A monitor analysis of the impact of the ECB’s
unconventional monetary policy in response to the crisis
Name :
Esther Hoefsmit
Supervisor: Ron van Maurik
Date:
17-02-2017
Bachelor thesis Economics and Finance
Abstract: This thesis investigates the impact of the ECB’s unconventional monetary policy in
response to the crisis, by monitoring the transmission of the ECB’s implemented policy into the real economy from different perspectives. Based on this monitor analysis, has the ECB reduced negative effects of the crisis and the economy seems slightly growing. However, several factors seems to have complicated the transmission of the ECB’s implemented policy into the real economy. In particular, asymmetries across countries in the euro area have hampered the transmission. In addition, it seems that a part of the monetary expansion remains ‘stuck’ in the banking sector. The results suggest that the ECB has inhibit negative effects of the crisis, rather than stimulating the euro area economy.
Hierbij verklaar ik, Esther Hoefsmit, dat ik deze scriptie zelf geschreven heb en dat ik de volledige verantwoordelijkheid op me neem voor de inhoud ervan.
Ik bevestig dat de tekst en het werk dat in deze scriptie gepresenteerd wordt origineel is en dat ik geen gebruik heb gemaakt van andere bronnen dan die welke in de tekst en in de referenties worden genoemd.
De Faculteit Economie en Bedrijfskunde is alleen verantwoordelijk voor de begeleiding tot het inleveren van de scriptie, niet voor de inhoud.
Contents
1 Introduction
4
2 The ECB’s unconventional monetary policy
6
3 Literature review
9
3.1 Conducted studies
9
3.2 Transmission mechanism
14
3.2.1 Portfolio rebalancing channel
14
3.2.2 Expectations channel
16
3.2.3 Credit channel
16
3.2.4 Exchange rate channel
17
4 Method
19
5 Results
22
5.1 Monitor of the euro area
22
5.2 Monitor across countries in the euro area
25
5.3 The ECB’s response to the crisis
27
5.4 The ECB’s monetary policy compared to the Fed, BoE and BoJ
28
6 Conclusion
30
7 Discussion
32
8 Appendix
33
8.1 Euro area
34
8.2 Across countries in the euro area
41
8.3
Related to the crisis
50
8.4 The ECB compared to the US, UK and Japan
51
9 References
56
1 Introduction
The European Central Bank (ECB) implemented an unconventional monetary policy in response to the global financial crisis in 2007. This unconventional monetary policy includes mostly large monetary expansions. This expansionary policy have been extended many times. More than nine years after the start of the first programme, the ECB is still easing money and no end date is specified. The magnitude and duration of the ECB’s expansionary policy are subject to much criticism. For instance, by the DNB president Klaas Knot and the Bundesbank president Weidmann1 (DNB, 2016; Hank & Thier, 2016).
In response to the global financial crisis, the ECB and the three other most influential central banks of the world implemented an unconventional monetary policy. The impact of large monetary expansions in response to the global financial crisis has been examined in several studies. Because the transmission of unconventional monetary policy into the real economy is not clear, are conducted studies divided into different divisions. Some studies aimed to quantify the effects into the (real) economy, and other studies are rather focused on unconventional monetary policy in general.
In order to assess the impact of the ECB’s unconventional monetary policy, I include both. Since the ECB cannot influence the real economy directly, I will first assort how the ECB attempted to influence the real economy with the implemented policy. Based on the ECB’s objective, conducted studies about the general transmission and conducted studies about the historical evidence in the euro area, I will assess how the ECB’s implemented policy could have been transmitted into the economy. On the basis of this analysis, I will monitor whether these developments have actually occurred.
Conducted studies that include a monitor analysis in order to determine developments as a result of the ECB’s unconventional monetary policy are Dunne, Everett and Stuart (2015), and Demertzis and Wolff (2016). Besides, Joyce, Miles, Scott and Vayanos (2012) made use of a monitor study to assess the effects in the UK. In addition, several central banks conducted or published similar studies and some empirical studies made use of graphical analyses as a complement. These conducted studies monitored in particular general developments related to a small part of the monetary
expansion. Because many empirical literature took the crisis into account and with the knowledge that members of the Governing Council and General Council of the ECB itself have expressed their doubts about the effectiveness, I expect the transmission is not straightforward and there may be other factors that hamper economic recovery. In order to obtain a finer view of the impact, I attempt to assess what hamper the transmission in the euro area by taking the crisis into account and by analyzing
developments across countries. In order to put the impact in the euro area in perspective, I will compare the euro area with developments in the US, UK and Japan
1 Knot and Weidmann hold both seats on the Governing Council and General Council of the ECB. In addition, they are also member of the
Financial Stability Board and governor of the International Monetary Fund (De Nederlandsche Bank, 2017; Deutsche Bundesbank, 2017).
4
This study suggests that several factors have complicated the transmission of the ECB’s implemented unconventional monetary policy. In particular, asymmetries across countries in the euro area have hampered the transmission. In addition, it seems that a part of the monetary expansion remains ‘stuck’ in the banking sector. Nevertheless, the ECB has inhibit negative effects of the crisis and the economy seems slightly growing.
The remainder of this study is organized as follows. Section 2 presents the ECB’s
implemented unconventional monetary policy since the global financial crisis. Section 3 includes a literature review of conducted studies as well as a general consensus of the transmission mechanism of unconventional monetary policy into the real economy. Section 4 presents the method and section 5 the results. The results are organized in four parts from different perspectives. The first part outlines the developments based on the transmission mechanism. The second part focuses on the transmission across countries in the euro area, and, in particular, asymmetries in the transmission. The third part takes the crisis into account and the fourth part compares the euro area with the US, UK and Japan. Section 6 concludes and section 7 outlines a discussion. Section 8 includes the appendix with the figures. The references are in section 9.
2 The ECB’s unconventional monetary policy
The primary objective of the Eurosystem is to achieve price stability (ECB, 2017a). This is defined by the Governing Council of the ECB as a “year-on-year increase in the Harmonized Index of Consumer Prices for the euro area of below 2%” (ECB, 2017b). The Eurosystem adds to this, that the general monetary policies are for the purpose of contributing to the achievement of full employment and balanced economic growth (ECB, 2017c).
The ECB cannot influence their primary objective directly, instead it can influence
intermediate targets (e.g. monetary aggregates, market interest rates, exchange rate). The ECB can use different instruments. The conventional way to influence inflation is by setting the key interest rates. This is no longer an option since the key interest rates in the euro area have come near to their effective lower bound. Consequently, the ECB implemented an unconventional monetary policy to limit the consequences of the global financial crisis started in 2007. This unconventional monetary policy taken by the ECB comprises mainly large asset purchases, but negative interest rates as well. The Federal Reserve Bank (Fed), Bank of England (BoE), and the Bank of Japan (BoJ) implemented a similar unconventional monetary in response to the global financial crisis. The monetary expansion is partly internationally coordinated (G20 London Summit, 2009). Figure 1 in the appendix gives an overview of the expansion of the ECB’s balance sheet.
The ECB’s implemented policy contains many programmes, adjustments of programmes and changes in regulation. Therefore, this section is focused on the main programmes and changes.
First, the ECB started in 2007 with the implementation of ‘long-term refinancing operations’ (LTRO). The maturity of the LTRO’s has been extended several times (together approximately €1 trillion). In 2008 started the ECB with ‘main refinancing operations’ (MRO). As the name implies, LTRO’s were focused on providing liquidity in the long-term. The MRO’s were used to manage liquidity, to control interest rates in the short-term and to signal the stance of the monetary policy (ECB, 2017d).
In 2009 started the ECB with the first ‘covered bond purchase programme’ (CBPP). The aim of CBPP was to support banks that had been affected by the crisis (€60 billion aggregate volume within a period of 1 year, non-sterilized) (ECB, 2010). In 2010 started the ECB with the ‘Securities Markets Programme’ (SMP). This programme included public- and private debt securities to ensure liquidity in dysfunctional markets (Greece, Portugal, Ireland, Spain and Italy; approximately €220 billion, sterilized) (ECB, 2017e). In 2012 stopped the SMP and the ‘Outright Monetary Transactions’ (OMT) were announced. Preceding the announcement of OMT, the president of the ECB Mario Draghi made a remarkable statement in his speech: “Within our mandate, the ECB is ready to do whatever it takes to preserve the euro. And believe me, it will be enough.” (ECB, 2012a). This
statement is often repeated in journals and newspapers. The OMT allows the ECB to purchase limitless of government bonds from countries that are already in the European Stability Mechanism (ESM). An import element of OMT was, that is was made conditional upon a structural IMF adjustment programme (ECB, 2012b).
In 2011 introduced the ECB the CBPP2, but this programme was stopped in 2012 (initially target was €40 billion, but €16.418 billion was purchased). The aim was to ease funding conditions for enterprises and credit institutions, and encouraging institutions to maintain and expand their lending (ECB, 2012c). In 2013, the ECB altered the monetary policy strategy with ‘forward guidance’ to deal with the macroeconomic consequences of the crisis. The aim of forward guidance was to curb interest rate volatility and to anchor interest rate expectations (ECB, 2013). In 2014 started the ECB with the implementation of ‘targeted longer-term refinancing operations’ (TLTRO). This programme was implemented in order to stimulate bank lending to the real economy and to further ease private sector credit conditions (ECB, 2017f).
On 22 January 2015 announced the ECB the ‘expanded asset purchase programme’ (APP). This programme includes the ‘asset-backed securities purchase programme’ (ABSPP) and CBPP3 (already started in the end of 2014), and the ‘public sector purchase programme’ (PSPP). The purchases under this programme started in March 2015 and the end date would initially be in
September 2016. However, the ECB Governing Council already indicated that the programme would continue until the primary objective is reached. Under the APP, the Eurosystem purchased bonds from governments in the euro area and securities from national agencies and European institutions. Initially, the APP began in March 2015 with the purchase of €60 billion assets monthly (ECB, 2015). These purchases are allocated to several assets. Initially, the monthly allocation was constructed as follow:
So the purchases are allocated to: debt of supranational institutions in the euro area; sovereign debt securities held by national central banks (NCBs) and the ECB; bonds issued by national agencies in
Germany, France and Spain; covered bonds (CBs) and asset-backed securities (ABSs). Some
allocations are based on ‘capital keys’ to compute the share of a NCB in the capital, based on GDP. In addition, the holdings have an issuing limit to keep the ECB neutral and the purchases must meet several criteria (ECB, 2017g).
The duration of the APP has been extended and criteria have been adjusted a several times. In March 2016 announced the ECB to expand the APP further with an increase of asset purchases to €80 billion per month and with the ‘corporate sector purchase programme’ (CSPP). The CSPP includes investment-grade euro-denominated bonds issued by non-bank corporations established in the euro area (ECB, 2016a). However, in December 2016, the further increase to €80 billion per month was changed to €60 billion per month and the asset purchases and the APP is extended until at least December 2017. Draghi emphasized that there is there is “...no question about tapering.” (ECB, 2016c).
3 Literature review
Many studies have been conducted to assess the impact of unconventional monetary policy, however the results and the methodology to assess the impact are mixed. Section 3.1 is a synthesis of conducted studies into the impact of the ECB’s implemented unconventional monetary policy. Section 3.2 takes studies abroad the euro area into account to assess the broad consensus of the transmission of
unconventional monetary policy into the real economy.
3.1 Conducted studies
Several studies evaluated the impact of the unconventional monetary policy implemented by the ECB. Researchers noticed that the effect of one programme is related to effects of other programmes, but some researchers focused rather on one programme. In addition, some studies are focused on specific segments, and others attempted to assess the effect to the real economy. However, Joyce et al. (2012) mention the difficulty to quantify the effects of unconventional monetary policy due the lack of historical evidence. They remarked that an event study analysis is not appropriate given the time-lag involved by the transmission to inflation and output. Nevertheless, an increasing literature attempt to determine the effects in the broader economy.
Peersman (2011) was one of the first who studied the impact of the ECB’s unconventional monetary policy. He included the period before the crisis (1999-2009), to compare both, conventional and unconventional monetary policy. Based on his findings, have had both policies a significant effect on economic activity and inflation. The study examined by Baumeister and Benati (2010) included a part of the same period, but their study was rather focused on the effect of the implemented unconventional monetary policy in context of the crisis (2007-2009). Besides, they focused on the effect of a
compression in the long-term bond yield spread into the macro economy. This effect is related to the ‘portfolio rebalancing channel’. Baumeister and Benati (2010) made a comparative analysis and concluded that the compression exerted a strong effect on inflation and output growth.
Gibson, Hall and Tavlas (2016) examined also the effect of bond yields, but without assessing whether this effect on yields have been transmitted into the real economy. They focused on the effects of CBPP, SMP and CBPP2 (2009-2012) on covered bond prices and sovereign bond spreads for the GIIPS countries (Greece, Ireland, Italy, Portugal and Spain). They concluded that the SMP modestly reduced spreads on sovereign bonds, but they noticed that the effect of Draghi’s speech in 2012 appears to have had a larger significant effect than SMP. Besides, they concluded that CBPP and CBPP2 have had an increasing effect on covered bond prices, although, also modest of magnitude. The impact of SMP on sovereign bond yields of the GIIPS countries are also examined by Eser and Schwaab (2016). Their results are comparable to the results of Gibson et al. (2016). However,
Eser and Schwaab (2016) made an additional remark that, notwithstanding of the large effects, the SMP was not sufficient to end the sovereign debt crisis.
Altavilla, Carboni and Motto (2015) evaluated the effect of the APP and concluded that the programme significantly lowered yields, reduced the term premium and supported the credit channels. This is in line with Demertzis and Wolff (2016), who concluded that the lowered long-term yields is one of the two major channels through which APP operates. Andrade, Breckenfelder, De Fiore, Karadi and Tristani (2016) also concluded that the APP had an desirable effect through the ‘portfolio
balancing channel’. The findings of Albertazzi, Becker and Boucinha (2016) suggest also that the effect through the ‘portfolio rebalancing channel’ for the transmission of the APP was significant, but only in the economies which were more affected by the crisis. Briciu and Lisi (2015) examined the effects of the main expansionary programme announcements till January 2015 and concluded as well that the effect of the APP was significant in lowering government bond yields (besides the (more) significant effects of the APP, but this will be explained later).
Wieladek and Garcia Pascual (2016) also included more transmission channels in their analysis, but they argued that their framework was not competent to explicit quantify the contribution of all the transmission channels. They focused on the period from June 2012 till April 2016 and attempted to assess the effect of the ‘portfolio balance channel’ (by analyzing the term premium of yields), the ‘credit easing channel’ (by analyzing interbank rates), the ‘signalling channel’ (by analyzing the spread of the future rate), the ‘uncertainty channel’ (by analyzing market volatilities in response to announcements), and the ‘exchange rate channel’ (by analyzing fluctuations in the effective exchange rate). They found support for the ‘portfolio balance channel’, the ‘credit easing channel’, the ‘signalling channel’, and the ‘exchange rate channel’. Especially the effect of the ‘portfolio balance channel’ was powerful. They did not study whether these particular channels have been transmitted into the real economy, but they assessed what core CPI and real GDP would have been in absence of the APP programme (till April 2016). Based on their study would core CPI have been 0.9% lower and would real GDP have been 1.3% lower in the euro area. The effects on real GDP was largest in Spain and smallest in Italy.
Other studies attempted as well to assess the effects from another perspective than the ‘portfolio rebalancing channel’. Several studies included the effect of the announcement of programmes or expectations. Bernoth, Hachula, Piffer and Rieth (2016) included this effect by analyzing unconventional ‘surprise’ expansions from 2007 till March 2016. Their results suggest that
unconventional ‘surprise’ expansions are effective in increasing inflation, stimulating real activity and reducing unemployment. Briciu and Lisi (2015) analyzed the announcement effect of expansions in the balance sheet of the ECB from October 2008 till January 2015. The main programmes started in this period are: MRO, expansions in maturity of LTRO, CBPP, SMP, CBPP2, OMT, TLTRO, ABSPP, CBPP3 and the announcement of APP. Only the effects of CBPP2, SMP, LTRO and APP were
significant in increasing market inflation expectations. Andrade et al. (2016) also found empirical evidence for the influence of expectations created with the announcement of the APP. Their evidence suggest that the APP have contributed to bring the long-term inflation expectations closer to the primary objective of the ECB.
Eser and Schwaab (2016) found that on the one hand, the announcement effect associated with SMP was large, but on the other hand, the signalling effect of future low interest rates was not
significant. This is not in line with the results of Sahuc (2016). His results suggest that the
macroeconomic impact of asset purchases are potentially influential when the porgramme is associated with ‘forward guidance’ on interest rates. However, he argues that he used a DSGE model and this model tend to overestimate the impact of ‘forward guidance’. Nonetheless, he concluded that the ‘signalling channel’ remains the prominent transmission channel. Sahuc (2016) focused on the APP and Eser and Schwaab (2016) on SMP, therefore, the importance of ‘forward guidance’ may depends on other factors as well.
The favorable effects on expectations are not in line with the study of Van den End and Pattipeilohy (2015). They analyzed the empirical effects on the size and composition of the ECB’s balance sheet (2007-2014), and found no substantial effect on long-term inflation expectations in the euro area and a negligible effect on inflation expectations. Briciu and Lisi (2015) emphasize that their results are based on event studies, and, therefore, these result are ‘snapshots’ of the immediate effects and not capture the full effects over time. The consideration of the announcements is a difference with the study of Van den End and Pattipeilohy (2015. Therefore, the announcement effect on expectations may be overestimated by Briciu and Lisi (2015), among others, or underestimated by Van den End and Pattipeilohy (2015).
The influence of the exchange rate, as a result of the unconventional monetary policy, have also been examined. Briciu and Lisi (2015) concluded that, besides the significant announcement effect of expansions in the ECB’s balance sheet related to APP on bond yields and on market inflation
expectations, the most powerful effect of APP was on the exchange rate. Wieladek and Garcia Pascual (2016) also found support for the ‘exhange rate channel’ and according to Demertzis and Wolff (2016) was this one of the two major channels through which APP operated.
This favorable evidences for the effect on the exchange rate, are not in line with the study examined by Bernoth et al. (2016). Their study suggests that, on average, a ‘surprise’ monetary expansion impulse have led to an appreciation of the euro. However, Bernoth et al. (2016) focused on the EUR/USD exchange rate and Wieladek and Garcia Pascual (2016) focused on the effective
exchange rate. This could have resulted in different conclusions. Besides they used a different methods to determine the effects, this may also have played a role.
The support for the ‘credit easing channel’ on the basis of the study carried out by Wieladek and Garcia Pascual (2016) is in line with the conclusion of Boeckx, Dossche and Peersman (2014). They examined the macroeconomic effects by focusing on changes of the ECB’s balance sheet (as a result of the monetary expansions) and concluded that an expansionary balance sheet shock stabilized financial markets, stimulated bank lending and had a positive impact on economic activity and prices. However, they also found that the effects on bank lending and output have been smaller in countries that have been more affected by the crisis. According to Boeckx et al. (2014) are these countries the Netherlands, Greece, Portugal, Spain and Cyprus. According to Demertzis and Wolff (2016) have lending, investment and housing increased a bit, but this was not a major channel through which APP have operated. Besides, Joyce et al. (2012) argued that instead of lending to the private sector, banks purchased government bonds with higher yields and deposited their liquidity
Giannone, Lenza, Pill and Reichlin (2012) focused on effects in the interbank market in times of crisis. They found a small significant effect on loans and on real economic acitivity (decrease in unemployment with 0.6%-points and increase of industrial production with 2%).
In addition, Baumeister and Benati (2010) compared their findings for the euro area with developments in the US, UK and Japan. They concluded that, compared to the US and UK, the monetary policy in the euro area do not seems to have been very powerful. According to Van den End and Pattipeilohy (2015) have had shocks in the composition or size of the balance sheet no substantial effect on long-term inflation expectations in the euro area, compared to the US, UK and Japan. Besides, Wieladek and Garcia Pascual (2016), argued, that the effect of the ‘uncertainty channel’ seems to have been strong in the US and UK, but the effect of uncertainties did not operated in the euro area.
Joyce et al. (2012) argued that analyses are difficult due the shortage in historical evidence, because they rely on constructing model-based policy or on counterfactual analysis. In addition, the estimates are based on different models, which makes comparison less pure. For instance, Van den End and Pattipeilohy (2015) and Bernoth et al. (2016) made a simulation with a vector autoregressive (VAR) model. Baumeister and Benati (2010), Wieladek and Garcia Pascual (2016) and Giannone et al. (2012) used a Bayesian vector autoregressive (BVAR) method. Boeckx, Dossche and Peersman (2014) used a structural vector autoregressive (SVAR) model. Sahuc (2016) assessed the effects in a dynamic stochastic general equilibrium (DSGE) model. Demertzis and Wolff (2016) based their findings on a monitor analysis. Gibson et al. (2016) used an autoregressive moving average (ARMA) model and remarked that their approach differs from previous studies. While some studies used dummy variables to capture the effects, did they used actual amounts. Besides, they mention the influence of rating downgrades on spreads during the crisis Many studies mentioned the importance of taking the crisis into account.
Eser and Schwaab (2016) argued that some studies used approaches in which they neglected the presence of important common factors, like the sovereign debt crisis. Besides they argued that, because the purchases of SMP were undertaken during the most intense phase of the sovereign debt crisis in the euro area, comparisons with other countries is difficult. Besides, Altavilla et al. (2015) mentioned that it is difficult to compare the APP with previous programmes, because, in contrast to the previous programmes, the APP was implemented during low financial distress and spreads were already lowered.
Moreover, the large quantity of the monetary expansion is without dangers. Andrade et al. (2016) and Demertzis and Wolff (2016) mentioned the potential of risks which may undermine the effectiveness. Because of the convergence of spreads, are the profits for banks and other institutions reduced. In addition, the ECB has the risk of losing capital and housing markets have potential risks. Joyce, Miles, Scott and Vayanos (2012), among others, argued that the increase in money supply may not lead to a decrease in bond yields (liquidity trap). These potential risks are mentioned among others, however, the included studies about the effects of the ECB’s unconventional monetary policy, did not accentuated the threat of potential risks.
Difficulties in the assessment of the impact of the ECB’s unconventional monetary policy raised due uncertainties about the transmission of this policy into the euro area, the lack of historical evidence, the precise impact of the crisis and the potential ‘unknown’ risks. Probably due these difficulties and due differences in estimations are the results mixed.
Overall, based on these studies, it seems that the effects into the real economy is positive. However, studies differ about the degree of effects into the real economy. In addition, most studies agreed about the desirable effects related to the ‘portfolio rebalancing channel’, although, the studies are divided into what extent this effect on yields and asset prices have been transmitted into the real economy. The effect of the increase in banks liquidity seems significant, but not the most powerful transmission channel. Conducted studies are mixed about the effects related to expectations and the exchange rate.
3.2 Transmission mechanism
The ECB cannot influence the inflation directly. The transmission mechanism describes the process in what ways the ECB can influence inflation. The scheme below illustrates this process in broad lines. The blue figures denote, the direct impact of the ECB and the orange figures denote other influences which may affect inflation outside the control of the ECB.
The process of the transmission mechanism are modeled in various transmission channels. The
literature is not clear about the transmission channels and named them different. According to Joyce et al. (2012) among others, is the debate at least as large as to which channels play a role, as the impact of massive asset purchases. However, it boils down to four broad transmission channels that are expected to be most influential: (1) portfolio rebalancing channel, (2) expectation channel, (3) credit channel and (4) exchange rate channel.
3.2.1 Portfolio rebalancing channel
The basis of the portfolio balance channel is created by Tobin (1961; 1963; 1969) and extend by Brunner and Meltzer (1973). With the asset purchases, the central bank increases the monetary base. Due the higher demand of these assets, investors are encouraged to rebalance their portfolios with close substitutes. However, according to the ‘preferred-habitat theory’, some investors prefer particular yield curves so they will pay a higher price (Modigliani & Sutch, 1966). In addition,
Vayanos and Villa (2009) noticed the ‘duration effect’ of yield curves. The large quantity of long-term bonds purchases decrease the returns of these bonds relative to expectations about future short-term interest rates. This lowers the term premium and this is the duration effect. This results in flattening yield curves and increasing asset prices.
Tobin (1969) described the impact of asset purchases on yields curves through valuation of equity by firms. This is modeled by Mishkin (2013): the expansionary monetary policy (MP) will result in an oversupply and this excess of money will be spent in stocks. In addition, the excess supply of money reduces bonds yields (lower interest rates ir ), which also makes equity more attractive. The increased
demand results in higher prices (Pstock), and consequently the market value of firms relative to
replacements cost of capital (q) rise, more investment spending (I) and hence an increase in aggregate demand (Y). Besides, Bosworth (1975) argued that the increase in stock prices reduces the yields and this makes investments financed by issuing equity more profitable.
MP↑ir↓ Pstock↑q↑I↑Y↑
However, according to Wallace (1981), are assets neutral and, therefore, perfect substitutes. In the case of ‘Wallace Neutrality’ and if interest rates are close to the zero bound, large monetary
expansion would have no influence on yields. This situation, in which the increase in monetary base is ineffective in increasing spending and economic activity, is named a ‘liquidity trap’ (Joyce, Miles, Scott, & Vayanos, 2012). Economists argues that quantitative easing is indeed not effective if the interest rates are at the zero lower bound due substitutability of bonds and money. However, According Eggertsson and Woodford (2003) and Krugman (2000), it is possible to avert a liquidity trap if the monetary policy is credible, because this creates higher inflation expectations (also related to the expectation channel) (Eggertsson & Krugman, 2012; Eggertsson, Ferrero, & Raffo, 2014) .
Other conditions for the portfolio rebalancing channel are: heterogeneity across investors and credit restrictions (Joyce et al., 2012). This means that prices need to change and investors have to hold different portfolios. The need for credit restrictions refers to the ‘Ricardian Equivalence
Theorem’, because without credit restrictions people expect the public debt spending will be offset by future taxations (also related to the expectation channel) (Barro & Gordon, 1983). This results in a lower present value wealth, which reduce spending.
In addition, sometimes economists make a distinction of the effects through the portfolio rebalancing channel. Mentioned channels are: ‘scarcity (risk) channel’, ‘duration channel’, ‘term premium’, ‘asset valuation’ (Joyce et al., 2011; Joyce & Tong, 2012; D’Amico, English, López-Salido, & Nelson, 2012; Andrade et al., 2016; among others)
3.2.2 Expectation channel
Expectations play an important role in the transmission. The effects of expectations are described in different ways. These way can be broken down into two categories. The first is the ‘signaling channel’ and refers to the future interest expectations. The central bank can influence expectations by providing information about their future monetary policy in a ‘forward guidance’. Investors adjust their
investments if they expect an expansion of the monetary policy and the lower bounds remains close to the zero lower bound. The second effect covers the effect of inflation expectations. These are known as the ‘inflation channel’, ‘confidence channel’ and ‘uncertainty channel’ (Krishnamurthy & Vissing-Jorgensen, 2011, 2013; Mishkin, 2013; among others).
Investors adjust their investments if they expect an expansion of the monetary policy and the lower bounds remains close to the zero lower bound. As mentoined, is the credibility of the monetary policy of the central bank important. This can be achieved by purchasing a large quantity assets with a long duration. Because if the central bank would raise the interest rates in the future, they will incur enormous loses on the assets they purchased (Clouse, Henderson, Orphanides, Small, & Tinsley, 2000; Bernanke, Reinhart, & Sack, 2004; Krishnamurthy & Vissing-Jorgensen, 2011).
However, too much transperancy would be determental to welfare (Morris & Shin, 2002). This finding provides much debate aabout the effectiveness of transparancy. In addition, expectations can cause an overshoot in inflation. But, according to Eggertson and Woodford (2003), accelarates inflation overshooting economic recovery (at the interest rate is at the zero lower bound). Such as accelarating the increase in employment on the long run (Erceg & Levin, 2014; Rudebusch & Williams, 2016).
Mishkin (2013) describes the influence of expectations in the ‘interest rates-channel’. He called this channel the key of the monetary transmission mechanism. If the nominal interest rates are zero and the ‘Taylor rule’ (ir=i- π
e
) is taken into account, then expectations should be transmitted as follow:
MP↑Pe↑πe↑ i
r↓I↑Y↑
3.2.3 Credit channel
The credit channel includes the ‘bank lending channel’ and the ‘balance sheet channel’. A central bank can ‘create’ money by purchasing government securities from banks (Dunne et al., 2015). In addition, if asset prices increase through the monetary expansion, the asset of banks will increase too and hence the value of banks (Deutsche Bundesbank, 2016). This encourage banks to provide credit by funding more loans and to meet stricter capital requirements. This leads to more investments (I) and
consumption (C), which ultimately increase inflation (bank lending channel). Mishkin (2013) modeled this process:
MP↑ liquidity for bank deposits ↑loans supply↑I↑C↑Y↑
Another result of the increase in asset prices, is that the increase in net worth of firms will reduce the credit default risk and because of this, the risk premium will be reduced (Deutsche Bundesbank, 2016). So this will reduce asymmetric information problems (moral hazard and adverse selection) and this stimulates lending, investments and finally this will raise the inflation (balance sheet channel). Mishkin (2013) modeled this process:
MP↑Pstock↑net worth of firms↑asymmetric information↓lending↑I↑Y↑ According to Mishkin (2013), the cash flow channel and household liquidity effect are part of the balance sheet channel. The contribution to inflation of both is as follows:
MP↑i↓Cash flow↑ asymmetric information↓lending ↑I↑Y↑ MP↑Pstock↑financial assets↑likelihood of financial distress↑
housing expenditure and consumer durable↑Y↑
In addition, this process increase households wealth (W) and this will also contribute to inflation (Mishkin, 2013).
MP↑ ir↓ Pstock↑W↑C↑Y↑
However, the effect of bank funding is possible not strong, because banks would increase their liquidity and insure against risk instead funding loans. But Joyce et al. (2012) argued that if the increase in liquidity is in the form of longer term funds, banks will probably expand their lending.
In addition, the money multiplier model predict to what extent the money supply will change for a given change in monetary base. As a result of large asset purchases, banks deposit these reserves (excess reserves ratio increases a lot) and this will lower the money multiplier (Mishkin, 2013). This means that the money supply will expand slightly, despite the large increase the monetary base, and this makes large monetary expansions less effective.
3.2.4 Exchange rate channel
If the exchange rate is flexible, a depreciation of the exchange rate (E↑) makes domestic goods and assets more competitive compared to the foreign. As a result, the net export (NX) increases and hence the inflation rises. This process is based on the IS-LM-BP model developed by Mundell (1963) and Fleming (1962). Mishkin (2013) included the ‘Mundell-Fleming model’ and describes the direct affect as follows:
MP ↑ir↓E↑ = depreciation (notation differ) NX↑Y↑
In addition, the depreciation has also an indirect effect on inflation, the price of imported intermediate goods used in the euro area domestic production becomes relative expensive, and this results in domestic prices to increase as well
4 Method
To examine the impact of the ECB’s implemented unconventional monetary policy, I will monitor the effects from different perspectives. In the literature review is explained through which channels the ECB’s policy should transmitted into the real economy to achieve their primary objective. As described is section 2 is the primary objective of the ECB is to maintain price stability and thereby contributing to full employment and balanced economic growth. I will use this information to monitor developments in the economy. This corresponds with the first pillar of the ECB’s analysis. The first pillar refers to an economic analysis by monitoring indicators to assess the short to medium-term risks to price stability (ECB, 2017h).
First, I will examine whether the expected effects through the transmission mechanism actually have occurred. I focus on specific announcements which are expected to have an influence because of large changes in the size or composition of the ECB’s balance sheet or because some announcement are highlighted in the literature. But since the programmes are related to each other and effects may not immediate visible, I consider the subsequent period. Table 1 shows the announcements.
Table 1: Policy announcements by the ECB
Date Policy announcement
2007 22 August LTRO (maturity 3 months)
2008 8 October Procedure of MROs
2009 4 June CBPP
2010 10 May SMP
2011 6 October CBPP2
2012 26 July Speech “Whatever it takes” by ECB President Draghi
2 August OMT
2013 4 July Forward guidance
2014 5 June TLTRO and negative deposit rates
4 September ABSPP and CBPP3 (started in : November 2014 and October 2014)
2015 22 January APP (ABSPP, CBPP3 and PSPP)→ QE 1 in graph
2016 10 March APP monthly purchases extended, CSPP, TLTRO II
→ QE 2 in graph
8 December APP monthly purchased reduced → Change QE in graph
Sources: see section 2
I start to analyze whether prices are stable. As described, this is defined as the yearly increase in the inflation (HICP) of below 2%, and this is the headline inflation. The core inflation exclude volatility in food price and energy developments and, therefore, core inflation might give a better impression of the effect of ECB monetary policy. This is in line with Bernoth et al. (2016), who remarked that “the core inflation qualitatively mirrors the development of headline inflation after the unconventional monetary
shock”. However, they also note “the effects are qualitatively smaller, not statistically significant, and materialize with a lag of up to one year, reflecting the higher degree of stickiness in core consumer prices”. Therefore, I consider both. Thereafter, I will monitor the transmission mechanism in order to assess whether developments have been transmitted and to which extent inflation developments could be related to the ECB. The process of the transmission channels are described in the literature review. However, since there are a lot economic developments which could be related to the channels, I will analyze the most crucial development. I will monitor these developments as follow:
Credit channel:
I will monitor whether money supply have increased, interest rates on loans have been lowered and if this have resulted in more lending.
Portfolio rebalancing channel:
I will monitor whether asset prices have risen, bond yields have declined and if the term premium is lowered. According to the capital key, purchased the ECB mostly German bonds. Therefore, I will monitor the German bunds with different maturities. The Euro Stoxx 50 index covers 50 stocks from 11 countries in the euro area. 2 These countries are a combination of stable and less stable countries and, in addition, the Euro Stoxx 50 serves as underlying for broad range of financial commodities. Therefore, the Euro Stoxx 50 index is likely to be representative for the euro area.
Expectation channel:
I will monitor the market inflation expectations by analyzing inflation linked swaps.
Exchange rate channel:
I will monitor if the effective exchange rate of the euro have depreciated and whether this have led to more net trade.
Second, I will analyze differences between countries within the euro area. This include a study whether the transmission of the monetary policy have been homogenous or not, whether it contributed to other economic developments (economic growth and lower unemployment) and to which extent the macroeconomic stimulus in countries can be explained by the ECB’s monetary policy. I will monitor this stimulus with the FMPI (Fathom’s Macroeconomic Policy Indicator). There euro area consist many countries, therefore I will report in particular the most stable and the least stable countries.
2 11 countries included in the Euro Stoxx 50 index: the Netherlands, Germany, France, Belgium, Italy, Ireland,
Portugal, Spain, Austria, Luxembourg and Finland
20
Third, I will analyze the effect of the ECB’s monetary policy by taking the crisis into account. Since the ECB started to implemented an unconventional monetary policy in response to the crisis and it is partially implemented during the crisis, the ECB may have influenced economic developments which contributed to limit the consequences of the crisis.
And finally, I will examine the impact of the ECB’s policy by comparing with other countries where the central bank implemented a similar monetary policy (the US, UK and Japan).
5 Results
5.1 Monitor of the euro area
Figures in this section can be found in appendix 8.1. The policy announcements described in the method are bold typed. Figure 1.1 shows the headline and core inflation. As mentioned, the ECB target their policy to the headline inflation, but sometimes may core inflation give a better indication of economic developments. This is visible in figure 1.2 and figure 1.3, which show that in particular fluctuations in oil caused volatility in headline inflation.
In the analysis of figure 1.1 is it remarkable that the inflation was near the target before the start of the first programme. After the announcement of the LTRO, the headline inflation increased considerably above target and core inflation remained relatively stable. After the MRO, core inflation decreased slightly and headline inflation decreased a lot. The decline continued after the introduction of the CBPP, but the decline was less sharply. In July 2009, the headline inflation reached its lowest point in years (-0.7), but increased quickly. At the same time, core inflation was still decreasing. Almost a year later, the ECB announced the SMP. In the subsequent period, both inflation parameters seems to have increased slightly. The two months preceding the CBPP2 core and headline inflation increased, the start of CBPP2 seems to have stopped this increasing trend. But since the headline inflation was above target, this was desirable. In the period after the “Whatever it takes” speech and the OMT, core and headline inflation decreased both. This decrease has led headline inflation closer to the target, but this decline was put through below target. The new strategic to forward guidance was followed by a decrease in both inflation parameters (below target). The TLTRO had no clear visible effect. The start of ABSPP and CBPP3 had a small visible increasing effect, although of very short duration. Both inflation parameters decreased further, but the announcement of the expanded asset purchase programme (QE 1) seems to have had a positive influence on this decreasing tendency. Since
QE 1, headline and core inflation increased slightly or remained stable, but considerably below target
(under 1%).
Credit channel
Figure 1.4 shows that core inflation could be related to bank lending to non-financial cooperation’s (NFCs) until 2013. Figure 1.5. shows the money supply and lending. During the crisis, M13 decreased substantially and M24 and M35 seem to have remained relatively constant. Between the end of 2007 and the beginning of 2008, M2 and M3 were relatively higher. It seems that loans to NFCs were also higher and this could be a result of the increase in money supply after the announcement of the
LTRO. Because LTRO was in particular focused to ease credit conditions, this policy seems to have
operated. Since May 2008, M2, M3 and loans to NFC’s decline sharply till January 2010. Therefore,
3 M1 = Currency + overnight and similar deposits
4 M2 = M1 + deposits (maturity of 2 years) + deposits redeemable in 3 months
5 M3 = M2 + repurchase agreements + MMF shares/unit + debt securities up to 2 years (ECB, 2017i)
22
the MRO had no visible effect on lending, despite of the remarkable visible effect on the reduction of interest rates to NFCs after the procedure of the MRO (figure 1.6). The CBPP seems to have had a small effect on lending to households. This increase seems mainly a result of the increase of house prices (figure 1.7). This could be as well a result of the ECB’s monetary policy, but this effect is more related to the very low interest rates set by the ECB and the fall in bond yields. After the
announcement of SMP, lending to NFCs increased visibly (figure 1.5), despite the increase of interest rates on loans (figure 1.6). After the CBPP2, the slightly increase in money supply was put through, but this did not resulted in more lending (lending decreased). Draghi’s “Whatever it takes” speech had also no visible affect through this channel. A month after the OMT the money supply increased, but loans to NFCs decreased even more and loans to households remained low (figure 1.4 and figure 1.5). The start of forward guidance had as well no visible effect in the short term, but soon after, lending started to increase. The LTRO had a small visible impact on lending to NFCs and the money supply. However, this effect seems weak, since the programme was in particular focused to ease credit conditions. Two months after CBPP2, interest rates on loans to NFCs fell again and this fall was put through by the subsequent programmes (figure 1.6).
It is remarkable, that since 2014, the pattern of core inflation and bank lending to NFCs changed. Before 2014, the percentage change in bank lending to NFCs was overall lower than the percentage change in core inflation (figure 1.4), while interest rates were still falling (figure 1.6). Therefore, it seems that the relative increase in lending did not resulted in higher core inflation. However, the percentage change in lending have become les negative during ABSPP and CBPP3,
QE1 and QE2.
In addition, it is remarkable that, despite the increase in monetary base (as a result of the monetary expansion by the ECB) and the creation of new lending facilities, M1 decreased from 2007 till the first half of 2008 and in the end of 2009 till 2011 (figure 1.5). Therefore, the huge increase in monetary base was not effective in increasing M1. According to the money multiplier, this can be explained due an enormous increase in excess reserves deposited at the ECB.
Portfolio rebalancing channel
Figure 1.8 shows bond yields with different maturities. After the start of LTRO, the yields edged down. After the MRO, yields of different maturities declined visible. However, the shorter the
maturity, the more the decline. This had the consequence that the yields changed in divergent direction and the spread became broader. The CBPP had a similar effect on yields, but smaller in magnitude and duration. After the SMP, the term premium lowered a bit. The CBPP2, the “Whatever it takes” speech and the OMT had no clear visible effect on these yields, but the process of declining yields and narrowing of the spread continued slowly. The effect of forward guidance was visible but of short duration and yields have converged a bit. After the LTRO, yields declined clearly and the term premium lowered. This was put through by ABSPP and CBPP3, QE1 and QE2, besides some yields
became negative. Especially QE1 seems to have had a strong effect, however, the decrease in yields and the lowered term premium seems temporary. The cut in monthly purchases (change QE) seems to have caused a broader spread. Nevertheless, all the yields are very low and some even negative.
Figure 1.9 shows the Euro Stoxx 50 index. The start of LTRO had a small visible increasing effect. Despite the sharp decline in prices in 2008 (during MRO), seems the prices have risen a bit in 2009. But this rise seems not direct related to SMP (possibly indirect in the longer-run). In 2010, the asset prices seems to have been stable, but the prices decline in the second half of 2011. The effect of the CBPP2 seems to have been of short duration and had less impact, but the fall in the preceding period seems considerably decelerated. The “Whatever it takes” speech seems to have had an increasing effect and the effect of OMT is not clear. After the start of forward guidance, prices started to increase again, but it seems that the LTRO and ABSPP and CBPP3 did not continue this upward effect. The QE 1 seems to have had a strong effect on the prices, but this sharp increase least for approximately four months and afterwards prices fell substantially. QE 2 had no clear effect on prices, but the fall in the preceding period seems to have slowed. The change of QE seems to have an increasing impact on prices
Expectations channel
The data about the market inflation expectations in figure 1.10 starts from August 2008, so it is not possible to monitor the LTRO from this figure. Remarkably, before the start of the MRO, inflation expectations were above target . After the MRO, the expectations declined sharply below target in the subsequent two months. But in 2009 inflation increased again and the CBPP seems to be related to a part of this increase. However, inflation increased above target. After the SMP expectations declined below target. These volatile fluctuations went through till 2012. The CBBP2, the “whatever it takes” speech, OMT, forward guidance, LTRO and ABSPP and CBPP3 are as well visible in the market expectation curves, but also have not clearly contributed to a stable expected inflation of below 2%. Market expectations declined further in 2014, but QE 1 seems to have stopped this decline. The effect of QE1 is clearly visible, however some months later, market inflation expectations declined again. After QE2 expectations seems to increase slightly. The change of QE seems to have no influence on expectations.
Exchange rate channel
Figure 1.11 shows the effective exchange rate of the euro. Overall, the euro depreciated substantially, but this the depreciation have been interrupted by periods of appreciation. After the start of CBPP,
SMP, CBPP2, TLTRO and QE1 the euro depreciated. Although, before the start of SMP, TLTRO
and QE1, the effective exchange rate was already depreciating. Therefore, was the effect of the programmes over time possibly more significant, than the effects of the announcements. Figure 1.12 shows the trade in terms of volume percentage changes. This gives an impression about whether the
weaker euro would have led to more trade volume. Since 2010 has the net trade volume been positive. Between 2011 and 2012 (short after CBPP2), the depreciation effect is clearly visible, because export volumes increased and import volumes decreased. Although, while the effective euro appreciated in 2012, remained the net trade positive. This pattern changed in 2014, despite the sharp depreciation of the euro became the net trade volumes negative. However, this net trade volumes are based on percentage changes, the visible trade balance in dollars is increasing since 2012 (figure 3.7).
5.2 Monitor across countries in the euro area
Figures in this section can be found in appendix 8.2. There are many differences between countries in the euro area. Most of these differences are from before 2007 (such as since the creation of the
Economic and Monetary Union). Therefore, these differences are more structural and the challenge for the ECB is to stimulate integration. This section takes the structural differences into account, since some structural differences are likely to have complicated the transmission, but is in particular focused on the transmission itself.
Figure 2.1 shows the core inflation for several countries in the euro area. It is visible that countries have responded differently to the ECB’s implemented monetary policy. Greece, Ireland, Portugal and Spain have experienced periods of core deflation in the past years. These four countries are together with Italy seen as the ‘periphery countries’ (and the PIIGS countries). In figure 2.2 is visible that the ‘periphery countries’ reacted more volatile to changes compared to the ‘core countries’ (Germany and France). Overall, it seems that economic developments have been lower in the
periphery countries. Some of these developments are due to the asymmetry in transmission and some are likely to be more structural .
Therefore, this first part of this section describes components of the transmission mechanism which are remarkable on the basis of asymmetry. The asymmetry can increase or reduce the
convergence between countries. Therefore, it would be desirable if the asymmetric effects have been favorable for the periphery countries. The second part of this section is focused on other economic indicators. The third part gives an impression of what degree in macroeconomic stimulus can be accounted to the ECB’s monetary policy.
Portfolio rebalancing channel
In figure 2.3 and figure 2.4 is visible that some yields of government and sovereign bonds diverged from 2007 to 2011. While the yields in the core countries have been slightly decreasing, have the yields of the periphery countries grown substantially. In 2012, most yields of the periphery countries converged again, but remained volatile. Especially Greek government bond yields were volatile. This means that the governments in the periphery countries had to borrow at much higher costs compared to the stable countries, and, therefore was the monetary policy more effective in the stable countries
than in the periphery countries. The transmission along the yield curve did not work in the periphery countries, while these countries in particular need a stimulus. So therefore was the effect perverse through the yield curves. Especially, because the SMP included public- and private debt securities to ensure liquidity in dysfunctional markets. The effect of SMP on sovereign bond yields is visible in figure 2.4abc, but this effect seems not effective in lowering the yields in the long-term.
However, from 2012, the yields of periphery countries seems to converge. This effect continued to the beginning of 2015 and most of the yields seems to have stabilized at a low rate. Greece is an exception of this and the yield of Portuguese government bonds is more volatile.
Expectations channel
The consumer confidence net balance (figure 2.5) and the business climate index (figure 2.6) give an impression about the division of expectations across countries. This division is the most visible in the consumer confidence figure, but in both figures, is the divergence the broadest between approximately 2010 and 2012. Since 2007 all the ‘expectations’ decreased. Although all the ‘expectations’ increased considerably in 2009, this effect was temporary. In 2010, the ‘expectations’ decreased in the periphery countries and increased in core countries. In the end of 2012, the expectations in all the countries seems to have increased in convergent direction. Especially, the expectations in Ireland, Italy, Portugal and Spain increased relatively compared to the rest of the euro area. Greece was an exception of this slightly increasing path.
Exchange rate channel
Figure 2.7 shows the real effective exchange rates across countries. Especially the real effective exchange rate of Spain, Italy and Greece have diverged, but since approximately 2014, these rates seems to converged a bit. Remarkable that the real effective exchange rate of Greece depreciated substantially since 2012. It would be desirable if this resulted in a better competitiveness. Figure 2.8 shows the ECB’s indicator for competitiveness. Before 2010 was a divergence in degree of
competitiveness across countries. Since 2010, the spread seems to have been converging, however, Germany have had considerably more competitiveness compared the rest of the euro area.
Figure 2.10 gives an impression of the trade in periphery countries. Relative to the euro area, do these countries seems to have profit more from the depreciation of the euro in 2014.This effect through the exchange rate seems also have had a positive effect on the current account deficits (figure 2.9ab). Though, the current account only gives an impression of what can be a result of the increase in competiveness in the periphery countries.
Figure 2.11 shows the GDP and gives an impression of the economic growth across countries. The recovery in growth in 2009, was in every country powerful. Although for a short period. From 2011 economic growth started to decrease again for most of the GIIPS countries. Figure 2.12a,b shows the unemployment rates in the euro area. The differences between stable and periphery countries have become large since the crisis, especially youth unemployment is very high. In particular, the
unemployment rates in Greece and Spain are high (in 2012 unemployment: more than 25%, and youth unemployment: more than 50%). Since 2013 seems unemployment have reduced a bit. This decrease have put through in the subsequent years.
Figure 2.13 shows the contribution of monetary policy and fiscal policy to the macroeconomic stimulus in countries. A high positive score implies tight policy and a negative score implies loose policy. This implies that a loose policy will boost growth in demand relative to the growth in supply. Before 2010, the distribution of the contribution of monetary and fiscal policy differ among countries. But overall contributed both policies to the same stimulus. Only in Germany seems the policies have ‘compensated’ each other’s. This means that in periods the fiscal policy have been tight and the monetary policy loose (or vice versa). But after 2010 this pattern seems to have changed in all the countries, the very loose monetary policy seems to be ‘compensated’ by very tight fiscal policy. This was in particular the case in Spain, Portugal and Greece. In Germany, Italy and France was the fiscal policy less tight. No data for other countries.
5.3 The ECB’s response to the crisis
Figures in this section can be found in appendix 8.3. As mentioned, the ECB started to increase liquidity in response to the global financial crisis, started in 2007. The crisis affected the euro area in several developments. First with the ‘liquidity crisis’ (hoarding liquidity in banking system, Q3 2007 – Q4 2008), second with the ‘solvency crisis’ (interbank market dysfunction, Q4 2008 – Q2 2010) and third with the ‘sovereign debt crisis’ (questions about the sustainability of the financial fragmentation, Q2 2010 – Q3 2013). These crises are clearly visible in figure 3.1 (lower economic activity).
The developed financial fragmentation in the euro area is remarkable since the crisis. The balanced liquidity in figure 3.2 reflects this financial fragmentation: high liquidity surpluses and high liquidity deficits. Since 2007, the financial fragmentation increased and became huge in the subsequent years. Especially, the periphery countries have built up large liquidity deficits. Part of the source of this financial fragmentation is due to the capital flight from the periphery countries to stable countries during the sovereign debt crisis (figure 3.3). The interbank market was dysfunctional. By providing full allotment, fixed rates and by guaranteeing that the ECB will do “Whatever it takes”, the ECB is likely to have inhibited some financial fragmentation from 2012. Although, this reduces have lasted
approximately two years. Therefore, QE1 and QE2, do not seems to have contributed by reducing this financial fragmentation.
The divergence and convergence in government and sovereign bond yields across countries (as mentioned in the previous section, figure 2.3 and figure 2.4) can also be related to the sovereign debt crisis and Draghi’s speech. Some yields increased substantially due the default risk, and this resulted in speculation. The ECB seems to have stopped this with Draghi’s “Whatever it takes” speech. By making a strong statement, that the ECB will always print more money to defend the euro, it was not useful anymore to speculate against the euro.
However, this statement also trespassed the boundary between monetary- and fiscal policy. Since the ECB guaranteed the debt sustainability immediately (with low interest rates and the purchase programmes, regardless whether the fiscal policy is sustainable), it reduced the pressure and weakened the incentive for governments for market discipline.
As described, the crisis in the euro area economy lasted until 2013, but there were certainly external shocks that could have hit the euro area economy. For instance, the banking crisis in China, the Brexit or the US presidential election in 2016. Therefore, it seems that the monetary policy of the ECB may have contributed to a more stable economy.
5.4 The ECB’s monetary policy compared to the Fed, BoE and BoJ
Figures in this section can be found in appendix 8.4. In response to the crisis, started the Fed, BoE and BoJ also with an expansionary policy. By analyzing the central bank’s balance sheet in figure 4.1, it seems that in particular the expansions of the BoE and the Fed are related to each other, the BoJ seems least related to the other central banks, and the ECB’s balance sheet is in wide lines comparable with the path of the BoE and Fed. In all the figures seems Japan the least related to the other central banks, therefore is Japan omitted in this report.
Compared to the CPI and core inflation in the UK before the crisis, seems the inflation in the euro area considerably lower in the past 10 years (figure 4.2 and figure 4.3). Although this difference became smaller. The path of the CPI inflation in the US is comparable, but core inflation is overall
substantially higher.
Figure 4.4 shows that economic growth decreased not only in the euro area economy, but also in the US and UK. Especially during the first period of the crisis. In 2010 and 2011, all economies seems to recover, but the recovery was interrupted in the euro area in 2011. The euro area diverged from the US and UK. But this was the beginning of ECB’s implemented policy, since 2015 seems the euro area have grown again in roughly the same path as the US and UK. However,
figure 4.5 shows that unemployment diverged substantially in 2012. Since 2014 is unemployment converging, but remains at a relative high level compared to the other countries (slightly below 10%).
Figure 4.6 shows the government bonds in the euro area haven been substantially lower in the euro area since the end of 2012. In addition, in contrast to the US and UK, the visible trade increased in the euro area since the end of 2011 (figure 4.7). But, on the other hand, inflation expectations have been decreasing more since 2014 (figure 4.8). In addition, figure 4.9 shows that the increase in lending to households and NFCs have been substantially higher in the US since 2013.
6 Conclusion
By analyzing only inflation, it may be questionable whether deflation really was a threat. But on the other hand, due the crisis developed the euro area structural problems that could have deteriorated the euro area economy and could have caused deflation in absence of the ECB’s unconventional monetary policy. Therefore, it is likely that the ECB contributed to the recovery of the euro area economy. However, the effects through the transmission mechanism seems not very powerful.
The transmission of the ECB’s implemented policy seems to have operated through the credit channel. However, the transmission of this large increase in monetary base (in combination with the created lending facilities) does not seem to have occurred in large size. Even though, lending have increased in 2010 and 2011, and the provision of loans appears to increase slowly since 2014, this increase seems minimal compared to the large increase in momentary base.
The effect through the portfolio rebalancing channel was clearly visible. Especially
government bond yields have decreased considerably compared to the US and UK. Although, before 2012, this desirable effect seems only has taken place in the more stable countries. This resulted in a divergence across euro area countries and this is not desirable. But since 2014, most of the government bond yields in the euro area seems to be lowered substantially. Based on this, the ECB’s monetary policy since 2012 (‘breakpoint’ from divergence into convergence) seems to have been effective. However, it is questionable whether this effect have been transmitted into the real economy, since headline and core inflation have been decreasing since approximately 2012 till 2014.The desirable effect on equity prices was as well visible in 2012, and especially in 2015 increased prices
considerably. Given that headline and core inflation have also increased slightly in 2015 this could be related to the ECB’s. However, the equity prices in 2016 seems even lower than before the expansion, so the impact seems not very powerful. Besides, the market expectations have been as well temporary higher in 2015. Therefore, the increase in inflation was probably also a result of the increased market expectations. Overall was the transmission of ECB’s monetary policy trough market expectations not substantial.
The effect through the exchange rate channel seems to have contributed to an increase in net trade, although temporary, since import volumes increased relative more than export volumes after the sharp depreciation of the exchange rate in 2014. But, on the other hand, the trade balance is
considerably higher in the euro area compared to the US and UK. In addition, there seems to have been some asymmetry in favor of the weaker countries, the effect through the exchange rate is small but seems to have contributed. However, because many influences can have an impact on the
exchange rate, it is questionable to which extent the depreciation of the euro was a result of monetary policy, but the overall weakening of the euro is likely linked to the ECB’s monetary policy.
