Skip to main content
Log in

The Impact of US Financial Uncertainty Shocks on Emerging Market Economies: An International Credit Channel

  • Research Article
  • Published:
Open Economies Review Aims and scope Submit manuscript


I document that US financial uncertainty shocks, measured by an increase in VIX, have a substantial impact on the output of emerging market economies (EMEs) without a material impact on US output during the last two decades. To understand this puzzling phenomenon, I propose a credit channel as a propagation mechanism of US financial uncertainty shocks to EMEs. I augment a boom-bust cycle model of EMEs by Schneider and Tornell (Rev Econ Stud 71(3):883–913 2004) with a portfolio choice model of constrained international investors. As international investors pull their money from EMEs—to satisfy their Value-at-Risk constraints—in response to financial uncertainty shocks, borrowing costs increase and domestic credit contracts. Higher borrowing costs and a decline in domestic credit, in turn, lead to a fall in investment in the non-tradable sector that causes a real depreciation via currency mismatch prevalent in EMEs and a decline in total output through sectoral linkages. The empirical regularity obtained by estimating structural VARs of 18 EMEs is consistent with the prediction of the model.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10

Similar content being viewed by others


  1. VIX is a measure of market expectations of near-term volatility, as implied by S &P 500 stock index option prices, which becomes a standard measure of uncertainty in financial markets. Here, I distinguish financial uncertainty from uncertainty regarding other dimensions of the economy, such as economic policy uncertainty by Baker et al. (2016) and focus on the former. The recent episodes of the Brexit and the US presidential election demonstrate how empirical proxies for each uncertainty can diverge dramatically from one another.

  2. Seeking the most parsimonious means of representation, the VAR model used here only includes the level of VIX and the log level of industrial production without de-trending, as described in Bachmann et al. (2013). VIX is ordered first in a recursive identification, and the VARs are estimated with six lags. However, the insignificant impact of financial uncertainty shocks on US output found in this paper is not driven by different identifying assumptions from an influential paper by Bloom (2009). By using the same identifying assumptions from Bloom (2009) and Choi (2013) finds that the impact of (financial) uncertainty shocks on US output and employment has substantially decreased since the mid-1980s. I still obtain a strong impact of an increase in VIX on US output from estimating the same bivariate VARs for an earlier period. See Section A.3.1 in the Online Appendix for further evidence.

  3. I thank to the anonymous referee for his suggestions on linking the risk aversion interpretation of VIX in this paper to Bekaert et al. (2013).

  4. This setup is intended to capture a common practice in EMEs where banks borrow in dollars and lend in local currencies (e.g., pesos).

  5. While the supply of non-tradable goods is predetermined—i.e., fixed in the current period—the demand for them decreases. To clear the market, the price of non-tradable goods falls, which leads to real depreciation (because the price of tradable goods is internationally given).

  6. Real domestic lending rates are measured by the difference between nominal domestic bank lending rates and expected inflation rates. See Section A.1.1 in the Online Appendix for how to measure expected inflation rates.

  7. I do not specify the problems of consumers (denoted by dashed circles in Fig. 2) in the model. As the economy is small and open, the destination of tradable goods is not important for the main implication of the model.

    Fig. 2
    figure 2

    Flowchart of the model

  8. The zero productivity in the bad state is just for simplicity, and any productivity lower than a would work as well.

  9. A large enough productivity shift in the final period induces a substantial real appreciation, encouraging non-tradable sector firms to take more debt along the equilibrium path.

  10. The bond price q t should be lower than \(\frac {1-u}{R}\) because international investors are concerned about the trade-off between the returns and risk of the portfolio and these bonds bear default risk. Otherwise, international investors never purchase these bonds in the presence of safe assets, which guarantee the gross risk-free rate of R. If international investors are risk neutral, then \(q_{t}=\frac {1-u}{R}\). Therefore, \(\frac {1-u}{R}-q_{t}\) denotes risk premium.

  11. 11Note that i t is a sufficient statistic for the expected returns and the variance of emerging market bonds due tothe property of independently, identically distributed (i.i.d.) binomial distribution of aggregateproductivity shocks.

  12. Positive correlations of asset returns are based on ample empirical evidence. If the correlations of asset returns are negative, then the VaR constraint is not necessary for the main result, and the simple mean-variance maximization will also deliver a reduced demand for emerging market bonds when the conditional variance of US stock returns increases.

  13. This is because R < 1 + i t , and \(E_{t}[p_{t + 1}](1+{i^{D}_{t}})= 1+i_{t}\) in any equilibrium. If h is large enough that the condition h ≥ 1 + i t is satisfied, then diversion becomes more expensive than the repayment, and the diversion costs have no effect on the lending decisions of domestic banks. I assume that this is the case for advanced economies, so borrowing constraints do not arise in advanced economies.

  14. Note that the representative investor’s expected returns of investing in emerging market bonds equal to the external cost of borrowing by domestic banks.

  15. The distinction between tradable and non-tradable goods is also key to understand real exchange rate fluctuations (Rabanal and Tuesta 2013).

  16. See Adrian and Shin (2014) for more realistic setups for studying the asset market implication and the microfoundations of the VaR constraint.

  17. As an alternative explanation, Gourio (2012) constructs a model in which VIX is driven by a disaster probability.

  18. The cross-country empirical analysis in Section 5.5 describes the role played by country-level credit market imperfections in explaining the impact of US financial uncertainty shocks on a domestic credit market.

  19. For the same reason, n t does not depend on the realization of a t (\(n_{t}=\tilde {n}_{t}\) for every t).

  20. As the number of variables (up to six) in the structural VARs is relatively smaller than the length of time-series data (over 200 periods), a panel VAR model is not considered to allow heterogenous dynamics of US financial uncertainty shocks.

  21. Relaxing the small open-economy assumption and letting the data free to speak regarding this assumption do not change the main results.

  22. Once the level of the US stock market is placed before VIX to control for the wealth effect, the impact on US output is no longer negative even in the short run. See Figure 6 in Choi (2013) for a similar finding.

  23. Although spreads from J.P. Morgan’s Emerging Markets Bond Index Plus (EMBI + ), corresponding to i t r t in the model, are often used to measure country spreads in EMEs (for example, Neumeyer and Perri 2005; Uribe and Yue 2006; Akıncı 2013), I do not use them for three reasons: (i) EMBI + index is based only on the spreads of dollar denominated sovereign bonds, which cannot capture the currency mismatch behavior in the private sector highlighted in the model. (ii) the EMBI+ index does not cover as many countries as bank lending rates, and (iii) it has substantially different starting and ending dates across countries, preventing a meaningful cross-country comparison. For example, the EMBI+ index for South Africa has only been available since 2002, and the index for Korea was no longer available after 2006. Nevertheless, Akıncı (2013) finds, from structural VARs on 6 EMEs, a similar degree of increase (0.3%) in the EMBI+ spreads after a one standard deviation increase in VIX.

  24. 24Bloom (2009)—further extended by Choi (2013)—identifies 17 exogenous events that led a spike in VIX since 1962. Among seven exogenous events during the sample period in this analysis, only one event (Asian Financial Crisis) is directly driven by EMEs, suggesting the implausibility of the reverse causality (fluctuations in VIX are driven by EME business cycles.

    Fig. 3
    figure 3

    The evolution of VIX, Notes: This figure plots the evolution of VIX between January 1994 and December 2013

  25. A recent study by Caldara et al. (2016) provide another explanation on the insignificant impact of uncertainty shocks on US output. They purge uncertainty shocks—also measured by VIX—from financial shocks—measured by the excess bond premium—and find that the effect of uncertainty shocks on US output is significant only if it is transmitted through a financial channel.

  26. This finding is consistent with Chudik and Fratzscher (2011), who state that EMEs have been more strongly affected by risk-appetite shocks—measured by VIX—than advanced economies during the global financial crisis.

    Fig. 5
    figure 5

    Historic decomposition of output during the Great Recession, Notes: This figure plots the contribution of US financial uncertainty shocks to changes in industrial production during the Great Recession for the US (left) and Korea (right). The solid lines indicate actual data; the blue bars indicate the simulated fluctuations in industrial production when all shocks except US financial uncertainty shocks (US output shocks and US financial uncertainty shocks for Korea) are turned on; and the red bars indicate the simulated fluctuations in industrial production conditional on the estimated US financial uncertainty shocks alone

  27. I do not plot confidence intervals for the average and the median response because US financial uncertainty shocks are not i.i.d. across countries. Common shocks to all the countries result in correlated error among countries, preventing a straightforward estimation of standard errors. See Carrière-Swallow and Céspedes (2013) for an alternative representation under similar circumstances.

    Fig. 6
    figure 6

    Responses to US financial uncertainty shocks: 18 EMEs, Notes: This figure plots the average (blue) and the median (red) responses of three macroeconomic variables (country borrowing spreads, real effective exchange rates, and industrial production) from the 18 EMEs in the sample and the corresponding interquartile range to a one standard deviation increase in VIX

  28. Arguably, the decline in EME output following US financial uncertainty shocks is rather large relative to their impact on financial market variables. This could be driven by omitted factors affecting both US financial uncertainty and EME output through other channels. Commodity prices are certainly such a factor, as a few countries in the sample are a commodity exporter and commodity prices are negatively correlated with VIX. To check this possibility, I re-estimate the baseline model by including the world commodity price and the US stock market index in the exogenous block. Section A.3.2 in the Online Appendix shows that this is indeed a case without affecting the conclusion of the paper. I appreciate an anonymous referee for pointing out potential omitted factors.

  29. Individual IRFs are available in the Online Appendix Section A.3.4.

    Fig. 9
    figure 9

    Response of the 11 EMEs to US financial uncertainty shocks in the extended VARs, Notes: This figure plots the average (blue), the median (red), and the interquartile range (shaded area) of the response of four domestic variables (country borrowing spreads, domestic credit, real effective exchange rates, and industrial production) in the 11 selected EMEs to a one standard deviation increase in VIX


  • Adrian T, Shin HS (2014) Procyclical leverage and value-at-risk. Rev Financ Stud 27(2):373–403

    Article  Google Scholar 

  • Akıncı Ö (2013) Global financial conditions, country spreads and macroeconomic fluctuations in emerging countries. J Int Econ 91(2):358–371

    Article  Google Scholar 

  • Bachmann R, Elstner S, Sims ER (2013) Uncertainty and economic activity: evidence from business survey data. Am Econ J Macroecon 5(2):217–249

    Article  Google Scholar 

  • Baker SR, Bloom N, Davis SJ (2016) Measuring economic policy uncertainty. Q J Econ 131(4):1593–1636

    Article  Google Scholar 

  • Bekaert G, Hoerova M, Duca ML (2013) Risk, uncertainty and monetary policy. J Monet Econ 60(7):771–788

    Article  Google Scholar 

  • Bloom N (2009) The impact of uncertainty shocks. Econometrica 77(3):623–685

    Article  Google Scholar 

  • Borio C, Zhu H (2012) Capital regulation, risk-taking and monetary policy: a missing link in the transmission mechanism? J Financ Stab 8(4):236–251

    Article  Google Scholar 

  • Caballero RJ, Krishnamurthy A (2001) International and domestic collateral constraints in a model of emerging market crises. J Monet Econ 48(3):513–548

    Article  Google Scholar 

  • Caldara D, Fuentes-Albero C, Gilchrist S, Zakrajšek E (2016) The macroeconomic impact of financial and uncertainty shocks. Eur Econ Rev 88:185–207

    Article  Google Scholar 

  • Carrière-Swallow Y, Céspedes LF (2013) The impact of uncertainty shocks in emerging economies. J Int Econ 90(2):316–325

    Article  Google Scholar 

  • Choi S (2013) Are the effects of Bloom’s uncertainty shocks robust? Econ Lett 119(2):216–220

    Article  Google Scholar 

  • Choi S, Loungani P (2015) Uncertainty and unemployment: the effects of aggregate and sectoral channels. J Macroecon 46(C):344–358

    Article  Google Scholar 

  • Choi S, Furceri D, Huang Y, Loungani P (Forthcoming) Aggregate uncertainty and sectoral productivity growth: the role of credit constraints. J Int Money Financ. 10.1016/j.jimonfin.2017.07.016 10.1016/j.jimonfin.2017.07.016

  • Christiano LJ, Motto R, Rostagno M (2014) Risk shocks. Am Econ Rev 104(1):27–65

    Article  Google Scholar 

  • Chudik A, Fratzscher M (2011) Identifying the global transmission of the 2007–2009 financial crisis in a gvar model. Eur Econ Rev 55(3):325–339

    Article  Google Scholar 

  • De Bock R, de Carvalho Filho I (2015) The behavior of currencies during risk-off episodes. J Int Money Financ 53:218–234

    Article  Google Scholar 

  • Djankov S, Hart O, McLiesh C, Shleifer A (2008) Debt enforcement around the world. J Polit Econ 116(6):1105–1149

    Article  Google Scholar 

  • Fratzscher M (2012) Capital flows, push versus pull factors and the global financial crisis. J Int Econ 88(2):341–356

    Article  Google Scholar 

  • González-Rozada M, Yeyati EL (2008) Global factors and emerging market spreads. Econ J 118(533):1917–1936

    Article  Google Scholar 

  • Gourio F (2012) Disaster risk and business cycles. Am Econ Rev 102(6):2734–2766

    Article  Google Scholar 

  • Kaminsky GL, Reinhart CM (2000) On crises, contagion, and confusion. J Int Econ 51(1):145–168

    Article  Google Scholar 

  • Kyle AS, Xiong W (2001) Contagion as a wealth effect. J Financ 56 (4):1401–1440

    Article  Google Scholar 

  • Matsumoto A (2011) Global liquidity: availability of funds for safe and risky assets. In: IMF Working Papers, pp 1–36

  • Neumeyer PA, Perri F (2005) Business cycles in emerging economies: the role of interest rates. J Monet Econ 52(2):345–380

    Article  Google Scholar 

  • Pavlova A, Rigobon R (2008) The role of portfolio constraints in the international propagation of shocks. Rev Econ Stud 75(4):1215–1256

    Article  Google Scholar 

  • Rabanal P, Tuesta V (2013) Nontradable goods and the real exchange rate. Open Econ Rev 24(3):495–535

    Article  Google Scholar 

  • Rey H (2015) Dilemma not trilemma: the global financial cycle and monetary policy independence. NBER Working Papers

  • Schinasi GJ, Smith RT (2000) Portfolio diversification, leverage, and financial contagion. IMF Staff Pap 47(2):159

    Google Scholar 

  • Schneider M, Tornell A (2004) Balance sheet effects, bailout guarantees and financial crises. Rev Econ Stud 71(3):883–913

    Article  Google Scholar 

  • Tornell A, Westermann F (2002) The credit channel in middle income countries. NBER Working Paper

  • Uribe M, Yue VZ (2006) Country spreads and emerging countries: who drives whom? J Int Econ 69(1):6–36

    Article  Google Scholar 

  • Yeyati EL (2006) Financial dollarization: evaluating the consequences. Econ Policy 21(45):61–118

    Article  Google Scholar 

Download references


I am deeply indebted to Aaron Tornell and Roger Farmer for their invaluable guidance and unwavering supports. I am thankful to Jinyong Hahn, Prakash Loungani, Andy Atkeson, Pablo Fajgelbaum, Zhipeng Liao, Francois Geerolf, Davide Furceri, Young Ju Kim, Jaeok Park, Kyle Herkenhoff, Yan Carrière–Swallow, Myungkyu Shim, and Jongho Park for insightful discussions, and participants at the Bank of Canada, International Monetary Fund, Korea Development Institute, Korea Institute International Economic Policy, Saint Louis University, SUNY Buffalo, and UCLA for helpful comments. I also have benefited from comments and suggestions by the associate editor (Apostolos Serletis) and one anonymous referee. All remaining errors are mine.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Sangyup Choi.

Additional information

An earlier version of this paper was circulated under the title “The Impact of VIX Shocks on Emerging Market Economies: AFlight to Quality Mechanism.”

Electronic supplementary material

Below is the link to the electronic supplementary material.

(PDF 0.98 MB)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Choi, S. The Impact of US Financial Uncertainty Shocks on Emerging Market Economies: An International Credit Channel. Open Econ Rev 29, 89–118 (2018).

Download citation

  • Published:

  • Issue Date:

  • DOI:


JEL Classification