Saving isn’t “setting money aside,” it’s BUILDING CAPITAL GOODS
People, I beg you to just stop. All the textbooks say S=I, so I thought people accepted this. I guess they don’t. Let’s start with some basic misconceptions:
1. We are interested in national income and output, not individual accounts. Yes, an individual can save without any change in national investment. You might buy a stock or bond. But in that case someone sold you the stock or bond, so they dis-saved exactly what you saved. THE ACT OF SETTING ASIDE MONEY HAS NOTHING TO DO WITH SAVING. SAVING IS BUILDING CAPITAL GOODS.
2. Let’s consider three types of goods; food, cars and houses. The government considers food and cars to be consumer goods, and houses to be capital goods. In fact all goods could be viewed as capital goods, albeit goods that depreciate at vastly different rates. Since the GDP data is annual, we might want to consider goods to be “capital” if they lasted more than a year. So cars and houses ought to be capital, but I believe the government views cars and home appliances as consumer goods, perhaps because they are too lazy to do all the calculations for depreciation.
3. Consider a $20,000 car that lasts 9 years, and depreciates in a straight line (yes, unrealistic.) Assume it sells for $2000 scrap in year ten. You just graduate from college, and blow $20,000 out of your $50,000 income (cash) for a new car. Technically you’ve saved $18,000 that year (assuming the rest of your income is spent on perishables like food and concerts and haircuts and rent. Yet you haven’t set aside any money–you’ve “spent” it all. Your car depreciates $2000 in that first year, but $18,000 in value carries over into future years, where it can be used for transport or sold. That should be called saving.
4. In each future year your car provides $2000 in transport services, but there is no cash outlay, you are drawing on your previous saving/investment. That frees up your future income for other uses.
5. Interestingly the same is true for houses. So why is blowing $40,000 on a new BMW considered wasteful consumption for a new grad, whereas buying a house is viewed as an investment? The BMW purchase basically commits you to consume $4000 a year for the next 9 years. So it is an investment, but also an implied commitment to consume at a pretty high rate. In contrast, the real value of houses tends to hold up quite well. My 90 year old house is worth much more in real terms than when it was first built. Yes, there is some opportunity cost, (stocks are probably a better investment in the long run), but otherwise it’s easy to see a house as an investment.
6. People become very confused about saving during recessions, because they wrongly associate saving with “putting aside money” and recessions with “people not spending.” But if I put aside money by buying a stock from you, then national saving and investment don’t rise at all. If I blow $100,000 cash on a new house, then I’ve saved close to $100,000 that year (minus only the tiny depreciation.) SAVING IS NOT SETTING ASIDE MONEY, IT’S SPENDING MONEY ON CAPITAL GOODS. PERIOD, END OF STORY.
7. Well not quite the end. If people hoard cash or bank reserves, i.e. try to hold larger amounts of cash in real terms, that’s deflationary. That seems kind of like saving, at least like the false definition of saving that is “putting money aside,” but it’s actually a mostly unrelated phenomenon. In fairness to Keynesians, in my previous post I explain why periods where people desire to save more at a given interest rate, often are periods of recession, where M*V falls. So they have good instincts. But accumulating lots of non-monetary financial assets is not, in and of itself, (aggregate) saving, nor is it money hoarding. It’s just debt. Suppose I give my brother a $1 quadrillion IOU, and at the same moment he gives one to me, with the same maturity date. There sure is a lot more debt in the economy, but has anything changed? Physical goods (and services) and the medium of account—focus like a laser on those two things, and macro becomes incredibly simple.
PS. David, you have no idea just how far I’m willing to go. 🙂
Tags:
12. January 2012 at 06:51
I always enjoy hearing talking heads blame the Fed for trashing grandma’s CD rates. She relies on that interest to live! Don’t blame the Fed, grandma. Blame the low real interest rate. Sorry our recession happened to coincide with your retirement.
12. January 2012 at 07:15
Scott, what do you call choosing longer production processes providing superior output over shorter processes providing inferior output, using the same production goods?
12. January 2012 at 07:36
If someone changes their choices, and begins choosing shorter production processes producing inferior output over longer production processes providing superior output, are they disinvesting?
If he switches to producing card board boxes over lumber and houses, what has taken place?
What if we go look at the data and find that that is what happened over the last 8 years?
What has happened?
What consequences will this have on other things?
12. January 2012 at 07:38
Reposted comment from Nick’s blog:
“Thank you. Helpful.
Of course, the textbooks are written, so maybe you need intermediate terms. Perhaps just leave it as “national saving” and “individual saving”. Or perhaps just call it non-consumption. Then you can say:
production = consumption + non-consumption
And who can disagree with that? At least it clarifies what Keynesians mean. Then at the national level we can have an argument about what “non-consumption” means.
If, in Steinbeck’s grapes of wrath, we dump truckloads of produce into the ocean to avoid depressing the price, is that consumption or investment?”
So I thought I had this square, until…
David Glasner writes:
“Savings is not identically equal to investment, the equality of savings and investment is an equilibrium condition.”
He asserts this partly based on the lag between production and consumption (inventory???)
Uh oh….
He also writes: “Keynesian fiscal stimulus works by transferring idle money balances in exchange for bonds at liquidity trap interest rate and using the proceeds to finance expenditure that goes into the pockets of people with finite (rather than infinite) money demand.”
Now THAT is interesting – he’s basically saying Keynesian fiscal stimulus (in the absence of monetary action) works by redistributing wealth to those who are willing to spend it immediately. (I would add that the multiplier is the impact on Y of the THREAT of redistribution.)
I would go one step further as well – the fundamental mechanic underlying monetarist stimulus is also the THREAT of redistribution – from those holding money to those holding productive assets.
12. January 2012 at 07:40
Oh, Scott, you should be aware that there is a perfect flipside to the argument that
“When the Fed targets inflation, fiscal policy has no effect”
That is the following: “When fiscal policy perfectly targets consumption, inflation targeting has no effect.”
Unstoppable Force, please meet Immovable Object
12. January 2012 at 07:44
yes, so when people stop building capital goods (houses and buildings, largely) that means that saving in the economy is broken. that’s why a housing price crash – which caused people to stop doing this – is such a big deal. people stopped saving in real terms (ie building capital goods) and switch to saving in purely nominal terms (ie cash and cash equivalents). how would you expect nominal yields – and prices in general – to respond to this – plummet of course.
12. January 2012 at 07:48
Thanks. Sorry if this has been already explained, but I was hoping to clarify one issue.
1) Since the great recession, I have been holding more cash deposits relative to consumption and “investments” like stocks. Most of the cash is in a “high” (approx 1% annual interest) interest savings account.
2) Is that hoarding or savings? I’m getting more interest than the bank gets on holding reserves, so I assume the bank has an incentive to invest it, but that the effort in finding a worthwhile investment slows down velocity. Do we have to know what the bank did with the money to know whether it was saved or hoarded?
3) I’m a novice, so I’m sure this is oversimplified, but I’ve always assumed that the paradox of thrift was a story about velocity — that because it takes more time to invest money than to spend it, and then for the sources of investment to spend that, what a household characterizes as “savings” generally slows down velocity, reducing the functional money supply. Is that how a Keynesian sees the paradox or am I missing something?
12. January 2012 at 07:50
@StatsGuy: but we have a perfectly coherent notion of what it means to say “holding fiscal policy constant”, whereas we don’t for monetary policy. We have to assume a central bank reaction function of some kind.
Not to mention, a fiscal reaction function that effectively targets consumption is… implausible. A central bank reaction function that effectively targets inflation is something we can actually see in the real world.
12. January 2012 at 07:53
I sympathize with your frustration, but I don’t think “building capital goods” is a good definition of saving. It’s true, from the point of view of whole economy, any saving has to take the form of building capital goods, but we also have to apply the concept to individuals. From an individual point of view, “setting aside money” is saving, although the relevant type of “setting aside” is a completely passive process that happens without any action on the part of the aside-setter. (Technically, it doesn’t necessarily involve money, but in practice individual saving almost always does.) It would be more precise to say that “receiving income” is saving. When someone receives income, they have no choice but to save it in the nanosecond after they receive it. Whether it represents saving in aggregate depends on whether the person who gave them the income was dissaving it or not. I had a blog post about this a few years ago.
12. January 2012 at 07:54
Scott writes,
“accumulating lots of non-monetary financial assets is not, in and of itself, (aggregate) saving”
So all those people in the 00’s who thought they were saving and investing in houses and housing mortgages were not actually saving?
Who is to decide this?
12. January 2012 at 07:56
Great stuff, Scott! I totally agree, from 1-6. All goods are capital goods which depreciate as they they deliver consumption (services). Even apples, until you eat them. (To my cousin, as a child, even a piece of gum was a capital good. Even *after* she had started chewing on it, she and her sister would share gum chewing consumption enjoyment. Yuck!). This is why (to use Nick’s favorite example) hoarding of antique furniture doesn’t cause recessions. An antique couch is just an elegant-seating-comfort factory.
But…
“In contrast, the real value of houses tends to hold up quite well.”
Not really. It’s the land that appreciates.
And 7 is muddled. There is nothing very different about bank money/liabilities and government and central bank money/liabilities. All nominal debt/money is a promise of a fixed quantity of future consumption that, depending on the promised rate of return, may be undeliverable. Holding capital assets, on the other hand, only promises whatever future output those assets happen to produce. When nominal instruments promise higher returns than can be delivered by the economy, they will fail. But under those circumstances it is still rational for *any given individual* to buy nominal assets, because the collective effect of everyone ceasing consumption and trying to increase holdings of nominal assets will cause consumption and the value of capital assets to collapse. There is no savings “rat hole.” Just a fixed (possibly rising in debt/deflation) quantity of consumption promises and collapsing output. Whoever hoards the most nominal instruments wins (or, actually, loses the least).
*That* is the real Keynesian paradox of thrift. No animal spirits, no savings rat hole. Just a bad equilibrium resulting from a coordination failure of rational, self serving consumers. The solution, of course, is for markets and central banks, never to mis-estimate the future return of capital assets, and thereby always get interest rates right, especially on long term debt instruments. Or to stop the profusion of nominal liabilities. *Or* to end the promise of money as a given quantity of consumption (inflation targeting). NGDP level targeting may be one possible step in the right direction.
12. January 2012 at 07:56
Thanks for highlighting that investing in stocks/bonds doesn’t count (directly) as savings. It’s easy to think that buying bonds is should count as investing. But in a GDP accounting sense, only the thing the company/state does with your money is an investment, the rest are just shuffling paper.
But your post still invites the question: What *is* setting money aside/ money hoarding counted as in a GDP sense? Your 6 and 7 kind of talk around it. What if I make $100000 income selling services, take that money as cash and put it under my bed for a few years? That doesn’t get counted as S (or I) for GDP accounting purposes, because nothing more is produced?
But it still has non-accounting effects on GDP by effectively reducing the monetary base?
12. January 2012 at 08:02
@ Alex
Sure we can develop a consistent notion of fiscal targeting – create a fiscal transfer system (like social security) that increase transfers from rich to poor as expected consumption (or, if you prefer, nominal income) decreases.
In other words, if we hold money SUPPLY to a fixed trajectory, and there is SOME reservoir of people who would spend money faster than the rate at which the taxed population decreases expenditures to offset tax expectations (which is likely due to differential utility curves, wealth stocks, and credit constraints), then fiscal policy can hit any nominal GDP target too… Heck, you can even use a NGDP futures targeting regime for transfer payments.
To be clear, I am NOT saying this is a good idea. I am saying it is a perfect mirror in the theoretical constructs that have been outlined. The choice of one method vs. another needs to be empirical.
12. January 2012 at 08:04
J Mann
Good question. You are “investing” in bank debt, but it has the same effect as hoarding cash under the bed.
Problem is the bank knows you are doing this for just the short term and not “investing” in the bank debt for your transactional business. With the latter type of balance they can easily behaviorally model the deposits and reinvest te majority of it private loans to give you a nice interest rate.
With your “hot” money they can only in invest in highly liquid, highly unproductive assets like excess reserves in the central bank or T-bills.
12. January 2012 at 08:07
Scott, You never cease to amaze me!
12. January 2012 at 08:10
During the boom, production goods (e.g. houses) were securitized and these assets were highly liquid and they played the role of money substitutes in many transactions, including credit transactions. The demand for other money instruments was lowered, including demand for money and credit instruments manipulated by the Fed.
When production goods are securitized and become highly liquid and take over many of the roles of money in credit and other transactions, lowering the demand for money, is that _de-hoarding_?
When securitized production goods (e.g. bundled housing mortgages) lose their status as highly liquid assets (because their value is negative or dropping off a cliff and radically uncertain, etc.) they lose many of the money-substitute roles they were playing in credit and other transactions, increasing the demand for actual money, can the consequences of that process, where people increase their stocks of money, be described as money hoarding?
12. January 2012 at 08:12
In the long run, income equals consumption. Saving is just deferred consumption.
The problem is that, holding the money supply constant, if everyone tries to defer consumption by increasing their money balances, then their efforts will be frustrated by falling incomes.
It’s a bit like that old chestnut about people standing up at concerts: if a few people do it, they will get a better view, but if everyone does it, nobody gets a better view. (Not exactly the same, but close enough).
Also, fiscal stimuli work by increasing the supply of a close substitute for money. If government bonds were not perceived as high risk assets, then fiscal stimulus would fail to change money demand at all and spending would remain constant.
12. January 2012 at 08:14
Oops! In the above comment, I meant to say:
‘If government bonds were perceived as high risk assets, then fiscal stimulus would fail to change money demand at all and spending would remain constant.’
12. January 2012 at 08:21
Since saving is just deferred consumption, buying a car is saving unless you plan to run it in a destruction derby, because most of the consumption of the vehicle will be deferred to future years.
12. January 2012 at 08:31
If the price of houses is doubling ever 4 years, and home buyers, home builders, and associated industries pull away 99% of the investment capital of the nation, and borrow vast amounts more from oversees, and the housing stock doubles in 4 years … and then it become clear significance percentages of those with mortgages did not have jobs with incomes big enough to cover those mortgages and debts, and the housing market then crashes, what do we call what has happened.
According to SCOTT, every dollar set aside to build those homes and bid away resources for all other industries was SAVINGS. PERIOD, END OF STORY.
SCOTT has a value production machine, like a donut production machine that extrudes donuts. What goes in — stuff valued in the PAST in dollar terms — is what what comes out the other side and off into the future forever (less depreciation), irrespective of all other changing and adapting relations of production and valuation anywhere else in the stream of production and consumption (allowing for only total changes in total aggregate values caused by total aggregate changes in the aggregate stock of money).
“Value” is simply extruded out of Scotts value extrusion machine. And the valuational principle — the economic way of thinking principle this machine works on is exactly RICARDO’s and KEYNES’S — the very principle refuted and rejected by Carl Menger, among others.
And, yes, this is just like rejecting Copernicus for Ptolemy. And it’s like returning to THE AGE OF ALCHEMY when lead could be turned to gold by the magic / naturology of the magicians philosopher’s stone — spending can be turned into economic value by the magic / econology of the macroeconomist’s math construct.
Scott,
“If I blow $100,000 cash on a new house, then I’ve saved close to $100,000 that year (minus only the tiny depreciation.) SAVING IS NOT SETTING ASIDE MONEY, IT’S SPENDING MONEY ON CAPITAL GOODS. PERIOD, END OF STORY.”
12. January 2012 at 08:32
Lee: What you are saying is a really good, shorter version of what I say above. And the concert analogy is great, except that in the paradox of thrift you end up with a much better view than other people if you stand up first, which encourages defection and permits the bad equilibrium.
12. January 2012 at 08:45
“We are interested in national income and output, not individual accounts.” What is the difference? I could, for example, declare myself to be a nation, like the last of the Mohicans. I think you mean global income and output, at least until we start trading with Mars.
Basically, in any open economy, there is bilateral saving with other economies and investment in capital.
12. January 2012 at 08:54
Ace. Dead solid perfect. Another blog by Scott Sumner.
12. January 2012 at 08:57
I don’t really understand why you don’t consider stocks, bonds and other financial assets capital goods? I mean they make up the ‘capital’ of banks.
12. January 2012 at 08:59
“THE ACT OF SETTING ASIDE MONEY HAS NOTHING TO DO WITH SAVING. SAVING IS BUILDING CAPITAL GOODS.”
If I have three gummy bears, and I eat one and set two aside, I will say “I saved two for later.”
You can define words how you want, of course, but your insistence on the above collapses what could be two useful concepts into one (so that one wonders why you are using two words at all), and runs totally contrary to everyday usage. And what’s more, you shout it at us (ALL CAPS!) as though it were some elementary truth of the universe rather than an arbitrary definition you are assigning.
12. January 2012 at 09:03
I don’t agree at all.
Saving is that part of income not spent on consumer goods.
Investment is spending on capital goods.
Total saving is the sum of individual saving. It can be positive or negative for individuals or in total.
Total investment is the sum of individual investment. While gross investment is greater than or equal to zero, net investmnet can be positive or negative. Net investment is gross investment less depreciation.
In equilibirum, saving equals investment. Of course, by this, I mean desired and actual saving and investment. At most interest rates, there is no equilibrium, and desired saving and desired investment are not equal in total.
That the amount actuallly saved and invested are always equal is no more interesting than the amount actually bought and sold are always equal.
For example, if the price of a good is above equilibrium, the result is a surplus. The amount households want to buy is less than the amount firms want to sell. Of course, the amount that is actually bought will match the amount actually sold, which would presumably be the amount households want to buy.
Only at equlibirum is the amount households want to buy, the amount they actually buy, the amount firms want to sell, and the amount they actually sell, all equal.
To say that supply always equals demand is pretty much _wrong_. To say that saving equals investment is the same thing. It is only true in the sense that actual purchases must always equal actual sales.
12. January 2012 at 09:12
Suppose you live in a barter economy. There are dollar bills, but they’re just pieces of paper and cannot be used as a medium of exchange. You don’t have any desire or use for them as bits of paper, but accept them in exchange anyway and set them aside for later. But for what later? You have no use for them or later — they’re junk. You have no use for dollar bills at any time; the notion of saving them for later is preposterous.
For this economy as a whole, saving by putting money aside is like the above: it literally doesn’t make sense.
One caveat: money created by something like a fractional reserve bank is different. By creating more money (as credit), they can overcome this problem.
12. January 2012 at 09:25
@StatsGuy:
Sure we can develop a consistent notion of fiscal targeting – create a fiscal transfer system (like social security) that increase transfers from rich to poor as expected consumption (or, if you prefer, nominal income) decreases.
That’s not what I was talking about. I mean there is an objectively reasonable meaning for the “ceteris paribus” as applies to the fiscal authority (it would roughly be the ‘current policy’ the OMB uses when it makes budget forecasts). There isn’t one for the central bank, which means you have to pick a reaction function and so you might as well pick one that actually describes reality.
This isn’t to say you aren’t allowed to include a fiscal reaction function in your model too – but if you do, you should try to use something realistic and a consumption-targeting fiscal authority is not a plausible description of Congress.
In other words, if we hold money SUPPLY to a fixed trajectory, and there is SOME reservoir of people who would spend money faster than the rate at which the taxed population decreases expenditures to offset tax expectations (which is likely due to differential utility curves, wealth stocks, and credit constraints), then fiscal policy can hit any nominal GDP target too… Heck, you can even use a NGDP futures targeting regime for transfer payments.
The issue right there is that you’ve decided that a constant money supply is an appropriate description of “holding monetary policy constant”, and yet that is very different from most people’s descriptions. Some people think it means constant interest rates, some people think it means constant inflation target, and a few people think it means a constant target for the level of nominal GDP (I won’t name names). Some people don’t actually have a coherent notion at all of what they mean by it. So if we have to sort through this mess anyway and pick one, we ought to spend the extra effort to pick one that actually describes the Fed.
To be clear, I am NOT saying this is a good idea. I am saying it is a perfect mirror in the theoretical constructs that have been outlined. The choice of one method vs. another needs to be empirical.
I agree with you, but for the reasons above I don’t think that there is an actual symmetry in the choice.
12. January 2012 at 09:28
@StatsGuy:
I just reread this part of your post:
In other words, if we hold money SUPPLY to a fixed trajectory, and there is SOME reservoir of people who would spend money faster than the rate at which the taxed population decreases expenditures to offset tax expectations (which is likely due to differential utility curves, wealth stocks, and credit constraints), then fiscal policy can hit any nominal GDP target too… Heck, you can even use a NGDP futures targeting regime for transfer payments.
and realized that I had misread it the first time. I think I agree with you (though this mechanism seems really difficult to implement), but I’m just not sure how relevant it is to Scott’s argument about the multiplier.
12. January 2012 at 09:46
Even Wikipedia doesn’t agree with Scott!
http://en.wikipedia.org/wiki/Saving
They even have a picture of a piggy bank.
Scott, you must now admit that you are wrong. 🙂
Seriously, though, the fact that Scott and Woolsey completely disagree on the definition of savings reinforces my concern with the state of academic macroeconomics. I wish there were some conferences focused on getting basic definitional consistency.
12. January 2012 at 09:54
Haven’t we known what a disaster the S and I language taken from Keynes is since at least Hayek’s review of Keynes _Treatise_ in the early 30s?
12. January 2012 at 09:56
P.S.
While I would be inclined to agree with Sumner that for the economy as a whole to shift consumption from the present to the future it is necessary to build capital goods, Rowe’s discussion of overlapping generations suggests that it isn’t quite so. Of course, Sumner does discuss consumer durables as investment. And producing nonperishable consumer goods and putting them in warehouses is a similar investment.
But I don’t think that this has much to do with the notion that saving equals investment always in the sense that actual purchases equals actual sales.
12. January 2012 at 09:57
@ Alex
“this mechanism seems really difficult to implement”
Yes! Probably impossible in the fiscal policy institutions in most democracies – perhaps less so in an autocracy.
But it does exist to a _limited_ and imperfect degree. I would, for example, cite the “automatic stabilizers” in europe, and the continuing extensions of extended unemployment benefits in the US. The cost of these is debt, however, with the concomitant risk that debt markets cease to function well due to default risk premia – although that doesn’t seem to happen in Japan, the UK, or the US in spite of very high debt/gdp.
All of this leaves aside the structural economic impacts of these policies, the international repercussions, relative ease of capital flight/labor flight, etc.
The primary point I wanted to make was that Scott’s argument is purely empirical – he’s saying fiscal policy doesn’t work because the Fed’s inflation target will neutralize it. Well, that assumes that the Fed is more powerful/committed/politically independent than the fiscal authority. I’d agree, but that’s an empirical and political/institutional argument. Economics, not so much.
12. January 2012 at 09:58
If one wants to change names, he is please to. Still, using name “savings” for “investments” leaves a blank term ready for binding. May I use the latter term for naming my dog?
12. January 2012 at 10:34
I=S isn’t an equilibrium condition, it’s an accounting identity. Desired saving = desired investment is an equilibrium condition. It is amazing how much people talk past each other by either mixing up or failing to specify which equality they’re talking about.
I think the problem with Scott’s definition of S is that it is indistinguishable from I. Investment is defined as the building/purchase of capital goods. If you define Saving to be the same thing, then it’s harder to see how desired saving could not equal desired investment.
12. January 2012 at 10:37
I feel that much of this “debate” is arguing over definitions.
“Savings” is that portion of income not spent on the consumption of perishable final goods. It can be used for capital investment, for durable consumption good investment, for hoarding cash, for buying antiques, for…
SCOTT READ THIS PART
Wait, here’s an easy example. Consider an OLG model, everyone lives for two periods, and we have a government that sells real bonds. Consumers have preferences over consumption today and consumption tomorrow.
Critically: this is an endowment economy, hence there is no production and no investment. Yet there are savings, defined as (endowment – consumption) and the equilibrium condition in the loan market is that aggregate desired savings equals the bond issue.
So I have a simple model in which there are savings, but not investment. How does this fit into your post?
12. January 2012 at 10:42
B, Exactly, but don’t forget the Fed caused the recession.
Greg, You asked:
“Scott, what do you call choosing longer production processes providing superior output over shorter processes providing inferior output, using the same production goods?”
I call it lengthening the production process.
If you mean “net investment” then disinvesting occurs when the capital stock gets smaller.
Statsguy, Saving equals investment for the same reason that buyer expenditure equals seller revenue.
Grapes dumped into the ocean out of inventory is depreciation, which makes net investment smaller than gross investment. (And net saving smaller than gross saving.)
You quoted Glasner:
“He also writes: “Keynesian fiscal stimulus works by transferring idle money balances in exchange for bonds at liquidity trap interest rate and using the proceeds to finance expenditure that goes into the pockets of people with finite (rather than infinite) money demand.””
He’s saying fiscal stimulus makes velocity rise, if it works. That’s true.
Regarding the battle between fiscal and monetary policy; the fiscal authorities can move V by a few percentage points. the Fed can move M by a billion percent. Guess who controls M*V?
q, I don’t think people were saving more in aggregate. But that covered up the fact that people privately saved more and the Federal government saved way less, so national saving fell, if I’m not mistaken.
J Mann, Complicated question. If the bank is lending money out for home construction, then your saving is financiag gross investment. But banks also hold lots of government bonds, so you might just be (indirectly) lending money to the Feds. In that case your saving is offset by their dissaving.
more to come . . .
12. January 2012 at 10:44
Scott,
Thought you might appreciate this Ryan Avent Tweet
“Scott Sumner would be angry at all of you. He’d say, “They were right about the economy looking fine, contingent on them not screwing up.””
Hoping to see your post on the newly published 06 Fed minutes soon!
12. January 2012 at 10:59
This post is exactly why those of us who aren’t as well versed in monetary theory are afraid of it.
In my family, we ‘save’ money by spending less of it on consumable goods, to hold against future spending or as security against risks to future income.
We do not even think of buying a house as savings, but rather as either an investment or as a large expense for which substantial savings acquired over a period of years were required.
The difference in our view between savings and investments is that savings are not supposed to be at risk, whereas investments we know are. We get upset, therefore, at the idea that the Fed would be empowered or encouraged not to protect the risklessness of our cash savings, but rather to increase it.
12. January 2012 at 11:08
Integral: “Wait, here’s an easy example. Consider an OLG model, everyone lives for two periods, and we have a government that sells real bonds. Consumers have preferences over consumption today and consumption tomorrow.”
I’m going to take that one. A cohort might be saving when young and dissaving when old. In any given time period aggregate (actual) saving is always zero in your model.
12. January 2012 at 11:13
Nick: if there is no government, yes. Throw in government, and there is aggregate (private) saving despite there being no production or investment going on.
Of course even with government, aggregate (public + private) saving still equals zero.
I’m perhaps missing something critical and elementary.
12. January 2012 at 11:13
Andy, I think 100% of the time the person that gave them the money is dissaving in a few nanoseconds, as nanoseconds are very short.
I don’t agree on “setting money aside,” unless you think plunking $40,000 cash down on a new BMW is “setting money aside.” If you do then I agree with you 100%. But my sense is the news media wants Americans to go out and “spend money” on new cars and houses, not “save” so I was using the terms as other people use them.
My point was that the everyday life meaning of “save” isn’t the concept economists talk about and model and think is important.
Greg, Oh they were saving and investing all right, but made poor investments that depreciated rapidly.
K, Well at least we agree on 1 to 6, I thought I was going crazy. But I strongly disagree on 7. What Keynesians don’t get is the importance of the medium of account. As long as the supply and demand for base money doesn’t change, the price level doesn’t change. You could triple the demand for stocks or T-bonds, and it wouldn’t matter unless the demand or supply of MB changed. In contrast, a change in the supply or demand for base money changes the price level, because goods are priced in terms of dollars, not T-bonds. The nominal price of a T-bond can change with demand, the nominal price of a dollar can’t change, all other goods prices must change. That why Keynesianism can’t tell us why the Japanese price level is roughly 100 times higher than the US price level.
I have a recent post that explains the paradox of thift in logical (i.e. monetarist) terms.
Cy, I remember people debating that when I was in grad school. I think there are two clear cases and one ambiguous one. Gold money is clearly capital, more total gold is more total capital. Interest bearing ERs and DDs are clearly debt. One person’s saving is offset by another person (or government’s) dissaving. But what about cash? Some view it as government debt, even w/o interest. So it’s not really saving. You hold more cash you save more, and the government saves less. Others see it as a sort of paper gold, just having a system of cash makes transactions more efficient. So then it’s capital. Of course if you double cash and prices double, then real capital hasn’t increased.
David; You said;
“You never cease to amaze me!”
In a good way or a bad way?
Greg, Money substitutes only matter to the extent they affect the demand for money (which admittedly may be quite important.)
Lee Kelly, I agree, but rather than increase close substitutes for money, I’d have the Fed increase the supply of money directly, and reduce the demand (via lower IOR and a higher NGDP growth target.)
Greg, You have me all wrong, see my earlier comments about housing depreciation. I may be stupid, but I’m not that stupid.
RebelEconomist, Good point. In earlier posts this week I referred to a closed economy Keynesian model. In a global economy, world S=I, but national amounts may differ–that’s correct. I meant closed economy here.
Thanks Ben.
Brito, It’s not me, unfortunately all economists define terms in different ways from everyday speech. We mean physical capital. If you counted stocks you’d be double counting the same capital goods. Stock in a company owning office buildings, and the value of the buildings themselves. Debt is an asset to one person and a liability to another, so no net wealth there.
more to come . . .
12. January 2012 at 11:15
Scott, I’m not going to go into details regarding the Keynesian model (you have plenty of readers with more interest and knowledge of the area), but i’ll post based on your response 2 blog posts bellow, where you said “S=I is a tautology. If Krugman being right depends on it being wrong, then he is in deep trouble”. This is an error of philosophy of science 101 level and since many more readers seem confused I’ll focus on it.
Now, that tautologies such as S=I are an analytic truth which is not subject to testing but true by definition, is something you could have argued in the early 20th century and get away with it, but it’s something nobody believes in science since the 50s. Before the 50s we were thinking in logical positivism terms, i.e. we had truths like “every bachelor is single” that were true because they were analytic (i.e. tautologies) and truths like “all swans are white”, which required empirical verification. That was before quantum mechanics, gödel’s theorem, etc. When we came across such weird stuff, two revolutions took place in philosophy of science. First, the Popperian revolution, i.e. the notion that we do not VERIFY a model (such as Newtonian physics) we simply fail to refute it. The second revolution came with Quine’s “two dogmas of empiricism”, one of which is that there are tautological truths such as “all bachelors are single”, where Quine showed that this simply is not true. You can never claim that S=I is a tautology hence true.
Now the above is accepted by all philosophers of science today. I don’t know even one professional philosopher in the English speaking world that would dispute this.
The above is also accepted by all natural science too, ESPECIALLY physics.
However, with economists things are different. They accepted the “Popperian” revolution out of necessity (because in economics it’s really hard to verify things, failing to refute sounded more convenient) but some (not all) chose to ignore the other one (even though I think Quine’s article is the most cited philosophy article in the last 100 years). You hear people saying “bite me, it’s a tautology”, i.e. telling you that IF my model is wrong, a number of things have gone wrong except that I=S was false; this is exactly the kind of error Quine seeked to prevent.
Now, I agree that it’s more useful to speak about the economy in M*V terms, but such a model by itself is not something that explains the world as we see it, because for the last 3 years increasing M has had no effect on NGDP. So the easiest way to save your model is by saying something like “the M that sits in fed reserves is not the M you should be looking at”, but this would be beside the point. The point is that if by assuming that S=I might not hold at the ZLB, i.e. that households want to spend money on capital goods (such as a car or a house) or consumption goods, but there are fewer people out there willing to lend them the money compared to last year, so the household has to buy less of something else to make up for the money they would normally be borrowing. This addition to the theory, first explains the buildup in fed reserves and second explains why M2 can go through the roof but NGDP won’t. It also proposes a solution to get out of the liquidity trap (have the government issue more bonds to mop up excess reserves and then spend them) which is not the optimal solution (I favor negative interest rates or helicopter drops) but it is something that is TESTABLE and REFUTABLE.
“S=I is a tautology so bite me”, is invalid, it’s not even wrong. So is most of the stuff that comes out of Chicago today (I remember Fama making a similar point to yours about Say’s law). But you can never test an element of the theory, you can only test the entire theory (another point of Quine), there is simpy no way to verify that S=I independently, all you can do is put this assumption in the pot and see if the food tastes better with it or without it.
Bottom line: It really doesn’t matter if you think S=I because there is no way to independently verify this (claiming is a tautology is NOT a verification), all that matters is whether this assumption makes your model’s predictions better or worse. Milton Friedman would have agreed with this.
12. January 2012 at 11:44
“Now, I agree that it’s more useful to speak about the economy in M*V terms, but such a model by itself is not something that explains the world as we see it, because for the last 3 years increasing M has had no effect on NGDP.”
I find it hard to believe you’re any kind of philosopher of science, because this is extremely unscientific. Where is the counterfactual analysis? How do you know that, were it not for increases in M, NGDP wouldn’t have dropped significantly in the counterfactual? That statement is almost impossible to verify, it is essentially unfalsifiable, save for some very clever statistics that might be possible.
“So the easiest way to save your model is by saying something like “the M that sits in fed reserves is not the M you should be looking at”, but this would be beside the point.”
Or, you could simply say that while M increased, V dropped massively, it’s really not that hard.
12. January 2012 at 11:51
Scott, I think as always it’s best to think of it in terms of Ramsey style economics, which Keynes admired.
Capital is what enters the production function to create output (mostly machinery, commodities etc..). Consumption is money used to increase individual utility (which can include setting money aside). I think this is the most consistent explanation out there.
Now, the reason I mentioned financial capital, is because financial capital is something used by banks to create their own output (lending), and it is not a 1 for 1 relationship because of leverage, so if you consider banking goods as part of ‘output’, then financial capital enters the production function of the economy. Agree?
12. January 2012 at 12:01
Gene, Sorry if the caps were insulting, but I get frustrated with commenters acting like I’m an idiot, when I’m simply using the term as it’s defined by economists. This blog is not written for the generally public, it’s written for people who know economics. I do things like the EMH and ratex, which are far higher level than S=I, which is intro textbook material. So is it unreasonable for me to assume we all know that economists regard S as equal to I?
Bill, I think I agree with you, so precisely which of my comments do you disagree with?
Ken, I actually agree with Bill, he just doesn’t know it. So please restore your blind faith in us economists.
And Wikipedia says saving is income not spent, money set aside for deferred consumption. Well which is it!!!
Take my BMW example, does Wikipedia consider that income not spent, but rather deferred consumption? (Like Andy Harless does) If so, then I also agree with Wikipedia. In other words, define “spent”. Is buying Microsoft stock the same as spending money? How about buying an office building?
Greg, I thought these definitions came from Fisher, not Keynes.
Andrew, You said;
“I=S isn’t an equilibrium condition, it’s an accounting identity. Desired saving = desired investment is an equilibrium condition. It is amazing how much people talk past each other by either mixing up or failing to specify which equality they’re talking about.”
I agree, but I don’t find the concepts of desired saving and investment useful, as quantities, but rather as functions of the interest rate or income.
Integral, I don’t disagree that if you spend $4000 on an antique clock you’ve saved $4000, but the seller dissaved the same amount, so national saving didn’t rise.
I don’t understand your endowment economy. What is the income? If there is no income then saving is negative C.
Marcelo, I already have a huge backlog; save me some time please and tell me the juicy parts.
Josh, Yes, those are the common sense views, but in economics we define lots of things differently from in everyday life. Most people think inflation hurts consumers, for instance, but it doesn’t.
Orionorbit, I’m still trying to figure how a bachelor can be married. Do you know of any?
You said;
“Now, I agree that it’s more useful to speak about the economy in M*V terms, but such a model by itself is not something that explains the world as we see it, because for the last 3 years increasing M has had no effect on NGDP.”
If you speak of the economy in M*V terms, then you aren’t speaking of the economy in M terms (like the old monetarist used to.) Hence your comment makes no sense. In any case, speaking of the economy in M*V terms is not really a model. The market monetarist model assumes that if M*V falls then RGDP is likely to fall. That model has held up pretty well over the past few years. As you probably know, I don’t favor targeting M, so I’m not concerned by the lack of correlation between M and NGDP.
I think you are confused about S=I. It’s a definition. Economists need definitions if they are going to have useful conversations, otherwise we’ll all be talking past each other, as I’ve been doing in recent comment sections.
12. January 2012 at 12:02
Err sorry, it’s dodgy to think of bank output as lending, I meant risk management, maturity transformation etc…
12. January 2012 at 12:04
Brito, I don’t agree that funds are capital that creates lending. The capital used for bank intermediation is things like bank buildings. In the same way that realtor’s offices are capital used to create housing transactions. But the houses themselves don’t create the transactions.
12. January 2012 at 12:06
Scott, short term duration lending (deposits) is what is used by banks to produce maturity transformation (combining them to make mortgages and other long term loans), deposits enter the maturity transformation (production) function, I still think it makes sense to think of that as capital.
12. January 2012 at 12:07
ssumner:
“Regarding the battle between fiscal and monetary policy; the fiscal authorities can move V by a few percentage points. the Fed can move M by a billion percent. Guess who controls M*V?”
Empirically, I agree with you. Theoretically, you are wrong.
If a fiscal authority decides to implement a policy that taxes all wealth at the end of period T that is not spent at 100%, and allocate it evenly across all people in the population, it wins. I don’t care WHAT interest rate you charge, or pay on savings – it’s all confiscated.
The only reason it can’t is because of institutional constraints… the same institutional constraints that say the Fed can’t create a billion percent inflation. The notion that the Fed is more powerful than Congress is illusory – it is because we say it is.
Glasner is correct that your argument against fiscal efficacy is circular in so far as it is framed in terms of what is possible, not in terms of what is practical. In the CURRENT institutional setup fiscal policy is constrained by the Fed because we have given the Fed more power than Congress. That is your argument, if it is true.
“Grapes dumped into the ocean out of inventory is depreciation”
That was my original thought, except that this implies that ALL current production passes through “inventory” and is thus temporarily considered an investment. In so far as fresh grapes would ever be considered an “investment” the accounting time frame is very short. But if GRAPES are an investment, then pretty much every good ever produced is (for some period) an investment. In that case, there is NO such thing as consumption of current production, merely depreciation of investments (eating food from the fridge is just like driving a car for a year, just faster). If that’s true, then inventories suddenly become critical to address in the core model.
This is just to make Nick Rowe’s point – savings is not a thing, it’s a non-thing (non-consumption) in that accounting identity.
12. January 2012 at 12:12
Scott,
I agree that new cars are (largely) savings, but I don’t agree that the act of buying a new car is an act of saving (except on the part of the car dealer, who saves the profit). Buying a new car (with cash) is changing the form of savings that you already have. (Actually, it also includes a significant immediate consumption component, because the car depreciates very rapidly in the first nanosecond of ownership, but that’s a side issue.) In practice, an act of saving is almost always an act of setting money aside, in the sense of receiving money but not immediately spending it.
“I think 100% of the time the person that gave them the money is dissaving in a few nanoseconds…
They’re only dissaving if the purchase was consumption. This is the flip-side of the car purchase: the purchaser is not dissaving but changing the form of her savings. She is dissaving if she purchases a haircut.
As a practical matter, if I choose to hoard money, I am saving. Depending on the central bank’s reaction function, I may be encouraging someone else to dissave by reducing their income, but for my part, I am saving, even if I don’t cause any capital goods to be produced. It’s meaningful to say that countries with trade surpluses are saving, even if they aren’t causing any capital goods to be produced.
12. January 2012 at 12:30
I used to think that the whole savings = investment thing was simply an innocuous simplifying assumption, just to make model computation easier, not that it was some fundamental and irrefutable truth. It just seemed reasonable to think that money not spent on purchasing output must eventually be channeled into investment into capital. Now I’m not sure.
12. January 2012 at 12:35
Setting money or even stockpiling consumer goods is the prerequisite for building capital goods. To see this more clearly, look at the one person “Robinson Crusoe” economy. In order for Robinson to build tools, basic capital goods, he has to underconsume and by either stockpiling food or reduce his food consumption while he builds his tools. The introduction of money doesn’t introduce any fundamental changes to this basic fact. You have to set money aside in order to construct capital goods (Money as a medium of exchange means that you don’t have to stockpile consumer goods in a monetary economy). Saying that the consumption of capital goods is saving is analogous to saying like eating food is producing food. A strange confusion about cause and effect.
12. January 2012 at 12:42
Here is another way of looking at it, the ‘circular flow’:
http://images.flatworldknowledge.com/suranovic/suranovic-fig02_003.jpg
Households buying from firms: consumption
Households putting money in financial institutions: saving (Shh)
Firms putting money in financial institutions: saving (Sb)
Firms using the rest of the money to pay wages: household disposable income
Financial institutions moving money into firms: investment = spending on capital
For saving to equal capital, Sb + Shh = I. This is a reasonable assumption, and again I think just a simple one made for the models. However in reality, I might be less than Sb + Shh, or I might be more than Sb + Shh.
12. January 2012 at 12:55
I don’t think anyone doubts the accounting identity I=S. It doesnt follow that putting money in stocks has anything to do with savings.
Show me the textbook that says that savings means building capital goods. That’s where the disconnect is
12. January 2012 at 12:55
Integral: “Of course even with government, aggregate (public + private) saving still equals zero.”
Agreed.
12. January 2012 at 13:17
Mike Sax: “It doesnt follow that putting money in stocks has anything to do with savings. ”
Why not? Do stocks vanish when you buy them? No, they are generally durable. People put money in stocks as they expect a return such that it can fund future consumption. Putting money away somewhere to fund future consumption is basically the definition of saving.
12. January 2012 at 13:22
Scott,
I chatted to you briefly when I was interning at the IEA in London, I now work for the (British) treasury, but posts like this make me think that if I’d taken the opportunity to pick your brains, I might have a much better understanding of how the world works…
PS – you seem to answer almost every comment, that’s really amazing.
12. January 2012 at 13:31
As to your three “capital goods” food, houses, and cars, the first is definitely consumption spending, houses can be seen as investing, cars I;m not sure whether it’s investment or saving but it certainly isn’t saving.
At best it seems that what Keynesians call saving you call hoarding. So we might speak of the paradox of hoarding.
12. January 2012 at 13:38
Brito, That’s fine, but again economists don’t consider financial assets to be capital. It is viewed as debt.
Statsguy, You said:
“The notion that the Fed is more powerful than Congress is illusory – it is because we say it is.”
Oh really. So we had the Great Inflation in 1965-1981, and then Congress decided to start targeting inflation at a low and steady rate. And Reagan did this by running huge budget de . . . oh wait, nevermind.
Let me try again. The Fed decided to target inflation at low and and steady rates by following the Taylor Rule. Which one sounds more plausible to you?
One of my professors said that a can of beer is a capital good that depreciates very rapidly as you drink it.
Andy, You said;
“I agree that new cars are (largely) savings, but I don’t agree that the act of buying a new car is an act of saving”
I think we are getting into semantics here. You point out that in the nanosecond after being paid almost all income is saved. That’s true. Then if you later buy consumer goods it’s regarded as consumption, and if you buy capital goods it’s still an asset that was previously saved. I agree. But these accounts are done on an annual basis, so it doesn’t matter what happens in a nanosecond, but rather what happens in a year. If at the end of the year you have earned $50,000 and all you have to show for it is a one year old car worth $18,000, then you’ve consumed $32,000 and saved $18,000 during that year. Can we agree on that?
Andy, You said;
“As a practical matter, if I choose to hoard money, I am saving. Depending on the central bank’s reaction function, I may be encouraging someone else to dissave by reducing their income, but for my part, I am saving, even if I don’t cause any capital goods to be produced. It’s meaningful to say that countries with trade surpluses are saving, even if they aren’t causing any capital goods to be produced.”
I’m going to quibble with both points. Global saving equals global investment. Everyone agrees that national saving differs from national investment by the CA. So no issues there. The saving of cash example is more complex issue. Cash can be viewed as government debt, or as a sort of capital good, like paper gold. Something that provides transactions services.
If it’s government debt then any extra cash you hold means equal dissaving by the government. National saving is unchanged. If it’s capital and your saving of cash causes the real quantity of cash to go up, then it is saving, but investment has also occurred. That interpretation makes most people queasy, but I throw it out there for completeness.
Brito, Nope, It’s a definition.
John, I agree, consumption of capital goods mean less net investment.
Mike Sax, You said;
“I don’t think anyone doubts the accounting identity I=S. It doesn’t follow that putting money into stocks has anything to do with savings.”
That’s what I’ve been saying too.
As far as the textbooks, both Mankiw and Krugman’s textbooks say S=I is an identity. I believ ethe others do as well. And I is defined as construction of new capital goods.
12. January 2012 at 13:40
Thanks Adam,
Mike Sax, Krugman says S=I by definition, so does Mankiw. I suppose they aren’t Keynesians then?
12. January 2012 at 13:46
“Brito, That’s fine, but again economists don’t consider financial assets to be capital. It is viewed as debt.”
Which economists? Maybe macroeconomists, certainly not economists in general. In the financial economics modules I took, it was taught by economists and required me to read papers by economists. There, the term capital was very frequently used to describe the assets on the balance sheets of banks. There is even a wikipedia article about it: http://en.wikipedia.org/wiki/Financial_capital
I mean have you not been paying attention to economists working in regulation, how they talk about ‘capital requirements’ etc..?
12. January 2012 at 13:52
Wikipedia has a nice definition:
“In a fundamental sense, capital consists of any produced thing that can enhance a person’s power to perform economically useful work””a stone or an arrow is capital for a caveman who can use it as a hunting instrument, and roads are capital for inhabitants of a city. Capital is an input in the production function.”
12. January 2012 at 14:03
“SAVING IS NOT SETTING ASIDE MONEY, IT’S SPENDING MONEY ON CAPITAL GOODS. PERIOD, END OF STORY.”
Scott, also pretty much every microeconomist would disagree with you here. Think of a typical intertemporal choice problem. There is consumption now, there is consumption in future periods. Saving is income not spent now to be used to fund future consumption, that is the definition. This money put aside can also earn an interest. There is no explicit mention of capital goods anywhere, the only relationship is that you could say that purchasing capital is ultimately required for saving to earn an interest.
12. January 2012 at 14:37
Scott how many times have I sadi that S=I is true but in the sense of an accounting identity.
If you can show me where Krugman says that savings is building capital goods and you’ll convince me.
I’d be shocked if you can find Krugman saying putting money in a stock is saving.
What your fellow market monetarist Bill Woolsley says above makes a lot of sense:
“I don’t agree at all.”
“Saving is that part of income not spent on consumer goods.”
“Investment is spending on capital goods.”
Again I’m not denying S=I in an accounting sense. In accounting there is another identity debits=credits+owner’s equity. Does this mean that debits which are largely assets are the same thing as credits which are largely libailities?
12. January 2012 at 14:38
I have some free time tomorrow and will try to look through it. If you open it up, I’m sure if you just (cntrl + F) Fischer you will find plenty of red meat!
12. January 2012 at 14:46
Ryan Avent has a ‘greatest hits’ here:
http://www.economist.com/blogs/freeexchange/2012/01/miscellany
12. January 2012 at 16:39
PERSONAL LEVEL
Buying a car is spending. Walking is saving.
Buying a house is spending, that you can sell it later for more, or owe on it what you cannot pay doesn’t change that money left your pocket, and will continue to leave your pocket, went into some other state, and is no longer money you have immediate access too.
Piggy banks ARE savings.
You might call the car an investment to keep you from walking, or you might be a pizza delivery guy, but again we are talking about saving more at a personal level.
Note the goal here:
1. to use words as regular folk understand them.
2. to recognize that words are not value neutral.
I’d argue that based simply on the absolute failure of the discussion here…
The important thing about the words chosen, is that the economics profession USE WORDS that everyone else use. The whole point of spreading a meme, is to make the idea digestable to the most people – to meet them on their terms.
Essentially, economists should write out everything in paragraph form the way a old woman with an 8th grade education can best understand it EVEN IF it slows down economists when they are talking to each other.
More words for economists to write and say = less complicated ideas for everyone to understand. Specialized language REALLY IS about exclusion.
It isn’t just this blog post… the facts show that economists have gained nothing over the long run of history doing it the way they do it now.
Instead you all should focus on meme spreading – deny to yourselves the idea that this stuff is complicated.
Also, you should be wary of pretending words are non-neutral values.
Ex: Since right now people are buying houses on credit, we should NEVER call buying a house an investment because it excuses people from the guilt that they are spending money, not putting it in a piggy bank. If later, it becomes standard to not buy a house until you have the cash, we can talk about it like an investment again.
When people “hoard cash” or put money in a piggy bank, or SAVE IT, they are doing what we (and by we I mean me) want them to do – which is stop taking on debt.
Now, if you buy the house for cash from your piggy bank, if you want to buy the car from your piggy bank, we can excuse that kind of spending…. we should even cheer it, buying things for cash not credit is a good thing!.
So now you know what to call:
1. saving – good
2. cash – good
3. credit – bad
4. spending – bad if done on credit, more acceptable if done from savings, but still bad
5. investment – non-cash savings where risk is taken to have more savings later – good
Ex:
When people get sick they SPEND MONEY (bad) to be healed, they do not INVEST (good) in their future selves.
When people get sick, and have to use debt / CREDIT to be healed, this is worse.
—–
The end result of such a system will be one where people save / hoard their cash, the price of things will dive, old companies will be out innovated and die, old labor skills will be made worthless, bankers will be less powerful, and there will be incredible demand that the government keep inflation down to nothing.
These are good things, whats more when these things happen, there will be no reduction in investment in real terms, in reality investment will increase.
The idea of investing based on credit should be abolished…
We want everyone hoarding money in piggy banks until they 1) spend it without credit or 2) invest it.
MACRO LEVEL
Who cares about the macro level, we want the Fed to stop printing money. There is no long run upside to credit markets supporting home buying and home values.
We should adopt language with the expressed intent of achieving these ends.
p.s. really truly you people should be nationally shamed into sacrificing all your special jargon, and doing more work so everyone else does less.
12. January 2012 at 17:46
Morgan that is ridiculous, credit is not inherently bad. Every single model of efficiency ever where this thing called the FUTURE exists would suggest that it is absolutely optimal to smooth consumption, which requires both credit in some periods and saving in other periods.
12. January 2012 at 18:59
With all that you say Morgan we should call you Dr. Austerity. Do you even believe in NGDP targeting?
12. January 2012 at 21:40
Mike,
I believe in NGDP level targeting at no more than 4% – preferably 3%. I think Scott’s plan is genius:
1. It will deliver far less inflation than we have had previously.
2. It will make CRYSTAL CLEAR that we have a macro planned economy and that as such growth should be only used for productive private RGDP and the unfortunate effects of commodity inflation – essentially it will make us hate rent seeking. It will make us hate public employees getting raises, we will despise Fannie and Freddie, we’ll even hate backstopping student loans.
Clarity changes everything Mike, when they say they are going to give someone a deficit financed hand out, we can attach instantly a “how much of this years NGDP does that eat up?” discussion to it.
Precisely when business is booming, and things are feeling good, there will be TERRIFIED SHRIEKING across the land if public employees get a raise…. we’ll also feel that way about war.
That is fundamentally different from what has happened since 1980.
With a level target at 4%, we’re one step closer to Austrianism.
Brito,
I understand that credit is at minimum helpful when people are desperate. Intellectually, I think “Natural Money” is a far more interesting theoretical construct than MMT.
But generally again, my point is really about the semiotics – we want to choose words that are most understandable to the most people, and give the most people the right kind of mental nudges when they read the words.
13. January 2012 at 00:02
Orionorbit claims: “Now the above is accepted by all philosophers of science today. I don’t know even one professional philosopher in the English speaking world that would dispute this.” By “the above” he means two points: “First, the Popperian revolution, i.e. the notion that we do not VERIFY a model (such as Newtonian physics) we simply fail to refute it. The second revolution came with Quine’s “two dogmas of empiricism”, one of which is that there are tautological truths such as “all bachelors are single”, where Quine showed that this simply is not true. You can never claim that S=I is a tautology hence true.”
This is ridiculous. Popper is now held in pretty low regard among most philosophers of science who discuss confirmation, theory testing, etc. And the dispute between Carnap and Quine about the analytic-synthetic distinction (more relevant here) is generally agreed to have been something of a standoff, in which (as so often in philosophy) the two sides were talking past each other.
Anyway, no one ever claimed (least of all Carnap) that a definition is TRUE by virtue of being a definition. It’s true relative to a language, of which the definition is constitutive, and (according to Carnap) there’s no such thing as truth outside some such defined or constituted language — there’s no non-language-relative truth.
This is all common knowledge. Type “Carnap Quine” or “analyticity” or “analytic synthetic” into Google scholar, and follow the links.
13. January 2012 at 01:49
“Dissaving”? So now we have S=I-Dissaving?
Why not S-Depreciation=I-Dissaving?
If this trend continues we might be in the real world after a while.
13. January 2012 at 04:38
The way you describe saving is not how my macro tutour explained it to me:
Saving includes: putting money in a bank, stuffing money in your matress AND buying a house.
S=I because: if I buy a house, that’s an investment. If I put the money in a bank someone else is investing it, and if I put the money under my mattress, then I’m not buying something from a store and then that store would have an extra stockpile of goods (on which i didn’t spend the money), and stockpiles count as investment in this identity.
13. January 2012 at 05:16
James in London: No. “Dissaving” just means negative saving.
13. January 2012 at 06:03
Nick Rowe: No. “Dissaving” can mean selling your investments (like mutual funds, drawing on savings accounts, house downsizing, even).
Or, perhaps, we agree, but mean “Net I” rather than I, ie gross investment less dissavings.
Or dissaving can mean borrowing, ie someone else’s saving.
13. January 2012 at 06:11
Brito, OK, I stand corrected, if they say “financial capital” then that’s OK, But I still say there aren’t any economists who think you can lump physical and financial capital together and call it “capital”. Suppose a bank had the following on its books:
a. Shares of stock claiming ownership of the Empire State Building ($500 million)
b. The Empire State building ($500 million)
They’d be criticized for double counting. I thought that was what you were trying to do.
Saving relates to physical capital, not financial capital.
Brito, You said;
“Saving is income not spent now to be used to fund future consumption, that is the definition.”
That’s flat out wrong. Take my car example. It also includes money spent right now on capital goods, which can be used for future consumption. Suppose the income was “spent” on shares of stock, are you claiming that’s not saving?
Mike Sax, I completely agree with Bill that saving is that part of income not spent on consumer goods, that’s what the post is all about. The part not spent on consumption is spent of capital goods. That’s my point. My criticism was aimed at those who claimed saving is money not spent at all!
If present Obama encourages Americans to go out and spent lots of money on new cars and homes, then he is, ipso facto, telling Americans to go out and save lots more.
Marcelo, Doesn’t take much to get them to laugh.
Morgan, We need words to have precise meanings, or else we can’t communicate in complex fields like macro. The public’s thinking is hopelessly confused, they should not be our inspiration. Remember the public thinks “inflation” hurts consumers. They think employers pay half of the payroll tax. They think imports cost jobs, etc.
A.W. Carus, I agree that Popper is now passe.
James, I love how you enjoying mocking things you know nothing about. Check out Nicks reply.
13. January 2012 at 06:14
Sam, He’s not wrong, but he’s describing individual saving. I’m talking about saving at a national level. Putting money in the bank only leads to more national saving to the extent that it leads to more capital being built. Otherwise your deposit is just a loan to someone else, who dissaves an equal amount.
13. January 2012 at 06:16
But Scott, Morgan believes that inflation is a bogey man too don’t you Morgan? That’s why you choose such a low NGDP target.
13. January 2012 at 06:23
Scott when you say, “If present Obama encourages Americans to go out and spent lots of money on new cars and homes, then he is, ipso facto, telling Americans to go out and save lots more.”
Does this mean that it’s simply an issue of vocabulary? If Obama ecnourages everyone to spend on cars and homes he probably thinks he’s encouraging them to spend more on consumption-as I think many do.
According to you he is actually telling them to go out and save some more. OK, assuming even that you’re right, what is the payoff in such a change of vocabulary? Is it simply precision-if so I don’t think it’s worth the heavy lifting.
What I wonder is that even if you’re right that when Keynesians talk about the paradox of thrift they should be talking about the paradox of hoarding, does this matter in any more than a linguistic sense?
What would change policy wise assuming you are right and people recongized it?
13. January 2012 at 06:40
I don’t know why you bother with I=S, since S defines I in macroeconomic theory. Why not just write I=I and be done with it.
13. January 2012 at 06:42
S and I are categories of our national accounts and are summed over a fiscal year. They can’t be used to show an equilibrium. Why?
1) Because income does not measure as equal to expenditure. There is a special reconciliation procedure which forces equality. This is desirable because the accounts are a form of double entry bookkeeping where debits = credits.
2) Because S = I isn’t an equilibrium. It’s like taking a pot of boiling water and saying that loss from the pot = gain of the air implies an equilibrium between the air and the pot.
S is a residual. S and I as defined in our national accounts can’t fail to be equal under any circumstance whatever.
Also the definitions of income and expenditure are filled with arguable points and grey areas. Is lumber a capital good or a consumer good? Implied rent!? What is the difference between
Selling a business
Selling the assets of a business to be used in a new business
It’s misleading to argue as if S and I were savings and investment as usually understood by non-economists, and economists do this all the time. I believe you’ve been guilty of this a time or two.
It would be nice to shed this baggage.
13. January 2012 at 06:49
Mike I keep saying this, I want a low target because it based on historical RGDP with little room left for inflation.
I want this not because I think inflation is boogie man, under Scott’s plan inflation is not one, but because I want to force an either / or choice where we all admit that gvt. spending is not a comparably productive use of our allotted growth.
I want us to encourage the creative destruction of say the Post Office, where everyone is RELIEVED we shut it down and fired everyone because that FREED UP some space for growth.
Instead of the economy taking a hit, we think shweet! now rates can be lowered for small businessmen, or kept lower for longer.
Right now we don’t have the either / or CLARITY so we don’t see the simple choices we should be making.
Notice again, under Scott’s plan inflation allows us to do far more radical forms of privatization and faster technical change, because there is no chance of deflation.
——-
Scott you are horribly wrong, the ideas are not complex – the policies are complex, we can force the policy choices to become less complex and thereby make the ideas far easier to grasp.
You are ALSO WRONG because complex ideas can be described with a more limited vocabulary, but it takes FAR MORE WORDS – like certain computer languages take more code / symbols to accomplish the same task.
What you don’t like about my policy is you suddenly in your profession, you don’t get to use short hand, instead we say your previous use of short hand is the REASON, the PROBLEM why your profession sucks so badly.
So we take your short hand away from you, and force you to say really long run on sentences with tons of repetitions, and qualifiers, in words based on what the folks at the bottom can understand.
13. January 2012 at 07:00
[…] Source […]
13. January 2012 at 08:25
Scott,
You have accepted the fact that the value of Investment depends on our choices now about future patterns of consumption and output across time, not requiring any changes in the sacrifice of current consumption in the present, i.e I can be expanded or contracted by choosing between alternative production processes which transform consumption and output patterns in the future, but NOT IN THE PRESENT.
You’ve also stipulated that S is a past measure of expenditure in money terms, set aside for Investment in capital goods.
In other words,
I is NOT fixed, its value depends on our choices and the uses of make of things.
S is a FIXED measure, a fixed stock, measure IN THE PAST.
Therefore, S does NOT equal I. And S does NOT determine I.
QED
13. January 2012 at 08:43
[…] Reading Sumner backs me up. Advertisement GA_googleAddAttr("AdOpt", "1"); GA_googleAddAttr("Origin", "other"); […]
13. January 2012 at 08:51
Also borrowing is the opposite of saving so paying down principle is saving.
People often assume that people with below median income have a higher marginal propensity to spend in a downturn but that is not true because the rich often spend any windfall on stocks while those of moderate income use it to pay down debt in a down turn.
13. January 2012 at 09:01
[…] and the ongoing “Great Recession.” It should be no surprise to anybody that when Scott Sumner blogs an extremely confused post on savings an Austrian, such as myself, is going to jump at the […]
13. January 2012 at 09:58
When the bank & financial institution capital reserves crash, when their liquidity evaporates, and when firm solvency dies, how does that not alter the supply of money — and the supply of what makes transactions and exchanges possible?
I want to know this world where the real economy is complete isolated from and causally independent of the pure world of money and it’s supply.
13. January 2012 at 10:59
Scott, read Quine’s article. The point is that your certainty that Bachelor=Unmarried man depends on you never having met a married bachelor, i.e. even “tautologies” are subject to empirical verification. On the other hand, you HAVE met situations when people might on average want to keep their spending on capital goods growing at a steady pace while their credit runs out. But then again, I’m not saying that this is in itself enough to justify Krugman’s view, but his more or less correct predictions are, until somebody comes up with an even more accurate model.
I would say that if you want to convince people that assuming S=I is a good idea, you will make your job much easier if you can show what kind of bad predictions a model will make if this assumption is ignored.
Britto, first, i am not a philosopher of science, or a philosopher of any kind. That said, you misunderstood my point about M*V. If the best you can do is to say “well, M went up but V went down even more” i’d say, you’re not wrong, you’re not even wrong, because in these terms there is no observed value of M that could refute your model, as you could always justify NGDP movements that are M-inconsistent in terms of some hypothetical movement of V. An irrefutable model is not even wrong, it is completely meaningless as far science is concerned and no, you don’t have to be a philosopher of science to understand that, just ask yourself why we think astrological models are unscientific.
But again, I am not saying that M*V is an unscientific way to look at things, quite the opposite, in fact i even conceded that when you use better proxies for M than M2 the model produces much better predictions. So focus on the issue we are discussing i.e., that in order to say that “Scott’s model is better than Paul’s” you need to show in which way’s Paul’s model produced worse predictions, just like you would have done if you wanted to show why general relativity is a better model to take your rocket to the moon than Hamiltonian mechanics.
13. January 2012 at 15:41
orionorbit,
good lord. It is purely definitional.
you cannot meet a married bachelor, because if M, not B.
it requires no empirical verification, if you met someone who claimed to be both, they would be wrong, your assumption can never ever be wrong, even if you never meet a single person.
13. January 2012 at 16:40
Mike Sax
“In accounting there is another identity debits=credits+owner’s equity.”
Nope sorry. In accounting, Assets=Liabilities+owners’s equity. DR=CR….period. Debits are not the same thing as assets.
14. January 2012 at 07:45
Mike Sax, To see the payoff read my recent posts where I show how sloppy thinking by Wren-Lewis allowed him to make embarrassing errors. Also, the saving/hoarding difference is hugely important, and one reason the old Keynesian model crashed and burned after the 1960s–they were focused on saving whereas they should have been focused on money supply and demand.
James, Because the saving and investment schedules are different. The desire to save rises with higher interest rates, whereas the desire to invest declines.
Peter, Yes, the government’s definitions are fuzzy and imprecise.
Greg, I agree that greater efficiency can allow C and I to rise at the same time. But if so, then C and S have also rising at the same time. S=I is an identity.
Floccina, Good point.
Greg, I define money as the monetary base–produced by the Fed.
Orionorbit, If I meet a married bachelor, I call him a liar.
You said;
“I would say that if you want to convince people that assuming S=I is a good idea, you will make your job much easier if you can show what kind of bad predictions a model will make if this assumption is ignored.”
I do that in recent posts crticizing Wren-Lewis.
14. January 2012 at 07:52
“That’s flat out wrong. Take my car example. It also includes money spent right now on capital goods, which can be used for future consumption. Suppose the income was “spent” on shares of stock, are you claiming that’s not saving?”
No, I’m not claiming its not saving. Saving is wealth set aside for future consumption, this definition is what is presented in every microeconomics textbook. It is what occurs in inter-temporal choice models and importantly, these inter-temporal choice models further form the basis of household optimality conditions in microfounded macroeconomic models. Assets are an imperfect and less liquid form of wealth than money, but if you’re using it as a store of value then it can still be saving.
14. January 2012 at 10:29
Scott: I re-read your post. Still didn’t like it. I re-read Nick’s post. Liked it. I re-re-read your post in light of his post. It looked a lot better. Then for good measure re-re-read Nick’s post.
I think I get it, but I’m filing this in the “general equilibrium is hard” and “macro aggregates do not work like micro” folders.
14. January 2012 at 22:51
This isn’t true, and the issue isn’t “efficiency”. In fact, depending on the choices made, (A) a fixed S can be structured to commit one to fixed C and I now, and lower C in the near term and increased C and I in the long term.
Or (B) S can be structured to commit one to fixed C and I now and the same the same C and I in both the near term and the long term.
This follows conceptually from the fact than no one chooses a longer production process unless that process promises greater output, and from the fact that such choices exist (read Bohm-Bawerk if you are not clear on this).
From this fact follows the alternative possibilities (A) and (B), among others.
“Greg, I agree that greater efficiency can allow C and I to rise at the same time. But if so, then C and S have also rising at the same time. S=I is an identity.”
The underlying problem is that S = I has no purchase on the economics of production goods using the logic of marginal valuation to make sense of the value of thing taking a forward looking perspective.
It’s a deeply misleading construct based in a fallacious use of backwards dollar-based or labor hour based (Keynes) measures of “value” crammed into an ersatz circular flow construct with fraudulently pretends to be to build within the parameters of the economic way of thinking — I.e. which pretends to be equivalent to and combinable with and combinable with forward-looking marginal valuation thinking.
15. January 2012 at 07:30
Brito, You said;
“No, I’m not claiming its not saving. Saving is wealth set aside for future consumption, this definition is what is presented in every microeconomics textbook.”
Yes, and spending on a car is clearly wealth set aside for future consumption (of transport services.) Ditto for spending on a house. Even the US government considers spending on a new house to be saving.
Integral, Thanks for those comments. I’m now discovering that even well-informed economists see these things differently from me. In a recent post I quote Wren-Lewis as saying saving isn’t spending on capital goods. I thought everyone accepted that, it’s in all the textbooks. It turns out that the popular view of saving is deeply entrenched.
Greg, Economists consider S=I by definition. If you don’t see it that way, then you are defining saving differently than we do. I find that debates over the meaning of words is sterile. There’s nothing to debate. If I say the sky is blue and you say it’s green, then we differ over the meaning of the term ‘green.’ End of debate.
15. January 2012 at 09:44
“Yes, and spending on a car is clearly wealth set aside for future consumption (of transport services.) Ditto for spending on a house. Even the US government considers spending on a new house to be saving.”
Scott, maybe you just misread me there? I said: “No, I’m not claiming its not saving.” As in, I do agree that buying cars and a house can count as saving, as these assets are a form of wealth. However it’s a really really bad and confusing example to use, and I don’t see it used in textbook macro.
By the way, I checked what the Blanchard macro textbook had to say about this (which is pretty much the most mainstream book in Europe), and he definitely does not define saving as ‘spending on capital goods’, he defines it as the sum of individual saving, which is the sum of income not spent on consumption. The S = I is an identity, but that’s not the same as a definition (true by definition =/= the definition), just like MV = PY is an identity, but the definitions of all those variables are different. S = I is true because any net saving must either be channelled directly into investment, or the fall in demand from less consumption causes an increase in business inventories, which is defined as investment in macro.
15. January 2012 at 09:51
“The underlying problem is that S = I has no purchase on the economics of production goods using the logic of marginal valuation to make sense of the value of thing taking a forward looking perspective.
It’s a deeply misleading construct based in a fallacious use of backwards dollar-based or labor hour based (Keynes) measures of “value” crammed into an ersatz circular flow construct with fraudulently pretends to be to build within the parameters of the economic way of thinking “” I.e. which pretends to be equivalent to and combinable with and combinable with forward-looking marginal valuation thinking.”
I refuse to believe you are speaking English here and not just randomly combining obscure words together.
16. January 2012 at 02:39
“James, Because the saving and investment schedules are different. The desire to save rises with higher interest rates, whereas the desire to invest declines”
I’m not at all sure what this means. You and others declare that S=I, and saving very definitely isn’t “putting money aside”. Why do higher interest rates increase the desire to save, under your and others definition of saving?
Paradoxically, “putting money aside” earns a higher nominal return with higher rates, for sure, but you actually need to put less aside as a result if nominal returns are what you are interested in. You need to put more aside when nominal rates are low, as indeed millions of savers are doing right now.
16. January 2012 at 12:33
Brito, S=I is deeply misleading to people who don’t know how to work with the concept. See my new post for a quotation that shows someone misusing the idea. I do know how to work with the identity, so it’s not deeply misleading to me.
Investment is production of capital. Gross investment is total production of physical capital; net investment is net increase in the capital stock. I’d guess that’s in Blanchard somewhere. I’ve taught macro for 30 years and quite frankly this is very basic stuff.
James, Most people assume the substitution effect outweighs the income effect, but of course nothing I’ve argued hinges on that.
16. January 2012 at 19:38
“Brito, S=I is deeply misleading to people who don’t know how to work with the concept. See my new post for a quotation that shows someone misusing the idea. I do know how to work with the identity, so it’s not deeply misleading to me.”
Scott that wasn’t me who said that, I was quoting Greg and ridiculing it.
17. January 2012 at 17:04
Sorry Brito, I get so many comments that sometimes I misread them.
17. January 2012 at 19:31
[…] by Scott Sumner // サムナーã®ãƒ–ãƒã‚°ã‹ã‚‰”Saving isn’t “setting money aside,” it’s BUILDING CAPITAL GOODS“(12. January […]
24. January 2012 at 07:31
[…] results in lots of confusion and logical error. Viz: careful economic thinkers like Nick Rowe, Scott Sumner, and Andy Harless feeling the need/inclination to write lengthy think-pieces on that nature of […]
24. January 2012 at 08:28
[…] results in lots of confusion and logical error. Viz: careful economic thinkers like Nick Rowe, Scott Sumner, and Andy Harless feeling the need/inclination to write lengthy think-pieces on that nature of […]
24. January 2012 at 08:45
[…] results in lots of confusion and logical error. Viz: careful economic thinkers like Nick Rowe, Scott Sumner, and Andy Harless feeling the need/inclination to write lengthy think-pieces on that nature of […]
22. February 2017 at 10:40
[…] http://www.themoneyillusion.com/?p=12617 […]