Divisia Money and “A Subjectivist Approach to the Demand for Money”

Recently Scott Sumner have brought up William Barnett’s new book “Getting it Wrong: How Faulty Monetary Statistics Undermine the Fed, the Financial System, and the Economy”. The theme in Barnett’s book is basically that “normal” money supply numbers where subcomponents of the money supply is added up with equal weight give wrong measure of the “real” money supply. Instead Barnett’s recommend using a so-called Divisia Money method of the money supply.

Here is a William Barnett’s discription of divisia money (from the comment section on Scott’s blog):

“Unlike the Fed’s simple-sum monetary aggregates, based on accounting conventions, my Divisia monetary aggregates are based on microeconomic aggregation theory. The accounting distinction between assets and liabilities is irrelevant and is not the same for all economic agents demanding monetary services in the economy. What is relevant is market data not accounting data.”

And here is the official book discription of Barnett’s book:

“Blame for the recent financial crisis and subsequent recession has commonly been assigned to everyone from Wall Street firms to individual homeowners. It has been widely argued that the crisis and recession were caused by “greed” and the failure of mainstream economics. In Getting It Wrong, leading economist William Barnett argues instead that there was too little use of the relevant economics, especially from the literature on economic measurement. Barnett contends that as financial instruments became more complex, the simple-sum monetary aggregation formulas used by central banks, including the U.S. Federal Reserve, became obsolete. Instead, a major increase in public availability of best-practice data was needed. Households, firms, and governments, lacking the requisite information, incorrectly assessed systemic risk and significantly increased their leverage and risk-taking activities. Better financial data, Barnett argues, could have signaled the misperceptions and prevented the erroneous systemic-risk assessments.

When extensive, best-practice information is not available from the central bank, increased regulation can constrain the adverse consequences of ill-informed decisions. Instead, there was deregulation. The result, Barnett argues, was a worst-case toxic mix: increasing complexity of financial instruments, inadequate and poor-quality data, and declining regulation. Following his accessible narrative of the deep causes of the crisis and the long history of private and public errors, Barnett provides technical appendixes, containing the mathematical analysis supporting his arguments.”

Needless to say I have ordered the book at look forward to reading. I am, however, already relatively well-read in the Divisia money literature and I have always intuitively found the Divisia concept interesting and useful and which that more central bank around the world had studied and published Divisia money supply numbers and fundamentally I think Divisia money is a good supplement to studying market data as Market Monetarists recommend. Furthermore, it should be noted that the weight of the different subcomponents in Divisia money is exactly based on market pricing of the return (the transaction service) of different components of the money supply.

My interest in Divisia money goes back more than 20 years (I am getting old…) and is really based on an article by Steven Horwitz from 1990. In the article “A Subjectivist Approach to the Demand for Money” Steve among other thing discusses the concept of “moneyness”. This discussion I think provide a very good background for understanding the concept of Divisia Money. Steve does not discuss Divisia Money in the article, but I fundamentally think he provides a theoretical justification for Divisa Money in his excellent article.

Here is a bit of Steve’s discussion of “moneyness”:

“Hicks argues that money is held because investing in interest-earning assets involves transactions costs ; the act of buying a bond involves sacrificing more real resources than does acquiring money. It is at least possible that the interest return minus the transactions costs could be negative, making money’s zero return preferred.

While this approach is consistent with the observed trade-off between interest rates and the demand for money (see below), it does not offer an explanation of what money does, nor what it provides to its holder, only that other relevant substitutes may be worse choices. By immediately portraying the choice between money and near-moneys as between barrenness and interest, Hicks starts off on the wrong track. When one “objectifies” the returns fro111each choice this way, one is led to both ignore the yield on money held as outlined above and misunderstand the choice between holding financial and non-financial assets. The notion of a subjective yield on money can help to explain better the relationship between money and near-moneys.

One way in which money differs from other goods is that it is much harder to identify any prticular good as money because goods can have aspects of money, yet not be full-blooded moneys. What can be said is that financial assets have degrees of “moneyness” about them, and that different financial assets can be placed along a moneyness continium. Hayek argues that: “it would be more helpful…if “money”were an adjective describing a property which different things could possess to varying degrees. A pure money asset is then defined as the generally accepted medium of exchange. Items which can he used as lnedia of exchange, but are somewhat or very much less accepted are classified as near-moneys.

Nonetheless, money and near-moneys share an important feature Like all other objects of exchange, their desirability is based o n their utility yield. However in the case of near-moneys, that yield is not simply availability. Near-moneys do yield some availability services, but not to the degree of pure money. ‘The explanation is that by definition, near-moneys are not as generally acceptable and therefore cannot he available for all the same contingencies as pure money. For example, as White argues, a passbook savings account is not the same as pure money because, aside from being not directly transferrable (one has to go to the hank and make a withdrawal, unlike a demand deposit), it is not generally acceptable. Even a demand deposit is not quite as available as currency or coin is – some places will not accept checks. These kinds of financial assets have lower availability yields than pure money because they are simply not as marketable.”

If you read Steve’s paper and then have a look at the Divisia numbers – then I am pretty sure that you will think that the concept makes perfect sense.

And now I have written a far too long post – and you should not really have wasted your time on reading my take on this issue as the always insightful Bill Woolsey has a much better discussion of the topic here.


Schuler on money demand – and a bit of Lithuanian memories…

Here is Kurt Schuler over at

“During the financial crisis of 2008-09, many central banks expanded the monetary base. In some countries, the base remains high; in the United States, for instance it is roughly triple its pre-crisis level. Such an expansion, unprecedented in peacetime, has convinced many observers that a bout of high inflation will occur in the near future. That leads us to the lesson of the day:

To talk intelligently about the money supply, you must also consider the demand for money. Starting from a situation where supply and demand are in balance, the supply can triple, but if demand quadruples, money is tight. Similarly, the supply can fall in half, but if demand is only one-quarter its previous level, money is loose.

In normal times, it is a fairly safe assumption that demand is roughly constant or changing predictably, but in abnormal times, it is a dangerous assumption. No high inflation occurred in any country that expanded the monetary base rapidly during the financial crisis. Evidently, demand expanded along with supply. In fact, Scott Sumner and other “market monetarists” think supply did not keep up with demand. Similarly, nobody should be perplexed if a case arises where the monetary base is constant or even falling but inflation is rising sharply. Absent a natural disaster or some other nonmonetary event, it is evidence that demand for the monetary base is falling but supply is not keeping pace.”

Kurt is of course very right – the way to see whether monetary policy is tight or loose is to look at the money supply relative to the money demand. Since, we can not observe the difference between the money supply and the money demand directly Market Monetarists recommend to look at asset prices. We know that tight money (stronger money demand growth than the money supply growth) leads to a drop in equity prices, lower bond yields (due to lower inflation expectations), a stronger currency and for large economies like the US or China lower commodity prices.

One thing Kurt did not mention – and I a bit puzzled about that as it is a very important argument for Free Banking – is that in a world with Free Banking the total privatisation of the money supply means that the money supply (ideally?) is perfectly elastic and that any increase in money demand is meet by a equally large increase in the money supply. The same will be the case in a world with central banks targeting the NGDP level. With a perfectly elastic money supply crisis like the Great Depression and the Great Recession is likely to be much more unlikely.

PS I am writing this while I am in Lithuania – a country where Kurt’s (and George Selgin’s) work played a key role nearly 20 years ago in the introduction of the country’s currency board system. See more on this here.

%d bloggers like this: