Central to arguments over macroeconomics—particularly concerning the role, if any, that the government or central bank should take during a deep recession—is the connection between saving and investment. For example, if most people in the community decide to “tighten their belts” and save more, will that help or hurt a depressed economy? Both classical and more modern, market-oriented economists tend to think saving is a good thing, particularly in the wake of a “profligate” but unsustainable boom.
On the other hand, Keynesian economists stress the so-called “paradox of thrift,” in which the community as a whole can’t cut spending in order to dig out of a financial whole, because one household’s spending is another household’s income. Thus, what makes sense for an individual doesn’t (necessarily) make sense for the group. (The Keynesians thus think that saving—at least in a depressed economy with insufficient Aggregate Demand—is an example of the “fallacy of composition,” similar to the situation where one person standing on his seat at a concert can give him a better view, but if everybody stands on his or her seat, nobody is better off.)
In these arguments over the market economy’s behavior during a recession, a key empirical dispute is the extent to which extra saving by households will be channeled into extra investment spending by businesses. Classical and modern market-friendly economists (such as the Austrians) tend to think that the households’ savings will be channeled (perhaps through banks and other intermediaries) into investment spending, while Keynesians ask, “Regardless of how low interest rates go, why would a business borrow more when the demand for its products is slack? Why build up more capacity when inventory is already accumulating in the warehouse?”
Against this backdrop, I remember being very surprised when I first read John Maynard Keynes’ General Theory, in which he argued that saving equals investment necessarily. Keynes used accounting definitions in order to derive this result; he wasn’t appealing to economic models of the credit market.
At this point, I realized we had two different conversations occurring. On the one hand, people were arguing about S=I in terms of macroeconomic theory, while on the other, they were appealing to the certainty of that relationship via accounting tautologies.
In this two-part series, I will first lay out the accounting approach, in which S necessarily equals I, for any time period and with no lag. Then, in the follow-up post, I will juxtapose this accounting tautology with the economic policy controversies, to understand how (if at all) the two approaches intersect.
As my guest Charles DuBois lays out in InFi episode #94, we can establish the identity S=I using standard definitions from economic theory and national income accounting. First we have that
SAVING = INCOME – CONSUMPTION
Few economists would quibble with this claim. It is the very definition of saving.
But we can also say with confidence that
SPENDING = INCOME
because any individual’s income must be originating from someone else’s spending.
Now we have
SAVING = SPENDING – CONSUMPTION
through substitution. Next consider the obvious fact that
SPENDING = SPENDING ON INVESTMENT + SPENDING ON CONSUMPTION
since there are no other possibilities.
Then this gives us
SAVING = (SPENDING ON INVESTMENT + SPENDING ON CONSUMPTION) – CONSUMPTION
This of course leads to our desired result, namely that
SAVING = SPENDING ON INVESTMENT
or simply
S = I.
Related to this discussion is yet another accounting tautology, which says that aggregate financial saving must equal zero. Once we realize this—and I will spell it out in a moment—we can easily see, once again, that Saving = Investment in the aggregate. This is because:
TOTAL SAVING = FINANCIAL SAVING + NON-FINANCIAL SAVING
In this context, financial saving means spending money to accumulate financial assets, such as stocks or bonds, while non-financial saving means spending to accumulate physical assets such as a factory or a hammer.
However, for the economy as a whole, aggregate (or net) financial saving must equal zero, because anybody’s asset (if it takes the form of a financial asset) must correspond to someone else’s liability. If a household uses some of its income to acquire a corporate bond, then the household has definitely saved, but the corporation has dissaved by the same amount. (To see these ideas spelled out, try this intuitive article or, for the advanced reader, this detailed exposition.)
So if the reader can see the logic here, and understand why total financial saving must equal zero for a closed system, then we have:
TOTAL SAVING = NON-FINANCIAL SAVING
which we know also means
SAVING = SPENDING ON INVESTMENT GOODS
or simply
S = I.
That’s a good spot to wrap up this introductory post. See the hyperlinks I referenced above to dive deeper into the framework. In the follow-up post, I will relate these accounting tautologies to the macroeconomic theories that employ similar language. As we’ll see, much of the controversy in this space revolves around the inconsistent use of terms, flipping from one framework to the other.
Dr. Robert P. Murphy is the Chief Economist at infineo, bridging together Whole Life insurance policies and digital blockchain-based issuance.
Twitter: @infineogroup, @BobMurphyEcon
Linkedin: infineo group, Robert Murphy
Youtube: infineo group
To learn more about infineo, please visit the infineo website