: 19 98 -99

The devaluation of the Thai baht in

July 1997 sparked significant capitaloutflows from Southeast Asia, leadingto a fall in local equity prices andplunging exchange rates. To counterthese outflows of capital, the IMF urgedmany of the nations of the region toraise interest rates, making their securi-ties more attractive to internationalinvestors. Unfortunately, the higherinterest rates also slowed the domesticeconomies.1

In response to this dilemma,Malaysia imposed capital controls on

September 1, 1998. The controlsbanned transfers between domestic andforeign accounts and between foreignaccounts, eliminated credit facilities tooffshore parties, prevented repatriationof investment until September 1, 1999,and fixed the exchange rate at M3.8 perdollar. Foreign exchange transactionswere permitted only at authorized insti-tutions and required documentation toshow they were for current account pur-poses. The government enacted a fairlyintrusive set of financial regulationsdesigned to prevent evasion. InFebruary 1999, a system of taxes on out-flows replaced the prohibition on repa-triation of capital. While the details arecomplex, the net effect was to discour-age short-term capital flows but to freelypermit longer-term transactions(Blustein, 1998). By imposing the capi-tal controls, Malaysia hoped to gainsome monetary independence, to be ableto lower interest rates without provok-ing a plunge in the value of the currency

as investors fled Malaysian assets.

The Malaysian government and

business community claimed to bepleased with the effect of the controlsin increasing demand and returning sta-bility to the economy. Even economistswho oppose capital controls believethat they may have been of some use inbuying time to implement fundamentalreforms (Barro, 1998) .

Others fear, however, that the capi-tal controls have replaced reform,rather than buying time for reform. Asof May 1999, the Malaysian govern-ment does not appear to be using the

breathing space purchased by the capi-tal controls to make fundamentaladjustments to its fragile and highlyleveraged financial sector. Rather,Prime Minister Mahathir has sackedpolicymakers who advocate reformwhile aggressively lowering interestrates, loosening nonperforming loanclassification regulation and settingminimum lending targets for banks.This strategy may prove short-sighted,as much of the capital outflow wascaused by the recognition that assetprices were overvalued and the bankingsector was weak (Global Investor ,1998b). Although monetary stimulusmay be helpful in the short run, it mayexacerbate the underlying problems. Inaddition, the government must be con-cerned about the long-term impact thatthe controls will have on investors’ will-ingness to invest in the country.

1 For an overview of the causes and policy options in the Asian

financial crisis, see Corsetti, Pesenti, and Roubini (1998a and

1998b).



bans on long-term capital flows oftenreflect political sensitivity to foreign own-ership of domestic assets. For example,Article 27 of the Mexican constitution lim-

its foreign investment in Mexican realestate and natural resources.

Controls on capital inflows and outflowsprovide some slack for monetary policy discre-tion under fixed exchange rates, but in oppo-site directions. Controls on capital inflows,which allow for higher interest rates, havebeen used to try to prevent an expansion of the money supply and the accompanyinginflation, as were those of Germany in 1972-74(Marston, 1995) or Chile during the 1990s.19

In contrast, controls on capital outflows permitlower interest rates and higher money growth

than otherwise would be possible (Marston,1995). They most often have been used topostpone a choice between devaluation ortighter monetary policy, as they have been inMalaysia, for example (see the shaded insert).

Price vs. Quantity Controls

Capital controls also may be distin-guished by whether they limit asset trans-actions through price mechanisms (taxes)or by quantity controls (quotas or outrightprohibitions). Price controls may take theform of special taxes on returns to interna-tional investment ( like the U.S. interestequalization tax of the 1960s—see theshaded insert) , taxes on certain types of transactions, or a mandatory reserverequirement that functions as a tax.

One type of price mechanism to dis-courage short-term capital flows is the“Tobin” tax. Proposed by Nobel laureateJames Tobin in 1972, the Tobin tax wouldcharge participants a small percentage of all foreign exchange transactions (ul Haq,

Kaul and Grunberg, 1996; Kasa, 1999).Advocates of such a tax hope that it woulddiminish foreign exchange market volatilityby curtailing the incentive to switch posi-tions over short horizons in the foreignexchange market. There are many prob-lems with a Tobin tax, however. The taxmight reduce liquidity in foreign exchangemarkets or be evaded easily through deriv-ative instruments. It is uncertain who

would collect the tax or for what purposesthe revenue would be used. And, mostdauntingly, a Tobin tax would have to beenacted by widespread international agree-

ment to be successful.A mandatory reserve requirement is a

price-based capital control that commonlyhas been implemented to reduce capitalinflows. Such a requirement typically oblig-ates foreign parties who wish to depositmoney in a domestic bank account—or useanother form of inflow—to deposit somepercentage of the inflow with the centralbank for a minimum period. For example,from 1991 to 1998, Chile required foreigninvestors to leave a fraction of short -termbank deposits with the central bank, earn-

ing no interest.20 As the deposits earn nointerest and allow the central bank to buyforeign money market instruments, thereserve requirement effectively functionsas a tax on short-term capital inflows(Edwards, 1998b). See the shaded inserton the Chilean encaje of the 1990s.

Quantity restrictions on capital flowsmay include rules mandating ceilings orrequiring special authorization for new orexisting borrowing from foreign residents.There may be administrative controls oncross-border capital movements in which agovernment agency must approve transac-tions for certain types of assets. Certain typesof investment might be restricted altogetheras in Korea, where the government has, untilrecently, restricted long-term foreign invest-ment (Eichengreen, et al. 1999). Forbiddingor requiring special permission for repatria-tion of profits by foreign enterprises operat-ing domestically may restrict capital outflows.Capital controls may be more subtle: Domes-tic regulations on the portfolio choice of insti-tutional investors also may be used as a type

of capital control, as they have been in Italyand in South Korea in the past (Bakker, 1996;Park and Song, 1996).

EVALU ATING CAPITAL

CONTROLSThe conventional wisdom of the eco-

nomics profession has been—whatever theproblems with destabilizing capital flows or

19 Recall that capital inflows entai

foreign purchases of domestic

assets or foreign loans to

domestic residents while out-

flows entail domestic purchases

of foreign assets or loans to for

eign residents by domestic resi-

dents. Under a fixed exchangerate, persistent capital inflows

will require an expansion of the

money supply or a revaluation

while substantial capital out-

flows will require a contraction.

20 The Chilean reserve require-

ment applied not only to bank

deposits but to many types of

capital inflows.



23





24

THE U.S. INT EREST EQUA LIZ ATION TAX: 19 63-74

During the late 1950s and early1960s, the United States had both afixed-exchange rate regime for the dol-lar (the Bretton-Woods system) andchronic pressures toward balance-of-payments deficits. These strains result-ed partly from the fact that interestrates in the rest of the world—especial-ly those in Europe—tended to be high-er than those in the United States,making foreign assets look attractive toU.S. residents.1 Faced with theunpalatable alternatives of devaluing

the dollar or conducting contractionarypolicies, on July 19, 1963, PresidentKennedy proposed the InterestEqualization Tax (IET) to raise theprices that Americans would have topay for foreign assets ( Economist ,1964a).2

The IET imposed a variable sur-charge, ranging from 1.05 percent onone-year bills to 15 percent on equityand bonds of greater than 28.5 yearsmaturity, on U.S. purchases of stocksand bonds from Western Europe, Japan,Australia, South Africa, and NewZealand (Congressional QuarterlyService, 1969). Canada and the devel-oping world were exempted from thetax out of consideration for their spe-cial dependence on U.S. capital mar-kets. By raising the prices of foreignassets, it was hoped that demand forthose assets—and the consequent bal-ance-of-payments deficit—would bereduced or eliminated.

The IET reduced direct outflows to

the targeted countries but didn’t changetotal outflows much because investorswere able to evade the tax through thirdcountries, like Canada (Pearce, 1995;Kreinen, 1971; Stern, 1973). In addi-tion, because the tax did not coverloans, investment was diverted initiallyfrom bond and stock purchases to bank loans. Loans from American banks to

firms in Europe and Japan jumped from$150 million during the first half of 1963 to $400 million during the secondhalf ( Economist , 1964b).

To check bank loans to foreigncountries, the U.S. Congress enactedthe Voluntary Foreign Credit RestraintProgram (VFCRP) in February 1965,broadening it in 1966 to limit U.S.short-term capital outflows to otherdeveloped countr ies. In addition, U.S.corporations were asked to voluntarilylimit their direct foreign investment.

The program was made mandatory in1968 (Laffer and Miles, 1982; Kreinin,1971) . U.S. capital controls wererelaxed in 1969 and phased out in1974, after the United States left theBretton-Woods system of fixed-exchangerates (Congressional Quarterly Service,1973a, 1973b, 1977).

One un intended consequence of the IET was the growth of foreignfinancial markets—at the expense of U.S. markets—as they inherited the jobof intermediating international capitalflows. For example, the volume of international borrowing in London rosefrom $350 million in 1962 to morethan $1 billion in 1963 while the vol-ume of foreign flotations in New York fell from $2 billion in the first half of 1963 to just over $600 million in thenext nine months ( Economist , 1964b).

1 At this time, the United States had a current account surplus that

failed to fully offset private demand for foreign assets— the capi-

tal account deficit— resulting in the need for temporary mea-

sures to close the gap.

2 In August 1964, the U.S. Congress enacted this tax, making it

retroactive to the date it was proposed.





25

CHILE’S ENCAJE : 1991-98

During the late 1980s and early 1990s,

international capital began to return to Chile as

a result of slow growth and low interest rates in

the developed world and sound macroeconomic

policies, including reduced debt, in Chile

(Edwards, 1998a). The Chilean authorities

feared that these capital inflows would compli-

cate monetary policy decisions—perhaps caus-

ing real appreciation of the exchange rate—and

they also were wary of the danger of building up

short-term debt.

Chile had long restricted capital flows and

these limits were updated in the early 1990s to

deal with the surge in capital inflows. Directinvestment was made subject to a 10-year stay

requirement in 1982; this period was reduced to

three years in 1991 and to one-year in 1993.

Portfolio flows were made subject to the

encaje—a one-year, mandatory, non-interest pay-

ing deposit with the central bank—created in

1991 to regulate capital inflows.1 The encajewas

initially 20 percent but was increased to 30 per-

cent in 1992. The penalty for early withdrawal

was 3 percent.2

The effect of the encajewas to tax foreign

capital inflows, with short-term flows being

taxed much more heavily than long-term flows.For example, consider the choice of an

American buying a one-year discount bond with

a face value of 10,000 pesos for a price of 9,091

pesos, or a 10-year discount bond with the same

face value and a price of 3,855 pesos. Either

bond, if held to maturity, would yield a 10 per-

cent per annum return.3 In the presence of a 30

percent one-year reserve requirement, however,

the one-year bond’s annual yield would be 7.7

percent and the 10-year bond’s annual yield

would be 9.7 percent. Hence, the encaje acted as

a graduated tax on capital inflows.

Researchers disagree about the effectivenessof Chile’s capital controls. Valdes and Soto

(1996) concluded that they changed the compo-

sition but not the magnitude of the inflows. In

other words, investors substituted from heavily

taxed short-term flows to more lightly taxed

long-term inflows. They also found that the

controls were ineffective in preventing a real

appreciation of the exchange rate. Larraín B.,

Labán M., and Chumacero (1997) studied the

same issue with different methods and found

that, although there was considerable substitu-

tion in the short run, the controls did change the

magnitude of the inflows in the long run.

There is even more disagreement about

whether the capital controls were important in

keeping Chile insulated from the Asian crisis.

Many observers have cited Chile’s capital con-

trols in advocating more widespread restrictions

on capital controls for other developing coun-

tries (Bhagwati, 1998). Edwards (1998a), on

the other hand, points out that Chile also hadsubstantial capital controls during the late 1970s

and early 1980s, before its major banking crisis

that cost Chileans more than 20 percent of GDP

during 1982-83. The major difference between

then and now is that Chile now has a modern

and efficient system of banking regulation.

Others credit the participation of foreign banks

in strengthening the Chilean banking system by

providing experience and sophistication in

assessing risks and making loans. At the time of

the crisis, Chile had a high percentage of domes-

tic loans from foreign-owned banks—20 per-

cent, about the same as the United States and farhigher than South Korea, Thailand, and

Indonesia (5 percent) ( Economist , 1997). In

addition, Edwards (1998a) claims that the encaje

harmed the domestic financial services industry

and the small firms that could not borrow long

term on international markets to avoid the tax.

If Chile’s capital controls helped, it was to buy

time for structural reforms and effective finan-

cial regulation.

1 The word encaje means “ strongbox” in Spanish.

2 The encaje was reduced to 10 percent and the early withdrawal

penalty from 3 percent to 1 percent in June 1998. The encaje waseliminated entirely on September 16, 1998 (Torres, 1998).

3 The yield to maturity on a bond equates the initial outlay with the

present discounted value of its payoff s. The yields on the bonds are

determined by solving the following equations for i :

9091 = 10,000/ (1+i), 3855 = 10,000/ (1+i)10,

1.3∗9091 = .3∗9091/ (1+i)+10,000/ (1+i) and

1.3∗3855 = .3∗3855/ (1+i) + 10,000/ (1+i)10.



fixed exchange rates—that capital controlsare ineffective and impose substantial costson economies that outweigh any benefits.That generalization ignores distinctions

among types of capital controls and variedcriteria for success, however. Capital con-trols have many potential purposes andthus many potential standards by which to

judge their efficacy. Difficulties in separat-ing the effects of capital controls from thebalance of payments or capital flow prob-lems they were intended to alleviate com-plicates the empirical study of the effects of capital controls (Johnston and Tamirisa,1998). Also, generalizing about the effec-tiveness of capital controls from one coun-try—or even one period—to another is

risky because the effectiveness of capitalcontrols depends on the rigor with whichthey are enforced (Obstfeld and Taylor,1998). Governments that control substan-tial aspects of their citizens’ lives (e.g.,Cuba) find it easier to enforce controls ontrade in assets (Minton, 1999).

Are Capital Controls Effective?

Keeping in mind these difficulties,there are several possible ways to gauge theeffectiveness of capital controls. Perhapsthe most direct way is to measure whetherthe imposition of capital controls changesthe magnitude or composition of capitalflows, using some assumption about whatflows would have been without the capitalcontrols. Measuring the composition of capital flows always has been difficult,however, and it has become more so since theadvent of derivatives that can be used todisguise capital flows. For example, a U.S.firm may build a production facility in Mexicobut hedge the risk that the peso will decline—

reducing the dollar value of the invest-ment—by buying put options on the peso,which will increase in value if the pesofalls.21 The direct investment will be mea-surable as an inflow, but the correspondingoutflow—the put contract to potentiallysell pesos—may not be (Garber, 1998).

If capital controls are designed to per-mit monetary autonomy, one can examinethe extent to which onshore interest

rates—subject to capital controls—differfrom those found in offshore markets ordomestic currency returns on foreignassets. Such tests assume that returns on

comparable investments in the same cur-rency should be equal in the absence of effective capital controls. To the extentthat they differ, the capital controls areeffective (Harberger, 1980; Edwards,1998b). This research has shown that cap-ital controls have been able to create mod-est “wedges” of one to several percentagepoints between returns on similar domes-tic and international assets (Marston,1995).22 A related test to determine mone-tary autonomy is to measure the effective-ness of sterilization in preventing an

appreciation of the real exchange rate.Generally, controls on inflows have

been found to be more effective than thoseon outflows because there is less incentiveto evade controls on inflows (Reinhart andSmith, 1998; Eichengreen, et al. 1999).23

Evading controls on inflows ordinarily willprovide only marginal benefits for foreigninvestors, as the expected risk-adjusteddomestic return usually will be compara-ble to that on alternative internationalinvestments. On the other hand, in theevent of an expected devaluation, there isenormous incentive to avoid such a loss byevading controls on capital outflows. Theexpected loss on holding domestic assetscan be several hundred percent in annual-ized terms over a short horizon. Forexample, if one expects the Malaysianringgit to be devalued 10 percent in oneweek, the expected continuously com-pounded annual return associated withholding the currency through such adevaluation is almost –550 percent.24

Therefore, researchers like Obstfeld (1998)

and Eichengreen (1999) have found theidea of preventing destabilizing outflowsby limiting inflows to be more promisingthan directly trying to stop outflows.

In sum, the consensus of the researchon capital controls has been that they canalter the composition of capital flows ordrive a small, permanent wedge betweendomestic and offshore interest rates butthey cannot indefinitely sustain inconsis-

21 An (American) put option con-

fers on the holder the right, but

not the obligation, to sell a

specified quantity of pesos at a

specified price, called the strike

price or exercise price, on or

before a given date.

22 Fieleke (1994) finds that capi-

tal controls were of very limited

effectiveness in creating inter-

est differentials during the

European Monetary System

crises of 1992-93.

23 As will be discussed in the next

subsection, capital controls

often are evaded by changing

from prohibited to permitted

assets or by falsifying invoices

for traded goods.

24 The continuously compounded

annual return is computed from

52* ln( .9/ 1) .



26



tent policies, and their effectiveness tendsto erode over time as consumers and firmsbecome better at evading the controls(Marston, 1995). Outflow restrictions, in

particular, may buy breathing space, butthat is all. There are more researchers will-ing to defend inflow restrictions, however.Eichengreen (1999) argues that, to restraininflows, controls do not have to be perfect,they just need to make avoidance costlyenough to reduce destabilizing flows.

How Are Capital Controls Evaded?

Over time, consumers and firms real-ize that they can evade capital controlsthrough the channels used to permit trade

in goods. Firms, for example, may evadecontrols on capital flows by falsifyinginvoices for traded goods; they apply tobuy or sell more foreign exchange than thetransaction calls for. For example, adomestic firm wishing to evade limits oncapital inflows might claim that it export -ed $10 million worth of goods when itonly, in fact, exported $9 million. It mayuse the excess $1 million to invest indomestic assets and split the proceeds withthe foreign firm providing the capital.

Perhaps the most common method toevade controls on capital flows is through“leads and lags” in which t rading firmshasten or delay payments for imports orexports (Einzig, 1968). To evade controlson outflows, for example, importers payearly for imports (leads), in exchange for adiscount, and exporters allow delayed pay-ments for their goods (lags), in return for ahigher payment. This permits importersand exporters to effectively lend money tothe rest of the world, a capital outflow. Toevade controls on inflows, importers delay

payments while exporters demand acceler-ated payments. Thus, leads and lags permittrade credit to substitute for short-termcapital flows. Governments often attemptto close the leads/lags loophole on short-term capital flows with administrative con-trols on import/export financing.

Travel allowances for tourists areanother method by which capital controlsmay be evaded (Bakker, 1996). More

recently, financial innovation has spawnedfinancial instruments—derivatives—thatmay be used to mislead banking and finan-cial regulators to evade prudential regula-

tion and/or capital controls (Garber,1998). For example, derivatives may con-tain clauses that change payouts in theevent of defaults or the imposition of exchange controls (Garber, 1998).Improvements in information technologymake it easier to buy and sell assets andreduce the effectiveness of capital controls(Eichengreen, et al. 1999).

Capital controls also induce substitu-tion from prohibited to permitted assets(Goldstein, 1995). So, for example, theU.S. interest equalization tax was evaded

through trade in assets with Canada whileheavy Chilean taxes on short-term inflowsmay have induced a (desired) substitutionto more lightly taxed longer-term inflows(Valdes, 1998). Capital controls have beenmore successful in changing the composi-tion of asset trade than the volume.

Costs of Capital Controls

Although they often are evaded success-fully, capital controls nonetheless imposesubstantial costs in inhibiting internationaltrade in assets. Foremost among these costsare limiting the benefits of capital flows asdescribed in Section 2: r isk-sharing, diver-sification, growth, and technology transfer(Global Investor , 1998a). Capital export-ing countries see a lower return on theirsavings while capital importers receive lessinvestment and grow more slowly withcapital controls. Krugman (1998) arguesthat capital controls do the most harmwhen they are used to defend inconsistentpolicies that produce an overvalued cur-

rency—a currency that would tend todepreciate or be devalued in the absence of the controls. This attempt to free govern-ments from the discipline of the marketpermits poor or inconsistent policies to bemaintained longer than they otherwisewould, increasing the costs of these policies.

Poorly designed or administered capi-tal controls often adversely affect directinvestment and the ordinary financing of



27



trade deals ( Economist , 1998). Controlscan even worsen the problem of destabiliz-ing capital flows. For example, the Koreangovernment has acknowledged that the

restriction on offshore borrowing byKorean corporations contributed to its bal-ance of payments and banking crises in1997 (Global Investor , 1998a). The cor-ruption created by evasion and the admin-istrative costs of controls also is anunintended cost of the controls. Even asthe costs accumulate and their originalpurpose has ended, capital controls, likeany regulation, develop their own con-stituencies and become difficult to phaseout. The resumption of free capital flowsdoes not always end the costs of capital

controls. Specifically, blocking the depar-ture of capital temporarily subsidizesinvestment but raises the perception of risk, increasing a risk premium and/ordeterring future investment ( Economist ,1998; Goldstein, 1995).

Partly because the costs of capital con-trols are serious and tend to worsen overtime, economists have suggested attackingproblems at their source rather than withcapital controls (Krugman, 1998; Mishkin,1998). For example, to cope with banks’incentive to take on excessive risk, a gov-ernment might concentrate on reformingand strengthening the domestic financialstructure—especially regulations on foreignborrowing—as it slowly phases out capitalcontrols to derive the benefits of capitalflows (Goldstein, 1995).25 Or, to fight a realappreciation brought on by a capital inflow,a government might conduct contractionaryfiscal policy. In all circumstances, bettermacroeconomic policy is needed to avoidfinancial crises, such as those that affectedAsia in 1997. Countries must eschew over-

valued currencies, excessive foreign debt,and unsustainable consumption.

CONCLUSION

Recently, a number of opinion leaders,including some prominent economists,have suggested that developing countriesshould reconsider capital controls. Thisarticle has reviewed the issues associated

with capital controls. Controls most oftenhave been used to permit more freedomfor monetary policy during balance of pay-ments crises in the context of fixed

exchange rates. Restrictions on inflowshave been implemented to prevent realappreciation of the exchange rate or tocorrect other pre-existing distortions, likethe incentives for financial institutions totake excessive risk. Although controls oncapital flows may change the compositionof flows, they impose substantial costs onthe economy and cannot be used to indefi-nitely sustain inconsistent policies. Undermost circumstances, it is better to attack the source of the distortion or inconsistentpolicy at the source rather than treating

symptoms through capital controls.Although the worst of the Asian finan-

cial crisis seems to be over, it—like the pesocrisis of December 1994—has been a sober-ing lesson in the volatility of capital flowsand the fragility of emerging market finan-cial systems. It also has raised questionsfor future research: Are limits on capitalinflows the best solution to protectingdomestic financial systems from the distor-tions inherent in banking? What is theproper sequence for economic reforms of the capital account, the current account,and the banking system?

REFERENCES

Alesina, Alberto, Vittorio Grilli, and Gian Maria Milesi-Ferretti. “ The

Political Economy of Capital Controls,” in Capital Mobility : The Impact

on Consumption, Investment, and Growth , Leonardo Leiderman and

Assaf Razin, eds., Cambridge University Press, 1994, pp. 289-321.

Bakker, Age F.P. The Liberalization of Capital Movements in Europe ,

Kluwer Academic Publishers, 1996.

Barro, Robert. “ Malaysia Could Do Worse than this Economic Plan,”

Business Week , November 2, 1998 , p. 2 6.

Bernanke, Ben S. “ Nonmonetary Effects of the Financial Crisis in

Propagation of the Great Depression,” American Economic Review

(June 1983), pp. 257-76.

Bhagwati, Jagdish. “ Why Free Capital Mobili ty may be Hazardous to

Your Health: Lessons from the Latest Financial Crisis,” NBER

Conference on Capital Controls, November 7, 1998.

Blustein, Paul. “ Is Malaysia’s Reform Working? Capital Controls Appear

to Aid Economy, but Doubts Remain,” Washington Post , November

21, 1998, p. G01.

25 An important but unresolved

issue is the sequencing of

reforms of the current account,

the capital account, and the

financial sector. Many blamethe recent Asian crisis on the

fact that the Asian governments

moved faster to liberalize inter-

national capital flows than they

did to regulate their financial

system. In contrast, Sweeney

(1997) argues for early liberal-

ization of the capital account to

provide accurate pricing infor-

mation for business decisions.



28



Marston, Richard C. International Financial Integration: A Study of

Interest Differentials Between the Major Industrial Countries ,

Cambridge University Press, 1995.

McKinnon, Ronald I., and Wallace E. Oates. “ The Impli cations of

International Economic Integration for Monetary, Fiscal andExchange Rate Policies,” The International Finance Section,

Princeton University, 1966.

Minton, Zinny, “ Global Finance Survey: A Wealth of Blueprints,”

Economist , January 30, 1999, pp. S5-S8.

Mishkin, Frederic S. “ International Capital Movements, Financial

Volatility and Financial Instability,” NBER Working Paper 6390,

January 1998.

Mitchell, B.R. International Historical Stati stics: Africa, Asia & Oceania

1750-1993 , Macmillan Reference, 1998.

New York Times . “ How U.S. Wooed Asia to Let Cash Flow In,” February

16, 1999, p. A1.

Obstfeld, Maurice. “ The Global Capital Market: Benefactor or

Menace?” Working Paper, 1998 .

_ _ _ _ _ _ _ _ and Alan M. Taylor. “ The Great Depression as a

Watershed: International Capital Mobility over the Long Run,” in

The Defining Moment: The Great Depression and the American

Economy in the Twentieth Century , Michael D. Bordo, Claudia D.

Goldin, and Eugene N. White, eds., University of Chicago Press, 1998,

pp. 353-402.

Park, Yung Chul, and Chi-Young Song. “ Managing Foreign Capital

Flows: The Experiences of Korea, Thailand, Malaysia and Indonesia,”

Jerome Levy Economics Institute Working Paper 163, May 1996.

Pearce, David W., ed. MIT Dictionary of Modern Economics , MIT Press,

1995.

Reinhart, Carmen M., and R. Todd Smith. “ Too Much of a Good Thing:

The Macroeconomic Effects of Taxing Capital Inflow s,” in Managing

Capital Flows and Exchange Rates: Perspectives from the Pacific Basin , Reuven Glick, ed., Cambridge University Press, 1998,

pp. 436-64.

Stern, Robert. The Balance of Payments: Theory and Economic Policy ,

Aldine Publishing Company, 1973.

Sweeney, Richard J. “ The Information Costs of Capital Controls,” in

Capital Controls in Emerging Economies , Christine P. Ries and Richard

J. Sweeney, eds., Westview Press, 1997, pp. 45-61.

Torres, Craig. “ Chilean Bid to Boost Confidence Lauded,” Wall Street

Journal , June 29, 199 8, p. A14.

ul Haq, Mahbub, Inge Kaul, and Isabelle Grunberg, eds. The Tobin Tax:

Coping with Financial Volatility , Oxford University Press, 1996.

Valdes, Salvador. “ Capital Controls in Chile Were a Failure,” Wall Street

Journal , December 11, 1998, p. A15.

_ _ _ _ _ _ _ _ and Marcelo Soto. “ New Selective Capital Controls in

Chile: Are they Effective?” Catholic University of Chile Working

Paper, 19 96.

Wade, Robert, and Frank Veneroso. “ The Gathering Support for Capital

Controls,” Challenge (November/ December 199 8) , pp. 14-26.



Intro to Capital Controls

Documents

Transcript of Intro to Capital Controls