Government Spending or Lending Logically Precedes Tax Revenue

In a previous post, it was explained that enforcement of taxes (or some other financial obligation to the state) is fundamental to the viability of a national currency. Without such an obligation, widespread acceptance of the currency would not be assured. The currency might cease to serve as an effective mechanism for public provision of adequate infrastructure, education, health care, social security and much more.

But there is something interesting about this. If we think about the initial introduction of the currency, the government could not have simply imposed a tax on us and then sat around waiting for us to pay. Where would we have got the currency from to pay the tax?

Clearly, the currency originally has to come from the government (meaning the consolidated government sector). The government can issue its currency in one of two ways, either by spending or lending.

The above considerations lead us to two basic points. Guaranteed viability of a national currency requires that:

  1. The imposition of a tax obligation precedes government spending or lending.
  2. Government spending or lending logically precedes tax payments and tax revenue.

The logical sequence is clear: (i) government imposes taxes; (ii) government issues the currency; (iii) we use the currency to pay taxes as well as for other purposes.

Strictly speaking, point 1 is sufficient rather than necessary. That is, it is perhaps conceivable that a national currency could gain acceptance without being backed by taxes but the effective enforcement of a tax obligation guarantees acceptance of the currency, at least to the extent necessary to satisfy tax requirements. Only a very brave (or foolish) government would attempt to operate a national currency without a tax basis.

Point 2 always holds. Currency issuance through either government spending or lending must already have occurred before taxes can in fact be paid.

Clearly, then, government does not need our tax payments in order to spend. What it needs is our readiness to accept its currency. It guarantees this by imposing a tax obligation. But in order for us to pay our taxes, government must first spend or lend the currency into existence.

Reflecting on this for a moment exposes as silly a lot of what we hear from the media and politicians when it comes to fiscal policy. We are frequently asked to believe that the government cannot afford to provide adequate infrastructure and social services because it has supposedly “run out of money”. Yet, the only entity for which it is impossible to run out of the currency is government – the currency issuer. The rest of us can run out of money. But the currency issuer cannot.

For this reason, the true measure of fiscal “affordability” should never be framed in terms of money, but rather real resources.

If there is a shortage of trained doctors and nurses, it will not be possible to expand hospital care until some more have been trained. If there is a shortage of teachers, educational services will be similarly constrained. A shortage of engineers, trades people and raw materials will impact on the provision of infrastructure. And so on.

Lack of money, however, is never a legitimate excuse for government. It can always spend sufficiently to ensure full employment of workers and real resources.

Related Post

The above is intended as an introductory post and does not touch on the details of current monetary and fiscal operations. In practice, governments typically follow a rather convoluted series of steps when they spend. None of that changes the basic point that government spending or lending must precede tax payments. The following post provides further details:

Exercising Currency Sovereignty Under Self-Imposed Constraints 


26 thoughts on “Government Spending or Lending Logically Precedes Tax Revenue

  1. Loans create deposits. Firms take loans from banks, pay employees, create products, which are bought by households. Households pay taxes to the government on their salaries. The government receives the funds and spends.

    Spending came after taxing here.

    You could say TTL, TGA bla bla but let’s just say that I am the Reserve Bank Governor of a newly formed State trying to prove you wrong and I do not transfer the funds paid as taxes to the government’s account at the central bank and simply ask the Treasury which is managed by me to directly transfer the funds paid as taxes intra-bank to government accounts and then pay using these funds intra-bank.

    Spending came after taxes in this example.

  2. Lending came first in your example, that’s all.

    The tax payments won’t be cleared without reserves (government money). The reserves come from the monetary authority (part of the consolidated government).

    As noted at the end of the present post, greater detail is provided in a previous post. Tax and loan accounts are discussed in that post. Deposits in tax and loan accounts can’t exist prior to government spending or lending, and in fact are evidence that prior government spending or lending must have taken place.

    None of the details change the basic point of this post, which, as the title suggests, is that government spending or lending logically precedes tax revenue.

  3. Peter,

    Yes bank lending came first but doesn’t matter. It’s not the government which is lending.

    Proof is construction of a scenario where government spending is after taxes are collected. As far as reserves are concerned, let’s say there are 4 commercial banks and I as the Reserve Bank governor create government accounts at these 4 banks. If the account holder has an account at bank 2, I spend using bank 2 account of the government. No reserve involved as the transaction is an intra-bank transaction.

    My general point is that in the history of mankind, a social formation of government must have come in later times and people used credit for transactions way before that. There’s no need to state the way MMTers state it.

  4. In your first example (private bank lending), yes, my point is that taxes don’t settle without reserves, which requires central bank lending before taxes can be paid (settled).

    In your second (intrabank) example I’d say that taxes have not settled (since reserves are necessary for settlement), so government spending precedes tax payments (settlement) in that scenario.

    If you are just saying that a different requirement for settlement of taxes could be employed involving some other money, I agree. But the government will still be specifying what is required to pay taxes, and whatever is required must be issued before taxes can be paid.

    It would be foolish of government, though, to specify a money other than its own as required for settlement of taxes, because then the tax obligation would no longer ensure acceptance of the government’s money and the government would potentially lose its capacity to spend for the provision of infrastructure, social services and so on.

    I don’t believe you’d find an example of a national currency where that is the arrangement. If such a case were found, it would basically amount to the private bank (or whoever happened to issue the money) functioning as the state. It would effectively be a privatized, undemocratic state.

    Regarding the history of money, I agree there’s room for different interpretations, but in my view the available evidence comes down strongly in favor of the chartalist interpretation.

    For newer readers, I’ll link to a summary of the chartalist perspective on the history of money: Further reading can be found in the reference list provided at that link. Feel free, Ramanan, to post a link to papers presenting your preferred perspective. (Note that a comment with more than two links will get caught temporarily in the spam filter till I see it. You could split the post into more than one to get past the filter in the event that you have multiple links to share.)

    For me, the MMT framing is the most logical, clear and convincing. But there is no doubt room for different opinions on that.

  5. My understanding is an IOU is just an IOU – trust and discretion, honour, are involved (my idea of ‘human capital’ because without them all other capitals lead to a very nasty capital called Chaos, in a land called Greed)!

    If IOU’s were created only by the private sector, as one would suspect and as has happened historically, there arise problems of competition, manipulation, and exchange settlement between competitors; and loan cheating . Govt. would be left standing only on its political leg without an economic leg – which is absurd, with diminished and isolated regulatory control over IOU’s. Who then would own and ‘pay’ the Military? Private $purpose would overrule public $purpose which is also absurd because without humane Law, it leads straight back to the jungle and the monkey on our back.

    So the question of whether or not a monetary system can exist where spending and taxing are reversed in sequence is just one mirror looking at another. What is significant is the direction a Society takes (?)

    A monetary, religious, or political system are just aspects, vehicles for getting there.

    I think what the world needs is real human capital. The systems are too often distractions ….!

  6. Yeah Ramanan made the more general point, which you concede, that a government can always declare the tax obligation to be payable in something they don’t issue (say gold, food for soldiers, foreign currency, or the ECB’s euro currency). In which case, they’re just a user of that something, and logically tax revenue must precede spending.

    Of course, it’s pretty stupid for any authority to do this if they have the ability to set the obligation to anything they choose. It would be like a sports stadium using beanie babies for admission instead of printed tickets. Nevertheless it’s a stipulation you have to point out when you’re describing the underlying logic of government spending.

  7. My general point is that in the history of mankind, a social formation of government must have come in later times and people used credit for transactions way before that. There’s no need to state the way MMTers state it.

    That’s not how I understand it. MMT economists follow Innes’s credit theory as well as Knapp’s chartal theory on that point and admit that credit preceded chartal. What MMT economists claim is that modern money is chartal (state issued and tax-driven) as well as being credit (state liability or “IOU”). Currency is a tax-credit, in Mosler’s formulation.

    “Modern money” in MMT is state money, which appeared later than credit. Credit arose in something like the way Graeber presents it anthropologically. Biologists surmise that the antecedents of this are pre-human, and probably even pre-primate. Accounting records have been found as early as Sumerian civilization. Temple money seems to have preceded state money and may have been a precedent for it along with the existence of credit. While money is a social construct, it wasn’t anyone’s bright idea. It’s a social institution that developed about of custom and tradition wrt reciprocal relationships.

    The credit theory and chartal theory joined in the creation of state money as a unit of account, and both credit-debt relationships and money things (tokens) preceded this historically.

    The credit theory of money was elaborated in circuit theory. MMT claimis to supplement circuit theory rather than replace it with a different theory. MMT holds that modern money is chiefly endogenous, on one hand, and tax-driven on the other. MMT also holds that taxation is not a necessary condition for the use of state money, but rather a sufficient one, as Peter makes clear in the post. The history of the development of money could be different.

    However, given the history the logic seems clear. A state declares what it will accept at its payment offices in satisfaction of obligations to the state. A sovereign state can choose any monetary arrangement it wishes, as Peter also mentions.

    MMT claims that the general practice in modern times is for states to accept only their own currency at payment offices. This implies that the state must provide the wherewithal to meet obligations it imposes unless it delegates the issuance of its currency to non-state entities, or accepts other forms of payment. States can voluntarily abandon currency sovereignty, and some do.

  8. What is wrong in this story?
    If the government pays for real goods and services with its own currency, if the central bank does not sell the public or state loans to the private sector or to foreigners but but takes them in safekeeping, there will be no debt, there will be no debt claim for paying interest or payback of the “debt”; there is only a virtual debt, which never must be paid back – in the real world this virtual debt has become infrastructure, social capital, human capital etc.

  9. “In your second (intrabank) example I’d say that taxes have not settled (since reserves are necessary for settlement), so government spending precedes tax payments (settlement) in that scenario.”

    Obviously that is incorrect. Funds reach the government’s accounts and taxes are settled.

  10. Peter, sorry to interject, your exchange with Ramanan reminds me of the fundamental difference between MMT and circuit theory (associated with, say, Alain Parguez and Riccardo Bellafiore).

    According to Parguez, money has nothing to do with the State and taxation by the State isn’t a necessary condition for the viability of a monetary system. Bank lending and capitalism are the necessary elements.

    Here’s a few sentences from a lecture by Parguez which I translated from French and Italian which may be of interest to your readers. (I haven’t been able to find anything as detailed in English….It’s too bad because Parguez lays out his view so clearly):

    “Without money, without monetary creation, capitalism does not exist. In other words, it is not the State that imposes money. Money is necessary, [however, it is important to mention that] there have been examples of States in which money did not exist. The necessary condition, that which imposes the use of money, is not the State but the economic system itself. We can therefore deduce from this a first theoretical difference between MMT and theory of the monetary circuit. Money – and its fundamental role – has nothing to do with the State and nothing to do with taxes. The proposition which holds that we must have taxes for money to exist is logically and historically completely incorrect.”

    circuit here: I once raised this point during a conference on circuit theory. Some folks there have doubts about whether the correct ontology of money is necessary for understanding how the system works. (For what it’s worth, I think Parguez also sees it this way, which may explain why this difference of opinion hasn’t stopped him from supporting MMT’s various policy prescriptions.)

  11. Ramanan, can you demonstrate your scenario from inception? To me it seems that either: (i) the government in your example is not the currency issuer and instead is spending private bank money as a currency user; or (ii) you are considering an isolated part of daily spending and taxing rather than settlement for the day in aggregate, which when the government is a currency issuer will require government money (reserves, currently).

    If (i), I acknowledged in my previous comment, in the paragraph immediately following the one you quote, the possibility of a different monetary system in which taxes are not due in the government’s own money. If (ii), funds in tax and loan accounts due to tax payments and new treasury issues will be the result of prior government spending or lending, irrespective of whether tax and loan deposits are transferred to the Treasury’s account at the central bank prior to this round of spending.

  12. circuit, thanks for your input. Parguez seems to be referring to a necessary condition. MMT says enforced taxes are sufficient but not necessary for a viable currency. So when he stated, “The proposition which holds that we must have taxes for money to exist is logically and historically completely incorrect”, he may have intended it as being leveled at MMT, but it is not really. (See a previous post, Randall Wray on Tax Driven Money, which quotes Wray directly on the MMT position.)

    I think circuit theory and MMT are completely compatible, in spite of any differences in preferences over framing. I think there is room for more than one framing. Needless to say, compatibility does not dictate that agreeing with one approach necessarily implies or logically requires agreement with the other.

    Regarding money, Wray basically says (from 0:59 – 5:21) that all IOUs are money. It’s just some IOUs (such as private bank money and state money) are more widely accepted than others. A key distinction when it comes to the state is its capacity to impose and legally enforce obligations denominated in its IOU. This gives it certain advantages in generating acceptance of its IOU, but this does not mean it is the only way acceptance can be obtained or that money always has to be state money.

  13. The MMT claim is based on the assumption that if a particular condition is operative, e.g., imposition of obligations to X that are only satisfied in the medium that X specifies, then this necessarily results in a demand of the specified medium is X has the ability to require performance on obligations it imposes.

    As Parquez argues this is a necessary condition for capitalism. Capitalism is chiefly dependent on legal institutions that establish the requirements for a valid contract and also means for enforcing contracts, such as addressing putative breach and recovery.

    I don’t think that anyone argues with this.

    MMT adds that in a modern monetary system, there is a hierarchy in which the state stand at the apex legally. The state has the ability to impose obligations on non-state actors and the power to enforce those obligations over all other obligations among non-state actors.

    Therefore, if the state imposes a tax requirement and also an acceptable means of payment, those on whom the obligation is imposed must obtain that which the state requires when the state requires it, or face the legal consequences that is backed by the power of the state.

    There is no necessity for a state to do this, but if a state chooses to do so, as virtually modern states do, then the consequences follow necessarily.

    The Libertarian position is based on states not doing this and not being permitted to do it. Could that arrangement be imposed and if so, could capitalism function. Of course. There have been periods of free banking, for example, and weak states that could not impose obligations effectively.

    However, for many reasons the history has taken a particular path to the point at which humanity now finds itself and in general modern states impose financial obligations on non-state actors that are only payable in the unit of account that the state specifies and controls. A state could delegate the creation of the unit of account to a non-state agent such as a private central bank or to private banks acting without a central bank. In that case the government would have to fund itself through non-state sources.

    This is in fact what many claim is happening at present in the US, with the central bank being private, and indeed the courts have declared that the federal reserve banks individually are private institutions. Opponents claim that the federal reserve system is public. The response to this is that the payments system is run by the individual banks and only monetary policy is set by the board of governors.

    The upshot is that the MMT general operational description is manifested specifically in different ways, both operationally under the law and regulations and also in terms of political controls such as a debt limit in the US.

    So I conclude 1) that Parguez is correct about capitalism in a monetary economy not requiring state money. It can also be argued that capitalism can exist in a non-monetary economy, too, since a (Libertarian) bullion system is essentially a barter system using a commodity rather than “money” (pace, Alan Greenspan who thinks that gold is money, as do most Libertarians);

    2) that MMT is correct that modern economies in general are monetary economies that are legally constituted along Chartalist lines — states impose obligations in the state’s unit of account (currency) and only accept it in payment.

    3) that different monetary arrangements exist specifically in the laws and institutions of various jurisdictions, some of which are sovereign and some not, some of which establish more policy space and some less;

    4) that a key contribution of MMT is to show the differences that various possible institutional arrangements make regarding available policy space. In my view MMT monetary analysis is essentially about policy space and it’s implications for macro and policy formulation. Operations follow from instituting legal arrangements.

    In this sense, MMT is basically an institutional approach that looks at possible and actual institutional arrangements for operating a monetary system and the likely outcomes.

  14. Funds reach the government’s accounts and taxes are settled.

    True, but that just save the Fed the step of using OMO to adjust the MB to hit its target rate. The bank is still on the hook for the rb due on taxes after netting, and only the Fed issues rb, which exist only on the Fed’s spreadsheet. Of course, banks obtain rb in the normal course of operations, but the source of rb is ultimately the Fed just as the source of coin and tsy securities is only the Treasury. Reserve balances and federal reserve notes are liabilities of the Fed (created by the Fed) and coin and tsy securities are liabilities of the Treasury (created by the Treasury). The Department of the Treasury is an agency of the US government and the Fed is a designated agent of the US government.

    Does anyone doubt that the USD is presently issued by the US government and only the US government using agency it controls?

  15. I really struggle to understand the prioritisation with systems.

    An IOU is an IOU. There is the skin of the earth and its resources and human values, we like to produce and exchange things; the IOU’s keep score (ho hum)! The ultimate expression of this is to build an Empire (which lasts for a nanosecond on the earth’s clock – ho ho hum)! In the process we turn a beautiful earth, an incredibly unique planet, into a polluted factory; fight over resources, and turn the factory into a war-zone, destroying, denigrating, demeaning, the most incredible miracle that is right before our eyes – one we refuse to acknowledge or plumb, that took billions and billions of (light~life) years to evolve – a human existence; a human being. We do not understand what it means to be human, nor appreciate it. We do not understand life is our only treasure!

    Human beings have created more IOU’s circulating on this earth right now, than ever before: where has that got us? Recycling Empire like caged rats on a wheel (ho bloody hum)! Now, as the Sun comes up each day, we start up like little robots and slave our guts out and sweat to earn IOU’s and spill our blood – for Empire! (?????) Is that really the goal of evolution? That portrays an incredible lack of imagination and misuse and abuse of the gift of intelligence.

    For me, the monetary system is just a tool, and like any tool it is limited: – it’s what you build with a limited tool that counts. The most significant factor is always – the human being, each one carrying within them something that is unlimited; something to do with our existence, and the real and natural reason for why we are here; and way beyond the reach of the earth’s clock!

    Why aren’t we interested in that!! Oh Yes – the systems …… the Empires; castles in the sand all erased by the tide of Time, each flow. We are the grains of sand on that beach and each one of us, though the Universes come and go, are absolutely unique! Never ever to be repeated. Surely we should be looking at things from the perspective of life? Not that I’m saying we shouldn’t work out the best way to run a monetary system – but how is it that Empire has been elevated above the miracle that is human existence. That’s what I would like to see turned back up onto its feet! Restored to its rightful place. There is no one human being on the face of this earth that has ever lived up to the fullness of the gift that has been bestowed upon us, the privilege granted. Then when we build, even though temporary, it may have some elements of that beauty. Our ‘humanity’ is meant to be an expression of what is within. Without kindness, we are not even alive! (Thanks Peter)!

  16. I can’t add much to the great posts but it strikes me that the idea of the government as lender is quite profound. Unless it ‘lends’ first (even the spending is a loan) then repayment via payment of taxes or purchasing government bonds can’t happen given the proviso that both those repayments must be made in State money.

  17. Agh! When will this blog comment on some recent blogs made by Yves Smith?

    When I cited Paul Cockshott’s opposition to basic income (in an either/or framework), he mentioned downward pressure on wages. I did not know until today that this, in fact, had a historical precendent!

    Yves Smith cited Karl Polyani’s account of the Speenhamland system, an industrial basic income experiment which benefit employers and depressed wages:

  18. Thanks for that link, hadn’t seen it. But that Speenhamland system doesn’t sound like any version of UBI I’ve heard advocated in modern times.

    Apparently that would set a specific income and then “top up” anyone making less than that, up to exactly that level. So if they set it at $1000 a month, and you work at walmart, then walmart can feel free to pay lower & lower wages while the employees can care less & less about their work, and they still get topped up to $1000 wages a month by the government.

    That’s considerably different than the government giving everyone $1000 a month no questions asked, and then anything you make on top of that is extra for you.

  19. I don’t see the big deal & I’m with Tom on the overall structure of Chartal and Circuit Theory. Of course credit existed pre State money.

    I mow Tom’s lawn. Tom picks Ramanan’s apples from the apple tree. Ramanan does Peter’s gardening. Peter picks up my groceries. Great when we know each other. That’s a social credit transaction & can easily precede monetary transaction existence.

    Fast forward, we don’t know each other, why would any of us trust the other. Along comes money and facilitates those exact same transactions.

  20. Pete,

    Only now did I find this post. Very good.

    But I’d like to second Jacob Richter’s comment: what about the Speenhamland system?

  21. I don’t really have anything to add to what I wrote in a previous post on my preference for a combined ‘job or income guarantee’ (JIG) (see in particular the section, ‘A JIG Would Have Built-in Safeguards’)::

    Some Reasons for Guaranteeing Both an Income and a Job

    Basically, my view is that the basic income component would help to protect the job guarantee from being reduced to workfare and the job guarantee component would help to protect wages.

  22. Perhaps I should add (it is alluded to briefly in the earlier post) that although the inclusion of a job guarantee is ideal (IMO), other safeguards to wages are available. Most notably, centralized wage bargaining (of which the minimum wage is a minimalist example) and a strong role for public sector employment (which defines various pay scales against which private-sector employers need to compete) would help to safeguard wages in the event of a standalone basic income being introduced.

  23. Peter, given the Yves Smith material you might wish to reconsider the JIG “in its leanest form” (your Feb. 2014 blog). I know that Yves Smith used Speenhamland as a wage price polemic against standalone basic income, but the details of Speenhamland – a basic income clawed back by wages earned during the period – look too damn similar to the “leanest form” JIG, where you stated “the option of accepting a job-guarantee position or, by opting out of the labor force, a means-tested but otherwise unconditional income payment.”

    Also, could you please explain “centralized wage bargaining” further?

  24. Actually, I meant to write “centralized wage determination”, not “centralized wage bargaining”. Sorry for any confusion caused. Centrally determined pay and conditions for various categories of workers (we had this in Australia for quite some time) would help protect against any tendency for wages to fall in the presence of a basic income. Fully disclosing public sector pay and conditions also creates a competitive pressure on other employers to offer jobs of comparable appeal.

    Regarding clawback, basic income proponents have sometimes argued that the policy would increase incentives to find employment relative to the present situation because of currently high effective marginal tax rates on benefit recipients moving from unemployment to employment. Their argument is based on standard neoclassical labor supply analysis. I don’t find it particularly convincing, but nor do I find incentive arguments claiming the opposite convincing.

    When it comes to shaping incentives, I don’t think we should necessarily be incentivizing formal employment over voluntary productive activity that takes place outside the workforce. Further, I don’t think we should necessarily be incentivizing production over leisure unless the former is defined so broadly as to entail the production of life in its fullest sense, including through leisure. I think we should be progressively separating labor time and income rather than reinforcing the wage-labor relation.

  25. Sorry if this point has already been made but I’ve got a curry to eat.

    Unless there is only one private bank, what do the banks settle up with without underlying currency ? I suppose we could always go back to gold.

Comments are closed.