Spillovers of US Unconventional Monetary Policy to Emerging Markets: Evidence from Egypt

This paper studies the Spillover effect of US unconventional monetary policy (UMP) on Egypt as a case study of an emerging market and a small open economy The authors adopts structural vector autoregressive (SVAR) model with variable lag structure. The Wu and Xia (2016)’s shadow interest rate is used as a measure of U.S. unconventional monetary policy. In case of Egypt, we use Short Interest rate (r) as a measure of monetary policy rate; our empirical results reveal that US unconventional monetary policies significantly affect the monetary policy of Egypt but this effect is less on other macroeconomic variables. The main recommendation of the paper is that monetary authority in Egypt should take into consideration the conflict effect of US monetary policy on Egyptian economic indicators, and at the same time it should implement suitable policies coincide with it to achieve the economic stability and targeting inflation. This study analyses the Spillover effect of US unconventional monetary policy (UMP) on Egypt as a case study of an emerging market and a small open economy. We have used the Wu and Xia (2016)’s shadow interest rate as a measure of U.S. monetary policy. To check the Spillover effect of US UMP on Egypt Economy, we used structural vector autoregressive (SVAR) model with variable lag structure. Our results show strong relationship between US monetary policy and Egypt policy rate. The variance decomposition indicates that 48% of changes in policy rate in Egypt return to US UMP. Other variables of Egypt economy such as GDP and inflation responds less and lagged more than 2 quarters. In addition, Egypt GDP do not respond quickly to the UMP, and the effect of USA shadow interest rate on inflation in Egypt is weak, and the effect of USA shadow interest rate on the Egyptian exchange rate is weak. which


Introduction
The traditional tools of monetary policy lost their effectiveness during the 2008 global financial crisis, when there was a recession with low inflation rates below the target and interest rates close to zero. The term Unconventional Monetary Policy (UMP) appeared on the surface as one of the mechanisms for facing the effects of the financial crisis.
The mechanisms of these policies were manifested in a number of related concepts such as quantitative easing (QE), which as explained by (Bank of England, 2013), that the central bank undertakes large-scale purchases of financial assets from commercial banks and other financial institutions, which is reflected in higher assets prices. These higher prices and lower returns on assets, at the same time working to inject more money into the economy. Quantitative easing differs from the open market operations followed in traditional monetary policies. The traditional monetary policies deal mainly with short-term government bonds and treasury bills, and aims to influence short-term interest rates. While in quantitative easing, the financial assets purchased by the central bank extend to long-term bonds. This policy aims to increase the money supply while affecting long-term interest rates. Buiter (2008) distinguished between the term quantitative easing and qualitative easing, which aims not only to add more financial assets to the balance sheet of the central bank, but also to diversify the structure of assets owned to include assets with higher risk. first round of quantitative easing (QE1), with large-scale asset purchases represented by the purchase of $600 billion in mortgage-backed securities (MBS), whereas before the financial crisis of 2007/2008 The US Federal Bank kept in its balance sheet the range of 700 to 800 billion dollars of treasury bills. In March 2009, it held $1.75 trillion in bank debt, mortgage-backed securities (MBS), and Treasury bills, reaching a peak of $2.1 trillion in June 2010, and buying more when the economy started to improve, it resumed in August 2010 when the Federal Reserve decided that the economy was not growing strongly. Further, In November 2010, the Federal Reserve announced its second round of quantitative easing (QE2), buying $600 billion of Treasury bills by the end of the second quarter of 2011. Furthermore, the third round of quantitative easing (QE3) was announced on September 13, 2012, as the Federal Reserve decided to launch a new cycle of buying $40 billion in mortgage-backed securities (MBS) each month.
Unconventional monetary policies contributed to supporting economic activity and preventing the collapse of the financial system, and the quantitative easing (QE) policy in the US had an effective role in avoiding the effects of the 2008 financial crisis, and there are great efforts to study the effects of these policies during their application and in the post-application stage. Christine Lagarde, August 2013 stated that unconventional monetary policies had contributed to saving the world from falling into the crisis of the Great Depression once again.
The aim of this study is analyzing how applying UMP in the USA affects the emerging countries. The main contribution of this study is covering the case of Egypt as an emerging market, and to the best of the author knowledge it is the first time for a paper to study this case.
The rest of the paper is organized as follows. The next section is exploring literature review. Section 3 describes the estimation methodology, Section 4 presents the data we use and section 5 discusses the main results. Finally section 6 summarizes the conclusion of this study.

Literature Review
Some studies focused on the cross-border impact of quantitative easing policies on the financial markets and assets prices of developed countries only. Neely (2010), Glick and Leduc (2012), Scotti and Wright (2014) concluded that quantitative easing in the United States lowered assets prices in other advanced economies.
Another study by Lubys and Panda (2021) were using the same event study methodology of Neely (2010) but applied it on developing countries and examined the effects of unconventional policy announcements by Europe and US on emerging stock markets (BRICS). The study concluded that the policy announcements do influence the stock markets of the emerging countries.
Another studies focused on the impact of US UMP on the global economy, , Anaya et al. (2017) exploring the impact of quantitative easing (QE) policies in the United States on both emerging and advanced economies, using the Global Vector autoregressive model (GVAR).While , used US term and corporate spreads as indicators of US unconventional monetary policy, Anaya et al. (2017) used the changes in the central bank's balance sheet. Both studies found that US (UMP) had significant effects on the global economy, and those effects varied greatly from one economy to another.
In the same context, other studies using (GVAR) to explore the effects of UMP not only of US but also of other major central banks. Chen, Lombardi, Ross and Zhu (2015) explored the comparison of the international spillovers of unconventional monetary policies between the European Union and the United States. Inoue et al. (2020) examined the effects of unconventional monetary policies (UMPs) by the major central banks.
In another study by Ramos-Francia, Manuel, and Santiago Garcí a-Verdú (2015) focused on the transmission channels of US monetary policy shocks to emerging market economies, using linear regression models and factor-augmented vector models autoregression (FAVAR) for the United States and each emerging country. The study found that the effects of monetary policies in the United States on the economies of emerging countries have changed in terms of importance and relative strength since the third quarter of 2008.
Yildirim and Ivrendi (2021) investigated the international spillover effects of US unconventional monetary policy (UMP) on advanced and emerging market economies, using structural vector autoregressive models (SVAR). They used two spreads as indicators of US UMP: the mortgage and term spreads. The findings also reveal that US unconventional monetary policies significantly affect financial conditions in emerging and advanced countries by altering the risk-taking behavior of investors.
Fewer studies focused in a comparison between the impact of conventional monetary policies before the crisis and unconventional monetary policy after the crisis. monetary policies on bond yields in emerging countries, and the study concluded that the effect was clear only in the period of unconventional policies and was on long-term bonds. Covering more macroeconomic variables, Chen, Mancini-Griffoli, and Sahay (2014) conducted a comparative study of the effects of US monetary policies before and after the crisis; the study found that US monetary policy shocks in general affect capital flows and asset price movements. The study also found that the indirect effects of monetary policy are different and stronger during the unconventional policy period, compared to the conventional monetary policy phase before the crisis.
Some literature studied the spillover effects on specific country or region among them Lakdawala (2021) explored the effects on India using high frequency financial market data with a time-varying parameter approach, show that US monetary policy decisions have had significant effects on the Indian stock markets well before the use of unconventional policy tools and that these effects have gotten stronger during unconventional policy phase. Also, Ekeocha and Udeaja (2020) examined spillover effects of U.S monetary policy on macroeconomic variables in Nigeria using BEKK-VARMA-CCCMGARCH model. They also found significant spillover effects of U.S CMP and UMP on interest rate, exchange rate and inflation rate in Nigeria. Regionally Tran and Pham (2020) traced the monthly responses of equity prices, long-term interest rates, and exchange rates in Asian developing markets to US (UMP). Employing a panel vector autoregression with exogenous variables (Panel VARX) model, found that UMP shocks from the US have different effects than the conventional monetary policy shocks as UMP is associated with a surge in equity prices, a decline in long-term interest rates, and an appreciation of currencies.

Methodology
We use the quantitative analysis by employing Structural Vector Auto Regressive (SVAR) model to obtain non-recursive orthogonalization of the error terms for impulse response analysis. We assume two blocks, Y t which includes a vector of n domestic variables and X t includes a vector of m foreign variables at time t. In our case, Y t represents set of variables of the Egyptian economy and X t represents set of variables of US economy. As Egypt is a small open economy, we assume that the elements of Y t do not affect any of the values in X t+k for any k ≥ 0.
The equations of the SVAR model are represented as follow: Where k and c are vectors of constants, A, B and C are matrices of parameters. u t and ε t are error terms.
We conduct also two analyses which used usually with vector auto regression (VAR), impulse-response function (IRF) and the variance decomposition. The former describes the evolution of the variable of interest along a. specified time horizon after a shock in a given moment. IRF used usually with vector auto regression (VAR) to keep track on the response of variables to shocks. Variance decomposition is used to aid in the interpretation of a vector autoregression (VAR) model. The variance decomposition indicates the amount of information each variable contributes to the other variables in the autoregression. It determines how much of the forecast error variance of each of the variables can be explained by exogenous shocks to the other variables.

Data
The data set used in the estimation spans the period 2001Q1-2019Q4.For the US variables we use GDP, CPI and Equity prices (EQ) are obtained from (GVAR) Database. Mohaddes, and Raissi (2020 The set of Egypt variables using in the SVAR model is constructed as follow: Y = ln(GDP), Dp = ln(CPI t )ln(CPI t-1 ), Ep= ln(E t /CPI t ) Eq = ln(EQ t /CPI t ), R = 0.25ln(1+r/100) Where: Y: GDP of Egypt.
DP: the Egyptian inflation rate.
Ep: the real Egyptian exchange rate against USD.
Eq: the real Egyptian equity prices.
R: the short term interest rate.

Results
In this section, we present empirical results.

Unit Root and Optimal Lags Tests
Dickey-Fuller unit root test was used to check the stationarity of the different series. The null hypothesis was that the series was nonstationary, while the alternative was that the series was stationary. The unit root results are illustrated in Appendix A where all series are nonstationary except for (Dp, USADp and USASP) are stationary. Thus, we used the first difference of nonstationary series.
Second, the number of lags is determined using Akaike information criterion, Schwarz information criterion, Hannan-Quinn criterion, and forecast prediction error criterion. Majority of information criteria suggest a model with five lags is the selected one.  Figure 1 shows the impulse responses of Egyptian D(y)-GDP changes-to other variables shocks. In this case, there is an increase in GDP changes due to US shadow interest rate shock after three quarters. Figure 2 shows the impulse responses of Egyptian DP-inflation-to other variables shocks. In this case, there is a decrease in inflation changes due to the changes in the shadow interest rate till quarter 5 then the changes in the shadow interest rate affects positively the inflation in Egypt till quarter number 24. Figure 3 shows the impulse responses of Egyptian D(ep)-exchange rate changes-to other variables shocks. In this case, there is a +ve effect of USA shadow interest rate on the Egyptian exchange rate changes after 4 quarters till quarter number 10, then the effect turns to be -ve till quarter number 24 except quarter number 20 when there is no effect. Figure 4 shows the impulse responses of Egyptian D(eq)-equity prices changes-to other variables shocks. In this case, there is a +ve effect of USA shadow interest rate shock on the Egyptian equity prices changes after 1 ijef.ccsenet.org International Journal of Economics and Finance Vol. 14, No.6; 2022 quarter till quarter number 17, then the effect turns to be -ve till quarter number 24. Figure 5 shows how the Egyptian short term interest rate responds to the shadow interest rate shock in the USA, the relationship overall is +ve and fluctuates across the period of study. It means when the shadow interest rate in the USA increases, the short-term interest rate in Egypt increases and vice versa.

Variance Decomposition
In this section, we measure the effect of each shock in the USA to explain the variance of the Egyptian variables. Table2 to 6 show Variance Decomposition of Egyptian variables in terms of the shocks identified for the U.S and Egyptian shocks. Table 2 shows that the change in D(Y) is explained by its own shock by 68% of the variability, while less than0.9% is explained by USASP shock in the first quarter but from third quarter this percent is increased to 9% this means that Egypt GDP do not respond quickly to the UMP. Table 3 shows the highest effect which is that the shadow interest rate in the USA explains 16.9% from the change in the inflation in Egypt after 6 quarters, before the 6 quarters and after them the effect of USA shadow interest rate on inflation in Egypt is weak. Table 4 shows the highest effect which is that the shadow interest rate in the USA explains 9% from the change in the Egyptian exchange rate after 5 quarters, before the 5 quarters and after them the effect of USA shadow interest rate on the Egyptian exchange rate is weak. Table 5 shows the highest effect which is that the shadow interest rate in the USA explains 16.5% from the change in the Egyptian equity prices after 4 quarters, before the 4 quarters the effect of USA shadow interest rate on the Egyptian equity prices is weak and after them the effect is declining. Table 6 shows the highest effect which is that the shadow interest rate in the USA explains 48.2% from the change in the Egyptian interest rate from the 1st quarter, after that the effect is still high but declining in the next quarters till it is the lowest effect after 12 quarters.

Copyrights
Copyright for this article is retained by the author(s), with first publication rights granted to the journal.
This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/).