Friday, December 24, 2010

Government Debt as Private Savings

I've heard that Treasury debt (this is important) is akin to private savings. I'm pretty sure Greg (our most prolific commentator) has made this claim, but I've seen it elsewhere. If I'm not mistaken (and I hope Greg will clear up any misconceptions as I state the claim) this is the position of Modern Monetary Theorists, aka chartalists.

Note: I think they prefer MMT to chartalist since the latter easily translates to charlatanist. Hee hee!

So here is what I think the MMT position is: Treasury debt is private savings in the sense that, to place the debt people need to be willing to purchase the bonds. Thus Treasury debt is money owed to private individuals. This is 2/3rds true - of the 13.5 trillion debt outstanding (Sept. 2010, according to the Treasury bulletin), 9.1 trillion is held by "the public." The remainder is in "Securities held by Government Accounts," or inter-agency debts.

Let's focus on the debt to the public. For arguments' sake, I'm go to abstract away from the level of foreign holdings (e.g. China) of U.S. debt and assume that all debt is held by U.S. citizens. This avoids having to discuss trade issues, which are not part of the story.

So in this set up, every dollar of public debt issued by the Treasury is ultimately held by U.S. citizens. You can own Treasury debt directly, or through some agent (e.g. mutual fund). Thus Treasury debt is de facto private savings. So far, so good - I have no quarrel to this point.

One question that comes up is: why are people willing to hold Treasury debt? Well, it does promise a nominal return (or real, in the case of TIPS), so the interest induces people to hold the debt. Also, Treasury debt is default-free, so it is less risky than other similar types of debt (including municipal, corporate and foreign debt). So, as a finance guy, it makes sense to put some Treasury debt into your portfolio. This default-free feature becomes very important when people are more uncertain than usual about the future performance of other bond issuers in general.

Now bonds are debt, and so must be paid back eventually. Every bond issuer has two options to pay back existing debt: use cash on hand (from retained earnings in the case of corporations, or from taxes in the case of governments) or refinance the debt (issue new bonds and use the proceeds to pay off the old ones). Refis are pretty common - corporations for example like to target a certain cost of capital and also look to minimize cost of capital, so they may shorten or lengthen maturities using refis when the markets are favorable. Refis are also how the Treasury operates; it typically just rolls over the debt by issuing new bonds to pay off the old ones (in addition to new ones to pay for more spending).

But if you don't have cash, and can't refi, you're in default. The Treasury is default-free because of the third option: printing money (I've discussed this before). So as long as bonds are denominated in nominal terms, the Treasury can never default.

Well, enough with the institutional details. When the Treasury issues a new bond, some people decide to buy the bond. So in a realistic senses, Treasury debt is private savings.

But what if no one wants to buy the bond?


33 comments:

  1. Good post.

    I think you're framing it mostly correctly. Where I think you are a little "out of paradigm" so to speak,is in the notion that the govt must tax to pay back the bond. The important point that MMTers make is that taxes "fund" nothing at the federal level. At the state levels yes but not the federal level.
    The govt first issues the very money that is then used to buy the bond. It spends first, sells debt and then we have an increase in "net financial assets" ( this is the preferred term of MMTers) in the private sector. As Warren Mosler says, a govt debt account is simply a savings account and when they "pay off the debt" they move it back into a checking account.

    Yes the interest payments are "new" income flows but in the vast majority of cases they are small and simply rolled into more savings.

    So I think its not just akin to private savings, its THE measure of private savings. There is no other. All spending that has been done at any time through our last 80 years at least has been accompanied with debt issuance. When the govt "buys" a tank. The bank of the tank seller gets a deposit of say 1 million. Then, by law 1 million of debt must be issued to "offset" this spending, but the spending already occurred AND the reserves in the banking system have been increased by the exact amount of the spending, they had to, thats the official recording of the spending. When the debt is issued the reserves are reduced by 1 million again BUT there is now 1 million more of Treasury securities.

    Your last question is always the one the critics of MMT think is the fatal flaw in the paradigm. They usually concede every other point, because they really cant be refuted, they are operational realities. This supposedly fatal flaw though has a couple of responses

    1) Its never happened. Not in a country that floats its own currency and only borrows in its own currency. Russia got in trouble because they ran a peg, as does China. A peg takes away sovereignty.

    2) If it did happen the govt COULD buy its own debt, via the central bank. Which has happened before and is happening in Europe as we speak.

    Now of course there are limits to how far this can go, BUT as long as there is still productive activity, people are still trading things denominated in US dollars there wont be a crisis.

    As Cullen Roche (Pragmatic Capitalist) says, do you really think people are going to turn their back on 15 trillion dollars of productive output, 25% of the worlds output? Sure its in the realm of possibilities but its highly unlikely.

    ReplyDelete
  2. And just to add one thing.

    When you talk about debt owed to foreigners, the checking/savings account analogy is the most apropos. This is because China actually acquires dollars by selling us stuff. They then take these dollars and "save" them (they need a supply of them to maintain their peg too) in an account at the fed called a Treasury account, which pays interest.

    When you have a savings account at a bank and you move some to checking, or all of it, would you ever describe that operation as the bank "owing you" money? I would think not. Its simply moving money from an account that pays 2% to one that pays 0.25% or something. This is EXACTLY the operation that occurs when we pay off a bond for China. Does this sound potentially earth shattering? Even if it happened to all 4 trillion at once, all that the case would be is China now has to make decision of what to do with THEIR money and none of those things are bad for us. They can buy some US goods (great, will lead to US sales!), they can move it back to another interest bearing account (the fed will always give them that option.... might only pay 1% but...) or they can "buy" someone elses currency and save with them (which doesnt affect us at all).

    ReplyDelete
  3. Greg,

    Thanks for posting so quickly. Now that I think I have the correct view of the position, I can think more carefully about it.

    To be clear, this could not describe anything but a fiat money arrangement with a central bank, is that so?

    ReplyDelete
  4. I dont think thats true. Even in the 20s when we were on a gold standard, we ran deficits and accumulated debt. The govt spent beyond their gold store quite regularly as I understand it.

    Mosler says that the only difference between Treasury securities and cash, in the gold standard days, was that cash was convertible to gold on demand. Treasury securities needed to be held for the term and could be converted later (when, it was assumed, the gold would be available) but were often rolled over.

    The operation of spending then debt has been the way of govts all along
    as I understand.

    The salient point with all this is that since govt is the currency issuer, it spends first and then taxes or sells debt. There is no currency in existence UNTIL it is spent. Kind of hard for some folks to latch onto. Now this is even true with a commodity backed currency its just that the commodity backing theoretically limits the amount that is issued.

    ReplyDelete
  5. For MMT to be a true statement of the nature of money creation, the federal government must continuously run budget deficits. This is not the case, however: http://research.stlouisfed.org/fred2/series/FYFSD?cid=5

    What is MMT really trying to state? What is it a theory of? The nature of money, the nature of fiscal debt? It seems like an accounting statement combined with an operational view of debt that starts in the middle of the transaction.

    I must admit, having read things here and there, this MMT appears to be a collection of assertions without logical ties or theory supported by empirical evidence.

    Greg, if you can, please point out some journal articles or books that state the theory and show the support that this theory is correct. At this point, I don't even know what MMT is trying to say, beyond some operational statements of debt financing.

    ReplyDelete
  6. Treasury debt is private savings in the sense that, to place the debt people need to be willing to purchase the bonds.

    MMT'ers say that bond sales are a potentially inflationary hold over from the days we had to compensate savers during the gold standard. It's also used to manage reserve balances to maintain a stable target overnight rate. The real reason they say government spending adds to the savings of the non-government sector is because the only way for one sector to net save is to acquire the liabilities from another sector. Reserves and cash are federal reserve liabilities, treasury secs are liabilities of the treasury and assets to you and I- the non-government sectors.

    Whenever the government spends it does so by issuing a check to the recipient. When the check is deposited, the recipient's bank marks up the recipient's account and the Federal Reserve marks up the recipients bank's reserve account at the Fed the same amount. Whenever it deficit spends (G>T) it spends more than it injects more assets than it drains. That's the conclusion anyways, if you're still skeptical that's actually what happens I'd recommend this new article.

    But what if no one wants to buy the bond?
    I) First, I doubt it will happen- it would mean agents would voluntary forgo higher interest on their savings, and the arbitragers of the world wouldn't take advantage of the massive opportunity of treasury rates differing significantly from the fed targets.
    II) Second, even if it does, it would have to be sustained for a few years or more to have a significant effect on the overall cost of debt service.
    III) Even if it does happen, the treasury could issue short-term debt only.
    IV) The Fed could always target long-term treasury rates either directly or via swaps.
    V)Congress could require the Fed to give the Treasury overdrafts at the target rate, and then pay interest on the reserve balances created--net effect on the budget is the equivalent of debt service at the Fed's target.

    In short, even if it does happen and is sustained, it's still a choice by Treasury and/or Fed to allow it to matter. On the gold standard (or any fixed exchange rate regime) it was different because government debt was convertible to physical gold at a fixed exchanged rate. Public policy had to focus on defending the peg over serving the public purpose. This is why countries usually abandon their pegs when the need/desire arises.

    ReplyDelete
  7. ..operational view of debt that starts in the middle of the transaction.
    Chartalists are intensely interested in the origins and social nature of money, and MMTers have taken many (but not all) of their insights. Most MMTers and Greg mentioned that there is no possible way government could tax (or borrow) unless it first issues the currency as a matter of logic. Seems like a good starting point, no?

    For MMT to be a true statement of the nature of money creation, the federal government must continuously run budget deficits. This is not the case,

    Why must it continuously run budget deficits for MMT to be true? This is not true. MMT just gives the insight that when government runs a surplus it drains more financial assets than it adds- It drains the non-government sector's stock of savings. If you'll notice those few times when government did run a surplus it usually precipitated a recession or depression.

    What is MMT really trying to state? What is it a theory of? The nature of money, the nature of fiscal debt?
    There are two aspects to Modern Money Theory, descriptive and policy. Actually to call it a theory is a misnomer. There is nothing theoretical about it- it's just a description of operational realities as they exist. One part MMT describes how our monetary system operates, with a lot of focus on the transactions between the government and non-government sectors, the so called "vertical" relationship. For analysis purposes the non-government sectors are usually divided into private and foreign sectors. The second aspect of MMT gives policy recommendations given operational realities mostly how to achieve price stability AND full employment.

    I must admit, having read things here and there, this MMT appears to be a collection of assertions without logical ties or theory supported by empirical evidence.
    Hardly. MMTers have explored the empirical evidence more than any school of economic thought that I've come across. Where else will you hear about what role TT&L accounts play in the economy or the complex calculation of stocks flows between sectors? They also have a stunning track record of being right- the housing bubble and subsequent financial melt down, the euro's inability to cope with a severe recession, bank reserves not being inflationary, the implications of the fed paying interest on reserves... I can go on and on..

    I don't even know what MMT is trying to say, beyond some operational statements of debt financing.
    Books-"The nature of money" or "Capitalism" by Geoffrey Ingham. There are many more books, but this was how I learned about it. Others might suggest "Monetary economics: an integrated approach to credit, money, income, production and wealth" by Dr. Godley and Dr. Lavoie and Understanding Modern Money by Dr. L Randall Wray.

    Articles- Can Taxes and Bonds Finance Government Spending? By Dr. Stephanie Bell
    Interest Rates and Fiscal Sustainability By Dr. Scott Fullwiler
    Understanding Policy in a Floating Rate Regime By Dr. L. Randal Wray

    ReplyDelete
  8. Yeah!! What HE said!


    Thanks Tschaff!

    Let me add that I'm trying to elucidate some of this with your prior thought experiment post. Getting you to think about starting your own currency in your own country will get you to begin to see this.

    You took the route of completely private issued currencies through banks but I still think you have left a lot of holes, I hope to explore it more in that thread.

    So for the sake of argument take the route that you will issue a new state currency and either decide to peg it to gold or float it. It should be obvious to you that no citizen can pay a tax in your currency until you spend it first, and a deficit will be when you have in net collected less than you spent out.

    Ultimately though the main point of MMT is that the only considerations should not be some arbitrary level of deficit or debt but real economic factors like inflation or unemployment.

    ReplyDelete
  9. Ah, now I see. MMT is Post-Keynesian. More and more puzzle pieces are falling into place.

    There's also quite a bit of confusion here regarding the banking sector. I'm not going to address everything, just the more outstanding problems I see. This has all given me quite a headache.

    "MMT'ers say that bond sales are a potentially inflationary hold over from the days we had to compensate savers during the gold standard. It's also used to manage reserve balances to maintain a stable target overnight rate."

    1st sentence I don't understand at all. Who is "we?" Are you trying to say the inflation premium is weighted differently according to the monetary regime? 2nd sentence: overnight reserves are managed using interbank lending against acceptable collateral. These are Treasury bonds often, but this is a secondary market action. If there were no Treasury bonds, other collateral could be used. In fact, overnight repos can and are collateralized against a variety of securities. The Treasury doesn't do this, the Fed does. I am detecting some confusion on the MMT part between Treasury and the central bank.

    "the only way for one sector to net save is to acquire the liabilities from another sector."

    What is meant by net savings? Savings in excess of investment? If not, then this is stupendously false, since savings can become equity and equity is not, by any means, a liability. It is ownership. A world of legal and economic difference between the two.

    And then if so, net savings would be simply cash hoarding, it seems.

    ReplyDelete
  10. "When the check is deposited, the recipient's bank marks up the recipient's account and the Federal Reserve marks up the recipients bank's reserve account at the Fed the same amount. Whenever it deficit spends (G>T) it spends more than it injects more assets than it drains. "

    Take a very simple balance sheet for the Fed. On the asset side, put cash. On the liability side, put debt. The government has no equity. The income statement is revenue (tax + seignorage) - expenses (spending including interest). As the government gets money, the cash side goes up and deposit liabilities go up. As it spends, cash goes down and deposits go down. If it spends more than it gets in tax revenue, it must get more cash - by increasing liabilities. From this perspective, the only difference between liabilities is the maturity structure. So to say it injects more assets than it takes out if it deficit spends seems incorrect.

    "But what if no one wants to buy the bond?"

    Your response missed entirely the point of the question. I will try to come at it from a different angle below. I'm trying to get at understanding why MMT thinks government must deficit spend to create money.

    ReplyDelete
  11. "Chartalists are intensely interested in the origins and social nature of money, and MMTers have taken many (but not all) of their insights. Most MMTers and Greg mentioned that there is no possible way government could tax (or borrow) unless it first issues the currency as a matter of logic. Seems like a good starting point, no? "

    MMT doesn't seem to be studying the origin of money at all if you're using fiat money as the starting point. The crown used to tax without fiat money - it collected taxes according to the currency that was in use, like gold or silver. When paper money was first introduced, it was in the form of a warehouse receipt that indicated the bearer of the receipt had physical currency at the bank. Taxes weren't payable in paper - they were people in market-based currency.

    I've conceded the point before that the government can win a currency competition by making other currencies illegal. It doesn't mean it has to deficit spend to get the currency into circulation. In fact, it doesn't do this. Currency is sent to banks for the purpose of circulation. Especially now that most "money" is not paper currency or coin. The Fed can just through a few extra zeros on some accounts and be done with it.

    "MMT just gives the insight that when government runs a surplus it drains more financial assets than it adds- It drains the non-government sector's stock of savings. If you'll notice those few times when government did run a surplus it usually precipitated a recession or depression."

    I see. The last sentence is 100% false. I did some analysis and it turns out government surplus/deficit is unrelated to the next year's growth rate in real GDP. I even skewed in favor of finding a relationship by leaving G in there.

    ReplyDelete
  12. "There are two aspects to Modern Money Theory, descriptive and policy. Actually to call it a theory is a misnomer."

    Apparently.

    "There is nothing theoretical about it- it's just a description of operational realities as they exist. One part MMT describes how our monetary system operates, with a lot of focus on the transactions between the government and non-government sectors, the so called "vertical" relationship. For analysis purposes the non-government sectors are usually divided into private and foreign sectors. The second aspect of MMT gives policy recommendations given operational realities mostly how to achieve price stability AND full employment."

    It seems MMT picks up with an already existing economy (namely the U.S.) and starts trying to describe a circular relationship without really understanding the beginning stages. So really MMT is answering (incorrectly in my view) the question of how does new money get created in an economy? Not, how does money come to exist in the first place.

    And, if the U.S. is the test case, you have a currency that was originally backed by gold and because of U.S. economic supremacy the dollar became the world reserve currency. This is the most special case I've ever heard of. MMT, to be a theory of money, has to apply to money generally.

    "MMTers have explored the empirical evidence more than any school of economic thought that I've come across."

    Really? I doubt this very much. I'm sympathetic to heterodox schools but this is a very strong statement, and needs to be backed up.

    "Where else will you hear about what role TT&L accounts play in the economy"

    Well, you got me there. I have no idea what the abbreviation "TT&L" is supposed to be.

    "or the complex calculation of stocks flows between sectors?"

    Um... anywhere? And do you mean "stocks and flows?"

    "They also have a stunning track record of being right- the housing bubble and subsequent financial melt down,"

    Yeah, Austrians called this one too. I'm curious as to why MMT thought there was a house bubble+ensuing crash?

    "the euro's inability to cope with a severe recession,"

    Yeah, Friedman called that one, although he got the timing wrong.

    "bank reserves not being inflationary,"

    No kidding. Everyone knows money has to circulate for prices to change.

    "the implications of the fed paying interest on reserves..."

    Yeah... MMT wasn't unique on this score either. Any neoclassicist can tell you that story.

    ReplyDelete
  13. "Net savings" is when the non govt sector as a whole is spending or paying in taxes less than it is receiving in income. Its a flow thing.

    When the non govt sector is spending more than it is receiving in income it is dis-saving or spending down its stock of savings.


    Yes savings for an individual can become equity. I can "buy" something with my money market cash, but the seller now has that money market cash when he "sells". The amount of money market cash does not change in that transaction. I spend 10,000 he he sells 10,000. My money market or checking or savings account goes down by 10,000 his goes up by the same amount and no net change. The stock he sold me has simply gone up in price but there is no more money in the system. Only the price of the equity vehicle has changed. At any point in time, "savings" can be measured and in fact is measured, on a macro level, as the level of national debt held domestically and the amount of cash in reserve accounts. It cant be anywhere else. All spending is accounted for in the banking system. Otherwise there has been counterfeiting.

    ReplyDelete
  14. Prof, your making the error in thinking that only paper money not backed by gold is fiat. The crown decided by fiat what was necessary to settle tax liabilities and THAT, thusly became money. It may in fact been in use as money already and it may have in fact been gold coins but thats irrelevant to our discussion. The state CAN make dandelions money by requiring them for tax purposes, and if people cant grow them for what ever reason, the state must issue them first so they can be in circulation.

    The fact remains that when a "State" makes something money, requires it to be paid in tax settlement, it must first spend it into the system in order for the taxpayers to acquire it. The direction is always from state TO taxpayers.

    Your aversion to this is mostly ideological, I get it. But its how our modern system operates. If you get your chance to have your own country and simply have multiple competing currencies, issued by private parties, so be it. But that is not what we have.

    ReplyDelete
  15. ""Net savings" is when the non govt sector as a whole is spending or paying in taxes less than it is receiving in income. Its a flow thing."

    Okay. Much confusion can be avoided if you MMTers call this what everyone else does: savings.

    "Yes savings for an individual can become equity."

    Equity means ownership in an enterprise. Hope I'm being clear here.

    "At any point in time, "savings" can be measured and in fact is measured, on a macro level, as the level of national debt held domestically and the amount of cash in reserve accounts."

    No. This is your conclusion, and it is what you must argue to be true. You cannot use it as a premise. To do so is a logical fallacy.

    Also, here you define savings as a stock. Above you claim it is a flow. Make up your mind.

    "It cant be anywhere else. All spending is accounted for in the banking system. Otherwise there has been counterfeiting."

    That's fine. But the increase in the value of my equity ownership is not. Nor are changes in the value of, say, derivative contracts. There is more to the monetary system than the banks.

    ReplyDelete
  16. "Prof, your making the error in thinking that only paper money not backed by gold is fiat."

    No - fiat just means that some authority declares something to be so, and backs that authority with coercive force.

    I'm trying to understand what monetary system MMT is talking about. You say taxation gives money is value. This is not so, which is why I raise the gold standard, which is where paper money originally got its value.

    "The crown decided by fiat what was necessary to settle tax liabilities and THAT, thusly became money."

    One form of money, which may or may not coincide with the money used in the market.

    "It may in fact been in use as money already and it may have in fact been gold coins but thats irrelevant to our discussion. The state CAN make dandelions money by requiring them for tax purposes, and if people cant grow them for what ever reason, the state must issue them first so they can be in circulation."

    What nitwittery would possess a state to tax something that wasn't already in circulation? Find some instance where this has happened and I'll give it some consideration.

    "The fact remains that when a "State" makes something money, requires it to be paid in tax settlement, it must first spend it into the system in order for the taxpayers to acquire it. The direction is always from state TO taxpayers."

    See above. Why would any state do this? Sure, issue greenbacks. Nobody will use them if you just issue them - unless they represent a claim to something that is already money. The money was already held by people. The state causes a standardized 'warehouse receipt' to be issued.

    Money, currency, whatever - is issued by the Fed, not the Treasury. It doesn't have to be "spent" into the economy. It can just be issued. After all, there have been times of no Federal debt and money has existed.

    "Your aversion to this is mostly ideological, I get it."


    No, but you can think that if it makes you feel better.

    "But its how our modern system operates."

    I don't think so. You've got some operational aspects correct, but there rest is illogical drawn from some operational and accounting facts. I've said that before, and I have yet to be correct.

    "If you get your chance to have your own country and simply have multiple competing currencies, issued by private parties, so be it. But that is not what we have."

    But here's exactly my point: an economic theory needs to be able to handle the history of money. If you want to describe why money has value, or why Federal debt doesn't matter, you have to explain why non-state money has existed, and why high Federal debt is related to, for example, low economic growth.

    ReplyDelete
  17. Have a quick read of 'The 7 Deadly Innocent Frauds of Economic Policy' at: http://www.moslereconomics.com/?p=8662/ thanks!

    It should answer a lot of your questions.

    There are also a few journal papers at www.moslereconomics.com as well.

    ReplyDelete
  18. Alright Jeff

    Where is savings measured then? How do we know how much savings, form a macro perspective, there is?

    If its not registered in the banking system where is it registered?

    "Savings" is a stock, saving is a flow. Its the rate at which savings is increasing. Dis saving is the rate at which savings are being depleted.

    You save out of income. Anything you dont spend (a flow) you, by definition, are saving. Your accumulation of saving (savings) can be used to purchase equity, buy a commodity or sit as cash in a money market.

    This;
    "Money, currency, whatever - is issued by the Fed, not the Treasury. It doesn't have to be "spent" into the economy. It can just be issued. After all, there have been times of no Federal debt and money has existed."

    So the fed just gives money to someone? The fed exchanges reserves for collateral. They give you a more liquid form of something you hold that has value but is hard to sell. There is no net addition of financial assets to the economy except as a fiscal operation. Either spending is increasing or taxation is decreasing.


    And this;

    "What nitwittery would possess a state to tax something that wasn't already in circulation? Find some instance where this has happened and I'll give it some consideration."

    THE EURO! Prior to the late 80s it didnt exist. It was created, exchanged for existing currencies and then collected in taxes. It wasnt already circulating until the EU put it into circulation.

    The point about taxation giving fiat value is not that it is what solely gives value but that it will immediately HAVE value once you assign a tax to it. The dollar is the highest on the hierarchy of currencies here because the IRS requires it. Eliminate taxation and something else might replace the dollar.

    ReplyDelete
  19. Prof J,

    You may find the following links to be of use.
    http://tinyurl.com/279u5da
    http://tinyurl.com/2akm2jc

    ReplyDelete
  20. The basic answer to why the national (government) debt is the record of nongovernment net saving is that only the federal government can increase in nongovernment net financial assets (through deficit spending), since all money creation in the private sector through lending nets to zero (loans create deposits). The $-4-$ deficit offset with tsy issuance rule means that net financial asset increases through deficit spending are necessarily "saved" by nongovernment in the form of tsys.

    Thus, the statement that national debt is equivalent to nongovernment saving of net financial assets is an ex post facto accounting identity.

    I suggest reading Warren Mosler's Soft Currency Economics for a detailed explanation of how the current (post 1971) US monetary system operates.

    ReplyDelete
  21. I am detecting some confusion on the MMT part between Treasury and the central bank.

    This is not a matter of confusion. It's a matter of the central bank being a wholly owned subsidiary of the Treasury (both institutionally and financially). Exchanges between the Treasury and the CB are similar to exchanges between the Treasury and the Ministry of the Interior (or exchanges between Accounting and Operations in General Electric, for that matter).

    Now, it may be politically expedient to pretend that the central bank is a distinct entity, just as it may be politically expedient to pretend that the public pension system is a "trust fund." But that is propaganda for public consumption, not a matter of economic necessity.

    Take a very simple balance sheet for the Fed. On the asset side, put cash. On the liability side, put debt. The government has no equity.

    Actually, most central banks do have an equity buffer against rediscounts that go bad. This is another convenient fiction, because there is nothing, operationally, that prevents the central bank from operating with zero or negative equity. It's just that the central bank is usually staffed by bankers, and zero or negative equity makes bankers lose sleep.

    If [the government] spends more than it gets in tax revenue, it must get more cash - by increasing liabilities. From this perspective, the only difference between liabilities is the maturity structure. So to say it injects more assets than it takes out if it deficit spends seems incorrect.

    I'm not sure I follow you here. When the sovereign runs a deficit, total consolidated sovereign (Treasury plus Central Bank) liabilities to the rest of the world increase, no? Sovereign liabilities to tRotW is private sector (and foreign) assets, no? So deficit spending increases private sector plus foreign assets, no?

    So really MMT is answering (incorrectly in my view) the question of how does new money get created in an economy? Not, how does money come to exist in the first place.

    Yes. The question of how the institution of money came to be is, while not without meaning and not without interest, of only historical relevance. A proposed treatment protocol for asthma does not need to digress into the evolutionary history of the mammalian respiratory system, even though a solid grounding in evolutionary history probably helps to develop the hypothesis.

    - Jake

    ReplyDelete
  22. And, if the U.S. is the test case, you have a currency that was originally backed by gold and because of U.S. economic supremacy the dollar became the world reserve currency. This is the most special case I've ever heard of. MMT, to be a theory of money, has to apply to money generally.

    No. It only has to apply to money in mature credit economies, because MMT'ers are only attempting to describe mature credit economies. That they do not attempt to describe barter economies is not a flaw in the theory, it is a recognition of the fact that in order to say anything interesting about a subject, you have to narrow down your area of interest.

    MMT'ers seem to narrow down their subject of interest to only encompass the credit/financial part of the economy. And for most uses, that's close enough for corporate work, because in a mature credit economy the credit part of the economy is the main story. It is the (private) financial sector that drives the business cycle, and all production and trade in the physical sector is accompanied by financial stocks and flows anyway (but the inverse is not necessarily true - a naked credit default swap is an example of a financial transaction that has no physical companion).

    Now, if you want to describe an emerging economy, you can't rely exclusively on MMT, because an emerging economy will have a large sector of wholly non-monetised economic activity (subsistence farming and household handicrafts). But MMT'ers are mainly attempting to describe the contemporary state of affairs in Europe, the US and the Commonwealth. And for that purpose, MMT models remain a distinct improvement over Marginalist models, because the latter are not stock/flow consistent.

    My gut feeling, which I am currently trying to flesh out a bit, is that even the description of mature credit economies can benefit from an integrated treatment of the physical and financial sectors. But that is a refinement, not a fundamental disagreement - the credit part of the economy is still going to be the main show.

    But here's exactly my point: an economic theory needs to be able to handle the history of money.

    Why? That's a different institutional setup. Why would you expect any single theory to be able to describe all possible institutional setups and still be able to say anything non-trivial? Would you also expect a single economic theory to be able to describe both a property-based sedentary economy and a communal hunter-gatherer economy?

    If you want to describe why money has value, or why Federal debt doesn't matter, you have to explain why non-state money has existed,

    Not necessarily. Non-state money still exists, but it's a different sort of beast. State money is a new invention, in the same way that the transistor is a new invention. Studying radio tubes may still be interesting and instructive, but that doesn't mean you'll use the same theory to describe both them and transistors.

    and why high Federal debt is related to, for example, low economic growth.

    It isn't. Not directly, at least, though it will probably coincide with high foreign debt, which tells you that you're doing something wrong in terms of industrial policy.

    - Jake

    ReplyDelete
  23. And, if the U.S. is the test case, you have a currency that was originally backed by gold and because of U.S. economic supremacy the dollar became the world reserve currency. This is the most special case I've ever heard of. MMT, to be a theory of money, has to apply to money generally.


    No. It only has to apply to money in mature credit economies, because MMT'ers are only attempting to describe mature credit economies. That they do not attempt to describe barter economies is not a flaw in the theory, it is a recognition of the fact that in order to say anything interesting about a subject, you have to narrow down your area of interest.

    MMT'ers seem to narrow down their subject of interest to only encompass the credit/financial part of the economy. And for most uses, that's close enough for corporate work, because in a mature credit economy the credit part of the economy is the main story. It is the (private) financial sector that drives the business cycle, and all production and trade in the physical sector is accompanied by financial stocks and flows anyway (but the inverse is not necessarily true - a naked credit default swap is an example of a financial transaction that has no physical companion).

    Now, if you want to describe an emerging economy, you can't rely exclusively on MMT, because an emerging economy will have a large sector of wholly non-monetised economic activity (subsistence farming and household handicrafts). But MMT'ers are mainly attempting to describe the contemporary state of affairs in Europe, the US and the Commonwealth. And for that purpose, MMT models remain a distinct improvement over Marginalist models, because the former are stock/flow consistent and the latter are not.

    My gut feeling, which I am currently trying to flesh out a bit, is that even the description of mature credit economies can benefit from an integrated treatment of the physical and financial sectors. But that is a refinement, not a fundamental disagreement - the credit part of the economy is still going to be the main show.

    But here's exactly my point: an economic theory needs to be able to handle the history of money.

    Why? That's a different institutional setup. Why would you expect any single theory to be able to describe all possible institutional setups and still be able to say anything non-trivial? Would you also expect a single economic theory to be able to describe both a property-based sedentary economy and a communal hunter-gatherer economy?

    If you want to describe why money has value, or why Federal debt doesn't matter, you have to explain why non-state money has existed,

    Not necessarily. Non-state money still exists, but it's a different sort of beast. State money is a new invention, in the same way that the transistor is a new invention. Studying radio tubes may still be interesting and instructive, but that doesn't mean you'll use the same theory to describe both them and transistors.

    and why high Federal debt is related to, for example, low economic growth.

    It isn't. Not directly, at least, though it will probably coincide with high foreign debt, which tells you that you're doing something wrong in terms of industrial policy.

    - Jake

    PS: My apologies if this makes a duplicate post - I think (but am not completely certain) that the software ate my previous attempt.

    ReplyDelete
  24. "But what if no one wants to buy the bond?"

    They will have to stick to cash then. What is the problem?

    Any transaction with cash within private sector just swaps the cash. If you bring cash to your bank, your bank will bring to central bank. Moreover FED now issues overnight "bonds" exclusively to banks with 0.25% guaranteed coupon. Banks are super-happy. But banks are still private sector and Fed is still part of the government. And everything that FED makes above fixed 6% belong to Treasury.

    So what is the problem with noone buying bonds? They will stick to zero-income zero-maturity financial government liabilities called cash.

    Ah, there is another way to get rid of cash. Spend on consumption! Which increases business activity, taxes and lower budget deficits and finally even brings surplus. So go and spend your cash but then it is not savings.

    ReplyDelete
  25. One other thing Jeff.

    When you say this: "high Federal debt is related to, for example, low economic growth." ---- define "high". We have been accumulating federal debt at a pretty consistent rate since our countries inception and have had the most vigorous economy in the history of the world ....no?

    Times when we have made special targeted efforts to reduce debt have been followed by recessions. Every time of surplus has been followed by a recession, not the next year necessarily, but soon.

    I think you are looking at the last 3 years and seeing the debt to GDP ratio rising and concluding (falsely in my view) that the rising debt is causing the low GDP. In fact the debt is a stock and it has risen in relation to GDP because the GDP has fallen. The GDP fell first, causing the rise in the ratio.

    The debt is a source of income for some in the non govt sector is it not?
    That income can be spent or saved. If GDP falls towards zero it cannot be because the level of income from govt bonds is too high, can it? How can an increase in income lead to decrease in spending?

    Think of our debt as the stock of "miles of roads constructed in the US" and our GDP as the flow of "miles driven by US drivers". As the flow falls how can the answer be to destroy more roads?

    ReplyDelete
  26. Well, the MMTers are certainly coming out of the woodwork!

    I won't respond to every comment, since a lot of them are repetitive and stuff I've addressed here and elsewhere on this blog.

    I did suspect that this was supposed to describe money creation in the post-1971 U.S. That means you are starting from a point well past "go" with an already-accepted currency floating around domestically. Moreover it is the currency standard for the world. Thus MMT is trying to describe something, like I said, starting in the middle of the game. It seems more and more than MMT is an apologia for the U.S. to have huge debts.

    Regarding debt/GDP and slow growth, Rogoff and Reinhart put the historically acceptable level at 90%.

    In my analysis of deficit/surplus and GDP growth, I specifically excluded the last three years, again to bias in MMT's favor. No relationship was found. So how soon is soon?

    The Fed is not a wholly owned subsidiary of the Treasury. It is a joint venture of the government and the banks.

    When the Treasury issues new bonds, where does the money to buy the bonds come from? Oh, wait, I know! The Treasury printed it up and gave it to the banks ahead of time! Gotcha, gotcha. So if I use all my money to buy bonds, then, how do I pay my taxes? Oh right, more money from Treasury. Right....

    The Fed manages the money supply through the branches and the member banks. The Fed cannot buy bonds directly from the Treasury. All primary Treasury auctions are conducted with private banks.

    Finally, you all seem to be unaware the public debt crowds out private investment. There is mucho evidence for this, but it's mostly Austrian or neoclassical, so I'm sure you think it's nonsense.

    How can the debt be a source of income? The spending that the debt pays for is.

    And, finally, the Euro. A negotiated record, that took years and years, and is an even worse idea than a national currency. But note that was a conversion of already really existing currencies, so it is not the same as inventing a new one out of thin air. And, the Euro is issued by the ECB, not the governments. So, does MMT apply to them? Or only when the Federal government is the money monopolist.

    I will post no more replies here. This discussion, like public debt, is crowding out more important investments.

    ReplyDelete
  27. Here’s a summary of sorts:

    Divide an economy (e.g. US) into three sectors – government, private, and foreign.

    Each sector accumulates a flow of financial assets (liabilities issued by US entities) and liabilities (claims on US entities). Each sector develops a NET financial asset position (positive or negative). A positive NFA position increases the equity of that sector. It represents accumulated saving.

    E.g. the foreign sector accumulates an NFA position on US entities as a result of the US current account deficit. That’s accumulated foreign saving.

    Consolidate the private and foreign sectors as the “non government sector”. That leaves two sectors making up the whole – government and non government.

    Within the consolidated non government sector, all financial assets and liabilities net to zero, except those with the government as counterparty. MMT sometimes refers to this financial interface as the “vertical” position.

    The “vertical” position of non government with government is usually a positive net financial asset position, which corresponds to a cumulative government deficit.

    Within the government sector, consolidate the treasury with the central bank. Assume the central bank only buys government bonds for its assets – i.e. a “pure” central bank operation, without private sector assets (i.e. assume away QE1 type assets or regular central bank last resort loans to banks, for simplicity).

    Then, FOR THE MOST PART, when the consolidated state (treasury and central bank) runs a deficit, it issues a mix of bank reserves (central bank liability), currency (central bank liability), and PUBLICALLY HELD bonds (treasury liability). Those liabilities become the net financial assets of non government. That mix of state liabilities corresponds to net saving for non government.

    Net saving means saving net of investment.

    ... continued

    ReplyDelete
  28. ...

    In a recession, aggregate demand from non government declines due to excessive desired saving (relative to aggregate demand potential). Another way of saying this is that desired saving exceeds the saving that is associated with investment at the prevailing level of GDP.

    Government can respond to this excess demand for saving by providing a unique source of net saving – deficits.

    In other words, by accounting identity, government must run a deficit in order for non government to run a surplus in terms of net saving and net financial asset accumulation. This is the only way for non government to achieve a saving level that exceeds investment, which is what it effectively wants to do before it will boost spending further in a recession (or in any economy where resources are underutilized relative to full employment objectives).

    Forget Chartalism and taxation causalities for now – it’s interesting, but it’s not absolutely essential, at least for understanding the main thrust of MMT. Chartalism is effectively a stand-alone theory of money that has been adopted as a subset within the full scope of MMT.

    Finally, a key idea of MMT is that an independent central bank is a self-imposed constraint. A sovereign government always has the (ultimate) option of making consolidation of treasury and the central bank fully operational. This means it has ultimate control over the mix of reserves, currency, and bonds issued. And that means it has the option of replacing bond issuance with reserve issuance - for any reason – including the unlikely event that it can’t sell bonds. A sovereign government can feed deficit spending directly into the banking system by crediting bank accounts and bank reserves, without issuing bonds, if it chooses to do so. All of this reflects the fact that the state is essentially a currency issuer rather than a currency user. It only appears to be a currency user (e.g. “borrowing” via bonds) because of the self-imposed constraint of institutionally separate treasury and central bank functions. Also, it is a fallacy that government debt has to be paid back. It never has been. This fact of history squares with the MMT view that non government tends to have a persistent appetite for net financial assets.

    ReplyDelete
  29. And, if the U.S. is the test case, you have a currency that was originally backed by gold and because of U.S. economic supremacy the dollar became the world reserve currency. This is the most special case I've ever heard of. MMT, to be a theory of money, has to apply to money generally.


    No. It only has to apply to money in mature credit economies, because MMT'ers are only attempting to describe mature credit economies. That they do not attempt to describe barter economies is not a flaw in the theory, it is a recognition of the fact that in order to say anything interesting about a subject, you have to narrow down your area of interest.

    MMT'ers seem to narrow down their subject of interest to only encompass the credit/financial part of the economy. And for most uses, that's close enough for corporate work, because in a mature credit economy the credit part of the economy is the main story. It is the (private) financial sector that drives the business cycle, and all production and trade in the physical sector is accompanied by financial stocks and flows anyway (but the inverse is not necessarily true - a naked credit default swap is an example of a financial transaction that has no physical companion).

    Now, if you want to describe an emerging economy, you can't rely exclusively on MMT, because an emerging economy will have a large sector of wholly non-monetised economic activity (subsistence farming and household handicrafts). But MMT'ers are mainly attempting to describe the contemporary state of affairs in Europe, the US and the Commonwealth. And for that purpose, MMT models remain a distinct improvement over Marginalist models, because the former are stock/flow consistent and the latter are not.

    My gut feeling, which I am currently trying to flesh out a bit, is that even the description of mature credit economies can benefit from an integrated treatment of the physical and financial sectors. But that is a refinement, not a fundamental disagreement - the credit part of the economy is still going to be the main show.

    But here's exactly my point: an economic theory needs to be able to handle the history of money.

    Why? That's a different institutional setup. Why would you expect any single theory to be able to describe all possible institutional setups and still be able to say anything non-trivial? Would you also expect a single economic theory to be able to describe both a property-based sedentary economy and a communal hunter-gatherer economy?

    If you want to describe why money has value, or why Federal debt doesn't matter, you have to explain why non-state money has existed,

    Not necessarily. Non-state money still exists, but it's a different sort of beast. State money is a new invention, in the same way that the transistor is a new invention. Studying radio tubes may still be interesting and instructive, but that doesn't mean you'll use the same theory to describe both them and transistors.

    and why high Federal debt is related to, for example, low economic growth.

    It isn't. Not directly, at least, though it will probably coincide with high foreign debt, which tells you that you're doing something wrong in terms of industrial policy.

    - Jake

    ReplyDelete
  30. Prof J,
    It's a shame you're leaving the discussion right as some of the sharpest tools in the shed join us. Anyways, you made many responses to my response that need to be answered.

    1) I did suspect that this was supposed to describe money creation in the post-1971 U.S. That means you are starting from a point well past "go" with an already-accepted currency floating around domestically. Moreover it is the currency standard for the world. Thus MMT is trying to describe something, like I said, starting in the middle of the game. It seems more and more than MMT is an apologia for the U.S. to have huge debts.
    How about going back to Adam Smith and his dried cod? Too recent? How about the Hazelwood tally stick? Still too modern? There is the brittle metal ingots of 1000 and 2000BC. You can read all about them here.

    The disagreement you and I probably have goes back to an 16th and 17th century wonderful debate between Metallists vs Chartalists view of money that still being argued to this day. The Neoclassical and Austrian (I think) school of thought adopting the Metallist view. I read this wonderful paper recently that recaps the debate. It goes on to explain that all money is an asset (one is promised something) AND a liability (someone has promised something). Double entry book keeping will help you find a creditor for every debtor and keep us from saying crazy things. :) Demand deposits are liabilities on a bank's balance sheet. The bank promises to convert its IOU to you to the state's liabilities (at a 1:1 ratio). All the cash in your wallet are liabilities on the federal reserve's balance sheet. See FRB balance sheet. Typically the Fed purchases Treasury debt (treasury liability/fed asset) and creates a new liability on its balance sheet, its an asset swap. Fiscal operations, not monetary are helicopter drops. See here. Without government debt you'd have no cash in your wallet, no treasuries in your retirement account, and your bank would have no reserves.

    So really MMT is answering (incorrectly in my view) the question of how does new money get created in an economy? Not, how does money come to exist in the first place.
    Money is a nebulous concept. I find it's best to avoid it as much as possible and speak strictly of what we are talking about i.e. treasuries, high powered money, commercial paper etc. Bank liabilities (deposits) can expand as loans are made. Bank liabilities are a sort of privileged money. The other sort of privileged money is govt (central bank+treasury) liabilities. So to answer your question bank credit money comes into existence when a bank makes a loan (and creates a deposit). It also can come into existence when the federal government spends. Similarly, bank credit money is destroyed as loans are repaid and government liabilities are destroyed when taxes are collected. JKH spelled this out more clearly than I did above maybe. Anyone is capable of making money, Mosler has a great story where he turns his business cards into money in his classic Soft Currency Economics The key though is getting the money to be accepted.

    ReplyDelete
  31. "bank reserves not being inflationary,"
    No kidding. Everyone knows money has to circulate for prices to change.

    Reserves settle payments and meet reserve requirements, that's it. Bank lending isn't reserve constrained (although it is capital constrained).
    Also, here you define savings as a stock. Above you claim it is a flow. Make up your mind.
    Saving= flow
    Savings = stock
    Regarding debt/GDP and slow growth, Rogoff and Reinhart put the historically acceptable level at 90%.
    Their analysis was a disgrace. Nowhere did they separate countries with sovereign fiat non-convertible floating currencies from those with fixed fx or other monetary arrangements. Without that you'll not be able to differentiate currency users from currency issuers, and think the US federal Government is headed to the same fate as Greece. It's not an apples-to-apples comparison. Nowhere in their book you are referring to could they identify countries with monetary arrangements like ours that became involuntarily insolvent, nor do they propose any explanation why >90% is the magic number to be avoided. For more see this. This isn't to say government debt doesn't matter, or inflation can't occur. Nowhere in the academic literature or the blogs will you find that. We say it matters, but just not in the way most people think.

    When the Treasury issues new bonds, where does the money to buy the bonds come from?
    If you bothered to read one of my original links you would know this.

    Oh, wait, I know! The Treasury printed it up and gave it to the banks ahead of time! Gotcha, gotcha. So if I use all my money to buy bonds, then, how do I pay my taxes? Oh right, more money from Treasury. Right....
    Why is it that some academics think it is intellectually honest to criticize things openly that they have barely studied, and apparently don't even want to.
    . "If you'll notice those few times when government did run a surplus it usually precipitated a recession or depression."

    I see. The last sentence is 100% false. I did some analysis and it turns out government surplus/deficit is unrelated to the next year's growth rate in real GDP.

    I'll not comment on your analytical skills, but lets see the data:
    1817-1821: U. S. Federal Debt reduced 29%. Depression began 1819.
    1823-1836: U. S. Federal Debt reduced 99%. Depression began 1837.
    1852-1857: U. S. Federal Debt reduced 59%. Depression began 1857.
    1867-1873: U. S. Federal Debt reduced 27%. Depression began 1873.
    1880-1893: U. S. Federal Debt reduced 57%. Depression began 1893.
    1920-1930: U. S. Federal Debt reduced 36%. Depression began 1929.
    1998-2001: U. S. Federal Debt reduced 9%. Recession began 2001
    this billy blog post. In the last decade aggregate savings desires of the private sector shrank/private debt soared so the govt could get away with smaller deficits without causing a recession. The current account balance also impacts the consequence of the federal govt budget. See Scott's <a href="http://neweconomicperspectives.blogspot.com/2009/07/sector-financial-balances-model-of_17.html>sectoral balance model of aggregate demand</a> for more on this.

    ReplyDelete
  32. Finally, you all seem to be unaware the public debt crowds out private investment.
    If you were still using loanable funds models you'd be right. However they are particularly inapplicable post-gold standard. Federal government demand for real physical resources can crowd out what's available for the private sector indeed. In a full employment world there are many either/or decisions to be made. In the current environment of tremendous idle productive capacity govt debt is "crowding in" private sector businesses. Cut that injection (govt spending) out of the income stream or raise the leakage (taxes) significanly and you're guaranteed a negative feedback loop of Irving Fischer style debt deflation. As for crowding out investment, what evidence is there? Interest rates are rock bottom and the Fed is able to set them anywhere they want. Net bank balances still haven't shrunk (despite the de-leveraging). Banks aren't lending, but not because they aren't able or interest rates are too high, but rather they don't see credit worthy borrowers (supply side) and businesses are strongly reluctant to take loans (demand side)and expand if they don't anticipate future demand. If you were an airline would you add seats to routes if you don't anticipate them being filled? Furthermore many corporations have been saving far more than their historic norms this is also reduces their demand for credit.

    You have a lot to learn about what you are criticizing. It will take a long time, after reading all your comments, if I can recommend just one paper for you to read let it be Interest Rates and Fiscal Sustainability: http://www.cfeps.org/pubs/wp-pdf/WP53-Fullwiler.pdf It will address the prevailing counterarguments in great detail and lay out a framework for an accurate understanding of actual monetary operations.

    ReplyDelete