Exchange rates and insulation in emerging markets

Eichengreen, Barry; Park, Donghyun; Ramayandi, Arief; Shin, Kwanho Working Paper Exchange rates and insulation in emerging markets ADB Economics Working Paper Series, No. 610 Provided in Cooperation with: Asian Development Bank (ADB), Manila Suggested Citation: Eichengreen, Barry; Park, Donghyun; Ramayandi, Arief; Shin, Kwanho (2020) : Exchange rates and insulation in emerging markets, ADB Economics Working Paper Series, No. 610, Asian Development Bank (ADB), Manila, https://doi.org/10.22617/WPS200078-2 This Version is available at: https://hdl.handle.net/10419/230361 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/3.0/igo/ ASIAN DEVELOPMENT BANK ADB ECONOMICS WORKING PAPER SERIES NO. 610 February 2020 EXCHANGE RATES AND INSULATION IN EMERGING MARKETS Barry Eichengreen, Donghyun Park, Arief Ramayandi, and Kwanho Shin ASIAN DEVELOPMENT BANK ADB Economics Working Paper Series Exchange Rates and Insulation in Emerging Markets Barry Eichengreen, Donghyun Park, Arief Ramayandi, and Kwanho Shin No. 610 | February 2020 Barry Eichengreen ([email protected]) is a professor at the Department of Economics, University of California Berkeley. Donghyun Park ([email protected]g) and Arief Ramayandi (arama[email protected]) are principal economists at the Economic Research and Regional Cooperation Department, Asian Development Bank. Kwanho Shin ([email protected]) is a professor at the Department of Economics, Korea University. This paper was prepared as background material for the Asian Development Outlook 2018 Update on “Maintaining Stability amid Heightened Uncertainty.” For comments we thank Ila Patnaik and other workshop participants in the Asian Development Outlook Update Midterm Workshop. We also thank Dohoon Kim for excellent research assistance; and Mahvash S. Qureshi, Beth Anne Wilson, and Andrea Raffo for help with the data. Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) © 2020 Asian Development Bank 6 ADB Avenue, Mandaluyong City, 1550 Metro Manila, Philippines Tel +63 2 8632 4444; Fax +63 2 8636 2444 www.adb.org Some rights reserved. Published in 2020. ISSN 2313-6537 (print), 2313-6545 (electronic) Publication Stock No. WPS200078-2 DOI: http://dx.doi.org/10.22617/WPS200078-2 The views expressed in this publication are those of the authors and do not necessarily reflect the views and policies ofthe Asian Development Bank (ADB) or its Board of Governors or the governments they represent. ADB does not guarantee the accuracy of the data included in this publication and accepts no responsibility for any consequence of their use. The mention of specific companies or products of manufacturers does not imply that they are endorsed or recommended by ADB in preference to others of a similar nature that are not mentioned. By making any designation of or reference to a particular territory or geographic area, or by using the term “country” inthis document, ADB does not intend to make any judgments as to the legal or other status of any territory or area. This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) https://creativecommons.org/licenses/by/3.0/igo/. By using the content of this publication, you agree to be bound bytheterms of this license. For attribution, translations, adaptations, and permissions, please read the provisions andterms of use at https://www.adb.org/terms-use#openaccess. This CC license does not apply to non-ADB copyright materials in this publication. If the material is attributed toanother source, please contact the copyright owner or publisher of that source for permission to reproduce it. ADB cannot be held liable for any claims that arise as a result of your use of the material. Please contact [email protected] if you have questions or comments with respect to content, or if you wish toobtain copyright permission for your intended use that does not fall within these terms, or for permission to use theADB logo. Corrigenda to ADB publications may be found at http://www.adb.org/publications/corrigenda. Notes: In this publication, “$” refers to United States dollars. ADB recognizes “Korea” as the Republic of Korea. The ADB Economics Working Paper Series presents data, information, and/or findings from ongoing research and studies to encourage exchange of ideas and to elicit comment and feedback about development issues in Asia and the Pacific. Since papers in this series are intended for quick and easy dissemination, the content may or may not be fully edited and may later be modified for final publication. CONTENTS TABLES AND FIGURE iii ABSTRACT iv I. INTRODUCTION 1 II. RELATED LITERATURE 4 III. DATA AND METHODS 5 IV. FINDINGS 7 V. CONCLUSION 31 APPENDIXES 33 REFERENCES 35 TABLES AND FIGURE TABLES 1 Impact of VXO on Various Variables in Emerging Market Economies: Propensity Score 8 Weighted Regressions 1.1 Impact of VXO on Real Domestic Credit Growth 8 1.2 Impact of VXO on Real House Price Growth 10 1.3 Impact of VXO on Real Stock Returns 11 1.4 Impact of VXO on the Change in the Loan-to-Deposit Ratio 12 1.5 Impact of VXO on Net Capital Flows 13 1.6 Impact of VXO on Liability Flows 14 1.7 Impact of VXO on Asset Flows 15 1.8 Impact of VXO on Foreign Direct Investment, Portfolio, and Other Investment 16 Liability Flows 1.9 Impact of VXO on Real Gross Domestic Product Growth 18 2 Impact of the Londono–Wilson Index on Various Variables in Emerging Market Economies 19 3 Impact of the United States Industrial Production Growth on Various Variables 21 in Emerging Market Economies 4 Impact of the Expected Probability of a United States Recession on Various Variables 23 in Emerging Market Economies 5 Impact of the United States Economic Policy Uncertainty Index on Various Variables 24 in Emerging Market Economies 6 Impact of the Federal Funds Rate on Various Variables in Emerging Market Economies 25 7 Impact of the Change in Total Assets of the Federal Reserve on Various Variables in 26 Emerging Market Economies 8 Impact of Non-United States Industrial Production Growth on Various Variables in 28 Emerging Market Economies 9 Impact of the Global Economic Policy Uncertainty Index on Various Variables in 29 Emerging Market Economies 10 Impact of the Expected Probability of Recessions Outside the United States on 30 Various Variables in Emerging Market Economies A.1 Description and Sources for the Variables Used by Obstfeld, Ostry, and Qureshi 33 A.2 Description and Sources for the Variables Used by Londono and Wilson 34 FIGURE Exchange Rate Regimes in Asian Countries 6 ABSTRACT The insulating properties of flexible exchange rates have long been a highly contentious issue in emerging markets—not least in Asian emerging markets. A number of recent theoretical and empirical studies question whether a trade-off exists between rigid exchange rate regimes and insulation from foreign shocks when the degree of international capital mobility is high. On the other hand, Obstfeld, Ostry, and Qureshi (2017) find that countries with flexible exchange rate regimes experience less real and financial instability in the face of global financial volatility. We contribute to this empirical debate by significantly extending their analysis. Overall, our findings are broadly consistent with their results, suggesting that flexible exchange rate regimes are better at insulating emerging markets from external shocks. There are, however, a few subtle differences. In particular, we find somewhat less robust evidence that limited flexibility is enough to insulate emerging markets from shocks. Keywords: exchange rate, exchange rate regime, fixed, flexible, insulate, intermediate, shock JEL code: F31 I.  INTRODUCTION The insulating properties of flexible exchange rates have long been a contentious issue in emerging markets—not least in Asian emerging markets. The debate goes back to the 1930s, when Japan—Asia’s first emerging market—quickly recovered from the Great Depression by abandoning gold convertibility and depreciating the yen (Yasuba 1988). In the 1990s, Asia’s emerging markets resisted international calls to move to greater exchange rate flexibility and experienced large financial inflows from abroad before being hit by an appreciating dollar–yen rate, higher oil prices, and a weakening global semiconductor market, resulting in the Asian financial crisis. This experience accentuated calls for greater exchange rate flexibility in Asia and in emerging markets generally (Eichengreen 1999). Accordingly, many Asian economies moved in the direction of greater flexibility, although with notable exceptions, such as Hong Kong, China—and there may still be a gap between rhetoric and reality. According to Pontines and Rajan (2011), intervention in foreign exchange markets is extensive, suggesting that policy makers still have doubts about the insulating properties of full exchange rate flexibility, or at least they question whether the benefits in terms of insulation justify the costs. These costs can be reductions in exports and export-led growth if a competitively valued peg is sacrificed or financial fragility in the presence of currency mismatches on corporate, bank, and public balance sheets. For some, these doubts are not entirely unfounded. The most recent theoretical and empirical literature questions whether a trade-off exists between exchange rate stability and insulation from foreign shocks in a setting of high international capital mobility. The traditional approach, grounded in the Mundell–Fleming model that posits the existence of this trade-off, is framed in terms of a trilemma (Mundell 1963). In this view, policy makers can attain only two of three desiderata: exchange rate stability, international capital mobility, and monetary autonomy. Here, it is the autonomy to alter domestic monetary policy that facilitates domestic adjustment and insulates against foreign shocks. The more recently observed global financial cycles which are characterized by large common movements in asset prices, gross flows, and leverage, have challenged this traditional view and argue for a policy dilemma rather than trilemma (Rey 2015, 2016). This posits a global financial cycle or disturbance that affects all countries, regardless of their exchange rate regime. Therefore, independent monetary policies are only possible if and only if the capital account is managed either directly or indirectly. A second, related strand of recent research focuses on the international role of the dollar. This argues that exchange rate flexibility provides emerging markets with at best limited flexibility from shocks emanating from markets in dollar-denominated assets. Shin (2016) argues that dollardenominated credit is the dominant form of funding in the international interbank market, so that changes in interest rates in the United States (US) (and in euro–dollar markets) can strongly affect financial conditions in emerging markets regardless of their exchange rate regime. Gopinath (2017a, 2017b) emphasizes the prevalence of dollar pricing in international merchandise transactions, and the implication that nominal exchange rate changes will not translate into commensurate real exchange rate changes, thereby limiting the stabilization benefits of currency flexibility in developing countries. This recent revisionism concerning the benefits of exchange rate flexibility has not gone unchallenged. Obstfeld, Ostry, and Qureshi (2017), in their major empirical analysis, marshal evidence of the stabilizing properties of flexible exchange rates. They study some 40 emerging markets over 1986– 2013, distinguishing countries with fixed, intermediate, and flexible exchange rates, and consider the transmission of global financial shocks to domestic credit growth, house price growth, capital flows, and 2 | ADB Economics Working Paper Series No. 610 gross domestic product (GDP) growth, among other variables. Their results support the hypothesis that countries with fixed exchange rates experience greater real and financial instability and cyclicality because of global financial volatility. But they do not find that countries with intermediate exchange rate regimes with a limited degree of exchange rate flexibility are more susceptible to global financial volatility than countries with flexible exchange rates. Their conclusions suggest that even limited degrees of exchange rate flexibility can have useful insulation properties and thus support the approach of many Asian countries. The inconclusive nature of this debate poses a dilemma for policy makers in Asia. Should they continue to move in the direction of exchange rate flexibility to enhance their insulation from global financial shocks? Or have the insulating properties of flexible exchange rates been oversold? There is no certain answer to either question because of the very different conclusions of the various studies. This paper aims to shed further light on these issues by revisiting and extending Obstfeld, Ostry, and Qureshi (2017). Specifically, we consider the following extensions and sensitivity checks: (i)We consider the robustness of the findings of Obstfeld, Ostry, and Qureshi (2017) to alternative data sources. As some of the authors’ series are proprietary, we have to consider using different proprietary or publicly available data in our analysis. 1 The comparison of our data sources to theirs is summarized in Appendix Table A.1. (ii)We consider a wider variety of global financial shocks. Obstfeld, Ostry, and Qureshi (2017) consider the VXO (the precursor of the Chicago Board Options Exchange’s Volatility Index, which is constructed using the implied volatility of a range of S&P 500 index options). We also consider other measures of global volatility. These include, following Londono and Wilson (2018), a global volatility index calculated as the market-value-weighted average of the implied volatility of equity options for seven countries (France, Germany, Japan, the Netherlands, Switzerland, the United Kingdom, and the US). We also consider three categories of fundamental drivers of global shocks, also following the Londono–Wilson approach. The first category comprises three US economic and risk variables (industrial production growth, the expected probability of a US recession within the next quarter, and economic policy uncertainty). The second is a pair of US monetary policy shocks (changes in the federal funds rate and changes in Federal Reserve assets, the latter as a way of capturing unconventional monetary policies). The third is a pair of global factors not emanating from the US (non-US industrial production growth and the Global Economic Policy Uncertainty Index). (iii)We also use an alternative empirical methodology designed to better control for the ways in which treatment-group countries (those with pegged exchange rates) and control-group countries (those with flexible exchange rate) differ. Ordinary regression methods may not be ideal for estimating the impact of alternative exchange rate regimes insofar as countries with different (observable and unobservable) characteristics are weighted equally. We instead use propensity score weighted regression methods; these are expressly designed to give more weight to comparable observations within treatment and control groups. 1 Most nonproprietary data are from Obstfeld, Ostry, and Qureshi (2017), and we greatly appreciate the authors sharing their data. Exchange Rates and Insulation in Emerging Markets | 9  1986– 2013 1986– 2013 1986– 2013 1986– 2013 1986– 2013 2000– 2013 1986– 2013 Variables (1) (2) (3) (4) (5) (6) (7) Lagged private credit/GDP –0.111*** –0.113*** –0.116*** –0.120*** –0.103*** –0.139*** –0.095*** (0.013) (0.014) (0.016) (0.017) (0.013) (0.016) (0.021) Real US T-bill rate 0.049 (0.064) Intermediate × real US T-bill rate 0.050 (0.065) Fixed × real US T-bill rate –0.159 (0.206) Real shadow federal funds –0.006 (0.153) Intermediate × real shadow rate 0.321* (0.165) Fixed × real shadow rate –0.014 (0.288) Lagged net capital flows/GDP –0.001 (0.007) Lagged central bank policy rate –0.326*** (0.114) Linear trend 0.029** 0.030** 0.032** 0.043** (0.014) (0.014) (0.013) (0.016) Global financial crisis 2.632*** 2.597*** 2.556*** 2.542*** (0.534) (0.503) (0.511) (0.501) Quarter-year effects No No No No Yes Yes Yes Observations 2,484 2,484 2,484 2,449 2,484 1,828 1,600 Adjusted R 2 0.260 0.267 0.270 0.279 0.309 0.497 0.572 No. of countries 43 43 43 43 43 42 35 GDP = gross domestic product, US = United States. Notes: 1. This is identical to Table 1 in Obstfeld, Maurice, Jonathan Ostry, and Mahvash Qureshi. 2017. “A Tie That Binds: Revisiting the Trilemma in Emerging Market Economies.” IMF Working Paper WP/17/130. Washington, DC: International Monetary Fund. The one exception is that we rely on an alternative methodology of propensity score weighted regressions. 2. The dependent variable is the three-quarter moving average of quarterly real domestic private sector credit growth rate (in percent) in emerging market economies. 3. VXO is the precursor of the Chicago Board Options Exchange’s Volatility Index or VIX. 4. Numbers in parentheses are clustered standard errors (by country). ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively. Source: Authors’ calculations.      Table 1.1 continued 10 | ADB Economics Working Paper Series No. 610 Table 1.2: Impact of VXO on Real House Price Growth  1986– 2013 1986– 2013 1986– 2013 1986– 2013 1986– 2013 2000– 2013 1986– 2013 Variables (1) (2) (3) (4) (5) (6) (7) Intermediate regime 1.361** 2.154 2.264 2.702 1.302 2.174 2.056 (0.602) (4.630) (4.831) (4.704) (4.796) (4.768) (5.595) Fixed regime 1.384 20.836*** 20.990*** 20.522*** 18.186*** 18.325*** 19.817*** (1.027) (5.493) (5.648) (6.230) (5.018) (5.048) (5.465) Log(VXO) –2.327*** 0.100 0.139 0.020 (0.802) (0.937) (0.985) (1.108) Intermediate × log(VXO) –0.099 –0.133 –0.285 0.167 0.106 0.272 (1.439) (1.499) (1.480) (1.553) (1.544) (1.817) Fixed × log(VXO) –6.187*** –6.222*** –6.171*** –5.444*** –5.258*** –5.081** (1.652) (1.711) (1.830) (1.618) (1.624) (1.757) Lagged real GDP growth 1.155*** 1.014*** 1.003*** 1.028*** 1.410*** 1.419*** 1.463*** (0.246) (0.261) (0.265) (0.268) (0.158) (0.157) (0.301) Lagged domestic credit growth 0.213*** 0.162** 0.163* 0.160** 0.198** 0.197** 0.187** (0.066) (0.076) (0.078) (0.072) (0.073) (0.073) (0.079) Real US T-bill rate –0.003 (0.070) Intermediate × real US T-bill rate 0.067 (0.124) Fixed × real US T-bill rate 0.027 (0.068) Real shadow federal funds –0.114 (0.118) Intermediate × real shadow rate –0.003 (0.289) Fixed × real shadow rate 0.067 (0.177) Lagged net capital flows/GDP 0.007 (0.021) Lagged central bank policy rate 0.063 (0.103) Linear trend –0.029 –0.033 –0.031 –0.043 (0.018) (0.022) (0.024) (0.029) Global financial crisis –1.452 –1.224 –1.256 –1.149 (1.182) (1.109) (1.154) (1.118) Quarter-year effects No No No No Yes Yes Yes Observations 579 579 579 562 579 566 497 Adjusted R 2 0.436 0.519 0.515 0.511 0.556 0.573 0.541 No. of countries 18 18 18 18 18 18 17 GDP = gross domestic product, US = United States. Notes: 1. This is identical to Table 2 in Obstfeld, Maurice, Jonathan Ostry, and Mahvash Qureshi. 2017. “A Tie That Binds: Revisiting the Trilemma in Emerging Market Economies.” IMF Working Paper WP/17/130. Washington, DC: International Monetary Fund. The one exception is that we rely on an alternative methodology of propensity score weighted regressions. 2. The dependent variable is the three-quarter moving average of quarterly real house price growth rate (in percent) in emerging market economies. 3. VXO is the precursor of the Chicago Board Options Exchange’s Volatility Index or VIX. 4. Numbers in parentheses are clustered standard errors (by country). ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively. Source: Authors’ calculations.  Exchange Rates and Insulation in Emerging Markets | 11 Table 1.3: Impact of VXO on Real Stock Returns  1986– 2013 1986– 2013 1986– 2013 1986– 2013 1986– 2013 2000– 2013 1986– 2013 Variables (1) (2) (3) (4) (5) (6) (7) Intermediate regime 0.054 –9.926 –11.631* –11.332 0.717 –0.268 –0.922 (0.776) (6.261) (6.262) (6.902) (6.553) (6.848) (6.742) Fixed regime –2.730* 1.387 1.036 2.801 7.328 5.201 5.936 (1.593) (7.360) (7.111) (7.776) (6.324) (6.479) (7.413) Log(VXO) –8.808*** –9.529*** –10.017*** –11.160*** (0.869) (1.183) (1.119) (1.144) Intermediate × log(VXO) 3.310 3.761* 3.724 0.623 1.120 1.304 (2.059) (2.067) (2.229) (2.095) (2.167) (2.131) Fixed × log(VXO) –1.357 –1.415 –1.598 –2.449 –1.584 –2.219 (2.441) (2.405) (2.441) (2.185) (2.141) (2.357) Lagged real GDP growth –1.540*** –1.569*** –1.313*** –1.308*** –0.018 –0.163 –0.063 (0.372) (0.378) (0.362) (0.382) (0.226) (0.293) (0.330) Lagged domestic credit growth 0.115 0.104 0.115 0.170* 0.049 –0.002 –0.019 (0.087) (0.088) (0.085) (0.095) (0.074) (0.085) (0.107) Real US T-bill rate –0.766*** (0.102) Intermediate × real US T-bill rate –0.324 (0.196) Fixed × real US T-bill rate –0.485** (0.180) Real shadow federal funds –1.061*** (0.181) Intermediate × real shadow rate –0.266 (0.413) Fixed × real shadow rate –0.504 (0.308) Lagged net capital flows/GDP –0.119*** (0.033) Lagged central bank policy rate –0.113 (0.105) Linear trend –0.060*** –0.063*** –0.122*** –0.163*** (0.018) (0.018) (0.018) (0.034) Global financial crisis –9.352*** –9.272*** –7.404*** –9.111*** (2.372) (2.383) (2.307) (2.367) Quarter-year effects No No No No Yes Yes Yes Observations 1,766 1,766 1,766 1,737 1,766 1,385 1,233 Adjusted R 2 0.204 0.208 0.270 0.249 0.555 0.589 0.591 No. of countries 34 34 34 34 34 33 29 GDP = gross domestic product, US = United States. Notes: 1. This is identical to Table 3 in Obstfeld, Maurice, Jonathan Ostry, and Mahvash Qureshi. 2017. “A Tie That Binds: Revisiting the Trilemma in Emerging Market Economies.” IMF Working Paper WP/17/130. Washington, DC: International Monetary Fund. The one exception is that we rely on an alternative methodology of propensity score weighted regressions. 2. The dependent variable is the three-quarter moving average of quarterly real stock price growth rate (in percent) in emerging market economies. 3. VXO is the precursor of the Chicago Board Options Exchange’s Volatility Index or VIX. 4. Numbers in parentheses are clustered standard errors (by country). ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively. Source: Authors’ calculations.  12 | ADB Economics Working Paper Series No. 610 Table 1.4: Impact of VXO on the Change in the Loan-to-Deposit Ratio  1986– 2013 1986– 2013 1986– 2013 1986– 2013 1986– 2013 2000– 2013 1986– 2013 Variables (1) (2) (3) (4) (5) (6) (7) Intermediate regime 1.622** 4.431 4.731* 4.354 5.721** 4.973** 3.318 (0.744) (2.704) (2.655) (2.706) (2.361) (2.446) (2.411) Fixed regime 4.332** 8.384** 8.676*** 8.711*** 7.780*** 6.712** 4.150 (1.724) (3.134) (3.195) (2.872) (2.301) (2.654) (2.769) Log(VXO) –0.704 0.002 0.095 0.422 (0.433) (0.604) (0.564) (0.567) Intermediate × log(VXO) –0.927 –1.030 –1.121 –1.527** –1.652** –1.389* (0.816) (0.796) (0.800) (0.752) (0.658) (0.690) Fixed × log(VXO) –1.343* –1.436* –1.640** –1.668** –1.738** –1.569* (0.744) (0.734) (0.721) (0.685) (0.690) (0.783) Lagged real GDP growth 0.531*** 0.527*** 0.519*** 0.476*** 0.450*** 0.366* 0.512*** (0.141) (0.144) (0.149) (0.150) (0.141) (0.199) (0.153) Lagged LTD ratio –0.065*** –0.066*** –0.066*** –0.070*** –0.065*** –0.085*** –0.058*** (0.013) (0.014) (0.013) (0.013) (0.011) (0.018) (0.017) Real US T-bill rate –0.008 (0.066) Intermediate × real US T-bill rate 0.086 (0.075) Fixed × real US T-bill rate 0.044 (0.160) Real shadow federal funds 0.081 (0.107) Intermediate × real shadow rate 0.251* (0.140) Fixed × real shadow rate 0.226 (0.288) Lagged net capital flows/GDP 0.001 (0.018) Lagged central bank policy rate –0.158*** (0.037) Linear trend 0.010 0.010 0.013* 0.032*** (0.007) (0.007) (0.007) (0.009) Global financial crisis 1.667*** 1.651*** 1.585** 1.639*** (0.613) (0.599) (0.614) (0.582) Quarter-year effects No No No No Yes Yes Yes Observations 2,484 2,484 2,484 2,449 2,484 1,828 1,600 Adjusted R 2 0.193 0.195 0.196 0.215 0.369 0.325 0.336 No. of countries 43 43 43 43 43 42 35 GDP = gross domestic product, LTD = loan-to-deposit, US = United States. Notes: This is identical to Table 4 in Obstfeld, Maurice, Jonathan Ostry, and Mahvash Qureshi. 2017. “A Tie That Binds: Revisiting the Trilemma in Emerging Market Economies.” IMF Working Paper WP/17/130. Washington, DC: International Monetary Fund. The one exception is that we rely on an alternative methodology of propensity score weighted regressions. 2. The dependent variable is the three-quarter moving average of change in the loan-to-deposit ratio in emerging market economies. 3. VXO is the precursor of the Chicago Board Options Exchange’s Volatility Index or VIX. 4. Numbers in parentheses are clustered standard errors (by country). ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively. Source: Authors’ calculations.  Exchange Rates and Insulation in Emerging Markets | 13 Table 1.5: Impact of VXO on Net Capital Flows  1986–2013 1986–2013 1986–2013 1986–2013 1986–2013 2000–2013 Variables (1) (2) (3) (4) (5) (6) Intermediate regime 1.726** 5.853 5.925 6.469* 7.124* 8.375** (0.657) (3.717) (3.755) (3.444) (3.587) (3.566) Fixed regime 1.184 15.231** 15.338** 14.693** 15.942** 16.714*** (1.994) (5.960) (5.945) (5.435) (5.995) (5.696) Log(VXO) –1.832*** 0.012 0.054 0.406 (0.656) (0.844) (0.857) (0.831) Intermediate × log(VXO) –1.343 –1.378 –1.678 –1.616 –1.975 (1.177) (1.194) (1.123) (1.155) (1.208) Fixed × log(VXO) –4.664** –4.706** –4.545*** –4.622** –4.493** (1.758) (1.761) (1.667) (1.760) (1.660) Lagged real GDP growth 0.331*** 0.315*** 0.313*** 0.288*** 0.319*** 0.393* (0.109) (0.105) (0.106) (0.102) (0.115) (0.204) Lagged institutional quality 20.900*** 22.387*** 21.992*** 21.928*** 15.900** 20.027** (5.530) (5.495) (5.352) (5.587) (6.016) (9.729) Lagged domestic credit/GDP –0.039 –0.047* –0.047* –0.053** –0.048* –0.089*** (0.025) (0.024) (0.024) (0.025) (0.026) (0.027) Real US T-bill rate 0.012 (0.086) Intermediate × real US T-bill rate 0.034 (0.119) Fixed × real US T-bill rate 0.165 (0.245) Real shadow federal funds 0.170 (0.173) Intermediate × real shadow rate 0.177 (0.209) Fixed × real shadow rate 0.144 (0.481) Linear trend 0.000 0.002 0.006 0.028 (0.017) (0.016) (0.018) (0.027) Global financial crisis –1.439 –1.322 –1.458 –1.334 (0.939) (0.982) (0.977) (0.963) Quarter-year effects No No No No Yes Yes Observations 2,083 2,083 2,083 2,051 2,083 1,629 Adjusted R 2 0.380 0.391 0.390 0.392 0.436 0.474 No. of countries 38 38 38 38 38 38 GDP = gross domestic product, US = United States. Notes: 1. This is identical to Table 5 in Obstfeld, Maurice, Jonathan Ostry, and Mahvash Qureshi. 2017. “A Tie That Binds: Revisiting the Trilemma in Emerging Market Economies.” IMF Working Paper WP/17/130. Washington, DC: International Monetary Fund. The one exception is that we rely on an alternative methodology of propensity score weighted regressions. 2. The dependent variable is the three-quarter moving average of quarterly net capital flows (in percent of GDP) in emerging market economies. 3. VXO is the precursor of the Chicago Board Options Exchange’s Volatility Index or VIX. 4. Numbers in parentheses are clustered standard errors (by country). ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively. Source: Authors’ calculations.  14 | ADB Economics Working Paper Series No. 610 Table 1.6: Impact of VXO on Liability Flows  1986–2013 1986–2013 1986–2013 1986–2013 1986–2013 2000–2013 Variables (1) (2) (3) (4) (5) (6) Intermediate regime 0.472 –3.228 –2.813 –1.863 1.979 1.781 (0.627) (3.243) (3.391) (3.054) (4.232) (4.831) Fixed regime 1.559 13.894** 14.144** 13.756** 17.506** 17.441** (1.634) (6.304) (6.296) (5.813) (7.150) (7.124) Log(VXO) –4.497*** –3.568*** –3.426*** –2.655*** (0.864) (0.787) (0.770) (0.850) Intermediate × log(VXO) 1.253 1.090 0.554 0.026 0.234 (1.055) (1.115) (1.098) (1.291) (1.455) Fixed × log(VXO) –4.103* –4.195** –4.160** –4.581** –3.966* (2.042) (2.047) (1.966) (2.226) (2.181) Lagged real GDP growth 0.385*** 0.369*** 0.366*** 0.314*** 0.332*** 0.407* (0.121) (0.115) (0.116) (0.110) (0.119) (0.212) Lagged institutional quality 24.804*** 26.040*** 25.307*** 24.331*** 18.016*** 23.860** (5.474) (5.541) (5.517) (5.767) (6.597) (9.679) Lagged domestic credit/GDP –0.009 –0.013 –0.012 –0.020 –0.008 –0.027 (0.026) (0.024) (0.024) (0.026) (0.026) (0.028) Real US T-bill rate 0.000 (0.136) Intermediate × real US T-bill rate 0.145 (0.216) Fixed × real US T-bill rate 0.256 (0.275) Real shadow federal funds 0.486** (0.180) Intermediate × real shadow rate 0.337 (0.360) Fixed × real shadow rate 0.270 (0.506) Linear trend –0.021 –0.022 –0.013 0.040 (0.018) (0.017) (0.017) (0.029) Global financial crisis –3.243** –3.270* –3.503** –3.160* (1.541) (1.617) (1.589) (1.634) Quarter-year effects No No No No Yes Yes Observations 2,083 2,083 2,083 2,051 2,083 1,629 Adjusted R 2 0.392 0.402 0.402 0.413 0.467 0.501 No. of countries 38 38 38 38 38 38 GDP = gross domestic product, US = United States. Notes: 1. This is identical to Table 6 in Obstfeld, Maurice, Jonathan Ostry, and Mahvash Qureshi. 2017. “A Tie That Binds: Revisiting the Trilemma in Emerging Market Economies.” IMF Working Paper WP/17/130. Washington, DC: International Monetary Fund. The one exception is that we rely on an alternative methodology of propensity score weighted regressions. 2. The dependent variable is the three-quarter moving average of quarterly liability flows (in percent of GDP) in emerging market economies. 3. VXO is the precursor of the Chicago Board Options Exchange’s Volatility Index or VIX. 4. Numbers in parentheses are clustered standard errors (by country). ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively. Source: Authors’ calculations.   Exchange Rates and Insulation in Emerging Markets | 15 Table 1.7: Impact of VXO on Asset Flows  1986–2013 1986–2013 1986–2013 1986–2013 1986–2013 2000–2013 Variables (1) (2) (3) (4) (5) (6) Intermediate regime 1.247** 9.122*** 8.789*** 8.364*** 5.156 6.624* (0.559) (2.673) (2.862) (2.792) (3.400) (3.593) Fixed regime –0.319 1.366 1.221 0.996 –1.534 –0.626 (0.719) (3.685) (3.659) (3.930) (3.907) (4.262) Log(VXO) 2.660*** 3.583*** 3.485*** 3.066*** (0.459) (0.567) (0.541) (0.631) Intermediate × log(VXO) –2.612*** –2.488** –2.249** –1.651 –2.220** (0.865) (0.927) (0.968) (1.050) (1.059) Fixed × log(VXO) –0.553 –0.501 –0.378 –0.022 –0.510 (1.221) (1.206) (1.243) (1.278) (1.292) Lagged real GDP growth –0.055* –0.054* –0.053* –0.027 –0.012 –0.013 (0.031) (0.030) (0.029) (0.026) (0.026) (0.039) Lagged institutional quality –3.757 –3.502 –3.160 –2.231 –1.927 –3.513 (3.609) (3.544) (3.590) (3.544) (3.911) (6.102) Lagged domestic credit/GDP –0.031 –0.034 –0.035 –0.033 –0.040* –0.062** (0.023) (0.022) (0.022) (0.023) (0.023) (0.023) Real US T-bill rate 0.012 (0.100) Intermediate × real US T-bill rate –0.109 (0.157) Fixed × real US T-bill rate –0.094 (0.124) Real shadow federal funds –0.315* (0.166) Intermediate × real shadow rate –0.155 (0.305) Fixed × real shadow rate –0.131 (0.197) Linear trend 0.021 0.024 0.020 –0.012 (0.015) (0.014) (0.014) (0.022) Global financial crisis 1.803 1.936 2.031* 1.816 (1.202) (1.165) (1.139) (1.140) Quarter-year effects No No No No Yes Yes Observations 2,103 2,103 2,103 2,070 2,103 1,649 Adjusted R 2 0.313 0.319 0.320 0.326 0.347 0.377 No. of countries 39 39 39 39 39 39 GDP = gross domestic product, US = United States. Notes: 1. This is identical to Table 7 in Obstfeld, Maurice, Jonathan Ostry, and Mahvash Qureshi. 2017. “A Tie That Binds: Revisiting the Trilemma in Emerging Market Economies.” IMF Working Paper WP/17/130. Washington, DC: International Monetary Fund. The one exception is that we rely on an alternative methodology of propensity score weighted regressions. 2. The dependent variable is the three-quarter moving average of quarterly net capital flows (in percent of GDP) in emerging market economies. 3. VXO is the precursor of the Chicago Board Options Exchange’s Volatility Index or VIX. 4. Numbers in parentheses are clustered standard errors (by country). ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively. Source: Authors’ calculations.   16 | ADB Economics Working Paper Series No. 610 Table 1.8: Impact of VXO on Foreign Direct Investment, Portfolio, and Other Investment Liability Flows  Real Credit Growth Real House Price Growth Change in LTD Ratio Variables (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Intermediate regime 0.848 1.155 1.209 4.780* 1.414 1.362 1.195 –1.076 –2.740 –2.736 –1.875 1.522 (1.988) (2.104) (1.581) (2.616) (1.659) (1.689) (1.365) (1.983) (2.420) (2.424) (2.765) (2.961) Fixed regime 6.912* 6.929* 6.683* 9.591** 0.787 0.599 1.332 –0.700 8.541 8.953* 7.125* 11.170* (3.747) (3.745) (3.510) (4.451) (2.637) (2.567) (3.233) (2.566) (5.134) (5.151) (4.056) (5.800) Log(VXO) –0.110 –0.061 0.441 –0.947** –0.967** –1.092*** –1.744*** –1.662*** –1.319** (0.368) (0.360) (0.403) (0.442) (0.458) (0.353) (0.595) (0.603) (0.645) Intermediate × log(VXO) –0.118 –0.214 –0.464 –1.020 –0.385 –0.374 –0.305 0.156 0.902 0.879 0.511 –0.195 (0.680) (0.726) (0.694) (0.848) (0.523) (0.542) (0.456) (0.618) (0.735) (0.740) (0.806) (0.851) Fixed × log(VXO) –2.360* –2.340* –2.394* –2.819* 0.063 0.115 0.121 0.347 –2.558 –2.728 –2.543 –3.047* (1.238) (1.237) (1.218) (1.418) (0.914) (0.885) (0.905) (0.957) (1.635) (1.675) (1.521) (1.765) Lagged real GDP growth 0.101** 0.098** 0.069 0.051 0.018 0.019 0.026 0.065 0.241*** 0.242*** 0.214*** 0.201*** (0.049) (0.048) (0.044) (0.045) (0.031) (0.031) (0.035) (0.040) (0.075) (0.076) (0.072) (0.071) Lagged institutional quality 7.112*** 7.028** 5.709** 2.940 2.956 3.232 4.261 4.282 16.127*** 15.019** 13.568** 11.954** (2.613) (2.606) (2.331) (3.939) (3.276) (3.468) (3.799) (3.781) (5.808) (5.642) (5.758) (4.453) Lagged domestic credit/GDP 0.014 0.013 0.010 0.025* –0.012 –0.012 –0.012 –0.024* –0.009 –0.007 –0.010 –0.007 (0.012) (0.012) (0.012) (0.013) (0.012) (0.012) (0.013) (0.012) (0.016) (0.016) (0.019) (0.018) Real US T-bill rate 0.096*** –0.048 –0.048 (0.032) (0.091) (0.060) Intermediate × real US T-bill rate 0.041 0.017 0.040 (0.107) (0.108) (0.087) Fixed × real US T-bill rate –0.019 –0.064 0.381* (0.079) (0.152) (0.201) Real shadow federal funds 0.302*** –0.098 0.193** (0.088) (0.154) (0.092) Intermediate × real shadow rate 0.293 0.000 0.102 (0.241) (0.175) (0.129) Fixed × real shadow rate 0.196 –0.243 0.510 (0.165) (0.291) (0.335) continued on next page Exchange Rates and Insulation in Emerging Markets | 17  Real Credit Growth Real House Price Growth Change in LTD Ratio Variables (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Linear trend 0.000 0.007 0.043** 0.018** 0.014** 0.004 –0.027** –0.023* 0.005 (0.006) (0.007) (0.017) (0.007) (0.007) (0.010) (0.012) (0.012) (0.015) Global financial crisis 0.450 0.292 0.528 –1.669*** –1.594*** –1.815*** –1.128 –1.284 –0.971 (0.580) (0.535) (0.570) (0.468) (0.463) (0.429) (0.877) (0.891) (0.903) Quarter-year effects No No No Yes No No No Yes No No No Yes Observations 2,118 2,118 2,084 2,118 2,118 2,118 2,084 2,118 2,083 2,083 2,051 2,083 Adjusted R 2 0.384 0.387 0.405 0.425 0.190 0.191 0.203 0.230 0.340 0.346 0.360 0.435 No. of countries 39 39 39 39 39 39 39 39 38 38 38 38 GDP= gross domestic product, LTD = loan-to-deposit, US = United States. Notes: 1. This is identical to Table 8 in Obstfeld, Maurice, Jonathan Ostry, and Mahvash Qureshi. 2017. “A Tie That Binds: Revisiting the Trilemma in Emerging Market Economies.” IMF Working Paper WP/17/130. Washington, DC: International Monetary Fund. The one exception is that we rely on an alternative methodology of propensity score weighted regressions. 2. The dependent variable is the three-quarter moving average in emerging market economies’ quarterly foreign direct investment liability flows (in percent of GDP) in columns (1)–(4), portfolio liability flows (in percent of GDP) in columns (5)–(8), and other investment liability flows (in percent of GDP) in columns (9)–(12). 3. VXO is the precursor of the Chicago Board Options Exchange’s Volatility Index or VIX. 4. Numbers in parentheses are clustered standard errors (by country). ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively. Source: Authors’ calculations.  Table 1.8 continued 18 | ADB Economics Working Paper Series No. 610 Table 1.9: Impact of VXO on Real Gross Domestic Product Growth  1986–2013 1986–2013 1986–2013 1986–2013 1986–2013 Variables (1) (2) (3) (4) (5) Intermediate regime 0.574 0.474 1.068* 1.239** 1.332*** (0.513) (0.513) (0.583) (0.556) (0.473) Fixed regime 2.664** 2.639** 2.838** 2.922*** 2.654** (1.047) (1.040) (1.130) (0.997) (1.028) Log(VXO) –0.487*** –0.515*** –0.499*** (0.122) (0.122) (0.141) Intermediate × log(VXO) –0.196 –0.158 –0.286 –0.390** –0.356** (0.171) (0.168) (0.177) (0.189) (0.165) Fixed × log(VXO) –0.811** –0.807** –0.876** –0.882** –0.752** (0.334) (0.332) (0.347) (0.333) (0.324) Real US T-bill rate –0.004 (0.019) Intermediate × real US T-bill rate –0.057** (0.027) Fixed × real US T-bill rate –0.012 (0.028) Real shadow federal funds 0.017 (0.029) Intermediate × real shadow rate –0.079* (0.046) Fixed × real shadow rate –0.002 (0.051) Lagged net capital flows/GDP –0.004 –0.004 –0.004 –0.008** –0.008** (0.004) (0.004) (0.004) (0.004) (0.003) Lagged institutional quality –0.055 0.124 –0.077 0.638 2.475 (1.225) (1.158) (1.085) (1.324) (1.494) Lagged private credit/GDP –0.014*** –0.013*** –0.013*** –0.012** –0.017** (0.004) (0.004) (0.004) (0.005) (0.007) Lagged real GDP per capita –2.326*** –2.374*** –2.565*** –2.416*** –2.549*** (0.711) (0.687) (0.742) (0.676) (0.771) Linear trend 0.014** 0.012** 0.014** (0.006) (0.006) (0.006) Global financial crisis –1.351*** –1.303*** –1.287*** (0.332) (0.332) (0.327) Quarter-year effects No No No Yes Yes Observations 2,103 2,103 2,071 2,103 1,635 Adjusted R 2 0.313 0.318 0.325 0.430 0.486 No. of countries 38 38 38 38 38 GDP = gross domestic product, US = United States. Notes: 1. This is identical to Table 9 in Obstfeld, Maurice, Jonathan Ostry, and Mahvash Qureshi. 2017. “A Tie That Binds: Revisiting the Trilemma in Emerging Market Economies.” IMF Working Paper WP/17/130. Washington, DC: International Monetary Fund. The one exception is that we rely on an alternative methodology of propensity score weighted regressions. 2. The dependent variable is the three-quarter moving average of quarterly (seasonally adjusted) real GDP growth rate (in percent) in emerging market economies. 3. VXO is the precursor of the Chicago Board Options Exchange’s Volatility Index or VIX. 4. Numbers in parentheses are clustered standard errors (by country). ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively. Source: Authors’ calculations. Exchange Rates and Insulation in Emerging Markets | 25 Table 6: Impact of the Federal Funds Rate on Various Variables in Emerging Market Economies Dependent Variables Explained Variables (1) (2) (3) (4) (5) [1] Real domestic credit growth Intermediate 1.6** 0.7 0.4 0.4 Fixed 3.1*** 2.7** 2.0* 2.0* FFR 0.2* 0.0 –0.1 Inter × FFR 0.4*** 0.5*** 0.2 Fix × FFR 0.2 0.4** 0.2 [2] Real house price growth Intermediate 1.3*** 1.8*** 1.9*** 1.7*** Fixed 1.0 1.3 1.2 1.2 FFR 0.1 0.2 0.2 Inter × FFR –0.2 –0.3 –0.3 Fix × FFR –0.1 –0.1 –0.2 [3] Real stock returns Intermediate 0.1 –0.2 0.4 3.5** Fixed –2.1 –1.3 –1.3 2.4 FFR –0.5** –0.5*** 0.2 Inter × FFR 0.2 0.1 –0.0 Fix × FFR –0.2 –0.2 –0.7** [4] Change in loan-todeposit ratio Intermediate 1.8** 1.2* 1.1 1.0 Fixed 4.5*** 3.7*** 3.4*** 2.6** FFR 0.2** 0.1 0.1 Inter × FFR 0.2* 0.3* 0.1 Fix × FFR 0.3 0.4* 0.1 [5] Net capital flows Intermediate 2.0*** 1.0 0.8 1.5 Fixed 1.5 0.5 0.5 1.2 FFR 0.4 0.2 0.1 Inter × FFR 0.3* 0.4* 0.1 Fix × FFR 0.4 0.4 0.3 [6] Liability flows Intermediate 1.4* –0.2 0.1 0.8 Fixed 2.7 1.3 1.6 2.6 FFR 0.9*** 0.6*** 0.8*** Inter × FFR 0.6* 0.4 0.3 Fix × FFR 0.5 0.4 0.4 [7] Asset flows Intermediate 0.6 1.2 0.7 0.6 Fixed –1.1 –0.7 –1.0 –1.3 FFR –0.5*** –0.4*** –0.7*** Inter × FFR –0.2 –0.0 –0.2 Fix × FFR –0.2 –0.0 –0.1 [8] Real GDP growth Intermediate 0.1 0.1 0.1 0.2 Fixed 0.4* 0.3 0.3 0.3 FFR 0.0 0.0 0.1 Inter × FFR –0.0 0.0 –0.1 Fix × FFR 0.0 0.0 –0.0 FFR = federal funds rate, GDP = gross domestic product. Notes: 1 Only the coefficients of intermediate and fixed exchange rate regimes, the federal funds rate, and their interactions are reported, estimated using propensity score weighted regressions and the same dependent variables as in Table 1, Tables 1.1–1.7, and 1.9, which are listed in the first column. 2. The results in column (4) are not reported because the real shadow funds rate is also included as a regressor. ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively, based on clustered standard errors (by country). Source: Author’s calculations. 26 | ADB Economics Working Paper Series No. 610 Table 7: Impact of the Change in Total Assets of the Federal Reserve on Various Variables in Emerging Market Economies Dependent Variables Explained Variables (1) (2) (3) (4) (5) [1] Real domestic credit growth Intermediate 1.0 1.0 1.0 1.2 Fixed 2.4** 2.7** 2.8** 2.8** ǻFedA/Y –2.6 34.7*** 31.8** Inter × ǻFedA/Y –34.7** –35.4** –41.9** Fix × ǻFedA/Y –82.8*** –85.5*** –78.6*** [2] Real house price growth Intermediate 1.4** 1.3* 1.3* 0.9 Fixed 0.8 1.1 1.1 0.7 ǻFedA/Y 4.8 36.0* 35.1* Inter × ǻFedA/Y –7.0 –9.0 15.5 Fix × ǻFedA/Y –75.7*** –77.1*** –45.6* [3] Real stock returns Intermediate 1.3 1.4 0.8 2.4* Fixed 0.9 1.1 0.2 2.1 ǻFedA/Y –128.3*** –102.0* –82.6 Inter × ǻFedA/Y –29.5 –14.0 –17.7 Fix × ǻFedA/Y –57.9 –33.0 –51.4 [4] Change in loan-todeposit ratio Intermediate 0.9 0.8 0.8 0.6 Fixed 2.3* 2.3* 2.5* 2.1* ǻFedA/Y –7.3 –9.6 –12.7 Inter × ǻFedA/Y 18.8 20.3 15.7 Fix × ǻFedA/Y –13.0 –14.9 –15.3 [5] Net capital flows Intermediate 2.1** 2.2** 2.3*** 2.6** Fixed 4.3** 4.6** 5.8*** 5.1** ǻFedA/Y 16.6 77.2* 80.0* Inter × ǻFedA/Y –74.1 –80.1* –83.9* Fix × ǻFedA/Y –86.0 –116.1* –92.4 [6] Liability flows Intermediate 1.0 0.7 0.9 2.1 Fixed 5.0** 4.9** 6.2*** 6.9*** ǻFedA/Y –19.6 –44.3 –43.1 Inter × ǻFedA/Y 59.7 47.2 44.8 Fix × ǻFedA/Y 6.4 –26.7 –3.6 [7] Asset flows Intermediate 1.1 1.5* 1.4 0.5 Fixed –0.5 –0.1 –0.2 –1.6 ǻFedA/Y 36.2*** 121.8*** 123.4*** Inter × ǻFedA/Y –133.5*** –127.2*** –128.1*** Fix × ǻFedA/Y –92.5* –89.6* –88.8* [8] Real GDP growth Intermediate 0.1 0.1 0.1 0.3** Fixed 0.1 0.2 0.2 0.4 ǻFedA/Y –4.8** 4.1 4.3 Inter × ǻFedA/Y –4.5 –3.9 –6.0 Fix × ǻFedA/Y –18.5* –18.7* –20.2* GDP = gross domestic product. Notes: 1. Only the coefficients of intermediate and fixed exchange rate regimes, ǻFedA/Y (change in total assets of the Federal Reserve as share of GDP), and their interactions are reported, estimated using propensity score weighted regressions and the same dependent variables as in Table 1, Tables 1.1–1.7, and 1.9, which are listed in the first column. 2. The results in column (4) are not reported because the real shadow funds rate is also included as a regressor. ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively, based on clustered standard errors (by country). Source: Author’s calculations. Exchange Rates and Insulation in Emerging Markets | 27 Table 10 examines the impact of an increase in the expected probability of a recession outside the US. This is analogous to the expected probability of a US recession in the coming quarter considered in Table 4. Again, the results are broadly consistent with our earlier findings. For example, domestic growth, the change in the loan-to-deposit ratio, and asset flows are equally sensitive in countries with intermediate and fixed exchange rate regimes. In sum, these new results that consider a broader array of measures and constituents of external volatility are consistent with earlier results; namely, that the exchange rate regime still has an effect in accentuating or dampening global volatility. But the results are less consistent with the strong conclusion that only fixed rates significantly diminish volatility relative to the freely floating alternative. Compared with this alternative, intermediate exchange rate regimes also accentuate the impact of real US shocks on a range of financial aggregates in emerging markets. This is also true of the impact of shocks not sourced from the US on not just financial variables in emerging markets but also, in at least one case, on real GDP growth. This is also true for the impact on at least some emerging market financial aggregates of shocks to the federal funds rate and to US unconventional monetary policy. In some of these cases, it appears that intermediate regimes perform even more poorly than fixed exchange rate regimes in providing insulation. One school of thought, going back to the mid-1990s, is particularly critical of intermediate regimes (Eichengreen 1994, 1999). Intermediate regimes, in this view, are more fragile than hard pegs and free floats, because the commitment of governments to limit the variability of the exchange rate is itself limited. Because of this, an external shock can destabilize the exchange rate regime in addition to destabilizing other domestic real or financial variables. An external shock can force the government and central bank to harden its soft peg or, more likely, to abandon that peg for a freer float. The collapse of the exchange rate regime, actual or anticipated (with some probability), may then further destabilize the other domestic real or financial variables of interest. Intermediate regimes were criticized on these very grounds after the Asian financial crisis—for example, when they showed this kind of fragility and, in the course of doing so, amplified the impact of external volatility on real and financial variables. This is one way of understanding our results suggesting that intermediate regimes, in some cases, have an outside impact on domestic economic and financial conditions. There are two caveats about this conclusion: First, it is not necessary to defend the strong form bipolar view (Fischer 2001) that pegged and freely floating exchange rates are always better than intermediate regimes to accept some of these conclusions. The point here is not that intermediate regimes are certain to die off or about to disappear, only that they can be fragile and are capable, in some circumstances, of amplifying external disturbances. Second, for some variables and in some specifications, the simple textbook hierarchy that pegged, intermediate, and floating rates provide declining levels of insulation from external disturbances continues to hold. In some cases, however, Obstfeld, Ostry, and Qureshi’s (2017) conclusion that mainly pegged rates diminish insulation survives these further robustness checks. 28 | ADB Economics Working Paper Series No. 610 Table 8: Impact of Non-United States Industrial Production Growth on Various Variables in Emerging Market Economies Dependent Variables Explained Variables (1) (2) (3) (4) (5) [1] Real domestic credit growth Intermediate 1.6* 1.5* 1.5* 0.5 0.9 Fixed 3.2*** 3.2*** 3.1** 3.0** 2.4** NUSIP 1.6 –0.8 –1.4 –1.3 Inter × NUSIP 5.7 5.9 1.5 9.5** Fix × NUSIP 0.6 2.3 1.2 3.1 [2] Real house price growth Intermediate 1.4*** 1.4*** 1.4*** 2.1*** 1.0* Fixed 0.9 1.3 1.3 1.8 1.0 NUSIP –8.8 –24.2** –24.8** –26.7** Inter × NUSIP 10.6 11.1 12.7 9.0 Fix × NUSIP 39.0** 39.3** 39.7** 37.0** [3] Real stock returns Intermediate 0.6 0.6 0.3 –0.1 2.6*** Fixed –1.6 –1.6 –2.1 –1.0 0.2 NUSIP –36.6*** –39.9** –31.8** –28.0 Inter × NUSIP 4.2 1.8 –4.9 7.8 Fix × NUSIP 6.1 4.7 6.4 12.3 [4] Change in loan-todeposit ratio Intermediate 1.7** 1.6** 1.6** 1.0 1.0 Fixed 4.5** 4.4** 4.4** 3.9*** 2.7** NUSIP 3.0 –5.5 –5.6 –7.9 Inter × NUSIP 12.5* 12.2 10.8 10.8* Fix × NUSIP 12.4 12.6 10.2 8.3 [5] Net capital flows Intermediate 1.8*** 1.7*** 1.7** 1.3 2.0** Fixed 1.2 1.1 1.1 1.0 1.9 NUSIP 1.9 –6.0 –6.8 –11.6* Inter × NUSIP 9.4 9.6 9.8 15.1* Fix × NUSIP 13.7 13.6 11.8 16.7 [6] Liability flows Intermediate 0.9 1.0 0.9 –0.5 1.9** Fixed 2.1 2.0 1.9 1.5 3.6* NUSIP 15.7** 18.5** 18.7** 8.1 Inter × NUSIP –12.1 –13.3 –13.9* 2.1 Fix × NUSIP 5.4 4.1 1.5 12.7 [7] Asset flows Intermediate 0.9 0.7 0.8 1.7 0.0 Fixed –0.8 –0.9 –0.8 –0.4 –1.6* NUSIP –13.9*** –24.6*** –25.6*** –19.9*** Inter × NUSIP 21.5*** 22.9*** 23.6*** 13.0 Fix × NUSIP 8.1 9.3 9.9 3.4 [8] Real GDP growth Intermediate 0.1 0.1 0.1 0.1 0.1 Fixed 0.4* 0.3 0.3 0.4 0.2 NUSIP 5.0*** 3.0** 3.5** 3.6*** Inter × NUSIP –0.1 –0.3 –0.4 –0.5 Fix × NUSIP 6.4* 6.3* 6.3* 6.2** GDP = gross domestic product, NUSIP = non-United States industrial production. Notes: Only the coefficients of intermediate and fixed exchange rate regimes, non-US industrial production growth, and their interactions are reported, estimated using propensity score weighted regressions and the same dependent variables as in Table 1, Tables 1.1–1.7, and 1.9, which are listed in the first column. ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively, based on clustered standard errors (by country). Source: Author’s calculations.  Exchange Rates and Insulation in Emerging Markets | 29 Table 9: Impact of the Global Economic Policy Uncertainty Index on Various Variables in Emerging Market Economies Dependent Variables Explained Variables (1) (2) (3) (4) (5) [1] Real domestic credit growth Intermediate 0.8 5.4 4.9 2.6 4.7 Fixed 1.6* 14.0** 16.0** 25.4*** 14.5** GEPU –2.1*** –1.0 –0.8 –0.2 Inter × GEPU –1.0 –0.9 –0.5 –0.9 Fix × GEPU –2.7* –3.2** –4.9*** –2.8** [2] Real house price growth Intermediate 0.9** –6.6 –6.7 –7.3** –12.0 Fixed 0.7 10.3** 10.9** 11.5 3.8 GEPU –1.9*** –1.7*** –1.6*** –1.8*** Inter × GEPU 1.7 1.7 1.8*** 2.9 Fix × GEPU –2.0** –2.1** –2.3 –0.6 [3] Real stock returns Intermediate 1.4* 8.1 7.8 7.5 7.2 Fixed –1.2 –7.5 –6.5 0.9 –10.3 GEPU –9.8*** –9.8*** –10.2*** –15.1*** Inter × GEPU –1.4 –1.3 –1.3 –1.0 Fix × GEPU 1.4 1.1 –0.3 2.3 [4] Change in loan-todeposit ratio Intermediate 0.4 4.8 4.3 3.9 4.4 Fixed 1.2 8.0 8.7* 14.5** 8.5* GEPU –0.7 0.0 0.1 1.0 Inter × GEPU –1.0 –0.9 –0.8 –0.9 Fix × GEPU –1.5 –1.7* –2.8** –1.6* [5] Net capital flows Intermediate 1.9*** 17.6** 17.5** 15.1** 14.9* Fixed 1.3 28.9* 28.5* 35.1*** 28.1* GEPU –2.7** 0.2 0.2 0.8 Inter × GEPU –3.5** –3.5** –3.0* –2.8 Fix × GEPU –6.0* –5.9* –7.2*** –5.7* [6] Liability flows Intermediate 1.3** 14.6 11.9 5.1 13.7 Fixed 2.8* 33.2** 31.6** 37.5*** 34.4* GEPU –6.5*** –3.5*** –3.8*** –2.6* Inter × GEPU –3.0 –2.4 –1.1 –2.6 Fix × GEPU –6.6* –6.2* –7.4** –6.6* [7] Asset flows Intermediate 0.6 3.5 6.1 10.4* 1.5 Fixed –1.4 –4.0 –2.8 –2.0 –6.1 GEPU 3.7*** 3.7*** 4.0*** 3.4*** Inter × GEPU –0.6 –1.2 –1.9 –0.3 Fix × GEPU 0.6 0.3 0.2 0.9 [8] Real GDP growth Intermediate 0.0 0.1 0.3 1.5 0.2 Fixed 0.2 1.6 1.6 0.9 1.8 GEPU –0.8*** –0.7*** –0.7*** –0.7*** Inter × GEPU –0.0 –0.0 –0.3 –0.0 Fix × GEPU –0.3 –0.3 –0.1 –0.3 GDP = gross domestic product, GEPU = Global Economic Policy Uncertainty Index. Notes: Only the coefficients of intermediate and fixed exchange rate regimes, the Global Economic Policy Uncertainty Index, and their interactions are reported, estimated using propensity score weighted regressions and the same dependent variables as in Table 1, Tables 1.1–1.7, and 1.9, which are listed in the first column. ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively, based on clustered standard errors (by country). Source: Author’s calculations. 30 | ADB Economics Working Paper Series No. 610 Table 10: Impact of the Expected Probability of Recessions Outside the United States on Various Variables in Emerging Market Economies Dependent Variables Explained Variables (1) (2) (3) (4) (5) [1] Real domestic credit growth Intermediate 1.56* 1.87** 1.86** 0.71 1.46* Fixed 3.13** 3.49*** 3.41** 3.27** 2.86*** EPROUS –0.96 0.53 0.40 0.53 Inter × EPROUS –1.86* –2.11* –2.08* –2.26* Fix × EPROUS –2.38* –2.13 –2.55* –2.26* [2] Real house price growth Intermediate 1.41*** 1.92** 1.93** 1.62*** 1.17 Fixed 1.26 3.18** 3.37** 2.31 2.36* EPROUS –1.60 1.98 2.09 1.99 Inter × EPROUS –1.74 –2.03 –1.79 0.65 Fix × EPROUS –6.71** –7.09** –6.81** –5.02 [3] Real stock returns Intermediate –1.00 –1.64 –1.83 –1.39 2.45** Fixed –4.31** –3.95** –4.13** –3.94** 0.73 EPROUS –14.99*** –15.52*** –14.51*** –15.14*** Inter × EPROUS 3.64 4.52 3.61 1.20 Fix × EPROUS –2.72 –1.64 –2.78 –4.19 [4] Change in loan-todeposit ratio Intermediate 1.82** 2.14** 2.15** 1.21 1.63** Fixed 4.51** 4.75*** 4.75*** 3.92*** 3.12** EPROUS 0.48 1.59*** 1.72*** 1.54*** Inter × EPROUS –1.94** –2.28** –2.15** –2.65*** Fix × EPROUS –1.51*** –1.69** –1.69*** –1.75*** [5] Net capital flows Intermediate 1.77*** 1.82** 1.81** 1.21 2.65*** Fixed 1.21 1.23 1.25 0.70 2.38 EPROUS –0.45 –0.30 –0.41 –0.41 Inter × EPROUS –0.28 –0.44 –0.60 –2.33 Fix × EPROUS –0.19 –0.52 –0.29 –1.36 [6] Liability flows Intermediate 0.61 –0.02 –0.02 –0.80 1.74 Fixed 1.70 1.34 1.38 0.81 3.60* EPROUS –1.79 –3.68** –3.99*** –4.47*** Inter × EPROUS 3.56** 3.24* 3.26** 1.37 Fix × EPROUS 2.40 2.06 2.39 1.29 [7] Asset flows Intermediate 1.16* 1.83** 1.81** 1.99* 0.89 Fixed –0.43 –0.02 –0.06 –0.00 –1.13 EPROUS 1.34** 3.38*** 3.60*** 4.06*** Inter × EPROUS –3.82*** –3.66*** –3.84*** –3.67*** Fix × EPROUS –2.60* –2.58* –2.67* –2.65* [8] Real GDP growth Intermediate 0.04 0.03 0.03 0.08 0.07 Fixed 0.28 0.28 0.27 0.20 0.30 EPROUS –0.61*** –0.62*** –0.65*** –0.66*** Inter × EPROUS 0.02 0.13 0.02 –0.00 Fix × EPROUS –0.00 0.06 –0.04 –0.12 EPROUS = expected probability of recessions outside the United States, GDP =gross domestic product. Notes: 1. Only the coefficients of intermediate and fixed exchange rate regimes, the expected probability of recessions outside of the United States, and their interactions are reported, estimated using propensity score weighted regressions and the same dependent variables as in Table 1, Tables 1.1–1.7, and 1.9, which are listed in the first column. 2. The expected probability of recessions outside of the United States within the next 12 months is calculated by Federal Reserve Board staff using excess bond premium and the global condition index. The authors thank Andrea Raffo for providing the estimates. ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels, respectively, based on clustered standard errors by country). Source: Author’s calculations. Exchange Rates and Insulation in Emerging Markets | 31 V. CONCLUSION Recent literature on the global financial cycle, the role of the dollar in bank funding markets, and the role of dollar invoicing and settlements have raised anew old questions about the insulating properties of flexible exchange rates in general and for emerging markets in particular. Some studies that respond to this question marshal evidence that more flexible exchange rates do in fact provide insulation—and that countries with pegged exchange rates experience more domestic volatility in response to global shocks. The principal these studies produce also leads to a further strong conclusion: that countries sacrifice the insulating properties of flexible exchange rates only when they move all the way to the other extreme—that is, to a currency peg—and not when they move only part way to intermediate regimes of managed flexibility. Clearly, the stakes are high, for emerging markets in general and Asian ones in particular. Global volatility is a problem for these countries. Whether the exchange rate regime can be tailored to limit the domestic impact of that volatility is an important policy question. The subsidiary question then becomes how exactly the exchange rate regime should be tailored. Asian countries have extensive experience with global volatility and its effects. And they have received much advice, in the last 2 decades in particular, about how to tailor their exchange rate regimes. Our empirical analysis revisits these questions, building directly on the recent literature, but using alternative measures of global volatility and alternative empirical methods. We come down on the side that the exchange rate regime matters, and that more flexible exchange rates provide emerging markets with more insulation from global volatility shocks. Here, our results are consistent with the conventional wisdom and with the findings of Obstfeld, Ostry, and Qureshi (2017), and inconsistent with the theory of the global financial cycle and related analyses that deny the existence of those insulating properties. Our results provide some support for the notion that both fixed and intermediate exchange rate regimes diminish insulation relative to the alternative of fully flexible rates. In some cases, intermediate regimes accentuate the impact of global volatility on domestic financial variables as much as pegged regimes. According to Obstfeld, Ostry, and Qureshi (2017), limited flexibility was enough if policy makers want to avoid accentuating the impact of global volatility on domestic conditions through their choice of exchange rate regime. But our findings suggest this is not always the case. To this end, we would also like to emphasize that despite the insulation properties of flexible exchange rate regimes, adopting them must be strongly supported by a domestic economic environment that is conducive for relatively stable currency movements. These are important for promoting healthy domestic credit conditions, asset price development, capital flows, and GDP growth. So, while a stable exchange rate environment promotes the development of domestic financial conditions, its potential to move flexibly in responding to external shocks will help an economy to insulate the unwanted effects of these shocks. APPENDIXES Table A.1: Description and Sources for the Variables Used by Obstfeld, Ostry, and Qureshi Variables Description Source Capital account openness Index (high = liberalized, low = closed) Quinn and Toyoda (2008) Capital flows In billions of dollars (BPM5 presentation). Net financial flows exclude financing items and other investment liabilities of general government; i.e., the difference between the International Monetary Fund’s International Financial Statistics (IFS) series codes “… 4995W.9” and “… 4753ZB9.” Liability flows and other investment liability flows also exclude other investment liabilities of the general government IFS database Consumer price index (CPI) Index IFS database Domestic private sector credit In local currency IFS database Exchange rate regime De facto, de jure Ghosh, Ostry, and Qureshi (2015) and Reinhart and Rogoff (2004) updated data from http://personal.lse.ac.uk/ilzetzki/ index.htm/Data.htm. Gross domestic product current and constant prices In billions of dollars (or local currency); seasonally adjusted observations for quarterly data Haver Analytics, CEIC, and IFS database House prices Index (in real terms) Bank for International Settlements Institutional quality Index (average of International Country Risk Guide’s [ICRG] 12 political risk components) Political Risk Group Loan-to-deposit ratio In percent IFS database Policy rate Policy rate or discount rate (in percent) IFS database Reserve requirements Average of reserve requirements on local currency demand, saving, and term deposits (in percent) Authors' calculations based on data from Federico et al. (2014) Shadow federal funds rate In percent. In real terms computed as [(1+nominal interest rate)/(1+expected inflation)]–1, where expected inflation is one-period ahead inflation Federal Reserve Economic Data Stock prices (in real terms) Stock price index deflated by quarterly CPI CEIC and author's calculations United States (US) interest rate US 3-month Treasury bill rate and 10-year government bond yield (in percent) IFS database and Bloomberg VXO and VIX Indexes Chicago Board Options Exchange Market Volatility Index Bloomberg Notes: 1. Description and source of all variables not shaded are identical to those in Obstfeld, Ostry, and Qureshi (2017). Since proprietary data series are not provided by these authors, we have collected these data. 2. Except for institutional quality, proxied by the ICRG, which is collected from the same source as that in Obstfeld, Ostry, and Qureshi (2017), other data denoted by shade are collected from different sources. 3. Although CPI and the shadow federal funds rate are not proprietary data, since they were not provided by Obstfeld, Ostry, and Qureshi, the compilers of this table collected them from publicly available sources. Sources: Authors’ compilation based on Federico, Pablo, Carlos Vegh, and Guillermo Vuletin. 2014. “Reserve Requirements over the Business Cycle.” NBER Working Paper No. 20612. Cambridge, MA: National Bureau of Economic Research; Ghosh, Atish, Jonathan Ostry, and Mahvesh Qureshi. 2015. “Exchange Rate Management and Crisis Susceptibility: A Reassessment.” IMF Economic Review 63: 238–76; Obstfeld, Maurice, Jonathan Ostry, and Mahvash Qureshi. 2017. “A Tie That Binds: Revisiting the Trilemma in Emerging Market Economies.” IMF Working Paper WP/17/130. Washington, DC: International Monetary Fund; Reinhart, Carmen, and Kenneth Rogoff. 2004. “The Modern History of Exchange Rate Arrangements: A Reinterpretation.” Quarterly Journal of Economics 199: 1–48; Quinn, Dennis, and Maria Toyoda. 2008. “Does Capital Account Liberalization Lead to Economic Growth?” Review of Financial Studies 21 (3): 1403–49. 34 | Appendixes Table A.2: Description and Sources for the Variables Used by Londono and Wilson Variables Description Source Londono and Wilson index Market-value-weighted average of the equity optionimplied volatility for seven countries: France, Germany, Japan, the Netherlands, Switzerland, the United Kingdom, and the United States (US) Chicago Board Options Exchange; Datastream; Osaka University, Center for the Study of Finance and Insurance US Industrial Production Index Industrial Production Index, index 2012 = 100, quarterly average of monthly data, seasonally adjusted Federal Reserve Economic Data Expected probability of a US recession Probability that real gross national product/GDP will decline (quarter over quarter) in the following quarter the survey was conducted Survey of Professional Forecasters US Economic Policy Uncertainty Index Economic Policy Uncertainty Index, seasonally adjusted, following definition in Baker, Bloom, and Davis (2016) Federal Reserve Economic Data Federal funds rate Fed funds rate when this rate is above zero and the shadow fed funds rate (Wu and Xia 2016) in the period when the zero-lower bound is binding Normal time rate: Federal Reserve Economic Data; shadow rate: Wu and Xia (2016) Change in Federal Reserve assets as a share of US gross domestic product (GDP) Change in total assets of all Federal Reserve Banks (quarterly average data) divided by GDP (annual data), seasonally adjusted Total Federal Reserve assets: Federal Reserve Economic Data; GDP: US Bureau of Economic Analysis, retrieved from Federal Reserve Economic Data Non-US industrial production growth Industrial production, seasonally adjusted, index Average of six countries: France, Germany, Japan, the Netherlands, Switzerland, and the United Kingdom International Monetary Fund’s (IMF) International Financial Statistics database; Federal Reserve Economic Data Global Economic Policy Uncertainty Index GDP-weighted average of the 18 national Economic Policy Uncertainty Index values, using GDP data from the World Economic Outlook database of the IMF Global Economic Policy Uncertainty Index (www.PolicyUncertainty.com) Expected probability of recessions outside the US The expected probability of recessions outside the US within the next 12 months, calculated by the Federal Reserve Board staff using excess bond premium and Global Conditions Index Cuba-Borda, Mechanik, and Raffo (2018) Sources: Authors’ compilation based on Londono, Juan, and Beth Anne Wilson. 2018. “Understanding Global Volatility.” IFDP Notes. Washington, DC: Board of Governors of the Federal Reserve System; Baker, Scott R., Nicholas Bloom, and Steven J. Davis. 2016. “Measuring Economic Policy Uncertainty.” Quarterly Journal of Economics 131 (4): 1593–636; Cuba-Borda, Pablo, Alexander Mechanik, and Andrea Raffo. 2018. “Monitoring the World Economy: A Global Condition Index.” IFDP Notes. Washington, DC: Board of Governors of the Federal Reserve System; Wu, Jing Cynthia, and Fan Dora Xia. 2016. “Measuring the Macroeconomic Impact of Monetary Policy at the Zero Lower Bound.” Journal of Money, Credit and Banking 48 (2–3): 253–91. REFERENCES Accominotti, Olivier, and Barry Eichengreen. 2016. “The Mother of All Sudden Stops: Capital Flows and Reversals, in Europe, 1919–32.” Economic History Review 69 (2): 469–92. Aizenman, Joshua, Menzie Chinn, and Hiro Ito. 2015. “Monetary Policy Spillovers and the Trilemma in the New Normal: Periphery Country Sensitivity to Core Country Conditions.” NBER Working Paper No. 21128. Cambridge, MA: National Bureau of Economic Research. Baker, Scott R., Nicholas Bloom, and Steven J. Davis. 2016. “Measuring Economic Policy Uncertainty.” Quarterly Journal of Economics 131 (4): 1593–636. Bank for International Settlements. https://www.bis.org/ (accessed 18 August 2018). Baxter, Marianne, and Alan Stockman. 1989. “Business Cycles and the Exchange-Rate Regime: Some International Evidence.” Journal of Monetary Economics 23 (3): 377–400. Bekaert, Geert, and Arnaud Mehl. 2017. “On the Global Financial Market Integration ‘Swoosh’ and the Trilemma.” NBER Working Paper No. 23124. Cambridge, MA: National Bureau of Economic Research. Bluedorn, John C., and Christopher Bowdler. 2010. “The Empirics of International Monetary Transmission: Identification and the Impossible Trinity.” Journal of Money, Credit and Banking 42 (4): 679–713. Caceres, Carlos, Yan Carriere-Swallow, and Bertrand Gruss. 2016. “Global Financial Conditions and Monetary Policy Autonomy.” IMF Working Paper WP/16/108. Washington, DC: International Monetary Fund. Choudhri, Ehsan, and Levis Kochin. 1980. “The Exchange Rate and the International Transmission of Business Cycle Disturbances: Some Evidence from the Great Depression.” Journal of Money, Credit and Banking 12 (4): 565–74. Cuba-Borda, Pablo, Alexander Mechanik, and Andrea Raffo. 2018. “Monitoring the World Economy: A Global Condition Index.” IFDP Notes. Washington, DC: Board of Governors of the Federal Reserve System. Cushman, David, and Glauco De Vita. 2017. “Exchange Rate Regimes and FDI in Developing Countries: A Propensity Score Matching Approach.” Journal of International Money and Finance 77 (C): 143–63. Edwards, Sebastian. 2015. “Monetary Policy Independence under Flexible Exchange Rates: An Illusion?” World Economy 38 (5): 773–87. Eichengreen, Barry. 1992. “The Comparative Performance of Fixed and Flexible Exchange Rate Regimes: Interwar Evidence.” In Business Cycles, Theories, Evidence and Analysis, edited by Niels Thygesen, Kumaraswamy Velupillai, and Stefano Zambelli, 229–72. New York: St. Martin’s Press.