Elsevier

Economic Modelling

Volume 102, September 2021, 105573
Economic Modelling

Fiscal stance and the sovereign risk pass-through

https://doi.org/10.1016/j.econmod.2021.105573Get rights and content

Highlights

  • Surges in spreads display nonlinearities in fundamental shocks' transmission.

  • Stress states of heightened amplification are detected.

  • Fundamentals play a modest role in explaining the historical spreads' dynamics.

  • A tight sovereign-bank risk pass-through emerges under high bank stress.

Abstract

Sovereign risk surges are tightly linked to bank risk and primary deficits during crises. While the literature documents this unconditional evidence, identification of the main channels driving the state of the sovereign-bank risk pass-through and its fiscal premia component remains an open issue. We estimate a Markov-switching VAR on Italian data for the period 1990–2019 to describe the run up of sovereign and bank credit risk in an environment where regime switches determine the extent to which the fiscal stance feeds risk. We find that a model displaying recurrent regimes affecting both shocks’ sizes and the transmission mechanism between fundamentals and spreads best explain the data. Stress states of heightened risk amplification and a modest role for fundamentals are historically identified. These states feature increased risk sensitivity to primary deficits and to fiscal multipliers, and a tighter sovereign-bank risk pass-through.

Introduction

Following the global financial crisis, several European countries experienced acute peaks in both sovereign and bank credit yields, leading to substantial misalignments with respect to risk-free rates. The surge in spreads is commonly interpreted as reflecting a vicious cycle of fiscal and financial distress (Diebold and Yilmaz, 2009). Nevertheless, the debate on the identification of the key drivers of these abrupt changes remains, and, calls for an assessment of the role of macroeconomic fundamentals in their determination and transmission. Some studies (Mendoza and Yue (2012); Corsetti et al. (2013)) interpret the deterioration of the fiscal outlook as driven by worsening fundamentals (debt, debt-to-GDP ratio). Others assign a key role to expectations, which may be related to debt dynamics (Calvo, 1988), but directly trigger confidence crises in sovereign and private bond markets. Also, identifying the contribution of fundamentals to spreads’ fluctuations advise the policy making, when selecting the fiscal stance.

Another line of research links variations in sovereign risk to bank stress and vice versa, namely through the “doom loop” (Farhi and Tirole (2018); Bocola (2016)). On the one hand, sovereign risk feeds into bank fragility through the balance sheets' exposure to government imbalances, impairing banks’ ability to rollover liquidity. This operates also through expectations of higher risk on bank holdings, implied by a recessionary outlook. On the other hand, a banking crisis spills over to sovereign risk when government guarantees are expected. This two-sided feedback mechanism generates a cycle of comovements explaining sovereign and bank yields beyond the contribution of fundamentals.

This paper jointly addresses the analysis of the drivers of sovereign risk and its pass-through to the bank credit market using Italian data for the period 1990–2019.1 We estimate a Bayesian Markov-switching vector autoregressive model (MS-VAR),2 including three domestic macro fundamentals -the cumulative primary deficit to one-month-lagged debt ratio (measuring the fiscal stance (Sims, 2011)), GDP and price levels- and two measures of risk, sovereign and bank spreads. Given our recursive identification strategy following Kliem et al. (2016), the model efficiently describes the nonlinear (state-dependent) spread dynamics over the last 2 decades, their relationship with macroeconomic fundamentals, and the latent sources of variability underlying regimes. The model's properties are evaluated among tested competitors according to Bayesian measures of fit, and select two independent sources of macro-financial structural instability. The latter separately affect the systematic determination of sovereign and bank credit yields, and the stochastic model component.

A preview of the results is as follows. First, the identified regimes have a story to tell, as they are consistent with abrupt historical events -financial, macroeconomic, and political- which characterize Italian macroeconomic developments. Changing institutional arrangements in the monetary policy, fiscal governance, and exchange rate regimes should all be conceived as contributing to the switches characterizing the structural dynamics of sovereign and bank spreads. Based on the coincidence of these events, we term the states in the residual variance variance states, those in the sovereign spread equation sovereign stress states, and those in the bank spread equation bank stress states. Descriptive statistics provide a first interpretation of regime switches, documenting the emergence of high-stress states when spreads are high and volatile. During states of high sovereign stress, sovereign spreads are positively correlated with fiscal stance, in turn displaying low correlation to output growth. Instead, states of high bank stress define periods of procyclical fiscal stance and negative correlation between bank spreads and fundamentals. The relationship between the two spreads is strong and positive during high bank stress states, and low and negligible during high sovereign stress. This suggests that the feedback loop is stronger from the sovereign to banks, rather than vice versa (Accornero et al., 2017).

A second set of results explains the transmission of identified structural shocks for sovereign and bank spread determination. State-dependent impulse responses show that a fiscal shock, with expansionary effects on output, induces substantial variation in the dynamics of sovereign and bank credit spreads during crises. This is consistent with higher sensitivity of yields to fundamentals, heightened amplification (Delatte et al., 2017), and higher fiscal multipliers (Auerbach and Gorodnichenko (2012); Canzoneri et al. (2015), among others), especially when during crises additional factors (related to policy uncertainty, self-fulfilling beliefs, etc.) come into play. Specifically, during high-stress states, sovereign spreads tend to increase in response to expansionary fiscal shocks, while bank spreads drop, given their tighter link to output realizations. This result is confirmed by simulating regime-specific responses to output shocks.

Finally, historical decompositions assess the role of model's structural shocks in conditioning spreads over time. During low-stress sovereign states, shocks to fundamentals contribute to sovereign spreads, while during high-stress states, the latter are mainly determined by idiosyncratic sources of variation. Instead, bank spread variability incorporates a substantial sovereign spread shock component, confirming the existence of a sovereign-bank pass-through, especially channeling from sovereign to banks. As for sovereign spreads, and especially after the global financial crisis, during low-stress states, shocks to fundamentals play a key role in determining bank spreads.

Our study contributes to two strands of the literature. First, by exploiting the properties of a MS-VAR, we complement earlier research on the role of fundamentals in sovereign risk pricing, considering the state-dependent sensitivity of both sovereign and bank spreads to fundamentals. Costantini et al. (2014), Aizenman et al. (2013), and Afonso et al. (2014) demonstrate the existence of an extra premium on fiscal imbalances for Italy, Spain, and Portugal during euro crises, while Bocola and Dovis (2019) decompose Italian sovereign spreads to quantify the fundamental and nonfundamental components of the Economic and Monetary Union (EMU) crisis. Second, this study contributes to the literature on the links between sovereign and bank distress, highlighting the regime-dependent nature of the risk loop, where a sizable sovereign-bank transmission channel during crises alternates with dampened pass-through in low-stress states. Empirical evidence is provided by Gennaioli et al. (2018) and Keddad and Schalck (2020), among others, using cross-country, bank-level panel data. The theoretical literature rationalizes different transmission mechanisms: Bocola (2016) investigates the impact of Italian sovereign risk on bank balance sheets, credit provision, and output losses; Faia (2017) identifies balance sheet, collateral, and liquidity channels; Acharya et al. (2014), Cooper and Nikolov (2018), and Farhi and Tirole (2018) analyze the pass-through through the lens of bank bailout incentives and costs.

The remainder of the paper is organized as follows. Section 2 presents some stylized facts about sovereign and bank risk. Here, inherent data features demonstrate strong nonlinearities characterizing the determination of spreads, thus motivating a Markov-switching model for the analysis. Section 3 describes the MS-VAR empirical model, data used, identification strategy, and model selection. The results are discussed in Section 4. Section 5 concludes the paper. A technical appendix, including data description, estimation details, additional results, and robustness exercises, follows.

Section snippets

Stylized facts about sovereign and bank risk

To provide an intuition on the mechanisms under study, Fig. 1 contextualizes the events featuring Italy's recent historical experience, reporting the 1990–2019 evolution of sovereign and bank spreads, along with that of domestic macroeconomic fundamentals, summarized by GDP growth, the general government primary deficit over one-month lagged debt, and -for comparison- the primary deficit-to-GDP ratio.3

Empirical model

Our main objective is to shed light on the relevance of fundamentals for Italian sovereign and bank risk dynamics over the past 3 decades. This section illustrates the empirical methodology, the data, and the identification strategy we adopt to document how the transmission mechanics operate. The selection of the best regime-switching structure delivers the closing details for model estimation and simulation.

Results

In this section, we discuss the main results derived from the estimation and simulation of our benchmark model. We first extract and interpret regimes to identify the historical events characterizing the Italian economy over the last 30 years. The state-dependent dynamic properties of the model are then evaluated by simulating impulse response functions (IRFs) to a fiscal shock. Historical and forecast error variance decompositions then quantify the relative shock contribution to the historical

Conclusions

Fiscal and financial factors jointly contributed to several crisis episodes in the Italian history (i.e, the EMS crisis, global financial crisis, and sovereign debt crisis). Sovereign and bank credit risk indicators often spiked together. Although this evidence is well established in the literature, identifying the actual sources of risk and their transmission remain a key issue. How much of the recognized risk surge comes from factors related to macroeconomic fundamentals? How can the emerging

Declaration of competing interest

The authors whose names are listed immediately below certify that they have NO affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in

Acknowledgments

We gratefully acknowledge financial support from the Sapienza Research grant n. 1529/2019.

References (70)

  • K. Hubrich et al.

    Financial stress and economic dynamics: the transmission of crises

    J. Monetary Econ.

    (2015)
  • B. Keddad et al.

    Evaluating sovereign risk spillovers on domestic banks during the European debt crisis

    Econ. Modell.

    (2020)
  • C.J. Kim

    Dynamic linear models with Markov-switching

    J. Econom.

    (1994)
  • V. Patella et al.

    Confidence swings and sovereign risk dynamics

    Struct. Change Econ. Dynam.

    (2021)
  • C.A. Sims

    Stepping on a rake: the role of fiscal policy in the inflation of the 1970s

    Eur. Econ. Rev.

    (2011)
  • C.A. Sims et al.

    Methods for inference in large multiple-equation markov-switching models

    J. Econom.

    (2008)
  • T. Theobald et al.

    Euro area sovereign yield spreads as determinants of private sector borrowing costs

    Econ. Modell.

    (2020)
  • M. Accornero et al.

    Non-Performing Loans and the Supply of Bank Credit: Evidence from Italy

    (2017)
  • V. Acharya et al.

    A pyrrhic victory? Bank bailouts and sovereign credit risk

    J. Finance

    (2014)
  • A. Afonso et al.

    Pricing sovereign bond risk in the EMU area: an empirical investigation

    Int. J. Finance Econ.

    (2014)
  • C. Altavilla et al.

    The financial and macroeconomic effects of the OMT announcements

    Int. J. Cent. Bank.

    (2016)
  • Andrade, P., J. Breckenfelder, F. de Fiore, and O. Tristani. 2016. “The reanchoring channel of QE: the ECB asset...
  • A.J. Auerbach et al.

    Measuring the output responses to fiscal policy

    Am. Econ. J. Econ. Pol.

    (2012)
  • F. Balassone et al.

    Public Debt and Economic Growth: Italy's First 150 Years

    (2013)
  • E. Beqiraj et al.

    Fiscal retrenchments and the transmission mechanism of the sovereign risk channel for highly indebted countries

    N. Am. J. Econ. Finance

    (2021)
  • B.S. Bernanke et al.

    Measuring the effects of monetary policy: a factor-augmented vector autoregressive (FAVAR) approach

    Q. J. Econ.

    (2005)
  • F. Bianchi et al.

    Escaping the great recession

    Am. Econ. Rev.

    (2017)
  • J. Bianchi

    Over-borrowing and systemic externalities in the business cycle

    Am. Econ. Rev.

    (2011)
  • O. Blanchard et al.

    An empirical characterization of the dynamic effects of changes in government spending and taxes on output

    Q. J. Econ.

    (2002)
  • L. Bocola

    The pass-through of sovereign risk

    J. Polit. Econ.

    (2016)
  • L. Bocola et al.

    Self-fulfilling debt crises: a quantitative analysis

    Am. Econ. Rev.

    (2019)
  • BoI

    Bank of Italy

    (2018)
  • D. Caldara et al.

    The analytics of SVAR: a unified framework to measure fiscal multipliers

    Rev. Econ. Stud.

    (2017)
  • G.A. Calvo

    Servicing the public debt: the role of expectations

    Am. Econ. Rev.

    (1988)
  • M. Canzoneri et al.

    Fiscal multipliers in recessions

    Econ. J.

    (2015)
  • Cited by (0)

    We thank the Editor and the anonymous reviewers for their thoughtful comments and suggestions. We are also grateful to António Afonso, Jaromír Baxa, Michael Donadelli, Ester Faia and Gert Peersman for their useful discussions. We thank the participants at various conferences and seminars.

    View full text