Title: Determinants of the outstanding volume of dollar bankers' acceptances
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00098675/00001
 Material Information
Title: Determinants of the outstanding volume of dollar bankers' acceptances
Physical Description: viii, 148 leaves : diagrs. ; 28cm.
Language: English
Creator: Kundey, Gary Eugene, 1947-
Copyright Date: 1975
 Subjects
Subject: Acceptances   ( lcsh )
Finance, Insurance, and Real Estate thesis Ph. D
Dissertations, Academic -- Finance, Insurance, and Real Estate -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Statement of Responsibility: by Gary Eugene Kundey.
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 143-147.
General Note: Typescript.
General Note: Vita.
 Record Information
Bibliographic ID: UF00098675
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000167625
oclc - 02861120
notis - AAT4016

Downloads

This item has the following downloads:

determinantsofou00kund ( PDF )


Full Text























DETERMINANTS OF THE OUTSTANDING VOLUME
OF DOLLAR BANKERS' ACCEPTANCE








By



GARY EUGENE KUNDEY


A DISSERTATION PRESENTED TO TIE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF TIE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY






UNIVERSITY OF FLORIDA

1975















TABLE OF CONTENTS


List of Tables . . . . . . . . . . .. . ... iv

List of Figures . . . . . . .. . . . v

Abstract . . . . . . . . . . . . . . vi

Chapter I: Introduction . . . . . . . . ... . 1

Brief History of the Market . . . . . . . ... 1
Literature on the Bankers' Acceptance Market . . . . 4
1910-1930 . . .. . .. . . . . . 4
1955-1974 . . . . . . ... . . . . 6
Goals of the Dissertation . . . . . . . . 14
Synopsis of Dissertation . . . . . . .... . . 15

Chapter II: The United States Dollar Bankers' Acceptance Market 16

Definition . . . . . . . .... . . . 16
Origination . .. .. . . . . . . . . . 17
Major Participants in the Market . . . . . ... 19
Borrowers . . . . .... .. . . . . . 19
The Creating Banks. ... . . . . . . . 23
The Acceptance Dealers . . . . .. . . . 25
Investors . . . . . . . . . . . 27
Characteristics of the Bankers' Acceptance Market Crucial
to the Volume of Acceptances Outstanding . . . ... 30
Factors Affecting Borrowers' Demand for Acceptance
Financing . . . . . . . . . . . 31
Volume of transactions . . . . . .. . 31
International trade . . . . . ... 32
Storage of goods . . . . . . ... 35
Creation of dollar exchange . . . ... 36
Costs of financing . . . . . . ... 38
Miscellaneous factors . . . . . . ... 38
Factors Affecting the Willingness of Creating Banks
to Extend Acceptance Financing . . . . ... 39
Role of the Dealer's Inventory in the Operation of
the Market .... . ..................... 41
Characteristics of Dollar Bankers' AcceptancesSigni-
ficant to Investors . . . . . . . ... 43
Advantages of investment in acceptance ...... 44
Disadvantages of investment in acceptance . . 45









Chapter III: Formulation of the Hypothesis . . . . ... 47

International Trade . . . . . . . . ... .. . 47
Cost Differentials in Alternative Methods of Financing . 57
Credit Availability . . . . . . . . ... .. . 64
Expected Exchange Rate Changes . . . . . . ... .68
International Access of Borrowers and Lenders . . ... 74
Availability of Credit Information . . . . . . 76
Hypothesis of the Dissertation . . . . . . .. 76

Chapter IV: Model Construction . . . . . . . . 78

Specification of the Model's Structure . . . ... 78
Notation Used in the Model . . . . . . . 78
Structure of the Model . . . . . . . .. 81
Derivation of the Data Series . . . . . . . .. 82

Chapter V: Results and Implications of Our Statistical Analysis 86

Reformulation of the Hypothesis . . . . . . ... 86
Initial Statistical Analysis . . . . . . . .. 87
Statistical Analysis Using the Revised Quarterly Model 89
International Trade . . . . . ..... . 90
Cost Differentials Between Alternative Methods of


Financing Trade . .
Credit Availability . .
Expectations of Changes in
Dollar . . . . .


the Exchange Rate of the


Chapter VI: Summary . . . . . . . . . . . .

Data Series Appendix . . . . . . . . . . . .

Bibliography . . . . . . . . . . . . . .

Biographical Sketch . . . . . . . . . . . .















LIST OF TABLES


Table 1: Volume of Dollar Bankers' Acceptance Held by
Various Types of Investors
Selected Years, 1954-1974 . . . . . . . 9

Table 2: Volume of Dollar Bankers' Acceptances Outstanding
Created for Residents of Various Countries and Regions
Selected Years, 1955-1973 . . . . . ... 12

Table 3: Percentage of Dollar Bankers' Acceptances Outstanding
Created for Residents of Various Countries and Regions
Selected fears, 1955-1973 . . . . . . ... 21

Table 4: Volume of Dollar Bankers' Acceptances Outstanding
Classified by Type of Transaction
Selected Years, 1955-1973 . . . . . . ... 33

Table 5: Percentage of Dollar Bankers' Acceptances Outstanding
Classified by Type of Transaction
Selected Years, 1955-1973 . . . . . . ... 34

Table 6: United States Liquid Liabilities and Dollar Bankers'
Acceptances Held by Foreigners . . . . . ... 46

Table 7: Results of Regression Analysis . . . . . ... 88

Table 8: Results of Regression Analysis Applied to Revised Model. 91

Table 9: Results of Regression Analysis Applied to Various
Sectors of Revised Model for Periods of Extreme and
Non-extreme Values of Lt . . . . . . . .. 96

Table 10: Results of Regression Analysis Applied to Revised
Model for Periods of Ht(-250 Million Dollars .... 101

Table 11: Results of Regression Analysis Applied to the Latin
American Sector for Periods of Extreme and Non-Extreme
Values of Rt . . . . . . . . ... . .104

Table 12: Results of Regression Analysis Applied to United
States and Japanese Sectors for Periods of Extreme
and Non-Extreme Values of Pt . . . . . ... .108














LIST OF FIGURES


Graph 1: Year-End Monthly Volume of United States Imports
and Outstanding Volume of Dollar Bankers'
Acceptances Created for United States Residents,
1955-1972 . . . . . . . . .. . . 49

Graph 2: Year-End Monthly Volume of Japanese Imports and
Outstanding Volume of Dollar Bankers' Acceptances
Created for Japanese Residents, 1955-1972 ...... 50

Graph 3: Year-End Monthly Volume of Imports for Developed
Nations (Excluding United States and Japan) and
Outstanding Volume of Dollar Bankers' Acceptances
Created for Residents of Developed Countries
(Excluding the United States and Japan)
1955-1972 . . . . . . . . ... . . 51

Graph 4: Year-End Monthly "olume of Imports and Exports
for Latin America and Outstanding Volume of Dollar
Bankers' Acceptances Created for Residents of
Latin America, 1955-1972 . . . . . . ... 52

Graph 5: Year-End Monthly Volume of Imports for Less-Developed
Countries (Excluding Latin America) and Outstanding
Volume of Dollar Bankers' Acceptances Created for
Residents of Less-Developed Countries (Excluding
Latin America) 1955-1972 . . . . . . ... 53

Graph 6: 1955-1973, Cost Differentials Between Dollar Bankers'
Acceptances and Short-Term Direct Loans in New
York . . . . . . . . ... ... . .. 61

Graph 7: 1955-1974, Cost Differentials Between Dollar Bankers'
Acceptances and Sterling Bankers' Acceptances .... 62

Graph 8: 1955-1973, Cost Differentials Between Dollar Bankers'
Acceptances and Short-Term Direct Loans in the
Euro-Dollar Market . . . . . . . . ... 63

Graph 9: 1955-1973, Average Quarterly Level of Free Reserves . 67















Abstract of Dissertation Presented to the Graduate Council
of thie University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy


DETERMINANTS OF THE OUTSTANDING VOLUME
OF DOLLAR BANKERS' ACCEPTANCE

By

GARY EUGENE KUNDEY

June, 1975

Chairman: C. Arnold Matthews
Major Department: Finance, Insurance, Real Estate and Urban Land Studies

The previous literature on the dollar bankers' acceptance market

provides a comprehensive description of the market and proposes three

determinants of the volume of bankers' acceptance outstanding which are:

(1) volume of international trade, (2) cost differentials in the financing

of trade between dollar bankers' acceptance and alternative sources of

financing, and (3) credit availability in the United States. The previous

literature does not establish these factors as being statistically signi-

ficant to the volume of acceptance outstanding. The previous literature

does not address itself to the possible effects of expectations of changes

in exchange rates on the volume of acceptance financing since the previous

literature predates the recent movement away from fixed exchange rates in

the international monetary system.

In an attempt to supplement the bankers' acceptance literature on

these two particular points, the goal of this dissertation was to test

the statistical significance of the previously proposed determinants, as










well as expectations on changes in exchange rates, on the volume of

acceptance outstanding. Graphical analysis of indicators of interna-

tional trade, cost differentials, credit availability, and the volume

of dollar bankers' acceptance is given as a preliminary test of the

significance of the determinants proposed in the previous literature.

From this analysis, a general direct relationship is observed in each

major geographic sector of this market between the volume of international

trade and the corresponding volume of dollar bankers' acceptance out-

standing. Further, an inverse relationship is observed between credit

availability in the United States and the volume of acceptance out-

standing (i.e., as credit availability lessens, the volume of accep-

tances increases). The graphical analysis failed to show a consistent

relationship between acceptance volume and the cost differentials in the

financing of trade.

Multivariate regression analysis was used to test further the signi-

ficance of the previously proposed determinants and to test for the

significance of expectations of changes in exchange rates. For each

major geographic sector of the dollar bankers' acceptance market, measures

of international trade, cost differentials, credit availability in the

United States, and expectations of changes in exchange rates were used as

independent variables with the dependent variable being the correspond-

ing sector's volume of acceptance outstanding. The regression results

confirmed the positive relationship previously observed between interna-

tional trade and the volume of acceptance outstanding. The regression

results indicated that as expectations on the depreciation of the dollar

increased, the volume of dollar bankers' acceptance outstanding increases.









The regression analysis did not find the indicators of cost differentials

and credit availability in the United States to be significant determi-

nants of acceptance volume. The lack of significance could be due to

the deficiencies of the available data used in the regressions and is

not a refutation of these factors being determinants of the volume of

hankers' acceptance outstanding.















CHAPTER I
INTRODUCTION


Brief History of the Market


The United States dollar bankers' acceptance market is a segment

of the money market and centered in New York.1 An acceptance market

did not exist in the United States before 1914 because national banks

and most state banks were prohibited from engaging in the creation of

bankers' acceptances2

In the early part of the twentieth century there was a movement in

United States financial circles to increase the liquidity of commercial

banks. Proponents of the establishment of an acceptance market main-

tained that a dollar acceptance market would increase the liquidity of

the commercial banking system since acceptance instruments would be

excellent secondary reserves for commercial banks.3 One result of the



1Hereafter the United States dollar bankers' acceptance market will
be referred to as the bankers' acceptance market.

2Federal Reserve Bank of San Francisco, "The Role of Bankers' Accep-
tance Financing in International Trade and Finance," Monthly Review (San
Francisco: Federal Reserve Bank of San Francisco, July, 1955), p. 84.

3Lawrence M. Jacobs, "Bank Acceptances," Monetary Commission Docu-
ments (Washingtn, D.C.: United States Senate, 1910, number 569, p. 10).









movement for more liquidity for the banking system was the Federal

Reserve being authorized to establish an acceptance market in the

United States.4

The Federal Reserve from its inception to 1929 spurred the growth

of the acceptance market by absorbing large numbers of acceptances5

In order to encourage the growth of the market, the Federal Reserve

System posted discount rates for acceptance below the discount rates

offered by acceptance dealers. This resulted in the Federal Reserve

System purchasing a large volume of acceptance and directly supplying a

large amount of funds to the market. The volume of dollar bankers'

acceptance outstanding grew at a rapid rate from 1915 to 1929.6 By

the end of 1929 the volume outstanding was 1,732 million dollars, an

impressive amount for a market which had been nonexistent before 1915.7

Acceptance financing was used principally for international trade

(1,348 million dollars in December 1929), especially trade involving

the United States (907 million dollars), and domestic trade and storage

(308 million dollars).8



4Federal Reserve Act, Section 13, paragraph 6, 1913.

5Joy S. Joines, "Bankers' Acceptances," Instruments of the Money
Market (Richmond: Federal Reserve Bank of Richmond, February, 1968),
p. 69.

6Board of Governors of the Federal Reserve System, "Bankers' Accep-
tances," Federal Reserve Bulletin (Washington, D.C.: Federal Reserve
System, December, 1916), p. 717.

7Ibid., October, 1930, p. 621.

8Robert Cooper, "Bankers' Acceptances," Monthly Review (New York:
Federal Reserve Bank of New York, June, 1968), p. 130.










The use of acceptance financing declined drastically during the

1930s In December 1939 the volume of acceptance outstanding was

239 million dollars, a decline of 86 per cent from the December 1929

volume of 1,732 million dollars.9 The reduction resulted from the

decline in business activity during the 1930s especially to the de-

crease in the level of international trade, and the growing government

interference in, and regulation of the exchange markets.

The volume of financing through the acceptance market did not re-

cover substantially until well into the 1950s .0 With the restoration

of convertibility of currencies, the decline in government financing of

international trade, and the increase in international trade, the volume

of bankers' acceptance outstanding increased substantiallyi From a

volume of 642 million dollars at the end of 1955, the amount outstand-

ing increased, reaching a peak of 8,432 million dollars at the end of

1973.12

The largest increases in volume have been for financing of inter-

national trade for the residents of the United States and Japan.13 In



Board of Governors of the Federal Reserve System, "Bankers' Accep-
tances," Federal Reserve Bulletin (Washington, D.C.: Federal Reserve
System, December, 1940), p. 130.

10Joines, "Bankers' Acceptances," p. 70.

liFederal Reserve Bank of San Francisco, "Role of Bankers' Accep-
tance Financing in International Trade and Finance," p. 84.
12
New York Federal Reserve Bank, Monthly Acceptance Survey (New
York: Federal Reserve Bank of New York, 1955-1973).

13Dcpartment of the Treasury, "Acceptances Created for the Accounts
of Foreigners," United States Treasury Report (Washington, D.C.: Depart-
ment of the Treasury, March, 1955 and February, 1974).








January 1955, the volume of dollar acceptance financing being used

by United States residents was 583 million dollars, but in December

1973, it had reached 4,595 million dollars. In January 1955 only four

million dollars of the outstanding volume of dollar acceptance had

been created for Japanese residents. In December 1973, the volume of

acceptance outstanding created for Japanese residents was 1,802

million dollars. During the same time period acceptance used to

finance United States domestic trade and to create dollar exchange
14
actually declined. Of the 8,493 million dollars of acceptance out-

standing at the end of 1973, only 295 million dollars were created to

finance domestic trade (as compared to the December 1955 level of 308

million dollars) and 52 million dollars were for dollar exchange (as

compared to the December 1955 volume of 76 million dollars).


Literature on the Bankers'
Acceptance Market


Market-related research has paralleled the trends in the volume of

acceptance outstanding with most of the work being done in the period

1910 to 1930 and since 1955.


1910 1930

Lawrence Jacobs in his 1910 brief to the National Monetary

Commission outlined the reasons for the establishment of a dollar
15
bankers' acceptance market in the United States. Jacobs maintained



14Ibid., February, 1956 and February, 1974.

15Jacobs, "Bank Acceptances," pp. 1-20.









that two major benefits would result from such a market. The first was

increased liquidity of commercial banks since acceptance are more

negotiable and marketable than commercial bank customer loans.16 The

second would be an increase in financial services available in the

United States, especially in New York. It was hoped that the dollar

acceptance market would become a major competitor to the sterling

acceptance market. Since the acceptance market was established, it

can be assumed that Jacob's brief was influential.

In the early part of the 1920s most commercial banks were un-

familiar with the use of acceptance financing. A reference guide was

needed by the commercial banks to provide information on the nature of

acceptance financing, terminology, governmental regulations, and the

mechanics of acceptance creation. This need was fulfilled by Park

Mathewson's Acceptances: Trade and Bankers' published in 1921.17

The only other work of major importance to appear in the 1920s

was Jerone Thrall's "The American Discount Market."18 Thrall detailed

the growth of the acceptance market up to 1925. He also posed two main

questions that were to occupy the literature on this market for the

remainder of the 1920s. The first was: Why had not bankers' accep-

tances become a major secondary reserve instrument as envisioned in the



16
This will be expanded upon and substantiated in Chapter II, page
44.
17
Park Mathewson, Acceptances: Trade and Bankers' (New York:
Appleton Company, 1921), pp. 1-183.

J1erone Thrall, "The American Discount Market," Bankers' Magazine
(New York: The Bankers' Publishing Company, April, 1925), pp. 621-625.









founding of the Federal Reserve System? Even at this early date (1925),

it was apparent that acceptance were not and would not be used widely

enough to become a major secondary reserve asset for the banking system.

The second question was: Why had not the market expanded beyond the

confines of New York? The acceptance market in 1925, as today, was

centered in New York and seemed unlikely to disperse itself evenly

across the country.

The literature of this period was descriptive in nature and dealt

with matters connected with establishing the market. Little, if any,

attention was directed to describing cause and effect relationships

within the market. Due to the decline in acceptance financing in the

1930s and the stagnation of the acceptance market that lasted until

the mid-1950s little was written on the subject between 1930 and 1955.


1955-1974

The rapid growth in the volume of bankers' acceptance outstanding

in 1954 (from 574 million dollars at the end of December 1953 to 873

million dollars at the end of December 1954, an increase of 52 per

cent) triggered a renewed interest in the market.19 Two major articles

were published in 1955. The first of these, entitled "Bankers' Accep-

tance Financing in the United States," made two significant contribu-

tions.20 The first was to provide general and basic information about

a mode of financing which had been infrequently used for twenty-five



19New York Federal Reserve Bank, Monthly Acceptance Survey (New
York: Federal Reserve Bank of New York, January and December, 1954),
p. 1.

20Robert Solomon and Frank Tamagna, "Bankers' Acceptance Financing
in the United States," Federal Reserve Bulletin (Washington, D.C.:
Federal Reserve System, May, 1955), pp. 482-494.









years. It thus served to familiarize the financial and business

communities with a growing and promising method of financing.

The second was to present basic cause and effect relationships

which the authors thought existed. They stated that the volume of

acceptance outstanding was a function of at least three forces.

Since the reason for borrowing through the use of acceptance was to

finance trade transactions, they hypothesized that as the level of

trade increased so would the volume of bankers' acceptance out-

standing. They further stated that since a major portion of

bankers' acceptance outstanding was used to finance international

trade, the total volume outstanding should vary directly with the

variation in international transactions.

A second stated determinant of the volume of acceptance out-

standing was the relative cost of financing trade. As bankers'

acceptance became a relatively less expensive source of funds, then

the volume of bankers' acceptance outstanding should increase. The

third, and believed by the authors to be the least significant deter-

minant, was credit availability. As credit availability decreased,

commercial banks would find it advantageous to create more bankers'

acceptance since these instruments are more readily marketable than

are regular customer loans and provide a vehicle by which banks can

service their customers' needs without committing the funds of the

bank.

The authors also reviewed the mode of operation of acceptance

dealers and investors. They found that the most important group of

acceptance investors was foreign banks (both central and commercial),

but they failed to mention the importance of the commercial banks









which create, in conjunction with borrowers, the acceptance instru-

ments. This latter group of banks is known as 'accepting' banks or

'creating' banks, and their importance as investors can be seen in

Table 1.

A striking characteristic of this article, as with much of the

financial literature of the mid and early 1950s is the lack of statis-

tical substantiation of various cause and effect relationships. The

cause and effect relationships which the authors discuss have a

logical base in theory, but they made no attempt to ascertain how

strong these relationships were in the real world.

The second article to appear in 1955 was entitled "The Role of

Bankers' Acceptances in International Trade and Finance."21 Much of

the information contained in "Bankers' Acceptances Financing in the

United States" was restated. But it did make three significant con-

tributions.

First was a review of the history of the market. Second was

a description of the then recent changes in the relationship of the

Federal Reserve to this market. Prior to 1955 the Federal Reserve

had purchased acceptance instruments at posted discount rates. In

1955 the Federal Reserve began to play a more active role in the

market than it had in recent years and began to make its purchases
22
at the acceptance dealers' discount rates. The Federal Reserve



21Federal Reserve Bank of San Francisco, "The Role of Bankers'
Acceptance Financing in International Trade and Finance," pp. 84-90.

22Ibid., p. 89.














Table 1
Volume of Dollar Bankers' Acceptances Held by Various Types of Investors
Selected Years, 1954-1974
(In Millions of Dollars)


1954 1955 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973

Accepting Banks (Total) 289 175 663 1272 1153 1291 1671 1223 1198 1906 1544 1567 2694 3480 2706 2566
Own Bills 203 126 489 896 865 1031 1301 1094 983 1447 1344 1318 1960 2689 2006 2129
Bills of Other Accepting
Banks 86 49 173 376 288 260 370 129 215 459 200 249 734 791 700 430


Federal Reserve System
(Total)
Own Account
Held for Foreign
Correspondents

Bankers' Acceptance
Dealers' Inventory*

Other Investors (Mainly
Foreign Investors)

TOTAL


19 61 304 177 196 254 216 331 384 320 167 210 307 515 285 270
0 28 74 51 110 162 94 187 193 164 58 64 57 261 106 77

19 33 230 126 86 97 122 144 191 156 109 146 250 254 179 193


15 11 76 40 133 293 212 412 307 512 319 349 468 524 817 726


550 395 985 1194 1168 1052 1286 1426 1715 1577 2398 3324 3588 3370 3090 4931

873 642 2027 2683 2650 2890 3385 3392 3604 4315 4428 5450 7057 7880 6898 8493


NOTE: Values are for end of December.
* "Bankers' Acceptance Dealers' Inventory" obtained from correspondence with Federal Reserve Bank of New York.
Source: "Commercial and Finance Company Paper and Bankers' Acceptances Outstanding," Federal Reserve Bulletin (Washington, D.C.:
Federal Reserve System, compiled from February issues for the years 1955, 1956, 1961-1974).








wanted this market to increase in volume because it believed that this

market could make a contribution to the short-term flow of inter-

national funds and the restoration of convertibility of currencies.

The last contribution of this article was the citation of addition-

al factors affecting the behavior of borrowers, accepting banks, and

investors. Borrowers were said to view acceptance financing as a more

complex and less flexible source of financing as compared to direct

loans. Commercial banks were depicted as considering acceptance

financing as a more complex method of providing for their customers'

needs. Investors were pictured as having difficulties finding desirable
23
acceptance investments due to their odd denominations.

Following the two 1955 articles, no significant research was done

on the market until 1960 and 1961. Significant growth, however, occurred

in the acceptance market between 1955 and 1961. The first article to

discuss this growth was entitled "Rebound in the Use of Bankers' Accep-
24
tances." During 1960 the volume of acceptance outstanding increased

at a very rapid rate. At the end of December 1959 the volume of

acceptance outstanding was 1,151 million dollars; at the end of

December 1960 it had reached 2,027 million dollars, an increase of
25
76 per cent. Three factors were cited as causing the growth: (1) in-



23Bankers' acceptance are created in amounts equal to the costs of
the transactions being financed which accounts for the odd denominations
in this market.

24Federal Reserve Bank of Cleveland, "Rebound in the Use of Bankers'
Acceptances, Monthly Business Review (Cleveland: Federal Reserve Bank
of Cleveland, January, 1961), pp. 5-10.

25Federal Reserve Bank of New York, Monthly Acceptance Survey,
December, 1959 and December, 1960, p.l.









creases in the level of international trade, (2) an increase in the

availability for and use of acceptance financing by non-United States

importers and exporters, and (3) the relative costs of financing trade

between dollar acceptance and alternative sources of financing.

There was increased foreign participation in this market as

demonstrated in Table 2. Of significance was the marked increase

in acceptance created for foreigners, especially residents of Japan,

during the time period 1955-1961. The period 1960 through 1961 was a

period when the cost of acceptance financing became more competitive

with other major sources of funds.26 The more competitive position of

the acceptance market provided an incentive for an increased volume

of acceptance outstanding.27 The major role of creating banks as

investors in acceptance was also stressed.

In 1961, a second and more comprehensive article was published

by Robert L. Cooper entitled "Bankers' Acceptances."28 Its main

contribution was a restatement and expansion of the determinants of the

volume of acceptance outstanding. Special attention was given to the

relative costs of financing trade and the role of monetary policy.

Cooper also discussed the reason why dollar bankers' acceptance



26This will be expanded upon in Chapter III, p. 60.

27Federal Reserve Bank of Cleveland, "Rebound in the Use of
Bankers' Acceptances," p. 9.

28Federal Reserve Bank of New York, "Bankers' Acceptances,"
Monthly Review (New York: Federal Reserve Bank of New York, June, 1961),
pp. 94-100.





















Table 2
Volume of Dollar Bankers' Acceptances Outstanding Created for Residents of
Various Countries and Regions
Selected Years, 1955-1973
(In Millions of Dollars)


1955 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973

Country or Region

United States 348 835 894 793 697 785 829 1161 1301 1574 2529 3092 3619 3683 4595

Japan 28 555 999 945 1400 1607 1572 1306 1829 1720 1592 2140 2117 1360 1802

Developed Nations
Excluding United States
and Japan 118 129 230 273 212 246 308 465 333 341 399 515 707 554 715

Latin America 141 469 498 548 458 609 530 633 637 599 594 861 815 648 799

Less-Developed Nations
Excluding Latin
America 7 39 62 91 123 137 153 138 216 194 236 449 631 653 582

TOTAL OUTSTANDING VOLUME 642 2027 2683 2650 2890 3385 3392 3604 4315 4428 5450 7057 7889 6898 8493

NOTE: Values are for end of December.
Source: "Commercial and Finance Company Paper and Bankers' Acceptances Outstanding," Federal Reserve Bulletin and "Acceptances
Created for the Accounts of Foreigners," United States Treasury Report (Washington, D.C.: Department of the Treasury,
Feb., 1956 and 1961-1974).









were not more generally used to finance European trade and trade within

the United States. The former was due to relative inexpensive and

readily available alternative sources of financing in the European

financial market. In the case of United States trade, it was attributed

to the widespread use of trade credit. The article also gave a list

of specific kinds of goods that were most frequently financed through

the use of acceptance. The last contribution of this article was a

statement of the major geographic sectors of the bankers' acceptance

market. The regional breakdown given in Cooper's article is the basis

of Table 2. This geographic breakdown will be continued as an integral

part of this study.

The 1961 articles have the same deficiency as the articles that

appeared in 1955. Cause and effect relationships are stated, but little

support is given to establish if the hypothesized relationships are

statistically significant.

In response to this defect, R. Eldridge in his 1966 dissertation

tried to establish statistically significant causal relationships for
29
this market. After documenting the 1955-1962 resurgence of the

dollar bankers' acceptance market, bivariate linear regression

analysis was used to regress the volume of bankers' acceptance

outstanding for the years 1955 through 1962 against the volume of

imports and exports of various geographic regions. From the resulting

regression analysis, and his accompanying graphs of the volume of

acceptance and international trade, he concluded that as the volume of



29R. Eldridge, The Revival of Bankers' Acceptance Financing in the
United States (New York: Columbia University, 1966), pp. 1-252.










international trade increased so did the volume of acceptance out-

standing. His analysis is not conclusive since it contains several

defects, among which are non-use of multivariate analysis (i.e.,

Eldridge ignored in these regressions independent variables other

than the volume of trade) and the problem of autocorrelation is not

accounted for.

His dissertation presented graphs for 1955 through 1962 of

acceptance volume and cost differentials between dollar bankers'

acceptance and likely financing alternatives. From these Eldridge

concluded that acceptance volume increased as acceptance became a

less expensive source of funds. This analysis ignored variables other

than cost differentials and did not establish these relationships as

being statistically significant; therefore, the analysis is not con-

clusive. Eldridge did cite various institutional factors that could

possibly explain part of the rapid growth of acceptance volume in the

1955-1962 period.

Since Eldridge's work, no other major research projects have been

completed on this market. The literature to date has provided us with

a comprehensive description of the market and its components and with

a theoretical explanation of the determinants of the volume of bankers'

acceptance outstanding at a given time. These determinants have not

been fully verified with data from the real world.


Goals of the Dissertation


This study will attempt to verify statistically for the period

1955-1973 the effect of international trade, relative costs of financing

trade, and credit availability on the volume of bankers' acceptance









outstanding. Additionally an attempt will be made to verify statis-

tically for the same time period the effect of expectations of changes

in exchange rates on the volume of dollar bankers' acceptance out-

standing.


Synopsis of Dissertation


A brief description of the bankers' acceptance market and of the

participants in the market will be presented in Chapter II. In the

third chapter the hypothesis will be developed. Chapter IV will

describe the data series used to represent the variables in the

hypothesis and the structure of the model. We will present our

statistical analysis and the study's findings in Chapter V. Chapter

VI is a summary of our work.














CHAPTER II
THE UNITED STATES DOLLAR BANKERS'
ACCEPTANCE MARKET


The bankers' acceptance market is a financial market which performs

two basic functions. The first is to channel funds from lenders to

users. The second is to serve as a part of the larger mechanism for

allocating funds between alternative uses and users. The financial

instrument through which these functions are performed is the dollar

bankers' acceptance.


Definition


Bankers' acceptance are a type of bill of exchange. They are
1
considered of excellent quality and are readily marketable. They

usually arise in connection with commercial transactions where a
2
buyer of goods gives as payment for the goods the acceptance instrument.

The four important items specified in an acceptance are the amount

to be paid, the date of payment, to whom payment is to be made, and
3
who is to make the payment.



1Solomon and Tamagna, "Bankers' Acceptance Financing in the United
States," p. 482.
2
Joines, "Bankers' Acceptances," p. 67.

For a more extensive coverage of the mechanics of a bankers' accep-
tance, consult Wilbert Ward and H. Harfield, Bank Credits and Acceptances
(4th ed.; New York: Toronto Press Co., 1958).

16









Bankers' acceptance are time drafts. The time between the date

of origination and the date of payment enables bankers' acceptance

to be used for extensions of credit. In the majority of instances

the longest maturity permitted is 180 days.

Bankers' acceptance are negotiable instruments with the buyer

of the instrument having the rights of a "holder in due course." A

small number of the largest commercial banks in the United States are

responsible for the payment of a large portion of bankers' acceptance.

Since these banks have excellent credit reputations and since bankers'

acceptance are unqualified obligations of the bank on which they are

drawn, the majority of bankers' acceptance are regarded as second

only to United States Treasury Bills in quality and enjoy a high degree

of marketability.5


Origination


A bankers' acceptance involves three parties: the drawer (the

one who draws the draft); the drawee (the one on whom the draft is

drawn); and, the payee (the one to whom payment is made). The creation

of a bankers' acceptance usually begins when a purchaser of merchandise

applies to a commercial bank for an extension of credit in the form of

a bankers' acceptance. If his application is approved, a letter of



4Federal Reserve Act, 12 U.S.C., Sec. 13, Para. 7 and 12, last
revision, August 31, 1946, Board of Governors of the Federal Reserve
System.

5Cooper, "Bankers' Acceptances," p. 131.










credit is issued in favor of the buyer of the goods.6

The letter of credit is sent to the seller of the goods. It

specifies the amount of credit authorized, the purpose of the credit,

and the documents required to accompany the drafts to be created. It

also authorizes the seller of the goods to draw a draft on the bank

thereby making the commercial bank the drawee of the draft.

The next step in the creation of a bankers' acceptance is for the

seller of the goods to draw a draft on the commercial bank issuing the

letter of credit with the payee being the seller of the goods. The

seller then ships the merchandise under a negotiable bill of lading.

He attaches the necessary shipping documents to the draft and usually

sells the draft to his commercial bank at a discount from its maturity

value. The seller's bank will then forward the draft and attached

documents to the drawee.

The drawee upon receipt of the draft and attached documents will

ascertain if the draft and documents are consistent with the letter of

credit. If they are, the shipping documents are released to the buyer

of the goods in exchange for his signature on a trust receipt for the

goods. The trust receipt gives the drawee bank a lien on the goods.

The buyer of the goods will also be required to deposit with the



For an expanded treatment of the origination process, consult
Joines, "Bankers' Acceptances," p. 68; Cooper, "Bankers' Acceptances,"
pp. 127-128; Federal Reserve Bank of San Francisco, "The Role of Bankers'
Acceptance Financing in International Trade and Finance," pp. 85-86;
Solomon and Tamagna, "Bankers' Acceptance Financing in the United States,"
pp. 485-486; and, Federal Reserve Bank of New York, "Bankers Acceptances,"
p. 94.









drawee bank sufficient funds to cover payment of the draft or drafts

on or slightly before the maturity date.

The drawee will, if it approves the terms of the draft and the

attached documents, acknowledge its unconditional obligation to pay

the draft when due by "accepting" the draft. Acceptance of the draft

is formally acknowledged by the word "accepted" written on the instru-

ment and the signature of an officer of the drawee bank. "Acceptance"

is the distinctive characteristic of the banker's acceptance which

sets it apart from other time drafts. It is the crucial factor that

makes bankers' acceptance highly marketable money market instruments.

Upon "acceptance," the instrument is returned to the legal owner who

may either retain or sell the instrument. Normally, the drawee bank

will have been requested by the owner of the instrument to purchase

it at the time of acceptance. If so requested, the accepting or

drawee bank will routinely purchase the instrument at the discount

rate quoted by dealers in bankers' acceptances7


Major Participants in the Market


There are four groups of participants in this market: borrowers,

creating banks, dealers, and investors.


Borrowers

The creation of any dollar bankers' acceptance is primarily the

result of the interaction between two major participants:



7Arthur Bardenhagen, Expanding the Use of Bankers' Acceptances
(New York: Irving Trust Company, 1965), p. 24.









the borrower and a commercial bank. The acceptance borrower obtains

an extension of credit since he receives goods prior to the time he

provides the payment of the draft to the bank. The funds are loaned

to the borrower by any party who holds the banker's acceptance. Who

extends the funds to the borrower is immaterial to the borrower. His

only requirement for repayment of the instrument is to cover the

amount of the acceptance at the creating bank on or slightly before

the maturity date.

In this study acceptance borrowers are classified according to

their residence in one of the following geographic regions: United

States, Japan, Latin America, developed nations (excluding United

States and Japan), and less-developed nations (excluding Latin America).8

The United States and Japan are separate regions owing to the fact that

during the time period of our study residents of these regions have

been the major borrowers in the dollar bankers' acceptance market.

Between 1955 and 1973, the percentage of the total volume of bankers'

acceptance outstanding created for residents of the United States

ranged between 23 and 54 per cent, as shown in Table 3. During the

same time frame, the percentage created for residents of Japan ranged

from 21 to 48 per cent. Furthermore, the volume of acceptance created

for Japanese residents increased from 28 million dollars at the end of

December 1955 to 1,802 million dollars at the end of December 1973.



8Cooper, "Bankers' Acceptances," p. 130.

9United States Treasury Department, "Acceptances Created for the
Accounts of Foreigners," February, 1956, p. 52 and February, 1974,
p. 96.



















Table 3
Percentage of Dollar Bankers' Acceptances Outstanding Created for
Residents of Various Countries and Regions
Selected Years, 1955-1973


1955 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973

Country or Region

United States 54 41 33 30 24 23 24 32 30 36 46 44 46 53 54

Japan 4 27 37 36 48 47 46 36 42 39 29 30 27 20 21

Developed Nations
Excluding United States
and Japan 18 6 9 10 7 7 9 10 8 8 7 7 9 8 8

Latin America 22 23 19 21 16 18 16 18 15 14 13 12 10 9 9

Less-Developed Nations
Excluding Latin
America 1 2 2 3 4 4 5 4 5 4 4 6 8 9 7


TOTAL 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100

NOTE: Values are for end of December. Because of rounding, figures do not necessarily add to totals.

SOURCE: "Commercial and Finance Company Paper and Bankers' Acceptances Outstanding," Federal Reserve Bulletin and "Acceptances
Created for the Accounts of Foreigners," United States Treasury Report.









In addition to the United States and Japan, residents in the

developed nations of the world, exclusive of the communist countries,

are significant borrowers in the dollar acceptance market. The

percentage of outstanding volume written for residents of this group

ranged from 6 to 18 per cent over the period 1955-1973 but mostly

fluctuated between 7 and 10 per cent. The countries included in this

group are those of Western Europe, Australia, New Zealand, South

Africa, and Canada. This region is of special significance in our

study for two reasons: residents of this region have long used accep-

tance financing in the form of sterling acceptance, and many of these
10
countries have well-developed money markets.

The less-developed countries of the world constitute the other

two groups of borrowers. The countries of Latin America have close

economic ties with the United States and their residents are major

borrowers in this market. The percentage of acceptance outstanding

written for this group has ranged between 9 and 23 per cent, but

has tended to decrease in recent years. The less-developed countries

of the rest of the world comprise our last group of borrowers. While

the percentage of acceptance outstanding written for this group has

been small, the dollar amount has increased substantially: from 7

million dollars at the end of December 1955 to 582 million at the end
11
of 1973.



10Federal Reserve Bank of Cleveland, "Rebound in the Use of Bankers'
Acceptances" p.8.

11United States Treasury Department, "Acceptances Created for the
Accounts of Foreigners," February, 1956, p.52, and February, 1974,
p. 96.









The Creating Banks

The major link between a borrower in the dollar bankers' accep-

tance market and the suppliers of funds is the creating bank. In the

description of the creation of a dollar banker's acceptance, two

Factors relating to the creating bank's role were omitted. These are

the regulations imposed on the creating bank and the composition of the

cost of borrowing through acceptance.

Federal Reserve member banks originate the vast majority of accep-

tances measured either by number or by dollar volume. These banks must

comply with the provisions of the Federal Reserve Act which places

three types of restrictions on acceptance written by member banks.

These are (1) an upper limit on the amount of acceptance that a

creating bank can write for all their customers, (2) a maximum

limit on the amount of acceptance that a creating bank can write for

a particular customer, and (3) restrictions on the maturity of the
12
instruments.

A member bank can create acceptance in an amount up to 50 per

cent of its unimpaired capital and surplus for the purpose of storage

or shipment of goods, and an additional amount not to exceed 50 per

cent of its unimpaired capital and surplus for the purpose of creating

dollar exchange.

The upper limit to the amount of acceptance that can be created

for any one customer is 10 per cent of the bank's unimpaired capital



12Federal Reserve Act, U.S.C. 12, Sec. 13, para. 7 and 12, last
revision August 31, 1946, Board of Governors of the Federal Reserve
System.









and surplus. In this respect acceptance are similar to commercial

bank customer loans.

The longest maturity of any acceptance created for the purpose

of financing the shipment and/or storage of goods is 180 days, or

approximately six months. The maturity of bankers' acceptance used

to create dollar exchange is limited to 90 days, or roughly three

months.

As can be seen from this brief summary of acceptance regulations,

the amount of acceptance created for a single borrower is not any

more restricted than for short-term commercial bank loans. The limits

on acceptance maturities and the total acceptance which a bank may

create are more restricted than the limits on commercial bank loans.

It can be assumed that these regulations have not restricted the

volume of acceptance created since little pressure has been brought by

the creating banks to have the limits changed.

The cost of acceptance borrowing is in two parts. The creating bank

charges a service fee which for "prime rate" borrowers is 1/8 per cent

of the face amount of the acceptance for each month that the acceptance

is outstanding.13 This fee compensates the commercial bank for clerical

costs and the risk it bears by placing itself unconditionally liable

for payment of the draft.

The second cost of borrowing is the amount of discount from the

instrument's face value computed from the day of its acceptance.14 This

discount compensates the lender for his temporary loss of the use of his



138ardenhagen, Expanding the Use of Bankers' Acceptances, p. 59.

14Cooper, "Bankers' Acceptances," p. 129.










funds and for any risks he assumes by lending the funds. Admittedly

the risks to the lender are small since normally the party liable for

the instrument is a large commercial bank. The rate of return re-

quired by investors to invest in bankers' acceptance is the rate

of discount at which dealers in bankers' acceptance sell these

instruments to investors. The dealers' rates are adjusted for dif-

ferent maturities by a fairly fixed format. The rate on acceptance

with less than 90 days to maturity is the base rate with 1/8 per cent

added to obtain the rate on acceptance with a maturity of 90 to 120

days, and an additional 1/8 per cent to obtain the rate on acceptance
15
with a maturity of 120 to 180 days.


The Acceptance Dealers

At end of May 1974, there were six dealers in dollar bankers'

acceptance. These were Briggs, Schaedle and Company, Inc.; Discount

Corporation of New York; First Boston Corporation; M and T Discount Corp-

oration; Salomon Brothers and Hutzler; and Merrill Lynch, Pierce,
16
Fenner and Smith. Of these six dealers, only M and T Discount Corp-
17
oration operates exclusively in bankers' acceptance. All of the

dealers' operations are headquartered in New York.

A dealer in bankers' acceptance operates similarly to dealers in



15New York Federal Reserve Bank, Monthly Acceptance Survey, 1955-
1973.

16Merrill Lynch, Pierce, Fenner and Smith Become the Sixth Dealer
in Bankers' Acceptances, Wall Street Journal (New York: Dow Jones and
Company, Inc., March 30, 1966), p.4.

17Bardenhagen, Expanding the Use of Bankers' Acceptances, p.64.










any money market instrument. He stands ready to buy or sell at

given rates into and out of his inventory. The "buy" rate of the

dealer always exceed his "sell" rate; the differential covers his

profit and operating expenses. The difference is called a "spread"

and in the case of bankers' acceptance is 1/8 per cent on an annual

basis.19 This "spread" is large as compared to "spreads" on other

money market instruments because of the high selling costs caused by

the odd sizes and maturities of the instruments in this market.20

Acceptance dealers can continue dealing in acceptance only by

having a continuous flow of securities. The primary source of accep-

tances to the dealers is the creating banks. Creating banks routinely

purchase their own instruments immediately after "acceptance." Sub-

sequently, they may discount the instruments to customers or to accep-

tance dealers. The dealers then sell the instruments to investors who

usually hold acceptance to maturity. The last characteristic of the

acceptance market makes creating banks the dealers' major source of

acceptances.21

Creating banks depend upon dealers to be a "residual market" for

the sale of acceptance which the banks do not wish to hold. If the

demand for acceptance is high, the creating banks sell directly to



18The dealer's "sell" rate is the discount rate at which the dealer
will sell instruments. The dealer's "buy" rate is the discount rate
at which the dealer will purchase instruments.

19New York Federal Reserve Bank, Monthly Acceptance Survey,
1955-1973.

20Cooper, "Bankers' Acceptances," p. 132.

21Ibid., p. 132.









investors because they collect the dealer's "spread" on these trans-

actions. Creating banks acquire acceptance at the dealer's "buy"

rate and sales to buyers other than dealers are at the dealer's "sell"

rate. When the demand for acceptance is low, banks find it difficult

to sell directly to investors and turn to dealers for the disposition

of their unwanted instruments. These transactions are at the dealers'

22
"buy" rate; thus, the banks do not collect the "spread.22 The net

effect is that during periods of strong demand, dealers find it dif-

ficult to maintain an adequate inventory to assure their customers

of a good selection of maturity dates and denominations. During

periods of slack demand, the opposite occurs; pressure is placed on

the dealers to absorb an increasing supply of acceptance and to hold

larger inventories.


Investors

For our purpose an investor in a banker's acceptance is anyone

who holds the instrument. Investors may be divided into two groups:

non-permanent investors (those who do not hold the instruments until

maturity) and permanent investors (those who hold the instruments until

maturity). Both non-permanent and permanent investors provide financ-

ing while they hold the instruments, no matter how short the interval

of possession. By our definition acceptance dealers and creating

banks are investors in acceptance, at least on a non-permanent basis.

Acceptance dealers are investors in acceptance even though they

do not normally hold instruments until maturity. Their role as investors



22 d
Ibid.









stems from their commitment of funds to finance their inventories.

An acceptance dealer provides funds to this market in an amount equal

to his inventory.

Creating banks can be non-permanent or permanent investors in

acceptance. When a creating bank discounts to dealers or to its

customers acceptance it has purchased, then it is a non-permanent

investor in these instruments. It only invests for the time period

required to purchase and dispose of the instruments. Creating banks

which hold acceptance until maturity are considered permanent in-

vestors. They may be permanent investors in either their own accep-

tances or in acceptance of another creating bank. The acceptance of

other creating banks are normally acquired through the process of

"swapping."

"Swapping" in the vast majority of cases is done to acquire
23
additional endorsers on the acceptance. If a creating bank exchanges

acceptance created under its own letters of credit for instruments of

another creating bank, it will acquire instruments of better quality

since other banks will have "endorsed" these instruments and can be

held liable for Inyment. These exchanges are done either directly

between banks or through dealers; in either case the process is called

"swapping." If the swapping is done through a dealer, the dealer will

purchase at his "buy" rate the instruments of the creating bank which

desires additional endorsements and sell at his "sell" rate to this

bank the instruments of another creating bank. Since the dealer's

"buy" rate is 1/8 per cent higher than his "sell" rate, the effective



23Federal Reserve Bank of New York, "Bankers' Acceptances," p. 98.









charge for this service is 1/8 per cent to the creating bank that

desires instruments with additional endorsements. Swapping has in

recent years lost some of its previous popularity with creating

banks.24

In addition to creating banks, other permanent investors in

acceptance are the Federal Reserve System, non-United Sates in-

vestors; and non-creating bank United States investors. The Federal

Reserve System normally purchases dollar bankers' acceptance from

acceptance dealers. These purchases are for monetary policy reasons

(although a very minor role is played by the bankers' acceptance

market in monetary management), for foreign correspondents, and to

support dealers for the purpose of insuring an orderly market.25

Foreign investors are the largest investors in dollar bankers'

acceptance (see Table 1). This group has, during the period of our
26
study, held roughly one-half of the outstanding volume. Further

illustration of the regard which foreign investors have for acceptance

is shown by the fact that bankers' acceptance rank third in volume of

all dollar claims held by foreigners. United States Treasury Bills are

first, followed by demand deposits.

The most important investors within this group of foreign individ-

uals, corporations, and financial institutions are central banks. Foreign

central banks obtain their instruments from either the Federal Reserve system



24Cooper, "Bankers' Acceptances," p.134.

2Bardenhagen, Expanding the Useof Bankers' Acceptances, pp. 65 and
70.
26
Board of Governors of the Federal Reserve System, "Bankers' Accep-
tances Outstanding," 1955-1973.









(the majority of Federal Reserve purchases are made for foreign cor-

respondents who are normally foreign central banks), correspondent

banks in New York City who often are creating banks, and occasionally

from dealers.

Moving from foreign investors to United States investors, we find

that the principal domestic investors in bankers' acceptance have

already been discussed; these investors are creating banks, the Federal

Reserve System, and acceptance dealers (see Table 1). Domestic in-

vestor interest from other sources is lacking, although in recent

27
years more interest has been shown, especially by savings banks.

Much of the reason for the lack of interest can be traced to the

general lack of knowledge of domestic investors about bankers' accep-

28
tances as an investment media.

Characteristics of the Bankers' Acceptance Market
Crucial to the Volume of
Acceptances Outstanding


Since the objective of this dissertation is to verify the major

determinants of the volume of bankers' acceptance outstanding, we need

to survey those characteristics of the market which seem to have signi-

ficant implications for this purpose. One approach is to consider the

reasons why borrowers, creating banks, dealers, and investors resort to

this market.



27Cooper, "Bankers' Acceptances," p.135.

28Joines, "Bankers' Acceptances," p.74.








Factors Affecting Borrowers'
Demand for Acceptance Financing

Borrowers' influence in this market is through their decision on:

how much financing to seek? This decision is based on all factors

that normally enter financing decisions (i.e., volume of transactions

to be financed, costs of financing, alternative sources of financing,

etc.).


Volume of Transactions

The first factor to consider is, obviously, the volume of trade

transactions which possess the characteristics necessary for financing

with bankers' acceptance. As this volume varies, the volume of

bankers' acceptance financing desired by borrowers will also vary.

Some categories of commercial transactions are particularly adaptable

to financing through the use of bankers' acceptance. A few of these

account for most of the volume of bankers' acceptance created. These

have certain characteristics which encourage the use of bankers'

acceptance.

The first characteristic is that the seller of goods (i.e., the

payee) requires unquestioned confirmation of the obligation to pay.

If no formal obligation to pay is required, then trade credit or

29
another alternative could be used.29 Bankers' acceptance financing

does represent a formal acknowledgement on the part of the drawee to

pay the draft at maturity.

The second characteristic is that the documents necessary for col-

lateral be routinely produced.30 In those transactions that account for


29
Cooper, "Bankers' Acceptances," p.129.

3Ibid.














the vast majority of bankers' acceptance financing, documents such

as bills of lading, warehouse receipts, etc., are routinely created.

The third characteristic is that the seller finds it difficult
31
to obtain adequate credit information about the buyer. Through

the use of the letter of credit, the creating bank authorizes a draft

to be drawn in exchange for goods to be sold to the buyer, and acknowl-

ledges its obligation to accept and hence to pay the draft. The

sellers of goods are thus willing to complete transactions with

buyers whose credit status is unknown since they rely on the credit

reputation of the bank that issues the letter of credit. The lack

of credit information about the buyer is really the major reason for

acceptance financing.

These characteristics are found in three categories of business

transactions that account for the vast majority of financing done with

bankers' acceptance. These are international trade, domestic or

international storage of readily marketable commodities, and the

creation of dollar exchange (see Table 4).


International trade.--International trade accounts for the largest

volume of commercial transactions financed by dollar bankers' accep-

tances (see Tables 4 and 5). From 1955 through 1973, 79 to 98 per cent

of the bankers' acceptance created were used to finance international

trade. Sufficient documentation for acceptance financing is routinely



3hIbid, p.127

















Table 4
Volume of Dollar Bankers' Acceptances Outstanding Classified by Type of Transaction
(In Millions of Dollars)
Selected Years, 1955-1973


Type of Transaction 1955 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973

Total International Trade: 562 1596 2273 2293 2792 3321 3330 3421 4117 4309 5310 6840 7730 6734 8198


Imports into the
United States 252 403 485 541 567 667 792 997 1086 1423 1889 2601 2834 2531 2320

Exports from the
United States 210 669 969 778 908 999 974 829 989 952 1153 1561 1546 1909 3340

Goods stored in or
shipped between
foreign countries 100 524 819 974 1317 1565 1564 1595 2042 1934 2268 2678 3350 2294 2538

Total Domestic Trade: 63 308 293 171 41 43 35 80 161 67 112 188 122 156 243

Shipments 9 13 13 12 9 12 11 15 19 8 13 36 28 24 43

Storage 54 295 275 159 32 31 24 65 142 59 99 152 94 123 200

Dollar Exchange 17 122 117 186 56 111 27 103 37 52 28 29 37 7 52

TOTAL 642 2027 2683 2650 2890 3385 3392 3604 4315 4428 5450 7057 7889 6898 8493
NOTE: Values are for the end of December. Because of rounding, figures do not necessarily add to totals.
SOURCE: "Commercial and Finance Company Paper and Bankers' Acceptances Outstanding," Federal Reserve Bulletin.


















Table 5
Percentage of Dollar Bankers' Acceptances Outstanding Classified by Type of Transaction
Selected Years, 1955-1973


Type of Transaction


Total International Trade:

Imports into the
United States

Exports from the
United States

Goods stored in or
shipped between


1955 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973


88 79 85 87 97 98 98 95 95 97 97 97 98 98 97


39 20 18 20 20 20 23 28 25 32 35 37 36 37 37


33 33 36 29 31 30 29 23 23 21 21 22 20 28 39


foreign countries 16 26 31 37 46 46 46 44 47 44 42 38 42 33 30

Total Domestic Trade: 10 15 11 6 1 1 1 2 4 2 2 3 2 2 3

Shipments 1 1 0 0 0 0 0 0 0 0 0 1 0 0 1

Storage 8 15 10 6 1 1 1 2 3 1 2 2 1 2 2

Dollar Exchange 3 6 4 7 2 3 1 3 1 1 1 0 0 0 1

TOTAL 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100
NOTE: Values are for the end of December. Because of rounding, figures do not necessarily add to totals.
SOURCE: "Commercial and Finance Company Paper and Bankers' Acceptances Outstanding," Federal Reserve Bulletin.









produced in these transactions, lack of information of trading part-

ners is fairly common, and formal acknowledgement of the obligation

is desired.

The following products are the most common United States imports

financed through bankers' acceptance: coffee, iron, rubber, metal

ores, jute, sugar, steel, automobiles, cocoa, textiles, wool, and

crude oil.32 The following products are the most common United States

exports financed through bankers' acceptance: grain, cotton, machin-

ery and parts, ores, oil, automobile products, chemicals, iron, and

steel products.33

Bankers' acceptance financing is not restricted to trade involv-

ing the United States. In the recent past a significant percentage

of the volume of acceptance outstanding has been used to finance

trade not involving the United States (see Tables 4 and 5). From

1960 through 1973, 26 to 47 per cent of all acceptance created were

to finance trade not involving the United States. Products financed

most frequently include oil, ores and metals, wool, cotton, grains,

automobiles, sugar, and rubber.34

Storage of goods.--Bankers' acceptance are used to finance the

storage of many readily marketable commodities. Sellers of commodi-



32Cooper, "Bankers' Acceptances," p. 129.

33bid.

341bid., p. 130.










ties normally do not want to finance the storage of the goods for the

buyer. The sellers desire cash or debt instruments which can be

easily converted to cash. For a debt instrument to be readily con-

verted, it must be the obligation of an unquestionable credit risk.

Warehouse receipts, which can be used in acceptance financing, are

routinely produced in the storage of readily marketable commodities.

Sellers and buyers in these transactions often do not have a con-

tinuing relationship; therefore, sellers often lack sufficient credit

information about buyers.

The following list of staple commodities are those most frequent-

ly stored through the use of this type of financing: cotton, rice,

flour, grain, wool, sodium nitrate, peanuts, and tobacco.3 The

volume of acceptance created for the purpose of financing storage of

goods is a small proportion of the total volume created (see Tables 4

and 5). Since 1965 only 1 to 3 per cent of the bankers' acceptance

volume was created to finance domestic storage.


Creation of dollar exchange.--Only a small part of the volume of

bankers' acceptance outstanding has been used to create dollar exchange

(see Tables 4 and 5). Since 1965 the amount has varied between 1 and 3

per cent. The use of bankers' acceptance for this purpose has been

for the benefit of commercial and central banks in a limited number

of Latin American countries.

This particular type of acceptance has as the drawer-payee a

foreign commercial or central bank which is located in one of the



35Ibid.









countries designated by the Federal Reserve System to be eligible

to receive dollar exchange through the creation of bankers' accep-

tances.36 The instrument is sold by the drawer-payee as a money

market instrument after it has been accepted by the drawee, thereby

providing dollar exchange for the foreign commercial or central bank.

Creation of dollar exchange by this process requires formal acknowl-

edgement of the debt by an institution having a high credit reputa-

tion in the money market. Substitution of the credit status of the

creating bank for the borrower via an acceptance is required since the

credit status of these borrowers is often not known or is questionable

to suppliers of funds in the money market.

Acceptances creating dollar exchange, unlike other acceptance,

do not require bills of lading or warehouse receipts since they are

not based on the shipment or storage of goods. The reason for this

special category of bankers' acceptance is the need, in certain Latin

American countries, for a mechanism to create dollar exchange. This

need comes from the failure of exports and imports to vary together.

Exports fail to generate foreign exchange at the time needed to

finance imports. This places pressure on exchange rates to fluctuate

seasonally during the year. The creation of dollar exchange moderates

the seasonal fluctuations in exchange rates by providing exchange when

exports are low, and using the surplus supply for repayment when ex-



3Federal Reserve Act, 12 U.S.C., Sec. 13, para. 7 and 12, last
revision August 31, 1946, Board of Governors of the Federal Reserve
System.










ports are high.37


Costs of financing

Any borrower should attempt to obtain borrowed funds at the

lowest cost. Therefore, the demand for bankers' acceptance financing

should increase as its cost decreases relative to the costs of alter-

native sources.

The most probable alternative sources of financing to dollar

bankers' acceptance are short-term direct loans in the borrower's

home country, direct short-term loans in New York City, short-term

direct loans through the euro-dollar market, and borrowing in other

bankers' acceptance markets.


Miscellaneous factors

In addition to the volume of transactions and the relative cost

of financing, there are other factors which borrowers may consider in

their financing decisions. One of these is the continuing availability

of an adequate flow of funds. If the volume of financing available

from a particular source of funds is restricted, then the amount of

financing done through this market should decrease and an increase

occur in the financing from other sources.

Another factor to consider is that an attractive source of

financing should be flexible not only in its creation but also in its

repayment. Acceptances are flexible in the former, but not in the



3Ward and Harfield, Bank Credits and Acceptances, p. 104









latter since the instruments are generally negotiated into the hands

of third parties. Lenders and borrowers should be familiar with a

particular form of financing if it is to be attractive. Many

borrowers, especially in the United States, are not familiar with

acceptance as a method of financing.38 Acceptance financing also

tends to be more complex than various types of direct loans which

does not count in its favor.

Borrowers are not the only participants in the acceptance market,

and, therefore, do not exert the only influences on the volume of

acceptance outstanding. The main link between borrowers and the

acceptance market is the creating bank; therefore, we would expect

the creating bank to play a major role in determining the volume of

acceptance created.


Factors Affecting the
Willingness of Creating
Banks to Extend Acceptance
Financing

Certain factors should encourage "accepting" banks to create

bankers' acceptance while othersshould not. A commercial bank would

want to create these instruments because of the "acceptance" or

service fee of 1/8 per cent per month of the face value of the in-

strument. This fee seems quite attractive to many banks judging by

the growing number of banks offering acceptance credit facilities to

their customers. In 1960 there were only 95 "accepting" banks; by



38Joines, "Bankers' Acceptances," p. 74.









1968 there were 150 "accepting" banks.39 Another illustration of the

attractiveness of the service fee is that of the instruments purchased

by creating banks at the time of acceptance, which is the majority of

the instruments created, at most only 1/3 to 1/2 are held for invest-

ment purposes by the creating banks (see Table 1). Therefore, the

majority of acceptance are not created by the bank for the purpose

of holding them as investments. One explanation of this is that the

service fee is attractive, but the return on the instruments, as

compared to commercial bank customer loans, is not sufficient to

encourage the creating banks to hold all the instruments they create.

Creation of acceptance can be attractive to creating banks

during periods of restrictive monetary policy. If bankers' accep-

tances are not held by the creating banks, they fulfill the loan

demand of the banks' customers without requiring commitment of the

banks' limited resources. Through the use of acceptance financing

the banks can satisfy a larger loan demand than they could through

the use of direct loans only.

Several factors adversely affect the willingness of banks to

extend credit through acceptance. To place short-term loans in the

form of acceptance raises the cost and the complexity of credit

extensions.40 Regulations and customs impose an upper limit on the



39"International: Acceptance Business Luring Additional Banks,"
Burroughs Clearing House (New York: Burroughs Corporation, September,
1968), p. 55.

40Federal Reserve Bank of San Francisco, "The Role of Bankers'
Acceptance Financing in International Trade and Finance," p. 87.









amount of transactions that can be financed by bankers' acceptance.

For most commercial banks, the low amount of transactions financiable

through the use of acceptance does not justify the added costs and

complexities. Therefore, the majority of commercial banks do not

extend credit to their customers through the use of bankers accep-

tances.

Another factor which affects acceptance financing from the

commercial banks's point of view is the reception of its acceptance

as money market instruments. If a bank wishes to use acceptance to

convert a portion of its short-term customer loans into readily market-

able money market instruments, the bank must have a good reputation

and be known to dealers and investors. If a bank is unknown or does

not possess a sound reputation in the acceptance market than its

acceptance will not be readily marketable at rates competitive with

those of its better-known competitors.


Role of the Dealer's
Inventory in the Operation
of the Market

A major portion of the instruments not held in the portfolios of

the creating banks flows to investors through the acceptance dealers'

inventories. The acceptance dealer's inventory is one of the most

crucial aspects of a dealer's operation. The dealer is confronted with

two objectives: to keep the inventory to a minimum level (thus mini-

mizing his capital investment) and to maintain an inventory sufficiently

large to include a wide variety of denominations and maturities to meet

his customers' needs. The three factors believed to dominate the size

of the inventory decision are (1) the cost of financing the inventory,









(2) the actual supply of acceptance, and (3) the expected demand for

acceptance. The relationship between the cost of financing and the

size of the inventory should be negative (i.e., as the cost of financ-

ing increases, the size of the inventory should decrease), while

the relationship between the volume created and the size of the

inventory should be positive.41 A dealer would be expected to increase

his inventory if he expects future demand for bankers' acceptance to

increase. The increased inventory would provide for the expected

heavier drain on his inventory as demand increased.

The dealer changes his inventory by varying the discount rates at

which he buys and sells acceptance relative to the rates quoted on

other money market instruments. For example, if a dealer's inventory

is low (i.e., a high demand or a low supply of instruments has depleted

his inventory) the dealer would raise his inventory by lowering his buy

and sell discount rates. One effect of this would be to discourage

investors from purchasing the instruments in his inventory since their

relative rate of return would be lower (assuming that other money

market rates have not changed as a reaction to the change in acceptance

rates). Creating banks would discount more of their instruments since

the price the dealer offers for these instruments has increased due to

the lowering of his discount rates. A longer run effect of the lowering

of the discount rates should be a greater supply of acceptance as

borrowers find the cost of acceptance financing to be lower because of

the dealer's lowered discount rates. The net effect of the rate changes



41Bardenhagen, Expanding the Use of Bankers' Acceptances, p. 29.









on the dealer's operation is to decrease demand and increase supply;

thus, increasing his inventory from both sides of his operation. The

opposite effect will occur if the dealer's discount rates are raised.

Dealer's discount rates may be changed to maintain a constant

inventory as well as to change its size. To keep his inventory

constant, an acceptance dealer will be forced to change his acceptance

'buy' and 'sell' rates as competitive money market rates change. A

change in the acceptance dealer's inventory would occur, if his accep-

tance discount rates are not adjusted as other money market rates

change, due to the shift in the rate of return differentials between

bankers' acceptance and competitive instruments. The change in the

rate differential would cause investors to either favor bankers'

acceptance as an investment media and lower the dealer's inventory,

or cause investors to favor competitive investment media and increase

his inventory.

The dealer's operations have two influences on the market. The

first is to provide a reservoir of instruments, which allows an effi-

cient flow of acceptance from creating banks to investors. The

second influence is to provide a source of funds for the market by an

amount equal to their inventory. Therefore the dealers are investors

in acceptance and have the effect that other investors in acceptance

have.


Characteristics of Dollar
Bankers' Acceptances
Significant to Investors

Investors influence the level of acceptance outstanding by the

rate of return they require on investments in acceptance (i.e., the










supply of funds extended by investors at given rates of return). The

investors' supply of funds curve is seen as interacting with the

borrowers' demand for funds curve. The interaction gives the amount

of funds lent (borrowed) and the rate of return (cost of borrowing)

of the funds. The rate of return investors require for investment in

acceptance is based on investors assessing the various alternative

investments available to them. Bankers' acceptance must compete with

other investment media for the investors' favor. Acceptances have

advantages and disadvantages relative to alternative investments.


Advantages of investment
in acceptance

Two major advantages of an acceptance to an investor are its

quality and liquidity. The quality of these instruments is based on

the primary liability of the creating bank and the self liquidating

transactions financed by the majority of these instruments. Bankers'

acceptance are highly marketable due to the quality of the instrument

which assures the existence of a market willing and able to absorb

large amounts of these instruments. The rates on acceptance are

attractive since yields are generally 1/4 to 1/2 per cent higher than

United States Treasury Bills.42 The majority of potential investors

in acceptance are seeking short-term investments for their portfolios.

Foreigners have additional reasons to favor bankers' acceptance.

First is the growth of the dollar holdings of foreigners, especially



42Cooper, "Bankers' Acceptances," p. 131.









foreign central banks, between 1955 and 1973. This trend is shown in

Table 6. Any high grade investment media denominated in United States

dollars is viewed favorably as an investment outlet by foreign inves-

tors with a large volume of United States dollars to invest.

In addition, foreign individuals, corporations, and institutions

usually possess greater familiarity with acceptance financing than

their counterparts in the United States. The familiarity stems from

the use for several centuries of the acceptance instrument in Europe

as a major financing device. Thus, bankers' acceptance are not the

strange and unknown money market instrument to foreign investors as

they are to United States investors.43


Disadvantages of
investment in acceptance

Bankers' acceptance do have unfavorable features which lessen

their attractiveness to investors. These instruments are denominated

in odd amounts and it is difficult to tailor maturities to the needs of

investors. During periods of large demand for dollar acceptance,

non-creating bank investors in acceptance often find it difficult to

obtain sufficient volume to fill their requirements for these instru-

ments.4 Another principal disadvantage is that the rates of return

on acceptance, though better than United States Treasury Bills, are

often significantly lower than other money market instruments such as

commercial paper and certificates of deposit.



43Cooper, "Bankers' Acceptances," p. 131.

44Federal Reserve Bank of New York, "Bankers' Acceptances," p. 99.

45Cooper, "Bankers' Acceptances," p. 134.















Table 6

United States Liquid Liabilities and Dollar
Bankers' Acceptances Held by Foreigners
(In Millions of Dollars)


Dollar Bankers'
Acceptances


1,215

1,320

1,254

1,149

1,408

1,570

1,906

1,733

2,507

3,470

3,838

3,624

3,269

5,124


United States
Liquid Liabilities


21,027

22,936

24,068

26,322

29,082

29,115

29,779

33,119

33,614

41,894

43,242

67,681

82,901

93,121


NOTE: Data are for the end of December.
SOURCE: "U.S. Liquid Liabilities to Foreign Official Institutions and
Liquid Liabilities to All Other Foreigners," and "Commercial
and Finance Company Paper and Bankers' Acceptances Outstanding,"
Federal Reserve Bulletin (Washington, D.C.: Federal Reserve
System, February, 1961-1974).


Year


1960


1969


~~















CHAPTER III
FORMULATION OF THE HYPOTHESIS


Our summary of the literature on the bankers' acceptance market

and our discussion of the market and its participants suggest six

major determinants of the volume of acceptance outstanding. These

are international trade, cost differentials between alternative

methods of financing trade, credit availability, expected changes in

exchange rates, international access of borrowers and lenders to the

acceptance market, and availability of credit information on importers

and exporters.

This chapter will discuss the theoretical basis for these deter-

minants, and present some empirical evidence to support their inclusion

in our hypothesis which will be formally presented in the last section

of this chapter.


International Trade

All articles cited in the literature review which were written after

1955 state that international trade is an important determinant of the

outstanding volume in this market.1 The reason for the hypothesized link

between the volume of acceptance created and the level of inter-



Consult pages 7, 10, 11, and 13 of the first chapter for several
articles supporting this statement. Also consult Joines, "Bankers'
Acceptances," pp. 68 and 70, as well as Bardenhagen, Expanding the Use
of Bankers' Acceptances, p. 74.









national trade is that the vast majority of acceptance finance

international trade. Table 5 indicates that the proportion of the

volume of acceptance outstanding used to finance international

transactions for the period 1955-1973 varied between 79 to 98 per

cent. It is only logical to conclude that as the volume of interna-

tional transactions increases for a given region, ceteris paribus, so

will the volume of financing required by importers and exporters of

that region. One of the sources of financing available is bankers'

acceptance. We conclude that a positive relationship should exist

between the volume of international trade and the volume of bankers'

acceptance outstanding.

In support of this relationship the reader is referred to Graphs

1, 2, 3, 4, and 5. These graphs show for each geographic region used

in the study the volume of bankers' acceptance outstanding created

for the residents of that region, and the amount of the region's

imports denominated in United States dollars.2 With the exception of

Latin America, our analysis relates to imports rather than the total

value of international trade since little use is made of pre-export

acceptance financing in most regions, and the burden of arranging

financing is generally borne by importers.3 Latin America is an ex-

ception since pre-export acceptance financing is more common there.4



2For this study, all data will be expressed in United States
dollars.

3Bardenhagen, Expanding the Use of Bankers' Acceptances, p. 29.

41bid., p. 30.









In Graph 4 the sum of the value of Latin American imports and exports

is plotted along with the volume of acceptance outstanding created

for residents of Latin America. The reader will observe for each

region a general positive relationship between the volume of accep-

tances outstanding and the amount of international trade.

In Graph 1 a general positive relationship seems to exist

between United States imports and the volume of acceptance outstand-

ing created for United States residents. However, there are periods

when this relationship does not hold. In 1960 imports decreased while

the volume of acceptance outstanding increased; in 1962 and 1963 im-

ports increased but acceptance decreased. In 1969 United States

imports increased slightly while the volume of acceptance outstanding

increased substantially. The opposite occurred in 1972; imports in-

creased substantially but acceptance increased only slightly.

Graph 2 demonstrates a positive relationship between Japanese

imports and the volume of acceptance outstanding created for Japanese

residents. From 1960 through 1972 the year-end volume of acceptance

outstanding exceeded the year-end monthly volume of imports. This is

possible because bankers' acceptance can be written with up to six

months maturity. Therefore, when acceptance finance a large portion

of imports, the volume of acceptance outstanding, which are a result

of the last six months of imports, can exceed the amount of the most

recent month's imports. In the other regions, the percentage of

imports (and exports in the case of Latin America) financed by bankers'

acceptance is not large enough for this to occur. Graph 2, as Graph 1

discloses periods when a general positive relationship does not exist.









In 1966, 1968, 1969, 1971, and 1972, the volume of Japanese imports

increased but the volume of acceptance outstanding declined. In 1960

and 1961 imports increased at a moderate rate while the volume of

acceptance increased at a much more rapid rate. At the beginning of

1960, 202 million dollars of bankers' acceptance were outstanding

but by the end of 1961 there were 999 million dollars outstanding, an

increase of 395 per cent. This rapid increase in volume was probably

partially due to the cost differentials during this period which

favored bankers' acceptance financing over alternative financing

media. These alternatives will be the subject of the next section of

this chapter.

A general positive relationship between imports and bankers'

acceptance outstanding is also demonstrated in Graph 3. Although

dollar acceptance outstanding created for residents of developed

nations other than the United States and Japan have increased with

increases in imports, the rate of increase does not appear to have

been as proportionate as for the residents of these two countries.

Furthermore, there are periods when the positive relationship does

not hold as in 1959, 1963, 1967, and 1972, when the volume of accep-

tances outstanding decreased although the volume of imports increased.

A positive relationship between Latin American international trade

and the volume of acceptance outstanding created for Latin American

residents is demonstrated in Graph 4. However, during a number of

periods the general positive relationship does not exist; in 1961,

1965, 1968, 1971, and 1972 international trade increased while outstand-

ing acceptance decreased. In 1961, acceptance increased slightly,

but the volume of international trade decreased significantly. In 1962,









international trade increased rapidly; however, acceptance increased

at a very slow rate. There appears to be a change about 1967 in the

proportion of Latin American imports and exports financed with

acceptance.

In Graph 5 we observe a general positive relationship between

imports and acceptance until the period 1968 through 1972. In 1968

and 1969, imports increased at a moderate rate but the volume of

acceptance outstanding created for residents of the less-developed

countries, excluding Latin America, stayed relatively constant. In

1970 and 1971, imports for this region increased at a moderate rate

with the volume of acceptance outstanding increasing at a much more

substantial rate. In 1972, there were substantial increases in imports

for this region while the volume of acceptance remained about constant.

In this period, 1955-1972, there were sub-periods when the volume

of bankers' acceptance outstanding increased substantially: 1960-

1971, 1969-1971, and 1973 (See Table 2). During the 1969-1971 period,

some regions (Japan in 1969 and 1971 and Latin America in 1971) ex-

perienced a decline in acceptance outstanding even though the volume

of their international trade increased, while other regions (United

States in 1960, Latin America in 1961, and the less-developed countries

excluding Latin America, in 1960-1961) in the 1960-1961 period experi-

enced an increase in acceptance outstanding even though their volume

of international trade decreased. A substantial decrease in the volume

of acceptance outstanding occurred in 1972: a decline experienced by

three of the regions (Japan, developed nations excluding the United

States and Japan, and Latin America) while the other two (United States









and the less-developed countries excluding Latin America) experienced

very little increase. This was a year of substantial increase in the

volume of international trade in which all of the regions participated.

Thus while the volume of international trade seems to be an important

determinant of the volume of bankers' acceptance outstanding, it does

not explain all of the variations which occur.


Cost Differentials in Alternative
Methods of Financing


Cost differentials between alternative methods of financing inter-

national trade have been advanced as a factor likely to influence the

volume of bankers' acceptance outstanding.5 It is assumed that im-

porters and exporters are profit maximizers, which implies that they

seek to minimize the cost associated with financing their shipments

and storage. From this profit maximization principle, the conclusion

drawn is that dollar bankers' acceptance volume should increase as

acceptance become a less costly method of financing international

trade. Four alternative sources of funds are considered in this study:

direct short-term loans in New York City, sterling bankers' acceptance,

direct short-term loans in the euro-dollar market, and direct short-

term loans in the borrower's home country.

Direct short-term loans in New York are likely to be an alternative

source available to the majority of importers and exporters in all regions,

since the commercial banks which engage in acceptance financing also



Refer to pages 7, 10, 11, and 14 of Chapter I for articles that
support this statement.








6
make direct short-term loans. Therefore, it is probable that import-

ers and exporters which have access to acceptance financing would

also have access to direct short-term loans from the same banks.

Another alternative to acceptance financing likely to be available

to many importers and exporters in all regions is sterling bankers'
7
acceptance. The basis for this is twofold. The first is the great

similarity between the sterling and dollar acceptance instruments and

the respective markets. The second is that the sterling bankers'

acceptance market is as important, as an acceptance market, as the

dollar acceptance market.

Direct short-term loans in the euro-dollar market are another

alternative available to many importers and exporters using dollar

acceptance financing. The euro-dollar market headquartered in the

second largest financial center in the world, London, has grown

rapidly in the 1960's. Many importers and exporters using dollar

acceptance financing with contacts in New York would also have access

to this market. Since a substantial portion of euro-dollar loans



6This is supported by: Federal Reserve Bank of San Francisco,
"The Role of Bankers' Acceptance Financing in International Trade,"
p. 86; Solomon and Tamagna, "Bankers' Acceptance Financing in the
United States," p. 487; Federal Reserve Bank of New York, "Bankers'
Acceptances," p. 97; Cooper, "Bankers' Acceptances," p. 130; and
Federal Reserve Bank of Cleveland, "Rebound in the Use of Bankers'
Acceptances," p.9.

This is supported by: Solomon and Tamagna, "Bankers' Acceptance
Financing in the United States, p. 488; Federal Reserve Bank of New
York, Bankers' Acceptances," p. 97; and Federal Reserve Bank of
Cleveland, "Rebound in the Use of Bankers' Acceptance," p.9.









are short-term direct loans, this should be a competitive alternative

to dollar acceptance financing.

The last alternative source of funds considered is direct short-

term loans in the borrower's home country.9 If an importer or exporter

can borrow through the use of acceptance financing, probably he can

also borrow by short-term direct loans from a domestic bank in his

home country. The extent to which this is a viable alternative will

vary, of course, with the nature of the financial system and the

acceptance of the country's currency in international trade. This

alternative will be considered in regions consisting of one country:

Japan and the United States. Domestic short-term loans to United States

residents will be reflected by short-term direct loans in New York;

while for Japanese residents, direct loans from Japanese banks by im-

port bills will be used. For the other regions, domestic direct

short-term loans are not considered since their inclusion would cause

a statistically unmanageable number of alternative sources of funds,

since each of these regions includes at least 30 countries. Addition-

ally, statistics to reflect costs of direct short-term loans in

economies other than the major economic powers are often lacking.

Thirdly, the acceptability in international trade of the currencies

of the countries in these regions would often be lacking.



8Friedrich Klaus, The Euro-Dollar System (Ann Arbor: Cornell
University, 1968), p. 11.

9This statement is supported by: Federal Reserve Bank of New
York, "Bankers' Acceptances," p. 97; Cooper, "Bankers' Acceptances,"
p. 131.






60



The volume of acceptance outstanding for a particular region

should increase as dollar acceptance become a less expensive source

of financing; i.e., as the differential between the rate charged for

dollar acceptance subtracted from the rate charged for alternative

sources becomes larger and the sign is positive, the volume of dollar

acceptance outstanding, ceteris paribus, should become larger.

As a preliminary test of this hypothesis, we have plotted quarter-

ly differentials in the cost of dollar acceptance financing relative

to short-term direct loans in New York, sterling bankers' acceptance,

and euro-dollar short-term direct loans. These values are then com-

pared with the quarterly changes in the volume of dollar acceptance

outstanding and recorded in Graphs 6, 7, and 8, respectively.

An examination of these graphs fails to reveal any significant

consistent relationship between cost differentials in the financing

of trade and the volume of dollar acceptance outstanding. There are

periods when the relationship was as hypothesized: from the last

half of 1960 through 1962 (Graph 6), the last half of 1970 through

1971 (Graph 7) and 1969-1971 (Graph 8). There are periods during which

the observed relationships were the opposite of those hypothesized:

1956-1957 and 1964-1969 (Graph 6); 1956-1957 (Graph 7); and 1961, 1969

and 1971-1972 (Graph 8).

There are a number of problems with this type of analysis. The

alternative is not dollar bankers' acceptance and one of the other

sources but between dollar acceptance and any of the other sources.

To the extent that the cost of financing using other sources does not

move uniformly relative to the cost of dollar acceptance financing, the










movement of the volume of dollar acceptance plotted against each may

not accurately measure the impact of changes in relative cost of

alternative sources of financing on the volume of dollar acceptance

outstanding. In addition, our data are quarterly changes and the

impact will be due to daily or even hourly fluctuations that may not

be the same as that depicted by quarterly changes. Finally, the

impact of relative cost changes will be reflected on the ending

quarterly volume of acceptance outstanding only through the volume

of newly created acceptance which may not be adequately reflected in

the quarterly ending aggregate outstanding.

While the empirical data do not lend significant support to this

hypothesis, it will be retained for further analysis because of the

strong logic supporting the hypothesis, namely, that importers and

exporters attempt to maximize profits.


Credit Availability


The third factor cited as likely to have an influence on the

volume of bankers' acceptance outstanding is the stance of monetary

policy in the United States.10 Restrictive monetary policy has at

least two effects on the acceptance market. The first is through an

increase in interest rates and the second is to limit credit avail-

ability. An easy monetary policy has the opposite effects.

Rising interest rates in the United States would be expected to

decrease financing sought by borrowers in United States financial



1Consult pages 7 and 11 of the first chapter for articles that
support this statement.








markets for two reasons. Higher interest rates mean higher costs and

therefore some trade transactions would not be profitable. This

possible effect of monetary policy on the volume of acceptance would

be reflected through the volume of international trade. Interest

rates do not increase equally and simultaneously in all United States

financial markets nor do foreign rates necessarily move in step with

rates in the United States. This would mean a shift in the cost

differentials between alternative methods of financing trade. These

shifts can be expected to have an effect on the volume of acceptance

created as explained in the preceding section of this chapter.

If monetary policy affects credit availability, it should have
11
an influence on the volume of acceptance outstanding. As the degree

of monetary restraint increases and credit availability lessens, com-

mercial banks (including creating banks) will be less able to meet

their customers' loan demand from existing sources of funds and will

attempt to ration funds. Converting a portion of a bank's commercial

loans into dollar banker's acceptance which are discounted in the

money market enables the creating bank to meet a segment of its loan

demand without committing its own funds. Thus the bank is able to

satisfy a larger loan demand than would be possible without the use of

acceptance. Creating banks, therefore, have an incentive to increase

the creation of bankers' acceptance during periods of shrinking avail-
12
ability of credit.



11
Ibid.

12
12James B. Ludike, The American Financial System (Boston: Allyn and
Bacon, 1968), p.113.









The relationship described is negative between the volume of

bankers' acceptance and credit availability in the United States. As

the level of credit availability decreases, the level of bankers'

acceptance should increase. The volume of acceptance created should

decrease as credit becomes more available. To illustrate the plausi-

bility of this relationship, Graph 9 is presented. In the top half

of the graph, we have plotted the level of free reserves (quarterly

average). Market forces seem to follow this data series as an indi-

cation of the stance of monetary policy and credit availability. A

quarterly average was chosen to eliminate variations due to market

forces and to more accurately reflect the effect of Federal Reserve

operations. A policy of monetary restraint and lessening credit

availability are indicated by an increasing volume of borrowed reserves

(i.e.,negative free reserves) which should increase the volume of

acceptance outstanding. In the bottom half of the graph we have

quarterly changes in the volume of bankers' acceptance outstanding.

The data presented do indicate support of the hypothesis.

During periods of net borrowed reserves, namely 1955-1957, 1959-1960,

1966, 1968-1970, and 1973, the volume of bankers' acceptance outstand-

ing increased; and during the periods of positive free reserves, namely

1958, 1961-1964, 1967 and the first part of 1972, the total volume of

acceptance outstanding tended to decrease.









Expected Exchange Rate Changes


A fourth factor that may influence the volume of bankers' accep-

tances outstanding is the expectations of importers and exporters of

probably changes in the exchange rate of the United States dollar.

This determinant is based on the assumption that importers and export-

ers are aware of factors influencing exchange rates and of the impact

of changes in exchange rates on the profitability of their transactions.

Consequently, when events point to substantial adjustments in exchange

rates, they will react in such a way as to benefit from the anticipated

changes. Such action will be triggered by expectations of changes of a

magnitude which differs from the premium or discount of currencies cal-

culated by comparing actual spot and forward rates.

Experience indicates that when conditions persist which lead to

changes in official par values of currencies, central banks interfere

in the foreign exchange markets. Central banks of countries whose

currencies are in a relatively strong position internationally have

defended the present exchange rates of their currencies by absorbing

large amounts of currencies that are expected to depreciate. As an

example of this activity, we cite the amount of United States dollars

absorbed by the Bundesbank, the central bank of West Germany, during

1971. During this period, when there was an expected depreciation of

the dollar, the Bundesbank acquired foreign exchange (mainly dollars):

increasing from a January 1971 level of 8.6 billion dollars to 13.6

billion dollars in May 1971, an increase of 57 per cent.13 These



13International Monetary Fund, "International Liquidity: Foreign
Exchange Holdings," International Monetary Statistics (New York:
International Monetary Fund, 1971).









acquisitions were made to defend the then existing rate of exchange of
14
the German mark and the United States dollar. By the end of May

1971, expectations of further depreciation of the dollar relative to

the mark had declined, and the Bundesbank was able to reduce its foreign
15
exchange holdings to approximately 11.5 billion for the rest of 1971.

Central banks' interference in the exchange markets gives support

in the forward exchange market to the currencies expected to depreciate.

This support causes the actual forward discounts of the currencies

expected to depreciate to be less than the discounts which would pre-

vail in the exchange markets without the central banks' support. For

evidence of this, we again look at the 1971 depreciation of the dollar.

The actual depreciation of the United States dollar in terms of the

German mark was 10 per cent, but the three-month forward discount of the
16
dollar to the mark never exceeded 4.83 per cent on an annual basis.6

This effect will exist as long as central banks defend specific exchange

rates; evidence suggests that central banks continue to do this even with

the recent changes in the international monetary system.17



14"Germany Buys Dollars," Wall Street Journal (New York: Dow, Jones
and Company, Inc., May 5, 1971), p.2.

15International Monetary Fund, "International Liquidity: Foreign
Exchange Holdings," 1971.

16International Monetary Fund, "Exchange Rates," and "Forward Rates,"
International Monetary Statistics (New York: International Monetary
Fund, 1971).

17"Germany Buys 1.2 Billions of Dollars," Wall Street Journal (New
York: Dow, Jones and Company, Inc., February 23, 1973), p.3.









There are two possible situations where expectations of changes

in exchange rates could affect the volume of dollar acceptance out-

standing. The first is growing expectations of a depreciation of the

United States dollar, and the second is growing expectations of an

appreciation of the dollar. An expected depreciation of the dollar

should lead to an increase in the volume of acceptance created. This

situation can be illustrated with the use of a numeric example. Assume

that a transaction can be financed with dollar acceptance at 12 per

cent per annum (including the service fee) and that the only alterna-

tive is sterling acceptance at 9 per cent per annum. Assume further

that importers and exporters expect the dollar to depreciate 6 per

cent per annum within the next three months. Finally, assume that the

three-month forward discount of the dollar is only 3 per cent due to

the central banks' efforts to defend the existing rates of exchange.

Under these circumstances exporters and importers receiving pay-

ment in United States dollars will find the cost of financing to be

the same with both sterling and dollar acceptance (i.e., 12 per cent

on an annual basis) if they cover their exchange rate risks. If $1,000

is borrowed to finance a trade transaction lasting three months and

with the above listed assumptions, $1,030 would be paid by the borrower

when the acceptance matured if a dollar acceptance is used. If the

transaction is financed by a sterling acceptance and the current spot

rate for the pound is $2.50, then 400 pounds would be borrowed and 409

pounds would be paid at maturity. Since the actual three-month forward

discount of the dollar in terms of the pound is 3 per cent (on an annual

basis) the quoted forward rate for the pound is $2.52 which allows cov-

erage of the sterling acceptance for $1,030. Therefore, the sterling








acceptance cost is the same as the dollar acceptance.

Importers and exporters receiving payments in pounds sterling

will be motivated to seek dollar acceptance financing and not cover

their exchange rate risk. This occurs even though usual methods,as

used in the second section of this chapter, of calculating the cost

differential between these alternate sources of financing indicate a

9 per cent cost on an annual basis from both sterling and dollar

bankers' acceptance financing (assuming exchange rate risk coverage).

Dollar acceptance financing is favored because its cost is 6 per cent

if the expectations about the depreciation of the dollar come true, and

if exchange rate risks are not covered (i.e., 12 per cent interest costs

minus the 6 per cent expected depreciation of the United States dollar

relative to the British pound). If 1,000 pounds is borrowed to finance

a three-month trade transaction and with the above listed assumptions,

1,022.50 pounds would be paid by the borrower when the acceptance

matured if a sterling acceptance is used. If the transaction is financ-

ed by a dollar acceptance with a current spot rate of the pound of $2.50,

then $2,500 would be borrowed and $2,575 would be paid at maturity.

If the dollar depreciates relative to the poind by 6 per cent on an

annual basis during the three months the dollar acceptance is outstanding,

only 1,015 pounds would be needed at the maturity of the dollar accep-

tance. Therefore, the net effect of rising expectations of a depreciation

of the dollar should be to increase the outstanding volume of dollar

acceptance created for importers and exporters receiving ultimate pay-

ment for their trade transactions in currencies expected to appreciate

relative to the United States dollar.










When expectations of an appreciation of the United States dollar

are rising, the volume of dollar acceptance outstanding should de-

crease. A numeric example will illustrate this. Assume that a trans-

action can be financed with dollar acceptance at 6 per cent per annum

(including the service fee) and that the only alternative is sterling

acceptance at 8 per cent per annum. Assume further that importers

and exporters expect the dollar to appreciate by 5 per cent per annum

within the next three months. Finally, assume that the three-month

forward premium of the dollar is only 2 per cent due to the central

banks' efforts to defend the existing rates of exchange.

Under these circumstances, importers and exporters receiving pay-

ment in pounds will find the cost of sterling and dollar acceptance

financing to be 8 per cent on an annual basis if they cover their ex-

change rate risk. If 1,000 pounds is borrowed to finance a trade

transaction lasting three months and with the above listed assumptions,

1,020 pounds would be paid by the borrower when the acceptance matured

if a sterling acceptance is used. If the transaction is financed by a

dollar acceptance and the current spot rate for the pound is $2.50,

then $2,500 would be borrowed and $2,537.50 paid at maturity. Since

the actual three-month forward premium of the dollar in terms of the

pound is 2 per cent on an annual basis, the quoted forward rate of the

pound is $2.4875 which allows coverage of the dollar acceptance for

1,020 pounds. Therefore, the cost of sterling and dollar acceptance

financing are the same.

Importers and exporters receiving final payment in United States

dollars will be motivated to seek sterling acceptance financing and not

cover their exchange rate risk even though usual methods of calculating










the cost differential between these alternate sources of financing

indicate a 6 per cent cost from both sterling and dollar acceptance

(assuming exchange rate risk coverage). The reason for the preference

is that the effective cost of sterling acceptance financing would be

only 4 per cent on an annual basis if the expectations about the

appreciation of the dollar come true, and if they do not cover their

exchange rate risk (i.e., 8 per cent interest cost minus 4 per cent

appreciation of the dollar relative to the pound). If $1,000 is

borrowed to finance a three-month trade transactions and with the above

listed assumptions, $1,015 would be paid by the borrower when the

acceptance matured if a dollar acceptance is used. If the transaction

is financed by a sterling acceptance and with a current spot rate of

the pound of $2.50, then 400 pounds would be borrowed and 408 pounds

would be paid at maturity. If the dollar appreciates by 4 per cent on

an annual basis during the three months the sterling acceptance is out-

standing, only $1,010 would be needed at the maturity of the sterling

acceptance. The net effect of the rising expectations of an appreciation

of the dollar, therefore, should be a decreased volume of dollar accept-

ances.

The following two factors are often sighted as being influential

in forming the expectations of foreign exchange speculators about de-

preciation or appreciation of the United States dollar: changes in

international reserves held by the United States and the net liquidity

balance of payments of the United States.18 As each of these becomes



18Warren L. Smith, "The Present System and Its Defects,"
International Monetary System (New York: Prentice Hall, 1969), p. 16.










more negative (i.e., as the international reserve loss of the United

States grows and as the deficit of the United States balance of pay-

ments increases) expectations of depreciation of the dollar should

increase and, therefore, increase the volume of acceptance outstand-

ing. As each of these becomes more positive, expectations of apprecia-

tion of the dollar should increase and result in a decreasing volume

of acceptance outstanding. Evidence illustrating the strength of the

relationship between these measures and the volume of bankers' accep-

tances outstanding will be given in Chapter V.


International Access of Borrowers
and Lenders


The dollar bankers' acceptance market is international in scope

as indicated by the number of foreign investors and the proportion of

acceptance created for residents of the various regions of the world.

Any factor promoting the access of borrowers and lenders to this market

should increase the volume of dollar bankers' acceptance outstanding.

Several trends in international trade and finance in the last two

decades have improved the access of borrowers and lenders to this

market. Some of these are removal or reduction of foreign exchange

restrictions in many countries, growth of United States dollar balances

held by foreigners, and the growth of an international point of view

among lenders and borrowers.19



19G.W. Woodworth, The Money Market and Monetary Management
(New York: Harper and Row, 1965), p.124.









The removal of exchange restrictions should increase international

trade and permit greater freedom of its financing. Either of these

should increase the volume of dollar bankers' acceptance financing. The

removal of restrictions allows borrowers and lenders easier access to

alternative international financial markets. The increased access to

all financial markets would facilitate the flow of funds between markets.

This would be reflected in a growing sensitivity of acceptance volume

to cost differentials between dollar acceptance and alternative sources

of financing trade not denominated in United States dollars.

The growth of United States dollar balances held by foreign institu-

tions, corporations, and individuals should increase the volume of dollar

bankers' acceptance outstanding since any financial market with instru-

ments denominated in dollars, such as the dollar acceptance market, would

be looked upon favorably by foreign investors with large United States

dollar balances.20 This likely influence would be incorporated by the

cost differentials between alternate financing methods since this influ-

ence deals with the supply of funds to the dollar acceptance market.

Another trend which has probably had an influence in the dollar

acceptance market is the development of an international point of view

of borrowers and lenders. The growth and expansion of the multi-national

corporation is a factor in this trend. It has been accompanied not only

by the recognition of borrowers and lenders of their ability to use inter-

national, as opposed to only national financial markets, but by the nec-

essity to be .conversant with trends and developments in all of the markets

which are viable alternatives as sources of financing. This increased



20See discussion of this point on page 44 of Chapter II.









awareness and sophistication by a number of lenders and borrowers in this

market should be reflected in the cost differentials that exist between

this market and other alternative methods of financing trade and contri-

butes to the growing sensitivity of the volume of dollar bankers' accep-

tances outstanding to forces outside the United States.


Availability of Credit
Information


Bankers' acceptance financing has a major advantage when little knowl-

edge is possessed by importers and exporters about their trading partners.

The trading partners may have little knowledge about each other, but a

transaction can still be financed since a well-known and respected creat-

ing bank places itself unconditionally liable for the payment of the accep-

tance. A change in the level of knowledge about trading partners would

logically have an effect on the volume of acceptance outstanding. The

availability of information on commercial enterprise in the last two

decades has probably increased which should have a dampening effect on

the growth of the dollar acceptance market. This factor cannot be ad-

equately documented (i.e., a data series cannot be constructed to re-

flect growth in the availability of credit information on an international

scale) since little knowledge, except for general trends, exists about

the quality or quantity of credit information distributed internationally.


Hypothesis of the Dissertation


The preliminary empirical evidence presented in this chapter suggests

that the first three factors discussed should be included in our model of

the determinants of the volume in the bankers acceptance market. Expecta-

tions of changes in exchange rates will also be included due to the








logical basis of this factor and because of the importance in our

present international monetary system of discovering if expected changes

in exchange rates affect the volume of dollar acceptance. The effects

on acceptance volume of our fifth hypothesized determinant (international

access of borrowers and lenders) are incorporated in the first two

factors discussed, and, therefore, need not be tested separately.

Our last hypothesized determinant lacks a data series to measure its

changes in magnitude; therefore, it cannot be included in the model.

The hypothesis we propose to test is: The volume of United States

dollar bankers' acceptance outstanding will vary directly with (1) the

level of international trade, (2) relative advantages in the cost of

financing with dollar acceptance, (3) the degree of restriction in

United States monetary policy, and (4) expectations of depreciation of

the dollar; and will vary inversely with expectations of appreciation

of the dollar.














CHAPTER IV
MODEL CONSTRUCTION

We can now proceed with the construction of a model of the hypoth-

esis. This chapter will consist of: (1) specification of the model's

structure, and (2) derivation of certain data series.

Specification of the Model's Structure

The model used is a linear model, which assumes that the effect

on the dependent variable of changes in two or more of the independent

variables can be expressed as a simple sum of the effects that would

occur if each independent variable changed with the others remaining

constant. If the effect on the dependent variable of the same magni-

tude of change in an independent variable differed with the level of

the other independent variables, then the effect would partially be

due to interaction between the independent variables.

Notation Used
in the Model

t : a subscription referring to a specific quarter within the time

period January 1955 through December 1973.

i : subscript referring to one of our five regions. When i=1,

United States; i=2, developed nations excluding the United

States and Japan; i=3, Japan; i=4, Latin America; i=5, less-

developed nations excluding Latin America.







Bt,i: Volume of United States dollar bankers' acceptance out-

standing at the end of time period t that have been cre-

ated for residents of region i.

S .: Value of imports of region i during time period t ex-
t,i
pressed in United States dollars.

S't,i: Value of imports and exports of region i during period t

expressed in United States dollars.

G : Average cost (expressed on an annual percentage basis)

during the middle month of time period t of financing

trade with 90 day United States dollar bankers' accep-

tances.

Nt : Cost differential (expressed on an annual percentage

basis) in the financing of international trade obtained

by subtracting Gt from the cost at the mid-point of

time period t of prime direct short term loans from New

York City banks.2

Et : Cost differential (expressed on an annual percentage

basis) in the financing of international trade obtained

by subtracting Gt from the cost at the mid-point of

time period t of prime direct short-term loans in the

London euro-dollar market.



1Data and data sources are presented in the appendix. This model
is constructed to test quarterly data. A monthly model was also
tested but the results were not as significant as the results of the
quarterly model.

2All financing costs should be averages for time period t. How-
ever, costs of financing were not available in this form, and an
average could not be accurately constructed from the available statis-
tical series.









Qt : Cost differential (expressed on an annual percentage

basis) in the financing of international trade obtained

by subtracting Gt from the cost at the mid-point of time

period t of 90 day prime sterling bankers' acceptance.

D : Cost differential (expressed on an annual percentage

basis) in the financing of international trade obtained

by subtracting Gt from the average cost for the middle

month of time period t of direct short-term loans from

Japanese banks through import bills.

Lt : The most favorable cost differential for creation of

United States dollar bankers' acceptance, of the dif-

ferentials Nt, Et, and Qt, divided by Gt.

Mt : The most favorable cost differential for creation of

United States dollar bankers' acceptance, of the dif-

ferentials Nt, Et, Q and Dt, divided by Gt.

Rt : Change during time period t in the volume of interna-

tional reserves (gold, foreign exchange, and special

drawing rights) held by the United States.

Pt : Net liquidity balance of payments (quarterly basis) of

the United States during time period t.

Ht : The average volume of free reserves of Federal Reserve

member banks during time period t.

Ut : Random disturbance or error term during time period t.

fz,i: The functional relationship for region i between variable

z ( z= Si ,L ,M ,R ,P ,H ) and Bt,i.

Ai : Constant term for region i.








Structure of
the Model

The model's structure is based, for the most part, on material

presented in Chapter I, II, and III. The model includes the following

variables which are not inferred from Chapters I, II, and III: (a) Lt

and Mt, and (b) independent variables lagged one quarter.

To reduce the number of independent variables in our model, our

measures of cost differentials in the financing of trade, Lt and Mt,

only include the most favorable cost differential for the creation of

dollar bankers' acceptance of Nt, Et, Qt, and Dt. During the time

period of the study interest rates have generally risen. If the mag-

nitude of the differentials Nt, Et, Qt, and Dt was the same in 1955

and in 1973, its significance for influencing financing decisions would

be understated for 1955 and overstated for 1973 because of the general

increase in the level of interest rates. To adjust for changes in the

overall level of interest rates, we need only to divide by Gt.3

The volume of bankers' acceptance outstanding at the end of

quarter t should be dependent on the values of the independent variables

during the previous quarter, t-l, as well as on their values during

quarter t. This relationship should exist since bankers' acceptance

at time of creation can have a maturity of up to six months. This

means that some of the acceptance created during period t-1 should be

outstanding at the end of period t. To take this factor into account

our model has for each independent variable value at time period t, a



3A model using absolute instead of relative cost differentials was
tried and the results are summarized in Chapter V., p. 95,








corresponding value lagged one quarter (i.e. t-l value).

The model is as follows:

ti = Ai + f s + St +f Lt +
ti i i St,i + fSt-l,i St-l,i fLti Lt fLt-'1i Lt-1


t, -fRt-, iRt- -ft,i t t- ,i t-

-fti Ht -ft + Ut


i=1,2,5

Bt,i = Ai + 1 + f +S
i A + ft,i St,i St-li St-li M,,i Mt Mt-l,iMt-


Rt ,i R -fRt-1 i Rt-1 -Pt,i Pt -Pt-l,i Pt-1

-Hti Ht t-f 1i Ht-1 +Ut


i=3

Bti = Ai + ti S i + f't-,i S't-li + f Lt


Sf L -f R f R
Lt-1,i t-1 Rti t Rt-1,i t-


Pt -fP i -f H Ht -f ,i Ht-l + Ut

i =4

Derivation of the Data Series

Some of the data series were not available in the form needed for

our analysis. This necessitated conversion of the data. Two major

types of conversions were made: converting data series from a monthly

basis to a quarterly basis, and the construction of Lt and Mt from

existing data series.

All data series on a monthly basis were converted to a quarterly

basis since our model is formulated on a quarterly basis. These con-

versions were made by summing for each quarter the monthly values of









the affected data series. The data series affected by these conversions

are the volume of imports of the United States and Japan, change in the

volume of international reserves held by the United States, and the

average volume of free reserves of Federal Reserve member banks. In

the latter case, the quarterly figure derived was divided by three

since we desire a measure of the quarterly average volume of free re-

serves.

The first component of Lt and Mt is the differential between the

cost of financing with dollar bankers' acceptance and with direct loans

from New York banks. The first step in calculating the differential in

the cost of financing with dollar bankers' acceptance and with short-

term direct loans was to combine the average rate on prime bankers'

acceptance, three-month maturity, for the middle month of the quarter

and the acceptance fee charged by United States accepting banks, which

is 1.5 percent per annum for prime risks. This sum was then subtracted

from the rate that prevailed on short-term loans from New York City

banks at the mid-point of the quarter.4 The cost differential for

financing with dollar bankers' acceptance and direct short-term loans

in the London euro-dollar market was created by combining the average

prime rate, for the middle month of the quarter, of bankers' accep-

tunces (three-month maturity) with the service fee for acceptance fi-

nancing, and subtracting this sum from the short-term London euro-

dollar deposit rate that existed at the mid-point of the quarter, ad-

justed by the service fee for prime direct short-term euro-dollar loans.5



4 Cooper, "Bankers' Acceptances," p. 131.

5 Obtained from personal correspondence with officials of the Bank
of England.








The differential in the cost of financing with United States

dollar bankers' acceptance and sterling bankers' acceptance is de-

rived by subtracting the average rate prevailing in the middle month

of the quarter on 90 day prime bankers' acceptance, adjusted by the

addition of the 1.5 percent service charge, from the costs of financing

that existed with sterling bankers' acceptance at the mid-point of the

quarter. The cost of financing with sterling acceptance includes (1)

the annual rate on prime sterling bankers' acceptance, 90 day maturity,

(2) the standard service fee, (3) the tax charge of .2 percent, and

(4) the three-month forward discount or premium of the pound sterling

relative to the United States dollar.6

The differential for a given quarter between dollar bankers' ac-

ceptances and direct short-term loans from Japanese banks at the import

bill rate was calculated by subtracting from the average short-term

Japanese import bill rate prevailing in the middle month of the quarter,

the sum of the average prime rate existing in the middle month of the

quarter on 90 day United States dollar bankers' acceptance and the

service charge of 1.5 percent. This approach is required for years

before 1971 because a statistical series of the forward premium (or

discount) of the Japanese yen relative to the United States dollar was

not available until 1971. Significant bias should not be introduced

through this technique since revaluation of the yen relative to the

dollar was considered unlikely during this period.7 For quarters in

the years 1971, 1972, and 1973 the differential was calculated by



6 loc. cit. Obtained from personal correspondence

7 This statement is based on personal correspondence with officials
of the Bank of Japan.






85



subtracting the average rate of dollar acceptance financing from the

average Japanese import bill rate adjusted appropriately for the pre-

mium (or discount) of the Japanese yen relative to the United States

dollar.















CHAPTER V
RESULTS AND IMPLICATIONS
OF OUR STATISTICAL
ANALYSIS


In this chapter we will reformulate the hypothesis to reflect the

data series used to measure the variables. We will then present our

statistical analysis and a discussion of its implications.


Reformulation of the Hypothesis

The hypothesis states that the volume of dollar bankers' accep-

tances outstanding varies directly with changes in the volume of inter-

national trade. Imports are used as our measure of international trade

in all regions except in Latin America where imports and exports are

used because of the widespread use of pre-export financing in Latin

America. The hypothesis reformulated to reflect this states that as a

sector's volume of imports (and exports in the case of Latin America),

St,i, changes, the volume of bankers' acceptance outstanding created

for its residents will change in the same direction.

We have hypothesized that the volume of bankers' acceptance out-

standing should vary directly with favorable cost differentials for

dollar bankers' acceptance relative to alternative methods of fi-

nancing. We have measured the cost differentials, Lt and Mt, by sub-

tracting the cost of financing with dollar bankers' acceptance from

the cost of financing with alternative sources. Thus, positive

values of Lt and Mt indicate favorable conditions for the creation of














dollar bankers' acceptance. Therefore, the reformulated hypothesis

states that with positive values of Lt and Mt, the volume of dollar

bankers' acceptance outstanding, Bt,i, will increase.

The hypothesis states that the volume of bankers' acceptance

outstanding created for each region will increase as monetary re-

straint in the United States increases. Negative values of Ht (the

average volume of free reserves during quarter t) indicate monetary

restraint. The reformulated hypothesis states that the volume of

bankers' acceptance outstanding will vary inversely with the volume

of free reserves (i.e., a negative relationship is expected between Ht

and Bt,i in all sectors).

Our hypothesis states that the volume of bankers' acceptance

outstanding will vary directly with expectations of a depreciation of

the United States dollar. Since we measure the change in expectations

of a depreciation by changes in international reserves held by the

United States (Rt) and by changes in the net liquidity balance of

payments of the United States (P ), and since decreases in either of
t
these measures increase the expectations of a depreciation of the

United States dollar, we expect a negative relationship of Rt and Pt

to Bi.

Initial Statistical Analysis

Multivariate regression analysis was applied to our quarterly model

for the 1955-1973 period, and the results are summarized in Table 7.

Standard 't' tests were used to test for the significance of the coef-










TABLE 7
Results of Regression Analysis (Quarterly Basis, 1955 Through 1973)


Variable

Ai


United
States

-629.*
(-3.58)


Japan

881.*
(2.377)


Developed Na-
tions Excluding
US & Japan


93.46*
(9 11n


Latin
America


772. *
(4.30)


Less-Developed
Nations Exclud-
ing Latin Americz


-361.*
(-12.9)


t,i .153 .2438* -.0014 -.012* .0133*
(4.10) (3.82) (-.298) (-2.69) (4.23)


st-li .155* -.074 .0085 .0051 .0606*
(3.76) (-1.025) (1.72) (.4588) (14.13)

Lt (Mt 210. -41.01 .94 97. 105.
in Japan (.643) (-.648) (1.085) (1.34) (1.35)



tin 225. -177.* 68. 113. -5.28
Japan) (.6959) (-2.79) (.383) (1.57) (-.068)



Rt -.11 -.059 -.040 -.011 .012
(-2.119) (-1.65) (-1.38) (-.95) (.944)

Rt- -.026 -.085 .027 .010 -.0092
(-.501) (-2.39) (.900) (.87) (-.70)

Pt .00075 .002984 .00096 -.0005 -.00257
(.124) -.71) (.2742) (-.397) (-1.67)

Pt- .0013 -.0054 .0017 -.00039 -.0022
t-1 (.23) (-1.35) (.5140) (-.301) (-1.52)

Ht .283* .163 -.048 -.027 .060
(2.16) (1.83) (-.722) (-.905) (2.00)

H -.1296 .0892 .0822 -.027 .011
t (-.9783) (.8766) (1.23) (-.73) (.387)

Coefficient
of determi- .9764 .9617 .5944 .9478 .9597
nation

Durbin-
Watson 1.82 2.04 1.97 2.23 1.813
Statistic


Note: First entry in each cell
test result.


is regression coefficient; second is 't'


*: Significant to the .05 level of significance, two-tail test.


1 '


7


-^--------------------------









ficients derived from the regression analysis.1 As indicated in Table

7, Lt-l,Rt-,P t-, and Ht-, are not statistically significant in any

sector at the .05 level of significance (the sole exception is Rt-1 in

the Japanese sector). These variables were eliminated (except Rt-1 in

the Japanese sector) from the model.

A suggested reason for the general non-significance of Ltl,Rt_-1

Pt-1, and Ht-1 is that the impact of these variables is through the

actual cost of financing or the expected changes in the costs or avail-

ability of financing. The impact of either of these should be on the

current volume of bankers' acceptance created. Therefore, the main

impact of t-l values of these independent variables should be on the

t-l volume of bankers' acceptance outstanding. While some of the ac-

ceptances created due to period t-l values of these variables will be

outstanding during period t, their relative importance will be reduced

due to: (1) the volume of period t's new creations and (2) maturing of

acceptance created during period t-l.


Statistical Analysis Using
the Revised Quarterly Model

The regression analysis as applied to our quarterly model as re-

vised with the deletion of Lt-1,Rt-1,Pt-1, and Ht-1 (except for Rtlin



IA two stage iterative process was used to overcome autocorrelation
difficulties. For an explanation of this process refer to D. Cochrane
and G.H. Orcutt, "Application of Least Square Regression to Relation-
ships Containing Autocorrelated Error Terms," JASA (Washington, D.C.:
American Statistics Association, March 1949), pp. 305-310.
The likelihood of autocorrelation was constantly monitored through the
use of the Durbin-Watson statistic, but with the iterative process the
Durbin-Watson statistic was not significant (therefore, not indicating
autocorrelation) at the .05 level of significance.










the Japanese sector) is summarized in Table 8. These results, along

with further analysis, are used to test each part of our hypothesis.


Intternat ional Trade

Our hypothesis states that an increasing volume of international

trade should be accompanied by an increase in the volume of acceptance

outstanding; i.e.,the sign of the coefficients of St,i and St-1,i should

be positive. Significant positive coefficients were found for St,i in

the United States: .1468(4.11), Japan: .2330(3.79), and the less-

developed countries excluding Latin America: .014(4.73).2 These

support our hypothesis. Further support to the hypothesis is given

by significant positive coefficients for St-1,i in the United States:

.17(4.55), less-developed countries excluding Latin America: .059(14.86),

and the developed nations excluding the United States and Japan: .011

(2.19).

The significant positive coefficients for St,i and St-_,i for the

United States and for the less-developed nations excluding Latin

America suggest that at creation the maturities of the acceptance

created for these sectors are skewed towards the longest allowed

maturities (i.e.,skewed towards six months in maturity). The volume

of acceptance outstanding at the end of time period t is significantly

affected by the value of international trade for periods t and t-l.

The statistically non-significant coefficient of St-l,i for Japan,while

its St,i coefficient is statistically significant, suggests that the

maturity of creations for this sector may be skewed towards the shorter

maturities.



2Corresponding 't' test results are shown in parenthesis beside the
coefficients.









TABLE 8
Results of Regression Analysis Applied to Revised Model
(Quarterly Basis, 1955 Through 1973)


Variahle


Ai


United
States

-661.*
(-3.70


Japan

919.*
(2.85)


Developed Nations
Excluding U.S.
and Japan

84.88*
(2.32)


Latin
America

832.*
(4.17)


Less-Developed
Nations Exclud-
ing Latin
America


-356.*
(-12.92)


Is .1468* .2330* -.004 -.012* .014*
ti (4.11) (3.79) (-.809) (-2.64) (4.73)

S. 170* -.8505 .011* .0044 .059*
t-i (4.55) (-1.11) (2.19) (.407) (14.86)


Lt (Mt 249. -22.81 1.82 80.27 105.4
in Japan) (.79) (-.46) (1.14) (1.13) (1.44)

Mt-1 N.I. -144.77* N.I. N.I. N.I.
(-2.51)

Rt -.0907 -.0675 -.044 -.012 .0109
(-1.92) (-1.90) (-1.61) (-1.14) (.821)

Rt- N.I. -.0855* N.I. N.I. N.I.
(-2.38)

t .0024 -.0011 -.0014 -.00008 -.0019
(.48) (-.33) (-.428) (-.076) (-1.43)

Ht .233* .169* .02169 .031 .066*
(2.11) (2.02) (.533) (1.15) (3.30)

Coefficient
of determi- .9758 .9604 .5806 .9432 .9576
nation

Durbin-
Watson 1.76 1.98 1.95 2.23 1.81
Statistic

Note: First entry in each cell is regression coefficient; second is 't'
test result.
*: Significant to the .05 level of significance, two-tail test.
NI: Not included.









For the developed countries, excluding the United States and Japan,

St-l,i has a significant positive coefficient: .01(2.19), while its

St,[ coefficient is negative and not significant: -.004(-.809). A

possible explanation of this lag effect of Sti is that accept-

ances are used in this sector mainly for the storage of imports, not

for shipment of imports. If the shipment of a sector's t-1 imports,

St_,i' was not significantly financed by dollar acceptance, but the

storage of these imports during period t was, then St-l,i should not

be statistically significant to the volume of acceptance outstanding

at the end of period t-1 since acceptance were not created during t-1

to finance these imports. If acceptance are created during period t

to finance the storage of the imports which were shipped during t-l,

then St-_ would affect the volume of bankers' acceptance outstanding

with a one period lag (i.e., St-i would not be statistically significant

to Bt,i but St-l,i would be). This explanation of the lag effect

cannot be tested using the available data.

Latin America has a significant negative coefficient for St,i:

-.012(-2.64). A possible cause of this negative coefficient is the

unique use in this sector of dollar bankers' acceptance to create

dollar exchange combined with our use of imports and exports as the

relevant measure of international trade for this sector.3 Dollar ex-

change acceptance are created to finance a portion of this region's

imports during its low export season; exchange created during the high

export season is used to repay the maturing dollar exchange acceptances4



3Exports are included as a component of relevant international
trade since pre-export financing is more common in the Latin American
sector than in the other sectors.

4For a more complete explanation of dollar exchange acceptance,
see pp. 36-37.




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs