V Financial Sector Development: Public Debt Markets
Author: Mr. Hemant Shah
Language:
English
Keywords:
OP; country; debt; government; debt issuance; market participant; market-clearing interest rates; securities exchange; market access; Government debt management; Securities; Securities markets; Government securities; Central America
Download PDF (258.8 KB)

Abstract

The development of public debt markets in Central America is critically important to the overall development of capital markets. As with many emerging markets, the Central American capital markets are somewhat underdeveloped. Problems in public debt management and underdevelopment of capital markets are related. The underdevelopment of capital markets, particularly institutional investors, limits the amount and maturity of funding available to the government locally, and can substantially increase the rollover and currency risks in managing public debt. At the same time, weak debt management practices, which result in fragmented issuance and a lack of a liquid benchmark yield curve, make it difficult for all borrowers (including corporate and home loan borrowers) to obtain long-term funding, and for institutional investors to undertake appropriate risk management. The broader development of capital markets is a complex and multifaceted problem that also needs to be addressed, but given the dominance of public debt in capital markets in the Central American countries, improving the public debt market is their logical first priority.

The development of public debt markets in Central America is critically important to the overall development of capital markets. As with many emerging markets, the Central American capital markets are somewhat underdeveloped. Problems in public debt management and underdevelopment of capital markets are related. The underdevelopment of capital markets, particularly institutional investors, limits the amount and maturity of funding available to the government locally, and can substantially increase the rollover and currency risks in managing public debt. At the same time, weak debt management practices, which result in fragmented issuance and a lack of a liquid benchmark yield curve, make it difficult for all borrowers (including corporate and home loan borrowers) to obtain long-term funding, and for institutional investors to undertake appropriate risk management. The broader development of capital markets is a complex and multifaceted problem that also needs to be addressed, but given the dominance of public debt in capital markets in the Central American countries, improving the public debt market is their logical first priority.

Even for the relatively small Central American economies, developing local public debt markets is important. Sovereign issuers in smaller economies suffer from many problems in terms of a limited local investor base, higher transaction costs, limited trading, and economies of scale. El Salvador and Panama are dollarized and several other countries have a significant amount of de facto dollarization. A high degree of domestic dollarization may force issuers—including sovereign ones—to rely on foreign currency funding for a significant part of their financing requirement, and create some preference for doing so from the more developed external markets and investors. A country with current account deficits may also need to rely on external capital flows to close the balance of payments. However, external debt financing in foreign currency cannot be obtained without incurring some risks of currency mismatch and sudden capital outflows. Recognizing this, virtually all sovereigns in the region have already undertaken domestic issuance programs, and from a sovereign debt management perspective, it would be appropriate to strengthen rather than abandon them, even if external debt issuance has to continue.

Development of a more integrated regional capital market, including that in public debt, entails complex benefits and difficulties that require further study. Given the small size of the Central American economies, a more integrated financial market may seem to offer substantial benefits to investors and issuers alike. For investors, it might mean more efficient capital allocation and risk diversification. For issuers, it might imply greater and more liquid access to regional savings. However, despite the relative openness of regional capital accounts and the substantial presence of regional financial groups,2 sovereign debt markets (and, more generally, capital markets) remain divided in important respects by currency, creditworthiness, regulation, restrictions on domestic institutional investors, and the absence of a regional exchange. To advance toward a regional sovereign debt market requires significant convergence and cooperation in terms of economic policies and regulation of financial markets, and potentially the creation of regional exchange or at least regional trading platforms. A full examination of such issues is worthwhile, but beyond the scope of this study, which focuses on the more pressing task of improving individual sovereign debt markets. Of course, several aspects of development of national debt markets discussed here, and in particular standardization of issues, will facilitate a more active regional debt market.

While developing an integrated regional market remains a longer-term goal, substantial progress is being made under a project to harmonize public debt market standards. As part of a project of the Central American Monetary Council, supported by the Inter-American Bank, the authorities of the seven countries signed a resolution to advance adoption of common standards in public debt issuance across the region in November 2003.3 The effort involves promotion of common standards and good practices in such areas as organizing the primary market, public debt management, money markets and monetary policy operations, wholesale secondary market in public debt, payment and settlement systems, conventions in calculations relating to public debt, regulation of collective investments and secondary markets in public debt, and developing a regional wholesale/over-the-counter market in public debt. The program has led to important advancements in establishing common market conventions and calculation standards, adopting standardized securities in new issuance, building consensus toward analytical approaches and solutions to common problems, and offering training to debt managers. Further progress in implementing the 2003 resolution would depend on substantial additional technical and political support.

This chapter examines the strategic context of public debt management and the functioning of primary and secondary markets for domestic public debt. It proposes measures to improve the functioning of these markets and promote a more efficient regional market in public debt.4

Financial Sector in Central America

The financial sector in Central America is dominated by banks, which are the largest financial intermediaries throughout the region (Table 5.1). The Dominican Republic and Guatemala have somewhat low levels of financial penetration in terms of a bank-assets-to-GDP ratio of around 35 to 40 percent. The numbers for Costa Rica, El Salvador, Honduras, and Nicaragua range from 50 to 90 percent and compare well with other emerging or developing markets. Panama, with a dollarized and open economy, has very high levels of financial penetration (250 percent), reflecting its offshore activities. Deposit levels in the region are typically about 60 to 80 percent of total assets, except in Costa Rica, where they are only a third of total assets. Private credit generally accounts for about half of the assets.

Table 5.1.

Structure of Financial System

article image
Source: IMF estimates based on official sources.Note: Data are from December 2005, except for data with an asterisk, which indicates December 2004. AUM = assets under management.

By contrast, capital markets are underdeveloped throughout the region. Except for Panama, the number of equity and corporate bond listings are generally in the single digits, and market-capitalization-to-GDP ratios are quite low. While complete data are unavailable, institutional investors (pension funds, mutual funds, and insurance companies) also are underdeveloped. In some countries, the enabling laws and regulation for important sectors such as pension and mutual funds need to be significantly upgraded, or remain to be implemented altogether. In all cases, these institutions intermediate a relatively small share of national savings. Thus, the principal investors in government securities are commercial banks and public sector entities such as pension funds.

Issuance of corporate securities is stymied by a variety of factors. Leading Central American businesses are often family-owned, and their reluctance to share even minority control or to comply with extensive transparency and disclosure requirements seems to be the principal hurdle. Anecdotal evidence suggests that the large majority of local businesses needing financing of under $10 million find it cheaper to seek bank financing, while the larger companies seeking from $25 million to $50 million source it overseas. Thus, domestic bond markets are approached for a relatively narrow range of financing needs. Bank loans are preferred because they have fewer information disclosure requirements, and those requirements are private rather than public. At times, this situation has been compounded by cyclical excess liquidity in the banking system that lowers bank lending rates. In some cases, brokerage and exchange fees and procedural requirements of the exchange are also deterrents to corporate debt or equity issuance. Mortgage- or asset-backed securities have yet to emerge in these markets. While the private financial sector has issued some bonds, this activity is also limited in light of the relatively liquid condition of many regional banking systems.

Thus, public debt issues dominate Central American capital markets. The capital markets of Costa Rica, the Dominican Republic, El Salvador, Guatemala, Honduras, and Nicaragua consist almost entirely of public debt instruments, and public debt also dominates capital markets in Panama (see Chapter IV). Despite this, problems of fragmented domestic debt issuance and the large stock of nontradable public debt mean that there is insufficient trading and liquidity in domestic public debt markets.

Organization of Public Debt Management

None of the countries in the region has yet developed a clear debt management strategy that articulates the long-term role of domestic markets, domestic currency, and resident investors in funding public debt. Many countries depend heavily on external sources of funding or on foreign currency funding, and lack clarity regarding the strategic role of local currency and resident investors in funding public debt, goals regarding the amount of domestic currency issuance, and the preferred tenor up to which the sovereign desires building a yield curve. In the absence of a well-developed strategy, domestic debt issuance decisions are taken in a somewhat ad hoc fashion. There is little systematic dialog between issuers and investors concerning the medium-term goals of the former or the problems facing the latter.

The capacity of the Central American countries’ debt management units needs to be substantially strengthened. Most units have grown out of traditional back-office operations. They typically have a weak front office and lack a clearly identifiable middle office charged with risk management. They generally do not have a regular investor relations program that involves meetings with key investors and institutions to incorporate market views or resolve problems relating to public debt issuance. That said, debt management unit staff are aware of the need to standardize debt issuance, facilitate secondary market trading, and extend yield curve. They also meet regularly at the regional level to discuss these issues as part of the Central American Monetary Council’s harmonization program.

By and large, debt management does not adequately aim to manage the stock of existing debt. Except for Panama and, to some extent, Costa Rica, debt managers focus primarily on securing current funding needs, and secondarily on standardizing future issuance or extending the maturity of the yield curve. Most debt management units have limited capacity to undertake the more complex management of the overall stock of existing domestic debt. This is particularly important given the high debt-to-GDP ratio in many countries and the sizable nonstandard and nontradable domestic debt that requires restructuring.

Coordinating debt issuance of the sovereign and the central bank is an important challenge. In many ways, the debt management problems of the smaller Central American economies are more complex than those of the larger emerging markets. One problem is the sizable debt issuance by the central banks that requires considerable coordination with the ministry of finance. While some coordination is the norm, it is typically confined to operational issues such as the issuance calendar, standardization of instruments, and auction logistics.

In several countries there are some problems in the structure of authority and responsibility between the legislature and administration. First, debt managers sometimes have only limited clarity or authority to conduct debt management operations. In the Dominican Republic, Honduras, and, to a lesser extent, Nicaragua, issuance is authorized under somewhat restrictive laws that define the operational terms but can clash with current market conditions, leaving debt managers with limited discretion.5 Second, in most countries debt managers lack the authority to conduct overall liability management operations that go beyond the financing of current cash flow needs. Third, as much of the nonstandard debt has been issued under provisions of specific laws, restructuring of that debt is likely to require more than general authority to finance the government or conduct liability management operations. Finally, the transfer of central bank losses to the sovereign is not routine and involves significant intervention by the legislature.

These problems are reflected in the differences between domestic and external market access. Several countries have access to external bond markets, and thus to advice from international underwriters and advisors. Their successes in the external markets often stand in sharp contrast to their efforts in domestic markets. For example, Panama has possibly the best developed domestic financial sector in the region, with considerable success in issuing eurobonds globally and in arranging a $980 million global bond swap in January 2006, issuing a 30-year bond in the process. At the same time, Panama is actually curtailing its domestic medium-term issuance program. Similarly, the Dominican Republic successfully issued a 20-year $300 million eurobond in March 2006, attracting the lowest coupon in the country’s history and an oversubscription of 3.3 times. However, the Dominican Republic has not made a single “regular” debt issue in the domestic market.

Dual Sovereign Issuers: Government and the Central Bank

Public debt issuance is complicated by the simultaneous and often competing presence of the sovereign and the central bank as issuers of public debt. Ideally, public debt should be issued by the sovereign alone, with the central bank using a stock of sovereign public debt instruments to conduct open-market operations for monetary policy purposes. With the exception of Panama, the central banks in all five countries (Costa Rica, the Dominican Republic, El Salvador, Honduras, and Nicaragua) have sizable accumulated deficits arising from previous financial crises, financial system bailouts, and even just from normal central bank operations. This makes the central bank a frequent issuer of debt in its own name. For example, in the Dominican Republic, the stock of central bank domestic debt is about 5½ times as large as that of the government (Figure 5.1). Moreover, as the bulk of government debt is not issued in the form of securities, the outstanding stock of central bank securities is about 17 times as large as the stock of government securities. In both Honduras and Nicaragua, the central bank is a far larger issuer, while in El Salvador and Costa Rica, central bank issuance is between a quarter and half that of the sovereign. Such large issuances by central banks, coupled with the frequency of issuance and a concentration in short-term tenors by both issuers, creates a number of problems.

Figure 5.1.
Figure 5.1.

Domestic Central Bank Debt/Domestic Government Debt, 2005

(In percent)

First and foremost, debt issuance by a central bank to finance its own operations compromises monetary policy operations. The continuous need for the central bank to issue debt to finance its own operations or to roll over maturing debt may cause doubts in the markets about the bank’s ability to conduct monetary policy consistent with the objectives of price stability and the proper choice of instruments and interest rate paths.6 While this chapter does not focus on such monetary policy issues, this is perhaps the most important reason to recapitalize the central bank to a point that it can have a stock of suitable public debt instruments with which to conduct open market operations.

The substantial and simultaneous issuance by central banks also creates a significant burden in terms of coordinating the issuance of public debt. While the concerned authorities in the countries of the region seem to have the will to cooperate effectively, the need for extensive coordination on a calendar of issues, demarcation of maturities, issuance terms, eligible issuers, issuance/trading platforms, and the manner in which auctions are conducted adds many layers of complexity compared to a situation in which the sovereign is the sole issuer and the central bank at best a fiscal agent. In many countries, there are gaps in technical sophistication, familiarity with market conventions, and in some cases pay scales between central bank and finance ministry staff that deal with debt issuance, with the latter usually lagging behind. Often there are differences in the issuance practices of the two issuers, and the sensible need to harmonize them, even when addressed, nonetheless slows the adoption of desirable innovations. For example, in Guatemala, while the ministry of finance can issue standard instruments (carrying a common coupon), such issuance has been delayed until the somewhat greater legal difficulties in issuance of similar standard instruments by the central bank can be resolved.

The sizable issuance of central bank debt creates a potentially large rollover risk and a short-term orientation toward public debt in the market. For monetary operations, central banks have a preference for issuing short-term debt. This bias tends to be carried over in financing central banks’ quasi-fiscal deficits that are not quickly or fully compensated for by the sovereign (which is the root cause of central bank debt issuance in Central America). Such decapitalized central banks are generally resource-constrained and loath to issue longer-term debt that, all else being equal, entails higher interest rates and thus larger deficits. The sovereign authorities also are generally wary of issuance by the central banks at longer maturities, as they themselves intend to tap these maturities and prefer not to have competing paper from the central bank.

Resolution of a central bank’s accumulated losses is typically delayed and inefficient. In many countries, the authorities do recognize the need to recapitalize the central bank, and, in turn, transfer the central bank losses to the fiscal accounts. However, this is not an easy process given the politics involved, the level of public deficits, and legislative complexities. Thus, such transfers—if they occur at all—are delayed. In some cases, they are still ad hoc—in the Dominican Republic, for example, the first such issuance of the recapitalization bond was authorized for 2005.7 Even in countries where transfers are now more routine (e.g., Costa Rica, Guatemala, and Honduras) and losses are supposed to be transferred annually, the accounting and legislative process of determining the actual central bank losses and authorizing the issuance of government bonds nonetheless takes one to two years. As recapitalization entails a considerable legislative effort, a partial recapitalization that does not eliminate future operational losses allows a difficult problem to persist.

When it does occur, recapitalization of central banks is more often apparent than real. Typically, recapitalization bonds have unusually long tenors on the order of 25 to 100 years, with zero or below-market interest rates. In present value terms, this results in an effective recapitalization well short of the face value of the bond. For instance, in Guatemala, a recapitalization bond issued in 2005 to cover losses incurred in 2003 had a 30-year maturity and a 3.87 percent coupon rate, well below the 9.6 percent rate on the 10-year bonds issued during the same period. Another bond expected to cover central bank losses for 1982–2002 is expected to have even more off-market terms: a maturity of 100 years and a zero interest rate, resulting in a bond with a present value of approximately 1 percent of the face value. In Honduras, central bank losses in 2003 were covered with a 50-year bond with zero interest rate, while 2004 losses were covered with a 25-year bond, paying a market interest rate only on one-tenth of the bond issued.

Partial recapitalization with deep discount bonds may actually inflate the nominal level of public debt, while creating weak incentives to check future public deficits. Positive capitalization of a central bank is desirable to ensure full recognition of its losses by the sovereign and to preserve the central bank’s independence.8 The current practice of recapitalizing central banks with zero coupon or deep discount bonds not only fails to adequately reduce future central bank losses, but also overstates the widely followed nominal debt-to-GDP ratios than would have been the case with bonds paying current market interest rates.9 The resulting reduction in future interest bills also probably weakens fiscal discipline.

The issuance of deep discount debt to the central bank has created a sizable stock of nonstandard and nontradable debt that reduces the liquidity of public bonds. Such debt is rarely traded due to its unusually long maturity, absence of liquid yield benchmarks at such maturities, and, above all, the central bank’s incentives to not recognize the large capital losses if it were to sell such debt.

The division of sovereign debt issuance between the two issuers hinders development of larger benchmark issues and lowers liquidity. Central American sovereign debt markets are small, and the market for longer tenors has yet to develop adequately. Liquidity of most bonds is related to the size of individual tradable issues and aggregate tradable stock. The division of issuance between the two sovereigns and the nontradable nature of debt issued to the central banks prevents most issues from reaching volumes needed to become liquid benchmarks and reduces liquidity. These factors are also generally detrimental to development of a long-term yield curve and widening the investor base, particularly to foreign investors in local issues.

Primary Issuance of Public Debt

The primary debt issuance process throughout Central America suffers from several problems. Despite progress made under the harmonization project, many problems remain in the form of poor issuance mechanisms, issuance of nonstandard debt, fragmentation of issuance, and weak auction practices.

Issuance Mechanisms

Several countries use multiple issuance mechanisms, thus draining volume, liquidity, and competitiveness from auctions. In addition to regular auctions, several issuers (Costa Rica, the Dominican Republic, Guatemala) also rely on post-auction windows (Table 5.2) offered by the central bank or the ministry of finance.10 The presence of such windows, for extended periods of time, drains liquidity from auctions themselves, and complicates the bidding and acceptance strategies for investors and issuers, respectively. In the Dominican Republic, the window creates a further segmentation of the market, as the fixed interest rates offered in the window are only loosely guided by the preceding auction or current secondary prices, the window is open only to retail investors, and longer tenor maturities are offered through the window than are auctioned. Securities issued through these windows are typically nonstandard and nontradable. Interestingly, in Costa Rica, the finance ministry takes advantage of the statutory restrictions on nonfinancial public sector entities by issuing public debt outside the competitive auction process at “negotiated” below-market yields and tenors, creating an additional source of nonstandard debt.

Table 5.2.

Securities Issued by the Ministry of Finance and Central Bank

article image
Sources: Central Banks, Ministries of Finance.

There are no primary dealer systems in Central America, although issuers often rely on brokers and exchanges to conduct primary auctions. None of the countries employ a true primary dealer system with clear market-making obligations. Nor are brokers or dealers capable of many of the functions performed by the primary dealers in larger government securities markets. However, reflecting the traditional dominance of exchange brokers (the largest of whom are typically affiliated with major banks), many countries employ a somewhat unusual process for public debt issuance by relying on brokers and exchanges to conduct the auction. These arrangements vary widely. In Panama, all auctions must be conducted through exchange brokers. In Costa Rica, banks can participate directly in an auction, while other investors must submit their bids through brokers and pay related commissions. In El Salvador, ministry of finance auctions are open not only to brokers but to other approved investors, including domestic and foreign banks, whereas central bank auctions are only open to brokers. As a general rule, it is difficult to make the case for value added by brokers during the primary auction process, particularly for institutional investors. As the banks typically have in-house brokers, as a group they tend to be neutral about the exclusive rights of the brokers and the fees charged by them, with the pension funds objecting most to such a monopoly.

Issuance on Nonstandard Debt

The issuance of “nonstandard” debt without suitable characteristics that facilitate trading has been diminished, but not eliminated. In addition to issuance in irregular transactions (such as compensation for properties taken over by the state, arrears to suppliers, and issuance to central banks at off-market terms), several countries issue debt carrying amounts and interest rates specific to each holder (such as bank term deposits) rather than through standardized debt instruments, or they issue debt in physical (not dematerialized) form. The last two problems are directly addressed by the Central American Monetary Council’s harmonization project. While all countries have agreed to start issuing standardized securities, the Dominican Republic, Costa Rica, El Salvador, Guatemala, and Nicaragua continue as of this writing to regularly issue some nonstandard securities. In addition, all recapitalization bonds issued or planned for issuance by the sovereign to the central banks in the Dominican Republic, Costa Rica, Guatemala, Honduras, and Nicaragua remain nonstandard.

Fragmentation of Public Debt Issuance

Despite the relatively small volume of domestic issuance, public debt is issued in an extremely fragmented fashion across Central America for a variety of reasons (Table 5.3). First, as discussed earlier, debt issuance is undertaken by the sovereign and the central bank in five countries. Second, there is a high frequency of issuance—all countries schedule auctions on a weekly or biweekly basis, except for Panama, which conducts auctions monthly (Figures 5.2 and 5.3). While many issues are re-opened and consolidated, their eventual total size is unclear. Third, Central American public issuers attempt to issue a large number of maturities relative to their total domestic issuance. Fourth, some of the nondollarized issuers (e.g., Nicaragua) are unable to issue all of their financing requirements in local currency, and while attempting to develop a local currency market have to divide their local issuance in different currencies. Finally, issuance of nonstandard debt with different interest rates to winning bidders (rather than different prices) obviously creates extreme fragmentation.

Table 5.3.

Auction Mechanisms for Public Debt Securities

article image
Sources: Central Banks, Ministries of Finance.

The auction of BEMs in Costa Rica is competitive for zero coupon and fixed-rate BEMs denominated in local currency, and noncompetitive for variable rate BEMs denominated in U.S. dollars.

BPIs in Nicaragua are partially standardized.

Figure 5.2.
Figure 5.2.

Potential Debt Auctions Per Year

(Number)

Figure 5.3.
Figure 5.3.

Actual Debt Auctions Per Year

(Number)

Conduct of Auctions

Investors throughout the region complain of gaps in information regarding primary auctions. As part of the harmonization initiative, in all countries there is now a pre-announced calendar of issues from both issuers. However, the overall annual calendar announcement is rather generic, and the amount offered is typically announced only about a week before the actual auction. There are considerable uncertainties about the actual issuance volume, as in some countries (e.g., Costa Rica and Honduras) the government and central bank often issue the same paper jointly. Allocation rules are not always fully formalized, and at times are perceived as murky by both investors and issuers. Investors are not always assured which issuer’s securities will be awarded to them, which is a nontrivial problem in markets where the prices differ, or of the eventual size of a particular bond issue.

Fragmented and frequent issuance of a small quantity of debt appears to be partly driven by the desire of issuers to systematically capture the most aggressive bids or beat down the yields. Issuers routinely complain that the market is not sufficiently competitive, and that domestic investors, particularly those in the private sector, “overcharge” the government. There is thus a widespread reluctance on the part of the issuers to offer larger auction volumes, or commit to issue the announced volume when auctions are fully subscribed. Some of the most aggressive bids often come from banks that temporarily are very liquid, or from pension funds that, for regulatory reasons, must heavily invest in government paper. From time to time, this results in bids that other market participants regard as unrepresentative for the full spectrum of investors. Debt managers with weak authority and lacking a clear strategy to develop domestic markets and an investor base find it easier to defend small issuances or large rejections than a higher interest rate.

While there may be some validity to these concerns, they ignore the fact that investors will eventually factor in such defensive behavior of the issuers and adjust their bids accordingly. Smaller volumes can be placed at seemingly lower yields from time to time. But unless the overall volume of sovereign and domestic issuance is reduced, smaller issuance in individual auctions itself cannot reduce the interest rate in the long run. Moreover, as smaller issues are less liquid, the market is likely to charge a premium for such issues. As the domestic rates in both primary auctions and secondary markets are generally higher than the yields for comparable eurobonds, where they exist, small and fragmented local issues ultimately appear to carry a higher cost.

This widespread lack of agreement on “market clearing prices” between issuers and investors results in frequent auction failures or rejections. Auctions are often undersubscribed and at times are “deserted” (especially in Nicaragua)—that is, they receive no bids at all. Neither is it uncommon for the issuers to reject many bids in a fully or oversubscribed auction, while not filling the entire amount offered (e.g., in Panama). Apart from the uncertainties they imply in public financing, such auction failures or rejections signal market-wide dis-equilibria that should best be avoided.

There are also concerns about the role of dominant public sector investors. There is some consensus that public sector investors should be excluded from competitive bidding, as they could be influenced by the issuers and may result in an unrepresentative market-clearing interest rate or at least the perception thereof. However, the Central American experience indicates that there may not be a simple solution to this situation. In Panama, public sector bidders are allowed to participate only through noncompetitive bids, whereas elsewhere they are allowed to participate with competitive bids as well (e.g., in Costa Rica). Frequent and sizable participation of public sector investors through noncompetitive bids is likely to influence the issuer’s acceptance of competitive bids, and, in turn, the competitive bids by the private bidders. There thus may be a need to frankly discuss the role of the public sector vis-à-vis the private sector, provide public sector investors enough regulatory leeway and operational autonomy to bid as much on an arms-length basis as possible, and encourage the largest possible participation in competitive bids.

Secondary Markets

Public debt securities account for the lion’s share of secondary market trading throughout Central America. Given the relative underdevelopment of corporate securities markets, trading in public securities dominates overall trading in regional exchanges (Table 5.4).

Table 5.4.

Secondary Trading of Fixed-Income Domestic Securities

(In millions of U.S. dollars, 2005)

article image
Source: National stock exchanges.

However, the bulk of domestic issuance in Central America is in short-term paper, and most of the domestic trading is in repo transactions, resulting in a lack of liquid, long-term secondary market yield curve. Few of the issuers (barring Panama and Guatemala) have managed to issue domestic debt for longer than a few years. This failure contrasts with the issuance of up to 30 years in the eurobond markets by Costa Rica, the Dominican Republic, El Salvador, Guatemala, and Panama. Moreover, the bulk of the trading is in repo transactions of a few days. Thus, despite accounting for 90 percent or more of local exchange trading, government securities do not provide a meaningful benchmark yield curve, and even for the short durations involved, secondary prices and yields are not sufficiently current.

Transaction costs in the smaller local securities exchanges—including brokerage and exchange fees involved in trading public bonds—as a rule are extremely high. On an annualized basis, they amount to 1 to 3 percentage points in the Dominican Republic, 1 to 2.5 percent in Nicaragua for an outright transaction and 1.25 percent for a repo transaction, and 0.1 to 1 percent in Guatemala.11 Transaction costs are somewhat lower in Costa Rica, Guatemala, and Panama. High brokerage and bolsa fees may be justified by the low trading volumes and the small size of many issues, but they create a chicken-and-egg problem. More importantly, low liquidity eventually must raise the market-clearing interest rates demanded by investors.

Central American brokerages and exchanges face an uncertain future given their limited viability and difficulties in terms of consolidation. It should be kept in mind, of course, that brokers and exchanges may have played an important role historically in raising the transparency of government securities trading and, to a limited extent, popularizing investment in public securities by retail investors. In an environment of extremely limited private sector security issuance, exchanges may not have developed without some support from public issuers. Nonetheless, continued support of Central American exchanges and brokerages through higher charges and fees on public debt issuance and trading may have questionable value, especially considering the ownership of many brokerages by banks. The low trading income, the opposition of institutional investors to paying fees, and the growing reluctance of public issuers to pay issuance fees may make consolidation both necessary and desirable. There is a need to examine the consolidation of the brokerage industry within the region and encourage well-capitalized foreign brokerage houses, cross-listings across countries, and, eventually, consolidation of the trading platforms themselves. Consolidation of domestic exchanges is also worth considering in Guatemala, where there are two competing exchanges within a small market. However, cross-border consolidation entails many important challenges in countries without a common currency or regulatory bodies. While these issues need to be examined in detail, the global impetus toward consolidation of smaller bourses needs to be recognized and harmonization of listing and trading practices within the region needs to be speeded up.

Investor Universe and Regulation of Investments in Public Debt

A preliminary examination of the regulation of principal institutional investors in public debt markets finds three broad problems: inadequacies in the regulatory framework for the securities markets themselves, limitations on the institutional investors, and limitations on those investors’ investment portfolios.

The enabling framework for securities issuances, markets, and regulation of markets is either lacking or in need of upgrading in many countries. In Nicaragua, a draft securities market law pending congressional approval for some years would allow the dematerialization of securities and the issuance of asset-backed instruments, thereby widening the set of available securities and facilitating development of a liquid secondary market. In Guatemala, weak institutions (e.g., a securities registry) act as a temporary substitute for a proper securities regulator.12 The absence of a superintendency of securities has also seriously hindered the creation of a central depository of securities.

Throughout the region there are also gaps in the legal and regulatory framework for institutional investors. In turn, these gaps affect development of local government securities markets. In particular, the regulation of private pension funds and mutual funds is inadequately developed in most countries. In Nicaragua, design of a law to regulate the creation of private pension funds was abandoned in 2003 following a vigorous debate in congress. In the Dominican Republic, pension funds are not permitted to invest in government or central bank securities, mainly because of opposition from other investors who fear losing a lucrative investment opportunity. In the absence of well-developed mutual funds, brokers in many countries are offering less well-regulated variations such as participations in so-called “investment portfolios.” Investors are not always aware of the liquidity and pricing of such investments, and at times bank deposit clients are drawn into such investments (e.g., in Costa Rica) without fully realizing the possibility of capital losses or the illiquidity of such products. The introduction of well-developed mutual fund laws, together with sound governance, pricing, and mark-to-market practices, would help develop the investor base for government securities in the long run.

The Challenge of Creating a Regional Market

Despite trade and political links, Central America’s regional public debt markets and capital markets more generally are not well integrated. There are many benefits of greater regional integration of financial markets, but getting there involves addressing several important challenges and complex issues. The paragraphs that follow sketch some of the issues involved in the context of sovereign debt.

Smaller economies such as those in Central America appear to have much to gain from regional financial integration. Small financial markets suffer from high transaction costs, small volumes, a small investor base, and a small universe of investment products and instruments. For issuers of securities, including the sovereign, small domestic markets may limit the size and tenor of issues, competitive pricing, or the speed of execution. For investors as well as financial intermediaries, the limited investment universe may result in suboptimal diversification and risk concentration. Brokers and markets, particularly security exchanges, may find it difficult to survive or grow without a critical mass of trading. Smaller economies can partly offset these well-known problems by integrating within a common financial space. Within Central America, these benefits are relatively easy to obtain given the geographical proximity, shared language, common time zones, relatively free trade and capital movements, presence of several regional financial groups, and political will to promote regional ties.

But in the absence of further economic and regulatory convergence, significant problems hamper further integration of regional capital markets in general and the sovereign debt markets in particular. Central America remains divided by currencies. Macroeconomic policies, debt burdens, sovereign ratings, and debt sustainability are materially different across countries. Thus, sovereign debt issued by different countries remains as securities with very different risk characteristics. Eurobonds issued by the regional sovereigns can be easily invested in and traded by regional investors. But underwriting and over-the-counter trading of such issues will remain located largely in global financial centers, given the advantages of the large global investor universe, speed of execution, competitiveness, and low cost they offer. Thus, it is more meaningful to think of regional integration within sovereign debt markets in terms of a common market and investor base for sovereign debt issued in regional currencies. As the situation currently stands, further integration of the regional currency debt is likely to be hampered by two important factors: the separation of regional security exchanges, and the restrictions on institutional investors such as banks, insurers, and pension funds.

First, tradable domestic debt of each issuer is typically listed only in local bourses and often in local currencies. In the absence of any centralized links between regional exchanges themselves, and the absence of regional brokerage firms, nonresident transactions in such locally listed securities typically entail working through a local broker. Such transactions generally are less efficient than the purchase of eurobonds through a broker based in Miami or New York. The listing of government securities across regional exchanges is often cumbersome, as the exchanges do not have completely common listing standards. While some bilateral arrangements to overcome these problems have been undertaken,13 they fall short of a regional securities exchange, coupled with a regional custodian and clearing and settlement systems. The latter is a major initiative that would require substantial effort, study, and cooperation by the private sector participants.

Second, institutional investors face substantial restrictions on the purchase of regional foreign government securities that cannot be readily harmonized. Pension funds, and sometimes banks, are subject to prohibitions or tight limits on the purchase of foreign government securities. In all cases, banks face higher capital adequacy requirements on foreign than on domestic government securities, which carry zero risk weighting. In Nicaragua, the 1995 regulatory framework on the investment portfolio of commercial banks requires the prior authorization of the bank regulator to acquire sovereign securities (with the exception of Nicaraguan and U.S. instruments). Similarly, the 2005 bill regulating the investment portfolio of insurance and reinsurance companies requires the prior authorization of the regulator for the acquisition of all foreign sovereign securities. In a nutshell, prudential restrictions on foreign government securities are tight. Capital adequacy standards cannot be readily harmonized given the substantial differences in sovereign credit ratings, as well as implications for currency risk. As for pension fund investments, most countries would have to carefully balance the cost and benefits of foreign investments and consider the interests of the fund beneficiaries rather than sovereign debt issuers. While eurobond issues may be readily sold to a global investor base, local currency issues cannot be so seamlessly sold to the regional investor base.

Key Problems and Recommendations

Central American countries face many important challenges in developing local public debt markets. Establishing ready sovereign access to domestic funding and a liquid and efficient domestic secondary market are at best medium-term goals. This section highlights the key problems in public debt management and the steps needed to resolve them.14 While the problems are not unknown to regional officials, significant political, legal, and fiscal constraints have prevented more effective resolution to date. Thus, it is worth emphasizing at the outset that successful implementation of the recommended measures would require strong macroeconomic performance, political stability, a clear political consensus for the desired reforms between the legislative and executive branches, and substantial technical capacity or technical assistance.15 The measures recommended below are grouped thematically, and while certain key actions are singled out as particularly important, further detailed work must be done to identify more precise country-specific action plans and timing for implementation. While substantial progress in implementing these measures can be made in two to three years, it is important to establish a clear, ambitious, and unambiguous goal leading to restoration of ready sovereign access to domestic markets at a reasonable cost, and efficient functioning of the secondary markets.

The most critical and strategic steps for improving public debt management in virtually all the countries in the region involve (1) developing a medium-term debt management strategy and formally defining and building capacity for debt management units; (2) eliminating the central bank’s quasi-sovereign deficit; (3) converting a large stock of nontradable and nonstandard domestic debt through liability management operations; and (4) eliminating future issuance of nonstandard debt. These four steps need to be undertaken as an integral and well-coordinated package that can be labeled “strategic debt management.” A variety of other measures are also needed to simplify and improve primary market issuance, make price discovery more efficient, and improve secondary market trading. But without the four strategic debt management measures, additional measures would only make a marginal contribution to solving the larger problems at hand.

Strategic Debt Management

It is important for all countries to develop a formal, medium-term debt management strategy, as advocated by the IMF as part of a debt substainability framework. Several Central American countries have large debt problems relative to their economic aggregates. Moreover, their debt problems are exceedingly complex. The present approach of managing these problems through routine issuance decisions with inadequate debt management units does not fully recognize the gravity of the problem. The inability of virtually all countries to place in local markets, say, issues on the order of $100 million for five years with a reasonable assurance of full subscription, is a strategic limitation. It also stands in sharp contrast to improvements in domestic market access by some other highly-indebted countries (e.g., Jamaica and Uruguay) and to the considerable successes of Panama, Guatemala, El Salvador, and the Dominican Republic themselves in accessing eurobond markets. Again, while highly desirable and feasible, the task of developing an adequate medium-term strategy cannot be underestimated. It would need substantial country-specific technical assistance, with an important role again for the international financial institutions.

The overall capacity for debt management needs to be significantly strengthened in all Central American countries. While this chapter has not focused much on institutional aspects, it is generally clear that the regional debt management units within the ministries of finance have significant weaknesses in terms of mandate, skills, staff size, and resources. Strengthening these entities requires adequate formal recognition of them as debt management units, definition of their mandate and accountability requirements, and adequate staffing and budget support. Such units need not require large staff; most offices may not need more than five employees. But staffing with the right personnel is critical.

The five countries with significant accumulated central bank debt or continuing central bank deficits—Costa Rica, the Dominican Republic, El Salvador, Honduras, and Nicaragua—need to consider full recapitalization of central banks at market terms so as to eliminate or substantially reduce central bank debt issuance in a phased manner.16 While the sustainability of overall public finances is an important macroeconomic challenge beyond the scope of this chapter, recognition of central bank debt is more of a political, legal, and technical problem. While such problems ought not to be minimized, recapitalization of the central bank should be addressed as part of a comprehensive strategy to eliminate future central bank losses and integrate the fragmented sovereign debt market. Most countries have already taken important first steps in recognizing such losses and establishing precedents for recapitalization. It is now desirable to speed up this process, issuing marketable bonds in future recapitalizations, and replacing existing recapitalization bonds with marketable instruments. Both new issuance at market terms and conversion may need to be suitably phased in to accommodate certain fiscal and legal constraints. But the goal should be to clearly restore the central bank’s solvency and future profitability, and eliminate the perpetual problem of dual sovereign issuers in the medium term. Designing appropriate options, assessing the related evolution of central bank profit and capital and of sovereign debt, and reconciling the latter with the current program targets of international financial institutions may require substantial technical assistance. Institutions such as the IMF can play an important role in supporting this effort.

Where needed, central banks could be provided tradable government securities for open market operations. In some cases, recapitalization of the central bank (i.e., transfer of losses) may not leave it with a sufficient stock of public debt with which to carry out monetary policy operations effectively. In these circumstances, consideration could also be given to issuing a stock of “sequestered” government securities to the central bank for such purposes, in lieu of an additional capital injection or issuance of central bank securities. Important details—including the appropriate treatment for fiscal accounting, public debt statistics, and debt issuance authority—would need to be customized and fleshed out for individual country situations. But the broad outline of such an operation would entail issuance of standard government securities to the central bank, with the bank maintaining “contra” accounts in its books. The latter would reflect both the acquisition of government securities as assets as well as a corresponding liability to repay the government with identical interest and amortization terms. The transactions between the central bank and the government would thus not entail any net cash flow. The central bank could then use these special government securities, which would be identical to other public debt, for open market operations or repos.17

Conversion of existing nonstandard debt issued to the central bank and other creditors is another highly promising but technically challenging effort that warrants high priority. In addition to the nonstandard debt issued to the central bank, virtually all countries have a substantial stock of nonstandard and nontradable debt issued to private holders in lieu of arrears, confiscation of property, and other measures. In the rare instances that such debt trades, it carries huge discounts. Trading benefits primarily the few market intermediaries, typically at the expense of the often unsophisticated holders of such debt. While conversion of such debt into marketable debt is not easy, several recent examples of complex domestic debt transformations by Brazil, Colombia, and Mexico, and of external debt by the Dominican Republic and Panama, suggest the fundamental feasibility of such transactions. Conversion of current nonmarketable debt also holds major benefits for the sovereign. First, most conversion has the potential for significant reduction in the nominal value of public debt, thus reducing the well-published level of debt. Second, conversions could be particularly useful to introduce new longer-term maturities in the domestic market and extend the current yield curve gradually.18 Fleshing out such internal debt restructuring may also require substantial country-specific technical assistance for designing appropriate options and implementing the same.

Issuance Process and Auction Mechanisms

In countries that continue to issue securities under a discriminatory price auction, the practice of issuing nonstandard securities with different rates of interest (rather than at different prices) should be stopped forthwith. Technically, this change is easily accommodated in the existing auction arrangements, and market participants would be able to adapt to it easily. Internal procedural rules or laws could be modified—with interim decrees, where needed—to expedite the transition.

Concurrent issuance of government and central bank paper in a single series and auction should be phased out. While central bank issuance in its own name should be eliminated altogether, if it does have to be continued, the two issuers should not seek to force a single auction upon investors. Such commingled auctions, such as those in Costa Rica, are more likely to raise the cost of funding for both issuers and cannot prevent secondary market prices from diverging. That said, in several countries, at present, the sovereign faces considerably more difficult access to the market than the central bank, and this transition would take time (perhaps a few years) to improve market access. However, there needs to be a recognition that improved sovereign access to the market must come about from fundamental improvements in debt sustainability, developing a strict culture of honoring debt-service payments, and more generally all sovereign payment obligations, rather than “association” with the central bank in a joint auction.

Throughout the region, the issuers should slowly develop a commitment to accepting market discipline. Regional sovereign issuers should naturally retain the legal right to reject bids they consider inconsistent with the market rate. There are also concerns in many markets that the bidders, both from public and private sectors, could exercise oligopolistic power and raise interest rates payable by the sovereign. However, the issuers’ behavior at auctions must also be guided by a long-term commitment to open and transparent auctions as the principal mechanism to raise funding domestically. Most sovereigns also have a compelling financing need that requires them to re-enter the auction market in a matter of days after undersubscribed or even deserted auctions. Thus, all things considered, there seems to be a stronger case for issuers to phase in greater acceptance of market bids, developing a long-term and steady investor base, and developing a convergence about the appropriate level of interest rates. In particular, once an auction amount is announced and is covered (i.e., bids cover the amount offered), rejection of competitive bids short of the offered volume (minus any noncompetitive bids) should be an exception rather than the rule. To avoid excessive uncertainty and costs for competitive market participants19 and the resulting long-term loss of investor interest, convergence of interest rate expectations of issuers and investors should be encouraged through several measures. These include developing a more liquid secondary market that would provide continuous quotations on current interest rates, holding a regular dialog with market participants, especially concerning new maturities, widening the investor pool by developing domestic institutional investors, and attracting regional and foreign investors. Where rejections are considered necessary, an effort should be made to explain the issuers’ stance and provide guidance to the market for future auctions. Again, while such a commitment to market discipline would be an appropriate medium-term goal for most countries, the pace of transition would need to be carefully analyzed and implemented in a country-specific manner.

Multiple avenues of noncompetitive bids should be closed, while all noncompetitive bidders should be encouraged to participate in the announced auctions. As a rule, the reliance on separate post-auction windows should be stopped. At times, issuers justify such windows on the grounds that auctions are insufficiently competitive, and such specialized windows (say, to retail investors or public sector investors) enable issuers to fulfill financing needs at more reasonable interest rates.20 However, when it is a routine practice, the use of such windows eventually drains interest and competitiveness from the “mainstream” auctions, reduces the volume of securities issued in large, competitive, and publicly announced auctions, and signals the unwillingness of issuers to accept market rates in competitive auctions. While noncompetitive bids per se can be allowed, noncompetitive bidders should be encouraged to participate in thesame auction. This should also apply to the forced placement of government paper with public sector enterprises, as in Costa Rica.

Consideration should also be given to allowing competitive bids by public sector borrowers. The standard advice to keep public sector investors out of competitive bids, and thus prevent them from influencing market clearing prices, has validity in certain situations. However, in the specific context of the region, and Panama in particular, it appears not to serve a useful purpose, as the participation of such bidders is large and indirectly affects the cutoffs set at competitive auction as well. Accordingly, it appears preferable to allow all public sector investors—including banks and pension funds—to participate in competitive bids, but with enough operational autonomy for them to set their own bids without “guidance” from the government.

Countries that use discriminatory price auctions—Costa Rica, El Salvador, Honduras, and Panama—could consider adopting uniform price auctions. While there is enough precedent for both auctions globally, there is a stronger theoretical justification on balance to adopt uniform price auctions. While discriminatory price auctions need not involve nonstandard issuance, adoption of uniform price auctions would allow a faster transition away from nonstandard issuance. Again, it is important to recognize the variation in convergence on market rates across the region, take measures to educate and consult with market participants, and phase in the adoption of such changes at a pace appropriate to specific country circumstances.

Several steps are needed to eliminate fragmentation of current securities issuance and extend yield curves. As discussed, an important step would be the recapitalization of central banks with the issuance of longer-term, standardized, and tradable bonds, and the elimination of short-term central bank issuance for its own financing purposes. Second, conversion of the large stock of existing nontradable securities would enable most governments to jumpstart the creation of medium- to long-term securities of up to 10 years, which would start generating secondary market quotes and a benchmark yield curve. Subsequent medium-to-long-term placements could be considered through the re-opening of these issues. Issuance calendars could be consolidated, particularly for longer-term securities. As sovereigns become the principal issuers of long-term debt, they need to commit to issuance in minimum volumes, which is one of the key goals of the Central American Monetary Council’s harmonization project. For the region, a minimum domestic issuance size of $100 million equivalent would seem an appropriate target.

Secondary Market Trading

Brokerage fees need to be reduced in most countries. The simultaneous solution to the chicken-and-egg problem of low trading volume and high transaction costs requires a multipronged approach that includes (1) reduction of exchange and brokerage fees; (2) equal access to all institutional investors in primary auctions with uniform participation rules for government and central bank auctions; and (3) other measures outlined above for extension of the yield curve and conversion of nonstandard government debt into more standardized and tradable instruments. Using stock exchanges and brokers for primary auctions of sovereign or central bank debt is of somewhat dubious value. While the still-limited stage of development of local securities exchanges must be kept in mind, it would be preferable to shift attention to promoting secondary market trading through the measures discussed above.

Certain institutional problems in specific markets need to be resolved expediently, perhaps with some gentle public encouragement. There may be merit in encouraging the various parties involved in Guatemala to consider uniting the two exchanges. Authorities in the Dominican Republic need to facilitate a solution to the dispute between the two competing custodial agencies. Issuance of dematerialized securities needs to become the norm in countries that do not yet issue them in this manner.

Other Measures

There is a need to expedite the passage of broader securities market legislation in several countries. Across the region, there are bills to improve the regulatory framework for the securities market, replace ad hoc presidential decrees, create interbank trading platforms, dematerialize securities, introduce asset-backed securities, and facilitate the formation and regulation of institutional investors. These measures would have profound beneficial effects on the depth and efficiency of public debt markets.

Conclusions

The countries of Central America face substantial challenges in managing their public debt burden and ensuring debt sustainability. Development of domestic public debt markets is of crucial importance for developing capital markets. While macroeconomic and fiscal discipline are of fundamental importance, regional public debt managers also address several important technical problems to achieve reliable access to domestic financing and create wider and more liquid public debt markets. The most important responses include developing and implementing a medium-term debt management strategy, building the technical capacity of debt management units, transferring the quasi-fiscal debt of central banks to the government, restructuring nonstandard and nontradable debt, and taking a series of steps to promote capital markets generally and pension and mutual funds in particular. While such a comprehensive reform must be implemented carefully, over the medium term, and at a pace consistent with the particularities of individual markets, a clear vision and political commitment at the highest levels are needed in most countries. Taking action on the measures that have been proposed may require substantial technical assistance to individual countries and sustained political will. Developing an efficient regional public debt market that overcomes the limitations of size and scale is an important long-term goal. Development of a more solid domestic debt market would provide the foundation on which such an integrated regional market could be built in the future.

1

The underlying country studies were prepared by Andy Jobst, Laura Valderrama, and Ivan Guerra. The drafts of this study were presented to the technical staff of the central banks and ministries of finance at a conference in Guatemala in June 2006, and then to senior policymakers at another conference in the Dominican Republic in June 2006. Their comments have been incorporated. Generally, the chapter reflects data and public debt market conditions in Central America as of end-2005. In addition to the large number of public and private sector officials in the seven countries studied who generously provided information to the team, we would like to thank Alfredo Blanco, Jorge Barboza, Catiana García Kilroy (Central American Monetary Council), Elizabeth Curry (World Bank), Eliot Kalter, Ceyla Pazarbasioglu, Guy Meredith, Dominique Desruelle, Alfred Schipke, Humberto Arbulu-Neira, Ousmene Mandeng, Hunter Monroe, Erik Offerdal, Anthony Pellechio, and Parmeshwar Ramlogan for extensive assistance. For their comments on the chapter, we would like to thank Alfred Schipke, Humberto ArbuluNeira, Francesc Balcells, Alona Cebotari, Ana Lucia Coronel, Luis Cubeddu, Rodrigo Cubero, Vikram Haksar, Geoffrey Heenan, Luis Jácome, Gabriel Lopetegui, Carla Macio, Rodolfo Maino, Rogelio Morales, Marina Moretti, Christian Mulder, Marco Rodriguez, Neil Saker, and Wendell Samuel.

2

A 2006 IMF study, Central America: Structural Foundations of Regional Financial Integration, prepared by a staff team led by Patricia Brenner, discusses regional integration issues relating to consolidated supervision, the insurance sector, and payment and securities settlement systems.

3

See Consejo Monetario Centroamericano, Secretaría Ejecutiva, Resolución CMCA/CMH-RE-02/11/03, “Aprobación de los Estándares para la Armonización de los Mercados de Deuda Pública,” November 2003, and progress reports. Available via the Internet: http://www.secmca.org/Proyectos_DeudaPublica.htm.

4

A more in-depth examination of both institutional and market-related issues concerning debt management can be found in World Bank reports on Guatemala, Costa Rica, and Nicaragua. Related work by the Central American Monetary Council and the Inter-American Development Bank promotes harmonization of technical standards of public debt issuance, and provides diagnostic analyses and technical assistance, much of which has benefited this chapter.

5

In Honduras, the parliament required issuance of bonds by a public electricity company specifying low (below-market) interest rates, resulting in failure of the issue. In the Dominican Republic, individual debt contracts need to be authorized by the Congress.

6

Note that the rollover of debt to finance accumulated losses is more problematic than, say, debt issuance to mop up excess liquidity and build up reserves, in which case the rollover risk is mitigated by the presence of reserves.

7

In 2006, the authorities in the Dominican Republic prepared a comprehensive plan to recapitalize the central bank, which is scheduled to be submitted to Congress in 2007.

8

However, this may often be a higher standard of capitalization than needed solely to eliminate future central bank losses. As central banks do not always pay market interest rates on all their liabilities (due to reserve requirements), the level of recapitalization needed to eliminate future deficits may be lower than that implied by full transfer of their accumulated losses.

9

While the true value of the net public debt of the central bank and the government remains unchanged, such an increase in the nominal debt-to-GDP ratio may add to a lack of transparency about public debt, and possibly to the cost of public debt and problems of access.

10

Often such windows are used for noncompetitive bids, although the point here pertains to their separation from the regular auction rather than the noncompetitiveness of the bids.

11

By comparison, the bid-ask spreads for government securities in the larger, well-developed emerging markets are on the order of 5 to 10 basis points.

12

A securities law was passed in 2006.

13

Various bilateral agreements have been undertaken to facilitate the cross-listing of securities, as in the case of El Salvador, Costa Rica, and Panama.

14

While this regional study naturally focuses on the more common problems of the region, and the authorities have a strong interest in regional solutions, a full program of local sovereign debt market development would address individual country situations.

15

The authorities have accepted most of the recommendations of this study and formally requested further technical assistance to improve the region’s debt management capacity and debt structure. In early 2007, there was an agreement in principle between the IMF and the regional authorities on a multi-year technical assistance program with potential participation of the World Bank as well.

16

While this chapter does not deal with central bank autonomy issues, such recapitalization should be part of an overall plan to ensure adequate independence of the central banks.

17

In fact, such securities of an appropriate tenor may be issued to the central bank at market-clearing prices determined in the relevant standard competitive auctions.

18

For instance, suppose it is difficult to introduce a 10-year maturity local bond because the domestic market is illiquid and the sovereign, to date, has issued bonds only up to 5-year maturity to investors. Conversion of, say, a 30-year bond issued to the central bank into 7-year and 10-year bonds with reasonable extrapolation of current yield curve could start some trading up to 10-year maturity, allowing the “market rates” for 7-year and 10-year maturities to be gradually discovered.

19

Market participants complain of the need to maintain very short-term liquidity ahead of auctions and the opportunity cost thereof in case of rejection of their bids.

20

Some issuers justify the use of windows on the grounds that they use the immediately preceding cut-off or average interest rates in the preceding auctions as reference rates. In active trading environments, the rate determined in an auction a week or two earlier would not necessarily constitute the current “market” rate. More importantly, the existence of post-auction windows either provides incentives to eligible private participants in the window to drop out of auctions, thus compounding the loss of competitiveness in auctions, or creates mechanisms to give public sector investors access to below-market interest rates. In some cases, windows are used to offer securities not generally available in regular auctions, as is the case of the retail window for longer-term central bank paper offered by the Dominican Republic.

Cited By

Within Same Series

Other IMF Content

Other Publishers Content

Cover Economic Growth and Integration in Central America
Economic Growth and Integration in Central America
Author: Mr. Alfred Schipke and Mr. Dominique Desruelle
ISBN:
9781589066168
ISSN:
0251-6365
Publication Date:
13 Jun 2007
Publisher:
International Monetary Fund
DOI:
https://doi.org/10.5089/9781589066168.084
Pages:
110

Domestic Central Bank Debt/Domestic Government Debt, 2005

(In percent)

Potential Debt Auctions Per Year

(Number)

Actual Debt Auctions Per Year

(Number)