Monday, Mar. 08, 1937

Trust Performance

Well into its second year is SEC's investment trust investigation, the specimen under the microscope last week being Harrison Williams' Central States Electric Corp. (see below). SEC's findings to date have not been precisely flattering to investment trust management--which was not surprising in view of the fact that the investigation's prime purpose was to find out why investors lost upwards of $5,000,000,000 in investment trust securities during Depression. But in 1936 the trust managers demonstrated that they had learned a lot since they took off with ceiling zero in the late 19205. Not only did they mount the witness stand in Washington with penitence and goodwill, but also wrote a good record in their annual reports. In a study of 47 general management trusts, Tri-Continental Corp. last week announced that the average gain for the year had been 32%.

Trusts have hitherto objected strenuously to comparisons with stockmarket averages. This year Tri-Continental was only too glad to point out that in comparison with an average gain of 32% for general management trusts, Standard Statistics' 90-stock average rose only 27.9% the New York Times' 50-stock average only 21.1%, the 782 common stocks listed on the New York Stock Exchange about 25%. The Dow-Jones averages showed a 24.5% rise in industrials, 32.5% in rails, 17.9% in utilities. Actually the comparison was even more favorable to the trusts because part of their assets were in cash bonds, or preferred stocks. Moreover, the trusts paid taxes and set up tax reserves which, as Tri-Continental dutifully observed, "do not enter into the market averages." Tri-Continental's own improvement was 32.1%.

Mutual type trusts did a little less well than general investment companies, Tri-Continental's average for 22 mutuals being 30.9%. A list of 15 specialized trusts with portfolios concentrated in one industry varied widely in performance, showed a gain on the average of only 22.2%.

This file is automatically generated by a robot program, so reader's discretion is required.