A critique of the Quantity Theory of Credit

P1050533 Another sunset


For our students I am pleased to say that Richard`s rebuttal has arrived and I include it in red after my comment:-

I have great respect for Richard Werner`s research abilities and attention to detail, but we have some differences of opinion about the assumptions used in his Quantity Theory of Credit (QTC) and New Paradigm. I have raised these points in tweets and Richard`s response is always have you read my QTC and my reply is yes. Last time this point was raised I said I would provide Richard with a critique of QTC and he passively agreed that it should be shared in a form that our students could read and reflect upon the debate. They like to see academics that they respect arguing rather than just sitting in their ivory towers. Below is what I sent to Richard:-

As promised a few thoughts on your 1997 Quantity Theory of Credit.

I see your disaggregation of the Quantity Theory of Money (QTM) as a step forward to grasping the different impacts of credit money used for financial transactions (Cf) and credit money used for real transactions (Cr) and I have no doubt that it explains perfectly what has happened, in this low interest rate environment, since the financial crisis.

R – It has worked perfectly in the 25 years that I have been using it, and interest rates have been far higher – even double digits in the UK at the time.

As I read it your explanation of changes in velocity (V) seem to disaggregate V in such a way that an increase in V caused by a growth in Cg will bring about a reduction in V for Cf and vice versa. Overall the impact on V will be much less than the change in each component.

R – No this works differently. What I show, eg in Werner 1997, is that there are 2 different velocities, namely of credit for the real economy, and credit for asset purchases. I show that the former velocity is very stable. There was no velocity decline, as others had argued, but based on the erroneous quantity equation that assumes all money is used for the real economy (not true!).

You prove growth in loans to the real economy (productive investment) determines economic growth. I agree but I would like to see a further separation between growth in loans for consumption and growth in loans for productive investment.

R- What I show is sufficient for the claims I make. Werner 2012 lists further empirical work, including some that goes in the direction of making this further disaggregation.

You say “The results imply for policy makers that it is imperative to monitor the allocation of credit”- I totally agree- and then you say “and intervene”. Here my concern is that it has been the wrong sort of intervention that has created this misallocation or as the Austrians would say malinvestment.

R- Sure, that’s why I am suggesting the right sort of intervention.

It is here that we seem to have a totally different emphasis on the role of interest rates. You seem to see a very passive accommodating role for interest rates whereas I see a very active and potentially damaging role for interest rates particularly when they are officially set at rates that are far removed from a free market rate. I see the current problem as a growth in Cf caused by a Bank Rate that is much too low and has distorted both the structure of interest rates with some further distortion also to the term structure. The solution then is to free interest rates and control quantity of money not its price.

Overall a big tick for your model and its predictive and analytical capacity, but now for a little nit-picking:-

I think your new paradigm builds on, rather than replaces, QTM.

I do not see a conflict between a money supply based on deposits when you consider that it is loan deposits, which are on the liability side of the balance sheet, that are the money created by banks. Also new money created is only created at the margin otherwise it is old money that is being recycled. I worry about people who think every loan created is new money in the economy.

I would not use M.V = P.Y as Y already incorporates P. Much easier to see how M.V = P.T (where PxT = Y).

R- Y is defined as real GDP, and does not incorporate P. To the contrary, P is estimated by subtracting something (based on assumptions) from PY (=nominal GDP). PY is observable, Y is not.

I do not see a stable V as a prerequisite of QTM.

R- 100% of economists who use the quantity equation however do. See the vast literature on this, and on the alleged ‘velocity decline’.

V is stable in a stable world, but becomes unstable in an unstable world (where instability is determined by volatile average prices). Accelerating inflation increases V while deflation slows V down.

R- I am dealing with the world we live in. The correctly defined real velocity is stable (actually for many reasons, including of technical nature).

A point that has been raised between us before is whether or not the money demand function is endogenous, and the money supply function is exogenous (which is my interpretation) and if correct means that they have different roles to play in any analysis.

R- The demand for money and credit is infinite, the supply is limited and controlled by the suppliers of money, namely the banks. All markets are rationed and the short side determines the outcome, which is the supply of money, i.e. bank credit creation and allocation decisions are key.

I do not know as much as you about the land price bubble in Japan, but I think you mistakenly held interest rates as a constant and unimportant variable in your analysis.

R- I do not assume anything ex ante about any variable. But the data showed that interest rates, which were included as potential explanatory variable, drop out as insignificant and adding no information value in a parsimonious model of nominal GDP growth of Japan. True also in the UK, Spain, and other countries where I tested this. See, e.g. here:


I always take money M as a stock and say that it does not become effective until it is a flow MxV. This includes credit money which is not a flow until it moves and this movement is intermittent not continuous.

R- I agree and implement in the analysis that we should look at flows. That’s why the model is based on credit creation, which explains economic growth, asset prices etc.

I do not see that all money is credit.

R- All money is credit. Which money is not credit? Explain.

I see all money created by credit but once created it takes on the role of being a medium of exchange, unit of account etc, as it is used again and again.

R- The use of money is not a definition of money.

It is only created and destroyed at the margin which means expansions and contractions are a very small proportion of the total.

R- As we agreed above, we need to look at growth, which nets out any unchanging stock/scale variable. The ‘total’ you mention may be unchanged and hence is not included after calculating the first difference.

You make reference to time deposits as money whereas I see these as near money or one step removed from money.

R- It is standard central bank and academic economics terminology to use ‘money supply’ aggregates as definitions of money. M3, M4 include time deposits.

You also include CDs, bonds and real estate as money whereas they are assets and definitely not money.

R- I do not include them, but I point out that analytically one could include them, as I describe the problem of the conventional definition of the money supply, namely to use private sector assets to define money. I think it is the wrong approach. Friedman admitted that there is no obvious cut-off as you lengthen the degree of liquidity. The solution is to drop private assets and look at private sector liabilities to the bank, ie. bank credit.

Money is what is immediately available to use in a transaction.

R- Forrest Gump’s definition of money.

A further point which creates confusion when you move to a policy based upon QTC is your assumption that there is a deficient demand output gap that can be closed by expansionary monetary policy. As every supposed output gap since WW2 has been accompanied by inflation in the UK the only way you can argue that this is deficient demand is if you accept the fallacy of cost push inflation. If, as I do, you accept that all inflations are the result of excessive monetary demand it is difficult to argue for an expansionary role for monetary policy. This does not invalidate your argument for a more effective allocation of credit, but that is something that will improve without interest rate manipulation and definitely does not require government or Central Bank intervention.

I look forward to your rebuttal.

John Hearn 29/1/2018 revised 6/2/18





Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s