The preponderance of the argument appears to be in favor of a reduction in the size of the specific exemption. The precise extent of that reduction, however, is subject to conjecture. Since the principle of granting exemptions emanates from the desire to provide for the dependents of the decedent, it has been suggested that the exemption be that amount which if invested, would yield the amount of income which it is desired to exempt from income taxation. /2/ A 5 percent investment of $20,000, for instance, would yield $1,000 annually, or an amount equal to the relatively high exemption afforded a single individual under the Federal income tax. Whether this be an acceptable basis for determining the size of the estate tax exemption or not (and there is much that may be said in criticism of it), the $20,000 figure is not unreasonable. A reduction to $25,000 was actually approved by the House as early as 1917. Walradt in his confidential report to the Secretary, entitled "The Federal Estate and the Federal Gift Taxes," recommended a reduction to $30,000.

A reduction of the estate tax exemption below $20,000 has little in its favor. The desire to bequeath a substantial tax-free land to one's heirs is too deeply entrenched in American social and economic concepts to be readily dislodged. Furthermore, the accompanying administrative task involved in the greatly increased number of tax returns would be disproportionately costly. Finally, the elimination of the insurance exemption recommended on page 14 below together with the proposed reduction of the specific exemption represents an effective exemption reduction to one-fourth of its present level. For these reasons it is recommended that the estate tax exemption be reduced to $20,000.

2. Exemptions varying in accordance with number of beneficiaries, and their relationship to the decedent.

The Federal estate tax recognizes neither the number of beneficiaries nor the relationship of said beneficiaries to the decedent in the computation of the tax burden. Some students of the problem consider this a serious deficiency which should be corrected. Their grievance originates with the contention that the specific exemptions provided by death tax laws are designed to protect those who were dependents of the decedent prior to his death. In the guise of this reasoning, though actually in the hope of reducing tax liability, there. has developed from time to time pressure for converting the present estate tax to the inheritance tax type of levy.

Others have suggested that the Federal estate tax be supplemented by an inheritance tax in the manner at present employed in a number of foreign countries and one or two of our own States. Great Britain, for instance, imposes in addition to the estate tax a legacy duty on personal property and a succession duty on real estate. Both of these are graduated but the degree of graduation is based on consanguinity and not on the size of each beneficiary's share. A similar situation prevails in France, in Germany, and some of the Canadian and Australian provinces. In the case of the American States, Oregon and Rhode Island supplement their inheritance taxes with estate levies.

In his message to Congress, dated June 19, 1935, the President said, in part:

"I recommend * * * that in addition to the present estate tax there should be levied an inheritance, succession and legacy tax in respect to all very large amounts received by any one legatee or beneficiary."

Following the President's recommendation, the House actually passed H.R. 8794 which provided for an inheritance tax at rates progressing from 4 percent for the first $10,000 to 75 percent for shares over $10,000,000, in excess of a $50,000 exemption to spouse and near relatives and $10,000 exemption for all others. When the bill reached the Senate, however, the Administration and House proposal was rejected and in its place a further upward revision of the estate tax substituted.

Without entering here upon a detailed discussion of the relative merits of estate and inheritance taxes, it is suggested that the use of the inheritance tax either as a substitute for or as a supplement to the estate tax, would at present be undesirable. Inheritance taxes, to be sure, are more equitable but that additional equity is of a rather fine order. If equity is desired, the estate tax itself offers numerous opportunities which should be explored first. Furthermore, the estate tax is easier to administer. To quote the Joint Committee on Internal Revenue with respect to the estate tax:

"(It) eliminates the necessity for considering the many complicated problems which arise in connection with the construction of wills and trusts, the application of probate laws, and the determination of the rights of the particular legatees. These problems would be especially difficult in the case of the Federal Government, as the regulatory control over the passing of property at death is reserved to the States and there is a great divergence in the various State laws."

It was largely its ease of administration that gained the estate tax the support of the National Chamber of Commerce and the National Tax Association. Furthermore, in order to yield the same revenue as an estate tax, the rates applicable to distributive shares under a graduated system of inheritance taxation would have to be substantially higher than those required by an estate tax in all cases and confiscatory in others.

However, required recognition of the number of beneficiaries and their relationship to the decedent need not necessarily imply the use of an inheritance tax. The same objective could be obtained by providing that type of an exemption in the estate tax. This is a procedure now employed in the North Dakota and New York estate taxes, and is one which is frequently used in foreign countries, particularly in New Zealand and in some Canadian and Australian states. In the New York law, for instance, a surviving spouse is allowed a $20,000 exemption, and each of the lineal descendants or ascendants are permitted an additional exemption of $5,000. All of these must, however, be taken from the first bracket and therefore cannot exceed $150,000. In the North Dakota law, the respective exemptions are $20,000 for surviving spouse, $5,000 to minor descendants, and $2,000 to adult descendants or ascendants.

It can be readily seen that exemptions provided in this manner could exceed in the aggregate the $20,000 total exemption recommended above. To guard against this contingency, and to forestall possible tax evasion by spreading legacies widely among members of the family, the above arrangement could be combined with a limit of $20,000 upon the aggregate exemption allowed against any single estate. If the exemptions were limited to surviving spouse and minor children of the decedent, the necessity for the maximum limit would not be absolutely essential, because of the relatively small size of the American family.

Proceeding from the philosophy that specific exemptions are intended to protect the decedent's dependents, it has been suggested that specific exemptions not only recognize the relationship of the beneficiary to the decedent, but that they recognize the life expectancy of the beneficiary. The plan was first proposed by Simeon Leland /3/ and is explored in detail by Walradt in his confidential report on death and gift taxes, cited supra. As a corollary to this principle, it has been suggested that the case for an exemption afforded a child disappears when the child becomes of age, and therefore, that the exemption should be made to vary inversely with the age of the child, being maximum in the case of newly-born babes and disappearing entirely in the case of children 21 years old. By the same token, in the case of children who are mentally or physically incapacitated from ever earning a living, the exemption afforded should be large enough to yield a reasonable annual income during the life expectancy of the child -- enough to make it unnecessary for them to become public charges.

That the arrangement described above, involving the recognition of the number of beneficiaries and their relationship to the decedent in determining the amount of the specific exemption would be desirable, is probably true. Nevertheless, because such a revision would involve a refinement of a finer order than that commensurate with the existing features of the estate tax, as well as with the existing features of our tax system as a whole, and because its operation would involve a more cumbersome administrative task, its adoption at present is not deemed desirable.

3. Constant, conditional, and vanishing exemptions

The $40,000 specific exemption provided by the present estate tax, the reduction of which is urged above, is a constant exemption; it is, in other words, unaffected by the size of the estate and is deducted from the total estate before determining net taxable value, no matter how large the estate.

The constant or continuous form of specific exemption is common, not only among the American States (Table 6), but also in some foreign countries. It is employed by the Commonwealth of Australia and by several Canadian and Australian states. Its popularity notwithstanding, this form of exemption has little to recommend it other than simplicity. As already noted, a specific exemption can primarily be justified on the ground of the dependence of the beneficiary upon the decedent. Supplementary support may also be found for it in the ability to pay principle. When, however, the size of the estate exceeds the amount required for the support of the beneficiary, the need for and justification of the specific exemption disappears.

Two alternatives may be suggested for the rectification of this weakness of constant exemptions. One of these is the so-called conditional exemption. In this form of an exemption, an estate is untaxed as long as it does not amount to more than the amount of the exemption. When, however, the value of the estate exceeds the amount of the exemption, the entire estate is taxed. Maryland and Massachusetts and to a limited extent New Jersey death tax laws employ this feature at present. In its commonest forms, this type of exemption fails to provide for a gradual transition from nontaxable to taxable estates and therefore tends to work injustice. Thus, in many cases, an addition of $1 in excess of the exemption to the value of the estate increases tax liability by a substantial sum. In the case of the Commonwealth of Australia, for instance, an estate not exceeding A$1,000 in value is wholly exempt, but an estate of A$1,001 is dutiable in full at the appropriate rate, namely, 1 percent. To quote the Royal Commission, "The difference in valuation of A$ 1 in these cases makes all the difference between total exemption and total dutiability." This difficulty of the conditional exemption may be overcome by the so-called step system, which provides that where estates not exceeding fixed amount are exempt from duty, the duty imposed on estates exceeding such amounts should not be greater than the excess of the estate over that amount. This is the principle employed in the British income and estate taxes.

A theoretically more acceptable alternative is the so-called vanishing exemption which provides for a gradual diminution of the specific exemption as the size of the estate increases. Let it be assumed that the specific exemption is $20,000. Under the vanishing exemption principle, estates amounting to less than $20,000 would be untaxed; but if the net value exceeds $20,000, the exemption would diminish in accord with a prescribed scale, until it reached the vanishing point. If, for example, the exemption diminishes $1 for every $1 by which the value of the estate exceeds $20,000, then if the total value were $21,000, the exemption would be $19,000 and the taxable amount $2,000; but if the value were $40,000 (which in that case would be the vanishing point) the whole amount would be taxable.

Obviously, this form of an exemption need not necessarily diminish dollar for dollar, but may just as easily decline at a lesser rate. A diminution of one dollar for every two dollars, for example, would eliminate all exemption at $60,000; one for every four dollars, at $100,000. The vanishing exemption, already employed in two estate tax laws (Kentucky and West Virginia), has the advantage of affording substantial justice to the beneficiary and at the same time assuring a marked increase in revenue by eliminating the exemption afforded medium- and large-size estates by the constant exemption.

Sight should not be lost of the fact that the use of the vanishing exemption complicates the task of tax computation, especially in instances of deficiency assessments or refund claims on estates not in excess of the vanishing point. In comparison with the gains in equity inherent in the vanishing exemption however these administrative difficulties are of secondary significance. It is therefore recommended that the present constant exemption be replaced by one based on the vanishing principle and that the exemption diminish dollar for dollar as the net estate exceeds $20,000, eliminating the exemption entirely on estates amounting to $40,000 or more.

4. Manner of taking and manner of providing exemption.

The present method of including the specific exemption as one of the deductions to be made from gross estate in order to arrive at taxable net estate, is undesirable. It is undesirable from the point of view of equity because the exemption is in fact taken at the rate applicable to the highest bracket and thus represents an unequal tax reduction which increases with the size of the taxable estate. The present $40,000 exemption represents an $800 tax benefit to a $50,000 estate as compared with a $20,000 benefit to a $50,000,000 estate.

In addition to its inequitable features, the present arrangement is undesirable because it involves a loss of revenue. It reduces the amount of the net estate in the top bracket that is subject to the highest rate of the progressive rate schedule applicable to a given estate.

That a specific exemption which is taken from the highest bracket is inequitable is obvious. It should be noted, however, that it works the greatest injustice in those cases in which the specific exemption is of the constant type. It disappears entirely in the case of the conditional exemption and is present in the case of the vanishing exemption only in the event that more than one rate is applicable to the lowest brackets before the exemption vanishes.

It should be recognized that this inequitable feature of specific exemptions is not confined to the present Federal estate tax and, in fact, is not confined to this particular feature of the Federal tax structure. It is present in the death tax laws of many States and many foreign governments and is also present in the Federal income tax.

In the case of State death taxes, twenty States provide that the specific exemption be deducted from the first bracket; twenty-one allow it to be taken from the last bracket; two have vanishing exemptions; two, conditional exemptions; and two others a combination of methods (Table 6). In Germany, where Class I and Class II beneficiaries are granted a continuous exemption, the exemption is deductible from the last bracket.

Obviously, the difficulty here presented will lose its significance if the above recommendation for vanishing exemptions is adopted. If, however, the present form of constant exemptions is retained, the inequity resulting from this difficulty should be rectified. So long as the present bracketed rate structure is retained, this can be accomplished by specifying that the exemption be taken out of the lowest bracket. The proposal involves no new principle. As already observed, it is incorporated in the death tax laws of almost half of the States.

If, however, the existing rate structure is replaced by one on the totality plan as recommended on page 21 below, then the inequity can be removed only by providing the exemption in terms of tax credit rather than in an amount deductible from the taxable base. Under this arrangement the tax liability is computed after deductions but before exemptions, and from the tax liability computed in that manner there is deducted the tax exemption. This is a procedure which has been followed in connection with certain State income taxes (Arizona, Iowa, South Dakota, Wisconsin), and may well be adopted in connection with the estate tax. It has the advantage of providing a uniform tax benefit to estates irrespective of size but, of course, does not incorporate the principle of vanishing exemption. The two could be combined, but that possibility is here no further explored since the vanishing exemption discussed above is more desirable. An exemption provided in terms of tax credit is psychologically undesirable in that it seems smaller to the taxpayer than one deductible from the tax base.

However, in the event that a totality rate schedule is adopted but the specific exemption is retained without the incorporation of the vanishing exemption principle, the present method of providing an exemption in terms of a deduction from the tax base ought to be eliminated and in its place, an exemption in terms of tax credit substituted.

5. Insurance exemption

Under the present Federal estate tax, the first $40,000 of insurance payable to specified beneficiaries other than the executor, under policies taken out by the decedent upon his own life, is tax exempt. This provision was first introduced in 1918 and has been retained since. Its quantitative importance is revealed in Table 7.

From the social standpoint it is no doubt highly desirable that a man provide his wife and children with an assurance of a cash fund payable at his death. It does not follow, however, that such a fund should be free from the estate tax. If the fund is in the form of a bank account, it is subject to the estate tax. If it is in the form of insurance and does not exceed $40,000, it is tax exempt. Valid reasons for this discrimination have never been advanced. M. H. Hunter is convincing in saying, "The payment of insurance premiums is often looked upon primarily as an investment, with the payment at death as a return. This is true to a greater extent with the extremely large policies and no good reason appears for allowing their deduction." (Underscoring supplied.) /4/ Mr. H. F. Walradt, in his memorandum to the Secretary, previously cited, points out that "the main value of the present exemption is that it furnishes a selling point to insurance companies, whose agents do not fail to inform prospects that they can create an estate up to $40,000 by taking out life insurance and not have it subject to the Federal estate tax."

The practice followed by other countries with respect to insurance is varied. In the majority of cases, however, insurance is taxable in the same manner as other assets. Great Britain allows no discriminatory treatment in favor of insurance payable to named beneficiaries; neither do Australia, New Brunswick, Alberta, Ontario, or Quebec. Manitoba, on the other hand, allows a maximum exclusion of $10,000 of insurance payable to the widow and/or two children under eighteen years of age. Nova Scotia and Saskatchewan allow a maximum exemption of $5,000 for insurance payable to Class A beneficiaries.

In the case of the American States the pattern is not different from that found in foreign countries. New York's provision is novel in that it allows an exemption for insurance payable to named beneficiaries of an amount equal to the extent to which $100,000 exceeds the specific exemption claimed. Alabama, Florida and Georgia follow the Federal law; Mississippi allows a $20,000 and California a $50,000 exemption for insurance payable to named beneficiaries; Pennsylvania, Massachusetts, and Illinois exempt such items in their entirety. Wisconsin includes in the taxable estate all insurance payable at death.

The $40,000 exclusion for insurance payable to named beneficiaries allowed by the present Federal law is significant in that it permits a 100 percent potential increase in the specific exemption allowable. Furthermore, this additional exemption is not uniformly applicable. A net estate consisting of $40,000 in general assets and $40,000 in insurance payable to named beneficiaries is not subject to the Federal estate tax. A net estate consisting of $80,000 in general assets is subject to an estate tax of $2,000. The value of this insurance exclusion increases with the size of the estate inasmuch as the exclusion is in effect deducted from the highest bracket subject to the tax. Thus, at $1,000,000, the insurance exclusion is worth $10,400 in tax dollars; at $5,000,000, $21,200; at $10,000,000, $26,000. The insurance exclusion serves to place a great number of estates in the nontaxable category, thus effecting a significant reduction in aggregate Federal revenues available from this source and should therefore be eliminated.

6. Insurance for the payment of estate taxes

It has been urged that insurance payable to the estate for purposes of paying the estate tax should be exempt. Section 801 of H. R. 12395 (74th Congress, 2nd Session), for instance, proposed that insurance be exempt to the extent of the actual estate tax liability but not in excess of $250,000. Such insurance is at present taxable under both the Federal and all but one (Arizona) of the State laws. Whatever the decedent leaves at death that has value represents taxpaying ability regardless of its form. To allow an exclusion for such insurance would discriminate in favor of the group of taxpayers who had seen fit to provide liquid assets for the payment of the tax in this form rather than in any number of other ways. Such an exclusion would, furthermore, defeat progression and thus relegate death taxes to a subsidiary position among the taxes in the fiscal system of the country; whereas the continuation of the recent efforts to increase the revenue from this tax source would seem to possess greater claim for support. As long as circumstances compel the extraction of more revenue from the several tax sources available to Government, it becomes difficult to justify raising the regressive excises, or even the progressive income taxes, while at the same time taking steps towards the effective reduction of the death taxes. In other words, the approach to the upper extreme in taxation, if it must be made, can be made most fittingly with the death taxes leading the procession.

It is argued in favor of excluding insurance for the purpose of paying estate taxes that where such insurance is not provided, it may be necessary for the decedent's beneficiaries to suffer heavy losses through forced liquidation. That such insurance should be provided is not questioned. What is questioned is making the proceeds from such insurance tax exempt. The British Committee on National Debt and Taxation, after noting the argument presented in favor of exempting life insurance which had been provided by the insured to meet death duties, stated: "Altogether, it is our opinion that differential relief in favor of this particular form of saving would not make for equity." It is concluded that insurance designed to pay the estate tax should continue to be included in the taxable estate.

7. Property previously taxed

The Federal estate tax exempts property previously taxed at the death of a prior decedent where the prior decedent died within five years of the second death and was subjected to the estate tax. This provision was first added to the estate tax in 1918, in part as a concession to soldiers, and has been retained to date.

The exemption of property previously taxed is usually based on the argument that "the decedent in paying the tax on the first succession paid for property from which he was unable to obtain his full quota of enjoyment and that such rapidly repeated taxation must necessarily wipe out family properties that otherwise would be handed down from generation to generation. /*/ Furthermore, it is sometimes maintained that in large families, because of the high frequency of deaths, property transfers are more numerous leading to the excessive depletion of family fortunes through death taxes.

Strong argument can be made against this form of exemption. Shultz /5/ puts the case as follows:

"From a logical point of view, only the doctrine that an inheritance tax is a deferred property or income tax justifies such rebates; in fact, from the point of view of 'ability,' or 'equality of sacrifice,' a rapid series of successions would justify a higher tax on the later series of heirs, since the coming of the later heirs into possession of the property would be all the more unexpected. Moreover, a short possession of the property by the decedent is likely to be followed by a long possession by the next heir, since the latter comes into possession so much earlier, with the result that the amount of taxes collected evens out in the long run."

Tax relief for quick deaths in computing death tax liability is not confined to the United States. The British death tax has recognized the problem continually since 1914. In that case, however, the exemption provided varies inversely with the interval between the two deaths and, furthermore, is confined only to property consisting of land, business carried on by a company or any interest in such land or business. The duty payable on the second death under the British tax is reduced as follows:

Where second death occurs within 1 year of the first death - by 50%
Where second death occurs within 2 years of the first death - by 40%
Where second death occurs within 3 years of the first death - by 30%
Where second death occurs within 4 years of the first death - by 20%
Where second death occurs within 5 years of the first death - by 10%

The estate tax of the Commonwealth of Australia takes no recognition of the problem. In the case of the Australian States, it is recognized only by Tasmania where it is provided that for estates not exceeding 4,000 no duty is payable with respect to any real estate which within five years before death passed to the deceased from his spouse, parents or child, and was subjected at that time to a death duty.

The Germans afford preferential treatment to quick deaths in the case of Class I and II beneficiaries. Their death tax law provides that property so situated which had been inherited within the preceding five-year period, and was then taxed in conformity with the present law, is at the time of the second death entitled to a 50 percent reduction; if succession takes place within five and ten years after the prior decedent's death, the tax is reduced 25 percent. It should be noted also that the recognition of quick deaths both in the English and the Tasmanian Law accord favorable treatment to real estate. This discrimination is based upon the theory that the capitalized (and taxable) value of real estate bears a particularly high ratio to current income and that favored treatment is therefore warranted. At the time of the Colwyn investigation, it was in fact urged that the exemption for quick deaths be extended in the case of agricultural real estate to deaths occurring within fifteen years.

The treatment of quick deaths in the Federal tax has had some influence upon State Statutes. Sixteen States, including New York, now make some recognition of such occurrence. In these, exemptions range from quick death occurring within one-year interval to those occurring within a six-year interval, and from those restricted to Class I beneficiaries to those wholly unrestricted.

It would appear that the case for the recognition of quick deaths is not very strong. At the time of the Colwyn investigation, in response to the Chairman's inquiry, "do you feel that it is a hardship in regard to estates where there happen to be rather frequent deaths in the course of a few years?" Mr. Dalton replied, "No doubt there are hard cases from time to time but it is very much harder if your parent die and leave you nothing than if they die and leave you an estate upon which you have to pay a tax. It is not a grievance which I should put in the first rank of those to be relieved." The opinion expressed by Dalton was subsequently accepted both by the Colwyn Committee and the Royal Commission of Australia. The former said in part:

"To a certain extent variations equalise out in the long run. This point, however, must not be pressed too far. There are so many variable factors in the history of estates, and the rates of duty are (to judge from the past) so liable to variation, that the future cannot be trusted to make amends for any present inequality. Nevertheless, it is fair to observe that quick succession may often be due to the first of the two persons deceased having enjoyed the estate for an exceptionally long period. He may have built up a business over a long term, and have died at the age, say, of 80, leaving his property to his son, then aged 50, who may have died within the next five or ten years, being succeeded in turn by his son, a young man of 25 or 30 with a life expectation of 40 or 35 years. Against the repetition of the burden within five or ten years must be set the long freedom of the estate from duty during the life of the first deceased, and the prospect of a further good period of immunity."

The recognition of quick deaths for tax purposes enjoys popular support. A sharp depletion of well established family fortunes through estate taxes occasioned by a quick succession of deaths is generally viewed with disfavor. It may be advisable that some recognition of the problem be retained.

It may be noted that quantitatively the situation is not very significant. During the two fiscal years, 1933 and 1934, when the Bureau of Internal Revenue handled a total of 15,872 returns, only 479 claimed deductions on account of property previously taxed for amounts representing less than two percent of reported gross estate. The British Inland Revenue Department estimated in 1927 that out of 6,000 or so estates exceeding L10,000 which pass each year, about 5 percent pass again within the next five years.

While granting the advisability of continuing to recognize the special status of quick deaths, it should be emphasized that the present arrangement is arbitrary for it provides a 100 percent exemption in the case of a second death within five years and provides. No exemption where the second death occurred one day after the five-year interval. It would be far more equitable if the degree of exemption were graduated by providing for a reduction of 100 percent where the second death occurred in the course of the first year; of 80 percent where the second death occurred in the course of the second year; of 60 percent where the second death occurred in the course of the third year; of 40 percent where the second death occurred in the course of the fourth year; and of 20 percent where the second death occurred in the course of the fifth year. Accordingly, it is recommended that the present uniform exemption be replaced by one graduated in accordance with the length of time intervening between the two deaths.

g. Charitable bequests

The exemption of charitable bequests under the present law, which has been in effect since December 31, 1917, serves to make the United States Treasury a major benefactor for charities. A decedent with an amount available for transfer equal in value to $50,000,000 can give away $20,000,000 at an effective cost to his heirs of but $6,200,000. The remaining $13,800,000 is in fact contributed by the Treasury. Thus, the present law respecting charitable bequests places the control of funds otherwise destined for public revenue in the hands of private persons. The Federal Government is allowing others to direct its expenditures. The desirability of such a procedure is, of course, subject to controversy. The important fact is that the upper brackets of the present rate structure are ineffective in producing revenue partly because of the complete exemption granted charitable bequests. For statistic on the relative significance of charitable bequests in the various "size of estate" groups see Table 8.

The inequity of the present exemption of charitable bequests as between taxpayers is shown in the following table:

Amount of :  Amount of : Rate applicable: Cost of bequest : Cost of
estate    :    bequest :     to bequest :     to taxpayer : bequest

100,000           $10,000       12%         $8,800           $1,200
500,000            60,000       23%         46,200           13,800
1,000,000         160,000       29%        113,600           46,400
2,000,000         460,000       35%        299,000          161,000
5,000,000         460,000       53%       216,2000          243,800
10,000,000        960,000       65%        336,000          624,000
20,000,000      9,960,000       67%      3,286,800        6,673,200
50,000,000     29,960,000       69%      9,287,600       20,672,400

The foreign practice in respect to charitable bequests varies. Great Britain, Saskatchewan and Newfoundland tax charitable bequests. France, Germany, Quebec, Ontario and Australia exempt such bequests from taxation. Alberta, Yukon, and Manitoba allow an exemption of $2,000 for any one charitable bequest. Nova Scotia allows an overall exemption of $25,000 for charitable bequests to be used within the Province.

In the case of the American States, thirty exempt bequests destined for use within the State, six grant unqualified exemption, the others grant either conditional exemptions or no exemptions at all.

It would be unwise in the case of the Federal law to eliminate completely the incentive for charitable bequests. It may be argued that such bequests serve a useful and necessary purpose in a well-ordered society. The present provision, however, has two faults. It defeats the primary purpose (revenue) of the progressive rate schedule and is inequitable between taxpayers by granting an advantage which increases with the size of the estate.

It would therefore appear desirable that the exemption granted charitable bequests be modified, and, in some acceptable manner, limited. Such limitations, however, should not be stated in terms of fixed number of dollars but rather as a fixed or varying percentage of the otherwise taxable value of the estate. The allowance of a charitable bequest exemption could be limited to a fixed percentage of the net estate, in a manner similar to that applicable to gifts by individuals, under the income tax law.

Conceivably, the proportion of the total estate thus made exempt may be varied inversely to the size of the estate. For purposes of illustration, estates whose value, including charitable bequests, exceeds: $10,000,000 may claim 20 percent of such bequests as an exemption; those between $10,000,000 and $5,000,000 may claim 25 percent of such bequests as an exemption; those between $5,000,000 and $1,000,000 may claim 50 percent of such bequests as an exemption; those between $1,000,000 and $500,000 may claim 75 percent of such bequests as an exemption; and those below $500,000 may claim 90 percent of such bequests as an exemption. A schedule designed in this manner, with appropriate provision at the breakover points, would have much to recommend it.

The first of the above proposals has the advantage of simplicity but remains subject to the criticism that the benefits accruing from the allowance provided in that manner would still be greater in the case of wealthy than poor donors. The second proposal may be so designed as to surmount this difficulty but remains subject to the criticism that the schedule, however designed, is arbitrary. It follows that a procedure must be sought which has both the virtue of simplicity and some semblance of equity.

To fulfill these requirements, it is proposed that charitable bequests be included in the taxable base for computation of tax liability but that the taxpayer be permitted a credit for charitable bequests in accordance with the proportion that 50 percent of the charitable bequests bears to the entire estate. Such a procedure would retain the benefits of the progressive rate schedule and at the same time afford an easily administered and uniform charitable deduction.

II. Rate Structure

1. Totality vs. bracket rates

The Federal estate tax achieves progression by the use of a bracket rate structure as opposed to the totality rate method employed by Great Britain, Australia and the Canadian provinces. Progression by totality has the advantage of producing greater revenue than a similar rate schedule on the bracket system, since the high rates of the upper schedule in a totality scheme apply to the entire estate instead of only to its limited fractions. The difference has great significance from the point of view of the public which looks only at the statutory rate schedule, ignoring the consideration whether it applies to the entire estates or only to its fractional parts, and is unable to realize that one form of tax is heavier than the other.

The totality rate method has one disadvantage which, however, can readily be corrected. Since the increased rate incidental to an estate falling in the next higher bracket applies not only to the amount in excess of that at which the rate changes but retrospectively to the first taxable dollar, it is possible that in the absence of provision for the amelioration of break-over points, the increased tax may be greater than the increase in the amount of the estate; and in consequence the beneficiary will actually receive less from a larger than from a smaller bequest. For example:

$250,000 taxable at 9 percent                    $22,500
$250,100 taxable at 10 percent                    25,010

Increased tax due to the addition                 $2,510
of $100 to the estate

This difficulty is eliminated in the British tax by a provision that the tax cannot exceed an amount which is the sum of the tax at the next lower rate plus the excess in the value of the taxable estate over the maximum amount to which that lower rate applies. In other words, the tax in the $250,100 estate cited above would not be $25,010 but only $22,600 ($22,500 + $100).

It will be readily apparent that the British provision for the breakover points leaves much to be desired for at certain levels it works complete confiscation of sizeable added segments to the taxable base.

A more acceptable procedure is the "step-up" system employed in some of the Canadian laws. This provides for a gradual spanning of the rate differential between any two brackets of the rate schedule. Occasional inequities of secondary magnitude appear even in the "step-up" system but these are not significant. Under the proposed schedule, for example, an increase in the taxable estate from $59,999 to $60,000 would increase tax liability by about $12. Similar one dollar additions for the taxable base at $999,999 and $74,999,999 would increase tax liability by about $167, and $75, respectively. Because these occasional inequities are of minor quantitative importance in relation to the taxable estate and because the nature of the item taxed is such as to preclude the possibility of tax minimization on the part of the taxpayer, it is concluded that the "step-up" system should be employed, the above-cited difficulty notwithstanding. At all events the possibility of eliminating these minor inequities is always available in the adoption of the British amelioration device.

Since the totality rate method would tend to lessen the hostility of the general public to the necessarily heavy tax burden, providing at the same time greater productivity and increased equity, its adoption would be a worthwhile innovation in the Federal estate tax and is therefore recommended.

2. Upward revision of the rate structure.

The bracketed rate structure of the Revenue Act of 1935 ranges from 2 percent on the first $10,000 of the net estate to 70 percent on that portion of an estate in excess of $50,000,000. Only relatively few estates are subject to the tax since the $40,000 exemption and the $40,000 insurance exclusion serve to make the great majority of property transfers nontaxable. This is apparent from Table 9 -- Number of Taxable Estate Tax Returns Filed by Net Estate Classes for years 1927 to 1935. In the interest of productivity it would therefore be desirable to make the estate tax rate structure more effective.

The charge that the present progressive rate schedule makes for the destruction of accumulated capital is unwarranted. In the first place, it is possible to extend the payment of the tax over several year. where justifying circumstances can be demonstrated. Secondly, a wealthy man may purchase life insurance during his lifetime to pay the estate tax at his death. In that event the tax becomes little more than an auxiliary income tax. Furthermore, the basis of the capital destruction argument -- that all revenues received by the Government represent total economic less to the country -- is unsound. To the extent that the Government continues to devote an increasing portion of its revenues to the construction of capital equipment and to the retirement of domestically-held national debt, there is no economic loss to the country from the payment of taxes to the Government.

The fact remains that the American Federal, State, and local tax picture is a conglomerate with an over-all tone of regressivity. Sales taxes, business taxes, and the so-called nuisance taxes are clearly regressive. The general property tax, the very backbone of the local tax structure, is a regressive tax. To redistribute the present tax burden more equitably, it is desirable that taxes which lend themselves to progression be more widely utilized. Death taxes clearly belong to this category. Unfortunately, in determining the degree of progression of a rate schedule, there is no true and unchanging theory or doctrine to guide us. There is no standard yardstick which may be applied.

The accompanying chart shows that the Federal estate duty in more lenient on the lower estate brackets and more harsh upon the upper estate brackets than any of the other (foreign) death duties plotted. The insert of that chart shows in greater detail the comparative treatment accorded the smaller estates by the death tax laws of some of the foreign countries and the United States. Experience has shown that the high rates on estates of $10,000,000 and over are practically inoperative due to the small number of returns subject thereto. Only two returns out of a total of 11,110 returns filed in 1935 reported a net estate of over $10,000,000. To increase governmental revenue there must be a stiffening of the lenient attitude towards estates of less than $1,000,000. The figures reported in Statistics of Income for 1934 show that for every decedent leaving a taxable estate of over $1,000,000, there were forty-two leaving taxable estates of less than $1,000,000. To increase revenue it is necessary to tax these forty-two more heavily. Increasing the rate applicable to the one million dollar estate and over can have very little revenue significance. Taxing the forty-two more heavily involves no injustice, but merely the principle of ability to pay. Any decedent who leaves an estate subject to the present Federal estate tax has left more to his heirs than 110 of his fellow decedents, who died without a taxable estate. The present Federal tax, therefore, hits only the wealthiest one percent of our population.

It is proposed that a totality rate schedule as outlined in the following table be adopted. The schedule recommended imposes a greater effective rate throughout the entire scale. It will, therefore, materially increase the revenue derived from this tax at the same time that the nominal rates are substantially reduced. The proposed schedule has been so prepared as to give the curve of progression a maximum smoothness. The division points and the rate fractions employed allow for quick computation of the tax. The rate is retarded in the first bracket date to the operation therein of the vanishing exemption.