Why the Government Must Keep Running Deficits. Forever.
Imagine an economy that consists of two households, one firm, one bank, and one government.
The government issues $50 to each household (maybe they do some work for it), crediting their bank accounts and running a $100 deficit. Voila! There’s money!
Now one household works for the firm, creating $50 in value, goods. The firm gives the household $50 in equity — company stock — basically a promise to give them some amount of money in the future. (The firm posts the $50 in newly created value as an asset on the lefthand side of their balance sheet, and $50 as shareholder equity on the righthand side — a liability).
The household can’t use that equity to buy a pack of gum today, so they want to monetize it — sell it to someone else. There’s only one “someone” — the other household.
But what if the other household doesn’t want to buy it because they’ve only got $50 and want hold it for the future? (It’s the babysitting coop dilemma.)
This is why in a growing economy where extra value is being created through people’s efforts, the government has to run deficits — creating money by crediting people’s/firms’ accounts with newly “printed” dollars.
If people can’t convert that extra value they’ve created into general-purpose “credit” (dollars) that can be exchanged for a variety of goods, they can’t spend. Which 1) gives them notably less incentive to create the value in the first place, and 2) prevents them from continuing the buying/selling log-rolling exercise that is our economy.
You’ve got a lot of newly created value/goods, but nobody with money to buy them.
Imagine if the cumulative government deficits today — the stock of money that government has spent into existence — were at the same level it was in 1900. The economy would be completely inoperable, locked up in primitive barter arrangements for lack of general-purpose money.
This is why government debt is not never, has not ever been, cannot ever be, paid off.
Another way to think about this: money — created, provided, by government as a public good — is a means for us to save consumption for the future. (“Saving consumption” is a funny concept, but it’s what we do when we put dollars under a mattress.) The only other way to do so is to create consumable real assets that will last, including those that can be used to create more consumables in the future (themselves being consumed in the process). It’s pretty impractical for a household to save all their consumption in this way, for all sorts of physical, personal, and logistical reasons.
Cross-posted at Asymptosis.
About paying off gov’t debt: I don’t see any gov’t debt in your scenario. 🙂
This is a ridiculous argument, similar to “there cannot be a gold standard because there is not enough gold in existence to back paper money.” To refute the answer would take too long, but I suggest you read Economics in One Easy Lesson or How An Economy Grows and Why It Crashes. It will explain economics to you in simple terms.
Interesting argument but you seem to be ignoring the financial system. In the normal course of things it is bank-created credit money and not government deficits that provide the liquidity required by expansions in activity. You’re imagining a world in which no private entity can emit liabilities that can be used as means of payment by third parties (“the household can’t use equity to buy gum”), so transactions have to settle in government issued money. But this is not how actual economies work. The household certainly can use newly created bank deposits to buy gum!
As a matter of fact, I agree with you that modern capitalist economies often will have trouble growing if there’s not sufficient government borrowing. But the reasons are complex and in part sociological – it’s not something that can be proved in a simple a priori way. I might add this is my general friendly criticism of MMT – the main arguments seem important and true, but they’re advanced in an overly deductive way.
“This is why in a growing economy where extra value is being created through people’s efforts, the government has to run deficits — creating money by crediting people’s/firms’ accounts with newly “printed” dollars.”
Newsflash. Printing money != running a deficit.
Most of us are agreed (other than the gold bugs and to hell with them) that an expanding economy requires an increased money supply. This is at the heart of Freidman’s analysis for example, or of Scott Sumner now.
But increasing the money supply does not require that government spend more than it collects in taxes.
You seem to have got very confused there between the monetary and fiscal worlds.
Tim Worstall said…
“Newsflash. Printing money != running a deficit.”
Yes, it is. If you mean mean money that people own rather than owe.
I assume “!=” means “not equal to”
Eric: “This is a ridiculous argument, similar to “there cannot be a gold standard because there is not enough gold in existence to back paper money.”
That’s a large part of the Long Depression in late 19th century. Halting the free coinage of silver helped kick it off, and discoveries of gold helped end it.
Tim Worstall: “But increasing the money supply does not require that government spend more than it collects in taxes.”
True enough. But how do you get the extra money into circulation?
So Steve Roth says that because the government is the monopoly supplier of meas of payment, it must run deficits forever. I say that’s not true, because the government isn’t really th emonopoly supplier of meas of payment. Tim Worstall says that not true, because the government can supply means of payment without running a deficit. That sounds like a second, independent criique, but it’s really not, and it’s interesting to think through why.
Say that on Planet Monetaria, all transactions must be settled with government-issued chits. There are no checing accounts,c redit cards, etc. Since there’s a limit to how fast chits can circulate, the government must increase the supply of them steadily, since the size and number of transqctions increases over time (i.e. the economy grows). How does it do this? If it’s via ope-market operations, it buys government debt (or lends against it as collateral), which of course requires that the stock of government debt in private hands is growing too. So, deficits forever. Alternatively, it could exchange chits for private goods, services and assets, also known as buying stuff. These urcahses will be in addition to those financed by taxes, so again deficits forever. Or, it can simply hand out the chits. This shows up as a transfer payment, so again deficits forever. So in this world, indeed, printig money = running a deficit.
Now, the difference between Planet Monetari and Planet Earth is that Earth has a banking system, so the private sector is able to produce its own means ofpayment. But this is an argument against monetarists like Sumner every bit as much as it is an argument against Steve Roth. If there really were a stock of money in the way that Monetarists claim to beleive, then it really would be necessary for governments to run deficits forever. Monetarists avoid this conclusion by an inconsistent use of the term money. When they are talking about money in relation to the economy, they mean broad money. But when they are talking about it in relation to policy, they mean base money. (You can sometimes see Nick Rowe, who is an exceptionally logical thinker, struggling with this inconsistency, e.g. by insisting there *must* be somethig special about currency so that inside money cannot substitute for it.) So you could think of Steve’s argument here as a kind of reductio of the Monetarist position.
Now that said, i do think there is good reason to believe that growth in modern eocnomies requires more r less contunuous governtment deficits. It just can’t be proven in this way.
Two better arguments are (1) the old Keynesian view that macroeconomic equilibrium requires desired saving to equal investment, but that — taking into account various frictions — the investment opportunities that will produce a return acceptable to wealth holders will be less than desired saving at full employment, and this gap is likely to grow over time. Mainstream theory says that the interest rate will automatically (or at least eventually) fall to the point at which S=I at full employmen. But there are good reasons (going well beyond the ZLB) to think that that interest rate cannot fall sufficiently. So government deficits must absorb the excess savings. (2) Government is not the monopoly supplier of means of payment, as both monetarist and (some) MMT arguments assume. But in advanced countries, it is the monopoly supplier of the safest and most liquid assets. A lack of safe assets can make wealth holders unwilling to commit themselves to inherently risky, illiquid investment projects. (If you want the 1% to buy factories, you need to let them also buy some T-Bonds.) While this is certainly true, the problem in practice is that both the relative advantage of the goverment in supplying […]
Isn’t the difference that private (bank) money is fragile? Much of it disappears during financial crises. Currency doesn’t.
Both arguments are valid … in a commodity money world. But we live in a credit money world.
That is a logical argument in principle, and it might have been true to some extent in the 19th century. But it really isn’t true today. Downturns are not assocaited with bank failures on anything like the scale required to produce an appreciable fall in the money supply.
Open the discount window.
The Federal Reserve Bank(s) is (are) a very strange institution. But it is what it is and does have the ability to print money or just create electronic bits that represent such and turn them over to banks. Bingo-money in circulation.
Now this fact makes Paulites want to claw out their eye sockets while whimpering ‘fiat money’ but just because they tend to be cranks doesn’t mean they don’t understand the mechanism. For example you could Google the term ‘quantitative easement’.
For bonus points you could look into the effects of QE on overall debt service given that the Fed rebates its profits to Treasury. Because at some level debt that requires no real servicing it’s not really debt at all. A point I have made in relation to the Social Security Trust Funds but which is just as applicable to Fed holdings of Treasuries. Which from what numbers are available are roughly double that of all Chinese entities combined.
But to reiterate the point you can increase money supply without changing federal spending, just allow/tell Helicopter Ben to wind up the rotors and open the cargo hatch.
Just try not to wake the Confidence Fairies or the Invisible Bond Vigilantes. Who by all evidence of our Magical Efficient Markets are snoozing away under the Troll Bridge. (despite the best efforts of the trolls at Club for Growth and AEI to rouse them)
Not really true, Bruce. As a matter of historical curiosity, do you know what the actual “helicopter drop” proposal was? It was for a *tax cut* financed by an expansion of the money supply.
To get back to Steve’s point on the impossibility of total debt pay down.
Greenspan wasn’t crazy in 1999-2000. Venal, self-serving and equally the willing serving of the top 1% yes. Crazy to warn about the elimination of the long bond? Not so much.
For better or worse the U.S. dollar is the major world reserve currency, not least because for now most commodities are priced in dollars. Which in practice means a certain irreducible supply of U.S. debt. Which after all are assets to the counter parties. Moreover assets that most can’t actually sell off, not at least to the last dollar equivalent.
For example if you go down the list of Major Foreign Holders of Treasuries and then look at the ‘Other’ category you realize that on paper every country in the world, perhaps without exception and irrespective of their foreign aid relation to the U.S. is technically a ‘creditor’ of the U.S. in that they are holding SOME paper backed by Full Faith and Credit. Even if that is just some dictators private stash of Benjamins. Or that drug dealer. Or arms smuggler. Every tattered $100 bill around the world (except the ones the North Koreans print, topic for another time) is a claim on U.S. productivity. If it could just be patriated.
But it can’t. Some irreducible quantity of Federal Reserve Notes and Treasury Notes and Bonds will NEVER be redeemed for goods and services. Because foreign criminals and Central Bankers (to the extent those are two different categories) and commodity buyers (ditto) need to keep some on account. Meaning that some fraction of even foreign debt holdings isn’t really debt at all, and certainly not at current levels of 10 Year’s.
But this is also true of domestic holdings. For example the Social Security Trustees are mandated to preserve a 100% of cost reserve. And by a 73 year old law in instruments fully guaranteed by the federal government. Meaning that right on $1 trillion of Trust Fund assets (out of $2.5 tn) are NEVER SUPPOSED TO BE REDEEMED. That is such failure would be legal violation of the Trustees’ fiduciary responsibility. And they are not alone. Banks have reserve requirements, pension plans have requirements to keep a certain percentage of assets in Treasuries, and ditto for certain balanced portfolio mutual funds.
So in toto a huge fraction of U.S. debt is legally tied up in reserve requirements and an additional fraction has to be in circulation to provide flight to safety/liquidity functions. What those fractions are is a good question, but Greenspan in 2000 had a point as does Steve today (though my point may not be his), a certain amount of long term U.S. debt is necessary just to grease certain legal and trade gears world-wide. And in nominal terms that irreducible minimum needs to increase every year. Now is the sweet spot $15 trillion? Probably not. Then again unlike countries that do not have currencies that function in reserve capacities, it can never go to zero.
The U.S. ain’t Greece. It ain’t even the State of California. Different rules apply to a country that can still pay its debts in its own paper.
Bruce Webb said…
“But to reiterate the point you can increase money supply without changing federal spending, just allow/tell Helicopter Ben to wind up the rotors and open the cargo hatch. “
What transfer mechanism would you prescribe that would make this work?
Money supply is a very vague term. Net federal (deficit) spending creates dollars and dollar-denominated financial assets in the non-government that are “owned” outright with no claims (liabilities) against them.
We know how federal net-spending transfers new money into the non-government. What other means is available to make fiscal transfers?
We’re you asleep the last few years as the Fed simply pumped up the big banks balance sheets by opening the discount window? Including banks that had not been eligible before, boom all of a sudden everyone was a bank holding company.
Banks were able to borrow trillions at effective negative interest rates, in what way DIDN’T that inject money OUTSIDE government purchases.
And beyond that the concept ‘quantity’ is inherent in Quantitative easement. I am not a finance guy, a fact that probably shows, but you don’t have to be a mastermind to know that the money supply is not exactly the purview of the Treasury and hasn’t been since 1913. Like I said Paulites are cranks but there really os something you can call ‘fiat money’ and the path to it runs through the Open Credit committee, which is a weird hybrid but definitely not a collaboration of Treasury and Congress.
What am I missing?
There was no liquidity crisis? Lending didn’t freeze up starting late in 2007? Whether that hard freeze actually limited various M measures is one thing, but claiming the downturn wasn’t caused or perhaps causing bank failures Ned’s to answer the question “Whence Lehman and WaMu and Countrywide?”
Banks failed. Credit froze. And some smart people labeled the result the Second Great Depression. Yet your tone suggests the whole thing was some hiccup.
If one supposes, and one should since it is already a fact, that the Fed can monetize any credit, not just government credits, then one can throw all this modern monetary stuff right out the window.
It is systematic credit expansion, period, which determines GDP growth, admittedly not in a linear fashion. It isn’t government borrowing per say but all sectors that counts. That isn’t to say that government credit demand does not have a special place but that is only because of the Fed traditon of buying only government paper. A tradition now broken.
Then too there is the complication that most ‘money’, with more and more all the time, dedicated to supporting asset prices. Leading to the current absurdity of trillions of dollars ‘invested’ by banks in negative return deposits and government securities. Monetary policy as the worlds biggest circle jerk.
It is systematic credit expansion, period, which determines GDP growth, admittedly not in a linear fashion.
This ought to be 1) a testable hypothesis, and 2) capable of being as a mathematical model. And since it’s univariate, the model should be rather simple.
Data on gdp growth and credit, broken down into quite a few categories, is available from FRED.
I’d love to see you mathematical presentation of the hypothesis, and graphic demonstration of its validity.
Bruce, come on. We can measure this stuff. There was no decline in the money supply. And “caused” and “causing” are two different things.
It just is not the case that the great recession was the result of a lack of means of payment. There really used to be recessions liek that — you can read about them. Creditors demanding payment in gold.
We had a collapse in aggregate demand, due to the end of the housing bubble. We also had a fall in asset values. We also had a financial crisis. Sorting out the causal connections betwen these three things is HARD. Talking as if the answers are obvious is not helpful. But one thing we do know is that the economy did not suffer from a lack of currnecy — which again, has been a real phenomenon in other times and places.
I’m a little confused. Are you supporting Steve R. here, or disagreeing with him?
Right, there are lots of reasons that the economy needs a substantial stock of federal debt. It’s jsut that the reason Steve gave — for use as means of payment — is not one of them.
Neither. Min asked a question about how you inject money into an economy absent action in the form of government spending. To the best of my ability and knowledge (which admittedly with $1.25 would buy you the smallest size Starbucks coffee-the one they don’t list on the menu but exists) I essayed an answer which rested on the fact (I think) that money supply is rather independent of the federal budgeting process but is instead in the hands of Central Bankers.
But my answer to Min didn’t have anything directly to do with MMT or Steve, which might have been clearer if JS-Kit comment threading was more robust.
On the other hand my past postings have advanced the claim that large parts of government debt, and particularly much of that debt represented by the Social Security Trust Funds, were never intended to be paid back. Ever. And even more oddly even the interest on that debt is discounted by some 50%’ or would be if we properly addressed SS finance. Which is to say Steve and I dispatched trains on parallel tracks leading to the same destination but maybe have different color schemes on the locomotives.
But the takeaway is something you conceded: a certain proportion of U.S. debt never needs to be redeemed, not on net. Which leaves the question of what is the right metric and measure of that debt, that is where is the sweet spot. But certainly the answer is not to take $15 tn and divide it by 330 million or so Americans and claim that every man, woman, and babe is on the hook for that amount of money. You might as well take M-1 or M-3(whatever the hell those are) divide it by the same denominator and make the claim that is our share of Full Faith and Credit. True but meaningless.
So while I am not ready to go Full Monty MMT that may just be that I am roughly 10% as smart as Jamie Galbraith and don’t know enough to agree.
Ok, thanks. I didn’t “concede,” tho – I enthusiastically agreed! 🙂
My problem is not with the destination, it’s with the train. More exactly, it’s with the attempt to demonstrate on purely formal, logical grounds claims that need a much more concrete institutional foundation. I wish MMT people would stop doing this – itmakes their (genuinely important) arguments weaker than they should be.
More like ‘Bottoms Up!’
First positing nonlinear causation pretty much rules out falsification a priori. You can always ring in a variable to ‘explain’ deviations. Kind of like modern Ptolemaic epicycles.
Second I am a little leery of discussions of productivity that discount labor input AND innovation to zero in favor of gnomes moving around counters. Because I retain a naive faith that the Industrial Revolution happened. To say nothing of the rent extraction of Colonialism more broadly. I don’t think we get anywhere attributing the last four hundred years of economic history to the actions of bankers working in the square mile that is formally The City of London. Or the similar square miles surrounding Wall Street or the Chicago Board of Trade. That seems to concede WAY too much to the MOTUs, who Lord knows are arrogant enough already.
JW I suspect I agree. But need to think it through.
But at some level it seems that Steve’s argument could reduce to the assertion that we had no economic exchange prior to establishment of the First Bank of Ur by Hammurabi the Great in 8000 BC or perhaps the Reserve Bank of Olduvai Gorge a few hundred thousand or million years earlier. Which is not to appeal to some preceding pure barter system, but an acknowledgment that exchange systems certainly preceded anything we would recognize as state banking systems. Or even state issued value tokens. For example back in the day I was quite the student of ancient Celtic law codes and even where Kings and Chieftains purported to have the authority to establish monetary penalties it wasnt like the slave girls or cattle that established the tariff had to come with state issued brands. That is bankers and still less state bankers were not the inventors of exchange and attendant liquidity.
Bruce Webb: “turn them over to banks. Bingo-money in circulation.”
But if the banks are not lending?
IIU J. S. Mill (thanks to DeLong) correctly, a drop in aggregate demand **means** an insufficient supply of money. OC, that does not mean any particular drop in the money supply, as it could manifest as an oversupply of other things. But it is measurable, right?
Bruce Webb said…
“What am I missing?”
The fundamental difference between money that is “given” to the non-government and credit money that is loaned (as in temporary, unable to be kept or hoarded) to the non-government. Would you rather have $1000.00 given to you or loaned to you?
Bruce Webb: “ Min asked a question about how you inject money into an economy”
Not exactly. I asked how you get it circulating. Tim Worstall said you could increase the money supply without gov’t spending. But simply increasing the money supply does not necessarily mean a corresponding increase in money in circulation. Quantitative Easing (TM), IIUC, increased the money supply like crazy, but does not seem to have had much effect upon money in circulation. Any recovery in bank lending (which would increase money in circulation) seems to be as slow, if not slower than it was in earlier times before QE.
If the gov’t had been lending money to regular people over the past three years, we would be much better off. 🙂 The same goes for giving money to regular people, too, but maybe not as much.
This is kind of what I’m getting at except my preference is reversed.
The only way to get out of this over-leveraged mess is deficit spending. The number of dollars that exist in the non-government necessary to satify existing outstanding loans is nowhere near adequate to the tune of $Trillions.
Americans have borrowed way beyond their ability to pay and the dollars needed to extinguish the liabilities have been hoarded through disproportionale accumulation of financial wealth. There is no way politically possible to claw back that wealth.
The trade deficit has further exacerbated the situation as a leakage of dollars currently in the $0.5 Trillion range annually. If the government doesn’t deficit spend. within say four years there would be no net cash left in the U.S (assumption: there is ~$2 Trillion in cash in the U.S. before people will have to start spending their savings) – we will then become a 100% credit-based economy, and in that case we would be like Greece. Obviously we have a choice and Greece doesn’t.
If the government doesn’t deficit spend the only option is mass default, which is defacto deficit spending.
If the government loans money to you at effective negative interest rates it is de facto a gift. Which is what the Fed was doing, bank ‘holding companies’, including some that had a few months before been pure investment banks, were borrowing money at the Window for 1% and lending it at 3%. Basically a license to print money.
As to banks not lending that money I would note there seems to be no lack of liquidity in various government bond markets, that is somebody seems to have enough access to loose change to keep 10 Year Treasuries and equivalent bonds from Germany, Japan and amazingly even France and Spain at close to the lower bound. Which given the price/yield inversion in bonds implies that the Big Boyz aren’t being frozen out, even as underwater homeowners and small business are having their nuts squeezed.
It is hard to suggest we have a universal liquidity crunch given current 15 and 30 year mortgage rates. There just seem to be some ‘haves’ and ‘have nots’ in relation to credit access. But the ‘haves’ seem to have pretty easy access to nearly free money.
“If the government loans money to you at effective negative interest rates it is de facto a gift.”
A gift is a loan that doesn’t have to be paid back.
If you borrow money you have to pay it back. This requires work and a living wage that allows one to have disposable income. But working for a wage doesn’t create new money, it just re-distributes existing money. Credit “creates” new money but it simultaneously “creates” offsetting liabilities. This “money” eventually disappears from the economy as the loans are paid back.
It’s mathematically impossible to have a stable growing credit-based economy over the long term (the Euro is making long-term seem like maybe 10 years). I know that probably sounds outrageous to most. Some entity has to monetize the growth in real wealth created and it can’t be done with a credit circuit..
The money we accumulate in our savings and pension plans for the future can’t come from credit. How would that work? I would think that would be un-controversial.
The Eurozone is the perfect example of a system based on credit.
Min, banks *are* lending. The perception that they aren’t is a result of the fact that we are coming off a period of excess credit.
We should expect, then, that lending will appear constricted.