THE DATA
For example, if we know the amount of revenue the federal government receives we can guess state and local governments also receive revenue that is approximately proportional to the federal receipts. Likewise if we know how much money is on deposit in banks we can guess there is a proportional amount deposited with other savings institutions. These proportions are not known, nor is there any reason to believe they remain constant. But, if prices and the available data provides a good enough correlation to demonstrate the truth of the equation of exchange without specifically knowing these things it then becomes only a matter of finding a control item that can be used to maintain the value of money. Among the purposes of this book is to show both the high correlation with the selected data and the methods available to control the value of money.
Two publications of the Federal Reserve Bank of St. Louis provide substantial data on money, prices, bank deposits, bank loans, and other measures of economic activity. The publications are National Economic Trends and Monetary Trends. It is these sources that provide the necessary data for our correlations.
The first data set required is a measure of the price level. National Economic Trends contains five indices of prices. They are:
The producer price indices are not particularly useful for this demonstration. These indices are averages of prices for intermediate goods and goods held for sale. Although conventional wisdom holds that a rise in producer prices will result in higher consumer prices, the actuality is that they cause fewer goods to be available not higher prices for those goods. But, the causes of economic growth and stagnation are items for later discussion. Here we are only concerned with the value of money.(1) Consumer Price Index (CPI); (2) Producer Price Index, Finished Goods; (3) Producer Price Index, Finished Consumer Goods; (4) Producer Price Index, Finished Goods, excluding foods; (5) Implicit Price Deflator for Gross National Product.
The remaining two indices, Consumer Price Index and Gross National Product Deflator, both represent relevant measures of the value of the dollar. There are three practical reasons to select the Consumer Price Index. They are:
The significance of the more frequent measurement of the Consumer Price Index is that the more often the price level is determined the sooner corrective action can be taken to restore the value of the dollar. If three months pass before corrective action is taken the correction required must be substantially more than if only one month passes. Considering the wild changes in inflation rates that we have experienced in recent years this may seem like a minor point. But, since there is a choice, why not take the better choice?(1) The CPI is compiled monthly while the GNP Deflator is only available every three months. (2) Most contracts calling for payment adjustments are based on the CPI, not the GNP Deflator. (3) The choice makes little difference so long as a choice is made.
The second reason deals with public perceptions. The usual method of expressing the rate of inflation is the Consumer Price index. Also, most inflation adjustments use the Consumer Price Index. These include adjustments to Social Security, Medicare, Welfare, pensions, union wage contracts and myriad other payment schedules. Therefore, the Consumer Price Index is a good choice in terms of public recognition of monetary value.
The third reason may, at first, seem a bit obscure. But, a reality of price indices is they are substantially interrelated. The Consumer Price Index includes prices of some goods that are not included in the Deflator. The Deflator includes some prices that are not included in the Consumer Price Index. But these two indicies are 70% to 80% based on the same prices. Because of this the choice makes little difference.
The next data set required is a measure of money supply. Several conventional measures are listed in Monetary Trends. These are:
With this variety of choices there could be some concern in selecting which of these measures to call money. However, because all of the first three measures and none of the fourth are used, the choice of which to call money is of no real consequence.(1) M1 - Consisting of cash held by the public, plus demand deposits (checking accounts) and other checkable deposits (NOW accounts and other non-conventional transferable deposits) plus travellers checks (perhaps best described as "signature money" in that once signed by the owner, travelers checks are negotiable); (2) M2 - Consisting of M1 plus savings accounts and small time deposits, plus overnight repurchase agreements (the very short term loans made by banks and other high rollers), plus money-market mutual fund balances (those mutual funds that let the small fry play the high-rollers game); (3) Total checkable deposits - that part of M1 that can be transferred by check; and, (4) Several other listings of financial data considered by some economists to be a form of money.
Payments by check were classified earlier as exchanges of goods or services for a promise. They were then included in the category of exchanges without money. When these data items are correlated with the price level they are used without labels. So, even though there is almost universal agreement among economists that demand deposits are money (I have not seen any economist's writings other than Irving Fisher's that do not agree.) checkable deposits will be called a source of payment for exchanges without money. This choice is simply a personal preference and a small tribute to Irving Fisher.
Savings are listed as transactions not included in GNP. The difference between M1 and M2 is the addition of those accounts that are generally regarded as savings. Economists cannot agree whether or not savings accounts are money. The difference is one of semantic choice and not a difference of substance. For that matter "checkable deposits" could just as well be called savings. About the only difference between savings and checkable deposits is the time preference and convenience of the owners of deposits.
If an owner intends to exchange his deposits in the near future, the holdings can be called "a source of payment without money." If the owner intends to use his deposits for exchange at some uncertain or distant future time then these holdings are more appropriately called "savings." The difference is not some innate characteristic of the holdings but only the intent of the owners.
At the risk of being repetitious, the choice of labeling is only one of convenience and not one of substance. So long as the data sets used produce a good correlation, errors introduced by inaccuracy that is caused by uncertainty of the intent of people can be ignored. And, what these individual measurements are called is of no significance whatever. So, the difference between M1 and M2 is labelled savings and is included with transactions not in GNP.
All that remains from the first three measures of money supply - M1, M2, and Demand Deposits - is the difference between M1 and total checkable deposits. This consists of coins, currency, and traveler's checks. There is no reason to call this item - M1 minus total checkable deposits - anything but money.
The next item to be considered can be called transactions not in GNP, exchanges without money or both. It all depends on the perspective. The subject is credit or loans.
Monetary Trends has two listings that measure credit. "Total Loans, excluding domestic commercial interbank loans" is an inclusive measure of both consumer credit extensions and commercial or business loans. "Business Loans, all commercial banks" measures only investment credit extensions. One can also expect that credit by savings and loan associations, mortgage lending institutions, and other finance companies and lenders will be sympathetic to bank credit extensions.
There seems to be a general agreement among economists that commercial or investment credit has a different effect on the economy than consumer credit. Less clear is which they consider better. There may be some economic law to explain this complex problem. There is also economist's answer for everything - "It depends." For lack of a reason, both are used. Business Loans, are called Investment Credit and Total Loans minus Business Loans are called Consumer Credit.
Government receipts, expenditures and debt are listed in Monetary Trends. Unfortunately these measures, as is the Gross National Product listed in National Economic Trends, are listed only on a quarterly basis. To make these measures compatible with the monthly measures of all other values used in the equation of exchange it is necessary to interpolate these listings.
For example, the first quarter 1983 value for government receipts is $623.3 billion and the second quarter is $652.6 billion. The first quarter covers the months January, February, and March; the second quarter covers April, May, and June. Since all other data is monthly, we must turn these quarterly values into monthly values.
The most direct way of doing this is to assume the values change in a uniform manner from one quarter to the next. First the quarter value is assigned to the mid-month of the quarter. That is, the first quarter value is assigned to February and the second quarter value to May. Next the values for March and April are estimated.
As is obvious, March is only one month from February but two months from May. Likewise April is one month from May and two months from February. Therefore the value for March should be much like February and a little like May and the value for April much like May and a little like February. So, March is made equal to two-thirds of the value for February plus one-third the value of May or, $633.4 billion. April is made equal to two-thirds of May plus one-third of February. The same interpolation process is applied to the other quarterly data and to other months from second to third quarter, third to fourth and fourth to first.
Government receipts are mostly taxes. There are some revenue items that are not properly called taxes. These include earnings of the Federal Reserve System and income from the sale or lease of government property. However, these data items are not listed separately. Also, government spending is not separated into its specific parts - defense, welfare, etc. So, only receipts and borrowing, the total of which equals spending, will be used in this version of the equation.
There is a difference between the three quarterly data listings of government receipts, government borrowing and GNP and the fourth quarterly data item, government debt. The first three of these items are annual rates that are averages for the quarter. This is why their values were placed at mid-month of the quarter. Government debt is listed as the amount existing at the end of the quarter. Therefore, the quarterly values of government debt are placed in the end months (March, June, September, and December) of each quarter and the first and second months are interpolated.
The data choices for Federal Debt include the following:
The difference between Gross and Net Federal debt is the inclusion or exclusion of government-owes-government items such as the various trust funds like Social Security and Civil Service pension funds. The portion of the Federal Debt held by trusts and agencies within the government represent nothing more than accounting practices and do not result in any real transactions. This eliminates the first of the above items as a reasonable choice for inclusion in the equation of exchange.(1) Gross Federal Debt. (2) Net Federal Debt (3) Federal Debt held by Federal Reserve Banks. (4) Federal Debt held by private investors. (5) Federal Debt held by foreign investors.
The debt held by Federal Reserve Banks does require the government to pay interest on this debt. However, any profit of the Federal Reserve Banks over a fixed amount is returned to the Treasury. Therefore, even though there are real transactions involved here and this data item is relevant to the method of controlling the money supply, the transaction effect of this item is the same as debt held by government agencies and trusts. That is, in terms of the equation of exchange, there is no effect. Later we shall find this item quite useful for another purpose.
The remaining items, debt held by private investors and debt held by foreign investors, both require meaningful transactions. Differences between them are meaningless from the perspective of the interest paid on this debt. But, when the recipients of this interest in turn buy goods and services, the distinction may make a difference in demand for domestic products. Therefore, both measures will be used.
How to classify government revenue and borrowing in the equation of exchange depends substantially on one's perspective. That is, when taxes are paid they are transactions not counted in GNP. But when the government buys goods or services with these revenues the transactions are counted in GNP.
By now it should be apparent that the semantic choice of what to call each item used to account for transactions is quite arbitrary. The relevance of this will be discussed in the next chapter - "Time, Sympathy, and Opposition."
Before going on to those subjects the use of GNP in the equation of exchange needs to be addressed. At the beginning of this discussion the economists' use of GNP for PQ was introduced. Dividing the equation:
by P gives:PQ = GNP
This is, of course, the definition of Q.Q = GNP/P.
Return to Contents Page
Return to Home Page