key: cord-0062510-gh57i4f7 authors: Bateman, Will title: The Law of Monetary Finance under Unconventional Monetary Policy date: 2021-04-21 journal: Oxf J Leg Stud DOI: 10.1093/ojls/gqab008 sha: efbd9cd76d7bbab558506551346bb603f4b5bf98 doc_id: 62510 cord_uid: gh57i4f7 Monetary finance (money creation by central banks to fund public expenditure) is a high-profile part of economic, political and policy debates concerning the legitimacy of central banks in liberal economies and democracies. This article makes a distinctively legal contribution to those debates by analysing the legal frameworks governing monetary finance in three prominent central banking systems between 2008 and 2020: the Federal Reserve System, the Eurosystem and the Bank of England. It begins by explaining the law governing central bank and national treasury relations in the United States, the EU and the UK. It then examines how that law operated under the unconventional monetary policies adopted by central banks in response to the financial crisis and the COVID-19 pandemic. The article concludes by reflecting on the challenges monetary finance presents to the sui generis position of central banks in the liberal constitutional order. The virtues and vices of monetary finance (financial support to public sector agencies provided by central banks) are deeply contested topics in modern central banking and economics. Prominent voices have spoken in favour of monetary financing, particularly during times of crisis and economic underperformance, 1 while others have urged extreme caution. 2 Interest in the *Associate Professor, Law School, Australian National University. Email: will.bateman@anu.edu.au. The author acknowledges the support and inspiration provided by numerous people, in commenting both on this manuscript and on related research projects on central banks and public finance: Professor Ramu de Bellescize (Rouen), Professor Peter Cane (Cambridge), Professor David Feldman (Cambridge), Professor David Howarth (Cambridge), Professor Rosa Lastra (QMUL), Professor Tony Prosser (Bristol), Dr Jens van t' Klooster (KU Leuven), Dr Nik de Boer (Amsterdam) and Dr Steffen Murau (Boston). I am particularly grateful to Mr George Blades (Allens-Linklaters) for his assistance in translating and explaining Japanese legislation, to Ms Karen Chow and Mr Sebastian Mazay (ANU Law students) for providing excellent editorial assistance and to the anonymous OJLS referees whose thoughtful comments improved this work considerably. 1 financial relationship between national treasuries and central banks has increased with the ascendance of quantitative easing (QE) programmes as central banks' principal monetary policy tool; 3 it has spiked with the expansion of QE during the COVID-19 pandemic. 4 Perhaps anomalously in central banking debates, monetary finance has become a litigated topic: 5 bringing considerations of the legal limits on central bank powers to the forefront. While legal scholars of central banking have long observed the tension between fiscal and monetary power, 6 no detailed legal analysis of monetary finance has appeared. Perhaps because of that absence, high-profile economic and policy accounts of the topic appear to assume that the economic 'taboo' of monetary finance is backed up by a legal prohibition: 'all advanced economies and most emerging ones constrain governments' ability to finance public deficits with fiat money'. 7 As this article explains, that assumption has little support in the design and operation of law governing the financial relationships between central banks and national treasuries. Not only do central banks in advanced economies have significant legal authority to provide monetary finance, but those legal authorities have been exercised in ways which have made large contributions to national treasuries since the advent of QE. Focusing on the legal frameworks governing three major central banking systems (the Federal Reserve System, the Eurosystem or European Central Bank (ECB) and the Bank of England), this article elucidates the law governing monetary finance, and explains how that law has operated in the past decade of unconventional monetary policy. Following the method of leading legal scholars of central banking, 8 that multi-jurisdictional approach illuminates the role of legal norms in economic and policy debates on monetary finance, while ensuring that the distinct legal and institutional identities of individual central banks are not blurred. Section 2 explains the legal and financial mechanics of 'monetary policy' and 'monetary finance' for a general legal reader-a necessary prefatory exercise, given the obscurity of the legal and transactional structure of central bank operations. It also explains the core reasons why monetary finance is understood as a 'taboo', both economically and constitutionally: inflation, currency debasement and blurring the constitutional boundary between fiscal and monetary authority. Section 3 analyses the design of the legal frameworks governing monetary finance in the United States, the Eurosystem and the UK. Monetary law in each jurisdiction empowers and constrains central banks' authority to extend credit to government entities, purchase government debt securities in primary markets, remit profits derived from expanded balance sheets to national treasuries, and guarantee reinvestment of maturing government bonds and bills. Understanding those different legal frameworks is a critical first step toward appreciating the legitimate scope of central banks' financial relationship with governments and appreciating that the 'taboo' of monetary finance has no meaningful foundation in law. Section 4 shows how the law of monetary finance operated in the Federal Reserve, the Eurosystem and the Bank of England by studying the financial flows between those central banks and their national treasuries between the (2007-08) financial crisis and the COVID-19 pandemic. Over that period of unconventional monetary policy, the legal instruments governing those central banks facilitated significant levels of monetary finance. Some types of monetary finance occurred through laws conferring discretionary power on central banks to fund public expenditure, such as the Bank of England's extension of an unsecured £20 billion loan to the UK Treasury during the financial crisis. Other types of monetary finance result directly from QE: via mandatory 'profit transfer' laws, which require central banks to remit revenues derived from QE portfolios to national treasuries; and the reinvestment of QE portfolios in government debt securities. Those transactions between treasuries and central banks reveal the lawful flouting of the 'taboo' of monetary finance in the United States, the EU and the UK. Section 5 closes the article by evaluating the impact of those legal analyses of monetary finance on central banks' institutional position in liberal constitutional systems. Sustained monetary finance blurs the divide between monetary and fiscal powers and places central banks' sui generis position (remote from political and legal accountability mechanisms) in the liberal constitutional order under increasing pressure. A very prominent illustration of that pressure appears in the Eurosystem through the tripartite dispute between the German Federal Constitutional Court (Bundesverfassungsgericht), the ECB and the Court of Justice of the European Union. 9 Although that dispute stems from the anomalously juridicalised attitude to central bank powers in the EU, it illustrates more general tensions arising from the absence of meaningful constitutional accountability mechanisms in central banking. The article concludes by mooting some potential legal and institutional reforms to resolve those tensions. At the outset, it is necessary to illuminate several matters which may be obscure to the general legal reader: the mechanics and legal basis of 'monetary policy'; the definition of 'monetary finance'; and the reasons why monetary finance is understood as a 'taboo' from the vantage point of orthodox economics and liberal constitutional theory. Those matters provide critical context to the article's analyses of the legislative frameworks governing monetary finance (section 3), the way those legal frameworks have operated since the financial crisis and during the COVID-19 pandemic (section 4), and the constitutional position of central banks (section 5). Central banks mainly exist to provide banking services to commercial banks and governments: 10 liquidity insurance (issuing money or 'credit') 11 and settlement facilities (payments systems allowing commercial banks and governments to settle their transactions and store their money). 12 'Monetary policy' describes the policies adopted by central banks in the provision of those banking services, with the goal of influencing real economic activity, including production, employment and investment. 13 The design of monetary policy has varied significantly since the inception of modern central banking; 14 from 2008, central bankers tend to divide monetary policy into 'conventional' (prefinancial crisis) and 'unconventional' (post-financial crisis) strains. 15 9 Weiss (CJEU) (n 5); Weiss (BVerfG) (n 5); ECB, 'Account of the Monetary Policy Meeting of the Governing Council of the European Central Bank' (3-4 June 2020). 10 This is a necessary heuristic simplification. For more in-depth legal analyses of central banks' many functions, see Lastra Before the financial crisis, conventional monetary policy targeted the achievement of a stable growth in the price of consumer goods (ie inflation) of around 2%: 'price stability'. 16 The principal method of hitting that target was to set short-term interest rates in the financial and non-financial economy by manipulating the price at which commercial banks could obtain credit on the inter-bank market, including from the central bank. 17 Legally, interest rate setting was executed through central banks' statutory powers to 'buy and sell' assets and 'lend' money to counterparties in the financial sector. 18 Exercising those legal powers, central banks provided credit to the financial sector by executing three types of transactions: sale and repurchase or 'repo' transactions (a form of fully collateralised lending); 19 collateralised credit facilities; 20 and the 'outright' purchase and sale of financial assets. Via those transactions, central banks influenced the volume of credit available on the inter-bank market, and interest rates in the financial and non-financial economy through supply-anddemand dynamics. 21 Conventional monetary policy relied on commercial banks: (i) being liquid; and (ii) lending to one another and the non-financial economy. Those conditions disappeared during the financial crisis when commercial banks' liquidity supply failed and they drastically reduced lending as large segments of their assets were revealed to be valueless. 22 Thereafter, monetary policy pursued a more diffuse set of objectives, including: 23 ensuring the inter-bank credit market continued to operate; opposing deflationary pressures; boosting credit provision to the real economy; and backstopping the payments system. 24 banks in the United States, the Eurosystem and the UK pursued those objectives by creating enormous and sustained expansions of the money supply through large-scale outright purchases of high-quality financial assets: that is, QE. 25 QE was legally executed via central banks' statutory powers to purchase securities, and those purchases were funded by the creation of credit balances ('reserves') in the accounts held by commercial banks at central banks: 26 viz an expansion of the money supply. Critically for monetary finance, the largest volume of assets purchased through QE were government debt securities. 27 As the COVID-19 pandemic began in 2020, QE programmes were expanded, both increasing the money supply and the volume of government debt owned by central banks. 28 As those debt securities paid interest and matured, central bank revenue and profit increased significantly, creating the conditions for large financial transfers to national treasuries and the 'taboo' of monetary finance. That definition concentrates attention on the budgetary impact of monetary finance, reflecting the importance of liquidity provided by central banks to treasury expenditure and the highly politicised matter of fiscal responsibility. 33 The budgetary definition of monetary finance has three main elements. Those three criteria are evident in 'direct' and 'indirect' types of monetary finance. Direct monetary finance occurs when a central bank: (i) credits a government account via an (un)secured loan transaction; or (ii) purchases a government debt security on the primary market. 40 In both cases, the central bank expands the money supply by creating new reserves that are directly transferred to the relevant government account. Although a formal 'debt' may be created, no net liabilities are created within the consolidated public sector, 41 and the loan proceeds are available for general government expenditure. Indirect monetary finance may occur in several different ways. (1) Remittance financing occurs when central banks transfer surplus profits which derive from earnings on securities acquired through their market operations. '[P]ast deficits and debts can be monetized' through those profit transfers because: 42 (i) interest payments ('coupons') on government debt securities are, at least, partially refunded to treasuries via a central bank's balance sheet; and (ii) national treasuries obtain the benefit of yields on private sector debt securities without any corresponding outlay. In both cases, the money supply is expanded by central banks crediting the reserve accounts of asset-purchase counterparties, the remitted yield of monetised debt purchases is available for general government expenditure and the transfer from central bank to national treasury does not create any form of debt liability. (2) Reinvestment finance occurs when central banks reinvest the proceeds of maturing government debt securities via purchases of new government debt on the primary or secondary market. When that type of reinvestment programme is announced, central banks' monetary powers are used to inject additional demand into primary markets for sovereign debt, thereby lowering the borrowing costs of debt-issuing sovereigns. 43 C. 45 From that economic perspective, monetary finance is forbidden because it fatally distorts the price system by artificially creating an overabundance of money, destroying the credibility of the national currency. Monetary finance could also be understood as a 'taboo' from a constitutional perspective: monetary finance represents a constitutionally illegitimate exercise of 'fiscal' powers (to tax and spend) by 'monetary' authorities. In orthodox liberal constitutional models, parliaments, as the apex democratic institution, have a monopoly on fiscal powers through legislative authorisation of taxation and national budget expenditure, and treasuries are formally or informally accountable to parliaments in their execution of financial plans. 46 Judiciaries stand ready to enforce democratic accountability over fiscal powers by holding treasuries (and tax agencies) to the boundaries of parliamentary legislation. For a variety of reasons, 47 central banks fall outside those traditional lines of constitutional accountability: they are remote from treasuries and parliaments, and largely immune to judicial review. 48 Alert to those design features of liberal constitutionalism, monetary finance appears as a constitutional taboo because it represents the illegitimate exercise of fiscal powers to spend and tax. 49 The veracity of the economic taboo of monetary finance has been strongly contested, with theoretical and empirical attacks being mounted against the idea that it necessarily causes inflation. 50 finance has been less overtly contested, although leading scholars of central banking have long noted the need to bring central banks within broad avenues of constitutional legitimacy which attach to fiscal institutions. 51 A major reason behind the low visibility of monetary finance in constitutional discussions of central banks is a general gap in the literature concerning the design of legal norms governing monetary finance, and the manner in which those norms operate. Sections 3 and 4 of this article fill that gap by explaining how legal frameworks can empower, or require, central banks to provide monetary finance and how those laws operate under the unconventional monetary policies adopted by central banks in the United States, the EU and the UK in response to the financial crisis and the COVID-19 pandemic. Together, those analyses reveal that the 'taboo' status of monetary finance is not reflected in the law which governs the relationship between central banks and treasuries in the most influential monetary systems of the advanced economies. The constitutional and policy impacts of that revelation are discussed in section 5. Although other aspects of central banking law are covered by a sophisticated literature, 52 the law governing financial relations between central banks and treasuries has hitherto been neglected. Perhaps because of that neglect, economic and policy debates about central banking often proceed on the premise that legal rules prohibit monetary finance, particularly in advanced economies. 53 The legal materials do not support that conventional wisdom: the law governing central banks and treasuries in the United States, the Eurosystem and the UK provide significant authority to engage in monetary finance. The Federal Reserve Act (FRA) gives each FRB general corporate powers 'to make contracts', including contracts for the provision of credit and purchase of assets. 54 Authority to use that general power to supply credit to the financial sector is conferred via bespoke legislative rules concerning 'advances' to commercial banks and, more importantly, 55 the execution of open-market operations (OMOs) by the FRB of New York (FRBNY) on the instruction of the Federal Open Market Committee (FOMC). 56 The FRBNY relies on that statutory authority to carry out the repurchase and outright purchase/sale transactions which facilitated conventional (interest rate setting) and unconventional (QE) monetary policy operations. 57 In executing OMOs, legal power is conferred on the FRBNY to purchase US debt securities on the condition that such power not be exercised in a bilateral or 'closed-market' operation with the US federal government. That constraint on direct monetary finance is contained in the final words of section 14 (2) The 'open-market' constraint first appeared in 1935 and post hoc explanations for its enactment have included discouraging government deficits, controlling the growth of the FRBs' balance sheets and protecting the margins of primary dealers in government securities. 58 Before its enactment, the US Treasury borrowed relatively large sums on a bilateral basis from the FRBs. 59 During World War Two, the open-market constraint was replaced by a $5 billion debt ceiling on the buying of bonds 'directly from . . . the United States'. 60 That wartime permission for direct lending to the US Treasury remained in place after the cessation of hostilities, and continued to provide a legal basis for the FRBNY to extend bilateral credit to the US Treasury until its repeal in 1981. Thereafter, the FRBNY was only authorised to engage in primary market transactions with the US Treasury through an 'open' tender process. Using that authority, the FRBNY acquires US government bonds and bills directly from the Treasury via non-competitive bids at auction by 'exchanging' shortly expiring US government debt securities (acquired through OMOs) with equivalent newly issued securities. 61 That legal authority underlies the FRBNY's reinvestment of maturing US government securities in its QE portfolio. 62 Indirect monetary finance is also permitted by the 'division of earnings' rules in section 7 of the FRA, which requires transfers of 'surplus funds' from the Federal Reserve to the US Treasury. The provision requires that the amounts transferred to the US Treasury be calculated by deducting from total FRB earnings: (i) a dividend that each FRB pays from its net profits to its commercial banks shareholders; 63 and (ii) other surplus funds up to an aggregate limit of $6.825 billion. 64 Any surplus above that limit is transferred 'to the Secretary of the Treasury for deposit in the general fund of the Treasury'. 65 Once transferred, the FRA imposes limitations on how the US Treasury may use the Federal Reserve's net surplus: 66 'The net earnings derived by the United States from Federal reserve banks shall, in the discretion of the Secretary . . . be applied to the reduction of the outstanding bonded indebtedness of the United States.' In the Federal Reserve's accounts, the surplus funds transferred to the US Treasury are called 'remittances'. 67 In conventional times, remittances may yield only a trivial amount of monetary finance, being the share of profits flowing from the short-term credit operations (mainly sale and repurchase agreements, and collateralised lines of credit), which represent relatively small expansions of the money supply. 68 limits on budget deficits of individual member countries' and 'limits on borrowing in non-Community currencies'. 74 Thereby, article 123 shared a similar origin with the United States' prohibition of bilateral dealings with the US Treasury: ensuring budget discipline. As the Court of Justice of the European Union (CJEU) explained in two cases on the lawfulness of the ECB's asset purchase programmes: when the ECB purchases government bonds on secondary markets, safeguards must be built into its intervention which are, on the one hand, sufficient to ensure that it does not fall foul of the prohibition of monetary financing . . . and, on the other, intended to limit the impact of the ECB['s QE] programme on the impetus to follow a sound budgetary policy. 75 In that sense, the Eurosystem's prohibition on direct monetary finance is far stricter than the US open-market constraint in limiting the legal powers of the national central banks of the Eurosystem (NCBs), which carry out all significant monetary policy operations, to engage directly with national treasuries. It also constrains QE reinvestment programmes, which may only occur through secondary market purchases which do not operate as de facto primary market purchases. 76 However, like the US FRBs, most NCBs are explicitly required to provide indirect monetary finance to their national treasuries by legal rules for the remittance of profits. The legal framework for remittances in the Eurosystem is complex, but can be distilled to five main steps undertaken annually. 77 First, NCBs earmark their income from monetary policy operations: 'monetary income', which includes income earned from credit facilities and assets purchased via QE programmes. 78 Secondly, the monetary income of all NCBs is pooled and then distributed among the NCBs according to their capital contribution to the ECB (capital key), which can create a net gain/loss of monetary income. 79 Thirdly, the ECB determines its net profit, deducts reserves and distributes the balance to NCBs according to the capital key. 80 Fourthly, the NCBs mix their portion of pooled monetary income and ECB distributions with other income. 81 Fifthly, the NCBs apply their domestic legal rules for the 74 ibid 24. 75 Weiss (Advocate General Opinion) (n 5) para 51; Weiss (CJEU) (n 5) para 107; Gauweiler and Others (n 5) paras 102, 115. 76 As section 5 below explains, the CJEU's interpretation of art 123 of the TFEU was heavily contested by the German Bundesverfassungsgericht, in a dispute which illustrated the constitutional controversy of monetary financing in the EU: Weiss (BVerfG) (n 5). 77 calculation and distribution of profits, including remission of surplus profits to national government treasuries. The effect of that sequence of legal rules is to push the ultimate question of the remittance of surplus profits down from the supranational European level to the domestic statutory law governing the NCBs that permits or requires the remission of surplus profits to national treasuries. 82 Because unconventional monetary policy loaded the Eurosystem's balance sheet with debt securities and credit facilities, interest revenue was the largest income item for the NCBs and accounted for much of the surplus profits remitted to national treasuries. 83 Like remittance finance in the Federal Reserve, a significant proportion of those larger profits represents indirect monetary financing in the form of debt relief to sovereigns in the Eurosystem 84 and monetised yield of corporate debt purchases. Unlike the Federal Reserve or the ECB, the Bank of England's legal framework contains express legal authority to provide direct and indirect monetary finance to the UK Treasury. Section 12(7) of the National Loans Act 1968 (UK) permits the Bank of England to provide uncollateralised credit to the UK central government up to the point of fiscal deficit and to ensure 'sound monetary conditions'. That financing permission operates by reference to the current balance of the UK's public finances and the Bank's assessment of 'sound' monetary conditions. 85 No limit is imposed on the commercial structure of loans under section 12(7) of the National Loans Act: they may be secured or unsecured, may occur through a balance transfer or by purchase of debt securities and may be unlimited as to maturity. The Bank of England has traditionally provided unsecured loans through a credit facility called the 'Ways and Means Advance'. 86 As section 3 explains, that line of credit was most recently used in 2008 to fund the UK Treasury's bailout of an ailing British bank, 87 but it was also 81 Which may include fees, commissions, charges and earnings on financial assets which are not held for monetary policy purposes: see European Central Bank, 'Agreement on Net Financial Assets' (25 October 2019) . 82 Illustrations of the operation of these legal rules appear in section 4 below. 83 As is recognised in general terms by Buiter (n 1) ch 2. 84 used extensively during the late 19th century and during wartime. 88 That central bank credit facility was expressly exempt from the prohibitions on monetary financing contained in article 123 of the TFEU during the UK's membership of the EU, 89 and has been advertised as a potential source of emergency funding during the COVID-19 pandemic. 90 UK law also provides for indirect monetary finance by remittances of cash balances accruing from government and corporate bonds purchased through its QE programme. Unlike the QE programmes undertaken by the Federal Reserve and the ECB, the Bank of England executes its asset purchases via a wholly owned, special purpose corporate vehicle: Bank of England Asset Purchase Facility Fund Limited (BEAPFFL). The BEAPFFL was incorporated under UK legislation as a wholly owned subsidiary of the Bank, with the generic legal powers given to corporations to buy, hold, sell and encumber property (including government debt securities), to enter into agreements with governments and to transfer money. 91 After its incorporation, the BEAPFFL received a 'loan' from the Bank of England equal to the amount of assets intended to be purchased under the UK's QE programme; that loan was funded by newly created central bank reserves. 92 The BEAPFFL then used those loan proceeds to purchase UK sovereign bonds and corporate bonds on the secondary market. As coupons are paid on those bonds, the BEAPFFL accrues cash surpluses in its own account, which are then transferred to the UK government's consolidated fund account. The legal basis for those transfers is a private contract between the UK treasury and the BEAPFFL, 93 as opposed to the statutory profit transfer rule applying to the Bank of England. 94 In that way, the UK's national treasury directly receives all profits from the Bank of England's QE portfolio. Table 1 Four different types of legal frameworks governing monetary finance are identifiable. The first type is an expressly permissive legal framework: the central bank is expressly authorised to engage in direct and indirect monetary finance. The Bank of England operates within that type of legal framework, with authority to provide unsecured loans to the national government with the only limits being the existence of a cash or structural deficit in the UK government's consolidated accounts and the promotion of 'sound monetary conditions'. The second type of legal framework is an implied permissive legal framework: the central bank is prohibited from some forms of direct monetary finance but is permitted by remittance rules to provide debt relief to central governments. The US Federal Reserve System is governed by that type of monetary finance framework. The FRBs are prohibited from providing the US Treasury with bilateral (closed-market) finance, but are required to remit the profits of interest earned on QE portfolios and are permitted to monetise demand in primary government debt markets through QE portfolio reinvestment programmes. The third type is a covertly permissive legal framework: the central bank is covertly permitted to engage in indirect monetary finance (by remittance and reinvestment rules) despite direct monetary financing being strongly prohibited. The ECB operates within such a legal framework: forthright rules against the provision of direct monetary finance exist; however, NCBs are permitted to provide remittance and reinvestment finance to their national treasuries. The fourth type is an outright prohibition legal framework: no such framework exists in any of the studied central banking systems. That conclusion is significant because it demonstrates the absence of any strict legal censure of monetary finance at the apex of the global financial system, indicating a far more fluid attitude to the distinction between monetary and fiscal authority than is implied by some accounts of the institutional position of central banks. Section 5 returns to that topic, after section 4 analyses the operation of the legal frameworks governing monetary finance by focusing on the flow of funds between central banks and treasuries from the financial crisis to the COVID-19 pandemic (2008-20). 96 The size of the monetary injection from the Bank to the UK Treasury is best appreciated in relative terms. The £19.5 billion monetary injection by the Bank was 200þ% of the UK's monthly income tax receipts in the preceding month and 300% of the UK's monthly VAT receipts in the same month. Although the expansion of the money supply to fund the WMA drawing was non-trivial, the Bank of England's credit line to the UK government was repaid relatively promptly. That monetary injection was also deployed to achieve a critical emergency measure (preventing a bank run) and, in that sense, could be understood as authorised by the legal permission in section 12(7) of the National Loans Act 1968 (UK) to ensure 'sound monetary conditions'. As the following section explains, the extent of that monetary finance paled in comparison with the budgetary accommodation provided to national treasuries from QE programmes adopted during the financial crisis. 95 Other central banks also provide direct finance to their central governments, eg the Bank Indirect monetary finance via the remittance of sovereign debt interest (coupon) payments from central banks to national treasuries was a common feature of the QE programmes operating in the United States, the Eurosystem and the UK. The same basic sequence was followed in each jurisdiction: (1) central banks purchased (majority) government and (minority) private debt securities on the secondary market; 97 (2) central banks' profits increased as government and corporate treasuries paid coupons on those securities directly to central banks; (3) legislative rules required central banks to remit a share of those profits to national treasuries; and (4) remittances provided monetary finance by: (i) providing debt relief to governments to the value of remittances attributable to that governments' coupon payments; and (ii) providing a monetised yield to governments to the value of remittances attributable to corporate coupon payments. Close analysis of the annual accounts of the Federal Reserve System, the ECB and the Bank of England illustrates the significant absolute levels of remittance finance which flowed from central banks to national treasuries from the inception of QE. 98 In the following analyses of each jurisdiction, the relative significance of that monetary finance is illustrated by reference to two core budgetary indicators: sovereign debt interest liability and the fiscal deficit/ surplus. In each jurisdiction, the role of legislative rules in facilitating monetary finance is vividly illustrated. Figure 2 shows the flow of funds from the Federal Reserve's consolidated account, SOMA, to the US Treasury pursuant to section 7 of the FRA, which provides that: 'The net earnings derived by the United States from Federal reserve banks . . . shall be applied to the reduction of the outstanding bonded indebtedness of the United States' (left axis). The figure also shows the proportional contribution of those funds to the US government's budget deficit and its debt interest liability (right axis). 97 The assets purchased in the studied jurisdictions broadly consisted of: sovereign bonds (United States, Eurosystem and UK); sovereign money-market securities (United States); non-sovereign government debt securities (such as US government agency debt and US government-guaranteed residential mortgage-backed securities and debt securities issued by state-owned entities in the Eurosystem); debt issued by international or supranational institutions (particularly in the Eurosystem); and corporate debt securities (in the United States, the Eurosystem and the UK). Some central banks purchased a broader range of instruments: see eg Nakaso (n 25) 5, describing the Bank of Japan's purchases of exchange traded funds. 98 In each jurisdiction, the amount of monetary finance from remittances is calculated using a formula MFr ¼ Pt * Rm, where Pt is the annual profit transferred from the central bank to the national government (including taxation where appropriate) and Rm is the percentage of the central bank's annual revenue attributable to expansion of the money supply, identified by calculating the proportion of the central bank's revenue from 'interest' (as opposed to dividends, fees and commissions) to total revenue. It begins in 2009, as the United States' first QE programme, Large-Scale Asset-Purchase Program (LSAP1), leads to the purchase of $2 trillion of assets. 99 By mid-2010, LSAP1 has concluded and the SOMA stabilises at slightly over $2 trillion. Between 2010 and mid-2011, the United States' second QE programme (LSAP2) adds $600 billion of US Treasury securities to SOMA. Remittances to the US Treasury track that expansion in the money supply, as they increase from 2009 to 2010, then plateau. Between September 2011 and September 2012, asset purchases continue without any balance sheet expansion as the FRBNY implements the maturity extension programme (MEP). 100 Predictably, as the MEP did not involve an expansion of the money supply, remittances to the Treasury decreased slightly between 2011 and 2012. In September 2012, the FOMC announced a third QE programme (LSAP3), adding $1.5 trillion to SOMA until its completion in December 2013. Remittances as a result of that monetary expansion increased significantly between 2013 and 2015: from <$80 billion to >$120 billion, amounting to 20-30% of the annual US national debt interest liability and 12-25% of the United States' fiscal deficit. Part of that increase was the result of a special legislative impost on the Federal Reserve's consolidated assets to fund infrastructure development projects. 101 Between 2015 and 2017, the Federal Reserve's QE holdings were maintained through reinvestment programmes, and from late 2017, SOMA began to shrink due to the reinvestment of maturing securities through 'normalisation' programmes. Around $800 billion of assets were removed from SOMA between 2017 and 2019. That contraction in the money supply is reflected in the proportional reduction of remittances to the US Treasury over the same period: reducing from $80 billion to $50 billion; from 20% to 10% of US debt interest liabilities; and from 15% to 5% of the fiscal deficit. The same basic relationship between expansion of the money supply by QE purchases and increased monetary finance via remittances to national treasuries is identifiable in the Eurosystem. The four largest NCBs are selected to prove that point: 102 the Bundesbank, the Banque de France, the Banca d'Italia and the Banco d'España. 103 For each NCB, the three critical date ranges are: (i) 2010-12, when the Securities Markets Programme (SMP) was active and the ECB's balance sheet expanded by e1 trillion; 104 (ii) 2012-15, when the ECB's balance sheet contracted by e1 trillion when no QE-like programme was operational; and (iii) 2015-19, when the Asset Purchase Program (APP) was active and the ECB's balance sheet expanded by e2.5 trillion. 105 Within those date ranges, remittances to national governments tracked the expansion and contraction of the studied NCBs' balance sheets in fiscally-significant ways. 106 Figure 3 shows the flow of funds authorised by article 27 of the Bundesbank Act (Bundesbankgesetz), which provides that the Bundesbank may retain 20% of its annual profits for reserves (until a limit of e2.5 billion is reached) and thereafter all profit is paid to the Federal Republic of Germany. The figure commences in a period (2009-11) when the Bundesbank's balance sheet was expanding with asset purchases under the SMP and expanded credit provision. The remittance spike from 2012 to 2014 (from e500 million to e4 billion) reflects increased interest earnings on EU Member State debt securities acquired via SMP purchases (and credit facilities), evidencing the connection between the Bundesbank's participation in the Eurosystem's expanded money supply and its increased interest earnings and profit transfers to the German state. The stunningly large contribution of those monetised remittances to Germany's fiscal position (amounting to 230% of the 2012 fiscal surplus and 370% of the 2013 surplus) reflects Germany's move to a modest fiscal surplus in 2012. In budgetary terms: the entirety of Germany's surpluses in those years, at the height of the EU sovereign debt crisis, was monetised through remittance attributable to interest repayments on other Eurosystem Member State debt. In legal terms, the Bundesbank Act required the Bundesbank to monetise Germany's budget surplus. As the Bundesbank's SMP portfolio contracted between 2013 and 2015, interest earnings and remittances contracted accordingly. However, the proceeds of the Bundesbank's SMP assets continued to make non-trivial contributions to the German government's debt interest payments (5%). From 2015, the Bundesbank began participating in the ECB's APP and its balance sheet expanded in a sustained manner for the remainder of the studied period. Remittance payments boomed as a result of the increased yield on securities purchased in the APP: the expansion in the Eurosystem money supply boosted remittances from the Bundesbank to the German government from e5 billion. The same basic trend between expansions of the money supply through asset purchases and increased remittances to NCBs' national treasuries is identifiable in the Banque de France, the Banca d'Italia and the Banco d'España. In France, the Monetary and Financial Code requires the Banque de France to pay corporate income tax and the balance of its after-tax profit to the French state; 107 monetised remittances paid under those legal rules stayed between e4 billion (pre-SMP earnings) and e7 billion (post-APP earnings), reaching high points of þ12% of annual debt interest payments and 10% of the fiscal deficit in 2018. In Italy, the Statute of the Bank of Italy requires that a maximum of 40% of profit be maintained for reserves and the balance distributed to the state, 108 resulting in monetised remittances between 2% and 11% of national debt interest liabilities and 1% and 5% of the fiscal deficit. In Spain, a Royal Decree provided that all profits be distributed to the Treasury of Spain, 109 resulting in the same link between expanded money supply, increased remittances and large contributions to the national budget (5-14% of debt interest and 3-8% of the fiscal deficit). Unlike in Germany, monetised remittances in France, Italy and Spain never contributed to surplus expenditure, and their lower contribution to debt interest repayments and fiscal deficit (relative to Germany) reflected, relative to Germany: (i) higher levels of debt and deficit; and (ii) lower total profit from QE portfolios under the Eurosystem's legal rules for profit distribution by reference to the ECB's capital key. 110 Predictably, transfer of profits from the Bank of England's QE subsidiary the BEAPFFL stuck to the American and European trend: a strong correlation is observed between the expansion of the Bank of England's balance sheet and the size of remittance payments to the UK Treasury. 111 As section 3 noted, remittances from the UK's QE programme were governed by a private law contract between the special purpose vehicle which carried out QE for the Bank of England and the UK's treasury, rather than a public law statute. Under that contract, all QE earnings were remitted to the UK's treasury. 107 110 See the explanation of the eurozone profit transfer law in section 3 above. That law will determine the amount of monetary finance via remittance which each Eurosystem Member State can receive, and favours states with large capital contributions to the ECB rather than states with large sovereign debt issues. 111 For operational details, see Joyce, Tong and Woods (n 25). The Bank of England began its QE programme in 2009, but earnings on QE portfolio assets were hoarded until 2013, when the UK Treasurer made a written request for the transfer of the cash balances held in the BEAPFFL's accounts to be used for debt-servicing. 112 The enormous spike in remittances in FY2013-14 represents the accumulated earnings of the UK's QE programme. Its contribution in that fiscal year to the UK's debt interest liability was profound (þ60%), as was its contribution to deficit financing (40%). From 2015 to 2019, remittances remained largely stable as the Bank of England's QE subsidiary maintained its balance sheet position. The fiscal benchmarks indicate the continued materiality of those remittances. The monetised contribution to UK interest repayments stayed constant at 20%, while the relative contribution to deficit financing increased (from <10% to >30%) as the UK Treasury implemented a deficit reduction policy. The final example of monetary finance occurred through central banks' reinvestment of the proceeds of maturing government debt securities in their QE portfolios. Simplifying the complexity of central banks' market operations, monetary finance occurs via QE reinvestment programmes in the following steps: (1) the treasury issues debt security (Bond A) to a private sector body with special privileges to purchase government debt on the primary market which participates in central bank asset purchase programmes (primary dealer); 113 A simple measure of the extent of reinvestment finance is the total amount of principal reinvested in government securities on the primary market, a practice followed by the FRBNY. In those circumstances, the gross volume of government debt purchased represents monetary finance because the treasury is a guaranteed purchaser up to the reinvestment limit. Figure 6 Where reinvestment programmes occur on the secondary market (as in the Eurosystem and the UK), measuring the monetised demand they create, and thus the amount of monetary finance provided, is far less straightforward. As section 3 noted, legal instruments limit the Eurosystem and (until Brexit) the Bank of England from directly purchasing government debt. That limit does not, however, wholly prevent those central banks from providing reinvestment finance, because a reinvestment announcement provides artificial demand in the secondary market for government debt, which lowers funding costs for sovereigns in primary markets. Determining the precise monetary yield of reinvestment programmes requires quantifying the relationship between primary market borrowing costs and the impact of central bank operations on market makers in the secondary market, both of which are beyond the scope of this article. However, the extensive empirical literature on the impact of asset purchase programmes on bond yields and liquidity supports the proposition that reinvestment programmes provide a significant measure of indirect monetary finance. 114 from the COVID-19 QE programmes will depend, inter alia, on the interest rate on government debt purchased by central banks, and the central banks' policies concerning reinvestment of their QE portfolios. It is, however, clear that monetary finance will continue to be a feature of central banks' institutional position after the COVID-19 pandemic has abated. This section's analysis of the operation of the legal frameworks governing monetary finance, through the flow of funds between central banks and national treasuries, discloses several important matters concerning central banking in the United States, the EU and the UK. First, legal authority to provide monetary finance is not purely formal or vestigial: it has been exercised extensively since the financial crisis and will continue to feature following the COVID-19 pandemic. Secondly, the levels of monetary finance provided via those legal authorities are not trivial: the legal frameworks governing central bank and treasury relations provide very meaningful monetised support for national expenditure. Thirdly, the most powerful examples of monetary finance derive from QE, indicating a link between central banks' emergency crisis fighting powers and additional monetised relief to sovereigns. Finally, and most importantly, the 'taboo' of monetary finance has clearly been flouted on a large and sustained basis since the financial crisis. The article closes in section 5 by reflecting on the impact of those propositions on the constitutional position of central banks. As section 2 observed, monetary finance can be understood as constitutionally problematic because it blurs the distinction between fiscal and monetary institutions. Fiscal authority (to tax and spend) is concentrated in parliaments and requires parliamentary approval before its exercise: witness the annual parliamentary budgetary process and the need for legislation to support lawful tax collection. Judiciaries stand behind those democratic representative bodies to, in principle, enforce the will of the representative majority enshrined in legislation on executive, particularly tax, agencies. That depiction of the constitutional position of fiscal institutions is a simplification, but it captures the conventional wisdom regarding the legitimate place of fiscal institutions in liberal constitutional systems. 121 Monetary authority (to create and control the value of the national currency) has, for most of the 20th century, been concentrated in central banks which have a very different constitutional position to fiscal authorities. At least from the 1950s, central banks in the United States, Europe and the UK have enjoyed varying levels of de facto or de jure independence from direction by politicians and treasury officials. 122 As Lastra has explained, there are good reasons for insulating central banks from the pressures of democratic politics, particularly preventing inflation (and its attendant social ills) through 'easy money' policies and debasement of the currency through hyperinflation. 123 Those justifications for central bank independence assume that monetary institutions do not directly act in support of fiscal institutions. 124 Central banks also enjoy significant formal and functional immunities from judicial oversight. Obscure (and perhaps outdated) legal authorities in prominent jurisdictions insulate central banks from ordinary judicial review in relation to a number of their functions. 125 In addition to those formal legal protections, there are powerful practical considerations which functionally insulate central banks from ordinary judicial review, including: the secrecy which attends their activities; the time sensitivity and commercial sensitivity of their operations; and the difficulty in identifying a clear contradictor to commence proceedings against a central bank within adversarial litigation systems. 126 The exceptional case appears in the EU, where the ECB has been subject to high-profile litigation; but, even there, the European judiciary has accorded high levels of legal autonomy (or deference) to the ECB in passing judgment on its monetary policy activities. 127 Although a complex set of factors underlie the non-justiciable nature of many central bank activities, 128 the institutional distinction between monetary and fiscal institutions is surely influential. The fact that central banks have engaged in monetary finance challenges the rationales for their sui generis constitutional position. 129 Providing financial accommodation to national governments entangles central banks in the exercise of fiscal authority, including: backstopping significant public expenditure programmes (through direct and indirect monetary financing of fiscal deficits); and acting as de facto sovereign debt managers (as QE portfolios warehouse and retire sovereign debt). The import of those fiscal powers will only increase in the wake of the enormous fiscal rescue programmes which have been announced to fight the economic harm caused by the COVID-19 pandemic. The predictable result of integrating central banks into the fiscal apparatus of government is increased tension concerning their constitutional position. Those tensions have most clearly surfaced in the eurozone, through the Weiss dispute (litigated in the CJEU and the German Federal Constitutional Court-the BVerfG), where the relationship between monetary finance and QE arose in a challenge by German politicians and economists to the legality of the ECB's government-debt purchasing QE programme: the Public Sector Asset Purchase Programme (PSPP). 130 In the Weiss litigation, the PSPP was challenged as exceeding the ECB's legal mandate to engage in 'monetary policy' and a violation of the prohibition on monetary financing contained in article 123 of the TFEU. 131 The CJEU held that the PSPP was intra vires its mandate, and a proportionate exercise of the ECB's powers to engage in monetary policy operations. 132 It also held, as section 3 noted, that the PSPP did not violate article 123 because the ECB had built certain 'safeguards' into the PSPP which prevented secondary market purchases of Eurosystem Member State debt from acting as disguised 'direct' purchases, and thereby was not direct monetary financing in violation of article 123. 133 Subsequent to the CJEU's decision, the BVerfG took the extraordinary step of hearing a challenge to the legality of the PSPP under the German Basic Law, and found that the CJEU had failed properly to scrutinise the legality of the PSPP on both 'mandate' and monetary finance grounds. 134 The BVerfG then decided that ECB had failed to 'substantiate' that the PSPP was proportionate to a monetary (rather than fiscal) policy objective under the TFEU 135 because the ECB failed, inter alia, to properly consider the PSPP's fiscal impacts as indirect monetary finance. 136 The BVerfG then, also extraordinarily, ordered that the Bundesbank could not participate in the PSPP until the ECB had provided further evidence that the PSPP was a proportionate monetary policy measure to the Bundestag and the German Finance Ministry, and those two institutions had signalled their approval of the proportionality of the PSPP. 137 Those steps occurred, and the Bundesbank is now permitted to participate in the ECB's QE programme. 138 The BVerfG's Weiss judgment also held that the PSPP did not manifestly violate the monetary financing prohibition in article 123, although it was deeply critical of the deference the CJEU displayed towards the ECB. 139 Throughout the BVerfG's Weiss judgment, the constitutional tensions caused by the ECB's entanglement in fiscal policy through its QE programme are clearly discernible. The court observed that the PSPP could be 'essentially . . . an economic and fiscal policy agenda' which uses 'the purported monetary policy' as a 'disguise'. 140 Recognising the constitutionally problematic nature of a central bank exercising fiscal powers, the BVerfG held that: The distinction between economic policy and monetary policy is a fundamental political decision with implications beyond the individual case . . . The classification of a measure as a monetary policy matter as opposed to an economic or fiscal policy matter bears not only on the division of competences between the European Union and the Member States; it also determines the level of democratic legitimation and oversight of the respective policy area. The court's concern with democratic legitimacy was reflected in the remedy it granted: it required the German parliament and the politically accountable Federal Ministry of Finance to deliberate on the ECB's justification of the PSPP as a monetary (rather than fiscal) policy measure. The Weiss controversy could only have arisen in the eurozone, with its uniquely juridicalised attitude to central bank powers, 141 but it displays the broader constitutional tensions raised by the existence of monetary finance under unconventional monetary policies. In concluding, it is worth mooting a set of institutional reforms which could potentially defuse those tensions. Governance changes could be made within central banks to provide more open and transparent treatment of monetary finance. Bespoke committees could be created to determine the impact of central bank operations on treasury finance, which communicate their deliberations and decisions to the public, modelled on the committees which determine monetary policy: the FOMC in the United States, the Governing Council of the ECB and the Monetary Policy Committee of the Bank of England. 142 Alternatively, the monetary financing activities of central banks could be more closely overseen by parliamentary bodies. Of course, parliamentary committees already hold central bank officials to account, but whether those committees result in meaningful impacts on central bank behaviour is questionable. 143 The desirability of each option would depend on a complex balancing of constitutional legitimacy, institutional independence, economic effectiveness and technical expertise. A higher-impact reform option would involve amending the legislative frameworks governing central banks and treasuries to provide clear quantitative and qualitative criteria to govern the provision of monetary finance, which are linked to annual parliamentary processes. For example, article 34 of the Bank of Japan Act requires direct parliamentary approval, including by reference to quantum, before monetary finance can be provided to the Japanese treasury: "As the central bank of Japan, the Bank of Japan may . . . conduct the following business with the national government:" (i) Making uncollateralized loans within the limit decided by the Diet as prescribed in . . . Article 5 of the Fiscal Act; 141 For a recent engagement with the legal limits on the ECB's mandate, see Rosa Lastra and Alexander Kern, 'The ECB Mandate: Perspectives on Sustainability and Solidarity' (Report to the European Parliament, June 2020). 142 I am indebted to one of the anonymous referees for raising this point. 143 For an excellent discussion of the parliamentary accountability mechanisms in the UK and Eurosystem, see Rosa Lastra, 'Accountability Mechanisms of the Bank of England and the European Central Bank' (Report to the European Parliament, September 2020). (ii) Making uncollateralized loans for the national government's temporary borrowing permitted under the Fiscal Act or other acts concerning the national government's accounting; (iii) Subscribing or underwriting national government securities within the limit decided by the Diet as prescribed in . . . Article 5 of the Fiscal Act; (iv) Subscribing or underwriting financing bills and other financing securities Under articles 5 and 7 of the Fiscal Act, the Diet (Japan's national legislature) must prescribe the 'special grounds' for unsecured loans under article 34(i) and (ii), and the debt purchases referred to in article 34(ii) can only be issued to finance cash defects in the 'national treasury' and must mature within a single fiscal year. Similar explicit regulation of the conditions under which monetary finance appears in the Bank of Canada Act, which permits the Canadian central bank to provide monetary finance to the national treasury, albeit to the limit of one-third of the 'estimated revenue of the Government of Canada for its fiscal year', and any monetary finance 'must be repaid before the end of the first quarter after the end of the fiscal year of the government that has contracted the loan'. 144 Legal frameworks resembling the Japanese and Canadian precedents could provide additional democratic legitimacy to central banks' provision of financial support to national treasuries and, thereby, reduce the constitutional tension caused by the blurring of monetary and fiscal power which arises from monetary finance. There are, of course, legitimate trade-offs between democratic accountability and economic management, 145 and hopefully this article will assist future work on the optimal balance between constitutional legitimacy and economic success in the design of monetary institutions. Scarcity and Spotlight Effects on Liquidity and Yield: Quantitative Easing in Japan' (Institute for Monetary and Economic Studies, Bank of Japan Board of Governors of the Federal Reserve System Federal Reserve System Assets What Is Quantitative Easing?' HM Treasury and Bank of England Announce Temporary Extension to Ways and Means Facility' (Press Release The Bank of England only releases data on 'weekly amounts outstanding' under the WMA, which omits any extensions of credit to the UK Treasury which are repaid within a weekly period: see Bank of England Dataset (RPWB72A), 'Weekly Amounts Outstanding of Central Bank Sterling Ways and Means Advances to HM Government (in sterling millions) Not Seasonally Adjusted'