International Transmission of Conventional and Unconventional Monetary Policy and Financial Stress Shocks from the Euro Area to Russia 1

: This paper studies the international transmission of the euro area´s monetary policy and financial stress to Russia. The results show that financial stress in the euro area damages Russian economic activity and stock prices, but not its trade balance. The contractionary euro area monetary policy shock decreases Russian GDP, leads to real appreciation of the euro against the Russian rouble, damages Russian stock prices, but does not significantly affect the trade balance between countries. We also found that the Central Bank of the Russian Federation adjusts to monetary policy shocks in the euro area.


Introduction
It has recently been documented that increasing global trade and financial flows have made economies around the world more susceptible to monetary policy actions in advanced economies (Cushman & Zha, 1997 Chen & Semmler, 2018). Some studies have examined in a unified framework the transmission of conventional and unconventional monetary policy and financial stress within the advanced economies of the US and the euro area (Hubrich & Tetlow, 2015;Kremer, 2016). The international transmission of both types of monetary policy and of financial stress shocks from the euro area to Russia in a unified framework has not been studied. We address this question here.
The euro area and Russia are strongly connected via trade. 2 Moreover, one should bear in mind a possible complexity of international monetary policy transmission of the euro area monetary policy to Russia given the new unconventional monetary policy. 3 Therefore, understanding the transmission of monetary policy and financial stress from the euro area to Russia is of great importance for the design of economic policy.  (Chen et al., 2017). To fill this gap, we investigate the transmission of euro area (conventional and unconventional) monetary policy to Russia and the relevance of individual channels of transmission.
With respect to international transmission of monetary policy, this paper extends Kremer´s (2016) one-country model to a two-country macroeconomic model. The model enables us not only to assess the effects of euro area conventional and unconventional monetary policy shocks, but also the effects of euro area financial stress shocks for a set of Russian economic variables. These variables include economic activity (GDP), price level, monetary policy rate, bilateral trade with the euro area, the bilateral exchange rate, stock prices and government bonds yield.

Literature review
The literature traditionally argues that international transmission of monetary policy works via the impact of exchange rates on trade flows and other macroeconomic variables (Cushman & Zha, 1997;Koray & McMillin, 1999;Kim, 2001a, b;Maćkowiak, 2007;Vespignani, 2015;Shobande & Shodipe, 2021). Mohanty (2014) notes that central bank practitioners identify the adjustment in the exchange rate and the monetary policy rate as the prime channels of monetary policy transmission from advanced to emerging economies: the former is singled out as the main channel in economies with a floating exchange rate regime, the latter with a fixed exchange rate regime. Rey (2016) and Evgenidis et al. (2019) point out that in a floating exchange rate regime, monetary policy impulses can exert influence on the trade balance of domestic and foreign economies in two opposing ways: demand-augmenting effects and expenditure-switching effects. A contractionary domestic monetary policy impulse weakens domestic demand including imports, which improves the domestic trade balance (demand-augmenting effect) and, at the same time, an increase in domestic policy rate may induce a real appreciation of the domestic currency, which in turn worsens the domestic trade balance (expenditure-switching effect) and affects other domestic and foreign macroeconomic variables (ibidem). Takáts and Vela (2014) and Rey (2016) show that a floating exchange rate regime cannot insulate domestic economies from monetary spillovers from advanced economies, while Bluwstein and Canova (2016) contend that the exchange rate is the most potent international transmission channel of conventional ECB´s monetary policy for non-euro European countries, irrespective of the exchange rate regime. Takáts and Vela (2014) reason that with fixed exchange rates, the international spillover of policy rates from advanced to emerging economies is direct.  Kremer, 2016). MacDonald and Popiel (2017) point out that each approach has its own advantages and disadvantages. For instance, the balance sheet cannot capture the central bank´s forward guidance, while the shadow rate can (ibidem). However, its weakness stems from the fact that if this rate alone is used to represent the stance of monetary policy, we cannot separate conventional from unconventional monetary policy during the period when the central bank´s official policy rate has not yet reached the zero lower bound (ibidem).
The literature on the transmission of unconventional monetary policy to Russia is sparse. Takáts and Vela (2014) have studied repercussions of US unconventional and conventional monetary policy in 24 emerging economies. They find that the Russian policy rate responds positively to shifts in the US policy rate, but not to shifts in the US shadow rate. Further, they find a significant positive relationship between the US and Russian long-term interest rates. Chen et al. (2017) have analysed and compared the impact of the Fed´s and the ECB´s unconventional monetary policies, proxied by their shadow interest rates, in 24 advanced and emerging economies, including Russia. They find that upon an expansionary Fed policy shock, Russia´s GDP growth and inflation rates increase, while upon the ECB´s monetary policy shock, only credit growth responds positively and significantly. Ono (2018) concentrates on Fed´s conventional and unconventional monetary policy shocks, proxied commonly by the shadow rate and assesses the impact on Russian stock prices, policy rate and bilateral exchange rate. The main finding of the study is that an increase in the shadow rate reduces Russian stock prices and interest rates and depreciates the Russian rouble. Kruglova et al. (2018) apply bank-level data to appraise the impact of the US unconventional monetary policy. They find that monetary tightening is associated with a reduction in bank lending in Russia. Specifically, banks that rely more heavily on international financing cut their loans more heavily. They also note that a structural change occurred in the Russian macroeconomic environment in 2014 and that economic modelling of the Russian economy should take this into account.
Our work is also related to studies of the transmission of financial stress in otherwise standard monetary Vector Autoregressive (VAR) models (Hubrich & Tetlow, 2015;Kremer, 2016). To analyse euro area stress and monetary policy transmission to Russia, we extended the model of Kremer (2016) who studied the macroeconomic effects of conventional and unconventional ECB´s monetary policy and financial stress (proxied by the Composite Indicator of Systemic Stress) and found that financial stress significantly affects output and inflation.

Methodology
The impact of euro area conventional and unconventional monetary policy and financial stress shocks on the Russian economy is assessed by fitting a two-country VAR model and computing impulse responses. The estimated VAR can be written as (see e.g. Christiano, Eichenbaum & Evans, 1996Luetkepohl, 2011;Kremer, 2016) 6 : (1) where is vector of endogenous variables partitioned into 3 blocks is matrix of regression coefficients, l denotes the lag (1,…,p), is vector of regression constants and is a vector of errors.
The first block of endogenous variables consists of euro area variables , the Russian variables are contained in the vector , and the variables related to foreign trade in the vector . is the real GDP for the euro area (Russia, respectively), is consumer price index, is the Composite Indicator of Systemic Stress in the euro area, is monetary policy rate, capturing the stance of conventional monetary policy in the euro area and in Russia respectively, is the Eurosystem´s balance sheet capturing the stance of unconventional monetary policy in the euro area, is stock prices index for Russia, is the yield on long-term Russian government bonds, is the real exchange rate between the rouble (RUB) and the euro (EUR), i.e., the price of one euro in roubles, and is the bilateral trade balance of the euro area with Russia in real terms. and are included in the Russian block variables to identify whether ECB´s monetary policy is transmitted to the Russian economy via the wealth and the portfolio rebalancing channels. The block of variables related to foreign trade helps to identify whether the international transmission of ECB´s monetary policy via trade linkages between the economies is operational. To control for a possible structural change in the Russian economy after 2014 7 , an 6 Compare also to Cushman and Zha´s (1997) model. 7 In 2014, at least three important factors may have contributed to this change (Kruglova et al., 2018; Central Bank of the Russian Federation, 2014): i) new monetary policy regime (managed exchange rate regime was substituted by formal inflation targeting), ii) a large fall in the price of oil, the main Russian export commodity and iii) the imposition of economic sanctions amongst others by the euro area countries. exogenous dummy variable was added 8 in model (1) VAR equations for the and blocks of variables. In addition, the world oil price ( ) was added as an exogenous variable 9 , entering all equations in VAR model (1). All variables, except , , and enter equation (1) in natural logarithm of levels 10 . The lag, , is determined by information criteria.
In equation (1) An ECB´s unconventional monetary policy shock is identified by assuming that, in a particular month, the central bank evaluates that months´ indicators of the euro area economic activity, price dynamics, financial stability and the current level of its policy rate before taking any action that will have an impact on the size of Eurosystem´s balance sheet, , while other variables impact Eurosystem´s balance sheet with a lag. We identify the Central Bank of the Russian Federation monetary policy shock by assuming that it considers contemporaneously all euro area economic and financial sector variables included in the model (1)

Data and empirical results
Model (1) is estimated on monthly data. Although for some variables data availability started in 1999M1, the availability of data for some others was shorter and given the estimated model characteristics the actual period of VAR model (1) estimation was 2000M10-2018M6. A detailed specification of variables is presented in Table 1.

Variable notation
Description Monthly index of seasonally adjusted real GDP for euro area and Russia. Monthly time series was estimated from monthly seasonally adjusted industrial production index for euro area and Russia (total industry, excluding construction; OECD (2019a) data was used) and quarterly (seasonally adjusted) GDP index series for Russia (OECD (2019b) data) and euro area (seasonally and calendar adjusted) (Eurostat (2019b) data) by temporal disaggregation method of Chow-Lin (1971) a . Computation is based on the sum method (index for each month was obtained by summing the estimates obtained by the method for the current and past two months) and as a robustness test the time series was estimated with the average method (see Quilis, 2019). The natural logarithm of the time series enters model (1). Consumer price index for the euro area and Russia. For euro area the seasonally and working day harmonized index of consumer prices (HICP) is used (European Central Bank -ECB (2019a) data), while for Russia the consumer price index (OECD (2019a) data). In robustness test we also seasonally adjusted the series for Russia by using the X-13ARIMA-SEATS method b . The natural logarithm of the time series enters model (1). Notes: a For this intention, the Matlab code of Quilis (2019) was applied. b We utilized the JDemetra+ software (see Grundowska, 2016), available at: https://ec.europa.eu/eurostat/cros/ content/software-jdemetra_en.

The monthly level (index) of the Composite Indicator of Systemic
Next, we present the impulse responses computed from the results of model (1). Figure 1 presents the orthogonal impulse responses of Russian variables to a one standard deviation shock to CISS for the euro area. The plots in the first row convey the response of GDP, CPI and bilateral trade balance in real terms. Evidently, the financial stress shock harms Russian economic activity: GDP drops almost by impact and 10 months from the shock contracts by approximately 0.2%, while recovery is gradual. The response of CPI and the trade balance is rather muted and non-significant. These results corroborate the empirical literature on detrimental domestic effects (Hakkio & Keeton, 2009;Kremer, 2016) and international effects of financial stress on economic activity and non-significant effect on price dynamics (Dovern & van Roye, 2014).

Figure 1: The response of the Russian economy to euro area financial stress (CISS) shock
Source: Own calculations Notes: Model (1) with two lags (indicated by HQIC information criteria) was estimated. The mean (orthogonal) impulse response of the variables to a positive one standard deviation shock to CISS for euro area and 95% confidence intervals (shaded area) are drawn. We used Stata degrees-of-freedom adjustment. The responses are changes in logarithm values of variables with the exception of policy rates and government bonds yield, which are in percentage points.
Graphs in the second and the third rows display the response of Russian financial variables. Only the impulse response for the share prices index is significant and conveys that Russian stock prices tumble upon a financial stress shock in the euro area. Response to a shock is immediate and culminates within a month from the shock when stock prices decline by approximately 3%. Whilst ECB responds 12 to 12 We do not present the impulse responses for the euro area variables but note that the euro area response to the domestic financial stress shock is almost identical to Kremer (2016): GDP and ECB policy rate drop, while the Eurosystem´s balance sheet expands.
counter the negative impact of the shock for the euro area´s economy by reducing its policy rate, the response of policy rate in Russia is not significant. The euro in real terms depreciates against the rouble on impact, but also this response is not significant 13 , possibly reflecting the exchange rate stabilizing efforts of the Russian central bank in the observed time period (see e.g. Central Bank of the Russian Federation, 2013, 2014). The impulse response of yield on Russian government bonds is not significant.
Possibly due to sanctions against Russian financial institutions, which drastically reduced borrowing from abroad (see e.g. Gurvich & Prilepskiy, 2015), the transmission channel of financial stress to the Russian financial market was rather muted. is observed, indicating (following e.g. Falagiarda et al. (2015) and Varghese & Zhang (2018)) that the portfolio rebalancing channel is not an operational channel of ECB´s conventional monetary policy transmission to Russian economy.  Figure 3 shows the response of the Russian economy to a positive ECB unconventional monetary policy shock that expands the volume of the Eurosystem´s balance sheet.

Figure 3: The response of the Russian economy to Eurosystem´s balance sheet (ECB unconventional monetary policy) shocks
Source: Own calculations Notes: Model (1) with two lags was estimated. The shock relates to Eurosystem´s balance sheet (unconventional) monetary policy. Other notes from Figure 1 apply.
An expansionary unconventional monetary policy shock is transmitted to Russia´s CPI, stock prices and the bilateral exchange rate. Following a one standard deviation increase in the Eurosystem´s balance sheet, CPI in Russia increases by approximately 0.15% two months from the shock. Compared to the conventional monetary policy shock (Figure 2), the effect is relatively small and shorter in duration. The response of the Russian stock market follows within a month of the shock when stock prices fall by approximately 1.2 % and then recover approximately one year from the shock. It implies that the wealth effect is operational. However, the result of the transmission of unconventional monetary policy shock through this channel differs from the conventional monetary policy transmission. Whereas a reduction in the ECB's policy rate is beneficial (note that Figure 2 shows the response to an increase in ECB´s policy rate), an expansion of the Eurosystem´s balance sheet is detrimental to the Russian stock market. Negative impact of the ECB´s expansionary unconventional (balance sheet) policy on domestic and foreign stock prices is reported in the literature (Bluwstein & Canova, 2016). In contrast, Ono (2018), who studied the impact of the Fed´s shadow rate shocks, found that contractionary policy shock of the latter results in a reduction in Russian stock prices. A significant response in the real bilateral exchange rate is notable. In line with theoretical predictions, a Eurosystem´s balance sheet shock leads to real depreciation of the euro against the rouble by a maximum of approximately 1% 3 months from the shock. The trade balance is significantly impacted by nonconventional monetary policy shock one month from impact only, worsening the bilateral trade balance ratio of the euro area with Russia.
Comparing the responses of Russian variables to monetary policy shocks in the euro area (Figures 2 and 3), we assert that conventional monetary policy has a larger impact on the Russian economy than unconventional monetary policy. In this regard, two observations are relevant. First, the euro area conventional monetary policy is transmitted to Russia via policy rate, exchange rate and wealth channels, while unconventional monetary policy is transmitted via exchange rate and wealth channels. Second, the former type of monetary policy affects Russian GDP and CPI, whereas the latter only affects CPI.
As a robustness test, an alternative ordering of share prices index, government bonds yield and exchange rate was tried, and we found that the above presented results are robust. Next, we estimated model (1) with alternative estimate of GDP series (see Table 1 for explanation) and found that the above presented results are not impacted, except the bilateral trade balance response to unconventional monetary policy shock that becomes non-significant. Using seasonally adjusted CPI time series for Russia instead of CPI (see Table 1 above) makes the responses of Russian GDP, CPI, and share prices to ECB´s policy rate shock and the CPI to ECB´s unconventional policy shock non-significant while other findings presented above are not impacted.

Conclusion
The scope of international trade and financial links between Russia and the euro area countries implies a potential importance of international transmission of monetary policy and financial stress to Russia. This paper fills the gap in quantitative analysis of these issues. Some of our results are in line with the reviewed literature. The reaction of Russian GDP and CPI to a euro area financial stress and conventional and unconventional monetary policy shocks qualitatively corresponds to the evidence in other countries. However, the reactions of stock prices, bilateral exchange rate, policy rate of the Central Bank of the Russian Federation and Russian government bonds yield show some specific dissimilarities. In particular, government bonds yield does not significantly react to a euro area shocks. The real exchange rate significantly responds to conventional and unconventional ECB's policy shocks, but not to a euro area financial stress. This reassesses the role of financial linkages. Among other findings, we also confirm the importance of the policy rate channel for the transmission of shocks to Russia and show the crucial difference between the impact of conventional and unconventional euro area monetary policy on the Russian stock market.