[발췌: H. Arndt's] Measuring Trade in Financial Services (1984)

출처: H. Arndt. “Measuring Trade in Financial Services.” ^Banca Nazionale del Lavoro Quarterly Review^, No. 149, June 1984.
자료: PDF Link


※ 발췌 (excerpt):

 There has always been trade in financial services.[n.1]  Bankers have provided financial intermediation, foreign exchange market and other financial services across national boundaries. But the importance of such trade has increased greatly in the past two decades withe internationalisation of banking and the growth of international financial centres as the bases for offshore currency markets and other activities linking national capital markets. The question naturally arises how to measure the value and the growth of this trade in financial services. The issue involved in this question are closely related to the difficulties that have been encountered in the treatment of financial enterprises in social accounts for national economies. It is best approached by going bank to these issues.

[n.1] This paper deals with financial services other than insurance.

Financial Services in Closed Economies

The crux of the problem was clearly explained in the 1947 memorandum by Richard Stone which laid the foundations for the original United Nations system of national accounts. "If we treated banks (and other financial intermediaries) like ordinary businesses, we should show their sales proceeds simply their charges to customers, as a consequence, a deficit rather than a surplus would appear on the other side of the operating account. In practice, this deficit would be so large that the property income generated in banking and even perhaps the whole income generated in banking would appear to be negative. This is clearly unsatisfactory."[n.2]

[n.2] ^Measurement of National Income and the Construction of Social Accounts^, "Appendix: Definition and Measurement of the National Income and Related Totals" By Richard Stone, United Nations, Geneva, 1947 (hereafter cited as "Stone Memorandum"), p. 40.

The solutions to this conundrum which have been adopted by social accountants fall into three classes corresponding to the three main functions of banks, the creation of money, the provision of payments mechanism and related services, and financial intermediation; or, in the words of an early contributor to the debate, "loan services, clearance or transfer of circulating medium services, and the creation and maintenance of circulating medium."[n.3]  Those who have put the emphasis on the creation of money function have regarded the services of banks as being provided to the community at large and have treated them, by analogy with government services, as final products and thus as contributing to gross domestic product (GDP). Those who have emphasised the payments mechanism function have regarded the services of banks as being rendered primarily to depositors and have divided them into those rendered to households, considered as final products and therefore included in GDP, and those rendered to enterprises, considered as intermediate products and therefore excluded from GDP. Those, finally, who have focused on the intermediation function have viewed bank services as being rendered mainly to borrowers and have therefore treated them wholly intermediate products used as inputs by enterprises (except sometimes for those associated with consumer loans, considered as a service to households).[n.4]

Services to the Community. The first approach has not been widely used. But it was proposed in Australia by H. P. Brown and was employed in the Australian national accounts from 1947 until 1972.[n.5]  Brown argued that the function of banks was essentially similar to a major functin of government, "oiling the wheels of industry and the community generally".[n.6]  He therefore proposed that "as for governments, the contribution of banks to the national income is equal to the wages and salaries paid by banks."[n.7] In effect, though not in rationale, Brown's approach was similar to the "aggregation of individuals" approach originally proposed by Kuznets and used by the US Department of Commerce until 1947, except that Brown excluded bank profits on the ground that they were merely a channel through which surplus of trading enterprises was passed on to shareholder-depositors, so that their inclusion would have involved double counting."[n.8]  The crucial feature of both versions was that the output of banks, valued at factor incomes (including or excluding profits) in banking, was included in national income without any deductions in other parts of the economy. In treating the whole output of financial enterprises as final products, this approach certainly involved double counting (though not more so than in the case of government services) and has therefore been generally discarded.

Services to Depositors. The second approach was first worked out in the US Department Commerce and then adopted in the first United Nations system of national accounts on the recommendation of a committee chaired by Richard Stone. Its rationale was explained by Stone as follows: “Financial enterprises require special treatment in view of the different functions they perform and the method they adopt in charging for services. Consider commercial banks as a typical example. On the one hand, they provide services to their customers in the form of keeping their accounts for them and providing advice on various financial matters. For this they make a charge which, in many cases, is inadequate. On the other hand, they lend the money deposited with them, whether as a result of their own activities or not, and from this receive a net return large enough to enable them to subsidise the other aspect of their business.”[n.9]

  To overcome this difficulty, he proposed the following procedure: “(a) An income is imputed to bank depositors for the use of their money equal to the excess of interest and dividends received by banks over interest paid out and (b) this income is assumed to be used in ‘paying’ for uncharged banking services. 

• In the case of persons, this imputed income and outlay appears on either side of the revenue account of persons, 
• but, in the case of enterprises of all kinds, the imputed outlay is charged to operating account, thus diminishing the surplus of the enterprises, while the imputed income is credited to the appropriation account, thus restricting the effect of the adjustment to the operating surplus alone... 

The allocation of the total amount imputed between persons and businesses can only be based on knowledge of the bank expenses incurred in respect of, but not charged to, these two types of depositors.”[n.10]  In effect, depositors were regarded as receiving, in return for depositing their money with banks, an income in kind in the form of payments mechanism and other services, in much the same way as factory workers may receive, in addition to their cash wages, free canteen meals.

  When this approach was adopted by the US department of Commerce the allocation of the imputed bank service charge between households and enterprises, and thus between final and intermediate products, was based on deposit ownership.[n.11]  When the same approach was adopted in the first UN system of national accounts (SNA), the same procedure was recommended, though the difficulty of inadequate statistics of deposit ownership in many countries was acknowledged. “The contribution of banks, etc. to gross domestic product is here evaluated by imputing to depositors a service charge equal to the excess of investment income accruing to these institutions and by imputing at the same time a corresponding amount of income to depositors. Thus these imputations do not change the income of banks or of other enterprises, but they result in a change in the industrial classification of domestic product (from other enterprises to banks) in so far as the imputation is made in respect of business deposits, and in an increase in the domestic product to the extent that the imputation is made in respect of the deposits of household, etc. The main problem in applying this rule is of a statistical character, but the total amount involved in most countries is small” and inaccuracies hardly serious.[n.12]

[n.11] U.S. Department of Commerce, ^National Income 1951 Edition^, cited R.E. Speagle & L. Silverman, op. cit., p. 130.
[n.12] ^A System of National Accounts and Supporting Tables^, United Nations, New York, 1953, [ST/STAT/SER F./No.2], p. 32.

  In the following years, the second approach came under heavy criticism, chiefly on the ground that it misconceived the functions of banks. One exposition of this view criticised the Department of Commerce interpretation of banking function as “primarily a matter of keeping accounts for depositors, including the mechanical operation of monetary transfers... As far as these lending institutions are concerned, lending and investing functions are performed free of charge to borrowers. The creation of purchasing power through the lending process is passed over.”[n.13]  In the view of these critics, "the chief business of commercial banks ... consists of two things: the expert management and investment of funds belonging to the banks' owners and creditors and, of equal importance under a fractional reserve system, the simultaneous creation of money." Since the capacity of banks to provide credit depends on their ability to maintain the convertibility of deposits, even "the expense of an elaborate clearing machinery set up expressly to keep deposit money freely convertible" should be regarded as part, indeed "a major part of lending costs."[n.14]

  Apart from its complexity and its neglect of bank service to borrowers, the Department of Commerce (and old SNA) approach was criticised also on the ground that ownership of deposits was a very inadequate guide to the relative costs of services provided by banks to household and business depositors, if only because of economies of scale.[n.15]

Services to Borrowers. The third view, that bank services should be regarded as rendered primarily to borrowers rather than depositors, because the primary function of banks is financial intermediation, prevailed in the 1960s. In 1968 it was incorporated in a revised UN system of national accounts. “The imputed service charge is to be treated as intermediate consumption of industries for a number of reasons” of which the first was that “a key service performed by banks and similar institutions is to channel the savings of other economic agents into loans to industries.”[n.16] Or, as it was put by the Australian government statistician who changed over to the new SNA procedure in 1973, the imputed bank service charge “measures the expenses associated with organising borrowing and lending.”[n.17]

  The Australian explanation gives a clear account of the mechanics of this approach. “Interest received is viewed as consisting of a pure interest component and a service charge for oranising the funds. It is not practicable to allocate all the service charge to customers [i.e. among borrowers by industry]. 

• The part relating to consumer loans (including hire-purchase) is treated as being paid by the customer and included in private consumption expenditure. 
• The remainder, termed the imputed bank service charge, is not allocated to customers but treated as being paid by a ‘nominal industry’ which accordingly has a negative operating surplus of this amount.”[n.18]

[n.17] ^Australian National Accounts: Concepts, Sources and Methods^, Australian Bureau of Statistics, Canberra, p. 108.
[n.18] Ibid., p. 109.

  Thus, whereas on the second approach{SNA 53} the output of banks (other than services explicitly charged for) is divided between final products (included in GDP) and intermediate products (excluded from GDP) in proportion to household and business ownership of deposits, on this third approach{SNA 68}, most of the output of banks (and other financial intermediaries) is treated as consisting of intermediate products which enter into the costs of enterprises and are therefore excluded from GDP, the only exceptions being services to household depositors for which banks make explicit charges and services to household borrowers for which banks receive interest on consumer loans.

The Primacy of the Financial Intermediation Function.  Two surprising facts stand out from this summary history of the treatment of financial enterprises in national accounts. One is that each school of thought focused on only one of the three major functions of banks, ^either^ services rendered to the community at large, such as the creation of money, ^or^ services to depositors, such as keeping accounts an providing payment facilities, ^or^ services to borrowers, such as financial intermediation. It would seem obvious that banks perform all three functions and, more particularly, provide services to both depositors and borrowers. Ideally, therefore, one would look for an allocation of the imputed bank service charge between these two main categories of bank customers and then, within each category, between services which some part of the cost of bank administration as a payment to demand deposits in lieu of interest.

( ... ... )