The author would like to thank Messrs. Gerrit Bilkes, Warren Coats, John Dalton, Thomas Ferguson, Omotunde Johnson, Jang-Yung Lee, and Bill Melbourn, and Ms. Micheline Lefebvre-Manthorp for their helpful comments and suggestions. Any remaining mistakes are, naturally, his own.
Russia’s demonetization of pre-1993 ruble bank notes in July 1993 forced seven of the countries of the former Soviet Union--Azerbaijan, Armenia, Belarus, Kazakhstan, Tajikistan, Turkmenistan, and Uzbekistan--to suddenly face a situation where their domestic currency had been demonetized by its issuer. This led these countries to scramble to produce their own currencies. In its efforts to help these countries, IMF staff discovered the absence of a coherent literature on this subject.
A paper by Abrams and Cortes-Douglas (1993) attempted to address other aspects of introducing a new currency by examining the policy, institutional, and technical issues associated with the process of introducing a new currency.
Many often favor producing a national currency that serves as an elegant national “calling card,” but this is likely to be an expensive mistake, particularly if inflation is not yet fully under control.
Possible themes are a person, an object, an animal, an event, or a subject that the citizenry find to be appropriate.
Disputes over subject matter, such as the exclusion or under-representation of women and/or indigenous minorities, have at times given rise to significant delays in the release of new currency issues.
Normally either the central bank, the Government, the Parliament or a certain minister is legally empowered to choose the design of the national currency.
Allowing different themes on each denomination of bank note also makes it easier to agree to give a discouraged group consideration when the next bank note is introduced.
For example, the design of Lithuania’s bank notes was influenced by Finnish notes.
For a useful discussion of the color combinations, including complementary colors and colors wheels, see van Renesse (1993), pp. 45-49.
For example, since 1904, all 10-guilder notes issued in the Netherlands have been blue.
The signatures are generally printed in letterpress under them.
Van Renesse (1993) covers many aspects of bank note security in much greater detail than in paper, notably pp. 111-126.
Although vending machines probably do not warrant consideration in the short term in many countries.
Public concern about counterfeit notes varies markedly between countries, and, therefore, so does the appropriate breakdown between security features desired by the general public and those desired by currency handlers.
In the intaglio printing process, thick paste inks on deeply engraved printing plates are put under high pressure to transfer the ink from the plate onto the paper. This results in raised print, and a potentially great range of tonal variations. Intaglio printing can also be used to produce latent/transitory images, usually numbers or letters, that appear/disappear when the note is viewed at an angle. These features cannot be reproduced using planar printing processes such as lithography and letterpress. For details, see van Renesse, pp. 112-113 and 116-117.
Watermarks may be continuous or registered (local). Both can use light and dark parts (mold-made) or only light parts (Fourdrinier). They are among the best security features because they are easily recognized by the general public. For more details, see van Renesse, pp. 96-98.
Photocopying machines are not designed to perfectly align both sides of a document, so these patterns are extremely difficult to counterfeit on such machines.
Plastic strips are usually marked with a repeating pattern and are visible only when held to light. Metal or foil strips, perhaps plated with cellophane, may be used to make electrical conductivity an additional test of authenticity. Iridescent strips that produce a single noniridescent color when copied with a photocopier have recently gained in popularity.
Serial numbers are particularly useful because groups of counterfeit bank notes often share the same serial number.
For a detailed discussion of OVDs, see van Renesse, pp. 207-225.
High-pressure printing processes, such as intaglio, compress the paper fibers, strengthening them and slowing the unraveling of the fibers.
Payne and Morgan (1981), pp. 45, 47, estimate the ideal value of the smallest currency unit is between one two-thousandth and one five-thousandth of the average day’s pay.
Currencies are often redenominated when the smallest unit of currency in use becomes “excessively small.” This is normally done by renaming the currency unit (and subunit) and defining it as some multiple of the old currency, 10, 100, etc.
Later on, decisions will be needed on when to change the denominations of coins and bank notes, including the denomination of the smallest bank note and the largest coin. For a discussion of such dynamic currency issues, see Payne and Morgan (1981).
In part, because coins are generally easier to counterfeit than quality bank notes.
The profit or seigniorage from issuing currency may, for the purposes of this note, be defined as the current value of the currency at issuance (what it can purchase, plus the discounted cost of the foregone interest from not having to borrow to make the purchase) less the discounted cost of keeping the currency in circulation. For coins, which are long-lived, seigniorage is mainly value less production cost. For bank notes, seigniorage includes the cost of keeping them in circulation, which includes the costs of destroying old notes and printing new ones.
For example, this may be why the two-dollar bill was unpopular in the United States.
For example, see Payne (1980). Others have suggested using a ratio to per capita GDP to take account of nonwage income.
Since it is often more convenient, and safe, to use noncash payments instruments for larger payments, such as mortgage or rent payments or purchases of durables.
While binary denominations (powers of 2) minimize the number of notes/coins exchanged in transactions, decimal denominations are a close approximation and are generally better understood and easier to work with.
The public generally prefer bank notes to coins, so they will wish the smallest note to be of a relatively low value. However, lower-value bank notes tend to wear out quickly--low-denomination notes often wear out in as few as six to eight months--so the long-term cost of issuing them can be high relative to the value. Thus, the central bank and the Government usually prefer the largest-value coin to be relatively high in value.
For a discussion of how these estimates might be made, see Bilkes and Bennett (1993), appendix IV.
For studies on the demand for currency in a particular country, see Fase (1981) and Cramer (1983) for the Netherlands, Kimball (1981) for the United States, Manski and Goldin (1982) for Israel, and Payne and Morgan (1981) for the United Kingdom. A planning model for central banks is presented in Fase, van der Hoeven, and van Nieuwkerk (1979).
Another benefit of producing bank notes locally is the employment it provides.
The demand for new bank notes includes bank notes issued both to accommodate changes in the public’s demand and to replace bank notes.
Estimates of the minimum efficient scale for producing bank notes ranged from a population of around 20 million to a demand of 250 million bank notes annually. Estimates of the cost of setting up such an operation were US$35-45 million.
Printing contracts tend to be large and long because this causes average costs to fall sharply owing to high fixed costs, such as note design and setting up the printing machines.
The contract terms must be clear and detailed in these areas. Since the contract is generally with a foreign company, it may be difficult and time-consuming to retrieve the printing plates and obtain the required materials if the contract is ambiguous on these points.
Assuming the number of bank notes printed is large enough to bring the cost of production close to the minimum average cost.
However, such notes wear out quickly and are easily forged.
For example, in 1986, Finland’s effective publicity campaign for its new note issue included distributing brochures to every home in Finland, with small brochures made available in banks and post offices. Heinonen (1986), p.2.
Sometimes one country demands that the other cease using its currency. While such a demand is difficult to enforce, the country issuing the currency can refuse to supply the other country with bank notes.
The estimated number of bank notes needed to meet demand in an “average” country of about eight million people.
Otherwise, the stamped notes must be gradually exchanged for unstamped notes. During the interim, parallel circulation of stamped and unstamped bank notes must be allowed to avoid the risk of a bank note shortage.
Once again, the process will be slowed if the inventory of bank notes is not at least equal to the stock of notes that is to be replaced.
The cost figures in this section are estimated in the broadest of magnitudes.
If 500 high-speed machines are bought to produce 200 million stamps and then they are scrapped, the cost of using the machines to affix the stamps would be on the order of US$75/thousand, plus labor.
Overprinting to change the value of a note will be riskier if there is a stock of circulating notes that have not been overprinted.
Preferably from an older series that has been demonetized or retired, so that old notes are in short supply.