Appendix J contains a large number of excerpts, taken from financial journals during 1933, which reflect in chronological order the American and leading foreign security markets' appraisal of the year's transactions insofar as they affected the United States. The principal emphasis during the year as a whole was placed on American purchases of foreign securities, usually viewed as a direct result of the depreciation of the dollar. That abnormal factors influenced this class of transactions is suggested by the fact that in 1933 the item was second in importance in the international securities movement, whereas in 1931 and 1932 these transactions were distinctly the least important of the four classes of operations. On the other hand, it is apparent that there was a widespread overestimating of the relative statistical importance of the effects of the depreciation of the dollar. The "flight" of American- and foreign-owned capital from the United States to foreign countries through the medium of the security markets apparently was fully offset during the year as a whole by a reverse flow of investment funds to the United States. Estimated foreign purchases of American securities aggregating $760,000,000 exceeded repurchases of American securities by $180,000,000, while American purchases of foreign securities exceeded purchases by foreigners (not to be confused with repatriations, which do not include all such purchases by foreigners) by $120,000,000.

That it would be exceedingly difficult, in the absence of exact statistical data, to make a correct appraisal of the international security movements throughout 1933 is indicated by an examination of the excerpts quoted in appendix J. On several occasions simultaneous statements reported contrary tendencies. Declines in American shares were attributed to foreign selling, until statistical evidence appeared which modified the impression. (See excerpts from New York Times, Oct. 14, 1933.) In the light of the actual data it cannot be said that the reported appraisals were in error as much as they were incomplete. Public attention throughout the year was riveted largely on the factors which particularly influenced two classes of international security transactions: (1) repatriations of German dollar securities and (2) American purchases of foreigncurrency securities. Both probably were items of large proportions, although the reported data do not make possible an exact evaluation. The latter class of purchases has been discussed in an earlier section 18 of this bulletin and requires no additional comment.

REPATRIATIONS OF FOREIGN SECURITIES

In the discussion of security repatriations at least two important facts must be constantly borne in mind. It must again be noted that foreign securities held in the United States and purchased by foreigners are not necessarily "repatriated" by the transaction. For example, European investors have at various times purchased LatinAmerican dollar bonds in the American market. While it is probably safe to assume-especially in 1933-that the transactions reported as "repurchases of foreign securities by foreigners" represent in large part actual "repatriations", it is important to bear in mind that the latter term is less inclusive than the former.

18 The Dollar in Foreign Exchange in 1933.

Also, the fact is often overlooked that the data collected for balanceof-payments purposes necessarily apply only to the aggregate market values at time of sale of the securities involved. The group of foreign dollar securities which have constituted the most important class of repatriated issues during the last 2 years have been quoted on the average at prices ranging roughly from one-fourth to two-fifths of par. It is possible that, even after allowing for repurchases by investors in third countries, the actual repatriations on a par-value basis were much higher than the reported repurchases of foreign securities, estimated at $565,000,000.19

Various factors support the popular assumption that dollar-bond repatriations have been taking place in substantial volume. For example, foreign mortgage banks whose bonds were floated in the American market have in some cases accepted their own obligations at par or at comparatively small discounts from their debtors, and the latter thus received, in time of low quotations in the American market, special inducements to purchase such securities. Again, the systems of exchange control set up in numerous countries during the past 2 years have either restricted or totally prevented security arbitrage transactions between important security markets, with the result that in some cases this situation was successfully employed by foreign authorities, notably by Germany, as a means of effecting repatriation on a profitable basis. German exporters have been permitted to use part of the dollar proceeds of their sales in the United States to purchase, for the purpose of resale in the German market, such German dollar bonds as could be purchased in the United States at prices below quotations within Germany. This differential accrued to the advantage of the German exporter, and thus enabled him to quote lower prices on merchandise than he could without such a subsidy." The self-generative character of this procedure, particularly when supplemented by the use of scrip issued in lieu of partial interest payments on outstanding German bonds as well as of blocked marks, has undoubtedly become an important factor in the repatriation of German dollar issues.20 The large discounts at which German corporations and other debtors could retire their outstanding obligations were also an important factor. The German Ministry of Economy has recently reported 21 that between November 1931 and November 1933 the total amount of repurchased bonds amounted to 781,000,000 marks, of which 547,000,000 marks were repurchased under supplementary export procedure and 121,000,000 marks taken in exchange for other "foreign" securities. These figures presumably represent par value and include, of course, repatriations of German securities from other countries than the United States.

Since the greater part of these securities were floated in the United States, it is safe to assume that the greater part of the repurchases were made in the United States. When account is taken of the fact that the figures of the German Ministry of Economy apply only to a 2-year period, it appears probable that total repurchases by foreigners,

19 It is, of course, possible that certain securities enter into more than one international transaction during the year. It must be borne in mind that all figures reported in the international securities movement, except arbitrage transactions, are gross rather than net figures. During a year when markets are influenced by numerous cross-currents, as in 1933, extreme caution must be used in drawing conclusions from the mere size of a given item.

20 See also section on Short-Term Capital Movements, and appendix K.

21 See also excerpt from New York Herald Tribune, Jan. 1, 1933, in Notes on International Securities Movement, appendix J.

to date, of German dollar issues originally floated in this country (approximately $1,450,000,000) aggregate somewhere between $350,000,000 and $450,000,000 par value.22

Various other foreign dollar issues originally floated in the United States have no doubt been sold back to foreigners in substantial volume. For example, when the United States suspended gold payments the Belgian Government altered the amortization scheme of Belgian dollar loans and invited the holders of such bonds to have them stamped at the National Bank of Belgium under conditions with which it was impossible for American holders to comply. Out of approximately $150,000,000 outstanding of four issues, nearly $100,000,000 were stamped, indicating the extent to which these issues had actually been repurchased by foreigners. Attention has already been called to the Swiss Government 51⁄2-percent $30,000,000 issue which has passed almost entirely out of American hands. It is therefore reasonable to believe that repurchases by foreigners and actual repatriations of issues other than the German have been quite substantial.

PURCHASES OF AMERICAN SECURITIES BY FOREIGNERS

With a view to supplementing the results of the questionnaire survey of the year's international security transactions, the Finance and Investment Division compiled, with the cooperation of 52 representative American corporations, several tables designed to show the changes during recent years in the foreign-held shares of American railway, utility, and industrial enterprises. The year-to-year changes as reported by these corporations (16 railroads, 9 utilities, and 27 industrials) are shown in table 18.

No attempt is made to draw far-reaching conclusions from the results of the survey. The records of the corporations, or transfer agents, do not show shareholdings by foreigners in the name of American nominees, and the reported figures are, therefore, not necessarily complete. Shares held abroad in arbitrage accounts fail to represent the true foreign investment interest in American equities. In comparing the results with the figures in the year's reported international securities movement, it must also be remembered that the trend in the number of shares is not necessarily an accurate measure of the trend in the dollar volume of investment.

TABLE 18.-CHANGES IN FOREIGN HOLDINGS OF COMMON SHARES OF AMERICAN CORPORATIONS, 1930-33

22 Owing to redemptions and cancelations, it is probable that the amount outstanding is not over $250,000,000 to $350,000,000.

These qualifying factors are probably of minor importance, however, when applied to as widely representative a group of corporations as is covered in this survey. The latter include practically every American enterprise whose common shares have been regarded for years as special investment favorites abroad. The majority of the industrials and half of the utilities are still on a dividend-paying basis. The sampling is believed to be sufficient particularly in rails and industrials to provide an accurate indicator of the general trend in foreign holdings of American corporate shares. It is therefore believed safe to conclude that the results of the survey are much more than an accidental corroboration of the results of the questionnaire survey of international security transactions relative to the purchases by foreigners of American securities during 1933.

Among the individual corporate groups, foreign-owned common shares of American railroads showed a net liquidation, as contrasted with utility and industrial equities. The net selling of rail shares by foreigners apparently reflects the general foreign appraisal of the railroad share market, inasmuch as nearly all the reporting companies reported a decline. All but 2 of the 16 rail issues are no longer on a dividend-paying basis. In the case of foreign-held utility stocks the situation appeared more mixed. Although foreign holdings showed a net increase during 1933, this change took place in the face of declines reported by several outstanding companies. Among the industrial corporations, the majority reported that their foreign-held shares increased in 1933, and it was this group which accounted chiefly for the increase reported by all corporations.

It is impossible to determine to what extent the depreciation of the dollar attracted foreign funds into American equities during 1933, a year featured by intermittent flights from the dollar. Reported data on foreign-owned preferred shares, while considerably less complete than the statistics on common shares, do not reveal an especially marked tendency of foreign investors to withdraw from fixed-dividend shares. Of 17 companies which reported the changes during 1933 in the shares of their preferred stock held abroad, 6 stated that the year's change was upward, 8 reported a decline, and 4 reported no change.

Nine of the companies reported the annual changes of their foreignheld common shares by individual countries. The results are shown in table 19.

TABLE 19.-CHANGES IN GEOGRAPHIC DISTRIBUTION OF FOREIGN-OWNED COMMON SHARES OF AMERICAN CORPORATIONS 1

1 The figures in this table are the aggregate reported by 9 corporations as follows: 2 steel companies; 1 leading concern in the mineral industry; 1 outstanding meat-packing concern; 1 leading oil producer; 1 large office-equipment manufacturer; 1 outstanding utility; 1 important railroad; and 1 leading food-manufacturing concern.

2 The year's sharp increase was due largely to one company's issue of shares to a foreign corporation in exchange for other securities.

3 This increase was due largely to the rise reported by one company, and included holdings in some European countries not separately reported by this particular company.

INSTITUTIONAL TRANSACTIONS

No data are available which make possible a classification of the types of buyers and sellers who dominate the international securities movement. Such institutional investors as banks, life-insurance companies, and investment trusts frequently determine the more pronounced trends because of the large individual transactions involved. Private investors are undoubtedly a strong influence at certain periods, particularly in times of banking crises and monetary instability.

Two excerpts in appendix J under date of November 19, 1933, call attention to a canvass made of American investment trusts relative to possible shifts of funds by them from American to foreign securities. The conclusion that American investment trusts did not participate in any flight from the dollar is supported by the information reported to the Finance and Investment Division by several of the leading trusts known to hold foreign issues in their portfolios, which indicated that on the whole the foreign holdings of investment trusts in this country underwent little change. No adequate data are available bearing on the operations in American markets by foreign trusts.

Activity by American life-insurance companies was apparently limited to possible shifts from one foreign security to another. The net changes in foreign holdings were negligible. The investments and reserves of 51 legal reserve life-insurance companies, as reported by the Association of Life Insurance Presidents, 23 showed an increase in the holdings of Canadian Government bonds from $448,403,000 on December 31, 1932, to $450,290,000 on September 30, 1933. Holdings of other foreign government bonds declined during the same period from $24,712,000 to $18,547,000.24 Data on portfolio changes of foreign life-insurance companies normally effected through the American markets are not sufficiently complete to warrant any conclusions.

Licensed member banks of the Federal Reserve System reported holdings of foreign government and other securities of $313,000,000 at the end of 1933, as compared with reported holdings of $450,000,000 at the end of 1932. These figures, however, overemphasize the extent of the year's net liquidation of such securities, since 529 member banks, exclusive of banks placed in liquidation or receivership, were still unlicensed to renew operations on December 27, 1933.25

Data relative to other institutional dealings are incomplete or totally lacking. In some cases issuing corporations both here and abroad took advantage of the prevailing low quotations to purchase their own issues. As already indicated, German exporters of merchandise to the United States engaged in the purchase of German dollar issues in the American market as a result of special advantages accruing to them under German transfer regulations. It appears certain that in the year's international security movements as a whole, private buying and selling was an important factor.

23 Walker, Bradford H., Life Insurance Investments as a National Resource, address delivered at the 27th Annual Convention of the Association of Life Insurance Presidents, New York, Dec. 7, 1933.

24 These changes may have been influenced also by changes in the basis of valuation.

25 See footnote 24.