to justify counting both the earnings on the policyholders' money and the lower premiums made possible by these earnings. Such a calculation is not correct.

Mr. Bailey concludes his analysis of the alleged bias in the ADL formula by noting that companies which have greater ratios of writing to surplus or higher ratios of reserves to surplus (such as mutuals and reciprocals) are reported by ADL as showing lower returns than companies with lower writing and reserve ratios (stock companies). Careful analysis, however, will show that unfortunately it is not any bias inherent in the ADL formula which produces these results, but the inherent nature of the present insurance industry that causes companies who do more writing and/or keep larger proportions of their assets in bonds to earn lower rates of return. The data strongly suggest that this is due to underwritings being relatively unprofitable and bond investment yielding, in total, less than stock investments. Under such circumstances, we do not understand what Mr. Bailey means by "efficiently" when he states that insurance companies "use their resources most efficiently by maintaining the highest leverage of premiums and reserves to net worth" (p.9). By such reasoning the buggy whip maker who around 1910 channeled his resources into more plant and equipment rather than out of the buggy whip industry would have been considered to be making the most efficient use of his resources. Likewise for the insurance investment manager who supplied his manufacturers with capital. As an economist I cannot agree with these propositions.

Mr. Bailey demonstrates far better than anything in the ADL report or anything I could say that it is the inherent nature and structure of the insurance industry, and not any bias in the ADL formula, which places insurance returns at the bottom of all other industry returns. Mr. Bailey notes that a company which received payments as earned and paid losses as incurred would have changed profits (its premiums would increase and its losses be reduced by an amount equal to the investment profit it used to earn under the old system). Mr. Bailey continues, "Such an insurance company would have no reserves for unearned premiums or unpaid losses. Its rate of return calculated by the formula used by the ADL report would be higher than the rate of return for a competing insurance company that collected annual premiums in advance . . (p.9). Mr. Bailey elegantly shows his reader that leaving the numerator (net income) unchanged and lowering the denominator (invested funds) increases the value of the fraction (rate of return). We agree.

We

Mr. Bailey's example has also shown something more revealing. It should be recalled that the ADL report stated only that the present rate of return in the insurance industry appears to be low in certain senses. did not state how this situation ought to be corrected. We did not say, for example, that profit (the numerator) should be raised nor that invested funds (the denominator) should be lowered. All we said was that the way the insurance industry is currently run produces an unsatisfactory rate of return; unsatisfactory that is, from the point of view of society. Mr. Bailey's example of changing the payments pattern and his remarks (p.8) concerning

"overcapitalization" shows how a fundamental, institutional change in

the operations of the insurance industry is likely to produce a marked change in its rate of return. I have urged on numerous occasions that those who are seriously concerned with the problems of the insurance industry turn their sights to the basic institutions and structure of the insurance industry for it is through changes in those areas that relief may well be forthcoming. Juggling with profit and rate making formulas, will produce no

relief for the insurance consumer.

We conclude that Mr. Bailey's allegation of bias in our formula is untrue. Our formula measures the return generated by all funds flowing into an insurance company. It neither penalizes nor rewards companies with larger reserves or premium to surplus ratios. If such companies show up as being less profitable, we suggest that it might be because their investments produced less income and/or they suffer higher underwriting losses. We believe that it is the inherent structure of present insurance operations and

not accounting or actuarial phenomena which produce the current unsatisfactory rates of return in the industry.

IV. RETURN ON NET WORTH

While we cannot accept his justifications, we can accept and do appreciate Mr. Bailey's desire to use return on net worth as a measure of comparable earnings between insurance companies and other industries. We feel the return on net worth measure inappropriate when discussing problems of insurance capacity and problems of stockholder owned insurance companies. However, in relying exclusively on this measure, Mr. Bailey leaves unanswered questions concerning the measurement of return on mutual and other non-stock insurance enterprises, the social reasons for measuring the efficiency of all assets employed as distinct from the efficiency of the employment of equity financed assets, and the effect of comparing industries with differing capital structures. How would the return to net worth measure be useful in these cases? Even when using return on net worth is appropriate,

we must emphasize one guiding principle in its use: the return on net worth for stock insurance companies must be compared with the return on net worth for other industrial or financial enterprises and, further, such comparisons must give due consideration to alternate employments of capital within a risk/return framework.

Most practitioners of financial analysis, as well as professors of finance and economics, regard the text Security Analysis--Principles and Techniques, by Graham, Dodd, and Cottle, as the "Bible of Security Analysis." The entire viewpoint of the text is parochial in nature; that is, it offers advice to investors seeking the profitable employment of their funds. Yet,

when they discuss profitability ratios, Graham and Dodd prefer to use the total return on invested funds rather than the return to net worth. The

authors note:

"The best gauge of the success of an enterprise is the
percentage earned on invested capital, i.e., on the long-
term (non-current) debt and preferred stock plus the book
value of the common stock. This percentage, or rate of
return, is the ratio to total capital of the final net
profit available for capital funds. Thus it reflects all
recurrent items of profit and loss, including income tax,
but not deducting interest on funded debt. The fundamental
merit of return-on-invested-capital ratio is that it
measures the basic or over-all performance of a business
in terms of the total funds provided by all long-term
investors--rather than a single class."

The editors of Forbes Magazine, who report the return on equity figure, clearly indicate that this statistic measures only the efficiency with which corporations employ owners funds and tell nothing about the corporation's

total efficiency. They explain the net income to net worth measures as follows: "By comparing equity capital with net earnings, we are showing how efficiently management is managing stockholders' property."*

Mr. Bailey claims that ADL calculated the return on net worth for all types of insurance companies. He quotes our figures for return on Policyholders' Surplus, and identifies them as the ADL-calculated values for return to Net Worth. It is inaccurate to say that we in any way implied that our N4/DI measure (Net Income/Policyholders' Surplus) was a measure of return on net worth.

**

*B. Graham, D. L. Dodd, and S. Cottle, Security Analysis--Priciples and Techniques McGraw-Hill Book Company, Inc., New York, 1962, Fourth Edition, pp. 233-34.

For bes, Jan. 1, 1969, p.37, emphasis added.