zimbabwe: hyperinflation to growth

36

Click here to load reader

Upload: haminh

Post on 04-Jan-2017

234 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Zimbabwe: Hyperinflation to Growth

The hallmark of Zimbabwe’s economic collapse ishyperinflation. The most recent official inflation fig-ure is for February 2008: a whopping 165,000 percent

year-over-year. At present (early June 2008), inflation is unof-ficially about 2.5 million percent a year. Not surprisingly, theZimbabwe dollar has lost more than 99.9 percent of its valueagainst the U.S. dollar during the past year.

Zimbabwe’s hyperinflation is destroying the economy,pushing more of its inhabitants into poverty, and forcingmillions of Zimbabweans to emigrate. Between 1997 and2007, cumulative inflation was nearly 3.8 billion percent,while living standards fell by 38 percent.

The source of Zimbabwe’s hyperinflation is the ReserveBank of Zimbabwe’s money machine. The government spends,and the RBZ finances the spending by printing money. TheRBZ has no ability in practice to resist the government’sdemands for cash. Accordingly, the RBZ cannot hope to regaincredibility anytime soon. To stop hyperinflation, Zimbabweneeds to immediately adopt a different monetary system.

Any one of three options can rapidly slash the inflationrate and restore stability and growth to the Zimbabweaneconomy. First is “dollarization.” This option would replacethe discredited Zimbabwe dollar with a foreign currency, suchas the U.S. dollar or the South African rand. Second is a cur-rency board. Under that system, the Zimbabwe dollar wouldbe credible because it would be fully backed by a foreignreserve currency and would be freely convertible into thereserve currency at a fixed rate on demand. Third is free bank-ing. This option would allow commercial banks to issue theirown private notes and other liabilities with minimum gov-ernment regulation.

Central banking is the only monetary system that hasever created hyperinflation and instability in Zimbabwe.Prior to central banking, Zimbabwe had a rich monetaryexperience in which a free banking system and a currencyboard system performed well. It is time for Zimbabwe toadopt one of these proven monetary systems and discardits failed experiment with central banking.

the cato institute1000 Massachusetts Avenue, N.W., Washington, D.C. 20001-5403

www.cato.orgPhone (202) 842-0200 Fax (202) 842-3490

J u n e 2 5 , 2 0 0 8 ● n o . 6

ZimbabweFrom Hyperinflation to Growth

by Steve H. Hanke

Steve H. Hanke is a professor of applied economics at Johns Hopkins University in Baltimore and a senior fellow at the Cato Institute.

Executive Summary

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 1

Page 2: Zimbabwe: Hyperinflation to Growth

HyperinflationIs Destroying

Zimbabwe’s EconomyWhen properly applied, the rule of law

guarantees freedoms in the economic, politi-cal, intellectual, and moral spheres. In the eco-nomic sphere, money constitutes an impor-tant element. Ludwig von Mises, one of themost important economists of the 20th cen-tury, dealt at length with this issue in his trea-tise The Theory of Money and Credit, publishedoriginally in 1912:

It is impossible to grasp the meaning ofthe idea of sound money if one does notrealize that it was devised as an instru-ment for the protection of civil libertiesagainst despotic inroads on the part ofgovernments. Ideologically it belongs inthe same class with political constitutionsand bills of rights. The demand for con-stitutional guarantees and for bills ofrights was a reaction against arbitrary ruleand the non-observance of old customs

by kings. The postulate of sound moneywas first brought up as a response to theprincely practice of debasing the coinage.It was later carefully elaborated and per-fected in the age which—through theexperience of the American ContinentalCurrency, the paper money of the FrenchRevolution and the British Restrictionperiod—had learned what a governmentcan do to a nation’s currency system.1

Since March 2007 Zimbabwe has been inthe midst of a hyperinflation—defined as a rateof inflation per month that exceeds 50 percent.Hyperinflations are rare: there have been only29 other cases in history, with the most recentone occurring in Bulgaria in 1997.2

Zimbabwe’s hyperinflation is destroyingthe economy, pushing more of its inhabitantsinto poverty, and forcing millions of Zimbab-weans to emigrate. In the 1997–2007 period,cumulative inflation was nearly 3.8 billion per-cent, while living standards (as measured byreal gross domestic product [GDP] per capita)fell by 38 percent (see Table 1). In addition,

2

Hyperinflation is destroying Zimbabwe’s

economy, pushing more of

its inhabitantsinto poverty, andforcing millions

to emigrate.

1

1,000

1,000,000

1,000,000,000

1,000,000,000,000

1,000,000,000,000,000

Zimbabwe (ZWD / USD)

Germany (Marks / USD)

Figure 1Exchange Rates

Sources: Thomas J. Sargent, Rational Expectations and Inflation, 2nd ed. (New York: Harper Collins, 1993); ImaraAsset Management Zimbabwe, May 1, 2008; and author’s calculations.Note: A logarithmic scale is used so that the chart can fit on one page.

ZW

D a

nd M

arks

per

USD

(Log

arith

mic

Sca

le)

Zimbabwe 2005Germany 1921

Zimbabwe 2006Germany 1922

Zimbabwe 2007Germany 1923

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 2

Page 3: Zimbabwe: Hyperinflation to Growth

hyperinflation has robbed people of their sav-ings and financial institutions of their capitalthrough real (inflation-adjusted) interest ratesthat are actually negative (see Table 1). Thisform of theft occurs, in large part, because thelaws and regulations governing financial insti-tutions (pension funds, insurance companies,building societies, and banks) force them toeither purchase government treasury bills thatyield only a small fraction of the current infla-tion rate or make deposits at the Reserve Bankof Zimbabwe (RBZ) that pay no interest.

The value of the Zimbabwe dollar has beenwiped out. Figure 1 tells the devastating story—one that is ominously following the same plot

as that followed by the German mark duringthe great German hyperinflation of the 1920s.

Worse is yet to come. Most private observersof the Zimbabwean economy believe that thedata reported by the International MonetaryFund (see Table 1) are too conservative.Moreover, they believe that in the ramp up tothe March elections and the June 2008 presi-dential runoff, the government has forced theRBZ to accelerate the printing presses.

The Reserve Bank of Zimbabwe HasFacilitated the Hyperinflation

The root cause of the hyperinflation isthat government policies have forced the

3

In the 1997–2007 period,cumulative inflation wasnearly 3.8 billionpercent, while living standardsfell by 38 percent.

Table 1Zimbabwe Economic Data (Percent)

Year Real GDP Per Capita Growth Consumer Price Inflation Lending Ratea Real Interest Rateb

1990 3.7 15.5 11.7 -5.61991 3.8 46.5 15.5 -8.51992 -11.2 46.3 19.7 -21.81993 -1.4 18.6 36.3 8.11994 2.3 21.1 34.8 12.61995 -3.1 25.8 34.7 12.21996 6.2 16.4 34.2 12.61997 2.4 20.1 32.5 13.71998 0.4 46.7 42.0 10.71999 -3.3 56.9 55.3 -2.62000 -7.0 55.2 68.2 12.62001 -2.4 112.1 38.0 -35.32002 -4.1 198.9 36.4 -96.72003 -11.3 598.7 97.2 -267.72004 -3.3 132.7 278.9 -71.02005 -4.0 585.8 235.6 -2.12006 -5.4 1,281.1 496.4 -520.22007 -6.1 108,844.1 N/A N/A

Sources: International Monetary Fund, World Economic Outlook database, April 2008, http://www.imf.org/external/pubs/ft/weo/2008/01/weodata/index.aspx; International Financial Statistics database, May 2008; author’s calcula-tions.a Rate charged by commercial banks on loans (International Financial Statistics database).b Lending rate adjusted for inflation.Note: Consumer price inflation, the lending rate, and the real interest rate are reported on an end-of-year basis. The2007 data fail to accurately reflect inflation pressures because the government mandated sharp reductions in adminis-tered prices. In addition, most observers believe that real GDP per capita declined by more than the -6.1 percent report-ed by the IMF for 2007.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 3

Page 4: Zimbabwe: Hyperinflation to Growth

RBZ to print money. From January 2005 toMay 2007, the RBZ issued currency at a ratethat even exceeded that of Germany’s centralbank from January 1921 to May 1923, theramp-up phase of the great German hyperin-flation (Figure 2).

Hyperinflation isn’t the only thing producedby the RBZ. It is also a proficient producer ofjobs, funded at the expense of every Zimbab-wean who uses money. In the 2001–2007 peri-od, the RBZ’s staff more than doubled, from618 to 1,360 employees.3 This staff increase of120 percent was the largest of any central bankin the world during this period. Despite thisincrease, the staff is not producing accurate andtimely data. The RBZ has fallen months, if notyears, behind in reporting even the most stan-dard economic and financial statistics.

Replacing the Central Bank Is the BestWay to Stop Hyperinflation

The most rapid and reliable way to stophyperinflation in Zimbabwe is to replace cen-

tral banking with a new monetary regime.That would signal a clean break with the prac-tices that have created hyperinflation andwould assure Zimbabweans that inflation willhenceforth be controlled.

Some may wonder whether it is necessaryto replace central banking with another mone-tary system in Zimbabwe. After all, other coun-tries with inflation in the hundreds or thou-sands of percent a year—including Angola,Mozambique, the Democratic Republic of theCongo, and Zambia—have checked high infla-tions without replacing their central banks.Instead, they changed their policies. Whycouldn’t Zimbabwe do likewise?

It could, but so far it hasn’t. In Zimbabwe’shistorical experience with a variety of mone-tary systems, only central banking has pro-duced hyperinflation. Similarly, throughoutthe world, hyperinflation has been a phenom-enon linked to central banking or its closecousin, the direct issue of currency by a gov-ernment’s treasury.

4

The root cause of the

hyperinflation is government

policies that haveforced the RBZ to

print money.

1.0E+00

1.0E+03

1.0E+06

1.0E+09

1.0E+12

1.0E+15

Germany

Zimbabwe

Figure 2Currency in Circulation

Sources: Thomas J. Sargent, Rational Expectations and Inflation, 2nd ed. (New York: Harper Collins, 1993); ReserveBank of Zimbabwe, “Statistics, Monthly Review” (February 2008), p. 26, http://www.rbz.co.zw; and author’s calcula-tions.Note: A logarithmic scale is used so that the chart can fit on one page.

Perc

ent G

row

th (

Log

arith

mic

Sca

le)

Zimbabwe 2005Germany 1921

Zimbabwe 2006Germany 1922

Zimbabwe 2007Germany 1923

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 4

Page 5: Zimbabwe: Hyperinflation to Growth

Central banks can put a stop to inflation asfast as they can fuel it. All they have to do is stopthe printing presses. Unfortunately, they canswitch course again with ease. Consequently,under central banking, inflation can return aseasily as it was snuffed out. Many countries,including the African countries mentionedabove, have had prolonged double-digit infla-tion or multiple bouts of very high inflationunder central banking. Central banks that havemade a seemingly permanent transition to lowinflation have usually required a long transitionperiod in which to establish their credibility asinflation fighters. During the transition, realinterest rates have often been punishingly high,and long-term loans in local currency have beendifficult to obtain. As a result, economic growthhas been relatively slow, general living stan-dards have remained stagnant, and the scourgeof poverty has spread. Given the current state ofaffairs in Zimbabwe and the dramatic hyperin-flation, the only way for Zimbabwe to make acredible commitment to stopping the hyperin-flation rapidly and avoid high transition costsis to replace central banking with a differenttype of monetary regime.

Options for Replacing the Central Bank4

Three options exist for quickly replacingthe RBZ with monetary regimes that wouldend hyperinflation and provide monetary sta-bility: (1) official “dollarization,” (2) free bank-ing, and (3) a currency board. These optionsare not mutually exclusive. For example, a cur-rency board system could be combined withofficial dollarization, as in the monetary sys-tems of Lesotho, Namibia, and Swaziland.5 (Afourth option—formal membership in theSouth African rand’s Common MonetaryArea—is not dealt with explicitly in this studysimply because it would probably involve pro-tracted political negotiations and time, whichZimbabwe can ill afford.)

Dollarization occurs when residents of acountry extensively use the U.S. dollar oranother foreign currency alongside, or insteadof, the domestic currency. Unofficial dollariza-tion occurs when individuals hold foreign-cur-rency bank deposits or notes (paper money) to

protect against high inflation in the domesticcurrency. Zimbabwe is already unofficially dol-larized to the extent that Zimbabweans holdSouth African rand, U.S. dollars, pounds ster-ling, and other foreign currencies as stores ofvalue. Official dollarization occurs when acountry uses a foreign currency as the main, oronly, component of the monetary base. (Themonetary base is the medium accepted forfinal settlement of payments in the local finan-cial system; in other words, it is what bankstypically use for clearing.)

For Zimbabwe, official dollarization couldbe based on, for example, the South Africanrand, the U.S. dollar, or the euro. A later sectiondiscusses the advantages and disadvantages ofeach currency. If Zimbabwe used the rand, itcould negotiate a profit-sharing agreementsuch as Lesotho and Namibia now have, andwhich Botswana and Swaziland formerly had.Under these agreements, South Africa sharesthe profit (seigniorage) it derives from issuingcurrency, according to estimates of how manyrand notes (paper money) and coins are in cir-culation in the partner country.6 In the late1920s, when Zimbabwe (then called SouthernRhodesia) used South African coins, it appar-ently had a profit-sharing arrangement withSouth Africa. The agreement applied only tocoins in circulation, because Zimbabwe had itsown notes.

Those notes were issued under “free bank-ing,” a system of competitive issuance by pri-vate commercial banks of notes and other lia-bilities with minimal regulation. A completelyfree banking system has no central bank, nolender of last resort, no reserve requirements,and no legal restrictions on bank portfolios,interest rates, or branch banking. Free bank-ing systems existed in nearly 60 countries dur-ing the 1800s and early 1900s. In general, thesesystems were relatively stable, issued curren-cies convertible into gold or silver at fixedexchange rates, and were not purveyors ofinflation.7

Zimbabwe had free banking from thetime its first bank was established in 1892until the government replaced free bankingwith a currency board in 1940. Zimbabwe’s

5

Throughout the world, hyperinflationhas been a phenomenonlinked to centralbanking.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 5

Page 6: Zimbabwe: Hyperinflation to Growth

free banking system was among the leastrestricted that ever existed. Reflecting thelimited use of modern money and creditamong the population at the time, the coun-try had only two commercial banks, theStandard Bank of South Africa and the Bankof Africa (later part of Barclays Bank). Theyissued notes denominated in pounds, andkept their privately issued pounds equal tothe pound sterling—except during the FirstWorld War and for a few years afterwards,when the local pound floated along with theSouth African pound (the predecessor to therand) against the pound sterling. Free bank-ing ended in Zimbabwe not because it per-formed poorly, but because the governmentdesired the profits from issuing notes.

Currency boards or central banks replacedfree banking systems because intellectual andpolitical fashions favored monopolizing theissuance of notes and coins by a governmentbody. Today, no free banking systems exist. Inthe last 30 years, economists’ interest in freebanking has revived because of dissatisfactionwith the performance of central banks.8 Morerecently, the possibility that electronic moneywill make notes and coins obsolete, enablingbanks to offer full-fledged rivals to govern-ment-issued currencies, has generated consid-erable interest.9

The final option for quickly replacing theRBZ is a currency board. Currency boardshave existed in more than 70 countries, and anumber are in operation today.10 A currencyboard is a monetary institution that issuesnotes and coins. (Even though some currencyboards accept deposits, the board this studyproposes for Zimbabwe would be prohibitedfrom doing so.). A currency board’s monetaryliabilities are fully backed by a foreign reservecurrency (also called the anchor currency) andare freely convertible into the reserve currencyat a fixed rate on demand. The reserve curren-cy is a convertible foreign currency or a com-modity chosen for its expected stability. Asreserves, a currency board holds low-risk,interest-earning securities and other assetspayable in the reserve currency. A currencyboard holds reserves equal to 100 percent or

slightly more of its notes and coins in circula-tion, as set by law.

Zimbabwe established the Southern Rho-desia Currency Board in 1940. The boardreplaced free banking, a system that had experi-enced no severe problems and had servedZimbabwe well. In establishing the currencyboard, the colonial government was acting inaccord with the prevailing belief of the timethat issuing notes and coins should be a gov-ernment monopoly. Officials were also awarethat a currency board would generate profitsand, therefore, revenue for the government.

The notes and coins of the SouthernRhodesia Currency Board were legal tender inZambia (then called Northern Rhodesia) andMalawi (then called Nyasaland) as well as inZimbabwe. The board existed until 1954, whenit was replaced by the Central Africa CurrencyBoard. The main feature distinguishing theCentral Africa board from its predecessor wasthat it allowed representation on its governingbody by Zambians and Malawians as well asZimbabweans.

The three governments of the Federationof Rhodesia and Nyasaland transformed theCentral Africa Currency Board into the Bankof Rhodesia and Nyasaland, a central bank,in 1956. The central bank was one of severalinstitutions intended to bind Zimbabwe,Zambia, and Malawi into a single economicunit and ultimately perhaps a single politicalunit. As with the transition from free bank-ing to a currency board, the transition from acurrency board to a central bank did notoccur because the currency board had experi-enced any severe problems. Rather, centralbanking had become the intellectual fashionof the time for countries that were indepen-dent or wanted to become independent soon.There was as yet little recognition of the dan-gers that inept central banking could bring:high inflation, exchange controls, and finan-cial underdevelopment. When the federationdissolved in 1964, its central bank was divid-ed into three separate central banks for eachcountry that had been a member of the fed-eration. Thus the Reserve Bank of Rhodesiacame into being.

6

To restore economic growth,

hyperinflationmust be extin-

guished rapidly,and people musthave confidence

that inflation willnot return.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 6

Page 7: Zimbabwe: Hyperinflation to Growth

Official dollarization, free banking, and acurrency board are all proven systems withrecords of success in providing reliable, low-inflation currencies in Zimbabwe and elsewhere.Any one of these systems, or a combination ofthem, could be successfully implementedimmediately, without preconditions, and wouldtherefore quickly put an end to hyperinflationand produce stable money.11 Although the dif-ferences among the three options are worthpondering, it must be stressed that any onewould represent an enormous improvementover the monetary policy that the governmenthas forced the RBZ to produce or that the RBZis likely to produce in the near future.

Financial LiberalizationFinancial liberalization is a vital companion

policy to dollarization, free banking, or a cur-rency board system. With the elimination ofhyperinflation, the rationale for price controlsand foreign exchange controls no longer exists.Both should be prohibited. Other forms of“financial repression,” including interest rateceilings, the forced purchase of governmentbonds, minimum reserve requirements forfinancial institutions, and the compulsory allo-cation of credit to favored borrowers should alsobe prohibited. This liberalization would permitthe free flow of capital into and out of

Zimbabwe. It would also increase the return onsavings and reduce the cost of capital inZimbabwe, removing major impediments toeconomic growth and improved living stan-dards. Such has been the experience of othercountries that have ended financial repression.12

Financial liberalization is critically impor-tant for Zimbabwe’s economic recovery. Depre-ciation of assets has exceeded investment forsome time, resulting in capital consumptionand the atrophy of the nation’s plants andequipment. In addition, the combination ofhyperinflation and price controls has wiped outmuch of the country’s stock of working capital.The quickest way to replenish Zimbabwe’s capi-tal stock is to import capital. To accomplish this,stable money and a liberal financial regime areimportant prerequisites.

Where Is the Money?Any analysis of alternative monetary ar-

rangements for Zimbabwe must involve a lookat the financial condition of the RBZ. The RBZreleases its balance sheet with a lag of at leastthree months. In the present hyperinflation,that lag is so long as to render balance sheet fig-ures almost useless. Even so, Table 2 presentsthe RBZ’s balance sheet as of January 2008.

It is remarkable that the RBZ classifies somany assets and liabilities on its balance sheet

7

The most rapidand reliable way to stop hyperinflation inZimbabwe is toreplace centralbanking with anew monetaryregime.

Table 2RBZ Balance Sheet, January 2008 (in Trillions of Zimbabwe Dollars)

Assets Liabilities

Gold 0.1 Notes and coins in circulation 339.8Other foreign assets 5.2 Bankers’ deposits 108.5Treasury bills discounted 0.0 Government deposits 0.0Other bills discounted 0.0 Other deposits 108.5Loans to government 39.4 Foreign loans 16.7Other loans 64.2 Other 0.0Investments—government stock 0.0 Capital and reserve 0.0Other investments 298.3 Other 609.9Other 776.2Total 1183.4 Total 1183.4

Source: Reserve Bank of Zimbabwe, “Statistics, Monthly Review” (February 2008), pp. 26–27, http://www.rbz.co.zw.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 7

Page 8: Zimbabwe: Hyperinflation to Growth

as “other.” Those assets most likely representvarious forms of credit to the government andto state-owned enterprises, while the liabilitiesrepresent their counterparts. The lumping ofthe majority of assets and liabilities into acatch-all category is yet another reason whythe balance sheet is so unilluminating. Thatsaid, the central bank clearly has no foreignreserves to speak of; they exist but are so smallas to represent nothing more than a roundingerror. The balance sheet does, however, revealone big thing: the RBZ is operating as the gov-ernment’s printing press.

As is the case with the RBZ’s financial state-ments, the government’s disclosures of its rev-enue and spending are not frequent enough tobe of much use in the current hyperinflation-ary environment. Indeed, figures more than aweek old are seriously out of date. Despite thelack of timely data, it is apparent that the gov-ernment is financing most of its spendingthrough money that the RBZ prints and lendsto it. It is also apparent that inflation hasreached the point where a large burst of addi-tional money printing is necessary to finance arelatively small increase in spending.13

Zimbabwe is in the late stages of a classichyperinflation. Consequently, tax revenuefrom sources other than inflation is shrinkingbecause lags between collection and spendingerode the real value of money. Inflation is gal-loping ahead as the supply of Zimbabwe dol-lars surges and the demand for them shrinks.Eventually, the currency will totally collapse aspeople simply refuse to accept it.

The RBZ nevertheless proudly proclaimsthat its vision is “to become the financial cor-nerstone around which Zimbabwe’s econom-ic fortunes and developmental aspirationsare anchored,” and that “the pursuit of theBank’s vision will express itself through lead-ership in the formulation, implementationand monitoring of policies and action plansfor fighting inflation, stabilisation of theinternal and external value of Zimbabwe’scurrency and of the financial system in amanner that gives pride of achievement toZimbabweans across the board.” These arenow little more than surreal slogans.14

Dollarization

Let us now consider the options for replac-ing the Reserve Bank of Zimbabwe. The first isdollarization. The three types of dollarizationare unofficial, semiofficial, and official. Takentogether, these forms of dollarization accountfor 55 to 70 percent of U.S. dollar notes in cir-culation,15 a significant portion of the stock ofthe euro cash, and small portions of other cur-rencies, including the South African rand.

Unofficial dollarization occurs when peo-ple hold much of their financial wealth inforeign assets even though foreign currencyis not legal tender. Unofficial dollarizationcan include the holding of foreign bonds andother nonmonetary assets, foreign-currencydeposits (either abroad or domestically), andforeign notes (paper money) in wallets, undermattresses, and in safe-deposit boxes.

More than a dozen countries have whatmight be called semiofficial dollarization orofficially bimonetary systems. Under semioffi-cial dollarization, foreign currency is legal ten-der and may even dominate bank deposits,but it plays a secondary role to domestic cur-rency in paying wages, taxes, and everydayexpenses such as grocery and electric bills.Unlike officially dollarized countries, semiof-ficially dollarized ones retain a domestic cen-tral bank or other monetary authority andhave corresponding latitude to conduct theirown monetary policy. Lesotho, Namibia, andSwaziland may be described as semiofficiallydollarized in that they allow the South Africanrand to circulate as legal tender alongside theirdomestically issued currencies.

Official dollarization, also called full dol-larization, occurs when a foreign currency (orcurrencies) has exclusive or predominant sta-tus as full legal tender. Not only is a foreigncurrency (or currencies) legal for use in con-tracts between private parties, but the gov-ernment uses it (them) for payments, too. Ifdomestic currency exists, it is confined to asecondary role, such as being issued only inthe form of coins having small value. Today,32 countries are officially dollarized.

8

Confidence canbest be restored

if Zimbabwe’s central banking

system is replacedwith dollarization,

a currency boardsystem, or

free banking.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 8

Page 9: Zimbabwe: Hyperinflation to Growth

How Official Dollarization WorksAn officially dollarized country is part of a

unified currency zone with the country whosecurrency it uses, hereafter called the anchor-currency country. Panama, for example, hasbeen officially dollarized and part of the U.S.dollar zone since 1904. An officially dollarizedcountry relinquishes an independent mone-tary policy and “imports” the monetary policyof the country whose currency it uses. Withinthe unified currency zone, arbitrage—buyingand selling to take advantage of differences inprices—tends to keep prices of similar goodswithin a narrow range. If a computer costs$500 in the United States, in Panama it cannotcost more than $500 plus extra taxes and ship-ping costs, otherwise it becomes profitable toship computers from the United States toPanama until the difference in price vanishes.16

Because arbitrage tends to keep prices ofsimilar, internationally traded goods within anarrow range throughout the unified curren-cy zone, inflation rates tend to be broadlysimilar throughout the zone. However, infla-tion need not be exactly the same through-out the zone. For example, in fast-growingregions, prices of goods that are not mobile,particularly real estate and labor, can risemuch more rapidly than the zonal average.Rapidly growing regions within the zone cantherefore have higher inflation than otherareas. There is nothing unusual about this;the same thing occurs in different regions ofa single country.

Interest rates also tend to be broadly simi-lar throughout a unified currency zone: if 30-year mortgages have an interest rate of 8 per-cent in the United States, the rate cannot betoo much higher in Panama; otherwise itbecomes profitable for U.S. banks to supplyfunds for mortgages in Panama until the dif-ference in rates vanishes. Some difference ininterest rates can persist, however, because ofcountry risk (political factors that affect thesecurity of property rights and credit risks).Interest rate differentials between the anchor-currency country and the dollarized countrycan also occur when their financial systemsaren’t fully integrated.

In an officially dollarized country, thesupply of money is determined “automatical-ly” by that country’s balance of payments,which itself reflects people’s preferences forholding versus spending money. The anchor-currency country determines the amount ofthe monetary base in existence (notes andcoins in circulation, plus bank reserves). Themonetary base is then held by people in vari-ous regions within the anchor-currencycountry or other countries in the unified cur-rency zone according to the intensity of theirdemand for base money. If people want toacquire more anchor-currency notes, theyhave to spend less, other things being equal;if they have more anchor-currency notes thanthey want, they can get rid of them by spend-ing more.

The current-account balance (trade ingoods and services) does not, however, rigidlydetermine the supply of money in an officiallydollarized country. People can also acquire ordispose of spending power through capital-account transactions (trade in financial assets,such as obtaining or making loans). Suppose,for example, that in one year Panama sells $6billion of goods and services to the rest of theworld and buys $7 billion; then its current-account deficit for the year is $1 billion. Thatdoes not mean its money supply must con-tract by $1 billion. If during the same year,Panamanians invest nothing abroad and for-eigners invest $2 billion in Panama, the capi-tal-account surplus is $2 billion, resulting in acombined surplus of $1 billion. In conse-quence, the money supply would expand,rather than contract.

When countries experience real economicshocks—those originating from changes insupply and demand (such as increases ordecreases in oil prices)—economic adjust-ments must follow. In an officially dollarizedcountry, adjustments can occur via all theusual channels except one: alterations in theexchange rate of its currency. Evidence fromdeveloping countries indicates that the lackof the exchange-rate adjustment channelenhances rather than impairs economic per-formance.17

9

Free banking systems existed in nearly 60 countries during the 1800sand early 1900s.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 9

Page 10: Zimbabwe: Hyperinflation to Growth

Beyond Simple Dollarization toFinancial Integration

If official dollarization goes no further thanthe use of a foreign currency, it does notachieve its full potential. While an officiallydollarized country has a unified currency withthe anchor-currency country, it does not neces-sarily have a financial system integrated withthat of the anchor-currency country. Fullfinancial integration occurs when the lawallows financial institutions extensive freedomof action to compete and does not discrimi-nate against foreign institutions. In particular,it means that foreign financial institutions canestablish branches, accept deposits and makeloans, obtain full ownership of domestic insti-tutions, and move funds freely into and out ofthe country. (Similar considerations also applyto currency-board and free-banking systems,but they are most pronounced in the case ofdollarization; to avoid repetition they are dis-cussed only in this section.)

Combining financial integration with offi-cial dollarization, in which a leading interna-tional or regional currency serves as theanchor currency, helps a dollarized countrytap into a large and liquid international poolof funds. Therefore, the location of loansneed not be closely linked to the location ofdeposits. Citibank, for example, does notneed to balance its loans and deposits inPanama any more than it needs to balance itsloans and deposits within the city of NewYork, where its headquarters are located. Itcan borrow where the cost of funds is lowestand lend where the risk-adjusted potential forprofit is highest anywhere in the dollar zone.The ability to allocate funds within an inte-grated financial system without exchange riskbetween an officially dollarized country andthe anchor-currency country reduces thebooms and busts of foreign capital that oftenarise in countries with independent monetarypolicies and financial systems that are notwell integrated into international capital mar-kets. That, in turn, acts to stabilize the realexchange rate (a measure of the impact ofchanges in the exchange rate and inflation onthe competitiveness of exports). With finan-

cial integration, domestic interbank marketsbecome much more liquid and reliablebecause foreign financial institutions canlend funds to domestic institutions. Readyaccess to foreign funds offers financially inte-grated, dollarized countries a substitute for acentral bank’s lender-of-last-resort facility. InPanama, the interbank market and the inte-grated financial system in general haveproven to be quite robust, even during timesof political turmoil, such as the 1989 U.S.invasion to oust Panama’s dictator.

Besides helping to stabilize the financialsystem and the economy, financial integra-tion improves the quality of the financial sys-tem by allowing consumers access to financialinstitutions that have proved their compe-tence internationally. Indeed, the exposure tointernational competition forces domesticfinancial institutions to reduce their costsand improve their services or lose marketshare.

The leading example of financial integra-tion in recent years has been the euro area.The currencies that preceded the euro experi-enced a number of crises. In particular, theso-called Exchange Rate Mechanism of theEuropean Monetary System experienced asevere crisis in 1992. The crisis erupted whenspeculators correctly concluded that the gov-ernments of most of the countries involvedwere not fully committed to maintaining theexchange rates of their currencies to theGerman mark, which was the de facto anchorfor the system.

When the euro replaced national curren-cies in 1999, no crisis occurred; nor has theeuro area experienced any internal currencycrises since 1999. Problems have been limitedto particular financial institutions that madebad decisions. Monetary unification andfinancial integration are responsible, in largepart, for the absence of crises. With a singlecurrency Germans cannot speculate against aseparate French currency, and people outsidethe euro area cannot speculate against Franceseparately from Germany in matters of mone-tary policy. An increasingly integrated finan-cial system, in which banks do much of their

10

Zimbabwe hadfree banking

from the time itsfirst bank wasestablished in

1892 until 1940.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 1

Page 11: Zimbabwe: Hyperinflation to Growth

business across national borders, means thatfinancial shocks tend to be distributed broad-ly and evenly over the whole euro area, ratherthan narrowly and deeply within one country.Consequently, the financial system of the euroarea is more robust than were the separatenational financial systems that existed before1999.

Which Currency to Choose?Which currency should Zimbabwe use for

official dollarization? The obvious choicesare the U.S. dollar, the euro, or the SouthAfrican rand. The considerations for choos-ing a currency are similar under dollariza-tion, a currency board, or free banking. Toavoid needless repetition, the currency choiceissue is discussed only in this section.

The U.S. dollar is the most widely used cur-rency in international trade and finance. Rawmaterials, such as Zimbabwe’s major exports—tobacco, cotton, gold, chromium, and nickel—are predominantly priced in dollars on worldmarkets. The U.S. dollar has gained its statusas the world’s predominant currency becauseit is fully convertible. In addition, inflationand real interest rates in the world’s largesteconomy have remained relatively low and sta-ble, giving credibility to U.S. monetary policyand the U.S. dollar.

The euro is another possibility. Although ithas existed as a unit of account only since1999, and as notes and coins since 2002, it hasquickly established itself as the second mostimportant international currency. During itsshort history, it has performed well, with lowinflation, high credibility, full convertibility,and low real interest rates. Many African coun-tries, notably the former French and Portu-guese colonies belonging to the CFA franczone, already link their currencies rigidly tothe euro. In addition, the euro area is a moreimportant trading partner than the UnitedStates for most African countries.

The final possibility, which seems the mostnatural, is the South African rand. SouthAfrica is Zimbabwe’s largest trading partner. Itis also the country with the greatest number ofZimbabwean expatriates, many of whom

transfer funds to relatives back in Zimbabwe.South Africa has a sophisticated financial sys-tem, to which Zimbabwe’s system has manylinks.

The rand does not have as good a long-term record as the U.S. dollar. However, bythe standards of emerging markets, its recordis not bad, and in recent years it has been fair-ly good. Unlike the U.S. dollar or the euro,the rand has exchange controls. But since1995, successive South African governmentshave reduced those controls, and furtherreductions are probable. Adopting the randwould make Zimbabwe part of a unified cur-rency zone with Lesotho, Namibia, andSwaziland as well as South Africa.

Adopting one currency as the main cur-rency for government transactions underofficial dollarization, or as the anchor for acurrency board, need not preclude peoplefrom using other currencies. In fact, it isdesirable to allow people to make contractsin any currency they find mutually agreeable.However, because of economies that arisewhen most people use the same currency, theusual tendency is for one currency to pre-dominate even where people can chooseamong many.

Were Zimbabwe to establish a currencyboard or free banking rather than dollariza-tion, the board or banks could use a basket ofcurrencies rather than a single currency asthe anchor. But baskets are less transparentto the public than a single anchor currency,and thus may make it more difficult toachieve high credibility and confidence. Abasket also imposes greater costs on the cur-rency board or a free banking system in termsof management time and transaction fees.Perhaps that explains why no previous cur-rency board or free banking system has useda basket as its anchor.

How to Establish DollarizationZimbabwe could implement official dollar-

ization by taking a series of steps (see Appendixfor more detail). For concreteness, the steps list-ed below assume that the South African randreplaces the Zimbabwe dollar. Accordingly, the

11

In the last 30 years, economists’interest in freebanking hasrevived becauseof dissatisfactionwith the performance ofcentral banks.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 11

Page 12: Zimbabwe: Hyperinflation to Growth

term “randization” is used rather than dollar-ization. The example assumes that the RBZ hasno foreign reserves and that foreign currencycannot immediately be obtained to exchangefor Zimbabwe dollars. On this point, Zimbab-we would differ from Ecuador and El Salvador;they began their dollarizations in 2000 and2001, respectively, with enough U.S. dollarreserves to convert their monetary bases intoU.S. dollars.

1. Freeze the Zimbabwe dollar monetarybase and other liabilities of the RBZ. The RBZshould cease increasing the Zimbabwe dollarmonetary base (also known as reserve mon-ey—notes in circulation plus deposits offinancial institutions at the RBZ). However,it may continue to exchange new notes forworn-out old notes of equivalent value.

2. Abolish exchange controls and allow theZimbabwe dollar to float cleanly for a brief, pre-established period. (Note: This step can be tak-en simultaneously with the first step.) TheRBZ should set an appropriate fixed ex-change rate at which to convert Zimbabwedollar prices to rand prices. The best indicatoris the market rate that will evolve once peopleknow that the monetary base is frozen, thatthe value of the Zimbabwe dollar will soon befixed, and that the rand will then replace theZimbabwe dollar. The demand for Zimbabwedollars may well increase, in which case theexchange rate will appreciate. The floatshould be “clean,” that is, the governmentand the RBZ should not try to manipulate theexchange rate to achieve any particular level;they should let market participants deter-mine the level. Manipulating the exchangerate is costly. A highly overvalued exchangerate will price exports out of world marketsand may create a recession, while a highlyundervalued exchange rate will make importsexpensive and prolong inflation.

The exchange rate should float for a pre-established period that probably need notexceed 30 days.

3. At the end of the period of floating, declarea fixed exchange rate between the Zimbabwe dol-lar and the South African rand and announcethat, effective immediately, the rand is legal ten-

der alongside the frozen stock of Zimbabwe dol-lars. For example, declare that henceforth theexchange rate is one million Zimbabwe dollarsper rand, or some other rate determined to besuitable. The fixed rate should be somewherewithin the range of market rates obtained dur-ing the period of floating, particularly towardthe end of the period. Setting exchange rates isan art rather than a science, with no mechani-cal formula for making the transition from afloating rate to an appropriate fixed rate. If indoubt about the appropriate rate, it is better toerr on the side of an apparent slight underval-uation compared to recent market rates, so asnot to cause a slowdown in economic growth.Again, a large deliberate overvaluation orundervaluation is undesirable because it willrequire unnecessarily large economic adjust-ments after the exchange rate is fixed.

The rand will be declared the “domestic”currency, with all the legal tender rights theZimbabwe dollar has. All payments in Zim-babwe dollars may be made in rand at thefixed exchange rate. The government will con-tinue to accept Zimbabwe dollars at the fixedrate in payment of taxes until the dollars areretired from circulation. Zimbabwe dollarswill also continue to be legal tender for pay-ments between private parties unless theyspecify otherwise. While Zimbabwe dollarswill not be compulsory or forced tender in theprivate sector, the government may continuepaying government workers at least partly inZimbabwe dollars. By ensuring some demandfor the Zimbabwe dollar, these steps will helpensure that its market rate remains close to thefixed official rate.

4. After the government sets a fixedexchange rate between the Zimbabwe dollarand the South African rand, announce that,effective immediately, all Zimbabwe dollarassets and liabilities, with the exception of thecomponents of the monetary base, are randassets and liabilities at the fixed exchange rate.Announce a transition period of no more than90 days for replacing quotations of wages andprices in Zimbabwe dollars with quotations inrand. After the period of floating has endedand the exchange rate has been fixed, both

12

With the elimination of

hyperinflation,the rationale for

price controls andforeign exchange

controls nolonger exists.

Both should beprohibited.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 12

Page 13: Zimbabwe: Hyperinflation to Growth

Zimbabwe dollar bank deposits and loanswill be converted to deposits and loans inrand at the fixed rate. Banks will charge nocommission fees for the conversion.

During the transition period, wages cancontinue to be quoted optionally in Zimbabwedollars at the fixed rate so that employers andbanks have time to modify their bookkeepingand computer systems. Prices may also contin-ue to be quoted optionally in Zimbabwe dollarsduring the transition period, so as to spare mer-chants the trouble of repricing the goods ontheir shelves. After the transition period, wagesand prices will cease to be quoted in Zimbabwedollars. Again, though, Zimbabwe dollars willcontinue to be legal tender at the fixed rate forall payments within the economy unless theparties to a payment specify otherwise.

5. Retire the Zimbabwe dollar monetary basefrom circulation as government finances permit.As the Zimbabwean economy resumes growth,the government may periodically accept por-tions of the Zimbabwe dollar monetary base forcredit against tax liabilities or exchange themfor government bonds as its finances permit. Itmay take some years to retire the monetary basecompletely. As the economy grows, though, thesize of the frozen Zimbabwe dollar monetarybase in relation to the economy should quicklyshrink to manageable proportions.

6. Reorganize the RBZ. The RBZ will cease tobe an institution making monetary policy orissuing currency. Its assets and liabilities will betransferred to the government or to a commer-cial bank operating as a trustee for the govern-ment. Employees working on financial statistics,regulation of financial institutions, economicanalysis, and accounting may be transferred tothe Ministry of Finance and Economic Develop-ment, the Central Statistics Office, or a banksupervisory agency. Alternatively, the centralbank may be converted into a new independentauthority in charge of gathering and analyzingfinancial statistics and financial regulation.Considerable economies in staffing should beachievable, given how rapidly the RBZ’s staff hasgrown recently.

In the dollarized monetary systems ofEcuador and El Salvador, the vestigial “cen-

tral banks” continue to issue coins, but notnotes. One rationale for that practice is thatthese countries are somewhat far away fromthe United States, so shipping coins is costlybecause coins are bulkier than notes. IfZimbabwe uses the rand, no such argumentfor a local coinage will apply. That said, it isimportant to stress that for dollarized coun-tries (and countries with currency boards orfree banking), all vestiges of a central bankshould be eliminated to make it more diffi-cult for central banking to return throughthe back door. As the late Milton Friedmanput it when advocating unified currencies:dollarization and currency boards were goodideas for developing countries with “one veryimportant proviso, that they do not have acentral bank.”18

7. If the South African rand is the currencyused, request an agreement to share in the profitsfrom issuing the rand. As mentioned above,South Africa already has such agreementswith Lesotho and Namibia. If South Africa isreluctant to share the profits, however, Zim-babweans should not be too concerned. Thegains from ending hyperinflation and provid-ing a solid basis for renewed economic growthwill far outweigh the loss from not sharing inthe profits from issuing the rand.

The central banks that issue the U.S. dollarand the euro—the other currencies Zimbabwewould most likely adopt—do not offer profit-sharing agreements, so the issue would proba-bly be moot should Zimbabwe adopt a cur-rency other than the rand.

Performance of Recent Dollarized SystemsAs previously mentioned, 32 dependencies

and independent countries currently haveofficial dollarization. Among recent experi-ences with dollarized monetary systems, themost relevant to Zimbabwe are those ofPanama, El Salvador, and especially Ecuador—the three most populous, officially dollarizedcountries. All three use the U.S. dollar as theirofficial note currency (legal tender), althoughall issue their own coins.

Panama dollarized in 1904, shortly afterbecoming independent from Colombia. As a

13

The quickest way to replenishZimbabwe’s capital stock is to import capital.To accomplishthis, stable money and a liberal financialregime areimportant prerequisites.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 13

Page 14: Zimbabwe: Hyperinflation to Growth

unit of account, Panama uses the balboa,named after an early Spanish explorer. Onebalboa is equal to one U.S. dollar, so when aPanamanian store owner asks for five balboasfor a sack of rice, he is really asking for a U.S.five-dollar bill, which is legal tender in Panama.

Panama is the only Latin American coun-try whose currency has not depreciatedagainst the U.S. dollar in the last 20 years, letalone in the period since 1904. Panama’s infla-tion performance over the past two decadeshas actually been better than that of theUnited States, with an average annual infla-tion rate of 1.3 percent in Panama versus 3.1percent in the United States. Since 1970, whenPanama passed a law that opened its bankingsystem to extensive foreign participation, thePanamanian financial system has becomeintegrated into international capital marketsand Panama has developed into a highly effi-cient regional financial center.19 Growth hasbeen satisfactory, though not spectacular. Inthe last few years Panama has implementedgrowth-promoting policies, and growth hasaccelerated: the rate for 2007 was 11.2 percent.Panama seems to have a chance at becoming arich country in a generation, rather thanremaining a middle-income country.

Ecuador began dollarizing in January 2000as a way of ending accelerating inflation anddepreciation of the national currency, thesucre.20 At the time of dollarization, distrust ofthe sucre was becoming so great that someshopkeepers were beginning to quote prices inU.S. dollars and to prefer payment in dollarsto sucres. The currency had lost about two-thirds of its value against the U.S. dollar overthe previous 12 months, and consumer priceshad risen by 61 percent in that period.Dollarization brought immediate relief toEcuador’s economy: the exchange rate stabi-lized at 25,000 sucres per U.S. dollar, the leveldecreed by the president, and within a weekthe overnight interest rate fell from 200 per-cent to 20 percent a year before falling evenfurther shortly thereafter (see Table 3).

Ecuador’s Congress passed a law in March2000 that removed lingering doubts aboutwhether it would allow dollarization to per-

sist. The sucre, Ecuador’s former currency,became less and less important and wasretired in September 2000. The economyresumed growth by the fourth quarter of theyear and has continued growing. The bank-ing system was in distress when dollarizationoccurred—despite a government rescue onlymonths earlier—but it made a strong recov-ery as the economy revived. After an initialperiod in which inflation caught up to thedepreciation of the sucre that had occurredbefore dollarization, inflation fell to low sin-gle digits. (To conserve the central bank’s for-eign reserves, the president deliberatelydecreed a rate that he suspected to be a largeundervaluation. If the sucre-dollar rate hadbeen set according to the procedures recom-mended in this study, the rate would proba-bly have been lower and inflation would haveabated much more rapidly in Ecuador.)21

Dollarization has provided Ecuador with astable currency and a foundation for eco-nomic growth despite a highly unstable polit-ical system. Dollarization has so far lastedthrough six presidents and more than adozen ministers of finance. Ecuador’s currentgovernment favors the kind of populistsocialism that has prevented other LatinAmerican countries from growing rich, and itdislikes dollarization. Even so, the govern-ment recognizes that the economic stabilitydollarization has brought is popular, withover 80 percent of the public approving ofdollarization. Therefore, it has been reluctantto criticize dollarization directly.

El Salvador dollarized at the start of 2001after pegging the Salvador colón to the U.S.dollar for more than seven years. Its motive fordollarizing was to strengthen its integrationinto the international financial system. Eventhough the Salvador colón had a good long-term record by Latin American standards, dol-larization resulted in an almost immediatedrop in certain interest rates, particularly thosecharged by credit card issuers.22 Dollarizationmade little difference to El Salvador’s rate ofeconomic growth, which was sluggish becauseof natural disasters, crime, and policies that thegovernment has subsequently tried to address.

14

“Dollarization”would replace the

discreditedZimbabwe dollar

with a foreigncurrency.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 14

Page 15: Zimbabwe: Hyperinflation to Growth

However, dollarization has eliminated the con-cern that previously existed regarding currencydevaluation.

In no country has dollarization by itselfbeen all that was needed to foster sustained

economic growth. It has, however, providedreliability in one important area of economicpolicy. A good currency is not sufficient forgrowth, but a very bad currency is so destruc-tive that it can wreck the economy (see Table 4).

15

Table 3Dollarization in Ecuador (Fully Implemented September 13, 2000, %)

Year Real GDP per Capita CPI Inflation (End of Period)

1990 0.6 49.51991 2.7 49.01992 1.3 60.21993 -0.2 32.01994 2.5 25.41995 -0.4 22.81996 0.3 25.51997 2.0 30.71998 0.1 43.41999 -8.1 60.72000 0.9 91.02001 9.6 22.42002 0.1 9.42003 2.1 6.12004 6.5 1.92005 4.5 3.12006 1.4 2.92007 0.5 2.9

Source: International Monetary Fund, World Economic Outlook database, April 2008, http://www.imf.org/external/pubs/ft/weo/2008/01/weodata/index.aspx.Note: Dollarization period below double border.

Table 4Cumulative Changes in GDP per Capita and Inflation in Selected Dollarized Countriesand Zimbabwe (Percent)

Average AnnualGDP per Capita, GDP Per Capita Inflation, Average Annual

Country Period Total Change Change Total Change Inflation

Ecuador 2000–2007 27.0 3.5 58.1 6.8El Salvador 2001–2007 7.5 1.2 27.3 4.1Panama 2000–2007 31.6 4.0 15.8 2.1Zimbabwe 2000–2007 -31.6 -5.3 1,064,067,691.1 908.9

Source: International Monetary Fund, World Economic Outlook database, April 2008, http://www.imf.org/external/pubs/ft/weo/2008/01/weodata/index.aspx.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 15

Page 16: Zimbabwe: Hyperinflation to Growth

A Currency Board

How a Currency Board WorksAn orthodox currency board system relies

entirely on market forces to determine theamount of notes and coins that the currencyboard supplies. Market forces also determineother components of the supply of money inthe broad sense.

In a currency board system and in a centralbanking system alike, commercial banks areentrepreneurs of credit. A commercial bankcannot for long lend more to borrowers thanit can borrow from depositors (or wholesalecredit markets), in the form of deposits heldinstead of spent. If a commercial bank lendsexcessively, the borrowers spend the excess, forinstance, by writing checks. In the paymentssystem, more funds flow out of the bank thanflow into the bank. To prevent the outflowfrom bankrupting it, a commercial bankholds reserves. Commercial banks must main-tain sufficient reserves to enable depositors toconvert deposits into cash (or reserves) ondemand and to withstand outflows of reservesthrough the payments system.

An orthodox currency board has no activerole in determining the monetary base. A fixedexchange rate with the reserve currency and therequirement to hold foreign reserves equal to100 percent of the monetary base prevent thecurrency board from increasing or decreasingthe monetary base at its own discretion. Nordoes a typical currency board influence therelationship between the monetary base andthe money supply by imposing reserve ratios orotherwise regulating commercial banks.Regardless of the metric used, the money sup-ply in a typical currency board system, there-fore, is determined entirely by market forces. Atypical central bank, in contrast, can increaseor decrease the monetary base at its discretion.It can lend to commercial banks, creatingreserves for them, even if its foreign reserves aredecreasing. More reserves tend to enable com-mercial banks to make more loans, which theydo by creating deposits for borrowers. Themoney supply then increases. Decreasing the

monetary base tends to have the oppositeeffect. Besides changing the monetary base, atypical central bank can also influence the sup-ply of commercial bank loans by changing thereserve requirements for commercial banks.

Even though an orthodox currency boardcannot create reserves for commercial banks atits own discretion, the money supply in a typ-ical currency board system is quite elastic—responsive to changes in demand—becausethe system can acquire foreign reserves. Therules governing a currency board merely pre-vent it from creating reserves for commercialbanks in an inflationary manner, as a centralbank can. Other sources of elasticity in themoney supply are variability in commercialbanks’ ratio of reserves to deposits, the pool-ing of reserves among branches of commercialbanks in the currency board country and thereserve country, interbank lending, and vari-ability in the public’s deposit-to-cash ratio.23

How to Establish a Currency BoardEstablishing a currency board would require

steps that in many ways are similar to those forimplementing dollarization. A currency boardcould be established very quickly—within amonth of passing appropriate legislation.

1. Establish a currency board. The Appendixcontains a model currency board law.

2. Freeze the Zimbabwe dollar monetarybase and other liabilities of the RBZ. This stepis like its counterpart under establishing dol-larization.

3. Abolish exchange controls and allow theZimbabwe dollar to float cleanly. This step isalso like its counterpart under establishingdollarization.

4. At the end of the period of floating, declarea fixed exchange rate between the Zimbabwe dol-lar and the anchor currency chosen. Again, thisstep is like its counterpart under establishingdollarization.

5. Allow the Zimbabwe dollar monetary baseto increase only to the extent that the increase isbacked 100 percent by liquid foreign assets. Inshort, the currency board will only be allowedto increase the supply of Zimbabwe dollars bya dollar if it receives a Zimbabwe dollar’s

16

Full dollarizationoccurs when a

foreign currencyhas exclusive or

predominant status as full legal

tender.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 16

Page 17: Zimbabwe: Hyperinflation to Growth

worth of the anchor currency in exchange,converted at the fixed exchange rate.

Under this rule, the Zimbabwe currencyboard will operate “at the margin” with 100percent foreign reserves. It may or may nothave enough foreign reserves to fully back itstotal (both new and old) Zimbabwe dollarliabilities. To establish a credible, stable mon-ey commitment, the important point is thatnew monetary liabilities must be backed 100percent by liquid foreign assets.

Two currency boards—Argentina’s of a cen-tury ago and the East African Currency Boardof the first half of the 20th century—beganwith less than 100 percent foreign reserve back-ing of their total monetary liabilities, but bothrequired 100 percent backing of their newmonetary liabilities. In both cases, the currencyboards were credible; the demand for their cur-rencies was robust; and their foreign reservesaccumulated rapidly. In Zimbabwe, thedemand for the currency board’s moneyshould likewise increase rapidly as the econo-my stabilizes and begins to grow. Given the 100percent marginal reserve rule, the currencyboard’s foreign reserves will also grow rapidly.

The currency board should retain all prof-its from its operations until its foreign-curren-cy reserves equal 100 percent of the monetarybase. After that, it may be prudent to establisha reserve fund equal to a modest, specified per-centage of the board’s assets—say 10 percent(total reserves would then equal 110 percent ofthe monetary base), as was the case withZimbabwe’s previous currency board. Thereserve fund would guard against the possibil-ity of the currency board’s foreign-currencyreserves falling below 100 percent because of adefault on the foreign securities it holds. Thecurrency board would pay to the governmentall earnings other than those needed to oper-ate the currency board and its reserve fundand maintain the specified ratio of foreign-currency reserves to base money.

Additionally, to help increase foreign-cur-rency assets to 100 percent of the monetarybase, the government could repay the loans theRBZ has made to it or the RBZ could sell itsgovernment securities to the private sector.

Depending on the government’s financial con-dition, these actions might not be feasible forsome time. Yet another possibility would be aforeign loan, such as from the InternationalMonetary Fund or a foreign grant to bolsterthe board’s initial foreign reserves. Therewould be no need to delay establishing a cur-rency board while waiting for a foreign loan orgrant, though; it is better to establish monetarystability immediately.

6. Reorganize the RBZ. The RBZ will ceaseto be an institution making independentmonetary policy. Its financial activities willlikewise cease. The RBZ will transfer owner-ship of its remaining assets and liabilities toother banks or the Ministry of Finance andEconomic Development, as the case may be.The Ministry of Finance will become the solemanager of government debt, rather thanusing the services of the RBZ as it does now.

RBZ employees who now issue the mone-tary base and manage foreign-currency re-serves will become employees of the currencyboard. Employees who now deal with othermatters may be transferred to the Ministry ofFinance and Economic Development, theCentral Statistics Office, or a new superinten-dency of financial institutions.

How to Operate a Currency BoardA typical currency board is simple to oper-

ate. Past currency boards, such as Zimbabwe’s,have usually had staffs of 10 or fewer people.Operational details are addressed in otherpublications.24 Some noteworthy points arepresented below.

Constitution. The Appendix contains amodel currency board law that distills fea-tures of past currency board constitutionsinto a form that will both enable the curren-cy board to operate efficiently and insulate itfrom political interference.

Exchange policy, clientele, and fees. The cur-rency board should exchange its notes andcoins on demand at a fixed rate into (or from)the reserve currency at its offices or agencies.Although the currency board should encour-age a wholesale currency exchange businesswith commercial banks for the sake of effi-

17

Which currency shouldZimbabwe use?The obviouschoices are theU.S. dollar, theeuro, or theSouth Africanrand.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 17

Page 18: Zimbabwe: Hyperinflation to Growth

ciency, the public should also be allowed todeal directly with the board. In dealing withthe public, the currency board’s minimumshould be zero or low, such as 100 rand. Also,there should be no upper limit to the amountof the anchor currency or to its own notes andcoins in circulation that the currency boardaccepts for exchange. Some currency boardshave charged commission fees of one-eighthpercent to 1 percent per transaction. ForZimbabwe, the social benefits of not chargingfees exceed the pecuniary benefits to the cur-rency board of levying fees. In addition, com-mission fees would loosen the link to theanchor currency, especially for short-term cap-ital movements, because they would imposehigh costs relative to the benefits of arbitrage.Experience also indicates that most of the cur-rency board’s income is likely to come frominvested assets, not from fees. Exchanges bythe currency board should be exempt fromtaxation, to prevent the government fromattempting to tax the currency board out ofexistence. The currency board should also beexempt from other legal barriers that mighthinder its exchange operations.

Offices. The currency board should have itsmain office in Harare and perhaps a branchagency in Bulawayo, which it may operate orsubcontract to a bank. The currency boardshould also have an office abroad, in theanchor country or in a safe-haven financialcenter such as Switzerland. The office abroadshould provide a backup location for redeem-ing notes and coins in case the governmentthreatens to harass the domestic offices of thecurrency board.

Management. The currency board shouldhave a small board of directors—say, five peo-ple—to supervise the board’s staff. The powersof the board of directors and of the staffshould be limited. Unlike their counterpartsin central banks, they should have no discre-tionary control over the monetary base. Toprotect the board of directors from politicalpressure to convert the currency board into acentral bank, directors should serve staggeredterms. Directorship should be open to quali-fied foreigners as well as to Zimbabweans.

Composition of reserves. The currency boardshould hold its foreign reserves in low-riskassets payable in the anchor currency only.Most of its foreign reserves should be low-risk,interest-earning securities. It could also holdsome foreign reserves in interest-bearingdeposits at reputable commercial banks in theanchor country, or in anchor-currency notesor noninterest-earning deposits at the centralbank of the anchor country. As much as pos-sible, the currency board should avoid holdingassets that earn no interest. The currencyboard should hold no assets issued in Zimbab-we dollars or by Zimbabweans. Doing sowould both open the way to central banking-type operations and endanger the quality ofassets by making them domestic rather thanforeign.

In the past, many currency boards dividedtheir foreign reserves into a pool of short-term and a pool of higher-yielding longer-term securities. The size of the long-termpool was determined by estimates of the pub-lic’s minimum, “hard-core” demand for cur-rency board notes and coins.

Expenses and profits. Judging from theexperience of past currency boards, the boardshould earn a return of at least 4 percent ayear on its reserves and have operatingexpenses of no more than 1 percent of totalassets, perhaps even 0.5 percent. The largestexpense will be printing notes and mintingcoins. Salaries will probably be the nextlargest expense. Rent, utilities, and remainingcosts should be small.

The currency board’s profits equal theinterest earned on its foreign reserves minusits operating expenses. Typically, the profitsfrom a currency board are roughly 1 percentof GDP.

How to Protect the Currency BoardProtecting the currency board and the

notes and coins it issues from the governmentof Zimbabwe requires both a credible commit-ment and competition. The idea and impor-tance of credible commitments have beenwith us through the ages, with the most com-monly cited example being that of Ulysses in

18

Under a currency boardthe Zimbabwe

dollar would befully backed by a

foreign reservecurrency and

would be freelyconvertible at a

fixed rate ondemand.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 18

Page 19: Zimbabwe: Hyperinflation to Growth

the Odyssey, who commanded his men to tiehim to the mast of his ship so he would notjump overboard and drown when lured by thesweet but fatal song of the Sirens.

The currency board law is of utmostimportance. It must be designed so that thegovernment is bound by a credible commit-ment to protect the currency board. The mod-el currency board law in the appendix offersdetails for creating such a credible commit-ment. The model law ties the hands of the gov-ernment, making it difficult for the govern-ment to interfere with the smooth workings ofthe currency board.

One feature of the law, designed to reducethe possibility of government meddling, re-quires the currency board to incorporate inSwitzerland. In addition, the law mandates thata majority of the board of directors be foreignersappointed by the Bank for InternationalSettlements in Basel, Switzerland. That will helpprevent the government from bending or alter-ing the rules of the currency board. Precedentsexist for such an arrangement. For example,only three of the eight directors of the LibyanCurrency Board of the 1950s were Libyans; theothers were Britons, French, Italians, andEgyptians chosen by their respective govern-ments. Most recently, the Dayton-Paris peaceaccords that ended the civil war in Bosnia andHerzegovina in 1995 mandated that the newmonetary authority in Bosnia-Herzegovinaoperate as a currency board. It also mandatedthat the first governor of the board of directorsnot be a citizen of Bosnia-Herzegovina or anyneighboring state and that the governor beappointed by the International MonetaryFund.25 To reduce the political influence of thegovernment on the domestic directors of thecurrency board, only one of the Zimbabweandirectors will be appointed by the governmentof Zimbabwe, with the other appointed by theBankers Association of Zimbabwe.

Under the model law, the currency boardwill also gain credibility because its assets will beheld at the Bank for International Settlements.This feature will guarantee that the notes andcoins issued by the currency board can alwaysbe redeemed for the anchor currency at the

fixed rate in Switzerland, even if the govern-ment of Zimbabwe obstructed redemption inZimbabwe. This fall-back factor is an impor-tant feature of the currency board and is some-what analogous to the fall-back factor thatexisted under gold and silver standards whennotes were convertible into a commodity.

Yet another way to strengthen the credi-bility of a currency board is for its notes tocontain a legally binding statement that theyare convertible into the reserve currency at afixed rate at the board’s offices domesticallyand abroad. Whether or not notes and coinsissued by the currency board contain anexplicit statement of convertibility, they willbe considered a type of contract promising afixed exchange rate, unlike notes and coinsissued by a typical central bank. Holders ofnotes and coins will have the right to sue thecurrency board for breach of contract in thevery unlikely event that it fails to redeem itsnotes and coins at the fixed exchange rate ondemand. (As discussed in the next section,either the exchange rate or the anchor cur-rency can be legally changed in extraordinarycircumstances.)

To prevent the domestic governmentfrom seizing the printing presses and over-turning the currency board system by print-ing notes unbacked by foreign reserves, thenotes issued by the currency board will beprinted abroad.

The currency board will be subjected tocompetition to induce it to maintain high-quality service. To enhance currency compe-tition, people will be allowed to make con-tracts, payments, and deposits in anycurrency they wish. In addition, the anchorcurrency chosen for the currency board willbe granted joint legal tender status, alongwith the Zimbabwe dollars issued by the cur-rency board.

How to Change the Anchor Currency, IfNecessary

Changing the anchor currency is benefi-cial if the existing anchor currency becomesquite unstable, because otherwise the curren-cy board system will suffer the monetary

19

Currency boardshave been highlysuccessful inmaintainingfixed exchangerates with theiranchor currencies.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 19

Page 20: Zimbabwe: Hyperinflation to Growth

problems afflicting the anchor country.Currency competition—introduced by thefreedom to make contracts and payments inother currencies—also offers an escape fromthis potential problem.

If the currency board is given the power tochange the anchor currency, the currencyboard law must carefully specify the procedure.Also, a change in anchor currency should comefrom the currency board itself, rather thanbeing a somewhat arbitrary government deci-sion as has happened with past currencyboards.

The currency board should not be allowedto change the anchor currency unless theannualized change in the consumer priceindex of the anchor country exceeds the rangeof -2 percent to 20 percent for two consecutiveyears, or -5 percent to 40 percent for three con-secutive months. These rates of change in con-sumer prices have historically caused substan-tial economic disruption when exceeded. Ifdeflation or inflation in the anchor countryexceeds the specified range, the currency boardwill be allowed to offset the change, partially orfully, by devaluing or revaluing its currency interms of the anchor currency. The limit to theoffsetting change will be the amount of theinflation rate in the reserve country for theperiod just specified (two years or threemonths). Alternatively, the currency board willbe allowed to choose a new, more stable anchorcurrency and set a new fixed exchange rate atthe rate then prevailing between that currencyand the original anchor currency.

Allowing the currency board to reset theexchange rate or change anchor currenciesmay trigger speculative flows into or out ofthe Zimbabwe dollar if the anchor currencyapproaches the specified limits. No perfectsolution exists. Here, at last, is a place for dis-cretion: the board of directors will not berequired to alter the exchange rate; it willsimply have the power to do so, within limits.

Performance of Recent Currency Board-Like Systems

More than 70 countries have had curren-cy boards. As in Zimbabwe during its curren-

cy board period of 1940–56, currency boardselsewhere have been highly successful inmaintaining fixed exchange rates with theiranchor currencies. When compared to theircentral bank counterparts, they have alsobeen successful in promoting fiscal disci-pline, low inflation, economic growth, andfinancial integration within their regions orwith the world.26 Despite the economic suc-cess of currency board systems, national gov-ernments converted most currency boardsinto central banks in the late 1950s and1960s. Some governments were influencedby theoretical arguments claiming that a cen-tral bank could promote stability and eco-nomic growth better than a currency board.But political considerations carried the mostweight. Most newly independent countriesabandoned currency boards because of theirassociation with colonial rule. Moreover, old-er, more established countries had centralbanks. A central bank was seen as a symbol ofindependence, like a national flag.

Today, 13 currency boards or currencyboard-like systems still exist. Three are locatedin noted international financial centers: theCayman Islands, Bermuda, and Hong Kong.The Cayman system is orthodox, while Ber-muda’s operates with rather loose capital con-trols. In Hong Kong, since the early 1990s, thesystem has wavered between orthodoxy whenin trouble, as in the aftermath of the Asian cur-rency crisis of 1997–98, and deviations fromorthodoxy in less troubled times.27 The 10 oth-er systems are in Brunei, Bosnia-Herzegovina,Bulgaria, Djibouti, Estonia, Falkland Islands,Faroe Islands, Gibraltar, Lithuania, and St.Helena.

In the 1990s several countries reformedtheir central banks or established new mone-tary authorities, giving them some but not allof the characteristics of orthodox currencyboards. Countries that established such cur-rency board-like systems are Argentina (whosesystem lasted from 1991 to early 2002),Estonia (1992), Lithuania (1994), and Bulgaria(1997). Bosnia-Herzegovina (1997) estab-lished a system more orthodox than the oth-ers.28

20

Currency boardshave been

successful in promoting fiscal

discipline, low inflation,

economic growth,and financial

integration within their

regions or withthe world.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 2

Page 21: Zimbabwe: Hyperinflation to Growth

It is widely acknowledged that the euro-based currency board-like systems of Estonia,Lithuania, and Bulgaria have performed well.The same can be said for the more orthodoxeuro-based system in Bosnia-Herzegovina.Indeed, all those countries rapidly stabilized

their post-communist economies and promot-ed growth with positive confidence shocks (seeTable 5).

Bulgaria is noteworthy because it record-ed the last hyperinflation of the 20th centuryin 1997. The introduction of its currency

21

Table 5Cumulative Changes in GDP per Capita and Inflation in Selected Currency BoardCountries and Zimbabwe, 1997 2007 (Percent)

Country GDP per Capita, Total Change Inflation, Total Change

Bosnia 96 27Bulgaria 75 84Estonia 114 54Lithuania 99 26Zimbabwe -38 3,800,542,354

Sources: International Monetary Fund, World Economic Outlook database, April 2008, http://www.imf.org/external/pubs/ft/weo/2008/01/weodata/index.aspx; and author’s calculations; Note: Inflation data are calculated from end-of-year data except for Bosnia, where only average annual figures areavailable.

Table 6Currency Board-Like System in Bulgaria (Installed July 1, 1997, %)

Year Real GDP per Capita CPI Inflation (End of Period)

1990 -8.4 64.31991 -9.9 4027.81992 -7.5 79.41993 -10.7 63.81994 -2.7 121.91995 -0.7 32.91996 -7.3 310.81997 -4.9 549.21998 4.7 1.71999 2.8 7.02000 6.2 10.02001 4.8 4.82002 5.2 3.82003 5.7 5.62004 7.3 4.02005 6.9 6.52006 7.0 6.52007 6.9 11.6

Sources: International Monetary Fund, World Economic Outlook database, April 2008, http://www.imf.org/external/pubs/ft/weo/2008/01/weodata/index.aspx; and author’s calculations.Note: Currency board-like period below double border.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 21

Page 22: Zimbabwe: Hyperinflation to Growth

board-like system on July 1, 1997, immedi-ately put an end to that painful episode (seeTable 6).

Currency board-like systems differ fromorthodox currency boards with respect to theirreserve ratios and their power to act as lendersof last resort. None of the currency board-likesystems have a maximum reserve ratio. If anorthodox currency board is allowed to accumu-late foreign reserves in excess of 100 percent ofthe monetary base, the amount of the surplushas a definite upper limit, which historically hasbeen 10 percent. Today, Bosnia-Herzegovinahas such a 10 percent upper limit.29 The pur-pose of the surplus reserves is to guarantee thatreserves are always at least 100 percent by pro-viding a cushion against losses in the securitiesin which the currency board invests. An ortho-dox currency board is not allowed to use thesurplus in a discretionary manner, and all prof-its beyond those necessary to maintain thesmall surplus go to the government. Most cur-rency board-like systems, in contrast, areallowed to accumulate profits (surplus reserves)unchecked, though in practice there is politicalpressure to contribute some reserves to the gen-eral government budget. Currency board-likesystems are also allowed to use their surplusreserves in a discretionary manner to act aslenders of last resort to commercial banks. Insome cases, they can also use their main reservesin the same manner.

These loopholes can be fairly harmless ingood economic times but can cause great mis-chief in bad times. Argentina’s currency board-like system, which looked like a great successuntil 1999, ended in a spectacular economiccrash and currency depreciation in January2002. Figure 3 shows how unorthodoxArgentina’s system was. For much of its life,the system operated like some central banks,not like a currency board. For an orthodox cur-rency board, net foreign reserves (foreign assetsminus foreign liabilities) should be close to 100percent of the monetary base. Moreover,“reserve pass-though,” defined as the change inthe monetary base divided by the change in netforeign reserves over the period in question,should also be close to 100 percent.

Argentina pegged its currency at oneArgentine peso per U.S. dollar. A reserve pass-through ratio of 100 percent means that if netforeign reserves rise (or fall) by, say, US$100million, the Argentine peso monetary baseshould also rise (or fall) by 100 million pesos.That was rarely the case in Argentina, but fewcommentators have appreciated how unortho-dox the system was.30

In Figure 3, reserve pass-through above 100percent means that the monetary base and netforeign reserves changed in the same direction,but the monetary base changed more than netforeign reserves. One might call this the “zoneof magnified foreign reserve effects.” Reservepass-through between 0 percent and 100 per-cent means that the monetary base and net for-eign reserves changed in the same direction, butthe monetary base changed less than net for-eign reserves. One might call this the “zone ofordinary sterilization.” In this zone, not all ofthe change in foreign reserves is allowed to flowthrough to the monetary base. Some of thechange in foreign reserves is sterilized by eithersales or purchases of bills or bonds denominat-ed in Argentine pesos. Reserve pass-throughbelow 0 percent means that the monetary baseand net foreign reserves changed in oppositedirections. One might call this the “zone ofsuper sterilization.” Both at 100 percent and at0 percent, no sterilization occurs. An orthodoxcurrency board with a fixed exchange rate hasreserve pass-through close to 100 percent. Themonetary base changes passively in response tochanges in the public’s desired holdings of basemoney, which occur through exchanging thenotes, coins, or deposits of the currency boardfor the anchor currency and vice versa at thefixed exchange rate the board maintains.(Reserve pass-through may not be exactly 100percent because of accounting and valuationissues concerning the board’s accumulation ofprofits and payment of expenses.) A centralbank with a clean floating exchange rate has areserve pass-through often close to 0 percent,because it rarely has reason to buy or sell its cur-rency for foreign reserves. However, if the cen-tral bank also holds foreign reserves for govern-ment accounts that are active but not related to

22

The monetarysovereignty

argument is asmokescreen to

conceal the RBZ’swretched

performance.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 22

Page 23: Zimbabwe: Hyperinflation to Growth

monetary policy, reserve pass-through mayoften be far from 0 percent.

Free Banking

Free banking is a system of competitiveissue of notes and other liabilities by privatecommercial banks with minimal regulation. Acompletely free banking system has no centralbank, no lender of last resort, no reserverequirements, and no legal restrictions onbank portfolios, interest rates, or branch bank-ing. Commercial banks are only restricted intheir issuance of deposits and currency by con-tracts and market practices, not by legislation.Despite the unfamiliarity of free banking tomost people today, it has deep historical rootsand a record of working well in a wide range ofcountries, including Zimbabwe.31

How Free Banking WorksAlthough free banking is unfamiliar, the

principles of competition that underlie it arenot. Indeed, they are already at work in depositbanking. People do not usually think of

deposits from different banks as being differ-ent types of currency, but in effect they are—atleast, they are different brands of a commonunit of account. By holding a deposit at onebank rather than others, a depositor is choos-ing that bank’s management, portfolio, andservices over those of its competitors.

Free banking extends competition fromdeposits to notes. In practice, multiple brandsof notes have generally not created problemsfor free banking systems any more than multi-ple brands of deposits create problems in cen-tral banking or other systems. (The issue ofcoins by banks has been much rarer thanissuance of notes, because governments havetraditionally monopolized coinage even wherethey have not monopolized note issue. Butmultiple brands of coinage present no specialproblems, particularly in a country such asZimbabwe where inflation has driven coinsout of circulation and hence no vendingmachines use coins.)

In a system of free banking, the field ofcompetitors includes banks, domestic or for-eign, which meet the requirements commonto other businesses: registering a place of busi-

23

The effect of dollarization or acurrency board isnot to create a colonial relationship, butto achieve morecredibility than alocal central bankcan.

-100

0

100

200

300

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Net Foreign Reserves (% of reserve money)Reserve Pass-Through (year over year)Line of Orthodoxy (100%)

Figure 3Was Argentina Orthodox?

Sources: International Monetary Fund, International Financial Statistics, database and printed volumes (Washington:International Monetary Fund, August 2007); and author’s calculations.

Perc

ent

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 23

Page 24: Zimbabwe: Hyperinflation to Growth

ness, stating who the officers are, listing theshareholders periodically, and publishingfinancial statements if the company is not aprivate partnership. Competition weeds outfirms that are less astute at delivering whatconsumers want. Abundant experience indi-cates that depositors want assurance that theyare placing their funds with a financially solidbank. Accordingly, the tendency almost every-where has been for a few large but highly com-petitive banks to dominate the market, whileleaving niches for small banks to serve special-ized clienteles. As with deposit banking, then,historical experience suggests that eventuallyor perhaps even immediately, note issuancewill be dominated by a small number of largebanks.

Under free banking, banks would have theliberty to issue deposits and circulate notes inany currency: the U.S. dollar, the euro, theSouth African rand, gold, etc. In past freebanking systems, issuance has converged on asingle unit of account: typically gold or a for-eign currency. In Zimbabwe, quite possibly,issuers would converge on the South Africanrand. However, free banking leaves it forbanks and customers to discover what worksbest for them; it does not presume that thegovernment already knows the answers.

How to Establish Free BankingThe steps for establishing free banking are

presented below. The first three can be imple-mented simultaneously.

1. Freeze the Zimbabwe dollar monetary baseand other liabilities of the RBZ. This step is likeits counterparts for the establishment of dollar-ization or a currency board. Accordingly, thereis nothing to add in the case of free banking.

2. Abolish exchange controls and allow theZimbabwe dollar to float cleanly. This step islike its counterparts under the establishmentof dollarization or a currency board.

3. Remove all barriers to entry into the busi-ness of banking, including restrictions againstissuance of notes and coins. The main effect ofthis step will be to remove reserve require-ments and minimum capital requirements.But because banking is a business where rep-

utation is of utmost importance, removingbarriers to entry will not necessarily result in arush of new entrants. A consolidation into afew large “financial supermarkets” may evenoccur as financial institutions become able tooffer services in lines of business where regu-lations formerly blocked their competitiveentry.

A fully free banking system has no reserverequirements. Currently, required reservesare 10 to 40 percent of deposits in Zimbabwe,depending on the kind of financial institu-tion and the kind of deposit. It may be desir-able to convert some or all required reservesinto freely tradable government bonds. Thiswould ensure that eliminating the currenthigh reserve requirements would not gener-ate a one-time inflationary surge as a legacyof central banking.

4. At the end of the period of floating, declarea fixed exchange rate between the Zimbabwe dol-lar and a suitable foreign currency. The pur-pose of this step is to give RBZ currency a reli-able value during its remaining time incirculation. This step is much like its counter-parts under the establishment of dollarizationor a currency board. But there are some differ-ences, which are explained below.

5. After the government sets a fixed exchangerate, announce that, effective immediately, allZimbabwe dollar assets and liabilities with theexception of the components of the monetarybase are foreign-currency assets and liabilities atthe fixed exchange rate. Under the dollarizationor currency board options, nothing moreshould happen with respect to the redenomi-nation of assets and liabilities because the gov-ernment’s setting of a fixed rate provides astrong “pole of attraction,” influencing mosttransactions to be denominated in the samecurrency as long as it is reliable. In a free bank-ing system, however, people may wish to rede-nominate again. For example, if the govern-ment chooses the U.S. dollar as the currencywith which it sets a fixed rate, but the marketpreference is for most assets and liabilities tobe denominated in South African rand, euros,or some mixture of other currencies, peoplewill convert into those currencies.

24

Less flexiblemonetary regimes

have generallyhad much lessinflation thanmore flexible

regimes.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 24

Page 25: Zimbabwe: Hyperinflation to Growth

As under dollarization or a currencyboard, the government will accept RBZ notesin payment for taxes, and RBZ notes will belegal tender but not compulsory tender fortransactions in the private sector.

6. Retire the Zimbabwe dollar monetary basefrom circulation as government finances permit.This step is similar to its counterpart underthe dollarization option. However, even beforethe government begins retiring the Zimbabwedollar monetary base, banks may decide thatto promote their own note issues, they willaccept RBZ notes in exchange for their ownnotes. In past free banking systems, solventbanks sometimes accepted at face value thenotes of banks that had failed, giving note-holders their own notes in exchange and hop-ing thereby to generate goodwill and stimu-late a demand for their own notes.

7. Reorganize the RBZ. As under the dol-larization or currency board options, theRBZ will cease to be an institution makingmonetary policy or issuing currency. Thereorganization will be similar to that underthe dollarization option. However, under freebanking, little special financial regulationwill be needed. Financial institutions will besubject to general corporate law but will notface minimum required reserves or variousother regulations that do not exist for firmsin other lines of business. Accordingly, theresidual functions of the RBZ will be limitedto the collection and analysis of economicand financial data. These functions will betransferred to the Ministry of Finance andEconomic Development or the CentralStatistics Office, as appropriate.

Objections to Replacing theReserve Bank of Zimbabwe

Dollarization, a currency board, or freebanking would perform better than theReserve Bank of Zimbabwe. To investigatewhether these options have disadvantagescompared to allowing the RBZ to continue toexist, this section considers the main objec-tions to all three options. Because the options

have many similarities, the objections to themalso have many similarities.

Much of what is written about centralbanking and its challenges is by central bankstaffs or external researchers funded by centralbanks. Moreover, it is published in journalssponsored and financed by central banks. Astudy of the U.S. Federal Reserve System con-firmed these conclusions.32 In light of theextremely poor performance of the RBZ,Zimbabweans should be skeptical of objec-tions to replacing it that originate from staff ofthe RBZ. They should also be skeptical ofobjections that originate from sources thathave a strong institutional bias in favor of cen-tral banking. The International MonetaryFund, for example, has often advised againstreplacing central banks even in countrieswhere their performance has been poor. TheIMF was established at a time when econo-mists and policymakers thought every countryshould have a central bank, and the institu-tional bias present at its creation has persistedfor more than 60 years, despite the many caseswhere central banks (often established orreformed with IMF help) have created highinflation, inconvertibility, and other problems.

Loss of SovereigntyThe most frequent objection to dollariza-

tion or a currency board is that they wouldreduce Zimbabwe’s sovereignty by deprivingZimbabwe of an independent monetary poli-cy. However, dollarization or a currency boardby itself would create no colonial subjugation.Panama, Ecuador, and El Salvador are no lesssovereign for being dollarized; Estonia, Lithu-ania, Bulgaria, and Bosnia-Herzegovina are noless sovereign for having currency board orcurrency board-like systems. Consider alsoMontenegro. It became an independent sover-eign state in 2006 largely because it replacedthe Yugoslav dinar with the German mark(now the euro) in 1999.33 Using the euroenabled Montenegro to break loose from thebad economic policies undertaken at the timeby Serbia, its much larger partner in theFederal Republic of Yugoslavia. The effect ofdollarization or a currency board is not to cre-

25

An importantsource of stabilityfor banks operating underdollarization,currency boards,and free bankingalike has been theability to borrowabroad.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 25

Page 26: Zimbabwe: Hyperinflation to Growth

ate a colonial relationship, but to achieve morecredibility than a local central bank can. Themonetary sovereignty argument is a smoke-screen to conceal the RBZ’s wretched perfor-mance. Under a currency board, which somemight term “colonialist,” the Zimbabwepound maintained a fixed exchange rate withits anchor currency, the pound sterling. Underthe RBZ, the Zimbabwe dollar has becomeworthless.

A Step Backwards Another objection to the three alternatives

to central banking which have been evaluatedis that drastically reorganizing or abolishingthe RBZ would be a step backwards in the evo-lution of Zimbabwe’s monetary regime. Zim-babwe has expended much effort building itsown central bank, and it would be a shame towaste that effort, the argument goes.

The real step backwards was to establishthe RBZ. Had Zimbabwe continued with acurrency board, or returned to free banking,it would have had lower inflation and highereconomic growth than it has had with theRBZ. The options for monetary reform thatthis study presents would be a step forwardbecause they would provide a stable currency,which Zimbabwe now lacks.

No FlexibilityAnother frequent objection to all three

options is that they would deprive the gov-ernment of the flexibility to make changes inthe supply of money to offset external orinternal shocks to the economy, such as aplunge in the price of tobacco.

Consider the record of flexible monetarypolicy in Zimbabwe. Whatever the govern-ment has wanted, the RBZ has given. Zim-babwe had better economic performancewhen monetary policy was less flexible.Indeed, less flexible monetary regimes havegenerally had much less inflation than moreflexible regimes.34 The former chairman ofthe U.S. Federal Reserve System, Paul Volcker,has remarked that “if the overriding objectiveis price stability, we did better with the nine-teenth-century gold standard and passive

central banks, with currency boards, or evenwith ‘free banking’.”35 Putting some back-bone into Zimbabwe’s monetary structureswould reduce or eliminate the possibility ofthe government using flexible monetary poli-cy as a way of mismanaging the economy.

No Lender of Last ResortA closely related objection is that monetary

systems lacking a central bank would be sus-ceptible to financial panics because they wouldhave no lender of last resort to financial institu-tions in danger of failing. Lack of a central bankas a lender of last resort does not seem to haveharmed countries with dollarization or curren-cy boards. Zimbabwe’s financial system wasadmirably stable before central banking began.An important source of stability for banks oper-ating under dollarization, currency boards, andfree banking alike has been the ability to borrowabroad. Because dollarization and currencyboards eliminate risk with the anchor currency,they facilitate access to international financialmarkets for banks that are temporarily short ofcash but fundamentally sound.

If Zimbabweans are worried about the pos-sibility of a financial panic, they might consid-er a deposit insurance scheme. The schemeshould be private and voluntary. Switzerland,Germany, and other countries have privatedeposit insurance schemes that could serve asmodels. Insurance should cover at most, say,80 percent of the value of large deposits, sothat depositors have an incentive to avoidimprudently managed banks that pay unsus-tainably high interest rates.

Another way to reduce the risk of financialpanics would be for commercial banks toinclude a “notice of withdrawal clause”(option clause) in their contracts with deposi-tors. The notice of withdrawal clause wouldallow a commercial bank to delay for a set peri-od the requests of depositors to convertdeposits into notes and coins. In return, thebank would pay a penalty rate of interest; forexample, 3 percent above the rate prevailingbefore it exercised the notice of withdrawalclause. Banks would be free to offer a notice ofwithdrawal clause, and depositors would be

26

The strongestform of legal

protection wouldbe to make the

monetary reformpart of the

constitution.

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 26

Page 27: Zimbabwe: Hyperinflation to Growth

free to do business with such banks. Notices ofwithdrawal clauses have precedents; for exam-ple, they were widespread among savingsbanks in the United States until the 1970s.

Zimbabwe Is So Big, It Must Have ItsOwn Central Bank

The International Monetary Fund’s WorldEconomic Outlook database of April 2008 reportsthat Zimbabwe’s economy is much smaller thanthe economies of Bosnia-Herzegovina, Bulgaria,Ecuador, El Salvador, Estonia, Lithuania, andPanama. If those countries can prosper withouttypical central banks, so can Zimbabwe.

IneffectivenessYet another objection is that the options

for monetary reform that this study propos-es would not restrain the government’sdeficit spending.

The possibility that the governmentmight try to undermine any of the optionsfor monetary reform emphasizes the need forstrong legal protection—a credible commit-ment—to insulate the monetary reform frompolitical pressure. The strongest form of legalprotection would be to make the monetaryreform part of the constitution. But in mostrecent cases, the kind of far-reaching mone-tary reforms discussed here have been estab-lished by statute, not by national constitu-tions. Even so, most countries that have beendollarized or established currency boards orfree banking have had smaller budget deficitsthan comparable countries with centralbanks; and they have not used regulatorytricks to force financial markets to hold theirdebt.36 Instead they have had much lowerdeficit spending (as a percentage of GDP)than Zimbabwe; and when they have issued

27

Far-reachingmonetary reformis possible andcan be politicallypopular.

Table 7Comparison of Options for Monetary Reform

Benefits Compared to Current System Costs Compared to Current System

Dollarization No discretionary monetary policy;* no Loss of seigniorage from notes and ability to finance government deficits;* coins (except perhaps if rand is used)no possibility of devaluation; much lower inflation; positive, but low and stable, real interest rates

Currency board No discretionary monetary policy;* no ability to finance government deficits;*much smaller possibility of devaluation; much lower inflation; positive, but low and stable, real interest rates

Free banking No discretionary monetary policy;* no Seigniorage accrues to private ability to finance government deficits;* banks rather than to governmentmuch smaller possibility of devaluation; much lower inflation; positive, but low and stable, real interest rates; most effective system for promoting financial development because it imposes the fewest restrictions on banks

*Some argue that these features represent costs rather than benefits. The evidence from developing countries does notsupport that conclusion. The ability of central banks to engage in discretionary monetary policy and finance govern-ment deficits is associated with relatively high rates of inflation, relatively large fiscal deficits, and relatively slow GDPgrowth rates (Steve H. Hanke, “Currency Boards,” Annals of the American Academy of Political and Social Science579 (January 2002): pp. 87–105).

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 27

Page 28: Zimbabwe: Hyperinflation to Growth

debt, financial markets have been willing tohold it voluntarily. Any of the three optionspresented here would be a stronger barrieragainst deficit spending than the RBZ is now.Deficit spending would still be possible, butit would be more difficult without a centralbank; and when it occurred, it would be morevisible to public scrutiny.

Final Observations

Since 1998, Zimbabwe’s economy has beenon a road to ruin. Per capita GDP has con-tracted in each year; inflation has surged; realinterest rates have been negative; the value ofthe Zimbabwe dollar has been wiped out; per-sonal savings have been destroyed; the nation’scapital stock has shrunk; and businesses’ work-ing capital has been decimated. This state ofeconomic ruin has forced millions of Zimbab-weans to lose hope for Zim-babwe’s future andto emigrate.

To restore economic growth, hyperinfla-tion must be extinguished rapidly, and peo-ple must have confidence that inflation willnot return. Otherwise, unstable expectationsand uncertainty will arise and undermine thestabilization program and economic growth.

Confidence can best be restored if Zimbab-we’s central banking system is replaced withone of the three options presented in thisstudy: dollarization, a currency board system,or free banking. None of these optionsrequires preconditions prior to its implemen-tation, and any one of them would establishstability and restore economic growth. As KarlSchiller, a former German Minister of Eco-nomic Affairs, put it: “Stability is not every-thing, but without stability, everything isnothing.”37

In Table 7, I briefly review the options formonetary reform, considering their benefitsand costs compared to the current system,the RBZ.

Other countries have overcome politicalobstacles that protected unsatisfactory mone-tary regimes and have made far-reachingreforms like those this study has evaluated.

Far-reaching monetary reform is possible andcan be politically popular.

Appendix: Model Statutes

Dollarization Law1. The Reserve Bank of Zimbabwe shall

cease to issue Zimbabwe dollars except asreplacements for equal amounts of old cur-rency that become worn out.

2. Except as specified in paragraph 3,wages, prices, assets, and liabilities shall beconverted from Zimbabwe dollars to [nameof currency] (“the replacement currency”) atthe conversion rate chosen in the law thataccompanies this law. By 60 days after thislaw enters into force, wages and prices shallcease to be quoted in Zimbabwe dollars.

3a. Interest rates shall be converted intothe replacement currency by the followingprocedure. The independent committee ofexperts specified in the law accompanyingthis law shall choose benchmark interestrates in the Zimbabwe dollar and replace-ment currency, having similar characteristicswith respect to maturity and liquidity insofaras possible. The ratio between existing inter-est rates in Zimbabwe dollars and the bench-mark interest rate in the Zimbabwe dollarshall determine the interest rate in thereplacement currency, which shall bear thesame ratio to the benchmark rate in thereplacement currency.

3b. In no case, however, shall new interestrates in the replacement currency resultingfrom the conversion procedure exceed 50 per-cent a year.

4. The president may appoint a commit-tee of experts on technical issues connectedwith this law to recommend changes in regu-lations that may be necessary.

5. Nothing in this law shall prevent partiesto a transaction from using any currency that ismutually agreeable. However, the replacementcurrency may be established as the default cur-rency where no other currency is specified.

6. While Zimbabwe dollars remain in cir-culation, the government shall accept them

28

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 28

Page 29: Zimbabwe: Hyperinflation to Growth

in payment of taxes at no premium to theconversion rate with the replacement curren-cy. Acceptance of Zimbabwe dollars shall notbe obligatory for any other party.

7. Within five years after this law takes effect,the government shall redeem all outstandingZimbabwe dollars for the replacement currencyor exchange it for government debt bearing amarket-determined rate of interest.

8. Existing laws that conflict with this laware void.

9. This law takes effect immediately uponpublication.

Currency Board Law1. The Zimbabwe Currency Board (“the

Board”) is hereby created. The purpose of theBoard is to issue notes and coins in newZimbabwe dollars, and to hold foreign reservessufficient to maintain them fully convertible ata fixed exchange rate into an anchor currencyspecified in paragraph 6.

2. The Board shall have its legal seat inSwitzerland and shall be subject to the lawsof Switzerland.

3a. The Board shall be governed by five direc-tors. Three directors shall be citizens of coun-tries other than Zimbabwe, appointed by theBank for International Settlements in Basel,Switzerland. Two directors shall be Zimbab-wean, one appointed by the Government ofZimbabwe and one by the Bankers Associationof Zimbabwe. The directors appointed by theBIS shall not be employees of governments ormultigovernmental organizations.

3b. A quorum shall consist of three of theBoard’s directors, including the director cho-sen by the Government of Zimbabwe or thedirector chosen by the Bankers Association ofZimbabwe. Decisions shall be made by major-ity vote, except as specified in paragraph 15.

3c. The first director appointed by theGovernment of Zimbabwe shall serve a termof one year. The first director appointed bythe Zimbabwe Bankers Association shallserve a term of four years. The first threedirectors appointed by the BIS shall serveterms of two, three, and five years. Subse-quent directors shall serve terms of five years.

Directors may be reappointed once.3d. Should a director resign or die, the

organization that selected that director shallselect a replacement to serve the remainingterm.

4. The board of directors shall have thepower to hire and fire the Board’s staff, andto determine salaries for the staff. The bylawsof the Board shall determine salaries for thedirectors.

5. The Board shall issue notes and coinsdenominated in new Zimbabwe dollars. Thenotes and coins shall be fully convertible intothe anchor currency. The notes shall be print-ed outside Zimbabwe.

6a. Initially, the anchor currency shall be[name of currency], and the fixed exchange rateshall be determined 30 days after the promul-gation of this law is announced. The proce-dures for determining the fixed exchange rateare contained in a separate law that accompa-nies this law. The fixed exchange rate so deter-mined will be used as the fixed exchange ratefor the duration of the currency board arrange-ment, subject to changes of the anchor curren-cy in accordance with paragraph 13.

6b. Failure to maintain the fixed exchangerate with the anchor currency shall make theBoard and its directors subject to legal actionfor breach of contract according to the lawsof Switzerland. This provision does not applyto attempts to redeem embezzled, mutilated,or counterfeited notes, coins, and deposits.Nor does this provision apply to changes ofthe anchor currency in accordance with para-graph 13.

7. The Board shall charge no commissionfor exchanging new Zimbabwe dollars for theanchor currency.

8a. The Board may assume the monetaryliabilities and corresponding assets of theReserve Bank of Zimbabwe. The Board neednot initially hold foreign reserves against thisstock of monetary liabilities if it has inherit-ed no corresponding readily saleable assetsfrom the Reserve Bank of Zimbabwe. Overtime it shall dispose of any domestic assetsinitially held as counterparts to its monetaryliabilities.

29

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 29

Page 30: Zimbabwe: Hyperinflation to Growth

8b. The Board may not increase its mone-tary liabilities without foreign reserves equalto 100 percent of the amount of the increase.

8c. The Board shall hold its foreign reservesin highly rated and liquid securities, or otherforms payable only in the anchor currency.These reserves shall be on deposit at the BIS.The Board shall not hold securities issued bythe national or local governments of Zimbab-we, or by enterprises owned by those govern-ments. The reserves of the Board are the prop-erty of the holders of the Board’s monetaryliabilities and may not be appropriated by theGovernment of Zimbabwe.

9. The Board shall pay all net seigniorage(profits) into a reserve fund until its unbor-rowed foreign reserves equal 110 percent ofits notes and coins in circulation anddeposits. It shall remit to the Government ofZimbabwe all net seigniorage beyond thatnecessary to maintain 110 percent foreignreserves. The distribution of net seigniorageshall occur annually.

10. The head office of the Board shall bein Harare. The Board shall establish a branchin Bulawayo and may establish branches orappoint agents in other cities of Zimbabwe.The Board shall also maintain a branch inSwitzerland.

11. The Board shall publish a financialstatement, attested by the directors, monthlyor more often on a publicly accessible Internetsite. The statement shall appraise the Board’sholdings of securities at their market value. Anannual audit of the Board shall be made by aninternational audit firm and shall be pub-lished by the Board.

12. The Board may issue notes and coinsin such denominations as it judges to beappropriate.

13. The Board may take the followingsteps to alleviate disruptions to the purchas-ing power of its currency resulting fromdeflation or inflation in the anchor currency.

(a) If the year-over-year change in the con-sumer price index of the country issuingthe anchor currency is below -2 percentfor two consecutive years, or below -5percent for three consecutive months,

the Board may devalue its currencyagainst the anchor currency by no morethan the cumulative amount of defla-tion since the consumer price index ofthe country issuing the anchor currencybegan to decline.

(b) If the year-over-year change in the con-sumer price index of the country issu-ing the anchor currency exceeds 20 per-cent for two consecutive years, or 40percent for three consecutive months,the Board may revalue its currencyagainst the anchor currency by no morethan the difference between the changein the consumer price index of thecountry issuing the anchor currencyand the rate of 20 percent or 40 percentthat triggers this clause.

(c) Alternatively, in the cases of deflationor inflation as defined above, theBoard may choose a new anchor cur-rency and fix the exchange rate of theCurrency Board currency to the newcurrency at the rate then prevailingbetween the new anchor currency andthe former anchor currency.

(d) The Board may only take these stepswithin 60 days of the consumer pricestatistics being reported in the countryissuing the anchor currency.

14. If the Board chooses a new anchor cur-rency in accordance with paragraph 13, itmust convert all its foreign reserves intoassets payable in the new anchor currencywithin two years.

15. The Board may not be dissolved, normay its assets be transferred to a successororganization, unless all of the following con-ditions are satisfied: two-thirds of the mem-bers of the Parliament of Zimbabwe approve,the President of Zimbabwe approves, all thedirectors of the Board approve, and all claimsagainst its monetary liabilities can be satisfied.

16. The Board may accept loans or grantsof reserves from multigovernmental organiza-tions or foreign governments to establish ini-tial foreign reserve backing of up to 100 per-cent of the monetary base. The loans shall notexceed 100 percent of the monetary base. After

30

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 3

Page 31: Zimbabwe: Hyperinflation to Growth

establishing the initial backing, the Board maynot accept loans.

17. Exchanges of currency by the Boardshall be exempt from taxation by the govern-ment of Zimbabwe and all its subdivisions.

18. Old Zimbabwe dollars, new Zimbabwedollars, and the anchor currency shall belegal tender for paying taxes and settlingdebts in Zimbabwe. However, private partiesshall be free to contract among themselves inany currencies they wish to specify, and nocurrency shall be forced tender for such con-tracts.

19. The Board may not perform bankingservices for the Government of Zimbabwe,and it shall not be responsible for the finan-cial obligations of the government.

20. Existing laws that conflict with thislaw are void.

21. This law takes effect immediatelyupon publication.

Law for Determining the Rate ofExchange between the Zimbabwe Dollarand the [name of anchor currency forcurrency board or replacement currencyfor dollarization]

1. This law shall take effect upon the pro-mulgation of the Zimbabwe Currency BoardLaw or the Zimbabwean Dollarization Law.

2. After the promulgation of the ZimbabweCurrency Board Law or the Zimbabwe Dollari-zation Law, the Zimbabwe dollar shall beallowed to trade for 30 days on the open mar-ket without intervention from or restrictionby the Government of Zimbabwe. The pricesand quantities of Zimbabwe dollars tradedshall be tabulated by dealers in foreign curren-cy for the duration of the 30-day period.

3. An independent accounting firm willuse foreign currency dealers’ tabulated trad-ing records to calculate the weighted averageof the Zimbabwe dollar-[name of currency]exchange rate at the end of each day in the30-day trading period, and for the overallperiod. The results of these calculations willbe taken into consideration by an indepen-dent committee of experts to determine thefixed exchange rate. The rate will be chosen

to best represent the fair market value of theZimbabwe dollar, facilitate economic calcula-tion, and allow for rapid implementation ofthe Zimbabwe Currency Board Law or theZimbabwe Dollarization Law.

4. The appropriate international crosscurrency rates shall be used to convert trans-actions in currencies other than the [name ofcurrency] into [name of currency] terms.

5. The fixed exchange rate determinedaccording to the procedures above will be usedas the fixed exchange rate in the ZimbabweCurrency Board Law or the conversion rate inthe Zimbabwe Dollarization Law.

Free Banking Law1. Any person may issue circulating notes

and coins denominated in any unit ofaccount.

2. All restrictions on the holding, use inpayment, and quotation of prices in curren-cies other than the Zimbabwe dollar are abol-ished.

3. Existing regulations that impose mini-mum reserve requirements and other special-ized regulatory burdens and taxes on bank-ing institutions that do not apply to otherindustries are abolished.

4. [Name of currency] is hereby declaredlegal tender for payment of all debts inZimbabwe alongside the Zimbabwe dollar ata fixed exchange rate of [number] Zimbabwedollars per unit of [name of currency].

5. All interest rates in Zimbabwe dollarsagreed to before this law takes effect shall beconverted according to the following proce-dure:

(a) The government, in consultation withthe Bankers Association of Zimbabwe,shall select or calculate a benchmark,market-determined interest rate inZimbabwe dollars and an analogousbenchmark interest rate of similarmaturity in [name of currency].

(b) Parties to a contract requiring paymentof interest in Zimbabwe dollars shallcalculate the ratio between the interestrate they have specified and the bench-mark interest rate in Zimbabwe dollars.

31

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 31

Page 32: Zimbabwe: Hyperinflation to Growth

(c) The new interest rate, which shall bethe rate for making all future pay-ments during the life of the contract,shall equal the benchmark interest ratein [name of currency] multiplied by theratio determined in the precedingparagraph.

(d) In no case, however, shall the new annu-alized interest rate exceed 50 percent.

NotesI would like to thank John Legat and JohnRobertson for their encouragement and support inthe preparation of this study, my research assis-tants Andrew Burgess and Alex Kwok, and mylong-time collaborator Kurt Schuler for commentson the manuscript. An earlier version of this man-uscript was published in Harare, Zimbabwe, by theNew Zanj Publishing House and was sponsored byImara Holdings Limited.

1. Ludwig von Mises, The Theory of Money and Credit(Irvington-on-Hudson, New York: Foundation forEconomic Education, 1971 [1912]), p. 414.

2. Peter Bernholz, Monetary Regimes and Inflation:History, Economic and Political Relationships (North-ampton, Massachusetts: Edward Elgar, 2003).

3. Robert Pringle, ed., The Morgan Stanley CentralBank Directory 2008 (London: Central BankingPublications, 2008).

4. I have relied on the following works for informa-tion concerning the monetary history of Zimbabwe:Bank of Rhodesia and Nyasaland, 1956/1961–1964annual reports; Salisbury: Bank of Rhodesia andNyasaland (the first annual report covers the period1956–1961); Patrick Bond, Uneven Zimbabwe: A Studyof Finance, Development, and Underdevelopment(Trenton, New Jersey: Africa World Press, 1996); SirRobert Chalmers, A History of Currency in the BritishColonies (London: Eyre and Spottiswoode for HerMajesty’s Stationery Office, 1893); Sir Julian StanleyCrossley and John Blandford, The DCO Story: AHistory of Banking in Many Countries 1925–27 (London:Barclays Bank International, 1975); James A. Henryand H. A. Siepmann, The First Hundred Years of theStandard Bank (London: Oxford University Press;Reserve Bank of Zimbabwe, 1983, The Currency Mediaof Southern Rhodesia (from the Time of the Charter to1953), of the Federation of Rhodesia and Nyasaland (from1954 to 1963), of Rhodesia (from 1964 to 1979), and ofZimbabwe (from 1980) (Harare: Reserve Bank ofZimbabwe); Reserve Bank of Zimbabwe, FinalAccounts (Harare, Zimbabwe, 1993); Reserve Bank ofZimbabwe, Annual Report (Harare: Reserve Bank of

Zimbabwe 1994-present (although the Reserve Bankof Rhodesia became the Reserve Bank of Zimbabwein 1980, the successor central bank did not publishits first annual report until 1994); Reserve Bank ofZimbabwe, Quarterly Economic and Statistical Review(later called Monetary and Economic Developments)(Harare: Reserve Bank of Zimbabwe, 1980–present);Reserve Bank of Zimbabwe, Selected Economic Indi-cators (Harare: Reserve Bank of Zimbabwe, monthly),http://www.rbz.co.zw; Reserve Bank of Zimbabwe,Weekly Economic Highlights (Harare: Reserve Bank ofZimbabwe), http:// www.rbz.co.zw; Rhodesia andNyasaland (Federation), Central African CurrencyBoard, Annual Report, 1953/1954–1955/1956. First[Second, Third] Report of the Central African CurrencyBoard (Salisbury: Central African Currency Board)(printed by Government Printing and StationeryOffice); Southern Rhodesia Currency Board: annualreport, 1939/1940–1952/ 1953, First Report of theSouthern Rhodesia Currency Board, Covering the PeriodEnded 31 March, 1940 (1939–1940); Second [etc.] Reportof the Southern Rhodesia Currency Board, Year Ended 31March, 1941 [etc.] (1940/1941–1952/1953); the 7th,9th, 10th, and 11th reports have the word “annual”in their titles, but the rest do not. Salisbury: Govern-ment Stationery Office; Rafik Ahmed Sowelem,Towards Financial Independence in a DevelopingEconomy: An Analysis of the Monetary Experience of theFederation of Rhodesia and Nyasaland, 1952–63(London: Allen and Unwin, 1967).

5. Benjamin Cohen, The Future of Money (Princeton,New Jersey: Princeton University Press, 2004); PaulR. Masson and Catherine Pattillo, The MonetaryGeography of Africa (Washington: Brookings Institu-tion, 2005).

6. Jian-Ye Wang, Iyabo Masha, Kazuko Shirono,and Leighton Harris, “The Common MonetaryArea in Southern Africa: Shocks, Adjustment, andPolicy Challenges,” International Monetary FundWorking Paper no. 158 (July 2007), http://www.imf.org/external/pubs/ft/wp/2007/wp07158.pdf.

7. Kevin Dowd, ed., The Experience of Free Banking(London: Routledge, 1992).

8. Kurt Schuler, Should Developing Countries HaveCentral Banks? (London: Institute of EconomicAffairs, 1996).

9. Lawrence H. White, “In What Respects Will theInformation Age Make Central Banks Obsolete?”Cato Journal 21, no. 2 (Fall 2001): 127–40, http://www.cato.org/pubs/journal/cj21n2/cj21n2-7.pdf.

10. Alan A. Walters and Steve H. Hanke, “CurrencyBoards,” in, The New Palgrave Dictionary of Money andFinance, ed. Peter Newman, Murray Milgate, andJohn Eatwell (London: Macmillan, 1992).

32

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 32

Page 33: Zimbabwe: Hyperinflation to Growth

11. Steve H. Hanke, “On Dollarization andCurrency Boards: Error and Deception,” Journal ofPolicy Reform 5, no. 4 (2002): 203–22.

12. Maxwell J. Fry, Money, Interest, and Banking inEconomic Development (Baltimore, Maryland: JohnsHopkins University Press, 1988).

13. Sonia Muñoz, “Central Bank Quasi-FiscalLosses and High Inflation in Zimbabwe: A Note,”International Monetary Fund Working Paper no.98 (April 2007), http://www.imf.org/external/pubs/ft/wp/2007/wp0798.pdf.

14. Reserve Bank of Zimbabwe, http://www.rbz.co.zw/about/vision.asp.

15. Richard D. Porter and Ruth A. Judson, “TheLocation of U.S. Currency: How Much Is Abroad?”Federal Reserve Bulletin 82 (October 1996): 883–903,http://www.federalreserve.gov/pubs/bulletin/1996/1096lead.pdf.

16. Dominick Salvatore, James W. Dean, andThomas D. Willett, eds., The Dollarization Debate(Oxford and New York: Oxford University Press,2003).

17. Steve H. Hanke, “Some Reflections on CurrencyBoards,” in Central Banking, Monetary Policies, and theImplications for Transition Economies, ed. Mario I.Blejer and Marko Skreb (Norwell, Massachusetts:Kluwer Academic Publishers, 1999); Steve H.Hanke, “Currency Boards,” Annals of the AmericanAcademy of Political and Social Science 579 (January2002): pp. 87–105.

18. Milton Friedman, “Do We Need CentralBanks?” Central Banking 5, no. 1 (1994): 55–58.

19. Juan Luis Moreno-Villalaz, “The MonetaryExperience of Panama: A Dollar Economy withFinancial Integration.” Cato Journal 18, no. 3 (Winter1999): 421–39; Juan Luis Moreno-Villalaz, “FinancialIntegration and Dollarization: The Case of Panama,”Cato Journal 25, no. 1 (Winter 2005): 127–40.

20. Steve H. Hanke, “Money and the Rule of Law inEcuador,” Journal of Policy Reform 6, no. 3 (2003):131–45.

21. Kurt Schuler, El futuro de la dolarizacion en Ecuador(Guayaquil: Instituto Ecuatoriano de EconomiaPolitica, 2002): 10–11.

22. Manuel Hinds, Playing Monopoly with the Devil:Dollarization and Domestic Currencies in DevelopingCountries (New Haven, Connecticut: Yale Univers-ity Press, 2006).

23. Steve H. Hanke and Kurt Schuler, Currency

Boards for Developing Countries: A Handbook (SanFrancisco: ICS Press [Institute for ContemporaryStudies], 1994). A more recent electronic versionis available on Kurt Schuler’s website, http://www.dollarization.org.

24. Ibid.

25. Warren Coats, One Currency for Bosnia (Ottawa,Illinois: Jameson Books, 2007).

26. Steve H. Hanke, “Currency Boards,” pp. 87–105.

27. John Greenwood, Hong Kong’s Link to the U.S.Dollar: Origins and Evolution (Hong Kong UniversityPress, 2007); Tony Latter, Hong Kong’s Money: TheHistory, Logic and Operation of the Currency Peg (HongKong University Press, 2007).

28. Steve H. Hanke, “Some Reflections on CurrencyBoards”; Steve H. Hanke,“Currency Boards,” pp.87–105; Steve H. Hanke, “On Dollarization andCurrency Boards: Error and Deception,” pp. 203–22.

29. Coats, One Currency for Bosnia.

30. Steve H. Hanke, “On Dollarization and CurrencyBoards: Error and Deception,” pp. 203–22; KurtSchuler, “Ignorance and Influence: U.S. Economistson Argentina’s Depres-sion of 1998–2000,” EconJournal Watch 2, no. 2 (August 2005): 234–78, http://www.econjournalwatch.org/pdf/CompleteIssueAugust2005.pdf, viewed June 23, 2008.

31. Dowd, The Experience of Free Banking; George A.Selgin, The Theory of Free Banking: Money SupplyUnder Competitive Note Issue (Totowa, New Jersey:Rowman and Littlefield, 1988); George A. Selginand Lawrence H. White, “How Would the InvisibleHand Handle Money?” Journal of Economic Literature32, no. 4 (December, 1994): 1718–49.

32. Lawrence H. White, “The Federal ReserveSystem’s Influence on Research in Monetary Eco-nomics,” Econ Journal Watch 2, no. 2 (August 2005):325–54, http://www.econjournalwatch.org/pdf/–InvestigatingAugust2005.pdf.

33. Steve H. Hanke, “Hyperinflation,” Forbes (June4, 2007), p. 200.

34. Steve H. Hanke. “Currency Boards,” pp. 87–105.

35. Paul Volcker, “Foreword,” in The Central Banks,ed. Marjorie Deane and Robert Pringle (NewYork: Viking Penguin, 1995), p. vii–viii.

36. Steve H. Hanke. “Currency Boards,” pp. 87–105.

37. David Marsh, The Bundesbank: The Bank that RulesEurope (London: Mandarin Paperbacks, 1992): 30.

33

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 33

Page 34: Zimbabwe: Hyperinflation to Growth

STUDIES IN THE POLICY ANALYSIS SERIES

617. Roadmap to Gridlock: The Failure of Long-Range Metropolitan Transportation Planning by Randal O’Toole (May 27, 2008)

616. Dismal Science: The Shortcomings of U.S. School Choice Research andHow to Address Them by John Merrifield (April 16, 2008)

615. Does Rail Transit Save Energy or Reduce Greenhouse Gas Emissions? by Randal O’Toole (April 14, 2008)

614. Organ Sales and Moral Travails: Lessons from the Living Kidney Vendor Program in Iran by Benjamin E. Hippen (March 20, 2008)

613. The Grass Is Not Always Greener: A Look at National Health Care Systems Around the World by Michael Tanner (March 18, 2008)

612. Electronic Employment Eligibility Verification: Franz Kafka’s Solution to Illegal Immigration by Jim Harper (March 5, 2008)

611. Parting with Illusions: Developing a Realistic Approach to Relations with Russia by Nikolas Gvosdev (February 29, 2008)

610. Learning the Right Lessons from Iraq by Benjamin H. Friedman, Harvey M. Sapolsky, and Christopher Preble (February 13, 2008)

609. What to Do about Climate Change by Indur M. Goklany (February 5, 2008)

608. Cracks in the Foundation: NATO’s New Troubles by Stanley Kober (January 15, 2008)

607. The Connection between Wage Growth and Social Security’s FinancialCondition by Jagadeesh Gokhale (December 10, 2007)

606. The Planning Tax: The Case against Regional Growth-Management Planning by Randal O’Toole (December 6, 2007)

605. The Public Education Tax Credit by Adam B. Schaeffer (December 5, 2007)

604. A Gift of Life Deserves Compensation: How to Increase Living KidneyDonation with Realistic Incentives by Arthur J. Matas (November 7, 2007)

603. What Can the United States Learn from the Nordic Model? by Daniel J. Mitchell (November 5, 2007)

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 34

Page 35: Zimbabwe: Hyperinflation to Growth

602. Do You Know the Way to L.A.? San Jose Shows How to Turn an UrbanArea into Los Angeles in Three Stressful Decades by Randal O’Toole (October 17, 2007)

601. The Freedom to Spend Your Own Money on Medical Care: A Common Casualty of Universal Coverage by Kent Masterson Brown (October 15, 2007)

600. Taiwan’s Defense Budget: How Taipei’s Free Riding Risks War by Justin Logan and Ted Galen Carpenter (September 13, 2007)

599. End It, Don’t Mend It: What to Do with No Child Left Behind by Neal McCluskey and Andrew J. Coulson (September 5, 2007)

598. Don’t Increase Federal Gasoline Taxes—Abolish Them by Jerry Taylor and Peter Van Doren (August 7, 2007)

597. Medicaid’s Soaring Cost: Time to Step on the Brakes by Jagadeesh Gokhale (July 19, 2007)

596. Debunking Portland: The City That Doesn’t Work by Randal O’Toole (July 9, 2007)

595. The Massachusetts Health Plan: The Good, the Bad, and the Ugly by David A. Hyman (June 28, 2007)

594. The Myth of the Rational Voter: Why Democracies Choose Bad Policiesby Bryan Caplan (May 29, 2007)

593. Federal Aid to the States: Historical Cause of Government Growth and Bureaucracy by Chris Edwards (May 22, 2007)

592. The Corporate Welfare State: How the Federal Government Subsidizes U.S. Businesses by Stephen Slivinski (May 14, 2007)

591. The Perfect Firestorm: Bringing Forest Service Wildfire Costs under Control by Randal O’Toole (April 30, 2007)

590. In Pursuit of Happiness Research: Is It Reliable? What Does It Imply for Policy? by Will Wilkinson (April 11, 2007)

589. Energy Alarmism: The Myths That Make Americans Worry about Oil by Eugene Gholz and Daryl G. Press (April 5, 2007)

588. Escaping the Trap: Why the United States Must Leave Iraq by Ted Galen Carpenter (February 14, 2007)

35354 _DPA6_1stclass.qxp 6/25/2 8 4:18 PM Page 35

Page 36: Zimbabwe: Hyperinflation to Growth

ANNE APPLEBAUMWASHINGTON POST

GURCHARAN DASFORMER CEO, PROCTER

& GAMBLE, INDIA

ARNOLD HARBERGERUNIVERSITY OF CALIFORNIA

AT LOS ANGELES

FRED HUGOLDMAN SACHS, ASIA

PEDRO-PABLO KUCZYNSKIFORMER PRIME MINISTER OF PERU

DEEPAK LALUNIVERSITY OF CALIFORNIA

AT LOS ANGELES

JOSÉ PIÑERAFORMER MINISTER OF LABOR AND

SOCIAL SECURITY, CHILE

T he Center for Global Liberty and Prosperity was established to promotea better understanding around the world of the benefits of market-lib-eral solutions to some of the most pressing problems faced by develop-

ing nations. In particular, the center seeks to advance policies that protect humanrights, extend the range of personal choice, and support the central role of eco-nomic freedom in ending poverty. Scholars in the center address a range of economic development issues, including economic growth, international finan-cial crises, the informal economy, policy reform, the effectiveness of official aid agencies, public pension privatization, the transition from socialism to the mar-ket, and globalization.

For more information on the Center for Global Liberty and Prosperity, visit www.cato.org/economicliberty/.

Nothing in this Development Policy Analysis should be construed as necessarily reflecting the views of the

Center for Global Liberty and Prosperity or the Cato Institute or as an attempt to aid or hinder the passage of

any bill before Congress. Contact the Cato Institute for reprint permission. Additional copies of Development

Policy Analysis are $6 each ($3 for five or more). To order, contact the Cato Institute, 1000 Massachusetts

Avenue, N.W., Washington, DC, 20001, (202) 842-0200, fax (202) 842-3490, www.cato.org.

OTHER STUDIES ON DEVELOPMENT FROM THE CATO INSTITUTE

BOARD OF ADVISERS

“A Decade of Suffering in Zimbabwe: Economic Collapse and Political Repression under RobertMugabe” by David Coltart, Development Policy Analysis no. 5 (March 24, 2008)

“Fifteen Years of Transformation in the Post-Communist World: Rapid Reformers OutperformedGradualists” by Oleh Havrylyshyn, Development Policy Analysis no. 4 (November 9, 2007)

“Troubling Signs for South African Democracy under the ANC” by Marian L. Tupy, DevelopmentPolicy Briefing Paper no. 3 (April 25, 2007)

“Kenya’s Fight against Corruption: An Uneven Path to Political Accountability” by John Githongo,Development Policy Briefing Paper no. 2 (March 15, 2007)

“A Second Look at Microfinance: The Sequence of Growth and Credit in Economic History” byThomas Dichter, Development Policy Briefing Paper no. 1 (February 15, 2007)

“Corruption, Mismanagement, and Abuse of Power in Hugo Chávez’s Venezuela” by GustavoCoronel, Development Policy Analysis no. 2 (November 27, 2006)

“The Rise of Populist Parties in Central Europe: Big Government, Corruption, and the Threat toLiberalism” by Marian L. Tupy, Development Policy Analysis no. 1 (November 8, 2006)

“Foreign Aid and the Weakening of Democratic Accountability in Uganda” by Andrew Mwenda,Foreign Policy Briefing no. 88 (July 12, 2006)