The state spends first: Logic, facts, fictions, open questions

ABSTRACT

Keynesian (or Kaleckian) logic leads post Keynesian economists to presume that a variation of state revenues from taxes and sales of Treasury bonds are the result of a variation in state spending and not the other way around. In the past two decades, the exponents of modern monetary theory (MMT) have been at the forefront in asserting the Keynesian (or Kaleckian) logic of this proposition, filling a theoretical vacuum in post Keynesian thinking. The question is that MMT consolidates the Treasury and Central Bank (CB) so that the latter automatically creates purchasing power in favor of decisions of the former to spend. Critics, however, point out that most institutional arrangements forbid CBs to finance the Treasury directly. After Lavoie (2013), the debate has moved forward and seen some convergence. The present paper critically reviews for unfamiliar readers an otherwise almost esoteric but fundamental discussion.

KEYWORDS:

• Chartalism
• endogenous money
• Keynesian theory
• Modern monetary theory
• post-chartalism
• state spending

JEL CLASSIFICATIONS:

• B5
• E12
• E42
• E5
• H6

Notes

¹I use the term post Keynesian in the sense of Lavoie (2014). Independently rediscovered by the Sraffian economists Bortis (1997) and Serrano (see Freitas and Serrano 2015), the super multiplier is not shared by all Sraffians (Cesaratto, 2015; Cesaratto and Mongiovi, 2015), but is raising.influential endorsement elsewhere (Allain, 2015; Lavoie, 2015).

²Notably, Wray (2012) and Cesaratto (2005, ch. 1), both inspired by Abba Lerner and Robert Eisner, were on the same side in debunking the Samuelsonian fictional representation of a PAYG social security scheme.

³See Cesaratto (forthcoming) for a tentative extension of endogenous credit/money theory to the diverse autonomous components of AD and for a discussion of various related complications.

⁴Fullwiler et al. (2012, pp. 17–18) and Tymoigne and Wray (2013, p. 4) refer to the circuit approach as the standard post Keynesian way to include endogenous credit/money in the theory of output determination, and intend to “bring the State into the circuit.” Cesaratto (forthcoming) adopts the super-multiplier analysis as an ordered Keynesian–Kaleckian premise to the investigation of the generation of monetary aggregate demand also in view of the shortcomings of the “monetary circuit,” in particular its insufficient integration with the short- and long-period theory of effective demand.

⁵The CB normally wishes to provide an amount of reserves such as to satisfy the aggregate banks’ liquidity needs. Banks with excess reserves will then trade them with banks short of reserves in the interbank market. In “normal times” in the Eurosystem, the European Central Bank (ECB) also offers overnight loans (marginal lending facility) at a rate that is relatively higher than the main refinancing rate (MRR), the basic rate at which it provides weekly reserves to commercial banks; and provides an overnight deposit facility at a rate that is relatively lower than the MRR. Typically, banks will profitably trade reserves at a rate (Eonia in the eurozone) which is in the middle of the last two rates (the so-called corridor) and that approximates the MMR. The U.S. system is basically similar. Government spending would inject reserves in excess of banks’ liquidity needs.

⁶According to Lavoie (working paper version of Lavoie [2013, p. 15]), this argument would not apply in “countries where reserves generate interest payments, usually close to the interest rate paid on short-term Treasury bills, then it is obvious that it makes little difference whether the debt is held in the form of bank reserves or in the form of Treasury bills.” In the Eurosystem, however, although mandatory reserves are remunerated at the rate on the main refinancing operations (MRO), normally excess reserves held in the reserve account at the ECB are not remunerated. In normal times, excess reserves are thus more conveniently held in the “marginal deposit facility” where they get a positive remuneration, albeit lower than the rate on the MRO. Therefore, a stimulus to purchase government bonds would remain even in regimes that normally remunerate excess reserves held in the deposit facility. I said “normally” since at the time of writing, excess reserves held both in the reserve account or in the deposit facility are remunerated at a negative rate.

⁷A reviewer suggested that a consolidation of the two branches of the private sector would not bring about much loss of information either.

⁸As is known, the basic model of monetary regulation of transactions suggests that economic subjects rely on monetary units issued by a third subject that both sides trust, that is, by somebody “higher in the hierarchy.” So private citizens trade using banknotes or transfer banks’ deposits, and banks make reciprocal payments using reserves, that is, HPM issued by the CB (intrabank payments between customers of the same institute do not use reserves since each bank trusts itself). In their transactions, CBs use gold or international (reserve) currencies. In recent times, interbank transactions are increasingly taking place through private payment systems based on the credibility of the counterparts. These systems concerns both petty and large payments. However, “central bank systems … continue to be the backbone for the rest of payments” (Kahn et al., 2014, p. 1).

⁹The former top ECB economist Ulrich Bindseil (2004, p. 46) suggests that governments may prefer to hold tax revenues at the CB, “the bank of the government,” rather than at commercial banks to avoid “the credit risk it would have to bear if it held its liquid funds in the market.” Government deposits with the CB represent “the most volatile and most unpredictable” autonomous factor in the CB balance sheet and, of course, the CB “has to precisely neutralize autonomous factor changes” as a “necessary condition to stabilizing short-term interest rates” (ibid., p. 60).

¹⁰As blogger JKH (2012) points out: “there is nothing special about the fact that the Fed supplies the reserves that enable tax and bond payments … . Some neo-Chartalists emphasize the uniqueness of Fed provisioning for tax and bond settlements. But there is no such uniqueness behind Fed operational … . That the Fed may or may not supply extra reserves to enable tax and bond payments against the backdrop of interest rate targeting doesn’t demonstrate anything of extraordinary relevance, because the same evidence and argument applies to payment circumstances more generally.” For instance, CBs also routinely intervene to meet the seasonal fluctuations in the absorption of HPM as banknotes.

¹¹The same terminology widely accepted for investment, initially financed by credit/money creation, finally funded by saving is used (see Cesaratto, forthcoming). If taxes totally “fund” spending, we would be in the famous balanced budget theorem. I put “funded” in quotation marks to indicate that in no way does state spending depend on tax revenues (or household saving). The state spends first.

¹²Similarly, JKH (2012) argues that: “Because the US Treasury is not an operational currency issuer, it obviously does not issue currency in conjunction with spending. It uses currency when it spends … . Thus, Treasury is a depositor with both the central bank and the commercial banks, with a centralized account (two accounts currently) at the Fed and peripheral tax and loan accounts with the banks. It is an operational currency user and cash manager, making disbursements from and taking receipts into its bank deposit account(s).”

¹³I noted above that this is done in order to stabilize the amount of reserves in the banking system. Since tax payments cluster around a few main dates, if they were immediately transferred to the TA at the Fed this would mean a major fall in banks’ reserves. JKH (2012) aptly describes the T&LA as a “feeder” of the TA at the Fed: “Tax and loan accounts constitute a feeder deposit system for the so-called Treasury general account at the central bank. These T&LA accounts exist for two reasons. First, they are convenient in terms of Treasury’s large scale cash management operation. It is more effective for Treasury to gather funds locally and subsequently import them into the central account, which in turn is the focal point of most disbursements. Second, there is a corollary benefit in terms of enabling operational co-ordination with central bank reserve management”.

¹⁴Actually, “the state spends first” does not refer to deficit spending only, as if tax receipts funded the rest of government spending. The question under discussion concerns government spending tout court. Even if the government is in surplus, spending logically precedes taxation.

¹⁵In the original post-chartalist presentation by Lavoie (2005, table 13; 2013, table 2) the state spends directly from its commercial bank account.

¹⁶Referring to some of the main publications, an explicit treatment of the actual ways the government finances its spending seems hard to find (e.g., in Wray, 1998). Bell (1998) and Bell and Wray (2003) deal with this, but the reasoning seems less elaborate and clear than in Fullwiler et al. (2012) and Tymoigne and Wray (2013), at least to the present author.

¹⁷“The basic objective of a government debt manager—a Bank of England handbook on PDs writes—is to cover the government’s borrowing needs as cheaply as possible. To accomplish this objective, both primary and secondary markets need to be broad and efficient and the secondary market … deep and liquid … . In return for a set of obligations, such as making continuous bid and offer prices in marketable government securities or submitting reasonable bids in the auctions, [PDs] receive a set of privileges in the market” (McConnachie, 1996, p. 6). PDs are said to be used in Canada, France, Italy, Spain, the UK, and the United States. In other countries such as Australia, Germany, and Japan “there are no formally designated [PDs], although in these countries a group of firms do collaborate in the allocation and proper development of the market in an informal way” (ibid., p. 7). Among the privileges, a most important one “is probably the facility to borrow stock and/or money, normally from the central bank … . It is certainly desirable that market-makers should have the facility to borrow money when they need to in order to discharge their market-making function” (ibid., p. 18). In the United States, the PDs cannot borrow funds from the Fed. Nonetheless, they are the only institution authorized to deal directly with the New York Fed in its open market transactions—repo and outright market operations (World Bank, 2010, pp. 26–27). In 2008–10 in response to the severe strains in the market and the resulting liquidity pressures faced by the PDs, the Fed established the Primary Dealer Credit Facility (PDCF). The PDCF functioned as a (fully collateralized) overnight loan facility for primary dealers, similar to the way the Fed’s discount window provides an ultimate source of funding to the banking system (Federal Reserve, 2013).

¹⁸Fullwiler et al. (2012, p. 24), for instance, argue: “Unfortunately, and most importantly, the added complexity is counter-productive because it leads to poor understanding among economists, poor modelling, and bad policy choices.”

¹⁹A referee also suggested that those who argue that “‘‘Governments that issue their own currency cannot default’ has at least implicitly endorsed consolidation.” See, for example, Paul De Grauwe (2015), who argues that member countries of the eurozone “could no longer guarantee that the cash would always be available to roll over the government debt.” What De Grauwe and others imply is that the government should act as a lender of last resort whenever a crisis of confidence increases the liquidity preference of the public. Looking at this through the MMT lens, it means that the monetary authorities (the CB and a cooperating Treasury) reverse the defense action undertaken to absorb excess liquidity (which is no longer in excess given the higher liquidity preference). Implicitly, therefore, those who invoke these measures admit the consolidation hypothesis that the CB initially financed government spending and that monetary authorities issued Treasury bonds to absorb excess liquidity, ready to repurchase them if the liquidity preference changes.

²⁰Reply to a letter to Chair Janet Yellen reported by Modern Money (2015, comment, 26/08/2015, 12:53 pm).

²¹Eccles was not a formally trained economist; his personal and professional experiences during the Great Depression made him a Keynesian, one would say, a Lernerian ante litteram. On the complex figure of Eccles, see Grugn Moe (2013). According to Mike Norman (2012), Fullwiler was the first to quote Eccles in the context of the present debate.

²²See also Gnos and Rochon (2002, p. 42); Febrero (2009, p. 536); and Lavoie (2013, p. 6).

²³Wray (undated, p. 23) argues that “private money things [deposits] are endogenously created as a leveraging of HPM.” This sounds like an oxymoron, although the author clarifies two lines later that “The central bank supplies reserves on demand.”

²⁴See, for example, Wray (2007, p. 12): “state liabilities (HPM) are destroyed when they return to the state, mostly in tax payments or bond purchases by the non-government sector … . Moore’s fundamental point remains: at the end of the day, the quantity of HPM remaining in private hands (as bank reserves and as cash) is demand-determined, and is not a discretionary variable from the point of view of the central bank that is targeting an overnight interest rate.”

²⁵“Private creation of money things can be characterized as a ‘leveraging’ of HPM precisely because they are convertible” (Wray, undated, p. 24).

²⁶“While we do not normally include central bank or treasury purchases of gold or foreign currency as a part of ‘spending’, it has the same impact on the quantity of outside HPM held by nongovernment sectors as any other kind of government spending. However, the impacts of lending (by the treasury or the central bank) are perhaps less expansionary than fiscal deficits because provision of HPM through lending does not directly increase nongovernment sector income, nor does it increase ‘net saving’ or ‘outside wealth’” (ibid.).

²⁷The verbal description of the stages is the following (Fullwiler, 2011, pp. 3–4):

The six transactions for Treasury debt operations for the purpose of deficit spending in the base case conditions are the following:

1. The Fed undertakes repurchase agreement operations with primary dealers (in which the Fed purchases Treasury securities from primary dealers with a promise to buy it back on a specific date) to ensure sufficient reserve balances are circulating for settlement of the Treasury’s auction (which will debit reserve balances in bank accounts as the Treasury’s account is credited) while also achieving the Fed’s target rate. It is well-known that settlement of Treasury auctions are “high payment flow days” that necessitate a larger quantity of balances circulating than other days … .

2. The Treasury’s auction settles as Treasury securities are exchanged for reserve balances…, bank reserve accounts are debited to credit the Treasury’s account…, and dealer accounts at banks are debited … .

3. The Treasury adds balances credited to its account from the auction settlement to tax and loan accounts … . This credits the reserve accounts of the banks holding the credited tax and loan accounts … .

4. (Transactions D and E are interchangeable; that is, in practice, transaction E might occur before transaction D.) The Fed’s repurchase agreement is reversed, or, otherwise stated, the second leg of the repurchase agreement occurs in which a primary dealer purchases Treasury securities back from the Fed. Transactions in A above are reversed.

5. Prior to spending, the Treasury calls in balances from its tax and loan accounts at banks. This reverses the transactions in C.

6. The Treasury deficit spends by debiting its account at the Fed, resulting in a credit to bank reserve accounts at the Fed and the bank accounts of spending recipients.

²⁸In Table 6 I have thus also simplified the ping-pong between the T&LA and TA of the Treasury that characterizes Fullwiler’s account.