The ‘Good News’: New Monetary Policies and Macro-Prudential Instruments for Dealing with Crises

In order to enlarge the number of existing policy instruments, monetary policy has searched for unconventional tools (what Blanchard, Dell'Ariccia and Mauro (2010: 10) call ‘exotic' instruments) to add to the usual interventions in the short­term market, with the intention of lowering long-term interest rates, which influence investment and thus recovery and growth.

The ECB and the central banks of other non-EMU countries (e.g. Denmark, Sweden, Switzerland and Japan) have intro­duced the practice of negative interest rates since 2014 in order to encourage banks to increase lending. The action, how­ever, has been almost ineffective, and the enduring deflation has really hindered the effort to expand the economy. Instead, two (partially) new instruments, quantitative easing and for­ward guidance, have been implemented, and another one has been proposed again (‘helicopter money’), that deserve more attention for their effectiveness.

5.3.1 QuantitativeEasing

‘Quantitative easing’ is only a partially new monetary policy tool, since a form of it, Operation Twist, had already been used in the 1960s. It tends to lower long-term interest rates in order to increase private investment. The first such measures (usually referred to as Quantitative Easing 1 (QE1)) - addressed to large-scale asset purchases (LSAP) of govern- ment-sponsored-enterprise (GSE) debt, agency debt, mort­gage-backed securities (MBS) and Treasury securities - began towards the end of 2008 in the United States. The second type of these measures (QE2) started in the second half of 2010 and concentrated on purchases of US Treasury securities.

These unconventional operations have been very effective in raising bond values and lowering sovereign yields (Cecioni, Ferrero and Secchi 2011; IMF 2013; Santor and Suchanek 2013) not only in the United States but also in other countries, as an effect of portfolio reallocation and repricing of the risk in global financial markets (Fratzscher, Lo Duca and Straub 2013). They have also helped to lower bank funding volatility and to increase the loan supply (Carpenter et al. 2013). Weale and Wieladek (2014) find that announcements of these opera­tions are associated with increases in output and inflation in the United States and the United Kingdom.

However, relying too much on them for too long can cause a liquidity glut, speculative activities - as the banks can find these operations profitable only for the capital account appreciation of their portfolio - distributional effects and capital movements both in the country where they are imple­mented and abroad. Moreover, the assets bought by central banks as part of their QE are certainly riskier than short-term government bonds, which could raise problems for the soundness of central banks' balance sheets.

As for distributional effects, QE can affect some persons and institutions positively and some others - at home or abroad - negatively (Bossone 2016; Den Haan 2016).7 By pushing up asset prices, it benefits richer people and, in particular, those households holding a larger share of financial assets.

Some - especially Germans, e.g. the Deutsche Bundesbank (2012), the German Constitutional Court (Bundesverfassungsgerich 2014), Sinn (2013) and Weber (2010), cited by De Grauwe and Ji (2015) - have criticised QE for having negative fiscal implications, as sounder coun­tries can be asked to pay the bill in the case of default of some sovereign states. According to Benigno, Nistico (2015), QE can cause some losses to the central bank if it relieves risky assets from the economy. These measures

For this, see also the international spill-overs thatwe mention below. would entail some kind of fiscal implication, unless the central bank implements a more inflationary policy to reduce its losses. These losses would increase if the assets bought by the central bank are issued by unsafe borrowers, such as Greece (for this, see Frankel 20148). De Grauwe and Ji (2013, 2015) have shown, however, that QE for the Euro area can be designed in such a way as not to have such implica­tions, at least for all countries in the area, e.g. by establishing that the sales of sovereign bonds are in proportion to the equity shares used for the purchase. In any case, as Armstrong et al. (2015) show, the risk deriving from the QE operations begun in 2015 are to be shared to a limited extent between EU institutions, the Euro system according to share­holdings and, mostly, the national central bank of the issuer. This would limit their burden for stronger countries.

International spill-overs (and thus, again, distributional effects) can derive from QE and other expansionary monetary policy tools. First, they act through the current account, as they imply positive direct and indirect spill-overs of such operations on the GDP of other economies, in particular, of LDCs.

8 Frankel advises the ECB to buy US bonds, which would have the advan­tage of lowering the value of the euro, with beneficial effects on exports.

9 An echo of the instability caused by the cessation of unconventional operations can be found in Cova, Pagano and Pisani (2016). unconventional policies implemented in this country, put a relative halt to capital inflows. These inflows, however, sud­denly appeared again when in 2013 the Fed announced a progressive upholding of the expansionary measures, causing an increase in long-term interest rates in the United States (the so-called taper tantrum), a rise in capital inflows also from the emerging markets and capital losses in these countries. Cooperation could mitigate the undesired effects of negative spill-overs not by bottling them up within their country of origin but by sharing them, in exchange for reciprocity (Engel 2015). However, some authors are sceptical about the possibi­lity of implementing coordination of policies to pursue an international equilibrium that is more efficient than the decen­tralised one (Blanchard 2016b). Restrictions on capital flows could be implemented by each country to reduce spill-overs, but they are difficult to be introduced by a single country, since it would fear the negative consequences of discrimination from possibly beneficial foreign investment and outflows of foreign capital after the innovation.

The need and the form for international cooperation are highly dependent on circumstances. However, there is at least a case for contacts, exchange of information and dialo­gue between them

under normal circumstances that make coordinated action possi­ble in exceptional circumstances... Global institutions and fora play an essential role for three reasons... First, they achieve a better understanding and a common assessment of the global spil­lovers from domestic policy actions and the potential policy trade­offs. Second, even if the scope for coordination is limited in good times, they make coordination possible in bad times. And third, they prompt work at (international) regulatory bodies... In so doing, they foster risk-sharing through decentralised markets, avoiding the need to coordinate in the first place. (Creure 2016:10) As a way of concluding on the benefits and dangers of QE, we can say that it is certainly effective and has indeed con­tributed to the expansion of the United States and the British economy, as well as avoiding a disruptive crisis in the Euro area. However, it should not be relied on too much and for too long, and a smooth exit from it is also necessary for its possibly overall negative international spill-overs.

5.3.2 Forward Guidance

‘Forward guidance’ may be defined as announcements made about the future stance of policy with the intention of influen­cing or managing expectations of future policy interventions and hence the expected path of future interest rates and out­comes for the economy.^[68] It may be unconditional or indefinite (e.g. ‘interest rates will remain low for a considerable - or for an indefinite - period of time’) or time dependent (‘for two years’) or conditional (‘interest rates will remain low until unemploy­ment falls below 6.5 per cent’). It can be applied to achieve controllability and stabilisability of the economic system.

A form of forward guidance was first implemented by the Federal Reserve in August 2003 and again in December 2008 and by the ECB in July 2013.

Forward guidance is therefore designed to make the private sector's expectations consistent with the policy intentions of the central bank or government with respect to interest rates, growth and investment - especially at the ZLB (Cmure 2013). There is some evidence that it has reduced the volatility of expectations (Cmure 2013; Filardo and Hofmann 2014). On this view, the governor of the Bank of Canada has argued that forward guidance is best used for stabilisation, especially at the ZLB.^{[69] [70] Various papers provide greater detail on how this policy instrument works.[71]}

Forward guidance is not a duplicate of QE but can or should complement it. Suppose that only the latter were implemented. QE should directly act on long-term interest rates, bypassing expectations on them. However, if no assur­ance were given on future interest rates, the risk might arise of unexpected rises in the interest rate. This would act con­trary to the supposed effects of QE. Then forward guidance is useful in that it can or should directly influence the expectations of future short-term interest rates by giving assurance on them. Obviously, forward guidance may fail if the expectations are not in line with QE. We will deal with forward guidance longer in Chapter 6.

5.3.3 HeIicopterMoney

The last type of unconventional monetary policies is ‘helicop­ter money’, initially suggested by Friedman (1969) and others (e.g. Haberler 1952) before him.^[72] Not very different from heli­copter money is the Keynesian idea of digging holes in the ground to bury banknotes and later digging the notes up again (Keynes 1936: 129). Both consist in distribution of money to citizens or to specific categories of agents, under various forms, by base money, grants or tax rebates, possibly financed by money issuance. It could be particularly effective in the case of interest rates stuck at the ZLB. Due to the form that it can take (especially in the case of tax grants), it can also be con­sidered under the heading of fiscal policy and, just because of its mixed, monetary-fiscal, nature, can be more effective than other unconventional operations. In fact, differently from QE, which implies a true asset swap, it injects purchasing power with no counterpart from the agents. More recently, a number of economists have written in favour of helicopter money (Buiter 2004; Reichlin, Turner and Woodford 2013; Bernanke 2016; Draghi 2016; Bossone 2016) as an effective way out of a crisis.^[73] Others - e.g. the head of the Reserve Bank of India, Raghuram Rajan (see Rajan 2013) - have indicated their fears of an inflationary pressure - which, frankly, at least in Europe and Japan, is currently not a problem - deriving from this type of unconventional monetary policy.

5.3.4 Instruments for Financial Stability

The recent financial crisis has highlighted the strengthening of financial regulation and supervision, which has gone beyond a purely microeconomic approach. The literature on this topic has soared, involving the foundations of macro-prudential tools - their nature, implementation and effectiveness - together with the need for relations with monetary policy and international coordination.

In this section we deal with the following specific issues that have been the object of study:

1. Identifying asset price bubbles and the unintended consequences of unconventional policies due to the possibility of financial instability;

2. Defining macro-prudential policies beyond micro action to counter instability;

3. Implementing macro-prudential policies;

4. Determining the consequences and effectiveness of these policies; and

5. Coordinating macro-prudential policies at an interna­tional level to meet cross-border spill-over effects.

We defer consideration of issues of coordination of all macro­economic policies to Section 5.5.

5.3.4.1 Identifying Asset Pricing Bubbles and Unintended Consequences of Unconventional Policies

Galati and Moessner (2013) attempted to explain some of the roots of the financial crisis. Overconfidence in the self­adjusting ability of financial markets - helped by improper policies of financial deregulation - led to an underestimation of the consequences of a growing financial sector, increasing indebtedness and financial innovation and reduced risk pre­mia. Soaring asset prices, bubble formation and financial imbalances were some of the consequences.

Some of the underlying theoretical reasons can be traced back to the implausible assumption of complete market par­ticipation underlying traditional asset-pricing theories. Limited participation implies a complete change in the prop­erties of models, as even small shocks cause relevant price volatility and multiplicity of equilibria (Allen and Gale 1994). Interbank loans can transmit and amplify shocks and cross-border spill-overs and transmit them from one country to another (Allen and Gale 2000; Nocciola and Zochowski 2016). The size and extent of unconventional policies to tame the crisis, which have sharply increased liquidity, put emphasis on the risk of new financial booms and asset bub­bles and thus on the need to devise proper regulation.

When monetary policy was practically the only possible action for dealing with asset bubbles, the issue had been raised whether it should counteract them. A conclusion derived by Filardo (2001) was that consistently with the prescription of control theory, monetary policy should include asset-price bubbles in its reaction function, if asset prices contain reliable information about the state of the economy, in particular, inflation and output. Now it has been realised that monetary policy can be relieved of some burden if complemented by a suitable regulatory policy. Then the two can be geared together to deal with asset bubbles (see Section 5.5).

5.3.4.2 Defining Macro-Prudential Policies: Beyond Micro Action

Micro-prudential regulation - e.g. quality/quantity of capi­tal, leverage ratio, liquidity requirements - aims at forcing banks to internalise possible losses on their assets to protect deposit insurance funds and reduce moral hazard. This kind of regulation - even if of general application to all banks - does not consider the links between the various financial institutions as well as the pro-cyclicality of the operation of the financial system, which can lead to financial imbalances and systemic risk that cannot be counteracted by simple micro-prudential regulation. In fact, this improves the resi­lience of individual financial institutions but disregards the issues of the financial system as a whole and can even have pro-cyclical negative effects. Systemic risk can arise from different types of externalities, such as the strategic interac­tion of financial institutions, which can lead to complemen­tarities during expansions; a generalised sell-off (‘fire sales') of financial assets initially due to asset sales of one bank facing difficulties; self-fulfilling equilibria generated by exo­genous shocks; amplification mechanisms of negative shocks (e.g. contagion as an effect of linkages); endogenous financial instability; and the propagation of shocks through various kinds of financial links. Macro-prudential regulation has exactly the objective of coping with these externalities and thus increasing the resilience to systemic risk. It consists in many cases of the same tools as micro-prudential regulation, e.g. ceilings on credit, on credit growth or on the loan-to- value ratio, or foreign currency lending, or the debt-to- income ratio, or minimum values of liquidity-related values. These measures can be applied to specific sectors - e.g. a mortgage cap - or to the whole economic system. The former can be particularly efficient because they target the origin of the imbalance. Thus, their potential positive effects are enhanced, while the negative effects on output are lessened. Macro-prudential policies can also be discretionary or rule based - i.e. automatic stabilisers - such as countercyclical capital requirements (higher in times of prosperity, lower during contractions) and cyclically dependent liquidity requirements. In any case, also discretionary macro­prudential policies can and need to be adjusted at different phases of the cycle exactly in order to smooth out cyclicality. The various macro-prudential policies can be combined to better correct systemic risk, but capital requirements are likely to play a crucial role (De Nicolo, Favara and Ratnovski 2012).

5.3.4.3 Implementing Macro-Prudential Policies

Basel III (i.e. the Third Basel Accord) is a global voluntary regulatory agreement on bank capital requirements, agreed to in 2010-11, to be implemented by steps and finally by 31 March 2019. It arose as a reply to the deficiencies in financial regulation revealed by the financial crisis that began in 2007-8. It increases bank liquidity and decreases bank leverage by strengthening capital requirements. In particular, it deals with systemic risk in stating the potential for national financial regulators to introduce ‘discretionary counter-cyclical buf­fers’. This can be done by requiring up to 2.5 per cent of capital during periods of high credit growth. Doubts, however, have been raised about the size of the requirements being enough to address systemic risks (Schoenmaker and Wierts 2015). Other doubts can also arise with reference to the discretionary nat­ure of the change in requirements. We deal with them below.

Any indication about the practical effectiveness of this agree­ment - even in the process of its implementation - is obviously premature. As to previous tools of macro-prudential policies, Lim et al. (2011) provide a first exploration of available data from forty-nine countries. Many instruments have been shown to be effective in reducing pro-cyclicality, even if their effec­tiveness is different based on the type of shock faced by the financial sector. Cerutti, Claessens and Laeven (2017) use an even richer data set, exploiting data for 119 countries in the period 2000-13, and find that these policies tend to lower credit supply, the more so in emerging countries. Borrower­based measures - such as loan-to-value and debt-to-income ratios - are effective for most countries, and foreign currency- related measures are effective mainly for emerging markets. Ba⅝kaya, Kenc et al. (2016) first draw a distinction between price- and quantity-based tools, the former immediately lead­ing to a change in the price/cost of the various alternatives and the latter putting a limit directly on the amount of the targeted item, either asset or credit. Quantity-based measures reduce total credit growth for all levels of financial development, whereas price-based tools are relatively more effective in the case of financially developed systems.^[74]

In addition to the choice of tools, an issue arises about the governance of supervision. Since supervisors have to look at various microeconomic and macroeconomic indicators in order to implement the most appropriate tools, the mandate to be conferred on them is necessarily of the kind of an incomplete contract. Then an issue of principal agent arises, which has consequences on the effectiveness of macro­prudential policies (Masciandaro and Quintyn 2016).

5.3.4.4 Determining the Consequences and Effectiveness of These Policies

When macro-prudential policies of the kind required by the Basel III agreement are implemented, the risk arises as to their effectiveness being impaired as a consequence of Goodhart’s law. Rational agents can, in fact, react to the rules and make them ineffective or even dangerous. Moreover, they could anticipate that regulators will tend to be lenient in their countercyclical policies, as an effect of pressures from the financial industry and politicians, thus creating ex ante moral hazard (Horvat and Wagner 2016). Finally, some macro-prudential policies may be not only ineffective but also counterproductive, as they can increase systemic risk.

Countercyclical capital requirements can add to the pro­blem of excessive correlation of risks, since these can be endogenous. In fact, such requirements ‘protect banks against common shocks, but not against bank-specific ones’, thus lowering costs from aggregate risk and increasing their incen­tives to invest in common projects. This can increase systemic risk rather than decrease it. A solution could be that capital requirements are raised for systemic banks (Horvat and Wagner 2013: 5). This could have the effect of reducing the internationalisation of banking, which is really due also to other factors (Forbes, Reinhardt and Wieladek 2016).

However, in general, macro-prudential policies can reduce the negative effects of capital flows volatility (see e.g. Ba⅝kaya, di Giovanni et al. 2016).

5.3.4.5 CoordinatingMacro-Prudential Policies at an International Level to Meet Cross-Border Spill-Over Effects Cross-border financial operations raise international spill­overs, whose sign and magnitude depend - inter alia - on the ownership structure of financial institutions and their linkages. Cross-border financial linkages are the channel through which not only shocks and bubbles but also policies (in particular, macro-prudential policies) in one country can transmit to other countries, as shown by the experience of the financial crisis that began in 2007 and of the ensuing policies. This raises a number of issues about how to coordi­nate national macro-prudential policies in a way to avoid excessive negative spill-overs and reduce their impact. In particular, the existence of a common prudential standard and supervisory body can be useful (Cecchetti and Tucker 2015).^[75] In any case, bilateral agreements for stronger reci­procity arrangements between countries and - at least - some notification of the looming bubbles and policy actions are necessary for mitigating leakages. An (un-)intended conse­quence of monetary and regulatory policies can be banking de-globalisation (Forbes, Reinhardt and Wieladek 2016).

5.4