![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
Abstract
This paper investigates a long-run relationship between public debt accumulation and default risk, represented by Eurozone countries’ bond rates minus German benchmark bond rates for 9 Eurozone countries under the ECB’s conventional vs. nonstandard monetary policy for the period 2000–2015. Along with various unit root tests and cointegration tests, Dynamic OLS (DOLS) and Fully Modified OLS (FMOLS) methods are applied in order to examine a common long-run linkage between bond rates spreads and macro variables concerned without ignoring heterogeneous short-run dynamics. Such techniques directly address the endogeneity issue often encountered in analyses of economic fundamental variables. The empirical evidence reveals that a positive relationship between the bond rates spread and the debt-to-GDP ratio is found during the European Sovereign Debt Crisis and before the Global Financial Crisis in which the conventional monetary policy prevailed, reflecting negative market sentiments on default risk and market discipline. A negative long-run relationship is, by contrast, shown under the effective unconventional monetary policy, implying that overstated (exaggerated) default risk diminished after the ECB’s nonstandard measures. This phenomenon is especially apparent in, but not restricted to, peripheral Eurozone countries where most of the new monetary measures were targeted.
Notes
1 Public debt burden is a cost of paying interest on public debt along with the principal. In section 3, it will be specified how it is paid under the ECB’s conventional vs. nonstandard monetary policy.
2 Conventional monetary policy uses its main instrument, setting a target short-term interest rate and, if necessary, providing a short-term liquidity to banks and the interbank market through open market operations, presuming that longer-term rates will behave in similar way. After it was proven ineffective as the official rate hits a zero lower bound and the usual monetary transmission mechanism from the official rate to market rates stopped working, unconventional monetary policy was implemented with the focus on the size of central bank balance sheets (Joyce et al. Citation2012). It involves purchases of assets in an unconventionally large amount and in a different kind usually with attempts to influence interest rates beyond the short-term target rate. In a nutshell, the focus of monetary policy has shifted from prices to quantities. Refer to the next section for more detail.
3 9 European Monetary Union member countries (Austria, Belgium, Finland, France, Italy, the Netherlands, Portugal, Spain, and Ireland) are used for study over the period 2000 to 2015 while Greece and Luxembourg are not included due to data availability. It is also not appropriate to include Greece and Luxembourg since the former opted out of the ECB’s expanded asset purchase program and the outstanding government debt, and the associated market of the latter are very small. Other Eurozone countries such as Cyprus, Estonia, Latvia, Malta, Slovakia, and Slovenia are also excluded for the balanced panel data since they have adopted euro in later periods unlike other countries that did in 1999. Finally, Germany is used as a benchmark country.
4 Eurosystem consists of the ECB and (currently 19) national central banks of the EMU countries.
5 LTROs allowed financial institutions to have almost unlimited access to central bank liquidity, subject to an adequate collateral in two ways: first, the maturity of LTRO was extended to ensure that banks would have longer liquidity planning horizons (i.e. initially six months in 2008 and then up to three years in 2011); the list of eligible collateral was, secondly, stretched to include lower quality ones.
6 Debt-monetization is defined as the currency-issuing authority (i.e. ECB) purchasing public bonds with their IOUs (reserves) newly created by the authority, leaving the system with an increased supply of liquidity. Looking at the end results, QE is qualified to be debt-monetization.
7 The ECB would intervene in secondary sovereign debt markets only if countries concerned comply with a precautionary macroeconomic adjustment program set by the European Financial Stability Facility and the European Stability Mechanism (i.e. austerity program).
8 Hale and Obstfeld (Citation2016) specified the four factors that provided financial institutions in export-surplus countries with a comparative advantage in lending to the periphery. It can be argued that bond markets, before the crisis, already reflected the likelihood that the ECB would use unconventional monetary policy for a tough time. Although it has an element of truth, it cannot explain why bond rates abruptly skyrocketed even after an extended version of LTROs and SMP were in place.
9 Default risk is defined as a probability that a country fails to fulfill its debt obligations in full at due time. Thus, it involves debtors’ (funding) liquidity risk and redenomination risk whereby a member country exits the euro and pay in a different, lower-valued currency.
10 Fullwiler (Citation2010, 5) asserts that the interest rate on the national bond market is “a policy variable not a market-set rate or at the very least could be if the government so desires” for a sovereign currency issuer under flexible currency rate in strong, semi-strong, and weak-form of the efficient market hypothesis.
11 From the perspective of Keynes’s interest rate theory of liquidity preference, Mario Draghi’s statement and the announcement of OMTs, by themselves, did not change money supply in the system. All they did was to calm down the exaggerated liquidity preference of bond holders by giving them confidence that ECB would protect struggling bond markets and thus bond prices would not fall. Through effective, verbal communication backed by the tangible instrument, the ECB could successfully stabilize sovereign bond markets.
12 See Lenza, Pill, and Reichlin (Citation2010), Aït -sahalia et al. (Citation2012), Fahr et al. (Citation2013), De Grauwe and Ji (Citation2013), Matei and Cheptea (Citation2013), and Darracq-Paries and De Santis (Citation2015).
13 The former case causes the multicollinearity issue and thus loss of efficiency while the latter case the simultaneous (endogeneity) bias.
14 De Grauwe and Ji (Citation2013) select Australia, Canada, Czech, Denmark, Hungary, Japan, Norway, Poland, Singapore, South Korea, Sweden, Switzerland, UK, and US as stand-alone countries.
15 Data have been obtained from the OECD data (https://data.oecd.org/) and confirmed by Eurostat (http://ec.europa.eu/eurostat/data/database). The selection of the sample period is limited by the availability of data on relevant variables.
16 Hausman test verifies that the fixed effect model is more appropriate than the random effect model.
17 The ECB announced 6-month LTROs on March 27, 2008, regarded as the first nonstandard monetary policy instrument, and the balance sheet of the Eurosystem has dramatically expanded since then, as found in . Hence, we can safely set 2007Q4 as the end of the ECB’s conventional monetary policy. Also, 2009Q3–2012Q1 is widely acknowledged as the European sovereign debt crisis in the literature. Lastly, more aggressive unconventional policy, the Outright Monetary Transactions (OMTs) was implemented from September 2012, followed by the expanded asset purchase program since the late 2014. Such categorization of three sub-periods is statistically confirmed by the Chow breakpoint test.
18 Core European countries consist of Austria, Belgium, Finland, France, and the Netherlands while peripheral Eurozone countries are made up of Italy, Portugal, Spain, and Ireland.
19 It does not mean we are totally free from the identification problem. Regulatory change, fiscal and structurally adjustment, potentially long and variable policy lags, spillover effects across markets and countries, and changing impact of policy initiative complicates the simultaneity issue (Kozicki, Santor, and Suchanek Citation2011). With this caveat in mind, we should study the effects of unconventional monetary policy.
20 The unit root test results are available from the author upon request.
21 The null hypothesis of no cointegration is rejected only in the sub-period of Post-OMT in both tests for core Eurozone countries. For peripheral countries, the null hypothesis fails to be rejected only in the sub-period of Pre-Crisis in Pedroni Residual Cointegration Tests.
22 A possible reverse causality is that higher bond rates spreads are most likely to increase debt-to-GDP ratios and decrease real GDP.
23 Mark and Sul (Citation2003) states that DOLS allows for a limited degree of cross-sectional dependence through the presence of time-specific effects.
24 The author is certain about how heterogeneous slope coefficients are among countries, thus adopting both within- and between-dimensional approaches.
25 Current account-to-GDP ratio does not exhibit a consistent trend.
26 Austerity policy may work if the real effective exchange rate of a country declines excessively and thus it achieves trade surplus through an increase in external competitiveness. However, when every country pursues the same policy, it is an unlikely event since the real effective exchange rate is a relative term. It would rather hurt domestic economic growth, which leads to higher bond rates spreads.
27 It is interesting that four ECB economists (Bunea et al. Citation2016) recently claims that the ECB faces no financial constraint since it can create money and operate with negative equity, and thus their performance should be judged by policy objectives other than profitability.