(Shaking head) Noah Smith is a professor of finance and he doesn't know the answer to this — although granted it is not very clearly stated as "government deficits equal private surpluses," as Jan Hatzius does, whom Noah quotes.
It is clearer to say that the government fiscal deficit equals the consolidated nongovernment fiscal surplus in aggregate, since the government balance plus the consolidated domestic private sector balance plus the external balance must sum to zero as an accounting identity. Transposing, the government balance will be the inverse of the consolidated nongovernment balance in aggregate.
If the external balance is zero, then the government balance and the domestic private sector balance must sum to zero, therefore, must be in inverse relationship. In this case, if government net spends by running a deficit, then the domestic private sector saves the same amount in aggregate wrt net financial assets. And vice versa.
If the both the external sector and the government sector are in balance, then the domestic private sector must also as an identity. There can be net savers but not net savers in aggregate, since accounts within the sector must balance, i.e., net to zero.
Hint to Noah: Read Godley and Lavoie, Monetary Economics for an explication of stock-flow consistency in sectoral balance accounting. Jan Haztius learned this from Wynne Godley, and it is the basis of MMT SFC macro modeling. This is really simple when you do the accounting properly.
While accounting identities don't say anything about the causality, they do reveal what is necessary and what is impossible in terms of stock flow consistency. Mixing up stocks and flows is a novice error.
Noahpinion
Thinking out loud: Do government deficits equal private surpluses?
Noah Smith | Assistant Professor of Finance, Stony Brook University
25 comments:
Krugman's reaction sometime:
"Oh it is Econ 101. It's all there in the textbooks"
... then goes on to muddle few lines later.
Pretty amazing. This should be macro 101. But nooo, you can get a teaching position and be clueless about it. Bill Mitchell and R Wray start with these identities i think.
AND he has Physics degree also I believe and he STILL has to question this math ... how does THAT happen????
Matt,
I have a physics PhD and I am embarrassed for him ;)
A physics PhD can be a disadvantage in econ and social science in general if one thinks of social systems in terms of physical models — which is the mistake of neoclassical (equilibrium) econ.
It's more appropriate to think if social sciences in terms of biological models involving complexity, autocatalysis, emergence, and reflexivity, which is why we are fortunate to have Roger here.
But Noah is smart enough to figure that out. Anyone who has a degree in physics should be handle the modeling and the advances math for dealing with complexity. And Noah has posted on deficiencies in the conventional approach based on his undergrad work in physics.
But accounting is another thing. Very smart people get tripped up with SFC. He needs to integrate accounting, finance and economics, and I am sure he is quite capable of doing it if he can get beyond the programming of an econ PhD.
I would say that double-entry in accounting is sort of analogous to the conservation principle in physics. Everything is accounted for step by step, and nothing gets lost in transitions.
Yes, but he prefers his intuitions and gets lost in distinctions real-financial, too bad b/c physics ppl and economists are taught to trust math and use it to cut through the fog of confusion, it is very effective (if one stays conscious of assumptions, which Noah knows well how to do, plus accounting is a man-made system, no assumptions are needed to be made, just follow the rules).
Part of the problem here is that there a number of advocates of the sectoral balances approach who seem confused about the difference between savings and net lending. Noah Smith can hardly be blamed if there are people continually using one term when they mean the other. If those who think correct accounting matters want to be taken seriously, they really need to start being more rigorous in their own use of terminology.
Hello,
A question - (probably stupid)
Isn't the point of Noah's post that the government deficit needn't equal private surplus if we simply choose a different monetary system?
If we decided that lumps of rock were money then wouldn't it be possible for the supply of money to increase without any corresponding debt - a commodity money system doesn't require a debt?
Likewise, if we simply decided not to account for the government deficit and instead recorded it as only a positive supply of money, what difference would it make?
Doesn't government produced fiat money have more in common with commodity based money than private credit money? If so why bother recording a government deficit?
Noah's point was weak. Even if some people mistake "savings" for "surplus", Hatzius did not do so (the reporter in the article did).
I am unsure whether there is a formal definition of "surplus", but I believe it is the "net lending" in the flow of funds accounts. By construction, the government deficit has to equal the surplus of all the other entities on those accounts. This will hold regardless of the monetary system, to respond to the point of Mark above.
Savings is another economic concept. I don't want to get mired in the swamp of the debate about the meaning of "savings", but the summary of the debate (as far as I can tell) is that people often use "savings" to follow standard English usage of the term when writing, which roughly corresponds to "net lending", and not the formal definition used in macroeconomics. This is unfortunate, but that's the problem with taking a word with well developed usage and then attempting to force it into a restricted formal definition.
Brian... "by construction" the government deficit equals the private sector surplus - but only because we choose to treat the government as an entity which can have a deficit.
if the nature of government is very different from that of an individual, if the nature of the money it produces is different, is it helpful to account for government in the same way as we account for individuals?
My money might represent my expectation that another individual in society is obliged to do something for me (they have a debt to me), and it is therefore necessary for those with debts to record them, so that they are aware of the claims which are due on them, but in what sense is this true of government?
If the government itself will never have to repay me, why is it necessary for us to record a deficit or debt on the governments account?
I hadn't meant to suggest that Hatzius was one of those who confuse the terms - I did look at the article and he seems to be fairly precise in what he is saying.
It certainly creates problems when economists use terms which have a precise meaning within economics, but also have a common usage which is vaguer and maybe different - the term "money" seems an excellent example. But perhaps this is all the more reason to be careful how these terms are used.
The difference between saving and net lending is crucial to an understanding of sectoral balances. If someone wants to use the term with a different meaning, that requires some explanation. Even worse is when the term "saving" is used with two different meanings within the same discussion.
I'm also not sure if surplus and deficit have a clear definition in terms of flow of funds. Saving and net lending are defined terms in the SNA, surplus and deficit are not.
If the government's "deficit" comes about by spending money it borrowed, somebody else in the "private sector" is going to have to pay it back. If it comes from squirting some new funny money upon a lucky recipient, that lucky guy makes off for free with the goods working people had planned to buy. Nobody has saved anything other than a purchaser of a bond who should have instead been investing in a business but for the attractive crowding out caused by the bond.
Mark, as long as double-entry accounting is used, both sides of the ledge must balance. TINA. Within one sector there cannot be a net surplus in toto. There can be net savers but the net is offset by someone else's liabilities.
Across sectors, net surplus in one sector can only occur if there is a net deficit in the total of the other sectors.
Of course, double entry is just a rule, but there's a reason that it is used in accounting. Nothing can "disappear."
The record can always be traced, which makes it possible to discover embezzlement, for example.
as long as double-entry accounting is used, both sides of the ledge must balance. TINA. Within one sector there cannot be a net surplus in toto. There can be net savers but the net is offset by someone else's liabilities.
When you squirt some new funny money to a government supporter (a new "net financial asset"), nothing has changed except that the government hack has now eaten the steak a working person planned to buy. And you don't even think about accounting for that.
@ Bob
There are costs associated with government. The question is whether the net benefit exceeds the net cost.
Sometimes it does, and sometime it doesn't. When it doesn't. then the government either falls or must be maintained by force of those in power.
There are tradeoffs in choice of monetary system, which can also be evaluated in terms of cost-benefit. Which is what we are doing here. You are on one side and most of us are on the other because our POVs are differentiated by different values, priorities, and assumptions about theory of money.
Nick,
So a *professor* of econ has the right to ve confused about basics because some *laymen* who understand this stuff better sometimes are unclear about terminology??? Sorry, it doesn't work this way. He should be teaching, not learning.
Mark,
Rocks are real assets. Financial assets are always someone's liability. So when a new financial asset is created ex nihilo the creator increases its luability by exactly as much as the receiver sees his assets increase. So for the private sector as a whole to get more fin. assets than liabilities it must by accounting/math necessity increase holdings of liabilities of someone outside the private sector: the govt.
Peter,
Well, Noah kicks off by questioning two statements "government deficits equal private sector surpluses" and "government deficits equal private sector savings".
That's not an unreasonable position. The latter one is pretty clearly wrong. Private sector savings are equal to public sector borrowing plus private sector fixed capital formation, so you have to have a pretty weird definition of government deficit to make that correct. The former is more tricky because surpluses and deficits can mean various things. If you take it to mean net lending or borrowing then it will be true, but it's not obvious you should take it to mean that.
However, on the whole I'd agree with you. There are plenty of good sources for the sectoral balances approach, so there's no excuse for Noah not knowing what it is when it's explained properly.
I'm not sure who are the laymen you had in mind, but I was thinking of people who should know better. Bill Mitchell (who on the whole is very good) seems to frequently confuse saving and net lending, even using the term "saving" to mean two different things in the course of the same post. I assume he understands the difference, so I guess this just comes down to carelessness, but it's not very helpful.
On reflection, there's a bigger point here. A professor of macroeconomics should really have a basic familiarity with the structure of national accounts, at least to understand when and how sectoral balances will sum to zero. As such Noah Smith should at least be able guess what people are trying to say when they make these sorts of statements, even if they don't articulate it very well.
Peter P,
At the very least, it isn't very helpful to use the same terminology for the government as it is individuals. In Noah's post you have people claiming that the fish he is trying to use as money are aquired by "natures deficit" and that the stock of fish represent "nature's debt" - this is clearly a completely unhelpful use of terminology, but is it any more helpful to refer to a "government debt" or "deficit"?
If we must keep a record for administrative purposes, perhaps we should call it "money produced" and "stock of money produced".
The financial assets created by government are not liabilities in the same sense that assets created by individuals are liabilities.
At best they are similar (to perhaps purely theoretical) commodity money, such as gold bars, since people simply decide to hold them for the sake of it, and worst they are held because the have a "avoid getting you head bashed in by government" function, in being used to pay tax.
Either way, it isn't the governments liability.
(This is a reply to Mark's response above.)
You could have "my" (private) money versus "government" money. What would happen is that you end up with two (or more) currency systems coexisting in the same country. In which case, you would have two sets of financial accounts which will each balance. I.e. transactions will appear in the accounts associated with the currency used. For example, people in Canada could start transacting in gold, and so you would need to track gold-denominated transaction accounts, plus the CAD dollar transaction accounts.
In which case, you could probably get some wierd effects in how they add up (entities may be transacting in both currency systems, and thus end up needing to convert values to create aggregate accounts). You would end up with a lot of very complicated currency valuation calculations...
However, you don't see that very often in practice in the developed economies. You do see things like that when the US dollar is used along side a local currency in less developed countries. For the developed countries, the tax system greatly discourages this sort of behavior (you have to pay sales taxes (VAT) in the local currency). So you need to get your hands on the government currency if you want to run a business.
@ Mark
In a convertible system, government liabilities are actual ("real") liabilities at the rate of conversion. So governments do function similarly to households and firms in that they have to obtain something that they do not create.
In a floating non-convertible system, that is not the case since govts that are currency sovereigns not borrowing in a medium that htey do not create themselves.
However, since currency is exchangeable for goods, the availability of real resources in an actual constraint.
When this is understood, the points you make become obvious. Until then, most people are under the illusion of government being analogous to households and firms, which currency sovereigns are not.
I am not sure that calling currency something else that a government liability or bonds something other than public debt would be a great help if people don't understand how the system works. They will just hink it is a rhetorical poly. And if they do understand how the system works, changing the terminology is not all that significant. If ir would happen, it would likely happen of its own accord as people started using more precise terms.
Anyway, it is highly doubtful that the accounting terminology is going to be changed, so rather than lobby for changing the terms, it is probably more practical to educate people in how the system actually works.
We've been around the block several times on this since it keeps coming up.
Let's suppose there is no government
No. Let's not suppose that. That would be like supposing there is no banks. This allows you to assume away everything that does not fit your model. If you give away the starting assumptions, you'll reach the conclusion that the person assuming it away wants to reach.
If you examine the suppositions carefully, there’s usually one of them which provides the function of a great big rug under which all the points you might want to make have been pre-swept.
Crooked Timber - http://crookedtimber.org/2004/08/19/the-correct-way-to-argue-with-milton-friedman/
Having accused Bill Mitchell of inconsistency over the use of the word "saving", I was pleased to note that in his most recent post (http://bilbo.economicoutlook.net/blog/?p=24953), he goes out of his way to explain the different ways he is using the word (paragraph one of answer to question 2). I'm still not sure I like his terminology, but credit to him for making it clear what he's saying.
Mark,
"Either way, it isn't the governments liability."
That is incorrect. You correctly wrote above that government money is a liability in a different sense but a liability of the government it is.
"Liability" is something that when presented to you forces you to do something which you otherwise wouldn't do. "Will pay back $100k before 2030" is a liability. "Will smile when presented this IOU" is a liability - but very easy to fulfill, you just smile to the holder and done. If you were a celebrity and you wrote on an autograph "I will give you another autograph when you show me this one" is a liability, but again very easy to pay off, just write another autograph and done. Government liabilities are of the same type: the government only promises to exchange them to other similar liabilities and accept them in payment of taxes and fines: "when you give us this piece of paper, we will cancel your taxes of $100". But the government is obliged to undertake some actions it wouldn't otherwise undertake so they are liabilities. Of course when people think of them as being as burdensome as dollar liabilities are for households and firms this is asinine, but so pervasive that it is hard to kill.
Post a Comment