The export credit world has been changing for as long as most people who work in it can remember. However, the changes that have taken place in recent years have seemed more rapid, more frequent, and more significant than before. Some of the principal changes, especially those that relate to the role and position of governments, have been mentioned in previous chapters. This chapter reviews these changes in more detail, first in relation to the short-term business of export credit agencies, and second in relation to their medium- and long-term business, since rather different considerations apply in the two areas.
Political Risk: A Popular Misconception
A surprising amount of confusion seems to prevail between, on the one hand, the notion of political risk insurance and, on the other, that of investment insurance. In practice, the two terms are not synonymous, and indeed, the vast bulk of insurance against political risk (especially that supplied by export credit agencies) has been, and is still, issued under export credit arrangements.
This confusion has, however, led a number of people to the view that export credit agencies provide relatively little cover for political risk, since they equate this type of cover with investment insurance coverage. The figures given in Chapter 1 and in Appendix I conclusively disprove this view, showing that, in practice, the political risk cover provided under both export credit insurance and investment insurance is substantial.
Underwriting Political Risk: Public and Private Sector Insurers
With a few exceptions, the conventional wisdom has been that political risks—especially those outside the area of short-term trade finance—are an area for the public sector rather than for private sector insurers and, especially, reinsurers. It is less easy to say with any precision why this has been the case, but at least 10 possibly significant reasons can be adduced for this view:
-
The risks to be covered have been felt to be unpredictable and thus not subject to commercial actuarial analysis and pricing.
-
These risks often come in very large lumps, in the form of high-value contracts, projects, or investments.
-
These risks frequently have very long tails: cover might be sought for a period of 15 years, taking both the construction or manufacturing period and the credit period together.
-
The cases to be insured often arise in difficult circumstances, in high-risk emerging markets or other developing countries, where the private insurance market has had both little experience and little appetite for risk.
-
In both developed and developing countries where cover has been sought, potential insurers perceived an unacceptable level of risk aggregation. That is, overall exposure to a country could build up rather quickly and take too long to run off. This has perhaps been one of the major inhibiting factors.
-
There has also been the belief (quite wrong in the author’s view) that if a buying country should encounter problems, all business done in that country would go into default of various kinds, leading to claims on every piece of business underwritten on the country.
-
Taken together, these factors could produce unbalanced portfolios and an inadequate spread of risk for the insurer.
-
There was perhaps also some feeling that the nature of the major risks involved meant that salvage or recovery was better carried out by government, either bilaterally, on a government-to-government basis, or under the auspices of a formal multinational group (e.g., the Paris Club), rather than by private loss adjusters. A variation of this view was that government insurers (or the governments themselves) could take stronger or more effective action, retaliatory or otherwise, toward host governments to prevent claims or to secure recoveries.
-
The relatively small number of insurers and the almost total absence of a substantial or credible reinsurance market meant that private insurers could not apply their normal approach, namely, breaking the risks into pieces so that each need only take a small share of any large case or of any “catastrophe risks.”
-
Arguably, there was also a self-fulfilling aspect to the absence of private insurance and reinsurance capacity. This kind of underwriting had always been done largely in the public sector, and even where it was not, some of the key decisions (including which risks to accept and what level of premium to charge) were made by governments. This probably helped inspire, or at least harden, the view that such risks were somehow inherently not commercial, and thus not appropriate to the private sector.
As already suggested, one of the main results of this line of reasoning over the last 50 years has been the establishment and spread of export credit agencies and investment insurers that either were public sector institutions themselves or wrote their business on behalf of, or on the account of, their governments. Political risk was simply felt to be too unpredictable for the private sector to insure. Against this background, it is hardly surprising that there was, until the last few years, little real competition from the private sector, especially for medium- and long-term credit business. But are political risks really any more unpredictable than earthquakes or hurricanes or tidal waves, or two jumbo jets crashing over New York? Indeed, it could be argued that many political risks are rather more predictable than such catastrophes.
Changes in Short-Term Trade Finance
One of the most interesting and important developments of the last few years has been the increased willingness of private sector insurers and, particularly, reinsurers to cover the political risks of short-term trade finance. The extent to which this trend may be slowed or reversed by recent events in Southeast Asia and elsewhere is unclear and will depend crucially on the extent of claims and losses that arise on existing business in that region.
As noted in Chapter 2, the maximum credit period for short-term business is normally six months, which has meant that some of the traditional difficulties perceived by the private sector about operating in this area have become less relevant. In particular, there has also been a growing willingness in the private sector—both insurers and reinsurers—to consider underwriting political risks. One tendency, therefore, has been for those insurers seeking reinsurance to combine the political and commercial risks into a single reinsurance package. This is also relevant to the point made elsewhere in this volume that the distinction is increasingly blurred between the traditional definitions of political and commercial risk.
Thus, it is perhaps not totally surprising that this whole area, including both investment insurance and project business and long-term credit (see below), is undergoing significant change. Nor is it surprising that the private sector, in the form not only of capital market investors and lenders but also of private insurers and reinsurers, has been increasingly willing to consider accepting political and other risks without seeking any involvement from export credit agencies or their governments.
There may later be scope for extending the new techniques being developed for increasing private sector capacity for insuring risks on natural catastrophes, such as hurricanes and tidal waves (through bonds or alternative risk transfer mechanisms), into the area of political risk insurance. These techniques would be an addition or an alternative to traditional reinsurance.
Perhaps the fundamental change is that not only are private insurers now prepared to look at these risks, especially short-term political risks, but that there has been a growing capacity and willingness in the reinsurance market to take on some political risks. This offers the possibility of expanding private involvement from the short-term area into the medium- and long-term end of the business, by breaking large transactions (or political risk catastrophes) into relatively small packages and spreading them over a large number of underwriters, both insurers and reinsurers. Arguably, the basic question is not so much “Why does the private sector now seem willing to underwrite political risks?” as “Why has it taken the private sector so long?”
Interestingly, this development has more or less coincided with the wish of a growing number of governments to reduce their provision or support of short-term export credit insurance. Increasingly their policy stance is that, if an activity can be carried out in the private sector, it should not be carried out in the public sector. The public sector, in this view, should not crowd out or compete with the private sector. This notion of “complement, not compete” has long been, as noted earlier, the theoretical position of at least some governments, but it has only become a practical possibility as the capacity and willingness of private sector reinsurers to provide political risk insurance have increased.
This trend was given impetus by developments within the European Union and, especially, by the decision in the United Kingdom in 1990-91 to privatize the whole of the short-term business of its export credit agency. In the European Union these developments led to the evolution of the concept of “marketable risks,” defined as those risks that are acceptable to private sector insurers or reinsurers. The remaining “nonmarketable risks” are those felt still to be appropriate for governments to insure.
However, these concepts, if they are to be of any practical value, cannot be static but rather must reflect the evolution of the private sector. They should not in any sense stultify or inhibit that evolution. It is the private market itself that will decide what risks it will accept—not governments or European Commission officials. It is one thing for the European Commission to say what cannot be done by or in the governments of EU member states. But it is quite another thing for the Commission to purport to say what categories of risk will or will not be underwritten by insurance and reinsurance companies in the private sector. That is simply not the way markets work. These are business decisions that will be made by the private sector parties themselves, on the basis of whether or not they find certain risks acceptable. And as recent events in Southeast Asia and Russia demonstrate, these are not static concepts. Thus, when private sector insurers or reinsurers see the underwriting of some political risks as acceptable business, or when there are strong marketing or commercial reasons for finding this business attractive, they will do this business. They will not be guided by what the Commission may have called at some point in time a “nonmarketable” risk.
A parallel and very significant development has been the emergence of credit insurers operating across borders, offering their customers—especially the large multinational companies—facilities that embrace not only domestic and export trade within and from one particular country, but also domestic trade within and exports between other countries. Such arrangements could hardly be structured on the basis of government or public sector reinsurance, which is normally or primarily available only for national goods and services, with the exceptions referred to elsewhere in this volume.
In addition, the need for high-capacity computer systems to obtain, store, and update credit information on literally hundreds of thousands, if not millions, of buyers worldwide imposes huge costs. (One large credit insurer recently claimed that it held information on no fewer than 20 million buyers worldwide.) This means not just that only the largest insurers will be able to support and bear such expenditure, but also that they will wish to spread this cost over the largest possible amount of business. For example, if an insurer is willing to underwrite the political or commercial risks on a buyer in, say, South Africa or Peru, it will not matter whether the exports originate in Germany or the United Kingdom or the United States.
Two figures illustrate these developments quite starkly. First, it is now estimated that up to 20 underwriters are offering credit insurance cover in the London market alone. Second, between 85 and 95 percent of short-term export credit insurance business within and beyond the European Union is now underwritten by insurers and reinsurers in the private sector, without the involvement of governments.
Changes in Medium- and Long-Term Project Business
In considering recent changes in the area of medium- and long-term business, three background factors are important: the debt crisis of the 1980s, international discipline, and competition.
The Debt Crisis
As the appendices to this volume show, many export credit agencies have suffered severe consequences from the international debt crises of the 1980s. In some years they paid claims significantly in excess of their premium income, and most of them—at least with respect to their medium- and long-term credit activities—exhausted their reserves and began to accumulate deficits.
In addition, in recent years the trend has been toward forgiveness of countries’ external debt, in the form of Paris Club reschedulings for low-income countries. Most recently these negotiations have been conducted under so-called Naples terms (which call for 50 to 67 percent reductions in the net present value of debt) and Lyon terms (up to 80 percent reduction). For this and other reasons, the recoverability of some claims that export credit agencies have paid has worsened or disappeared, and this has had an impact on their balance sheets. Moreover, as export credit agencies and their governments have steadfastly refused to change the cutoff dates for debts that may be rescheduled under Paris Club arrangements, preferring instead to rereschedule only preexisting Paris Club debt, the recoverability of some of that debt has not improved.
As already noted, in these circumstances most export credit agencies and their governments or guardian authorities have had to review both their premium rates and the amount of new cover they can provide. These factors have opened a window of opportunity for private insurers and reinsurers to seek out new business.
International Discipline
For some time there has been more or less general agreement that a credit war or international export credit subsidy war is not a great idea, especially from the point of view of taxpayers in the exporting countries. There is also now the discipline provided by the WTO Agreement on Subsidies and Countervailing Measures, which requires export credit agencies to break even over the long run. Export credit agencies may be subject to action in the WTO if they do not conform to these rules.
Meanwhile, in the forum of the OECD Arrangement there has been, as already noted, a fairly consistent trend toward squeezing interest rate subsidies out of the system. Thus, the special, fixed interest rates offered by export credit agencies for credits of more than two years, which had acted as a magnet attracting business, have been brought closer and closer to market rates. This trend continues.
However, as the blanket subsidies were squeezed out, the temptation arose to introduce selective subsidies, largely through the blending of bilateral aid and export credit into so-called mixed credits. But these arrangements, too, are now subject to increasing scrutiny, prior notification, reporting, and control within the OECD. This trend also continues.
A recent development has been the recognition that the premium rates charged by different export credit agencies for medium- and long-term credit have differed widely, causing distortions. Although this development is not likely, in the short run at least, to produce fully harmonized premium rates, there is now agreement within the OECD on minimum or floor rates. And after a short transition period, premium rates from different export credit agencies for political risk cover will be much more alike.
Two other factors are important in making official export credit agency facilities for project financings less attractive and private facilities more attractive. The first is that, until very recently, the credit terms mandated under the OECD Arrangement (especially the starting point of credit and the repayment profile) were too rigid to meet the cash flow and amortization needs of many project financings. In these cases, where repayment depends on the profile of cash flows generated by the project itself rather than on the financial strength of the buyer or borrower, the need for specially tailored repayment profiles is acute.
The second factor is the increasing tendency of official export credit agencies to add various conditions and requirements to their facilities, such as environmental requirements, standards on the use of child labor, or the obligation that the host government recognize trade unions. One result of these conditions, and the delays and the administrative burden they cause, is that lenders and investors may investigate the possibility of getting private market cover before turning to the export credit agencies. In other words, for practical business reasons, exporters and their banks may themselves begin to regard export credit agency cover as a last resort.
Competition
In the past, export credit agencies tended to be national, and so competition in the export credit business was between countries. However, as noted earlier, in the area of short-term export credit the role of government is being reviewed in many countries, and competition is increasingly between insurers, and thus between export credit agencies. Contrary to the belief of some, this phenomenon is not restricted to the United Kingdom or the European Union.
There is also a trend, or was before the recent events in Southeast Asia, Russia, and Latin America, for some lenders and investors and capital market funders to perceive less of a need for insurance protection from the export credit agencies—that is, to rely on “self-insurance.” At the same time, the private sector (both insurers and reinsurers) has shown increasing interest and activity in covering at least some political risks related to project finance and investment insurance.
This trend, coupled with the potential appetite for risk on the part of banks and, especially, capital market lenders, means that export credit agencies no longer have a monopoly product. Indeed, as discussed earlier, some developments such as the phasing out of subsidies, the raising of premium rates, and the adding of onerous conditions have made export credit agencies’ facilities less attractive and private market facilities more so.
Recent Trends in Project Financing
The traditional security for export credit agencies in large projects in non-OECD countries has been government or sovereign guarantees. However, one important background event, noted above, has been the large number of buying and borrowing countries that defaulted in the 1980s and have gone to the Paris Club to reschedule their bilateral debts. This and the trend toward debt reduction and debt forgiveness for low-income countries have led to export credit agencies paying huge claims ($150 billion over the last 10 years or so), and most have also suffered substantial losses. The impact on export credit agencies and their governments remains substantial. And again, export credit agencies are not aid agencies—they are expected to break even over time.
The inevitable disenchantment of the export credit agencies with sovereign guarantees from many borrowing countries has also coincided with a shift in many countries toward privatization. This, too, is partly the result of disenchantment with sovereign guarantees, on the part of those who formerly provided them. Privatization in its many manifestations has had a major impact on how projects are now financed, not least because, with increasing frequency, host governments are no longer willing to be the borrower or guarantor of repayment for lending to projects, even infrastructure projects. Together, these factors have resulted in much greater interest in project financing (also known as limited recourse financing), where the main security is the viability and cash flow of the project itself rather than the general financial strength of the buyer or guarantor.
However, as with many things in life, converting theory into practice is easier said than done. Project financings present a whole new range of challenges and problems, not only for export credit agencies and banks, but also for exporters, contractors, and investors. One key question is whether it really makes sense for borrowers, host governments, lenders, investors, and those who insure them to finance large numbers of infrastructure and other projects in foreign currency. Most of these projects, after all, generate little if any foreign exchange to repay the debt incurred. This question is likely to be raised much more frequently and vigorously in the wake of recent events in Asia and elsewhere. It should also be asked of the international financial institutions, which encourage such financing through their guarantee programs. There should surely be much greater emphasis on encouraging the local financing (or at least the financing in local currency) of large projects that do not earn foreign currency.
Project financings are now a major activity for export credit agencies, especially the largest, all of which have now brought several cases to financial completion. As noted above, because (among other things) of the international debt crisis of the 1980s, there is significant disenchantment with sovereign guarantees, both on the part of those who gave those guarantees and on the part of those who received them as the main security for their activities. In addition, the international financial institutions are encouraging more and more governments to privatize and decentralize and to disengage from industrial and commercial activities. This has led to a proliferation of project financings.
It has also led, however, to some projects that are inherently government projects masquerading instead as private projects. It is not clear that disguising public projects as private ones helps anyone, especially when problems arise, for example when a devaluation brings about a need for large tariff increases to repay foreign creditors. This is a different point from the one about the dangers of financing externally projects that do not earn foreign exchange. Here the point is that some projects are inherently and intrinsically government projects, and that where, consequently, the host government is asked, as part of the security package, to accept a large number of understandings, undertakings, and obligations, the practical result is, when taken together, a host government guarantee, but one with maximum complexity and minimum clarity.
In any case, should not the international financial institutions ensure that such undertakings are carefully recorded and “scored,” so that the true extent of the government’s obligations is known and counted against IMF-agreed limits or ceilings on public sector borrowing? These undertakings are, strictly speaking, contingent, but they are nonetheless real, and at the very least they should be acknowledged in the form of footnotes to financial statements or perhaps as a separate category of obligation.
Impact on Export Credit Agencies
The fact that an increasingly large percentage of project business is now primarily structured on a private sector or privatized basis has produced significant challenges for all parties, including both export credit insurers and investment insurers. One challenge is that the composition of political risks has changed. Not only do many of the traditional risks remain, but a range of new risks have arisen, which could affect the ability of borrowers in a project financing to generate sufficient income—in particular, foreign currency income—to repay loans. For example, in many telecommunications, electric power, and water projects, a key part of the security package is the willingness of the host government to allow tariffs to be adjusted to the level necessary to repay foreign creditors. But when, for whatever reason, a shortage of foreign exchange materializes, or when huge pressures on the exchange rate lead to a significant depreciation, this willingness to raise tariffs inevitably comes into question. Governments may be quite reluctant to see tariffs for sensitive items like water or power increase substantially for domestic consumers (who are also voters), simply to repay foreign creditors. The government may even lose its enthusiasm to continue with the project at all. In any case, sharply higher tariffs will choke off local demand for the project’s output, and the general recession or slowdown that often follows a sharp devaluation adds to this effect, thus raising questions about the viability of the project.
More generally, as countries liberalize their economies (including their capital accounts) and attract large capital inflows, the risk of equally large and possibly sudden capital outflows looms large. Exchange rates will increasingly take the initial strain both of foreign exchange shortages and of short-term capital outflows. But a significant devaluation will have a serious impact on all local importers with foreign currency-denominated debts to overseas suppliers. More defaults and insolvencies can be expected, and this will have an early and substantial effect on the export credit agencies that have insured these debts—especially as regards their providing new cover except as part of a workout to protect existing exposure.
But a devaluation may also lead to pressure on host governments to take over at least some of the responsibility for importers or projects or bank debts (or to give them exchange rate protection) in exchange for rescheduling their debts. This raises a range of practical issues, and not just that of moral hazard.
Defining and Apportioning Risks
As a general rule, it is in the area of political risk that the export credit agencies are most experienced. However, that is far from saying that there is a clearly defined set of risks universally accepted as political. It is also misleading to think that political risks arise only in emerging markets or other developing countries. Also in this category are such things as a radical new environmental law or regulation imposed on a half-completed power project in, say, California.
Some differences among export credit agencies are relevant here. There are, perhaps, two main approaches. Some export credit agencies, including those of Canada, Japan, the United Kingdom, and the United States, either lend directly or, when the lending is done by others, usually give 100 percent and unconditional guarantees. (As explained in Chapter 2, this means 100 percent of the 85 percent of the amount financed.) Other agencies, such as those of France, Germany, and Italy, provide conditional insurance and less than 100 percent cover.
But in such cases, and especially where risk sharing with banks or noninsured lenders or investors is involved, it is necessary to be as clear as possible about which risks are covered. Here, as in so many other contexts, it is not a good idea to wait for a default to find out whether a particular risk is insured and thus who is responsible for the loss.
At the risk of caricaturing the position, the banks that receive export credit agency facilities would probably prefer a risk-sharing approach that specifies the risks not covered. These they would regard as commercial risks and all other risks as political and consequently for the export credit agencies to cover. The export credit agencies, on the other hand, would prefer the political risks covered to be listed and all other risks to be commercial and for the banks.
Two examples may help illustrate some of the problems.
Documentation Risk. For many export credit agencies the traditional policy has been that exporters and their banks are responsible for their own documents. Thus, if a claim arises because a document is faulty or not enforceable, this is not an insured risk and the claim is not payable. If an export credit agency stipulates a particular kind of guarantee, it will normally not examine or approve the guarantee at the time it is obtained. The key stage is then the claims stage.
Obviously, for project financings this is a very important issue, not only because the documents can be a meter high, but because in some buying countries—for example, in Eastern and Central Europe—key aspects of the new legal framework may remain unclear or untested in the courts.
Export credit agencies naturally worry that, if they do inspect the documents, for example a take-or-pay contract, it will be very hard for them later to disown responsibility if a claim arises in which the legal enforceability of the documents becomes a key issue in court. As with so many issues in this area, it is very dangerous to generalize, and thus the one thing one can say about the question “Who bears the documentation risk?” is that it is a very good one. But it is a question that is best faced at the outset, not later when problems arise.
Actions and Inactions by Governments. The issue of action or inaction by a host government giving rise to nonpayment and thus to a claim is also much more difficult in project financings. This is especially true of infrastructure projects (but not restricted to them), because in such projects different agencies and levels of the host government can play so many different roles. The viability of a project can depend, for example, on certain tax breaks, and thus on the government not changing its tax policy. Or the project may depend on an export license for the output being approved, or on certain export or import duties being waived or set at a fixed level. Or the government may be the supplier of crucial feedstocks or of fuel or raw materials. It may be the main purchaser of a project’s output, either directly or through a wholly owned or controlled vehicle company. It may be the ultimate decision maker about tariffs or output prices. Or it may be necessary that the government sanction the transfer of repayments, profits, or dividends.
This is not, of course, an exhaustive list, but it demonstrates that the role of government is much more complicated than that of a simple buyer or borrower. Actions or inactions in any of these contexts could raise problems for the project, resulting in failure of the project to earn the income or profits it might have otherwise, possibly resulting in default.
This brings us back, of course, to documentation risk. How does one ensure that all the things that government says it will and will not do are catalogued, preferably in a single undertaking or letter? Apart from the clear risk of leaving something out, who should sign such a letter, and is it clear that those persons can commit the full faith and credit of the host government in all relevant contexts? This is particularly important where different levels of government—central, regional, local—are involved, as the relations between these levels vary from country to country. Finally, and importantly, should the letter specify any kind of penalty on the government if it should breach any of the undertakings? A third example of the blurred risks that arise in both short-term and medium-term business is what happens when a country experiences a foreign exchange shortage. In the past such shortages led to transfer delays and so to political risk claims. Now and in the future, however, they are likely to lead to currency depreciation and to the default or insolvency of buyers who can no longer afford to purchase the foreign currency they need to repay external creditors. Are these political or commercial risk claims?
These considerations indicate that there is now a very important—and growing—gray area of risks that are not easily categorized as either political or commercial. Many risks are both political and commercial, depending on the circumstances, and this has led some people to argue that the distinction between them is artificial or out of date. In practice, the key point is identifying and defining the various risks and deciding who shall carry them. This seems more important than categorizing a given risk as either political or commercial. It is not helpful to wait for problems or claims to arise before deciding who is carrying this or that risk.
Infrastructure projects present special problems in this regard, as indeed do any projects that do not earn foreign currency that can be held offshore in an escrow account. Developing an acceptable security package can then become even more difficult and time-consuming. Problems can also arise if some lenders or investors—including the international financial institutions—are not prepared to accept a pari passu position (that is, equal treatment with other lenders and investors).
Against this background, export credit agencies face certain constraints and certain problems. For example:
-
Expertise. Export credit agencies cannot have experts in all sectors.
-
Analytical capacity. Export credit agencies are limited in the number of projects they can examine at any one time. And projects whose financing takes four or five years to structure (e.g., the Hub River project in Pakistan) are a dubious model for anybody or anything.
-
The effects of the debt crisis. Export credit agencies bear the scars of the 1980s debt crisis, and many have lost substantial reserves. This has led to unprecedented levels of scrutiny, by legislatures and ministries of finance, in most exporting countries.
-
The overenthusiasm of some project sponsors and advisers. Advisers to a project may be paid on a daily basis rather than by results. Their eagerness to earn fees may clog up the system by lodging a host of impractical applications to be examined (see the section below on “Projects That Do Not Happen”).
-
Coordination. Export credit agencies often encounter difficulty in designing and implementing risk-sharing and multisourcing arrangements (i.e., projects where suppliers and insurers in a number of countries are involved).
-
Lack of understanding. Export credit agencies must deal with a continuing misunderstanding on the part of many—some of whom should know better—about their role. Again, export credit agencies are not sources of aid, nor are they sources of untied finance. They are not solely or even primarily vehicles for industrial support or foreign policy.
Projects That Do Not Happen
Projects that are brought to financial closure are in some ways the easy ones, since there is at least a chance that those involved will recover their costs. Much more difficult are the projects to which agencies and their clients devote time and resources but that ultimately do not happen. And for any export credit agency this can be seven out of eight. Who can or should bear the costs of these projects—the costs of analysis, specialist advice, and the like? There is no common policy on this among export credit agencies, but it seems likely to become a question of increasing importance and difficulty. The temptation is, of course, always to see this as a problem for someone else.
One possible answer is for more export credit agencies to levy nonrefundable charges of various kinds for the preliminary work they undertake on projects. All or part of such charges could then be netted off against the premiums on those projects that do happen.
A similar problem arises where an export credit agency works on a project for months, if not years, and finally the project comes to fruition—but the lenders or sponsors decide not to use export credit agency support but instead turn to the capital markets. Or they may switch the funding base from an export credit agency package to a capital market package when construction is completed. In either case the export credit agency receives no premium—or a much reduced premium—for its work. It must therefore look for some way to recover these costs, perhaps by way of a charge where loans are prepaid.
These problems may not be discussed frequently or publicly, but they do arise and need solving. This will not happen if they are ignored or swept under the carpet.
Precompletion Risks
Another difficult area is that of precompletion risk: who takes the risks of noncompletion of a project, and what is the proper role of export credit agencies in this area? This has been an area of some change in recent years. Initially, some export credit agencies did not wish to cover any risks until after projects were commissioned. Their feeling was that the project sponsors should take the precompletion risks, relying on completion guarantees of various sorts from contractors and other suppliers. This partly reflected a view that precompletion risks were essentially commercial risks and thus more appropriate to the commercial banks and to contractors than to export credit agencies.
However, this has proved not to be a sustainable position—not least because political risks can clearly arise in the precompletion phase. More export credit agencies have become willing to cover at least political risks prior to completion, and others will look at a wider spectrum of risks during this phase. In this area, as in some others, a more common and consistent approach among export credit agencies has begun to evolve.
Past Debt Situations
In the past, mechanisms and structures have been developed for dealing with countries experiencing problems in repaying large debts. Although a detailed description of these is beyond the scope of this volume, certain points are relevant.
First, the London Club has become a forum where commercial banks negotiate with debtor countries. Second, the Paris Club is where official debts, and especially debts owed to export credit agencies, are rescheduled. It is sometimes said that only debt owed by public sector debtors is covered in this forum. However, if an export credit agency has paid a transfer claim on a debt owed by a private entity, and if the debt meets the criteria of the relevant debt rescheduling agreement (e.g., if it has been guaranteed or assumed by the debtor government), it can, in principle, be included. In practice, little private sector medium- and long-term debt has been included in Paris Club negotiations. But this mostly reflects the fact that, in the past, most of the medium- and long-term debt of countries going to the Paris Club has been government or public sector debt.
In addition, short-term trade debt is normally excluded from Paris Club negotiations. One reason is that, traditionally, this debt is smaller than medium-and long-term debt. But another reason is the sheer administrative burden on debtors and creditors of including such debts, which can number in the thousands, and many of which are of relatively low value. Another reason may be that there is some hope that, if such debts are paid, then trade facilities may be kept in place. However, arrears on short-term debt have been deferred over relatively short periods (one to three years) and so included in a number of Paris Club rescheduling agreements over the last few years, especially for the poorer countries. Investment insurance debts are also excluded, by convention.
Other creditors that have tended to escape countrywide reschedulings under Paris and London Club auspices include the holders of forfait debts and bonds as well as equity investors. Preferred creditors and those benefiting from the umbrella of the international financial institutions (e.g., commercial banks under World Bank, International Finance Corporation, or European Bank for Development and Reconstruction “B loan” arrangements) are also excluded.
The position of uninsured trade creditors is more complicated. They, like some other non-London Club and non-Paris Club creditors, apart from those enjoying preferred creditor status, are sometimes caught by the debtor country simply not having the money to pay them, and/or by the Paris Club’s comparability-of-treatment requirement. Under this requirement, for a Paris Club rescheduling to remain in force, the debtor must agree not to pay other “similar creditors” on terms more favorable to the creditor than those agreed with Paris Club creditors.
So much for the past. A number of interesting questions arise for the future. Some of them stem from the fact that, in the past, the “normal” debt problem has been one relating to medium- and long-term public sector or sovereign debt. In the future, however, when account is taken of a number of factors already discussed, problems may arise in countries where most of the difficult debt is short-term private sector debt or debt related to large project financings. It would, in addition, be fairly common ground among many creditors (and especially official creditors) both that moral hazard should be avoided to the extent possible and that no creditors should be given a free ride. In other words, lenders, purchasers of bonds or other securities, investors, and insurers alike should bear the consequences of their own decisions and actions and should not expect to be bailed out by others.
In this and other contexts, however, it is almost certainly not helpful to debtor countries to build up a large burden of external preferred creditor debt, and therefore the international financial institutions should be very cautious in extending their umbrellas or their guarantees to other creditors. Not only does a large volume of preferred creditor debt restrict the freedom of action of debtors, but potential future creditors will almost certainly not see a large volume of such debt as a positive feature.
In addition, in a liberalized world where economies are more open and exchange rates more flexible, shortages of foreign exchange may manifest themselves in different ways. In the past they tended to result in delays and in sovereign default, and ultimately in debt reschedulings under Paris and London Club arrangements, where, among other things, the export credit agencies paid political risk and transfer claims. However, foreign exchange shortages today may lead first to depreciation of the currency. Such shortages can, of course, arise from a range of reasons, not all of which are due to failures of policy on the part of the debtor country government. This may lead to individual buyers being unable (or unwilling) to purchase foreign exchange at the new, higher rate. There may also be problems for banks that have opened letters of credit.
The result may then be buyer default or insolvency, or the failure of banks that have opened letters of credit on buyers’ behalf. Export credit agencies may then face large claims, but these will almost certainly be commercial rather than political risk claims—that is, claims in respect of default or insolvency rather than convertibility or transfer claims. This point is also of relevance to reinsurers, both governments and those in the private sector.
In addition, for large projects—and especially those structured on a project finance basis—workouts may have to be on a case-by-case basis rather than through across-the-board access to the Paris or the London Club. They will be especially relevant for projects that do not generate foreign currency and where, for whatever reason, tariffs or other prices may not be raised to offset a significant depreciation. These will be made more complicated, difficult, and controversial if some creditors refuse to accept pari passu treatment.
None of this is to say that the Paris Club has totally lost its relevance and has no part to play in future debt situations. As Indonesia, Pakistan, and Russia have recently demonstrated (in different ways), there may well need to be a government-to-government forum—even if it has to be reconstructed—for discussing debt. All of these points together suggest the need to review the future role of the Paris Club and to decide whether to continue to exclude or, preferably, not exclude some categories of creditors, especially capital market lenders, from equal burden sharing. Also needed, more immediately and more importantly, is an early and radical reappraisal of the whole structure and mechanisms for dealing with countries with debt problems. Of course, this is much easier said than done, but it will have to include how to treat bondholders, who cannot expect other creditors to carry the total burden simply because of the way the bond documentation has been prepared or because no provision has been made for what to do in the event of problems.
Setting Country Policy and Limits and Premium Rates
As already noted, export credit agencies are such a heterogeneous group that it is dangerous to generalize about them. This is especially true in the area of country limits and controls and in the area of premium rates. Moreover, rather different considerations apply in short-term business than in medium- and long-term business.
Short-Term Business
Most export credit agencies that underwrite short-term business do so on a comprehensive basis. A framework policy is issued, and individual buyers are then normally underwritten subject to individual credit limits, which are usually valid for more than one shipment or contract. Exporters are often given discretion to perform their own credit assessment up to certain limits (so-called discretionary limits) without prior or specific approval by the export credit agency. In other words, the business is not usually underwritten on a case-by-case basis but on something approaching a conveyer belt basis. As far as is practicable, premiums are collected on the same basis. The aim is to keep administration as streamlined as possible in what has traditionally been a rather labor-intensive activity. The ratio of administrative costs to premium income in this area tends in any case to be higher than in other categories of insurance. One result is that it is not always easy for an insurer to know with any degree of precision exactly what its exposure on short-term business is on an individual importing country, much less an individual buyer or sector, at any point in time.
Country Policy and Country Limits
For these kinds of reasons, it is not usually practicable for an export credit agency to exercise control over its exposure in a country on short-term business by means of a country limit, although this may be possible in exceptional circumstances. And as already noted, short-term cover has normally been excluded from Paris Club rescheduling agreements.
Thus, country policies for short-term business tend to concentrate on whether it is possible to offer any cover at all on the country concerned. Except where there is actual default or some other very serious situation that throws significant doubt on all buyers, underwriting policy tends to concentrate on a number of underwriting controls on individual cases, rather than on overall country limits.
For example, in serious circumstances, all existing credit limits on a country may be withdrawn, and exporters may be required to apply for limits only when they have a contract in hand. Any limit then issued will apply only to that contract or shipment. Or the export credit agency will make all or most cover conditional on there being letters of credit, which reduces the commercial risks on individual buyers. Or cover may be restricted to government buyers or, if the host government itself has a bad record, to private buyers. Or it may be restricted to certain sectors (e.g., essential imports such as food or Pharmaceuticals), or to the largest or most creditworthy customers. Agencies may also make cover available only where the exporter has a long and good track record with the buyer concerned. Cover may also be made subject to payment by letters of credit from specified “acceptable” banks. Or in extreme circumstances, cover may only be available where letters of credit are confirmed by banks outside the buying country; in these cases the cover may not go into effect until after the letter of credit has been opened or confirmed. It is rare, but not unknown, for cover to be subject to host government guarantees or letters of credit issued by the central bank. Sometimes (see below) premium rates may be increased or subject to a surcharge. This may of course choke off some business, but not necessarily the highest-risk business, which can often bear the greatest margins.
All of these underwriting measures probably have a blend of five objectives:
-
To quantify existing exposure
-
To reduce overall exposure
-
To maintain the lowest-risk business while avoiding the worst risks
-
To find a way to maintain at least some cover without causing losses or problems with reinsurers (be they private or public), and
-
To protect existing exposure, for example on outstanding payments on goods and services already delivered.
However, the balance among these objectives may vary from country to country and in the same country under different circumstances.
Premiums
Most short-term insurers will look to adopt a premium system that is easy and cheap to administer and so will not be keen to have a system where individual cases are rated separately. In this area, as in others, the approach will differ from that for medium- and long-term business (see below). But all insurers will look for sufficient premium income (but not necessarily a premium on each and every case) to cover the following:
-
Administrative expenses
-
Expected losses (often for this class of business as a whole rather than for each buyer or buying country), and
-
Some contribution to reserves, should forecasts of expected losses prove to have been overoptimistic.
As noted above, for some buying countries that are perceived as involving higher but still acceptable risks, an addition or surcharge may be made to normal premium rates.
A continuing problem is how to cater to small exporters, a group that most private insurers are not overly anxious to have as customers. The dilemma is whether, on the one hand, to offer small exporters some kind of preferential premium rate or, on the other, to charge them for the higher administrative costs that they normally involve for insurers. This question, which is examined in more detail in Chapter 4, is an important one, especially for government export credit agencies. There is no simple answer, but one thing is certain: if export credit agencies are required by their governments or guardian authorities to charge premium rates that do not cover costs, the shortfall has to be made up elsewhere. Moreover, a government subsidy to do this would be contrary to WTO rules, and any attempt to pass the costs on to larger exporters would simply cause them to look for other insurers. Many export credit agencies find themselves in this dilemma, and some, such as the U.S. Eximbank, have lost many of their large short-term customers partly as a result of this.
There are no other international (OECD or Berne Union) agreements or understandings or requirements on short-term premium rates, and given the number of underwriters prepared to do this kind of business, it is subject to very strong competition, especially on premium rates. Overall, it is probably unusual for short-term premium rates to be high (i.e., above 1 percent). In any case, raising rates is not seen as an effective way of avoiding the riskiest business. After all, it is precisely the highest-risk business that offers the highest potential profit margins, and thus the greatest ability to pay higher premium rates. Risk is better avoided by means of sound underwriting techniques.
Medium- and Long-Term Business
Capital goods or projects sold on medium- or long-term credit are a very different area from short-term business, for several reasons. First, cases tend to be large and to be underwritten individually, as the subject of separate facilities. Even so, because contract values are larger and premium rates higher, administrative costs represent a much smaller percentage of premium income than for short-term business. In addition, as noted earlier, medium- and long-term business tends to be subject to Paris Club settlements, which apply to all appropriate contracts signed before the agreed cutoff date. Meanwhile the whole area of project business is undergoing significant change, with the decline of sovereign risk business and the growth of project financings of all kinds, even in infrastructure. It is also in this area (for credits of two years’ maturity or more) that the discipline of the OECD Arrangement applies.
Country Policy
Assessing political risks is always difficult. But if it is a challenge for short-term business, it is even more of a challenge where the credit period spreads over 10 years or more. It is clearly impossible to do this kind of assessment in a way that allows total confidence in the results. But this does not mean that risk assessment should not be attempted at all.
The key task in such an assessment is to evaluate the prospects of a country being unable at some future date to service its external debts in a timely way (i.e., the transfer risks referred to earlier). Part of the analysis will be backward looking, to review the country’s track record of debt repayment and its past debt-service ratios. But part of it must inevitably be forward looking as well, with all the uncertainties that implies.
This volume is not intended as a textbook on country risk analysis. Suffice it to say that those charged with this task, in the export credit agencies and elsewhere, should seek out the best, most reliable, and most comprehensive data, particularly on countries’ external debt and, especially, its repayment profile. Particularly important, under today’s circumstances, is whether the external debt is mainly short term or mainly medium to long term. External debt obligations must be measured against possible future export earnings, capital flows, and a series of ratios on debt and debt repayments to GDP, reserves, export earnings, and the like.
As described elsewhere in this volume, forecasting default is even more complicated today than in the past. At the risk of caricaturing the position, in the past a shortfall in foreign exchange receipts tended to lead to repayment delays and eventually to reschedulings in the Paris and London Clubs. However, in a more liberalized world, and under flexible exchange rates, foreign exchange outflows and shortfalls may arise suddenly and may primarily reflect short-term capital movements. The strains of such a shortfall will tend to manifest themselves—initially at any rate—in problems stemming from currency depreciation, as buyers and borrowers are forced to come up with additional local currency to purchase foreign exchange to pay their debts, rather than in payment delays. For the export credit agencies, one of the implications is that, again, the traditional distinction between political and commercial risks will be blurred.
The traditional and probably still the most common method of applying country policy in the area of medium- and long-term credit is to grade countries into one of four to six country categories and to apply a separate ceiling or limit on one’s exposure to each country or, possibly, to each category. A country limit can be defined in terms of new business signed in a particular year, or in terms of total exposure outstanding, or by the amount of repayments falling due in any particular future year or run of years. The ceiling may also be set with reference to the volume of business the agency has covered in the past, to possible demand, to a maximum percentage of the agency’s total worldwide exposure, or some combination of these. But setting such ceilings is not a very exact science, and they are normally reviewed when demand for cover from exporters and their banks exceeds supply under existing ceilings. History and experience show that it is not easy for guardian authorities or export credit agencies themselves to turn away business on a country that is not experiencing payment difficulties simply for reasons of portfolio balance. Thus, although it is unusual for an export credit agency to be totally off cover for a buying country, it is equally unusual to be totally on cover, without limits or underwriting restrictions of any kind.
Of course, external debt repayment problems can be caused by problems other than foreign exchange shortfalls, and therefore an agency’s country analysis must also consider the risks of political instability and war and civil war. But the key point, based not least on the experience in the Paris Club over the last 20 years, is that the major sources of political risk claims are transfer and convertibility factors.
Two vital tasks in assessing country risk and in making country policy are, first (as noted earlier), to gather the most timely, full, and authoritative information possible, and second, to determine the positions of other export credit agencies doing business with the countries in question. Thus, it is not helpful for debtor countries to be other than as open as possible on the classification and publication of information. This, as noted elsewhere in this volume, is very important for all actual and potential creditors, not just official ones, and so in the context of moral hazard (i.e., if creditors are to share fully and equally in the burden of debt problems, they should also have full and equal access to relevant information).
The international financial institutions—and the IMF in particular—have a role here, for example in facilitating the flow of information and in verifying its accuracy. These institutions should also be involved not only in their current activities of setting limits on particular categories of debt but also in enabling creditors and potential creditors to access the “scores” and in trying to ensure that debtor countries’ scorekeeping in this regard is as accurate as possible. This means, among other things, that so-called quasi-government debt should not be omitted from public sector debt statistics or disregarded in setting debt ceilings. Such debt, which stems from undertakings and letters of comfort signed by governments with regard to project financings, can represent very large contingent obligations.
As part of the new OECD procedures discussed below, there is now a model to be used in assigning buying countries to country gradings or categories. This will have an impact directly on the minimum premium rate to be charged for political risks, but individual export credit agencies and countries will still have to decide unilaterally both whether to be on cover or off, and if they remain on cover, up to what amount.
Premiums
In the past, most export credit agencies set their own premium rate schedules, which sought to reflect not only the contract value and the length of credit involved, but also the country category to which the buying or borrowing country was assigned. The number of categories varied from agency to agency: for example, for many years the export credit agencies of Germany and Austria had, in effect, a single category, while the United Kingdom moved ever closer to rating every case separately under its Portfolio Management System.
However, over the last 10 years, as other kinds of medium- and long-term export credit subsidies (e.g., on interest rates) were reduced or negotiated away within the OECD, it became ever more apparent that there were gross disparities in the premium rates charged by different export credit agencies. For example, one agency might have been charging a premium rate of 1.5 percent for a 10-year credit on India, while another might have been charging 15 percent.
Thus, within the OECD considerable work has been done toward developing minimum premium rates for medium- and long-term political risks. This work has now borne fruit, and minimum rates will almost certainly be phased in over the year or so beginning in April 1999. It will still be up to individual export credit agencies to charge premium rates above these floor rates if they wish, and to decide whether or not to make cover available either generally for a country or only for particular contracts.
Inevitably, these minimum premium rates will represent something of a compromise, and there is probably no very precise actuarial relationship between risks and rates. But no doubt this will change and develop over the next few years, as agencies and their customers build up experience with, and thus greater confidence in, the new system.
There is fairly common ground on the basic notion that premium rates should cover administrative costs and expected net losses (i.e., political risk claims paid less recoveries made), however difficult the second concept is to quantify with precision. But there is perhaps still some disagreement over the incremental premium that should be charged to make some contribution to reserves, against the possibility that future net losses have been underestimated. This is, of course, a very complicated and technical area, but the essential points are that henceforth there should both be little or no competition among export credit agencies on premium rates and that, over time, a more common and consistent approach to country risk analysis should emerge.