Revisiting a previous post, “Saving†≠“Saving Resourcesâ€*, wherein I question Scott Sumner’s notion that people who spend and consume more (save less) take resources “out of society.”
Try this:
John works for Debbie, and Debbie works for John.
They each start out with $100 in dollar bills, $200 total.
They pay each other in dollar bills: $100 a year, each direction.
Between them, through their labor, each year they produce $200 in real resources — things that humans can consume to derive human utility (or to produce more consumables in the future).
But: This year Debbie decides to save money, so she doesn’t hire John for as many hours, and only pays him $80. She leaves $20 sitting in her drawer; she doesn’t circulate it this year.
At the end of the year Debbie has $120, and John has $80.
Debbie has produced $100 worth of real resources, and John has produced $80 worth. $180 total, instead of $200 the year before.
Did Debbie “take those $20 in real resources ‘out of society’”? (Or was it John — lazy, feckless soul that he is — who didn’t do that $20 in resource-creation?)
We can certainly say that Debbie’s decision to leave the $20 sitting in her drawer instead of circulating (spending) it caused “society” (read: John) to produce less resources than it would have if she had circulated (spent) it.
Is Debbie a “taker”?
Cross-posted at Angry Bear.
Comments
25 responses to “Do Savers “Take Resources out of Society”?”
This isn’t how Sumner approaches such a problem.
I can’t match his voice, but I can sing a similar tune from the same album. Saving = Investment. (This is his language. It’s a common textbook language.) Using this terminology, we should note immediately that your economy does not have more saving. Debbie is micro-saving, but John is micro-dissaving to the exact same extent. It cancels out, as usual. Net micro-saving is zero. The only way to have true saving is to build investment goods.
Using this language, I could rephrase your main point by saying that an attempt to save does not necessarily succeed. An attempt to save succeeds when it results in more investment goods, and an attempt to save fails when other people decide to dissave at the exact same rate that others are saving. (Sumner has written posts that touch a bit on this idea, by the way. I can dig one up if you want.)
He advocates consuming less because it frees up production to focus on investment. If we consume less, but work the same amount, then we must necessarily be devoting more of our effort to preparing for the future. There is a transition here. People have to move out of consumption jobs, and move into investment jobs, but he doesn’t think that’s a big deal… with proper central bank policy.
He’s a market monetarist. If Debbie would feel more comfortable having another piece of paper, Sumner would recommend printing one up for her. But that by itself doesn’t change the future productivity of this economy. It doesn’t even necessarily change her underlying negotiation with John. Given identical productivity levels, he might simply increase his prices until they are once again trading the same amount of goods for the same amount of goods.
For them to have more stuff in the future, given the same total production today, they need to devote more of today’s production to investment and not consumption. That is the underlying idea. When people are panicking about not having enough paper, well, there’s an easy solution to that.
@Hellestal “consuming less because it frees up production to focus on investment.”
That’s the crux of the misconception. It’s confusing a backward-looking accounting identity with an assertion about what causes what.
Looking back, with a given Y (since we’re looking back, it is fixed), any spending that’s not C is obviously I. Voila. Less C means more I!
But that actually tells us exactly nothing about what caused that spending mix. Less C doesn’t cause I to increase, magically. QTC in fact.
Looking forward, we can say that if people spend more (on C, or some mix of C and I), producers will produce more, and they will invest more in production capacity to satisfy that increased demand for C. (If sales are good this year, they’ll have higher expectations for next year.)
C increases. I increases. Y increases.
Sure, there may be some optimum C/I proportion for a given economy at a given time, a proportion that will result in the greatest accumulation of real goods over time. Obviously we don’t want 100% I or C.
But it’s not even vaguely obvious that policies encouraging people to not spend out of their wealth will deliver that optimum proportion, much less maximum GDP.
Mary can swap her $20 bill with John in exchange for his labor, or she can swap it in exchange for his $20 in gold coins. Which creates real resources?
@Hellestal
An observation regarding your recent, rather offensive post:
http://www.hellestal.com/?p=1226
Where you said:
“C + I + G + (X – M) = C + S + T
I’ve never seen this before. It’s bizarre. If you isolate S algebraically, you get:
S = I + (G – T) + (X – M)â€
This defines saving as investment (okay), plus the government budget deficit (!) plus net exports. This is saying that the government spending more money than it takes in is saving, rather than dissaving. That is the exact opposite of the conventional definition of public saving.
They call this the MMT definition. Fine, fine. Definitions aren’t right or wrong. They’re either useful or not useful.
This? I can’t see how it’s useful. This definition of saving combines two heterogeneous categories as if they were equivalent. This definition of saving is real capital equipment plus pieces of paper created by the government (again, net exports should cancel if we look worldwide). There’s no purpose to this definition, from my perspective. I don’t want saving defined as the combination of paper and machines. That’s just my opinion, but it’s an opinion backed with some years of study.â€
And where you also said:
“Naturally, the professional economist Nick Rowe was completely correct, and this guy is wrong. (He is right, though, that Nick Rowe is very smart.)â€
Well, this is what Nick Rowe actually said here:
http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/08/is.html
“Substitute equation 1 into 2 to eliminate Y and you get:
3. S = C + I + G + X – M – T – C
You can eliminate C in 3, and rearrange terms to get:
4. I – S + G – T + X – M = 0â€
Either of which in fact are the very same as:
S = I + (G – T) + (X – M)
Which is the well-known equation of sector financial balances, not only used by MMT, but by others such as Wynne Godley, an acknowledged early foreseer of the financial crisis, and the highly respected Jan Hatzias, chief economist of Goldman Sachs, and many others as well.
@JKH
Oh well, I guess I need to move a term over to make it more glaringly obvious as a sector construction:
(S – I) = (G – T) + (X – M)
@Hellestal
“”That’s the crux of the misconception. It’s confusing a backward-looking accounting identity with an assertion about what causes what.”
It isn’t. He has an (implicit) model. You have a different model. I have yet another. I wouldn’t go any further putting words in his mouth, but I can outline a horribly simplified version of what one might look like.
You have Y, which represents the natural limit of output. There are limitations to resources available, which means limitations to what we can produce. Y is a hard limit here. Some of that production is for consumption goods, and some for investment goods. Model: People decide to save more. They spend less on consumption, which means less income for the producers of consumption goods. (Income = Expenditure.) People get fired, which drops GDP below its potential. Eventually the owners of investment industries see the available resources and start hiring. People change jobs, moving into investment industries, and output returns to its potential at Y.
This isn’t something I would expect to be convincing. I don’t believe it myself. The point is that you can see the gears of causality turning, above and beyond the identity which is true by definition.
JKH, you’re right. It was offensive and I didn’t read Rowe carefully. My only defensive was that I was intentionally writing on my own blog in order to vent, and I didn’t think about comment trackback. I’m glad you spotted that and corrected it, and I’ll try to be more careful in the future.
[…] In the comments at this Asymptosis post, JKH points out several embarrassing errors in this post, and takes issue with the tone here, […]
@Hellestal
Your error is that you use saving freely. In one terminology (rather than “model”), S is private saving and in another it is national saving.
Each equation is alright by itself but the reason you see inconsistency is that you mix different Ss.
You look at S = I + CAB and then
S = I + G – T + CAB
and say first is right and hence the second cannot be.
But the S in the second is a different S than the first.
If you want to use them simultaneously, use subscripts.
Economists use π (pi) for inflation and also for the irrational number which is approximately equal to 3.14159.
So according to your analysis, economists think inflation is approximately equal to 3.14159?
“Debbie has produced $100 worth of real resources, and John has produced $80 worth. $180 total, instead of $200 the year before.”
What happens the next year assuming John does not dissave?
@Hellestal
Ramanan said: “You look at S = I + CAB and then
S = I + G – T + CAB
and say first is right and hence the second cannot be.
But the S in the second is a different S than the first.”
I believe the I’s are different too.
The first S and I are the private and gov’t sectors combined.
The second S and I are just the private sector.
You deaggregated the first one to get the second one.
@Hellestal
“Debbie is micro-saving, but John is micro-dissaving to the exact same extent. It cancels out, as usual. Net micro-saving is zero.”
I’m going to try to construct a scenario where that is not true.
John works for Debbie (Deb), and Deb works for John.
Deb starts out with $200 in dollar bills, $200 total.
They pay each other in dollar bills: $200 a year, each direction. Velocity = 2.
Between them, through their labor, each year they produce $200 in real resources ($100 each)– things that humans can consume to derive human utility (or to produce more consumables in the future). They both charge $2.
MV = PY ; $200 times 2 = $2 times 200
Neither save or dissave, balanced budgets.
But: This year Deb decides to save money, so she doesn’t hire John for as many hours, and only pays him $180 instead of $200. She leaves $20 sitting in her drawer; she doesn’t circulate it this year.
John still balances his budget. He spends $180.
Deb has produced $90 worth of real resources at $2, and John has produced $90 worth at $2. $360 total, instead of $400 the year before.
MV = PY ; ($180 times 2) plus ($20 times 0) = $2 times 180
Deb saved $20. John still has a balanced budget (no dissaving, $0).
Saving in the MOE/MOA caused a “recession”.
@JKH
I find it easier to say/think:
current account deficit = gov’t deficit plus private deficit
The real question is is there an entity missing in it.
@Fed Up
“I’m going to try to construct a scenario where that is not true.”
It’s easy to formulate this “micro-saving” concept as (yet another) identity, and identities don’t have scenarios where they aren’t true. They are always true, by definition.
“Deb saved $20. John still has a balanced budget (no dissaving, $0).”
You said Deb started with $200. She spent $180 of that, then received $180 back again.
200 – 180 + 180 = 200. She ends where she started, same as John. They both saved nothing, under the definition I was using.
My point wasn’t to argue for the perfect clarity of this, just to say that this is very similar to how some people think. (I can cite examples, if you like.) If we want to understand what they’re saying, we can’t impose our preferred meanings on their words, and then tell them they’re wrong. We have to dig in a little deeper to see what their actual point is. It’s not easy, but it has to be done, because otherwise, we’re stuck forever on word games.
@Hellestal
In S = I, I (as in “me”) believe S equals Y – C, Deb: $200 – $180. S = $20, I = $0. John: $180 – $180. S = $0, I = $0. John has a balanced budget and started with $0 and ended with $0. John did not micro dissave, while Deb did micro save. Overall, S = $20 and I = $0.
@Fed Up
Normally Y is defined as income.
S = Y – C. The flow of saving equals the flow of income minus the flow of consumption. They’re all flows. But you said her starting stock of cash is $200. So you have S = Stock of cash – Flow of Consumption. You’re mixing the stock with the flow. Her income is clearly $180, and you don’t account for that because you use Y as her stock of cash instead.
This is nonstandard. That’s fine if you’re comfortable with it, but are you sure that’s what you want to say?
@Hellestal
That’s probably true, but isn’t it also true above?
“They each start out with $100 in dollar bills, $200 total.”
Also, the model has to start somewhere.
@Hellestal
Thinking about it, I’m not sure.
What about a capital gain or inheritance?
@Fed Up
Well, I can only tell you how I would personally classify things, and I’m weird, so listening to me might not be your best bet.
I’d look at an inheritance as a transfer of an already-existing pool of savings (subtly different from a flow of saving) from one person to another. No change in anything, just a transfer from the dead possessor to the living one, subtracting from one balance sheet and adding to the other.
Capital gains would be trickier. It clearly does increase the pool of nominal savings, as individuals calculate these things. If I were trying to defend a cherished identity, I’d simply define it out of existence in sort of the same way that GDP doesn’t count capital gains. But if it were important to my analysis, I’d account for it in some fashion. That’s a circular answer, but sometimes that’s all we’re left with. I’m not talented enough to write an econo-manifesto of everything, so I don’t try. I just try to do the cleanest analysis I can that tackles the question in front of me.
@Hellestal
Here is a scenario to think about. Assume Deb and John had a barter economy between the two of them. At a future date, they decide to switch to a medium of exchange economy. How would they get the MOE circulating?
@Fed Up
Honestly, that would depend on what dynamic of their exchange I’d want to think about. If I just wanted a unit of account to keep track of their transactions, then any random initial allocation would be okay. It would just be a bookkeeping measure.
Or for another situation, we could explore a division of labor, such as where John specializes in investment goods, like better fish nets, while Deb specializes in consumption goods, like catching fish. Deb gives her fish to John while she receives tokens representing IOUs to keep track of his debt. When he completes a better net (or a better fish spear, or whatever), they can evaluate whether the device is good enough to consider the debt paid. But in this case, the tokens don’t circulate back and forth indefinitely. They represent a specific debt from John to Deb only, because that’s the dynamic that’s most interesting in this scenario.
@Hellestal
I want to try to stick to a scenario where there is no debt.
@Fed Up
Without debt, money is likely to serve as only an accounting placeholder.
Deb might build up a lot of cash, but the “prices” between the two could possibly be renegotiated so that her apparent advantage doesn’t hold, i.e. the money isn’t worth as much when it’s held by one person. This is why I would only want to explore introducing an MOE to illuminate a specific dynamic in their relationship.
Well I believe an economy should have zero debt so I’ll let you think about that. I’m not sure what will happen.
[…] “Saving†≠“Saving Resources†[…]
[…] “Saving†≠“Saving Resources†[…]
[…] “Saving†≠“Saving Resources†[…]