It is not necessary that the bank hold excess reserves for the effect of increased saving to be lost to investment. All that is necessary is if reserves are required against relatively inactive time (saving) deposits as well as against much more highly active demand (transactions) deposits.

If the bank uses some of its demand deposit liabities to purchase extant bonds rather than to finance investment spending, and the seller places the proceeds of the sale into a time deposit (which may be the next best investment for the seller if she wishes to alter the composition of her investment portfolio rather than requiring money to spend), then the bank is required to hold reserves against these time deposits. This means that some of the nation's money supply (defined as cash plus active demand deposits; i.e., actual means of payment) is sterilised and aggregate spending falls. The intermediation of savings to investors is blocked and a mere change in the form in which savings are held prevents some of them from remaining in circulation.

This process was examined by Lauchlin Currie, _The Control of the Supply of Money in the United States_ (Harvard UP, 1934) and in his subsequent memoranda to Marriner Eccles when he was Eccles's chief adviser at the Fed, 1934-39. His policy advice (not taken up) was that reserves against time deposits be abolished and 100% reserves be imposed against demand deposits in order to strengthen monetary control and avoid a situation in which a change in the ratio of cash to deposits alters the money supply inappropriately. (One of the reasons for the calamitous drop in the money supply, 1929-32, was an increase in the public's preference for cash, which the Fed failed to offset.)

- Roger Sandilands


________________________________________
From: Societies for the History of Economics [[log in to unmask]] On Behalf Of Robert Leeson [[log in to unmask]]
Sent: Tuesday, October 18, 2011 4:46 AM
To: [log in to unmask]
Subject: [SHOE] S = I?

Keynes' (GT 1936, 81, 83) alleged "optical illusion" of intermediation appears to be an elementary but fundamental error: "It is supposed that a depositor and his bank can somehow contrive between them to perform an operation by which savings can dissapear into the banking system so that they are lost to investment ..." One counter example to Keynes' assertion is excess reserves.

Keynes also asserted that if banks use the deposit to buy a second hand bond the bond seller would be "dis-saving ... he must be spending it on current consumption in excess of current income ... It follows that the agregate saving of the first individual and of others taken together must necessarily be equal to the amount of current new investment."

Yet no new net investment would take place (in the relevant time period) if the bond seller used the proceeds to

1. hold money for speculative purposes (and the bank held this money demand as excess reserves) or

2. buy another second hand bond, and the seller of that second hand bond used the proceeds to buy another second hand bond and so on throughout the relevant time period.

I would be grateful to be directed to the secondary literature on this topic.

RL