Equipo Interdisciplinario De Investigación

According to the conventional approach a reaction of monetary policy to asset-price misalignments is justified only when the latter are known to provide useful information about the future course of inflation. In particular, so long as monetary policy maintains price stability, it promotes financial stability as well (see, for example, Schwartz (1995) and Bordo, Dueker, and Wheelock (2002), (2003)). This view holds that financial crises (or simply “financial imbalances”) need to be tackled by lender-of-last-resort practices or regulatory policies (as in Schwartz (2002))151. Bernanke and Gertler (1999) express this view in the following concise way: “The inflation targeting approach dictates that central banks should adjust monetary policy actively and pre-emptively to offset incipient inflationary and deflationary pressures. Importantly, for present purposes, it also implies that policy should not respond to

151

This approach is famously proposed in Bernanke and Gertler (1999), (2001) in the context of a Taylor rule, as well as among others in Schinasi and Hargraves (1993), Bullard and Schaling (2002), and White (2004).

changes in asset prices, except insofar as they signal changes in expected inflation” [emphasis is in the original] (Bernanke and Gertler (1999), p. 78).

The conventional approach, however, has been (at least partially) criticised by several economists. For example, according to Smets (1997) in the presence of ‘irrational exuberance’ in the financial markets a monetary policy tightening is actually optimal. In a similar manner, Cecchetti, Genberg, Lipsky, and Wadhwani (2000) suggest a monetary policy reaction to asset-price misalignments. However, Bernanke and Gertler (2001) doubt whether policymakers are in fact capable of reliable judgement of whether asset price movements reflect “irrational exuberance” or that the possibility of a collapse in asset prices is substantial and imminent. Nevertheless, on this account, Cecchetti (2003) argues that the task of identifying promptly and accurately asset-price misalignments is not more challenging than other elements of policy-design, such as potential output.

Bordo and Jeanne (2002a), (2002b) are in favour of a monetary policy reaction to asset price booms, as they view that pre-emptive actions so as to contain asset-price misalignments, in fact, provide insurance against the high economy-wide costs of lost output in the event of the bubble bursting. They argue that policymakers must try to defuse asset-price booms either when there is a high risk of a bust (and the consequences it may bring considerably damaging to the economy) or when the cost of such an attempt is estimated to be low in terms of foregone output. They point-out that the more optimistic investors get, the higher the risk of a market-sentiment reversal becomes. However, a higher cost is attached to monetary policy actions of ‘leaning against the wind of investor optimism’. Therefore, they contend that

monetary policymakers need to evaluate not only the probability of a crisis occurring, but also the extent to which monetary policy is capable of reducing this probability.

Greenspan famously saw a conundrum in the attempts of monetary policymakers to defuse an asset-price boom. He argued that the likelihood of stock-market booms occurring is relatively high in low inflation environments:

“…We have a very great difficulty in monetary policy when we confront stock market bubbles. That is because, to the extent that we are successful in keeping product price inflation down, history tells us that price-earnings ratios under those conditions go through the roof. What is really needed to keep stock market bubbles from occurring is a lot of product price inflation, which historically has tended to undercut stock markets almost everywhere. There is a clear trade-off. If monetary policy succeeds in one, it fails in the other. Now, unless we have the capability of playing in between and managing to know exactly when to push a little here and to pull a little there, it is not obvious to me that there is a simple set of monetary policy solutions that deflate the bubble.” (Federal Open Market Committee meeting transcript, 24th September, 1996, pp. 30-31).

Driffill et al. (2005) analyse the argument that central banks tend to alter their interest-rate instrument gradually so as to protect banks from the vulnerability caused by sharp increases in short-term interest rates, and thus promote their macroeconomic objectives further than inflation targeting, such as that of financial stability. They prove theoretically that smoothing interest rates may lead to indeterminacy of an economy’s rational expectations equilibrium. In particular, they extend the analysis of determinacy of equilibrium as in Bullard and Schaling (2002) and by investigating the inclusion of futures prices, and the associated basis risk, in the central bank’s reaction function, they prove the existence of a trade-off between macroeconomic and financial stability. They, also, argue that this trade-off calls for

caution, but does not necessarily imply that the central bank cannot smooth interest rates to reduce basis risk.

Bordo and Wheelock (2004) survey US stock-market booms and find that “booms do not occur in the absence of increases in real economic growth and perhaps productivity growth” (Bordo and Wheelock (2004), p. 41). Their analysis does not support the proposition that excessive money or credit growth leads to a boom in asset prices. However, they report that most asset-price booms that took place during the 19th century occurred in times of monetary expansions. Notably, they strongly disregard the view that monetary policy can trigger speculation in the asset markets when it fails to control the credit supply. However, they present anecdotal evidence suggesting that at times the stock-market does rise above the levels justified by fundamentals. Bordo and Wheelock (2004) argue that “although one can offer plausible theoretical arguments for responding proactively to an asset price boom”, as their survey suggests, “policymakers should be cautious about attempting to deflate asset prices without strong evidence that a collapse of asset prices would have severe macroeconomic costs” (Bordo and Wheelock (2004), p. 41).

Nevertheless, any attempt to evaluate the appropriate monetary policy response to asset price bubbles should not fail to consider primarily the explicit objectives of monetary policy (stabilising inflation and economic activity), and its ultimate aim to promote public welfare by fostering economic prosperity. Since asset price movements lead to macroeconomic fluctuations affecting prices and employment, the monetary authorities are bound to be concerned. It is also considered that if monetary policy manages to give a fitting and prompt response to the asset-price

misalignments under consideration, the adverse macroeconomic consequences of the latter will be neither severe, nor long-lasting. In this line of argument, Mishkin (2008) advocates that “whether an asset price bubble is occurring or not, as asset prices rise and boost the outlook for economic activity and inflation, monetary policy should respond by moving to a more restrictive stance, [while] after a bubble bursts and the outlook for economic activity deteriorates, policy should become more accommodative” (Mishkin (2008), p. 2). He, nevertheless, makes the distinction that “… in most cases, monetary policy should not respond to asset prices per se, but rather to changes in the outlook for inflation and aggregate demand resulting from asset price movements, [which] implies that actions, such as attempting to ‘prick’ an asset price bubble, should be avoided” (Mishkin (2008), p. 2)152 .

Nevertheless, Issing (2003) points out that the choice of the monetary policy strategy imposes a considerable influence on the stability of the financial system. He argues that if the primary objective of the central bank is to maintain price stability over the medium-term, then to pursue an inflation-targeting strategy with respect to a forecast of inflation spanning over one or two years may not be the optimal policy strategy at all times. He views that in a limited-horizon inflation forecast overall costs (that in terms of the central bank most commonly used tend to be future deflation that succeeds a financial crisis) may not receive the appropriate weight (see also Borio (2005) for further discussion on this point). Optimal monetary policy should, at times, under considerable strains in the financial system, accept deviations from the desired inflation rate over shorter periods so as to preserve price stability over the

152

Mishkin (2007c) and Kohn (2006) also presents similar views on the response of monetary policy to asset prices.

medium to long run. Following this argument, one can also reach the conclusion that monetary policy decisions actually bear on the state of the financial system and may even enable the system to avoid a crisis and gain overall recovery. Equally, taking into consideration risk asymmetries, one can reach the same conclusions. Since a systemic financial crisis tends to produce rather substantial effects, optimal monetary policy may at times deviate by being considerably cautious, so as to reduce the likelihood of a crisis. Eventually, actual inflation will exceed expected inflation (optimally) for a period of time, due to the asymmetries involved, namely the very low probability of a very large loss (Issing (2003), p. 17).

Following the above line of argument, Issing (2003) also expresses the robustness of the ECB stability-oriented two-pillar monetary policy strategy over inflation (forecast) targeting. He stresses that “explicitly focusing on monetary and credit developments in order to form a judgment on consumer price inflation in the medium to long run forces the ECB to take a sufficiently forward-looking perspective…. This longer perspective highlights risks to price stability stemming from financial imbalances,… [and, thus,] the optimal price stability-oriented policy reaction based on monetary and credit developments is likely to diminish financial imbalances” (Issing (2003), p. 17)153. Trichet (2003) also recognises that the first pillar of the monetary policy strategy of the ECB enables an evaluation of the amount of liquidity within the euro area (by monitoring the deviations of the broad monetary aggregate (M3) deviates from its reference value), as well as its allocation. Since, in this way, credit and loan developments in addition to portfolio shifts are carefully monitored

153

See also Issing (2002a) and (2002b) for a discussion on how the first pillar of the ECB monetary policy strategy can take into account financial imbalances.

and scrutinised according to the relevant financial and economic developments, potential risks of a bubble formation are thought to be signalled (Trichet (2003), p. 18).

The pre-emptive role of the first pillar of the ECB monetary policy strategy is also identified by Borio, English and Filardo (2003), who argue that: “…policy frameworks in which monetary aggregates still play a prominent role can more naturally accommodate policies aimed at addressing the build-up of financial imbalances… No doubt, such frameworks can make it easier to justify interest rate increases even in the absence of near-term inflationary pressures as long as the corresponding monetary aggregates are growing fast. … Pillar I in the ECB strategy is rationalised precisely in terms of providing better signals about inflationary pressures beyond short horizons, complementing the assessment of more near-term inflation pressures based largely on real-side indicators under Pillar II” (Borio, English and Filardo (2003), p. 43)154.

Therefore, the debatable issue still remains of whether monetary policy should react directly to asset prices or, even, if asset prices need to appear in some form in a reaction function a central bank uses as a guide for monetary policy. Trichet (2003) suggests that “we should remain cautious about it, perhaps because it would be like opening Pandora’s Box if we started setting our key policy rates according to asset price changes” (Trichet (2003), p. 16). He argues that extreme caution needs to be

154

Nevertheless, Borio, English and Filardo (2003) also accept that even though “responding to rapid monetary growth … foreshadows inflationary pressures … financial imbalances can also herald recessionary, and so potentially deflationary, pressures down the road. This could raise delicate issues of communication and transparency unless a broader set of potential outcomes was explicitly considered” [emphasis is in the original] (Borio, English and Filardo (2003), p. 43).

exercised by monetary policymakers on this issue since it is considerably difficult both to assess asset price valuations and mostly to determine and measure fundamental asset-price values, which he views as highly hypothetical (Trichet (2003), p. 16). Therefore, he remarks that any evaluation whether asset-price changes reflect deep fundamental change or not is a challenging task and gives the NASDAQ tech-stock bubble as an illustration155. Mishkin (2008) also shares the same view concerning the difficulty in accurately identifying asset price bubbles and argues that if a monetary tightening is exercised in order to restrain an asset price bubble that is falsely identified, the prospects of economic growth may substantially deteriorate. Furthermore, Bernanke (2002) points out that when the monetary authorities are not confident in their estimations about both the presence of a bubble and its amplitude, the monetary policy actions aiming to affect asset price developments may lead to a misallocation of resources.

Nevertheless, Trichet (2003) stresses that a crucial issue is whether the central bank’s stance should be different in a reversal of expectations due, for example, to a reassessment of expected profitability asset prices decline, in the effort to foster monetary and financial stability (Trichet (2003), p. 17).

It has been suggested that the monetary policy response to an asset-price bubble should be asymmetric in that when asset prices rise, central banks need not react provided there is no deviation from the price stability objective, while in the opposite

155

Trichet (2003), in particular, points out that even though during 1996 market participants were thought to act in “irrational exuberance”, capital spending increases due to the development of new technologies led to faster productivity growth that resulted in increases in equity prices. Relatively accurate asset valuations were difficult to undertake at the time due to uncertainties about fundamentals, even though central banks were actively concerned with the large movements in asset prices.

case during the unwinding a possible reaction is justified if it is estimated that monetary and financial stability are endangered. However, this asymmetric reaction, especially if it is held is a systematic manner, bears the cost of giving rise to moral hazard (as explained in Chapter 5)156. Therefore, it is questioned if a symmetric (systematic) monetary policy reaction to asset price changes is the appropriate solution. Trichet (2003) does not share that view based on the opinion expressed above that central banks cannot accurately assess asset-price deviations from their fundamental value. He considers that if the monetary authorities do not diagnose the presence of a bubble, thus, not devising an appropriate reaction, they actually encourage market participants to undertake riskier projects based on their false perception of a sound financial and economic environment, bringing about a problem similar to “disaster myopia” (as in Guttentag and Herring (1986)), (Trichet (2003), p. 17).

Nevertheless, De la Dehesa (2002) expresses the opposite view. He argues that the above view that central banks may create moral hazard by cultivating expectations that remedial monetary policy action will be taken if a bubble bursts, probably stems from the fact that market-price changes are, actually, asymmetric. He suggests that “this perception may not happen or may be reduced if the central bank reacts to asset price movements in a symmetric and transparent way” and “the instrument that seems to be more adequate to respond to asset prices developments is the conventional interest-rate policy, [since] experience has proven that traditional

156

Research that analyses the incentive and moral-hazard effects that may arise by a central bank’s considerably aggressive response to a collapse in asset prices includes, for example, Miller et al. (2000), and Caballero and Krishnamurthy (2003).

monetary policy is easier to implement and more effective than other alternative instruments such as the increase in margin requirements or policy signals to influence those movements” (De la Dehesa (2002), p. 2).

Furthermore, when policymakers face large fluctuations in asset prices but muted inflation expectations, it is considered whether inflation is measured accurately, as well as whether price stability is ensured. It is debated whether asset prices should be taken into account when defining price stability, and, generally, whether asset prices may play a significant role in the conduct of monetary policy. The underlying concept is that since valuations of assets are computed in a forward-looking manner, eventually asset prices incorporate expectations referring to future inflation and economic growth. Nevertheless, Trichet (2003) remarks that this view is undermined by the fact that it has been difficult to establish wealth effects in a definitive manner, which although it seems not to apply to the U.S., it is probably the case in the euro area. Additionally, he expresses concerns about the observed divergence of asset prices from the CPI and points out the potential of an “internal conflict, if the objective of price stability is defined by aggregating the changes in the CPI and the changes in asset prices … since the nature of both types of prices is fairly different” (Trichet (2003), p. 18).

Referring to the ECB monetary policy strategy, Trichet (2003) remarks that if monetary policymakers decide not to respond directly to asset price developments, after evaluating all the elements described by the indications of the first pillar (mentioned above), they certainly need to consider the consequences of these developments on aggregate supply and demand, as well as on the confidence and

expectations of economic agents, as they may eventually exercise an effect on price developments (Trichet (2003), p. 18).

Furthermore, Mishkin (2008) argues that actions aiming to “prick” an asset price bubble should better be avoided for (at least) three reasons. Initially, because he accepts that it is difficult to identify asset price bubbles accurately and promptly, and secondly, because he contends that even if this is not the case, still the effect that interest-rate policy can impose on asset price bubbles is highly uncertain. He stresses that tightening interest-rate policy may prove considerably ineffective in its effort to restrain the bubble, since “market participants expect such high rates of return from buying bubble-driven assets”157(Mishkin (2008), p. 2). Since he views bubbles as “departures from normal behaviour” he accepts that “it is unrealistic to expect that the usual tools of monetary policy will be effective in abnormal conditions” (Mishkin (2008), p. 2). Finally, he points out that since asset prices span over a large spectrum of assets trading in the relevant markets, and since in a single period a bubble may be present only in a segment of those assets, monetary policy actions may prove to be “a very blunt instrument in such a case, as such actions would be likely to affect asset prices in general, rather than solely those in a bubble”158 (Mishkin (2008), p. 3).

Finally, Bernanke, Gertler, and Gilchrist (1999), Bernanke and Gertler (2001) and Gruen, Plumb, and Stone (2005) support the above view that monetary policy should not attempt to defuse asset price bubbles and only respond to changes in the

157

Greenspan (2002) also further discusses this point.

158

An additional reason proposed has been that many crashes of asset prices which have become associated with asset price bubbles have had very limited effects on the economy (see the propositions of Mishkin and White (2003) in Chapter 5).

prospects for inflation and aggregate demand as it is thought to lead to superior results even in the event of a bubble.