Journal Issue
Share
Article

Budgetary Implications of a Devaluation

Author(s):
International Monetary Fund. External Relations Dept.
Published Date:
March 1973
Share
  • ShareShare
Show Summary Details

The author describes a method for estimating the budgetary impact of a devaluation and identifies the chief factors to be considered.

A devaluation is in certain circumstances the best policy alternative for a country beset by persistent balance of payments deficits. However, it is sometimes avoided because governments feel that it may cause economic or political difficulties more serious than those it is intended to cure. On the basis of various misconceptions, it is argued, inter alia, that a reform of the exchange system would increase the budget deficit. Perhaps the most important misunderstanding of this type arises when the authorities overlook the fact that the central bank’s profits on its foreign exchange operations are, or could be, captured by the government. For this reason, I am treating central banks as part of a “public sector” which should be understood as also including the government proper as well as dependent government agencies and enterprises.

Balance of Payments Projections

Any attempt to appraise the budgetary implications of an adjustment in the exchange rate should start with projections of balance of payments developments both without and with such an adjustment. An example of the required forecasts is given in Table 1. According to column one, there would be a total deficit of $10 million if no devaluation were carried out. Column two assumes that, as a result of the reform envisaged by the authorities, private sector exports and other foreign exchange receipts would increase from $90 million to $100 million. Imports and other payments would decline by $10 million to $70 million, since foreign goods and services would become more expensive to the domestic buyer. (This is usually but not always the case. If a devaluation is accompanied by a liberalization of import restrictions, it does not necessarily lead to an increase in import prices.)

Table 1.Balance of Payments Without and With Devaluation 1(In millions of U. S. dollars)
Without

Devaluation
With

Devaluation
A. Private sector
1. Receipts90100
2. Payments−80−70
3. Surplus or deficit 21030
B. Government
1. Receipts2020
2. Payments−40−40
3. Surplus or deficit 2−20−20
C. Total
1. Receipts110120
2. Payments−120−110
3. Surplus or deficit 2−1010

Excluding foreign economic assistance and imports financed with such assistance. Table 3, however, assumes an amount of $10 million of foreign (nonproject) aid.

Minus sign denotes a deficit.

Excluding foreign economic assistance and imports financed with such assistance. Table 3, however, assumes an amount of $10 million of foreign (nonproject) aid.

Minus sign denotes a deficit.

It is assumed, moreover, that the foreign exchange receipts and payments of the government would remain unchanged at $20 million and $40 million, respectively; such receipts are probably not responsive to adjustments in the exchange rate, while government payments (e.g., defense imports and debt service) should presumably not be increased at a time when, owing to the unfavorable balance of payments situation, the authorities are obliged to adopt the drastic, belt-tightening measure of devaluation. Thus, according to the example in Table 1, the government’s deficit on its receipts and payments in foreign exchange would be $20 million, as it would have been without a reform. Overall, however, the external sector would register a surplus of $10 million. In the interest of simplification, a devaluation is thus assumed to have an immediate impact on the balance of payments position; in fact, it may take a considerable period of time to achieve the substantial improvement reflected in this example.

Basic Budgetary Effect

The terms “devaluation,” “reform of the exchange system,” and “adjustment of the exchange rate” are used synonymously in this article.

“Profits on foreign exchange operations” refers to profits from current purchases and sales of foreign exchange, not to those obtained by revaluing foreign exchange reserves at the post-devaluation rate. The former have a real impact on the economy (they are tantamount to a tax on the private sector, if derived from transactions with that sector), while the latter represent entries of a strictly bookeeping nature and are therefore ignored in this article.

Now suppose that the devaluation whose hypothetical results have just been described raises the price of foreign exchange from US$1 = 5 domestic currency units (DCUs) to US$1 = DCU 10; on this assumption, one can calculate its basic effect on public sector finances. As indicated in Table 2, A and B, the cost of the government’s $20 million deficit on its receipts and payments in foreign exchange will increase from (5 × 20=) DCU 100 million to (10 × 20=) DCU 200 million. Will the central bank realize any profits or incur any losses on its purchases and sales of foreign exchange as a result of the reform? The answer is clearly no. Without a reform the bank would be selling on a net basis foreign exchange in an amount equal to $10 million and acquiring instead DCUs amounting to 50 million; it would show on its books an exchange of two different assets of an equal value at the prereform exchange rate. With a devaluation there would be an exchange of assets of equal value at the postdevaluation rate; the bank would be purchasing foreign exchange of $10 million for DCU 100 million. Thus the total impact of the reform will be to raise the deficit of the public sector as a whole from DCU 100 million to DCU 200 million.

Table 2.Basic Budgetary Effect

(In millions of DCUs) 1

A. Without devaluation (US$1 = DCU 5 for all transactions)
1. Government deficit on its receipts and payments in foreign
exchange (5 × −20 =)−100
2. Central bank purchases and sales of foreign exchange
a. Receipts from net sales of foreign exchange (5 × 10 =)50
b. Cost of exchange sold at official rate (5 × 10 =)50
c. Profit or loss 2 on exchange operations (c = a−b)0
3. Public sector deficit 2 (3 = 1+2c)−100
B. With devaluation (US$1 = DCU 10 for all transactions)
1. Government deficit on its receipts and payments in foreign
exchange (10 × −20 =)−200
2. Central bank purchases and sales of foreign exchange
a. Expenditures on net purchases of foreign exchange
(10 × 10 =)100
b. Cost of exchange purchased at official rate (10 × 10 =)100
c. Profit or loss 2 on exchange operations (c = b−a)0
3. Public sector deficit 2 (3 = 1-2c)−200
C. With devaluation (US$1 = DCU 10 for private sector and US$1
= DCU 5 for government transactions)
1. Government deficit on its receipts and payments in foreign
exchange (5 × −20 =)−100
2. Central bank purchases and sales of foreign exchange
a. Expenditures on net purchases of foreign exchange
(10 × 30 − 5 × 20 =)200
b. Cost of exchange purchased at official rate (10 × 10 =)100
c. Profit or loss 2 on exchange operations (c = b−a)−100
3. Public sector deficit 2 (3 = 1−2c)−200

Domestic currency units.

Minus sign denotes loss or deficit.

Domestic currency units.

Minus sign denotes loss or deficit.

Special Rate for Government Transactions

Suppose that the authorities, while still setting the new official rate at the level of US$1 = DCU 10, decide to adopt a special rate of US$ 1 = DCU 5 for government transactions. Will this arrangement alleviate the burden of the devaluation on public sector finances? This question is answered in Table 2C. The cost of the government’s deficit on its foreign exchange transactions would be reduced to DCU 100 million. In this example, however, the central bank would incur a loss on its foreign exchange operations. For the $10 million of foreign exchange it would acquire, it would pay out DCU 200 million, or DCU 100 million more than the value of the exchange purchased at the new rate. Thus the gain to the government from such a special arrangement would be exactly offset by a loss to the central bank. It is evident, therefore, that the accounting rate used for government foreign exchange transactions does not affect the public sector deficit—so long as one considers both the government and the central bank as part of a public sector whose finances are inseparable. This is not tantamount to denying that proper valuation is of the utmost importance. When the government applies artificially low exchange rates to defense imports, expenditures of embassies abroad, travel of its staff in foreign countries, and service payments on its foreign debt, the real costs involved are not immediately apparent. As a result, such spending is not curtailed as much as it should be, and it becomes difficult to allocate scarce foreign exchange resources properly among competing public and private uses.

Taxation of External Transactions

A devaluation may lead to, or be accompanied by, changes in taxation on the foreign exchange transactions of the private sector. Thus, there may be an increase in the proceeds from taxes on payments, such as tariffs, import surcharges, or exchange taxes. Even if the relevant rates are not changed, it is likely that collections would rise, because the same tax rates would apply to a much higher value of payments in terms of DCUs. To illustrate this point, Table 3 assumes that a tax of 10 per cent on private payments would be imposed both without and with a devaluation; if so, in the former case government revenues would be equal to 10 per cent of $80 million at US$1 = DCU 5, or DCU 40 million, while in the latter they would amount to 10 per cent of $70 million at US$1 = DCU 10, or DCU 70 million. Moreover, taxes on some private sector receipts (e.g., exports) may be imposed (or continued) after a devaluation, at least on a temporary basis, in order to prevent exporters from realizing windfall profits. On the other hand, the authorities may wish to maintain after a reform, or introduce, subsidies on payments (e.g., on imports of essential commodities of mass consumption) or on receipts (e.g., on nontraditional exports that are not regarded as receiving sufficient incentive even with the postdevaluation rate); Table 3 shows the effect of a hypothetical subsidy averaging 5 per cent on both payments and receipts of the private sector.

Table 3.Taxation of External Transactions and Counterpart Funds(In millions of DCUs)
Without

Devaluation
With

Devaluation
A. Revenue
1. Taxes on private sector payments in
foreign exchange (at a rate of 10
per cent)4070
2. Taxes on private sector receipts in
foreign exchange (at a rate of 5
per cent)2250
3. Counterpart funds (assuming foreign
aid of $10 million sold at prevailing
exchange rates)50100
4. Total112220
B. Expenditures
1. Various subsidies on private sector
payments in foreign exchange (at
a rate of 5 per cent)2035
2. Various subsidies on private sector
receipts in foreign exchange (at
a rate of 5 per cent)2250
3. Total4285
C. Net revenue (C=A4−B3)70135

For those countries that are the recipients of foreign nonproject assistance (grants or loans to be used for purchasing essential consumer goods or raw materials) an especially important result of a devaluation may be an increase in counterpart funds, i.e., the proceeds from the sale of these commodities to the private sector. Table 3 assumes that such assistance amounts to $10 million and that the government would realize revenues of DCU 50 million without a reform and of DCU 100 million with a reform, provided, in the latter case, that selling prices of the commodities involved are adjusted in line with the devaluation.

Allowing for both revenues and expenditures, the example of Table 3 concludes that if there is a devaluation the government’s net revenues from taxes and counterpart funds minus subsidies would amount to DCU 135 million, or DCU 65 million more than if no devaluation is undertaken.

Sometimes, reforms of the foreign exchange system do not adopt a unitary rate, but establish two or more rates for receipts and payments. Thus, the authorities may devalue from a level of US$1 = DCU 5 to a new official rate of US$1 = DCU 10, applicable to most private transactions and to the government, and also decree special rates of US$1 = DCU 8 and of US$1 = DCU 9 for certain categories of private sector exports and imports, respectively. These special rates constitute in fact taxes or subsidies (the rate of US$1 = DCU 8 involves a tax of DCU 2 per U. S. dollar on exports and that of US$1 = DCU 9 a subsidy of DCU 1 per U.S. dollar on imports) and should be treated as such in any computation of the budgetary effects of a devaluation, since they obviously affect the profits or losses of the central bank on its exchange operations.

As an exercise in the application of the above analysis, let us examine a case of some practical importance. Suppose it has been decided to adopt a rate of US$1 = DCU 8 for half of the private sector’s receipts and half of its payments and one of US$1 = DCU 10 for the remaining halves. Assume also that the authorities are trying to determine the levels in that range at which they should set the official rate used in the valuation of the central bank’s holdings of exchange reserves and the rate for government transactions. What are the considerations that should guide their decisions? To begin with, there is no good justification for having different rates for these two purposes; nothing can be gained thereby by way of reducing the public sector deficit, while something may be lost perhaps by needlessly complicating the exchange system. As indicated earlier, a realistic accounting rate should be applied to government transactions in foreign exchange, and this is also true for similar reasons as regards the valuation of reserves. It follows that these two rates should be set at the same level and as close as possible to the equilibrium rate, i.e., the unitary rate needed in order to eliminate the balance of payments deficit without undue reliance on restrictions. Precisely how such a rule should be applied in the case at hand would require careful study of the circumstances involved. It may be noted, however, that in many instances in which a dual rate structure is decreed, the lower (more appreciated) rate is adopted as a temporary arrangement in the knowledge that it overvalues the DCU, whereas the higher rate is regarded as more or less realistic; the intention is to unify the exchange system eventually at this higher rate. If so, the two rates discussed here should be fixed at US$1 = DCU 10.

Surprising though it may seem, in the above example the authorities can realize profits of an accounting nature if they in fact decide on this rate of US$1 = DCU 10 rather than some lower rate, say US$1 = DCU 8. Using the methods shown in Tables 2 and 3, it can be computed that with a rate of US$1 = DCU 10, the cost of the government’s deficit on its exchange transactions would be ($20 million × 10 =) DCU 200 million. On the other hand, according to the arrangements described in the preceding paragraph, i.e., that half of private sector receipts and payments would be at the rate of US$1 = DCU 10 and the remaining half at US$1 = DCU 8, the central bank would be making a profit on its exchange operations with the private sector, since it would be acquiring from that sector $30 million of foreign exchange and paying DCU 270 million, i.e., DCU 30 million less than the cost of this exchange at the official rate. Thus the overall public sector deficit would amount to DCU 170 million. With a rate of US$1 = DCU 8 the cost of the government’s deficit would be equal to DCU 160 million, but the central bank would be incurring a deficit of DCU 30 million on its foreign exchange operations with the private sector, so that the overall public sector deficit would total DCU 190 million. The general rule, which the reader can easily verify, is that with a higher (more depreciated) rate, the overall public sector deficit would always be smaller, if the balance of payments were in surplus after a reform of the exchange system.

Internal Taxation

A devaluation normally leads to increases in prices and incomes and, therefore, to larger receipts from internal taxation (defined for present purposes as taxation other than the above-mentioned taxes on foreign exchange receipts and payments of the private sector). While it is not possible here to treat this complex subject exhaustively, an attempt is made in Table 4 to outline the principal relationships involved. The table assumes a devaluation of the extent indicated in Table 2. It divides the economy into two sectors, one producing exportable goods (exported or consumed at home) and import-competing goods and another referred to as the “domestic” sector, the output of which is neither exportable nor import-competing. Given the figures shown in the table for gross domestic product (GDP) and its components and assuming that the average tax rate is 10 per cent, uniform for both sectors, the total proceeds from internal taxes without a devaluation would amount to DCU 400 million.

Table 4.Internal Taxation 1(In millions of DCUs)
Export and Import

Competing Sector
Domestic

Sector
Total
A. Without devaluation
1. Value of output6003,4004,000
2. Proceeds from internal taxation
(at a tax rate of 10 per cent)60340400
B. With devaluation
1. Output at predevaluation prices9003,2004,100
2. Output at postdevaluation prices1,8003,2005,000
3. Proceeds from internal taxation
(at a tax rate of 10 per cent)180320500

This table assumes a devaluation of the form and extent described in Table 2B or 2C.

This table assumes a devaluation of the form and extent described in Table 2B or 2C.

In response to the price incentives or disincentives created by the devaluation, there would probably be a substantial increase in output in the sector producing exportable and import-competing goods and a relatively smaller decline in the domestic sector; the former sector would be likely to attract capital and labor from the latter and also to mobilize productive resources which would have been unemployed (or underemployed) without a reform. According to the example shown in Table 4, the devaluation would lead to an increase in real output of 2.5 per cent. This example, like all others used here, is intended only as an illustrative device and the actual increase in real GDP could be considerably less or more than this. It could be more if a sizable fraction of the country’s productive capacity is idle or underemployed before a reform.

Prices would remain unchanged in the domestic sector, because it is postulated in Table 4 that the authorities would follow the recommended policy of not permitting wage increases after the devaluation. (This may be an oversimplification. The government may find it inevitable to allow some wage increases and there may also be adjustments in profit margins. If domestic prices do, in fact, rise there may be additional effects on both budget revenues and expenditures not taken into account in the remainder of this paragraph.) On the other hand, prices would rise 100 per cent in the other sector in line with the change in the effective exchange rate. (It should be noted, however, that the increase in the volume of exports associated with a devaluation might bring about a decline in export prices in terms of foreign currencies; if so, prices of exportable goods would rise by less than the increase in the price of foreign exchange.) On the basis of the data given in Table 4 for GDP in current prices and the components thereof following a devaluation, and assuming that the average tax rate would continue to be a uniform 10 per cent, the total yield of internal taxation would amount to DCU 500 million or DCU 100 million more than without a devaluation. However, it is possible, perhaps likely, that the marginal tax rate would prove higher than the average. If so, the increase in revenues associated with devaluation would be higher than indicated above.

The Total Effect

Table 5, which summarizes the results obtained from the examples of Tables 2, 3, and 4, indicates that the assumed devaluation would improve the position of the public sector by DCU 65 million. It should be emphasized that these examples have been designed with a view to describing a correct method of calculation; they are not intended as a basis for generalizing on the possible budgetary impact of a reform. In point of fact a devaluation need not always result in an improvement. If the government’s deficit on its transactions in foreign exchange is substantial and the offsets indicated above are of a relatively small order of magnitude, the overall effect will be unfavorable. However, in most cases, the sum of the offsets is likely to be quite significant. Tariffs and internal taxation will often yield considerable additional revenue in the event of devaluation. Moreover, for a number of countries, particularly in the less developed group, receipts from temporary export taxation and counterpart funds may increase appreciably. A priori, therefore, public finances should not be expected to deteriorate much as a consequence of a reform, and they may very well improve. It should be pointed out, however, that a substantial deterioration could occur if the government tried to neutralize the price effects of a devaluation, by, inter alia, granting subsidies, reducing tariff rates, or failing to raise prices of aid-financed imports sold to the public; such measures are sometimes adopted for political reasons, even though they undermine the balance of payments objective for which a reform is undertaken.

Table 5.Total Effects of Devaluation and Accompanying Measures on Public Sector Deficit(In millions of DCUs)
A. Increased revenues
1. From external taxation (Table 3)65
2. From internal taxation (Table 4)100
3. Total165
B. Increased expenditure
1. Increase in cost of public sector deficit on its
payments and receipts in foreign exchange (Table 2)100
C.Net impact of devaluation (C = A3 − B1)65

BALANCE OF PAYMENTS YEARBOOK

The Fund publishes statistics on the international transactions of about 100 countries in the Balance of Payments Yearbook. Each volume of the Yearbook is first available in loose-leaf form, issued in monthly parts, at an annual cost of US$11.00 (including binder). The volumes are then published in bound form in hard covers at a price of $6.00 to subscribers to the loose-leaf version or $12.50 to nonsubscribers.

Orders may be sent to

The Secretary

International Monetary Fund

19th and H Streets, N.W.

Washington, D.C. 20431

U.S.A.

Bound volumes are available as follows:

Volume 20 covering 1963–67

Volume 21 covering 1964–68

Volume 22 covering 1965–69

The loose-leaf version of Volume 23 commenced publication in December 1971; it covers 1966-70.

University libraries, faculty members, and students may purchase loose-leaf volumes at a special price of $3.00.

Other Resources Citing This Publication