IT HAS FREQUENTLY BEEN OBSERVED that, while the national income accounts are based on a usable economic theory, the national financing accounts, whose construction is under way or in prospect in many countries of the world, have no such basis.1


IT HAS FREQUENTLY BEEN OBSERVED that, while the national income accounts are based on a usable economic theory, the national financing accounts, whose construction is under way or in prospect in many countries of the world, have no such basis.1

IT HAS FREQUENTLY BEEN OBSERVED that, while the national income accounts are based on a usable economic theory, the national financing accounts, whose construction is under way or in prospect in many countries of the world, have no such basis.1

The framework of the income accounts lies in the Keynesian analysis of saving and investment as an explanation of the cause of the expansion or contraction of income. The key to the need for financing accounts and for integrated income and financing accounts is to be found in the fact that the income accounts are deficient in two respects as a source of data on the variables in the Keynesian analysis. Investment as measured in the income accounts is not a wholly satisfactory measure of the investment variable of income analysis; and the income accounts omit entirely data on money and other financial assets, which are variables that play roles in the income analysis as necessary as those of saving and investment. The need for financing accounts is the need to measure the strategic variable, money, and to provide data on other financial assets in a form in which the causes and effects of changes in the economy’s preferences for money and other types of financial asset can be analyzed. The need for integrated income and financing accounts is the need to provide a single set of data that will include measurements of income, savings, investments, money, and other financial assets in a form in which the effects on income of autonomous changes in any of the last four can be analyzed.

Both investment and money are necessary to the logic of the analysis. Which of the two might at any time explain better the source of expansion or contraction, and which is therefore the more useful object of policy, depends greatly on the situation. In a climate that is inflationary, money is the dominant autonomous variable: the economy will expand if the monetary system expands. In a deflationary climate, investment is the dominant variable: the economy is not likely to expand unless investment expands. The Keynesian analysis was developed in a period of depression. During all of the postwar years, and probably at most times and in most countries, borrowers have been plentiful and it is more nearly true to say that the willingness of the money and banking system, and ultimately of the monetary authorities, to lend is the limiting, and hence the autonomous, factor. Money as a variable in analysis and policy formation has the further advantages of being measured relatively easily and subject to control by the monetary authorities. It is, in any event, important that the national accounts include measurements of both investment and money and the items through which either may be seen to work.

Sector Net Lendings and Net Borrowings

Investment in the Keynesian analysis has no very exact meaning. It is a shorthand expression for supposedly autonomous sources of expenditure. The income accounts, however, give to investment the precise and literal meaning of real capital formation. The government’s deficit appears as a negative saving rather than as a source of expansion; exports, which are largely an autonomous force from abroad, are netted against imports as foreign investment; and other classifications are made that cause the data to measure something other than Keynes’ “saving” and “investment.” Measuring real capital formation provides a useful part of a set of social accounts, not only because we want to know the amount of capital formation in order to assess real progress in the economy, but also because in the search for the source of economic expansion and contraction capital formation provides a valuable element. But real capital formation, literally defined, does not measure Keynes’ investment.

Integrated income and financing statistics require that the income accounts be sectored and that capital transfers and transactions in so-called “existing assets” be added to obtain sector totals of net lending, the link with financing data. The measurements of sector net lendings that result are useful supplements to data on capital formation in seeking measurements of expansionary forces. The government deficit appears as a source of expansion. Expenditures for the construction of privately owned housing, which may be a large part of real capital formation, can be attributed properly to the household sector rather than to a fictitious “business” sector, and purchases and sales of real estate, which cannot play any role in over-all accounts, can be entered in the picture. Since some sectors (primarily the household sector) are persistent surplus sectors and others (primarily the business sectors) are persistent deficit sectors, time series of net lendings by sectors are useful for observing changes in the usual positions of the sectors. Measurements of sector surpluses and deficits do not provide, any more than do capital formation and savings statistics, unambiguous measures of the variables needed for Keynesian analysis, but they do provide information for use in examining capital formation totals for some of the causes of excessive or deficient expenditure.

Because they are the link between income data and financing data, sector net lendings can be calculated by extending “downward” the income accounts, or by extending “sidewise,” to the accounts of all sectors, the types of data on financial assets and liabilities found in money and banking statistics. The accounts of the money and banking and financial system can provide a large part of the necessary data and in a highly reliable form. In all economies, a substantial amount of borrowing and lending is indirect. The institutions in the middle of the indirect process are necessarily few, compared with the numbers of primary borrowers and lenders, and they necessarily keep accurate records, which the primary borrowers and lenders are less apt to do. Financing accounts constructed from the “sidewise” extension of money and banking statistics to include the accounts of the other parts of the financial system with assets (and so far as possible, liabilities) classified by the economic sector against which they constitute claims (or toward which they constitute liabilities) go a long way toward the completion of a matrix of all borrowings and lendings. Although data on the assets and liabilities of financial institutions are relatively easily available and provide a large part of the data required for financing statistics, they have not—except for the accounts of the money and banking system—been assembled in most countries. The first conclusion in respect of the construction of financing accounts seems to be that work should begin here in order to provide much of the necessary data. Such differences as may exist between (1) the money and banking and insurance sector accounts needed for the financing accounts and (2) the data on these sectors now assembled by the International Monetary Fund in the Monetary Surveys and the Life Insurance accounts in International Financial Statistics should be examined, and if possible reconciled, and work on the assembly of data for all other financial institutions should be begun.

The purpose for which financing accounts are constructed is not revealed either by an examination of the usefulness of statistics of sector net lendings and net borrowings or by the fact that the accounts of the money and banking and financial systems are useful in the construction of financing accounts. Nor does either of these considerations provide a guide to the construction of financing accounts. If financing data were useful only for analyzing sector net lendings and net borrowings, no use would be made of the components. The role of the money and banking system and of all other financial institutions would be obscured instead of clarified since, for the financial institutions, net lending and net borrowing are essentially zero. The significance of the money and banking and financial system lies in the fact that what it lends and what it borrows are different things. Its net lendings reflect only the fact that financial institutions are also businesses. It would therefore be an aid to the understanding of integrated income and financing accounts to include the income transactions of financial institutions with those of the business sector so that there would be no near zero figure reported as the net lendings of the financial institutions. The combined financial and nonfinancial business sector could then be divided in the financing accounts into its nonfinancial and financial parts.

Components of Financing Accounts

There are three possible systems of components for financing accounts. The accounts might be directed to the measurement of inter-sector finance: they would then measure either each sector’s lendings to and borrowings from the other sectors or the increase or decrease of each sector’s claims on and liabilities toward the other sectors. Because of the existence of capital market securities, these two things are not the same. Alternatively, the accounts might be directed to the measurement of sector liquidity, measuring for each sector increases or decreases in each of several types of financial asset and financial liability. There are strong reasons for preferring the last of these three possibilities.

Savings not directly invested must take the form of some financial asset (or of a reduction of some financial liability). Investment not financed from income must be financed by the sale of some financial asset or by incurring some financial liability. Savers do not decide to finance others’ deficits, and surplus sectors do not decide to finance deficit sectors. Savers, or sectors with surpluses, decide to buy financial assets or to reduce their financial liabilities to suit their preferences for liquidity and yield. The financing accounts can tell us the kinds of asset that were bought by surplus sectors, the kinds of financial liability that deficit sectors issued, and the kinds of financial liability that financial institutions issued. It is here that the interest and importance of financing accounts lies and that guides to their construction can be found.

The asset preferences of those with surpluses are made to accord with the types of liability that those with deficits are able and willing to incur, partly through changes in relative interest rates and security prices and partly through the operations of the financial system. The financial sectors of the economy can be defined for the purpose of making financing accounts as that part of the economy whose business is the creation of types of asset that the economy wants to hold, primarily to satisfy its preferences for liquidity. The financial sectors are those parts of the economy which are willing and able to have quick liabilities greater than their quick assets. Their quick liabilities provide the economy’s liquidity and are therefore in demand.

Among the assets that the economy holds is money, and the role of money is special. Money is the most liquid asset. Money carries no interest rate and has no price, and, except in times of deflation, its quantity is fixed by the actions of the money and banking system. The economy can satisfy an increased preference for other types of financial asset by moving funds from one type of holding to another and/or by raising the value of its existing holdings through the attempt to acquire more. For example, the economy can satisfy an increased preference for savings and loan shares by buying more of them and increasing the size of the savings and loan institutions in the economy’s financial system. It can satisfy an increased preference for bonds by attempting to buy them and hence raising their prices. Raising their prices will increase the supply of bonds by inducing borrowers to offer bonds rather than other types of security, and at the same time it will directly increase the money value of all the economy’s holdings of bonds acquired in earlier periods. An increased preference for money, however, cannot be satisfied by inducing an increase in the quantity of money; neither, since it has no price, can its price be raised. It must instead be satisfied by deflating the economy through a fall in prices and incomes, a rise in interest rates, and a fall in the values of all other assets.

Financing statistics are valuable largely for the data they provide for analyzing the consequences of changes in the supply of, or preferences for, money and other types of financial asset. It therefore follows that

(1) Financing accounts should record assets and liabilities in a form which makes it possible to measure the liquidity and yield choices of holders. To do this, the sectors distinguished in the accounts should be calculated by summing groups of persons or institutions. Conceptual entities made by splitting the accounts of persons or institutions that keep single accounts and act as a single unit cannot have financing accounts that show the assets that the members of the sector choose to hold. A sector of this kind might have a calculable surplus or deficit, but there would be no useful way of determining which borrowings and lendings or which changes in assets and liabilities financed its surplus or deficit.

The accounts of sectors that represent sums of persons or institutions should include intrasector financial transactions. The data should include corporate securities held by corporations, and mortgages and other liabilities of households held by households. Claims of any given type against other members of one’s own sector serve the purpose of providing their holder with liquidity and yield quite as well as claims of the same type against other sectors. Without their inclusion, the accounts would not be addressed to the question of why the sectors hold the assets they do.

Such accounts should also be gross in a second sense: liabilities are not offsets to assets. The sectors obtain the liquidity they require through changes in their assets and changes in their liabilities. Measurement of the motivations of their actions and of the effects of their actions on the economy depends on the separate recording of both changes in assets and changes in liabilities.

(2) The accounts should separate money from other financial assets, use the word “money” to describe the assets included, and provide in the entries for the money and banking system’s liabilities the total of the economy’s money supply. The measurement of the role of money is the most important single purpose of financing accounts. To name the entry “money” is consistent with the useful practice, elsewhere in the national accounts, of identifying significant economic variables by commonsense names, e.g., saving, investment, and income, and takes cognizance of the fact that money cannot be usefully defined for all countries of the world as the sum of any particular combination of currency and deposits. Use of the word “money” would permit the legal definition to vary from country to country and yet maintain consistency in its economic definition.

To include in the accounts the total for money, it is necessary to record under other headings holdings by the government and by foreign governments and foreign banks of the items that would otherwise be called money, and to exclude by consolidation, or to include under a separate heading, intrabank claims.

Holdings of currency and demand deposits by foreign governments and foreign banks are foreign exchange liabilities that should be recorded as such rather than as money. Government holdings of currency and demand deposits should be excluded from the definition of money since the government is not a sector that can be presumed to be motivated by considerations of liquidity. While there are some who argue that government holdings of currency and deposits are money, others maintain that a government’s financial transactions with its central bank are in effect a single set of transactions and that government holdings of claims on the central bank should be netted against its debts to the central bank. The recording of government holdings under a separate heading serves both views. Intrabank claims consist of reserve money and deposits of commercial banks with other commercial banks. The separate identification of reserve money would be a useful feature of the accounts.

(3) The financing accounts should provide sectors for financial institutions and should separate the money and banking system from other parts of the financial system. Financial institutions may be defined as those whose principal function is the provision of assets that the economy wishes to hold, primarily to satisfy its needs for liquidity. The money and banking system may be defined as that part of the financial system which provides the assets called money and its closest substitutes, such as savings deposits, which can be called quasimoney. The accounts should also be constructed in such a way as to facilitate further classification of the money and banking sector into its three principal components: the monetary authorities, the deposit money banks, and the savings banks and similar institutions.

The definition of money and the definition of the money and banking system involve departures from the rules recommended in (1) above. They involve the creation of a split personality if the accounts of the money-issuing authority are to be reported in a single sector, since in most countries the government is the creator of part of the money supply. They also involve the exclusion of some intrasector claims if deposits between commercial banks are not to be included.

(4) Life insurance and pension funds fall under the definition of financial institutions as institutions whose business is the creation of assets that the economy wishes to hold. There are a number of reasons why their accounts should be reported as those of a separate sector.

Their receipts are partly receipts of others’ savings and partly payments for a service. On the part of the payers, there is question whether premiums are expenditures or purchases of assets, and insofar as they are thought to be acquisitions of assets, there is question about the value of the assets acquired: the face value, the actuarial value, the surrender value, or the loan value of the policy. Insurance premiums are, moreover, contractual payments which it is desirable to separate from other savings and asset holdings. For these reasons, it seems preferable to make life insurance and pension funds a sector, to record their income transactions, and to record in that sector the savings of others that take the form of insurance holdings. The life insurance and pension sector would therefore appear to be a large net saver, with large holdings of financial assets, and with essentially no financial liabilities. This treatment of insurance would be parallel to the inclusion of the social security system in the government’s accounts. The principle of not splitting personalities leads to the consolidation of the social security system with the central government account, with premiums recorded as government income and benefits recorded as government expenditure.

Casualty insurance companies fall entirely outside the definition of financial institutions. Like many other businesses, individuals, and nonprofit organizations they hold large assets, but unlike financial institutions they do not create assets for others to hold.

(5) Among the economy’s real assets, real estate is that which is most similar to financial assets in that it is frequently held as an alternative to other financial assets rather than for direct use. It would be desirable, therefore, if the capital reconciliation account distinguished real estate transactions from other investment transactions. To sector the accounts and calculate sector net lendings, transactions in so-called “existing assets” must be added to capital formation expenditures. Real estate is the principal component of existing assets. For the economy as a whole, there are no transactions in “existing assets,” and capital formation is the relevant concept. For the sector accounts, however, interest lies in the identification of sources of expansion and contraction; if these are to be properly attributed to the sectors, purchases and sales of existing assets must be included on the same footing with expenditures that result in new capital formation. The sectors choose to acquire capital assets; they do not choose to add to capital formation. The measurement of transactions in existing assets serves no purpose other than to complete sector accounts made in the first instance with capital formation entries appropriate to the total account rather than to the sector accounts. The measurement of real estate transactions, the principal component, serves the purpose of separating from other capital expenditures that which may in many cases be a transaction in a financial asset rather than in capital goods. It would therefore add to the usefulness of the accounts if, in the capital reconciliation account, fixed investment expenditures were classified as real estate and other fixed capital expenditures, rather than as new capital formation expenditures and transactions in “existing assets.”

(6) Liquidity is an asset and not a transaction. Financing data can measure the transactions that the sectors made to obtain liquidity and can show how these led to the finance of investment. They cannot, however, measure either the amount of the sectors’ liquidity or the effects on the values of assets of changes in asset prices and interest rates. That the economy’s wish to maintain a relationship between the values of its various assets is an important source of its actions implies that a complete set of national accounts needs not only financing accounts but also asset and liability accounts, at least for financial assets and financial liabilities. Inasmuch as the balance sheets of the financial institutions and others are a major source of the available data on finance, asset and liability data for the principal financial sectors should be as easily available as financing data, and their construction might well proceed simultaneously.


Mr. Hicks, who is a graduate of the University of North Carolina, is Assistant Director of the Research and Statistics Department.


This paper is one of a series prepared for the Expert Group on Statistics of Changes in Financial Assets and Liabilities, convened by the Conference of European Statisticians, Geneva, Switzerland, February 23–27, 1959. The series includes the paper by Graeme S. Dorrance, pages 168–209 of this issue of Staff Papers, and one by Poul Høst-Madsen which will be published in a later issue.