In explaining why it is investment that creates saving, and not the other way around, different economists have offered different ways of looking at the matter. Keynes (see previous posts here and here) distinguished planned (ex ante) from actual (ex post) investment and saving to note the way in which planned saving adjusts to planned investment via changes in income. Kalecki (discussed previously here) emphasized the initiating role of decision making. Firms can decide how much to invest, but not how much profit they will make, because profit depends on the spending decisions of others. Post Keynesian analysis of institutions and monetary operations suggests an explanation summed up with characteristic insight and clarity by Warren Mosler in the short video embedded below. The purpose of the present post is just to flesh out the explanation a little and highlight a few of its implications.
The process is basically as follows:
A firm decides it would like to invest in production. It believes it can do so profitably. But it needs financing. One possibility is to draw upon past savings, whether its own (retained earnings) or the savings of others (e.g. by issuing shares). But this merely pushes the question back a step, because past savings are the result of past income not yet spent. The other possibility is for the firm to borrow from a bank. Let’s say the firm applies for a bank loan.
The bank assesses the loan application. If it shares the firm’s optimism concerning the profitability of the investment, it will extend the loan. As Post Keynesian analysis of banking and finance makes clear, the loan simultaneously creates a deposit. Specifically, the loan is an asset of the bank and liability of the firm; the deposit is an asset of the firm and liability of the bank.
At this stage, investment has not yet occurred. Nor has income been created. The deposit, at this point, does not constitute either income or saving. Income only results from spending on goods or services, and saving is a portion of income. So far, no spending has actually occurred.
Having obtained the loan, the firm is now in a position to invest. It draws down its account in order to purchase investment goods. The bank account of the firm supplying the investment goods is credited. In other words, the investment spending of the investing firm simultaneously creates income and saving for the supplier of the investment goods of an amount equal to the investment spending.
Of course, the supplier will subsequently use some of the income to spend, some to pay taxes and some to save. But the total amount of leakage from the circular flow of income that is generated by this particular act of investment spending is equal at all times to the amount invested. Warren sometimes sums this up as, “Saving is the accounting record of investment spending”. This is the immediate impact. Ultimately, part of the leakage will go to taxes and imports. The marginal propensity to leak to taxes, saving and imports will determine the size of the expenditure multiplier, and so the ultimate impact on income, but it will not affect the amount of leakage.
The causation is clear, and reinforces Kalecki’s point about decision making:
It is also clear that no prior saving was needed to finance the investment. Nor could anybody’s savings in a bank account be the source of finance. Banks do not – and cannot – lend out deposits. It would be a nasty shock to log in to your bank’s website one day to find that the funds you thought were safe in your personal account had gone missing, lent out to somebody else!
For this reason, the notion of a ‘market for loanable funds’ makes little sense. There is no supply of loans independent of the demand for loans. Deposits are created by banks out of nothing (ex nihilo) in the act of lending to credit-worthy borrowers. If, upon lending, a bank finds itself short of reserves, it can obtain them after the fact from other banks or, if the banking system as a whole is in deficit, from the central bank, either by exchanging government bonds for reserves or by borrowing (i.e. incurring an overdraft on its reserve account) at a penalty rate determined by the central bank. The cost of obtaining reserves will be factored in by the bank in evaluating the profitability of the loan, but the bank’s lending behavior is not constrained by reserves.
This also undermines the idea of interest rates being determined through the interaction of the supply and demand for loans. In particular, it undermines the argument that investment will have an automatic tendency to adjust to a level consistent with full-employment output (and full-employment saving) through interest-rate movements.
It is more likely that the interest rates charged on loans can be a consequence, rather than a cause, of the demand for loans. But it is not an automatic consequence. It is enabled by central bank policy. When the economy is strong, demand for loans from credit-worthy borrowers is likely to be high. The prospect arises for banks to charge higher interest rates without dissuading businesses (and also households) from borrowing.
But assuming bank lending is subject to competition, the rise in interest rates will require a supportive central bank policy. Since loans create deposits, there is no dwindling supply of “loanable funds” and excess demand for “loanable funds” to drive up interest rates.
On the occasions when central banks do adopt rentier-friendly policies, it is typically framed as “inflation targeting”. Lifting the short-term policy rate tends to bring about, via arbitrage, a rise in all rates. In this way, rentiers are able to step up their demands when economic activity is strong, aided by central bank policy.