Martin Wolf's recent public adoption of the Positive Money group's full reserve banking policy has led to quite a bit of discussion in the econoblogosphere:
"Strip private banks of their power to create money"
I do not necessarily agree with the full reserve banking position, as I've never read any literature that satisfactorily explains how it would work. So here are some questions that maybe a FRBing advocate would be able to address....follow below the fold for more.
First of all, in order to examine the potential impacts of a full reserve banking system (subsequently referred to as FRB), we must understand the operations of the current institutional system. So let's start there. Here are some fundamental operational facts about the way the system currently works:
1) Banks do not lend out reserves to non-financial institutions. There is never a corresponding debit to any commercial deposit account when a bank expands it's balance sheet through making a loan. This is the mechanism by which banks create their form of money aka bank deposits.
2) The Fed is the monopoly supplier of reserves
3) Fractional Reserve Banking is largely a myth. Especially the 10% reserve requirement meme, there is no evidence for that position. Yes, reserves (more accurately described as settlement balances) increase slowly over time as more and more transactions occur and their value increases. But Reserves are in no way a constraint to bank lending.
4) Before QE, the level of reserves was always the level of required reserves necessary to maintain the Fed's interest rate target and satisfy settlement balance requirements through the payments system.
5) All Govt spending, taxing, and bond issuance is done only through the reserve system.
So how do these facts play out in the operations?
We wont concern ourselves with equity because that's largely irrelevant to the functioning of reserves in this discussion and are beyond the scope of this post. Lets assume that the Fed's target interest rate is 5% (Obviously its not 5% right now because there really is no 10% reserve requirement, but this is the argument that FRBers are making)
Private banks:
$42T in liabilities (bank deposits = private debt)
$42T in assets ($4T reserves, $38T in other assets)
Citizens:
$42T in liabilities (Bank Loans)
$42T in assets (Bank deposits)
Govt:
$18T in liabilities ($4T in reserves and $14T in securities held by the Non-Fed)
Govt dollar assets are by definition infinite.
Here's our baseline. What happens when bank loans increase $1T in a given period (say 1 year)?
Bank loans and deposits will both increase by $1T, and the amount of required reserves would go up by $100B. But where do the extra reserves come from? If the Fed does not supply them through repos and open market operations, then the supply would be less then the demand and interest rates (the price of reserves) would go up from 5%. If the Fed supplied too many reserves, then the supply would be greater then the demand and the interest rate (price of reserves) would go down from the target of 5%.
So these are the operations and accounting of the way the current system operates (minus the fact that the reserve requirement is not really 10%).
Now lets look at the FRB proposal of 100% reserves:
In the Positive Money proposal, instead of bank lending increasing first and the level of reserves second. The reserves would have to come first through deficit spending, and then banks could make loans up to that amount of reserves. This part is not hard to understand. In this framework, bank lending really would be constrained to the level of sovereign deficits (net reserve addition).
My first question is: how would we create the $38T in reserves necessary to match private debt on a 1 to 1 or 100% basis?
QE could only account for a possible $14T in additional reserves. The Fed would buy every TSY security in existence. Which means they would debit $14T worth of securities accounts and credit $14T worth of reserve accounts. After this first step, the accounting would be:
$42T in bank loans and deposits
$18T in Govt liabilities ($18T reserves and $0 in securities)
So that gets us part way to our goal of $42T in reserves, only $24 T to go.
There is no current legal mechanism for the Fed to add this many reserves once the securities are all gone as the Fed can only exchange financial assets, they are not allowed to create new NET financial assets as this is the domain of fiscal policy by Congress and the TSY.
But our current securities model of Govt deficit spending means that deficit spending has a net zero impact on reserve levels. So FRBers advocate for a reserve injection only model of deficit spending. Thats all well and good, but how long will it take to reach the requisite level of reserves? If we do $2T (almost 13% deficit to GDP) in deficit spending every year, then we could get the additional $24T in reserves within 12 years, but this assumes that bank debt doesn't increase over this 12 year period. How would that be accomplished?
I'm not so much worried about the Fed's ability to set the interest rate since they can always pay Interest on Reserves in a non-securities model of deficit spending.
But where would the reserves come from to get to a 100% reserve system?