wp5 W.xl¸ POLICY RESEARCH WORKING PAPER 2815 Pricing Currency Risk Facts and Puzzles from Currency Boards Sergio L. Schmukler Luis Serven The World Bank Development Research Group Macroeconomics and Growth March 2002 I POLICY RESEARCH WORKING PAPER 2815 Abstract Schmukler and Serven investigate the patterns and derived from interbank rates, particularly during times of determinants of the currency risk premium in two crisis. The large magnitude of these cross-market currency boards-Argentina and Hong Kong. Despite differences can be the consequence of unexploited the presumed rigidity of currency boards, currency arbitrage opportunities, market segmentation, or other premium is almost always positive and at times very risks embedded in typical measures of currency risk. The large. Its term structure is usually upward sloping, but premium and its term structure depend on domestic and flattens out or even becomes inverted at times of global factors related to devaluation expectations and turbulence. Currency premia differ across markets. The risk perceptions. forward discount typically exceeds the currency premium This paper-a product of the Macroeconomics and Growth, Development Research Group-is part of a larger effort in the group to understand how capital market and exchange rate regimes work. Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact Emily Khine, room MC3-347, telephone 202- 473-7471, fax 202-522-3518, email address kkhine@worldbank.org. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at sschmukler@worldbank.org or Iserven@worldbank.org. March 2002. (76 pages) The Policy Research Working Paper Seoes disseminates the findings of work in progress to encourage the excaange of ideas about devJelopment issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Produced by the Research Advisory Staff Pricing Currency Risk: Facts and Puzzles from Currency Boards * Sergio L. Schmukler World Bank and Luis Serven World Bank JEL classification codes: F31, F36, G12, G15 Keywords: currency risk; currency premium; forward discount; currency board; term structure; covered interest parity; market segmentation; financial crises 4 We are grateful to Gustavo Cafionero, Stijn Claessens, James Conklin, Sebastian Edwards, Chang Hsieh, Marcio Garcia, Michael Kumhof, Eduardo Levy Yeyati, John Merrick, Maury Obstfeld, Roberto Rigobon, and David Sekiguchi for their valuable comments and suggestions. We also benefited from feedback received at presentations held at the NBER Inter-American Seminar on Economics in Cambridge, the Latin American meetings of the Econometric Society in Buenos Aires, the LACEA meetings in Montevideo, and the World Bank/IMF Joint Research Seminar. For excellent research assistance, we are particularly grateful to Leonor Coutinho Gouveia. We also thank Andrea Bubula, Tatiana Didier, Penny Lermanometee, and Yaye Sakho, who helped us at different stages of the project. For help in obtaining data we thank Alberto Ades, Gustavo Cafionero, Starla Cohen, Kevin Fan, Ricardo Martinez, Daniel Oks, and Daniel Tenengauzer. For financial support we thank the World Bank Latin American Regional Studies Program, PREM anchor unit, and the Research Support Budget. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors and do not necessarily represent the views of the World Bank. Contact address: World Bank, 1818 H Street NW, Washington, DC 20433. Phone (202) 458- 4167. Fax: (202) 522-3518. Email addresses: sschmuklergworldbank.org, lserven(aworldbank.org I. Introduction Interest rate differentials, the spread between local interest rates and international interest rates, are a key variable for emerging countries. Spreads are usually positive for these countries, which implies that they face a higher cost of capital than developed economies. Interest rate differentials vary substantially over time and increase in periods of local and foreign financial turmoil, and this has lead to "sudden stops" of economic activity in emerging economies.' Lower spreads typically translate into lower borrowing costs, for both the public and private sector, and higher growth. For that reason, the ways to achieve a reduction in interest rate differentials have been recently at the center of academic debate and have been a major concern for policy makers Conceptually, the total differential between interest rates on domestic currency- loans issued by local borrowers and those on foreign-currency loans issued by foreign borrowers reflects both country and currency premia. The former refers to the gap between the borrowing costs of domestic and foreign borrowers in a common currency. The latter, on which this paper focuses, refers to the gap between the domestic-currency and foreign-currency interest rates faced by a given borrower; it is often called "currency risk premium" and, less precisely but more popularly, currency risk. Of the two components of interest rate differentials, the country risk premium has been intensively studied, perhaps due to the availability of daily cross-country data. Indexes of yield spreads on emerging market bonds (EMBIs) are compiled by JP Morgan. Data on primary issues also exist. The literature has studied the behavior of yield spreads See Calvo (1998). including their time pattern, determinants, and cross-country comovement.2 Some papers also study the country risk premium in relation to the currency premium.3 The other component of interest rate differentials, the currency premium, has received less direct empirical attention in the context of emerging economies. Still, the currency premium is relevant to several strands of the literature - like those on exchange rate determination, uncovered interest parity, and real interest parity. The present study relates directly to at least four different strands of the international finance literature: the debate on the choice of exchange rate regime, the assessment of economic perfornance under currency boards, the term structure of currency premia, and covered interest parity. First, the debate on the choice of exchange rate regime pays particular attention to the currency premium. Participants in this debate, which intensified during the currency crises of the 1 990s, have claimed that countries should opt for either hard pegs or floating regimes. Proponents of hard pegs argue that, other things equal, the adoption of a rigid parity - such as a currency board - should reduce the currency premium, even eliminating it entirely if the peg is viewed as irrevocable. In this view, hard pegs are thought to be credible and transparent, and this yields financial stability and low inflation. As a consequence, hard pegs would reduce the level of domestic interest rates.4 Credible hard pegs would also reduce the probability of currency attacks and contagion effects. 2 See, for example, Edwards (1984), Edwards (1986), Favero, Giavazzi, and Spaventa (1997), Eichengreen and Mody (1998), Kamin and von Kleist (1999), Mauro, Sussman, and Yafeh (2000), Kaminsky and Schmukler (2001), Merrick (2001), and Rigobon (2001). 3 See Domowitz, Glen, and Madhavan (1998), Sturzenegger and Powell (2000), Didier and Garcia (2001), and Druck, Moron, and Stein (2001). 4 Note, however, that even if the currency premium declines the country premium could rise if adopting a rigid peg is perceived to weaken the country's solvency. In such case, the net effect on the level of borrowing cots would be ambiguous. 2 But as Edwards (2000) suggests, the currency premium can still be significantly positive even in hard pegs, if they are not fully credible. Second, the debate on exchange rate regimes has generated a related literature on economic performance under currency boards. Ghosh, Gulde, and Wolf (1998) find that currency boards are associated with better inflation performance and higher output growth. Kwan and Lui (1996) argue that currency boards tend to slow down output growth but reduce inflation. They also claim that currency boards might result in higher output volatility than flexible regimes. Rivera Batiz and Sy (2000) argue that currency boards yield more credibility and better economic performance than simple pegs. Hausmann (2001) discusses the conditions that might help alleviate potential problems due to the rigidity of currency boards. Calomiris and Powell (2001) describe how the Argentine currency board helped in the development of the financial system. The third strand of the literature relevant to this paper is the one that studies the term structure of currency premia. The term structure reflects markets' perception of depreciation and exchange risk at different horizons, and has been studied mostly in the literature on target zones. For example, Svenson (1991) shows that under a credible target zone the absolute value of the interest rate differential is decreasing in the time to maturity, since the expected depreciation until maturity is bounded by the exchange rate band. Bartolini and Bodnar (1992) study the term structure of forward premia to assess the implied credibility of the French/German target zone under the European Monetary System. Weak currencies are found to be associated with upward-sloping forward premia, as investors forecast a further depreciation of the currency over the next periods. Their results also show that the short-term premium fluctuates more than the long-term 3 premium. Domowitz, Glen, and Madhavan (1998) examine the term structure of the currency premium in the case of Mexico up to the Tequila crisis. They show that the term premium turned negative before and during the crisis. The fourth strand of the literature directly relevant to this paper is the one that studies covered interest parity. This literature shows that, in the absence of country barriers or other risks, interest rate differentials are equal to the forward discount implied by the future and spot exchange rates. This fact is generally supported by the literature on industrial economies. The evidence for emerging markets is much more limited and concentrated on few countries.5 The present paper sheds new light on these strands of the international finance literature by providing a comprehensive characterization of the currency premium in two currency boards, Argentina and Hong Kong. Focusing on these two economies has two major advantages. First, these two currency boards have a rich history, which pernits analyzing how domestic and international events impact on the currency premium. Second, these two cases offer a wide range of data not available for other economies. This paper explores five major dimensions of the currency premium. First, we provide an analytical characterization of the various components of the total interest differential and, in particular, of the currency premium. We also draw a distinction between "strict" and "broad" versions of covered interest parity, that has been overlooked in much of the empirical literature. Second, we assess the extent to which hard pegs have in fact resulted in low and/or stable currency premia, an aspect of currency boards that 5 See, for example, Branson (1969), Frenkel and Levich (1977), Deardorff (1973), Dooley and Isaard (1980), Giavazzi and Pagano (1985), Artis and Taylor (1990), Frankel (1992), Chinn and Frankel (1994), Obstfeld (1995), and Kumhof (2000). 4 has so far received little attention in the debate on exchange rate regimes. We document the time pattern of the currency premium and its response to major domestic and foreign events. Third, we study the term structure of the currency premium in different markets - the money market and the foreign exchange market. We characterize its behavior during tranquil and turbulent times to gauge investors' expectations about the future of rigid currency pegs. This is possible because in these economies the most important financial contracts are denominated in both local currency and U.S. dollars. Fourth, we show how different financial instruments embody diverging assessments of the currency premium, particularly at times of financial stress. These cross-market discrepancies pose a puzzle that might reflect market segmentation, unexploited arbitrage opportunities, or the presence of other risks embedded in the commonly used measures of currency risk. Fifth, we study the determinants of the currency premium and its term structure, using detailed daily domestic and international financial data as well as political and economic events. The rest of the paper is organized as follows. Section II provides a simple analytical framework for the paper. Section III documents the empirical regularities of the currency premium and its term structure in Argentina and Hong Kong over time and across instruments and maturities. The section also presents information on the institutional features of forward markets. Section IV empirically studies the determinants of the premium and its term structure. Section V concludes. The appendix assesses the extent of cross-market differences in currency premia. 5 1I. A simple analytical framework Consider the interest differential between assets that may differ in terms of issuer, currency of denomination, and jurisdiction of issue - but are identical in other respects. Formally, let R,,k denote the annualized gross yield (i.e., one plus the interest rate) at time t on local-currency debt issued in the home country with k-period maturity; let R;k denote the gross yield on foreign-currency debt of the same maturity issued at home by the same debtor (or, more precisely, posing identical default risk as the local-currency debt); and let R,f denote the gross yield paid abroad on foreign-currency debt with the same maturity, issued by some benchmark foreign debtor (in the context of sovereign debt, typically taken to be the U.S. government). To break down the total yield differential into its two components we start from the identity R,k R,k R,w' ( Rk Rk R(,k Taking logs, letting i,.k In (Rlk), and similarly with the other yields, we can write (il'k i4,k) (t- itk) + (i,*, - i,*f )(2) currency premium1,k country premium,k II.a The currency premium Let's ignore for the moment the country premium and focus on the currency premium. The latter refers to the difference between the returns on two securities identical in all respects except for their currency denomination - i.e., they are issued in the same jurisdiction and involve identical (or are free from) default risk. Speculation across these two assets by risk-neutral investors would result in the well-known uncovered interest parity condition: 6 RIk R R E IS,k (3) where E,S,,k is the expectation at time t of the exchange rate at time t+k, and the exchange rate is defined as local currency per unit of foreign currency. Letting As' denote the (per period) anticipated percentage change in the spot exchange rate r,k kIn ' S+k , we can rewrite (3) as (i,k -it,k )i AS,k ( so that the currency premium equals the anticipated rate of change of the exchange rate. A considerable empirical literature has investigated the consistency of the data with (3) or its equivalent (4). The frequent failure of uncovered interest parity to hold in practice has been traced to two main sources (Lewis 1995): persistent expectation errors - due to irrationality, agent heterogeneity, or peso problems - and risk aversion, what is more important for our purposes. Under risk aversion, investors demand a compensation for the risk of exchange rate changes, and in such case the interest differential (4) has to be expanded to include also an exchange rate risk premium. Thus, in the general case the currency premium consists of two components: (itk -itk) = N5 e + errp, k (5) currency premium,,k anticipated devaluation,k exchange risk premium/k where errp denotes the exchange risk premium. There is a literature that attempts to break down empirically the currency premium into these two components, using survey 7 data on exchange rate forecasts (Frankel 1991) or Kalman filter techniques (Wolf 1987, Cheung 1993). II.b The country premium The country premium can also be broken down into two terms: the pure default premium and what we shall label the "onshore premium." These two premia are associated with default and transaction risks related to cross-country transactions. The pure default premium refers to the return differential between identical assets issued in the same jurisdiction by two different borrowers posing different default risk. Hence it reflects the possibility that borrowers may not honor their debts. In turn, the onshore premium refers to the return differential between assets issued in two different jurisdictions (onshore and offshore) by the same borrower, and reflects the cost and risk derived from shifting assets across jurisdictions (Aliber 1973). Hence, it relates to ingredients such as capital controls, differential taxation, commissions, and fees, as well as the risk of changes in regulations (e.g., changes in the status of capital controls) or in the market conditions that affect the transaction cost. Further, it may also reflect the differential legal treatment of default in the home and foreign jurisdictions - which can make a given borrower more likely to default in one jurisdiction (typically onshore) than in the other (offshore).6 Formally: R, Rk offshoreR* ( offshore R*f (6) R,*,fk R~t*,k Rz,k 6 Default regulations in major financial centers such as New York and London are stricter than those in many emerging markets, making the costs of default on offshore instruments much larger than those on onshore instruments. This issue has recently become prominent in the context of external payments difficulties, such as the Ecuador default and Argentina's "debt swap." 8 where ffshreR*t,,k denotes the gross yield on foreign-currency instruments issued abroad by domestic debtors with the same characteristics as those issued at home (which yield R*t,k). Taking logs and using the same notation as before, we have (itk k- i,k) = (it -ogshoreit) + (offshore i i ) (7) country premium,k onshore premiuMr,k pure default premium,k II.c Strict and broad covered interest parity If a forward exchange market exists, then risk-free arbitrage between domestic and foreign-currency securities yields what we shall label the "broad" version of the covered interest parity condition: R,#k =R f t+k (8) where F., is the k-period forward exchange rate at time t. This is a broad version of covered interest parity because the assets involved may differ in currency of denomination, issuer (domestic versus foreign) and jurisdiction of issue (onshore versus offshore). As before, equation (8) can be rewritten to show that the interest rate differential equals the forward discount: (itk -'I,k) fd,k, (9) where fd, k = I ln[ h{k 1. Thus, under broad covered interest parity there are in principle two identical measures of the currency premium, (it,k-i,k ) and fj' ,k A considerable empirical literature tests the broad version of covered interest parity, comparing (i,k - i{*) with the forward discount. It is clear from (7) that nonzero 9 onshore premia (due for example to existing or anticipated capital controls) and/or pure default premia (due to the differential default risk of local and Ibrcign borrowers) will lead to the failure of broad covered interest parity, a result comnonly fourLd in studies using emerging market data. In contrast, the "strict" version of the covered interest parity condition states that (i,# i,t =fd,,k . (10) In this version, the assets involved differ only in their currency of denomination but not in their issuer. Although empirical tests of the "strict" version of covered interest parity are hard to find in the literature, in principle one would expect it to hold up more generally than the broad version. But in practice several factors can cause even the strict parity condition to fail. First, default risk may differ across instruments issued in alternative currencies, even when issued by the same borrower in the same jurisdiction. This might reflect, for example, a threat of mandatory re-denomination of foreign-currency assets into local currency assets (akin to partial confiscation in the case of a devaluation), or also the fact that the government can print only local currency, so that it can redeem its local-currency obligations more easily than its foreign currency ones (or those of any debtor in need of bailout). In these circumstances, observed asset yields do not equal anticipated ones, and strict covered interest parity can fail to hold. A second factor that can potentially affect the strict version of covered interest parity is transaction costs. Aside from default risk, arbitrage across onshore instruments in different currencies might involve potentially large costs resulting from various market imperfections - such as the impossibility of shorting certain assets, or the presence of 10 large bid-ask spreads reflecting market illiquidity. This can also lead to a failure of the strict version of covered interest parity.7 In such case, deviations f.rom strict covered interest parity would be bounded by the magnitude of transaction costs. In the appendix we provide a more detailed analytical and empirical discussion of these issues. We conclude this section with a final point on the exchange rate risk premium. Ignoring for the moment default risk and transaction costs - so that strict covered parity holds -- equations (7) and (10) together imply that the exchange risk premium equals the difference between the forward premium and anticipated depreciation: errpt k = (Jdt, k k) ( 1) The patterns and determinants of the exchange risk premium have received considerable attention in the literature (e.g., Engel 1992, 1996; Lewis 1995). In a context of intertemporally-optimizing investors, it can be shown that the risk premium arises from the covariance between exchange rates and real consumption when investors are risk averse.8 The premium can be positive or negative, which roughly speaking can be viewed as reflecting whether the domestic currency is perceived as more or less risky than the foreign currency, respectively. Several papers have explored how the magnitude of the risk premium is affected by investors' preferences towards risk. On analytical grounds the result is ambiguoi s, and depends on the specifics of the model at hand (see Engel 7 A considerable literature has explored how various forms of transaction costs may lead to market segmentation and impact on covered interest arbitrage; see for example Blenman (1991). 8 See for example Obsfeld and Rogoff (1998) and Engel (1999). More precisely, the exchange risk premium, typically measured as E,[S,+1 - F,+,]l S, (the "nominal premium") or as E4[(S,., - F,+,)IP,+,] / S, (the "real premium", see e.g., Hakkio and Siebert 1995) generally involves two terms: one that depends on the degree of investors' risk aversion and the covariance mentioned in the text (which can be interpreted as the risk premium proper), plus another term reflecting nonlinearity of the premium in its defining variables. The latter term is independent of risk preferences and is generally presumed to be small in magnitude. 11 1999). Numerical simulations find more often than not that higher degrees of risk aversion lead to larger (in absolute terms) risk premia.9 IId The term structure of currency premia Finally, we consider briefly the term structure of currency premia (obviously, similar considerations can be made for country premia, but we will not pursue them here). For two different maturities k and k' we can write from (10) and (11) above (itk l-k) (i ,k i,,k )= fdk) fd1,k' =: s/k - A k' )+ (errp,,k -errp,k,) (12) This equation characterizes the term structure of currency premia. It reflects both the time path of anticipated depreciation and the term structure of the exchange risk premia. The literature has focused mostly on the former. Expected depreciation can be further decomposed into the perceived probability of devaluation and the magnitude of the devaluation, conditional on devaluation taking place. The time paths of these two factors shape the term structure of anticipated depreciation and thereby the term structure of currency premia. Alternative trajectories of the subjective probability and the conditional magnitude of devaluation can result in very different term structures. In particular, the term structure can become inverted if the bulk of anticipated depreciation is concentrated in the near rather than the distant future.'° This may happen, for example, when there is a 9See for example Hakkio and Siebert (1995), Siebert (1996), and Evans and Kenc (2001). The latter authors also find that the risk premium is relatively insensitive to changes in the pattern of correlations among the forcing variables in their model. 0 As an example, consider a fixed exchange rate regime where at time t the (log) exchange rate is so, and devaluation can happen at some uncertain future time r. Let s,+,k>so denote the exchange rate holding at time t+k if devaluation has already happened (otherwise the exchange rate stays unchanged at so). 12 perceived probability of collapse of a fixed exchange rate regime, and the exchange rate after the collapse is expected to overshoot - so that the magnitude of the conditional depreciation is larger in the short than in the long run. Overshooting aside, term structure inversion is also more likely if agents expect devaluation to take place in the near rather than the long term - e.g., they expect either an immediate devaluation or no devaluation at all. Along these lines, there is some literature that attempts to identify the likely term structure of expected depreciation under alternative currency regimes. For example, Favero, Giavazzi, and Spaventa (1997) argue that under floating exchange rates the term structure of anticipated depreciation tends to be flatter (even inverted) than under pegged rates, as in the former regime the bulk of depreciation may be projected to occur in the near future, while in the latter a plausible scenario may be an eventual abandonment of the peg, along with a cumulative devaluation. Let P[T >u] denote the subjective probability that devaluation will not happen prior to time u. The term structure of anticipated depreciation between t+k and t+k+j is: 1j E [St+k+j -So] -IEt[s,+k -so] = I Plira*) and/or a higher recovery ratio (9 > 9) than foreign-currency assets. If the reverse is true, then the observed interest rate differential exceeds the forward premium.34 A.III Unexploited arbitrage opportunities? We next review in more detail how different markets price currency risk. First we assess whether the evidence from Argentina seems consistent with no-arbitrage opportunities. To do this, Appendix Figure 1 displays three charts. The top panel plots the 1-month forward discount along with the currency premium derived from 1-month lending and deposit rates. The middle panel plots the forward discount along with the upper and lower bands described in equation (A2), while the bottom panel uses the bands displayed in equation (A3). Appendix Figure 1 shows that the forward discount differs from the currency premium derived from interbank rates. For most of the sample the two measures are roughly similar, but in many instances the forward discount is significantly different from the interbank market currency premium. This is especially the case during turbulent times and at the end of the sample, when the forward discount becomes considerably larger than the currency premium derived from interbank rates. The forward discount has very few values below the interbank currency premium. A similar picture is displayed in the middle and lower panels of Appendix Figure 1, corresponding to cases 2 and 3 above. In these panels the forward discount lies for the most part within the no-arbitrage bands. But in some observations, particularly at the end 34 On this point, see also Broda and Levy Yeyati (2001). 56 of the sample, the forward discount jumps above the upper band. As we shall discuss below, these are not just one-day events. The lower panel uses offshore deposit and lending rates. Relative to the middle panel, the bands shift upward because country risk is not present in the offshore rates. These rates are lower tharl, domestic interbank rates, so the differential shrinks. Despite the upward shift in the band, the forward discount still lies above the band during crisis times. But in this case there exist a few observations in which the forward discount lies below the lower band. Does this evidence imply that arbitrage opportunities exist? The answer is just maybe. There are three alternative explanations for the evidence founid; we have already mentioned two of them.35 One explanation is that in fact there are unexploited arbitrage opportunities in the short run and, thus, covered interest parity fails to hold. The currency premium derived from the exchange market is significantly different from the one derived from the money market, and is larger than any existing transaction costs. For some reason, arbitrage does not take place.36 A second possible explanation is that unobserved transaction costs - aside from the spread between lending and borrowing rates considered above -- are large enough to rule out profitable arbitrage opportunities. However, this argument does not explain why 35 In fact, there is a fourth explanation, which claims that data might not be wel) aligned by time of day. This can generate differences across markets, as shown by McCormick (1979). In our case, we use closing daily data, which in terms of trading hours are reasonably aligned. The trading hours in Buenos Aires are the same as those in New York, while trading hours in Hong Kong are similar to those in other financial centers in Asia. Given the regularities found in the data for both Argentina and Hong Kong, we believe that lack of perfect data alignment is not explaining the cross-market differences. We thank Maury Obstfeld for raising this point. 36 Informal evidence gathered from market participants suggests that at times of turbulence, when the forward premium becomes quite large, domestic banks refrain from getting involved in short-term arbitrage operations and prefer to "stay liquid." While this is consistent with the opening of a gap between the interbank and forward currency premia, the precise reasons for this decision are not known to us. 57 the different markets exhibit systematically different currency premia. The cross-market differentials might reflect the action of heterogeneous agents, endowed with different expectations, in the various markets. In this case, the cross-market differential will lie within bounds determined by the magnitude of these unspecified transaction costs, similarly to (A2) and (A3) above. In the case of Argentina, however, there are no obvious transaction costs to support this explanation. There are no restrictions on capital movements, and local residents can operate in the local and foreign markets without being taxed on interest, dividend, or capital gains.37 Yet the fact that currency premia are not equal across markets suggests that other types of transaction costs or market imperfections leading to market segmentation could be responsible for our findings. For example, there could be large bid-ask spreads unknown to us in the forward market,38 or it might not be possible to perfonn transactions at quoted prices. In view of the large volume of transactions in the NDF market, however, this explanation does not seem very convincing. A third possible explanation for our findings is that the differences in currency premia reflect in fact differences in other risks across markets. In other words, the measures of the currency premium that we (and the rest of the literature) use embed other types of risks and do not solely measure "currency risk." In such case, the cross-market currency premium differential could reflect default risk. Specifically, borrowers or banks 37 In April 2001 a small transaction tax of 0.2 percent was imposed on some financial transactions in Argentina, what cannot explain the large differentials found before and after that date. In countries with substantial explicit costs, like capital controls, there is a wedge between local and foreign rates, as shown by Herrera and Valdes (2001) for the case of Chile. 38 Data on bid-ask spreads from NDFs are not available. However, we were able to obtain data on bid-ask spreads from the peso spot market. The maximum annualized spread from this market is 240 basis points. Unless spreads on NDFs are much larger than those on spot transactions, they will be unable to 58 might be more likely to default on dollar loans and deposits than on peso contracts.39 As a result, the forward discount would exceed the peso-dollar interest differential by a default premium along the lines of equation (A4) in the text.40 In sum, at one extreme the discrepancy between the forward discount and the currency premium derived from interbank rates might reflect divergent expectations that are not or cannot be arbitraged away. At the other extreme, the discrepancy might reflect the perceived default risks in the interbank market. Transaction costs might also play a role. Of course, it is also possible that the three explanations are simultaneously behind the cross-market differentials. It is difficult to disentangle these alternative explanations without much more detailed information on market transactions. A.IV Convergence of the cross-market currency premium differential Though we cannot detennine exactly the source of the spread between the forward discount and the interbank currency premium, it is still interesting to study its behavior, as it provides information about the differential behavior of both markets. Formally, we define the spread between the forward and the interbank market as follows Y,k -fdftk -(i,k i',k). (A5) explain the observed cross-market differences. We thank Amadou Sy for generously providing us the data. Collin-Dufresne and Solnik (2001) describe a similar case for the swap and LIBOR markets. 40 An alternative version of the same argument would resort to systematic differences in default risks between domestic banks operating in dollars and those operating in pesos. In Argentina, however, virtually all banks borrow and lend in both currencies. 59 These interbank rates can be either lending or deposit rates. Here we shall work with lending rates, since they display much more variability than deposit rates (see Figure 2 in the text) and hence their behavior resembles more closely that of the NDF premium. The literature on covered interest parity has employed two main approaches to analyzing this spread. The first one simply takes the observed deviations from covered interest parity and assesses whether they frequently exceed what would be justified by transaction costs. The second approach performs unit root tests on the covered interest differential to determine whether non-stationarity can be rejected; failure to reject non- stationarity implies that the covered differential persists indefinitely and thus covered interest parity fails to hold. Still, even if non-stationarity is rejected the covered differential may converge very slowly to its mean, reflecting persistent (albeit not permanent) failures of covered interest parity. In our case, we take the case most favorable to covered interest parity - namely that in which the arbitrageurs have no funds (Cases 2 and 3 above). We also work with the onshore rates used in the strict form of covered interest parity described in Section II. We then examine the behavior of the forward discount relative to the no-arbitrage band defined by borrowing and lending rates, using the band displayed in the middle panel of Appendix Figure 1. For those observations where the forward discount lies above the band, we examine the dynamics of the differential between the forward discount and the upper band. For the observations inside the band, we study the dynamics of the forward discount.4' 41 Since we have no observations below the lower band, we ignore this case. 60 This approach is in the spirit of the threshold autoregression (TAR) models used to study arbitrage in goods and assets markets.42 These models typically need to estimate the "commodity points" or thresholds of no arbitrage. In our case, however, the problem is simpler because the thresholds are known, and given by the no-arbitrage bands in Figure Al. Therefore, we estimate the following model: A(fd1, - (it k - ik c)) "t + 2 d,k k -(it - k ;-i,k ))+ ek if fdI,k > (i -i,k )(A Afd1 k C= + Xfdt,k + E7'k if fd,,k < (it,k - it ) Note that the mean and the speed of adjustment, as well as the variance of the disturbance, are allowed to differ across equations. Our primary concern is to assess the speed of adjustment 2 both within and outside the band. The top panel of Appendix Table 1 shows the number of observations for which the forward discount lies outside and inside the no-arbitrage band. Around 49 percent of the observations are above the no-arbitrage bands (345 observations). The histogram displayed in the table shows the distribution of observations relative to the upper band. Negative numbers represent observations below the upper band, while positive numbers are observations above the band. The histogram shows that the observations above the band can take very large values - their median exceeds 1,000 basis points. The bottom of Appendix Table 1 reports the results from estimating the TAR model. Since the observations above the upper band clearly become more abundant in the latter part of the sample, we perform the estimation on two different samples: the full 42 These models have been used to examine issues such as the validity of purchasing power parity, or the extent of arbitrage under the gold standard. See for example, Obstfeld and Taylor (1997), Prakash and Taylor (1997), Taylor and Peel (2000), and Taylor, Peel, and Sarno (2001). 61 sample of available data, and a subsample ending at the time of the cut in international financing (early July 2001). For the observations outside the band we can reject nonstationarity of the dependent variable. The estimates reflect reversion to the mean. Interestingly, for the observations inside the band we cannot reject nonstationarity. The speed of adjustment to the mean appears to have declined dramatically after the cut in international lending of July 2001. This is particularly evident from the half-life of the differential [calculated as ln(0.5)/ln(l+A)] outside the band, which is one and a half day in the restricted sample and nearly five days in the full sample. Hence, departures from the arbitrage band appear to have become much more lasting - indeed, much more lasting than should be expected under perfect arbitrage. To conclude, we document the main features of the cross-market currency premium differential, as a reflection of unspecified heterogeneity across markets rather than specifically as a potential failure os covered arbitrage. Appendix Figure 2 plots the 1-month forward-interbank spread 6,,i, while Appendix Table 2 displays summary statistics of the spread for the 1- and 12-month maturities. The spread is on average positive, particularly for the 1-month maturity. But the mean is affected by large positive values reaching as high as 11,000 basis points during the 2001 crisis. The distribution is skewed to the right, so that the median is smaller than the mean, but still positive. It is noteworthy that all the large values take place during crisis times. In tranquil times, the cross-market difference is also positive but smaller. The differential takes on only a few negative numbers, reaching at most -200 basis points. 62 Figure 1 Currency Premia under Currency Boards The figure shows historical values of daily currency premia in Argentina (top panel) and Hong Kong (lower panel). The currency premium for Argentina is calculated as the spread of local peso time deposit rates over local U.S. dollar interbank deposit rates, with maturities up to 2 months. The currency premium for Hong Kong is calculated with the forward discount, the forward exchange rate minus the spot exchange rate, using I-month contracts. The sample for Argentina covers the period 4/1/93 - 9/25/01, for Hong Kong it cosers the period 1/4/93 - 9/25/01. All rates are in basis points, annualized, and continuously compounded. Currency Risk Premium from Argentina's Time Deposit Rates 2,000 1,600 0 w) ~~~~~~~~~. 0; J ^ ~ ~ ~ ~ ~ C oC4t t t * 2 r 0 0~~~~~~~~~~~~~~~~ 8001.0 O > y 400 f 5I -so ON ON 0' ' ON 0g 0i g 0' rg x g 0' 0' 0 0 8 Currency Risk Premium from tong Kong's Forward Exchange Rates 3,0001 2.5001j0 . 2,000-0 0 C 0 1,5000 ~. 5,00 0 500~~~~~~~~~~~~~~~~~~~~~~5 -500 Cl N ' 'CCl 0 C C Cl 00 C 0 0' 0 0 - Figure 2 Alternative of Measures of Currency Premia The figure shows different measures of currency premia for Argentina and Hong Kong, using daily 1-month interest rate premia and 1-month forward exchange rate discounts. The interest rates used for Argentina include the interbank offer rate (BAIBOR), the interbank deposit rate index or money market index (MMR) consisting of rates paid on deposits of more than I million pesos or U.S. dollars, and the time deposit rate for deposits up to I million pesos or U.S. dollars. Both the BAIBOR and the MMR are from Bloomberg and the time deposit rates were obtained from the Central Bank of Argentina. For each type of rate, the currency premium for Argentina is measured by the difference between the rate denominated in domestic currency and that in U.S. dollars. The forward exchange rates for Argentina are non-deliverable forward (NDF) rates and the NDF currency premium is measured by the forward discount. NDF forward rates are from two sources: Deutsche Bank and Bloomberg. The interest rates used for Hong Kong are interbank offer rates (HIBOR), obtained from the Hong Kong Monetary Authority. The premium is obtained over the U.S.-dollar LIBOR. The forward rates used for Hong Kong are from Bloomberg and correspond to deliverable contracts. All the rates are annualized, in basis points, and continuously compounded. Argentina 6,000 Interbank Offer Rates Premium 6,000 Interbank Deposit Rates Premium 5,000 5,000 4,000 4,000 3,000 3,000 2,000 2,000 1,000- 1,000 0 0 1 0 X 55 N 0 ol a, et > ol es e ol a 0 c v 4 , _ I_ x o O Ej 00 XN 00 AN x00Gx ¢ G x t S x R N X t N O Time Deposit Rates Premium Non-Deliverable Forward Discount 6,000 16,000 5,000 14,000- 12,000- 4,000 10,000 - 3,000 8,000 - 2,000- 6,000 - 4,000- 1,000 , , 2,000 0 - @ t A% x% '0 N N 00 0% % x' (N x 0% A% x0 N 00 0% N x 0% 0% ~~~~~~~~~~~ 8 0%~~~~~~ 0 % 0% 0% Hong Kong Interbank Offer Rates Premium Deliverable Forward Discount 3,000 3,000 2,500 2,500 - 2,000 2,000 - 1,500 1,500 - 1.000 1,000 500 500 - 0 ~~~~~~~~~~~~~~~~~~0 I 14A -500 -0 - - - v - (4 00 t N 00 > (N 00 -50 -N 0 Figure 3 Term Structure of Currency Premia The figure shows the term structure of the currency premium in Argentina (top panel) and Hong Kong (bottom panel), calculated with the difference between the 12-month and the l-month currency premium. For each country, the currency premium is measured using both interbank offer rates and forward exchange rates. For Argentina the peso interbank offer premium is obtained with the spread of peso over dollar denominated Argentine interbank offer rates, while for Hong Kong the interbank offer premium is measured with the spread of the Hong Kong interbank offer rates over the U.S.-dollar LIBOR. The forward discounts are measured by the spreads of non-deliverable forward (NDF) exchange rates, for Argentina, and deliverable forward rates, for Hong Kong, over the spot exchange rate, respectively. The interest rates for Hong-Kong were obtained from the Hong Kong Monetary Authority and the exchange rates from Bloomberg. The Argentine interbank rates are from Bloomberg and the NDF rates were obtained from Deutchbank and Bloomberg. All the rates are annualized, in basis points, and continuously compounded. Argentina's Interbank Offer Term Structure Argentina's Non-Deliverable Forward Discount Term Structure 2,000 2,000 0 - 0 -2,000 - -2,000 4,000 4,000 -6,000 l -6,000 -8,000 -8,000 - 0,000 -10,000- -12,000 - .12,000 - Hong Kong's Interbank Offer Term Structure Hong Kong's Forward Discount Term Structure 1,000 I 1,000 500 00500 0 00 0 00 0 00 0 00 000 0 00 000 0Li- - - -50°0 0 500 0 ,o -2 -1 ,000 0 -1,500 -1,500 -2,000 -2,000 Ne 00 st 00 x0 ,O N- 00 C00 000 00 _ t 00 '3 '3 N 00 00 00 o - a 00 00 00 00 00 00 00 a 00 0 00 00 00 00 00 00 00 00 00 00 00 00s 00 00 00 00 00 00 _0 00 00 _ 00 00 00 0v0 0 0 0 0 0 0 0 Table 1 History of Currency Premia in Argentina and Hong Kong Summary Statistics The table shows sumnmary statistics of daily currency premnia in Argentina and Hong Kong for different samples. The currency premium for Argentina is calculated as the spread of local peso time deposit rates over local U.S. dollar interbank deposit rates, with maturities up to 2 months. The currency premium for Hong Kong is calculated with the 1-month forward discount. The crisis periods for Argentina are the following: (i) Mexican crisis, (ii) Attack on the Hong Kong dollar, (iii) Russia's default, (iv) Devaluation of the Brazilian real, (v) Financial Times article and presidential elections, (vi) Vice president resigns, (vii) Changes of finance minister, and (viii) Cut of international credit. The crisis periods for Hong-Kong are: (i) The Mexican crisis, (ii) the Financial distress in KoTea, (iii) the Attacks on the Thai baht and the four attacks on the Hong Kong dollar. See text for a description of all the events. All rates are in basis points, annualized, and continuously compounded. Number of . Standard Dates Nube of Mean Median Stnad Min Max Observations Deviation Argentina Time Deposit Rates Up to 2 month Total 04/01/93 - 06/05/01 2,121 211 148 204 1 1,986 Tranquil Periods (after the Mexican Crisis) 1,228 126 98 97 1 812 Crisis Periods 446 383 252 333 25 1,986 Mexican crisis 01/10/95 - 04/10/95 63 603 520 243 264 1,316 Attack on the Hong Kong dollar 10/29/97 - 11/26/97 21 223 223 90 83 393 Russia's default 08/19/98 - 10/16/98 42 201 193 69 65 405 Devaluation of the Brazilian real 01/13/99 - 02/12/99 23 210 206 56 97 315 Financial Times article and elections 05/17/99 - 12/17/99 149 186 155 74 88 397 Vice president resigns 10/06/00 - 12/29/00 56 216 220 90 25 397 Changes of finance minister 03/16/01 - 05/18/01 38 548 480 252 202 1,308 Cut ofinternational credit 07/10/01 - 09/25/01 54 1,005 940 359 441 1,986 Hong Kong Forward Rates Total 01/04/93 - 06/05/01 2,240 58 12 165 -111 2,840 Tranquil Periods 1,908 15 6 53 -111 325 Crisis Periods 332 301 217 313 -47 2,840 Mexican crisis 01/10/95 - 04/10/95 62 73 38 106 -47 341 Early signs of financial distress 01/27/97 - 02/21/97 19 50 9 112 -19 376 Attack on the Thai Bath 05/14/97 - 7/24/97 51 182 101 163 23 621 Ist Attack on the Hong-Kong dollar 08/15/97 - 12/15/97 87 386 225 398 70 2,840 2nd Attack on the Hong-Kong dollar 01/05/98 - 02/04/98 22 598 491 293 147 1,165 3rd Attack on the Hong-Kong dollar 05/27/98 - 07/06/98 29 418 348 205 212 1,119 4th Attack on the Hong-Kong dollar 07/10/98 - 10/06/98 62 426 306 272 178 1,309 Table 2 Maximum Implied Devaluation in Argentina and Hong Kong Summary Statistics The table shows the maximum devaluation implied by the currency premium in Argentina and Hong Kong, for different probabilities of devaluation and different sample periods. Under risk neutrality, the currency premium is equal to the probability of devaluation times the value of the devaluation. Therefore, the maximum implied devaluation is obtained by dividing the maximum currency premium observed in each sample period by the assumed probability of devaluation. The currency premium for Argentina is calculated as the spread of local peso time deposit rates over local U.S. dollar interbank deposit rates, with maturities up to 2 months. The currency premium for Hong Kong is calculated with the 1-month forward discount. The crisis periods for Argentina are the following: (i) Mexican crisis, (ii) Attack on the Hong Kong dollar, (iii) Russia's default, (iv) Devaluation of the Brazilian real, (v) Financial Times article and presidential elections, (vi) Vice president resigns, (vii) Changes of finance minister, and (viii) Cut of international credit. The crisis periods for Hong-Kong are: (i) The Mexican crisis, (ii) the Financial distress in Korea, (iii) the Attacks on the Thai baht and the four attacks on the Hong Kong dollar. See text for a description of all the events. All rates are in basis points, annualized, and continuously compounded. Dates 10% 25% 50% 75% 90% Argentina Time Deposit Rates Up to 2 month Total 04/01/93 - 06/05/01 19,861 7,944 3,972 2,648 2,207 Tranquil Periods (after the Mexican Crisis) 8,122 3,249 1,624 1,083 902 Crisis Periods 19,861 7,944 3,972 2,648 2,207 Mexican crisis 01/10/95 - 04/10/95 13,165 5,266 2,633 1,755 1,463 Attack on the Hong Kong dollar 10/29/97 - 11/26/97 3,934 1,574 787 525 437 Russia's default 08/19/98 - 10/16/98 4,052 1,621 810 540 450 DevaluationoftheBrazilianreal 01/13/99-02/12/99 3,146 1,258 629 419 350 Financial Times article and elections 05/17/99 - 12/17/99 3,968 1,587 794 529 441 Vice president resigns 10/06/00 - 12/29/00 3,974 1,590 795 530 442 Changes of finance minister 03/16/01 - 05/18101 13,078 5,231 2,616 1,744 1,453 Cut of intemational credit 07/10/01 - 09/25/01 19,861 7,944 3,972 2,648 2,207 Hong Kong Forward Rates Total 01/04/93 - 06/05/01 28,398 11,359 5,680 3,786 3,155 Tranquil Periods 3.252 1,301 650 434 361 Crisis Periods 28,398 11,359 5,680 3,786 3,155 Mexican crisis 01/10/95 - 04/10/95 3,409 1,364 682 455 379 Early signs of financial distress 01/27/97 -02/21/97 3,760 1,504 752 501 418 Attack on the Thai Bath 05/14/97 - 7/24/97 6,213 2,485 1,243 828 690 1st Attack on the Hong-Kong dollar 08/15/97 - 12/15/97 28,398 11,359 5,680 3,786 3,155 2nd Attack on the Hong-Kongdollar 01/05/98 -02/04/98 11,651 4,660 2,330 1,553 1,295 3rd Attack on the Hong-Kong dollar 05/27/98 - 07/06/98 11,187 4,475 2,237 1,492 1,243 4th Attack on the Hong-Kong dollar 07/10/98 - 10/06/98 13,090 5,236 2,618 1,745 1,454 Table 3 Alternative Measures of Currency Premia Summary Statistics The table shows summary statistics for different measures of currency premia for Argentina and Hong Kong, using daily 1-month interest rates and 1- month forward exchange rate discounts from 9/9198 to 9/25/01 for Argentina and 1/4/93 to 9/25/01 for Hong Kong. The differences in the number of observations are due to missing values. The interest rates used for Argentina include the interbank offer rate (BAIBOR), the interbank deposit rate index or money market index (MMR) consisting of rates paid on deposits of more than I million pesos or U.S. dollars, and the time deposit rate for deposits up to I million pesos or U.S. dollars. Both the BAIBOR and the MMR are from Bloomberg and the time deposit rates were obtained from the Central Bank of Argentina. FoT each type of rate, the currency premium for Argentina is measured by the difference between the rate denominated in domestic currency and that in U.S. dollars. The forward exchange rates for Argentina are non-deliverable forward (NDF) rates and the NDF currency premium is measured by the forward discount NDF forward rates are from two sources: Deutsche Bank and Bloomberg. The interest rates used for Hong Kong are interbank offer rates (HIBOR), obtained from the Hong Kong Monetary Authority. The premium is obtained over the U.S.-dollar LIBOR. The forward rates used for Hong Kong are obtained from Bloomberg and correspond to deliverable contracts. All the rates are annualized, in percentages, and continuously compounded. Number of Mean Median Standard Min Max Observations Deviation Argentina Interbank Offer Rates 720 318 144 487 29 5,217 Time Deposit Rates 726 214 126 271 -4 2,067 Interbank Deposit Rates 704 206 93 312 0 2,305 Non-Deliverable Forward Rates 737 857 228 1,772 42 14,726 Hong Kong Interbank Offer Rates 2,090 39 1 150 -115 3,336 Deliverable Forward Rates 2,240 58 12 165 -111 2,840 Correlations Interbank Offer Time Deposit Interbank Non-Deliverable Rates Rates Deposit Rates Forward Rates Interbank Offer Rates I Time Deposit Rates 0.83 1 Interbank Deposit Rates 0.89 0.78 1 Non-Deliverable Forward Rates 0.91 0.82 0.85 1 Interbank Offer Deliverable Rates Forward Rates Interbank Offer Rates I Deliverable Forward Rates 0.94 1 Table 4 Term Structure of Currency Premia Summary Statistics The table shows the summary statistics for the term premia, spread between the currency premium measured with 12-month rates over the one from 1-month rates, during crisis and tranquil periods, for Argentina and Hong Kong. The crisis periods for Argentina are the following: (i) Mexican crisis, (ii) Attack on the Hong Kong dollar, (iii) Russia's default, (iv) Devaluation of the Brazilian real, (v) Financial Times article and presidential elections, (vi) Vice president resigns, (vii) Changes of finance minister, and (viii) Cut of international credit. 7he crisis periods for Hong-Kong are: (i) The Mexican crisis, (ii) the Financial distress in Korea, (iii) the Attacks on the Thai baht and the four attacks on the Hiong Kong dollar. See text for a description of all the events. All rates are in basis points, annualized and continuously compounded. Dates Number of Mean Median Dt.nMin Mi max Observations Deviation Argentina Interbank Offer Rates Total 04/21/97-06/05/01 1,098 168 151 204 -2,283 864 Tranquil Periods 720 139 134 95 -238 442 Crisis Periods 378 222 198 315 -2,283 864 Attack on the Hong Kong dollar 10/29/97 - 11/26/97 21 -24 -30 35 -81 54 Russia's default 08/19/98 - 10/16/98 42 81 49 95 -59 258 Devaluation of the Brazilianreal 01/13/99- 02/12/99 23 130 131 31 54 193 Financial Times article and elections 05/17/99 - 12/17/99 149 444 486 126 200 627 Vice president resigns 10/06/00 - 12/29/00 56 124 130 47 -41 188 Changes of finance minister 03/16/01 -05/18/01 38 -73 -114 249 -491 311 Cut of international credit 07/10/01 - 09/25/01 49 158 222 621 -2,283 864 NDF Rates Total 10/20/97-06/05/01 732 -12 270 1,198 -11,720 984 Tranquil Periods (after the Mexican Crisis) 393 231 282 423 -3,667 643 Crisis Periods 339 -293 160 1,657 -11,720 984 Russia's default 08/19/98 - 10/16/98 27 66 -1 278 -481 579 DevaluationoftheBrazilianreal 01/13/99-02/12/99 22 157 135 139 -120 337 Financial Times article and elections 05/17/99 - 12/17/99 143 502 490 276 44 984 Vice president resigns 10/06/00 - 12/29/00 56 33 46 159 -485 297 Changes of finance minister 03/16/01 - 05/18/01 38 -1,295 -846 1,648 -7,200 251 2 Cut ofinternational credit 07/10/01 - 09/25/01 53 -2,430 -2,348 2,872 -11,720 941 Hong Kong Interbank Offer Rates Total 01/04/93-06/05/1)1 2,065 49 30 81 -836 358 Tranquil Periods 1,754 50 32 69 -51 358 Crisis Periods 311 44 17 130 -836 346 Mexican crisis 01/10/95 - 04/10,95 59 24 58 65 -175 99 Early signs of financial distress 01/27/97 - 02/21/97 18 18 19 9 -7 38 Attack on the Thai Bath 05/14/97 - 7/24/97 46 -8 -11 20 -44 25 1st Attack on the Hong-Kong dollar 08/15/97 - 12/15/97 81 6 -34 152 -836 345 2nd Attack on the Hong-Kong dollar 01/05/98 - 02/04/98 20 137 169 139 -24 346 3rd Attack on the Hong-Kong dollar 05/27/98 - 07/06/98 28 81 104 129 -437 269 4th Attack on the Hong-Kong dollar 07/10/98 - 10/06/98 59 118 156 158 -483 316 Forward Rates Total 01/04/93-06/05,'01 2,152 37 25 118 -1,854 481 Tranquil Periods 1,820 45 26 81 -269 481 Crisis Periods 332 -9 -6 228 -1,854 430 Mexican crisis 01/10/95 - 04/10/95 62 -16 11 79 -217 108 Early signs of financial distress 01/27/97 - 02/21/97 19 -37 7 113 -366 28 Attack on the Thai Bath 05/14/97 - 7/24/97 51 -143 -55 160 -570 3 IstAttackontheHong-Kongdollar 08/15/97- 12/15/97 87 -76 -37 312 -1,854 354 2nd Attack on the Hong-Kong dollar 01/05/98 - 02/04/98 22 139 263 219 -205 430 3rd Attack on the Hong-Kong dollar 05/27/98 - 07/06/98 29 93 125 163 -405 263 4th Attack on the Hong-Kong dollar 07/10/98 - 10/06/98 62 114 175 186 -532 334 Table 5 Determinants of Currency Premia in Argentina The columns of the table show the results for regressions of the 1-month currency premium on a set o explanatory variables. Columns I to 5 correspond to ordinary least squares (OLS) regressions and columns 6 and 7 extend the model to an exponential garch (EGARCH). The models estimated are described in equations (15) and (16) in the paper. The first column of the table shows the results for the basic specification, including as regressors (i) the spread of a high yield bond index over a comparable U.S. government bond, (ii) the EMBI spread of Latin American countries excluding Argentina, and (iii) the ratio of total reserves of the central bank to total deposits and (iv) the ratio of total cash held by the financial system over total deposits. The second specification, column (2), adds the Golman Sachs measure of currency mislignment. Specification (3) disaggregates the reserves of the central bank into the ratio of reserves held in the form of govemment bonds to total deposits asd the ratio of hard currency reserves to total deposits. Specification (4) tests the effect of Argentina's average foreign currency credit rating. The last OLS specification, in column (5), tests the effects of external and internal shocks on the conditional mean with two durnmy variables. The extemal shocks dummy captuTes the effect of the Russian defalut and the Brasilian devaluation, while the internal shocks dummy captures the effects of the internal crisis periods during 1999 and 2001: Financial Times article and presidential elections, Vice president resigns, Changes of Finance Minister, and Cut of international credit. In all OLS specifications 4 lags in differences of the dependent variable and all regressors were included. In the EGARCH specifications the lags for the reserves and cash ratios could be treammed down to two, allowing for a larger sample which improves convergence. The EGARCH specifications use the basic set of regressors and a dummy variable for all crisis (internal and external): column (6) tests for the effects of the crisis dummy in the conditional mean and the last specification, column (7), tests for its effects also in the conditional variance. Standard errors are in parentheses. * significant at 10%, ** at 5%, and +** at 1%. I-month NDF Discount Dependent variable in First Differences: (fl-s) (1) (2) (3) (4) (5) (6) (7) Lagged Dependent Variable -0.326 -0.331 "'* -0.319 -0.369 -0.369 -0.129 -0.140 (0.066) (0.067) (0.064) (0.073) (0.074) (0.010) (0.005) High Yield Spread -0.161 1.035 -0.424 1.343 1.071 0.219 ** -0.635 (0.952) (1.095) (O.g55) (0.902) (1.025) (0.101) (0.045) GS Misalignment Measure 0.079 (0.131) LAC EMBI Spread 0.640 0.369 ** 0.666 0.438 ** 0.287 0.013 0.033 (0.236) (0.115) (0.230) (0.211) (0.189) (0.016) (0.010) Total Reserves to Deposits Ratio (t-3) -0.327 *+* -0.441 *' -0.185 -0.248 * -0.109 * -0.009 ** (0.117) (0.112) (0.133) (0.128) (0.007) (0.003) Bomd Reserves to Deposits Ratio (t-3) -0.040 (0.032) Hard Currency Reserves to Deposits -0.292 Ratio (1-3) (0.114) Banks'Cash overDeposits Ratio (t-3) 0.169 *** 0.162 *4* 0.161 * 0.115 ** 0.283 * 0.048 t 0.007 (0.049) (0.055) (0.046) (0.050) (0.082) (0.003) (0.001) Average Foreign Currency Credit Rating -8.598 ** (3.892) External Shocks (Russia and Brasil) 1.957 * (1.004) Internal Shocks 3.358 * (1.465) Crisis Dummy * 0.494 * 0.660 4*4 (0.061) (0.046) constant 9.584 -1.398 -6.313 47.254 * 55.715 * 3.125 t 1.772 # (13.462) (16.047) (15.902) (22.297) (29.490) (0.976) (0.131) time 0.007 0.007 0.000 0.007 0.000 W 0.001 * (0.005) (0.005) (0.004) (0.005) (0.000) (0.000) earch -0.468 + 0.045 (0.064) (0.091) arch-abs. 2.590 * 2.746 *44 (0.108) (0.097) egarch 0.918 * 0.828 *4* (0.016) (0.018) Crisis Dummy 1.611 "'* (0.129) constant -0.029 -0.692 * (0.079) (0.098) Numberof observations 378 358 378 378 378 454 454 R2 0.40 0.42 0.41 0.41 0.42 Maximum numberoflags 4 4 4 4 4 4 4 Q(S) p-value 0.88 0.92 0.85 0.86 0.91 0.08 0.56 Q(10) p-value 0.72 0.78 0.76 0.67 0.70 0.23 0.59 Log-likelihood -1203.63 -1145.88 -1201.02 -1201.76 -1197.90 -824.71 -740.44 Wald test ofjoint significance 0.00 * 0.00 ' 0.00 *"* 0.00 *$ 0.00 0.00 0.00 *" (p-value) Table 6 Determinants of Term Premia in Argentina The columns of the table show the results for regressions of the term premium on a set o explanatory variables. Columns I to S correspond to ordinary least squares (OLS) regressions and columns 6 and 7 extend the model to an exponential garch (EGARCH). The models estimated are described in equations (15) and (16) in the paper. The first column of the table shows the results for the basic specification, including as regressors (i) the US dollar LIBOR term premium, (ii) the spread of a high yield bond index over a comparable U.S. government bond, (iii) the EMBI spread of Latin American countries excluding Argentina, and (iv) the ratio of total reserves of the central bank to total deposits and (v) the ratio of total cash held by the financial system over total deposits. The second specification, column (2), adds the Golman Sachs measure of currency mislignment. Specification (3) disaggregates the reserves of the central bank into the ratio of reserves held in the form of government bonds to total deposits and the ratio of hard currency reserves to total deposits. Specification (4) tests the effect of Argentina's average foreign currency credit rating. The last OLS specification, in column (5), tests the effects of extemal and internal shocks on the conditional mean with two dummy variables. The external shocks dummy captures the effect of the Russian default and the Brasilian devaluation, while the intemal shocks dummy captures the effects of the intemal crisis periods during 1999 and 2001: Financial Times article and presidential elections, Vice president resigns, Changes of Finance Minister, and Cut of intemational credit, In all OLS specifications 4 lags in differences of the dependent variable and all regressors were included. In the EGARCH specifications the lags for the reserves and cash ratios could be treammed down to two, allowing for a larger sample which improves convergence. The EGARCH specifications use the basic set of regressors and a dummy variable for all crisis (intemal and extemal): column (6) tests for the effects of the crisis dummy in the conditional mean and the last specification, column (7), tests for its effects also in the conditional variance. Standard errors are in parentheses. * significant at 10%, ** at 5%1, and t** at 1%. Dependent Variable In First NDF Term Sructure Differences: (fl 2-s) (1) (2) (3) (4) (5) (6) (7) Lagged Dependent Variable -0.483 -0.508 *'* -0.482 * -0.510 ** -0.523 *' -0.204 *** -0.216 (0.078) (0.081) (0.076) (0.086) (0.084) (0.009) (0.009) US Term Structure 0.717 0.947 0.686 0.731 0.827 0.227 ** 0.226 *5* (0.664) (0.856) (0.694) (0.663) (0.723) (0.046) (0.038) High Yield Spread -0.622 -1.056 0.1)06 -1.509 -1.967 -0.082 -0.031 (1.130) (0.849) (1.045) (1.089) (1.100) (0.098) (0.084) GS Misalignment Measure 0.032 (0.133) LAC EMBI Spread -0.104 0.013 -0.209 0.062 0.313 0.139 ** 0,139 (0.166) (0.117) (0.166) (0.161) (0.217) (0.015) (0.018) Total Reserves to Deposits Ratio (t-3) 0.237 ** 0,317 * 0.143 0.145 0.017 ** 0.019 * (0.098) (0.095) (0.113) (0.120) (0.007) (0.006) Bomd Reserves to Deposits Ratio (t-3) 0 058 * (0.031) Hard Currency Reserves to Deposits 0.232 * Ratio (t-3) (0.096) Banks' Cash over Deposits Ratio (t-3) -0.165 *5* -0.173 * -0.165 * -0.131 ' -0.256 *5* -0.046 * -0.045 * (0.047) (0,050) (0.043) (0.047) (0.066) (0.002) (0.003) Average Foreign Currency Credit Rating 4.916 (3.618) Extemal Shocks (Russia and Brasil) -2.095 t (1.008) Intemal Shocks -2.931 I (1.292) Crisis Dummy -0.552 *"" -0.559 * (0.049) (0.062) constant -22.729 * -18.560 1.253 -45.397 ** -63.063 ** -13.500 - -0.523 ** (13.448) (15.735) (15.732) (22.529) (26.537) (0.288) (0.079) tiTne -0.003 -0.003 0.002 -0.002 0.000 0.000 (0.004) (0.004) (0.004) (0.004) (0.000) (0.000) earch 0.529 *** 0.478 * (0.069) (0.079) arch-abs. 2.051 *** 2.032 + (0.104) (0.112) egarch 0.993 *i* 0.929 * (0.014) (0.017) Crisis Dummy 0.786 * (0.113) constant -0.152 + -13.268 * (0.062) (1.091) Numberofobservations 351 331 351 351 351 415 415 R2 0.46 0.47 0.47 0.46 0.47 Maximum number of lags 4 4 4 4 4 4 4 Q(5) p-value 0.48 0.73 0.49 0.44 0.69 0.32 0.88 Q(10) p- value 0.52 0.73 0.56 0.46 0.64 0.67 0.99 Log-likelihood -1070.58 -1013.29 -1067.26 -1069.79 -1065.11 -694.61 -672.76 Wald test ofjoint significance 0.00 0 0.00 .00.00 0.00 0.00 ** 0.00 O*' 0.00'*** (p-value) Appendix Figure 1 Non-Deliverable Forward Discount and No-Arbitrage Bands The figure shows the 1-month Non-Deliverable Forward (NDF) discount and two threshholds defining no-arbitrage bands. The top panel displays the forward discount and two currency premia, the spread between peso and dollar denominated Argentine interbank rates and the spread between peso and dollar Argentine time deposit rates. In the middle panel, the upper threshhold consists of the difference between the interbank offer rate in pesos and the time deposit rate in dollars; the lower threshhold is the difference between the time deposit rate in pesos and the interbank offer rate in dollars. In the lower panel, the upper threshold is the difference between the BAIBOR in pesos and the U.S. deposit rate in dollars; the lower threshold is the spread of the Argentine time deposit rate in dollars over the U.S. Federal Funds Rate. All rates are annualized, in basis points, and continuously compounded. See text for a discussion on arbitrage in case 1, 2 and 3. Case 1: (i j& dsos.) fd, = ( dipdl id.okiK.) 14,000 12,000 - 10,000 - 8,000 6,000 4,000 - 2,000 - 0 -2,000 x0 00 ON oN ONON ON oN *> o o Case 2: - i >")ˇ 2 f (iw -°.iioi.) upperband lowerband 14,000 12,000 10,000 4,000 2,000 400 -2,000 -000 0 N O N O N O ON - 2. ON N ON - -> o o o o oN o - - - - ON Case 3: ( g j * r j >fd_ ˇ upperband lowerband 14,000 - 12,000 - 10,000 8.000 6,0001 4.000 2,000 0 -2.000 -NDF discount Appendix Figure 2 Cross-Market Currency Premium Differential The figure shows the spread between the I -month forward discounts and the interbank currency premium for Argentina and Hong Kong. The interbank rate for Argentina is the interbank offer rate (BAIBOR) obtained from Bloomberg. Since in Hong Kong these rates are denominated on]y in Hong Kong dollars, the premium is measured by the spread of these over the U.S.-dollar LIBOR. The forward exchange rates for Argentina are non-deliverable forward (NDF) rates obtained from Deutsche Bank and Bloomberg. The forward rates for Hong Kong correspond to deliverable contracts and come from Bloomberg. The interbank rates used for Hong Kong are interbank offer rate (HIBOR), obtained from the Hong Kong Monetary Authority. All the rates are annualized, in basis points, and continuously compounded. 12,0007 Spread of NDF over BAIBOR Currency Premia, Argentina 10,000 8,000p looo 1 6,000 4,000 ~2,0001 -2,000 N ON ON ON ON ON S ZN 0 N Z Z O ON ON ON O7 ON ON ZN ZN 0N 0s ZN ZN ZN soo Spread of FWD over HAIBOR Currency Premia, Hong Kong 600 400 -600 9 Z Z, ZN ON ON NO NO N N ON N O ON ZN o o _ ON ON ON _ _ _ > o o > 9 <> g g 0 8 o ON ON ON ON ON ON ON ON ON ON ON ON Z £ £t r b _ b _ b ._ F _ b _ t_ _ I_ _ {_~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~N O O N ON O N N O Z N Z Appendix Table 1 Currency Premia and the No-Arbitrage Band The top panel shows summary statistics fo the 1-month Non-Deliverable Forward (NDF) discount and two threshholds defining the no-arbitrage band. The upper threshhold consists of the difference between the interbank offer rate in pesos and the time deposit rate in dollars. The lower threshhold is the difference between the time deposit rate in pesos and the interbank offer rate in dollars. The mid panel shows the histogram of tbe difference between the NDF discount and the upper band. Since there are no observations bellow the lower band, negative observations correspond to observations inside the no-arbitrage band, while positive observations correspond to deviations from the no-arbitrage condition. The lower panel shows economnetric estimations of revertion to the no-arbitrage band for observations of the NDF discount outside the band and to the conditional mean for observations of the NDF discount inside the band. All rates are annualized, in basis points, and continuously compounded. Summary Statistics Number of Obs. in Observations Percent of Mean Median Std. Dev. Min Max Observations Total NDF discout 737 - 857 228 1,772 42 14,726 Upper band 1,098 458 287 568 100 6,702 Lower band 1,098 - -39 -49 148 -1,042 1,253 Obs. above the no-arbitrage band 220 30% 2,380 1,049 2,673 120 14,726 Obs. inside the no-arbitrage band 517 70% 209 150 189 42 2,141 140 Distribution of Deviations from the Upper Band 120 j 60 l Observations above the upper band Observations inside the band 401 20 o_~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1 111 1 . || 0 ,, vD No 0' 0 -¢ ^ < '0 *. ,', "'w s r , Revertion of Forward Discount [A(fd,-(i,-;*)) =c<+*"tAd,,.-(i,,*-i,,)+g,J if fd,. > (i,. - iJ) Afd,k c'+AZfd,,+e,;f if £,t -N(O, &-') and e,X inN rO,ain) Including the "Cut of intemational credit" crisis period Excluding the "Cut of intemational credit" crisis period Dickey- Dickey-Fuller Ai t-ratio Fuller 1% Half-life Ai t-ratio 1% critical Half-life critical value value outside no-arbitrage band' -0.16 -4.00 -3.49 3.86 -0.45 -7.24 -3.50 1.17 (0.04) (0.06) -0.07 -6.28 -3.44 9.68 -0.07 -6.28 -3.44 9.68 inside no-arbitrage band- (0.01) (0.01) I The half-life corresponds to the convergence to the non-arbitrage band 2 The half-life corresponds to the convergence to the conditional mean Appendix Table 2 Cross-Market Currency E'remium Differential Summary Statistics The table shows summary statistics of the spread between the I - andl 2-month forward discounts and interbank currency premia for Argentina and Hong Kong, using different samples. The currency premium for Argentina is calculated as the spread of local peso over local U.S. dollar interbank offer rates. The currency premium for Hong Kong is the spread of Hong Kong dollar interbank offer rates over the U.S.-dollar LIBOR. The crisis periods for Argentina are the following: (i) Mexican crisis, (ii) Attack on the Honlg Kong dollar, (iii) Russia's default, (iv) Devaluation of the Brazilian real, (v) Financial Times article and presidential elections, (vi) Vice president resigns, (vit) Changes of finance minister, and (viii) Cut of intemational credit. The crisis periods for Hong-Kong are: (i) The Mex;can cr;s;s, (ii) the Financia) distleSg in Korea, (iii) the Attacks on the Thai baht and the four attacks on the Hong Kong dollar. See text for a description of all the events. All rates are in basis points, annualized, and continuously compounded. Dates Obervof Mean Median Standard Min Max Obserralia,ss Deviation Argentina NDF-BAIBOR 1-month Total 04/21/97-06/05/01 720 509 64 1,319 -109 11,221 Tranquil Periods 392 154 37 493 -105 4,520 Crisis Periods 328 934 183 1,790 -109 11,221 Russia's default 08/19/98 - 10/16/98 27 618 674 377 1 1,407 Devaluation of the Brazilian real 01/13/99 - 02/12/99 22 343 374 225 5 673 Financial Times article and elections 05/17/99 - 12/17/99 141 79 60 109 -109 580 Vice president resigns 10/06/00 - 12/29/00 54 231 158 183 -14 712 Changes of finance minister 03/16/01 - 05/18/01 36 1,651 1,108 1,695 29 7,102 Cut of international credit 07/10/01 - 09/25/01 48 4,147 4,054 2,462 632 11,221 NDF-BAIBOR 12-month Total 04/21/97-06/05/01 905 285 161 381 -202 2,252 Tranquil Periods 546 201 142 229 -202 1,340 Crisis Periods 359 413 268 509 -154 2,252 Attack on the Hong Kong dollar 10/29/97 - 11/26/97 13 302 252 126 122 568 Russia's default 08/19/98 - 10/16/98 41 662 654 189 385 1,121 Devaluation of the Brazilian real 01/13/99 - 02/12/99 22 369 369 150 150 730 Financial Times article and elections 05/17/99 - 12/17/99 143 131 68 184 -154 595 Vice president resigns 10/06/00 - 12/29/00 54 141 127 90 -15 362 Changes of finance minister 03/16/01 - 05/18/01 38 374 352 245 -83 916 Cut of international credit 07/10/01 -09/25/01 48 1,431 1,510 569 -63 2,252 Hong Kong FWD-HIBOR 1-month Total 01/04/93 - 06/05/01 2,090 18 6 57 -523 688 Tranquil Periods 1,782 11 5 30 -85 306 Crisis Periods 308 57 24 123 -523 688 Mexican crisis 01/10/95 - 04/10/95 59 36 22 48 -111 175 Early signs of financial distress 01/27/97 - 02/21/97 17 53 1 119 -20 386 Attack on the Thai Bath 05/14/97 - 7/24/97 46 131 35 163 -26 535 1st Attack on the Hong-Kong dollar 08/15/97 - 12/15/97 81 44 23 132 -497 688 2nd Attack on the Hong-Kong dollar 01/05/98 - 02/04/98 19 61 44 132 -106 481 3rd Attack on the Hong-Kong dollar 05/27/98 - 07/06/98 28 46 26 94 -181 317 4th Attack on the Hong-Kong dollar 07/10/98 - 10/06198 58 46 26 120 -523 404 FWD-HIBOR 12-month Total 01/04/93 - 06/05/01 2,106 -5 -3 28 -833 150 Tranquil Periods 1,795 -4 -3 14 -69 99 Crisis Periods 311 -12 -9 66 -833 150 Mexican crisis 01/10/95 - 04/10/95 60 -24 -18 30 -194 43 Early signs of financial distress 01/27/97 - 02/21/97 18 -9 -10 8 -22 5 Attack on the Thai Bath 05/14/97 - 7/24/97 46 -13 -13 8 -32 1 1st Attack on the Hong-Kong dollar 08/15/97 - 12/15/97 81 -29 -8 110 -833 150 2nd Attack on the Hong-Kong dollar 01/05/98 - 02/04/98 20 4 19 60 -140 89 3rd Attack on the Hong-Kong dollar 05/27/98 - 07/06/98 28 16 13 31 -43 88 4th Attack on the Hiong-Kong dollar 07/10/98 - 10/06/98 58 4 3 49 -257 136 Appendix Table 3 Data Description Deseription Maturly Sample Source Argentine data Non-deliverable forward Pesos per U.S. dollar; for some I and 12 months 9/9/98-9/25/01 Bloomberg (I and 12 exchange rates (NDF) dates the rates are reported as (I-month) months from 9/9/98); and points for others as outright. 10/20/97-9/25/01 Deutsche Bank (12-moonth (12-month) before 9/9/98) Spot exchange rate Pesos per U.S. dollars - 10/20/97 - 9/25/01 Bloomberg Interbank offer rates in Annualized rate I and 12 months 4/21/97 - 9/25/01 Bloomberg pesos and dollars (BAIBOR) Time deposit rates in pesos Annualized rate I month and up to 2 1/4/1993 - 6/5/01 central bank of Argentina and dollars months Interbank deposit rates in Annualized rate I month 1/13/1995 - 9/25/01 Bloomberg pesos and dollars Total reserves of the central Total reserves held by the central - 12/29/1994 - 9/25/01 Bloomberg bank bank of Argentina (government (original source: central bonds and hard currency, U.S. bank of Agentina) dollar, billions) Bond reserves of the central Reserves in Argentine government - 12/29/1994 - 9/25/01 Bloomberg bank bonds, (U.S. dollar, billions) (original source: central bank of Agentina) Hard currency reserves Reserves in hold, currency and - 12/29/1994 - 9/25/01 Bloomberg short and long-term deposits (U.S. (original source: central dollar, billions) bank of Agentina) Total deposits of the Total Argentine bank deposits - 12/29/1994 - 9/25/01 Bloomberg financial system (U.S. dollar, millions) (original source: central bank of Agentina) Total cash holdings of the Cash holdings (effectivo) in local - 12/29/1994 - 9/25/01 Bloomberg financial system and foreign currency (U.S. dollars, (original source: central millions) bank of Agentina) Hong Kong data Deliverable forward Hong Kong dollars per U.S. dollar I and 12 months 1/4/93 - 9/25/01 Bloomberg exchange rate (FWD) Spot exchange rate Hong Kong dollars per U.S. dollar - 1/4/93 - 9/25/01 Bloomberg Interbank rate in Hong Kong Annualized rate I and 12 months 1/4/93 - 9/25/01 Hong Kong Monetary dollars (HIBOR) Authority International data U.S. Federal Funds rates Annualized rate I month 1/4/93 - 9/25/01 Bloomberg (FFR) U.S. Deposit rates Annualized rate I month 9/25/96-9/25/01 Bloomberg U.S. Treasury bill rates Annualized rate 3 and 12 months 1/4/93 - 9/25/01 Bloomberg U.S. dollar LIBOR Annlualized rate I and 12 months 1/4/93 - 9/25/01 Bloomberg High yield spread Spread of Moodys junk bond - 1/4/93 - 9/25/01 Bloomberg index over the U.S. 30-year government bond yield EMBI spread for Latin Weighted average of the EMBI - 1/4/93 - 9/25/01 JP Morgan American Countries spreads of Latin American countries excluding Argentina, using 1999 GDP weights Policy Research Working Paper Series Contact Title Author Date for paper WPS2791 The Static and Dynamic Incidence of Dominique van de Walle February 2002 H. Sladovich Vietnam's Public Safety Net 37698 WPS2792 Determinants of Life Insurance Thorsten Beck February 2002 A. Yaptenco Consumption across Countries Ian Webb 31823 WPS2793 Agricultural Markets and Risks: Panos Varangis February 2002 P. Kokila Management of the Latter, Not the Donald Larson 33716 Former Jock R. Anderson WPS2794 Land Policies and Evolving Farm Zvi Lerman February 2002 M. Fernandez Structures in Transition Countries Csaba Csaki 33766 Gershon Feder WPS2795 Inequalities in Health in Developing Adam Wagstaff February 2002 H. Sladovich Countries: Swimming against the Tide? 37698 WPS2796 Do Rural Infrastructure Investments Jocelyn A. Songco February 2002 H. Sutrisna Benefit the Poor? Evaluating Linkages: 88032 A Global View, A Focus on Vietnam WPS2797 Regional Integration and Development Maurice Schiff February 2002 P. Flewitt in Small States 32724 WPS2798 Fever and Its Treatment among the Deon Filmer March 2002 H. Sladovich More or Less Poor in Sub-Saharan 37698 Africa WPS2799 The Impact of the Indonesian Lisa A. Cameron March 2002 P. Sader Financial Crisis on Children: Data from 33902 100 Villages Survey WPS2800 Did Social Safety Net Scholarships Lisa A. Cameron March 2002 P. Sader Reduce Drop-Out Rates during the 33902 Indonesian Economic Crisis? WPS2801 Policies to Promote Saving for Dimitri Vittas March 2002 P. Infante Retirement: A Synthetic Overview 37642 WPS2802 Telecommunication Reforms, Access Antonio Estache March 2002 G. Chenet-Smith Regulation, and Internet Adoption Marco Manacorda 36370 in Latin America Tommaso M. Valletti WPS2803 Determinants of Agricultural Growth Yair Mundlak March 2002 P. Kokila in Indonesia, the Philippines, and Donald F. Larson 33716 Thailand Rita Butzer WPS2804 Liberalizing Trade in Agriculture: John S. Wilson March 2002 P. Flewitt Developing Countries in Asia and the 32724 Post-Doha Agenda WPS2805 To Spray or Not to Spray? Pesticides, John S. Wilson March 2002 P. Flewitt Banana Exports, and Food Safety Tsunehiro Otsuki 32724 Policy Research Working Paper Series Contact Title Author Date for paper WPS2806 Dirty Exports and Environmental John S. Wilson March 2002 P. Flewitt Regulation: Do Standards Matter to Tsunehiro Otsuki 32724 Trade? WPS 2807 The Role of Natural Resources in Benoit Bosquet March 2002 D. Duff Fundamental Tax Reform in the 39506 Russian Federation WPS2808 A Capital Accord for Emerging Andrew Powell March 2002 E. Mekhova Economies 85984 WPS2809 On the Measurement and Impact of Robert D. Ebel March 2002 M. Morris Fiscal Decentralization Serdar Yilmaz 37285 WPS2810 Growth, Distribution, and Poverty Luc Christiaensen March 2002 N. Nouviale in Africa: Messages from the 1990s Lionel Demery 34514 Stefano Paternostro WPS2811 The Epidemiological Impact of an John Stover March 2002 H. Sladovich HIV/AIDS Vaccine in Developing Geoff P. Garnett 37698 Countries Steve Seitz Steven Forsythe WPS2812 Can Financial Markets be Tapped to Jerry Skees March 2002 P. Kokila Help Poor People Cope with Weather Panos Varangis 33716 Risks? Donald Larson Paul Siegel WPS2813 The Collective Model of the Household Kaushik Basu March 2002 N. Jameson and an Unexpected Implication for Ranjan Ray 30677 Child Labor: Hypothesis and an Empirical Test WPS2814 Estimating the Endogenously Gayatri Koolwal March 2002 N. Jameson Determined Intrahousehold Balance Ranjan Ray 30677 of Power and Its Impact on Expenditure Pattern: Evidence from Nepal