Quantitative easing (QE), which started in 2008, swelled the Federal Reserve’s balance sheet to an unprecedented $3.4 trillion. In May 2013, the Fed announced that it would evaluate the possibility of a reversal of its unconventional monetary policies – QE in particular .
The event, which has come to be known as ‘tapering’, prompted a sharp, negative response from financial markets (the so-called ‘taper tantrum’):
- US long rates rose by almost one percentage point between late May and August, and
- The concomitant rebalancing of global portfolios away from emerging-market assets resulted in capital outflows and currency depreciations in several large emerging-market countries.
Brazil, India, Indonesia, South Africa, and Turkey were particularly affected.
Surprised by the strength of the market response – and further bolstered by somewhat tepid labour market data – the Fed held back on actual tapering action over the course of the rest of the year. In the interim, it pressed on with conditioning market expectations for an eventual slowdown in large-scale asset purchases. The long-awaited taper eventually began in early January 2014.
Figure 1. Capital inflows to developing countries
Source: World Bank staff calculations, from IMF Balance of Payments and BIS Locational Banking Statistics.
Note: Cumulative gross inflows computed as the sum of quarterly changes in foreign holdings of direct investment, portfolio (BOP), and bank lending (LBS) flows, net of disinvestment.
How will Fed policy normalisation unfold in the years ahead?
This question is crucial for developing economies, since they have benefitted substantially from increased inflows over the period in which QE policies have been in place – total gross inflows as a share of GDP appear to have picked up over the course of all three QE episodes (see Figure 1). The risk of reversals in such inflows is therefore a genuine concern as the Fed embarks on its normalisation plans.
Much work has been done on identifying factors associated with financial inflows (Alfaro et al. 2008, Bruno and Shin 2013, Forbes and Warnock 2012, Fratzscher 2012, Gelos et al. 2011). Our recent research builds on this to address the effects of monetary policy normalisation on financial flows to developing countries (World Bank 2014). Our approach relies on a suite of three models for financial flows and crisis that incorporate elements designed to capture the effects of QE unwinding.
The first model that we use to establish our baseline scenario is a dynamic panel model.1 This model allows for the tremendous cross-country heterogeneity in gross financial flows, while also accounting for the effects accruing to global and domestic factors that can potentially affect inflows. These include real (growth and growth expectations) and financial (interest rates, interest rate differentials, and the VIX index) conditions, alongside institutional drivers (such as country credit ratings). Crucially, in addition to (time-invariant) country fixed effects, we also include a series of indicator variables that are designed to capture whether episodes of QE may have had an effect on financial flows over and above the observable channels.2
The estimates from this model are summarised in Figure 2, which shows the response of capital inflows to a change of one standard deviation in each of the explanatory variables. The estimates – which are broadly consistent with the existing literature on factors associated with financial inflows (Alfaro et al. 2008, Bruno and Shin 2013, Forbes and Warnock 2012, Fratzscher 2012 Gelos et al. 2011) – indicate that while observable factors at both the global and domestic level account for much of the cross-country variation in flows, an (unobservable) QE-specific effect remains, which can account for the larger-than-expected financial flows during the period in which unconventional monetary policies were in place. Between the first half of 2009 and first half of 2013, our estimates indicate that observable global factors explained slightly less than two-thirds of the increase in inflows, of which around a fifth was due to this QE-specific effect (see Figure 2).
Figure 2. Factors linked to capital inflows
Source: World Bank staff calculations.
Note: Estimates of the relative contribution of different factors were calculated by multiplying the observed changes in short-term policy rates, yield curve, the QE episode dummy, and the risk index between the first half of 2009 and the first half of 2013 by the coefficient estimates obtained from the benchmark model.
We then use this model to simulate a baseline scenario for financial flows as global conditions normalise. These simulations are conditioned on the following underlying assumptions:3
- Developing and high-income country GDP growth gradually strengthens in line with the projections reported in World Bank (2014);
- QE tapering by the US Federal Reserve spans from January to December 2014, and has a very gradual effect on market conditions. It adds 50 basis points to US long-term interest rates by the end of 2015 and a cumulative 100 basis points by the end of 2016. Policy rates in the US start to increase in Q3 2015, from 0.25% to 2% by the end of 2016;
- The ECB, the Bank of Japan, and the Bank of England start to unwind their own quantitative/qualitative policies in the course of 2015–2016, adding 50 basis points to their long-term yields by the end of the forecast horizon, and tighten policy rates later than the US Fed does.
The results for the baseline are a decline of capital inflows – relative to a ‘no change’ status quo – of about 10% by 2016, or 0.6% of developing-country GDP by 2016 (see Table 1).
Table 1. Baseline simulations for monetary policy normalisation, 2014-16
The foregoing results assume that monetary authorities in high-income countries are able to engineer a gradual increase in long-term interest rates as quantitative easing is withdrawn in line with improved growth conditions. However, the ‘taper tantrum’ in the middle of 2013 suggests that a smooth market reaction to the actual tapering of quantitative easing is far from assured. The second model we consider is a six-dimensional vector autoregression model (VAR) – comprising gross inflows, developing-country GDP growth, G4 GDP growth, short-term interest rates, yield curves, and the VIX – that offers greater flexibility in capturing such disequilibrium scenarios. The two scenarios that we explore with this model are:4
- Fast normalisation: long-term interest rates snap up by 100 basis points in the first half of 2014, before gradually converging back to baseline levels over the subsequent two years; and
- Overshooting: market reactions are assumed to be more abrupt, resulting in a sharp, 200 basis-point increase in long-term interest rates in first half of 2014, followed by a more protracted adjustment back to the baseline.
In the fast normalisation scenario, the resulting increase in market volatility and rising risk aversion leads to a sharper but partially temporary correction in flows. In this context, private capital inflows drop by an average 30% in 2014, with a peak impact of 50% toward the end of the year. In the overshooting scenario, flows drop by 45% in 2014 as whole, and up to 80% at the peak impact. These simulations are captured in Figure 3.
Figure 3. Simulation results
Source: World Bank staff calculations.
Note: Baseline paths are from the panel model, and fast normalisation and overshooting paths are from the VAR.
The preceding analysis suggests that in the long run, the withdrawal of quantitative easing and a return to a tighter monetary policy in high-income countries will have a relatively small impact on capital inflows, reducing them from 4.6% of developing-country GDP in Q3 2013 to 4.0% by the end of 2016. However, the path to this new normal level of flows will matter. If market reactions to tapering decisions are precipitous, developing countries could see flows decline by as much as 80% for several months. That would raise the likelihood of abrupt stops at the country level, with more than 25% of individual economies experiencing such an episode in these circumstances.
We thus rely on a third model, a (pooled) Probit for banking crises, to examine the vulnerability of countries in the extreme case where monetary policy normalisation precipitates crises in emerging markets.5 The probability that a country suffers a banking crisis is modelled as a function of global factors (such as global risk appetite and liquidity), contagion factors (such as trade and financial linkages), and domestic factors (such as the current account and fiscal balance).
The model points to all three of these sets of factors contributing to increased banking crisis risk. Based on these estimates, we construct representative radar charts to illustrate domestic sources of risk, by region (see Figure 4).
Figure 4. Domestic sources of risk
Source: World Bank staff calculations.
Notes: Radar charts summarise areas of elevated risk in each region. Each segment corresponds to significant domestic risk factors from the Probit analysis. The centre is the least risky area, and the further away from the centre, the greater the risk. The thick line in each region represents the average value of each indicator among the countries whose predicted crisis risk is particularly elevated (one standard deviation above the average predicted risk of the entire sample). The teal area represents the average values of each indicator for the region as a whole. There are no countries whose predicted risk is more than one standard deviation above the average predicted risk in South Asia. Indicator values are standardised using percentile ranks.
Although conditions on the ground will vary and the indicators need to be interpreted with caution, the results are suggestive that:
- In the East Asia and Pacific region, rapid credit expansions over the past five years and a rising ratio of short-term debt to total debt are common areas of concern.
- A high external debt to GDP ratio, which exposes countries to exchange-rate and rollover risk, is an issue in several Central and Eastern European economies.
- In Latin America and the Caribbean, fewer countries appear to be at immediate risk, with rapid credit growth combining with significant short-term debt ratios as the main sources of risk.
- In the Middle East and North Africa, risks stem mainly from exposure to domestic credit quality and government financing needs, against the background of a deterioration in current-account positions.
- Based on existing data, risks in South Asia and Sub-Saharan Africa appear low, but there are concerns that non-performing loans in India have increased, and several Sub-Saharan African economies appear to have elevated risk, with deteriorating reserve positions a common thread.
Although the probability of disorderly adjustments remains low at present, policymakers in developing countries need to make contingency plans and be prepared for the inexorable tightening of global financing conditions. Countries with adequate policy buffers and investor confidence may be able to rely on market mechanisms, and counter-cyclical macroeconomic and prudential policies, to deal with a retrenchment of foreign capital. In other cases, where the scope for manoeuvre is more limited, countries may be forced to tighten fiscal and monetary policy to reduce financing needs and attract additional inflows. Where adequate foreign reserves exist, these can be used to moderate the pace of exchange-rate depreciation, while a loosening of capital inflow regulation and incentives for foreign direct investment might help smooth adjustments. Eventually, reforming domestic economies by improving the efficiency of labour markets, fiscal management, the breadth and depth of institutions, governance and infrastructure will be the most effective way to restore confidence and spur stability.
Disclaimer: The findings, interpretations, and conclusions expressed in this article are entirely those of the authors. They do not necessarily represent the views of the World Bank, its Executive Directors, or the countries they represent.
Alfaro, Laura, Sebnem Kalemli-Ozcan, and Vadym Volosovych (2008), “Why Doesn't Capital Flow from Rich to Poor Countries? An Empirical Investigation”, Review of Economics and Statistics, 90(2): 347–368.
Bruno, Valentina and Hyun Song Shin (2013), “Capital Flows, Cross-Border Banking and Global Liquidity”, NBER Working Paper 19038.
Forbes, Kristin J and Francis E Warnock (2012), “Capital Flow Waves: Surges, Stops, Flight, and Retrenchment,” Journal of International Economics, 88(2): 235–251.
Fratzscher, Marcel (2012), “Capital Flows, Push versus Pull Factors and the Global Financial Crisis”, Journal of International Economics, 88(2): 341–356.
Gelos, R Gaston, Ratna Sahay, and Guido Sandleris (2011), “Sovereign borrowing by developing countries: What determines market access?”, Journal of International Economics, 83(2): 243–254.
Glick, Reuven and Michael M Hutchison (1999), “Banking and Currency Crises: How Common Are Twins?”, Proceedings of the Federal Reserve Bank of San Francisco, September.
Lim, Jamus Jerome, Sanket Mohapatra, and Marc Stocker (2014), “Tinker, Taper, QE, Bye? The Effect of Quantitative Easing on Financial Flows to Developing Countries”, Background paper for Global Economic Prospects 2014, Washington, DC: World Bank.
World Bank (2014), Global Economic Prospects: Coping with Policy Normalization in High-Income Countries, Washington, DC.
1 Due to the well-known (Nickell) bias for panel estimates when lagged dependent variables are included, our estimates are obtained using bias-corrected least squares dummy variables (Bruno 2005).0
2 Additional modeling details are documented in Lim et al. (2014).
3 To discipline the dynamic interactions between the global factors and gross financial flows, we rely on a VAR model (described below) to generate paths for the relevant independent variables, which we then return to the panel model to obtain our baseline estimates.
4 As a consistency check, we also compared the baseline estimates for the VAR against those of the equilibrium panel model. The results are very similar to those in the panel model, with the share of capital inflows to GDP in developing countries declining by 0.5% over the projection horizon.
5 We consider banking, rather than other forms of crises (such as currency crises), for three main reasons. First, in a world in which developing countries are increasingly accumulating large reserve holdings and maintaining floating exchange-rate regimes, currency crises due to sudden stops in financial flows are likely less of a risk than they may have been historically. Second, emerging-market borrowing over the past decade has relied less on foreign currency-denominated debt, which reduces their vulnerability to currency crises. Third, to the extent that banking and currency crises often occur in tandem in emerging economies (Glick and Hutchison 1999), it is sufficient for our purposes to focus on one phenomenon.