Rome, 25 - 29 September 2000
Protection of the Organization's Programme of Work against Exchange Rate Fluctuations
1. Financial Regulation 5.6 states that annual contributions to the budget shall be assessed in United States dollars. It should be noted that, while FR 5.6 allows for contributions to be paid in convertible local currency, its essence is that the assessments are made in US dollars and that Member Nations' obligations are determined in US dollars regardless of the currency of payment.
2. Consequently, all of the Organization's income is effectively in US dollars, but expenditures can be in many currencies, in particular the currency of the host country.
3. The problem for the Organization is that the liability of Member Nations is for assessed contributions in US dollars, while the Organization's liabilities are in many other currencies although, in particular, in Lire/Euro. This problem can be conveniently classified as follows:
4. This is best illustrated by examining a typical biennium with an Appropriation of US$ 650 million approved with a budget rate of Lire 1875 = US$ 1. Such a budget will include expenditure in many currencies with a breakdown which will be something like the following:
|Currency||Percentage of Regular Programme Expenditures|
|Other European-based currencies||6%|
|Other non-European currencies||8%|
To simplify this example, the following comments deal only with the risk associated with the 46% of expenditure in Italian lire. Two possible scenarios are examined:
5. The Lire content of the Appropriation was 46% or the equivalent of US$ 299 million at the budget rate of Lire 1875 = US$ 1; that is, the budget assumed Lire expenditure of Lire 560.6 billion.
6. However, if on implementation, the Lire requirements were bought at the average market rate of Lire 2050 = US$ 1, then the cost in US dollars will be US$ 273.5 million. This implies a saving versus the budget of US$ 25.5 million.
7. The same budget assumptions apply. The only difference is on implementation when the Organization would need to expend US$ 311.5 million in order to buy the required amount of Lire 560.6 billion. In this scenario, a loss of US$ 12.5 million would be incurred versus the budget assumption.
8. In both cases the losses or gains are real as the assessment in US dollars is fixed at the budget rate.
9. The problem between biennia results from the same mechanism but reveals itself in the difference between the budget rate of one biennium and the next.
10. Hence, if in Biennium 1 the budget is as described above (i.e. an Appropriation of US$ 650 million approved with a budget rate of Lire 1875) and then for Biennium 2 the budget rate is different, a loss or gain in purchasing power arising from the exchange rate difference will be experienced. Thus, even assuming zero inflation over the two biennia, maintenance of the same programme (and therefore the same purchasing power) will require a different US dollar Appropriation. Using the same rates of exchange as in the example above, we can construct the following budgets:
|Biennium 1||Biennium 2|
|Lire 1875 = US$ 1||Lire 1800 = US$ 1||Lire 2050 = US$ 1|
|Lire requirements (millions)||560,625||560,625||560,625|
|US$ equivalent of lire requirement (millions)||299.0||311.5||273.5|
|US$ requirement for non-Lire currencies (millions)1||351.0||351.0||351.0|
|US$ Appropriation required to maintain the same Programme of Work||650.0||662.5||624.5|
|Increase/(Decrease) in Appropriation arising from difference in approved budget rate||zero||12.5||(25.5)|
11. The Secretariat, in developing and proposing a budget, and the membership, in approving it, expect the associated Programme of Work to be implemented. As fluctuations in exchange rates, which are clearly external to the Organization, may affect the capacity to implement the Programme of Work, the policy objective is to seek to establish a situation where the effect of such fluctuations is eliminated or, at least, minimized.
12. It is also recognized that such a policy applies equally to avoidance of currency gains as it does to currency losses, particularly as it is assumed that a least-cost solution is being sought.
13. The two risks have been approached in different ways over the years.
14. Up until the 1990-91 biennium, the only protection for the budget was derived through the authority to charge the impact (i.e. either loss or gain) of any difference between the approved budget rate and the actual market (or UN) rate on staff costs to the Special Reserve Account.2
15. However, in 1986-87 the Organization had a very adverse experience. The budget had been approved at a rate of Lire 1760 = US$ 1 whereas implementation occurred at an average UN rate of Lire 1406 = US$ 1. This resulted in losses totalling US$ 31.2 million which eliminated the balance of the Special Reserve Account and forced the General Fund to absorb the remainder of the loss (i.e. US$ 9.4 million). It became clear that the Special Reserve Account was not a sufficiently robust means of protecting the budget and, moreover, its replenishment became problematic.
16. This experience eventually led to the adoption of the forward purchase technique for protecting the Programme of Work against the unforeseen effect of fluctuations in the exchange rate. This technique of protecting the budget has been fully endorsed by the Governing Bodies and by the FAO Advisory Committee on Investments whose guidance has always been to avoid exchange rate speculation.
17. Forward purchase is a form of "hedging" which allows the purchaser to meet future needs of a given currency at a guaranteed fixed rate of exchange. The form of contract undertaken by FAO was to buy forward all Lire requirements at, or close to, the rate at which the budget was approved. It recognized the fact that FAO would never have the cash available to purchase the entire biennium's needs at the spot rate immediately following budget approval. This involves entering into a contract which consists of 24 separate monthly transactions thus allowing the Organization to buy each month's needs at the time that they are actually required but always at the rate fixed in the contract and NOT at the spot rate.
18. The rate in the contract is settled through a telephone tendering process which attempts to obtain the most attractive price (i.e. the highest lire rate to the US$ 1). The factors affecting the rate are straightforward:
19. The former is by the far the most significant as can be seen from the fact that, for example, in November 1995 when the Organization bought forward for the 1996-97 biennium, when Lire interest rates were high (i.e. 3rd quarter 1995 average market rate 8.31%) versus US dollar interest rates (i.e. 3rd quarter 1995 average market rate 5.48%), the forward purchase contract automatically resulted in a gain on exchange of about US$ 14 million. The gain arose because the blended forward rate contracted with the bank was Lire 1669 = US$ 13 which was more favourable to the US dollar than the budget rate of Lire 1600 = US$ 1.
20. However, for the purposes of currency protection in 2000-01, the situation had reversed in that the Lire/Euro interest rates were lower than the US dollar rates at the time that the Organization entered into the forward purchase contract and hence the blended forward rate was Lire 1826 = US$ 14 versus a budget rate of Lire 1875 = US$ 1. The consequence is that currency protection in the current biennium is estimated to cost US$ 8.3 million. While this will be charged to the Special Reserve Account as an exchange loss, it should be recognized that the very favourable situation involving high Italian Lire interest rates experienced for most of the 1990s is not likely to repeat itself and thus reliance on the Special Reserve Account is not a sustainable method of covering the cost of forward purchase unless some other regular source of replenishment is identified.
21. Up until the first Zero Nominal Growth budget approval in 1994-95, the Conference had always followed a policy of separating the issues of "cost increases/decreases" from "programme increases/decreases".
22. Under this approach, budget proposals (i.e. the Outline and Summary Programme of Work and Budget) were all prepared at the exchange rate approved for the previous budget and excluding cost increases (i.e. inflation) from the programme proposals. This allowed consideration of the programme proposals in a truly comparable fashion. Thus, for example, changes in amounts proposed for a given programme were entirely due to the programme itself and not as a result of the different exchange rates or prices.
23. Cost increases were developed separately and considered in detail by the Finance Committee which validated the calculations as being based upon reasonable assumptions. These calculations were made at the Lire/US dollar rate of exchange adopted on the approval of the budget for the previous biennium.
24. The Governing Bodies were also provided with separate information on the impact of different US dollar/Lire exchange rates so that Members were in a position to anticipate the likely effect of the current rate upon their assessed contributions.
25. Finally, on the day of the vote on the Appropriations Resolution at the Conference, a budget rate was set at the "spot" market rate on that morning. The figures in the Resolution were changed to reflect the effect of the new rate under a long-standing methodology agreed with the Governing Bodies.
26. Under this approach, Member Nations were explicitly protecting the Organization's programme from effect of exchange losses and gains and accepting responsibility for such variations.
27. Zero Nominal Growth or negative growth budgets were introduced into FAO from 1994-95 onwards. The ZNG approach was seen as a way of reducing budgets across the UN system (and also some national government spending budgets). It was based on the simple premise that budgets for a future period should not exceed the absolute amount of the budget approved for the previous period. This obliged each programme to absorb the effect of inflationary increases.
28. While some questions were raised about the soundness of the overall rationale and its apparent contradiction with the need to set priorities between programmes, it was generally accepted as a way of reducing budgets and of making sure that the inflationary spiral was terminated.
29. However, in the international environment ZNG had an additional and much more serious effect; that is, it also required each agency to absorb the loss or gain that would automatically arise from any exchange rate variation from one budget period to the next. This inadvertently transferred the management of the inter-biennium risk from the Members to the detriment or benefit of the Organization's programmes. It did not, however, reduce the risk for most Members who still had to contend with the exchange differences arising between their own currencies and the assessed contributions which were stated in US dollars.
30. This approach would also appear to deny the importance of priority setting between the specialized agencies; that is, the relative importance of labour versus health versus food and agriculture was completely subsumed to the vagaries of the currency markets.
31. The current practices of forward purchase and ZNG budget do not provide a satisfactory response to the policy goal of protecting the Programme of Work from the vagaries of currency fluctuation at the lowest possible cost.
32. The options which can address both risks are quite limited and will only work in specific circumstances. These options are briefly described below with the aim of seeking the Committee's guidance on the most acceptable solution. Each of the options needs to be further developed before the full implications can be appreciated. However, there is a need to have the views of Members on the basic policy assumptions under which the options should be developed before further progress can be made.
33. Two alternatives have been identified as suitable for further consideration by the Organization. They are:
34. A further two options, namely increasing the Special Reserve Account and purchasing currency options on forward purchase, were identified but not considered suitable because the former is believed to be unacceptable to Member Nations and the latter has a high cost associated with it. They are, therefore, not pursued any further in this paper.
35. In both of the alternatives under consideration, the Organization's accounting will continue to be performed in one currency, the US dollar. Any attempt to track and maintain both currencies in the financial accounts will not only be tantamount to doubling our accounting effort in financial processing and reporting, but would also be very complex to implement in ORACLE Financials, leading to extensive customization of the package.
36. Each of the alternatives has implications in other areas which will require further consideration and elaboration over the coming months. These include:
37. This is the so-called "split currency system of assessment" which, in FAO's case, would result in a change to the Resolution for Assessed Contributions such that approximately 50% of the assessment would be made in Euro and the balance in US dollars. The split in the assessment would be declared at the time of presentation of the Budget to Conference and would be based on the best estimate of forecast expenditure in Euro currency. This method protects the purchasing power of contributions received for Euro expenses at no cost to the Organization (other than a small increase in administrative cost of the call for and receipt of contributions).
38. It would require a change in the Appropriation Resolution to reflect both currencies by Chapter and some agreed method of automatic adjustment to the Appropriation for the purposes of post facto reporting of budgetary performance both in Statement IV of the Accounts and in the Annual Report of Budgetary Performance.
39. This alternative fully protects FAO's programmes from exchange rate movements both within the biennium and between biennia.
40. Another advantage of this approach is that it is not discriminatory; that is, all Member Nations are subjected to precisely the same conditions as regards their assessed contributions.
41. A number of UN agencies already use this split assessment system with regard to their assessed contributions, namely IAEA (Austrian schillings and US dollars) and UNESCO (French francs and US dollars). UNIDO also uses this approach (Austrian schillings and US dollars) but has recently decided to adopt the Euro, which accounts for 82% of its expenditures, as its accounting currency and will therefore no longer apply the split assessment methodology with effect from 2002-2003.
42. Instead of making split assessments compulsory for all Member Countries, one could allow all Member Countries to pay their assessed contributions in either US dollars or in Lire/Euros at the budget rate of exchange. Financial Regulation 5.6 already allows contributions to be paid in convertible local currency. However, this represents a change in that the liability of those Members selecting Euros for their assessed contributions, would be stated in Euros and not US dollars. All other Members would have their liability expressed in US dollars as at present and regardless of the currency of payment.
43. Approximately 29.7% of total assessments come from the eleven Members which have adopted the Euro. If the UK is included, the total goes up to 34.8 % and if the entire European Region (FAO definition) is included, these countries would contribute just over 40% of the total. Conversely, it is estimated that 46% of FAO's total Regular Programme expenditures are currently paid in Italian Lire and 52% in all European currencies.
44. At present, these and other countries have to use their existing US dollar reserves or have to purchase US dollars with which to pay their current US dollar-based assessments, thereby exposing themselves to currency exchange fluctuations. Moreover, FAO then has to convert a substantial proportion of these US dollars to meet local Lire/Euro-based expenditures resulting in a second exposure to currency exchange fluctuations.
45. Now that the Euro has been introduced, it is expected that Euro-based countries, if given the option, would choose to pay their assessed contributions in Euros, thereby eliminating their exposure to currency exchange fluctuations. At the same time, through countries opting for payment of contributions in Euros, FAO would build up a certain level of cash reserves in Euros with which to meet its local currency (Euro) expenditures. Consequently, the US would be joined by a number of other Euro-based countries in enjoying the facility to make payments to FAO in their own currency and the double exposure to currency exchange rate fluctuations would be eliminated. Other countries outside the Euro area may also choose the Euro as their currency of payment.
46. On the day the budget is approved, Members would be called upon to elect which of the two currencies they choose for assessment, at the budget rate set by Conference, thereby fixing the amounts of contributions which FAO would receive for both US dollars and Euros. In the absence of election by any one country, rules such as the following might apply:
47. Members would not be permitted to change their elected currency until the subsequent biennium.
48. With the above approach, the Organization would know in advance the level of Euro-assessed contributions at the time the budget is approved. It would then be able to plan its Euro-based expenditure accordingly. The alignment of obligations as much as possible to currency inflows would reduce the level of hedging which the Organization currently undertakes. Following such an analysis, the Organization would, if necessary, proceed to forward purchase only for the difference between its expected Euro income and its forecast Euro expenditure, instead of the current practice which is to forward purchase 100% of its local (lire) expenditure. This, in itself, would reduce the amounts of gains or losses resulting from exchange movements although it is recognised that it would probably still require a forward purchase transaction for about 23% to 43% of Euro expenditures which could involve a substantial premium - as much as US$ 2.5 million based on our experience in 2000-01.
49. One of the pitfalls of the above approach is that it fragments the supply and demand of cash from the present single US dollar currency to the proposed two currencies. Consequently, this fragmentation may lead to more short-term exchange transactions and be prone to short-term market conditions.
50. In general, this option is less satisfactory as regards its effectiveness in protecting the Programme of Work from the impact of exchange rate movements and would continue to involve considerable additional cost to the Special Reserve Account.
51. A second alternative would be to obtain approval from Members to set the Budget Rate at the forward rate available in the market on the day of the approval of the Appropriation Resolution by the Conference for a 24 month forward purchase instead of at the "spot" market rate on that day.5 Following Conference approval of the budget, the Organization would immediately forward purchase the full Euro requirement for the biennium.
52. In this way, the Organization is guaranteed the local (Euro) currency equivalent with which to carry out its programme of work. This would fully protect the Programme of Work and Budget within the biennium since the full local currency requirement is forward purchased and any applicable margin funded by Members.
53. However, this approach will only protect the Programme of Work between biennia if Members adopt a policy of changing the amount of the budget to reflect the forward rate - whatever it happens to be. This requires Members to forego the option of pre-setting the budget level at a fixed absolute US dollar amount before the forward rate is known as to do so would put at risk the approved Programme of Work. It follows that the ZNG option in setting the budget would not be possible if this approach were to achieve its policy objective.
54. The Finance Committee may wish to advise the Council:
Financial Regulation 5.6 states that:
"Annual contributions to the budget shall be assessed in United States dollars. To the extent that the Conference, after ascertaining in what currencies Member Nations and Associate Members propose to make their contributions in the ensuing financial period, finds that anticipated United States dollar income will be inadequate to meet estimated United States dollar expenditures of the Organization as determined by the Conference, the Conference will determine the proportionate share of contribution that all Member Nations and Associate Members who do not pay their contributions in full in United States dollars shall pay in that currency. Each Member Nation and Associate Member shall pay the remainder of its contribution in lire, or in its own currency which, for the purposes of its contributions to the Organization, must be freely convertible into lire, the convertibility being the responsibility of the contributing government. The applicable rate shall be the official rate of the lire to the dollar on the first business day in January of the calendar year in which the contribution is due, or the rate in effect on the day the payment is made, whichever is the higher."
1 For simplification of the example, the balance is assumed to be US dollar denominated (i.e. ignoring other currencies)
2 Conference Resolution 13/81
3 The "spot" rate was Lire 1589.50 at the moment the contract was struck.
4 The "spot" rate was Lire 1874 at the moment the contract was struck.
5 The mechanism for doing this requires further elaboration.