Does Saving Cause Lending Cause Investment? (No.)

The idea that household saving funds business investment is both incoherent and empirically wrong.

Share with your friends

More share buttons
Share on Pinterest

By Steve Roth

Households save money and lend it to businesses, who invest it in productive enterprises. That’s the economic story you’ve been hearing your whole life, right? Or at least since Econ 101. Saving funds investment. That core idea is embedded (and unquestioned) in modern macro: the Solow growth model, IS/LM, the lot.

Put aside that the basic bookkeeping of this idea — that personal saving creates “savings” that “fund” lending and investment — doesn’t make any sense. (It’s an error of composition; you have more savings if you save, but the economy doesn’t.) Let’s look at history: when households save more, is there more lending and (business) investment — either immediately or a few quarters/years down the road?

Get Evonomics in your inbox

Mostly: no.

Here are some saving, lending, and investment measures for the U.S., post-war (starting Q1 1948), all divided by GDP to make them comparable:

The first thing to notice here: both lending and investment are vastly larger than household saving. They can’t be “funded” by that saving, or at least not much. (Think: bank lending and endogenous money.)

Otherwise, eyeballing this, it’s pretty much impossible to tell if these measures are correlated. When saving goes up or down, do lending and investment do likewise (concurrently, or some time later)? They’re all over the place. So let’s use software to look at correlations between them.

A correlation of 1.0 means the two measures move perfectly together — if X goes up, so does Y, by an equal amount. A correlation of -1 means they move perfectly in opposite directions — if X goes up, Y always goes down by an equal amount.

Of course, correlation doesn’t demonstrate causation. But lack of correlation, and especially negative correlation, does much to disprove causation. What kind of disproofs do we see here?

• Personal saving and commercial lending seem to be lightly correlated. The correlation declines over the course of a year, but then increases two or three years out. It’s an odd pattern, with a lot of possible causal stories that might explain it.

• Personal saving and private investment (including both residential and business investment) are very weakly correlated, and what correlation there is is mostly negative. More saving correlates with less investment.

• Commercial lending has medium-strong correlation with private investment in the short term, declining rapidly over time. This is not terribly surprising. But it has nothing to do with private saving.

• Perhaps the most telling result here: Personal saving has a significant and quite consistent negative correlation with business investment. Again: more saving, less investment. This directly contradicts what you learned in Econ 101.

• The last line — commercial lending versus business investment — is most interesting compared to line 3 (CommLending vs PrivInv). Changes in commercial lending seem to have their strongest short-term effects on residential investment, not business investment. But its effect on business investment seems more consistent and longer-term.

I’ll leave my gentle readers to ponder those only-somewhat stylized facts. Here’s the spreadsheet for any who want to explore further.

2017 January 30

Donating = Changing Economics. And Changing the World.

Evonomics is free, it’s a labor of love, and it's an expense. We spend hundreds of hours and lots of dollars each month creating, curating, and promoting content that drives the next evolution of economics. If you're like us — if you think there’s a key leverage point here for making the world a better place — please consider donating. We’ll use your donation to deliver even more game-changing content, and to spread the word about that content to influential thinkers far and wide.

 $3 / month
 $7 / month
 $10 / month
 $25 / month

You can also become a one-time patron with a single donation in any amount.

If you liked this article, you'll also like these other Evonomics articles...


We welcome you to take part in the next evolution of economics. Sign up now to be kept in the loop!

  • annamarie

    How do you address the international aspect of capital markets?
    Economic theory only teachers that savings and investment should be strongly correlated in a closed economy. In an open economy like the US where capital is mobile, there’s no reason to believe that savings and investment should be correlated. The puzzle is more savings and investment continue to be correlated so much in open economies (Feldstein-Horioka puzzle – – not that savings and investment aren’t correlated enough.

    • Marc Thornton

      Whenever someone brings the Feldstein-Horioka puzzle up, I like to point to this seminal contribution by Claudio Borio and Piti Disyatat: “Capital flows and the current account:
      Taking financing (more) seriously”:

      They explain in a language which is accessible to a trained economist that the FH-puzzle is a misconception. It does not exist.

      So why does the FH-puzzle not exist? From the abstract of an other paper:
      “This paper makes the case that a large part of the discussion about the Feldstein-Horioka puzzle is beside the point. On the one hand, the meaning of “saving” in theoretical models and its relation to statistical data often is not made clear, which leads to wrong conclusions when interpreting data and empirical evidence. On the other hand, in nearly all contributions to the debate there is no differentiation between “saving” and “financing”.

      Source: Solving the Feldstein-Horioka puzzle by Johannes Schmidt –

      Unfortunately both papers do not seem to be well known and we still have dozens of economist producing FH-regressions accompanied with explanations for their (supposed to be) puzzling results.

  • Colin Lloyd

    Dear Steve,

    An interesting observation indeed. I wonder whether this might be due to two (or more) factors:-

    1. The time lag between saving and investment
    2. The effects of fractional reserve lending, which may entirely eclipse the impact of saving on investment, especially in a fiat currency world.

    Many thanks for your thought provoking article. As soon as I read it I was stuck by the fact that I had observed this but not considered the disconnect between theory and practice.

    As Annamarie commented a few hours ago, the international effects may be a direct source of this lack of cause and effect. I recall the Asian crisis in the 1990’s being fueled by what Brian Reading described as “Up-hill capital flows, from the US to Asia”.

    My guess is, even if we could find reliable global data, the correlations would be at best unstable and more likely negligible.

    A final thought. The negative correlation observed between savings and business investment. I wonder to what extent this is due to business leaders response to central bank policy. When savings decline and unemployment increases, CBs normally adopt counter-cyclical policy to soften the blow of recessions. Firms, anticipating this, take advantage of lower interest rates to invest in expectation of the next recovery.

    Thanks again. Most stimulating.

  • Steve, this is an extraordinarily fascinating and provocative piece. Sure as heck caught my attention. Violative of much I’ve assumed. (And many others, I’m sure.)

    On examining it, I concluded that the conundrum is due mostly to the difference between the BEA definition of Personal Saving and “savings” as understood in the Solow etc. framework. By way of specific example, Gross Private Domestic Investment includes undistributed corporate profits. These are profits that have not been returned to shareholders/securities owners and instead are used for reinvestment. BEA does not include such in Personal Saving.

    But in a private ownership economy, households are the ultimate owners of assets operated by firms. By allowing these firms to forgo dividends/distributions in the current period, such households are effectively saving. These savings (forgone consumption) help fund investment indirectly.

    We could talk more generally about the other investment metrics showing in your superb graphic, but my deep sense is that the savings/investment identity remains fundamentally intact (time lags pointed to by Colin Lloyd aside) and that all investment in the economy ultimately derives from household savings in this larger sense. The alternative feels like a violation of some physical conservation principle.

    What a stimulating essay, Steve. Thank you.

  • Tyler

    You appear to have ignored the impact of international capital flows. America invests more than it saves and borrows the difference from abroad. Both our savings and investment rate have been falling since the 1980s with a slight rebound recently as the economy recovers from the financial crisis.

    Also, there doesn’t appear to be much of a global savings glut anymore. Global savings spiked for a couple years in 2005-2008 but we’re back to historic norms. The world seems to save around 24% of global world product every year, plus or minus 1 percentage point.

    • Tyler, do you happen to know how this 24% compares to global investment rates?

      • Tyler

        By definition, they are equivalent. If there is a high rate of savings and people are bidding up the price of existing assets, the real rate of return on capital would fall. No?

        So we really should look at whether the real rate of return is rising or falling.

        • Thanks, Tyler. Yes, it would fall in that case. In my picture of the world, the real rate of return to capital depends primarily on the relative supply of capital vs. labor. If capital is scarce, real returns to capital will rise and wage rates will be depressed (think early industrialization in England for the horrifying extreme of this); if labor is scarce, the opposite will happen. The dynamics of income equality/inequality, in other words.

          But I agree that the definitional equivalence of savings vs. investment rules the actuality, at least as I see it. I suspect Steve’s analysis somehow obscures this, though the obscurity is exceptionally subtle.