Open AccessArticle

Real Exchange Rate Channel of QE Monetary Transmission Mechanism in Selected EU Members: The Pooled Mean Group Panel Approach

Stefan Stojkov

^*,

Emilija Beker Pucar

and

Aleksandar Sekulić

Department of Economic Theory and Policy, Faculty of Economics, University of Novi Sad, Segedinski put 9-11, 24000 Subotica, Serbia

Author to whom correspondence should be addressed.

J. Risk Financial Manag. 2025, 18(1), 12; https://doi.org/10.3390/jrfm18010012

Submission received: 3 December 2024 / Revised: 22 December 2024 / Accepted: 27 December 2024 / Published: 30 December 2024

(This article belongs to the Special Issue Open Economy Macroeconomics)

Download

Browse Figures

Figure 1
APP structure and ECB balance sheet. (a) Largest asset purchases by ECB per country in the period 2014–2023; (b) ECB balance sheet assets. Source: author’s research according to FRED (2024) data. "> Figure 2
Fluctuation of REER and ECB balance sheet dynamics in the EZ member states (Austria, France, Germany, Netherlands, Italy, Spain, and Belgium) in the period 2014Q1–2023Q1. Source: author’s research according to quarterly IMF and FRED (2024) data. "> Figure 2 Cont.
Fluctuation of REER and ECB balance sheet dynamics in the EZ member states (Austria, France, Germany, Netherlands, Italy, Spain, and Belgium) in the period 2014Q1–2023Q1. Source: author’s research according to quarterly IMF and FRED (2024) data. "> Figure 3
Fluctuation of REER and central bank’s balance sheet dynamics in the monetary autonomous EU countries (Czech, Poland, Hungary, and Romania) in the period 2014Q1–2023Q1. Source: author’s research according to quarterly IMF and FRED (2024) data. "> Figure 4
Average current account position (% of GDP) for EZ members in the period 2014Q1–2023Q1. Source: author’s research according to quarterly OECD data. ">

Review Reports Versions Notes

Abstract

Since the Great 2008 Recession, central banks around the world have been coping with monetary consequences that highlight structural costs of the economic system and the rise of unconventional monetary measures. This research aims to capture the heterogeneous effects of expansionary balance sheet (Quantitative easing) policy on the real effective exchange rate and current account balance under the different exchange rate regimes in crisis circumstances. The sample is structured of two groups of EU countries differentiated by level of monetary autonomy: EZ members (Austria, Belgium, France, Germany, Netherlands, Italy, and Spain) are represented by countries with the highest level of asset purchases by ECB and emerging monetary autonomous EU economies (Czech, Hungary, Poland, and Romania). Empirical findings are based on the framework of cross-sectional dependent, non-stationary, heterogeneous, dynamic panels using the (Pooled) Mean Group estimator during the 2014Q1–2023Q1 time horizon. Results indicate a positive long-run relationship between the central bank balance sheet assets, the real interest rate, and the real effective exchange rate. A negative long-term relationship with the current account balance is confirmed, suggesting a diminishing external position. While error-correction parameters are significant and heterogeneous, research confirms higher real effective exchange rate reaction for the EZ members with higher adjustment toward worsening competitiveness along with external balance.

Keywords:

QE; REER; current account; heterogeneous panels; PMG

1. Introduction

Over the past decade, the macroeconomic policy of the European Union (EU) has encountered numerous challenges necessitating adjustments to monetary strategies. The neoliberal institutional framework, which places monetary policy at the forefront of crisis mitigation, has underscored the significance of programs such as Quantitative easing (QE) and low-interest rate policies (Stefanski, 2022). QE, involving the expansion of the central bank’s balance sheet through purchasing securities and other assets, aims to enhance liquidity and stimulate aggregate demand. However, it entails substantial implications for the exchange rate and competitiveness of member countries. A monetary union, exemplified by the Eurozone (EZ), exhibits certain vulnerabilities in terms of monetary autonomy. Fixed exchange rates hinder the adjustment to inflationary disparities thereby allowing for higher economic costs for extended periods (Beker Pucar & Glavaški, 2020; Beker Pucar & Glavaški, 2021; Glavaški et al., 2022). Conversely, countries with flexible exchange regimes, such as Poland or the Czech Republic, with autonomous monetary policies, have more instruments at their disposal for prompt reaction. This divergence in approach between EZ members and countries with monetary autonomy holds the key to comprehending the effects of QE policy.

The expansion of central banks’ balance sheets through QE had a significant impact on the international competitiveness of EU countries, especially on the real effective exchange rate (REER). Central banks across the world, including the European Central Bank (ECB), have insured by increasing liquidity to stimulate economic growth during crises. However, this approach is not without consequences. While an increase in liquidity can reduce real interest rates and make borrowing easier, it can also lead to an appreciation of REER due to rising prices. Countries with fixed exchange rates, such as members of the EZ, face limited flexibility in adapting to these changes, while countries with monetary autonomy can better absorb shocks and adjust. The key challenge in this macroeconomic environment is balancing internal price stability and external competitiveness. In addition to monetary policy, fiscal measures play a key role in mitigating the effects of QE on external competitiveness. Fiscal discipline can help limit the inflation caused by increased liquidity, thus mitigating negative effects on REER (Glavaški & Beker Pucar, 2021; Pejčić et al., 2024).

Despite rich literature regarding exchange rate regimes, there is a significant gap in the analysis of the QE policy transmission mechanism and relevant repercussions. Consequently, the effects of QE on REER and external balance must be investigated. This research contributes to the existing literature in various ways: First, deepening the understanding of the impact of QE policy on REER in different exchange rate regimes. Comparing EZ members and countries with monetary autonomy contributes to a better understanding of the differences in the adjustment of monetary policy and its consequences on competitiveness. This is particularly useful for designing tailored economic policies in heterogeneous monetary unions. Second, empirical confirmation of a prolonged period (time-lag) is necessary for QE policy to have fundamental effects on competitiveness. This finding contributes to the literature on monetary transmission, providing empirical evidence on the need for long-term adjustment and recalling the limited effectiveness of conventional monetary policy instruments in crisis conditions. Third, confirming that QE can worsen the CAB through increased imports due to higher domestic consumption contributes to the literature on balancing between internal and external stability with prompt policy responses.

The main idea of this research is to investigate how balance sheet expansion affects the REER and the competitiveness of EU member states, especially comparing countries with a fixed exchange rate regime (Austria, Belgium, France, Germany, the Netherlands, Italy, and Spain) and those with a flexible exchange rate regime (Czech, Poland, Hungary, and Romania) during the 2014Q1–2023Q1 time period. The aim is to determine if and how the QE policy, through the expansion of central banks’ balance sheets, affects the adjustment of REER and CAB by highlighting the differences between the two groups of countries. Therefore, the main hypothesis of this paper are as follows:

Hypothesis (H1):

The QE policy has a statistically significant effect on real effective exchange rate appreciation in selected EU countries through the exchange rate channel.

Hypothesis (H2):

The time-lag effects of QE policy significantly affect long-run competitiveness, with delayed effects on the real effective exchange rate and current account balance manifesting after the initial adjustment period.

Hypothesis (H3):

Competitiveness adjustments and external adjustment mechanisms through the exchange rate channel are more challenging for the EZ member states relative to monetary autonomous EU countries.

This paper is structured as follows: after Section 1, the following part of this paper presents a literature overview, and Section 3 of this paper investigates monetary strategies over the past decade with descriptive analyses of key variables of interest. In Section 4 of this paper, the methodological framework is explained through the panel ARDL model (PMG and MG estimators). Section 5 of this paper discusses empirical findings concerning relevant issues.

2. Literature Review

Neoliberal institutional design led to the highest level of economic integration with the creation of supranational monetary authority of the EZ. This integration created a climate of globalization and capital deregulation that preceded a chain of economic crises. Given the reduction in fiscal constraints on government spending, policymakers have found unconventional monetary policy as the key to macroeconomic stabilization. Neoliberal capitalism emphasizes the private sector and the market, while monetary policy becomes the primary toolkit for mitigating crises by encouraging credit and asset price growth. Fixing the exchange rate and abandoning monetary autonomy in a heterogeneous monetary union meant relinquishing a countercyclical buffer in the face of inflationary differentials. The European monetary strategy focusing on price stability and marginalizing the real economy entails certain costs. Depending on the effectiveness of the monetary transmission channels, policy efficiency can be determined. For instance, Weale and Wieladek (2016) examined how unconventional monetary transmission channels reduce economic uncertainty, thereby fostering higher private consumption resulting in higher prices. Unlike in floating regimes, where there are fewer constraints due to the exchange rate serving in a shock absorber role with fixers, inflationary differentials amplify the costs of longer adjustments (Fidora et al., 2017; Pierluigi & Sondermann, 2018; Rohit & Dash, 2019).

Based on the research of Buch et al. (2018), the relevance of a central bank’s balance sheet expansion can be highlighted, particularly within heterogeneous monetary unions, due to its dual influence on exchange rates. Asset purchases lead to an increase in the money supply, resulting in depreciation effects on the nominal exchange rate. This effect can be reversed through inflationary pressures and real exchange rate appreciation. These findings suggest that the long-term effects of QE measures lead to diminished competitiveness. Furthermore, Hohberger et al. (2019) emphasize the distinction between the short-run liquidity effects of QE on long-run financial conditions and cross-border capital outflows. Increased money market funds in the short run spill over to long-run expansionary effects through eased financial conditions. Lowering real interest rates generated by QE creates a macroeconomic environment where increasing aggregate demand becomes the central focus (Grauwe & Ji, 2024). Kinateder and Wagner (2017) suggest that in crisis circumstances for EZ members, QE serves as a fundamental instrument to mitigate the negative effects of external shocks on internal balances, but it entails a trade-off with regard to external balance deterioration.

Alternative monetary theories suggest that the primary objective of QE policy should be to prevent deflation rather than directly influencing real economic variables. This policy is considered a fundamental economic driver due to the increased liquidity it brings to the financial system (Kelton, 2020). This increased liquidity subsequently leads to credit easing, which stimulates aggregate consumption and investments, thereby preventing deflation during crisis conditions. Since the source of increased liquidity supply is financial asset purchases, this policy does not result in increased debt. Conversely, the basic restriction alternative to public debt and deficit is inflation. Further studies indicate that, since financial assets are the central point of interest, only the upper class will benefit, leading to increased income inequality. Therefore, coordination with fiscal policy is necessary, as QE should not only increase financial market liquidity but also include programs that assist lower-income classes.

This research paper delves into the consequences of unconventional monetary policy, specifically highlighting repercussions in different exchange rate regimes. The volatility of real exchange rate determinants implies that a country’s competitiveness is vulnerable to the dominance of price or nominal exchange rate fluctuations (Gnimasoun & Mignon, 2013; Kenourgios et al., 2015). Thus, Dedola et al. (2020) suggest that currency depreciation during central bank balance sheet expansion is imminent. The impact of this phenomenon on macroeconomic activity is uncertain and depends upon relative price increases. QE policies expand central bank assets and increase money supply by injecting liquidity into the financial system. Consequently, Claeys and Alvaro (2016) investigate characteristics of the ECB’s asset purchase program (APP), implying that effectiveness could be questionable due to certain constraints on asset purchases. Modifications to the regulations, such as elevating the limit on the percentage of bonds that the ECB can acquire, have somewhat enhanced flexibility, but have also resulted in fluctuations in the currency’s value due to increased market liquidity (Koijen et al., 2017). Higher money supply lowers real interest rates but increases aggregate spending, hence resulting in expansionary effects and increased prices (Stojkov et al., 2023). Conversely, a lower real interest rate reduces currency demand, leading to nominal exchange rate depreciation. From a policy perspective, if QE generates higher inflationary pressures than nominal exchange rate depreciation, it will lead to real exchange rate appreciation. Higher prices will consume nominal depreciation, increasing costs of worsening competitiveness and prolonging the adjustment process for fixers. Countries with floating exchange rate regimes can absorb higher prices more quickly, shortening the costs of the competitiveness adjustment and creating an environment for stable growth.

Spillover effects from QE to lower interest rates also generate portfolio rebalancing in terms of lower risk premiums, which tends to lead to capital outflow and nominal depreciation (Blanchard et al., 2016; Williamson, 2012). Higher nominal depreciation than price increases would lead to real exchange rate depreciation and improvement in competitiveness. This raises an incremental question in the sense that restrictive monetary policy could imply turbulent financial market reactions. Eichengreen and Gupta (2013) concluded that real exchange rate appreciation resulted in a current account deficit for countries with expansionary economic policies. Furthermore, Park et al. (2015) determined that QE effects, such as lowering long-term bond yields, were crucial factors in influencing exchange rate movement during crisis circumstances. These divergences amplify structural differences in balance sheet expansion in the sense that focusing monetary policy on bolstering economic activity would come with a cost. Highly indebted economies are more vulnerable to global shocks since asset prices fall during crises. The fundamental aspect of overcoming crisis conditions is the implementation of QE policy due to its high demand for private sector assets. Since prices responded more swiftly to unconventional measures, real exchange rate appreciation occurred in the EU with higher adjustment costs in relation to EZ members (Ivanova, 2018). These dynamics, which include adjustments in interest rates, changes in investment flows, and changes in the value of the domestic currency, create challenges for economic policies in open economies that often have limited options for exchange rate control.

3. European Monetary Strategies

Over the past decade, the ECB has encountered significant challenges that have necessitated a reevaluation of its monetary strategies. These challenges include low-interest rate conditions, global economic shocks, and accelerating inflation. A pivotal moment came with the revision of the 2021 strategy, which introduced a symmetrical inflation target of 2% and committed to a persistently accommodative monetary policy near the effective lower bound of interest rates (Baumeister & Benati, 2013). These policy changes enhance flexibility during periods of low interest rates without compromising the ability to respond effectively to inflationary shocks, as shown by Ferdinandusse et al. (2020). Moreover, during the pandemic economic crisis, the ECB employed unconventional measures in conjunction with a symmetrical approach to inflation targeting by focusing on market liquidity (Hesse et al., 2018). This combination was employed to mitigate the adverse effects of the recession and foster economic recovery. Prior to 2014, the ECB exhibited a more pronounced reaction to inflation exceeding the target level. Conversely, the more recent strategy has adopted a more symmetrical approach, demonstrating adaptability in periods characterized by low natural interest rates. Unconventional measures such as negative interest rates and balance sheet expansion were instrumental in this transition, while strategies like average inflation targeting further bolstered the resilience of monetary policy in intricate economic scenarios (Bernanke, 2020). The comprehensive approach is to strike a balance between the short-term objectives of price stability and the long-term preservation of the currency’s purchasing power.

Analyzing the RER within the EU illuminates the pivotal factors that contribute to economic imbalances and competitiveness. In a fixed exchange rate regime, a slower RER adjustment process leads to prolonged economic costs, including external imbalances and elevated prices (Curdia & Woodford, 2011). Striking a balance between the internal and external consequences of expansionary policies, such as QE, is a fundamental challenge in optimizing their efficiency. In contrast, emerging EU countries with autonomous monetary policy highlight the significance of external factors and their structural characteristics. For instance, Poland adjusts short-term interest rates based on factors such as terms of trade to influence exchange rate stability during crises (Bhattarai et al., 2021). Emerging countries are interconnected, and their growth is highly dependent on foreign capital. This is the primary reason why volatility in these countries often originates from global economic shocks. Hungary relies on foreign direct investment (FDI) from the EU, which, in turn, impacts the RER through increased productivity and technological advancements (Hashmi & Nsafoah, 2024). Consequently, there is a fundamental possibility of worsening competitiveness on the international stage if there is no appropriate domestic policy response. Mitigating external shocks through policy adjustments is the primary reason for preserving monetary autonomy in emerging EU countries. However, in dire circumstances, such as a global economic crisis, central banks resort to balance sheet expansion as a solution.

The programs that form the foundation for implementing the QE policy is APP which expands the central bank’s balance sheet assets. Figure 1a illustrates the trajectory of the APP program in European countries where the ECB acquired the most assets between 2015 and 2023 (Agba et al., 2022). Additionally, it provides insights into the asset composition of the ECB’s balance sheet. During the initial years (2015–2017), all observed European countries experienced a consistent growth in asset purchases. The primary driving force behind this exponential surge was the spillover of the Great Recession into Europe, coupled with the debt crisis that the EU faced during those periods. Germany stands out with the highest values, while Austria and the Netherlands exhibit a more moderate increase in purchases. Belgium and France display smaller fluctuations, while Spain and Italy maintain a steady growth trajectory. The fluctuations in the APP program across countries are influenced by various factors, including the public debt of each nation, the size of its economy, and the current economic circumstances (Bleaney & Tian, 2013). Economies with lower fiscal responsibility and those experiencing economic malaise are likely to derive the most significant benefits from the ECB’s programs. The APP program reached its peak in 2017, after which there was a gradual decline in asset purchases in all countries. However, the COVID-19 pandemic crisis brings about a reversal in this trend, resulting in a sudden drop in the APP program due to global restrictions and uncertainty. Consequently, the ECB shifted its focus to the pandemic emergency purchase program, which comes into effect with a certain time-lag (Bowdler & Radia, 2012). As a result, asset purchases are rehabilitated until 2021. Germany and Belgium demonstrate steady recoveries in asset purchases by the ECB, while Italy, Spain, and France experience a slower return to their previous levels. Manipulating a balance sheet, an unconventional measure employed during crisis circumstances, can have detrimental effects such as elevated prices and diminished competitiveness. Consequently, it is imperative to increase assets at a more gradual pace. Figure 1b illustrates this concept; despite a decline in the APP program, the ECB’s balance sheet exhibits a consistent upward trajectory. This suggests that the ECB is attempting to balance both the positive and negative repercussions of balance sheet expansion.

Analyzing the dynamics of QE and REER in selected EZ member states, Figure 2 clearly demonstrates a correlation between both trajectories. On the primary axis are shown values of the ECB balance sheet while on the secondary axis are values of the REER index. All countries exhibit a consistent upward trend in QE, while the REER exhibits varying patterns depending on the nation. While QE facilitated economic recovery and averted deflation, inflation in France remained comparatively low in relation to EZ countries. Consequently, France’s relative prices and costs became more competitive, resulting in an elevation of the REER as determined by these factors. Austria, Belgium, and Germany maintain relatively stable growth in their REER, experiencing only slight fluctuations. However, Austria’s REER eventually declined, suggesting a slight increase in competitiveness. In contrast, Germany demonstrates resilience by maintaining stability, highlighting its ability to adapt to changes in monetary policy by allowing REER to appreciate. France and Belgium, on the other hand, witness a clear rise in their REER, which could imply an increase in relative prices. This potential threat to their trade competitiveness is further exacerbated by the ongoing expansion of the QE policy. The Netherlands, Italy, and Spain exhibit slightly different patterns. The Netherlands experiences more volatile REER, which primarily depends on FDI influx. This is because their fiscal policy is structured to provide tax incentives to foreign investors (Horvath & Voslarova, 2016). Italy, on the other hand, observes a modest increase in REER, which could potentially indicate rising production costs due to a faster increase in real wages relative to productivity. Furthermore, domestic economic challenges such as labor market rigidities and high bureaucratic burdens affect companies’ ability to respond effectively to international competition (Kucharcukova et al., 2016). Spain, despite recording a decrease in its REER initially, experienced substantial effects of the QE policy. This highlights the necessity for robust monetary support to stimulate Spain’s economy in crisis conditions. Overall, countries exhibit diverse responses to the ECB’s monetary policy. In general, the expansion of QE policy is one of the fundamental factors that influence REER. However, internal factors, such as economic structures and trade patterns, also play a crucial role in determining the specific effects of these policies on each country’s competitiveness.

Based on Figure 3, analyzing the relationship between QE and changes in the REER in the Czech Republic, Poland, Hungary, and Romania provides insights into the impact of the central bank’s balance sheet expansion on these economies. On the primary axis are the central bank’s balance sheet asset values while on the secondary axis are values of the REER index. In all four countries, QE exhibits a direct or indirect correlation with REER. Generally, QE growth contributes to fluctuations in REER, although the specific links vary across nations. This is because, unlike in members of the EZ where there is a fixed exchange rate regime, here, monetary autonomy provides a shock absorber function. This implies that there is quicker REER adjustment in these countries which, in turn, lowers the costs of REER appreciation. For instance, in the Czech Republic, after the initial substantial QE operations in 2016, REER experienced a significant increase. After a period of adjustments, the trajectory of REER remains stable without prolonged costs of competitiveness worsening. Conversely, Poland exhibited a milder rise in REER despite moderate QE programs. Various economic factors, including the trade balance and domestic demand, may have influenced this outcome. However, exchange rate policy is primarily focused on maintaining a stable REER. Hungary and Romania exhibited distinct effects of QE on REER. Hungary experienced a slight appreciation in REER, which, after the adjustment process, was even oriented toward depreciation. In contrast, Romania’s REER exhibited more volatility, with occasional spikes in response to QE. This suggests a greater sensitivity of its economy to external shocks.

Overall, the growth of QE contributes to an increase in REER, but the magnitude of this change depends on specific economic conditions, including inflation, foreign exchange reserves, and prompt policy responses toward exchange rate adjustments. The analysis indicates that QE plays a crucial role in shaping REER, but the intensity and direction of its impact are different than in EZ countries because of the monetary autonomy.

Figure 4 illustrates the evolution of the current account averages (as a percentage of GDP) for Austria, Belgium, France, Germany, the Netherlands, Italy, and Spain over the time period from 2014Q1 to 2023Q1. A consistently positive current account balance has been observed throughout the majority of the observed time period, with the exception of the pandemic crisis, which underscores the competitive position of EZ members on the global market. The primary driver of current account fluctuations is the expansive QE policy implemented by the ECB as a response to the debt crisis. EZ countries lack the ability to adjust their exchange rates, resulting in prolonged economic costs. Examining the movement of the current account reveals a gradual decline from 4% to below 2% of GDP from the end of 2015 to 2020. While the expansive QE policy stimulates the economy and positively impacts the internal balance, it also leads to negative external repercussions, particularly in the form of increased imports due to the expansion of domestic consumption. The pandemic crisis precipitates a significant drop in the current account, which is primarily attributed to external shocks caused by the global closure of economies. After slight rehabilitation of the current account, the global inflationary environment caused a second shock, which led the current account position to negative values. The latter quarters are characterized by a recovery of the current account.

A descriptive analysis of the relationship between the central bank’s balance sheet and the REER suggests a positive correlation with a more pronounced impact on EZ members. The analysis underscores the fundamental distinction in exchange rate regimes where monetary autonomy is paramount for maintaining lower costs through prompt exchange rate adjustments. This relationship is further explored utilizing the Pooled Mean Group (PMG) model.

4. Methodological Framework

National REER indices of monetary autonomous and non-autonomous countries are linked with the balance sheet assets of central banks as expansionary measures implemented during crisis episodes. As is well known, the REER index serves as the foundational indicator of a country’s competitiveness and is determined by the level of inflation and the nominal exchange rate. Expansionary monetary policy typically leads to currency depreciation, which enhances competitiveness. However, the relationship becomes more complex when prices are taken into account. In this research, the authors aim to elucidate the impact of expansionary balance sheet policy on REER for each of the examined countries. The determination of short-run and long-run dynamics, including adjustments toward equilibrium (ECT), is paramount for policymaking decisions, particularly in mitigating economic costs during crisis conditions in different exchange rate regimes. The realization of competitiveness deterioration during crises represents a vulnerability that necessitates reassessment with the implementation of policies aimed at achieving internal balance.

Additionally, the (in)effectiveness of conventional monetary instruments (real interest rate) is also investigated, as it is one of the primary reasons for the rise of unconventional measures. As expected, the relation is positive indicating that a fall in real interest rates decreases the attractiveness of domestic assets leading to a decrease in currency demand and finally REER depreciation. The authors do not investigate motives for interest rate dynamics, such as capital influx or inflationary pressures.

Finally, a separate model captures the consequences of QE policy on the external balance. This time, the relationship is negative, as QE policy leads to a worsening of the current account. Injecting liquidity into the monetary system generates higher aggregate demand, which, in turn, results in higher prices. These higher prices negatively affect exports, further worsening the current account position. Since the shock absorber function has been locked, external balance will be marginalized in favor of internal balance.

Macro-panel data cover 11 monetary autonomous and non-autonomous members (N dimension) in the time framework 2014Q1–2023Q1 (37 quarters as the T dimension). The sample includes seven EZ members with the highest amount of ECB assets purchases: Austria, France, Germany, the Netherlands, Belgium, Italy, Spain, and monetary autonomous EU countries: Hungary, Poland, Romania, and the Czech Republic. The model estimation includes quarterly iterations of selected variables: real effective exchange rate—reer (source IMF International Financial Statistics), quantitative easing policy—lnqe (measured as log of central bank balance sheet asset; source FRED and IMF International Financial Statistics), real interest rate—rir (source OECD) and current account balance—cab (% of GDP, source OECD).

Pooled Mean Group vs. Mean Group Estimator

Since the central topic in the analysis is to determine the long-run and short-run dynamics between REER and QE, but also to differentiate adjustment mechanisms, the authors apply techniques introduced by Pesaran et al. and Pesaran et al. (Pesaran et al., 1999; Pesaran & Smith, 1995) based on non-stationary heterogeneous dynamic panels. The assumption of homogeneity of slope parameters and stationarity are frequently inappropriate in macro panels; therefore, we implemented the Mean Group (MG) and Pooled Mean Group (PMG) methods to estimate non-stationary heterogeneous dynamic panels. Mean-group estimation involves estimating N time-series regressions and averaging their coefficients. In contrast, PMG estimation assumes an equal long-run relationship across all panel units and averages the coefficients. Both models assume heterogeneous short-run adjustments, thus, the choice among these estimators is directed toward restriction in the long-run coefficients. The Hausman test is applied to determine which estimator is more efficient; if the long-run relationship is homogeneous, PMG is more efficient (Hausman, 1978). MG model assumes a heterogeneous long-run relationship, while short-run adjustments are heterogeneous in both models. Namely, these models are particularly useful in research where policy responses in the long-run are expected to be homogeneous while adjustments in the short run depend on dynamics in each state. The error correction parameter (speed of adjustment) when confirming long-run relation should be statistically significant and negative. Since these methods belong to the panel error-correction group, the model could be described as follows:

∆ y_{i t} = ϕ_{i} (y_{i t - 1} - θ_{i} X_{i t}) + \sum_{j = 1}^{p - 1} λ_{i j} ∆ y_{i t - 1} + \sum_{j = 0}^{q - 1} δ_{i j} Δ X_{i t - j} + µ_{i} + u_{i t}

(1)

where Y presents dependent variable while the cross-section units are represented by i = 1, 2, …, N; the number of periods t = 1, 2, …, T; X_it is a k × 1 vector of explanatory variables; ϕ_i is the error-correction parameter, which presents adjustment mechanism toward long-run equilibrium relationship for each monetary autonomous and nonautonomous EU economy—the error-correction parameter is expected to be negative under the assumption that long-run relationship exists and variables converge to long-run equilibrium; in contrast, ϕ_i = 0 means that there is no long-run equilibrium; θ_i is the long-run equilibrium relationship between variables; λ_ij is the coefficient of the lagged dependent variable, δ_ij is the short-run coefficient for each panel unit (EU economy), µ_i represents the individual effects, and u_it is the stochastic disturbance term. In this research, a real effective exchange rate (reer) represents the dependent variable investigated in relation to the impact of central bank balance sheet expansion policy (qe) and real interest rate (rir). Thus, our specification is as follows:

∆ {r e e r}_{i t} = ϕ_{i} ({r e e r}_{i t - 1} - θ_{i} {q e}_{i t} - θ_{i} {r i r}_{i t}) + \sum_{j = 1}^{p - 1} λ_{i j} ∆ {r e e r}_{i t - 1} + \sum_{j = 0}^{q - 1} δ_{i j} Δ {q e}_{i t - j} + \sum_{j = 0}^{q - 1} δ_{i j} Δ {r i r}_{i t - j} + u_{i} + u_{i t}

(2)

In our second model, testing the effects of balance sheet expansion on worsening current account position, the dependent variable is the current account balance (cab) while the independent variable is the central bank’s balance sheet expansion policy (qe). We can estimate the following model:

∆ {c a b}_{i t} = ϕ_{i} ({c a b}_{i t - 1} - θ_{i} {q e}_{i t}) + \sum_{j = 1}^{p - 1} λ_{i j} ∆ {c a b}_{i t - 1} + \sum_{j = 0}^{q - 1} δ_{i j} Δ {q e}_{i t - j} + u_{i} + u_{i t}

(3)

In order to obtain final estimation results regarding short-run and long-run relationships, certain empirical steps were performed. First, testing cross-sectional dependence (CSD) with the null hypothesis of cross-sectional independence among highly integrated EU economies is expected to be rejected. Second, the Panel unit root test is investigated (PURT) with Pesaran’s second-generation stationarity test accounting for cross-sectional dependency (Pesaran, 2007). Third, the Westerlund cointegration test is performed, with the null hypothesis suggesting the absence of a long-run cointegrated relationship between variables (Persyn & Westerlund, 2008). Finally, choosing which model is more efficient between MG and PMG estimators of the panel ARDL model is concluded with the Hausman test. In the following section, results of the stated pre-estimation tests along with panel ARDL with PMG estimator are presented.

5. Discussion and Results

5.1. Panel ARDL Model: PMG vs. MG Estimators Results

The econometric framework begins with the obtained results of the pre-estimation procedure, including the Pesaran CD test, the Pesaran PURT test in the level, and the first differences, as well as the Westerlund cointegration test, which are summarized in Table 1. The first pre-estimation test when analyzing heterogeneous nonstationary macro panels is the CSD test. Consequently, the null hypothesis implies cross-sectional independence, while the alternative hypothesis suggests cross-sectional dependence. The results are as expected with highly integrated countries; we reject the null hypothesis and, thus, we can conclude that CSD is present among EU economies. Since confirmation of CSD requires a second-generation stationarity test, PURT is implemented, which accounts for CSD. The null hypothesis indicates that all series’ are non-stationary in the level; therefore, the proceeding step is stationarity testing at the first difference. The results show that all series are stationary in the first differences. Further analysis includes the identification of long-run equilibrium relationships between selected variables. Since each variable is integrated at order 1, a cointegration test can be implemented. The absence of a long-run relationship led to the conclusion that alternative estimators could be more efficient in analysis. The null hypothesis, no cointegration, against H1, panels are cointegrated. The results of the cointegration test show that the p-value is below 0.05, which rejects the null hypothesis of no cointegration, meaning that REER, QE, and RIR are cointegrated.

The Hausman test serves as the basis for the selection of more efficient among two heterogeneous, non-stationary, and dynamic macro-panel estimators (MG vs. PMG). The assumption of long-run parameter homogeneity cannot be de facto implied. Therefore, we use the Hausman specification test to investigate the effects of heterogeneity on the coefficients. If the null hypothesis of parameter homogeneity cannot be rejected, the PMG estimator is more efficient. Accordingly, if the alternative hypothesis is accepted, then the efficient estimator is MG (p-value below 0.05). In our research, the p-value is 0.37, suggesting that the null hypothesis cannot be rejected (Table 2). We can conclude that the efficient estimator is PMG. Looking at the short-run dynamics, competitiveness is determined by conventional monetary instruments. Initial expansionary measures in crisis circumstances lead to capital outflows and currency depreciation; this is why, in the short run, competitiveness increases. The lagged effects of QE policy imply that, after a certain time period, money market liquidity and increased prices, in the long run, dominate, with competitiveness worsening. The error-correction term helps with analyzing the sustainability of this long-term relationship by indicating the speed of adjustment toward equilibrium. The coefficient of the error-correction term is significantly negative (−0.38), implying that competitiveness adjusts at a 38% speed of adjustment each quarter. A previously determined cointegrated relationship is, as expected, confirmed with estimated long-run coefficients as well as with an error correction term.

Distinguishing the effects of conventional from unconventional measures has been a key aspect in determining the fundamental repercussions of QE policy on monetary transmission channels. This has been the primary motive for authors to separately analyze MG and PMG estimator results on monetary autonomous and monetary non-autonomous countries. Table 3 presents estimates for seven EZ members during the 2014Q1–2023Q1 time horizon. Both MG and PMG estimators give similar results in terms of QE policy relevance in the long run. Short-run effects suggest that interest rate was a primary instrument in crisis reaction, which led to expected capital outflows and, consequently, positive change in reer. Looking at the long-run relationships, qe significantly affects competitiveness in terms of reer appreciation. The conventional monetary apparatus is not effective in the long run since the p-value is 0.680. This necessitates the need for unconventional measures. The lagged effects of QE policy dominate in EZ members since this policy needs time for excess liquidity to increase aggregate demand and prices. The error-correction term is significantly negative (−0.31), which confirms robust adjustment toward a long-term relationship.

By focusing on monetarily autonomous EU countries, we will try to shed light on how monetary adjustment functions in autonomous exchange rate regimes. In contrast to the EZ member countries, in the monetary autonomous regimes, the interest rate is not statistically significant for the impact on the reer, while the qe policy is, in the sense that the QE causes a depreciation of the reer in the short term (Table 4). In the long run, both instruments of monetary policy are statistically significant, with qe causing an expected reer appreciation, while interest rate cuts would cause reer to depreciate. EU members, who have retained monetary sovereignty, retain the effectiveness of both instruments of monetary policy in the long term, thereby expanding the room for maneuvering in order to influence the country’s competitiveness. While qe appreciates reer and worsens competitiveness, the lowering of the interest rate has the opposite effect, which can produce a balanced effect on reer, in contrast to EZ members, where the deterioration of competitiveness is inevitable in the long term. The presented results confirm the hypothesis (H1) assumption that a positive long-run relationship exists between QE and REER.

The heterogeneous coefficients presented in Table 5 confirm certain findings. In general, reer is not significantly affected by rir (except Hungary) or the central bank’s balance sheet expansion (except Belgium, Poland, and Hungary) in the short run. Insignificant short-run coefficients are in accordance with the estimated ARDL model. Investigating Belgium as a highly trade-dependent economy, there are statistically significant QE effects in the short run at a 5% threshold. This is because Belgium’s trade sector exhibits highly sensitive reactions to changes in monetary policy that can affect competitiveness. On the other hand, significant short-run effects for Poland are a reflection of their capital flow vulnerability. Balance sheet expansion policy in a monetary autonomous regime will lower interest rates and result in capital outflow; this is why Poland exhibits negative signs. Examining France as a core EZ member, the short-run interest rate dynamic exhibits weakly significant effects at the 10% threshold. Notably, structural rigidities, such as a rigid labor market and high production costs, coupled with its reliance on fiscal measures for competitiveness growth, suggest only marginalized effects of real interest rates in France. In contrast, statistically significant real interest rates in Hungary are indicative of their economic structure. Hungary’s economy is heavily reliant on FDI and external financing. Consequently, fluctuations in interest rates directly impact these flows. The results confirm the presence of time-lag effects when implementing unconventional monetary measures. Lagged effects come as a necessity since monetary policymakers need time to determine the appropriate amount of excess liquidity that serves as a demand driver. Focusing on the error-correction term, equilibrium adjustment is confirmed for every country (except France and Czech), while the speed of adjustment to a long-run relationship is higher for EZ members relative to monetary autonomous countries. This underpins the significance and amounts of asset purchases by the ECB implemented as an expansionary measure during crisis conditions but also the direct effects on competitiveness-worsening as a monetary transmission repercussion. Hypothesis (H2) is oriented toward the assumption that there are significant time-lag effects with the implementation of the QE policy. The presented short-run dynamic suggests, in general, the absence of short-run effects of QE policy, confirming the initial H2 assumption.

Further examining the external balance repercussions of unconventional monetary policy in Table 6, the panel ARDL model confirmed the relation between two variables with expected results. Since the Hausman specification test suggests that the null hypothesis of homogeneity cannot be rejected, the results support the PMG estimator as more efficient. Accordingly, asset purchases increase money market funds, thus increasing money supply. As a result, the higher demand generated will increase aggregate consumption, leading to higher prices and imports. Consequently, prolonged expansionary qe policy, if generating higher prices than nominal currency depreciation, will lead to external balance worsening. Findings suggest that short-term dynamics are, in general, not statistically significant. However, after adjustments of import and export variables with an inevitable time lag, qe will eventually have a negative effect on external balance in the long run.

Capturing the fundamental distinction between exchange rate regimes entails examining the external balance adjustment as a response to the QE policy. Table 7 presents current account adjustments to the QE policy for the seven EZ members with fixed exchange rate regimes. The panel ARDL model suggests an inverse relationship between the two variables: an increase in asset purchases injects more liquidity into the system, leading to consumption increases and a worsening of the external balance position. While the short-run coefficients are negative but statistically insignificant, indicating the presence of time-lag effects, the error-correction parameter is statistically significant at the 1% threshold. This suggests that implementing QE for EZ members to overcome internal shocks would entail a necessary trade-off with a worsening of the external balance. Stabilizing the internal balance would imply a worsening of the external position for fixed exchange rate EZ members. The Hausman test confirms that the null hypothesis cannot be rejected, thereby validating the PMG model.

Table 8 presents panel ARDL for the monetary autonomous EU economies. The long-run relationship between the two variables is negative, as expected, but statistically insignificant, unlike for the EZ members. Short-run dynamics also suggest that the immediate effects of QE on the external position are absent. The given results amplify the significance of monetary autonomy where the shock absorber function mitigates negative policy effects on the current account balance. The Hausman test confirms that the null hypothesis cannot be rejected, thereby validating the PMG model. The Panel ARDL model supports the hypothesis (H3) that the external position of EZ members deteriorates as a result of QE policy, in contrast to the external position of monetary autonomous EU countries.

Considering the policy rigidity imposed by the fixed exchange rate regime on EZ members, the long-term appreciation of the REER is evidently driven by QE. Policy reforms such as controlled exchange rate flexibility could potentially mitigate competitiveness losses for members experiencing prolonged exchange rate appreciation. Furthermore, increased fiscal and monetary policy coordination should be the primary focus when analyzing the macroeconomic consequences. Implementing a centralized fiscal policy framework at the EZ level, coupled with automatic stabilizers, would enable a progressive countercyclical approach to regions adversely affected by policies led by the ECB. Additionally, enhanced policy transparency would reduce delayed responses to crises and foster increased consumer trust. These reforms can enhance the resilience of monetary unions facing external shocks while simultaneously managing internal and external policy effects.

5.2. Robustness Check

5.2.1. Augmented Mean Group Estimator

In order to perform a robustness check, an alternative method of estimation is used. For confirmation of panel ARDL model results, we used the Augmented Mean Group (AMG) model. The model is appropriate for estimating dynamic heterogeneous macro-panels since it accounts for heterogeneity and cross-sectional dependence. AMG estimates regression for each cross-sectional unit but averages the coefficients to derive overall results. Coefficient averages are computed as unweighted means. Furthermore, the estimator includes correction terms for unobserved common factors that might influence all units with a heterogeneous impact, as well as the common dynamic process as an additional regressor (Eberdhardt, 2012). Mainly, we focus the analysis on the confirmation of three key factors: economic relation, statistical signification, and higher effect on the dependent variable. In general, the results (Table 9) suggest similar findings as with the PMG estimator. Both independent variables show a positive relationship with the dependent variable, as was expected due to previous confirmation of the same relations. The AMG estimator confirms worsening competitiveness as an incremental repercussion of the central bank’s asset purchases. Also, interest rates exhibit a positive relation in the sense that lowering interest rates results in improved competitiveness through nominal depreciation. Similar to PMG estimations, qe policy shows a higher coefficient relative to rir with statistical significance since the p-value is below 0.005. However, interest rates are not statistically significant since the p-value is above 0.05. Presented estimates confirm PMG findings that QE policy has negative repercussions through the REER channel of monetary transmission.

5.2.2. Dynamic Fixed Effects Estimator

To further confirm our findings, an alternative estimation method is used for robustness check. We used the Dynamic Fixed Effects (DFE) model as an appropriate alternative since it restricts the coefficients of the cointegrating vectors as well as the speed of adjustment to be equal (Blackburne & Frank, 2007). Similar to the PMG estimator, parameters in the DFE model suggest a positive long-run relationship between the presented variables. Error correction parameters, along with the homogeneous coefficients estimated by DFE, are statistically significant at the 10% level (Table 10). Even though the same dynamic of adjustment is presented, the magnitude of adjustment is lower in the model estimated by the DFE estimator (4.3% against 23.9%). Furthermore, between the DFE and PMG estimators, the Hausman test indicates that PMG is preferred over the DFE model. Since the DFE model showed that a homogeneous relationship is significant, with the same sign, we could claim that the results are stable.

6. Conclusions

This paper delves into the heterogeneity of the competitiveness (REER) and monetary adjustments of key EZ members as representatives of fixed exchange rate arrangements and monetary autonomous EU countries. REER dynamics are analyzed based on the influences of two relevant factors: (i) the central bank’s balance sheet expansion and (ii) conventional instruments of monetary policy such as interest rates. The fundamental problem for EZ members is the absence of sovereign monetary reaction needed for prompt adjustment to negative ECB policy repercussions. On the other hand, interest rates, as a conventional measure of policy, are not reliable supranational instruments in crisis circumstances anymore. For monetary autonomous countries, keeping the autonomy as a counter-cyclical buffer kept the adjustment mechanism unlocked in the long run for each instrument. However, in both cases, an adequate monetary–fiscal mix should maintain internal and external balance. In the case of this research, fiscal discipline would help control the worsening competitiveness caused by monetary instruments and keep the external position from failing.

The authors in this research try to understand the fundamental consequences of QE and RIR as independent variables in different exchange rate regimes on REER as a dependent variable. The estimated heterogeneous, dynamic, and non-stationary macro-panel of 11 European EU countries in the period 2014Q1–2023Q1 helps us to shed more light on short-run and long-run relations under the fixed, as well as flexible, exchange rate regime. Furthermore, the worsening of external balance is confirmed as a direct negative repercussion of QE policy during the same time horizon. Analysis of CSD, Pesaran’s PURT, and long-run cointegration tests were all performed as a pre-estimation procedure, while Hausman indicated the PMG estimator as a preferred one in both models. The authors discovered that monetary variables function with a certain time lag, thus, in the short run, relations are heterogeneous and weak. In the long run, causality and adjustments are supported. The key finding suggests that QE policy contributes to real effective exchange rate appreciation in the long run for both groups of countries with different exchange rate regimes. Fundamental differences are highlighted in the adjustment process based on monetary autonomy. For the EZ members, the fixed exchange rate amplifies the higher economic costs of REER appreciation for a longer period, unlike floaters, where the shock absorber function mitigates this problem. Furthermore, diminishing competitiveness contributes to the decline in the current account balance for the EZ members, highlighting the trade-off between internal and external balance. Conversely, a flexible exchange rate allows for a shorter adjustment process without the necessity for the worsening of external balance. These findings highlight the key problems for EZ members in terms of balance between internal and external repercussions.

For a sustainable European monetary union, it is of great purpose to lower these heterogeneous effects, especially in the short run. This can be achieved through further integration processes such as fiscal consolidation and political union. The absence of a key institutional framework underpins the fragility of EZ to macroeconomic shocks. A functional automatic stabilizer mechanism would imply prompt policy responses to competitiveness and external balance worsening.

Author Contributions

Conceptualization, S.S. and E.B.P.; methodology, E.B.P. and S.S.; software, E.B.P. and S.S.; validation, E.B.P. and A.S.; formal analysis, A.S.; investigation, S.S. and A.S.; resources, E.B.P.; data curation, S.S.; writing—original draft preparation, E.B.P. and S.S.; writing—review and editing, E.B.P. and S.S.; visualization, A.S. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data are contained within the article.

Conflicts of Interest

The authors declare no conflicts of interest.

Notes

1	All estimations are carried out in Stata 15.
2	Pre-estimation procedure can be additionaly requested.

References

Agba, E., Bennani, H., & Gnabo, J.-Y. (2022). Assessing the sources of heterogeneity in eurozone response to unconventional monetary policy. Applied Economics, 54(48), 5549–5574. [Google Scholar] [CrossRef]
Baumeister, C., & Benati, L. (2013). Unconventional monetary policy and the great recession: Estimating the macroeconomic effects of a spread compression at the zero lower bound. International Journal of Central Banking, 9, 165–212. [Google Scholar]
Beker Pucar, E., & Glavaški, O. (2020). Eurozone non-optimality: An OCA based analysis. Ekonomika, 66(2), 1–15. [Google Scholar] [CrossRef]
Beker Pucar, E., & Glavaški, O. (2021). Non-stationary heterogeneous panel approach of current account adjustments in the Euro-area. Economic Computation and Economic Cybernetics Studies and Research, 1(55), 185–199. [Google Scholar] [CrossRef]
Bernanke, B. (2020). The new tools of monetary policy. American Economic Review, 110(4), 943–983. [Google Scholar] [CrossRef]
Bhattarai, S., Chatterjee, A., & Park, W. Y. (2021). Effects of US quantitative easing on emerging market economies. Journal of Economic Dynamics and Control, 122, 104031. [Google Scholar] [CrossRef]
Blackburne, E., & Frank, M. (2007). Estimation of nonstationary heterogeneous panels. Stata Journal, 7(2), 197–208. [Google Scholar] [CrossRef]
Blanchard, O., Ostry, D. J., Atish, R. G., & Chamon, M. (2016). Capital flows: Expansionary or contractionary? The American Economic Review, 106(5), 565–569. [Google Scholar] [CrossRef]
Bleaney, M., & Tian, M. (2013). Net foreign assets, real exchange rates and net exports (CREDIT research paper, 13/04). The University of Nottingham. [Google Scholar]
Bowdler, S., & Radia, A. (2012). Unconventional monetary policy: The assessment. Oxford Review of Economic Policy, 28(4), 603–621. [Google Scholar] [CrossRef]
Buch, C., Bussiere, M., Goldberg, L., & Hills, R. (2018). The international transmission of monetary policy (NBER working papers, 24454). National Bureau of Economic Research. [Google Scholar]
Claeys, G., & Alvaro, L. (2016). The European central bank’s quantitative easing programme: Limits and risks (Bruegel policy contribution, 2016/04). Bruegel. [Google Scholar]
Curdia, V., & Woodford, M. (2011). The central-bank balance sheet as an instrument of monetary policy. Journal of Monetary Economics, 58(1), 54–79. [Google Scholar] [CrossRef]
Dedola, L., Georgiadis, G., Grab, J., & Mehl, A. (2020). Does a big bazooka matter? Quantitative easing policies and exchange rates. (European central bank research bulletin, 76). European Central Bank. [Google Scholar]
Eberdhardt, M. (2012). Estimating panel time-series model with heterogeneous slopes. The Stata Journal, 12, 61–71. [Google Scholar] [CrossRef]
Eichengreen, B., & Gupta, P. (2013). The real exchange rate and export growth: Are services different? (Policy research working paper series, 6629). World Bank Group. [Google Scholar]
Ferdinandusse, M., Freier, M., & Ristiniemi, A. (2020). Quantitative easing and the price liquidity trade-off (European central bank working paper series, 2399). European Central Bank. [Google Scholar]
Fidora, M., Giordano, C., & Schmitz, M. (2017). Real exchange rate misalignments in the Euro area (European central bank working paper series, 2108). European Central Bank. [Google Scholar]
Glavaški, O., & Beker Pucar, E. (2021). Real interest rate and exchange rate divergences within the EZ12: Evidence based at mean group estimator. Revista de Economia Mundial/Journal of World Economy, 58, 141–162. [Google Scholar] [CrossRef]
Glavaški, O., Beker Pucar, E., & Stojkov, S. (2022). Public revenues and public expenditure nexus: Evidence from Eurozone heterogeneity. Anali Ekonomskog fakulteta u Subotici, 58(48), 83–99. [Google Scholar] [CrossRef]
Gnimasoun, B., & Mignon, V. (2013). Current-account adjustments and exchange-rate misalignments. CEPII Research Centre. [Google Scholar]
Grauwe, D. P., & Ji, Y. (2024). How to conduct monetary policies. The ECB in the past, present and future. Journal of International Money and Finance, 143, 103048. [Google Scholar] [CrossRef]
Hashmi, A. R., & Nsafoah, D. (2024). International spillovers of conventional versus new monetary policy. Economic Modelling, 139, 106812. [Google Scholar] [CrossRef]
Hausman, J. A. (1978). Specification tests in econometrics. Econometrica, 46(6), 1251–1271. [Google Scholar] [CrossRef]
Hesse, H., Hofmannn, B., & Weber, J. M. (2018). The macroeconomic effects of asset purchases revisited. Journal of Macroeconomics, 58, 115–138. [Google Scholar] [CrossRef]
Hohberger, S., Priftis, R., & Vogel, L. (2019). The macroeconomic effects of quantitative easing in the euro area: Evidence from an estimated DSGE model. Journal of Economic Dynamics and Control, 108, 103756. [Google Scholar] [CrossRef]
Horvath, R., & Voslarova, K. (2016). International spillovers of ECB’s unconventional monetary policy: The effect on central Europe. Applied Economics, 49(24), 2352–2364. [Google Scholar] [CrossRef]
Ivanova, M. N. (2018). Quantitative easing: A postmortem. International Journal of Political Economy, 47, 253–280. [Google Scholar] [CrossRef]
Kelton, S. (2020). The deficit myth: Modern monetary theory and the birth of the people’s economy. PublicAffairs. [Google Scholar]
Kenourgios, K., Papadamou, S., & Dimitriou, D. (2015). On quantitative easing and high frequency exchange rate dynamics. Research in International Business and Finance, 34, 110–125. [Google Scholar] [CrossRef]
Kinateder, H., & Wagner, N. (2017). Quantitative easing and pricing of EMU sovereign debt. The Quarterly Review of Economics and Finance, 66, 1–12. [Google Scholar] [CrossRef]
Koijen, S. J. R., Koulischer, F., Nguyen, B., & Yogo, M. (2017). Euro-area quantitative easing and portfolio rebalancing. American Economic Review, 107(5), 621–627. [Google Scholar] [CrossRef]
Kucharcukova, O. B., Claeys, P., & Vasicek, B. (2016). Spillover of the ECB’s monetary policy outside the Euro area: How different is conventional from unconventional policy? Journal of Policy Modelling, 38(2), 199–225. [Google Scholar] [CrossRef]
Park, D., Ramayand, A., & Shin, K. (2015). Capital flows during quantitative easing: Experiences of developing countries. Emerging Markets Finance and Trade, 52(4), 886–903. [Google Scholar] [CrossRef]
Pejčić, J., Glavaški, O., & Beljić, M. (2024). Driving forces of the consumer price index during the crises in the Eurozone: Heterogeneous panel approach. Economies, 12, 292. [Google Scholar] [CrossRef]
Persyn, D., & Westerlund, J. (2008). Error-correction based cointegration tests for panel data. The Stata Journal, 8, 232–241. [Google Scholar] [CrossRef]
Pesaran, M. H. (2007). A simple panel unit root test in the preference of cross section dependence. Journal of Applied Econometrics, 22, 265–312. [Google Scholar] [CrossRef]
Pesaran, M. H., Shin, Y., & Smith, R. P. (1999). Pooled mean group estimation of dynamic heterogeneous panels. Journal of the American Statistical Association, 94, 621–634. [Google Scholar] [CrossRef]
Pesaran, M. H., & Smith, R. P. (1995). Estimating long-run relations from dynamic heterogeneous panels. Journal of Economics, 68, 79–113. [Google Scholar] [CrossRef]
Pierluigi, B., & Sondermann, D. (2018). Macroeconomic imbalances in the Euro area: Where do we stand? (European central bank occasional paper, 211). European Central Bank. [Google Scholar]
Rohit, A. K., & Dash, P. (2019). Dynamics of monetary policy spillover: The role of exchange rate regimes. Economic Modelling, 77, 276–278. [Google Scholar] [CrossRef]
Stefanski, M. (2022). Macroeconomic effects and transmission channels of quantitative easing. Economic Modelling, 114, 105943. [Google Scholar] [CrossRef]
Stojkov, S., Beker Pucar, E., Glavaški, O., & Beljic, M. (2023). Oil price spillover effects to the stock market sentiment: The case of higher vs. lower oil import EU countries. Economies, 11(11), 279. [Google Scholar] [CrossRef]
Weale, M., & Wieladek, T. (2016). What are the macroeconomic effects of asset purchases? Journal of Monetary Economics, 79, 81–93. [Google Scholar] [CrossRef]
Williamson, D. S. (2012). Liquidity, monetary policy, and the financial crisis: A new monetarist approach. American Economic Review, 102(6), 2570–2605. [Google Scholar] [CrossRef]

Figure 1. APP structure and ECB balance sheet. (a) Largest asset purchases by ECB per country in the period 2014–2023; (b) ECB balance sheet assets. Source: author’s research according to FRED (2024) data.

Figure 2. Fluctuation of REER and ECB balance sheet dynamics in the EZ member states (Austria, France, Germany, Netherlands, Italy, Spain, and Belgium) in the period 2014Q1–2023Q1. Source: author’s research according to quarterly IMF and FRED (2024) data.

Figure 3. Fluctuation of REER and central bank’s balance sheet dynamics in the monetary autonomous EU countries (Czech, Poland, Hungary, and Romania) in the period 2014Q1–2023Q1. Source: author’s research according to quarterly IMF and FRED (2024) data.

Figure 4. Average current account position (% of GDP) for EZ members in the period 2014Q1–2023Q1. Source: author’s research according to quarterly OECD data.

Table 1. Pesaran CD test, Pesaran CIPS test, and Westerlund cointegration test1.

Sample: 11 EU Economies; Period 2014Q1–2023Q1
	(a)		(b)					(c)
Variables	Pesaran CD Test	p-Values	Lags	Pesaran (PURT) Panel Unit Root Test in the Level	p-Values	Pesaran (PURT) Test at the First Differences	p-Values	Westerlund Cointegration Test	p-Value
reer	10.74	0.000	0	−0.825	1.000	−4.945	0.000	−4.216	0.000
			1	−0.625	1.000	−3.453	0.000
			2	−0.502	1.000	−2.682	0.001
lnqe	41.39	0.000	0	−0.797	1.000	−5.316	0.000
			1	−0.680	1.000	−3.530	0.000
			2	−0.768	1.000	−2.210	0.060
rir	37.75	0.000	0	−1.638	0.679	−3.988	0.000
			1	−1.809	0.446	−3.195	0.000
			2	−1.616	0.707	−2.262	0.041