Sovereign currency and long-term interest rates in advanced economies from 1879 to 2016

ABSTRACT

This paper explores the effects of government debts and deficits on long-term interest rates under different international monetary systems from the perspective of monetary sovereignty. The findings of this paper confirm conventional accounts of financial crowding-out and the adverse impact of growing public debt on interest rates among countries under the classical Gold Standard period (1879–1913) and non-sovereign countries under the recent fiat system (1973–2016). By contrast, countries in the Bretton Woods era (1959–1970) and sovereign countries in the contemporary post-Bretton Woods era (1973–2016) do not exhibit such positive debt-interest rates relationships while still showing crowding-out but with much less magnitude. Overall, the results support the Modern Money Theory (MMT) view that the consequence of monetary sovereignty from 1879 to 2016 is to remove, or greatly attenuate, the impact of default risk, fiscal balance, and bond vigilantes on the interest rate.

KEYWORDS:

• Sovereign currency
• government debt
• interest rate
• monetary system
• panel data

JEL CLASSIFICATION:

• E42
• E43
• E58
• H6

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. There is a large body of empirical literature that focuses on fiscal factors (Baldacci and Kumar 2010; Bernoth, Hagen, and Schuknecht 2012; Gruber and Kamin 2012; Laubach 2009) and non-domestic variables such as over-reaction of financial markets (Aizenman, Hutchison, and Jinjarak 2013; Ang and Longstaff 2013; Beirne and Fratzscher 2013; D’Agostino and Ehrmann 2014; Poghosyan 2014), global fiscal policy (Ardagna, Caselli, and Lane 2007; Dell’erba and Sola 2016), and external competitiveness/imbalances (Costantini, Fragetta, and Melina 2014; Salem and Castelletti-Font 2016), all of which have not taken into account underlying monetary and exchange rate arrangements of countries. Once recognising the importance of currency sovereignty, it is perhaps not surprising to see heterogeneous findings regarding the impact of fiscal deterioration on interest rates. See Kim (2021) for further theoretical and empirical studies on this topic.

2. Monetary sovereignty and MMT’s perspective on interest rates will be discussed in depth in the next section.

3. Contrary to the view of credit rating agencies (CRAs) (e.g. Standard and Poor’s 2007), such default risk must be clearly differentiated from the inflation risk for three reasons. First, a government usually does not promise a constant purchasing power on upcoming debt service payments so it is not defaulting if their purchasing power falls. Equally, government security holders are not required to compensate a government for any windfall of purchasing power due to deflation. Nominalism applies in the law so governments are under no obligation to provide a hedge against inflation; bond holders take upon themselves the inflation risk. Governments may promise to do so via inflation-indexed securities – in which case failure to maintain purchasing power is a default event – but they do not have to do it and usually do not. Second, CRAs see tax revenues as the sound means to service the public debt but inflation raises tax revenues and so raises the ability to service the debt, which lowers the default risk. Third, the CRA view of default risk implicitly assumes that the government is responsible for inflation. Not only is there no clear link between fiscal deficits and inflation, but also all forms of spending can be inflationary and there are other sources of inflation than spending (such as energy costs, medical costs, overhead corporate cost structure, among others). The central implication is that the interest rate on the domestic public debt of a monetary sovereign government is not influenced by the credit ratings of the public debt.

4. As the experience with US Treasury’s tax and loan accounts illustrates (Tymoigne 2014b), the government may decide to let proceeds from taxes and issuances of securities be credited to accounts it holds at private banks (e.g. in the example below, $75 of tax revenues would be credited on the asset side of the government via an increase in private bank accounts held by the government). Such financial operations, however, would be done for monetary policy purpose instead of financing purposes, as was the case for the US Treasury. A monetarily sovereign government, by definition, uses its own payment system (or more broadly a payment system it controls) to transact (tax, spend, issue securities), otherwise it is a user of a currency instead of its issuer.

5. While one may argue that the consolidated view is too much of a simplification that leads to counterintuitive conclusions, all these conclusions are grounded in the practical operations of the Treasury and central bank of a monetary sovereign government. The hypothesis is a theoretical simplification that is backed by a detailed analysis of the heavy coordination between the central bank and Treasury in monetary and fiscal policies. Treasury gets involved in monetary policy and the central bank routinely gets involved in fiscal policy to ensure that the monetary and fiscal goals are achieved (Bell 2000; Fullwiler 2011; Fullwiler, Kelton, and Wray 2012; Tymoigne 2014b). Taxes drain reserves while government spending adds reserves and so deficit spending leads to a net injection of reserves that needs to be neutralised by selling government securities. Both the Treasury and the central bank have been involved in making sure that reserve injections and removals are done in a way that allows monetary and fiscal goals to be met. With monetary sovereignty, the role of the central bank expands beyond providing an elastic currency to banks and towards ensuring that the Treasury can finance its expenditures. The central bank becomes the investment banker of the Treasury and works behind and in front of the scene to make sure that Treasuries auctions are orderly. On the other side, the role of the Treasury expands beyond implementing the legislative budget and collecting taxes and towards ensuring the central bank can meet its interest rate target. As such, the Treasury manages the public debt to fulfill broad macroeconomic goals that are not necessarily consistent with a budgetary logic (Tymoigne 2020). For example, the supply of treasuries has to be large enough to manage interest rates and the Treasury may continue to issue Treasuries even when running a fiscal surplus.

6. MMT argues that political independence has been granted to central banks in order to shield them from short-term political influence when fulfilling their mandate of ensuring price stability. Central banks also have independence in choosing tools (interest-rate setting and quantitive easing) to achieve goals (inflation, etc.). On the other hand, the Treasury and the central bank must work extensively together for fiscal and monetary policies to work properly. In that sense, the central bank cannot be fully independent of the Treasury (operational dependence). Similarly, the Treasury needs to work monetary-policy considerations into its daily activities otherwise interest-rate stability would not be maintained and inflationary pressures could more easily materialise.

7. Jordà, Schularick, and Taylor consulted a broad range of sources, such as economic and financial history volumes and journal articles, and various publications of statistical offices and central banks. For more information about the source(s) for each variable by a country, refer to their website: http://www.macrohistory.net/JST/JSTdocumentationR4.pdf.

8. We focus on nominal, not real interest rates because the former is generally the main policy target of the central bank and thus a proxy for monetary policy space. In addition, having a lagged variable for the debt ratio and real GDP growth addresses endogeneity that derives from reverse causality and simultaneity as commonly done in the relevant literature.

9. Public debt in foreign currency should be treated differently from that in local currency since the former limits the capacity of the sovereign government to service its debt, as discussed in MMT. Nevertheless, we do not include a measure of exposure to foreign currency-denominated debt because public debt in foreign currency issued by sovereign countries in the fiat systems is minimal. Claessens, Klingebiel, and Schmukler (2007) found that only a small fraction (2%) of government debt of advanced economies was issued in foreign currency as of December 31, 2000 and since then countries have issued relatively more public debt in local currency.

10. It also implies that an increase in the public debt ratio is negatively associated with interest rates when the stock of debt is below the tipping point for non-sovereign countries. This negative non-linear impacts of public debt on bonds rates can be explained with a portfolio effect (Caporale and Williams 2002). Portfolio theory states that when a government issues high-quality, low-risk debt, investors switch into them from bad-quality, high-risk debt, putting downward pressure on the bond yield of the former. However, further increases beyond the threshold are associated with higher interest rates as investors demand a higher default risk premium.

11. Controls on capital movements under the Bretton Woods regime were essential to its functioning, and thus cross-border flows of private capital were constrained. As a result, capital controls may have shielded domestic monetary policy to some extent when exchange rate expectations were stable (Bakker and Chapple 2002). The demise of the Bretton Woods system started from the failure to preserve exchange rate stability by tightening controls on capital outflows in the face of clear underlying imbalances. On the other hand, in the fiat system, monetary policy acting in support of government policies to achieve domestic economic goals would lead to exchange rate instability, thus endangering fixed exchange rates under, for instance, the European Monetary System (EMS) (Georgiadis and Zhu 2019). To achieve exchange rate stability under free capital flow, countries with a pegged currency had to follow base-country policy rates or otherwise, they were subject to speculative ‘hot-money’. In other words, if a country does not have monetary sovereignty, then its central bank cannot control short-term interest rates, let alone influence long-term interest rates and the complex structure of interest rates in the context of the topic of this paper. However, as Shambaugh (2004), Obstfeld et al. (2005), and Klein and Shambaugh (2015) show, imposing capital account restrictions leads to a significant increase in monetary policy autonomy in terms of the administration of short-term and long-term rates.

12. We could additionally exclude Germany but the results would not fundamentally change.