The politicisation of the European Central Bank and its emergency credit lines outside the Euro Area

ABSTRACT

Is the European Central Bank (ECB) increasingly acting on political – rather than technocratic – considerations? This question is of a central concern to students of European Union (EU) political economy. This article contributes to this debate by studying the ECB’s credit lines to the central banks of EU member states outside the Euro Area during the Global Financial Crisis and the COVID-19 crisis. Both times the ECB accorded selectively better borrowing conditions to some central banks. The article finds that its selection of who gets favourable borrowing terms has indeed become more political. In 2008, the ECB decided the credit terms based on technocratic criteria, but twelve years later, it granted better lending conditions to countries that were close to adopting the euro. How the ECB balances its mandate for price stability in the Euro Area and its role as a supranational EU institution decides whether it will become more politicised.

KEYWORDS:

• Euro adoption
• European Central Bank
• international role of the Euro
• politicisation
• swap lines

Introduction

The role of the European Central Bank (ECB) in European macroeconomic governance has changed profoundly since the onset of the Global Financial Crisis (GFC). Since then, the ECB has established a series of new policy instruments and, as member of the Troika, exercised considerable influence on macroeconomic policy in member states. In addition to its monetary policy responsibilities, the ECB sits at the centre of the Banking Union and supervises the Euro Area’s major banks. The success of these institutional reforms and the increasingly bold policy measures that the ECB has adopted to calm down bond markets during the COVID-19 crisis in 2020, have led many analysts to describe the ECB as the ‘only game in town’ for macroeconomic policy in the European Union (EU) (Koranyi & Canepa, 2020; Marsh & Kyriakopolou, 2020).

For political economists studying the ECB, an important question has been how the ECB has changed as an actor over time and various authors have argued that the ECB has become more politicised. It has been highlighted that the ECB has acted increasingly strategically (Torres, 2013) and played the role of a leader within the EU (Schoeller, 2018; Verdun, 2017). The increased salience of its decisions has led the ECB to draw on more political justifications for its policy decisions (Moschella et al., 2020; Rauh, 2021; Tortola & Pansardi, 2019) and to build strategic alliances to legitimise its actions (Collignon & Diessner, 2016). As the actions of the ECB have increasingly brushed up against its mandate, many authors have cautioned that this politicisation may threaten its democratic legitimacy (Högenauer & Howarth, 2016; Schmidt, 2016; van’t Klooster, 2018; van’t Klooster & Boer, 2020; Verdun, 1998).

This study contributes to this literature by examining a so far understudied aspect of ECB policy, namely international central bank cooperation. During both the GFC in late 2008 and the COVID-19 crisis in early 2020, central banks from non-Euro Area EU countries (henceforth: ‘Euro Area outsiders’) approached the ECB with requests for emergency credit lines. During both crises, the ECB selectively granted better credit conditions to some Euro Area outsiders than to others. Some central banks received swap lines, while others received repo lines. As section 2.2 will explain in more detail, repo lines offer less favourable conditions to the recipient than swaps. This article examines the criteria that the ECB used to determine which Euro Area outsiders could borrow under swap arrangements and which ones under repo lines in respectively 2008 and 2020. The article aims to find out whether the borrowing terms applied by the ECB have become more politicised over time.

International central bank cooperation has been instrumental in containing international financial instability during both the GFC (Tooze, 2018) and the COVID-19 crisis. So far, most of the literature on the issue has, however, focused on the provision of swap lines by the United States (US) Federal Reserve (Harris, 2015; McDowell, 2012; Pape, 2021; Sahasrabuddhe, 2019). Less attention has been paid to cooperation among central banks in Europe – despite the role of the ECB as issuer of the second most important international currency (Allen, 2013 offers an exception). Yet, the ECB decisions in 2008 have been politically controversial. Concretely, its reluctance to provide more support to central banks in Central and Eastern Europe has received comprehensive criticism afterwards (Åslund, 2010; Tooze, 2018, chap. 9). Whereas the literature on ECB politicisation has so far focused on the ECB’s policies inside the Euro Area, this article highlights the ECB’s scope of action beyond the Euro Area.

This article compares the ECB’s credit lines to EU central banks in 2008 and 2020 based on recently disclosed internal documents and elite interviews (a list of interviewees is provided in the online appendix). The analysis reveals that the borrowing terms have over time become politicised. In 2008, the ECB set the borrowing terms under the credit lines in line with narrow technical considerations, even if that meant that they were of little use for the East European central banks; in 2020, ECB decisions were influenced by political considerations around Euro Area enlargement and it granted surprisingly good borrowing terms to Bulgaria and Croatia, the two countries closest to adopting the euro. This finding suggests a new source of ECB politicisation which lies in the conflict between the monetary policy interests of the Euro Area and the role of the ECB as a supranational EU institution. Even in situations where the survival of the Euro Area was not a stake, the ECB decided to deviate from technocratic policy-making principles.

The remainder of the article is structured as follows. Section two reviews the literature on ECB politicisation and then operationalises this concept in the context of international monetary cooperation and the mandate of the ECB. Sections three and four respectively discuss the borrowing terms that the ECB accorded to Euro Area outsiders during the GFC and the COVID-19 crisis. The last section discusses the findings and concludes.

Central bank politicisation and the ECB’s international cooperation

Central bank politicisation

The term politicisation has been used to describe a variety of distinct processes in the study of European integration (de Wilde, 2011). Some authors apply it to the salience of topics pertaining to European integration in national and European discourse (Kriesi, 2016; Schmidt, 2019) and the polarisation of opinions around the issue (Hooghe & Marks, 2009). Others have applied the term to categorise the formal influence of elected officials on decision making in the EU (Radaelli, 1999) and have argued that decision making in the EU, especially in economic affairs has been de-politicised (Sánchez-Cuenca, 2017). This article, however, applies Tortola’s (2020) preference-based definition of the politicisation of technocratic institutions, such as the ECB. In this context, a technocratic institution, such as the ECB, becomes politicised when it ceases to follow ‘techno-scientific considerations and starts following preferences and motivations situated in the realm of partisan politics’ (Tortola, 2020, p. 507).

To see why the politicisation of central banks’ preferences might be problematic, a good starting point is to consider technocratic policymaking as a baseline (Radaelli, 2013). Technocrats are supposed to operate at arms’ length from their political principals; with clear mandates, but enough discretion to take decisions based on their technical expertise (Tucker, 2019). Though some have argued that monetary policy is inevitably political because of its distributive consequences (Dietsch, 2020), the original argument for central bank independence was that this redistribution can be tolerated, so long as society as a whole benefits (Majone, 2001; McNamara, 2002). As a result, central banking in the 2000s was ostensibly a-politicised and ‘scientised’ (Marcussen, 2009); monetary policy was supposed to ensure price stability and be decided based on technical considerations.

As soon as technocrats stop basing their policy decisions on mandates or expertise, however, the delegation of power to them loses its legitimacy (Tortola, 2020). Research so far suggests a few factors which may compromise technocratic decision-making. First, the central bank may be insufficiently independent from partisan influences to act in the public interest. Either the central bank enjoys too little independence from the government and implements policy with partisan interests in mind (Ennser-Jedenastik, 2014, 2016; Vuletin & Zhu, 2011). Or its capacity to set monetary policy is constrained by its reliance on the financial sector to achieve its policy goals (Braun, 2017, 2018; Diessner & Lisi, 2020). Second, the preferences of central bankers themselves may be affected by non-technical concerns, such as personal career objectives (Adolph, 2013) or their socialisation into epistemic communities (Verdun, 1999) and pressure to conform to dominant policy paradigms (Gabor, 2011; Johnson, 2016). The last source of central bank politicisation concerns its mandate: insufficiently clear mandates may leave technocrats with too much discretion, while conflicting objectives can result in multiple-mission constraints which turn any policy option inevitably into a political trade-off (Tucker, 2019).

In sum, the literature on central bank politicisation has identified several reasons for central banks’ policy decisions to reflect political interests rather than technical considerations. The international dimension of monetary policy has in this regard, however, not yet been considered. The next section proposes a way to operationalise politicisation in the context of international monetary policy and applies this understanding to ECB cooperation with other central banks in the EU.

The politicisation of international monetary cooperation

International central bank cooperation has a long history (Bordo et al., 2015). Over time these credit lines have served various specific purposes such as stabilising exchange rates under the original European Monetary System (EMS) in the 1980s (James, 2012) or backstopping the global dollar market in 2008 (Tooze, 2018). Each time, the basic idea was that central banks that held or issued international reserve currencies lent these to central banks that needed them to intervene in currency markets or to provide liquidity to banks. This way, the borrowing central banks would temporarily not be constrained by their own level of foreign reserves when they needed to stop a panic. Central bank credit lines offer a complement to International Monetary Fund (IMF) programmes because they can be set up and unwound quickly and informally, potentially in huge volumes, and without the government needing to sign up to conditionality.

The objectives that central banks pursue when they provide these credit lines can be categorised according to three technocratic rationales: interest rates, the exchange rate, and financial stability spillovers (Coeuré, 2019). First, central bank credit lines may be geared at facilitating the transmission of domestic monetary policy and alleviating market stresses. If banks in other jurisdictions struggle to raise funds, they will bid up international interest rates which may spill over and undermine a central bank’s domestic monetary policy targets, especially if it aims to lower interest rates. Second, monetary cooperation can prevent unwanted effects on the exchange rate. If international liquidity is scarce, the reserve currency appreciates with potential deflationary consequences for its issuer. Lastly, central bank swap lines may support financial stability when international market stress disrupts the functioning of domestic money markets or because losses abroad might destabilise domestic financial institutions (European Central Bank, 2014). Several studies have argued that the US Federal Reserve followed these defensive technocratic motivations when it provided swap lines during the GFC (Hardie & Maxfield, 2016; Hardie & Thompson, 2020; McDowell, 2012).

There might also be ‘textbook’ considerations that speak against providing a credit line to a central bank. Two considerations revolve around the issue of trust in the borrowing central bank (Bindseil, 2014, p. 285). A central bank may worry about credit risk and be wary of lending money to a financially unstable state. Or it might fear that credit lines might induce moral hazard and allow the borrowing state to delay necessary adjustments (Moessner & Allen, 2010). A last potential objection concerns the flipside of the interest rate target: if the borrowing central bank intervenes too heavily it might depress domestic interest rates below the lender’s target – a frequent complaint of the Bundesbank under the EMS (Bini Smaghi & Ferri, 2006; Marsh, 1992). Taken together, credit lines issued for technocratic reasons would serve clearly identifiable monetary policy interests of the issuer, but central banks might seek protection when lending to less trustworthy borrowers.

For the central issue of this article, the choice between swaps and repo lines, concerns about interest rate targets and the exchange rate can be expected play a subordinate role. A look at the operational differences between both instruments reveals instead that their main difference lies in the credit risk that the lender accepts. To set up a swap line, two central banks simply open accounts in each other’s names. Whenever the borrower draws the credit line, it must provide the equivalent amount its own currency as collateral on the lender’s account; once it repays the loan, both balances disappear. Sale-and-repurchase (or repo, for short) agreements, by contrast, do not increase the borrower’s foreign reserves because they require the borrower to pledge a foreign asset that they already hold as collateral if they want cash. While swap lines are secured merely on trust that the borrower will repay the loan, a repo agreement protects the lender’s balance sheet because it can seize the collateral asset in case of non-repayment. If a borrower has a weak credit rating, providing a repo, rather than a swap can be justified on technocratic grounds (Bindseil, 2014, p. 285). For the borrower, that choice is consequential because swap lines result in a temporary increase in its foreign reserves and augment its capacity to intervene in markets, whereas repo lines can, at best, allow it to access its existing reserves bit faster.

In some instances, central banks have issued credit lines where clear monetary policy rationales were missing. In these more ambiguous environments, the provision of credit lines can be shaped by various non-technocratic, political considerations. Work on the US Fed has argued that the swap lines in 2008/09 were also motivated by the intention to spare exposed US banks from losses abroad (Broz, 2014) and by diplomatic considerations regarding the recipient countries, such as capital account openness (Sahasrabuddhe, 2019). Absent clear monetary policy rationales, central banks may still provide credit lines if those further a wider national interest.

Another political factor especially relevant in the European context is also the participation in international agreements. Under the EMS, European central banks regularly intervened in support of each other’s currencies according to specified procedures and the adjustment of parities was decided at government level (James, 2012). Still, it can be questioned how closely central banks adhere to these agreements in practice (Cameron, 1993). For instance, in the context of the ERM crisis in 1992, it has been argued that the Bundesbank intervened more heavily in support of the French franc than other currencies in similar distress with an eye to Franco-German diplomatic relations (Bini Smaghi & Ferri, 2006; James, 2012).

The political salience of central bank cooperation during crises and the significance of credit lines for diplomatic relations may also lead to a more politicised decision-making process as the central bank coordinates more closely with the government. The US Federal Reserve, for instance, proactively sought political back-up when it provided swap lines during the GFC and each agreement was approved by the State Department (Harris, 2015; Tooze, 2018). After the GFC, the US Congress revised the Fed’s mandate to have more control over its swap line provision (Broz, 2014). Conversely, the Bundesbank’s extraordinary support measures for the Banque de France in 1992 are widely attributed to external political pressure, including from chancellor Kohl himself, to which it relented (James, 2012, pp. 360–361). While government influence may not be a permanent feature of central bank cooperation, in high-stakes situations, monetary policy considerations may be temporarily subordinated to the wishes of political principals.

To sum up this section, the literature on central banking has identified several monetary policy considerations that might compel a central bank to extend credit lines abroad to meet its domestic policy targets. But central bank credit cooperation is also part of a state’s foreign policy, and as such central banks may have to provide credit lines out of institutional obligation or under external political pressure (Table 1).

Table 1. Technical and political considerations around international central bank cooperation.

Technical considerations	Political considerations
Financial stability	Individual banks’ stability
Interest rates	Diplomatic considerations
Exchange rate	International agreements
Credit risk

The issue at hand: the ECB and central bank cooperation

The ECB’s policy mandate and its statutes provide some general guidelines for international monetary cooperation but leave it with considerable discretion. To start with, though price stability is the ECB’s primary target, one of its secondary objectives is to ‘contribute to financial stability’ (Art. 127.5, TFEU). The ECB can freely determine its external policy conduct by buying and selling ‘all types of foreign exchange assets’ and conduct ‘all types of banking transactions in relations with third countries and international organisations’ (Art 23, TFEU, Prot. IV). One constraint that might matter in this context is, however, the requirement to protect the ECB’s balance sheet by demanding adequate collateral (Art 1, TFEU, Prot. IV), a provision which the ECB has usually interpreted based on the credit rating of an asset by independent agencies (Cheun et al., 2009; Gabor & Ban, 2016).

The ECB lacks a direct counterpart to a foreign ministry which could instruct it how to treat different Euro Area outsiders. The relevant passages in the Treaties (Art. 138 and 219, 1.-3. TFEU) merely stipulate that the Council of the EU sets the overall priorities of exchange rate policy and concludes formal agreements. An initial power struggle with the Council was resolved in the Turku Agreement of 1999, where it was agreed that the ECB would de facto have sole competence for foreign exchange interventions (Henning, 2007). In addition, the members of the ECB Governing Council are barred from taking any formal instructions by virtue of Art. 130 TFEU – though they may still be attentive to domestic political sentiments (Moschella & Diodati, 2020). Against this backdrop, it seems unlikely that the ECB would follow any external orders when setting the borrowing terms for central banks.

But this does not mean that the relations of the ECB with the other central banks in the EU are free from political considerations. Since the Lisbon Treaty, the ECB is officially a supranational EU institution (Hodson, 2011, chap. 2) and all member states, inside or outside the Euro Area, own a share in its capital. Euro Area outsiders are excluded from the ECB’s decision-making organs, but they participate in a variety of technical committees in Frankfurt, as well as the ECB’s General Council (Umbach & Wessels, 2009). The ECB also reviews central bank legislation in all EU member states.

Moreover, the ECB is one of the actors involved in advising on the euro adoption process. Alongside the European Commission, it assesses whether a member state fulfils the conditions to adopt the euro before that is officially decided by other institutions (Art. 140 and Protocol 13, TFEU). The process of euro adoption gives the ECB an unquestionably political role. Many authors have examined the question why some countries have been keener to join the Euro Area than others (Dandashly & Verdun, 2018; Dyson, 2006; Epstein & Johnson, 2010; Johnson, 2008). Those member states that aspire to adopt the euro first join an exchange rate agreement with the ECB called the Exchange Rate Mechanism II (ERM II).¹ There they commit to keeping their exchange rate within the range of ±15per cent (±2.25 per cent in the case of Denmark) around a central value against the euro for at least two years. In return, the intergovernmental agreement on the ERM II includes not just a ‘Very Short-Term Financing Facility’ (which is in effect a swap line) but even a provision under which the ECB intervenes in potentially unlimited amounts if a currency threatens to devalue against the euro (Deutsche Bundesbank, 1998).

Taking stock, the ECB confronts two sets of considerations when deciding whether it should provide a swap line or a repo to another EU central bank. Its mandate does not impose any legal constraints on it and should allow it to grant credit lines, at least whenever necessary to ensure Euro Area objectives. However, the ECB may decide that the borrowing state is not creditworthy enough, and therefore, demand better collateral for ‘technocratic’ reasons. Conversely, it seems unlikely that the ECB decision could be directly influenced by the Council of Ministers. Still, the ECB might decide to take account of political factors related to European integration, given its role as a supranational EU institution and its position within the euro adoption process.

The ECB’s credit lines in 2008 – domestic stability first

In its early years, the ECB did not consider the international role of the euro to be an important consideration when determining its policy. Officially, it maintained a ‘neutral’ stance, arguing that currency internationalisation should be a ‘market-driven process, not to be steered by central banks or by political bodies’ (Duisenberg, 2000). This doctrine reflected the strong intellectual legacy of the Bundesbank’s orthodox views of monetary policy (Marsh, 1992) which the former Bundesbank officials Ottmar Issing and Jürgen Stark carried into the ECB’s Economics Department when they became Chief Economists (Lebaron, 2013). International developments would only become relevant for the ECB if they threatened its mandate for price stability. Just once, in autumn 2000, had the ECB intervened in currency markets for that purpose, targeting the euro/dollar exchange rate (Henning, 2007). Yet the ECB was not entirely oblivious to the potential need for international liquidity. After the 9/11 attacks, it set up a temporary swap line with the US Fed which it used to provide dollar liquidity to Euro Area banks. In the following years, it also put into place some precautionary swap agreements with other G10 central banks, including the Bank of England, where it would stand ready to provide euros (Interview ECB Board Member 2). Still, according to an external evaluation the ECB devoted too few resources to understanding the international financial system prior to the GFC (Freedman, 2011).

Among the credit lines that the ECB provided during the GFC, there is a clear geographic distinction between the Nordic and East European central banks. On the one hand, the central banks of Sweden and Denmark could both borrow under swap lines. The ECB considered the Swedish request for a swap arrangement in line with the existing precautionary agreements with G10 central banks (European Central Bank, 2007). The €10bn swap agreement between the ECB and the Swedish central bank was signed in December 2007, but both central banks agreed to keep it secret both from market participants and other central banks (Leung, 2020). The swap line with the Danish National Bank in October 2008, by contrast, was immediately made public and the Danish National Bank used it to provide euro liquidity to domestic banks (Danmarks Nationalbank, 2009). Both these credit lines were uncontroversial inside the ECB (Interviews Papadia, ECB Board Members 1, 2) – the Executive Board did not even seek formal approval from the Governing Council before concluding the swap agreement with the Swedish central bank (European Central Bank, 2007).

The ECB response to the requests from the East European member states was, however, more reluctant. When money markets in Hungary collapsed on 10 October, both the Hungarian and the Polish National Banks reached out to the ECB for swap lines to backstop foreign exchange markets and shore up confidence. The ECB first handled the more pressing Hungarian request, but only offered a repo arrangement which set the precedent for the credit lines with the other two East European central banks. The central bank of Latvia two weeks later agreed upfront to accept the same borrowing terms as Hungary (Rimšēvičs, 2008). The ECB decision to offer a repo line to the National Bank of Poland explicitly followed the precedent set by the credit line to the Hungarian National Bank (European Central Bank, 2008b).

The question of which central bank would receive which borrowing terms was a contested issue inside the ECB. When the crisis broke out in Europe, the ECB had no policy in place on how to respond to requests for liquidity assistance from Euro Area outsiders – it only agreed on a set of principles for liquidity provision on 20 October (European Central Bank, 2008b). In practice, though, the ECB devised its approach towards handling the requests for swap lines ad hoc (Interview Papadia). Initially, it was divided on how to respond to the requests from Hungary and Poland. On the side of the ECB staff, the Market Operations and International Departments were ready to agree to swaps. But they were met with opposition from the Economics Department – which advanced the ‘very German argument’ (Interview Nowotny) that the central bank’s balance sheet needed to be protected (Interviews Nauschnigg, Papadia). In the Governing Council, the Austrian National Bank, supported by Banque de France, unsuccessfully advocated swap lines for the Hungarian National Bank (Vallee, 2010; Interviews Nauschnigg, Nowotny). The Executive Board proposal to offer a repo struck a balance between offering some support and protecting the ECB against financial losses.

Officially the ECB has argued that it considered a variety of factors when setting the credit terms (European Central Bank, 2014). It required the East European central banks to show that there were exceptional market dysfunctions and document which measures they had already taken. But, as the then-Director for Market Operations explained, the ‘basic reason for doing repos instead of swaps was credit risk’ (Interview Papadia). Indeed, not only did both Sweden and Denmark have AAA sovereign credit ratings, Sweden’s participation in the G10 lent it additional credibility and Denmark was trusted thanks to its track record of cooperation inside the ERM II (Interviews ECB Board Members 1, 2). By contrast, Hungary and Latvia both required an IMF bailout and were seen as far less stable. Only the repo for Poland sits a bit uneasily with that story. On 15 October 2008, the ECB lowered its collateral standards to accept even assets rated BBB- as collateral on its balance sheet. Poland was rated A-, far above that and solidly investment grade. The ECB’s decision to offer just a repo nevertheless (despite heavy complaints by the Polish National Bank) reflected primarily the ECB’s intention to not treat Poland better than Hungary (Interview Nowotny; European Central Bank, 2008b).

At the same time, there are few indications that the ECB saw any of the credit lines as supporting domestic policy objectives. The financial crisis in the European periphery was a side issue compared with the funding stresses that the ECB confronted in the Eurodollar market (Interview Papadia; cf. Hardie & Thompson, 2020). While the Swedish krona, Hungarian forint, and Polish zloty all depreciated heavily in early 2009, the exchange rate was no central policy concern for the ECB, as the dollar and the Swiss franc appreciated against the euro at the same time (Bini Smaghi, 2009).

Moreover, the ECB remained surprisingly calm about financial instability spillovers from Eastern Europe. Major shares of the banking markets in Poland and Hungary – as in most other countries in the region – were controlled and funded by a handful of West European banking groups (Árvai et al., 2009; Mitra et al., 2009). However, the ECB, did not consider the risks from this exposure to be a threat to the stability of the Euro Area as a whole and financial supervision was still a member state’s competence at the time (Interview ECB Board Member 2).

The national banking supervisors, for their part, expressed worries about contagion. Given the huge regional exposure of Austrian banks, the Austrian National Bank was especially forceful in advocating swap lines for Hungary and Poland because it was worried that financial instability might spill over to the parent groups in Austria (Nauschnigg & Schieder, 2011). Even the Bundesbank, which had still reacted with reservation after the Hungarian request, supported a swap for the Polish central bank with a view to German investments in Poland (Interviews Nauschnigg, Nowotny). The ECB backstopped exposed banks by supplying the parent banks with liquidity in euros and Swiss francs (which it obtained through the swap with its Swiss colleagues). The banks could then forward this liquidity to their operations in Eastern Europe (Pann et al., 2010). These measures were deemed sufficient to ensure financial stability of the Euro Area as a whole and there was no reason for the ECB to take further steps.

The ECB likewise rejected calls from exposed Euro Area banks themselves to step up support for Eastern Europe. In late 2008 the largest Euro Area banking groups active in the region coordinated to set up an initiative to ensure support for financial stability – the so-called Vienna Initiative (Pistor, 2011). In January 2009, the CEOs of ten major EU banks wrote a joint letter in which they proposed several concrete measures as to how the ECB could support financial market conditions in the new member states, including swap lines and accepting local currency bonds as collateral (the letter is reproduced in EIB, 2019, p. 238). But this was seen as the problem of individual banks, rather than a threat to the stability of the Euro Area (Interview ECB Board Member 2). A few days after the letter, ECB Board member Yves Mersch responded that the ECB did not have ‘a mandate to be a regional United Nations agency’ and ‘cannot be a small god for everyone and for everything’ (Atkins & Wagstyl, 2009). The ECB could not be swayed by private interests.

A last surprising finding concerns the way the ECB handled the two ERM II participants Denmark and Latvia. Although the ERM II provided assistance facilities for crisis situations, both credit lines were legally outside the ERM II framework. In fact, the ECB insisted that the credit lines only be drawn for purposes of liquidity provision to banks and not to defend the exchange rate (European Central Bank, 2008a; González-Parámo, 2008). For Denmark, that was exactly what the central bank needed to do (Danmarks Nationalbank, 2009), but the ECB rejected the Bank of Latvia’s request to stabilise the exchange rate (González-Parámo, 2008) even though the devaluation risk was the major problem of the Latvian crisis (Åslund & Dombrovskis, 2011, p. 47).

The ECB had never had any intentions of supporting the currency pegs in the Baltic countries. When Latvia joined the ERM II in 2005, its central bank continued the 1 per cent fluctuation band against the euro, instead of the customary 15 per cent and the ECB insisted that this remained a unilateral commitment which placed no obligations on the ECB (European Central Bank, 2005; Interview Papadia). In November 2008, the Executive Board noted in an internal document that ‘the ECB’s policy line has always been that currency board or unilateral pegs by third countries are not backed in any way by policy commitments from the ECB’ (ECB Executive Board, 2008). In late 2008, the Bank of Latvia struggled to fend off a speculative attack against the currency, which would have delayed its euro adoption by years (Åslund & Dombrovskis, 2011, p. 47). But though the ECB argued that Latvia should defend the currency peg, it was unwilling to provide a swap line to support the central bank since that would have contradicted the disciplinarian logic of the ERM II (Interviews IMF official, ECB Board Members 1, 2).

The ECB’s approach towards central bank credit lines during the GFC conforms to a narrow technocratic reasoning. It set the borrowing terms based on its mandate for the Euro Area, sidestepped institutional obligations under the ERM II, and resisted political pressure from banks and governments. Its main concern in 2008 was sovereign credit risk: the Swedish and Danish central banks were seen as reputable long-standing partners, but when lending to East European central banks, the ECB sought protection for its balance sheet. In the aftermath of the crisis, the ECB has been criticised for failing to support the EU’s East European member states and to assume responsibility for the international role of the euro (e.g., Åslund, 2010; Gabor, 2016; Tooze, 2018). While there may have been a technocratic reasoning behind the ECB’s decisions, the outcome surely resulted in political disappointment among the EU’s East European member states.

The ECB’s credit lines in 2020: more for more

In the years prior to the COVID-19 crisis, the ECB’s stance on the international role of the euro and central bank cooperation underwent three profound changes. First, swap lines between central banks had become an established policy tool since 2008 (Helleiner, 2014; Mehrling, 2015) and the ECB was more familiar with these instruments. In 2011, the ECB and five other leading central banks had converted the informal credit lines from the early 2000s into a system of standing swap lines (Albrizio, 2021). Second, the ECB had re-assessed the costs and benefits of currency internationalisation and its staff had concluded that the international financial system had changed profoundly since 2010. Importantly, it now recognised the ‘exorbitant duty’ of providing liquidity during international financial crises and discussed the link between the availability of swap lines and the international attractiveness of a currency (European Central Bank, 2019). And third, when the Council of the EU appointed Philip Lane as Chief Economist and former IMF Director General Christine Lagarde as ECB President in 2019, these decisions were seen as indications that the ECB might pursue a more dovish and politically calibrated course in the future (Jones, 2019). The ECB’s new approach dovetailed with a more assertive general political climate in the EU and the call by the European Council in late 2019 for a stronger international role of the euro (Council of the European Union, 2019).

The COVID-19 panic in March 2020 triggered a spike in demand for international euro liquidity. First, investors reduced exposures to risky assets in a ‘flight to safety’, but later they needed to sell even safer assets to get hold of liquidity, in an episode dubbed the ‘dash for cash’ (FSB, 2020). Early into the crisis, the ECB and the Danish National Bank reactivated their previous swap line, though its volume was now doubled to €24bn. Soon after, in mid-April 2020, the central banks of Bulgaria and Croatia received swap lines of up to €2bn each. Over the summer, the ECB then provided repo lines to the Hungarian and Romanian central banks – once more it had offered better borrowing terms to some central banks than to others.

The technocratic considerations that the ECB had weighed before do not provide a clear explanation for why Bulgaria and Croatia should receive better borrowing terms. Their sovereign credit ratings (BBB and BBB-) were the same as those of Hungary and Romania, respectively. Besides, the central banks of Bulgaria and Croatia had higher foreign reserve levels than their Hungarian and Romanian colleagues (Lukinić Čardić, 2020). In terms of credit risk, it is not apparent why Hungary or Romania should be less trustworthy borrowers.

The first two columns in Table 2 show that most banking systems had massively reduced their reliance on euro funding from abroad since 2008. Out of the four East European member states, Hungary was the only one remaining with a net negative cross-border position in euros. In the other three countries, the banking sectors were international net creditors in euros and had solid liquidity buffers. Thus, if any country could potentially have disrupted monetary policy conduct in the Euro Area, it was none of the swap recipients. But instead, the Hungarian central bank had to piece together repo lines from the ECB, the Bank for International Settlements, and the US Fed to insure itself against liquidity risks over the summer of 2020 (Magyar Nemzeti Bank, 2020, p. 53).

Table 2. Cross-border risks for the Euro Area for selected Euro Area outsiders.

		Domestic banks’ net cross-border position in euro, USD m		Share of total assets of foreign banks		Financial exposure of largest 3 Euro Area creditors, USD m
		2008	2019	2008	2020	2008	2020
Member State	Sweden	−42,517	−3553	/	/	79 512	126,556
	Denmark	−38,013	−6882	/	/	126 302	107,506
	Hungary	−18,397	−2493	61.20%	43.50%	58 328	13,392
	Latvia^a	−16,286	/	65.90%	66.10%	8 850	/
	Poland	−290	−829	72.50%	46%	53 737	54,539
	Croatia	−588	989	90.60%	91.10%	35 289	10,694
	Bulgaria	−6581	2908	83.90%	77.70%	15 609	5681
	Romania	−28,865	3674	88.50%	73.60%	48 246	17,972

Sources: ECB Statistical Data Warehouse, CBD2.A.?.W0.11._Z._Z.A.A.I0002._Z._Z._Z._Z._Z._Z.PC; BIS locational banking statistics, Tables 6.1 and 6.2 (author’s calculations)

^a Foreign bank ownership in Latvia comes primarily from Swedish Banks.

Similarly, direct cross-border exposures from the Euro Area to Eastern Europe have been reduced as banks have relied more on local funding and less on their parent banks. Levels of foreign bank ownership have remained high except in Hungary, where the government has pursued an aggressive course of ‘financial nationalism’ after 2010 (Ban & Bohle, 2021; Johnson & Barnes, 2015; Oellerich, 2019). The rightmost columns in Table 2 document that Croatia and Bulgaria in 2020 represented only a small, insignificant financial risk to Euro Area. From a financial stability perspective, they would have been the least likely candidates for swap lines.

Instead, a more political reasoning around the prospect of euro adoption seems to have been decisive. Both Bulgaria and Croatia had started discussions about adopting the euro with the ECB and the European Commission back in 2018 (Dorrucci, 2020). They had applied to be included in the ERM II in 2019, fulfilled the technical convergence criteria, and were on track to join the ERM II in autumn 2020. Their applications had been held up by the ECB and the Commission due to concerns over the quality of financial supervision, as joining the ERM II nowadays entails becoming part of the Banking Union (Krasimirov, 2018). Both countries had already been operating rigid currency pegs against the euro, and the deutschmark before that.

But during the financial market panic in March 2020, the Croatian National Bank was suddenly faced with a speculative attack against its currency, the kuna, and was forced to spend €2.25bn to defend the exchange rate. In this context, the ECB’s swap line offered invaluable support in restoring financial stability in Croatia and protecting the central bank from having to devalue so shortly before entering the ERM II. Though two weeks passed between the Governing Council’s agreement to establish the facility and the press release (European Central Bank, 2020a), when the swap line for Croatia was eventually announced, the speculation against the currency stopped immediately (International Monetary Fund, 2021). The swap line for Bulgaria was both agreed and announced a week later, as a purely precautionary instrument, since there was no speculative attack against the Bulgarian leva. Both swap lines were instead supposed to signal to market participants that the ECB stood behind these currency pegs.

The swap announcements also stood out in style. The way in which the ECB communicated them leaves little doubt that these countries’ planned euro adoption was the crucial factor. In both press releases, the ECB mentioned that the recipient was preparing to participate in the ERM II which, it stressed, ‘is a prerequisite for a Member State to join the euro area’ (European Central Bank, 2020b, 2020c). According to Croatian central bank Governor Vujčić ‘even the way the announcement was phrased by the ECB […] was a very clear signal to the markets that the firepower of the Croatian National Bank is now much larger’ (Bank of Albania, 2020, p. 3:13:00).² The swap for Bulgaria was announced on the same day that ECB President Lagarde held a publicised phone call with prime minister Borisov where they discussed the prospect of euro adoption (BNT, 2020). Clearly, these announcements were geared at making an impression on market participants.

In July 2020, the ECB and the Commission put aside their remaining concerns and admitted both Croatia and Bulgaria into the ERM II – ahead of the original schedule. The concerns about financial supervision have not been resolved, but the institutions in effect accepted the Croatian and Bulgaria’s promise to do better (Dorrucci et al., 2020; European Commission, 2020). Since October 2020, the ECB directly supervises the major banks in both countries. For market participants, the accelerated ERM II and Banking Union entries were further steps to boost confidence and have brought potential upgrades of their credit ratings back on the cards (Tataru & Pham, 2020).

The repo lines for the central banks of Hungary and Romania appear plausible based on the same political logic. For them, ERM II entry remains a more remote prospect. The Romanian government has ambitions to enter the ERM II around 2024 and it remains doubtful whether the Hungarian government ever wants to adopt the euro (Arató et al., 2021). But given that the ECB in 2020 also provided repo lines to several central banks in the EU’s vicinity (such as the Albanian one), the repos for two EU members seem a bit out of line. Most of the other repo recipients had sub-investment grade sovereign credit ratings.

In summary, it seems that the ECB in 2020 did not set the borrowing terms for Euro Area outsiders based on technocratic considerations regarding its monetary policy mandate. The ECB’s swap lines for Bulgaria and Croatia are remarkable, not just because they were provided to aspiring ERM II participants despite their weak sovereign credit ratings, but also because the announcements during the crisis were aimed at restoring market confidence outside the Euro Area. To shepherd Bulgaria and Croatia into the ERM II, the ECB, together with the EU Commission and the Council, had to put aside technical details for wider political considerations. While the ECB has justified its more proactive extension of credit lines in 2020 based on the attractiveness of the euro as an international currency, it has so far not presented an official rationale for the discrimination between EU member states (Panetta & Schnabel, 2020). Its refusal to also provide swaps for Hungary and Romania suggests that it set credit conditions in line with institutional considerations around euro adoption.

Conclusion

This article started with the question whether ECB decisions on borrowing terms for Euro Area outsiders have become politicised between 2008 and 2020. It has first applied a preference-based definition of politicisation to the context of international central bank cooperation. The principal technocratic consideration regarding the ECB’s choice between swaps and repos concerns credit risk; decisions that consider other factors suggest that the ECB has acted as a more politicised agent. The two case studies have found that the ECB in 2008 conforms closely to the image of a technocratic agent and set tougher credit terms for countries with weaker credit ratings. Its swap lines in 2020 for Bulgaria and Croatia seem more motivated by an institutional interest in Euro Area enlargement, clearly a more political motivation. It contrasts with its behaviour in 2008 when it did not do the same for the Baltic states that were in a similar position. This finding suggests that the ECB has indeed over time become a more political actor in the field of international central bank cooperation and behaved more as a supranational institution in the EU context.

A logical follow-up to this conclusion is to ask which factors have changed in the twelve years between the two crises that might explain this politicisation. The argument advanced here is that the contexts of the two crises are comparable and that, by elimination, it might be more about the agency that the ECB exercised. In fact, based on structural conditions, one might – if anything – have expected the ECB to be more forthcoming in 2008 than in 2020: Hungary and Latvia had the same credit ratings as Bulgaria and Croatia; during the GFC bilateral financial exposures were considerably bigger (see Table 2), which led major Euro Area banks to pressure the ECB to provide more support; and Latvia’s failure to defend its currency peg would not only have derailed its own immediate plans for euro adoption, but would likely have sparked financial contagion across the Baltic region with reputational drawbacks for the EU. And yet, the ECB remained untouched by these concerns, focusing on the issues of Euro Area financial stability and potential risks to its balance sheet. In 2020 by contrast, the ECB provided swap lines to countries outside the ERM II framework that posed no conceivable risk to Euro Area financial stability. What had changed was not the choice before the ECB, but how it weighed these conflicting considerations.

In line with Tortola’s (2020) preference-based approach towards central bank politicisation employed here, this article argues that what changed instead was more likely the ECB’s discretionary choices as an actor navigating conflicting imperatives. The findings suggest that the ECB’s position regarding developments outside the Euro Area might have changed because of internal ideological contestation and incremental learning among its staff (Ban, 2015; Chwieroth, 2008; Ferrara, 2020). The Bundesbank-inspired doctrine of a ‘neutral stance’ towards the international role of the euro, which was prevalent in the early years of the ECB, has given way to a softer and more proactive stance in late 2019 as the ECB re-assessed the costs and benefits of currency internationalisation in broader terms (European Central Bank, 2019). Similarly, the appointment of the ‘dove’ Philip Lane as Chief Economist in 2019 contrasts with his more hawkish predecessor Jürgen Stark. Overall, the ECB’s justification of central bank credit lines in 2020 in terms of supporting the international role of the euro (Panetta & Schnabel, 2020) speaks to a wider trend where the ECB has increasingly considered reasons beyond its narrow policy mandate to legitimise its actions (Moschella et al., 2020).

On a final note, the politicisation of the ECB’s credit lines might have wider implications for the role of the ECB in the EU as a political system and for its legitimacy. This article has argued that the ECB needs to navigate a multiple-mission constraint in its cooperation with Euro Area outsider central banks (Tucker, 2019). Its formal independence leaves the ECB without an executive counterpart that could provide instructions or political backing on how to balance its mandate to support financial stability in the Euro Area and protect its balance sheet with the demands placed on it given its role as a supranational institution (Papadia, 2013). The comparison of both crises has shown that the ECB’s technocratic approach in 2008 has drawn considerable criticism for failing to take responsibility for the international role of the euro and smaller member states. Its preferential treatment for Bulgaria and Croatia in 2020, however, was swift and politically uncontroversial. Based on this, future research might investigate whether such a more politically considerate approach towards international cooperation during financial crises does really compromise the technocratic legitimacy of the ECB, as the preference-based framework of central bank politicisation implies. If monetary policy and European integration are hard to disentangle, the ECB might in the end benefit from showing that it can look beyond the immediate interests of the Euro Area.

Acknowledgements

The author is grateful for the suggestions by Matt DiGiuseppe, Tom Theuns, Amy Verdun, Dustin Voss, as well as the anonymous reviewers. Previous drafts of this paper were presented at the Leiden Political Economy PhD Seminar, the NKWP politicologenetmaal, the ECPR SGEU conference, the UACES Graduate Forum, and the CES conference 2021. The author acknowledges financial support as a UACES Scholar 2021.

Disclosure statement

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

Additional information

Notes on contributors

Lukas Spielberger

Lukas Spielberger is a PhD candidate at Leiden University, the Netherlands.

Notes

1 Denmark participates in the ERM-II but has an opt-out of the Treaty requirement to join the euro.

2 The author thanks Piroska Nagy for this reference.