Fiscal stress and monetary policy stance in oil-exporting countries

https://doi.org/10.1016/j.jimonfin.2020.102302Get rights and content

Abstract

We documented that for some oil-exporting countries, the correlation between exchange rates and oil prices is strongly negative during periods of significant oil price drop but is much weaker during other periods. To interpret this time-varying asymmetric correlation, we develop and estimate a Markov-switching small open economy New Keynesian model with oil income as a source of government revenue. In particular, we allow monetary and fiscal policy coefficients to switch across “active” and “passive” regimes. Using data on Russia, our result shows that the policy combinations fluctuate. We find that active monetary policy isolates the exchange rate from oil price variations but changes to passively tolerate depreciation and inflation to support government debt when oil price drops place fiscal policy in a state of stress. Counterfactual policy experiments suggest policy regime switching is crucial to account for the observed asymmetric impact of oil prices on the exchange rate and that the transmission channels of oil price shocks differ significantly across policy regimes.

Introduction

A large volume of empirical studies have found that oil price variations have an important effect on economic dynamics for small oil-exporting countries. However, whether and how the impact varies over time received less attention. Fig. 1 plots the dynamics of oil prices and exchange rates for six oil-exporting countries. The shaded areas indicate two periods of significant oil price drop.1 We can see that oil price changes are associated with substantial exchange rate movements during these two periods, but not during other periods, except for Norway. Table 1 further reports the statistical relationship between oil prices and exchange rates across different periods. It is evident that the correlation between exchange rates and oil prices varies substantially over time, which is highly negative for all countries during sharp oil-price-drop periods and becomes much weaker, or even positive, for all sample countries except Norway during other periods.2 A natural question then is why the effect of oil price variations is asymmetric across different periods for some oil-exporting countries?

A commonly held view for this time-varying asymmetric impact of oil prices on exchange rates for some oil-exporting countries is that they adopt managed floating exchange rate arrangements during normal times; however, oil price plunges generate fiscal stress and constrain the ability of monetary policy to stabilize exchange rates and the economy.3 To see the fiscal stress created by lower oil prices, Fig. 2 shows that oil income accounts for more than 50% of fiscal revenue and that the fall of oil income significantly drives down fiscal revenue for the sample countries.4 In contrast, oil price variations cause a time-invariant symmetric effect on exchange rates for oil exporters under a free-floating system, such as that seen in Norway. Therefore, taking into account the role of monetary and fiscal policy interaction in transmitting oil price shocks seems crucial to explain the business cycle dynamics for the oil-exporting countries under managed floating exchange rate system. However, the existing literature has not yet provided a formal quantitative evaluation of this view. In this paper, we attempt to take this task.

In particular, we develop and estimate a Markov-switching small open economy New Keynesian dynamic stochastic general equilibrium (DSGE) model with oil income as a source of government revenue. In our framework, the government collects tax revenue from oil production in foreign currency and issues long-term debt in domestic currency to finance its domestic currency expenditures. Importantly, to quantitatively analyze the role of monetary and fiscal policy interaction in transmitting oil price shocks, we allow the monetary and fiscal policy regime to switch stochastically. When oil prices are high, the budget perspective is bright, and investors expect future budget surplus is sufficient to retire government debt. The ample fiscal space in turn accommodates monetary policy to actively target exchange rate movements and inflation. We follow the terminology developed by Leeper (1991) to label this policy combination as “active monetary/passive fiscal” (AM/PF). On the contrary, lower oil prices adversely influence the fiscal budget, and investors expect that fiscal policy will be unable to adjust to repay government debt. In this case, monetary policy is restricted by fiscal policy and has to allow higher inflation to reduce the real value of outstanding government debt denominated in domestic currency, as well as exchange rate depreciation to increase the relative value of foreign-currency-denomiated oil income against domestic-currency-denominated government expenditures. We label this policy configuration as “passive monetary/active fiscal” (PM/AF). In addition, we follow Bianchi, 2012, Bhattarai et al., 2016, Bianchi and Ilut, 2017 to permit “active monetary/active fiscal” (AM/AF) and “passive monetary/passive fiscal” (PM/PF) regimes in the estimation. To perform sensible quantitative evaluation, our model also features consumption habit formation, endogenous capital utilization, real frictions, sticky prices and wages, and various shocks.

We estimate our model to Russia, a prototype oil exporter that manages its exchange rate and heavily relies on petroleum revenue to finance government expenditures. We find that monetary and fiscal policy regimes fluctuate over the sample period. During most of the sample period, the monetary policy is in the active regime, but it switches to the passive regime during periods of substantial oil price falls. Narrative evidence from the central bank of Russia suggests that the model-inferred policy regime switching periods correspond to stated changes of monetary policy. In particular, Russia increased the operational band of its managed floating exchange rate framework in late 2008 and attributed the reason to sharp changes in the global commodity markets. In November 2014, Russia cancelled the operational band for exchange rate interventions. In addition, the implied active fiscal policy regime aligns with the oil price plunge period.

The transmission channels of oil price variations on the macroeconomic variables differ significantly conditional on different policy combinations. A fall of oil price generates budget and trade deficits. In the AM/PF regime, the government is able to raise tax revenue in response to higher government debt and monetary policy can keep the exchange rate and inflation stable. As a result, real interest rate rises significantly. This in turn amplifies the initial negative oil price shock. On the other hand, in the PM/AF regime, monetary and fiscal policy jointly determine the macroeconomic dynamics and the price level. A negative oil price shock is inflationary and depreciationary. Since the nominal interest rate only weakly increases in response to the rise of inflation and exchange rate in this regime, real interest rate falls and real exchange rate depreciates significantly. These real price adjustments mitigate the unfavorable oil price shock. Thus, our result suggests the importance of considering policy regime switching when examine the impact of oil price shocks for oil-exporting countries.

Moreover, we perform counterfactural policy experiments to examine the importance of monetary and fiscal policy regime switching in producing the asymmetric relationship between exchange rates and oil prices over time. We find that conditional on the AM/PF regime over the oil price plunge period, the model generates only a weak correlation, which is inconsistent with the data. Our estimation result also shows that the macroeconomic stability implications differ across policy regimes. Theoretical variance decomposition suggests that the oil prices shocks contribute more to the volatilities of macroeconomic variables under the PM/AF regime relative to that under the AM/PF regime. As a result, ignoring the policy regime switching may lead to misguided conclusion on the main drivers of business cycles.

Section snippets

Literature review

Our work belongs to the literature that analyzes the effect of monetary and fiscal policy in transmitting external shocks for open economies. Many studies have been devoted to examining the role of monetary policy and exchange rate regimes. For example, Broda, 2004, Edwards and Yeyati, 2005, Devereux et al., 2006 demonstrate that the impacts of external shocks vary substantially across different exchange rate regimes.5

Model

In this section, we lay out a two-sector small open economy New Keynesian model. The model incorporates an oil production sector, whose products will be exported, and the government collects tax revenue from oil production. Moreover, we allow the monetary and fiscal policy rules to individually switch regimes according to the region of policy rule coefficients.

Quantitative analysis

In this section, we describe the data used in the quantitative analysis, the calibration strategy, and the prior and posterior distribution for our model parameters.

Empirical results

We present the results obtained from estimating the Markov-switching small open economy DSGE model in this section. We first report model fit, and then examine the theoretical variance decomposition of structural shocks, emphasizing the contribution of oil price shocks. We also report the regime probabilities and the regime-specific impulse responses to emphasize how the monetary and fiscal policy mix influences the dynamics of key macroeconomic variables, especially inflation and exchange rate.

Counterfactual analysis

In this section, we conduct two counterfactual policy experiments to illustrate how the policy regime switches affect the observed macroeconomic dynamics.

Robustness check

Next, we conduct a couple of robustness checks for the main results. First, we check the sensitivity of the estimation results to different prior distributions. Following Smets and Wouters (2007), we broaden the 90% probability intervals of the prior distributions by 50% for all the parameters. The estimation results do not change much.

Second, admittedly the Russian economy is in a state of transition, and there maybe uncertainty about which long-run equilibrium the economy is converging to. To

Conclusion

In this paper, we documented how oil price variations have time-varying asymmetric impacts on exchange rates for some oil-exporting countries. The correlation between oil prices and exchange rates is strongly negative during periods of significant oil price drop, but it becomes weak, or even positive, during other periods. To interpret the data pattern, we developed a Markov-switching small open economy New Keynesian model with tax income from oil production and estimated the model based on

References (35)

  • K. Hou et al.

    Oil price shocks and their transmission mechanism in an oil-exporting economy: a var analysis informed by a dsge model

    J. Int. Money Finance

    (2016)
  • E.M. Leeper

    Equilibria under active and passive monetary and fiscal policies

    J. Monet. Econ.

    (1991)
  • A. Pieschacón

    The value of fiscal discipline for oil-exporting countries

    J. Monet. Econ.

    (2012)
  • S. Schmitt-Grohé et al.

    Closing small open economy models

    J. Int. Econ.

    (2003)
  • A. Berg et al.

    Public investment in resource-abundant developing countries

    IMF Econ. Rev.

    (2013)
  • S. Bhattarai et al.

    Policy regimes, policy shifts, and us business cycles

    Rev. Econ. Stat.

    (2016)
  • F. Bianchi

    Regime switches, agents’ beliefs, and post-world war ii us macroeconomic dynamics

    Rev. Econ. Stud.

    (2012)
  • Cited by (11)

    • Is refined oil price regulation a “shock absorber” for crude oil price shocks?

      2023, Energy Policy
      Citation Excerpt :

      Based on the work of Hamilton (1983), scholars began to pay attention to the relationship between oil prices and economic growth (Jiménez-Rodríguez and Sánchez, 2005; Lardic and Mignon, 2006; Benhmad, 2013; Balcilar et al., 2017) and consider it from the perspective of nonlinearity (Mork, 1989; Hamilton, 1996; Lee et al., 1995; Balke et al., 2002; Kilian and Vigfusson, 2011; Jiménez-Rodríguez, 2009; De Oliveira et al., 2021). In addition to directly studying the relationship between oil prices and economic development, some scholars also pay attention to the role of government policies, especially monetary policy (Leduc and Sill, 2004; Kilian and Lewis, 2011; Herrera and Pesavento, 2009; Delpachitra et al., 2020; Jin and Xiong, 2021). Although a large number of pieces of literature have studied the relationship between crude oil price and economic operation, it was not until around 2010 that scholars gradually turned their attention to China to explore the impact of international crude oil price fluctuations on China's economic operation, but there are still differences on the relationship between the two.

    • Cross-category spillover effects of economic policy uncertainty between China and the US: Time and frequency evidence

      2022, Journal of Asian Economics
      Citation Excerpt :

      Blagrave et al., 2017) illustrated that when the source country is in an accommodative monetary policy, domestic fiscal policy shock implies greater demand for imports and subsequently transmits larger spillovers through expenditure shifting. Also, public debt pressures from fiscal shocks have forced monetary policy to adjust passively to tolerate currency depreciation and higher inflation (Jin & Xiong, 2021). Static analysis is the total amount of analysis for the full sample and all time periods.

    • Alternative monetary policy regimes in an oil-exporting economy

      2022, Quarterly Review of Economics and Finance
    • Do central banks respond to exchange rate movements? A Markov-switching structural investigation of commodity exporters and importers

      2021, Energy Economics
      Citation Excerpt :

      Our work also contributes more generally to a broader literature that emphasizes the importance of allowing for parameters and volatility of shocks to change when analysing policy questions. So far, only a few papers in the literature address this issue in open economies, see for instance Liu and Mumtaz (2011), Dybowski et al. (2018) and Jin and Xiong (2020).9 However, changes in shock variances and/or studies of the great moderation in closed economies such as the US have for some time been documented, see e.g. Stock and Watson (2005), Sims and Zha (2006), Lubik and Schorfheide (2004), Smets and Wouters (2007), and Bjørnland et al. (2018).

    View all citing articles on Scopus

    We would like to thank two anonymous referees, as well as participants at the 5th HenU/INFER Workshop on Applied Macroeconomics, the 2019 Asian Meeting of the Econometrics Society, the Tsinghua Workshop in International Finance 2019, and Xiamen University brownbag seminar for their helpful comments. Jin acknowledges financial support from China National Natural Science Foundation (Grant No. 72003160), China Ministry of Education Project of Humanities and Social Sciences (Grant No. 20YJC790054), and Basic Scientifc Center Project 71988101 of National Science Foundation of China . All errors are our own.

    View full text