The only thing we have to fear is investors’ fear in the market

4 August 2014

The Tapering Scare began in May 2013 sparking investors’ fears about unclear future US monetary policy shifts, resulting in increased risk aversion and downward pressure on asset prices. Which is the country that suffered the most in realized returns because of fear triggered by the Fed’s announcement?

What is the cost of the Fed-Tapering Scare? More precisely, how much have investors lost in financial markets as a result of increased risk aversion in response to growing expectations of unconventional monetary policy tightening? In times of economic stress, investors’ tolerance for risk decreases and puts downward pressure on asset prices. This was seen in May 2013 as a consequence of speeches by Fed officials, which provided a shock to the market (fears of unintended consequences of a disorderly exit from expansionary monetary policies), increased risk aversion and decreased market returns. Focusing on equity markets, we estimate how much of a loss investors suffered due to increased risk aversion from that shock and create a ranking of the markets that are most sensitive to these shocks.

To measure the total loss faced by investors we apply a two-step methodology. In the first step, we decompose the VIX[i] into its two components of uncertainty and risk aversion, selecting the best forecasting model for realized stock variance (which reflects stock market uncertainty) and estimate (squared) risk aversion as a residual.[ii] Then we calculate a real-time “fear” index that identifies periods of excessive risk aversion. In the second step we apply a simple counterfactual exercise in which we estimate a theoretical level of stock returns corresponding to a situation of “normal” risk aversion. In particular, unexpected shock returns are regressed to the fear index, removing the impact of exceptionally large risk aversion shocks. Finally, for each stock market, theoretical and actual equity indexes are then compared and the Tapering Scare cost estimated as the difference between the two.[iii] [iv]

The decomposition of the VIX into uncertainty and risk aversion illustrates that in times of economic calm, uncertainty is the larger component of the VIX (Figure 1). However, in times of economic and financial stress – such as the September 11, 2001 attacks, the 2008 Lehman Brothers Collapse or the 2010 Sovereign Debt Crises in Europe – risk aversion is the larger component of implied volatility.

Figure 1: VIX Decomposition into Uncertainty and Risk Aversion

From the beginning of the so-called Tapering Scare to March 14th 2014, before the FOMC meeting, investors in the emerging markets[v] have lost on average 5% in terms of Total Return, while developed economies have gained on average 10.9%. These performances reflect conventional wisdom that after Bernanke’s speech in May 2013, emerging markets faced capital flows reversal by investors that had previously been seeking yields in these markets because of the Fed’s asset purchase program. A natural question that arises is whether the Tapering Scare has also reduced investors’ tolerance for risk other than revising beliefs about probable returns from investing in emerging economies. Our results suggest that in this period risk aversion has been a global downward factor on equity returns, as well as, uncertainty, to a lesser extent. Based on our indexes we find that during the period from the FOMC meeting in May 2013 through the beginning of March 2014, the US had the greatest loss with a difference of 6.4% between realized and counterfactual returns, while China had the smallest one of 3.6% (Figure 2). The results show there were indeed losses from the Tapering Scare and they impacted developed economies on average more than emerging economies (the uncertainty indexes provide similar results).

Figure 2: Performances of stock return index and our indexes during the Tapering Scare and its implementation: May 17th 2013 – March 14th 2013

The main explanation for why the US, and not emerging markets, suffered the most from risk aversion relies on the fact that the Fed’s decisions have, in general, greater impact on the American economy than the rest of the world. Monetary policy shocks are then transmitted to other developed economies because of the high degree of both synchronization of business cycles and interconnection of domestic stock markets. In general we can observe that the impact of risk aversion is quite homogeneous across countries ranging from 3.6% to 6.4% in the period May 2013 – March 2014. More specifically, the impact of risk aversion is similar in US, Germany and Italy reflecting concerns about US monetary shocks affecting economic growth in the Euro Area, in particular for countries where Public Debt to GDP reaches abnormal levels. Interestingly, in the same period stock markets in both Germany and Italy increased by 16.2% and 23.9% respectively. According to our methodology, this difference in performance is not explained by a larger sensitiveness of Italian stock returns to investors’ risk tolerance but, more likely, by improved economic fundamentals of the Italian economy that are reflected in historical comovements.

In conclusion, our analysis shows that increased risk aversion and/or uncertainty in times of economic shock results in a loss of returns for investors. Our findings also highlight a relationship between monetary policy and risk aversion in which major shifts in US monetary policy have implications for global financial markets’ level of risk aversion and asset prices. Our methodology allows us to estimate a real-time “fear” index, identifying when we are facing periods of turbulence in stock markets. Furthermore it can help investors in gauging the level of concern they should have for risk aversion shocks and in creating a ranking of the countries that are most sensitive to them.


[i] The VIX is an indicator of implied volatility expectations of the S&P 500 Index Option for the next 30 days in the market. High levels of volatility indicate a market expectation for a large positive or negative shock in the short term, whereas low levels point towards expected stability.
[ii] The VIX could be interpreted as the sum of two components: expected uncertainty of equity markets (the quantity of risk) and risk aversion (the price of risk). The methodology for VIX decomposition involving the testing of 7 different forecasting models is based on the work of: Bekaert G., Hoerova M. and Lo Duca M. (2013), “Risk, Uncertainty and Monetary Policy”, European Central Bank Working Paper Series, No. 1565. *Note that we applied a GARCH to our realized variance data.
[iii] The counterfactual case approach using only exceptionally large changes is defined as the recorded value of uncertainty or risk aversion being two standard deviations above or below its previous 60 day mean. This approach is based on the work of: Groen J. and Peck R. (2014), “Risk Aversion, Global Asset Prices, and Fed Tightening Signals”, Liberty Street Economics, Federal Reserve Bank of New York.
[iv] The VAR (Vector Autoregressive Model) is estimated on the weekly equity returns for 22 counties: 8 advanced and 14 emerging economies. These countries are: Argentina, Australia, Brazil, China, Chile, France, Germany, Hong Kong, India, Indonesia, Italy, Japan, Malaysia, Mexico, Philippines, Singapore, South Korea, Taiwan, Thailand, Turkey, UK, and USA.
[v] The average realized return is computed from our dataset of emerging economies excluding Argentina that have performed 79% in the same period. The main explanation for this rally is attributed to the ban on buying dollars for locals that has boosted domestic demand in the equity market.


Read All