Redistributing central bank profits & losses across the eurosystem: The eurosystem's monetary income
Abstract
EconStor is a publication server for scholarly economic literature, provided as a non-commercial public service by the ZBW.
Full text
Cesaratto, Sergio; Febrero, Eladio; Pantelopoulos, George Working Paper Redistributing central bank profits & losses across the eurosystem: The eurosystem's monetary income FMM Working Paper, No. 104 Provided in Cooperation with: Macroeconomic Policy Institute (IMK) at the Hans Boeckler Foundation Suggested Citation: Cesaratto, Sergio; Febrero, Eladio; Pantelopoulos, George (2024) : Redistributing central bank profits & losses across the eurosystem: The eurosystem's monetary income, FMM Working Paper, No. 104, Hans-Böckler-Stiftung, Macroeconomic Policy Institute (IMK), Forum for Macroeconomics and Macroeconomic Policies (FMM), Düsseldorf This Version is available at: https://hdl.handle.net/10419/301005 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/legalcode
FMM WORKING PAPER No. 104 • July 2024 • Hans-Böckler-Stiftung REDISTRIBUTING CENTRAL BANK PROFITS & LOSSES ACROSS THE EU ROSYSTEM: THE EUROSYSTEM’S MONETARY INCOME Sergio Cesaratto 1 , Eladio Febrero 2 , George Pantelopoulos 3 ABSTRACT National Central Banks (NCBs) of the Eurosystem pool profits and losses related to monetary policy operations to form the Eurosystem’s so-called ‘monetary income’. This is then redistributed – i.e. allocated – among NCBs according to respective capital keys (the participation shares of each NCB to the ECB’s capital). Monetary income has relevance for current debates such as that concerning the high fiscal costs of an ample reserve regime as a result of the abundant reserves banks hold in the deposit facility of their respective NCBs. These costs are in fact redistributed through the allocation of monetary income. Nonetheless, exactly how monetary income is pooled and subsequently allocated between Eurosystem NCBs remains rather enigmatic. The aim of this paper is to explore how monetary income is both pooled and allocated. This seems a useful task beyond the aforementioned debate to dissipate other puzzling issues like the costs of TARGET2 imbalances. A more detailed dissemination from the relevant authorities as to the process by which profits/losses are pooled and subsequently allocated is however in our view warranted. ————————— 1 Università di Siena, Italy. Email: [email protected]. 2 University of Castilla-La Mancha, Spain. 3 University of Newcastle, Australia.
1 Redistribung central bank profits & losses across the Eurosystem: the Eurosystem’s monetary income Sergio Cesarao, Eladio Febrero, George Pantelopoulos* (This version June 2024) Abstract Naonal Central Banks (NCBs) of the Eurosystem pool profits and losses related to monetary policy operaons to form the Eurosystem’s so-called ‘monetary income’. This is then redistributed – i.e. allocated – among NCBs according to respecve capital keys (the parcipaon shares of each NCB to the ECB’s capital). Monetary income has relevance for current debates such as that concerning the high fiscal costs of an ample reserve regime as a result of the abundant reserves banks hold in the deposit facility of their respecve NCBs. These costs are in fact redistributed through the allocaon of monetary income. Nonetheless, exactly how monetary income is pooled and subsequently allocated between Eurosystem NCBs remains rather enigmac. The aim of this paper is to explore how monetary income is both pooled and allocated. This seems a useful task beyond the aforemenoned debate to dissipate other puzzling issues like the costs of TARGET2 imbalances. A more detailed disseminaon from the relevant authories as to the process by which profits/losses are pooled and subsequently allocated is however in our view warranted. * Sergio Cesarao is full professor of European monetary and fiscal policy, Diparmento di economia polica e stasca, Università di Siena (Italy); Eladio Febrero is associate professor of Economics at the University of Caslla-La Mancha (Spain); George Pantelopoulos, lecturer in economics, University of Newcastle (Australia). We thank Servizio Bilancio from the Bank of Italy for advice on specific points, and Giuseppe Ferrero and Marc Lavoie for some comments. The responsibility for any remaining errors or misunderstandings are, of course, exclusively ours. Corresponding author: Cesara[email protected]
2 1. Introducon Aer the rise of interest rates in 2022, naonal central banks (NCBs) belonging to the Eurosystem – similar to the other central banks – began paying considerable interest to banks on the excess reserves (ER)1 held by the laer in their respecve NCB’s deposit facility (DF) as a consequence of the several ‘balance sheet policies’ undertaken in former years.2 At the present rate, and given current stocks of excess reserves, banks are receiving slightly less than €130 billion per year. This has sparked off some debate on how to avoid taxpayers paying a hidden subsidy of sorts to commercial banks (e.g. European Parliament, 2023; De Grauwe and Ji, 2023; McCauley and Pinter, 2024; Tucker, 2022). A soluon a là De Grauwe and Ji (2024) of increasing the mandatory reserve coefficient and paying zero interest on required reserves (RR) is problemac, since reserves are unevenly distributed among the euro area jurisdicons.3 As a consequence, a higher reserve coefficient would in all likelihood adversely affect banks in Southern Europe where reserves are much less abundant than in Northern jurisdicons. Moreover, it would also appear that the ample reserve regime is here to stay given the enduring larger demand for reserves from banks (Åberg et al. 2021; Schnabel 2023, 2024; Altavilla et al. 2024) which compresses the space for increases to minimum reserve requirements (Hudepohl et al, 2024).4 A related issue concerns the distribuon of the high fiscal costs of the DF. In the Eurosystem profits and losses associated to monetary policy operaons are pooled by all NCBs to form the Eurosystem’s so-called ‘monetary income’, which is then subsequently allocated among NCBs according to their 1 Reserves are issued by the central bank and are used by commercial banks to execute interbank payments, to promptly meet cash withdrawals by customers, to comply with reserve requirements (if in place), as a financial buffer, and to fulfil regulaons concerning safe assets. Central Bank Digital Currencies (CBDC) would give the general public direct access to this kind of money (Cesarao and Febrero, 2023). For a further discussion of CBDC, see e.g. Pantelopoulos (2024a), Auer et al. (2024). 2 This includes the numerous large-scale asset purchase programs (aka Quantave Easing), in addion to other non-convenonal measures (e.g. Targeted Longer-Term Refinancing Operaons) (see e.g. Baglioni, 2023). 3 The maldistribuon may also concern different banks (say small and large) within a single jurisdicon (see e.g. Fricke, Greppmair, and Paludkiewicz, 2024). 4 Two opposite opinions on the feasibility of a return to a scarce reserve regime (or ‘corridor system’) are, respecvely, Borio (2023) and Altavilla et al. (2024).
3 respecve capital keys (the parcipaon shares of each NCB to the ECB’s capital).5 This pooling includes the interest payments on reserves held in the deposit facilies of NCBs.6 Given that – as noted above – excess reserves are maldistributed among the euro area jurisdicons, one may then infer that NCBs in the jurisdicons where the proporon of excess reserves is above the country’s capital key share the related ‘excess’ costs of the DF vis-à-vis the NCBs where the poron of excess reserves is below the country’s capital key. This leads to the queson of whether aer the pooling and allocaon of monetary income process some NCBs are indirectly subsidizing the banking system of other jurisdicons. This issue has been recently raised by Baglioni (2024). While we defer a full discussion of this issue to further research, in the present paper we deal with some preliminary methodological noons concerning monetary income that are oen sidestepped or deferred to a literature that we found to be rather lacking (e.g. Belhocine et al, 2023; Sonnemberg, 2023). In our opinion this preliminary work will shed some light on the pooling and sharing of the costs of the deposit facility, which sits within a complex web of financial flows involved in the Eurosystem’s monetary income, thereby allowing a more comprehensive analysis of the possible existence of unwarranted South-North financial flows. As a maer of fact, the process by which the monetary incomes of NCBs are pooled and subsequently allocated is sll a rather lile-known subject.7 In secon 2 we shall introduce the concept of monetary income, while secon 3 discusses in detail the single items that enter in its calculaon, with secons 4 and 5 providing some hypothecal examples as to how monetary incomes are pooled and allocated between Eurosystem NCBs. Secon 6 takes stock of the methodological results, with secon 7 providing an exploratory analysis of the 5 A NCBs’ share of ECB capital is calculated using a key which reflects the respecve country’s share in the total populaon and gross domesc product of the EU. 6 Broadly speaking, in the Eurosystem commercial banks have two major accounts at their respecve NCB where they hold their reserves: a current account in which minimum reserve requirements are fulfilled, and a deposit facility, where any excess reserves above and beyond minimum reserve requirements are kept. Unl the global financial crisis – in a way that was funconal to the tradional corridor system – liquidity within the minimum reserve deposit account were typically remunerated at a rate higher than that of the deposit facility. During the past decade or so (aer the adopon of a floor system), the remuneraon on excess reserves offered by the two accounts on ER has oen been idencal, meaning that banks were indifferent as to where to hold them. However, aer interest rates on ER held in the DF returned from mid-2022 in posive territory, in mid-2023, the interest rate paid on minimum reserves was brought to zero to somewhat reduce the costs of the ample reserve regime (ECB 2023a). 7 This reminds of the uneasiness of economists when the German economist Werner Sinn raised the queson of TARGET2, a topic that required someme to be fully understood (Bindseil and König, 2012; Cesarao, 2013; Febrero and Uxò 2013). Cesarao (2023) provided a preliminary exploraon of monetary income in the context of the debate on the fiscal costs of the ample reserve regime.
4 pooling and sharing of monetary income in the last ten years for three Eurosystem NCBs. Finally, the conclusion underlines the main methodological results and summarises some implicaons of our work for the ongoing debate with regard to the fiscal costs of excess reserves. 2. Monetary income Table 1 shows a synthesis of the Profit and Losses Accounts of the Bank of Italy, the Bank of Spain and the Bundesbank for the years 2021-2023. The (perhaps) most important item contained in the Profits & Losses (P&L) accounts is ‘net interest income’ (item 1). This is the result of interest payment flows, most of which are related to monetary policy operaons. Interest income and expenses related to monetary operaons are then pooled (item 5.1), following some specific rules, and contribute towards the Eurosystem’s ‘monetary income’ and then allocated (item 5.2) according to capital key. The net result of allocang monetary income is reported as item 5. The raonale of this process is of pooling and redistribung those interest flows that originated by decisions taken at the Eurosystem level and whose consequences on NCB’s P&L accounts must therefore be shared.8 Table 1 – A selected synthesis of the P&L account of three Eurosystem NCBs (euro millions) Source: CB Annual Accounts. The definions of monetary income provided by various NCBs are (expectedly) consistent, so we mainly refer to the one offered by the Bank of Italy (2023). The Eurosystem’s monetary income is the 8 The Bank of Spain and the Bundesbank transfer the bulk of item 11 (i.e. final profits) in total to their respecve Treasuries as dividends when the outcome is posive. By contrast as it is partly owned by private instuonal agents, the Bank of Italy pays taxes out of item 11, then transfers a proporon to those private owners and, finally, sends what remains to the government.
5 result of the pooling of NCBs’ own monetary income. This is derived from the profits or losses calculated based on certain interest income and expenses obtained from a list of earmarkable assets held by each NCB against a list composing its liability base. Both lists regard operaons associated to the implementaon of common monetary policy decisions, so it is understandable relave profits and losses and pooled are ulmately shared. The calculaon of a NCB’s own monetary income should be carefully disnguished from the NCB’s own P&L account, as exemplified in Table 1. If the analogy helps, the calculaon of a NCB’s own monetary income is similar to an income tax declaraon in which some profit and losses are reported to a higher authority that will eventually bestow a posive or negave tax (a rebate). For instance, for the Bank of Italy (2023, p. 73), “the net result of the allocaon of monetary income in 2022 [was] equal to €2,375 million (…).9 This was the difference between the monetary income pooled by the Bank, amounng to a negave €1,162 million, and that redistributed [i.e. allocated] to the Bank, equal to a posive €1,213 million”. This final posive or negave tax will contribute to the final financial result (item 11 of Table 1) of the NCB modifying the inial ‘net interest income’ (item 1). As the Deutsche Bundesbank (2024) explains: “The monetary income of the naonal central banks is inially reflected in profit and loss item 1 ‘Net interest income’ [see Table 1], while any unequal allocaon among naonal central banks is balanced out via profit and loss…‘Net result of pooling of monetary income’”. Importantly, any transfers to and from the ECB are effectuated as transcripons within the TARGET2 payment system.10 For instance, the pooling of monetary income by the Bank of Spain to the ECB is finalised by the former incurring a TARGET2 liability vis-a-vis the ECB. Also, it is important to note that the ECB’s own profit and losses do not contribute to the determinaon of the Eurosystem’s monetary income: the ECB just reallocates the pooled monetary income of NCBs according to 9 Net monetary income as shown in Table 1 includes €5 million, pertaining to the recalculaon of amounts for previous years. 10 TARGET2 is in principle a payment plaorm and T2 imbalances just accounng entries. Lato sensu, TARGET2 liabilies and claims may be however considered as central bank money that both the NCBs and the ECB issue (and accept) for their reciprocal intra-Eurosystem payments. In this capacity TARGET2 claims and liabilies contribute to the net internaonal investment posion (NIIP) of a country, and are recorded within the financial account balance in the balance of payments, in the sub-account ‘Other Investment’. For a further discussion of NIIP, see e.g. Pantelopoulos (2024b).
6 capital-key, as noted above. In case the ECB is making losses it might, however, acvate only a paral or a nil redistribuon.11 3. Earmarkable assets and the liability base To calculate its local monetary income to be pooled – that ulmately forms part of the Eurosystem’s monetary income – each NCB refers to interest income and expenses relave to a list of ‘earmarkable assets’ and to a ‘liability base’ both referring to monetary policy operaons. We may note here that these interest income and expenses are included in items 1.1 and 1.2 of Table 1. The process of pooling and allocaon of the Eurosystem’s monetary income as reflected in the ‘Net monetary income allocated’ (item 5 in Table 1) will modify the inial impact of these interest income and expenses. Table 2 details the components of both earmarkable assets and liability base with the respecve interest rate at which the income to be pooled is calculated. This rate is somemes convenonal and set equal to the rate on main refinancing operaons (MRO), and not the rate actually perceived (see e.g. Bank of Spain, 2023, p. 63; Bank of Finland 2023, pp. 212-213). Both lists concern operaons related to monetary policy (including the smooth funconing of the payment system). Table 2 – Earmarkable assets and liability base Earmarkable assets Liability base (a) lending to euro-area credit instuons relang to monetary policy operaons (effecve interest rate on refinancing operaons) (a) (a’) banknotes in circulaon (i = 0) (b) securies held for monetary policy purposes (public securies: iMRO; corporate bonds: effecve interest rate) (b’) liabilies to euro-area credit instuons related to monetary policy operaons denominated in euros (effecve interest rate: iRR and iDF on required and excess reserves, respecvely). (c) intra-Eurosystem claims arising from the transfer of reserves to the ECB (iMRO) (d) net intra-Eurosystem claims resulng from TARGET2 transacons (iMRO) (d’) net intra-Eurosystem liabilies resulng from TARGET2 transacons (iMRO) 11 Moreover, we shall note in secon 4 that in a couple of cases the ECB regulates interest payments with the NCBs via monetary income.
7 (e) net intra-Eurosystem claims related to the allocaon of euro banknotes within the Eurosystem (iMRO) (e’) net intra-Eurosystem liabilies related to the allocaon of euro banknotes within the Eurosystem (iMRO) (f) a pre-set amount of gold holdings and gold receivables in proporon to each NCB’s subscribed capital key (i = 0) Source: Bank of Italy (2024, p. 76; Bank of Spain, 2024, p.58) As a reference rule governing the difference between what a NCB contributes/pools to the Eurosystem’s monetary income and what it receives back, the Bank of Italy (2023, p. 74) suggests: “The difference between the monetary income pooled by each NCB and the amount redistributed to that NCB, which may be larger or smaller, …depends on two factors: (a) the first (income effect) relates to possible differences between NCBs regarding the interest income received on specific earmarkable assets and the interest expense due on some components of the liability base; (b) the second (composion effect) arises from the fact that the amounts of the above assets and liabilies in the NCBs’ balance sheets do not generally coincide with their capital keys.” In other words, if the composion of earmarkable assets and liability base for each NCB were in line with their respecve capital key, and if the same interest rate were applied on each asset or liability, then a NCB will pool the exact same amount of monetary income that it will subsequently receive back following the monetary income allocaon process. This is not always the case, as we shall see. We shall also note that the menoned ‘reference rule’ is not obviously (or only relavely) applicable to the cases of the net intra-Eurosystem claims and liabilies related to TARGET2 (items d and d’ of Table 2) and to the allocaon of banknotes (items d and d’) (concerning the funconing of the payment system rather than monetary policy operaons). It should finally be noted that it is not necessarily true that, respecvely, earmarkable assets bring profits and the liability base causes losses. For instance, with reference to the earmarkable assets listed in Table 2, item (a) contains refinancing operaons that in the last decade have been associated with negave interest rates, bringing losses to NCBs; conversely, bank reserves comprised in item (b’) on the liability side have been remunerated at negave rates again in the last decade, bringing profits to NCBs. As menoned, interest accrued/paid by a single NCB on earmarkable assets and liabilies are pooled contribung toward the Eurosystem’s monetary income, and then allocated according to capital key. However, an addional rule dictates that: “Where the value of an NCB’s earmarkable assets exceeds or falls short of the value of its liability base, the difference (i.e. the ‘GAP’) is considered to bear (or
14 important conclusion here is that since nothing is shared, the enre actual revenues obtained from public securies remains at the respecve NCBs. To elucidate the point, we provide a simple example precisely focused on public securies with the two NCBs of the previous example (Table 5). In this case, a large-scale asset purchase operaon is implemented where a total of €300 billion of public securies are purchased, where each NCB purchases securies in its own jurisdicon according to capital key, again 1/3 (NCB-A) and 2/3 (NCBB). This operaon leads to the creaon of reserves (a liability) deposited in the deposit facility where commercial banks earn a rate iR (reserves are included in the liability base – i.e. item (b), as we shall see in the next secon). Table 5 - A public securies purchase operaon NCB-A NCB-B Bonds: +100 R: +100 Bonds: +200 R: +200 We can now calculate the virtual net monetary income on public securies for each NCB that is pooled and subsequently allocated. We assume iMRO = 2% and, for the sake of the argument, iR = 0%.22 We have: NCA-A interest income to be pooled = 100 x 2% = 2; NCA-B interest income to be pooled = 200 x 2% = 4. Total monetary income pooled is = 6. This is redistributed according to capital key; i.e. 2 to NCB-A and 4 to NCB-B. Net monetary income is therefore zero for both. In other words, each NCB confers to the pool precisely what it receives back from it. In the end, in this case the exercise of pooling and allocaon has no effect, and not surprisingly since both the income and the composion effects are absent. NCB-A will retain the income on its capital key share of public securies purchases, calculated at the actual interest rates, and the same will apply for NCB-B. An income effect would be present if we had considered corporate assets whose income is pooled at the actual rates which plausibly differs among jurisdicons. Another case is if the public securies purchasing operaon had not been implemented according to capital keys but, say, to rescue a specific jurisdicon, as in the case of the menoned SMP or of Draghi’s famous OMT 2012 statement. 22 To reiterate, in our examples we assume that all the items, except for those under scruny, are irrelevant (or a zero-interest rate is applied to them).
15 In this case if, say, the Bank of Italy were authorised to buy a certain amount of domesc public bonds, with the risk shared over the enre Eurosystem, it seems natural that the Bank of Italy would have to share with other NCBs the relave revenues.23 During the APP (launched 2015) and PEPP (launched 2020) the idea was instead that risk for public securies remained mostly naonal. The modalies and results of Table 5 reflect these last experiences: no risk and correspondent revenues are shared on public securies. The Deutsche Bundesbank (2021, p. 75) seems to confirm this interpretaon when it writes: “It is assumed that no income is generated from […] the covered bonds purchased under the CBPP and CBPP2 as well as the government bonds (including regional government bonds and bonds issued by eligible agencies located in the euro area) purchased under the PSPP and PEPP generate income commensurate with the applicable main refinancing rate, as the ECB Governing Council has ruled out the possibility of pooling the risk and returns arising from these instruments among the naonal central banks.” A paral excepon to the uniform composion rule concerned PEPP purchases that, in some instances, could temporarily deviate from a NCB’s capital key to pacify tensions in specific markets (see ECB, 2023b, 2024a). Looking at Table 3 we may indeed observe that in 2023 what the Bank of Italy pooled as income from the ‘Securies held for monetary policy purposes (not subject to risk sharing)’ (€23,984 million), was only roughly returned to it (€21,546). In this specific case Italy’s net contribuon to the pool (€2,438) was presumably due to the sharing (at a convenonal rate) of interest revenues from the (authorised) purchase of domesc government assets above its assignment. A further complicaon regards the case in which asset purchases were to be implemented by a NCB in other jurisdicons of the monetary union, as transpired during the APP and PEPP. This generates TARGET2 imbalances and is dealt with below (point (d)). (c) Intra-Eurosystem claims arising from the transfer of reserves to the ECB. This item refers to the claims in euro that NCBs have recorded on their balance sheets vis-à-vis the ECB for the transfer of foreign reserves (according to capital key) in its favour when the Eurosystem was established (or when a new NCB enters the Eurosystem). In summary, the ECB includes on the assets side of its balance sheet the foreign reserves it has obtained and in counterpart a eurodenominated debt vis-à-vis the NCBs, which in turn report a claim in their balance sheet. This claim 23 We guess that the same applied to the Securies Market Programme, the limited program of purchases of government bonds by Southern NCBs launched during the course of 2010.
16 is included in earmarkable assets in the same way as any other intra-Eurosystem credit/debit (such as those related to TARGET2 and adjustments related to the issuance of banknotes, as we shall shortly see); it has no redistribuve effect as the income pooled on these claims (at the convenonal iMRO rate) and credit of each NCB are both aligned to its capital key. Table 3 shows that, for instance, in 2023 both the Bank of Italy and the Bank of Spain pooled an amount of income precisely equal to that redistributed to them. (d) Net intra-Eurosystem claims resulng from TARGET2 transacons In a parcular set of circumstances, payments across euro area jurisdicons may generate T2 imbalances consisng of claims held by some NCBs vis-à-vis the ECB mirrored by liabilies by the remaining NCBs vis-à-vis the ECB. On these liabilies NCBs pay monthly interest at the prevailing marginal interest rate that the ECB transfers to the remaining NCBs. Therefore at a first glance, it would be envisioned that T2 imbalances entail losses for some NCBs and symmetric profits for the remaining NCBs. However the rules concerning the pooling and allocaon of monetary income prescribe that at the end of the year each NCB can deduct (i.e. must transfer) its aggregate losses (profits) from (to) the Eurosystem’s monetary income. This has two results: (1) T2 imbalances have no final effect on the NCB own actual P&L account since the deducon (addion) from (to) the NCB’s transfer to the Eurosystem’s monetary income cancels out the inial T2 losses (profits) incurred over the year; (2) In the calculaon of the Eurosystem’s monetary income the losses brought by some NCBs are precisely cancelled out by the symmetrical profits brought by the remaining NCBs, so the net effect on monetary income is nil (i.e. nothing is redistributed). As Cour-Thimann (2013, p. 29) authoritavely explained when, a decade ago, the controversy over TARGET2 blew up: “Target balances are de facto not remunerated within a cohesive monetary union. (…) First, it is important to recall that the size or distribuon of Target balances have no impact on the monetary income of the individual NCBs within the Monetary Union. Target balances in first instance bear monthly payments at the prevailing marginal interest rate in the main refinancing operaons (in full allotment equal to the main refinancing rate). These interest payments flow from NCBs with Target liabilies via the ECB to NCBs with Target claims. However, at year-end, when the NCBs pool their monetary income net of expenses in the context of the income-sharing scheme, these interest payments are taken into account and thus offset. Sll, in the context of perceived risk on the cohesion of the Monetary Union, the fact that the Target balances accrue the monthly interest payments might be seen as remunerang such risk."
17 The raonale for this eventually ineffecve accounng rule is provided in the last passage: in the case where a country with T2 liabilies leaves the union during the year – thereby reneging its T2 debt – it would have at least paid an interest on its T2 liabilies during the year (which is ‘returned’ if it doesn’t eventually leave). Symmetrically, the NCB with T2 claims receives a remuneraon for the risk it has incurred over the year (of seeing its T2 claims renegaded), and ‘returns it’ when, at the end of the year, the risk has not materialised. We may modify the example of Table 5 in order to introduce some T2 imbalances (Table 6). In the modified example NCB-A purchases domesc government bonds in a financial market located in jurisdicon B – where internaonal investment funds presumably operate – as part of a public securies purchase programme. In pracce, NCB-B (say the Bundesbank) buys €100 billion of country A public securies in its jurisdicon (say Frankfurt) on the behalf of NCB-A (say the Bank of Italy) which ‘pays’ by booking a T2 liability. According to the ECB, this has been the main source of TARGET2 imbalances aer 2015 (e.g. Eisenschmidt et al. 2017). Table 6 – NCB-A purchases country A bonds in a market located in the jurisdicon B NCB-A NCB-B Bonds: +100 T2: +100 T2: +100 R: +100 Bonds: +200 R: +200 Commercial bank R: +300 D: +300 We assume that securies offer no return and reserves are not remunerated, but that T2 liabilies require paying say 2% (our assumed interest rate on MRO) over the year. If payments are made as ‘accounng annotaons’ (i.e. result in an increase in T2 liabilies), on 31 December, the two NCB balance sheets would appear as transcribed below in Table 7: Table 7 - NCBs pay/receive a return on T2 liabilies/claims, annotang such payment in their P&L accounts NCB-A NCB-B Bonds: +100 T2: +102 Equity: -2 T2: +102 R: +100 Equity +2 Bonds: +200 R: +200 Commercial bank R: +300 D: +300
18 We can now calculate each NCBs’ monetary income that must be pooled. Given the hypothesis, the only earmarkable asset (or liability) are TARGET2 claims/liabilies. With interest on T2 claims/liabilies set at iMRO = 2%, profits/losses reported to the Eurosystem would be: NCB-A= - 2% x T2 = – 2% x 100 = -2 (this loss corresponds to a new T2 claim for NCB-A) NCB-B = 2% x T2 = 2% x 100 = 2 (this profit corresponds to a new T2 liability for NCB-B) The Eurosystem’s monetary income would thereby equate to zero. There is nothing to share here. TARGET2 is also irrelevant for the ‘Net (or gross) profits for the year’ (Table 1) for both NCBs since their respecve loss or profits (‘posive or negave equity’) – although inially counted in the ‘net interest income’ (item 1, Table 1) – are later compensated by the corresponding deducon in the ‘income declaraon’ to the Eurosystem. De facto, interest on TARGET2 claims/liabilies are thereby irrelevant both in the NCBs’ own P&L account and for the Eurosystem’s monetary income: if a NCB has a T2 liability (claim) and makes monthly payments (receives profits) over the year, at the end of the same year it will deduct those payments (add those profits) from (to) the income it confers to the Eurosystem’s monetary income. This is shown below in Table 8. Finally, the TARGET2 losses and profits conferred to the Eurosystem’s monetary income cancel out so there is no ‘dividend’ (posive or negave) from TARGET2. Table 8 shows that the final outcome is analogous to that of Table 5 in which purchases were done by each NCB in its own jurisdicon.24 Table 8 - NCBs deduct/add losses/profits from/to the monetary income declared to the Eurosystem NCB-A NCB-B Bonds: +100 T2: +102 Equity: -2 T2: +102 R: +100 Equity: +2 T2: +2 (conferral of losses to MI) T2: -2 (conferral of profits to MI) NET NET Bonds: +100 T2: +100 T2: +100 Bonds: +200 R: +300 Table 3 (above) showed that in 2023 the Bank of Italy pooled a T2 loss of € 22,769, whereby it received back a ‘rebate’ of €2,388. Similar results are reported by the Bank of Spain. We must take a moment to explain such a result since we expected a zero ‘rebate’ (net-result). Let us first recall that while the monthly payments on the T2 liabilies were reported among the interest expenses in 24 It makes sense that NCB-A should not pay a penalty for having executed large-scale asset purchases in the most efficient way, e.g. in other jurisdicons.
19 the Bank of Italy’s ‘Net interest income’ (item 1 of Table 1), the ‘discharge’ of these payments in the calculaon of the monetary income to be pooled with the Eurosystem de facto zeroed those interest expenses for the calculaon of the NCB final financial result. This is considering only the NCBs, as there remains a posive difference, the menoned ‘rebate’ due to the ECB liability posion on TARGET2. It must be recalled that the ECB has a debit T2 balance vis-à-vis the NCBs (not specifically vis-à-vis the Banca d'Italia or the Bank of Spain) on which it pays interests at the iMRO rate. This interest income on TARGET balances is allocated on a capital key basis, within the net result of the redistribuon of monetary income. The Bank of Italy in 2023 recorded, for instance, a benefit of 16.85% of the total interest expense paid by the ECB which explains the above menoned ‘rebate’. The size of the ECB's T2 debit balances are mainly related to the purchases of monetary policy securies (the ECB parcipated in both APP and PEPP by buying 10% of bonds). Euro area commercial banks, however, have account relaonships only with the NCBs and not with the ECB. As a result, the purchases by the ECB were facilitated through NCBs, according to capital key, by booking a T2 liability vis-à-vis each NCB. On these liabilies the ECB paid an interest rate that account for the above ‘rebate’.25 It seems therefore that the ECB employs monetary income to regulate some payments with the NCBs. (e) Net intra-Eurosystem claims related to the allocaon of euro banknotes within the Eurosystem. The issuance of banknotes in the Eurosystem by each NCB should in principle be in line with their respecve capital key (with 8% of the issuance appanage of the ECB). Nonetheless, as the issuance of banknotes is dependent on the public’s demand (i.e. the demand for banknotes is endogenously determined), whether a single NCB has issued an amount of banknotes above or below its capital key entlement is only calculable ex post. 25 In the notes to the ECB's 2020 balance sheet, with reference to the item ‘Other liabilies within the Eurosystem (net)’, ECB (2021, p. 49) observes that “…in 2020 this item consisted predominantly of the TARGET2 balances of euro area NCBs vis-à-vis the ECB. […] The increase in the net TARGET2 liability resulted mainly from the net purchases of securies under the PEPP and the APP, which were seled via TARGET2 accounts.”
20 Any eventual over-issuance is penalised, and any under-issuance ‘rewarded’, at the rate applied to refinancing operaons.26 Yet as in the case of T2, any possible profit (or loss) as a consequence of a NCB under-issuing (over-issuing) banknotes realised over the year is de facto deleted, as profits made by under-issuing NCBs are precisely offset by the losses incurred by over-issuing NCBs, so from the point of view of pooled Eurosystem monetary income there is nothing le to redistribute once the pooling and allocaon process is complete. For instance, suppose that given the demand for banknotes in the respecve jurisdicons, NCB-A issues €80 and NCB-B €220 billion of banknotes. With the respecve capital keys of 1/3 and 2/3, the ex-post entlements would be 100 and 200 respecvely. NCB-B will then pay a penalty during the year to the Eurosystem on its over issue (20€) at the MRO rate (i.e. iMRO) by booking a TARGET2 liability and NCB-A will be remunerated in a symmetrical manner due to its under issuance (20€) (by way of posing a TARGET2 claim). Having paid a penalty, NCB-B will report a loss to the Eurosystem’s monetary income, while NCB-A will report a profit. The net effect for both NCB-A and NCB-B is nil. As in the case of T2, such accounng rules are enforced as if a euro area member were to leave the monetary union aer having over-issued euros during the year relave to its respecve capital key (in this case, NCB-B), it would have at least paid interest during the year (which is returned if it doesn’t leave). Interesngly, Germany overissues banknotes (see e.g. Deutsche Bundesbank, 2022). For example, at the end of 2023, the Deutsche Bundesbank (2024, p. N/A) reported that while in principle it could distribute €377,036 million of banknotes, the value of the euro banknotes actually issued was €920,705 million (€543,670 million more than its allocated amount). On this difference an interest rate equal to the iMRO rate was paid by the Bundesbank over the year (that via ECB was transferred to under-issuing NCBs). The Bundesbank subsequently pooled this loss via the pooling of its monetary income. On the other hand, the Bank of Italy in 2023 issued banknotes below its assignment. Table 3 above shows indeed that the Bank of Italy pooled €1,877 million of monetary income with regard to banknote issuance and received back a negave ‘rebate’ of €-812 million. To this end, the Italian central bank reported in its P&L account a loss concerning the allocaon of banknotes of €-2,689 million. An even larger post-pooling loss was reported by the Bank of Spain 26 ‘The respecve share of the total value of euro banknotes in circulaon due to each central bank in the Eurosystem is calculated on the last business day of each month in accordance with the key for allocang euro banknotes’ (Deutsche Bundesbank, 2022, p. 42).
21 that pooled €6,933 million and received back a negave ‘rebate’ of €-570 million for a final loss of €-7,503 million. Why the €-812 million and €570 million residuals for the Bank of Italy and the Bank of Spain, respecvely? As much as in the case of TARGET2 residuals, there should be zero residuals. We must refer again to the role of the ECB. Although the ECB is allocated an 8% share of the total value of the euro banknotes in circulaon, the ECB does not technically put banknotes into circulaon. Banknotes are always put into circulaon by the NCBs that have a relaonship with the euro area banking system. The value of banknotes in circulaon found in the balance sheets of the NCBs is a convenonally adjusted value: if the total value of banknotes in circulaon of the whole Eurosystem is 100, 8% is convenonally allocated to the ECB and the remaining 92% is allocated to the NCBs in proporon to their respecve capital keys. The difference between the amount of banknotes issued by each NCB according to capital key and the actual amount issued is offset by intra-Eurosystem claims/liabilies (different from TARGET2 balances but always remunerated at the MRO rate). The relave profits or losses are then pooled and shared via monetary income, an operaon that in pracce cancels them out. The ‘residual’ concerns the ECB that, against the 8% of banknotes on its balance sheet, matures a claim against the NCBs for the same amount. Against this claim, all NCBs pay interest to the ECB (€-4,817m in 2023) according to capital key; the Bank of Italy's share (capital key of 16.85%), for instance, corresponds precisely to a payment of €-812m. (f) A pre-set amount of gold holdings and gold receivables in proporon to each NCB’s subscribed capital key. It was decided to include gold among earmarkable assets, for a total amount fixed for the enre Eurosystem and broken down for each NCB according to the capital key. Therefore, since for the purposes of monetary income calculaon each NCB includes an amount of gold (and gold-equivalent claims transferred to the ECB) aligned with the capital key, there are no redistribuve effects – the more so since, as seen in Table 2 “Gold is not considered to generate interest” (Bank of Italy, 2023, p. 74). 5. Calculaon of net monetary income: liability base (a’) Banknotes in circulaon Refinancing operaons and the other sources of liquidity that we find among the earmarkable assets (point (a)) must be matched by corresponding liabilies. Refinancing operaons create base money
22 that consist of banknotes and reserves. Banknotes do not generate any sort of interest rate per se so this item is irrelevant for the actual formaon of monetary income (nonetheless as divulged above, the uneven allocaon of banknotes among NCBs leads to interest payments, but the related profits and losses are later zeroed through pooling etc.). (b’) Liabilies to euro-area credit instuons related to monetary policy operaons denominated in euros This item concerns the other component of base money consisng of mandatory (or required) reserves (RR) and excess reserves (ER). In a classical corridor system (or scarce reserve regime), RR are held by banks at their respecve NCB, and yield an interest rate equivalent to that applied to MROs, while excess reserves are normally held in the deposit facility (DF) but yield a lower interest rate. This applies also to a floor system (or ample reserve regime, see ECB, 2024b), but variaons are possible. As already noted, for instance, given the high returns banks were receiving on excess reserves held in the deposit facility in the Eurosystem, the ECB since July 2023 brought the remuneraon of RR to zero so as to reduce ‘the overall amount of interest that needs to be paid on reserves in order to implement the appropriate stance’ (ECB, 2023a) – a ‘mini tiering‘ a la De Grauwe (2023). Be this as it may, the losses NCBs bear on remunerated reserves are pooled and shared. Importantly, the distribution of excess reserves and related losses are not however in line with respective capital keys. And since the interest rate applied is uniform, a composition effect is active here, but not of an income effect. Let us take the example of Table 6 (reproduced below as Table 9) in which NCB-A purchases domesc government bonds in a financial market located in a foreign jurisdicon (i.e. in jurisdicon B) as part of a public securies purchase programme. For simplicity we suppose that mandatory reserves do not exist and that all reserves are deposited in a deposit facility that yield a posive interest rate, say, iDF = 1%. As usual we neglect the other items (or assume they yield a zero-interest rate). Table 9 – NCB-B bears the costs of excess reserves NCB-A NCB_B Bonds: +100 T2: +100 T2: +100 R(DF): +100 Bonds: +200 R(DF): +200 Commercial bank
23 R: +300 DF: +300 We can now calculate each NCBs’ monetary income to be pooled. Given the hypothesis, the only liability base’s component producing an income (or loss) is the deposit facility. If the DF yields a return to commercial banks of, say iDF = 1%, each NCB’s monetary expenses to be pooled would therefore be: NCB-A = - 1% x R(DF) = – 1% x 0 = 0 NCB-B = - 1% x R(DF) = -1% x 300 = -3 The Eurosystem’s monetary income would equate to -3, which will be redistributed according to capital keys; i.e. NCB-A would receive -1 and NCB-B -2. NCB-A will receive a net monetary income of -1 – 0 = - 1. Hence, even though NCB-A would initially report in its P&L account zero interest income, following the pooling and allocation process it would incur a net monetary income (a loss) of -1. By contrast, NCB-B will post a net monetary income of -2 – (-3) = + 1 in the aftermath of the pooling and allocation process; i.e. NCB-B reported in the P&L account an interest expense of -3, but after the positive result of the net monetary income will see the loss reduced to -2. In this way, the cost of the excess liquidity is re-proportioned so that each NCB bears its k-key share (in the period 2015 to mid-2022 a negative interest rate penalised excess reserves, so that NCBs collected a positive income; this was then pooled and allocated as per the usual process). (d’) Net intra-Eurosystem liabilies resulng from TARGET2 transacons This item has already been dealt with in the earmarkable asset side. (e’) Net intra-Eurosystem liabilies related to the allocaon of euro banknotes within the Eurosystem This item has also already been dealt with in the earmarkable asset side. 6. Some methodological results The Eurosystem's monetary income is the result of each NCB pooling their interest income and expenses associated with monetary policy operations or the functioning of the payment system. This pooled monetary income is then shared/allocated according to each respective NCB’s capital key. There will be a redistribution if either a composition and/or an income effect are present. A
30 Table 12: Bank of Spain – selected data from the P&L accounts (millions of euro) As is widely known, 2015 to mid-2022 was characterised by a zero rate on the MRO, a negave rate on the DF interest rate policy, and by an aggressive balance sheet policy (Rostagno et al., 2021). On the one hand, negave interest rates on excess liquidity produced an interest income for NCBs. On the other hand, this was accompanied by increasing interest expenses due to the negave rates on longer term refinancing operaons (this cost was parcularly acute between 2020 and mid-2022 when seeking recourse to ECB credit by banks was parcularly aracve). Over this period, the Bundesbank (Table 10) mostly incurred a negave net monetary income following the pooling and allocaon process, while net monetary income was instead posive for the other two NCBs (Tables 11 and 12). In this same period, interest payments on T2 imbalances (despite T2 imbalances increasing) and on under/over-issuance of banknotes had a negligible or nil role, given that iMRO was zero (the reader should recognise by now that, whatever the level of iMRO, the net impact of TARGET2 on the P&L account, once considered monetary income, is eventually zero). The reader will as well acknowledge that interest revenues on the no risk-sharing public securies purchases programs will not lead to any redistribuon. The menoned paern of monetary income was the likely result of a relavely larger resort of southern banks to TLTRO operaons (bringing interest expenses to southern NCBs), and of the modalies of APP and PEPP that generated abundant reserves in some northern jurisdicons (thereby producing interest
31 revenues to northern NCBs). Through the pooling and reallocaon of interest costs and revenues, this double movement generated a north to south NCB monetary income redistribuon. Line 21 of Tables 10-12 provides a rough indicaon of the net interest flows over the period between the local commercial banks and their respecve NCBs.29 Because of the reasons provided in the previous paragraph, during the period 2015 to mid-2022 the net interest flow was favourable to the Bundesbank vis-à-vis domesc banks, whereas for the two southern NCBs the interest flow was favourable to local banks. Things changed radically with the rise of interest rates from mid-2022. NCB revenues from refinancing operaons returned in 2023 to posive territory, parcularly for the Bank of Italy given the larger resort of local banks to TLTROs. The costs of excess reserves held in the deposit facility also became exorbitant. This outcome sparked off the debate regarding the fiscal costs of the ample reserve regime (as noted earlier in the paper). Costs regarding net transfers to banks (as shown in the last line in Tables 10-12) have been parcularly onerous for the Bundesbank, modest for the Bank of Spain, and negligible for the Bank of Italy (where recourse to TLTRO was extensive). With respect to the result of pooling and allocaon via monetary income of the high DF expenses (especially for the Bundesbank) and of the TLTRO revenues (parcularly for the Bank of Italy) from mid-2022, NCB monetary income was redistributed from the south to the north, at least for Italy, albeit not for Spain, which had an amount of reserves closer to its capital key.30 As explained above, with regard to T2 imbalances and the misallocaon of banknotes, the process by which monetary incomes are pooled and allocated has the funcon, de facto, of cancelling out profits or losses from the NCBs’ P&L accounts on these items. The reallocaon was, in a sense, unfavourable to the Bundesbank, parally migated by a reallocaon of monetary income in its favour due to the high interest expense on its over-issuance of banknotes. Despite sharing with the Bundesbank the high costs of excess reserves, the pooling of expenses associated with T2 liabilies, as well as the already menoned volume of reserves held in the deposit facility, maintained the Bank of Spain’s net monetary income in posive territory. Although the Bank 29 This is a rough indicator since the local interest income and expenses for the NCBs are then pooled and reallocated through monetary income (while of course symmetric income and expenses for commercial banks would not change). 30 Table 3 above shows that in 2023 the Bank of Italy entered in its P&L account a final loss, aer pooling and reallocaon, concerning the deposits of banks of €14,123 million, a figure larger than its inial loss of €7,850. The Bank of Spain reported a final loss €7,263 against an inial loss of €8,160 (Bank of Spain, 2024, p. 64).
32 of Italy shared with Spain a large debtor posion in terms of T2, a relavely low amount of pooled expenses on bank reserves plus a conspicuous ‘GAP’ brought Italian net monetary income for the first me into negave territory.31 Further, while remaining muted for the Bank of Italy, net transfers to banks (line 23) accelerated in the last two years for the Bank of Spain and especially for the Bundesbank. All in all, European banks have been, nevertheless, net winners vis-à-vis their respecve NCBs (in the sense that recent mes have more than compensated for the less ‘advantageous’ years – i.e. from 2015 to mid-2022).32 All this considered, both the Bundesbank, the Bank of Italy and the Bank of Spain have witnessed their final profits (line 18) worsening the inial net interest income (i.e. pre-pooling, line 1; see also Table 1). To this end, in 2023 all NCBs resorted to their accumulated financial buffers to bring transfers to the State to zero (Bundesbank and Bank of Spain), that remain negave for the Bank of Italy once aer a negave tax contribuon is considered.33 Both in 2021 and 2022 for which Bundesbank data are also available, the Bank of Spain and the Bank of Italy appear to have pursued, in connuity with previous years, a quite prudenal financial provision policy compared to Frankfurt. Regarding profit transfers to the State (line 20), over most of the period considered all NCBs posted favourable results, predominantly as a result of APP and PEPP policies; the Bundesbank and the Bank of Italy could return to the State the interest income earned on the purchased securies, neutralising the de facto fiscal costs associated with this poron of public debt (though not the Bank of Spain, with part of the difference funding financial buffers and another part its net balance with banks). Nonetheless in the last two years, such transfers to the government have dramacally shrunk and negave transfers have been avoided (but not for the Bank of Italy) so far by NCBs eroding their accumulated financial buffers, as noted above. Considering NCBs as part of the consolidated public sector, this erosion equates nonetheless to an overall loss of public net wealth (Cecche and Hilscher 2024, p. 6). The major factor underpinning this result has to do with the impressive remuneraon of the large stocks of excess reserves, which monetary income has spammed over all 31 This high GAP was possibly due to a high value of ANFA (see footnotes 12 and 14 above). The ‘Annual average Net Financial Assets according to ANFA’ in 2023 was (in EUR billion) €111.0 for Italy, €-124.0 for Germany, €-22.5 for Spain (source: ECB). 32 Italian banks are an excepon in 2023 because of their modest stock of excess reserves. 33 Final (gross) profits for the Bank of Italy were €-7,125 million in 2023. The tax deducon for the bank was €2,340 million and recourse to financial buffers €5,600 million. We thereby obtain a net (posive) profit of €815 million most of which the Bank of Italy transferred to the Treasury. However, without the negave tax, net profits would have been € -1,525 as indicated in Tables 1 and 10.
33 NCBs determining some not irrelevant subsidies from some NCBs to the banking system of other jurisdicons. 8. Conclusion This paper has shown the nature, composion and raonale behind the process by which monetary incomes are pooled and subsequently allocated between Eurosystem NCBs, which thus far has been an almost unknown element of Euro area monetary governance. Our contribuon will be helpful to dissipate many ambiguies/omissions present in some recent papers that have touched this topic upon with regard to, for instance, presumed interest payments associated with TARGET2 imbalances and to the uneven issuance of euro banknotes. Analogously, we also shed light on related ambiguies, namely surrounding the pooling of interest income from securies purchased during the course of various large-scale asset purchase programmes. Moreover, the paper has clarified quesons concerning the pooling and reallocaon of interest income and expenses associated with refinancing operaons and excess reserves. This last aspect has a direct impact on the ongoing debate over the exorbitant fiscal costs of excess reserves in the Euro area (and elsewhere), which has so far been omied a full discussion as to the role of how Eurosystem NCB monetary income’s are pooled and allocated. While postponing a complete policy discussion of this issue, we may argue here that once the pooling and allocaon of monetary income is given full consideraon, the uneven costs of excess reserves due to their irregular distribuon in the Euro area brings about a south to north redistribuon, as ancipated in Cesarao (2023) and argued by Baglioni (2024). We also show that negave interest rates on excess reserves inflicted on banks in the last decade cannot jusfy the present huge remuneraon which is currently leading to the depleon of NCBs accumulated financial buffers. Taking into account that banks benefited in the same period by negave rates on refinancing operaons, a preliminary glance of the data confirms that the present net transfers to banks is not jusfied at least from this point of view. Indeed, how to reconcile a monetary policy regime based on ample reserves while simultaneously minimising the fiscal implicaons of excess reserves is sll an open and challenging queson.
34 References Åberg, P., M. Corsi, V. Grossmann-Wirth, T. Hudepohl, Y. Mudde, T. Rosolin, F. Schobert (2021), Demand for central bank reserves and monetary policy implementation frameworks: the case of the Eurosystem, ECB Occasional Paper Series No 282 / September. Altavilla, C., M. Rostagno and J. Schumacher (2023), ‘DP18581 Anchoring QT: Liquidity, credit and monetary policy implementaon‘, CEPR Discussion Paper No. 18581. CEPR Press, Paris & London. Auer, S., Branzoli, N., Ferrero, G., Ilari, A., Palazzo, F., e Rainone, E. (2024) CBDC and the banking system, Banca d’ItaliaQuesoni di Economia e Finanza (Occasional Papers), n. 829. Baglioni, A. S. (2023). Monetary policy implementaon: Which "new normal"? Journal of Internaonal Money and Finance, vol. 141, 1-19. Baglioni, A. (2024) Implemenng monetary policy with excess reserves: fiscal implicaons for the euro area, IEP@BU Policy Brief, March. Bank of Italy (2023) Annual Accounts 2022, Rome. Bank of Italy (2024) Annual Accounts 2023, Rome.. Bank of Spain (2024), Annual Accounts 2023, Madrid. Bank of Finland (2023) Bank of Finland Annual Report 2022, Helsinki (available online). Belhocine, N., A.V. Bhatia, and J. Frie (2023) “Raising Rates with a Large Balance Sheet: The Eurosystem’s Net Income and its Fiscal Implications,” IMF Working Paper WP/23/145. Bindseil, U., & König, P. J. (2012). TARGET2 and the European sovereign debt crisis. Credit and Capital Markets–Kredit und Kapital, (2), 135-174. Borio, C. (2023) Getting up from the floor, BIS Working Papers No 1100 May. Cecchetti, S.G., and Hilscher, J. (2024) Fiscal Consequences of Central Bank Losses, NBER Working Paper Series, n. 32478. Cesarao, S. (2013) The implicaons of TARGET2 in the European balance of payment crisis and beyond, European Journal of Economics and Economic Policies: Intervenon, 10 (3), pp. 359– 382. Cesarao, S. (2020) Heterodox Challenges in Economics – Theorecal Issues and the Crisis of the Eurozone, Springer Nature Switzerland, Cham. Cesarao, S. (2023) Annotazioni sull’implementazione della polica monetaria: ieri, oggi, domani, Quaderni DEPS n. 904. Cesarao, S. and E. Febrero (2023) Central Bank Digital Currencies: a proper reacon to private digital money, Review of Keynesian Economics, 11 (4). Corsi, M., and Muddle, Y. (eds.) (2022) “The use of the Eurosystem’s monetary policy instruments and its monetary policy implementaon framework in 2020 and 2021”, European Central Bank Occasional Paper Series no. 304, September. Cour-Thimann, Philippine (2013) Target Balances and the Crisis in the Euro Area, Cesifo Forum, Volume 14, Special Issue April.
35 De Grauwe, P. and Ji, Y. (2023) Monetary Policies without Giveaways to Banks, CEPR Discussion Paper DP18103, April. De Grauwe, Paul and Ji, Yuemei (2024) How to conduct monetary policies: the ECB in the past, present and future. Journal of Internaonal Money and Finance, 143. Deutsche Bundesbank (2022) Demand for euro banknotes issued by the Bundesbank: current developments, Monthly Report April. Deutsche Bundesbank (2023) Annual Report 2022, Frankfurt (pdf available online) Deutsche Bundesbank (2024) Annual Report 2023, Frankfurt am Main (only available online) ECB (2021) Annual Accounts of the ECB 2020, Frankfurt am Main. ECB (2023a) ECB adjusts remuneration of minimum reserves, Press release, 27 July. ECB (2023b) FAQ on the public sector purchase programme, hps://www.ecb.europa.eu/mopo/implement/app/html/ecb.faq_pspp.en.html ECB (2024a) FAQ on the pandemic emergency purchase programme, https://www.ecb.europa.eu/mopo/implement/pepp/html/ecb.faq_pepp.en.html ECB (2024b) Changes to the operaonal framework for implemenng monetary policy, Statement by the Governing Council 13 March 2024, hps://www.ecb.europa.eu/press/pr/date/2024/html/ecb.pr240313~807e240020.en.html Eisenschmidt, J., Kedan, D., Schmitz, M., Adalid, R., and Papsdorf, P. (2017) The Eurosystem’s asset purchase programme and TARGET balances, Occasional Paper Series 196, European Central Bank. European Parliament (2023) Quantave ghtening in the euro area - Requested by the ECON commiee, Monetary Dialogue Papers, March Febrero, E., and J. Uxó (2013) Understanding TARGET2 Imbalances from an Endogenous Money View” Working paper DT-DAEF 2013/2, University of Caslla-La Mancha. Fricke, D., Greppmair, S. and Paludkiewicz, K. (2024) Excess reserves and monetary policy ghtening, Discussion Paper, Deutsche Bundesbank, No 05. Galli, G. e F. Neri (2023) https://osservatoriocpi.unicatt.it/ocpi-pubblicazioni-ha-senso-un-allarmeper-le-perdite-delle-banche-centralition of papers Hudepohl, T., Pamina, K., Linzert, T., Nguyen, B., Skrzypińska, M., & Vaz Cruz, L. (2024) How banks deal with declining excess liquidity, The ECB Blog, 18 June. McCauley R., Pinter, J. (2024) Unremunerated reserves in the Eurosystem, part 1: Heads I win, tails you lose; part 2: Tax incidence and deposit relocaon risks, Vox.eu, 15, 16 Jan. Pantelopoulos, G. (2024a) Navigang some misconcepons in CBDC. Mimeo. Pantelopoulos, G. (2024b) Can external sustainability be decoupled from the NIIP? Internaonal Economics and Economic Policy, (21), 89-116. Rostagno R., C. Altavilla, G. Carboni, W. Lemke, R. Motto, A. Saint Guilhem, J. Yiangou (2021) Monetary Policy in Times of Crisis: A Tale of Two Decades of the European Central Bank, Oxford University Press. Schnabel, I. (2023) Back to normal? Balance sheet size and interest rate control, Speech given at Columbia University, New York, 27 March 2023
36 Schnabel, I. (2024) The Eurosystem’s operaonal framework, Speech by Isabel Schnabel, Member of the Execuve Board of the ECB, at the Money Market Contact Group meeng Frankfurt am Main, 14 March 2024 Sonnenberg, N. (2023) ECB Stepping on the Brake(S): Monetary Tightening in an Abundant Reserve System, in European Parliament (2023), pp. 43-77. Tucker, P. (2022) Quantitative easing, monetary policy implementation, and the public finances, London: Institute for Fiscal Studies
Imprint Publisher Macroeconomic Policy Institute (IMK) of Hans-Böckler-Foundation, Georg-Glock-Str. 18, 40474 Düsseldorf, Contact: [email protected], https://www.fmm-macro.net FMM Working Paper is an irregular online publication series available at: https://www.boeckler.de/de/fmm-working-paper-22457.htm The views expressed in this paper do not necessarily reflect those of the IMK or the Hans-Böckler-Foundation. ISSN 2512-8655 This publication is licensed under the Creative commons license: Attribution 4.0 International (CC BY). Provided that the author's name is acknowledged, this license permits the editing, reproduction and distribution of the material in any format or medium for any purpose, including commercial use. The complete license text can be found here: https://creativecommons.org/licenses/by/4.0/legalcode The terms of the Creative Commons License apply to original material only. The re-use of material from other sources (marked with source) such as graphs, tables, photos and texts may require further permission from the copyright holder.