Introduction
The endogenous money approach has a long history and development, but the proponents of Modern Monetary Theory (MMT) point out that it can be extended by Chartalism and the leading role it gives to money created by the State. MMT shares the Keynesian view of endogenous money but in contrast to it, MMT sees the money created by banks as a kind of leverage on state money, all allowed and consented to by the authorities. MMTers argue that such an interpretation does not contradict the endogeneity of money approach, but rather complements, generalises, and extends it: ‘the endogenous money theory and the state theory of money are not contradictory, but rather complement one another to build an understanding of our modern monetary system’ (Nersisyan and Wray Reference Nersisyan and Wray2016, 1299). Indeed, one MMTer has highlighted this idea in the very title of one of his papers: ‘Government monetary and fiscal operations: generalising the endogenous money approach’ (Tymoigne Reference Tymoigne2016). In fact, MMTers believe that such integration is not only possible but desirable because the Chartalist view best explains the creation of money in modern economies: ‘MMT - by providing a more complete picture of how money is created in modern capitalist economies that is grounded in historical and institutional analysis - extends endogenous money theory, integrating it with the state money approach’ (Nersisyan and Wray Reference Nersisyan and Wray2016, 1299).
This idea is questioned by some post-Keynesian authors, who consider that the origin and value of money must be found in the private sector in general, and in banks in particular, so that state money is necessarily dependent on such bank money. This conception contests the MMT view that Chartalism extends the endogenous money view, since state money, being logically later than the bank money, brings nothing new to the endogenous money approach. Some other heterodox authors go further and directly question whether MMTers share the endogenous money view.
We try to shed light on this question, structuring the paper as follows. First, we review the theoretical approach to endogenous money. Second, we review how MMTers incorporate their Chartalist view into this approach. Third, we discuss the criticisms that this proposal has received, as well as the replies that have been given to it. Finally, we draw our conclusions.
Endogenous money
The conventional view of how the central bank supplies state money to banks is known as the monetary multiplier, which was explained in some detail by Paul Samuelson (Friedman Reference Friedman1960; Matthews Reference Matthews1961, Samuelson et al Reference Samuelson, Nordhaus and Enrri2003): first, the central bank would create a quantity of high-powered money, the monetary base, and then the private banks would create bank money from that quantity, thus obtaining the money supply, the total quantity of money. This quantity would be a multiplication of that first monetary base. Consequently, the monetary authorities, through the central bank, would be controlling the amount of money that is put into circulation through the private banks. If they wanted more money in circulation, they would create more high-powered money because banks could multiply it; if they wanted less, they would create little high-powered money because banks would be forced to create less bank money.
But this view has been strongly criticised by both Keynesian economists and central bankers (Minsky Reference Minsky1967; Kaldor Reference Kaldor1982; Goodhart Reference Goodhart1984; Moore Reference Moore1988; Wray Reference Wray1990). In one bulletin of the Bank of England, one can read: ‘neither step in that story represents an accurate description of the relationship between money and monetary policy in the modern economy’ (McLeay et al Reference McLeay, Amar and Ryland2014, 21). Even former Federal Reserve Chairman Ben Bernanke explained in a lecture that the money multiplier is not a faithful explanation of reality (Bernanke Reference Bernanke2007). The alternative view is that of endogenous money: first, banks would extend credit to customers they believe to be creditworthy, which would create bank deposits, and then, only in case they needed bank reserves to meet their obligations or legal requirements, would they take care to obtain them by borrowing them from other banks or, in the last case, directly from the central bank. This is something that even a vice-president of the New York Federal Reserve, Alan Holmes, acknowledged long time ago: ‘in the real world, banks extend credit, creating deposits in the process, and look for the reserves later’ (Holmes Reference Holmes1969, 73). The name ‘endogenous’ money refers to the fact that money would be created endogenously to economic activity (i.e. money would originate with economic activity, just when bank credit is granted because it is requested by someone to buy or invest), not exogenously or outside of it as the conventional view, which states that the central bank creates money independently of the purchases or investments taking place in the economy (Pollin Reference Pollin1991).
Therefore, the monetary authorities would not control the quantity of bank money, because it would not be limited by the state money created. On the contrary, banks would grant credit on their own, thus creating bank money, and then central banks would create state money to adapt it to the banks’ needs. It is the banks that would determine the total amount of money that will be available through the granting of credit, which, in turn, would depend on how much credit households and firms demand, and this in turn would depend on the economic cycle; in times of economic growth more credit would be demanded, and in times of crisis, less would be demanded. The central bank would not determine the quantity of money; it would only adapt to what is happening in the economy; it is the inverse view of the monetary multiplier. According to one of the most important proponents of the endogenous money view, Basil Moore (Reference Moore1988), the endogenous money approach should not be understood as a framework that presents the central bank as an agent that has no influence whatsoever on the quantity of money, but rather that this influence would be through prices rather than quantity, which would significantly change the analysis.
Whether banks need these bank reserves and ask the central bank for them or not will depend mainly on two factors: First, on what their customers do—because if they withdraw the money in cash or if they pay people whose bank is different, the banks will have to deliver bank reserves—and second, on banking regulations—such as the statutory cash ratio or credit provisions, which establish a minimum of bank reserves. Should they eventually need these bank reserves, then they would obtain them through the various avenues mentioned above, with the central bank as a last resort. Only when the bank was in a very weak situation, close to insolvency, would it not be able to obtain sufficient bank reserves. But if the bank was of considerable size and could destabilise the economy with its failure, the authorities would most likely do what was necessary for it to obtain bank reserves, either through bailouts, nationalisations or mechanisms for it to be absorbed by another healthier bank. However, some Post-Keynesian economists explain that ‘the central bank’s primary responsibility is to ensure that there is enough liquidity in the banking system to facilitate bank transfers’ (Culham and King Reference Culham and King2013, 395). Indeed, in a scarce reserve regime (classical corridor system), the central bank has always to provide reserves on demand. As economist Thomas Palley reminds us, failure to do so could cause banks to fail: ‘if central banks do not accommodate the increase in deposits that follows the increase in lending, the banking system, which will have become insolvent, will collapse’ (Palley Reference Palley, Piégay and Rochon2006, 71).
This view of endogenous money was pointed out long ago by some economists of the British School of Banking in the mid-19th century and used by economists such as John Maynard Keynes, Michael Kalecki, and Joan Robinson, although it was not explicitly and directly addressed until the late 1960s by the economist Nicholas Kaldor (1960). According to Wray (Reference Wray2015), Hyman Minsky also based his work in those years on the endogeneity view of bank money, something also discussed by Lavoie (Reference Lavoie, Cohen, Hagemann and Smithin1997). Years later, the economist Basil Moore developed this endogenous money view in detail and updated it to modern monetary systems, forming one of the best known explanations to date (Moore Reference Moore1988). Thereafter, the founders of MMT adopted the endogenous view of money (Wray Reference Wray1990, Reference Wray1992a, Reference Wray1992b; Mosler Reference Mosler1995, 3, 5; Tcherneva Reference Tcherneva2016).
Modern Monetary Theory’s view of endogenous money
MMTers consider bank money as a promissory note (or IOU) of the bank, and these promissory notes would be issued when they grant credits (Tcherneva Reference Tcherneva2016, 15). When a bank grants a loan to an economic agent, it is only committing to pay a certain amount of promissory notes from the state. If a family requests a credit of a certain number of monetary units, the bank would credit its bank account so that it would increase by that amount, which would be reflecting the amount of government promissory notes it has committed to deliver when it requests it. This is what was called stylograph money by James Tobin (Reference Tobin1963), referring to the fact that the money given on credit was simply what was written (or typed in modern times) by the bank worker. Some economists at the Bank of England have recognised that money creation works in this way and have explained it in detail: ‘whenever a bank makes a loan, it simultaneously creates a matching deposit in the borrower’s bank account, thereby creating new money’ (McLeay et al Reference McLeay, Amar and Ryland2014, 14). This was, after all, something even Schumpeter noted: ‘the banker plays the role of the “ephor” of capitalism. … the creation of new purchasing power out of nothing - out of nothing’ (Schumpeter Reference Schumpeter1912, 72-3). The bank’s promissory note would be extinguished when the loan is repaid and when the customer withdrew his entire balance in cash—since it is a government note in physical form; or when the customer made, for the full value of his balance, a transfer to another bank or an electronic payment to someone with an account at a different bank—so that the original bank’s note would be converted into a receiving bank’s note (Wray Reference Wray2012, 84-85). If these transactions were made in an amount less than the total balance, then the promissory note would not be extinguished, but would simply be smaller in magnitude (Wray Reference Wray1998, 77-78).
The key point of interest for this paper is that, according to the money-debt or money-credit view—shared by the MMT, any economic agent has the capacity to issue his own promissory notes; he can commit to deliver value (utility) in the future in exchange for receiving something of value (utility) in the present. Whether the counterparty accepts the proposal will depend fundamentally on the confidence in the issuer of the promissory note. After all, if the commitment is not honoured, there will no longer be any incentive to accept promissory notes from the defaulting party. This is none other than the keystone of Mitchell Innes’ credit theory: the necessary condition for any debt to have value is that the debtor must fulfil his promise, or else the creditor would not accept a new promise (Innes 1913, 53). Therefore, although everyone can issue promissory notes, the ability to get them used is limited to a very specific space: to those economic agents who accept the promise. This is what the economist Hyman Minsky referred to when he stated that ‘anyone can create money, the problem lies in getting it accepted’ (Minsky Reference Minsky1986, 228). However, as Stephanie Kelton reminds us, because these commitments necessarily involve two parties (the one who makes the commitment and the one who accepts it), it makes no sense to speak of their existence before the deal takes place. Therefore, it would be more accurate to say that ‘anyone can make promises or offer to go into debt but the “problem” is to find someone who is willing to become a creditor (i.e. to hold that promise or debt)’ (Bell Reference Bell (Kelton)2001, 151).
Some promissory notes or IOUs are widely accepted while others are only rarely accepted. According to this degree of acceptability, all promissory notes can be classified, thus forming what Hyman Minsky called a ‘hierarchy of money’ (1986, 228) or Foley a ‘debt pyramid’ (Foley Reference Foley, Vernengo, Pérez Caldentey and Rosser1987). In Randall Wray’s words, ‘there is a hierarchical arrangement whereby liabilities issued by those higher in the pyramid are generally more acceptable’ (Wray Reference Wray2012, 86).
At the top of this pyramid or hierarchy would be the IOUs issued by the state, as they would be the most widely accepted by the public, which, according to the Chartalist view, is because taxpayers are obliged to use them to settle their tax debts (Bell Reference Bell (Kelton)2001, 160). These state IOUs are also called high-powered money or base money, and have both a physical and an electronic form. The physical form is cash, coins, and banknotes (the latter accounting for approximately 95% of the entire value of physical IOUs). The electronic form is the IOUs held in the central bank’s electronic accounts, called bank reserves, which account for up to 5 times more than cash (McLeay et al Reference McLeay, Amar and Ryland2014).
Since in modern economies, government spending and revenue collection is done through the banking system, government spending increases the balance of the beneficiary’s bank account, and thus also the bank reserves of the bank concerned Conversely, tax collection reduces the balance on the taxpayer’s bank account, and thus also the bank reserves of the bank concerned. This positioning of government IOUs within the banking system is considered by many to be one of the main contributions of MMT, For example, Marc Lavoie asserts ‘their main contribution, both to monetary theory at large and to post-Keynesian economics in particular: to show and analyse the links between the central bank and the government within the context of the payment system’ (Lavoie Reference Lavoie2019, 97). Consequently, government deficits increase bank reserves and government surpluses reduce them (Forstater and Mosler, 156). And all this occurs without intervention by commercial banks or the central bank. This approach is not specific to MMT either, but rather, is shared by the Monetary Circuit Theory: ‘the higher the state deficit, the greater is the net increase in commercial bank reserves’ (Parguez & Seccareccia Reference Parguez, Seccareccia and Smithin2000, 112).
Precisely because all private sector economic agents are obliged to use government IOUs to pay taxes, private sector IOUs are also denominated in the government unit of account, as this would facilitate their acceptability. Thus, there is only one unit of account, not as many as there are economic agents issuing IOUs (Bell Reference Bell (Kelton)2001, 158).
In this pyramid or hierarchy of money, just below state IOUs would be bank IOUs, because they are also widely accepted by the whole population, even though they are not compulsory. There are two interrelated reasons for this. The first one would be that the state allows these IOUs to be used to pay taxes, thus conferring on them virtually the same Chartalist status as government IOUs (Mitchell et al Reference Mitchell, Wray and Watts2019, 155). The early Chartalists explained this ability of the state to determine that something is considered to be money in general use. For example, in Knapp’s words:
all means by which a payment can be made to the state are part of the monetary system. On this basis it is not the problem, but the acceptance, as we call it, that is decisive. State acceptance delimits the monetary system” (Knapp Reference Knapp1924, 95).
Lerner was also quite clear on this, stating that the modern state can, by accepting it in its offices of payment, ‘make anything it chooses generally acceptable as money’ (1947, 313). Finally, James Tobin himself, who did not recognise himself as a Chartalist but who studied money and the banking system in depth, also came to the same conclusion:
… by its willingness to accept a designated asset in settlement of taxes and other obligations, the government makes that asset acceptable to any who have such obligations, and in turn to others who have obligations to them, and so on (Tobin 1998, 27).
The second reason that would explain why bank IOUs are so widely accepted, and which is interrelated to the first, is that they could be easily convertible into government IOUs. No one would pay taxes directly with bank IOUs, but, in doing so, banks would be responsible for converting them into government IOUs, so it would be with them that taxes would actually be paid (Wray Reference Wray1998, 37). Consequently, private banks act as intermediaries between taxpayers and the government, making cash payments (reserves) on behalf of taxpayers. Therefore, however, much banks act as intermediaries, taxes could only be paid with state IOUs; hence, it could be argued that ‘in reality only the government’s money is definitive (finally discharging the tax liability)’ (Bell Reference Bell (Kelton)2001, 160).
The key to this whole issue would be that banks have a lot of facilities to convert their IOUs into government IOUs, equating the former with the latter, and so the taxpayer would not normally be aware that the bank is acting as an intermediary between him/her and the state. And this would not be because banks have little leverage or create few bank IOUs; rather, they do the opposite to maximise their profitability (Wray Reference Wray2012, 78). What would happen is that the state would make it very easy for such a conversion to take place because it would guarantee a healthy and stable banking system (Tcherneva Reference Tcherneva2016). If it were not for the state’s guarantee of bank deposits, bank IOUs would not be so widely accepted (Nersisyan and Wray Reference Nersisyan and Wray2016, 1302). In any case, it is important to note that both motives would be important. If bank promissory notes were easily convertible into other types of promissory notes, and not into ones that would settle the tax debt, then they would not be so widely used (Bell Reference Bell (Kelton)2001, 160).
Finally, below the IOUs of the state and banks, there are the IOUs of the other private economic agents (households and non-financial firms) (Bell Reference Bell (Kelton)2001, 159). The IOUs that any family or business might create would have very low acceptability because no one would be obliged to use them (as they are not used to pay taxes); only those with a close link to their issuers (e.g. a family member or a customer) would be willing to accept them. However, the acceptability of corporate promissory notes would normally be higher than household promissory notes, partly because firms tend to have greater economic weight and deeper markets: ‘firms’ promises are more readily convertible into the medium of exchange (i.e. are more liquid) than household’s promises because better secondary markets exist for their resale’ (Bell Reference Bell (Kelton)2001, 159).
Economic agents would promise to convert their own IOUs into IOUs higher up in this hierarchy of money; a household can promise to pay with a share in a company, a company with a bank balance, and a bank with coins and notes (Tcherneva Reference Tcherneva2016, 16). The key is that there would be nothing above the state money at the top of the pyramid; the state would be the only economic agent that does not deliver anything in exchange for its IOUs (Wray Reference Wray2012, 49). Therefore, government IOUs would form the core of the debt pyramid (Wray Reference Wray2012, 87). Government IOUs, because they should be used to settle tax debts and because they would not be convertible into anything, would form the type of money that would shape the rest of the IOUs, the most important money of all (Wray Reference Wray1998, 77).
Consequently, what MMT does is to combine the endogenous money approach with the Chartalist state money approach to provide a theoretical framework that, according to them, better describes the reality of modern monetary systems.
Criticisms to endogenous money’s view of Modern Monetary Theory
The main heterodox criticism of the MMT conception of bank money and endogenous money comes from post-Keynesian economists, the first to develop the endogeneity of money approach: ‘Wray’s account of the money-creation process is at odds with the post- Keynesian approach’ (Gnos and Rochon Reference Gnos and Rochon2002, 45). These authors question the idea that the money created by banks is a leveraging of state money, and they do so basically for two reasons: (1) private banks would have historically created their own IOUs without the need for state money and, therefore, without leveraging on it, and (2) bank money would be created from the demand for credit from the private sector, ergo the state would play no role in this process. In the words of Eladio Febrero: ‘we find the term “leverage,” relating bank money to state money, a rather troublesome term from a historical perspective (private money predates state money) and because of the logic of endogenous money (banks create deposits when they accept demands for credit and reserves are obtained later)’ (Febrero Reference Febrero2009, 524).
On the first point, Perry Mehrling notes that ‘private finance is a better logical place to start when trying to understand modern money’ (Merhlin, 402), and Rochon and Vernengo point out that ‘it is the fact that bank loans must be repaid that ensures the utilization of bank money, and money becomes a creature of banks rather than a creature of the state’ (Rochon and Vernengo Reference Rochon and Vernengo2003, 61). On the contrary, MMTers place the origin of state money much earlier than bank money, so that the latter would not only be later, but also dependent on the former: ‘states (and their predecessors) managed without banks for thousands of years and today the central banks and private banks operate within state money systems-with the state choosing the money of account and with currency and reserves occupying the top spot in the money hierarchy. (…) state money comes before bank money’ (Nersisyan and Wray Reference Nersisyan and Wray2016, 1304).
Parguez and Seccareccia (Reference Parguez, Seccareccia and Smithin2000) attempt to bridge the gap between the two views with the following approach. On the one hand, they agree with the MMTers that bank money could not exist without the state: ‘we have argued that credit money could not exist without the state, and that all credit-driven money is by its very nature a fiat money, irrespective of whether it takes the form of a commercial bank or central bank liability’ (Parguez & Seccareccia Reference Parguez, Seccareccia and Smithin2000, 107). On the other hand, they also believe that bank money is essential for state money to operate properly: ‘state power alone cannot guarantee the existence and survival of a viable monetary system’ (Parguez & Seccareccia Reference Parguez, Seccareccia and Smithin2000, 107). This would be so because what would give value to any kind of money, including state money, would be the future production that is financed by bank money:
money is at all times the liabilities issued by banking institutions which have been endorsed by the state primarily for the purpose of financing the formation of future real wealth. This money has a real extrinsic value because every holder of these liabilities has acquired a claim on the future physical wealth that results from the initial bank credit advances” (Parguez and Secareccia Reference Parguez, Seccareccia and Smithin2000, 107).
The experience of the European Economic Union is used by some authors to position themselves in this debate, in this case by differentiating between state money and central bank money. Due to the prohibition by the European Central Bank to directly finance member states, the key and original money of the economic process would be bank money, which would challenge the Chartalist view:
it is the cancellation of bank debt and not tax settlement that moves all agents in the European economy (national states included), in the last instance, to accept money which, in turn, is the logical consequence of (private) bank credit” (Febrero Reference Febrero2009, 530).
Some MMTers such as Ehnts (Reference Ehnts2014) have tried to respond to these criticisms by using the institutional design of the European Economic Union, but in general, this matter has not been addressed by MMTers as the European case would be only a specific one deriving from a general case that would be more interesting to describe and analyse.
With respect to the second point made, that bank money is endogenous and therefore cannot be a leverage of state money, Marc Lavoie considers that ‘if bank reserves are endogenous to required level, then the expression “leverage” does not seem appropriate’ (Lavoie 1999, 371). Parguez and Secareccia deduce that if, as the MMT says, bank money is leveraged over state money, then the former cannot be endogenous (Parguez & Seccareccia Reference Parguez, Seccareccia and Smithin2000, 120). However, this seems a wrong deduction since the MMTers do not say that bank money is constrained by state money: it would be one thing for banks to create IOUs referenced to state IOUs, which is what would explain the use of the word ‘leverage’, but it would be another thing that banks cannot create bank IOUs until they have state IOUs. As we have already pointed out, MMT considers that the bank first creates the credit, and only after that does it worry about getting the bank reserves it needs. Some MMTers charge that their critics have a misconception of the term ‘leverage’:
in the world of finance, “to leverage” signifies being able to take a position in an asset without having to provide all or any funds for the position. Banks necessarily leverage Central Bank (CB) currency, because they acquire asset position by issuing financial instruments that promise to deliver CB currency on demand or on some contingency at a later date (Tymoigne and Wray Reference Tymoigne and Wray2013, 37)
They insist that this is not the same as multiplying central bank-provided money, as some critics have said (Tymoigne and Wray Reference Tymoigne and Wray2013, 37). It seems that some critics such as Cesaratto agree with the MMTers on this point: ‘clearly, Wray’s meaning of “leveraging” must be understood in a broad sense, stressing the hierarchical primacy of the state in producing high powered money (HPM)’ (Cesaratto Reference Cesaratto2016, 63).
For their part, Gnos and Rochon point out that the high-powered money used by the central bank to balance transactions in the interbank market is not the same as the money used by the state for its fiscal operations (Gnos and Rochon Reference Gnos and Rochon2002, 46). Based on Graziani’s approach, the function of central bank high-powered money would be merely to compensate bank debts: ‘the role of the Central Bank is in fact of acting as a third party between single banks so far as their reciprocal payments are concerned’ (Graziani Reference Graziani1990, 18). Or, in the words of Rochon and Rossi, ‘transactions in the interbank market can only be settled using central bank money’ (Rochon and Rossi Reference Rochon and Rossi2004). For them, there is a crucial difference between central bank money and state-created money: the former would be endogenous in nature, while the latter would not:
if banks cannot borrow from other banks or sell assets to the central bank, then it can borrow high-powered money from the central bank as lender of last resort. In this sense, the supply of high-powered money is always endogenous, a result of the demand for it (Gnos and Rochon Reference Gnos and Rochon2002, 45).
In contrast, for MMT, state and central bank money are fully comparable:
it should be obvious, but it usually does not appear to be so … that central bank liabilities do not differ in any significant degree from treasury liabilities—in other words, we can treat both as essentially ‘high powered money’ or liabilities of the state (Wray Reference Wray, Rochon and Rossi2003, 87)
Therefore, the MMTers consider that state money can be created endogenously when banks demand reserves from central banks.
On the other hand, critics question whether bank money is widely used by the public because it is used to pay taxes as MMT claims: ‘it is one thing that the state encourages the acceptance of money, even that the “which thing” should be accepted, but it is quite another thing that the acceptance of money depends on the means to pay taxes’ (Febrero Reference Febrero2009, 537). Parguez and Secareccia point out that if banks’ promissory notes are denominated in the unit of account of the state, it is basically to facilitate their circulation (Parguez & Seccareccia Reference Parguez, Seccareccia and Smithin2000, 105). In fact, they understand the process in the opposite way to MMT: first, there is the creation of bank money—which has extrinsic value because it finances the production of future wealth—then, the state supports this process with its authority, and finally the state can create its own IOUs in an broad way because it makes them convertible into bank money:
the state’s power to create money at will is the logical consequence of its role in the endorsement of bank activity. Having the legal authority to bestow on bank debt the characteristic feature of money, the state has also the power to issue debt on itself that will be money, freely convertible into bank liability (Parguez & Seccareccia Reference Parguez, Seccareccia and Smithin2000, 106).
MMTers have responded to this type of criticism by pointing out that central banks must exogenously accommodate the number of bank reserves (government IOUs) that can be produced endogenously to economic activity:
it is an exogenous fluctuation (or a “vertical” injection of base) for the domestic non-federal sector even though it may be endogenous to the state of the economy (…) All this of course does not mean that MMT is throwing away the endogenous component of variations in L1. (Tymoigne and Wray Reference Tymoigne and Wray2013, 24)
That is, despite bank reserves changing considerably as a result of private economic activity, the central bank is obliged to accommodate them in detail: “any central bank that administers an overnight interest rate target must supply reserves on demand-for otherwise it would lose control of the interest rate. In the post-Keynesian literature, it is said that CB policy always ‘accommodates’ the demand for reserves. Given that this demand is highly interest-inelastic, there is little room for ‘error’ by the central bank. It must accommodate more-or-less exactly the demand” (Fullwiler et al Reference Fullwiler, Bell and Wray2012).
In other words, MMT considers the exogeneity of government IOUs injected through fiscal policy to be compatible with the endogeneity of government IOUs injected through the central bank in response to the evolution of bank IOUs:
one can conceive of a vertical component of the money supply process that consists of the government supply of fiat money; money drops vertically to the private sector from government. … On the other hand, the bankmoney- supply process is horizontal; it can be thought of as a type of ‘leveraging’ of the hoarded vertical fiat money (Wray Reference Wray1998, 111).
Similarly, in the words of Mosler and Forstater: ‘[h]orizontal activity represents leveraged activity of a vertical component. … The creation of bank loans and their corresponding deposits is a leveraging of the currency’ (Mosler and Forstater Reference Mosler, Forstater, Davidson and Kregel1999, 168). For this very reason, as we pointed above, MMTers argue that ‘the endogenous money theory and the state theory of money are not contradictory, but rather complement one another to build an understanding of our modern monetary system’ (Nersisyan and Wray Reference Nersisyan and Wray2016, 1299).
Nesiba (Reference Nesiba2013) and Febrero (Reference Febrero2009) point out that MMT’s interpretation of the link between banks and the central bank implies that banks would have no limits, apart from the demand for credit, in creating IOUs of their own: ‘Wray (Reference Wray1998: 111–115; 2004: 259) addresses the question of how state and bank money are related, but his discussion does not, at least in these two places, elucidate definitive financial limits to bank money creation’ (Nesiba Reference Nesiba2013, 52). From an institutionalist approach, MMT has been criticised for not attaching sufficient importance to the endogeneity of money and relying heavily on the exogenous capacity of the state to control the amount of money needed by the economy.
Conclusion
The MMTers adopt the post-Keynesian view of the endogeneity of money, but they incorporate nuances and elements to adapt it to their own Chartalist view, so that they end up drawing a holistic view of state money and bank money. In this way, bank money would be a promissory note that banks create autonomously when they grant credit; hence, its origin and quantity depend endogenously on economic activity. So far, this coincides with the conventional view. The novelty is that for the MMters these promissory notes are quantified in the unit of account of the state and are commitments to deliver state promissory notes in the future; hence, they are considered as a leveraging of state money. Bank promissory notes would be widely used because they are convertible into state promissory notes (the only ones that serve to settle tax debts, which is what would give them value) and because the state manoeuvres so that this conversion is always—or almost always—possible due to its interest in maintaining the health of the banking system. In short, banks have the autonomy to increase the amount of money used in the economy, but their own operations and survival ultimately depend on the state.
This holistic view of bank money and state money has been heavily criticised in heterodoxy, especially by post-Keynesian authors. The central criticism shared by all these critics is that bank money would logically and economically precede state money (even though bank money may need the support of the state to be widely accepted, as some point out), so that it would make no sense to speak of leverage. In support of this approach, it is often argued that entrepreneurs first plan an economic activity, then ask for bank loans to carry it out, and finally, money is created endogenously by banks, so that state money would play no role in this process. Some reinforce their arguments by pointing out that bank money has historically been independent of public authorities, so that the same would be true today; and that institutional configurations such as those of the European Monetary Union, due to the impossibility of monetising public deficits, would demonstrate empirically that bank money precedes state money.
But after conducting an analysis of both positions, we conclude that the MMT approach does not contradict the endogenous money vision and, in fact, seems to be more in line with economic reality. Without the support of the state and the existence of its own promissory notes, the widespread acceptance of bank money would not be possible, since it is denominated in the state unit of account, and is after all, a commitment to deliver state promissory notes in the future. Before the existence of bank money, there must necessarily be state-regulated institutions that allow its origin and widespread acceptance. The state has full control over the banking system: just as it usually acts to achieve its survival, it could also act to achieve the opposite (either through a ban on leveraging with state IOUs or through the total nationalisation of the sector, to give just two examples). As for the historical argument, it does not seem easy to take a clear-cut position, but it would not be necessary either: whatever the role of bank money historically, what matters is what its role is today, since modern economies have very different institutions and behaviours from those of the past, so any historical factor should not necessarily be extrapolated to the present.
On the other hand, the criticisms of the concept of leverage do not seem to be fair or appropriate since critics seem to wrongly equate ‘leverage’ with constraints on bank money creation. In reality, it is perfectly compatible for banks to have ample room to create bank money through their lending with all those IOUs being leveraged into government IOUs.
It seems that most of the discrepancies can be explained by the different conceptions of central bank money. Thus, while critics see money that is set in motion by monetary policy as something essentially different from money that is set in motion by fiscal policy, for MMTers, it is exactly the same. Moreover, this debate could be more adequately and accurately addressed if critics were more mindful of the theoretical contributions of the ‘hierarchy of monies’ by Minsky (Reference Minsky1986), Foley (Reference Foley, Vernengo, Pérez Caldentey and Rosser1987) or Bell (Reference Bell (Kelton)2001), as they often seem to ignore, or underestimate them. It is much easier to understand the MMT view on the endogeneity of money if one conceives of bank money as an IOU that is referenced in government IOUs. This thought is shared by authors such as Smithin (2016):
it should also be said at the outset that a concept introduced some years ago by another member of the MMT school-namely Bell’s (2001) notion of the ‘hierarchy of money’ might have been useful in reconciling some of these differences (Smithin 2016, 65)
and also by Nesiba, although he is not entirely convinced by the idea of the ‘hierarchy of money’. (Bell Reference Bell (Kelton)2001), in her discussion of money hierarchies, also provides background that can be used to inform this question [how state and bank money are related], but her arguments do not fully satisfy this reader’ (Nesiba Reference Nesiba2013, 52).
Finally, it is worth mentioning an issue that, agreeing with Nesiba (Reference Nesiba2013), is not fully addressed by the MMTers: that is, the limits that banks have in creating bank money through credit. Perhaps, a satisfactory and more complete answer can be found in the work of James Tobin (Reference Tobin1963) and McLeay et al (Reference McLeay, Amar and Ryland2014): banks, although they have no technical limits to create bank money by granting loans, would not be able to grant as many loans as they want because this would affect their profitability, which is their raison d’être. Moreover, by creating money through credit, they would need someone on the other side demanding credit, and this is what would explain why the amount of money created depends on the borrowing preferences of households and firms, which in turn depends on the economic cycle.
