Note To Economists: Saving Doesn’t Create Savings

Is Saving a Sin?

If you save more, if everybody collectively saves more, there are more savings, right? There’s more money that firms can borrow and invest to make us all more prosperous. Household saving “funds” business investment, so if we all save more, the world will be more productive and prosperous. You hear this all the time. And it makes sense, right?

Wrong. It’s hogwash. Incoherent codswollop. Gobbledegook faux-accounting-think. Bunkum. Think: fallacy of composition — believing something is true of the whole because it is true of the parts.

You may know the much-discussed paradox of thrift: If everybody saves more, there’s less spending, so less income, less aggregate demand. On the surface, at least, that seems rather straightforward. But the real paradox here lies in the plural: the stock of monetary “savings” that lurks, implicit, at the core of the paradox. If we save more, are there more savings?

Start by knowing this: “Savings” (the plural) is not a measure in the national accounts, even though economists and commentators use the word ubiquitously. (Think: Bernanke’s “global savings glut.”) Search the Fed’s quarterly Z.1 report, the Financial Accounts of the United States. You’ll see. There are various measures of “saving,” with various accounting definitions — “flow” measures — but there’s no named measure of our collective stock of savings. (This makes some “stock-flow consistent” models somewhat…problematic.)

Then ask yourself:

When you spend money — transferring it to someone else in return for newly-produced goods and services — does it affect our collective monetary savings? In strict accounting terms, obviously not. Your money just moves from your account to someone else’s account; it doesn’t disappear. Your bank has less deposits; the recipient’s bank has more deposits. Aggregate monetary savings is unchanged by that accounting event. (The economic effects of that transaction — what behaviors it triggers — are another matter entirely.) One person’s spending is another person’s income. And vice versa.

But what if you don’t spend? You “save” instead, leaving the money sitting untouched in your account instead of transferring it to somebody else’s account. Does that increase aggregate monetary savings? Even more obviously not. That “act” of saving (not-spending) is quite literally a non-event.

In the simplest accounting terms, household monetary saving is just a residual measure of two flows — income minus expenditures. It’s perfectly understandable on that individual, micro level of a household. Less so in the aggregate. Because in aggregate, income = expenditures. And since saving = income – expenditures, saving must equal zero. What’s with that? (Note that in the stylized world of the National Income and Product Accounts, households only consume; they don’t invest. All household spending is consumption spending. Only firms invest, and all their spending is investment spending.)

So how does saving actually work? What does it mean to “save” — as an individual or a household, as a country, or as a world? Start with individuals, and the vernacular understanding of “household saving.”

You work your whole life, spending somewhat less than you earn (“saving”), leaving the residual sitting in your checking account. Maybe you swap some of that checking-account money with others in exchange for a deed to a house, or a portfolio of stocks and bonds, which go up in value over the years. Eventually you retire and live off that stock of “savings” (plus ongoing returns from those savings). In that everyday, individual context, savings (the stock) means “net worth” — your balance sheet assets minus your balance sheet liabilities. When you hit retirement, net worth — your savings — is the financial indicator that really matters. Bottom line: “How much money do you have?”

But what about our collective monetary savings — the stock measure that’s missing from the national accounts? That’s also best represented by aggregate household assets, or net worth. (For a sector with no externally-held assets or liabilities, assets and net worth are the same. For the world, assets equals net worth. We don’t owe anything to the Martians. The righthand side of the world balance sheet is all net worth.) For reference, U.S. household assets are about $101 trillion. Net worth is about $87 trillion. The household sector owes about $14 trillion to other sectors. (Here.)

It’s important to remember: households own all firms, at zero or more removes. A company can be owned by a company, which can be owned by a company, but households are the ultimate owners. (Firms’ liabilities are netted out of their net worth, by definition.) This because: Households don’t issue equity shares — their liability-side balancing item is net worth, not shareholder equity. Firms don’t own households. (Yet.) Companies’ net worth is telescoped onto the lefthand, asset side of household balance sheets. So household net worth = private-sector net worth. When it comes to tallying up private asset ownership — claims on existing goods and future production — the accounting buck stops at households.

The monetary measure “household net worth” is national accountants’ best effort at tallying up the markets’ best estimate of what all our real stuff is worth, in dollars. (More precisely, what all the claims on those goods are worth.) It’s far from a perfect measure; its relationship to government net worth, for instance (if that’s even meaningfully measurable), is decidedly iffy. See in particular J. W. Mason’s article on Germany’s uncanilly low household net worth. But it’s pretty much the best, maybe the only, measure we have.

So if monetary saving doesn’t increase the stock of monetary savings, how do we “save,” collectively? By producing long-lived goods — goods that we don’t consume within the accounting period. Machinists create drill presses, carpenters create houses, inventors create inventions, businesspeople create companies, economists create textbooks (yeah, I know…), teachers and their students create knowledge, skills, and abilities. All that tangible and intangible stuff that we can use and consume in the future constitutes our collective “real” wealth.

The financial system creates claims on all that stuff (and on future production), in the form of financial instruments — from dollar bills to checking-account balances to deeds on houses to collateralized debt obligations. The markets assign and adjust dollar values for those claims. When you hold those instruments, you’re holding a promise that you can purchase and consume real goods in the future. They’re claims (again at zero or more removes) on existing goods and future production. The markets constantly adjust those instruments’ prices/values based on our collective expectations of future production — our optimism/pessimism, or “animal spirits.”

With that as background, here’s the crux of the “saving” problem: Economists confuse saving money with saving corn. They conflate stocks of money (claims on stuff) with stocks of stuff. Think: “financial capital.” It’s an oxymoron. Capital is real stuff — despite Piketty and others’ inconsistent and self-contradictory use of wealth and capital as synonyms. See for instance, “capital is imported (net) to fund the trade deficit,” here. (And again, see J.W. Mason on this conundrum.)

This confution of monetary and real saving — financial instruments versus real capital — is a problem because money/financial instruments are nothing like real goods. These promises, or claims, are created with zero resource inputs to production. (Promises are cheap — actually, free.) And they are not ever, cannot be, consumed or used as actual inputs to production. (You can’t eat promises, or feed them into an assembly line.) That stock of dollar-designated claims, monetary wealth, is simply created and destroyed — expanded and contracted — by the creation/destruction of financial instruments, and their repricing in the markets.

Consumption reduces our stock of stuff. If we eat less corn, we have a larger stock of corn remaining. Consumption spending doesn’t reduce the stock of anything. If we spend less money, we still have the same stock of money.

Corn is produced and consumed. Financial/monetary wealth — the netted-out, dollar-denominated value of our web of promises and claims — simply appears and vanishes. That’s the magic of this social-accounting construct we call money.

A semi-aside on the accumulation of real, long-lived goods, real wealth: There’s another widespread though often-implicit logical accounting error that merits enthusiastic eradication. Starting with accounting definitions: (C)onsumption spending is paying people to produce goods that will be consumed within the accounting period. (I)nvestment spending is paying people to produce goods which will exist beyond the accounting period. C + I = Y (GDP, or total spending).

The error: More consumption spending means less investment spending — less accumulation of real goods, real wealth. Right? Wrong. That thinking assumes Y is fixed, which is only a given in accounting retrospect. If we spend less on consumption goods, we might just spend less, total — less Y, with no effect on investment spending. Obvious behavioral/incentive thinking actually suggests even worse: if consumers spend less, firms will do less investment spending. Both categories of spending, and total Y, will be lower (relative to a counterfactual of more consumption spending).

So what are the modern mechanisms of this relationship, between monetary savings and real goods? How do we collectively “monetize” our ever-increasing stock of real goods, our “real” savings, to create monetary savings? Three ways (none of which is “personal saving”):

1. Government deficit-spends money into existence. Treasury simply deposits dollars, created out of thin air, into private-sector checking accounts, either as transfers or in return for goods and services. This increases both private-sector balance-sheet assets, and private sector net worth — because no private-sector liabilities are created in the process. (Treasury then selling bonds, and the Fed buying them back, doesn’t directly affect private-sector assets or net worth. It simply swaps asset for asset, and alters the private sector’s portfolio mix, bonds versus checking-account deposits — though again with potential carry-on economic effects.)

2. Banks issue new loans, dollars that are also created ab nihilo. This loan issuance increases private-sector assets, but the act of lending/borrowing itself does not increase net worth, because borrowers simultaneously takes on liabilities equal to the new assets. New loans from banks only increase net worth if the leveraging later pays off (an economic effect), via mechanism #3.

3. As we create more stuff and decide that existing stuff is worth more, the financial system creates new financial instruments, and the existing-asset markets bid up prices of existing instruments (stock shares, deeds, etc.) — expanding the stock of claims to approximate the expanded stock of stuff. Market runups increase household balance-sheet assets, with no increase in household liabilities. So like government deficit spending but unlike bank lending, market runups increase household net worth. Voila, households have more money. This is arguably the dominant financial mechanism for “money printing.” (This reality highlights the pervasive “conservation of money” fallacy that still plagues even much “stock-flow consistent” thinking, a fallacy that’s beautifully explicated in this paper by Charlotte Bruun and Carsten Heyn-Johnsen. It also points out the deep conceptual problems of “income” measures that don’t include capital gains income — where “saving” doesn’t equal change in net worth.)

That Econ 101 circular-flow diagram might need some rethinking.

Over the long run, our stock of real goods and our stock of dollar-valued claims on those goods (tallied as balance-sheet assets or net worth) go up roughly together. But it’s a very long-run thing, subject to variations spanning decades, even centuries, and contingent on shifts in societal attitudes, cultural norms, institutional structures, political power, monetary policy regimes, beliefs, geopolitical forces, environmental exigencies, and technological disruptions, among other things. If Piketty’s Capital depicts nothing else, it depicts that reality.

Many things affect our collective saving and savings, real and monetary. But one thing is sure: the non-act of personal monetary saving does not increase our collective monetary savings. Personal saving does not create funds (much less “loanable funds”). That notion is incoherent, in simple, straightforward accounting terms.

Rather, the economic effects work like this: more spending causes more production (incentives matter, right?), which creates more surplus and stuff — both long-lived and short-lived — more value. The value of the long-lived stuff is then monetized by the government/financial system through the creation of new dollar-denominated financial instruments/legal claims, and price runups on existing instruments/claims.

In three simple words: spending causes saving. Real, collective accumulation of real, long-lived stuff. Monetary saving — not-spending part of your income this year — doesn’t, collectively, create either real or monetary savings.

Maybe the Demon Debt is not the Great Evil after all. Maybe Selfish Saving — hoarding of claims against others — is actually the greater economic sin.

Cross-posted at Evonomics.