Euro area reform cannot ignore the monetary realm

Jérémie Cohen-Setton, Shahin Vallee 20 June 2018



This column is a lead commentary in the VoxEU Debate "Euro Area Reform"

The authors of CEPR Policy Insight 91 (Bénassy-Quéré et al. 2018) are right to argue that the euro area needs an alternative anchor than the current system of fiscal rules and financial penalties to discipline fiscally wayward members. But by refusing to complement their proposals with recommendations in the monetary realm, they miss an opportunity to provide a balanced reform package that would not only increase fiscal discipline and risk sharing, but also enhance liquidity provision. The responsibilities and operating procedures of the Eurosystem cannot stay outside of debates about the future architecture of EMU. 

The view that the proposals in Bénassy-Quéré et al. (2018) would increase rather than decrease “redenomination risks connected to acute liquidity and credibility crises” (Buti et al. 2018) results in part from the authors’ silence on the role of the ECB and national central banks in the provision of liquidity when needed. That 14 economists would have different views on the appropriate stance of monetary policy and their preferred choice of instruments is not surprising (Farhi and Martin 2018). But there should be broad agreement on some basic principles about ways to improve the lender of last resort function in a reformed EMU.

Principles for improving the Eurosystem

In our view, the following four principles should guide such rethinking:

  1. The central bank of a monetary union should be able to backstop financial markets in sovereign bonds if it considers that panicked investors are threatening the integrity of the single currency (Eichengreen 2016). Being an unconditional lender of last resort for solvent governments (Mody 2014) – instead of making the activation of Outright Monetary Transactions (OMT) conditional on a European Stability Mechanism (ESM) programme – necessarily requires drawing a difficult line between liquidity and solvency. If this judgement is completely left to governments, it creates a risk of fiscal dominance. If it is completely left to the ECB, it creates a risk of monetary dominance. If it is shared between several institutions, it creates risks of indecision, delays and uncertainty. Claeys and Goncalves Raposo (2018) recently proposed relying on a debt sustainability analysis, which would be done by the ESM, to determine eligibility of government bonds in refinancing operations. Under the current intergovernmental structure of the ESM with unanimity, this approach seems unsuitable for OMT. But options along these lines should be explored to improve predictability of the sovereign backstop and to remove the ECB from a position where its actions can be interpreted as attempts to discipline governments.1
  2. Monetary policy and liquidity operations should be fully centralised and mutualised. That “remnant[s] of national sovereignty in monetary policy” (Draghi, 2018) still exist in the risk-sharing arrangements of the Eurosystem governing the provision of emergency liquidity (Emergency Liquidity Assistance), some aspects of its collateral framework (Additional Credit Claims), and its most important asset purchases programme (the Public-Sector Purchase Program) is troubling and should be removed. The renationalisation of these policies has limited their effectiveness and increased redenomination risks by suggesting that there could be “conditions […] in which one [national central bank] would be reluctant to build up claims indefinitely on some other national central bank” (Kenen 1999).
  3. Balance sheet considerations should not constrain a central bank’s ability to perform its monetary and lender of last resort functions. The current ECB doctrine on financial independence should be revised and replaced by a Fiscal Carveout,as suggested byTucker (2018), which would specify how losses would be covered by the fiscal authority. Alternatively, it could be replaced by a more blanket statement of support from European governments for ECB actions and the potential risks associated with them. This would make clear that recapitalisation arrangements matter not because the notion of central bank solvency makes sense, but because they reveal the sovereign’s political support of the monetary authority. 
  4. The operational framework should not reinforce doom loops. Since 2005, the ECB has relied on ratings made by private credit-rating agencies to determine eligibility and haircuts. This reliance has led to abrupt swings in haircuts that have been destabilising (Claeys and Goncalves Raposo 2018). To counter some of the built-in procyclicality of its collateral framework, the ECB has often relaxed these constraints ex post by introducing waivers, lowering haircuts, and expanding the collateral pool (Cohen-Setton et al. 2013, Wolff 2014). A more robust, predictable and countercyclical framework could, however, be designed to deal with these problems ex ante rather than ex post. It would also help avoid the risk of the ECB appearing politicised in its eligibility decisions.

How the Great Depression changed the Federal Reserve System

As we argue in a recent paper (Cohen-Setton and Vallée 2018), these principles draw on important lessons from the history of the US Federal Reserve System, whose initial design was also problematic and was only corrected after its failures became obvious during the Great Depression.

Because the Federal Reserve Act of 1913 created a decentralised system of 12 regional Federal Reserve Banks (FRBs) owned by commercial banks in their respective districts, risk-sharing of profits and losses linked to monetary operations did not occur automatically. In fact, what was still a decentralised system of regional central banks with 12 individual balance sheets only became de facto unified when the federal government eventually clarified in 1933 that FRBs – if faced with important losses – would not have to be recapitalised by the member banks of their respective districts but by the US Treasury.

In the words of President Franklin D. Roosevelt, “it [was] inevitable that some losses may be made by the FRBs in loans to their member banks […] [but] there is definitely an obligation on the Federal Government to reimburse the 12 regional FRBs for losses which they may make on loans made under these emergency powers. I do not hesitate to assure you that I shall ask the Congress to indemnify any of the 12 FRBs for such losses” (Silber 2009). 

In fact, the preceding weeks had illustrated the extent to which autonomous regional FRBs could behave in uncooperative manners to protect their own reserves leading (Eichengreen 1992). In early 1933, a speculative attack against the New York Fed’s gold reserves had led to a reduction of its gold ratio towards the statutory limit. In refusing the New York Fed’s request to rediscount government securities, the Chicago Fed precipitated a cascade of state bank holidays that culminated with the National Banking Holiday (Wigmore 1987).

Only when regional monetary autonomy was eliminated,2 and when the compact between the central bank and its sovereign was clarified, did regional considerations eventually became subordinated to national ones. Together with these institutional changes, technical improvements in its operational framework were also critical in transforming theFederal Reserve into a fully-fledged lender of last resort. Recognising, for example, the procyclical bias of a collateral framework that “require[ed] substantial amounts of excess collateral” (McKinney 1960) against the deterioration of economic activity, the Federal Reserve Board adopted a new regulation specifying that credit be instead extended liberally “at times when the value of assets held by banks may be decreasing because of a downward turn in the nation’s business and a decrease in the national income” (Board of Governors of the Federal Reserve System 1937).

The initial design of the Eurosystem is stronger than that of the initial Federal Reserve System. The ECB has also clearly moved some distance to complement its framework and enhance its lender-of-last-resort function and to provide some level of backstop to the sovereign bond market. Yet this remains incomplete and the renationalisation of some aspects of its operational framework have generated new institutional risks that need to be addressed in the context of reforms of the architecture of the euro area. There can be no such thing as the completion of the monetary union without profound reforms in the conduct of monetary operations and its interactions with fiscal and political authorities.

Authors’ note: The authors thank Madona Devasahayam, Adam Posen, and Jeromin Zettelmeyer for useful feedback.


Bénassy-Quéré, A, M Brunnermeier, H Enderlein, E Farhi, M Fratzscher, C Fuest, P-O Gourinchas, P Martin, J Pisani-Ferry, H Rey, I Schnabel, N Véron, B Weder di Mauro, and J Zettelmeyer (2018), “Reconciling risk sharing with market discipline: A constructive approach to euro area reform”, CEPR Policy Insight No. 91. 

Board of Governors of the Federal Reserve System (1937), “Regulation on Discounts, by Federal Reserve Banks”, Federal Reserve Bulletin, October: 15-17.

Buti, M, G Giudice, and J Leandro (2018), “Deepening EMU Requires a Coherent and Well Sequenced Package”,, 25 April.

Claeys, G and I Goncalves Raposo (2018), “Is the ECB Collateral Framework Compormising the Safe-Asset Status of Euro-Area Sovereign Bonds?”, Bruegel Blog, 8 June.

Cohen-Setton, J, and S Vallée (2018), “Federalizing a Central Bank: A Comparative Study of the Early Years of the Federal Reserve and the European Central Bank”, in J F Kirkegaard and A S Posen (eds), Lessons for EU Integration from US History, Report to the European Commission.

Cohen-Setton, J, E Monnet and S Vallée (2013), “Deconstructing the ECB collateral framework: assessment and outlook”, ECB Watchers, 24 July.

DeGrauwe, P (2011), “The European Central Bank as a Lender of Last Resort”,, 18 August.

Draghi, M (2018), Committee on Economic and Monetary Affairs Monetary Dialogue, European Parliament, 26 February.

Mody, A (2014), “The ECB’s Bridge Too Far”, Project Syndicate, 11 February.

Eichengreen, B (1992), “Designing a Central Bank for Europe: A Cautionary Tale from The Early Years of the Federal Reserve System”, in M B Canzoneri, V Grilli, and P R Masson (eds), Establishing a Central Bank: Issues in Europe and Lessons from the U.S., Cambridge University Press.

Eichengreen, B (2016), “The European Central Bank: From Problem to Solution”, in The Search for Europe: Contrasting Approaches, BBVA Group.

Farhi, E, and P Martin (2018), “The Role of the ECB in the Reform Proposals in CEPR Policy Insight 91”, VoxEU, 19 April.

Kenen, P B (1999), “The Outlook for EMU. Eastern Economic Journal”, Eastern Economic Association 25(1): 109-115.

McKinney, G W (1960), The Federal Reserve Discount Window: Administration in the Fifth District, Rutgers University Press.

Silber, W S (2009), “Why Did FDR’s Bank Holiday Succeed?”, Economic Policy Review 15(1).

Tucker, P (2018), Unelected Power, Princeton University Press.

Tucker, P (2016), “The political economic of the central bank balance sheet management”, Speech at Columbia University, May.

Wigmore, B A (1987), “Was the Bank Holiday of 1933 Caused by a Run on the Dollar?”, Journal of Economic History 47(3): 739–55.

Wolff, G B (2014), “Eurosystem Collateral Policy and Framework: Was it Unduly Changed?”, Bruegel Policy Contribution 204/14.


[1] For this same reason, we think that the ECB should no longer participate in troika programmes.

[2] Between 1913 and the New Deal reforms of the Fed regional Feds enjoyed autonomy in setting discount rates, inparticipating in open market operations and in their lender of last resort policies.



Topics:  EU institutions EU policies

Tags:  euro area reform, VoxEU debates, lender of last resort, ECB, Federal Reserve

Research Fellow, Peterson Institute for International Economics

Former economic policy advisor, Economist, PhD candidate at the LSE European Institute


CEPR Policy Research