Chapter

Operation of Facility

Author(s):
International Monetary Fund
Published Date:
January 1980
Share
  • ShareShare
Show Summary Details

The speed of operations has been a major concern in setting up and administering the compensatory financing facility. A Fund mission does not normally need to visit the country presenting a request, and telex is the most usual means of communication. After the member has provided the necessary data, the staff requires approximately two weeks for analyzing them19 and, if the request appears justified, for preparing the report to be submitted to the Executive Board. The purpose of the report is to establish the amount which the member can draw and to explain why the staff considers that the various requirements specified in the compensatory financing decision are satisfied. The Executive Board decides on the request about two weeks after receiving the staff report and, if the request is approved, the country can draw within a matter of days. The entire operation can therefore be conducted in about a month.

The conditions under which a member can draw and the ways of calculating the amount it can draw are defined by the decision reproduced in Appendix I. They will be reviewed below in four sections dealing with coverage, maximum drawing, conditionality and requirement of need, and terms of purchase and repurchase.

Coverage

The facility, which initially covered shortfalls in earnings from merchandise exports only, was extended in August 1979 to cover both merchandise and services. The new decision did not affect the choice of the shortfall year or the measurement of export earnings in terms of SDRs.

Merchandise and Services

For calculation of the shortfall, exports are measured on a customs basis, whenever feasible, and are net of re-exports. In accordance with UN statistical practice, re-export is defined as the export of goods that have previously been imported and whose physical characteristics have not been modified while they have been within the country’s boundaries.20 Exported goods produced with imported raw materials are included in the calculation of the shortfall, and no deduction is made for the import content of exports, although a decrease in the volume of exports may be partly offset by a reduction in imports.21

Since the facility was liberalized in August 1979, receipts from travel and workers’ remittances can be added to earnings from merchandise exports under two conditions.22 First, the Fund must be satisfied that the statistics are reasonably accurate. Second, the member must opt for the inclusion or the exclusion of travel and workers’ remittances when it presents its first request after the end of 1979, and the option cannot be reversed for a period of five years. If a member opts for the inclusion of income from travel and workers’ remittances but does not have reasonably accurate statistics for both, only the item for which accurate data are available is included.

Choice of Shortfall Year

The decision prescribed the shortfall year as the latest 12-month period for which customs data are available, and the practice has been to require that the interval between the end of the shortfall year and the time of request not exceed 6 months. This requirement, however, gave rise to two difficulties. First, countries reporting their trade data with more than a 6-month lag could not have access to the facility. Second, even for a country with a short data lag, almost a year would elapse between the middle of the shortfall year and the time of drawing. These two drawbacks were largely eliminated by the introduction in December 1975 of the early drawing procedure, which allowed members to estimate export earnings for up to 6 months of the shortfall year.

By using the early drawing procedure, a member with a data lag of less than 6 months can make a choice among shortfall years ending in 7 different months. Consider, for example, a member having a data lag of 3 months and presenting a request in April 1980. The member could follow the standard procedure and select calendar year 1979 as the shortfall year (the latest 12-month period for which data are known). Alternatively, it could follow the early drawing procedure and select a shortfall year ending any month from January 1980 to June 1980; depending on the choice made, export earnings would have to be estimated for 1 to 6 months.

In addition, the early drawing procedure allows a member with a data lag of more than 6 months but no more than 12 months to use the facility. As the lag increases from 6 to 12 months, the choice among different shortfall years is progressively reduced. Thus, a member with a lag of 12 months would not have any choice. For it, the first half of the shortfall year would have to comprise the last 6 months for which data are available, as earnings in the shortfall year can be estimated for no more than 6 months.

As protection against exaggerated shortfall estimates, a special repurchase provision is associated with the early drawing procedure.23 Shortly after actual data become available for the part of the shortfall year that previously had been estimated, the amount of the shortfall is recalculated by substituting these data for the earlier estimates. The original forecast for the two following years is retained. If the new calculation shows that the member was overcompensated, the member is asked to repurchase the excess.

The early drawing procedure was applied to 30 of the 107 drawings made from 1976 through March 1980. At the end of that period, the actual value of export earnings in the shortfall year was known in 22 of the 30 cases.24 In 2 of these, the member was required to repurchase. In the remaining 20 cases, a repurchase was not requested, either because earnings in the shortfall year had not been underestimated, or because the margin between the calculated shortfall and the amount drawn was not sufficiently wide.

Receipts from travel and workers’ remittances raise special problems because they are often recorded only on the basis of either calendar or fiscal years, which may not correspond with the 12-month period selected as the shortfall year. Suppose that the shortfall year ends in March 1980 but that earnings from services are recorded only for calendar years. Suppose further that the latter are 100 in 1979 and are estimated to be 80 in 1980. Earnings from services in the shortfall year would be prorated from their calendar-year values and calculated as (100 × 3/4) + (80 × 1/4) = 95. Because of the special data problems for services, earnings from travel and workers’ remittances can be estimated for up to the full 12 months of the shortfall year. As in the case of merchandise exports, the member must represent that it shall repurchase any excess compensation which might result from an underestimation of earnings from services.

Measurement of Export Earnings

The effect of currency fluctuations on export earnings is reduced by measuring earnings in SDRs. For this purpose, monthly earnings recorded in U.S. dollars (or in another currency) are transformed into SDRs by applying the average monthly exchange rate between the U.S. dollar (or the other currency) and the SDR. Fluctuations in export earnings result, therefore, from the combination of the fluctuations in the volume of exports and fluctuations in average export unit values expressed in SDRs.

In the staff report submitted to the Executive Board, export earnings are shown for the 12-month period selected as the shortfall year and for the five preceding 12-month periods, with an analysis of total export earnings among main export commodities (e.g., sugar, copper) or groups of export items (e.g., manufactures, workers’ remittances). For the main export commodities, an analysis of export earnings between volume and unit values is also given.25

Maximum Drawing

The maximum amount which a member can draw under the facility is assessed in three steps. First, a forecast is made of export earnings in the two post-shortfall years in order to calculate the amount of the shortfall. Second, when necessary, a deduction is made from the shortfall to avoid double compensation with previous drawings made under the compensatory and buffer stock financing facilities. Third, this adjusted amount is reduced, whenever required, in order that the member’s outstanding drawings under the facility does not rise above 100 per cent of its quota in the Fund.26

Export Earnings Forecast

As the shortfall year always includes the latest data on export earnings, calculation of the trend value requires a forecast of export earnings during the two-year period following the end of the shortfall year. This projection is made in terms of both volume and price for major commodities and in terms of earnings only for other exports. It is based on past trends and changes from the trend likely to arise from factors within the member’s control (e.g., production and trade policies) and from other factors (e.g., world market prospects). The projection of export unit values relies heavily on the forecast of commodity prices made regularly by the staff of the Fund. Volume projections are made, when relevant, within the framework of a commodity balance showing domestic production, domestic consumption, exports, and changes in stocks.

Approximately two years after the request, or a little longer when the early drawing procedure was used, the error which had been made in forecasting export earnings can be measured. The amount drawn by the member can then be compared with the amount it would have drawn in the absence of any forecasting error. Such calculations were done for 53 purchases made in 1976 and 1977.27 In 37 of these, the amounts purchased would have been unchanged if the forecasts had been completely accurate; in 11 instances members were overcompensated because post-shortfall earnings had been overestimated, and in the remaining 5 cases members were undercompensated because post-shortfall earnings had been underestimated. The total amount drawn in these 53 cases was SDR 2.5 billion; it would have been SDR 2.3 billion had there been no forecasting errors.

One country accounted for more than a third of total overcompensation and another for almost two thirds of total undercompensation. For the first country, growth of 13.8 per cent a year had been projected for the two post-shortfall years, but actual growth was only 11.2 per cent a year. For the second country, growth of 15.4 per cent a year had been projected, but the actual growth rate was 19.9 per cent a year. As can be seen from these cases, forecasting errors that are relatively small in percentage terms may cause sizable differences in the amounts purchased.

Use of Geometric Average

The trend value used in assessing the amount of the shortfall was calculated as an arithmetic average until the adoption of the 1979 decision. The change to a geometric average was made because it was recognized that the nominal value of export earnings follows much more closely a geometric curve than an arithmetic curve. Calculating the trend as a geometric average instead of an arithmetic average hardly affects the amount of the shortfall when export earnings remain approximately constant, but it makes a substantial difference when export earnings follow a strong upward trend. This is illustrated in Table 10 by assuming that export earnings increase each year at a constant rate, taken alternatively as 0 per cent, 10 per cent, and 25 per cent. In the first example, earnings are 100 during each of the five years; whether the trend is calculated as a geometric average or as an arithmetic average, the trend value remains 100 and the amount of the shortfall is zero. In the last example, in which earnings increase by 25 per cent a year, the five-year geometric average centered on the shortfall year is, by definition, equal to the value of export earnings in that year and the shortfall is again zero. If instead the trend had been defined as a five-year arithmetic average, its value in the shortfall year would have been 105 and the amount of the shortfall would have been equivalent to 5 per cent of the value of export earnings in that year. In contrast, if export earnings had been increasing each year by a constant amount (instead of at a constant rate), the use of an arithmetic average would have shown a zero shortfall while the use of a geometric average would have shown an excess.

Table 10.Shortfalls Calculated with Arithmetic and Geometric Averages for Various Growth Rates of Export Earnings
Annual Growth Rates of Export Earnings in Per Cent
01025
Export earnings
Year −210082.664
Year −110090.980
Year 0100100100
Year +1100110125
Year +2100121156
Arithmetic average
Trend value100100.9105
Shortfall00.95
Geometric average
Trend value100100100
Shortfall000

Whether export earnings tend to increase geometrically (i.e., at a constant rate) or arithmetically (i.e., by a constant amount) was tested with a sample of 74 countries during 22 consecutive years (1957 through 1978). The statistical fit is clearly better with a geometric trend than with an arithmetic one, especially for the more developed countries (Table 11). Moreover, shortfalls and excesses tend to offset each other with a geometric average, while the sum of shortfalls is substantially greater than the sum of excesses with an arithmetic average (Table 12). The difference is particularly striking for Korea, the country with the fastest growth of export earnings (38 per cent a year). The sum of shortfalls is 100 times greater than that of excesses when the trend value is measured as an arithmetic average, while the sum of shortfalls exceeds the sum of the excesses by only 9 per cent when it is measured as a geometric average.

Table 11.Arithmetic and Geometric Trends, Relative Goodness of Fit for 74 Countries, 1957–781
ArithmeticGeometric
Best fit (highest R2)(Number of countries)
Trust Fund countries2534
Other less developed countries023
More developed countries012
All countries569
Average correlation coefficient(R¯2)
Trust Fund countries20.680.78
Other less developed countries0.730.86
More developed countries0.790.92
All countries0.710.83

Excluding the industrial and oil exporting countries listed in footnotes 2 and 3 of Table 1.

Countries with low per capita income eligible to borrow from the Trust Fund.

Excluding the industrial and oil exporting countries listed in footnotes 2 and 3 of Table 1.

Countries with low per capita income eligible to borrow from the Trust Fund.

Table 12.Amounts and Number of Shortfalls and Excesses Calculated with Arithmetic and Geometric Averages for 74 Countries, 1959–761
Arithmetic AverageGeometric Average
ShortfallsExcessesShortfallsExcesses
Amounts(Per cent of export earnings2)
Trust Fund countries5.303.624.214.45
Other less developed countries5.292.943.943.83
More developed countries4.122.152.872.95
All countries5.103.173.914.01
Number of years(Per cent of total)
Trust Fund countries57435248
Other less developed countries60405545
More developed countries64365743
All countries59415446

The basic data cover the 22-year period 1957–78 as in Table 11, but shortfalls and excesses cannot be calculated for either the first two years (1957 and 1958) or the last two (1977 and 1978).

Obtained by averaging shortfalls (or excesses) over trend value of export earnings.

The basic data cover the 22-year period 1957–78 as in Table 11, but shortfalls and excesses cannot be calculated for either the first two years (1957 and 1958) or the last two (1977 and 1978).

Obtained by averaging shortfalls (or excesses) over trend value of export earnings.

The use of a geometric average facilitates the analysis of the shortfall into its volume and price components, but it complicates the analysis of the shortfall in total export earnings into its commodity components. With an arithmetic average, the shortfall in total earnings is equal to the sum of the commodity shortfalls minus the sum of the commodity excesses but this equality is not true with a geometric average. With the latter, the shortfall in the aggregate is lower than the sum of shortfalls in the components minus the sum of excesses in the components, except when export earnings from each component increase at the same rate.28

Deductions for Avoiding Double Compensation

Once the amount of the shortfall has been calculated, deductions may have to be made in order to avoid double compensation which could arise from previous drawings made under the compensatory financing facility or the buffer stock facility. Deductions may also have to be made when stocks have increased during the shortfall year.

Previous compensatory drawing.

—Since members can purchase under the compensatory financing facility more than once within a 12-month period, the shortfall years associated with two consecutive compensatory purchases may overlap. It is clear that a member should not be compensated twice for the shortfall experienced during the overlapping months. This is avoided by deducting from the second shortfall the amount of the previous compensatory purchase prorated according to the number of overlapping months. For example, if there is a 3-month overlap, one fourth of the first purchase is taken as compensation already received for the second shortfall and deducted from the amount calculated for the second shortfall. This deduction does not, however, reduce the amount which a member can purchase if the shortfall thus adjusted exceeds the maximum amount which the member can purchase on account of quota limitations.

Previous buffer stock drawing.

—Under the terms of the 1977 International Sugar Agreement, exporting members have to constitute a minimum amount of stocks when prices remain below an agreed level. Such members, if they are also members of the Fund, may draw under the buffer stock facility an amount corresponding to the export earnings forgone by constituting these stocks. The maximum amount of the drawing is calculated by valuing the stocks at the price prevailing on the free market at the time of their constitution, or at the floor price of the Agreement if the latter is below the market price. For example, suppose that a member requests a drawing under the buffer stock facility in January 1980 on account of its stocking obligation for 1979, and that it simultaneously requests a drawing under the compensatory financing facility on account of a shortfall experienced in 1979. Part of the export shortfall calculated for that year would be due to constitution of the special stocks, which are not recorded as exports because they are held within the country’s boundaries.

If the member is compensated for the constitution of its special stocks under the buffer stock facility, it should not be compensated a second time under the compensatory financing facility for the part of its shortfall resulting from the constitution of the special stocks. The method used for avoiding double compensation is to treat the amount drawn under the buffer stock facility as unrecorded export earnings during the 12-month period preceding the buffer stock drawing, and to allocate ½ of the total amount drawn to each of these 12 months. Raising the value of export earnings during one of the months of the shortfall year would reduce the calculated shortfall, while raising it during one of the 24 months preceding the shortfall year or during one of the 24 months following it would increase the amount of the shortfall. When the buffer stock purchase is made in the period starting the fourth month of the shortfall year and ending at the time of the compensatory drawing, the adjustment results in a deduction from the shortfall.29 This deduction would not, however, affect the amount drawn, if the deduction was less than the excess of the calculated shortfall over the maximum amount which could be drawn on account of the quota limit.

The problem of double compensation does not arise for a buffer stock drawing made in connection with internationally financed stocks, as is the case for tin. As the stocks bought by the international buffer stock manager are normally recorded as exports, they do not affect the export shortfall and no adjustment is required.

Stock accumulation.

—Suppose that a country accumulates stocks in 1979 and sells them at the beginning of 1980 just before presenting a request to draw under the compensatory financing facility. Part of the shortfall calculated for 1979 would correspond to the export earnings forgone by having accumulated stocks in that year, but the member would have already been compensated for this part of the shortfall when it sold its stocks before requesting a drawing under the compensatory financing facility. The possibility of double compensation is avoided by adding the value of the stocks accumulated in the shortfall year to the earnings recorded in that year, and by subtracting the proceeds obtained by selling the stocks in the following year from the earnings projected for that year. If prices were to be unchanged, the adjustment would be equivalent to subtracting the value of the stocks accumulated in the shortfall year from the amount of the calculated shortfall.

At the time the compensatory financing request is made, it is generally not known when the stocks will be sold or at what price they will be sold. Sometimes, even the amount of stocks accumulated in the shortfall year is not known precisely. Moreover, the member may have to hold stocks for a long time due to circumstances beyond its control. In view of these uncertainties, adjustments for stock accumulation have been confined so far to instances where it was clear that double compensation should be avoided.

Before the decision of August 1979, deductions had also to be made in connection with credit tranche drawings made prior to the drawings under the facility. The rules applied were, however, complex and somewhat arbitrary, because it was not possible to determine ex post the part of a previous credit tranche drawing which ought to be considered as compensation for the export shortfall on which the compensatory financing request was based. Because of these difficulties and because the application of the rules had had little effect on the amounts drawn, it was decided to make a deduction from the calculated shortfall only in connection with previous drawings made under the compensatory and buffer stock facilities as described above. The fact that credit tranche drawings were taken into account when assessing the need for the member to draw was considered as sufficient protection against possible double compensation. A similar view was taken regarding transfers received by members under STABEX, the compensatory facility administered by the European Community. STABEX transfers are taken into account only in assessing the member’s need to draw on the basis of its balance of payments deficit and reserve position.

Quota Limits

The amount a member can draw under the facility can exceed neither the calculated shortfall net of adjustments for avoiding double compensation nor 100 per cent of the member’s quota in the Fund. Drawings have always been subject to quota limitations since the facility was established, but the maximum amount of outstanding drawings has been raised progressively from 25 per cent of quota under the 1963 decision to 50 per cent of quota under the 1966 decision, 75 per cent of quota under the 1975 decision, and 100 per cent of quota under the 1979 decision. The last decision also eliminated the additional constraint on drawings within a 12-month period, which were limited to 25 per cent of quota under the 1966 decision and to 50 per cent of quota under the 1975 decision.

The progressive relaxation of quota limitations has been partly offset by the erosion of Fund quotas resulting from inflation. As shown in Chart 4, the ratio of quotas to export earnings follows a cycle that rises sharply when new quotas go into effect and then falls progressively until the next quota increase. The peak, which should have coincided with the entry into effect of the sixth quota increase in 1978, is, however, hardly apparent on the chart because of the high rate of inflation in the five preceding years. From 1967 to 1972, the 50 per cent quota limit of the 1966 decision was equivalent to 8.4 per cent of export earnings for a sample of 74 countries. In 1978, the 75 per cent quota limit of the 1975 decision was equivalent to only 6 per cent of quota earnings for the same 74 countries. The 100 per cent quota limit of the 1979 decision is likely to be equivalent to less than 7 per cent of export earnings for the same group of countries in 1980. It would, however, rise in 1981 to almost 9 per cent of export earnings if Fund quotas were raised by one half in that year as envisaged under the Seventh General Review of Quotas.

Chart 4.Quotas as Percentage of Export Earnings for 74 Countries, 1963–811

1 Sum of quotas divided by sum of export earnings for 74 sample countries, assuming quotas are raised by 50 per cent in 1981 and earnings increase by 10 per cent a year from their estimated value in 1979.

↓ Quota increases associated with the periodic reviews of quotas.

Although the maximum amount of outstanding purchases was raised to 75 per cent of quota under the 1975 decision, quota limitations prevented the majority of the members which drew under that decision from being fully compensated for their shortfalls. In order to eliminate the effect of the forecasting errors made at the time of request, rates of compensation have been measured ex post by dividing the amount of the purchase by the amount of the shortfall calculated on the basis of actual export earnings in the two post-shortfall years. In April 1980 post-shortfall earnings were known for 53 purchases made under the 1975 decision, and average rates of compensation were derived from the rates calculated for each purchase by combining two different measurements of the shortfall with three types of averaging (Table 13). The amount of the shortfall was measured by calculating the trend value either as an arithmetic average or as a geometric average. The overall rate of compensation was then derived in three different ways. The first is the weighted average obtained by dividing the sum of purchases by the sum of shortfalls. The second is the median obtained by ranking individual rates by increasing order and selecting the rate appearing in the middle of the list. The third is the simple average calculated as the unweighted average of individual rates.

Table 13.Average Rates of Compensation for 53 Purchases, 1976 and 19771(In per cent)
Shortfall Calculated With
Nature of AverageArithmetic averageGeometric average
Weighted average232 (38)41 (48)
Median33654
Simple average46068

Purchases expressed as percentage of shortfalls calculated on the basis of actual earnings in the two years following the end of the shortfall year.

Sum of drawings as a percentage of sum of shortfalls. In the figures shown between parentheses, Korea and Malaysia are excluded from the calculation.

Rates of compensation were listed by increasing order and the rate corresponding to the purchase in the middle of the list was selected.

Unweighted average of rates of compensation calculated for each purchase, excluding rates higher than 500 per cent.

Purchases expressed as percentage of shortfalls calculated on the basis of actual earnings in the two years following the end of the shortfall year.

Sum of drawings as a percentage of sum of shortfalls. In the figures shown between parentheses, Korea and Malaysia are excluded from the calculation.

Rates of compensation were listed by increasing order and the rate corresponding to the purchase in the middle of the list was selected.

Unweighted average of rates of compensation calculated for each purchase, excluding rates higher than 500 per cent.

The weighted average is substantially lower than the simple average, because the former is heavily weighted by a few countries that experienced a very large shortfall and were compensated for only a very small part of their shortfall. Thus, when two countries (Korea and Malaysia) are excluded from the calculation, the weighted average is raised from 41 per cent to 48 per cent when a geometric trend value is used. On the other hand, the simple average is heavily influenced by the rates calculated for the countries that were overcompensated; this was the reason for eliminating from the calculations compensation rates in excess of 500 per cent.

Individual rates of compensation that are either very low or very high do not affect the value of the median which, for this reason, probably provides the best measurement. Similarly, calculating the trend as a geometric average provides a more accurate measurement of the shortfall than calculating it as an arithmetic average, since nominal export earnings tend to increase by a constant rate rather than by a constant amount each year. The median derived from individual rates calculated with a geometric average being equal to 54 per cent, it seems fair to say that, on the average, members were compensated for approximately half of their shortfalls. The fact that members were compensated for only half of their shortfalls does not mean, however, that the facility has achieved only half of what it was intended to achieve. There are two reasons for this.

First, the facility was established to provide assistance for the financing of balance of payments “deficits arising out of export shortfalls,”30 and export shortfalls are not always associated with a balance of payments deficit of an equal or greater magnitude. This is because the export shortfall may be partly or fully compensated for by offsetting movements in other items of the balance of payments. For example, among the 53 purchases covered in Table 13, the two largest shortfalls were those experienced by Korea (SDR 1,119 million) and Malaysia (SDR 552 million). When these two countries presented requests under the facility (in June and August 1976, respectively), their needs to draw were considered relatively small. The need could, nevertheless, justify purchases of SDR 40 million and SDR 93 million, respectively. Both amounts were equivalent to 50 per cent of quota, but they represented only a small fraction of the calculated shortfalls. As it happened, the reserve positions of both members improved substantially soon after the purchases, and both countries experienced a balance of payments surplus in 1976. For these two countries, even in the absence of any quota limitation, the need to draw could not have justified purchases equal to the full amount of the shortfall.31

Second, the compensatory financing facility was established as a special means of assisting countries adversely affected by unstable commodity markets and not as a substitute for other forms of Fund credit. The 61 members that received balance of payments assistance amounting to SDR 4 billion under the facility from January 1976 to March 1980 also received balance of payments assistance amounting to SDR 4.8 billion under other Fund facilities during the same period.32 It would clearly be inappropriate for the Fund to extend all its financial assistance in connection with the low conditionality associated with the compensatory financing facility, as many of the countries experiencing export shortfalls have balance of payments difficulties that arise partly from inappropriate policies and cannot be solved without adjustments in these policies. Providing large amounts of assistance without financial programs might be counterproductive in this situation; it might induce the member to delay the adoption of the necessary policy adjustments and, as a result, exacerbate its financial difficulties. A proper balance has to be found between the assistance extended by the Fund under the facility with low conditionality and that extended with higher conditionality under other Fund facilities.

Conditionality and Need

The member does not have to present a financial program for making a purchase under the compensatory financing facility, but some conditions need to be satisfied. The shortfall must be temporary and largely attributable to circumstances beyond the control of the member. Moreover, the member must cooperate with the Fund and have a balance of payments need.

Shortfall of a Temporary Character

The temporary character of the shortfall is ensured, to a large extent, by defining the shortfall as the downward deviation from the five-year average centered on the shortfall year. It has been sometimes argued that such a definition may not be appropriate when the five-year average is raised either by abnormally high prices for a key export product during the two pre-shortfall years, or by the anticipation of new exports resulting, for example, from the opening of a new mine during the two post-shortfall years. If adjustments were made for abnormal events, they should be made for abnormally low earnings as well as for abnormally high earnings, and they should be made whenever the abnormal event occurs in any of the five years included in the calculation of the trend value. Such adjustments would be difficult to make in a precise and objective manner, and they would have the effect of raising the size of the shortfall as often as of reducing it. Another solution for reducing the impact of abnormal events on the calculated shortfall would be to extend the length of the five-year average centered on the shortfall year to seven years or even to nine years. This possibility has been considered, but has not been adopted.

If the fall in export earnings were to last for several years, the shortfall as defined for the purpose of the facility would progressively disappear. This is illustrated in Chart 5 where export earnings increase steadily except during a shortfall period that lasts only one year in one case (top of the chart), but up to five years in another (bottom of the chart). When the shortfall lasts for a single year, its destabilizing effect on the economy of the country can be largely offset by the drawing under the facility, provided the drawing is not severely constrained by the quota limit and is made as soon as the shortfall is identified. When the shortfall lasts for two consecutive years, its destabilizing effects may still be alleviated by making use of the facility, and a number of countries have drawn on account of shortfalls encountered during two consecutive years. When the length of the export shortfall reaches five years, the compensatory financing shortfall disappears in the middle year and the facility, which has been designed to deal with shortfalls of a temporary character, would be clearly inappropriate. In the latter case, the member should apply for financial assistance from the Fund under stand-by arrangements that extend over two-year periods or extended Fund facility arrangements that cover a three-year period, and from the World Bank under structural adjustment loans. Such assistance would require the preparation of financial programs, but the very fact that the member’s balance of payments difficulties are likely to last for a number of consecutive years indicates that there is a need for policy adjustment.

Chart 5.Effect of Length of Shortfall Period on Calculated Shortfall

(Temporary decline from a linear long-term trend)

Shortfall Beyond Member’s Control

When the shortfall in export earnings results mainly from a decline in world market prices and when the member accounts for only a small share of the world exports of the commodities concerned, it is clear that the shortfall is attributable to circumstances beyond the member’s control. However, when the immediate cause of the shortfall is a decline in the volume of exports, the ultimate cause may be partly in the policies followed by the member. Where the volume decline results from the withholding of stocks in the expectation of higher world prices or of the depreciation of an overvalued currency, the amount of the shortfall may be adjusted by taking into account the increase in the level of stocks during the shortfall year.33 Where the member’s policies have been a factor contributing to the shortfall, an attempt is made by the staff to assess the part of the shortfall which is attributable to circumstances beyond the member’s control. Moreover, when the shortfall is partly attributable to inappropriate pricing policies or to unrealistic exchange rates, such policies are taken into account in assessing the adequacy of the member’s cooperation with the Fund.

Cooperation with International Monetary Fund

For making any purchase under the facility, the member must represent that it “will cooperate with the Fund in an effort to find, where required, appropriate solutions for its balance of payments difficulties.”34 The test of cooperation is stricter when the purchase has the effect of raising outstanding purchases above 50 per cent of the member’s quota, as the Fund must then be satisfied that past cooperation has been adequate. Although the extent of the cooperation required has not been codified, satisfactory performance in the context of a financial program supported by the Fund would be considered as evidence of past cooperation.

From the beginning of 1976 through March 1980, 107 purchases were made and outstanding purchases under the facility exceeded 50 per cent of quota in 37 cases (Table 14). In 11 of these, the member had no financial program supported by the Fund at the time of purchase; in 5 of the 11 cases, however, the member adopted a program supported by the use of the Fund’s resources within the 12 months following the drawing under the facility. In the other 26 cases where the stricter test of cooperation had to be applied, the member had a financial program in effect at the time of the purchase and this program was supported by the Fund. This program was supported by a Trust Fund loan or a first credit tranche drawing in 14 cases, and by drawings under second or higher credit tranches or under extended facility arrangements in 12 cases.

Table 14.Purchases Under Fund Facility and Test of Cooperation
Type of Compensatory Financing PurchaseNumber of PurchasesAmount Purchased (Million SDRs)
All purchases under facility1074,048
Purchases under facility not raising outstanding purchases above 50 per cent of quota702,488
Purchases under facility raising outstanding purchases above 50 per cent of quota1371,560
No financial program in effect at
time of facility purchase11546
−Nor within 12 months of facility
purchase6280
−Program within 12 months of facility
purchase5266
Financial program in effect at time of
facility purchase261,014
−Program associated with Trust Fund
or first credit tranche stand-by
arrangement14439
−Program associated with higher credit
tranche stand-by or extended
arrangement12575

Including countries with outstanding drawings that were already above 50 per cent of quota at the time of drawing.

Including countries with outstanding drawings that were already above 50 per cent of quota at the time of drawing.

The record clearly shows that a financial program in effect is not required for passing the stricter test of cooperation. In fact, some of the countries which passed this test had already taken appropriate measures to solve their balance of payments difficulties, even though they did not have a financial program supported by the Fund’s resources. On the other hand, the fact that a member has presented a program supported by the Fund cannot always be taken as evidence of past cooperation; it is also necessary that the country’s performance under the program has been satisfactory. For these reasons, the extent of the member’s cooperation with the Fund has to be reviewed on a case-by-case basis.

Requirement of Need

Purchases under the facility, like all other Fund purchases, are subject to the requirement of need, which is assessed on the basis of the member’s balance of payments or reserve position or developments in its reserves.35 In making this assessment, it is necessary to look both backward and forward. As assistance is provided under the compensatory financing facility to members experiencing deficits arising from export shortfalls, it is necessary to consider the overall balance of payments position of the member during the shortfall year which generally ends several months before the request is made. It is also necessary to consider the need of the member to draw at the time the request is made. Balance of payments statistics are normally available only on a calendar-year or fiscal-year basis. The balance of payments position of the member needs, therefore, to be reviewed for two consecutive calendar (or fiscal) years, in order to cover not only the shortfall year associated with the purchase under the facility but also the period during which the request is made and a few months beyond it. For example, suppose a request were to be made in June 1980 in connection with a shortfall experienced in the 12-month period ended March 1980; the relevant balance of payments deficits would be those relating to calendar years 1979 and 1980. The balance of payments for 1979 would normally be based on actual data, while that for 1980 would be essentially based on forecasts.

The particular balance of payments deficit used for assessing need is the one financed by reserve assets and selected liabilities. A deficit of this kind, which is the most comprehensive of the several measures of imbalance that are in common use, is often termed an “overall” or “official settlements,, deficit. Under this approach, an element of judgment is usually required in deciding whether certain capital flows should be regarded as contributing to the deficit or as helping to finance it. Decreases in international reserves, increases in uses of Fund credit, and the accumulation of payments arrears are always taken to be financing items. The motivation for much official borrowing, however, is ambiguous or unknown; a loan may be contracted, for example, to bolster reserves, to finance a development project, or to achieve both of these purposes. It has not been possible, therefore, to lay down hard and fast rules about how a balance of payments deficit is to be measured in all circumstances.36

The member’s reserve position normally relates to its gross international reserves which, for comparative purposes, are generally expressed in terms of weeks’ imports. Considering all purchases made under the facility since 1976, the reserve position was, on the average, equivalent to about ten weeks of imports. The average ratio in the three months following the purchase (excluding the amount of this purchase) was on the average one fourth below the average ratio recorded during the eight-year period 1971 through 1978. Members’ reserves were, therefore, at a low point when the purchase was made under the facility, as could have been expected. These averages are, however, only mentioned for illustrative purposes. It is not possible to define a threshold below which reserves should be considered as inadequate, because the adequate level of reserves may vary considerably from country to country. On the one hand, some countries belonging to a common currency union (e.g., African countries in the French franc area) may operate with very low reserves of their own, because they have easy access to a pool of reserves within the union. On the other hand, some countries subject to wide fluctuations in their export earnings consider that their reserves are inadequate if they do not exceed the equivalent of half a year of imports.

Repurchases

In order to make a purchase under the facility, members must have simultaneously an export shortfall and a need to draw. It could be argued by analogy that members should be expected to repurchase when they have simultaneously an export excess and the ability to repay. The Fund decided, however, not to link the repurchase expectation to the existence of an export excess and to treat repurchase under the facility in the same manner as any other repurchase of the Fund’s resources. The reason for not linking repurchase to export excess is a pragmatic one. Doing so would have required a forecast of export earnings at regular intervals (e.g., quarterly) for every country with outstanding drawings under the facility, and an agreement between the Fund and the member on such forecasts when they led to a repurchase expectation.

As in the case of repurchase under the normal credit tranches, repurchases of drawings under the facility are “made in equal quarterly installments during the period beginning three years and ending five years after the date of purchase unless the Fund approves a different schedule.”37 However, an expectation to repurchase earlier will normally arise on the basis of an improvement in the member’s balance of payments and reserve position, as in the case of any other purchase from the Fund.

A member’s balance of payments and reserve position would normally be deemed to have improved sufficiently for repurchases to be expected under Article V, Section 7(b), if the member’s position is judged sufficiently strong for inclusion in the quarterly designation plan or the operational budget. During the quarter following the adoption of the plan or budget, it would be expected that the member’s outstanding purchases would be reduced either by repurchases or by sales of the member’s currency, the amount of the reduction being related to the level of the member’s gross reserves and to changes in the level of these reserves during the previous six months.38

Another instance of early repurchase may arise in connection with drawings based on partly estimated data for the shortfall year. When the underestimation of export earnings in the shortfall year results in overcompensation, the member has to repurchase the excess promptly in accordance with its representation made at the time of purchase.39

Concluding Remarks

The compensatory financing facility was established in 1963 to provide additional balance of payments assistance to countries adversely affected by fluctuations in their export earnings, but its role remained relatively modest until the facility was liberalized at the end of 1975. Since then, it has become an important means of balance of payments assistance: in the four years following the liberalization, yearly drawings under the facility averaged SDR 1 billion, which represents approximately 30 per cent of all Fund credits.

The amount drawn varies widely from year to year; it follows a highly cyclical pattern which reflects that of the business cycle in industrial countries. In expansion years, most commodity prices rise, few countries experience export shortfalls, and the amount drawn is small. In recession years, most commodity prices fall, many countries experience export shortfalls and balance of payments difficulties, and the amount drawn is large. Compensatory financing may, therefore, have a sizable anticyclical effect, provided assistance is timely.

In order to provide assistance to members as soon as they experience an export shortfall, the 1975 decision introduced the early drawing procedure which has been widely used since 1976. When the facility was reviewed in 1979, various ways of measuring more accurately the effect of the shortfall on the balance of payments deficit were considered (e.g., measuring exports net of their import content), but the Executive Board decided against a number of possible refinements which would have required additional data and reduced the speed of operations. In fact, the 1979 decision simplified the modus operandi of the facility in several respects.

The compensatory financing facility is complementary to other forms of Fund credit. It provides balance of payments assistance expeditiously to members that have been adversely affected by temporary export shortfalls largely attributable to circumstances beyond their control. When the member’s balance of payments difficulties result from inappropriate policies, the member should not seek Fund assistance under the compensatory financing facility. Rather, it should draw under the Fund credit tranche policies on the basis of a program outlining the measures it intends to apply to find appropriate solutions to its balance of payments difficulties.

The facility is a means of cushioning export earning fluctuations, but it is not a substitute for price stabilization. It should, therefore, be regarded as complementary to other approaches, such as commodity agreements and international buffer stock schemes, which aim at reducing price fluctuations for selected commodities.

    Other Resources Citing This Publication