years. From 1917 to 1922 the United States Government made its first extensive foreign loans, those made to assist the Allies to procure ammunitions, foodstuffs, and other supplies from this country as well as advances for postwar relief and sales of surplus war materials on credit. The story of these loans and credits is told in chapter I.

From 1920 through 1930 the private investors in the United States subscribed to more than 10,000 million dollars of foreign bonds (after deducting refunding issues). About 45 percent of these were the issues of European countries, 25 percent Latin American, 20 percent Canadian, and 10 percent far eastern. Out of these and earlier flotations, and as a result of transactions in these and other foreign securities, the United States accumulated a portfolio which reached a peak about 1930 of approximately 7,000 million dollars.

Early in 1931, when there had been a serious decline in the prices of raw materials and the world was in the midst of depression, some of the Latin-American countries defaulted on their foreign debt. Within the next few years other borrowers, especially European, also defaulted. During the later years of that decade a few of the defaults were adjusted by temporary or by more permanent arrangements for the reduction of interest payments and extension of amortization over a longer period.

During the period of defaults, transactions in foreign dollar bonds 1 continued and resulted in net sales by investors in this country including substantial net sales of defaulted bonds to the issuing countries. These sales and the regular amortization of the nondefaulted issues reduced the par value of such bonds presently held in this country to about 2,000 million dollars, approximately two-thirds of which are Canadian issues. As of the end of 1946, 50 percent of the issues originally sold in this country were in default, either complete or partial (see table 30). As of the same date, only 25 percent of foreign dollar bonds held by persons resident in the United States were in default. The difference resulted from the large repatriations of the defaulted bonds reducing the holding of that type, and net purchases of Canadian issues which increased the investment in nondefaulted obligations. Data in table 30 are based on the total of foreign dollar bonds outstanding and take no account of the location of the owners. Furthermore, in some countries part or even all of the issues in default are those of state, municipal, or private corporate borrowers. Detailed information on the course and present status of defaults by countries is available in condensed form in reports of the Foreign Bondholders Protective Council, Inc.

1 Foreign dollar bonds are bonds of foreign governments and corporations expressed in terms of United States dollars. Usually they were originally sold to American investors in large blocks through public offerings by investment bankers.

difference in the tax burden indicates merely a different degree of government participation in the economic life of the country; and government expenditures, such as for social-security benefits, are merely an alternative to private expenditures for the same services in other countries. The Government of the United Kingdom, for instance, has held the cost of many of the necessities of life down to a low figure by absorbing a large part of the cost itself. This adds to the cost of government in the United Kingdom, while in countries that do not follow this practice the additional cost is borne by individuals as a personal expenditure outside of the tax structure. Instances of this kind can be multiplied; and, unless the data are limited to comparable functions of government, comparisons will in general only point to the differences in the services performed.

A further point to bear in mind is that in some countries the national governments perform many functions performed by local governments in other countries. For example, in times of peace the tax collections of the national government of the United States have been less than those of State and local governments combined (see table 38), and in Canada and Switzerland the national government receipts are normally the least important, while in France the national government is the dominant factor and local revenue receipts are a small part of taxes collected. Furthermore, since the war, many national governments act as purchasing organizations for some of the essential imports of foodstuffs and industrial raw materials. This makes their revenues and expenditures higher in proportion to their national incomes than would be the case for other countries.

Intercountry comparison of per capita tax and debt burdens is meaningful, therefore, only as a measure of the degree of participation of the government in the economy. Taxes and government debt are payments for goods, services, and additions to national wealth provided by governments. The greater the volume of goods provided and services rendered under government auspices, the higher the tax and/or debt burden.

Supplementary tables are included in this chapter, partly as a service to persons using the data to enable them to see in what manner the gross foreign currency data were put into the form of per capita amounts expressed in dollars. For this purpose table 33; Population and Exchange Rates, was prepared. In table 32, the national debt is divided into its internal and external components. The latter reveals the countries which are most dependent financially upon foreign countries. In a few cases, such as Hungary, the relationship between the internal and external debts is the result primarily of the postwar inflation which, for all practical purposes, wiped out the value of internal obligations.

Government revenue

In the answer to the item 11, the term tax burden was interpreted broadly to include other revenues which constitute a part of the burden imposed on the people for the support of the government. Insofar as possible data were collected regarding all foreign countries. However, it was impossible to obtain any substantial amount of information regarding state and local taxes and revenues and, accordingly, the data supplied relate exclusively to national governments.

The taxation and other revenue figures shown in table 31 are for the calendar years 1939 and 1946; or where the country is on a fiscal-year

basis, for the fiscal years which include the most months in 1939 and 1946. Actual taxes and revenues collected rather than budget estimates were used in all cases where the data were obtainable. Budget estimates are used when actual revenues are not available. These data were culled from a variety of sources including consular dispatches, official publications of the various governments, and Moody's Governments and Municipals. The publication, Tax Systems of the World, was especially useful for 1939 data concerning Latin America. Taxes include direct and indirect taxes such as excise, sales, estate, income, corporation, unemployment compensation and other socialsecurity taxes, customs duties, fees, and licenses. "Other revenues" include government royalties, interest, and dividends from government investments, proceeds from UNRRA sales, sales of public land, confiscated property and surplus property, but exclude loan proceeds, grants-in-aid, and reparation receipts. Net profits derived from government enterprises include such government undertakings and monopolies as the post office and railroads, telephone and telegraph, commodity monopolies, and exchange control. Each government enterprise in this category has been treated as a separate unit; consequently, where the expenditures exceeded receipts, the deficit was not deducted from the profits of some other unit. The total resultant revenue receipts, expressed in the local currency, was then converted to United States dollars at the average official exchange rate for the period.

Government Debt

The same procedures were followed in compiling the public debt of the various countries. Whenever available, the debt as of December 1939 and 1946, or the nearest available date, is broken down into its foreign and domestic components. Debts of government-owned corporations, to the extent that they are held outside of the government, are included. World War I debts to the United States have been excluded, as these are being shown separately in chapter I. Loans to private companies guaranteed by the government are excluded unless the government has been called upon to make good on its guaranty. Municipal debt guaranteed by the Federal government is considered as municipal rather than Federal. In addition to the above sources for the Government revenue, other main sources of debt data were the League of Nations' Statistical Yearbook and the Statesman's Yearbook.

National income

Data regarding the national incomes of various countries were obtained from official estimates where those were available. There is, however, considerable divergence among them with respect to the methods and principles of compilation and evaluation, particularly the inclusion and exclusion of a number of items. Sometimes these differences are due to the nature of the available statistics and sometimes they arise out of the peculiar structure of the economies to which they relate, and sometimes they are due to conceptual differences with regard to the content of national income.

Another difficulty marring the comparability of the per capita figures in dollar terms is the selecting of a proper exchange rate for translating the foreign currencies into dollars. As a consequence of government control of the exchange rates and the internal maladjust

TABLE 30.—Status of publicly offered foreign dollar bonds as of Dec. 31, 1946

1 Includes non-interest-bearing scrip issued in full or part payment of matured coupons.

2 Most issues in default as to interest are also in default as to sinking fund.

3 Includes $355,752 non-interest-bearing fractional certificates.

4 Includes $4,469 non-interest-bearing fractional certificates.

Estimated amount of American tranche of Kingdom of Rumania Monopolies Institute loan.

6 Bonds in default on sinking fund or principal only not included in the first 2 columns, amounted to $106,612,479, divided as follows: Dominican Republic, $3,640,500; Haiti, $6,094,479; Belgium, $9,544,000; Denmark, $81,788,500; Norway, $5,545,000.

Source: Institute of International Finance of New York University, Statistical Analysis of Foreign Dollar Bonds, appendix B, pp. 32-33, Bulletin No. 150, June 30, 1947.

CHAPTER IV. NATIONAL DEBT, REVENUE, AND INCOME

This chapter brings together the data in reply to five items in Senate Resolution 103 that relate to internal matters. These questions were as follows:

Item 10. The effects of the war on the public debt of the United States.

Item 11. The per capita tax burden of the people of the United States classified as (1) Federal, (2) State and local, and (3) total; and the total per capita tax burden of the people of each of the countries now in debt to the United States or with whom loan and investment discussions have been held by any American official since 1939.

Item 12. The total per capita debt burden for each of the countries mentioned in item 11.

Item 13. The latest reasonably reliable report on the national income, reduced to a per capita basis, for each country mentioned in item 11.

Item 14. The average interest rates for Government borrowing, according to the latest reasonably reliable report, in each of the countries mentioned in item 11.

GENERAL STATEMENT AND DEFINITIONS

The interpretation of data comparing the public debt, the per capita tax burden, or the national income of various countries is always subject to severe limitations. This is true not only because the data on which this information is based are usually incomplete and the series differ in their composition, but also because the same relative tax or debt figures do not necessarily imply a comparable economic impact or a comparable burden for the individual.

These difficulties are compounded at the present time when, as a result of the war, internal debts in some countries are being practically wiped out by inflation, when tax collections are slow and ineffective and cannot catch up with the progressive inflation of values, and when the yield of the tax system is greatly reduced by the low level of productive activity. Furthermore, in conditions of that kind, the measurement of government revenues and national income is most difficult and the results, particularly for purposes of comparison with prewar and with other countries, are of questionable validity.

All three sets of data-public debt, tax burden, and national income-have been brought together, in table 31, because the principal value of the data is in the relationship between them. The relationship between the tax burden and national income is probably most significant and that has been shown in the table in the form of a ratio. However, it should be pointed out that the economic impact of the same relative per capita burden might differ considerably as between countries, depending upon the structure of their tax system. The same total revenue could be raised in one case by a tax system in which a graduated income tax would provide the bulk of the revenue, while in another country the chief reliance would be put on customs and excise taxes. The effect in either case on production, consumption, and incomes might be quite different.

Furthermore, the type of expenditures to be financed from the tax revenue should also be taken into consideration. In some cases the