Ebook Liquidity, Inflation and Asset Prices in a Time-Varying Framework for the Euro Area
To achieve its primary objective of price stability, the European Central Bank (ECB) uses a strategy based on two "pillars". One of these pillars, referred to as the monetary analysis, exploits the long-run link between money and inflation. In particular, to signal its commitment to price stability and to provide a benchmark for the assessment of monetary developments, the ECB announces a reference value for the growth rate of the broad monetary aggregate M3. This prominent role assigned to money has been subject to intense criticism from the very beginning. Besides theoretical motivations for not considering monetary aggregates (e.g. Galí 2003, Woodford 2007), it has been frequently argued that money might be an unreliable indicator in an environment of low inflation (e.g. Estrella and Mishkin 1997, De Grauwe and Polan 2005). Since the introduction of the euro, the annual growth rate of M3 has almost continuously been above its reference value of 4.5% without a corresponding tightening of monetary policy or an acceleration of inflation, which supports doubts about the usefulness of money aggregates as an indicator of risks to price stability.
The ECB claims, however, that the analysis of monetary developments goes well beyond the assessment of M3 growth in relation to its reference value. The monetary analysis uses a comprehensive assessment of the liquidity situation based on information about the balance sheet context as well as the composition of M3 growth (ECB 2004). It is intended to shed light on the outlook for price stability and the implications for monetary policy eschewing a mechanical policy response to a monetary aggregate. For instance, the Governing Council has repeatedly stated that some episodes of rapid money growth were due to special factors and shifts in the demand for money arising from e.g. portfolio shifts or changes in the opportunity cost of holding money. As a consequence, such episodes were disregarded and not considered as signalling risks to price stability. On the other hand, there were cases where excess money growth did warrant a tightening of policy, especially when combined with information obtained from the other pillar of ECB’s monetary policy strategy, the economic analysis (Gerlach 2007). This illustrates that the link between excess money growth or excess liquidity and future inflation is probably not constant over time and depends on other factors as well, such as the source of increased liquidity and general economic conditions.
In defense of its two-pillar strategy, the ECB also often argues that asset price bubbles could be the result of strong and persistent growth in money and credit aggregates. Since developments of asset prices not in line with fundamentals are not captured by a pure inflation forecast, they do not trigger a policy reaction in a traditional Taylor rule framework (Issing 2002). A detailed monetary analysis could therefore provide early information on emerging financial imbalances which could have destabilizing effects on economic activity and inflation. Detken and Smets (2004) indeed show that high-cost booms in asset prices often follow rapid growth in money and credit stocks just before and at an early stage of the boom. Since financial assets are growing in importance and hence, asset price fluctuations increasingly affect the economy, monetary policy could be improved by taking account of the evolution of money and credit aggregates as a signal of financial imbalances (Hildebrand 2008). There are obviously also episodes in history during which money, temporarily growing out of line with fundamentals, did not coincide with asset price bubbles. Accordingly, the information for asset prices contained in these indicators may also vary over time and this suggests that the weight assigned by a central bank to the monetary analysis should be state dependent (Issing 2002).
In this paper, we focus more extensively on the complex link between money, economic activity, asset prices and inflation. In particular, we investigate the impact of liquidity shocks in a time-varying and state-dependent framework for the Euro area economy. Excess liquidity is identified as the deviation of road money from an equilibrium value in a structural VAR. We first estimate the impact of a shock to liquidity on a set of macroeconomic variables and asset prices within the VAR framework. This shock has a temporary effect on economic activity and a permanent impact on the price level which is less than proportional. Increased liquidity also creates temporary rises in real equity, property and aggregate asset prices. The economic consequences and the magnitude of the impact however depend heavily on the underlying source of increased liquidity. A 1 percent long-run increase in M1 has a considerable impact on economic activity, asset prices and inflation. The impact on inflation is even proportional. In contrast, a shock in M3-M1 has only minor economic consequences and results more in a permanent long-run increase in the real money stock. We observe that shocks to credit, which is the counterpart of the broad monetary aggregate, have rather similar effects as shocks in M3.
Using a simple sample split and more sophisticated Bayesian VARs with time-varying parameters, we also find considerable variation in the dynamic responses over time. On the one hand, inflationary consequences of a liquidity shock are much weaker since the mid-eighties resulting also in a more permanent shift of real money. In more recent times, however, there seems again to be a tendency for an increased impact on inflation. On the other hand, time variation with respect to the effect on output and asset prices is less clear. We find increased responsiveness during some periods but decreased reactions at other points in time. This is not surprising given the growing theoretical and empirical literature (as discussed below and in section 4) that argues that the macroeconomic impact depends on the state of the economy, e.g. asset price boom-busts, credit booms, the business cycle or the monetary policy stance. The final part of the paper analyses this in more detail. More specifically, we estimate the impact of excess liquidity shocks in a single equation approach allowing the coefficients to differ depending on the state of the economy. We find evidence that liquidity shocks have a stronger impact on economic activity and asset prices during asset price booms and busts, at times of credit booms (which are a proxy for financial innovations) and, to a lesser extent, during periods of tight monetary policy. Negative shocks to liquidity also exert stronger effects on real activity and asset prices than positive ones.
While real property prices are much more sensitive to excess liquidity when economic growth is above its trend value, the reaction of output is stronger during recessions. On the other hand, inflationary effects are greater during boom phases of asset prices, economic expansions and credit booms, but smaller when the monetary policy stance is restrictive. In sum, the impact heavily depends on the underlying state of the economy. The estimated differences are also economically important. The reaction of real asset prices to a liquidity shock during an asset price boom is, for instance, three times larger than the average estimated reaction for the whole sample period. Similar conclusions can be drawn for inflation and output growth.
Download
PDF Ebook Liquidity, Inflation and Asset Prices in a Time-Varying Framework for the Euro Area
Posted in :
