« Keynesian Beauty Contest SRAS shocks | Main


Feed You can follow this conversation by subscribing to the comment feed for this post.

As someone who responded "difficult" to your initial poll, let me say that this was a helpful post.

Category 2 was always easy to understand.
Category 3 now is crystal clear
Category 4 is now clearer, but still a little foggy

If I'm reading you correctly, in Category 3, for every bilateral transaction there is a buyer and a seller, which in other words means there are equal amounts of savings and dissavings. Category 4 is similar. For every singular, unilateral decision to keep money in a bank account or a wallet (savings), there is an equal and opposite amount of displaced money income (dissavings). Because you didn't spend $1 on apples, the apple seller is now $1 poorer in income.

So since Categories 3 and 4 in the aggregate cancel out, all that's left is Category 2.

Did I get that right?

I wonder to what extent this is similar to the 'immaculate rebalancing' school of balance of payments analyses. There, too, one is tempted to use accounting identities as theories, so people forget that things like exchange rates and interest rates have to move in order to get people to make the decisions that make things balance.

I think the reason that economic professors talk about inventories of unsold goods is because that is the area where the term Investment is used in a way that is contrary to everyday parlance.

If I produce 100 consumer goods to sell but only succeed in selling 80 of them then this would not sound like Investment to the average non-accountant. In my opinion it is the transformation of Potential C into Actual I (based upon what is sold) that is implicit in the equations that is non intuitive and confusing to people.

This confusion does not exist in economies where all goods are produced to order , or there are only services and in those economies the equations would not be at all non-intuitive.

Kyle: thanks! Yes, I think you've got that right. Category 4 is always the tricky one, because money is not like other assets. We use money as medium of exchange, so it flows both into and out of our pockets whenever we sell or buy anything else. So there are always two ways to accumulate money: increase the flow in (which is a bilateral decision); decrease the flow out (which is a unilateral decision).

Stephen: I think there's a similarity. But there's also a difference, because the open economy is, in a sense, a bit like an individual household. We Canadians *can* all save by buying used cars, if we buy them from the US (and export new cars or whatever in exchange).

MF: I think you might be right. And when I used to teach Intro, I would talk about inventory accumulation when I explained why production had to be equal to expenditure. "Accountants cheat, to make the accounting identities come true! They say you bought those unsold apples from yourself!"

Following the assumption that I earn $1000 this month, I much prefer to consider how the person who enriched me got this $1000 and why did he give it to me. This approach causes me to question the accounting identity S=I when applied to macroeconomics.

The giver could be taking the money from long term savings, from HIS last pay check (short term savings), or he could be borrowing from himself (as when government borrows from the central bank). I, as a receiver of money, could have earned it by building a new product but in this modern world, only a small number of people actually build products that actually have an inventory presence. Most 'inventory' is the result of constructive accounting.

By beginning with income and then discerning the source, we recognize that when we divide output "Y into consumption goods C and investment goods I, so Y=C+I, we are looking at a small part of the economy, albeit a very important part. Having defined Y as representing goods (either consumption or investment), we cannot go on to add government spending (as in "Y=C+I+G") because doing so would increase Y to the extend of government spending.

Of course, "Y=C+I+G" is accepted macro accounting so we have an incongruity. In my own mind, I redefine the macro investment term 'I' into a balancing term having meaning only so far as it balances the equation. I would conclude that term I has nothing to do with inventories.

Your explanation is different from mine. But I have a bit of a problem with yours in that it seems to be a story about the future, not about the present. "Our individual attempts to save by accumulating money will collectively fail, but will cause our money income to fall until we stop trying." But the identity has to work even in an arbitrarily short time period, where there might not be time for our money income to fall.

Of course you could argue that my story is a story about the past. "You did invest or dissave, and now that I am receiving income from your act of investment of dissaving, I'm saving that income." But the time span involved is inherently infinitessimal.

I'm surprised to hear that the standard teaching story is about inventory. Isn't this kind of narrow? What about infrastructure, equipment, buildings, factories, tools, upgrades, improvements, renovations, durable goods etc. Even things like education, teaching, skills building, scientific discovery or data gathering can be "stored" as knowledge in people's heads or on paper. Or are all these things considered "inventory" in economics? I guess they all vaguely fall under inventory or production capacity. "Capital" might be to polysemous of a word to use in teaching. It might add to the confusion.

To me, the point of S=I is that to be considered net S or I, it has to add to the worth of the stock of intrinsically valuable physical things or stored useful information and not be just promises, financial instruments or numbers in an account.

Roger: sorry, but you lost me there.

Andy: good point. I think I worded that wrong, because, as you say, this accounting identity says nothing about future (or past) time periods. Is this better?: "Our individual attempts to save by accumulating money must collectively fail, but will have caused our money income to have fallen to a point where we stop trying."

Benoit: "What about infrastructure, equipment, buildings, factories, tools, upgrades, improvements, renovations, durable goods etc. Even things like education, teaching, skills building, scientific discovery or data gathering can be "stored" as knowledge in people's heads or on paper."

Those are all investments. But those are desired or planned investments. What's different about inventory accumulation is: we don't normally think of it as "investment, so it's counter-intuitive, and needs explaining; it can be unplanned/undesired, like when you produce goods planning to sell them, but fewer get sold than you had expected.

Here’s the key conceptual difficulty IMO, which (as you're aware) I spent years struggling with:

Saving is not a flow. (Though it is a flow as opposed to a stock measure — tallied over a period of time, not at an instant.) It’s a residual of two flows — income minus (consumption) expenditure.

So you can’t point to any transaction and say “that’s saving.” Buying Apple shares or a work of art is just an asset swap. It has no effect on your total assets or net worth — only the composition of your portfolio assets. Depositing a paycheck in your checking account isn't saving, until you don't spend it. Saving is, by its very nature, a non- flow. That's its sine qua non.

You can point to a transaction and say “that’s investment (spending)”: paying people to create goods which will not be consumed within the accounting period (or in practice and measurement, buying such newly created long-lived goods).

But if you conflate this investment spending with saving, you have a very troublesome conundrum: spending is saving? That's a difficult thing to intuit, for undergrads inevitably thinking in terms of household "saving"; (almost) all household spending is consumption spending, so saving is just...income minus expenditures.

This embodies a broader problem, the almost universal confution of wealth with real stuff (wealth is market-traded and -priced claims on stuff) that I think is most perfectly expressed in the phrase "financial capital" a confusion which is pervasive across economics — from Marx and Ricardo to Lucas and (very much so) Piketty — he explicitly makes wealth and capital synonymous...except when he doesn't.

Steve: In my version I point to the transaction in which you received the income and say "that is saving" (on your part). Somebody else is spending, but you are saving what they have just spent.

Andy: So then, condensed, "income...is saving"? That seems deeply problematic.

Steve: I think I get what you are saying. We can identify the flow of income, and identify the flow of consumption, but saving is just the residual difference between the two. Saving is not a thing, it's an "anything else" category, a "non-thing".

And ordinary people's concept of saving (=not spending) isn't the same as economists', true.

But I think it's useful to break down that "non-thing" saving into: transactions where we buy new investment goods; transactions where we buy other assets (real and financial); the residual where we don't buy anything at all but just accumulate money. Because, even though it makes very little difference to me as an individual whether I buy a newly-produced house or a one-year old house, or whether I buy a bond or leave the money I earned in my pocket, it makes a macroeconomic difference.

"Financial capital/wealth" makes sense for an individual, but doesn't add up right. I hear it from Marxists, but didn't think Lucas uses that concept?

Nick: >Saving is not a thing, it's an "anything else" category, a "non-thing".

Sure. But when we resort to the word “thing,” we know we’re in trouble. I think it’s clearer to call it a non-flow. (While it’s obviously a flow measure.) Really, that’s what makes it what it is. Saving is not-spending. (Out of, relative to, income, over a period. I think spending out of wealth/assets is a more pertinent and transparent concept, but that’s a further discussion.)

Takeaway: saving, being a non-flow, does not, cannot, increase any imagined stock of “savings.” (Whatever stock measure that undefined but pervasive plural is referring to.)

When you “save” — spend less than you receive — you have more money/wealth. We don’t. When you eat less corn, we have more corn. When you spend less money, we have no more money.

Real goods and balance-sheet assets/claims are created and destroyed through utterly different mechanisms. Income - Consumption is trying to subtract apples from dollars. (Intuitive?)

Production creates stuff; it doesn’t create balance-sheet assets. Those are created by the government/financial system through three mechanisms: 1. bank net new lending, 2. gov def spending, and 3. (overwhelmingly dominant) holding gains. (That system is “looking at” the stock of real goods when it creates those assets. If there’s new stuff out there, it creates new claims, balance-sheet assets, reflecting ownership of that stuff.)

>[1.] transactions where we buy new investment goods; [2.] transactions where we buy other assets (real and financial); [3. ] the residual where we don't buy anything at all but just accumulate money.

My definition of “spending” might help here: “transferring assets from the lefthand side of one balance sheet to the lefthand side of another, in exchange for newly-produced goods and services.” (Physical cash is just a convenient technology for making that balance-sheet transfer.)

By this definition, #2 is not spending. It’s swapping existing claims/assets. That’s why GDP ignores these transactions.

Gonna stop here, but a last key point I think: national-accounts Saving (a residual of national accounts Income) doesn’t even come close to explaining balance sheet changes — ∆ total assets and net worth. Intuitively, it should, right? As in normal business accounting statements?

Maybe useful: images in this tweet and thread, responding to Josh Mason’s preceding tweet image from the BEA:


I don't see any reason to stop using the inventory story--its worked for me for a couple of decades. The important point is that *actual* investment is defined to include any change in inventories. That is what makes Actual I=S an identity. The inventory story comes in handy because it provides a (primitive) explanation of why AD=Y or Planned I=S describe equilibrium: if you produce too much (Y>AD), inventory accumulation sends a clear signal to reduce production.

Its true the inventory story would collapse in a pure service economy. But then you'd always be in goods market equilibrium with Y=AD and no distinction between actual and planned investment.

S=I because we mostly use double entry accounting. We use double entry accounting because any triple entry accounting system can be decomposed into a set of double entry accounts.

When S=I we are in compliance with the savings = loan ratio managed by central banker. Any deviation from S=I is considered a random mis-pricing and gets a separate liability account, and that account tracks the disequilibrium of the currency.

The real question is why do all natural processes appear as a yin and yang (look it up). That is a question for all things natural, not just accounting.

thomas: fair point. But I might be planning to invest in a machine, but be unable to buy one, because I can't find a willing seller.

Matt: for every apple sold, there's an apple bought. I think that's the origin of this particular identity.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name and email address are required. Email address will not be displayed with the comment.)

Search this site

  • Google

Blog powered by Typepad