Unconventional monetary and fiscal policies in interconnected economies: Do policy rules matter?

https://doi.org/10.1016/j.jedc.2018.01.028Get rights and content

Abstract

This paper examines the effects of adopting unconventional policies in a crisis environment characterised by the international transmission of negative financial intermediation and real capital quality shocks. Using a two-country model with financial frictions, we compare adjustments in both countries. We condition one country to adopt a credit easing rule as the monetary regime, regardless of the source of the crisis. We consider results when the other country does nothing, or implements a fiscal (tax-rate) policy rule. Our results show that when both countries experience massive asymmetric shocks, the adoption of optimal policy rules can be used effectively to mitigate negative consequences for both economies. However if the crisis events are due to similar shocks, unconventional monetary policy or fiscal policy can improve conditions in own country, whilst worsening economic conditions in the other country, generating beggar-thy-neighbor growth outcomes.

Introduction

Many advanced economies have been experiencing unusually deep and prolonged crises. Spain and Italy are in a growth slump, with an even slower pick-up in employment. Others like Greece and Japan are facing fiscal problems and in dire need to rein in deficits and reduce the size of their public debt. These debt-laden countries are further constrained by official interest rates at the zero lower bound rendering ineffective the use of monetary policy as a stabilization (and/or stimulatory) tool. Furthermore, for many economies the depreciation of the exchange rate is not available as a channel to facilitate growth. To compound the problem, economies are inter-linked by trade and financial flows, so that domestic policies in one country have effects on other countries and vice-versa.

The issue of a zero lower bound (ZLB) has resulted in countries adopting unconventional monetary policies to inject liquidity into the system.1 For example, four major developed economies, the United States, the United Kingdom, the Euro Area and Japan have engaged in massive expansions of domestic liquidity. In the US, for example, the expansion came in various waves after the onset of the financial crisis in 2008, with the monetary base, rising from just over $600 billion in 2000, to nearly $4 trillion in 2014. These liquidity injections have been labeled generically as Quantitative Easing (QE) policies, then as Large Scale Asset Purchases (LSAP), even though as noted in a speech by Bernanke (2009) the policy adopted in the US is more appropriately described as credit-easing as the intention is to swap types of assets to influence spreads. In the Euro Area such injections were called Long Term Refinancing Operations, Securities Market Programme and Outright Monetary Transactions. The first wave of QE in Japan was in 2001. In an effort to stimulate its economy, it was re-implemented in 2010 and even more aggressively in 2013, in order to generate an uptick in inflation.

The surge in liquidity has generated massive capital flows to many countries, leading to concerns about the volatility of such flows and problems of sudden stops.2 For many countries, these flows have lead to lower interest rates, in order to prevent exchange-rate appreciation. Fiscal policies to stimulate the economy were adopted but this only aggravated the fiscal debt problem. At the same time, higher government borrowing pushed up the credit rating, so the premium also increased. Overall, the prolonged years of crisis have yielded massive expansion in both debt and liquidity.

Needless to say, there has been much discussion of the effectiveness of QE policies both within the domestic country implementing QE and in the economies experiencing the spillover effects of the QE policies, in terms of surging capital inflows and low trade flows. The importance of studying policies in an interconnected world has also been highlighted by Collard et al. (2016), in their study about optimal monetary and prudential policies.

Less attention has been paid to the use of optimal tax rate rules as alternative options to unconventional monetary policy, based on the work by Correia et al. (2013). These authors found that optimal tax-rate rules can be effective substitutes for monetary policy when the interest rate is at the zero lower bound, since the tax rates enter the Euler equations (like interest rates) and affect the intertemporal allocation of resources. Just as Sims (2010) referred to QE polices as “quasi-fiscal” monetary policies, such tax-rate rules represent “quasi-monetary” fiscal policies.

This paper examines the effectiveness of unconventional monetary policy (here referred to as a QE regime), as well as a tax-rate rule in non-QE countries (for convenience of exposition, we have called this the FT regime). Given the prevalence of unconventional monetary policy, attention is paid to how the adoption of QE in one economy affects the other country, and how the other country can best respond, The central questions of this paper are: first, when are QE policies helpful (in the sense of lifting GDP) to the QE-implementing country? Second, under what conditions are tax-rate rules in non-QE implementing countries also helpful in times of prolonged crisis? We investigate the issue under different scenarios about the source of the shocks.

Crises come and go, some are deeper and more prolonged than others, and crises can originate from different sources. Furthermore crises have spatial repercussions and they could be closely correlated in time. Our aim is to contribute to an understanding of the effectiveness of unconventional policies using a two-country model where the economies are highly integrated in trade and finance. Our specific focus is on the mix of policies under alternative shock scenarios.

From a modeling perspective, Gertler and Karadi (2011) have examined unconventional monetary policy in a model with financial frictions. Dedola et al. (2013) extended this model to an open-economy two-country framework with flexible prices, with each economy of equal size. They found that unconventional policies work best if the objective is to maximize joint utility in a cooperative framework, for the calibration of optimal simple rules based on the spread between the return on equity and the risk-free rate. In further work, Kolasa and Lombardo (2014) explore the implications of such cooperative policies for price-stability targets of the monetary authorities in each country.

We adopt the Dedola et al. (2013) model, but extend the analysis to allow the economies to be subject to different types of shocks and to adopt alternative policy-response rules. In particular, we go beyond the case explored in DKL where both countries adopt unconventional monetary policy to allow one country to adopt fiscal options, specifically a tax-rate policy rule, while assuming that the other country adopts credit-easing. This allows us to investigate the interaction of unconventional monetary and fiscal policies, across two-countries, since fiscal adjustment and debt have also played major roles in policy discussion amongst crisis-prone countries. Our analysis considers whether policy rules in response to crises matter for two highly integrated economies subject to different recurring negative shocks.

We compare a number of scenarios, either with both countries subject to the same type of shocks – real capital quality shocks or financial intermediation shocks as well as cases when the economies are subjected to different types of shocks (for example one experiencing capital quality shocks which affect productivity while the other is subjected to financial shocks which affect the leverage ratios).

Our approach is to examine first the behavior of key variables in the base crisis scenario with no policy response in either country. Then we examine outcomes with an unconventional monetary policy rule in place in one country, while the other country is faced with two options – do nothing, or adopt an optimal tax-rate rule for labor income. In terms of methodology, we go beyond analyzing impulse responses of key variables in each country. We also analyze outcomes from a very long simulation. Specifically, the model is simulated for recurring shocks for a very long run of T = 10,000.

The long simulations allow us to do two types of analysis. First, the overall distributions of outcomes in each country under the base “no policy” response regime and under alternative policy regimes allow us to identify the probability of differing outcomes, namely, the likelihood of a severe negative outcome for output.

Secondly, in order to capture better the “disparate confounding dynamics” noted by Faust and Leeper (2015) and usually overlooked in conventional monetary analysis, we also adopt a variant of the Mendoza (2010) approach for the analysis of crisis events. In particular, we isolate sub-periods when the GDP is at low values (namely two standard deviations below its stochastic mean) in the base case of no policy response and then we examine the adjustment of key variables for three periods before and three periods after this worst case. Then, for the same sequence of shocks, we allow for policy responses, and examine the adjustment process pre-crisis and post-crisis under various policy scenarios. We refer to these periods of adjustment in crisis as “dark corners”, following Blanchard (2014). This approach allows us to capture not only the overall volatility of key variables, but also their adjustment during the worst-case crisis scenarios.

Our analysis shows that when both countries experience massive common real shocks (associated with falls in capital quality) neither QE or FT rules are particularly effective. In contrast, when the global shocks cause falls in the leverage ratios (due to drop in confidence), the adoption of optimal policy rules can be used effectively to mitigate negative consequences for the economy. However in this scenario, the adoption of an optimal QE or FT rule in one country shifts the distribution toward GDP growth in its own favour while creating a beggar-thy-neighbour lower GDP growth rate in the other country.

For the case of asymmetric shocks, the effectiveness of joint policies is clearer. We find that a tax policy in the country experiencing severe capital quality shocks and a QE policy rule in the country experiencing severe financial intermediation shocks can result in win-win outcomes. Interestingly, when the matching of policy and shocks is changed to – tax-rule to manage financial intermediation shocks and credit-easing rule to manage capital quality shocks – the combined QE and FT rules can yield growth outcomes that are beneficial to both, even though each policy rule, when used by itself, is not especially effective.

To be sure, in this paper, our modeling of spillover effects of non-traditional policies, in times of crisis, is not intended to capture the experiences of specific countries or regions, such as the effects of ECB policies on the highly indebted countries of the Eurozone, or on non-Eurozone European countries (whether pegged to the Euro, or not). Nor do we try to capture the effects of US or Japanese quantitative easing on emerging markets. We focus on understanding which conditions make such policies have “beggar thy neighbor” effects or create win-win outcomes.

Although our analysis is stylized, it appears to be consistent with recent empirical studies. For example, Bluwstein and Canova (2015) use a structural VAR to examine the effects of ECB unconventional monetary policies on nine non-Eurozone member European countries, grouped as advanced (Sweden, Norway, Denmark and Switzerland), Central Eastern European (Poland and the Czech Republic) and Southern Eastern European (Hungary, Romania and Bulgaria). They find that there are no generalized beggar-thy-neighbor effects of such policies. Using pairwise combinations between the Euro Zone and the non-Euro Zone countries, they find that some countries benefit and other countries were worse off, as a result of the ECB policies.3

Importantly, Bluwstein and Canova (2015) find that the exchange rate channel is not very important for shaping the responses in the foreign countries not following unconventional monetary policies, which they note is in sharp contrast to the international transmission effects of conventional monetary policy actions. They point out that the credit channel, working through the bank lending and balance-sheet sub-channels, as well as the confidence channel, emerge as the key links between the Euro Zone and the non-Euro Zone countries.

The next section describes the model specification, its calibration and the unconventional monetary policy (QE) and unconventional fiscal policy (FT) rules. The third section contains an analysis of the simulation results for recurring negative capital quality and financial intermediation shocks. The last section concludes.

Section snippets

The model

The model is fully described in Dedola et al. (2013), henceforth denoted as DKL. This paper is an open-economy extension of the model developed in Gertler and Karadi (2011), which we denote as GK. It is a two-country economy model with households, firms and financial intermediaries (bankers, which are a subset of householders). There is also a government sector which is responsible for monetary and fiscal policies. The model is described briefly in the next section and includes only the key

Simulations

The analysis in DKL is about the effect of shocks in the home country and the extent of spillovers. Here we are interested in shocks not just in the home country, but in the other country. Specifically, we are concerned with the choice of policies – monetary or fiscal – in an integrated world subjected to a series of shocks – but the shocks work through either real side (capital quality) or financial side (leverage) of the economy. We are interested in scenarios where both countries are subject

Conclusion

The focus of this paper is the management of crisis situations for countries that are open to repercussions from each other’s policies as well as common sources of economic risk. We investigated situations when the sources of shocks are the same and when they are different. The two policy options considered are credit easing which acts on investment and a tax rate rule which acts on labour income and on labour supply.

The shocks analyzed comes from two sources – changes in the quality of capital

References (27)

  • K. Forbes et al.

    Bubble thy Neighbor: Portfolio Effects and Externalities from Capital Controls

    (2012)
  • G. Kaminsky et al.

    When it rains, it pours: pro-cyclical capital flows and macroeconomic policies

    NBER Macroeconomics Annual 2004

    (2005)
  • R. Banerjee et al.

    Self-Oriented Monetary Policy, Global Financial Markets and Excess Volatility of International Capital Flows

    (2015)
  • B.S. Bernanke

    The Crisis and the Policy Response

    (2009)
  • G. Bernanke et al.

    The financial accelerator in a quantitative business cycle framework, chapter 21

    Handbook of Macroeconomics, Volume 1, Part C

    (1999)
  • O. Blanchard

    Where danger lurks

    Finance Develop.

    (2014)
  • K. Bluwstein et al.

    Beggar They Neighbor? The International Effects of ECB Unconventional Monetary Policy Measures

    (2015)
  • M. Brunnermeier et al.

    International credit flows and pecuniary externalities

    Am. Econ. J.

    (2015)
  • M.D. Chinn

    Global Spillovers and Domestic Monetary Policy

    (2013)
  • L.J. Christiano et al.

    Risk shocks

    Am. Econ. Rev.

    (2014)
  • F. Collard et al.

    Optimal monetary and prudential policies

    Am. Econ. J.

    (2016)
  • I. Correia et al.

    Unconventional fiscal policy at the zero bound

    Am. Econ. Rev.

    (2013)
  • L. Dedola et al.

    Global implications of national unconventional policies

    J. Monet. Econ.

    (2013)
  • Cited by (3)

    We would like to thank three anonymous referees for extremely useful comments. Discussion from participants at the 21st International Conference on Macroeconomic Analysis and International Finance (University of Crete, May 2017) and at the SCE 23rd International Conference on Computing in Economics and Finance (New York, June 2017) are also gratefully acknowledged.

    View full text