1999 Fiscal Conference Panel - Sustainable Public Sector Finance in Latin America: Stock - International Lending and Capital Flows
International Lending and Capital Flows
n this paper I will not attempt to provide a Wall Street version of the issues that we have been discussing at the seminar on Sustainable Public Sector Finance in Latin America. I am glad that this was not the topic entrusted to me either, since I see very little room for disagreement. I do not mean to trivialize the subject at all by saying merely that budgetary reform is still extremely necessary in most of Latin America, and extremely difficult to achieve. Instead I intend to provide a broad overview of international lending and capital flows, with particular reference to the financing of public sector deficits. I will take advantage of my position towards the end of the schedule, however, to make reference whenever possible to the capital markets issues that arose over the first day and a half of the meeting.
Recent Trends in Capital Flows
There have been important changes in the uses and sources of capital flows around the world over the past couple of years. The Asian crisis has prompted a sharp adjustment in external balances in that region, almost eliminating the current account deficit in emerging markets altogether in 1998 (see Chart 1). Resident net lending and unaccounted outflows remain a significant drain on resources for emerging markets as a whole, despite a drop from US$137 billion in 1998 to an estimated US$90 billion in 1999. In this regard, however, it is important to bear in mind that problems of poor disclosure and data quality, which bedevil accounting practices and fiscal accounts in emerging markets, are particularly acute when it comes to the balance of payments. “Resident net lending and unaccounted outflows,” as the Institute for International Finance (IIF) terms it—or capital flight, as is often the case—totaled US$140 billion in Russia in the 1990s, for example, and is still estimated to be running at around US$20 billion a year.
In Latin America, on the other hand, the adjustment has been less dramatic: the region has run down reserves and has continued to post substantial current account deficits (see Chart 2). Capital flight from Latin America appears to be subdued, reflecting well on structural changes during the 1990s and increased domestic confidence in economic prospects.
Capital flows to emerging market countries have also undergone a dramatic transformation in the 1990s and periodically during crises (see Chart 3). Official flows have taken up some of the slack when private sector debt flows have declined. This was particularly the case in Asia in 1997–98 and in Latin America in 1998–99 (see Chart 4). After the liquidity boom of 1996–97, foreign direct investment has resumed the upper hand over portfolio investment as a source of capital for emerging market countries.
Commercial bank credit has contracted for the past two years overall and fell sharply in Latin America in 1999, as we shall see in a moment. Latin America remains the main market for bond issuance as sovereign borrowing needs endure, but corporate issuance has slowed dramatically from the precrisis period in the first half of 1997. This is because companies are shut out by investors who are nervous about the true credit quality of enterprises operating in a deteriorating business environment or because they balk at paying the high financing costs demanded.
A closer look at capital flows to Latin America shows that the region’s current account deficit adjusted sharply in 1999 and is expected to stabilize in 2000, after having deteriorated steadily through the mid-1990s as a result of strong domestic demand (especially in post-real Brazil) and trade liberalization (see Chart 5). Despite shocks to confidence in the region, net private flows are still expected to provide more than enough finance for the current account deficit through the present postcrisis malaise (see Chart 6).
It is worth noting, however, that in the boom years of 1996–97 net private inflows were equivalent to more than double the sum of the region’s current account deficits. The persistence of strong flows of private capital to the region does suggest that the economic policies pursued both in the run-up to and during the 1998–99 crisis, buttressed by solid multilateral support for such orthodox responses, have helped to maintain the confidence of foreign investors in the future of Latin America. Within the private flows category, foreign direct investment has proved particularly stable (see Chart 7), although a modest decline is expected for 1999–2000 as the privatization schedule tapers off, as Fábio de Oliveira Barbosa mentioned earlier.
The other component of the region’s financing needs, namely, external debt amortizations, is playing against Latin America this year, however (see Chart 8). Total amortizations have been rising steadily and are set to peak in 1999 for two main reasons. On the one hand, the grace periods for Brady debt renegotiated in the late 1980s and early 1990s are gradually expiring; on the other hand the two- to three-year corporate loans and bonds extended and issued in 1996–97, when global liquidity was at a peak, are now falling due.
We have already seen how foreign direct investment, which is closely linked to the current account deficit and ideally provides much of its financing, is holding up well in Latin America. The same is not true of private debt-creating flows such as commercial bank lending and bond issuance. Starting in the first quarter of 1998, when it became clear that the region was not going to escape contagion from the devaluations and financial turmoil in Southeast Asia, commercial bank lending fell sharply (see Charts 9 and 10).
This phenomenon is particularly interesting in the case of Brazil, as seen in the previous two charts. You will remember that senior international bankers, occasionally under the auspices of the IIF, were prevailed upon to publicly support the multilateral assistance package for Brazil in the last quarter of 1998 and then the Brazilian government’s attempts to stabilize the economy in the wake of the devaluation of the real in January 1999. The banks pledged to roll over their lending to Brazil, particularly in the shape of trade lines, and it is possible that this contribution helped to keep the hot topic of involuntary private sector bail-ins off the table at the time.2 Certainly G7 and multilateral representatives have been heard since to laud the initiative as a worthy example of a private sector contribution to crisis resolution.
The data suggest, however, that bank exposure was not maintained at historic levels. In fact, it appears that the offer or pledge was made once the international banks had already scaled back their lending to Brazil and Argentina, in particular, through net outflows in the third and fourth quarters of 1998. The commitment to maintain credit lines going forward was therefore made from a low base in late 1998. It further appears that those credit lines that were maintained were available largely to big exporters and certainly not to the riskier small and medium-sized enterprises.
This feature of bank lending in Brazil in part explains the absence of a strong V-shaped trade recovery in the wake of the devaluation, since export credit was not available for those smaller companies that might easily have reoriented their production to external markets. Other factors, such as poor export infrastructure and the weakness of demand in neighboring countries, also clearly played a role, of course.
Bond issuance likewise fell sharply with the onset of the Asian crisis and its associated contagion effects (see Chart 11). Sovereign issuance only declined by around US$5 to US$6 billion from 1997 to 1998–99, but corporate issuance was hit particularly hard. This was in part because sovereign needs are more rigid, so governments are forced to accept higher borrowing costs to finance (higher cyclical) deficits; companies, on the other hand, can delay expansion plans or seek equity injections instead. Governments also have more opportunities to diversify their sources of bond financing, by tapping the retail investor base in Europe, for example, which, though it might not be familiar with Latin corporations, would have heard of Argentina. Governments can also diversify by securing multilateral guarantees for a bond issue, as Argentina did in October 1999.
Having examined the major trends in capital flows, the paper will now review some of the developments influencing the borrowing and lending communities in the Latin American debt market.
Evolution of the Borrowers
Changes in the regional and global environment have strongly affected the prospects for international lending and capital flows. Flows to emerging markets fluctuate with global market conditions, as we have just seen, but are also of course vulnerable to specific events in emerging markets. The improvement in sovereign credits through the 1990s has played a crucial role in attracting flows back to Latin America in particular, after lenders were left with severely burnt fingers in the regional debt crisis of 1982.
Reforms in three key areas have contributed to the improvement in sovereign credits:
The first area—public sector reforms—has of course been the focus of our meetings here in Atlanta. Every speaker has touched on some aspect of the reforms listed above. A theme running through that area and the other two, however, is simplification, and this has a specific relevance to the field of capital markets. Carlos Boloña mentioned cuts in the number of taxes imposed in Peru, and there are clearly administrative gains available from simplifying taxation structures. In pursuing this path, however, a government is also making it easier for the investment board or portfolio manager of a pension fund, for example, to follow the fiscal performance of the country. An investor who understands the credit risk in a given country is more likely to follow through with an investment, either by purchasing a new issue or by bidding up the price of a bond in the secondary market.
This process of reform and structural adjustment has been motivated primarily by the lessons of the 1980s. In Latin America’s case, the policy agenda of the 1990s conformed fairly closely to the so-called Washington Consensus, which emerged as multilateral lenders in particular drew a reduced set of conclusions from those broad lessons. The resulting policy framework, as adopted in Latin America, had three key elements that feed back into the question of sustainable public sector finance: (1) Countries’ inability to pay primarily reflects fiscal problems, not balance-of-payments problems. (2) Capital flight was a major factor in the failure of governments to stabilize their economies in the 1980s, and it usually reflected a lack of confidence in the government’s ability to meet its fiscal obligations, leading to fears of monetization and asset seizure (if not the actual event). (3) Finally, resolution of a fundamental fiscal imbalance is a domestic challenge and cannot be secured through increased external “support.”
This last point was illustrated by the failure of successive attempts to attract new lending from the private sector during the 1980s, as a result of the inadequacy of fiscal adjustment efforts in Latin America. Multilateral and bilateral lenders were instead forced to provide financing (often through the buildup of arrears) while the private sector waited for indications that the underlying credit quality was improving. Mexico’s successful Brady Plan in 1989, for example, which did finally bring in new money, was preceded by several years of impressive fiscal performance.
For many investors, the same fundamental problems and fears lay behind the most recent emerging market scares in Russia and Brazil, which can be classified as classic debt traps even though prompt action by the authorities saved Brazil from the worst-case scenarios painted by many in January 1999 after the devaluation of the real. In both cases, hikes in domestic interest rates (to defend currencies) compounded the government’s fiscal problem by producing an explosion in the domestic debt, which undermined the sovereign’s ability to pay. At this point in the story, the Russian government defaulted on its obligations and Brazil suffered downgrades from ratings agencies. The next stage in the classic debt trap, which materialized with a vengeance in Russia but proved subdued in Brazil, is that high interest costs lead to recession, which in turn impedes the rest of the economy’s ability to pay, thus depressing fiscal revenues and prompting pressure for bailouts.
As we saw earlier, these developments provoked severe capital flight in Russia. Although the support package sponsored by the International Monetary Fund (IMF) has reduced capital flight in Brazil, questions remain concerning the country’s ability to address its fiscal problems over the longer term and its capacity to reduce interest rates to sustainable levels.3 Capital flight remains a major threat for the Brazilian authorities since the much-vaunted ability of the local market to finance most of the government’s borrowing needs depends on continued confidence in the domestic financial system. Meanwhile, neighboring Argentina remains the most highly indebted large country in Latin America but is only gradually able to resort to the fledgling domestic market to satisfy financing needs. At times of external volatility, doubt over the government’s ability to refinance obligations will rear its head again.
While the governments of Latin America have adopted a new policy framework in response to the lessons of the 1980s, the external environment has not stood still. In particular, the international financial world of the 1990s is fundamentally different from that of the previous decade, demanding even higher standards of prudential control from both the public and private sectors. In a world of securitized finance, money moves quickly and often in huge amounts; this puts a premium on building “cushions” to absorb such swings and on keeping economic performance in line. Likewise, domestic banks are exposed to new risks, including volatility in asset prices and hence in their mark-to-market valuation. An institution’s ability to access capital markets may also fluctuate, often constrained by the investor’s perception of the sovereign government of the bank’s resident country. Against this backdrop of volatility and recurrent emerging market crises, investors’ perceptions of the current system’s stability have been tested by the evident inability of the IMF and major G-7 powers to dampen risk in the global economy.
Through all the changes, however, there are important elements of the puzzle that endure and are sometimes overlooked both in the frenzy of financial market contagion and in the search for scapegoats after a crisis. One of the most important is the fact that the traditional framework for sovereign credit analysis still holds, and it remains the “bread and butter” of the work of portfolio managers, research analysts, and credit strategists—on Wall Street and around the world. The two pillars for this analysis are a government’s willingness and ability to pay (see Box 1). Kurt Weyland yesterday expressed concern over the so-called efficiency of a capital market response at times of contagion, when it appears that investors are unable to differentiate between good and bad credits. Unfortunately it seems to me that this contagion arises precisely because market participants are engaging in traditional credit analysis and responding to fears that tight liquidity will jeopardize the ability to pay of a wide range of creditors in emerging markets. The effect of this group conclusion is intensified by the safety factor built into those credit decisions to allow for the margin of error in second-guessing the conclusions drawn by other players.
Evolution of the Lenders
Finally, I would like to provide a brief, analytical chronology of the development of the Latin American debt market and to highlight the evolution of perceptions of Latin American debt by U.S. investors. Essentially, changes in the U.S. investor base for emerging market debt can be divided into six stages.
Stage 1: The Pre-Brady Era (before 1993). The marketplace for emerging market debt evolved very slowly in the wake of the 1982 debt crisis. Over its first decade it was characterized mainly by interbank loan swapping and some limited secondary market trading of loans. The market was dominated by the banks themselves, which had large inventories of nonperforming or low-yielding debt as a result of the 1982 defaults and restructurings as well as by dealers, hedge funds, and a few specialist mutual funds. A thumbnail sketch of the market would depict shrewd investors looking to buy undervalued securities from skittish original lenders searching for a way out of the emerging market debt business. As the U.S. Treasury in particular took an interest in organizing a more sustainable solution for the troubled borrowers and a more permanent market for the debt instruments, the Brady plan was born.
Stage 2: The Boom Period (January 1993–February 1994). As more Brady deals were concluded, more players were attracted to the secondary market and liquidity improved. At the same time, improving fundamentals spurred renewed interest in emerging markets as the policy lessons of the 1980s were applied. In terms of the technical characteristics driving the market, the rally in developed country bond markets compressed yields and encouraged fixed-income investors to search further afield for juicy returns (effectively a “flight from quality”). The steep U.S. yield curve, with short rates at around just 3 percent, added extra fuel to this heady mix. It was temporarily possible for investors to profit from a carry trade, borrowing in the short term to make leveraged investments in long bonds that included the newly created Brady instruments.
Stage 3: The Impact of Five Fed Hikes (February 1994–June 1995). It quickly seemed that the boom had been premature. The U.S. Federal Reserve started to raise rates in February 1994, prompting drops in all bond markets. Speculative markets such as the infant Brady bond market fell further. Tighter global liquidity exposed the weaknesses of Mexico’s financing strategy, which relied on dollar-linked, short-term domestic debt, forcing first the devaluation of the peso in December 1994 and then domestic liquidity crises in both Mexico and Argentina. The latter was seen as particularly vulnerable due to its currency board-style arrangement and weak financial sector. To varying degrees, international assistance and disciplined domestic policy choices were required to prevent more extensive damage to emerging markets in general and the financial systems of those countries in particular.
Stage 4: A Diversified Investor Base Emerges (June 1995–June 1997). Credit fundamentals stabilized slowly after the Federal Reserve’s round of tightening and the steadying of the Mexican economy. Crucially, new investors started to enter the emerging debt market, including the long-awaited crossover buyers who augured greater integration with other credit spread markets. This development was interpreted as reducing sensitivity to U.S. interest rates, while helping to ensure that small sell-offs would be contained rather than mushrooming into precipitous declines, as buy-and-hold investors contributed to a sense that the debt was somehow in “firm hands.” Meanwhile emerging market credit fundamentals were back on an improving path: Mexico and Argentina recovered quickly, Brazil maintained its impressive trajectory in the wake of the 1994 stabilization plan, and Southeast Asia continued to boom. Against this backdrop, spreads tightened and yields converged around the world. A host of countries in Latin America were able to issue thirty-year bonds, including Argentina, Mexico, Panama, and Venezuela. In June 1997, for example, Brazil was able to issue thirty-year debt at just 395 basis points over U.S. Treasuries. Likewise, a Chilean electricity utility issued 100-year paper. With hindsight such issues look like clear evidence of euphoria, but at the time they seemed rational bets on the future prospects of developing countries in a liberalizing global economy.
Stage 5: All Bets are Off (July 1997–March 1999). Just when it seemed that the good times were there to stay, the next crisis emerged from an unexpected source. The devaluation of the Thai baht in July 1997 precipitated a series of Asian currency crises, which spread to the external debt markets of a wide range of developing countries in October 1997. The greater leverage at work in the market contributed to a far greater level of contagion than during the 1995 Tequila Crisis. Over an extended period, asset prices suffered extreme volatility across all sectors of emerging market debt, capped by extraordinary losses in Russia in August 1998. The pessimism at international financial gatherings was palpable, as new questions were raised by the crisis about the viability of emerging market debt as an asset class. In a more overt illustration that recovery value was overtaking the probability of default as a benchmark, the bonds of several countries started to trade on price rather than on a spread basis. The crisis itself also had its own victims. The rating agencies of course responded to the deterioration in sovereign creditworthiness with a wave of successive ratings downgrades, and the extended downturn forced speculative investors out of the market. The resulting lack of liquidity, at a time of heightened market stress, in turn further exacerbated other losses as well as reinforced the pessimism weighing on the asset class.
Stage 6: So Far, So Good (March 1999 to the present). After two disastrous years for emerging market fixed-income investors, 1999 finally turned out to offer some reward for those who stuck to their guns. Wider spreads offered attractive entry points for new investors, and technical factors in the market were broadly supportive since leveraged investors had largely been forced out. On the other hand, the very poor performance of dedicated emerging market funds in 1997–98 meant that very little new cash flowed to this kind of institutional investor. Instead, the marginal bid for the emerging market asset class came increasingly from high-grade investors with relatively lower risk tolerance who were looking to gain modest yield enhancement by dipping their toes into emerging market debt. These investors are known as “crossover” investors.
In order to focus this overview of international lending and capital inflows back onto the question of sustainable public sector finance in Latin America, it is worth drawing just a couple of conclusions from the gyrations in the emerging market debt universe in the second half of the 1990s. There are two important implications of this shifting investor base. First, it is worth bearing in mind that while varying degrees of leverage in the market can accentuate or dampen the contagion effects of financial crises, the ultimate focus of the institutional investor remains on traditional credit analysis reflecting the sovereign debtors’ willingness and ability to pay. This is particularly the case for so-called crossover investors who are forced to draw a sharp distinction between high-grade issues and non-investment-grade issues. The importance of sustainable public sector finance therefore extends beyond its inherent value for the country in and of itself and into financing costs for the government, which in turn helps to create a virtuous circle where the reward for good behavior in reducing noninterest expenditure, for example, is a further reduction in interest costs.
The second lesson at first glance appears to be a harsher one, however, in that no obvious remedy is at hand. The greater differentiation between those sovereign issuers that can attract crossover investors and those that cannot will condemn governments with weak public finances to alternative sources of funding. In the absence of developed local financial markets, this often means running arrears on payments to suppliers or on wages to public sector employees. Such a strategy undermines economic activity and is not sustainable, as Ecuador’s government discovered in the summer of 1999 (see Box 2). On the other hand, it is really only the legacy of those heady days of early 1997 that suggests that developing country governments are somehow entitled to tap international capital markets to fund budget deficits, and a dose of a harsher reality in this regard may also be the discipline necessary to push the more recalcitrant administrations in the direction of the reforms and improvements discussed at this conference.