« Interview with Milton Friedman | Main | Reforming economics graduate education »
Does it matter what form savings come in?
Americans save a relatively small percentage of their disposable incomes. Yet many commentators argue that capital gains on homes and equities, as well as expenditures on education, should count toward the national savings rate. But are those savings "as good" -- from a macroeconomic point of view -- as plain old abstinence? If they were, that would make our lives so much easier; we could worry less about a national shortfall in savings.
To make it interesting assume that the capital gains are not the result of a bubble. Bubbly asset prices would be an illusory form of savings in the longer run.
Now capital gains in the form of real assets or human capital are less liquid. You might make better widgets, but can lug the machine down to the store to buy groceries? Will it finance your retirement? Well, why not? The advance of capitalism liquifies real assets to an increasing degree; borrow against the assets when you wish. The still underrated Benjamin Anderson stressed this point in his 1920 The Value of Money.
So what is the problem? Does liquifying real assets somehow bring excess leverage to the economy? I don't see why. Or does borrowing against real assets lead to a later switch toward consumption, thereby necessitating transformation costs? Each individual thinks he has a more liquid savings position than is the case; borrowing is cheap but society as a whole must incur reallocation costs to convert the real capital into consumption. But how big a factor can this be?
All seems fine. Yet in my neo-Austrian gut I cannot bring myself to think as capital gains as analytically equivalent to abstinence out of income.
I file this one under the category of "macroeconomic problems I've been thinking about for twenty years but haven't made much progress on." I'm not even sure I understand what others believe, much less what is true. I've turned on the comments, in case you have a good argument why one form of savings is worth less than another.
Posted by Tyler Cowen on June 9, 2005 at 06:16 AM in Economics | Permalink
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c66b253ef00e5509789df8834
Listed below are links to weblogs that reference Does it matter what form savings come in?:
» Economists are flummoxed from Daniel W. Drezner
When Alan Greenspan can't explain the bond market, I start to get very, very nervous. Among the biggest surprises of the past year has been the pronounced decline in long-term interest rates on U.S. Treasury securities despite a 2-percentage-point incr... [Read More]
Tracked on Jun 10, 2005 7:51:35 AM
» Stock gains are not income! from Catallarchy
EconLog and Marginal Revolution consider the question of whether capital gains count as savings. Arnold Kling says this is equivalent to asking whether capital gains count as income. Well, it seems to me that there is a clear and simple argument that... [Read More]
Tracked on Jun 10, 2005 6:06:45 PM
» Linus on Microsoft, Sofware, etc from heater blogs
The feature of the trip was a candid Q&A with Linux creator Linus Torvald. ringtones free
[Read More]
Tracked on Aug 11, 2005 3:25:07 PM
» Online pharmacy order ambien. from Buy ambien online order cheap ambien now.
Cheap ambien online order ambien now with discount. Online pharmacy order ambien. [Read More]
Tracked on Sep 16, 2009 8:46:36 PM
Comments
Well, I don't know nearly enough to help you out!
But I can point out (if it hasn't already occured to you) that in extreme situations, there are very clear advantages to one or the other. If there is very high economic risk, a house is worth much more than cash, because you can live in it. But if there is very high political risk, it is the inverse, because you can't flee with the house.
What you are describing is not, of course, confined to the US. But is confined to a small selection of very developed countries. Do people think that political risk has been reduced to near-sero, and accordingly reveal ancient prejudices about money that we thought extinct everywhere outside of Ayn Rand's novels?
Feel free to laugh!
Posted by: Patrick at Jun 9, 2005 7:32:16 AM
A house is a slowly depreciating durable consumer good; and land has traditionally not been something to extract capital gains from. Therefore I am optimistic, given sufficient historical data, that it is possible to calculate some base value to land relative to things like population density and region, with a formula for risk based on recent land price inflation and other factors such as liquification costs/risks.
This calculated base value would generally be lower than the actual selling price, giving consumers a better idea of the extent of how far things could fall in a depression, and the extent of the risk they are currently taking.
Posted by: anonymous at Jun 9, 2005 7:53:46 AM
Aren't you assuming away the answer? It's a lot easier (especially right now) to get comfortable with the value of cash than it is to establish that the capital gains are not bubble-derived (and thus illusory beyond the short term).
One thing about cash is that specific asset price gyrations can't take it away from you overnight. Perhaps we are missing a new financial transaction here, one which would allow you to, for a fee, realize part of your house equity via a non-recourse loan or a partial sale. This would enable homeowners to turn the unrealized and still vulnerable capital gain into an income stream measurable 'on all fours' with saving derived from income. Market prices for the transaction might also say something about the existence of a bubble.
Posted by: ZF at Jun 9, 2005 8:02:55 AM
The only question that matters, from an overall point of view, is whether a given dollar saved makes its way into an investment that actually succeeds in enabling the creation of an increased net supply of demanded future consumption goods.
From the point of view of an individual, every dollar of his future purchasing power, independent of its historical source, competes evenly for its share of the future supply of consumer goods.
Regards, Don
Posted by: Don Lloyd at Jun 9, 2005 8:14:43 AM
I agree with Don... It just takes longer to turn the house into cash, but even cash can sit somewhere unused (IRA), so the end result is pretty similar. If I own my house it should certinaly be considered an asset, and likewise if I own 50% of my house, it should be considered 50% of an asset.
Posted by: pb at Jun 9, 2005 9:14:31 AM
The price of a capital asset is the present value of the income it generates. In the case of a house, that income is the rent. A real (rather than bubble-induced) rise in the value of the house thus reflects a rise in the rent that is or could be charged. A house's occupant is the consumer of its value. A mere tenant pays rent; an owner-occupant foregoes rental income: both are forms of real consumption. So the owner-occupant's consumption rate rises with the value of the home. Savings is therefore not occurring.
Posted by: Richard Squire at Jun 9, 2005 9:18:58 AM
Be aware: less-than-half-baked thoughts follow.
It seems to me that there are distinctions to be made between capital gains and abstinence out of income.
If the discount rate fell (and were to be sustained at the new, lower level), then capital gains would immediately be evident. However, there would be no immediate change in the real stock of productive capital or savings.
If all future oil exploration failed to turn up any new deposits then the understood scarcity of oil would increase and existing holdings would experience capital gains althougth the world would be understood to be less well-off then previously believed.
Where capital gains and abstinence out of income would seem more analytically similar would be situations where the capital gain is the result of an increase in productive capacity (e.g. availability of improved irrigation techniques rendering infertile land suitable for farming and hence more valuable, even though it is not yet farmed).
Posted by: Brent Buckner at Jun 9, 2005 9:20:24 AM
Tyler, in the entry that followed this one, you prescribed "Make them read
Adam Smith." Would Adam Smith say that a relative increase in the price of
a house represented an increase in the wealth of a nation? What did he say
the wealth of a nation was?
Posted by: Mark Shroder at Jun 9, 2005 9:46:05 AM
The problem is that mapping between capital and real assets can't capture important attributes.
Capital can "spoil". Capital can "ripen". Capital can "gather dust". Capital can be "targeted". Capital can be "protected".
My take on capital? "Footprints" and "smells" that allow for some prediction of where the economy is and where it may go.
Posted by: Huggy at Jun 9, 2005 10:00:30 AM
Forget for a moment "passive" capital gains. Consider building a swimming pool in your backyard. This is an investment, adding a swimming pool to your portfolio of assets, just like adding some IBM stock to your investments. Is it savings? Of course it is. The fact that you can swim in it does not make it cease to be savings, just as voting your IBM shares does not make them not be savings any more.
So then, it would seem logical that this investment, like any other, can appreciate or depreciate. The change in the value of your house or your swimming pool is exactly analogous to a change in the value of your IBM stock, and should be accounted in the same manner.
Sur, it falls a mit more toward the solid end of the continuum of liquidity. But unless you've got a really clever place to draw a line, that's no reason to exclude it.
Posted by: Grant Gould at Jun 9, 2005 10:46:18 AM
Continuing in the theme of half-formed, and possibly misguided thoughts...
A swimming pool, like housing, is different from IBM stock because it's a durable consumption good rather than, or as well as, an investment. As Richard so insightfully pointed out, price increases in durable consumption goods increase both consumption and assets (for owner-occupiers their own consumption increases, for landlords, their tenants' consumption increases, but the net effect on savings is the same) - so it's not clear how more saving is actually occurring.
However, it might still be possible that falls in measured savings rate could either:
(a) understate savings - if it counts one side of this ledger (increased consumption of housing services) and not the other (house price increases); or
(b) misrepresent the extent of the problem, because if all of the increase in PV wealth is being consumed, the savings rate will fall (rise) if there is net positive (negative) savings to begin with, despite the fact that absolute savings hasn't changed.
My other question would be what impact Richard's point has on the wisdom of owner-occupiers consuming out of House price increases, if they're not planning on cashing out and moving somewhere cheaper in the future?
Posted by: conchis at Jun 9, 2005 11:26:46 AM
Don Loyd and Richard Squire seem to have summed up the essential situation.
Of course, if rent prices *are* rising rapidly, that counts as inflation and should discourage savings by rational investors.
It looks to me like official inflation numbers are extremely unrealistic.
Rent is ~ 25% of GDP and rising at ~6%/year
Petrochemicals are ~5% of GDP and rising ~20%/year over the last 5 years.
Medical is 15% of GDP and rising at >10%/year
University Education is 3% of GDP and rising at ~7%/year
R&D costs are rising, GVMT spending is rising...
Real inflation must be at least 4% for all practical purposes.
Mark Schroder makes a great point about Smith. Endlessly insightful. Asking WWASS should be a reflex. Actually, making it one is probably a good part of the point of a liberal education.
Brent Buckner's substantive argument is correct and important, but I regret the economical terminology. Discount rates, while an established part of economic theory, appear to me to be a sleight of hand whereby normative concepts are substituted for descriptive. Behavioral economics has disproved beyond serious debate the hypothesis that real humans behave as if they had consistant discount rates, and the idea is kept around by financial economists in order to patch the countless empirical holes in theories of market aggregation of information, holes that can actually be patched very easily within the paradigms of utility theory or behavioral economics.
By the way, swimming pools don't statistically increase the value of a house, and do decrease liquidity. Technically, something's not an investment if expected return is negative unless it correlates negatively with the market aggregate.
Posted by: michael vassar at Jun 9, 2005 11:47:31 AM
From a policy perspective -- ie, I think low national savings is bad and wish to propose policy to change that -- what would seem to matter is households' perception of whether capital gains is savings. I would argue that they will perceive such gains to be savings. At the end of the year, they see that the balance in their retirement account has increased significantly; they see that they own a larger fraction of their home than their principal payments can account for; it seems inevitable that they will perceive that they have saved. Policy to increase actual savings (ie, decrease current consumption as a fraction of income) must overcome that perception. I think perhaps a more fundamental question that must be answered before taking on "Are capital gains savings?" would be "Are capital gains income?"
Posted by: Michael Cain at Jun 9, 2005 12:05:26 PM
Picking up on Richard's and particularly Conchis' comment, most people
own only the house or condo/coop they live in, so if they were to sell
their house, they'd have to buy an equally expensive alternative one,
or actually suffer lower consumption by moving to a smaller house or
a less desirable location. So how are they really made richer by
the nominal appreciation in the value of their domiciles?
And what is the average person doing with their increased nominal wealth?
Not, I would venture to guess, in capital expenditures to increase wealth
creation, but in consumption, consumption, consumption. One possible
exception--education costs for kids.
Greenspan, based on long personal interest in the role of household equity
on the economy, engineered the classic solution to preventing a depression:
he fooled people into thinking they were richer than they were so they would
keep spending. The classic solution is just the printing of money, but maybe
after decades of doing that, people can't be fooled that way any more.
Posted by: Cleon at Jun 9, 2005 12:13:03 PM
The question of housing assets being so difficult, I would rather concentrate on the other piece of the savings puzzle that Tyler mentioned - does expenditure on education count as savings? Education, especially at the post-secondary level, appears to be a combination of consumption (some classes are enjoyable but do not increase employability and future earnings) and investment (students obtain credentials and knowledge that increase their employability, and lead to higher lifetime incomes). It seems then that a substantial proportion of education should count as savings, to the extent that it leads to higher future levels of consumption (mostly of services) by the individual and society as a whole than would be possible without it. Thus, if Americans spend four percentage points of GDP more on post-secondary education than Europeans do, I would have to guess that at least two percentage points of that represent additional savings that are not missed in the official savings rate.
Posted by: Peter Fitton at Jun 9, 2005 1:07:37 PM
Isn't the entire issue one of
stocks vs flows.
You have a stock of capital assets
and you can add to that stock out
of current income -- the savings flow --
or reduce the stock to spend --dissavings.
It is like the comparison of wealth and income.
They are different things and the current
discussion is irrelevent and just
obscures the real issues.
The real issue is that
as a nation we are eating our seedcorn to
finance current consumption.
Maybe the real question is why are some
people trying to obscure the real issues?
Posted by: spencer at Jun 9, 2005 1:32:02 PM
Another half-awake thought, playing off the excellent points of Michael, Michael and Brent:
Side-stepping the argument of liquidity, consider that a house used as a house offers utility both in present and in future, and can increase offerings of utility without amending the asset itself (a Beckerian way of saying "the kids like the house too, and the more there are of them, the more utility is additive").
However, a house purchased with intent to re-sell for gain in the future is really an asset intended for intertemporal arbitrage; futures purchased at the Chicago Merc Exchange are an example of intertemporal arbitrage between markets with distinct perceptions driving a difference in expected value. Likewise, cigarettes purchased in Missouri and re-sold in Illinois (to take advantage of the $15/carton price difference) do not contribute to savings since the increase in value is only due to price and is illusory.
Consider it this way: say the tabulators of national accounting consider homeowner equity as savings, but housing prices decline. Then all house owners technically incur negative savings, which would also be illusory since a negative savings implies an increase to consumption. If housing prices drop, there is no correlated increased consumption; the illusion stems from the now top-heavy houses (since the value of the mortgage debt would remain constant but the value of collateral would decrease, and that liability on the right side would have to be offset by an asset on the left side, in this instance depreciation).
Thus, housing does not behave consistently with the concept of savings, both because it does not reliably store value (as when a housing market crumbles) nor does it increase value (except through arbitrage).
Posted by: Jeremy at Jun 9, 2005 1:56:19 PM
I think one of the four important points has been made above...
1) a house does not represent savings if your intention is to use those savings to buy another house. The argument has been made, but think of two identical houses next door to each other. If $100K can buy the one today and in 20 years that house is worth $250K, then there is no appreciation relative to the house next door which is also worth $250K. So you have to downsize in order to extract "savings" or you have to die (in which case a negative mortgage can be used to extract money until that point after which your heirs can profit).
2) You have to pay interest on your leverage in a house. A house only can serve as savings (given the above constraint) if the return on the capital appreciation exceeds the interest paid on the leverage;
3) A house's appreciation can only count as savings if (given the above two constraints) the total return on capital invested is higher than the average inflation on the basket of goods on which you plan to spend the "savings" when extracted.
4) Major shifts in demographics might compromise the ability to extract any savings using the negative mortgage option described in point number 1.
Given this, I think a weak argument could be made that it is savings, but a formula for the calculation of that across the nation would be difficult if not impossible.
Posted by: Dylan Barrell at Jun 9, 2005 2:23:18 PM
As to point 1, Why is a house not savings if you plan to buy another house? That makes no sense. If you didn't have the first house, you'd have to get the money for the second house from somewhere else. It doesn't matter that the first house has or has not appreciated relative to other houses.
As to point 2, Savings that lose value due to inflation still count as savings.
Richard Squire's explanation is interesting but by that standard, cash in the mattress or piggybank, gold, art, and other collectibles are not savings either.
Posted by: David at Jun 9, 2005 2:54:34 PM
If savings is simply the annual change in one's wealth - of course, capital gains is a form of savings. OK, this is an accounting truism but don't accoutants count R&D as an expense? It is a cash outflow which hopefully will generate future benefits. And the market price of a company's stock reflects its best expectation of the present value of these benefits. But now I'm getting close to answering Kevin Drum's question as to how I would change the tax code: tax all income when it accrues at the same rate regardless of source. Which means: (a) allow R&D to be immediately deducted, but (b) tax all capital gains as soon as they are accrued.
Posted by: pgl at Jun 9, 2005 4:15:31 PM
I'll leave comments on housing values to others, but it seems obvious to me that abstinence from spending all or part of your capital gains on equity investments does constitute savings (above and beyond mere price level changes). My income certainly consists of income earned from employment and earnings from past investments. Dividend yields are relatively low and I do expect the majority of my return to come from share appraciation.
I hope soon to provide most of my income from investments. Looking back, I could have reached that goal in a variety of ways: saving much at an early age, and letting compound interest carry me forward, saving continuously over my working career, or starting late and drastically cutting back my expenditures. At least in the first case, if you ignore savings from capital gains, I would appear to be saving very little now.
Posted by: Rich Berger at Jun 9, 2005 4:34:10 PM
Tyler, education is NOT an asset, you can't sell it to someone, there's no guarantee that it will even help you earn a higher salary.
Posted by: Half Sigma at Jun 9, 2005 5:41:28 PM
If saving is just avoiding or postponing consumption of present income, and savings is the accumlated sum of all acts of saving, then. . .
The negative of the sum of all historical house-related cash flows (in today's dollars)--including mortgage payments, down payment, closing costs, maintenance, taxes and tax credits, etc.--PLUS the sum of historical rent & renter's insurance for an equivalent property over the same period (in today's dollars), EQUALS the amount you have "saved." If this number is zero or less, you were not saving for some or all of the time you lived in the house.
The price you sell the house for is the value of those savings. This value may be less than or more than what you saved.
You could do a similar exercise with education, but it's much more difficult to get a good estimate of your opportunity cost.
Posted by: David at Jun 9, 2005 10:52:13 PM
It's important to draw a distinction between savings on a personal level (i.e., an increase in net worth) and savings on an economic level (i.e., an increase in capital stock). The reason we worry about low savings is not that we're afraid that people won't be able to afford luxurious retirements; it's because we worry that there won't be enough capital to fuel future growth. So it's economic savings, not personal savings, that we want to increase.
Home equity can be considered savings on a personal level, but it generally doesn't contribute to the capital stock. If I buy a $400,000 house instead of a $300,000 house, that doesn't increase anyone's productive capacity; it just means that I'll probably enjoy the house more.
Education can be considered as adding to the capital stock to the extent that it facilitates future production. Of course, this varies greatly, depending on the field of study and the aptitude of the student, among other things. A degree in electrical engineering might be considered an increase in the capital stock, while a degree in, say, art history, might not. Degrees in other fields might actually hinder future production.
To count unrealized capital gains on equities seems reasonable. If you took the money out of the equities market, that would drive down prices and make it harder for firms to raise money. It's absention in the sense that you could take your profits and buy consumer goods, but don't.
Posted by: Brandon Berg at Jun 10, 2005 1:10:42 AM
Brandon writes: "The reason we worry about low savings is not that we're afraid that people won't be able to afford luxurious retirements; it's because we worry that there won't be enough capital to fuel future growth. So it's economic savings, not personal savings, that we want to increase."
Really? I'm worried about both. Presumably we care about growth because we're worried about our future well-being (retirement or otherwise), rather than as an end in itself. So aren't these kind of the same issue?
Posted by: conchis at Jun 10, 2005 10:06:31 AM
