The Impact of ECB Monetary Policy on Stock and Bond Market Liquidity. The Case of Germany

Table of Contents

1. Introduction

2. Theory
2.1 Market Liquidity
2.2 Stock and Bond Market Liquidity
2.3 Market Liquidity Measures
2.4 Monetary Policy and Market Liquidity
2.5 Monetary Policy Measures

3. Timeline of Monetary Policy Events

4. Empirical Evidence
4.1 Descriptive Statistics on Liquidity Levels
4.2 Descriptive Statistics on ECB Monetary Policy Indictors
4.3 Vector Autoregressive Analysis

5. Conclusion

References

Abstract: During the financial crisis and the following Eurozone crisis, liquidity in financial markets basically froze and became a problem for the real economy. Therefore, market liquidity became one of the major concerns of the ECB, which applied non-standard measures, e.g. irregular asset purchasing programmes. This paper sheds light on the impact of monetary policy on liquidity levels of the DAX 30 equity index and German 10-year government bonds. For the following analysis, the monetary policy impacts are estimated using the base money growth rate and EONIA rate, whereas the relative bid-ask spread is employed for measuring liquidity levels. The research method includes literature-based research about common market liquidity theories, a short timeline of important ECB monetary policy decisions, descriptive statistics on liquidity levels and monetary policy variables and a VAR analysis, including variables spreads, returns, volatilities, industrial production and inflation. The results indicate that a decrease (increase) in stock market liquidity or an increase (decrease) in bond market volatility lead to a decrease (increase) of EONIA. Furthermore, decreases (increases) in stock return or industrial production result in a decrease (increase) of EONIA. However, base money growth is positively correlated only to changes in bond market volatility. Overall, the results suggest that the monetary policy decisions by the ECB are influenced by changing market conditions without the ability to forecast liquidity levels.

Table of Figures

Figure 1 DAX30 companies.

Figure 2 Liquidity spirals.

Figure 3 Additional liquidity

Figure 4 Flooded with cash.

Figure 5 ECB’s bond purchases

1. Introduction

In recently published headlines readers can feel a certain amount of distress concerning financial markets: “Bond market fears liquidity crunch repeat” (Financial Times, 2015) and “Frozen - Regulators have made banking safer. But has that made markets riskier?” (Economist, 2015). These alleged concerns about market liquidity are not a new phenomenon. Since the financial crisis back in 2008, market liquidity has been a major concern of central banks. During that time liquidity basically froze, viz. banks did not want to lend each other money. In 2008 the Economist published the article: “When the rivers run dry - Can bank regulators and central banks prevent future liquidity crises?” It deals with the sudden evaporation of liquidity and the tools central banks use to stabilize the economic system. According to the Financial Times (2015) article mentioned earlier, the current situation is worrying, due to the fact that the European Central Bank (ECB) imposed several rules on commercial banks after the financial crisis. As a result, these banks provide less liquidity to the markets. The same arguments are used in another article by the Economist (2015), saying more regulations on banks “have made such assets less liquid. Investors may not be able to buy and sell them quickly, cheaply and without moving the price. The consequences in a downturn, when markets are less liquid anyway, could be severe”. As market liquidity remains a major concern in 2015, it is important to understand what market liquidity is, how it is measured and if the ECB can impact the levels of liquidity.

Regularly, the ECB interferes in markets during times of crisis. In detail, a loosening of monetary policy, also called easing or expansionary fiscal stance, will be expressed by a decrease in the interest rate or an increase in base money growth (Fernández-Amador, et al., 2013, p.57). These measures affect both markets because assets seem more attractive to investors in times of a loose monetary stance. However, tradeoffs between markets might occur when monetary policy measures are in place, e.g. an increase in the interest rate could increase attractiveness of bonds, but might make borrowing costs more expansive. Therefore, investment in any market would not increase. In theory, central banks could increase the money supply by buying back government securities or by decreasing the reserve requirements for commercial banks. Hence, banks have more money to lend to investors, who can now borrow money more easily and invest it in markets. Therefore, market liquidity will increase. The opposite situation, also called tight monetary stance, is characterized by a high interest rate or a small rate of base money growth.

Another possibility to increase market liquidity is to lower the interest rate at which banks can borrow money from the central bank. The theoretical linkage is that banks will afterwards lower their interest rates for the private or industrial borrower. In that case, it is assumed that the beneficiaries are more willing to borrow money and invest into markets. This will result in increased market liquidity.

Previous research papers about monetary policy effects on liquidity primarily focus on equity markets only, including cross-sectional determinants rather than a macro perspective. However, they do not address the direct impact of policy measures on liquidity and are mostly carried out for US markets. This is what Chordia, Sarkar and Subrahmanyam (2005, p.126) argue in order to highlight the importance of further studies about theoretical linkages between monetary policy and market liquidity. Despite the fact that several papers have addressed the above-mentioned criticism since Chordia, Sarkar and Subrahmanyam published their research paper in 2005, among others Amihud, et al. (2005) and Fernández-Amador, et al. (2013), there is still little empirical evidence in the case of Germany. Therefore, the purpose of this paper is to examine how changes in ECB monetary policy impact the liquidity of the German DAX 30 equity market and the German 10-year government bond market. Furthermore, it will be investigated if additional variables such as volatility, return, industrial production and inflation influence German market liquidity . Several authors explored the correlations between these variables for foreign markets. Among them, Brunnermeier and Pedersen (2009) showed that past returns and volatility can influence stock market liquidity. According to Goyenko and Ukhov (2009, p. 194), productivity and inflation can influence liquidity indirectly via inventory risk with the latter increasing inventory holding costs and lowering liquidity. In times of high productivity, risky assets become more attractive to investors, which could be reflected in cash outflows from bond markets . In addition, if these factors are important pieces of information regarding macroeconomical development, they will therefore be part of the discussion about monetary policy in the ECB’s rulings.

In the following, the theory of market liquidity and common interlinkages with monetary policy are examined in Section 2. Section 3 emphasizes major events in Germany that relate to monetary policy decisions. The empirical evidence in the form of descriptive statistics and a Vector Autoregressive Analysis (VAR) is reviewed in Section 4. Finally, the results are analyzed in Section 5 and interrelations are explained, while linking back to the theory.

2. Theory

In the following paragraphs more insights about the theoretical framework of market liquidity shall be given with reference to previous research papers. This section outlines how market liquidity is defined (Section 2.1), what it is comprised of (Section 2.2), how it is measured (Section 2.3) and if previous research papers found correlations with ECB monetary policy (Section 2.4). Lastly, measures of ECB monetary policy variables are presented (Section 2.5).

2.1 Market Liquidity

According to Chordia, Sarkar and Subrahmanyam (2005, p.85), “liquidity, a fundamental concept in finance, can be defined as the ability to buy or sell large quantities of an asset quickly and at low cost.” In other words, it can be described as the ease of trading assets (Amihud, Mendelson and Pedersen, 2005, p. 2). Moreover, Fleming (2003, p. 85) mentions that low transactions costs determine a liquid market.

On the other hand, illiquidity is supposed to be a risk factor (Goyenko and Ukhov, 2009, p. 191). The literature considers illiquidity as a systematic risk, which investors take into account when making investment decisions. An illiquidity shock in one market will lower its attractiveness to investors. This will decrease demand, what could impact illiquidity.

This paper focuses on aggregated market liquidity of 10-year German treasury bonds and the DAX 30 equity index.

Abbildung in dieser Leseprobe nicht enthalten

Fig. 1: DAX30 companies (dieInfografik, 2015)

Nonetheless, individual stock level liquidity is also interesting to observe. According to Fernández-Amador, et al. (2013, p.64), smaller companies seem to be more sensitive to changes in liquidity levels than larger companies. This is due to their dependency on funding costs, which might be influenced by monetary policy. In detail, monetary policy influences the interest rate, which has an impact on the cost of bank financing. The effect of EONIA, for example, decreases with increasing company size. The same effect can be observed when increasing the base money growth rate, which increases liquidity. However, the higher market capitalization of stocks, the weaker the effect of monetary policy will be. The authors assume that bank lending is more crucial to smaller firms than larger ones in bank-based economies such as Germany. Higher costs of financing affect expected returns of stocks and affect market liquidity. A detailed description on an individual stock level, however, would go beyond the research scope of this paper.

Moreover, this paper focuses on the German 10-year treasury bonds. A further extension could be to add different maturities of government bonds to the analysis. However, this paper follows Brunnermeier and Pedersen (2009, p.2228), when assuming that market liquidity co-moves across assets and stock and bond markets, since “changes in funding conditions affect speculators’ market liquidity provision of all assets”.

2.2 Stock and Bond Market Liquidity

This paper focuses on two types of market liquidity, namely stock and bond market liquidity. Before continuing with the effects of monetary policy, correlations between these two markets should be established. Chordia, Sarkar and Subrahmanyam (2005, p.96) found similarities in both markets, such as timing commonalities: market liquidity is lower on Fridays and around the holidays in both markets. Moreover, liquidity peaks generally from July to September. During crisis periods, spreads tend to be higher, viz. market liquidity is lower.

Furthermore, cross-market dynamics flowing from volatility to liquidity suggest common influences across markets. In detail, “volatility in either market is significantly correlated with spreads in both markets” (Chordia, Sarkar and Subrahmanyam, 2005, p.101).

More empirical evidence about linkages is given by Fleming (1997). He provides a model about the linkages of volatility and market liquidity in the bond and stock market. Indeed, the author found strong connections. Fleming (1997, p. 136) argues that this is due to common information, which comes from events such as announcements of the inflation rate. Consequently, this simultaneously alters the expectations of investors across all markets. Afterwards, traders will adjust their holdings across markets, producing an information spillover, viz. cross-market hedging. Information is also a key factor in Goyenko’s and Ukhov’s (2009, pp.198-199) research paper. The authors find a strong linkage of liquidity between stock and treasury bond markets. This is due to the fact that illiquidity of bond markets predicts illiquidity of stock markets. In contrast, shocks to stock illiquidity are negatively correlated with shocks to bond market illiquidity, which is consistent with flight to quality and flight to liquidity periods.

Admittedly, these are two strategies in which one market is more favorable to the investors than the other. In uncertain market conditions, investors prefer assets which are less risky and therefore less profitable; those are usually bonds. These episodes are called flight to quality and often occur because a tight monetary policy leads to an increase in interest rates, which increases the attractiveness of bonds compared to equities. However, when the economy is flourishing investors feel safe while holding stocks. They are indifferent to taking risk as they want to make more profit, which creates an outflow from bond to stock markets. These episodes, which are called flight to liquidity, occur when markets are more liquid and returns are assumed to be higher in equity markets (Fernández-Amador, et al., 2013, p.66). These two strategies are relevant because an outflow of money from one market into another can cause price pressures and increase liquidity in one market while decreasing liquidity in the other market (Chordia, Sarkar and Subrahmanyam, 2005, p.87).

Nevertheless, central banks’ monetary policy decisions lead equity and bond markets to move similarly. During times of expansionary monetary policy, increases in funds of investors can cause more liquidity in both markets (Chordia, Sarkar and Subrahmanyam, 2005, p.87). They suggest that money flows, i.e., those generated by investments, are a commonality in stock and bond market liquidity.

2.3 Market Liquidity Measures

This paper measures stock and bond market liquidity by observing the bid and ask prices. The bid price is the maximum price at which an investor is willing to buy an asset. In other words, holders of an asset will receive that price as soon as they are willing to sell. The ask price is the minimum price a holder of an asset wants to receive in order to be willing to sell to the buyer. Therefore, investors willing to buy an asset have to pay the ask price. However, there will be a slight difference between those two prices. That gap is called “spread” and indicates the liquidity of an asset. A high spread indicates lower liquidity, because the asset cannot easily be traded on the market. The opposite, a low spread, is a sign of high liquidity (Investopedia, 2015). As a result, they are often associated with down markets (high spreads) and up markets (low spreads), e.g. by Chordia, Roll and Subrahmanyam (2001, p. 526). The spread is the compensation for the ‘market makers’ for the burden of providing liquidity and alternatively selling to buyers or buying from sellers. That is why ask prices are always higher than bid prices.

One of the few research papers dealing with monetary policy of the ECB and liquidity of European stock markets is Fernández-Amador, et al. (2013, p.67). In line with those scholars, and the approach of Amihud and Mendelson (1986), this paper uses the relative bid-ask spread to measure liquidity in equity markets. The figure is calculated by dividing the difference of the ask price and the bid price by the midquote, the latter being the average of the bid and ask prices. Therefore, the measure is taken in relation to the price, which makes it easier to compare across firms.

Moreover, Fleming (2003, pp. 98-99) observed the treasury market for the best liquidity measure. In detail, the author compared the bid-ask spread, quote size, trade size, the on-the-run/off the-run yield spread, trading volume and trading frequency. He suggests the bid-ask spread as an adequate measure of treasury market liquidity due to the fact that all the other indicators are only weak proxies for market liquidity. Since bonds are quoted in yields, the spread is calculated by bid yield minus ask yield.

2.4 Monetary Policy and Market Liquidity

This paper builds on a vast theoretical basis concerning connections between monetary policy and market liquidity. As Chordia, Roll and Subrahmanyam (2001, p. 512) point out, margin requirements, short-selling constraints and inventory paradigm should depend on interest rates and therefore monetary policy. The latter concept was first introduced by Demsetz (1968). It basically suggests that liquidity depends on inventory turnover rates and inventory risks. A policy triggered decrease in financing costs of inventory could increase trading activity, and as a result, market liquidity would increase.

Monetary policy decisions are announced mostly in regular intervals. For example, the target funds rate is announced periodically. This means that the policy decisions are partially predictable. As always in the financial markets, the expectations about what will occur are already included in the price of an asset. After announcements of central banks, markets only react to the surprise component (Chordia, Sarkar and Subrahmanyam, 2005, p.119). The same principle is valid for stock markets when the reporting season of the DAX 30 companies lies ahead. The price of the stock already includes the expectations of investors. This means that share prices will only move if a company performs better or worse than expected. One argument for a possible connection of market liquidity and monetary policy is that both are part of the business cycle. In this case the real economy would react as a transmission channel of liquidity (Fernández-Amador, et al., 2013, p.55).

The findings of Chordia, Sarkar and Subrahmanyam (2005, pp.101-124) suggest that monetary expansion increases market liquidity, which is characterized by reduced spreads. Their analysis was carried out in the US market including the Fed, which is the central bank of the USA and the equivalent of the ECB. The authors specifically mention the role of the central banks during a crisis: in stock and bond markets volatility is higher and liquidity is lower than in normal market conditions. On the other hand, an up market is characterized by decreased spreads, which means higher liquidity. During crisis periods monetary policy is supposedly able to forecast market liquidity, i.e. when the federal funds rate increases, the spreads increase. However, the central bank will usually perform a monetary expansion, not tightening, during a financial crisis, leading to increased market liquidity across both markets.

Furthermore, monetary policy indirectly influences market liquidity. According to the above-mentioned authors, the federal funds policy is positively correlated with volatility. This change in volatility will then affect spreads because they are positively correlated. In other words, volatility has a negatively correlated effect on liquidity. For example, an unexpected increase in the federal funds rate will result in an increase in volatility and spreads, meaning decreasing market liquidity. The authors mention a time lag of two months between the shock in bond flows until the peak of the effect in bond spreads. Other authors suggest that markets react in the mid-run, viz. the effects of monetary policy peak after one year (Fernández-Amador, et al., 2013, p.59).

Further research (Goyenko and Ukhov, 2009, p. 210) shows that the flow of funds can be described as a circle of strong causality, consisting of:

1. Monetary policy variables

2. Bond market liquidity

3. Stock market liquidity

[...]