ONGOING IMPROVEMENT OF COUNTRY RISK, YET OVERVALUED BY FINANCIAL MARKETS

Transcription

1 BFI Etudes Economiques February 10 th, 2004 Guy LONGUEVILLE Head of Country Risk Research ONGOING IMPROVEMENT OF COUNTRY RISK, YET OVERVALUED BY FINANCIAL MARKETS The decline in country risk continues, thanks to the favourable global economic environment and an ongoing improvement in LDCs fundamentals (I). We would like to focus, however, on certain hazards the low level of risk-free rates and spreads that might account for speculative bubbles and crises, reversal of the cycle in China, drastic increase in interest rates in the United States and their effects on emerging countries (II). I ONGOING DECLINE IN COUNTRY RISK LDCs have been benefiting in from a noticeably favourable global economic environment: - since 1999, oil prices have remained at a high level, with Brent crude averaging USD 27 in the period and expected to remain close to this level in High oil prices over such a long period have significantly reduced sovereign and non-transfer risks in oil-exporting countries 1, boosted their growth and prevented or curtailed risks of social explosion: Indonesia, Nigeria, Venezuela up to a point, etc.; - since 2002, there has been a significant rise in the price of most non-oil commodities: in USD, year-on-year, +39% in 2002 and 8% in A further rise is expected this year 2. This trend has noticeably helped Latin American countries, and to a smaller extent Sub-Saharan African countries, because their imports come mostly from the EU (effect of the appreciation in the EUR/USD) and certain exported commodities have risen only moderately: cacao, sugar, coffee; 1 For example, Russia s currency reserves are expected to stand at USD 100 billion at end-2004 (Source: IIF). 2 Chinese and US imports of commodities are growing, excess international liquidity seeking return are also heading into commodity markets, and the depreciation in the dollar offset by a rise in prices denominated in USD.

2 2.

3 3. - since 2003, growth in the OECD has picked up and Chinese growth has accelerated, implying an upturn in world trade, including manufactured products and the IT sector 3. The zones expected to benefit most are as follows: Latin America, Eastern Europe excluding CIS and above all Asia, with growth in intra-asian trade driven by China; In volume terms ,% chge GDP OECD US Import. of G & S OECD US Source: OECD, Dec since 2003, private capital has headed back into emerging countries, paving the way for a historic decline in risk premia (spreads on emerging-market high-yields in January 2004 under their trough of 1996) in a context of interest risk-free rates also down to a historically low level. Governments have seized this opportunity, tending to renew their debt denominated in foreign currencies at a lower cost and with a longer maturity rather than increase their indebtedness (Brazil, Bulgaria, Colombia, Mexico, Romania, South Africa, Turkey, Ukraine and Uruguay). Conversely, the private sector has increased in many cases its debt held in foreign currencies (Brazil, Chile, Croatia, Hungary, India, Russia, and Turkey). All in all, LDCs are benefiting in from a favourable macroeconomic and financial environment that had not been witnessed since , or very partly in LDCs are in a better position to benefit from the current global recovery than they did from previous ones This results from the following factors 5 : since , macroeconomic imbalances have declined, banking systems have consolidated, collaboration with international financial institutions has improved implying better support in the event of a problem breaking out, several countries have changed over to flexible exchange rate regimes, and others have successfully carried out a peaceful democratic transition: Argentina, Brazil, Mexico, Turkey, Indonesia, Thailand, notably. However, structural reforms have been implemented at a slow pace. This leads to: - better capacity to weather shocks because of lower external financial vulnerability than in ; 3 Global sales of semiconductors increased 15% in A 20% increase is expected in 2004 (source: Semi-conductor Industry Association) 4 In , the global recovery did not result in an upturn in capital inflows into LDCs, or in a significant rise in commodity prices. 5 Detailed in greater depth in previous country risk reports.

4 4. Liquidity and external solvency ratios of LDCs % of exp. of G&S % of imports of G&S (USD billion) Source: WEO Sept external debt external debt servicing Forex reserves (470) (641) (1119) - significant upside potential in imports before any external constraint kicks in. The surplus in the current-account balance of LDCs climbed to USD 98 billion in 2003 (1.8% of their GDP) while they were virtually balanced in 1998 (including a substantial deficit in Latin America). Furthermore, exports from LDCs should grow significantly (effect of robust global demand, greater competitiveness advantage); - greater potential to finance growth thanks to a rise in domestic savings (see the Table below) and a larger external financing capacity (better liquidity and external solvency ratios; excess global liquidity). Domestic savings in LDCs (as% of GDP) Total LDC Asia Lat. Am Source: IMF Sept.03 All in all, the recovery in growth in three major zones, USA, EU and Japan, should lead to or accelerate growth in LDCs; the latter, in contrast with , should not be rapidly hampered in their upswing by a deficit in credibility or a tightening in their external financial constraint; this point has to be nuanced naturally according to zones and countries (countries that could rapidly be hampered: Croatia, Egypt, Hungary, Romania and Turkey). Consequently, growth in LDCs should continue to accelerate in 2004, thanks to a sharp recovery in Latin America (Argentina, Brazil, Mexico, Uruguay and Venezuela) and without any significant deterioration in balances of payments. Real GDP growth (% change) Total LCDs Latin America Eastern Europe Africa & Middle East Asia Source: IIF January 2004

5 5. The reduction in country risk must be modulated geographically and according to risk category Southern and Eastern Asia, Eastern Europe and the CIS and, lastly, oilproducing countries in the Middle East are pressing ahead or consolidating their improvement in country risk. Africa as a whole is making no headway (with the exception of South Africa, and Angola thanks to its oil prospects). Latin America is experiencing contrasted trends but is improving overall: Brazil, Uruguay, and to a smaller extent Argentina and Colombia. The decline in non-transfer risk (rise in forex reserves, flexibility of exchange rates, cautious liberalisation of capital accounts) is indirectly improving sovereign risk in foreign currency, which is decreasing intrinsically. Despite the persistently high fiscal deficits of all LDCs (around 3.5% of GDP in 2004 as in according to the IMF), the proactive management of public debt is reducing the risk of non payment on the foreign currency-denominated part of this debt: the external debt is being refinanced at more attractive conditions, substitution of debt in local currency for debt held in foreign currencies thanks to abundant domestic liquidity (Colombia, Mexico, Venezuela, India, Malaysia, nearly all Central European countries etc) and deindexation of local currency debt, progress in HIPC initiatives in Sub-Saharan Africa, support provided by the IMF within the framework of orthodox policies that have acquired credibility. Large countries previously on the razor s edge (Brazil, Indonesia, Russia and Turkey) are now generating satisfactory recurrent primary budget surpluses. Reflecting these developments, rating agencies have steadily been upgrading sovereigns since However, the offhandedness of the Argentine government 7 is a disastrous example that other key countries (from a geopolitical point of view) could follow in case of major problems (Turkey?), given the lack of progress achieved with respect to improving the international financial architecture 8. The main country risk reductions Brazil, Russia and Turkey partly reflect the smaller risk of a trend reversal that would affect the favourable scenario under way for the last few years. Nonetheless, the factors that could trigger such a crisis have not disappeared: calling into question of economic policy under socio-political pressure in Brazil and Turkey, drastic decline in oil prices in Russia. But these evolutions reflect above all the better capacity of all three countries to weather a possible reversal of the current favourable scenario (which is difficult to foresee). In Brazil, this results from the smaller sensitivity of the public debt to market variables and from the reduction of the external financial constraint; in Turkey, from the cleaning-up of banks balance sheets and the support provided by the IMF and the US; in Russia it stems from the increase in assets in foreign currencies in a context of renewed growth. 6 For instance, for Standard & Poor s (report dated ), upgrades of sovereign debt in foreign currencies have outweighed downgrades since The ratio of upgrades relative to downgrades rose to a high level in 2003 (4:1). Given the current outlooks, this move should continue in The G-7 summit held on February 7, however, solemnly warned Mr Kirchner s government. 8 The restricted extension of CACs ( Collective Action Clauses ) and the talks still going on about a code of good behaviour point to a lack of progress. Furthermore, whereas the private debt held in foreign currency tends to be substituted for the public debt in FC (Russia, for example), the risks of a default in the medium term could concern private debt to a greater extent where as the IFI s are continuing to focus on the problems of public indebtedness.

6 6. II HAZARDS AND RISKS WEIGHING ON THE SCENARIO OF A GRADUAL REDUCTION IN COUNTRY RISK Does the excess reduction in spreads given the reality of underlying risks threaten to lead rapidly to financial crises, as was the case from 1996 to ? In the near term, we would think not. Admittedly, as in 1996, risk aversion decreased substantially in Last year, the contraction in high-yield spreads was significant for emerging-country sovereigns and corporates 9, and its magnitude was comparable to US and UE corporates. But, in contrast with 1996, in early 2004 no manifest asset overvaluation 10 has appeared in emerging countries (stock markets, exchange-rate markets and property) or excessively high debts, foreshadowing drastic corrections in the foreseeable future. This does not mean that a lengthy decline in spreads would not eventually lead to bubbles rising and result in crises. But this course of events does not seem likely in 2004 and to a certain extent in By comparing the two periods, the mitigating factors pointing to a low risk of asset bubbles are as follows: - net capital inflows into emerging countries, although increasing in , are half of what they were in 1996 (in relative value terms). Above all, the speculative or risky part of these flows, likely to feed bubbles (portfolio investment, short- term capital, bond issuance, remunerative investments in local banks), is far lower. In fact, trends in net flows in 2004 tend to reflect a return to normality, in line with the comparative growth prospects of developed countries and emerging countries; Net capital inflows into emerging countries USD billion % of GDP of emerging countries Source: IIF, January 04 - upstream from this slowdown in flows, the economic policies followed in large LDCs have become more responsible as a whole than they were in From a headlong flight forward into debt (example of Argentina), the main LDCs have changed over to proactive management of public debt that they are seeking to keep under control; - widely adopted after the crises of , flexible exchange rates do not encourage speculative investments. However, the forex reserves held by central banks of emerging countries have ballooned (to an even greater extent than in ), feeding their monetary base and domestic liquidity. More specifically: 9 Taking advantage of the low level of spreads and risk-free rates, emerging-country sovereigns completed in January one-quarter of the issues planned for the year. 10 This does not rule out, on a case-by-case basis, overvaluations: exchange rates of Turkey and South Africa, notably.

7 7. - the currency reserves of LDC central banks increased by about USD 200 billion per year in the period (Asia s contribution accounts for twothirds of this growth) compared with USD 50 billion per year in the period and USD 70 billion per year in ; - the faster increase in currency reserves of central banks in than in despite smaller net capital flows results from the improvement in current-account balances (currently surpluses versus deficits in the past) and, first and foremost, smaller net capital outflows carried out by residents who are regaining confidence in their country: Argentina, Brazil, China, Russia, and Turkey notably 11 ; - the increase in reserves is fuelling monetary bases and the money supply in LDCs but has not until now led to a marked upturn in domestic credits 12. Banking sectors are becoming over-liquid all in all, pointing to a rate cut rather than a massive upturn in credits. This phenomenon of rising monetary bases is particularly true for Asia. Central banks in Asian countries (that export manufactured products) massively buy US dollars to stabilise their currencies versus the US dollar, while their counterparts in Latin America (which primarily exports raw materials) allow their currencies to appreciate versus the US dollar and limit their US dollar purchases. Therefore, monetary bases grow faster in Asia than in Latin America (see box below). All in all, the transmission of speculative bubbles from developed countries to developing countries, as was witnessed in the past in the 1930s and mid-1990s is not yet on the agenda. The recoveries in stock markets of emerging countries in 2003 did not lead to high PER levels, unlike in Asia or Latin America. Moving in the wake of the dollar, exchange rates are not over-valued as a whole. There are still few real-estate bubbles. The increase in domestic credit remains moderate, despite the inflation in monetary bases. If the abnormality of spreads 13 were to persist, excessively high debt and bubbles could build up, but probably not before several years. In this respect, close attention needs to be paid to trends in domestic loans and the way they are used: in view of recent developments, domestic credit far more than external speculative capital could result in bubbles and an excessively high private debt load building up in some LDCs in the next few years. China (see below) is the archetypal example 14. Nevertheless, should speculative flows resume at a brisk pace, asset bubbles could inflate rapidly. A mitigating factor is the expected forthcoming rise in US rates, bound to widen spreads. 11 Estimated by the IIF, these net outflows did not exceed USD 30 billion per year in the period (bottoming out at USD 10 billion in 2003) against USD 160 billion per year in the period (high at USD 220 billion in 1996). 12 In the second half of 2003, in Asia, the year-on-year increase in loans to the private sector remained negative in the Philippines, Hong Kong, Singapore, and Taiwan. It was back to virtually zero in South Korea. The increase was close to 5% in Malaysia, Thailand, and Indonesia. Only China posted growth of more than 20%. 13 The causes have been analysed in previous country risk reports: against a context of international excess liquidity seeking a high return, the contraction in spreads has resulted from excess supply of the international capital market over the demand by LDCs. 14 Net growth in bank loans in China in stood at nearly 40% of 2002 GDP.

8 8. The possible swelling of a credit bubble: recent observations From last available data (4Q2003) on monetary basis, money supply, and domestic credit, the major emerging countries can be classified as follows: Group 1: strong growth in in monetary base, money supply and domestic credit: China, Russia. For the latter, this is a needed catch-up effect. Group 2: strong growth in in domestic credit, low growth in monetary base: Bulgaria, Czech Republic, Hungary, South Africa. Domestic demand fuels credit demand, through the growth in the credit multiplier. Group 3: strong growth in monetary base, low growth (or contraction) in domestic credit: Argentina (liquidity recovery), Indonesia, Malaysia, Mexico, South Korea, Thailand. Some other countries such as India, the Philippines and Turkey are in an intermediary state. All in all, growth in monetary base is fast in Asia, Russia and a few countries in Latin America, but does not fuel credit growth, except in China and Russia. What about the downturn in the economic cycle in China and its effects? 15 The risk of a downturn is fuelled by the concomitance of bottlenecks (transport, commodities, energy production) and current or future excess capacity (consumer goods and intermediate goods sectors), financed by excess credit growth. Assessing the likelihood and timing of such a downturn is difficult. At this stage, we can make the following comments: - The possible recessionary adjustment resulting from a correction of imbalances between supply and demand would be dampened by the underlying trend of sustained growth; - The effect of such a downturn on the banking system is difficult to assess. Of course, non-performing loans could grow further but the plentiful domestic savings could be partly invested in bond issuance aimed at recapitalising banks (this process begun in 2003); - For the reasons outlined above, Asia ex China and Japan (mainly South Korea, Indonesia, Hong Kong and Malaysia) should be able to absorb the indirect financial shock resulting from such a downturn. Conversely, growth in the real economy would be hurt 16, particularly in Hong Kong. 15 An in-depth study about this question is being drafted. 16 As a % of GDP, in 2002, exports to China accounted for 35.4% in Hong Kong, 7.5% in Singapore, 4.6% in Malaysia, circa 2.5% in Indonesia and Thailand. If we draw on intra-asian trade as a whole (with Japan included), the respective percentages are: 53%, 75%, 50%, 17% and 25%.

9 9. Effects of a possible dramatic rise in US bond yields. If this were the case and high-yield risk premia were to rise consequently, a few countries could face a tricky situation in terms of covering their external public financing in 2004: Ecuador, Indonesia, Philippines and South Africa, eventually Brazil. But the advance taken in their issuance programme allows them to curtail this risk. Sensitivity to a possible political slippage would be increased. Other hazards We had already mentioned another factor in a similar memo dated November 2003, i.e. the global recovery needs to firm for the countries or zones that have not yet managed to establish solid autonomous growth momentum: Latin America above all and, to some extent, Eastern Europe. In Asia, the outbreak of fresh epidemics needs to be monitored closely: SARS in the province of Guangdong in early February, bird flu epidemic especially in Vietnam and Thailand. Bird flu s spread is uncertain for now.