A NEW DATABASE ON THE CURRENCY COMPOSITION AND MATURITY STRUCTURE OF FIRMS BALANCE SHEETS IN LATIN AMERICA,

A NEW DATABASE ON THE CURRENCY COMPOSITION AND MATURITY STRUCTURE OF FIRMS’ BALANCE SHEETS IN LATIN AMERICA, 1990-2002 Definition of Variables, Methodology of Construction and Data Sources Herman Kamil

1

November 2004 Abstract

Given their potential implications for aggregate vulnerability and the conduct of exchange rate and monetary policy, understanding the magnitude, determinants and effects of currency and maturity imbalances at the firm-level has become a top priority for academics, country-policymakers and international financial institutions alike. To date, however, most efforts to conduct cross-country empirical research and policy analysis on these issues have been hampered by the paucity and lack of uniformity of microeconomic data on the currency and maturity composition of corporations’ asset-liability structures. In particular, new analytical approaches aimed at increasing surveillance of corporate sector indebtness and identifying financial vulnerabilities have remained less than fully operational due to the lack of crosssectional time series information on the extent of foreign currency denominated borrowing. This paper introduces a new cross-country database of firm-level data for public and non-publicly traded companies in Latin America that aims to start filling this information gap. This data set provides annual accounting and other relevant firm-specific information for approximately 2000 non-financial firms from ten Latin American countries, spanning the period 1990 to 2002. The database is unique in that it presents detailed, consistent and comparable information at the firm-level on the currency composition of assets and liabilities and the maturity profile of domestic and foreign currency denominated debt. The purpose of this document is twofold: (i) introduce this database as a new resource for academics and practitioners conducting research on issues of corporate finance, corporate governance and balance sheet effects in emerging markets and (ii) provide a set of benchmark regularities for policymakers interested in the design and implementation of macroeconomic policies and financial regulation in the region. We first describe the sources and methodology followed in the construction of the database. We then present summary statistics on key financial ratios and on the currency and maturity structure of the private sectors’ balance sheet that could serve as a basis for enhancing our understanding of corporate financial structures and firm dynamics in Latin America and other emerging markets.

1 Paper prepared for the Policy Seminar on Currency Mismatches organized by the Research Department, Inter-American Development Bank. This project was undertaken while the author was a Visting Scholar at the Research Department of the InterAmerican Development Bank. I am very grateful for their hospitality. I am especially indebted to Kevin Cowan for his help and encouragement through all stages of this project, and to Ugo Panizza, Arturo Galindo, Alejandro Izquierdo, Hermes Martinez, Erwin Hansen, Rudy Loo-Kung, Danielken Molina, Paula Auerbach, Patricia Yañez and Laura Clavijo for their help in putting this dataset together. I would also like to thank the following people for their cooperation in finding, collecting and/or interpreting data in each of the following countries: Sebastian Auguste, Sebastian Calonico, Patricia Hoffman, Yael Navarro and Florencia Garfinkel for Argentina; Carolina Villagra, Ana Maria Pacheco and Patricia Yañez for Bolivia; Patricia Berbia, Susana Okaze, Olavo Borges, Glaucia Ximenez, Silvio Zeitune and Rodrigo Ribeiro for Brazil; Roberto Alvarez, Christian Johnson, Procomer and Liliana Morales for Chile; Ana Fernanda Maihuasca and especially Leopoldo Fergusson for Colombia; Isabel Tan Chan for Costa Rica; Sangeeta Pratap, Carlos Alvarez, Lilia Bringas and Karla Siller for Mexico; Katharine Fierro, Luis Carranza, Juan Jose Cayo and Jose Berrospide for Peru; Ignacio Munyo, Pablo Miraballes, Martin Larzabal, Diego Melazzi, Gerardo Licandro, and Ana Alonzo for Uruguay; Cristina Santana and Maximiliano Gonzalez for Venezuela and Adair Morse, Paul Michaud and Brahima Coulibaly in the United States. Special thanks to Laura Vila for her unstinting support and valuable insights. I gratefully acknowledge financial support of the University of Michigan through the Rackham Pre-Doctoral Fellowship, the Discretionary Fund Fellowship, the Dean O.Bowan Research Fellowship, the Roosa Doctoral Fellowship in Monetary Economics, the Mitsui Fellowship on Firm Level Research in Emerging Markets and the CIBE Doctoral Research Fellowship on International Business. They were essential to complete this study. Andrea Pesce and Diego Guichon provided excellent research assistance. All remaining errors or omissions are my own. Contact address: 611 Tappan St., Department of Economics, University of Michigan, Ann Arbor, MI 48103. Email addresses: [email protected].

1. Motivation and Overview of the Database Mismatches in Theory The relationship between corporate balance sheets and a country's macroeconomy has received increased attention from academics, practitioners and policymakers in many emerging market countries. Underscoring this heightened scrutiny of firms’ financial structures is the prevailing view that currency and duration mismatches in firms’ balance sheets have increased both the likelihood and severity of recent financial crisis1. According to conventional textbook models, expansionary monetary policy and depreciation of the currency are optimal in response to an adverse foreign shock, like a sudden capital account reversal or a competitive devaluation of a main trading partner. These stabilizing properties of the exchange rate mechanism, however, can be offset or even reversed once we take into account the interplay between the economy’s adjustment to external shocks described above and the special characteristics of firm’s borrowing patterns in developing countries2. Mainly because of uncertainty about the future value of the domestic currency, most nonfinancial firms in emerging countries find it much easier to issue debt if the debt is denominated in foreign currencies and short term3. Thus, firms in emerging markets typically display a currency and maturity mismatch in their balance sheets. On the one hand, companies tend to hold large stocks of foreign currency-denominated liabilities and issue debt with relatively short maturity. On the other hand, firms’ revenue used to service and pay down this debt is primarily in domestic currency and linked to business assets installed for the long term and therefore illiquid. This particular asset -liability structure exposes firms to exchange rate and interest rate risk. After a sharp and sudden real exchange rate depreciation (exchange rate shock), firms which are highly leveraged in foreign currency but depend on local currency revenues (or, more precisely, whose revenues increase with the relative price of goods produced at home), will see the peso value of their debt expand more than the peso value of their assets or income. This would increase their real debt burden, probably pushing them into financially distress. A maturity mismatch, on the other hand, exposes a firm’s balance sheet to rollover and interest rate risk. If liquid assets do not cover maturing debts, a firm’s financial health is vulnerable to a rollover risk, especially during currency crisis when emerging markets’ firms can find themselves shut out of international capital markets. Furthermore, a sharp increase in interest rates (interest rate shock) can dramatically increase the cost of rolling over short-term liabilities, leading to a rapid increase in debt service4.

1

Hausmann, Panizza and Stein (2002) show that countries most likely to go into a crisis were those in which firms held substantial amounts of foreign currency-denominated debt. In empirical studies using a panel of macro data for developed and developing countries, Galindo, Panizza and Schiantarelli (2003) and Cespedes (2004) find that devaluations have a contractionary impact in countries with heavy liability dollarization. 2

This new view of currency crisis has centered on microeconomic corporate financial policies and has paid particular attention to the (changing) credit constraints faced by private sector firms during periods of steep exchange rate adjustments. Krugman (1999), Céspedes, Chang and Velasco (2000) and Aghion, Bacchetta and Banerjee (2000), among others, have stressed that - in the wake of large currency depreciations – widespread shorter-term liability dollarization increases the real debt service burden of the private sector , leading to an investment and output contraction. 3

The same holds true for banks and governments.

1

Any of the shocks described above can bring about a deterioration in the value of a company’s assets compared to its liabilities and hence a reduction of its net worth5. This has been specially so in many countries in Latin America, where the private sector has been unable to hedge exchange rate or interest rate movements due to inexistent or highly illiquid derivative markets6. The drop in a firm’s net wealth – and the ensuing deterioration of its creditworthiness and borrowing capacity in lending markets- affects the supply of credit for investment and the availability of short-term working capital, leading to an investment and output contraction. In cases were firms are highly leveraged, net worth may turn negative and the firm become insolvent. In the extreme case were the currency (or maturity) mismatch and devaluation (or interest rates) are large (or high) enough, many firms that were initially viable but inadequately hedged, can suddenly find themselves forced into capital liquidation or whipsawed into bankruptcy7. In summary, the interaction of short term dollarized debt and net worth complicates an economy’s response to external shocks, and may thus cause devaluations to be contractionary, not expansionary8. There is also evidence that monetary authorities do factor in this so-called “balance sheet effects” into their exchange rate and monetary policies. As documented by Hausmann et al. (2002) and Calvo and Reinhart (2002), the output costs associated to exchange rate fluctuations in highly dollarized countries have emerged as a prime reason why many Central Banks are reluctant to allow their currencies to devalue in response to external shocks. Thus, widespread liability dollarization has limited the capacity of the Central Bank to conduct stabilizing monetary policy. As emphasized in Goldstein and Turner (2004), to the extent that countries that officially claim to be adopting flexible exchange rate regime do not really float in practice, they give up the significant benefits of greater monetary policy independence and strain the flexibility of the domestic economy to self-correct a real exchange rate overvaluation. Theory Ahead of Practice As clearly summarized by Roubini and Setser (2004): ”Borrowers that depend on short term debt are in effect giving their creditors an option at par to exit when these debts mature. Firms (and countries) typically assume that they can refinance their existing debt rather than pay the debt as it matures. However, investors are – unsurprisingly- more inclined to exercise their option during bad times than in good times….Thus, a firm is likely to find that its costs of fund go up, even as an adverse shock may reduce its ability to pay. While the theoretical literature has mostly emphasized currency mismatches, maturity mismatches are also important because defense of a pegged exchange rate may lead to high real interest rates which are likely to harm firms that have short-term domestic currency debt. 4

5

Financial vulnerability arising from the interaction of currency and maturity imbalances and steep changes in relative prices, can be amplified by the existence of a capital structure mismatch , that is, too much debt relative to firm’s equity. In many countries, especially in Asia, the deterioration in non-financial balance sheets was also compounded by large drops in asset prices and the value of a firms’ collateral. There has been a significant growth, however, in foreign exchange rate derivative trading in recent years, especially in countries like Colombia, Mexico and Chile that have switched to floating exchange regimes. 6

Of course, if there is a financial crisis in the banking system, even firms who have no currency mismatch and see their profitability rise when a devaluation occurs may still be adversely affected if panic or financially stricken banks are no longer able to extend them credit (as witnessed recently in Argentina and Uruguay). 7

8

It should be noted that the possibility of contractionary devaluations was already analyzed - albeit in another context and through other channels- as early as 1963 by Diaz Alejandro, and later by Krugman and Taylor (1978) and Edwards (1989).

2

This financial vulnerability of dollarized countries was brought to the forefront in the late 1990s by the Asian crises and, more recently, by the banking crises in Argentina, Uruguay and Dominican Republic. Indeed, recent post-devaluation experiences in these economies have been associated with deep recessions rather than competitiveness-led booms. Many observers have argued that excessive reliance on short-term debt and un-hedged foreign currency borrowing in these countries, implied that firms (and their creditors) became financially distressed or suffered grave liquidity shocks when the currency depreciated and capital inflows came to a sudden stop. Deterioration of the corporate sector’s financial health, in turn, got spread quickly to all sectors of the economy through cross-firm interactions, financial disintermediation and government contingent liabilities. In light of these widespread collapses in economic activity and in an effort to detect and prevent future ones, understanding what drives firms in particular, and private agents in general, to choose the currency and maturity composition of their debt has become a key theoretical and empirical question for academics, market participants and policymakers alike. At the academic level, and stimulated by an influential paper by Eichengreen and Hausmann (1999) inquiring on the reasons why developing countries cannot borrow abroad in their own currencies and long term9, a recent stream of papers have used firm-level data in emerging markets to explore the determinants of debt dollarization10. These papers typically analyze whether the currency composition of debt at the firm level matches the firms’ sources of income, by testing if foreign currency leverage is related to the international tradability of output. At the policy level, recent crisis have prompted the development of new analytical approaches to the study of crisis detection and prevention. This new analytical framework is based on examination of stock variables (assets and liabilities) in the aggregate balance sheet of a country and the balance sheets of its main economic sectors, focusing on the risks created by maturity, currency, and capital structure mismatches11. Although these issues have generated an active academic and policy debate, they are only beginning to be studied empirically in the literature and in policy-making circles. The rub, of course, has been the data. Most efforts to conduct cross-country empirical research on (or intense surveillance of) the corporate sectors’ indebtness and exchange rate exposure have been hampered by the difficulty of assembling a comprehensive data set that contains consistent crosscountry, time-series firm-level data on the currency composition of assets, liabilities, sales and (potentially) off-balance sheet derivative positions. Good quality data on corporate sector balance sheets are hard to come by. For a start, balance sheet information has not been among the statistics routinely produced and disseminated by national authorities. Second, firms do not typically report the currency composition of their debt 9 The authors call this phenomenon original sin. (see also Eichengreen, Hausmann and Panizza (2003)).. In the original-sin interpretation, the problem is not that the private sector lacks incentives to borrow in domestic currency, but that it lacks the capacity to do so. A central empirical result on Eichengreen, Hausmann and Panizza (2003), however, is that the only variable that is robust in explaining country differences in original sin is economic size, measured by a country’s total GDP or total credit. All other macroeconomic variables fail to capture much of the cross-country variation in liability dollarization. See also Hausmann, Eichengreen and Panizza(2003), Goldstein and Turner (2004) and the references therein for a lively debate on the empirical and conceptual pertinence of this hypothesis.

Until very recently, empirical studies on the determinants on the currency composition of debt in emerging markets were focused solely on the currency denomination of sovereign debt (see, for example, Bonn (1990)). 10

11

See Allen, Rosenberg, Keller, Setser and Roubini (2002).

3

in their balance sheets on a voluntary basis, espec ially in countries like Bolivia, Uruguay, Costa Rica and Venezuela. Regulatory entities in most countries have only recently imposed the obligation to report disaggregated debt in standarized format in balance sheets, although in some countries this information is still subject to confidentiality restrictions12. Moreover, one of the most used commercial databases in the literature, Worldscope, has no information on the currency composition of assets and liabilities, and scant information on foreign sales for firms in developing countries. As a result, lack of firm level data on exchange rate and liquidity exposure variables has been pervasive in studies of emerging markets. Given the dearth of relevant data, is not surprising that little is known on the prec ise cross-country determinants of the currency composition of debt and currency imbalances at the microeconomic level13. Indeed, the few empirical studies on cross-country liability dollarization that exist have relied on indirect measures (such as the pass-through from the exchange rate to prices) rather than on quantity-based estimates of foreign currency liabilities. At the same time, available crosscountry measures of foreign currency debt do not fully capture the possible presence of currency mismatches (Eichengreen, Hausmann and Panizza, 2002 and Goldstein and Turner, 2004). The scarcity of relevant data has not been limited to the realm of academia. At a policy level, lack of information on the balance sheets of corporate borrowers has been flagged as a very important concern at international financial institutions (see Allen et al (2002)) and Goldstein and Tuner (2004))14. In most countries in the region, systemic risk emanating from corporations’ financial health is relatively high given that claims on the corporate sector represent large portions of bank’s assets15. On the other hand, and although much progress has been made in assessing the role of corporate balance sheet indicators in crisis detection and prevention (see Mulder, Perrelli and Rocha (2003)), lack of data on foreign currency borrowing may still hinder the predictive power of early warning indicators. This is specially important as recent experience suggest that currency and maturity imbalances in firms’ financial structures tend to remain hidden and unassessed during normal times, but have spelled disaster in the corporate sector following sharp

12

This is especially true in the case of Colombia and Bolivia.

One of the first exceptions is the paper by Bleakley and Cowan (2003), which looks at the cross-country determinants of dollar debt composition for a sample of 480 publicly traded firms in five countries in Latin America (Argentina, Brazil, Chile, Colombia and Mexico) between 1991 and 1999. The authors find that firms tend to match the currency composition of their liabilities with the ex-ante sensitivity of revenues to the real exchange rate. In other words, firms that produce tradable goods tend to hold more dollar debt than firms that produce tradable goods. As a consequence of this tendency towards natural hedging, Bleakley and Cowan find no evidence in support of a negative balance sheet. As discussed in detail in Galindo, Panizza and Schiantarelli (2003), the Bleakley and Cowan contribution is an important one. Yet, there are some limitations to this study. First of all, the panel is highly unbalanced: fifty percent of the observations come from Brazil. This is in itself would not be an important limitation , except for the fact that Brazil is a country where liability dollarization is fairly limited and where the government may have provided implicit hedges to firms that do hold dollar debt. Apart from issues on sample composition, the study does not control systematically for the presence of dollarized assets on the other side of the balance sheet, nor does it investigate the possibility of different determinants of dollar debt choice for different maturity profiles. Regarding the latter, one would expect, for example, that firms would be more prone to hedge short term dollar debt to avoid big liquidity shocks in the presence of a large devaluation. Allayanis et al (2003) and Luengnaruemitchai (2003) are two other recent studies that explore the cross-country determinants of the currency composition of debt for East Asia countries around the currency crises period. 13

14

In light of these data lacunae, the IMF has recently set new standards on data dissemination at the country-level that .

15 Recent studies from the IMF (2003) and IDB (2004) on banking stability in dollarized economies concluded that the main risk for banks in highly dollarized economies was the exposed position of their borrowers. The IMF reported that the share of dollar loans granted to borrowers in the non-tradable sector reached 50% in Costa Rica and 60% in Peru in 2002.

4

and unexpected changes in relative prices16. For this reasons, lack of data on the extent of foreign currency financing is a serious shortcoming for policy analysis in this area17. IDB’s Red de Centros Balance Sheet Project Recognizing this state of affairs, the Research Department of the Inter-American Development Bank (IDB) spearheaded in 2002 a Latin American research project called “Debt Composition and Balance Sheet effects of Exchange Rate Fluctuations in Latin America: A Firm Level Analysis”. One of the main goals of this project was to collect firm-level data on liability composition for a large sample of Latin American companies. Six independent research teams collected and analyzed balance sheet data for firms in Argentina, Brazil, Chile, Colombia, Mexico and Peru18. Very often these data was not readily available in electronic format and their collection required either buying expensive databases, having access to confidential information maintained by supervisory institutions or painstakingly collecting hard copy balance sheets and manually inputting the data. As a result of this project, new firm-level information was collected by the IDB for five Latin American countries: Argentina, Brazil, Chile, Mexico and Peru19. Subsequent efforts at the Research Department of the IDB focused on two goals: (1) assemble the different country data-sets while ensuring that variables’ definitions were uniform across countries, and that firm-level accounting information was accurate within countries, comparable across economies and consistent across time; and (2) pool new firm-level data sources to create a Final Database with a broader set of Latin American countries, a richer set of firm-variables (especially on the currency and maturity composition of balance sheet stocks) and a longer and more recent sample period. The end product of these undertakings is a new data set on annual accounting information covering roughly 2000 non-financial firms from ten Latin American countries, spanning the period 1990 to 2002. The database is unique in that it presents detailed, consistent and comparable information at the firm level on the currency composition of assets and liabilities and the maturity profile of domestic and foreign currency denominated debt. To our knowledge, this 16As discussed in Pettis (2001) and Roubini and Setser (2004), the true costs of both short term and foreign -currency debt are masked when times are good, growth is strong and capital inflows are plentiful. This macroeconomic environment was a defining feature of many Latin American countries that adopted exchange rate-stabilization plans during the 1990s. In these cases, short-term debt rolled over not only without difficulty but also often at lower interest rates based on expectations of dwindling inflation. In turn, expectations that interest rates would fall over time made economic actors reluctant to lock in contemporaneous interest rates. At the same time , private firms and banks may have interpreted relatively stable or fixed exchange rate regimes as a government’s promise to protect private borrower from currency risk, either by selling currency at a fixed rate or by providing a financial hedge (like in the case of Brazil) or an effective bailout in the event of a currency crisis (as it finally happened in Argentina). Thus, fixed or pegged exchange rate regimes may have made market participants complacent about currency risk. In addition, large capital inflows reinforced this perception by fueling a real appreciation of the currency and reducing the real burden of the outstanding foreign currency debt, providing firms with little incentives to hedge. As a result, the corporate sector ended up borrowing too much and underestimating future currency risk, leading to a large buildup of un-hedged foreign currency debt that left them expose to a sudden reversion of economic conditions .

Up until now, the only cross-country database available with a corporate breakdown of external debt is the one assembled by the Bank of International Settlements (see Moulder et al (2003) and Goldstein and Turner (2004) for an analysis of these data).

17

Information on the currency composition of debt of Colombian companies was not made available to the IDB at the firm-level for confidentiality reasons.

19

5

is the first time a comprehensive dataset has been put together for emerging market firms with information on the currency composition of stocks on both sides of the Balance Sheet. Distinguishing Features of the Database and Roadmap The dataset improves on existing publicly available data from commercial vendors and international institutions on several grounds. First, the databa se covers a broad set of ten Latin American countries: Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Mexico, Peru, Uruguay and Venezuela. Pooling these countries together ensures a wide variation in exchange rate regimes, trade openness, relative importance of bank (as opposed to equity) financing and regulations on dollar bank lending, among others. This high cross-country variation in economic and institutional structures provides an ideal testing ground to investigate the factors that affect currency composition of debt at the microeconomic level. Second, for each country in the sample, the database provides information on a substantial share of publicly traded firms (and not just the bigger or most liquid ones) covering all non-financial sectors of economic activity. More importantly, our database includes a variety of firms with nonpublicly traded shares20. The inclusion of this type of companies would allow analyzing the financial decisions of typically smaller and bank-dependent companies, and assessing the extent to which their financial choices differ from bigger and more financially sophisticated ones. Third, the discussion on most of the empirical literature on firm level response to sudden depreciations focuses on total dollar debt as the mechanism through which a change in the exchange rate can have balance sheet effects. By doing so, it is ignoring the fact that firms may also hold dollar denominated assets (be it productive assets, dollar-indexed government bonds, current assets in foreign banks or offshore investments) and may differ in their potential response of non-interest flows (like exports) to a change in the exchange rate21. In these cases, the inflated value of these sources of income following a depreciation could (partially) offset the negative balance sheet effect of dollar liabilities. Likewise, firms could differ in the maturity profile of dollar denominated debt and thus in their short term exchange rate exposure and financial vulnerability22. Thus, a good measure of currency mismatch has to consider both the fraction of foreign-currency denominated stocks on the asset and liability sides of the balance sheet and its maturity profile. In addition, it must take into account the ex-ante sensitivity of income flows to change in the real exchange rate23. Most commercial databases consist almost exclusively of publicly-traded companies, so that smaller and governmentowned companies are typically underrepresented.

20

For example, firms in Chile hold a significant amount of foreign assets. In our sample, the average ratio of dollar assets to total assets is 5.8%, very close to the 9.3% average of dollar liabilities. On the other hand, in the months leading up to the devaluation of the Real in 1999, the Brazilian government effectively provided exchange rate insurance to the private sector by increasing the issuance of domestic dollar-link bonds that were acquired by firms (and banks) in their asset portfolios. Therefore, in these cases, using total foreign debt to measure a company’s exchange rate exposure would overestimate the currency mismatch in the firms’ balance sheet. 21

These are clearly not the only factors affecting a firm’s exchange rate exposure. As argued by Adler and Dumas (1983), a firm’s exposure to exchange rate will be determined by at least 4 channels: (i) the impact of the exchange rate on nominal assets, (ii) the impact on the value of physical assets, (iii) the impact on sales price and unit costs and (iv) the indirect impact on sales volume.

22

Firm-level empirical evidence is far less conclusive regarding the contractionary impact of a depreciation on output and investment in the presence of currency mismatches. Using a sample of publicly listed Mexican firms in the 1990s, Aguiar (2002) finds that firms with large exposure to short-term foreign debt before the crisis showed a marked drop in investment after the devaluation. On the other hand, Bleakley and Cowan (2003) provide evidence that the negative 23

6

The distinguishing feature of this database is that provides detailed information on the currency composition of firms’ assets and liabilities, the duration of foreign currency liabilities and a breakdown of domestic and foreign sales. This information is crucial if we are to adequately measure the level of currency mismatch and thus the effects of this exposure on output and investment. Having access to a precise measure of foreign currency exposure at microeconomic and sectoral level across a wide range of countries is extremely important because the possible presence of negative balance sheet effects has important implications for the design economic policy, for the design of adequate regulatory frameworks to deal with dollarization risks, and for the debate on the optimal exchange rate regime.

2. Structure of the Database Description The database is un-balanced panel of annual firm-level observations spanning the period 1990 to 2002. It contains accounting and other relevant firm-specific information for approximately 2000 non-financial firms in ten Latin American countries: Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Mexico, Peru, Uruguay and Venezuela. The thrust of the information was collected from annual reports and corporate filings obtained from local stock markets, and financial statements from credit registries, regulatory agencies and/or business groups in each country. Where appropriate, we complemented these country-specific sources with data obtained from commercial data providers Economatica, Worldscope and Bloomberg. In addition to basic accounting data, the database also contains other key information about the firm that provides a picture of its production mix and export orientation, its access to international financial markets, ownership structure, multinational affiliation and a history of the main corporate events, including mergers, acquisitions and privatizations. Tables 1, 2.A and 2.B provide a description of all variables featured in the database, their coding in the dataset and corresponding sources. Table 3 shows the number of firm observations per country and year in the sample containing consistent balance sheet data. As shown, data for all countries except Bolivia, Costa Rica and Uruguay is available since 1990 and all countries - except Uruguay – have data available till 2002. Two things are worth pointing. First, the number of firms considered in the sample is substantially less for Bolivia, Costa Rica and Uruguay. In these countries, the reduced scale of the stock market and the opacity of financial information in the entrepreneurial sector imposed severe limitations on data collection24. Second, the size of the sample changes as new firms enter and exit the sample. To the extent that this entry/exit dynamics is governed by corporate events and not by firms’ delayed or erratic reporting of financial statements, it can provide valuable information regarding the interaction of firms’ financial structures and their economic performance. Thus, we track all firm-specific events balance-sheet effect is dominated by the competitiveness gains from a devaluation in a sample of publicly traded Latin American firms. More recent country-level studies summarized in Galindo, Panizza and Schiantarelli (2003) using micro data for Latin America, however, find that liability dollarization can reduce or possibly reverse the expansionary Mundell-Fleming effects of a devaluation. Evidence is also mixed in the case of Asia. On the one hand, Harvey and Roper (1999) find that balance sheets effects played a significant role in propagating the crisis, while Claessens, Djankov and Xu (2000) argue that inflated domestic debt and interest payments may have led to wide scale insolvency and liquidity problems in East Asian firms. On the other hand, Allayanis, Brown and Klapper (2003) and Luengnaruemitchai (2003) find no evidence that unhedged foreign currency debt was associated with significantly worse performance during the Asian crisis. 24 These caveats notwithstanding, this is the first time that information on the currency and maturity composition of firms in these smaller countries is made available in a standardized way.

7

related to bankruptcies, de-listing, privatizations or mergers and acquisitions when they occur. Table 4, for example, shows the fraction of firms (as a percentage of the total number of firms in the country-sample that year) that were merged or acquired by a foreign company on (or before) time t. Keeping track of corporate events helps to control more precisely for the potential accounting and selection bias built into the composition of the sample. For example, if those firms that go bankrupt or de-list and thus drop out of the sample are those with higher ratios of liability dollarization, then we would tend to observe an artificial reduction in the average foreign currency leverage due to changes in the composition of the sample. Likewise, firms that are merged generally experience a sharp increase in investment that may not be related to their operating performance but rather to consolidation of financial information25. This can be a potentially important empirical issue for studies of balance sheet effects around currency crisis relying on investment flows data. In particular, several recent studies have highlighted the fact that liquidity crisis could be consistent with an inflow of foreign capital, in the form of mergers and acquisitions (M&A) that seeks to take advantage of profitable investment opportunities in the hands of cash-strapped domestic corporations26. Firms in our sample that are merged or acquired by a foreign or domestic company are indicated by a dummy variable that takes the value of one starting on the year they were first bought. Thus, this identification provides one (possibly rudimentary) way to avoid or correct for the undue influence of mergers and acquisitions on the flow of investment27. We restrict our sample to non-financial companies. Given that currency mismatches are affected by banking regulation, the capital structure of banks is not comparable with the behavior of nonfinancial firms. Table 5 shows the distribution of firms by sector of economic activity, while Table 6 shows the fraction of firms with non-publicly traded shares in each country and year, that is, firms that are solely financed by private equity, bank loans or marketed debt. Table 7, in turn, provides summary statistics on the fraction of firms in each country-year that were exporters. Caveats and Data Limitations Several additional features of the dataset and some cautionary notes on data limitations or interpretation are worth highlighting from the outset. First, the data set contains detailed information on the capital structure of firms, but it does not include information on sources and uses-of-funds statements (most notably, investment). In addition, and as discussed in detail in the next section, although we attempted to correct for major differences in cross-country accounting and disclosure standards, significant differences may still persist for certain balance sheet variables. For example, maturity and currency mismatches are sometimes masked in indexed or floating rate debt instruments, making them 25

This is particularly relevant in the case of Mexican companies (see Pratap et al. (2003)).

Aguiar and Gopinath (2002) find that there was a substantial increase in M&A activity in South EastAsia between 1996 and 1998.

26

Bleakley and Cowan (2003) perform a more rigorous adjustment, whose implementation was beyond the scope of this study. In the event of a merger, a spin-off or a split, they construct an artificial firm that contains all of the component firms for the entire sample period. In the cases in which information on all component firms is not available they drop the firm from the sample. Given that Worldscope provides information on the reasons for which accounting data is no longer updated on all firms, they use this information to build the artificial firms. 27

8

less evident28. In some emerging market economies (e.g., Brazil) liabilities may be formally denominated in local currency, but indexed to the exchange rate. Similarly, the nominal maturity of an asset ma y be long but the interest rate it bears may be floating. Such indexation creates the same mismatches as if the debt were denominated in foreign currency or as if the maturity were as short as the frequency of the interest rate adjustments. Second, information on firm-specific characteristics like export shares, ownership structure, ADRs, privatizations, mergers and acquisitions, multinational affiliation, holdings of international assets and shareholder composition was typically not available, let alone presented in an uniform manner. Thus, linking this firm-specific data involved matching the baseline database of accounting information with numerous other sources of information that had no common identifier for a firm29. Since firms’ names are not reported identically across datasets (any may well vary over time, for example, following a merger, acquisition or privatization), merging these databases for all 2000 firms was far from a trivial exercise. Again, although we crosschecked the information across different data sources, potential pitfalls may remain. Third, given the difficulty in obtaining firm-level information in Latin America (especially on the currency composition of the balance sheet) sample selection was basically dictated by data availability. Table 8 presents information on the number of firm-year observations for which we have consistent data on the currency composition of firm-liabilities. Thus, the applied researcher should be aware that differences across and within countries in liability dollarization and other financial ratios could partly reflect the different nature of the country-samples. Two examples are noteworthy. First, due to restrictions on data availability, our measure of total dollar liabilities of Colombian firms only include financial debt, while excluding trade credit. This distinction is important, as trade related dollar debt is an important component of foreign debt in Colombia30. On the other hand, accounting figures for Argentina correspond to preliminary information on the third fiscal quarter of 2002, and thus do not capture the pesoification of the economy that occurred after the demise of the Convertibility regime. Quality and Consistency of Accounting Firm-Level Information The relevance of any empirical analysis - and especially the policy conclusions derived from it – can only be as good as the quality of the data upon which it is based. This is especially true in the case of research based on firm-level data in emerging countries, given the concerns that have been raised regarding differences in reporting accounting standards and the true state of firms’ balance sheets in developing countries31. In the particular case of Latin America, marked variations in accounting conventions across countries may raise concerns regarding differences in reported financial information of firms in different economies32. For example, most balance sheet statements in Brazil, Mexico and See a recent IMF(2003) report on The Balance Sheet Approach and its Applications at the Fund for a detailed discussion of this point.

28

Indeed, with the exception of Chile, firms in the original sample were only identified by name (and – in the case of Me Mexico- only with the stock market ticker). 29

30 31

See Echeverry et al. (2003) . On this issue, see Ratha, Shuttle and Mohapatra (2003).

32

As indicated by Kasa (2003), the accuracy of the balance sheet numbers themselves may not be as important as the need to control for different accounting regimes. A given set of accounting numbers can mean quite different things, depending on the specifics of bankruptcy and forbearance policies.

9

Venezuela are typically presented in consolidated fashion. Accounting conventions and disclosure requirements have also varied markedly across time within countries. For example, the accounting rules in Brazil were overhauled in Brazil in 1994 after the end of the hyperinflation, and Bolivia has only recently made it mandatory for firms to report disaggregated information on the currency composition of debt. Furthermore, given that in most part the construction of this database did not rely on commercial databases, the actual process of datacollection may have introduced potential measurement error s. Indeed, collection of key data on the currency denomination of balance sheets was rarely available in electronic format and thus required manually inputting the figures from hard copy, non-standardized templates33. With these potential pitfalls in mind, we have made every other effort to ensure consistency and accuracy between and within country comparisons of firm level financial information, as we describe in what follows. First, we have ensured a uniform accounting definition for a core set of variables across countries, relying on information on country-specific accounting practices and the Worldscope manual34. Second, and whenever possible, we crosschecked the basic accounting data information with Worldscope and Economatica. Since the two sources have information for some overlapping years, we were able check the consistency of the data and kept the companies for which the two sources reported the same information. Third, there were a number of extreme and unrealistic outliers that undoubtedly repr esent reporting or inputting errors. We addressed these and other miss-reporting problems by performing consistency checks based on different accounting identities. In this sense, we dropped all firm/year observations were the accounting data was not self-consistent. In particular, we drop observations if short-term liabilities (assets) exceed total liabilities (assets), when total dollar debt exceeded total debt, when financial ratios were clearly not correct35 or if accounting variables did not accord with sign conventions.

3. Additional Information on Selected Firm-Level Variables Table 1 provides the names, database-coding and detailed definition of every variable featured in the dataset. Below we provide additional information on a few selected variables on the methodology of construction that may be may be of interest to the applied researcher. Classification of Economic Sectors The classification of the economic sector for every firm in the dataset is based on the International Standard Industrial Classification, at three different levels of desegregation (ISIC1, ISIC2 and ISIC3). Thus, these differentiated classification of a firm’s production mix not only allows to control for a variety of industry effects at different levels of aggregation, but also provides more flexibility in matching this dataset with alternative sources of data (like industry output, labor hours or tariff barriers).

Data on the currency composition of debt for Argentina, Peru, Mexico and Chile draws heavily on the collection effort of the country research teams of the Red de Centros Project, as illustrated in Table 2. For the rest of the countries in the sample, we collected new and expanded information on liability and asset currency composition in the second stage of this project. 33

34 Accounting practices for Argentina, Brazil, and Mexico are described in Coopers and Lybrand(1993). Bavishi (1995) contains descriptions of accounting practices in the remaining countries.

35

We excluded firm-years observations were a company’s leverage was greater than 10.

10

Balance Sheet Variables The original accounting data in the database is reported in units of domestic currency, in current values and corresponds to last fiscal quarter. The key accounting variables are constructed as follows: Total Assets - a key variable in gauging the size of a firm – is defined as Sum of total current assets, long-term receivables, investment in unconsolidated subsidiaries, other investments, net property, plant and equipment, and other assets. Current Assets include highly liquid instruments such as cash as well as holdings that are normally liquidate rapidly, such as inventories and other intermediate goods. Current Liabilities include all liabilities coming due in the upcoming fiscal year. This measure includes debt issued at short maturities, as well as long-term issuances whose terminal date falls in the upcoming year. The measure of Foreign-Currency denominated liabilities is the book value of total foreigncurrency liabilities converted into local currency. In all countries in our sample, accounting standards dictate that conversion of debt from foreign to local currency values be carried out using the exchange rate for the period at the end of the fiscal year in which the balance sheet is reported. Several points regarding the definition and construction of this variable are worth highlighting. First we do not have a breakdown of foreign-currency denominated liabilities in terms of different currencies. As seems to be common usage in the literature on foreign-currency assets and liabilities, we use the term "dollar" to refer to any asset or liability denominated or indexed to a foreign currency. Second, data limitations also preclude identifying the jurisdiction dimension (i.e., the domestic/external origin) of dollar debt contracted by the firm. Finally, there is an important mechanical valuation effect linking exchange rate changes with measured dollarization shares. By construction, any dollarization ratio will increase after depreciation even if the stock of outstanding debt remains unchanged36.

36 Is important to note that valuation effects may be present regardless of the currency used to express the values of the variables.

11

4. Potential Uses of the Database This paper introduces a new and unique database containing detailed information on the currency and maturity composition of firms’ assets and liabilities in addition to data on firm-level exports for public and non-publicly traded firms in ten Latin-American countries. This new data available can extend the existing empirical and policy literature in potentially many directions. First, knowledge of the type, magnitude and duration of foreign currency debt outstanding and sectoral balance sheet mismatches can aid in the design and implementation of macroeconomic policies. For a start, as the corporate sector has increasingly become the main conduit for development finance, it has become increasingly important for policymakers and market participants to be aware of the scope and trends in corporate sector indebtness (both domestic and foreign) in emerging markets. Along the same lines, knowing whether open currency positions or maturity mismatches is the dominant source of corporate financial fragility can assist in making better informed policy choices in evaluating the stabilizing properties of exchange rate vis a vis interest rate adjustments37. At the same time, countries in which firms have large maturity mismatches are more willing to let the exchange rate float more freely and avoid the high real interest rates associated with a defense of the exchange rate. Therefore, the presence and magnitude of currency and maturity mismatches plays a determinant role in the assessment of the advantages and disadvantages of different exchange rate regimes. Along the same lines, accurate information on sectoral vulnerabilities arising from asymmetries in financing choices across tradable and non-tradable sectors, for example, can buttress policy advice in defining new directions in lending practices and banking supervision. For example, even if banks’ books are formally matched as a result of prudential measures that limit their net foreign exchange positions, banks may be still subject to substantial foreign exchange rate risk through their non-.financial borrowers’ risk of default. Indeed, as illustrated by a recent IDB(2004) study, banks in Latin America have faced an important credit risk (de facto a foreign exchange rate risk) arising from dollar-debtors whose earnings are not denominated or indexed to the dollar. Finally, a complete analysis of the financial vulnerability of the corporate level requires information on off-balance sheet positions, which can substantially alter the overall risk exposure of a firm. Due to important restrictions on availability of such specific data, our database does not include systematic information on off-balance sheet positions or off-balance sheet activities for the firms in the sample. Although financial transactions such as forwards, futures, swaps and other derivatives are not recorded on a balance sheet, they provide enhanced opportunities for risk sharing and thus can be used to effectively reduce the (recorded) risk created by balance sheet mismatches38. This issue is particularly important in light of the significant growth in foreign exchange rate derivative trading in recent years in Latin America. This trend has been

If the stock of short-term debt is large, tight monetary policy will both depress economic activity and increase the real burden of the domestic debt stock. If, on the other hand, the scale of the open currency positions is large, reducing the risk of a larger move in the exchange rate by allowing a rise of short-term interest rates may be the right answer.

37

By the same token, off-balance sheet activities can increase the risk exposure, if they are not used to hedge (taking a position that is negatively correlated to an existing balance sheet risk) but to speculate. 38

12

especially noticeable in countries that have switched to floating exchange regimes, as witnessed by the recent experiences in Brazil, Chile and Colombia39. Data on the use of derivatives at the firm level in Latin America is very limited, mostly because regulatory entities have imposed the obligation to report this kind of transactions only recently. Thus, identifying, collecting data on and assessing the off-balance sheet activities of Latin American companies (especially in Mexico, Brazil, Chile and Colombia) should be a fruitful area of future research as it would allow to sharpen our understanding of the corporate sector’s financial vulnerabilities.

5. Research that Has Used This Database Galindo A.; Panizza U.; Schiantarelli F. (2003) “Debt composition and balance sheet effects of currency depreciation: a summary of the micro evidence”. Emerging Markets Review, December 2003, vol. 4, iss. 4, pp. 330-339(10) [ download Science Direct] Galiani S.; Levy Yeyati E.; Schargrodsky E. (2003), “Financial dollarization and debt deflation under a currency board”. Emerging Markets Review, December 2003, vol. 4, iss. 4, pp. 340-367(28) [download Science Direct] Bonomo M.; Martins B.; Pinto R. (2003) “Debt composition and exchange rate balance sheet effect in Brazil: a firm level analysis.” Emerging Markets Review, December 2003, vol. 4, iss. 4, pp. 368-396(29) [download Science Direct] Benavente J.M.; Johnson C.A.; Morande F.G. (2003) “Debt composition and balance sheet effects of exchange rate depreciations: a firm-level analysis for Chile.” Emerging Markets Review, December 2003, vol. 4, iss. 4, pp. 397-416(20) [download Science Direct] Echeverry J.C.; Fergusson L.; Steiner R.; Aguilar C. (2003) “'Dollar' debt in Colombian firms: are sinners punished during devaluations?” Emerging Markets Review, December 2003, vol. 4, iss. 4, pp. 417-449(33) [ download Science Direct] Pratap S.; Lobato I.; Somuano A. (2003) “Debt composition and balance sheet effects of exchange rate volatility in Mexico: a firm level analysis. Emerging Markets Review, December 2003, vol. 4, iss. 4, pp. 450-471(22) [ download Science Direct] Carranza L.J.; Cayo J.M.; Galdon-Sanchez J.E. (2003) “Exchange rate volatility and economic performance in Peru: a firm level analysis.” Emerging Markets Review, December 2003, vol. 4, iss. 4, pp. 472-496(25) [download Science Direct] Cowan, K., E. Hansen and L.O. Herrera (2004) “Currency Mismatches, Balance Sheet E¤ects and Hedging in Chilean Non-Financial Corporations”, Prepared for the 8 th Annual Conferennce of the Central Bank of Chile. [download pdf] IADB (2005), Report on Economic and Social Progress in Latin America, Chapter 4. [link]

39 To the best of our knowledge, the only two studies that take into account corporate sector hedges through off-balance sheet transactions in the analysis of balance sheet effects are Cowan, Hansen and Herrera (2004) for Chile and Rossi (2004) for the case of Brazil.

13

14

References Adler, M. and B. Dumas (1983). “International Portfolio Choice and Corporation Finance: A Synthesis”. Journal of Finance 38 (3):925 - 84. Aghion, P., Bachetta, P. and A. Banerjee (2001a). “Currency Crises and Monetary Policy in an Economy with Credit Constraints”. European Economic Review 45 (7):1121 - 50. Aghion, P., Bachetta, P. and A. Banerjee (2001b). “A Corporate Balance-Sheet Approach to Currency Crises”. Forthcoming in the Journal of Economic Theory. Aguiar, M. (2002). “Investment, Devaluation, and Foreign Currency Exposure: the Case of Mexico”. Mimeo, Graduate School of Business, University of Chicago. Allayannis, G., Brown, G. and L. Klapper (2001). “Exchange Risk Management: Evidence from East Asia". Forthcoming in the Journal of Finance. Allayannis, G. and E. Ofek (2001). “Exchange Rate Exposure, Hedging, and the Use of Foreign Currency Derivatives”. Journal of International Money and Finance 20 (1):273 - 96. Arteta, C. O. (2002). “Exchange Rate Regimes and Financial Dollarization: Does Flexibility Reduce Bank Currency Mismatches?”. International Finance Discussion Papers # 738, Board of Governors of the Federal Reserve System. Arteta, C. (2003). “Are Financially Dollarized Countries More Prone to Costly Crises?”. International Finance Discussion Papers # 763, Board of Governors of the Federal Reserve System U.S. Baliño, T., Bennett, A. and E. Borensztein (1999). "Monetary Policy in Dollarized Economies". Occasional Paper No. 171, International Monetary Fund. Bavishi, V. B. (1995). International Accounting and Auditing Trends. Princeton, N.J., Center for International Financial Analysis & Research. Benavente, J. M., Johnson, C. and F. Morande (2003). “Debt Composition and Balance Sheet Effects of Exchange Rate Depreciations: a Firm-Level Analysis for Chile”. Emerging Markets Review 4:397 - 416. Bleakley, H. and K. Cowan (2002). “Dollar Debt and Devaluations: Much Ado About Nothing?”. Working Paper # 02-5, Federal Reserve Bank of Boston. Bohn, H. (1990). “A Positive Theory of Foreign Currency Debt”. Journal of International Economics, 29 (3):273-92. Bonomo, M., Martins, B. and R. Pinto (2003). “Debt Composition and Exchange Rate Balance Sheet Effect in Brazil: a Firm Level Analysis”. Emerging Markets Review 4:368 - 96. Calvo, G. A. (1999). "On Dollarization". Available at http://www.bsos.umd.edu/econ.

Calvo, G., Izquierdo, A. and L.F. Mejía (2004). “On the Empirics of Sudden Stops: the Relevance of Balance-Sheet”. Research Working Paper # 509, Inter-American Development Bank. Calvo, G. and C. Reinhart (2002). "Fear of Floating." The Quarterly Journal of Economics 67:379408. Carranza, L.J., Cayo, J.M. and J.E. Galdon-Sanchez (2003). “Exchange Rate Volatility and Economic Performance in Peru: a Firm Level Analysis”. Emerging Markets Review 4:472 - 96. Céspedes, L. F. (2004). “Financial Frictions and Real Devaluations”. Mimeo, Central Bank of Chile. Céspedes, L. F., Chang, R. and A. Velasco (2000). “Balance Sheets and Exchange Rate Policy”. NBER Working Paper # 7840. Claessens, S. and S. Djankov (2000). "Publicly Listed East Asian Corporates: Growth, Financing and Risks". In D. Dwor-Frecaut, F. Colaco and M. Hallward-Driemeier (eds.), Asian Corporate Recovery: Findings From Firm Level Studies in Five Countries. Washington D.C., World Bank:97-109. Claessens, S., Djankov, S. and L. Xu. (2000). "East Asian Corporations, Before and During the Recent Financial Crisis". World Bank Research Observer, v15, n1:23-46. Coopers and Lybrand (1993). “International accounting summaries”. New York: Wiley, John & Sons, Incorporated. Cowan, K. and Do, Q. (2003). "Financial Dollarization and Central Bank Credibility". Working Paper 3082, World Bank. Dwor-Frecaut, D., Colaco, F. and M. Hallward-Drienieier (2000). "Asian Manufacturing Recovery: A Firm Level Analysis". In D. Dwor-Frecaut, F. Colaco and M. Hallward-Driemeier (eds.), Asian Corporate Recovery: Findings From Firm Level Studies in Five Countries. Washington D.C., World Bank:1 - 17. Echeverry, J. C., Fergusson, L., Steiner, R. and C. Aguilar (2003). “Dollar Debt in Colombian Firms: are Sinners Punished During Devaluations”. Emerging Markets Review 4: 417 - 49. Fuentes, M. (2003). “Dollarization of Financial Contracts: Evidence from Chilean Firms”. Mimeo, Economics Department, University of California, Berkeley. Galiani, S., Levy-Yeyati, E. and E. Schargrodsky (2003). “Financial Dollarization and Debt De.ation Under a Currency Board”. Emerging Markets Review 4:340 - 67. Galindo, A. and L. Leiderman (2003). "Living with dollarization and the route to dedollarization". Conference on Financial Dedollarization: Policy Options, Inter-American Development Bank. Galindo, A., Panizza, U. and F. Schiantarelli (2003). “Debt Composition and Balance Sheet Effects of Currency Depreciation: a Summary of the Micro Evidence”. Emerging Markets Review 4:330 - 39.

Gelos, G (2003). "Foreign Currency Debt in Emerging Markets: Firm-Level Evidence from Mexico." Economics Letters, 78 (3):323 - 27. Goldstein, M. (1998). “The Asian Financial Crisis: Causes, Cures, and Systemic Implication”. Volume 55 of Policy Analyses in International Economics. Washington, DC: Institute for International Economics. Goldstein, M. and P. Turner (2004). “Controlling Currency Mismatches in Emerging Markets”. Institute for International Economics Publisher. Harvey, C. and A. Roper (1999). “The Asian bet”. In: Harwood, A., Litan, R., Pomerleano, M. (Eds.), The Crisis in Emerging Financial Markets. Brookings Institution Press 1999:29 - 115. Hausmann, R., Panizza, U. and E. Stein (1999). "Why Do Countries Float the Way They Float?". Working Paper No. 418, Inter-American Development Bank. Update. IMF (International Monetary Fund) (2003). “Financial Stability in Dollarized Economies”. Washington DC. Photocopy (April). Inter-American Development Bank (2002). “Debt Composition and Balance Sheet Effects of Exchange and Interest Rate Volatility: A Firm level Analysis”. Terms of Reference for Research Proposals. Kasa, K. (2001). Discussion on "How are Shocks Propagated Internationally? Firm-Level Evidence from the Russian and East Asian Crises". In R. Glick, R. Moreno and M. Spiegel, eds., Financial Crises in Emerging Markets. Krugman, P. (1999b). “Balance Sheet Effects, the Transfer Problem and Financial Crises”. In Isard, P., Razin, A., Rose, A. (Eds.), International Finance and Financial Crises. Kluwer Academic Publisher. Krugman, P. (1999). “Balance Sheets, the Transfer Problem, and Financial Crises”. International Tax and Public Finance 6 (4):459 - 72. Check Krugman, P., and L. Taylor (August 1978). "Contractionary Effects of Devaluation". Journal of International Economics 8:445 - 56. Levy-Yeyati, E. (2003). “Financial Dollarization: Where Do We Stand?”. Conference on Financial Dedollarization: Policy Options, Inter-American Development Bank. Luengnaruemitchai, P. (2004). “The Asian Crises and the Mystery of the Missing Balance Sheet Effect”. Mimeo, Economics Department, University of California, Berkeley. Martinez, L. and A. Werner (2001). “The Exchange Rate Regime and the Currency Composition of Corporate Debt: The Mexican Experience”. Presented in NBER Inter-American Seminar on Economics, July 20-21, 2001. Update. Pratap S., Lobato, I. and A. Somuano (2003). “Debt Composition and Balance Sheet Effects of Exchange Rate Volatility in Mexico: a Firm Level Analysis”. Emerging Markets Review 4 (2003):450 - 71.

Rajan, R. (2004). "How Useful Are Clever Solutions?". Finance and Development: 56 - 57. Schneider, M. and A. Tornell (2001) "Bailout Guarantees, Balance Sheet Effects, and Financial Crises". Review of Economic Studies, forthcoming. Tirole, J. (2002). “Financial Crises, Liquidity and the International Monetary System”. Princeton University Press. Tornell, A. and F. Westermann (2002). "The Credit Channel in Middle Income Countries". Working Paper No. 9355, National Bureau of Economic Research, November 2002.

Table 1

List of Variables in the Final DataSet Common Across All Countries Is their Information For All Firms in the Country Database? Name of Variable as Appears in Dataset

Definition

Argentina

Bolivia

Brazil

Chile

Colombia Costa Rica

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes* Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes*

Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes* Yes Yes* Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes* Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes* Yes Yes* Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes

Yes Yes*

Yes Yes Yes Yes Yes Yes Yes*

Yes Yes Yes Yes Yes Yes Yes*

Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes

Mexico

Peru

Uruguay

Venezuela

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes* Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes*

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes* Yes Yes* Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes*

Yes Yes Yes Yes Yes Yes Yes*

Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes

Firm Specific NAME YEAR ID COUNTRY_CODE TICKER ISIC_1 ISIC_2 ISIC_3 TRADABLE INTERNATIONAL_ASSETS EXPORTS AGE ADR+GDR ADR GDR PRIVTZ AFFMULT AFFMULTUSA AFFMULTRW PUBTRAD FOROWN MER_ACQ FORACQ FORACQ_50 TOTASST SHORTASST LONGASST TOTLIAB SHORTLIAB LONGLIAB TOTDOLASST SHORTDOLASST LONGDOLASST TOTDOLLIAB SHORTDOLLIAB LONGDOLLIAB SALES OPINCOME EBITDA CORPORATE_EVENT

Name of the Firm Year Firm Identification Code Identification Code for Country Stock Market Identification Code 1-Digit International Standard Industrial Classification 2-Digit International Standard Industrial Classification 3-Digit International Standard Industrial Classification Dummy=1 if Firm is in Tradable Sector (ISIC_1= 1, 2 or 3) Dummy=1 if Firm holds International Assets Total Exports of the Firm Age of the Firm each Year Dummy=1 starting on the year the Firm issued a Depositary Receipt Dummy=1 starting on the year the Firm issued an American Depositary Receipt Dummy=1 starting on the year the Firm issued a Global Depositary Receipt Dummy=1 starting on the year the Firm was Privatized Dummy Variable for a Firm being a Local Affiliate of a Foreign Multinational (FM) Dummy Variable for a Firm being a Local Affiliate of a US Multinational Dummy Variable for a Firm being a Local Affiliate of a FM other than the US Dummy Variable=1 if the Firm has Publicly Traded shares Dummy=1 if the Firm's main shareholder is a foreign company or conglomerate Indicator Variable indicating every time (if any) the firm in the sample was merged or acquired Dummy=1 starting on the year the Firm was first Merged or Acquired by a Foreign Firm Dummy=1 starting the year the firm was M & A by a Foreign Firm purchasing more than 50% of shares Total Assets Total Current Assets Total Long Term Assets Total Liabilities Total Current Liabilities Total Long Term Liabilities Total Dollar Assets Total Current Dollar Assets Total Long Term Dollar Assets Total Dollar Liabilities Total Current Dollar Liabilities Total Long Term Dollar Liabilities Sales Sales - (Costs+Administrative and Commercial Expenses) Earnings Before Interest, Taxes, Depreciation and Amortization Text tracking the ocurrence and date of a bankruptcy, delisting or merger or acquisition

Yes Yes* Yes* Yes Yes Yes Yes*

Consumer Price Index (Average). Base Year 1996 Consumer Price Index (End of Period). Base Year 1996 Domestic Currency Per Dollar (Average Nominal Exchange Rate) Domestic Currency Per Dollar (End of Period Nominal Exchange Rate) Lending interest rate in domestic currency (in %) Deposit interest rate in domestic currency (in %)

Yes Yes Yes Yes Yes Yes

Yes* Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes* Yes* Yes Yes Yes Yes

Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes* Yes* Yes* Yes Yes Yes Yes Yes Yes

Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes* Yes* Yes Yes Yes Yes Yes Yes

Yes* Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes* Yes* Yes* Yes Yes* Yes* Yes Yes

Country Specific CPI96AV CPI96END NOMEXCHAV NOMEXCHEND LENDRATE DEPRATE

Note: * denotes that there may be some firms with missing information

Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes

Table 2.A Sources and Definition of Variables in the Database Concept

Sectoral Classification

Age

Code

ISIC_1 ISIC_2 ISIC_3 TRADABLE

AGE

Definition

American or Global Depositary Receipt Issuance

PRIVATIZ

Privatization

Local Affiliation to Foreign Multinationals

Merger and Acquisitions

Foreign Ownership

Foreign Currency Denominated Balance Sheet Variables

Corporate Events

Firm Quotes in the Stock Market

TOTASST SHORTASST LONGASST TOTLIAB SHORTLIAB LONGLIAB SALES OPINCOME

Total Assets Total Current Assets Total Non-Current Assets Total Liabilities Total Current Liabilities Total Non-Current Liabilities Total Sales Operational Income Earnings Before Income, Taxes, Depreciation and Amortization

NOTES

Economatica Worldscope

SuperIntendencia de Valores de Colombia (SVCO)

Austin Asis

SuperIntendencia de Valores de Chile (SVC)

SVCO

Bloomberg

PROCOMER

SVCO

SVC

SVCO

2

SVSB and BVB

Exinet Database (Nosis)

Bloomberg Economatica Worldscope ADR-Universe (JPMorgan) World Bank Privatization Database Corporate Affiliations Database The Major Companies Database in E.Markets Directory of Foreign Firms Operating Abroad

SVC

SDC Platinum (Thompson Financials)

Bloomberg Economatica Worldscope Lexis Nexis Corporate Affiliations Database

Lexis Nexis Economatica Worldscope

Buenos Aires Stock Exchange

Bloomberg Economatica Worldscope

Buenos Aires Stock Exchange, Inspeccion General de Justicia and Other Regulatory Agencies

Total Dollar Assets Total Current DollarAssets Total Non-Current Dollar Assets Total Dollar Liabilities Total Current Dollar Liabilities Total Non-Current Dollar Liabilities Tracks the Ocurrence of Bankruptcies, Delistings or Mergers and Acquisitions

Notes: 1. Based on data compiled by Bleakley and Cowan (2003) : :

Ficha Estadistica Codificada Uniforme (FECUS)

2

SVC

SVCO

Bloomberg 1 Corporate Affiliations Database Lexis Nexis

PUBTRADE

EBITDA

Bloomberg

Buenos Aires Stock Exchange Inspeccion General de Justicia

ADR_GDR ADR GDR

TOTDOLASST SHORTDOLASST LONGDOLASST TOTDOLLIAB SHORTDOLLIAB LONGDOLLIAB

Colombia

Lexis Nexis

INTERNATIONAL_ASSETS

Aggregate Balance Sheet and Income Statement Variables

Chile

Date of Foundation

Ownership Structure

Equity Issuance

Brazil

SuperIntendencia de Valores y Seguros de Bolivia (SVSB) and Bolsa de Valores de Bolivia (BVB)

Buenos Aires Stock Exchange, Inspeccion General de Justicia and Other Regulatory Agencies

Exports

FOROWN

Bolivia

Lexis Nexis Economatica Worldscope

EXPORTS

MER_ACQ FOR_ACQ FORACQ_50

Argentina 1-Digit International Standard Industrial Classification 2-Digit International Standard Industrial Classification 3-Digit International Standard Industrial Classification Tradability of Output

Access to International Goods' and Capital Markets

AFFMULT AFFMULTUSA AFFMULTRW

Country - Specific Sources

Global Source

2. Based on data collected by Galiani, Yeyati and Schargrodsky (2003) 2. Based on data collected by Benavente, Johnson and Morande (2003)

Buenos Aires Stock Exchange, Inspeccion General de Justicia, Other Regulatory Agencies and Economatica 2

Worldscope

SVSB and BVB

Comision Nacional de Valores

SVC

SVCO

SVSB and BVB

Bloomberg

FECUS

SVCO

SVSB and BVB

Bloomberg

SVC 3

SVCO

2

Table 2.B Sources and Definition of Variables in the Database Concept

Code

Definition

Country - Specific Sources

Global Source Costa Rica

Sectoral Classification

Age

Access to International Goods' and Capital Markets

1-Digit International Standard Industrial Classification 2-Digit International Standard Industrial Classification 3-Digit International Standard Industrial Classification Tradability of Output

Lexis Nexis Economatica Worldscope

SuperIntendencia General de Valores de Costa Rica (SUGEVAL)

Date of Foundation

Lexis Nexis

SUGEVAL

EXPORTS

Exports

Economatica Worldscope

ADR_GDR ADR GDR

American or Global Depositary Receipt Issuance

PRIVATIZ

Privatization

ISIC_1 ISIC_2 ISIC_3 TRADABLE

AGE

AFFMULT AFFMULTUSA AFFMULTRW MER_ACQ FOR_ACQ FORACQ_50

Local Affiliation to Foreign Multinationals

Merger and Acquisitions

Ownership Structure FOROWN

Foreign Ownership

Aggregate Balance Sheet and Income Statement Variables

PUBTRADE

Firm Quotes in the Stock Market

TOTASST SHORTASST LONGASST TOTLIAB SHORTLIAB LONGLIAB SALES OPINCOME

Total Assets Total Current Assets Total Non-Current Assets Total Liabilities Total Current Liabilities Total Non-Current Liabilities Total Sales Operational Income Earnings Before Income, Taxes, Depreciation and Amortization

EBITDA

Foreign Currency Denominated Balance Sheet Variables

Corporate Events

TOTDOLASST SHORTDOLASST LONGDOLASST TOTDOLLIAB SHORTDOLLIAB LONGDOLLIAB NOTES

Bolsa Mexicana de Valores (BMV) and Comision Nacional Bancaria y de Valores (CNBV)

BMV and CNBV

Bloomberg Economatica Worldscope ADR-Universe (JPMorgan)

Peru

Uruguay

Venezuela

Comision Nacional de 4 Valores (CONASEV)

Bolsa de Valores de Montevideo (BVMO) and Auditoria General de la Nacion (AGN)

Bolsa de Valores de Caracas (BVC)

CONASEV

BVMO, AGN and MC Consultores

BVC

Peru Top1000

BVMO, AGN and Exinet Database (NOSIS)

BVC

CONASEV

BVC

World Bank Privatization Database Corporate Affiliations Database The Major Companies Database in E.Markets Directory of Foreign Firms Operating Abroad

CONASEV

BVMO

CONASEV

MC Consultores

BVC

SDC Platinum (Thompson Financials)

Bloomberg Economatica Worldscope Lexis Nexis Corporate Affiliations Database

Bloomberg 1 Corporate Affiliations Database Lexis Nexis

INTERNATIONAL_ASSETS

Equity Issuance

Mexico

Lexis Nexis Economatica Worldscope

Bloomberg Economatica Worldscope

Total Dollar Assets Total Current DollarAssets Total Non-Current Dollar Assets Total Dollar Liabilities Total Current Dollar Liabilities Total Non-Current Dollar Liabilities Tracks the Ocurrence of Bankruptcies, Delistings or Mergers and Acquisitions

Notes: 4. Based on data collected by Carranza, Cayo and Galdon-Sanchez (2003)

Worldscope

SUGEVAL

CNBV

CONASEV

BVMO

BVC

SUGEVAL

BMV and CNBV

CONASEV 4

BVMO and AGN

BVC

SUGEVAL

BMV and CNBV

CONASEV

4

BVMO and AGN

BVC

Table 3 Number of Firms Observations per Country and Year Country Argentina Bolivia Brazil Chile Colombia Costa Rica Mexico Peru Uruguay Venezuela Total

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Total

6

26

51

131

152

168

74 146

112 160

128 173

143 184

213

238

241 5

24 235 7

175 206 175 38 213 135 21 24

276 228 167 35 193 133 22 25

180 27 292 236 179 32 190 156 27 25

197 28 292 234 175 27 182 154 28 26

213 35 307 238 124 38 168 145 28 27

214 36 398 231 128 32 150 129 25 25

218 40 363 235 99 27 124 128 28 24

200 39 325 230 118 27 149 123 82 22

66 37 240 228 121 21 120 58 14

1,822 242 3,125 2,729 1,165 301 2,296 1,173 261 212

439

536

598

724

1139

1247

1344

1343

1323

1368

1286

1315

905

13,567

2002

Note: Each cell indicates the number of firm observations containing consistent balance sheet data

Table 4 Fraction of Firms Merged or Acquired by a Foreign Company (in %) Country Argentina Bolivia Brazil Chile Colombia Costa Rica Mexico Peru Uruguay Venezuela

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

9

11

6

5

9

12

14

15

17

18

18

1

1

1

1 2

1 1

2 1

5 2

1

1

2

3

5 2

6 5

8 4

7 3 2 4 8 5

4

4

4

4

9 3 5 5 7 6 4 4

11 4 6 9 10 7 4 8

14 5 6 7 13 7 4 8

14 6 6 7 10 9 2 18

Note: Each cell indicates the fraction of firms that had been acquired by a foreign company before or on that year, as a % of total firms in the : sample at time t

Table 5 Number of Firms by Productive Sector Sector Agriculture Mining Manufacturing Electricity, Gas and Water Construction Commerce Transport & Communications Services Miscellaneous Total

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Total

16 16 240 30 14 39 29 54 1

18 21 297 38 18 47 31 65 1

22 24 327 43 20 53 34 72 3

26 25 375 65 29 71 49 81 3

57 52 609 87 36 96 73 125 4

56 53 652 106 44 100 90 143 3

60 55 667 125 44 113 111 166 3

56 53 650 141 46 109 121 163 4

55 55 631 143 45 102 124 164 4

55 52 639 159 49 112 124 175 2

52 52 602 154 52 98 106 170

52 52 609 150 53 111 110 177 1

41 32 402 94 35 69 70 136 14

566 542 6,700 1,335 485 1,120 1,072 1,691 43

439

536

598

724

1139

1247

1344

1343

1323

1367

1286

1315

893

13,567

Table 6 Fraction of Firms with No Publicly Traded Shares (in %) Country Argentina Bolivia Brazil Chile Colombia Costa Rica Mexico Peru Uruguay Venezuela

1990 33

1991 27

1992 22

1993 62

1994 61

1995 64

39 52

35 49

31 49

29 45

98

88

85 60

0 82 29

27 48 85 0 75 56 5 21

31 50 86 0 20 53 0 20

1996 64 50 34 51 87 0 18 52 0 20

1997 64 42 32 49 85 0 14 49 0 19

1998 64 38 28 48 83 0 11 45 0 26

1999 63 38 32 47 83 0 9 36 0 28

2000 61 39 28 47 80 0 4 36 4 29

2001 62 43 24 45 83 0 13 38 67 36

2002 9 40 18 47 83 0 7 12 50

Note: Each cell indicates the fraction of firms which do not quote on the stock market at t, as fraction of total firms in the sample at time t

Table 7 Fraction of Firms with Exporter Status (in %) Country Argentina Bolivia Brazil Chile Colombia Costa Rica Mexico Peru Uruguay Venezuela

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

71

72

70

68

67

69 14 60 34 43

57 5 52 33 39

59 9 48 33 50

17 50 32 43

73 77 76

73 77 74

66 77 44

67

50

65

61 34

43 35

58 39 39

62 35 42

65 36 41

58 36 42

75 10 62 37 43

64

61

59

58

59 60 60

66 62 67

69 64 69

66 71 74

69 75 75

Note: Each cell indicates the fraction of firms which exported at t, as fraction of total firms in the sample at time t

Table 8 Number of Firms Observations per Country and Year Country Argentina Bolivia Brazil Chile Colombia Costa Rica Mexico Peru Uruguay Venezuela Total

1990 2

1991 17

1992 42

1993 124

1994 145

1995 161 232 205 159 30 191 126 14 17

1996 173 18 252 215 171 26 187 139 15 18

1997 189 19 251 211 172 22 179 146 24 20

1998 198 19 262 217 118 31 164 133 23 20

1999 196 16 255 209 123 27 143 122 23 14

2000 198 21 233 200 89 23 118 120 26 15

2001 181 29 215 206 107 23 142 115 67 5

2002 49 24 188 217 105 17 110

Total 1675 146 2347 1868 1204 243 2355 1024 205 124

90

106

120

210

235

236

14 231

143 188 160 30 209 123 13 15

212

342

384

489

1026

1135

1214

1233

1185

1128

1043

1090

700

11181

Note: Each cell indicates the number of firm observations containing consistent data on the currency composition of liabilities

ARGENTINA Country Summary Statistics

Table 1

Balance Sheet Data Whole Sample

Dollar Indebtness % Firms with dollar debt Debt dollarization ratio (%)

1

Short-dollarization ratio (%)

1

Long-dollarization ratio (%)

1

Asset dollarization ratio (%)

2

Debt Maturity Total debt maturity (%)

Dollar debt maturity (%)

3

4

Non-dollar debt maturity (%)

5

Debt maturity currency difference

Leverage Leverage ratio (%)

7

Exports As % of total assets

As % of total sales

6

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Mean No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

100.0 2 37.9 37.9 2 n.a. n.a. n.a. n.a. n.a. n.a. 9.4 9.4 2

100.0 17 48.5 46.3 17 n.a. n.a. n.a. n.a. n.a. n.a. 5.8 4.9 16

100.0 42 44.7 42.3 42 n.a. n.a. n.a. n.a. n.a. n.a. 7.2 5.3 42

96.0 124 50.6 53.8 124 44.1 43.8 49 51.2 62.1 49 7.6 3.6 54

92.4 145 51.6 58.6 145 38.5 38.1 57 55.4 85.8 57 8.8 4.4 59

95.7 161 55.5 62.1 161 45.9 47.2 68 58.0 86.6 68 8.6 4.1 62

94.8 173 55.1 61.8 173 42.6 42.9 75 64.4 89.0 75 8.7 4.2 67

95.8 189 56.5 63.2 189 42.5 38.3 88 68.5 92.7 88 8.7 3.3 81

96.0 198 56.3 60.7 198 45.9 44.9 115 59.7 87.0 115 7.8 3.3 104

96.4 196 57.3 64.0 196 48.1 49.2 136 60.0 87.5 136 7.3 3.3 108

95.5 198 58.5 64.7 198 48.8 50.2 134 62.8 84.8 134 8.4 4.0 111

95.6 181 60.1 67.3 181 53.3 57.2 121 58.7 77.5 121 9.0 4.5 103

98.0 49 55.2 63.4 49 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

21.1 20.6 6 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

33.8 36.2 26 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

28.7 22.1 50 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

22.3 13.9 131 29.0 16.7 46 11.4 2.4 49 18.8 5.4 46

24.7 17.6 152 39.2 41.0 51 10.9 3.7 57 29.2 27.0 51

24.0 15.7 168 35.7 26.8 65 10.7 5.7 68 25.2 24.2 65

28.9 20.7 180 49.5 57.6 72 13.6 8.0 75 36.7 39.4 72

32.6 26.0 195 54.3 62.7 85 14.3 8.8 88 40.6 46.7 85

30.9 24.9 212 45.3 47.2 111 17.0 8.3 115 28.4 28.8 111

30.2 22.9 213 41.4 43.2 130 16.6 8.2 136 24.5 22.6 130

30.9 24.2 218 44.6 50.4 129 16.1 11.2 134 28.4 28.0 129

32.4 27.5 200 42.0 42.6 117 19.5 12.6 121 23.0 22.8 117

31.1 24.9 66 n.a. n.a. n.a. 20.6 20.6 1 n.a. n.a. n.a.

Mean Median No. obs

41.7 48.0 6

33.5 29.3 26

33.8 32.3 51

43.9 40.2 131

47.2 45.2 152

47.8 45.3 168

50.6 47.9 180

50.9 48.8 197

51.1 48.8 213

53.6 54.3 214

53.7 54.6 218

53.9 54.3 200

54.9 55.1 66

Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

16.1 0.4 110 9.2 1.0 110

16.9 1.1 127 12.1 1.7 127

13.6 1.1 139 11.8 1.9 139

16.7 1.0 149 13.3 1.4 149

9.9 0.6 162 7.3 1.2 162

4.1 0.2 173 3.8 0.4 173

1.3 0.0 169 1.3 0.1 169

16.0 2.5 105 15.3 6.0 105

15.6 2.6 100 16.8 5.7 100

n.a. n.a. n.a. n.a. n.a. n.a.

Notes: Country summary statistics calculated using all the firms within the sample. 1 Dollar-linked debt as a percentage of total liabilities. 2 Dollar-linked assets as a percentage of total assets. 3 Debt (or asset) maturity = Long-term liabilities (assets) / total liabilities (assets). 4 Non-dollar debt (or asset) maturity = Long-term non-dollar liabilities (assets) / total non-dollar liabilities (assets). 5 Dollar debt (or asset) maturity = Long-term dollar liabilities (assets) / total dollar liabilities (assets). 6 Difference in maturity of dollar vis a vis non-dollar debt (assets). 7 Leverage = Total liabilities / total assets. Source: Economatica, Buenos Aires SE, Regulatory Agencies, Nosis External Trade and IDB calculations. Part of the data drawn from Galiani, Levy-Yeyati and Schargrodsky (2003).

BOLIVIA Country Summary Statistics

Table 2

Balance Sheet Data Whole Sample

Dollar Indebtness % Firms with dollar debt Debt dollarization ratio (%)

1

Short-dollarization ratio (%)

1

Long-dollarization ratio (%)

1

Asset dollarization ratio (%)

2

Debt Maturity Total debt maturity (%)

Dollar debt maturity (%)

3

4

Non-dollar debt maturity (%)

5

Debt maturity currency difference

Leverage Leverage ratio (%)

7

Exports As % of total assets

As % of total sales

6

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Mean No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

88.9 18 50.6 57.6 18 37.5 34.8 14 65.6 71.8 14 10.5 4.1 14

100.0 19 51.7 55.7 19 28.9 28.1 15 64.1 66.6 15 15.6 2.7 23

100.0 19 53.1 56.3 19 36.6 43.1 16 57.5 62.4 16 19.2 8.6 22

100.0 16 62.2 75.7 16 40.0 39.0 10 47.9 44.5 10 12.1 5.4 20

100.0 21 49.8 48.7 21 33.2 19.5 17 45.7 47.7 17 16.0 8.1 17

100.0 29 52.9 60.0 29 32.3 7.0 21 54.0 51.0 21 12.8 7.0 26

100.0 24 54.6 64.7 24 30.7 32.2 19 52.2 58.5 19 14.1 2.9 18

Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

49.6 53.2 27 65.8 70.3 14 34.2 34.9 14 31.63 31.16 14

50.8 50.0 28 71.4 89.3 15 29.6 24.1 15 41.77 49.95 15

44.4 39.4 35 56.0 57.9 16 30.4 27.7 16 25.63 17.68 16

46.2 42.9 36 55.6 61.4 10 31.3 33.2 10 24.33 23.09 10

47.7 52.2 40 64.1 82.1 17 41.2 37.2 17 22.83 45.37 17

45.0 42.8 39 71.9 79.4 21 35.1 31.7 21 36.79 52.46 21

49.4 51.1 35 70.4 78.7 19 42.9 42.6 19 27.44 35.77 19

Mean Median No. obs

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

43.0 42.3 27

39.6 37.9 28

43.7 40.9 35

41.8 40.9 36

41.5 42.2 40

41.6 40.6 39

44.6 48.9 37

Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

4.1 0.0 19 5.4 0.0 19

7.6 0.0 21 6.5 0.0 21

28.1 0.0 18 5.1 0.0 18

31.4 0.0 21 6.4 0.0 21

18.2 0.0 22 9.9 0.0 22

Notes: Country summary statistics calculated using all the firms within the sample. 1 Dollar-linked debt as a percentage of total liabilities. 2 Dollar-linked assets as a percentage of total assets. 3 Debt (or asset) maturity = Long-term liabilities (assets) / total liabilities (assets). 4 Non-dollar debt (or asset) maturity = Long-term non-dollar liabilities (assets) / total non-dollar liabilities (assets). 5 Dollar debt (or asset) maturity = Long-term dollar liabilities (assets) / total dollar liabilities (assets). 6 Difference in maturity of dollar vis a vis non-dollar debt (assets). 7 Leverage = Total liabilities / total assets. Source: Superintendencia de Valores y Seguros, Bolsa de Valores de Bolivia and IDB calculations.

BRAZIL Country Summary Statistics

Table 3

Balance Sheet Data Whole Sample

Dollar Indebtness % Firms with dollar debt Debt dollarization ratio (%)

1

Short-dollarization ratio (%)

1

Long-dollarization ratio (%)

1

Asset dollarization ratio (%)

2

Debt Maturity Total debt maturity (%)

Dollar debt maturity (%)

3

4

Non-dollar debt maturity (%)

5

Debt maturity currency difference

Leverage Leverage ratio (%)

7

Exports As % of total assets

As % of total sales

6

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Mean No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

51.1 90 11.8 1.1 90 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

50.9 106 11.5 1.2 106 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

55.8 120 12.4 3.3 120 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

58.7 143 11.7 3.6 143 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

51.3 232 11.6 0.3 232 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

56.0 252 13.8 4.0 252 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

55.0 251 15.7 4.2 251 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

56.9 262 17.1 5.8 262 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

58.8 255 17.0 3.8 255 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

63.9 233 19.1 15.2 233 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

66.5 215 20.4 16.4 215 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

69.7 188 20.2 14.9 188 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

32.2 31.8 73 n.a. n.a. n.a. 27.5 28.5 16 n.a. n.a. n.a.

33.0 30.2 109 n.a. n.a. n.a. 26.9 24.7 40 n.a. n.a. n.a.

34.8 34.1 125 n.a. n.a. n.a. 29.0 25.7 48 n.a. n.a. n.a.

36.5 34.9 139 n.a. n.a. n.a. 32.4 30.5 49 n.a. n.a. n.a.

37.3 36.7 171 n.a. n.a. n.a. 31.3 31.6 55 n.a. n.a. n.a.

36.1 35.2 271 n.a. n.a. n.a. 33.8 31.4 102 n.a. n.a. n.a.

39.5 41.4 283 n.a. n.a. n.a. 37.8 36.4 97 n.a. n.a. n.a.

41.4 43.4 283 n.a. n.a. n.a. 39.1 41.1 101 n.a. n.a. n.a.

40.7 42.9 298 n.a. n.a. n.a. 40.3 42.9 108 n.a. n.a. n.a.

41.4 43.2 381 n.a. n.a. n.a. 40.6 38.8 103 n.a. n.a. n.a.

45.0 46.3 356 n.a. n.a. n.a. 42.5 40.4 81 n.a. n.a. n.a.

45.0 47.4 317 n.a. n.a. n.a. 45.8 46.4 70 n.a. n.a. n.a.

44.1 44.6 233 n.a. n.a. n.a. 39.3 34.9 54 n.a. n.a. n.a.

Mean Median No. obs

44.1 41.2 74

33.5 33.1 112

37.3 35.8 128

39.4 38.1 143

39.4 37.4 175

44.7 40.8 276

50.8 46.3 292

54.8 48.9 292

60.6 50.5 307

70.2 57.7 398

74.5 59.8 363

78.7 60.6 325

80.2 63.2 240

Mean Median No. obs Mean Median No. obs

31.4 0.0 9 19.2 0.0 9

6.9 0.0 13 18.7 0.0 13

10.7 0.4 18 27.3 2.4 18

15.4 0.0 21 21.2 0.0 21

11.4 0.0 29 14.5 0.0 29

13.0 0.0 48 17.4 0.0 48

10.4 0.3 45 13.6 0.2 45

9.2 0.0 48 14.0 0.0 48

9.6 0.3 54 16.7 0.5 54

7.0 0.0 54 12.0 0.0 54

5.6 0.0 65 8.1 0.0 65

7.1 0.0 56 10.2 0.0 56

6.3 0.0 39 7.4 0.0 39

Notes: Country summary statistics calculated using all the firms within the sample. 1 Dollar-linked debt as a percentage of total liabilities. 2 Dollar-linked assets as a percentage of total assets. 3 Debt (or asset) maturity = Long-term liabilities (assets) / total liabilities (assets). 4 Non-dollar debt (or asset) maturity = Long-term non-dollar liabilities (assets) / total non-dollar liabilities (assets). 5 Dollar debt (or asset) maturity = Long-term dollar liabilities (assets) / total dollar liabilities (assets). 6 Difference in maturity of dollar vis a vis non-dollar debt (assets). 7 Leverage = Total liabilities / total assets. Source: Economatica, Bloomberg and IDB calculations.

CHILE Country Summary Statistics

Table 4

Balance Sheet Data Whole Sample

Dollar Indebtness % Firms with dollar debt Debt dollarization ratio (%)

1

Short-dollarization ratio (%)

1

Long-dollarization ratio (%)

1

Asset dollarization ratio (%)

2

Debt Maturity Total debt maturity (%)

Dollar debt maturity (%)

3

4

Non-dollar debt maturity (%)

5

Debt maturity currency difference

Leverage Leverage ratio (%)

7

Exports As % of total assets

As % of total sales

6

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Mean No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

62.2 188 21.3 4.5 188 19.5 6.8 160 16.4 0.0 160 5.1 0.0 181

58.0 205 20.6 2.3 205 19.9 4.2 184 18.7 0.0 184 4.7 0.0 202

59.1 215 23.0 5.5 215 20.5 6.8 190 23.1 0.0 190 4.4 0.0 208

63.5 211 25.2 7.0 211 20.4 5.4 190 27.1 0.0 190 4.8 0.0 208

63.6 217 27.0 7.3 217 22.2 5.8 188 28.7 0.0 188 6.0 0.0 211

61.2 209 26.4 6.8 209 19.4 4.4 181 30.5 0.0 181 4.5 0.0 204

52.5 200 20.5 0.3 200 18.1 0.4 179 21.1 0.0 179 6.7 0.0 207

60.7 206 20.5 2.4 206 17.0 1.5 180 20.0 0.0 180 6.8 0.0 207

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

33.7 25.6 146 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

37.3 32.2 160 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

36.2 29.3 173 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

35.9 28.9 184 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

36.7 30.5 206 29.6 4.8 106 42.9 40.4 160 -10.42 -12.65 106

37.5 31.5 228 31.7 12.9 114 42.0 35.8 184 -5.83 -8.50 114

38.5 34.1 236 36.1 20.5 121 40.6 38.8 190 -0.78 -1.08 121

42.0 41.9 234 41.0 35.9 129 40.5 34.4 190 5.08 -0.47 129

42.7 43.0 238 43.1 37.5 128 41.3 38.7 188 5.57 2.23 128

44.6 44.4 231 48.9 52.3 114 42.5 41.3 180 12.80 8.38 113

43.9 44.0 235 44.0 42.7 99 43.8 42.6 179 5.30 -0.67 99

42.2 39.9 230 39.1 21.8 112 45.3 43.6 180 -2.88 -7.59 112

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs

37.0 29.8 146

36.3 28.4 160

35.6 30.1 173

33.3 30.5 184

36.0 29.0 206

34.6 29.3 228

35.1 30.1 236

35.7 32.0 234

37.8 32.7 238

39.9 33.9 231

38.1 33.2 235

39.7 33.3 230

n.a. n.a. n.a.

Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

5.4 0.0 148 8.0 0.0 148

5.2 0.0 155 8.5 0.0 155

5.9 0.0 168 9.6 0.0 168

6.4 0.0 190 9.9 0.0 190

6.4 0.0 197 10.2 0.0 197

5.7 0.0 195 9.9 0.0 195

5.1 0.0 198 9.8 0.0 198

5.1 0.0 189 10.7 0.0 189

4.8 0.0 188 9.8 0.0 188

5.6 0.0 189 10.4 0.0 189

n.a. n.a. n.a. n.a. n.a. n.a.

Notes: Country summary statistics calculated using all the firms within the sample. 1 Dollar-linked debt as a percentage of total liabilities. 2 Dollar-linked assets as a percentage of total assets. 3 Debt (or asset) maturity = Long-term liabilities (assets) / total liabilities (assets). 4 Non-dollar debt (or asset) maturity = Long-term non-dollar liabilities (assets) / total non-dollar liabilities (assets). 5 Dollar debt (or asset) maturity = Long-term dollar liabilities (assets) / total dollar liabilities (assets). 6 Difference in maturity of dollar vis a vis non-dollar debt (assets). 7 Leverage = Total liabilities / total assets. Source: Ficha Estadística Codificada Uniforme (FECUS) and IDB calculations. Part of the data drawn from Benavente, Johnson and Moránde (2003).

COLOMBIA Country Summary Statistics

Table 5

Balance Sheet Data Whole Sample

Dollar Indebtness % Firms with dollar debt Debt dollarization ratio (%)

1

Short-dollarization ratio (%)

1

Long-dollarization ratio (%)

1

Asset dollarization ratio (%)

2

Debt Maturity Total debt maturity (%)

Dollar debt maturity (%)

3

4

Non-dollar debt maturity (%)

5

Debt maturity currency difference

Leverage Leverage ratio (%)

7

Exports As % of total assets

As % of total sales

6

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Mean No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

25.6 160 5.0 0.0 160 3.3 0.0 146 8.0 0.0 146 n.a. n.a. n.a.

23.3 159 5.1 0.0 159 3.7 0.0 107 9.2 0.0 107 n.a. n.a. n.a.

25.7 171 4.9 0.0 171 2.7 0.0 129 9.3 0.0 129 n.a. n.a. n.a.

27.9 172 5.5 0.0 172 2.1 0.0 133 10.5 0.0 133 n.a. n.a. n.a.

22.9 118 4.6 0.0 118 2.8 0.0 93 8.3 0.0 93 n.a. n.a. n.a.

26.0 123 7.1 0.0 123 5.6 0.0 97 9.5 0.0 97 n.a. n.a. n.a.

25.8 89 7.1 0.0 89 3.5 0.0 75 8.3 0.0 75 n.a. n.a. n.a.

28.0 107 6.4 0.0 107 4.3 0.0 89 8.7 0.0 89 n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

34.8 34.2 175 52.7 62.9 41 36.8 35.0 146 17.0 16.5 41

26.0 22.6 167 54.3 68.1 31 37.6 35.3 107 18.5 14.5 31

32.4 34.5 179 67.2 82.9 40 41.1 39.1 129 25.5 30.1 40

35.8 34.6 175 72.8 89.5 43 43.4 41.0 133 30.3 34.4 43

31.8 26.1 124 67.3 80.5 25 39.1 37.9 93 26.9 31.0 25

34.0 34.2 128 55.8 71.5 31 41.5 41.9 97 13.4 11.7 31

38.4 38.5 99 64.5 86.9 21 44.2 42.9 75 18.1 24.3 21

38.8 43.6 118 53.4 66.7 29 45.6 48.4 89 5.8 20.1 29

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

34.9 31.8 175

35.4 31.6 167

34.4 30.1 179

35.9 31.6 175

34.1 28.0 124

36.2 33.0 128

33.8 29.3 99

35.3 29.7 118

n.a. n.a. n.a.

Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

3.9 0.0 159 6.0 0.0 159

4.0 0.0 166 6.6 0.0 166

3.6 0.0 178 6.7 0.0 178

4.0 0.0 174 6.8 0.0 174

6.2 0.0 123 10.1 0.0 123

4.8 0.0 123 10.1 0.0 123

7.4 0.0 99 11.8 0.0 99

11.2 0.0 107 15.5 0.0 107

n.a. n.a. n.a. n.a. n.a. n.a.

Notes: Country summary statistics calculated using all the firms within the sample. 1 Dollar-linked debt as a percentage of total liabilities. 2 Dollar-linked assets as a percentage of total assets. 3 Debt (or asset) maturity = Long-term liabilities (assets) / total liabilities (assets). 4 Non-dollar debt (or asset) maturity = Long-term non-dollar liabilities (assets) / total non-dollar liabilities (assets). 5 Dollar debt (or asset) maturity = Long-term dollar liabilities (assets) / total dollar liabilities (assets). 6 Difference in maturity of dollar vis a vis non-dollar debt (assets). 7 Leverage = Total liabilities / total assets. Source: Superintendencia de Valores and IDB calculations.

COSTA RICA Country Summary Statistics

Table 6

Balance Sheet Data Whole Sample

Dollar Indebtness % Firms with dollar debt Debt dollarization ratio (%)

1

Short-dollarization ratio (%)

1

Long-dollarization ratio (%)

1

Asset dollarization ratio (%)

2

Debt Maturity Total debt maturity (%)

Dollar debt maturity (%)

3

4

Non-dollar debt maturity (%)

5

Debt maturity currency difference

Leverage Leverage ratio (%)

7

Exports As % of total assets

As % of total sales

6

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Mean No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

78.6 14 23.8 14.3 14 28.6 28.6 13 12.1 0.0 13 5.4 0.1 14

86.7 30 34.9 36.8 30 34.8 41.5 27 22.7 9.6 27 8.2 4.1 30

93.3 30 40.0 42.2 30 39.3 47.4 27 34.1 14.2 27 8.6 6.0 31

88.5 26 37.5 40.8 26 39.4 46.5 23 42.7 37.1 23 7.3 2.9 26

90.9 22 37.7 37.9 22 39.2 39.1 20 41.5 31.8 20 7.4 4.3 22

93.5 31 45.4 48.4 31 41.5 42.3 28 49.5 49.2 28 8.8 5.5 30

92.6 27 48.3 51.1 27 48.7 51.4 25 48.5 53.6 25 11.5 7.5 26

95.7 23 56.8 65.2 23 56.6 64.4 19 53.7 56.5 19 19.0 13.6 23

95.7 23 64.3 71.0 23 60.0 70.2 21 57.7 70.4 21 23.4 17.6 23

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

22.4 20.1 24 4.5 0.0 11 24.5 22.3 13 -16.87 -12.09 11

22.8 19.5 38 14.9 6.2 24 25.0 22.6 27 -9.37 -14.05 24

27.7 20.9 35 23.5 12.1 26 25.9 19.5 27 -0.16 -6.79 26

25.4 25.7 32 31.7 28.2 22 21.6 20.3 23 10.09 11.30 22

25.4 27.6 27 27.1 20.6 19 19.8 18.9 20 6.27 -7.04 19

30.6 28.7 38 38.5 31.5 27 19.6 20.3 28 18.46 7.07 27

29.2 22.3 32 37.5 33.2 24 23.7 18.2 25 12.90 2.35 24

30.1 29.0 27 33.7 25.5 19 24.6 15.4 18 9.53 2.27 18

29.2 18.3 27 38.0 31.3 21 27.9 23.6 20 10.54 8.15 20

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

46.6 57.8 24

46.2 56.4 38

51.3 56.7 35

50.0 56.0 32

49.8 56.0 27

46.0 55.1 38

45.4 52.7 32

43.1 48.2 27

42.5 44.0 27

n.a. n.a. n.a.

Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

Notes: Country summary statistics calculated using all the firms within the sample. 1 Dollar-linked debt as a percentage of total liabilities. 2 Dollar-linked assets as a percentage of total assets. 3 Debt (or asset) maturity = Long-term liabilities (assets) / total liabilities (assets). 4 Non-dollar debt (or asset) maturity = Long-term non-dollar liabilities (assets) / total non-dollar liabilities (assets). 5 Dollar debt (or asset) maturity = Long-term dollar liabilities (assets) / total dollar liabilities (assets). 6 Difference in maturity of dollar vis a vis non-dollar debt (assets). 7 Leverage = Total liabilities / total assets. Source: Superintendencia General de Valores and IDB calculations.

MEXICO Country Summary Statistics

Table 7

Balance Sheet Data Whole Sample

Dollar Indebtness % Firms with dollar debt Debt dollarization ratio (%)

1

Short-dollarization ratio (%)

1

Long-dollarization ratio (%)

1

Asset dollarization ratio (%)

2

Debt Maturity Total debt maturity (%)

Dollar debt maturity (%)

3

4

Non-dollar debt maturity (%)

5

Debt maturity currency difference

Leverage Leverage ratio (%)

7

Exports As % of total assets

As % of total sales

6

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Mean No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

81.9 210 29.8 23.0 210 29.7 24.1 195 26.9 0.0 195 4.7 0.6 7

81.7 235 30.8 25.2 235 33.2 30.3 212 28.0 0.0 212 4.6 1.7 93

80.9 236 31.0 24.1 236 31.7 24.8 220 27.7 0.0 220 3.2 1.2 110

84.0 231 33.9 28.3 231 32.1 27.0 219 33.9 12.8 219 4.1 1.2 120

85.2 209 42.0 40.0 209 39.3 39.4 198 41.9 29.5 198 7.1 2.6 108

89.5 191 45.0 43.8 191 43.6 43.6 182 42.9 32.6 182 6.8 3.7 92

89.3 187 42.6 41.4 187 38.3 37.5 171 45.7 48.0 171 9.0 2.7 116

87.7 179 45.4 48.9 179 37.9 35.4 168 49.9 60.9 168 10.8 3.4 132

89.6 164 45.4 44.4 164 38.7 36.8 155 50.9 58.8 155 11.1 3.7 117

89.5 143 42.2 43.3 143 34.1 31.0 134 51.1 64.7 134 13.0 10.9 13

92.4 118 37.1 37.6 118 37.7 35.9 115 34.7 25.8 115 9.9 11.6 11

88.0 142 33.3 33.5 142 33.5 26.1 139 30.3 15.8 139 0.0 0.0 0

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

25.8 22.7 213 26.0 4.9 163 25.7 21.7 195 1.0 -2.4 163

29.6 26.4 238 26.5 7.3 180 32.9 29.5 212 -6.0 -8.8 180

34.0 34.2 241 28.5 13.0 180 35.4 34.0 220 -6.7 -6.1 180

37.8 39.4 235 35.9 33.4 186 35.1 36.2 219 1.0 -2.1 186

36.5 36.1 213 37.5 36.0 172 31.7 29.8 198 7.1 6.8 172

35.8 32.8 193 35.6 32.6 166 30.8 27.2 182 5.7 2.3 166

39.2 40.5 190 41.5 47.2 156 32.3 27.1 171 9.2 8.4 156

39.5 40.7 182 45.1 53.6 150 28.1 23.1 168 16.9 20.9 150

37.3 38.2 168 45.4 53.5 143 25.9 19.1 155 20.2 18.6 143

36.0 37.0 150 44.9 54.4 122 23.8 16.2 134 21.0 20.2 122

46.6 48.8 124 43.1 48.2 107 46.9 46.0 115 -4.7 -3.3 107

46.4 48.0 149 40.4 43.8 123 47.1 47.3 139 -7.9 -9.3 123

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs

38.4 37.5 213

41.2 41.4 238

42.5 42.5 241

43.1 43.3 235

46.9 46.5 213

51.2 50.0 193

48.6 45.3 190

47.1 43.9 182

49.1 43.7 168

52.5 44.6 150

55.8 53.6 124

55.7 51.1 149

n.a. n.a. n.a.

Mean Median No. obs Mean Median No. obs

7.4 2.2 210 10.1 3.1 210

6.0 0.9 235 8.5 1.4 235

5.5 1.0 234 8.3 1.3 234

4.8 0.8 231 8.4 0.8 231

5.5 0.7 209 9.4 1.2 209

10.9 3.2 191 16.9 5.8 191

11.5 4.1 187 15.7 5.0 187

12.4 4.3 179 16.8 6.4 179

13.0 3.2 164 17.8 6.1 164

11.9 4.4 143 16.4 6.5 143

13.0 5.2 117 18.0 7.6 117

9.6 2.0 142 14.3 3.3 142

n.a. n.a. n.a. n.a. n.a. n.a.

Notes: Country summary statistics calculated using all the firms within the sample. 1 Dollar-linked debt as a percentage of total liabilities. 2 Dollar-linked assets as a percentage of total assets. 3 Debt (or asset) maturity = Long-term liabilities (assets) / total liabilities (assets). 4 Non-dollar debt (or asset) maturity = Long-term non-dollar liabilities (assets) / total non-dollar liabilities (assets). 5 Dollar debt (or asset) maturity = Long-term dollar liabilities (assets) / total dollar liabilities (assets). 6 Difference in maturity of dollar vis a vis non-dollar debt (assets). 7 Leverage = Total liabilities / total assets. Source: Mexican Stock Exchange and IDB calculations.

PERU Country Summary Statistics

Table 8

Balance Sheet Data Whole Sample

Dollar Indebtness % Firms with dollar debt Debt dollarization ratio (%)

1

Short-dollarization ratio (%)

1

Long-dollarization ratio (%)

1

Asset dollarization ratio (%)

2

Debt Maturity Total debt maturity (%)

Dollar debt maturity (%)

3

4

Non-dollar debt maturity (%)

5

Debt maturity currency difference

Leverage Leverage ratio (%)

7

Exports As % of total assets

As % of total sales

6

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Mean No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

100.0 123 53.1 52.5 123 58.1 61.0 45 41.8 42.7 45 16.1 12.7 120

99.2 126 57.4 62.2 126 53.3 54.4 45 43.3 32.8 45 16.5 13.2 124

97.8 139 58.8 62.0 139 64.0 69.4 44 44.8 40.0 44 19.2 17.4 139

100.0 146 59.2 62.8 146 56.4 60.0 52 42.8 30.0 52 19.1 16.9 141

100.0 133 67.2 73.8 133 69.4 74.6 45 56.4 68.2 45 22.7 20.9 133

100.0 122 64.8 73.1 122 60.7 66.7 52 52.5 71.3 52 21.2 19.1 120

100.0 120 62.9 66.7 120 52.9 51.6 59 47.3 51.9 59 21.6 20.5 118

100.0 115 63.5 71.6 115 54.1 53.0 58 51.7 53.4 58 21.8 20.4 113

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

21.5 21.1 4 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

20.8 22.1 6 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

25.9 22.3 135 24.3 19.5 45 36.2 32.6 45 -11.8 -16.3 45

24.9 21.7 133 22.5 16.5 44 24.6 20.9 45 -1.2 -5.1 44

25.2 23.8 155 22.8 15.8 42 29.3 23.8 44 -5.9 -9.3 42

27.2 24.5 153 28.8 12.2 52 27.7 20.7 52 1.0 -5.8 52

29.1 27.8 144 28.8 25.6 45 32.9 33.3 45 -4.2 -1.4 45

31.2 28.9 127 26.0 16.6 52 27.2 18.3 52 -1.3 -1.1 52

34.7 33.7 127 38.2 36.6 59 41.8 40.5 59 -3.6 -1.3 59

37.8 42.1 123 42.2 46.4 58 41.8 39.1 58 0.5 3.9 58

36.3 37.4 57 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs

n.a. n.a. n.a.

n.a. n.a. n.a.

42.3 29.3 5

38.1 31.7 7

45.7 45.6 135

46.9 47.4 133

46.2 45.4 156

46.4 47.3 154

49.3 48.1 145

47.8 44.6 129

49.0 45.1 128

49.1 43.6 123

48.4 44.6 58

Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

14.7 0.2 126 15.9 0.2 126

15.1 0.1 129 15.8 0.1 129

12.3 0.3 150 15.5 0.4 150

12.9 0.8 150 16.2 1.2 150

11.5 1.1 138 18.1 1.9 138

11.9 1.0 124 18.5 2.6 124

12.4 1.0 120 18.9 2.5 120

12.9 2.9 113 19.5 5.0 113

n.a. n.a. n.a. n.a. n.a. n.a.

Notes: Country summary statistics calculated using all the firms within the sample. 1 Dollar-linked debt as a percentage of total liabilities. 2 Dollar-linked assets as a percentage of total assets. 3 Debt (or asset) maturity = Long-term liabilities (assets) / total liabilities (assets). 4 Non-dollar debt (or asset) maturity = Long-term non-dollar liabilities (assets) / total non-dollar liabilities (assets). 5 Dollar debt (or asset) maturity = Long-term dollar liabilities (assets) / total dollar liabilities (assets). 6 Difference in maturity of dollar vis a vis non-dollar debt (assets). 7 Leverage = Total liabilities / total assets. Source: Comisión Nacional de Valores (CONASEV) and IDB calculations. Part of the data drawn from Carranza, Cayo and Galdon-Sanchez (2003).

URUGUAY Country Summary Statistics

Table 9

Balance Sheet Data Whole Sample

Dollar Indebtness % Firms with dollar debt Debt dollarization ratio (%)

1

Short-dollarization ratio (%)

1

Long-dollarization ratio (%)

1

Asset dollarization ratio (%)

2

Debt Maturity Total debt maturity (%)

Dollar debt maturity (%)

3

4

Non-dollar debt maturity (%)

5

Debt maturity currency difference

Leverage Leverage ratio (%)

7

Exports As % of total assets

As % of total sales

6

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Mean No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

100.0 13 76.5 87.7 13 75.3 84.5 6 90.6 99.9 6 10.9 6.6 13

100.0 14 80.5 89.1 14 75.2 80.5 6 92.4 100.0 6 14.2 11.9 14

100.0 15 74.3 90.5 15 78.7 87.6 8 71.5 100.0 8 14.0 7.8 16

100.0 24 77.9 85.2 24 76.4 78.9 16 91.0 100.0 16 20.2 14.6 24

100.0 23 80.7 87.4 23 79.3 83.3 15 87.7 100.0 15 18.6 17.1 23

100.0 23 82.3 88.5 23 74.4 78.7 13 96.0 100.0 13 15.1 6.8 23

100.0 26 83.9 88.8 26 80.4 85.5 18 90.7 100.0 18 16.2 10.2 26

100.0 67 77.6 85.7 67 80.9 92.0 34 89.3 100.0 34 26.8 17.2 76

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

20.8 11.8 21 23.1 19.9 6 3.6 0.6 5 23.6 21.6 5

25.5 19.1 22 54.6 56.6 6 3.4 0.0 6 51.1 56.6 6

28.5 12.2 27 35.3 32.7 8 17.2 0.0 8 18.1 20.7 8

30.4 24.2 28 42.7 44.4 16 3.0 0.0 14 41.8 37.8 14

31.3 32.2 28 38.7 36.0 15 9.5 0.0 14 29.4 34.6 14

32.3 35.1 25 46.4 40.7 13 8.2 0.0 13 38.1 34.1 13

29.2 23.1 28 33.2 28.1 18 14.4 0.0 18 18.8 16.4 18

17.3 2.9 82 36.7 33.3 34 6.2 0.0 32 30.5 29.2 32

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

52.9 53.5 21

51.9 53.7 22

55.8 54.7 27

57.1 58.2 28

58.1 59.8 28

58.5 53.9 25

60.7 57.1 28

65.4 58.3 82

n.a. n.a. n.a.

Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

17.4 2.3 18 22.7 9.1 18

24.1 5.2 19 25.6 16.2 19

19.5 6.3 24 26.0 8.4 24

18.9 4.8 26 25.0 6.4 26

17.7 6.1 27 26.2 10.2 27

13.3 7.9 25 24.5 15.0 25

13.3 5.7 27 25.5 10.9 27

11.7 0.0 72 14.5 0.0 72

n.a. n.a. n.a. n.a. n.a. n.a.

Notes: Country summary statistics calculated using all the firms within the sample. 1 Dollar-linked debt as a percentage of total liabilities. 2 Dollar-linked assets as a percentage of total assets. 3 Debt (or asset) maturity = Long-term liabilities (assets) / total liabilities (assets). 4 Non-dollar debt (or asset) maturity = Long-term non-dollar liabilities (assets) / total non-dollar liabilities (assets). 5 Dollar debt (or asset) maturity = Long-term dollar liabilities (assets) / total dollar liabilities (assets). 6 Difference in maturity of dollar vis a vis non-dollar debt (assets). 7 Leverage = Total liabilities / total assets. Source: Bolsa de Valores de Montevideo, Auditoría General de la Nación, and IDB calculations.

VENEZUELA Country Summary Statistics

Table 10

Balance Sheet Data Whole Sample

Dollar Indebtness % Firms with dollar debt Debt dollarization ratio (%)

1

Short-dollarization ratio (%)

1

Long-dollarization ratio (%)

1

Asset dollarization ratio (%)

2

Debt Maturity Total debt maturity (%)

Dollar debt maturity (%)

3

4

Non-dollar debt maturity (%)

5

Debt maturity currency difference

Leverage Leverage ratio (%)

7

Exports As % of total assets

As % of total sales

6

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Mean No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

93.3 15 36.0 34.0 15 n.a. n.a. n.a. n.a. n.a. n.a. 13.3 14.6 15

94.1 17 34.7 34.8 17 n.a. n.a. n.a. n.a. n.a. n.a. 13.0 10.6 16

94.4 18 25.5 20.6 18 n.a. n.a. n.a. n.a. n.a. n.a. 11.5 9.1 17

95.0 20 29.7 28.6 20 n.a. n.a. n.a. n.a. n.a. n.a. 7.8 6.4 19

100.0 20 26.6 17.2 20 n.a. n.a. n.a. n.a. n.a. n.a. 5.4 3.9 18

92.9 14 25.3 19.0 14 n.a. n.a. n.a. n.a. n.a. n.a. 4.6 4.2 14

93.3 15 26.0 18.4 15 n.a. n.a. n.a. n.a. n.a. n.a. 4.3 3.3 13

100.0 5 34.3 34.7 5 n.a. n.a. n.a. n.a. n.a. n.a. 6.6 1.9 7

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs Mean Median No. obs Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

38.0 36.4 23 n.a. n.a. n.a. 23.0 23.0 1 n.a. n.a. n.a.

37.9 30.7 24 n.a. n.a. n.a. 62.1 62.1 1 n.a. n.a. n.a.

36.6 33.8 24 n.a. n.a. n.a. 23.8 23.8 1 n.a. n.a. n.a.

36.6 32.7 26 n.a. n.a. n.a. 56.2 56.2 1 n.a. n.a. n.a.

25.3 23.6 27 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

30.1 32.1 25 n.a. n.a. n.a. 1.0 1.0 1 n.a. n.a. n.a.

28.7 26.2 24 n.a. n.a. n.a. 0.6 0.6 1 n.a. n.a. n.a.

29.0 26.7 22 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

35.8 35.2 14 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Mean Median No. obs

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

43.8 47.7 24

43.0 44.0 25

32.4 31.1 25

27.8 25.1 26

25.7 23.3 27

27.8 28.0 25

33.1 34.3 24

34.5 31.7 22

37.9 34.9 14

Mean Median No. obs Mean Median No. obs

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

n.a. n.a. n.a. n.a. n.a. n.a.

Notes: Country summary statistics calculated using all the firms within the sample. 1 Dollar-linked debt as a percentage of total liabilities. 2 Dollar-linked assets as a percentage of total assets. 3 Debt maturity = Long-term liabilities / total liabilities. 4 Non-dollar debt maturity = Long-term non-dollar liabilities / total non-dollar liabilities. 5 Dollar debt maturity = Long-term dollar liabilities / total dollar liabilities. 6 Difference in maturity of dollar vis a vis non-dollar debt. 7 Leverage = Total liabilities / total assets. Source: Bolsa de Valores de Caracas and IDB calculations.