Debt management when monetary and fiscal policies clash: Some empirical evidence

Hodula, Martin; Melecký, Aleš Article Debt management when monetary and fiscal policies clash: Some empirical evidence Journal of Applied Economics Provided in Cooperation with: University of CEMA, Buenos Aires Suggested Citation: Hodula, Martin; Melecký, Aleš (2020) : Debt management when monetary and fiscal policies clash: Some empirical evidence, Journal of Applied Economics, ISSN 1667-6726, Taylor & Francis, Abingdon, Vol. 23, Iss. 1, pp. 253-280, https://doi.org/10.1080/15140326.2020.1750120 This Version is available at: https://hdl.handle.net/10419/314091 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Journal of Applied Economics ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/recs20 Debt management when monetary and fiscal policies clash: some empirical evidence Martin Hodula & Aleš Melecký To cite this article: Martin Hodula & Aleš Melecký (2020) Debt management when monetary and fiscal policies clash: some empirical evidence, Journal of Applied Economics, 23:1, 253-280, DOI: 10.1080/15140326.2020.1750120 To link to this article: https://doi.org/10.1080/15140326.2020.1750120 © 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. Published online: 16 Apr 2020. Submit your article to this journal Article views: 7119 View related articles View Crossmark data Citing articles: 8 View citing articles Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=recs20 ARTICLE Debt management when monetary and fiscal policies clash: some empirical evidence Martin Hodula and AlešMelecký Department of Economics, Technical University of Ostrava, Ostrava, Czech Republic ABSTRACT We explore the effects of fiscal and monetary policy shocks on key debt management variables and provide empirical evidence supporting the notion of a strict separation of economic policy from the debt management agenda. We find that a tighter monetary policy coupled with fiscal expansion increases the risk that government debt will have to be rolled over at unusually high cost. This is especially the case in a downturn, where low or even negative interest rates often provide incentives for debt managers to invest predominantly in short-term bonds. Our findings echo the postcrisis environment of low or even negative interest rates, where many debt managers altered their portfolios’structure in favor of short-term bonds. In this respect, we argue that debt managers should use a longer optimization horizon and base their strategy on the mediumand long-term economic outlook. ARTICLE HISTORY Received 3 May 2019 Accepted 27 March 2020 KEYWORDS Czech Republic; debt management; monetary policy; fiscal policy; FAVAR 1. Introduction Historically, debt management was not a stand-alone policy, but was considered a part of fiscal or monetary policy. In 2001, the IMF and World Bank published a set of guidelines on public debt management for policymakers, which were later revised in response to financial sector regulatory changes and macroeconomic policy developments (IMF and WB, 2014). These guidelines stress the importance of formulating a sound debt management strategy for the optimal allocation of government debt and the need to separate debt management from other policies. In this paper, we assess how fiscal and monetary policy measures may influence public indebtedness, debt service costs, and sovereign default risk in a small open economy. We show that increasing government spending coupled with rise of the monetary policy rate (for instance, during an economic boom) could increase the risk that the government debt will have to be rolled over at unusually high cost if the economy slows down in the future. Our findings may have some major policy implications given the policy actions following the global financial crisis in 2008. After the crisis, debt managers took advantage of an all-time low nominal interest rate environment and altered the government debt portfolio structure in favor of short-term bonds. 1 However, such an investment CONTACT Martin Hodula [email protected],Department of Economics, Technical University of Ostrava, 15/2172 17. Listopadu St., Ostrava –Poruba 708 33, Czech Republic 1 Wheeler (2004) shows that short-term bonds are usually associated with higher market risk. JOURNAL OF APPLIED ECONOMICS 2020, VOL. 23, NO. 1, 253–280 https://doi.org/10.1080/15140326.2020.1750120 © 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/ licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. choice may increase the debt service costs and endanger debt sustainability once the monetary policy stance returns to neutral levels. 2 In this respect, we argue that debt managers should use a longer optimization horizon and base their strategy on the mediumand long-term economic outlook. 3 Moreover, our modelling approach may be of some appeal to debt managers as it allows taking into consideration a broad set of variables representing the macroeconomic environment. Quantification of the potential impacts of changes in macroeconomic policy on key debt variables may help debt managers better anticipate changes in the debt portfolio and debt service costs, allowing them to take timely steps to optimize the portfolio and achieve the primary debt management objective. For example, in the Czech Republic the goal is to cover the borrowing requirements and payment obligations of the government while reaching the lowest possible service cost in the mediumto long-term horizons at a specific level of accepted risk. Therefore, our model may be considered as a complement to the Asset Liability Management (ALM) approach applied by debt managers to achieve optimal debt composition. We rely mostly on Czech data; the Czech economy is an interesting example because of a long-standing environment of low interest rates on government bonds and recent experience with an exchange rate commitment, which was in place from 2013 to 2017. In the Czech Republic, a department for the management of government debt and financial assets was established in 2005 under the Ministry of Finance, and the Czech National Bank (CNB) now acts as the market supervisor. The primary objective of Czech monetary policy is price stability (inflation targeting). The CNB uses mostly the two-week repo rate to keep inflation close to the target. The remainder of the paper is organized as follows: Section 2 lists the existing studies and discusses our contribution to the literature. Section 3 provides stylized facts about the development of government debt, its servicing, and the macroeconomic environment in the Czech Republic; Section 4 outlines the theoretical underpinnings of the applied framework and describes the data employed; Section 5 discusses the empirical results, and Section 6 concludes. 2. Literature review Our research fits into the strand of literature focusing on policy coordination and potential conflicts among fiscal, monetary, and debt management policies, a topic that has been studied in various frameworks. Generally, a conflict between monetary policy and debt management may arise due to a shortage of independent policy instruments (Togo, 2007). In addition, procyclical fiscal policy, expansionary in booms and contractionary in recessions, may in the long run increase macroeconomic volatility and uncertainty, harm economic growth, and raise debt service costs. Therefore, we examine the effects of monetary and fiscal policy on variables relevant to the debt manager. Starting with the classic studies, Barro (1995)finds that debt management could be helpful in tax smoothing. Calvo and Guidotti (1990) point out the role of debt 2 Faraglia et al. (2008) state that policymakers are usually more concerned with minimizing costs instead of risk. Angeletos (2002) claims that public debt management can be an important tool for reducing fiscal vulnerability. 3 The Medium-Term Debt Management Strategy (MTDS) framework assessment can be found in IMF (2017). 254 M. HODULA AND A. MELECKÝ management as a commitment device in ensuring a time-consistent monetary policy. Among more recent studies, Canzoneri, Cumby, and Diba (2016) highlight the need for the debt manager to satisfy liquidity demand and accommodate a smooth tax rate. However, they add that if government bonds provide liquidity, conflicts may arise. Bianchi and Melosi (2019) claims that monetary and fiscal policies are not completely independent and there is a need for their coordination. Some public debt management choices and large fiscal deficits endanger inflation management and the interest rate policy, and even the independence of the central bank. The monetary policy setting may therefore influence the cost of deficit financing and even the size of the public debt. Moore and Skeete (2010)find that negative monetary policy shocks could significantly raise the debt service costs and the future setting of economic policy needs to be coordinated. Cavalcanti, Vereda, de Doctors, and Lima (2018) call for higher policy coordination and find high correlation between the monetary policy rate and the interest rate on public debt in Brazil. Togo (2007) illustrates the importance of policy separation and coordination to achieve a consistent policy mix. According to his study, separation does not preclude the need for policy coordination because of frequent policy interactions in the real world. Poor fiscal policy may produce high deficits that need to be financed by new debt. Excessive debt levels will lead to high risk premia demanded by investors and will limit the debt manager in issuing the debt instruments needed to achieve optimal debt composition. 4 Monetary policy may constrain debt management through exchange rate and interest rate policy, as these policies directly influence the amount of foreign currency and floating rate debt that can be issued. Togo argues that separation of debt management policy may reduce trade-offs among these three policy objectives and enhance the credibility and effectiveness of policy implementation. The presented ALM framework suggests that the debt manager should act counter-cyclically to help minimize the risk of tax increases, expenditure cuts, or debt defaults. Based on policy games under fiscal and monetary dominance, Togo concludes that weak debt management without a separate policy goal could produce an inconsistent policy mix. Moreover, as Sight (2015) states, the coordination of fiscal, monetary, and debt management policies is even more important in developing countries. To broaden the scope, Dottori and Manna (2016) study strategy and tactics in debt management and argue in favor of a broader perspective of coordination that also includes financial stability. According to Das, Papapioannou, Pedras, Ahmed, and Surti (2010), debt management influences financial stability through five channels –stock of debt, debt profile, investor base, debt market structure, and institutional aspects. We contribute to this strand of literature by studying the policy interactions in a data-rich environment while taking into consideration alternative debt manager goals. In the Czech Republic, numerous studies focus on fiscal discipline and debt sustainability, see e.g., Ambrisko et al. (2012), Bulir (2004), European Commision (2012), IMF (2013), Komarkova, Dingova, and Komarek (2013), or Dybczak and Melecky (2014). Yet there is a shortage of studies focusing directly on government debt management. The 4 Faraglia, Marcet, and Scott (2010) in their search for a theory of debt management criticize the idea of complete market approach to debt management. Specifically, that the fiscal policy and debt structure should be jointly determined. According to them, this theory focusses only on fiscal insurance and abstracts from fundamental features of market incompleteness. JOURNAL OF APPLIED ECONOMICS 255 Czech Ministry of Finance regularly publishes progress reports on debt management, 5 including results of the cost-at-risk model. Pavelek (2005) contributes to the debate on advances in risk management of government debt with the Czech experience and suggests a more proactive approach. Matalik and Slavik (2005) concentrate on government debt management in the Czech Republic and its development during the transition period (from the centrally-planned economy of the country’s communist past) of the 1990s and early 2000s. They stress the need for domestic capital market development to better manage the foreign exchange risk of government debt. We may summarize the findings from the literature review as follows: Individual policies should be politically separated in terms of their objectives and accountability, in line with the Tinbergen principle, to provide an optimal policy mix. Nevertheless, information sharing and cooperation should be further promoted, including sharing forecast data and developing strategies in coordination with other policies. The government debt management should not be neglected in this process. Our paper aims to address the gap in the literature on government debt management in the Czech Republic, help inform Czech debt management policy, and provide an example of the use of a theoretical framework for the practical allocation of government debt. 3. Stylized facts To achieve inflation targets in the post-crisis period, central banks created an environment of low or even negative interest rates and applied various unconventional monetary policy tools, such as quantitative easing, which influences government bond yield curves (Corradin & Maddaloni, 2017; Ferdinandusse, Freier, & Ristiniemi, 2017; Schlepper, Riordan, Hofer, & Schrimpf, 2017). In 2013, the CNB adopted an exchange rate commitment to intervene in the foreign exchange market. 6 The central bank had weakened the Czech koruna (CZK) exchange rate to close to 27 CZK/EUR (from pre-intervention levels around 25.50 CZK/EUR) in order to avoid deflation. Such actions influence the cost of unhedged foreign-currency-denominated debt. Melecky (2012) provides a review of policy approaches to the choice of currency structure of foreign currency debt. A change in inflation dynamics may also affect the cost of the debt portfolio and government debt through a change in short-term interest rates. In the Czech Republic, inflation-linked bonds are issued only within a retail program and make up a negligible part of the government debt portfolio. However, inflation-linked bonds may be at least partially substituted by bills indexed to the short-term interest rate, as is done in the Czech Republic. Debt management performance should be regularly evaluated, as is done by the Czech Ministry of Finance in annual reports. First of all, prediction of interest expenditure on debt service should be provided and relevant risks should be evaluated, including refinancing, interest rate, and currency risk. For instance, an excessive proportion of short-term floating rate debt could significantly increase refinancing costs or even make 5 Regular publications by the Czech Ministry of Finance regarding debt management include Funding and Debt Management Strategy, Government Debt Management Annual Report, Debt Portfolio Management Quarterly Report, and Performance Evaluation of Primary Dealers; for details please see http://www.mfcr.cz/en/themes/state-debt/pub lications-and-presentations. 6 For more details on the exchange rate commitment, please refer to Bruha and Tonner (2018). 256 M. HODULA AND A. MELECKÝ refinancing impossible in extreme cases of weak debt management. This happens especially in emerging or highly indebted countries. An excessive proportion of unhedged foreign currency bonds in the government debt portfolio may significantly influence debt service costs during currency depreciation. However, debt management should not be evaluated based solely on economic criteria. For the purpose of comprehensive evaluation of debt management, the World Bank in cooperation with other relevant institutions developed the Debt Management Performance Assessment (DeMPA) methodology. 7 Figure 1 shows the development of government debt and debt interest payments in the Czech Republic. During the 2000–2012 period, Czech government debt exhibited an obvious upward trend, accompanied by rising debt interest payments. After 2012, government debt growth was significantly reduced due to a concomitant rise of the economy and wealth of economic agents. Debt interest payments peaked in 2014 and thereafter declined due to the “escape from risk”effect, wherein investors buy bonds of relatively “safe” countries with stable currency, such as the Czech Republic was at the time. This effect was intensified by the existence of CNB’s one-sided foreign exchange commitment and investor speculation on future appreciation of the koruna. This environment led to negative bond yields. The Czech Ministry of Finance, which controls the majority of government debt as the debt manager, used the environment of negative yields on mediumand longterm government bonds to, in December 2015, auction offgovernment bonds maturing in 2017 at an all-time low yield of −0.35% p.a. The total public budget revenue from this investment activity, lending facilities with government bonds, and from the negative yields Figure 1. Government debt, deficit, and interest payments (CZK billion). Note: MA(4) denotes a four-quarter moving average; rhs = right-hand scale. 7 Although the primary focus of this tool is on developing countries, it can be applied to others as well. The methodology evaluates debt management according to fourteen performance indicators that can be summarized into five key areas (Governance and Strategy Development, Coordination with Macroeconomic Policies, Borrowing and Related Financing Activities, Cash Flow Forecasting and Cash Balance Management, and Debt Recording and Operational Risk Management). For more details, see Panzer et al. (2015). JOURNAL OF APPLIED ECONOMICS 257 of government bonds amounted to CZK 524.9 million (approximately EUR 19.4 million) in 2015 (Ministry of Finance, 2017) and a similar trend continued the following year. However, operations of this type significantly reduced the average maturity of government debt, down to about 5 years (Figure 2, panel A). The largest changes occurred in the segment with residual maturity of up to 3 years. Between 2010 and 2016, the shares in the debt portfolio changed as follows: T-bills dropped from 8.9% to 0.3%, the share of government bonds with residual time to maturity of up to 1 year increased from 8% to 14.4%, and bonds with residual time to maturity between 1 and 3 years increased from 17% to 29.6% (Figure 2, panel B). The value of the share of net foreign-currency exposure to government debt with impact on the level of interest expenditure on government debt reached 11.5% at the end of 2016 and remained under the strategic limit of the Czech Ministry of Finance (15% +2 p.p.). The net foreign-currency exposure of government debt with impact on government debt service was denominated solely in EUR at the end of 2016. Further details about the debt composition are available in Ministry of Finance (2017). Several empirical studies highlight the importance of strategic interaction between monetary and fiscal policies (Franta, Libich, & Stehlik, 2018; Hodula & Pfeifer, 2018; Kirsanova & le Roux, 2013). 8 Moreover, the relationship between monetary and fiscal policy is expected to grow over time because of increased budget financing pressures resulting from population ageing (see also Komarkova et al., 2013; Ambrisko et al., 2017). This could be a serious issue for the Czech Republic, as government spending on pensions jumped form CZK 222 bn. in 2008 to CZK 315 bn. in 2015, i.e., an increase of more than 40% over an 8-year period. Figure 2. Czech government bonds. Note: The data are taken from the Czech Republic government debt management annual report 2018. 8 The theoretical examination of monetary-fiscal interactions dates back to the seminal paper of Sargent and Wallace (1981). 258 M. HODULA AND A. MELECKÝ 4. Modelling approach and data To find out how changes in the economic policy setting may influence the debt management-related set of variables, we use the factor-augmented vector autoregression model (FAVAR) introduced in Bernanke, Boivin, and Eliasz (2005). The FAVAR model has several appealing properties that make it an ideal candidate for our empirical exercise. First, it allows us to avoid information bias when identifying the set of economic policy innovations. For example, when trying to identify a monetary policy shock in a VAR model, the shock may actually not be truly exogenous, as it may also capture instances when the central bank endogenously reacts to changing inflation expectations. 9 This gives rises to the infamous “price puzzle”. In contrast to the basic VAR model, the FAVAR model includes unobserved lowdimensional factors in the autoregression, reducing the information bias. The FAVAR model uses the advantages of a data-rich environment while remaining tractable in terms of the number of parameters to be estimated. The restriction of a small number of latent factors is consistent with standard dynamic equilibrium macroeconomic theories (Stock & Watson, 2016). To obtain the factors, we follow Bernanke et al. (2005) who employed the Stock and Watson (1998,1999,2002)) dynamic factor model, which is estimated using the static principal components approach. Here, it is crucial to distinguish between the static and dynamic representation of a dynamic factor model in order to avoid confusion. The static approach relies on the time-domain forecasting method of Stock and Watson (1998,1999,2002)). The estimates are based on contemporaneous covariances only and, as such, do not exploit the potential information contained in the leading-lagging relations between the variables. However, as the authors show, the space of factors is still consistently estimated by the static approach, provided the number of variables and the time dimension are large. The FAVAR modelling framework is used in many economic applications. For instance, Forni and Gambetti (2010) use a FAVAR model to study the effects of monetary policy in the US. Similarly, Eickmeier and Hofmann (2013) apply a FAVAR model to US data with the aim of analyzing monetary transmission via private sector balance sheets, credit risk spreads, and house prices and of exploring the role of monetary policy in the housing and credit boom prior to the global financial crisis. Hodula and Pfeifer (2018) extend the FAVAR framework to study strategic policy interactions between fiscal and monetary policy and financial stability. 4.1. The FAVAR model specification We specify an M1vector of macroeconomic time series Ytand a K1 vector of unobserved factors Ft. We assume that the joint dynamics of Yt;Ft ðÞare given by the following equation: Ft Yt  ¼ΦLðÞFt1 Yt1  þεt(1) 9 The use of many variables to span the space of the shocks mitigates the “invertibility problem”of structural vector autoregressions (SVARs). For more detailed discussions, see Forni, Giannone, Lippi, and Reichlin (2009) and Leeper, Walker, and Yang (2013). JOURNAL OF APPLIED ECONOMICS 259 response of the real effective exchange rate is rather ambiguous, with relatively wide confidence intervals. 5.2. Monetary restriction, fiscal expansion, and debt management Concerning responses of variables linked to debt management (Figure 4), the contractionary monetary policy shock produces an immediate increase of government bond yields, which further affects the yield curve of the government debt portfolio. Due to increase in government bond yields, debt service costs increase, peaking at 2 quarters after the shock. Higher debt service costs are also reflected in increased debt interest payments. Increased cost of financing, together with declining GDP growth, means higher government borrowing needs and greater indebtedness as measured by the debtto-GDP ratio. The aforementioned is in line with findings stemming from derivation of fiscal debt dynamics. The effect of tighter monetary conditions is stronger for higher initial government debt-to-GDP ratios and lower GDP dynamics, which are negatively affected by tighter monetary conditions. Note that the exchange rate appreciation following a contractionary monetary policy shock may suggest that some foreign-currency debt can be issued as a hedge against interest-rate risk affecting short-term debt. As expected, the IRFs projected in Figure 4 well document the faster and slightly higher reaction of 2Y bond yields (0.65 pp at impact) compared to yields of 5Y (0.43 pp at impact) and 10Y bonds (0.33 pp at impact). A higher share of 2Y bonds in the Figure 4. Monetary restriction, fiscal expansion, and debt management. Note: Median impulse responses of the FAVAR model with 2 lags are reported with 90% confidence bounds; responses are in percentage points; the x-axis is in quarters after the shock. 266 M. HODULA AND A. MELECKÝ government debt portfolio could increase its riskiness due to undesirable development of debt service costs. This finding supports the idea that short-term bonds are riskier than long-term bonds, especially because of higher interest rate and refinancing risks. This fact should be considered in the preparation of the government’s debt management strategy. Indeed, debt managers in many countries took advantage of the negative interest rate environment 13 to extend the average maturity of their underlying debt. For instance, Switzerland and the United Kingdom dramatically increased the average maturity of their debt, and a similar trend is seen in many other countries (see Figure 1C in the Appendix). However, the Czech Republic and a few other countries (Hungary and Sweden) adopted a different strategy and changed the structure of the debt portfolio in favor of short-term bonds, resulting in the Czech Republic in a drastic drop in average maturity from 6.4 years in 2007 to 5.1 years in 2016, 14 far below the international average. The only countries with lower average maturity in 2016 were Hungary and Norway. 15 The relatively small savings from the issuance of short-term bonds may be canceled out (or even overcome) by higher debt service costs in the long term due to accumulating debt issues and increased refinancing and interest rate risks. Still, the Czech Republic enjoys good credit standing and debt management remains prudent. Turing to fiscal expansion, simulated through an increase in total government expenditures, we document three effects. First, we identify a partial crowding-out effect, i.e., the increase in government spending is not completely canceled out by the decline of private expenditures and leads to GDP growth. Second, faster GDP growth decreases relative indebtedness as measured by the debt-to-GDP ratio. Nevertheless, a hike in government spending generates deficit, which needs to be financed by issuance of new bonds. Accordingly, we document an increase in government debt, both in absolute and relative terms, accompanied by rising debt interest payments and debt service costs. Unlike the first two effects, the third appears with a lag, probably because the market needs some time to appreciate the higher risk and reflect it in higher rates. Increasing both government expenditures and debt increases the risk associated with debt sustainability and leads to a higher sovereign default risk, which is reflected in higher bond yields. 5.3. Alternative objectives of the debt manager One must also bear in mind that the debt manager may follow secondary (alternative) goals. Debt management may support government investment plans focused on large infrastructure projects, construction of nuclear power plants, and other government priorities that requires large-scale and long-running investments. This is especially true if debt management is not completely independent (for instance, falls under the Treasury). Several studies also refer to tax smoothing as one of the government goals that may affect the debt management (see Canzoneri et al., 2016; Faraglia, Marcet, & Scott, 2008). The government may even directly cooperate with the debt manager in order to spread out the tax burden (which may be associated with tax reforms) into a longer time period. In addition, in countries with underdeveloped financial markets, 13 Figure 2C in the Appendix shows the average government bond yields in the Eurozone. 14 This drop in average maturity has continued in 2017. 15 Still, Norway is in a different position due to massive income from oil-related wealth funds. JOURNAL OF APPLIED ECONOMICS 267 the debt manager may stimulate the supply of investment products (especially government bonds) to deepen the financial market. To account for these alternative debt management goals, we consider a positive shock to the debt-to-GDP ratio and track the impulse responses of gross capital formation, tax burden (defined as tax revenues over GDP), and financial market depth index. The responses are depicted in Figure 5. First, following the immediate increase in the debt-to-GDP ratio, we document a gradual growth of gross capital formation. Judging by that, the simulated increase in indebtedness may bring along some positive long-run effects for the economy. Growth in the debt-to-GDP ratio may therefore be the result of government investment plans, which are beyond the decision-making power of the debt manager, but which it would have to take into account when deciding on the optimal debt strategy. Second, we focus on the reaction of tax burden to the debt shock. In terms of tax smoothing, we may conclude that government debt seems to be a suitable instrument for distributing the tax burden across generations. An increase in indebtedness did not lead to significant changes in taxation in the Czech Republic. Third, we consider the response of financial market depth, measured by the financial market depth index taken from the IMF. 16 Given the fact that one of the secondary goals of government debt management in the Czech Republic is to support the development of financial markets, a positive reaction of the financial market depth index shortly after the debt shock shows that it is possible to develop financial markets by issuing government bills and bonds. We conduct a number of robustness checks to verify our results. First, we test whether our results are affected by changing the number of factors that enter the FAVAR model. In particular, we try estimating the model with 5 and 7 factors. Most of the estimated median IRFs from the models with 5 and 7 factors lie inside the 90% confidence interval of the FAVAR model with 3 factors. The only difference is a slightly slower reaction of the variables to the monetary policy shock for models with a larger number of factors. Second, we use an alternative identification strategy for fiscal and monetary policy shocks. Results are robust to use of other plausible orderings of the variables in the Figure 5. Responses to a positive shock to debt-to-GDP ratio. Note: Median impulse responses of the FAVAR model with 2 lags are reported with 90% confidence bounds; responses are in percentage points; the x-axis is in quarters after the shock. 16 For details regarding the estimation procedure of the financial development index, please refer to Svirydzenka (2016). 268 M. HODULA AND A. MELECKÝ FAVARs. None of the above has a significant impact on our results. The associated impulse response functions are available in Appendix D, Figures 1D,2D,3D and 4D.We also compare the estimates from a large FAVAR model with those from small-scale VAR models to show the computational gains from modelling the system in a data-rich environment (see Appendix B for the estimates). Further robustness checks include an increased number of lags in the FAVAR model (up to 4), which yields results that are both qualitatively and quantitatively similar. 6. Conclusion The monetary and fiscal policies are inherently intertwined. This statement is generally accepted and holds even for countries where the central bank has a long history of independence from the government. In this paper, we take advantage of the mutually-reinforcing interlinkages of these economic policies and study the impact of fiscal and monetary policy shocks on debt management in the Czech Republic. We use a data-rich FAVAR model to study the responses of key debt management variables to economic shocks. The FAVAR model helps ensure that the estimated impulse responses are invariant with respect to extensions of the information set –an issue that often plagues impulse response results. The Czech Republic is an interesting case for a study as the national debt manager took advantage of the low-interest environment between 2010 and 2017 and altered the debt portfolio structure in favor of short-term bonds. With the upcoming normalization of nominal interest rates and continuous economic growth that allow fiscal policy to expand, we provide some key policy findings regarding the relationship between the main policy and debt management variables. We report several findings: (i) We find that a tightening of monetary policy alters the yield curve of government bonds, causing rapid yield increase;(ii)followingthis,debtservicecosts increase and, coupled with decreasing economic activity, lead to an increase in the overall public debt burden. Turning to the effects of fiscal policy expansion, (iii) it comes as no surprise that a hike in government expenditures increases the level of government debt; (iv) nevertheless, the effects might be even more pronounced in the Czech Republic, as we also find evidence for a partial crowding-out effect. In this respect, we identify an extremely unfavorable policy mix (monetary contraction with fiscal expansion) that might significantly increase debt service costs and, in the long-term, may endanger government debt sustainability. Furthermore, we show that increasing government indebtedness could cater to some of the alternative (parallel) goals of debt management, such as deepening the country’sfinancial market and supporting government investment plans. Our analysis suggests that changes in both monetary and fiscal policy shocks significantly affect debt management policy, but the time evolution of the studied impulse responses is different for each. This calls for a forward-looking debt management strategy, i.e., a longer optimization horizon. The debt manager should incorporate possible changes in current economic policy into its decisions about the debt portfolio structure and reflect the mediumand long-term outlook in its strategy. In doing so, the debt manager should intensively coordinate its actions with the central bank, e.g., through sharing economic forecasts of the likely future trajectory of interest rates. Any increase in the monetary policy rate or growth of government expenditures may quickly transmit to the service costs and size of government debt, especially when the debt manager favors short-term bills over long-term government bonds in the government JOURNAL OF APPLIED ECONOMICS 269 debt portfolio. Looking at recent data on the average term-to-maturity of outstanding government debt across multiple countries (Figure 2C), there are certain other countries (Hungary, Sweden) that took the same approach of lowering their average maturities; this has left them more susceptible to monetary and fiscal policy-like shocks in the long term. Our data might thus be of some value to these countries as well. Our results largely support the separation of the individual economic policies. They need to be separated politically in terms of their objectives and accountability, in line with the Tinbergen principle. This holds especially for the fiscal and debt management policies. However, the policymakers involved need to share information to make informed long-term decisions and coordinate their actions. Acknowledgments We would like to thank Jan Libich, LukášPfeifer, Zdeněk Pikhart and Gulcin Ozkan for their helpful comments. We also thank participants of the 8 th conference on “Managing and Modelling of Financial Risks”for their comments. We gratefully acknowledge financial support from the Czech Science Foundation (GA16-22540S) and the SGS project at VSB-TUO (SP2020/110). Disclosure statement No potential conflict of interest was reported by the authors. Funding This work was supported by the Czech Science Foundation Agency [GA16-22540S]; Vysoká Škola Bánská - Technická Univerzita Ostrava [SP2020/110]. Notes on contributors Martin Hodula is an Assistant Professor at the VSB - Technical University of Ostrava. His research interest include macroprudential policy, banking sector regulation and non-bank financial intermediation (shadow banking). Martin holds a PhD degree in Economics from the Technical University of Ostrava. AlešMelecký is an Associate Professor at the VSB - Technical University of Ostrava. His research interest include macroeconomic modelling, macroprudential policy, credit risk, non-bank financial intermediation (shadow banking), and government debt management. Ales holds a PhD degree in Economics from the VSB - Technical University of Ostrava. ORCID AlešMelecký http://orcid.org/0000-0002-4495-6296 References Alesina, A., Campante, F. R., & Tabellini, G. (2008). Why is fiscal policy often procyclical? Journal of the European Economic Association,6(5), 1006–1036. 270 M. HODULA AND A. MELECKÝ Ambrisko, R., Augusta, V., Hajkova, D., Kral, P., Netusilova, P., Rikovsky, M., & Soukup, P. (2012). Fiscal discretion in the Czech Republic in 2001–2011: Has it been stabilizing? (Czech National Bank Research and Policy Notes 2012/1). Ambrisko, R., Babecky, J., Rysanek, J., & Valenta, V. (2015). Assessing the impact of fiscal measures on the Czech economy. Economic Modelling,55(1), 350−357. Ambrisko,R.,Dingova,V.,Dvorak,M.,Hajkova,D.,Hromadkova,E.,Kulhava,K.,&Stikova,R.(2017). Assessing the fiscal sustainability of the Czech Republic (Czech National Bank Research and Policy Notes 2017/2). Angeletos, G. (2002). Fiscal policy with non-contingent debt and the optimal maturity structure. Quarterly Journal of Econmics,117(3), 1105–1131. Bai, J., & Ng, S. (2002). Determining the number of factors in approximate factor models. Econometrica,70(1), 191−221. Barro, R. J. (1995). Optimal debt management (NBER Working Paper 5327). Bernanke,S.B.,Boivin,J.,&Eliasz,P.(2005). Measuring the effects of monetary policy: A Factor-Augmented Vector Autoregressive (FAVAR) approach. Quarterly Journal of Economics,120(1), 387−422. Bianchi, F., & Melosi, L. (2019). The dire effects of the lack of monetary and fiscal coordination. Journal of Monetary Economics,104(C), 1–22. Blanchard, O., & Perotti, R. (2002). An empirical characterization of the dynamic effects of changes in government spending and taxes on output. The Quarterly Journal of Economics,117(4), 1329−1368. Borys, M., Franta, M., & Horváth, R. (2009). The effects of monetary policy in the Czech Republic: An empirical study. Empirica,36(1), 419−443. Bruha, J., & Tonner, J. (2018). An exchange rate floor as an instrument of monetary policy: An ex-post assessment of the Czech experience. Czech Journal of Economics and Finance,68(6), 537−549. Bulir, A. (2004). External and fiscal sustainability of the Czech economy: A quick look through the IMF’s night-vision goggles (Czech National Bank Internal Research and Policy Note 4/2004). Calvo, G. A., & Guidotti, P. (1990). Indexation and maturity of government bonds: An exploratory model. In R. Dornbusch & M. Draghi (Eds.), Capital markets and debt management. Cambridge University Press, 52−93. Canzoneri, M., Cumby, E., & Diba, B. (2016). Optimal money and debt management: Liquidity provision vs tax smoothing. Journal of Monetary Economics,83,39−53. Cavalcanti, M. A. F. H., Vereda, L., de Doctors, R. B., & Lima, F. C. (2018). The macroeconomic effects of monetary policy shocks under fiscal rules constrained by public debt sustainability de B.. Economic Modelling,71(2018), 184–201. Corradin, S., & Maddaloni, A. (2017). The importance of being special: Repo markets during the crisis (ECB Working Paper 2065). Das, U. S., Papapioannou, M., Pedras, G., Ahmed, F., & Surti, J. (2010). Managing public debt and its financial stability implications (IMF Working Paper 10/280). Dottori, D., & Manna, M. (2016). Strategy and tactics in public debt management. Journal of Policy Modeling,38(2016), 1–25. Dybczak, K., & Melecky, M. (2014). EU fiscal stance vulnerability: Are the old members the gold members? Economic Modelling,38(C), 87−101. Eickmeier, S., & Hofmann, B. (2013). Monetary policy, housing booms and financial (im)balances. Macroeconomic Dynamics,17(4), 830−860. European Commision (2012). Fiscal sustainability report 2012 (European Economy 8/2012). Economic and Financial Affairs. Faraglia, E., Marcet, A., & Scott, A. (2008). Fiscal insurance and debt management in OECD economies. The Economic Journal,118(527), 363−386. Faraglia, E., Marcet, A., & Scott, A. (2010). In search of a theory of debt management. Journal of Monetary Economics,57(7), 821−836. Ferdinandusse, M., Freier, M., & Ristiniemi, A. (2017). Quantitative easing and the price-liquidity tradeoff(ECB Working Paper 2021). Forni, M., & Gambetti, L. (2010). The dynamic effects of monetary policy: A structural factor model approach. Journal of Monetary Economics,57(1), 203−216. Forni, M., Giannone, D., Lippi, M., & Reichlin, L. (2009). Opening the black box: Structural factor models with large cross sections. Econometric Theory,25(5), 1319–1347. JOURNAL OF APPLIED ECONOMICS 271 Franta,M.(2012). Macroeconomic effects of fiscal policy in the Czech Republic: Evidence based on various identification approaches in a VAR framework (Czech National Bank Working Paper 13/2012). Franta, M., Libich, J., & Stehlik, P. (2018). Tracking monetary-fiscal interactions across time and space. International Journal of Central Banking,14(3), 167−227. Fry, R., & Pagan, A. (2011). Sign restrictions in structural vector autoregressions: A critical review. Journal of Economic Literature,49(4), 938−960. Hodula, M., & Pfeifer, L. (2018). Fiscal-monetary-financial stability interactions in a data-rich environment. Review of Economic Perspectives,18(3), 195−224. IMF (2013). Czech Republic, 2013 Article IV consultation (IMF Country Report No. 13/243). IMF (2017). The medium-term debt management strategy: An assessment of recent capacity building (IMF Policy Paper July 28). IMF and WB. (2014). Revised guidelines for public debt management. International Monetary Fund and the World Bank. Washington D.C. Kilian, L. (1998). Small-sample confidence intervals for impulse response functions. Review of Economics and Statistics,80(2), 218−230. Kirsanova, T., & le Roux, S. (2013). Commitment vs. discretion in the UK: An empirical investigation of the monetary and fiscal policy regime. International Journal of Central Banking,9(4), 99–152. Komarkova, Z., Dingova, V., & Komarek, L. (2013). Fiscal sustainability and financial stability (CNB Financial Stability Report 2012/2013). Prague: Czech National Bank. Leeper, E. M., Walker, T. B., & Yang, S.-C. S. (2013). Fiscal foresight and information flows. Econometrica,81, 1115–1145. Matalik, I., & Slavik, M. (2005). Debt management in the Czech Republic (formation in the 1990s and the current state). Prague Economic Papers,1(1), 33−50. Melecky, M. (2012). Choosing the currency structure of foreign-currency debt: A review of policy approaches. Journal of International Development,24(2), 133−151. Ministry of Finance. (2012). Methodological manual of the fiscal outlook CR. Prague: Ministry of Finance. Ministry of Finance (2017). The Czech Republic government debt management annual report 2016. Prague: Ministry of Finance. Moore, W., & Skeete, K. (2010). The implications of monetary policy shocks for government debt management in Barbados (CBB Working Paper 2010). Panzer, J., Thomas, M. R., Peuker, A. R., Razlog, L., Prasad, A., Anderson, P. R. D., & Li, Y. (2015). Debt management performance assessment (DeMPA) methodology. Washington, D.C.: World Bank Group. Retrieved from http://documents.worldbank.org/curated/en/ 305821468190742099/Debt-management-performance-assessment-DeMPA-methodology Pavelek, P. (2005). Risk management of government debt in the Czech Republic. In Advances in risk management of government debt (pp. 245−262). Paris: OECD. Ramey, V. A. (2011). Identifying government spending shocks: It’s all in the timing. The Quarterly Journal of Economics,126(1), 1–50. Romer, C. D., & Romer, D. H. (2010). The macroeconomic effects of tax changes: Estimates based on a new measure of fiscal shocks. American Economic Review,100(3), 763–801. Sargent, T. J., & Wallace, N. (1981). Some unpleasant monetarist arithmetic. Quarterly Review,5(3), 1–17. Schlepper, K., Riordan, R., Hofer, H., & Schrimpf, A. (2017). Scarcity effects of QE: A transaction-level analysis in the Bund market (BIS Working Paper No 25). Sight, C. (2015). Separation of debt and monetary management in India. IIMB Management Review,27,56−71. Soares, R. (2013). Assessing monetary policy in the euro area: A factor-augmented VAR approach. Applied Economics,45(19), 2724–2744. Stock, J., & Watson, M. (1998). Diffusion indexes (NBER Working Paper 6702). Stock, J., & Watson, M. (1999). Forecasting inflation. Journal of Monetary Economics,44(2), 293–335. Stock, J., & Watson, M. (2002). Forecasting using principal components from a large number of predictors. Journal of the American Statistical Association,97(1), 1167−1178. Stock, J., & Watson, M. (2016). Dynamic factor models, factor-augmented vector autoregressions, and structural vector autoregressions in macroeconomics. In J. B. Taylor & H. Uhlig (Eds.), Handbook of macroeconomics, ed. 1, chapter 0 (Vol. 2, pp. 415–525). Amsterdam: Elsevier. 272 M. HODULA AND A. MELECKÝ Svirydzenka, K. (2016). Introducing a new broad-based index of financial development (IMF Working Paper 16/5). Togo, E. (2007). Coordinating public debt management with fiscal and monetary policies: An analytical framework (World Bank Policy Research Working Paper 4369). Tuzcuoglu, K., & Hoke, S. H. (2016). Interpreting the latent dynamic factors by threshold FAVAR model (Bank of England Working Paper 622). Wheeler, G. (2004). Sound practice in government debt management. Washington D.C.: The World Bank. Appendices A. Data description Table 1A. Shows all time-series incorporated in the analysis. Abbreviations stand for: CSO –Czech Statistical Office, CNB –Czech National Bank database ARAD, IMF –International Monetary Fund database, ECB –European Central Bank Statistical Data Warehouse. The transformation codes (TC) are: 1–no transformation; 2 –first difference of logarithm. An asterisk, “*”, next to the transformation code number denotes seasonally-adjusted variables using CENSUS X13. S/F ranks variables as slow or fast moving in the estimation. Group No. Series description Unit Source TC S/F Real economy 1 Industrial production index, industry total 2010 = 100 CSO 2* S 2 Industrial production index, mining and quarrying 2010 = 100 CSO 2* S 3 Industrial production index, manufacturing 2010 = 100 CSO 2* S 4 Industrial production index, electricity, gas, steam and air conditioning 2010 = 100 CSO 2* S 5 Sales from industrial activity, industry total 2010 = 100 CSO 2* S 6 Sales from industrial activity, mining and quarrying 2010 = 100 CSO 2* S 7 Sales from industrial activity, manufacturing 2010 = 100 CSO 2* S 8 Sales from industrial activity, electricity, gas, steam 2010 = 100 CSO 2* S 9 Direct export sales, industry total 2010 = 100 CSO 2* S 10 Direct export sales, mining and quarrying 2010 = 100 CSO 2* S 11 Direct export sales, manufacturing 2010 = 100 CSO 2* S 12 Domestic sales, industry total 2010 = 100 CSO 2* S 13 Domestic sales, mining and quarrying 2010 = 100 CSO 2* S 14 Domestic sales, manufacturing 2010 = 100 CSO 2* S 15 Domestic sales, electricity, gas, steam 2010 = 100 CSO 2* S 16 New industrial orders, industry total 2010 = 100 CSO 2* S 17 Non-domestic new orders 2010 = 100 CSO 2* S 18 Domestic new orders 2010 = 100 CSO 2* S 19 Construction production index 2010 = 100 CSO 2* S 20 Construction production index, buildings 2010 = 100 CSO 2* S 21 Construction production index, civil engineering works 2010 = 100 CSO 2* S 22 Retail trade receipts 2010 = 100 CNB, ARAD 2* S 23 Gross domestic product, market prices Millions CZK CSO 2* S 24 Gross fixed capital formation Millions CZK CSO 2* S 25 GDP deflator 2010 = 100 CNB, ARAD 2* S 26 Final consumption expenditures, total, current prices Millions CZK CSO 2* S 27 Final consumption expenditures, households, current prices Millions CZK CSO 2* S 28 Final consumption expenditures, government, current prices Millions CZK CSO 2* S 29 Final consumption expenditures, non-profit organizations, current prices Millions CZK CSO 2* S 30 Gross capital formation, total, current prices Millions CZK CSO 2* S 31 Export, current prices Millions CZK CSO 2* S 32 Import, current prices Millions CZK CSO 2* S 33 Real gross domestic income Millions CZK CSO 2* S (Continued) JOURNAL OF APPLIED ECONOMICS 273 Table 1A. (Continued). Group No. Series description Unit Source TC S/F Labor market 34 Industry total, average number of persons employed (ANPE) No. of persons CSO 2* S 35 Industry, mining and quarrying, ANPE No. of persons CSO 2* S 36 Industry, manufacturing, ANPE No. of persons CSO 2* S 37 Industry, electricity, gas, steam and air conditioning supply, ANPE No. of persons CSO 2* S 38 Industry total, average gross nominal wage (AGNW) CZK per person CSO 2* S 39 Industry, mining and quarrying, AGNW CZK per person CSO 2* S 40 Industry, manufacturing, AGNW CZK per person CSO 2* S 41 Industry, electricity, gas, steam and air conditioning supply CZK per person CSO 2* S 42 Construction total, average number of persons employed (ANPE) No. of persons CSO 2* S 43 Construction total, average gross nominal wage (AGNW) CZK per person CSO 2* S 44 Employees total, hours worked Thousand hours CSO 2* S 45 Employees, Agriculture, forestry and fishing Thousand hours CSO 2* S 46 Employees, Manufacturing, mining and quarrying and other Thousand hours CSO 2* S 47 Employees, Construction Thousand hours CSO 2* S 48 General unemployment rate among those aged 15 to 64 % CNB, ARAD 1* S 49 Job Vacancies Thousand CNB, ARAD 2* S 50 Unplaced job seekers Thousand CNB, ARAD 2* S Government 51 Government debt, total Millions CZK CSO 2* S 52 Debt securities, total Millions CZK CSO 2* S 53 Debt securities, short-term Millions CZK CSO 2* S 54 Debt securities, long-term Millions CZK CSO 2* S 55 Government loans, total Millions CZK CSO 2* S 56 Government loans, short-term Millions CZK CSO 2* S 57 Government loans, long-term Millions CZK CSO 2* S 58 Debt interests payed Millions CZK CSO 2* S 59 Government expenditures, total Millions CZK CSO 2* S 60 Government revenue, total Millions CZK CSO 2* S 61 Debt to GDP ratio Ratio CSO 1 S 62 Debt service costs = interests payed in t/debt in t-1 Ratio CSO 1 S 63 Government revenues from taxes Millions CZK CSO 2 S 64 Tax quota = government revenues from taxes/GDP Ratio CSO 2 S Prices and price expectations 65 Consumer Price Index (CPI), total 2005 = 100 CNB, ARAD 2* S 66 CPI, food and non-alcoholic beverages 2005 = 100 CNB, ARAD 2* S 67 CPI, alcoholic beverages, tobacco 2005 = 100 CNB, ARAD 2* S 68 CPI, clothing and footwear 2005 = 100 CNB, ARAD 2* S 69 CPI, housing, water, electricity, gas and other fuels 2005 = 100 CNB, ARAD 2* S 70 CPI, furnishings, household equipment, routine maintenance 2005 = 100 CNB, ARAD 2* S 71 CPI, health 2005 = 100 CNB, ARAD 2* S 72 CPI, transport 2005 = 100 CNB, ARAD 2* S 73 CPI, communications 2005 = 100 CNB, ARAD 2* S 74 CPI, recreation and culture 2005 = 100 CNB, ARAD 2* S 75 CPI, education 2005 = 100 CNB, ARAD 2* S 76 CPI, restaurants and hotels 2005 = 100 CNB, ARAD 2* S 78 CPI, miscellaneous goods and services 2005 = 100 CNB, ARAD 2* S 79 Industrial Producer Prices (IPP), total 2005 = 100 CNB, ARAD 2* S (Continued) 274 M. HODULA AND A. MELECKÝ Table 1A. (Continued). Group No. Series description Unit Source TC S/F 80 IPP, mining and quarrying 2005 = 100 CNB, ARAD 2* S 81 IPP, manufacturing 2005 = 100 CNB, ARAD 2* S 82 IPP, electricity, gas, steam and air conditioning supply 2005 = 100 CNB, ARAD 2* S 83 IPP, water supply; sewerage, waste management and remediation 2005 = 100 CNB, ARAD 2* S 84 Inflation expectations of non-financial corporations for 1Y ahead % CNB, ARAD 1* F 85 Financial market inflation expectations for 1Y horizon % CNB, ARAD 1* F Interest rates and credits 86 Repo rate –2 weeks % CNB, ARAD 1 F 87 PRIBOR 3 M % CNB, ARAD 1 F 88 PRIBOR 1Y % CNB, ARAD 1 F 89 Government bond yield 2Y % CNB, ARAD 1 F 90 Government bond yield 5Y % CNB, ARAD 1 F 91 Government bond yield 10Y % CNB, ARAD 1 F 92 Bank interest rates on CZK-denominated loans, non-financial corporations % CNB, ARAD 1 F 93 Bank interest rates on CZK-denominated loans, households total % CNB, ARAD 1 F 94 Bank interest rates on CZK-denominated loans, consumer credit % CNB, ARAD 1 F 95 Bank interest rates on CZK-denominated loans, credit for house purchase % CNB, ARAD 1 F 96 Bank interest rates on CZK-denominated loans, other loans –total % CNB, ARAD 1 F 97 Bank interest rates on CZK-denominated loans, NFC, up to 1Y % CNB, ARAD 1 F 98 Bank interest rates on CZK-denominated loans, NFC, up to 5Y % CNB, ARAD 1 F 99 Bank interest rates on CZK-denominated loans, NFC, over 5Y % CNB, ARAD 1 F 100 Monetary base, monthly average Billions CZK CNB, ARAD 2 F 101 Monetary aggregate M1 Millions CZK CNB, ARAD 2 F 102 Monetary aggregate M2 Millions CZK CNB, ARAD 2 F 103 Loans to private sector, total Millions CZK CNB, ARAD 2 F 104 Net foreign assets Millions CZK CNB, ARAD 2 F 105 Loans to residents and non-residents –MFIs Millions CZK CNB, ARAD 2 F 106 Loans to non-financial corporations –MFIs Millions CZK CNB, ARAD 2 F 107 Loans to financial corporations –MFIs Millions CZK CNB, ARAD 2 F 108 Loans to government Millions CZK CNB, ARAD 2 F 109 Loans to households Millions CZK CNB, ARAD 2 F 110 Loans, short-term (up to 1Y) Millions CZK CNB, ARAD 2 F 111 Loans, medium-term (up to 5Y) Millions CZK CNB, ARAD 2 F 112 Loans, long-term (over 5Y) Millions CZK CNB, ARAD 2 F 113 Consumption loans, total Millions CZK CNB, ARAD 2 F 114 Mortgages, total Millions CZK CNB, ARAD 2 F Financial sector 115 Capital adequacy ratio, large banks % CNB 1* F 116 Leverage ratio, large banks % CNB 1* F 117 Risk-weighted assets, large banks % CNB 1 F 118 Non-performing loans, large banks % CNB 1 F 119 Credit spread, large banks % CNB 1 F 120 Composite indicator of sovereign stress 0–1 interval ECB 1 F 121 Financial cycle indicator 0-1 interval CNB 1 F 122 Difference between credit-toGDP ratio and the minimum ratio % CNB 1 F 123 Index PX Value PSE 2 F 124 Credit-to-GDP gap, (since 2000), HP filter, lambda = 200,000 p. p. Own calculation 1F 125 MFI total assets Millions CZK CNB –ARAD 2 F 126 House price index 2010 = 100 CSO 2 F 127 Financial markets development index IMF 2 F (Continued) JOURNAL OF APPLIED ECONOMICS 275