I think Tom and I are beginning to talk past each other now.  Had he 
addressed the criticisms I made against Wicksell's "pure credit" system 
directly, I wouldn't need to respond again.  Perhaps, this post will 
help to show clearly why he and I cannot agree.

First, money is the particular commodity that serves as a unit of 
account, according to the classical definition.  That money is used also 
as a medium of exchange is its secondary function. Somewhere along the 
line in the late nineteenth century economists began to cite the medium 
of exchange function as the first identifier of what money is.  They 
thus started to include instruments of credit, like checks or checkable 
deposits, in their definition of money.  As Francis A. Walker (1878) 
writes in objection to that practice, "Money is that which passes from 
hand to hand in final discharge of debts and full payment for goods. The 
bank-deposit system allows the mutual cancellation of vast bodies of 
indebtedness which would, without this agency, require the intervention 
of an actual medium of exchange; but deposits, like every other form of 
credit, save the use of money; they do not perform the functions of 
money. Money is what money does."  A pure credit system thus does not 
have money, properly so called.

Second, the price of a commodity is the quantity of money (the unit of 
account) that is exchanged for it.  Note that a check is a command to 
pay money; that is why a check is an instrument of credit rather than 
money itself.  So, in the absence of money, we would have relative 
exchange values for different commodities or services, amounting to 
N(N-1)/2 in all, where N = the number of goods and services in an 
economy.  But since none of those goods is the measure of value or unit 
of account, we cannot have a determinate common value (Price Level) that 
is the index of their weighted average, like the GDP deflator or the 
CPI.  Tom thinks that is obtainable; I don't see how.  He says: "In 
short, there exists both a unit of account (the dollar) and a means of 
exchange (checkable bank deposits) in Wicksell's pure credit case. Both 
assure a measurable, definable nominal dollar price of each good and 
also their aggregate, the general price level."

Third, when there is a unit of account, there is also an identifiable 
source of its supply.  And variations in the quantity of that unit of 
account relative to its demand determine the price level: P = H/kY, 
where H = the quantity of the unit of account, k = the proportion of 
income that the public wants to hold in the unit of account, and Y = 
aggregate of nominal incomes in the community.  Now this is the quantity 
theory of money, of which Wicksell had difficulty to understand fully.  
To then talk about variations in the price level in the absence of H 
again makes no sense to me.

Fourth, variations in the supply and demand for credit determine 
interest rates (according to classical analysis); Wicksell had a problem 
understanding that, too.  And in a commercial society, there are a 
variety of interest rates over which a central bank (the source of both 
money and some credit, not founded on savings) has no control.  To 
pretend otherwise in any modelling is unhelpful, I contend.  It may well 
be true that Wicksell "formulated an interest-rate-adjustment feedback 
policy rule much like those used by central bankers today," as Tom 
says.  Any wonder central bankers are messing up several economies 
around the world, including the USA?

Previously, Tom argued that the natural rate of interest is 
"unobservable."  When I pointed out that we can observe the rate of 
interest; we just can't tell whether it is the natural rate or not, he 
changes his position.  Now, it is whether prices move or not.  If prices 
don't move, we have a natural rate.  Otherwise, we don't have a natural 
rate and the central bank has to act.  Simply by calling out a different 
interest rate or changing the flow of credit (how)?  But for Wicksell's 
commingling of money with credit, perhaps this firm link between "money" 
and interest rates would be avoided.  Robert Greenfield and Leland 
Yeager (/SEJ/, 1986), "Money and Credit Confused: An Appraisal of 
Economic Doctrine And Federal Reserve Procedure," documents how the US 
Fed has operated inappropriately on the basis of their commingling of 
credit and money.  Alas, the message is yet to be fully appreciated.

I happily (and proudly) stand by the monetary analysis of the 
classicals, from David Hume, Adam Smith, David Ricardo, Henry Thornton, 
J.S. Mill, and carried into the neoclassical period by Alfred Marshall.  
Tom may keep his admiration of Wicksell's.

James Ahiakpor

Thomas Humphrey wrote:
> Much as I admire James Ahiakpor's persistence in showing that a long 
> list of classical and neoclassical predecessors scooped Wicksell's 
> cumulative process analysis, I'm afraid I still cannot concur with his 
> Wicksell criticism in all its particulars.
>
> Concerning Wicksell's pure credit economy in which the stock of 
> checkable bank deposits ("credit" in Wicksell's terminology) is 
> untethered by reserve constraints or metallic coin backing -- neither 
> of which exists --James writes that "without money (cash) we can't 
> tell what is the price level, let alone its movement."
>
> With all due respect, this argument can't be right. For contrary to 
> James, the price level and so its rate of change indeed are both 
> definable in Wicksell's pure credit economy.  While there is no cash 
> money in such an economy, there is a transaction medium of exchange in 
> the form of bank deposits transferable by check. Buyers purchase goods 
> paying for them with checks drawn upon their deposit accounts. Sellers 
> receive the checks which add dollars to their deposit balances. Goods 
> are paid for by checks drawn upon deposits at prices measured by the 
> number of dollars' worth of deposits given up in exchange for each 
> good. In short, there exists both a unit of account (the dollar) and a 
> means of exchange (checkable bank deposits) in Wicksell's pure credit 
> case. Both assure a measurable, definable nominal dollar price of each 
> good and also their aggregate, the general price level.
>
> What James must mean is that the price level may be */indeterminate/*, 
> not */undefinable/*, in the pure credit case. And here he is partly 
> right. He is right in the sense that with (1) no reserve constraint, 
> or (2) demand for metallic cash money to limit the stock of bank 
> deposits, which therefore can expand or contract without end, there is 
> nothing to pin down the price level. That is, there is nothing to pin 
> down the price level unless the central bank elects to pin it down by 
> means of a feedback policy rule requiring adjustment of the bank rate 
> in the same direction prices are moving, stopping only when they cease 
> to move. Indeed, that's the whole rationale for Wicksell's policy 
> rule, namely to pin down the price level and render it determinate in 
> circumstances when it otherwise wouldn't be. And here is where James's 
> price-indeterminacy claim falls down: For even in the pure credit 
> case, indeterminacy need not prevail provided the central bank adheres 
> to a stabilizing feedback policy rule.
>
> What about James's claim that "It is precisely because central bankers 
> don't know whether the observed market rate is the natural rate that 
> they should refrain from attempts to influence it." Again I must 
> demur. Central bankers */do/* know whether the market rate is or is 
> not at the natural rate. The behavior of the price level tells them. 
> If prices are moving, the market rate is not at the natural rate. But 
> if prices are stable, the market rate is at its natural equilibrium 
> level. Central bankers can use such information on price movements to 
> adjust the market rate until it conforms with the natural rate. 
> Wicksell's feedback policy rule is both feasible and stabilizing.
>
> Finally, even though he didn't invent cumulative process (CP) 
> analysis, Wicksell still merits a place in the pantheon of monetary 
> theorists. He contributed at least three innovations that advanced 
> monetary science beyond the teachings of his great classical 
> predecessors.
>
> 1. He applied the classical CP analysis to a whole spectrum of 
> alternative payments mechanisms, including pure cash, mixed 
> cash-credit, and pure credit regimes. By contrast, the classicals 
> limited their analysis to the first two of these.
>
> 2. He formulated an interest-rate-adjustment feedback policy rule much 
> like those used by central bankers today. By contrast, his classical 
> predecessors postulated only two policy rules, namely the gold 
> standard rule calling for conversion of paper into specie at a fixed 
> price upon demand, and the Currency School's 100% marginal gold 
> reserve requirement rule.
>
> 3. His money demand analysis went beyond anything his classical 
> counterparts had to offer. He resorted to probability theory to 
> examine the public's demand for cash balances. The public demands cash 
> to help bridge shortfalls caused by lack of synchronization between 
> receipts and payments. To explain this, Wicksell incorporated a 
> probability distribution's mean and standard deviation into the money 
> demand function. The distribution describes the frequency with which 
> cash shortfalls of various amounts are likely to occur. The 
> distribution's mean or central tendency measures the expected 
> shortfall. The distribution's spread or standard deviation measures 
> the risk that actual shortfalls will be larger than expected. 
> Classical money demand analysis offered no comparable probability 
> reasoning.
>
> Thomas Humphrey


-- 
James C.W. Ahiakpor, Ph.D.
Professor
Department of Economics
California State University, East Bay
Hayward, CA 94542

(510) 885-3137 Work
(510) 885-7175 Fax (Not Private)