From the Foreword to the first issue: “Among the responsibilities of the International Monetary Fund, as set forth in the Articles of Agreement, is the obligation to fact as a center for the collection and exchange of information on monetary and financial problems,’ and thereby to facilitate ‘the preparation of studies designed to assist members in developing policies which further the purposes of the Fund.’ The publications of the Fund are one way in which this responsibility is discharged. “Through the publication of Staff Papers, the Fund is making available some of the work of members of its staff. The Fund believes that these papers will be found helpful by government officials, by professional economists, and by others concerned with monetary and financial problems. Much of what is now presented is quite provisional. On some international monetary problems, final and definitive views are scarcely to be expected in the near future, and several alternative, or even conflicting, approaches may profitably be explored. The views presented in these papers are not, therefore, to be interpreted as necessarily indicating the position of the Executive Board or of the officials of the Fund.”


From the Foreword to the first issue: “Among the responsibilities of the International Monetary Fund, as set forth in the Articles of Agreement, is the obligation to fact as a center for the collection and exchange of information on monetary and financial problems,’ and thereby to facilitate ‘the preparation of studies designed to assist members in developing policies which further the purposes of the Fund.’ The publications of the Fund are one way in which this responsibility is discharged. “Through the publication of Staff Papers, the Fund is making available some of the work of members of its staff. The Fund believes that these papers will be found helpful by government officials, by professional economists, and by others concerned with monetary and financial problems. Much of what is now presented is quite provisional. On some international monetary problems, final and definitive views are scarcely to be expected in the near future, and several alternative, or even conflicting, approaches may profitably be explored. The views presented in these papers are not, therefore, to be interpreted as necessarily indicating the position of the Executive Board or of the officials of the Fund.”

Over seven years ago it was possible to believe that “a large and interesting development of financial statistics” had “breathed a welcome new life into social accounting statistics.”1 Unfortunately this promise has not really been fulfilled. Few countries have undertaken the work of developing financial statistics, which may be defined as attempting to “build up as complete a picture of borrowings and lend-ings as we can achieve”; and even fewer have developed “statistics of assets in terms of market values”2 and of the structure of the community’s liabilities.

It is easy to explain why progress has been so disappointing, but harder to indicate how we might move forward more rapidly in this field. “Facts arranged in the right way, speak for themselves; unarranged, they are as dead as mutton.” “As always it is the arrangement which is the delicate operation.”3 Arrangement in the right way requires a theoretical framework on which the statistical presentation may be built. Our present theory of finance can perhaps be best described, in terms used by one commentator to describe the financial theory used by one central bank, as “woefully inadequate, unverified and incapable of bearing the heavy burden that is placed upon it.”4

Any review of balance of payments statements or national income and expenditure accounts in nonsocialist countries would reveal marked similarities between the different analyses. They are all based, or appear to be based, on the International Monetary Fund’s Balance of Payments Manual,5 and the United Nations’ A System of National Accounts.6 These similarities reflect much more than international bureaucratic conformity: they reflect general agreement on important theoretical concepts. The Fund’s Manual is a direct descendant of the statistical explorations of Viner,7 White,8 and Williams,9 which were undertaken to examine the theoretical postulates set forth by Taussig.10 These concepts fitted neatly into the later principles underlying the national income and expenditure accounts. Similarly, for national income and expenditure, Keynes outlined the fundamental theoretical relations11 which could be used to provide meaningful statistical equations by individuals such as Denison, Gilbert, Luxton, Meade, Schwartz, and Stone.12

As yet, there has not been a Taussigian synthesis or a Keynesian revolution in financial theory that would provide the framework that is required for a satisfactory presentation of financial statistics. Reference need only be made to the controversy set off by the publication of the “Radcliffe Report”13 to indicate that economists hold a wide range of views on the mechanics of the operation of a community’s financial system, and, consequently, a wide range of views on the best statistical description of its operation.

Theoretical Foundations for Analyses

This lack of generally accepted theoretical principles is one of the reasons why a survey of financial accounts must present a picture of statisticians groping along different, never clearly marked routes, each hoping that he will be able to develop a satisfactory set of accounts. In general, it is possible to suggest five important paths that financial statisticians are following. Taking arbitrary origins for the concepts upon which these types of account are based, they may be classified as:

  • 1. The balance-sheet approach stemming from Hicks, The Social Framework, published in 1942.14

  • 2. The flow-of-funds approach stemming from an unpublished memorandum circulated by Wesley Mitchell in 1944.15

  • 3. The monetary-survey approach stemming from a paper presented by Triffin to the First Meeting of Central Bank Technicians of the American Continent, in Mexico City, in 1946.16

  • 4. The liquidity approach first outlined by Governor Holtrop of the Netherlands Bank in 1953 (12.3.1).17

  • 5. A financial surpluses and deficits of individual sectors approach, whose origins are rather difficult to trace, but which was probably first used in the Annual Report on the operations of the “Monnet Plan” in 1954 (5.3).

The Social Framework was a revolutionary book. It was the first undergraduate text to be based on the national income and expenditure accounts which now appear somewhere, even if not prominently enough, in all textbooks. However, Hicks found this framework inadequate for a complete description of the economy. He found it necessary to include a chapter based on the national balance sheet.18 From this, it might be concluded that a national balance sheet should be a component of any set of national accounts. Yet national balance sheets are very rare. As early as 1947, the Netherlands published estimates for 1938 (12.1.1). Again in 1950 (12.1.2), 1954 (12.1.3), and 1955 (12.1.4) statements for the period 1946-54 were issued. Since then, there have been reports of further investigation but no results. In this field, the effort has been left largely to private investigators, such as the Postwar Capital Markets Study group at the National Bureau of Economic Research, which produced the monumental Studies in the Balance Sheet of the United States, by Goldsmith, Lipsey, and Mendelson;19 the Wealth Inventory Planning Study group at George Washington University under the general direction of Kendrick, which presented Measuring the Nation’s Wealth20 to the Joint Economic Committee of the U.S. Congress; and the Department of Applied Economics of Cambridge University, where Revell’s national balance sheet was prepared.21

The U.S. Federal Reserve Board’s series on the flow of funds is perhaps the most publicized of the presently available sets of financial accounts. This work originated at the National Bureau and appeared in its early form as Copeland’s A Study of Moneyflows in the United States. He set out to measure “the complex made up of all the various outflows and inflows of all the various transactors.”22 The original Federal Reserve study (17.3.1) followed Copeland’s model rather closely. In fact, it was largely a continuation of Copeland’s work under his constant watch. As will be indicated later, the flow-of-funds accounts have changed over time and now bear little of their original Copelandian appearance. Despite the wide international use of the term “flow-of-funds accounts,” this type of accounting has not really been adopted permanently by any country. It is now recognized that an attempt to measure all the various outflows and inflows of all the various transactors is redundant. The national income and expenditure accounts have proved to be useful tools for economic analysis. They are well known, their uses and limitations are understood, and they have come to form an almost subconscious part of the thought processes of economists, civil servants, politicians, and journalists. However, because the present system of national accounts assumes that borrowing and lending transactions “cancel out and do not provide a net source for the finance of domestic capital formation,”23 it cannot fully perform its allotted role as “a practical means of describing what is taking place in an economic system insofar as this can be expressed in terms of transactions between a set of accounts drawn up on the double-entry principle.”24 If financial accounts are directed toward financial aggregates rather than the entire complex of transactions, they can contribute to the completion of an economic accounting structure as a record of the transactions in a community.

Triffin was faced with the problem of trying to assess the causes and effects of inflation in the financially simple immediate postwar economies of Latin America. He viewed “the money supply” as the key variable, and set forth to measure its “external” (i.e., balance of payments) and “internal” (i.e., domestic bank credit) origins. With the passing of time, interest in the internal and external origins has waned, but the view that relatively simple monetary surveys are useful financial statements has spread. Aside from the surveys for 89 countries published by the International Monetary Fund each month in International Financial Statistics, such surveys are available in national sources for at least 45 countries.25

It is reported that, when the first draft of the 1953 Centraal Economisch Plan (12.2.1) was circulated in the Netherlands, President Holtrop of the central bank commented on it that although the draft contained no monetary data, its monetary implications appeared inconsistent with its price assumptions. The staff of the Bureau was instructed to prepare a monetary analysis indicating the changes in bank credit and consequent changes in community liquidity that would be required to give effect to the Plan. This analysis confirmed Holtrop’s suspicions, and since then, the annual statement of the Plan, which has continued to be primarily a national income and expenditure plan, has contained a “monetary survey,” which measures past and prospective changes in money (with an attempt to separate active from inactive balances) and secondary liquid assets (primarily Treasury bills and time deposits) and assesses the significance of these changes. The annual report of de Nederlandsche Bank (e.g., 12.3.2) bases its discussion of monetary policy very largely on a similar analysis.

One of the most obvious extensions of the national income and expenditure accounts into a description of the financial aspects of the community has been to measure the financial surplus or deficit of each sector and the individual financial transactions producing this total. If adjustments can be made to the activity sectors that are used in national income and expenditure accounting, to give totals for the decision-making sectors that are relevant for financial analysis, consistent financing, income, and expenditure accounts may be derived. For any sector, its income plus transfers received less its consumption and transfers paid out are equal to its saving;26 its saving less its investment is equal to its financing balance. Similarly, for any sector its lending plus purchases of financial assets less its borrowing and sales of financial assets are equal to the same balance. Provided that all transactions are properly recorded, these two measures should be identical. In part, the income-expenditure and financing approaches reflect different approaches to the measure of expansionary and contractionary forces in an economy. For the income and expenditure approach, investment is expansionary and saving is contractionary, the financing surplus or deficit being very largely a rather irrelevant balancing transfer. For the financing approach, business and other entrepreneurs who finance investment out of their own savings are neutral in their effect on expansion or contraction.27 Their savings provide the resources from which their investments are financed, while the net expansionary pressures come from those who wish to borrow more than they lend and net contractionary pressures come from those who wish to lend more than they borrow during the period.28

While, by themselves, a wide variety of theoretical approaches would produce a wide variety of national accounts, there is a further difference between balance of payments accounts and income and expenditure accounts, on the one hand, and financial accounts, on the other, leading to variety among national statements, even if some of these differences may be more apparent than real. All countries import and export goods and services, borrow or lend internationally, and hold foreign balances. Hence, all balance of payments statements have similar appearances. The details on commodity export classifications, portfolio transactions, or changes in foreign-owned bank balances may be different in each national statement, but these differences are relatively small. Similarly, wages, profits, interest, and rents are paid and earned in all countries; producers and consumers consume, invest, and save: a standard system of national accounts can serve all countries. Financial systems, however, differ markedly from country to country. Even if the same principles were applied in both countries, an accounting system describing financial transactions in the United States, where private resident economic units other than financial institutions hold only 20 per cent of their financial assets in the form of money plus quasi-money (with corporate equities accounting for 30 per cent of the total, life insurance for 15, and bonds for over 10 per cent)29 would look quite different from one for Argentina, where money alone accounts for almost 60 per cent of the community’s financial assets, and quasi-money for almost another 30 per cent.30 It is no accident that the origins of the balance-sheet approach are found in a book describing the financially most complex society in the world,31 and the origins of the monetary survey approach in a paper dealing with Latin American problems.

Presently Available Analysis

The detail of this description of the origins of different types of financial accounts should not be allowed to lead to the conclusion that financial analyses are an important part of the economic accounts in many countries, or that, where they are compiled, they are always important instruments for the determination of policy. Aside from the countries producing monetary surveys already mentioned, it is probably fair to say that real financial analyses are prepared in no more than 14 countries (Denmark, Finland, France, the Federal Republic of Germany, Israel, Italy, Japan, the Netherlands, Norway, Sweden, the United Kingdom, the United States, Venezuela, and Yugoslavia). Rudimentary data are available for three more (Australia, India, and New Zealand). The possibilities of developing complete sets of accounts are being examined in Australia (1.1). On the other hand, Canada (2.1 and 2.2) and Mexico (11.1) have apparently abandoned work started earlier in this field. Even though only 5 classifications have been outlined here, or 4 if the Triffinian monetary surveys are excluded, and only 14—or at most, 19—countries may be included in this review, any classification of the presently available accounts is difficult.

Granting the difficulties of classification, and recognizing that some of the presently published statements may be considered as reflecting more than one of the approaches outlined here, it is possible to suggest that:

  • 1. Elementary balance-sheet statistics are published in Japan (10.1.2 and 10.1.4), Norway (14.1), and the United States (1.7.32). Further, corporate balance-sheet data are presented in Australia (1.2), India (7.1.1), Japan (10.1.1), New Zealand (13.1), and the United States (17.4 and 17.5).

  • 2. Flow-of-funds type data continue to be produced in: Japan (10.12) and the United States (17.32), and each year the statistical concepts underlying these presentations move farther from their original premises. In addition, the measures of total borrowing and lending are of an essentially flow-of-funds type as presented in the statements in the Federal Republic of Germany (6.1), India (7.12), Italy (9.1), Norway (14.1), Sweden (75.2), and Venezuela (18.1).

  • 3. The monetary-survey approach is, in addition to the countries already mentioned, the underlying philosophy of the statistical presentation in the Economic Survey of Denmark (3.1), and the Banca d’Italia’s Annual Report (9.1).

  • 4. The Holtropian approach of analyzing primary and secondary liquidities remains confined to the Netherlands, even though annual discussion of it continues to be one of the most stimulating events in the financial statistician’s year (12.32).

  • 5. Sector financing balances are the keynote of the analyses in Finland (4.1), France (5.2), the Federal Republic of Germany (62), Norway (14.1), Sweden (15.1), and the United Kingdom (16.1). For the annual reports of the Bank of Israel (8.1), and the National Bank of Yugoslavia (19.1), this type of analysis may be classified as more than the keystone: it is the fundamental framework on which the description of financial developments during the year is built.

This classification should not be allowed to give the impression that there has been either a full integration of financial statistics with other national accounts in all these countries, or that complete sets of financial statistics are presented in all of them. Integration with other national accounts is really attempted only in Finland (4.1), France (5.2), the Federal Republic of Germany (6.2), the Netherlands (12.2.2), Norway (14.1), Sweden (15.1), and the United Kingdom (16.1). Complete sets of financial statistics probably remain limited to the Federal Republic of Germany (6.2), Israel (8.1), Norway (14.1), the United Kingdom (16.1), the United States (17.3.2), and Yugoslavia (19.1).

Nor should this classification be allowed to give the impression that financial statistics are used for policy determination in all the countries where they are prepared. Thus, in the United States, the Council of Economic Advisers analyzes the borrowing and lending activities of corporations each year (17.1) on the basis of the national income and product accounts of the Department of Commerce (17.2) rather than the Federal Reserve Board’s flow-of-funds series. This is done even though the Reserve Board’s data not only frequently differ greatly in magnitude from those of the Department of Commerce but, for the crucially important figure of net borrowing and lending by individual sectors, sometimes show flows in the opposite direction. Moreover, they are probably superior in quality to those that the Council chooses to use.

The statistical analysis of the Bank of England’s Quarterly Bulletin (16.1) is now tightly knit around the financial surpluses and deficits of the important sectors. In fact, its writing provides a model of how financial statistics should be used. However, for the United Kingdom, in addition to the surpluses and deficits derivable from the national income and expenditure accounts, financial data are shown by the records of identifiable financial transactions. These two sets of data provide radically different pictures of the transactions of individual sectors. Not only are they usually of completely different magnitude, but, more frequently than in the similar U.S. comparisons, they are of the opposite sign. There are good grounds for believing that the records of identifiable financial transactions provide more reliable guides than the data used by the Bank; yet, the Bank continues to rely on the national income and expenditure accounts rather than on its own statistics.

The French national accounts show a similar difference between the data derived from income and expenditure accounts and those derived from financial accounts. In France, however, more credence is given to the financial data. The Federal Republic of Germany and India avoid this problem by making their estimates of saving and investment on the basis of financial data (6.2 and 7.1.1).32 When the Canadians were faced with this statistical problem, they stopped publishing the financial records (2.1 and 2.2).33

It is possible to suggest that these apparent preferences for national income and expenditure records over financial data reflect two views. In the first place, the national income and expenditure accounts of most countries are aesthetically pleasant statements, while their financial accounts tend to remain as honestly messy documents. In the second place, financial policies are frequently determined by individuals who are well aware of the statistical weaknesses in the structure of financial accounts and tend to have suspicions regarding their reliability. On the other hand, they receive the national income and expenditure accounts in finished form; all the entries balance and these statements contain few of the unexplained differences which occur frequently in financial accounts. Hence, these statements appear to be reliable. The truth usually is that the financial accountant has more complete data and hence finds that errors and omissions exist, while the national income and expenditure accountant is willing to rely on residual estimates for some of his most important aggregates. Thus, to quote one example, the latter presents to the policymaker, and the policymaker (and the academic econometrician) is willing to accept, estimates of personal saving that should properly be entitled, “errors and omissions (including personal saving).” Only a few individuals like Morris Copeland have had the courage to suggest that the present Department of Commerce’s residual estimate be dropped and the Securities and Exchange Commission’s direct estimate based on financial data be substituted for it.34 It is probably fair to say that this problem in national accounting, which was only made fully apparent with the compilation of financial data and the emergence of these discrepancies, has just begun to receive the attention which it deserves.35

The Prospects


This brief review of the present theory and practice of financial accounting indicates that this branch of economic accounting is in a very underdeveloped state. We cannot be satisfied with our present accounts and must be considering the direction in which they should be progressing. The evolution of three sets of accounts may indicate the direction in which we should move. The original U.S. flow-of-funds publication (17.3.1) was bereft of balance-sheet data. The sector statements were limited to flow information. All the aggregative stock data were presented in the transactions chapters, so that it was impossible to derive measures of the total asset and liability positions, and the interrelations between sector assets and liabilities, without considerable computation. Now the accounts bring all the available totals together for each sector and subsector and thus present partial financial balance sheet data (17.3.2).

A comparison of the original (1958) Japanese flow-of-funds publication (10.1.3) with their most recent (July 1965) publication on this subject (10.1.5) reveals a similar shift in emphasis. The 1958 study spoke solely in terms of flows: “Economic activities of a nation are developed through two flows, real and monetary, the former a flow of goods and services, and the latter a flow of currency, deposits, loans, etc.” The July 1965 article (p. 12) speaks in terms of balance-sheet aggregates, referring to “a slowdown in the increase of bank lending and of monetary liquidity of corporate businesses, an expansion of trade credit, a deterioration of banks’ cash positions, and a dullness in the stock market.”

The evolution of the Rapports Annuels of the Conseil National du Crédit in France reveals a similar trend. The early reports were limited to an analysis of flows of credit, with particular attention to the direction of loans to different types of activity (5.1.1). Now attention is largely directed to the total assets and liabilities of the private sector, and to changes in holdings of different types of financial asset, with emphasis on the more liquid of these (5.1.2).

If it be accepted that an economy may be viewed from two complementary aspects, a clear guide can be given for the development of financial accounts. The current demand for supply of goods and services may be analyzed in the income and product accounts: but if investment and saving are also considered as being partly the product of attempts to change asset and liability positions, the financial accounts should be directed toward the analysis of balance-sheet data.

However, the suggestion that balance sheets should be developed is little more than a vague indication of the direction in which we should move. As yet, the available balance-sheet statistics are limited to data on liabilities and financial assets for France (5.1.2), Japan (10.1.5), Norway (14.1), and the United States (17.3.2), with some data for Yugoslavia (19.1) and more complete balance-sheet data for the corporate sector for Australia (1.2), India (7.1.1), New Zealand (13.1), and the United States (17.4.1 and 17.4.2). The fundamental problem to be faced is whether attention should be directed toward changes in “liquid” positions, i.e., essentially financial data, or should so-called “financial accounts” be more inclusive.

In order to reach conclusions regarding not only the range of assets and liabilities to be covered in a set of balance sheets, but also other aspects of financial accounts, it is essential to have a view on the role of financial and other assets and liabilities in the determination of economic decisions. Economic analyses are often presented in a form which considers the inflation or deflation problem to be one relating to savings and investment. It is argued that if, ex ante, desired investment is greater than the volume of savings arising at current income and price levels, inflation will result. Similarly, the problem of deflation is seen as an insufficiency of investment to match the savings accruing at current income and price levels. In this form of analysis, investors are considered to exert an expansionary pressure on the economy. The actions of savers are viewed as contractionary. In a period of inflation, investors are viewed as the devils. In a period of deflation, savers are the devils. If this argument, presented earlier in the description of the “financial balance” approach to financial accounting, be correct, instead of measuring the “balance of resources” in a community, that is, the difference between ex ante savings and investment, economic statistics should measure the “balance of finance” in the community, that is, the totals of net lendings and net borrowings by the significant economic sectors.

Borrowings and lendings, however, are more than financial flows. They are changes in the assets and liabilities of individual economic units. As such they are significant. Only a miser is willing to continue accumulating assets for their own sake. Only an irresponsible spendthrift is willing to incur unlimited liabilities. Assets are only acquired if they are of use. Specific types of assets are only acquired if they are of use, and if their addition to the holder’s total stock of assets will make the distribution of his stock more attractive to him. Likewise, liabilities are only incurred if the expenditure which they make possible is more attractive than the eventual cost of the liabilities. Further, reasonable people will only assume liabilities if they are of a type which the borrowers may expect to be able to repay together with the other liabilities already assumed. This ability to repay may well be influenced by the assets held as a protection against obligation. Similarly, assets are only sold if their disposal does not endanger the position of the sellers.

This argument may be presented in symbolic terms by suggesting that the Keynesian system of equations should be extended to include a further set along lines similar to the following:


Sa =Δ Af

Ia =ΔL

A =f (A,L,..)

Af=f (A,L,..)

L =f (A,L,..),

where S represents savings; Sa, savings in financial form; la, net borrowing; A, assets; Af, financial assets (i.e., claims on others); and L, liabilities.

If lending and borrowing are, in fact, conscious decisions to change assets and liabilities, it is reasonable to assume that these decisions to modify the values of stocks are partly determined by the pre-existing values of the stocks themselves. That is, if lending and borrowing are conscious decisions to alter balance-sheet aggregates, it is reasonable to assume that they are partly determined by the existing structures of the balance sheets of the individual economic units making the decisions. Further, if this be true, it follows that balance-sheet criteria must be incorporated into our economic theory. Consequently, the aggregated balance sheets of the significant economic sectors must be essential parts of our economic statistics.

A detailed discussion of the form which these balance sheets should take, or of the sectors for which they should be compiled, would be out of place here. This has been undertaken elsewhere.36 However, a few points related to both general economic theory and economic accounting might well be raised.

In the first place, it has been suggested that changes in assets and liabilities are determined in part by desires to alter existing structures of the balance sheets of individual economic units. These changes in structure are of two types. They may represent changes in the totals of assets and liabilities. They may represent changes in the composition of assets and liabilities. Consequently, any set of balance-sheet statistics should record the gross assets and liabilities of the economic sectors which are identified.

At the same time, balance-sheet statistics should record specific types of assets classified by significant criteria. On both the asset and liability sides, two divisions are of prime significance. Assets may be physical possessions, or alternatively they may be claims on others; hence the asset accounts should be separated into these two parts (with fixed capital and inventories separately identified). Liability entries may record obligations to others, or residual net worth entries; hence the liability accounts should be separated into these two parts.

Both assets and liabilities vary in type, and may be classified by diverse criteria. Probably the most significant classification is one which identifies financial assets and liabilities by liquidity. So far as changes in individual assets and liabilities alter the structure of balance sheets, they primarily shift their liquidity structures. So far as the existing structures of balance sheets influence individual economic units to desire changes in these structures, it is primarily the liquidity structures which individual units wish to alter. So far as changes in assets and liabilities are coincident (e.g., when a unit borrows in order to finance the purchase of an asset), the relation between the liquidity attributes of the assets acquired and the liabilities incurred is significant. Hence balance-sheet statistics should record the liquidity attributes of assets and liabilities.

Asset records should identify the most liquid of all assets—money. They should recognize that many other assets are close substitutes for money. Other assets are readily salable, with a greater or lesser risk of capital gain or loss, and hence liquidity attributes of varying degree. Government securities have an assurance of repayment, and hence a liquidity value different from that of private obligations. Foreign assets have particular liquidity attributes. Life insurance policies have a liquidity value to their holders, in addition to their security attributes. These and other classifications should be recorded in balance-sheet statistics.

Similarly, liabilities have liquidity attributes. The willingness of any economic unit to invest will be determined, in part, by the structure of its liabilities. The decisions made by a unit with large short-term debts to a bank will be different from the decisions made by a unit whose major indebtedness is of a long-term nature. When any outstanding indebtedness to a bank is expected to be renewed, it differs in significance from indebtedness which the borrower knows must be met in money at maturity. Hence liability accounts also should recognize the liquidity aspects of different types of debt.

Valuation problems

One final problem should be mentioned. It relates to the valuation of balance-sheet entries. Financial assets are also liabilities. What is one man’s asset is another man’s debt. Hence, it might be thought that the records of these items should be entered at the same money value in the balance sheets of the asset holders and the debtors. This is a widely accepted view. Thus Stamp has used essentially the capitalized value of equity income as a measure of the national wealth.37 Hicks has been specific on this point: “The balance sheet of ‘companies’ and of ‘government’ must frankly be adjusted so as to maintain consistency with the personal sector. The shares and bonds, as they appear in the balance sheet of the ‘companies’ sector must be entered at the values which have been given them in the balance sheet of the personal sector, not at the values given them by the companies. In spite of this, we must hold to the principle that the net assets of companies are nil. This means that we must not attempt to value the real assets of companies directly. We must accept the ‘shareholders value’ of these real assets—not the value which is set upon them by the company, but the value which is implied in the market value of the shares.”38 Copeland has accepted the idea that items should be consistently valued in creditor and debtor records. In outlining the principles on which sector balance sheets should be compiled, he proposed to “either adjust the holders’ statement or agree with that of the holder.”39

Goldsmith has stated both that a set of sector balance sheets should be compiled so that “When all intragroup and intergroup relationships are eliminated we obtain instead of a combined national balance sheet a national wealth statement”40 and that those who support the view expressed in this paper are “subjectivist” as opposed to the “objectivists” referred to in this paragraph and in the preceding one.41 Edey and Peacock have referred to differences in creditor and debtor valuations as “inconsistencies [which] do, of course, occur, and we have to accept the fact that this is one of the aspects of life not susceptible to satisfactory treatment in accounts.”42 Revell has recognized some of the difficulties involved in this approach43 but adopts it because he believes that a market price valuation of liabilities is more valuable in determining debtor responses. The National Accounts Review Committee has gone as far as to say: “Statements, to make economic sense, must be based on balance sheets of the component units which are uniform with respect to scope of assets and liabilities and to their valuation.”44

Yet, if one is interested in significant economic relations, there is no reason to believe that asset holders and debtors have symmetrical views of the values of individual financial obligations. An attempt should be made to record the values which reflect the views that asset holders have of their wealth, and the views that debtors have of their liabilities. In particular, interest should center on the liquidity attributes which assets and liabilities have for their holders and debtors. For marketable assets (bonds, other securities, and even mortgages), market prices of the assets are probably the most significant to the holders of these assets. Hence, for these items, it is their market values which should be recorded in asset accounts. That is, in the equations suggested above, the assets at least should be measured at current market values. The market value of liabilities is probably of only remote significance to debtors. Hence liability accounts should probably record liabilities at their nominal value. For other financial claims, there are similar disparities between the appropriate asset record and the appropriate liability record. For example, the significant value of a life insurance contract to an insurance company may be the actuarial reserve associated with the contract. The actuarial reserve value of a life insurance contract is a meaningless figure to the policy holder. To him both the loan value and the face value are meaningful. Consequently, the asset records should try to accommodate both these values while the liability records should be limited to the reserves held against policies.

The different treatment of reserves against life insurance policies which are appropriate for balance-sheet statistics, and for the flow-of-savings statistics, provides an example of the differences between the proper approaches to the two types of statistics. Increases in insurance reserves provide an adequate measure of personal savings in the form of life insurance, and hence have a place in the savings records of the personal sector, even though they may be meaningless entries for a set of personal balance sheets.

The principles underlying these suggestions for different valuations of assets and liabilities lead to a consideration of the manner by which balance-sheet criteria may influence economic decisions. If anything occurs to alter the market values of assets, asset holders are likely to feel both wealthier (or poorer) and more liquid (or less liquid). That is, their balance-sheet positions will be modified. They are likely to react to these changes by trying to alter both the totals and the structures of both their assets and their liabilities. They are likely to wish to lend or to borrow, or to do both. In other words, on the basis of the earlier arguments, they are likely to try to exert expansionary or contractionary pressures on the economy. It is through this mechanism that the considerations outlined here, relating to both stocks and flows, are brought together. In symbolic terms, this may be expressed by saying that:

A=α (i, L..)


A=α’ (i,A,L..)


Sa=β (i,A,L..)


Ia=γ (A,L)=β (i,A,L..)

where i represents the rate of interest.

It is through its effects on balance sheets, and their subsequent effects on expenditure that financial policy can play part of its role in the general field of economic policy. The most important problem before the financial theorist, and perhaps the outstanding problem facing economists today, is to investigate the nature of the relations between changes in balance-sheet positions and the flow of resources in the economy.45


I. Alternative Measures of Borrowing and Lending

A number of countries provide alternative statistics on the financing of household saving and of company investment. A comparison of these alternative measures for five countries for which data are readily available indicates important discrepancies between them. The evidence of these discrepancies suggests that, for these countries at least, one or both of the sets of data must be treated with reserve. On the basis of the scanty, and admittedly conjectural, evidence presented in this appendix, it may be suggested that for the OECD countries, it is possible that the traditional national income and expenditure estimates of household saving and company investment tend to be underestimations of the order of 10-15 percent.


Up to 1956, the Bank of Canada published direct estimates of personal saving. These are presented in Table 1. They indicate that for the period covered, the direct estimates (largely derived from financial records) were, on the average, almost 25 per cent higher than the national income residual estimates.

Table 1.

Canada: Alternative Measures of Personal Sector Saving, 1950-55

article image
Source: 2.1 (Appendix II).


The “Comptes de la Nation” is a self-balancing set of accounts, except that its statement of “Créances et Dettes” contains an “Ajustement” item that is, in fact, a discrepancy entry. As indicated in Table 2, the income and expenditure data suggest that French income and expenditure estimates of household saving are approximately 15 per cent lower than the comparable data derived from financial data.

Table 2.

France: Alternative Measures of Household Saving, 1961-64

(In billions of francs)

article image
Source: Ministère de Finances, Statistiques et Etudes Financières, July-August 1965.

For companies, the difference appears to be of the order of 5 per cent (Tables 3 and 4) with the income and expenditure accounts suggesting that company investment was higher than the amount that would be consistent with the financing data. An “error” of this magnitude may not be regarded as serious. Yet, when the year-to-year movements in the adjustment entry are compared with similar movements in the total estimated investment, it appears not impossible that the estimated change in investment from one year to the next may be even more than one third in error.

Table 3.

France: Alternative Measures of Net Company Borrowing, 1961-64

(In billions of francs)

article image
Source: Ministère de Finances, Statistiques et Etudes Financières, July-August 1965.
Table 4.

France: Comparison of “Ajustement” and “Investissement” Entries in French Accounts, 1961-64

article image
Source: Ministère de Finances, Statistiques et Etudes Financières, July-August 1965.


Comparisons of the Japanese data indicate discrepancies between the income and expenditure data and the flow-of-funds data for financing by the household and company sectors. Table 5 suggests that the financial data show that households have tended to borrow, on balance, almost 15 per cent more than the amounts indicated by the income and expenditure accounts. If the financial data are added to the income and expenditure data on home building, to derive an estimate of total household saving, this total tends to be almost 10 per cent higher than that recorded in the income and expenditure accounts.

Table 5.

Japan: Comparison of National Income and Expenditure and Flow-of-Funds Data on Household Borrowing, 1957-64

(In billions of yen)

article image
Sources: Bank of Japan, Economic Statistics Monthly, March 1966, and Monthly Economic Review, July 1965.

Comparison of company borrowing and lending data (Table 6) suggests a discrepancy that tends to approximate 35 per cent between these two sets of data. If it is assumed that the income and expenditure records of company investment underestimate this flow by the amount of discrepancy in Table 6, this would mean that company investment in Japan tends to be understated by almost 10 per cent.

Table 6.

Japan: Comparison of National Income and Expenditure and Flow-of-Funds Data on Company Borrowing, 1957-64

(In billions of yen)

article image
Sources: Bank of Japan, Economic Statistics Monthly, March 1966, and Monthly Economic Review, July 1965.

United Kingdom

At the present time, the U.K. authorities are re-examining and revising their estimates of household saving.46 Hence, any observations on the presently available statistics must be tentative. Even so, the comparison of the income and expenditure statistics of household sector saving and the comparable estimates derived from financial data, presented in Table 7, indicate a large discrepancy between the two series.

Table 7.

United Kingdom: Comparison of Household Sector Savings Derived From Income and Expenditure and Financial Accounts, 1963–65

(In millions of pounds sterling)

article image
Source: U.K. Central Statistical Office, Financial Statistics, April 1966.

The United Kingdom presents data on company borrowing derived from the national income and expenditure data, and also from financial records. These two sets of data are compared in Table 8. In this juxtaposition, the national income and expenditure data suggest that British companies were net lenders in 1963-65,47 while the financial statistics show them as net borrowers in these years. The discrepancy between the two sets of data were of the order of magnitude of 20 per cent of estimated total investment. This discrepancy suggests that one or both sets of data may be subject to important error. It seems more likely that companies would be net borrowers than net lenders. If they were the latter, it would suggest that British industry was in the process of liquidation, converting itself from productive enterprises to investment trusts. Therefore, the income and expenditure data seem to be the ones where the most serious error is likely to be found.

Table 8.

United Kingdom: Comparison of Estimates of Net Company Borrowing, 1963-65

(In millions of pounds sterling)

article image
Source: U.K. Central Statistical Office, Financial Statistics, April 1966.

This suspicion is strengthened by a comparison of the partial and “complete” statistics on company financing. The Board of Trade provides an analysis of the accounts for about 2,000 of the largest manufacturing and distribution companies whose shares are quoted on U.K. stock exchanges. Only the 1963 data for this group of companies can be compared with the other data on company financing.48 For that year, it is possible to estimate “derived accounts” for the “quoted” and “unquoted” companies. This is done in Table 9 (derived from Tables 1012), where markedly different pictures are given by the income-expenditure and the financial data. The income-expenditure comparison suggests that whereas the “quoted” companies were, as might be expected, net borrowers on a fairly large scale, the “unquoted” companies were net lenders on an even larger scale. It is rather difficult to believe that these two groups of companies have behaved in such markedly different fashion in recent years. This suspicion is confirmed by the comparison of the data from financial sources for the two groups of companies, from which it appears that in 1963 the “quoted” companies were net borrowers, and the “unquoted” were small net lenders.

Table 9.

United Kingdom: Comparison of Estimates of Net Borrowing by Quoted and Other Companies, 1963

article image
Source: Tables 1012.
Table 10.

United Kingdom: Estimates of Expenditure and Saving of Quoted and Other Companies, 1963

article image
Sources: Table 8 and U.K. Central Statistical Office, Financial Statistics, February 1966.
Table 11.

United Kingdom: Estimates of Net Borrowing by quoted and Other Companies Based on Expenditure and Savings Data, 1963

article image
Source: Table 10.
Table 12.

United Kingdom: Estimates of Net Borrowing by Quoted and Other Companies Based on Financial Data, 1963

article image
Sources: Table 8 and U.K. Central Statistical Office, Financial Statistics, February 1966.

United States

The Federal Reserve Board’s flow-of-funds statistics are also a self-balancing set of accounts that contain, as discrepancy entries, the differences between the flows as measured from financial data and as measured from income and expenditure data. The comparison of the totals for household savings derived from these two sources is presented in Table 13, which indicates that the income and expenditure estimates are, on the average, approximately 5 per cent less than the corresponding flow-of-funds data.

Table 13.

United States: Alternative Measures of Household Sector Saving, 1953-65

(In billions of dollars)

article image
Sources: Federal Reserve System, Board of Governors, Division of Research and Statistics, Flow of Funds, Annual, 1953-64, and Flow of Funds Seasonally Adjusted, 1963-1965.

There is a similar discrepancy in the company accounts. As suggested in Table 14, the net borrowing of nonfinancial corporations as measured from an examination of their financial accounts is some four times the total estimated on the basis of income and expenditure data. The discrepancy between these two series is comparable to that for the series on net lending by households (Table 15). Thus, there is evidence that one or both of these series contains errors or omissions. If the errors are primarily in the income and expenditure data, the estimates of company investment should be raised by something in the order of the discrepancy in the flow-of-funds accounts, as is done in Table 16. A comparison of this “adjusted” estimate with the “original” estimate suggests that year-to-year changes in the widely accepted income and expenditure account estimates of company investment requiring financing from internal or external sources may be in the opposite direction from that of the changes actually taking place, and, if not in the wrong direction, may, on the average, well be over 100 per cent in error.

Table 14.

United States: Alternative Measures of Net Borrowing of Nonfinancial Corporations, 1953-65

(In billions of dollars)

article image
Sources: Federal Reserve System, Board of Governors, Division of Research and Statistics, Flow of Funds, Annual, 1953-64, and Flow of Funds Seasonally Adjusted, 1963-1965.
Table 15.

United States: Alternative Measures of Personal Sector Net Lending, 1953-65

(In billions of dollars)

article image
Sources: Federal Reserve System, Board of Governors, Division of Research and Statistics, Flow of Funds, Annual, 1953-64, and Flow of Funds Seasonally Adjusted, 1963-1965.
Table 16.

United States: Alternative Estimate of Capital Expenditure by Nonfinancial Corporations, 1953-65

(In billions of dollars)

article image
Sources: Federal Reserve System, Board of Governors, Division of Research and Statistics, Flow of Funds, Annual, 1953-64, and Flow of Funds Seasonally Adjusted, 1963-1965.

II. Bibliographical Appendix

The following official and semiofficial documents are referred to in this paper:


1.1 Reserve Bank of Australia, Flow-of-Funds, Australia, 1953/54 to 1961/62, Staff Paper (July 1965).

1.2 Reserve Bank of Australia, Flow-of-Funds, Australia. Statistical Bulletin, Company Supplement (October 1965).


2.1 Bank of Canada, “Direct Estimate of Personal Saving in Canada: 1950-1955,” Statistical Summary (October 1956, the last occasion on which this table was published), p. 304.

2.2 Royal Commission on Canada’s Economic Prospects, L.M. Read, S.J. Handfield-Jones, and F.W. Emmerson, “A Presentation of National Transactions Accounts for Canada, 1946-1954,” in Financing of Economic Activity in Canada, by W.C. Hood (1959).


3.1 Economic Survey of Denmark, 1965.


4.1 “Transfers of Income Between Public and Private Sectors, Their Disposable Income and Its Utilization in 1962-1964,” Economic Survey, 1964, p. 71.


5.1.1 Conseil National du Crédit, Premier Rapport (1947).

5.1.2 Conseil National du Crédit. Dix-Huitième Rapport Annuel, Année 1963 (1964).

5.2 Institut National de la Statistique et des Etudes Economiques, Etudes et Conjoncture (May-June 1964), pp. 268-77.

5.3 Ministère des Finances, Service des Etudes Economiques et Financières, Rapports sur les Comptes Provisoires de la Nation de I’Année 1953 et sur le Budget Economique de I’Année 1964 et Compte Rendu des Travaux de la Commission des Comptes et des Budgets Economiques de la Nation (1954).

Federal Republic of Germany

6.1 Deutsche Bundesbank, “Wealth Formation and Its Financing in 1964,” Monthly Report (April 1965).

6.2 Deutsches Institut für Wirtschaftsforschung, “Zur Wirtschaftslage der Bundesrepublik Deutschland,” Vierteljahrshefte zur Wirtschaftsforschung, Erstes Heft (1964).


7.1.1 Reserve Bank of India, “Finances of Private Limited Companies 1962-63,” Reserve Bank of India Bulletin (February 1965).

7.1.2 Reserve Bank of India, Reserve Bank of India Bulletin. “Estimates of Saving and Investment in the Indian Economy: 1950-51 to 1962-63,” Reserve Bank of India Bulletin (March 1965).


8.1 Bank of Israel Annual Report, 1963-1964.


9.1 Bank of Italy, “National Financial Assets and Liabilities,” Abridged Version of the Report for the Year 1964, p. 114.


10.1.1 Bank of Japan, Analysis of Financial Statements of Main Industrial Corporations in Japan (issued semiannually).

10.1.2 Bank of Japan. Flow of Funds Accounts in Japan, 1954-63.

10.1.3 Bank of Japan. Monetary Development, 1954-1957, and Essential Features of the Financial Structure of JapanA Study of Money flow in Japan (1958).

10.1.4 Bank of Japan. “Recent Developments of the Flow-of-Funds in Japan,” Monthly Economic Review (January 1965).

10.1.5 Bank of Japan. “Flow-of-Funds in the Japanese Economy in 1964,” Monthly Economic Review (July 1965).


11.1 Banco de México, “Activos Liquidos de Empresas y Particulares,” lnforme Anual, 1963, p. 119 (not repeated in lnforme Anual, 1964).


12.1.1 Central Bureau of Statistics, “Uitkomsten van enige berekeningen betreffonde het nationale vermogen in Nederland in 1938” [Results of some calculations on the National Wealth of the Netherlands in 1938], Statistische en Econometrische Onderzoekingen [Statistical and Econometric Studies], No. 3 (1947).

12.1.2 Central Bureau of Statistics. H. Rijken van Olst, B. Korn, and C.A. Oomens, “Het verband tussen de nationale balans en het stelsel der nationale jaarrekeningen” [The Connection Between the National Balance-Sheet and the System of National Accounts], Statistische en Econometrische Onderzoekingen, No. 3 (1950). (This paper was discussed and translations of its tables published in F. Sewell Bray, “A National Balance Sheet,” Accounting Research (July 1951), and reprinted as Reprint Series No. 44, University of Cambridge, Department of Applied Economics (1951).)

12.1.3 Central Bureau of Statistics. “Nationale Rekeningen voor Nederland 1938 en 1946-1953” [National Balance Sheet for the Netherlands 1938 and 1946-1953], No. 2 (1954).

12.1.4 Central Bureau of Statistics. “Nationale Rekeningen voor Nederland 1938 en 1947-1954,” No. 2 (1955).

12.2.1 Central Planning Bureau, Centraal Economisch Plan, 1953.

12.2.2 Central Planning Bureau. Centraal Economisch Plan, 1965.

12.3.1 De Nederlandsche Bank, Report for the Year 1953.

12.3.2 Central Planning Bureau. Report for the Year 1964.

New Zealand

13.1 Reserve Bank of New Zealand, New Zealand Public Company Financial Statistics, Supplement to Reserve Bank of New Zealand Bulletin, (December 1965).


14.1 Statistick Sentralbyrå, Kreditmarkedstatistikk 1963.


15.1 “Financial Saving,” Revised National Budget, 1965, p. 16.

15.2 Sveriges Riksbank, Riksbankens Förvaltningsberättelse för år 1965.

United Kingdom

16.1 Bank of England, “Analysis of Financial Statistics,” Quarterly Bulletin (March 1966).

16.2 Central Statistical Office, “More Light on Personal Saving,” Economic Trends (April 1965).

United States

17.1 Council of Economic Advisers, “Sources and Uses of Corporate Funds, 1954-65,” Annual Report, January 1966, p. 287.

17.2 Department of Commerce, “National Income and Product in 1965,” Survey of Current Business (January 1966).

17.3.1 Federal Reserve System, Board of Governors, Flow of Funds in the United States 1939-1953 (1955).

17.3.2 Federal Reserve System, Board of Governors. Federal Reserve Bulletin (January 1966), pp. 106-114.

17.4 Federal Trade Commission and Securities and Exchange Commission, Quarterly Financial Report for Manufacturing Corporations.

17.5 Securities and Exchange Commission, “Working Capital of U.S. Corporations,” Statistical Bulletin (April 1966).


18.1 Banco Central de Venezuela, Informe Económico 1964.


19.1 National Bank of Yugoslavia, Annual Report, 1964.

La comptabilite financiere: situation actuelle et perspectives d’avenir


Cet article passe en revue les statistiques nationales disponibles à I’heure actuelle sur les emprunts et les prêts à I’intérieur des collectivités et sur les bilans nationaux. Les 17 pays sur lesquels il porte comprennent pratiquement tous ceux pour lesquels on dispose actuelle-ment de ces données, et deux autres pour lesquels elles existaient dans le passé.

L’auteur avance quelques suggestions pour le développement futur des statistiques financières, en insistant, sur I’importance des renseignements fournis par le bilan dans Fanalyse des facteurs qui influencent les décisions des membres d’une collectivité.

L’article expose également les grandes lignes d’une théorie des rapports entre les modifications de la structure des bilans nationaux et les décisions d’investissement et d’épargne, avec leurs conséquences sur la production, les prix, etc.

Contabilidad financiera: estado actual y perspectivas


Este artículo reseña las estadísticas nacionales de que actualmente se dispone referentes a la obtención y concesión de préstamos en comu-nidades, así como a balances nacionales. Comprende 17 países, entre los cuales se cuentan, junto con dos países sobre los que se informara anteriormente, prácticamente todos aquellos de cuyos datos se dispone en la actualidad.

Se sugieren algunas ideas para el desarrollo futuro de las estadísticas financieras, subrayando la trascendencia que tienen los datos del balance cuando se analizan las fuerzas que influyen en las decisiones de los miembros de la comunidad.

El artículo propone una teoría—cuyos lineamientos generales avanza—sobre la relación entre los cambios de estructura de los balances nacionales y las decisiones conducentes a la inversión y el ahorro, con los consiguientes efectos sobre la producción, los precios, etc.


Mr. Dorrance, Chief of the Financial Studies Division, has been a lecturer at the London School of Economics and a member of the staff of the Bank of Canada. This article is an extended version of a lecture given on the International Dimensions Program, University of Pittsburgh, on June 11, 1965. The extensions of the Pittsburgh lecture draw heavily on two earlier articles: “Consideration de Algunas Relaciones Keynesianas Fundamentales,” Revista de Desarrollo Economico (Buenos Aires, January-March 1960); and “A Review of Some Fundamental Keynesian Relations,” Pakistan Economic Journal (Karachi, Summer 1961).


J.J. Polak, “Financial Statistics and Financial Policy,” Staff Papers, Vol. VII (1959-60), p. 1.


Ibid., p. 8.


J.R. Hicks, The Social Framework (Oxford, Clarendon Press, 1942), p. 3.


Allan H. Meltzer, in testimony before the Subcommittee on Domestic Finance, Committee on Banking and Currency, U.S. Congress, House of Representatives (88th Cong., 2nd Sess., February 11, 1964).


Third Edition, Washington, D.C., IMF, July 1961.


United Nations, Statistical Office, A System of National Accounts and Supporting Tables, Studies in Methods, St/Stat/Ser.F/No. 2 (New York, 1953; hereafter referred to as SNA).


Jacob Viner, Canada’s Balance of International Indebtedness 1900-1913 (Cambridge, Harvard University Press, 1924).


H.D. White, The French International Accounts 1880-1913 (Cambridge, Harvard University Press, 1933).


J.H. Williams, Argentine International Trade Under Inconvertible Paper Money, 1800-1900 (Cambridge, Harvard University Press, 1920).


See F.W. Taussig, International Trade (New York, Macmillan, 1927).


See J.M. Keynes, The General Theory of Employment, Interest and Money (London, Macmillan; New York, Harcourt Brace, 1936).


See E.F. Denison, “Report on Tripartite Discussions of National Income Measurement,” Studies in Income and Wealth, Vol. 10, National Bureau of Economic Research (New York, 1947) for a discussion of the development of these accounts.


Committee on the Working of the Monetary System, Report, Cmnd. 827 (London, 1959). The liveliest discussion of this report was conducted in The Banker in 1959 and 1960: “No Verdict on Money,” pp. 491-96; “The Report Analysed,” pp. 497-534; Sir Oscar Hobson, “As I See It,” pp. 542-46; E.V. Morgan, “What Role for Interest Rates?” pp. 584-90; F.W. Paish, “What is this Liquidity?” pp. 590-97; T. Balogh, “A Post-Mortem on Freedom?” pp. 597-604; J. Spraos, “Control by ‘Stickiness’ of Rates?” pp. 674-80; and Sir Dennis Robertson, “A Squeak from Aunt Sally,” pp. 718-22—all in Vol. CIX—and B. Tew, “A Case for Financial Controls?” pp. 28-33; H.B. Rose, “Another Look at ‘Liquidity’,” pp. 160-67; “Monetary Policy in Action,” pp. 223-40; Alan Day, “Now for its Critics,” pp. 404-409; and T. Wilson, “Radcliffe Revisited,” pp. 647-51—all in Vol. CX. Other important contributions to this discussion are R.S. Sayers, “Monetary Thought and Monetary Policy in England,” The Economic Journal, Vol. LXX (1960), pp. 710-24, and also in The Banker, October 1960, pp. 671-83; J.S. Gurley, “The Radcliffe Report and Evidence,” American Economic Review, Vol. L (1960), pp. 672-700; and E. P. Neufeld, “The Implications of the Radcliffe Report for Canada,” The Canadian Journal of Economics and Political Science, Vol. 26 (1960), pp. 413-27.


Hicks, op. cit.


See Morris A. Copeland, A Study of Money flows in the United States, National Bureau of Economic Research (New York, 1952), p. 3.


Robert Triffin, “Esbozo General de un Análisis de las Series Estadísticas Monetarias y Bancarias de América Latina Sobre Bases Uniformes y Comparables,” Memoria Primera Reunión de Técnicos Sobre Problemas de Banca Central del Continente Ámericano (Banco de Mexico, S.A., México City, 1946), pp. 410-14.


All the official and semiofficial documents referred to in this paper are listed alphabetically, by country, in Appendix II, Bibliographical Appendix (see p. 225), and identified in the text by their numbers in Appendix II.


Hicks, op. cit.


R.W. Goldsmith, R.E. Lipsey, and Morris Mendelson, Studies in the National Balance Sheet of the United States (Princeton, Princeton University Press, 1963).


Measuring the Nation’s Wealth (New York, Colombia University Press, 1964); U.S. Congress, Joint Economic Committee, Hearings Before the Subcommittee on Economic Statistics (89th Cong., 1st Sess., 1965), pp. 1-21.


J.R.S. Revell, “The National Balance Sheet of the United Kingdom,” Department of Applied Economics, Cambridge University, 1965 (privately circulated).


Copeland, p. 214 of work cited in fn. 15.


SNA, p. 10. It might be noted that this comment on canceling out might be made of almost any pair of entries in a system of double-entry accounts.


Richard Stone, “Functions and Criteria of a System of Social Accounting,” in Income and Wealth, International Association of Research in Income and Wealth, Series I (Cambridge, England, Bowes & Bowes, 1951), p. 1.


See Gerard R. Aubanel, Graeme S. Dorrance, and Earl Hicks, “Monetary Analyses,” Staff Papers, Vol. V (1956-57) for a description of most of these national statements.


Transfers may be considered gross as additions to income or current expenditure, or separated into transfers on current account (additions to income and consumption) and on capital account (additions to saving and investment). On either basis, they have the same effect on measures of financing.


There are other arguments regarding the economic significance of self-financed investment, but they are not relevant to the problem considered here.


Here, and subsequently, the term “lending” is used to comprise direct lending, the purchase of financial assets, and the repayment of debt. Consistently, “borrowing” includes direct borrowing, the sale of financial assets, and the realization of debts due from others.


These data are based on Goldsmith and others, op. cit.


These data are based on Banco Central de la República Argentina, Boletín Estadístico, and material supplied to the author by the Banco Central.


On the basis of Goldsmith’s measure of the financial interrelations ratio, the United Kingdom with an FIR of 2.15 is the financially most complex society in the world. (The FIR for the United States is 1.25; for Mexico 0.60, for Guatemala 0.30, and for Ethiopia 0.15.) See Organization for Economic Cooperation and Development, The Determinants of Financial Structure, Development Centre Studies No. 2 (Paris, 1965).


In the Federal Republic of Germany the estimates of the Bundesbank (6.1) differ only slightly from those of the Institut für Wirtschaftsforschung (6.2), and the Bank’s analysis is based on its own estimates.


The comparable national income and product data for Canada, France, the United Kingdom, and the United States are presented in Appendix I.


Morris A. Copeland, “Proposal for a Revised Set of Summary Accounts and Supporting Financial Details,” Studies in Income and Wealth, Vol. 22, National Bureau of Economic Research (Princeton, Princeton University Press, 1958) p. 340.


E.g., D.J. Smyth, “Saving and the Residual Error,” Bulletin of the Oxford University Institute of Economics and Statistics, Vol. 26 (August 1964).


See Graeme S. Dorrance, “Balance Sheets in a System of Economic Accounts,” Staff Papers, Vol. VII (1959), pp. 168-209.


Josiah Stamp, Wealth and Taxable Capacity (London, P.S. King & Son, 1936) and The National Capital and Other Statistical Studies (London, P.S. King & Son, 1937).


Hicks, op. cit. (2nd ed., 1952), pp. 277-78.


Morris A. Copeland, “The Feasibility of a Standard Comprehensive System of Social Accounts,” Problems in the International Comparison of Economic Accounts, in Studies in Income and Wealth, Vol. 20, National Bureau of Economic Research (Princeton, Princeton University Press, 1957), p. 77.


R.W. Goldsmith, A Study of Saving in the United States (Princeton, Princeton University Press, 1955-56), Vol. Ill, p. 4.


R.W. Goldsmith, “The Uses of National Balance Sheets” (a paper presented to the 1965 meeting of the International Association for Research in Income and Wealth).


H.C. Edey and A.T. Peacock, National Income and Social Accounting (London, Hutchison’s House, 1954), p. 214.


Revell, op. cit., p. 10.


National Accounts Review Committee of the National Bureau of Economic Research, “The Accounts, Review, Appraisal and Recommendations,” Hearings Before the Subcommittee on Economic Statistics, U.S. Congress, Joint Economic Committee (85th Cong., 1st Sess., October 29 and 30, 1957), pp. 28-29.


One subsidiary point might be mentioned. Is it unrealistic, or would it be too expensive in real resources, to contemplate the extension of economic statistics to include balance sheet records? It can reasonably be argued that it is neither unrealistic nor too expensive. If we are prepared to devote considerably more resources to the development of balance-sheet statistics than are now directed to this study, but considerably less than is now allocated to the production of national income statistics, success in this field should be relatively easy. The conceptual problems are not greater than many of the problems which have faced economists and statisticians in the past. A group of eminent U.S. economists might be quoted who declared that “the concepts of the national balance sheet and the national wealth statement are essentially not more difficult—indeed they are probably simpler—than those of national income and product” (The National Committee on Review and Appraisal of National Economic Accounts, op. cit., p. 250). On the basis of his experience with the U.K. data, Revell was “convinced that there are no inseparable problems in producing national balance sheets for the United Kingdom annually” (op. cit., p. 17).


See 16.2 (Appendix I)


1963, 1964, and 1965 are the only years compared here because they are the only ones for which adequate financial data are available. However, the income and expenditure data for all the postwar years provide the unlikely suggestion that U.K. manufacturing and distribution companies have been net lenders.


As of the time of writing (mid-May 1966).