A. General Relief for Irregular Incomes

Under a progressive income tax assessed on an annual basis, large payments of income in one year may push the taxpayer up into higher brackets. The result may be a tax burden that is unfairly high as compared with that borne by a taxpayer with the same total income who receives it more evenly over the years. Similarly, individuals suffering from large losses in a single year fail to obtain an adequate reduction in their tax burden as compared to taxpayers suffering losses of a similar aggregate amount but distributed over several years.

For example, authors, composers, and similar creative workers complain that the period of maximum earning power is comparatively short, and that under the progressive tax rates applied to high, though temporary, income they are unable to make adequate provision for later periods when earning power will be low. The earning power of other taxpayers of comparable economic capacity is more evenly distributed and does not become subject to the higher brackets. Income from fishing is also likely to fluctuate widely from year to year owing to variations in the run of fish, and to large concentrated losses due to storms and the like. Forestry income may also be realized very largely in a few years, although scientific cropping of timber lands would tend to minimize this lumpiness. Gains or losses from the sale of property are often realized in large lump sums in single years. And in an economy going through a period of rapid change and severe readjustment, extreme fluctuations in business incomes are likely to occur for a variety of reasons.

Undue differences in tax burden are especially apt to arise from such causes where, as in Japan, the income tax is steeply graduated over an income range that includes a large section of the population.

Some relief in such cases is indicated, but care must be taken to avoid unwarranted discrimination in favor of special kinds of income, that may become a loophole through which taxpayers may evade their just share of the tax burden. It is also important to keep the relief measure simple. However, since the relief will be applied in a relatively small number of cases, more complexity is tolerable than with a provision that must be applied to almost all income taxpayers.

Accordingly, the following treatment is recommended for those kinds of income that are unlikely to continue at a comparable rate. The taxpayer will be permitted to separate out and spread over several years' tax returns the amount received in one year from these kinds of income. In the meantime he will be required to pay a suitable tax deposit on the amount of income allotted to future years, in order that the bulk of the tax shall be collected at the time the taxpayer has the money. The appropriate number of years might be allowed to vary according to circumstances: for fishery income a five-years period might be sufficient; for capital gains and losses five or ten years might be appropriate; and for authors' royalties and forestry income the period might be as long as twenty years. The taxpayer might be given the option of using a shorter period than the maximum specified in the regulations.

For example, if a taxpayer has capital gains of 500,000 yen in one year, the taxpayer would include 1/5 of this, or 1000,000 yen, in his taxable income for the current year. Suppose that his other "normal" income for the year is 200,000 yen. A tax for the current year would be determined on the 300,000 yen total at the regular graduated rates, amounting to say 120,000 yen. On the remaining 4/5ths of the special income, or 400,000 yen, a tentative tax deposit would be payable immediately at a rate equal to the average rate of tax on the income for the current year. In this case this average rate would be 120,000/300,000 or 40%, and the tax deposit would then be 40% of 400,000 yen, or 160,000 yen.

In each of the following four years, one-fifth of the special income, or 100,000 yen, would be added to the other income for that year in applying the graduated rates. From the tax so computed, one fourth of the tentative tax deposit would be deducted as a tax credit. Thus if in a subsequent year regular income rose to 300,000 yen, a tax would be computed at the graduated rates on an income of 300,000 yen, plus 100,000 yen, or 400,000 yen; this tax might amount to say 170,000 yen. Against this tax a credit of one-fourth of the 160,000 yen tax deposit or 40,000 yen would be allowed, making the net tax payable 130,000 yen instead of the 120,000 yen that would have been payable had there been no carrying forward of the capital gain. The effect of carrying forward the capital gain is to push the 300,000 yen ordinary income a little higher up on the scale of graduated rates. If income falls to say 50,000 yen in a subsequent year, the tax credit may operate to reduce the net tax payable below the amount ordinarily payable on an income of this amount: The tax on 150,000 yen may be, say, 45,000 yen so the tax credit of 40,000 yen would bring the net tax down to 5,000 yen which may well be less than the tax on an ordinary 50,000 yen income. If income shall fall to zero in a subsequent year, a tax refund would normally be called for.

The administrative burden imposed by this treatment should not be as great as might appear at first. It will be applied only in a relatively few cases, on application by the taxpayer. If the taxpayer fails to request this special treatment the regular assessment can proceed, and in most cases no serious harm will be done. Certainly no taxpayer will be made worse off by the existence of such a provision. And it will afford a sound basis for relief in extreme cases, where otherwise the tax administration might be tempted to impair the uniformity of application of the law by condoning a partial evasion of the tax. The slight additional complication of the computation could well be provided for on a special form, in order not to complicate the regular form with instructions which would be used but seldom. This special form, or a copy of it, would then probably be kept in a special file which would serve as a means of checking to ensure that such taxpayers made the proper adjustments in the subsequent years. Thus it would seem relatively easy to prevent the abuse of this provision. And even if taxpayers fail to make the subsequent adjustment the amount of evasion involved will be minute.

But whether or not such a provision comes to be extensively used, its existence is much more important to a successful income tax than may appear at first. Without such a provision to take care of irregular incomes it may be difficult to secure the full inclusion of capital gains in taxable income, and without such inclusion of gains, the entire income tax structure is seriously weakened. Thus it is particularly important to have this provision in the law immediately, even though, if it were looked at as merely a device to make a minor adjustment in tax in a few select cases, introduction of such a refinement at this time might seem premature.

Indeed, although a superficially simpler method of treating such cases of irregular income might be devised, there is good reason for avoiding such short-sighted short-cuts. As experience has shown, too often a provision adopted in disregard of sound principles because of its superficial simplicity has proven to involve a host of hairline distinctions and irrational discriminations that turn the hoped for simplicity into a hopeless tangle.

Each element of the method here proposed has an important function and cannot be eliminated without danger of producing unwanted results. A reasonable measure is provided for a part of the tax to be paid immediately. This is necessary because otherwise the taxpayer might dissipate the large income and become unable to pay any substantial tax later on. On the other hand, it is necessary in later years to make adjustments, which in most cases will be relatively minor, in order to avoid cases in which an increase in the "normal" income in the year in which the extraordinary income is received would result in an unduly large increase in tax. In extreme cases, if no later adjustment were provided, the effective marginal rate could be over 100%, so that by working less and actually having less income a taxpayer could retain a greater amount net after taxes. The provision for the adjustments in later years also inhibits excessive claims for relief, since if too much of a year's income is shifted to future years the total tax will be increased. These adjustments thus aid administration as well as help to preserve incentives for productive effort. The added complication is not unduly great, in view of these advantages.

It is necessary to spread the income over future years rather than past years, partly because of the absence of adequate records, and partly because it will be easier to adjust future tax returns than to reopen past returns. Where the taxpayer does not survive to the end of the adjustment period, the remainder of the adjustment can be waived without serious effect.

B. Losses

A somewhat similar problem arises when an individual is subjected to some heavy loss in a given year. If he deducts the loss entirely in the year in which it occurs (or in which it is recognized for tax purposes) he may not have enough income to absorb the loss. Even if he does, the loss, when deducted in computing net income, will offset other income falling in lower brackets and will thus result in a smaller reduction in tax than it would if it were spread over several years. Accordingly, some adjustment to the tax is appropriate in this case too.

However, the appropriate adjustment for this case is not necessarily the exact counterpart of that described above for a gain. To permit an average rate of tax, computed on the current net income, to be applied to the amount of the loss to be carried forward is likely to involve a tax refund. This is not wrong in principle, but it does involve an administrative complication that had best be avoided. Also, five years seems an adequate period for the spreading of nearly all kinds of losses. Accordingly, the following scheme is suggested in the case of large losses: The current tax payment shall be computed by deducting the entire loss from the net income. A tax liability shall also be computed allowing only one fifth of the loss to be deducted, and the excess of this tax liability over the tax actually paid shall be considered a deferred liability to be liquidated by the taxpayer in the four subsequent years. In each of the four subsequent years, one fifth of the loss shall be carried over and deducted from the net income; the tax shall be computed on the balance, and to the tax so computed shall be added one fourth of the deferred liability to obtain the amount to be paid in the subsequent year. This treatment would be permitted only where the loss is of an unusual character and is larger than, say, one fourth of the entire income. Examples of such losses are those from storms, fires, natural disaster, or other non-recurrent phenomena.

Where there is an actual net loss for the year, this provision operates as a modified five-year carry-forward of loss. However, regardless of whether or not the loss is an extraordinary one, there should be provisions for carryover of net losses sufficient to insure that virtually all losses are allowed as offsets against income. Thus whenever there is a net loss for a year, (after spreading any extraordinary losses as above) the amount of this net loss should be treated as a deduction in computing the net income or loss of the following year, so that in effect net losses are carried forward indefinitely until offset against income. The carry-over of net business losses is further discussed in Chapter 7. However, it is not desirable to make any provision for the carrying forward of unused personal allowances; while net losses are sufficiently infrequent so that the tax office should be able to make an adequate check on the amount of the loss to be carried forward against future income, unused personal exemptions occur with such frequency that an adequate check of the income reported in such cases would be an undue burden on the tax administration.

Care should be taken in any case in defining the circumstances where this relief is to be applied to avoid making the definition so loose that the administration will become burdened with an undue number of applications. Nor should it be so vague that an undue amount of discretion will be left with the local tax office to accept or deny application for the spreading of income. For example, income from royalties, artistic efforts, and similar short-lived careers should be allowed to be spread only to the extent that it exceeds say 150,000 or 200,000 yen for the year. Spreading should be applicable to capital gains, unusual losses, and other lumpy items only if such items exceed say 30% of the net income. Or the spreading treatment may also be limited to cases where the income for the current year, if the lumpy item were include in full, would be at least 30% greater or smaller than the income reported for the preceding year, adjusted by the estimated overall increase in incomes. The determination of these conditions in detail may well be left to administrative ordinance. The net result of our recommendations should be a substantial increase in the definiteness and certainty of the tax law as compared with the present blanket authority given the tax officers to abate the tax when in their judgment the taxpayer has "badly lost his ability to pay". Grants of discretion as broad as this should be avoided wherever possible, since they generally produce an undesirably large amount of dispute, and provide the occasion for possible favoritism and even corruption.

For most cases of temporary, unusual, or "lumpy" incomes, windfalls, or losses, the above treatment will represent a substantial relief as compared with the rigid adherence to the annual basis of assessment, and should prove a readily acceptable solution to the problem. Indeed, such relief is about all that such taxpayers are equitably entitled to on grounds of irregularity or "lumpiness". To be sure, some additional complaints have been heard that inadequate allowance has been made for legitimate expenses in connection with the earning of special types of income, such as author's royalties and the like, but proper remedy in this case is a more reasonable administration rather than a change in the basic law.

C. Special Types of Irregular Income - Forestry

However, certain other types of income have been accorded special treatment under the income tax far beyond what equity requires. Capital gains, income from forestry, and certain other type of income are included in taxable income under the present law only to the extent of 50%. In some cases, as with forestry income, such favorable treatment may be partially justifiable as a means of indirectly subsidizing an industry in which there is a substantial public interest. However, it does not seem likely that a subsidy which increases as the income of the owner reaches higher brackets is the one that will accomplish most in return for a given sacrifice of public funds. Actually it seems likely that some form of direct and overt subsidy would be more effective and less subject to abuse than a tax exemption. And even if tax exemption has psychological advantages as a stimulus that cannot readily be obtained with other methods, it is likely that an exemption or special treatment expressed by providing a credit of a certain percentage of the forestry income from the tax would obtain better results per yen of revenue sacrificed.

The fact that such a subsidy has been adopted and that under its promise landowners have made investments in reforestation requires that the privileges thus implicitly promised not be lightly withdrawn or impaired. To some extent the good faith of the Japanese government may be considered by some to have become engaged. The degree to which a legitimate vested interest has been established in the continuation of this tax privilege is probably fairly slight and the promise is only implicit rather than direct, so perhaps this consideration should not be given too much weight if there are substantial considerations on the other side.

D. Capital Gains

In the case of capital gains, however, there is no special social purpose to be served by promoting speculative investment for appreciation in value at the expense of investment for steady earnings and dividends. The present favorable treatment appears to have been accorded to capital gains in part through adherence to the precedent set in the United States, in part in recognition of the principle that capital gains that merely reflect a severe inflation in the general price level do not represent real income, and in part because of the difficulties of administering a tax on capital gains at the rates of up to 85% applicable to the larger incomes where most of the capital gains would be found.

However, in this case the precedent of the United States is a bad one to follow. The special concessions to capital gains have been the least satisfactory parts of the American Income Tax Law, as is attested by the fact that these provisions have been changed at very frequent intervals. There are many ways in which income that might normally be received as dividends, royalties, business profits, and even salaries can be transformed so that in form it appears as a capital gain. If capital gains are subjected to lower rates than other forms of income, many taxpayers will attempt such transformations in order to avoid tax. Not only will revenue be lost but in the process much time and effort will be wasted and business relationship will often be distorted into forms that are not conducive to the most efficient operation of the economy.

Since there is no real economic distinction between capital gains and other income, attempts to draw the line between capital gains and other kinds of income for tax purposes must often make very arbitrary and artificial distinctions. In the unsuccessful attempt to restrict the conversion of other income into capital gains, many complications have been added to the law of the United States and it is not too much to say that the discount on capital gains is responsible for more complications in the law and for more dispute than almost any other single feature of the law. It is strongly recommended that Japan avoid proceeding any further along the path that has led the United State to this unsatisfactory result.

This requires that, as a permanent policy, no greater concessions be given to capital gains as a form of income than those dictated by the fact that such gains may in some cases be concentrated in a single year. The spreading treatment suggested above for lumpy incomes generally seems to meet such requirements adequately in the case of capital gains, and it is recommended that as a permanent program this be the only form of special treatment accorded to capital gains.

Indeed, it cannot be too strongly emphasized that inclusion of the full amount of capital gains, other than those attributed to inflation, is a basic cornerstone of our program of tax reform and that no departure from this principle can be admitted without seriously impairing the coherence of the program. Specifically, the corporation tax proposals would have to be considerably modified if they were to be a part of a system that did not include the full taxation of capital gains in the hands of individuals. Allowances for depreciation, both for the corporation and for individual enterprises, would have to be reexamined. A stricter examination into the expensing of repairs and minor improvements to property would be required. The entire rate structure would have to be reexamined, in order to avoid what one writer has termed "the ridiculous business of dipping deeply into high incomes with a sieve". Finally, if gains are not fully taxed, many restrictions will have to be imposed on the deduction of losses. Such restrictions would seriously impair the willingness of taxpayers to embark in risky enterprises because of the "heads I win, tails you lose" nature of the arrangements that would necessarily ensue.

In short, this is one point where no deviation or compromise can be tolerated if the objectives of the tax reform program are to be adequately realized.

According to the strict theory of the taxation of accrued income, the amount by which the market value of a taxpayers' assets increases during the year should be assessed and taxed each year. As this is difficult, in practice such gains are to be assessed only when the taxpayer sells the asset and realizes the gain in cash or other more liquid form. As long as this realization takes place within a reasonable period of time, the tax is merely postponed slightly, and no serious infringement of the basic principle results. However, it is necessary to prevent an indefinite postponement of the assessment of the gain, as this would permit the taxpayer to escape a substantial part of his proper tax burden. One of the most important methods of preventing this is to require that where property is disposed of by gift or by inheritance, the amount of increase in value shall be computed and included in the income of the donor or decedent, as the case may be. Computing the tax according to the method described above for irregular incomes will prevent an undue pushing of the income into the upper brackets. Consequently, although in the case of inheritance there will be no future adjustments of the tax, this will not be a serious matter. Of course, this tax, along with other income tax liabilities of the decedent, will be deductible before the inheritance tax is computed.

Taxing accrued capital gains at time of gift or transfer at death will also reduce the tendency of capital gains taxation to cause taxpayers to postpone selling property or shifting their investments. If no such tax were levied at gift or death, taxpayers would tend to postpone sale of property that had increased in value, in the hope that the tax could be indefinitely postponed and perhaps avoided altogether. If this provision makes it apparent that the tax will have to be paid sooner or later in any case, this reluctance will be much less and the markets for such assets will feel less of a disturbance on account of the tax.

E. Capital Losses

In principle, of course, if capital gains are to be included in full in the taxable income, capital losses should be allowed as a deduction. The present law appears to be silent on the matter of capital losses, and up to the present time, owing to the general inflation of prices and the methods of administration the problem has been almost entirely hypothetical. To be sure, paragraph 3 of article 9 seems open to the interpretation that losses are deductible only from gains and from forestry income, not from ordinary income, but there is no specific prohibition against deduction from ordinary income, and the law could also be interpreted as permitting only half of the losses to be so deducted.

Aside from problems connected with inflation, there is no serious objection to allowing the deduction of capital losses to the same extent as gains are taxed, provided that there is some assurance that the corresponding capital gains will be assessed and included in income within a reasonable period of time. Under the present United States law, gains accrued but not realized at the time of the taxpayer's death escape taxation entirely, and partly for this reason it is not possible to allow unrestricted deduction of losses without opening a substantial loophole. Taxpayers would be able to sell assets that showed losses, and retain those that showed gains and pass them on to their heirs. This would enable them to deduct a loss from their other income, even though their trading on the whole showed a profit.

Partly for this reason, the United States law imposes rather severe and arbitrary restrictions on the deduction of capital losses. These restrictions not only work occasional hardship, and produce unjustifiable discriminations among taxpayers in essentially similar circumstances, but they also discourage investment, particularly in risky projects. For example, an investment of 100 that has an equal chance of returning 150 (a net profit of 50), or a return of only 80 (a net loss of 20), may be considered a good risk. If there is an income tax of 40% with losses fully deductible, it will still be a good risk, since the net return after tax will be 130 (150 less the tax at 40% on the profit of 50) in the one case and 38 in the other (80 plus the 8 tax saving resulting from the deduction of the loss of 20 from other income taxed at 40%). The possible net profit will be smaller, but the possible net loss will also be smaller in the same proportion. But if the government plays "heads I win, tails you lose" with the taxpayer and takes its cut of the profits but allows nothing for the loss, the investment will return only 80 in case of loss, and only 130 in case of profit, so that the proposition no longer looks attractive, particularly if compared with an investment that would return 110 with complete security. Failure to allow for full deduction of losses can thus have quite serious effects on investment and enterprise.

However, under present Japanese law, there is no assurance that gains will be taxed within any reasonable length of time. For gains are not taxed until the property is sold, either by the taxpayer or his heirs. And if the property is held for several generations, taxation of the gain is postponed almost indefinitely. However, since we recommend that the accrued gains on property transferred by gift or bequest be included in income, we place a reasonable though rather loose limit on the postponement of the final reckoning. Accordingly, we also recommend the complete deduction of all capital losses, without restriction. And only with this complete allowance for capital gains and losses can the income tax approach the equitable, non-discriminatory and non-repressive tax that it should in principle be.

F. Evasion and Enforcement of Tax on Capital Gains

It has been argued, of course, that since it is rather more difficult to assess capital gains completely than other forms of income, the rate should be kept low so as to give less of an incentive for evasion and so as to produce less inequity between those who evade this tax and those who have similar amounts of capital gains but pay the tax. Incidentally it is interesting to note that this is just the reverse of the argument applied in the case of the earned income credit, where it has been asserted that the rate of tax on wage-earners should be abated and the nominal rates of tax on the self-assessed made relatively higher in order to offset the evasion. If this argument were applied to capital gains the result would be a heavier tax on such gains than on other income.

To be sure, if capital gains were a separate and distinct element of income, not interrelated with other types of income, one might, as with the tobacco tax or the liquor tax, urge that there is a point beyond which increasing the tax rates increase evasion to such an extent that the added revenue is too small, relative to the impairment of taxpayer morale and other detrimental effects of the increased evasion, to justify the tax. But capital gains are merely one of many forms that income may take. Decreasing the rates on such gains below those applicable to other types of income may decrease illegal evasion, but only at the expense of increasing legal avoidance of income taxes, as sophisticated taxpayers convert ordinary income into the form of capital gains. The knowledge that some individuals are avoiding their just share of the tax burden with impunity and in accordance with the law may be much more detrimental to taxpayers' morale than the knowledge that some individuals are legally evading taxes.

Moreover the loss of revenue through reducing the rate of income tax on capital gains is much greater than the amount by which receipts from this part of the tax diminish. Indeed, even though reducing the rate actually were found to increase the yield of the tax resulting from capital gains, this would still represent a substantial overall loss in revenue, since much of the increased revenue would be coming from income which otherwise would have been reported as dividends or profits so that the full income tax rates would have been paid.

And there is no assurance that reduction in the rates of the capital gins tax would actually reduce the amount of illegal evasion. Where capital gains rate are reduced, an individual who has been induced by the low capital gains rates to put his income in the form of capital gains (at some trouble and expense) may be tempted to go the rest of the way and fail to declare his income. But if there is no advantage to taking the initial step, a taxpayer may well hesitate to go to the trouble and expense of shifting his income to the capital gains form when this will be worth while only if he is able to evade the tax completely.

Thus there is no excuse for special reduced rates on capital gains merely because enforcement for the full income tax rates is particularly difficult at this point. A chain is no stronger than its weakest link, and a progressive income tax is but little more progressive than its widest loophole. If at a given level of rates the amount by which capital gains illegally escape assessment is intolerable, the proper remedy is to lower the entire rate structure to a point where an acceptable degree of enforcement can be obtained all along the line. If such an acceptable level of enforcement cannot be reached, then the question is raised whether the entire idea of an integrated progressive income tax should not be scrapped in favor of less equitable but more enforceable methods of obtaining revenues.

This does mean, however, that if an income tax of this kind is to be retained, substantial improvement in the degree to which capital gains are assessed will have to be achieved. To a large extent, the passing of the inflationary period, and the introduction of special provisions outlined elsewhere, will eliminate nominal inflation profits from the tax base. The net capital will thus be brought more nearly in line with a figure that will be acceptable to taxpayer and assessor alike as being a reasonable measure of real income or profit. But in addition increased effort on the part of the assessors will be required. In particular, limitations should be placed on the anonymous ownership of property of all kinds.
Endorsements of stock shares, for example, should be limited to a period of one month or so, after which time the new owner would be required to be registered on the books of the company. The adoption of the net worth tax, recommended elsewhere in this report, should prove of considerable assistance in securing the full taxation of capital gains. If adequate use is made of the resources available, it should be possible to raise the assessment of capital gains to an acceptable level of completeness.

G. Adjustment of Basis

When a piece of property is disposed of, the capital gain is computed by deducting from the amount received (or the market value, in case of a gift or bequest) an amount called the "basis" of the property. In general, this "basis" is what the property cost the taxpayer. But where a property has been used in business, or has produced an income, or has been otherwise the occasion for receipts, expenses, or adjustments to income accounts, certain adjustments must be made to the cost to arrive at this basis. In particular, all receipts from the specified property that are not reported as income must be deducted from the cost of the property in computing the basis to be sued in arriving at gain or loss. Likewise, all deductions connected with the purchase or holding of the property that have been claimed in computing net income but which do not correspond to actual outlays, must be deducted from cost in computing the basis.

On the other hand, actual outlays not deducted in computing income, and amounts reported as accruals of income not corresponding to actual cash receipts, should be added to the basis. Thus a payment for damage, or the proceeds of a partial sale, or depreciation charged off, must be deducted from the cost in computing the basis. The cost of extraordinary repairs not counted as a current expense, or an accrued rental included in income though not actually received, may be added. Only if the principles laid down above are adhered to can we be sure that there is no omission of income or double counting. If the above principles are followed, then no income can escape eventually being subjected to income tax, except through actual failure to report receipts or the reporting of fictitious outlays, or the deduction as necessary expense items which properly are personal expenditure. These three sources of error are relatively easy to discover and check, and it will be easy even for comparatively inexperienced tax agents to understand what constitutes a violation at these points.

To fail to establish such correct principles from the very outset, whether on the plea of inadequate records or of poor enforcement, would indefinitely postpone the time at which Japan can hope to have a modern, scientific, and equitable tax system.

H. Inflation Gains

There remains, however, the problem of what allowance, if any, should be made for the fact that a large portion of the capital gains that have been realized recently may represent merely the inflation of values rather than a net addition to the purchasing power to the taxpayer. If the inflation of value had been moderate (say 10% per year or less), relative to the number of years elapsed, or if the rates of tax were low, it might be proper to make no allowance for this factor. If the tax on capital gains were applied uniformly, the result would then approximate the imposition of a mild capital levy on top of the income tax. As long as such a levy is moderate as to rate and is uniformly applied, there is no inequity, and no undue impairment of incentives and rewards for productive investment. Indeed, any substantial concession to capital gains under these circumstances would be decidedly inequitable unless some similar concession is given to holders of life insurance, savings deposits, and bonds, to allow for the shrinkage in the purchasing power of these assets. And to make such an allowance would hopelessly complicate the administration of the tax. But where inflation has proceeded for several years at rates of 50% per year and up, the capital levy implicit in the taxation of nominal capital gains as income is too severe to be ignored. Worse, since capital gains may be realized and taxed at different periods, the tax will be payable in currency of widely varying purchasing power so that considerable inequities will arise in the application of the tax to various individuals. Finally, the administration of the tax in recent years, and the records available, are such that it seems impossible to provide any completely equitable treatment of past capital gains.

Accordingly it seems proper to eliminate, as far as this may be possible, from the capital gains to be included in the income tax base, any gain representing merely the inflation of the general price level up to some date in 1949. This would mean including only that part of a gain that represents an actual increase in the real purchasing power of the taxpayer. And in some cases, it may be proper to go even further and ignore any real gain or loss incurred prior to 1949, on the ground that adjustments to the cost basis of assets during this period were too often not reflected in taxable income, so that to recognize such gains and losses now would hardly improve the equity of the income tax in an amount commensurate with the administrative difficulties involved.

For this purpose it is appropriate to distinguish several types of assets: (a) depreciable business assets, with respect to which there are nearly always numerous items such as depreciation, maintenance, repairs, and the like that should be reflected in the current income reported by the taxpayer; (b) inventory, which because of its frequent turnover is already on books at a figure reasonably close to present value; (c) business land; (d) buildings other than business structures, which may need to be treated in a way that does not involve the often difficult separation between the business and residential parts of a structure; (e) farm land, which involves special problems connected with the restrictions on sale; (f) forest land, in which a special public interest exists from the point of view or erosion and flood control; (g) other land, which has also been subjected to extensive control and which has not in general risen in price as much as the general cost of living, and; (h) other assets, such as intangibles, personal belongings, furnishings, and treasures.

For most of these assets other than depreciable business assets, adjustments to be made to the costs because of expenses and depreciation allowances are few, and, when they do occur, usually do not figure in the computation of current net income. For depreciable business assets, it will in many cases be relatively difficult to decide on the basis of the available records exactly what adjustments should be made to the original cost to provide a basis for the computation of capital gains. In this case, however, the problem is intimately tied up with the problem of the revaluation of business assets to provide a basis for more realistic current depreciation allowances. Indeed, as a result of the revaluation procedure, there will be a value established as of some determined period in 1949, which will be used as book value and as a basis for depreciation. Such a value will be arrived at either by adjusting book values by means of a general price index, or, where records are inadequate for this, by means of the application of standards to achieve a comparable result. It is quite appropriate to use, in such cases, this revaluation figure as a basis for the computation of capital gains or loss in case of eventual sale or liquidation of assets. This would mean, in effect, that real capital gains or losses accrued prior to 1949 would be largely included in taxable income where adequate books were kept, and largely ignored, or included on a rough average, where standards are used. But in the latter case it would in any event be extremely difficult to assess such gains and losses with any great equity, and here it will be necessary to let bygones be bygones.

For non-depreciating assets, the problem is somewhat different. Here there is no immediate occasion for determining a value for income tax purposes since there is no question of depreciation allowance, and if fact the problem of determining such a value is not likely to become pressing until assets are actually sold. However, if an attempt were to be made when an asset is sold several years hence to determine retrospectively a 1949 value for such an asset for purposes of assessing the gain, the result might be not a little capricious. Even under stable economic conditions, it is difficult to determine value long after the event; and since Japan is just now emerging from a period of inflation and economic confusion, establishing retrospectively a reasonable 1949 market value is likely to cause much difficulty, controversy, and dispute.

Another factor to be considered is that if business accounting is to be placed on a rational basis, it will be desirable to have a reasonably current book value for all assets, or at least all operating assets. Accordingly it will be desirable to have a current value determined immediately for most, if not all, business assets. Also, since in many cases land and buildings are used partly for business and partly as a residence, it would be desirable to avoid the necessity of apportioning the value between the two uses, and at the same time to avoid discriminating between purely residential real estate and real estate used in part for business purposes. Also, for such assets a value, though not necessarily the identical value, has to be determined in any case for the application of the land and house tax. Thus it will be desirable to determine a 1949 value immediately not only for depreciable business assets, but also for all non-farm real estate, and for all other operating assets of corporations. This value will then be used for the purpose of computing further capital gain or loss.

For other assets the actual determination of the revalued basis may be deferred, either until the time of sale, or possibly to some period when the other demands on the tax administration are less pressing than they are at this time when many substantial changes are being introduced. In general, for non-depreciating assets there is less likelihood of there having been substantial depreciation charges, improvements, and other adjustments to the basis that would effect current income. The difficulties in the way of relating the capital gain to the actual cost are therefore much smaller than for depreciable assets. Moreover, in the case of a substantial proportion of individually owned assets that have not changed hands since 1946 there is a further bench mark in the value on which the capital levy was assessed. Thus, for these assets it may be possible to make a more complete and definite allowance for real gains and losses accrued prior to 1949.

For assets of individuals other than land and depreciable business assets, it is accordingly suggested that the basis for computing capital gains be set at a figure obtained by multiplying the value assessed for capital levy purposes by the ratio by which the cost of living increased from the valuation date of the capital levy to some appropriate period in 1949 or approximately by 10.76 times. This would have the advantage that any undervaluation for capital levy purposes would in effect be penalized by the lower basis for capital gains purposes, and vice versa. Indeed, if the added complexity seems worth while, a further element of compensation for past inequalities could be introduced by using the increase in the cost of living since the date on which the capital levy was actually paid, rather than the increase since the valuation date, since then an undue postponement of payment of the capital levy (which meant payment in more highly devalued yen) would also act to increase the amount of capital gain. Of course, in many cases only an overall assessment of aggregate capital is available, and in such cases a suitable apportionment of the assessed amount will have to be made along the various assets. The fact that any overallotment of one asset will reduce the amount allowable to others and correspondingly increase the gain when they are sold will be some protection to the revenue.

Also many individuals were exempt from the capital levy by reason of the specific exemption of 100,000 yen, and others did not file returns. In such cases it will be up to the taxpayer to produce evidence of value at the time of the levy or perhaps at some subsequent date up to 1949, and possibly no very good check is available here. But in any case such taxpayers will be stopped from claiming more than 1,076,000 yen (which is the present equivalent of 100,000 yen at the capital levy date) in the aggregate as a basis for all of their assets acquired previous to March 3, 1946.

For assets acquired between the date of the assessment for the capital levy and 1949, the cost of acquisition can be multiplied by the ratio of increase in the cost of living from the date of purchase to a 1949 date. In effect, for such assets, a capital gain would be assessed only to the extent that there had been an increase in the real purchasing power of the asset since the time of purchase.

Land, however presents a separate problem. Land in general has not risen in price nearly as much as other assets, in part because of special restrictions on rentals and on its use and transfer. Also land is subject to the land and house tax, and it is proposed to base this tax on a capital value. In order to avoid confusion, and at the same time provide some stimulus to the declaring of an adequate value for the land and house tax, it would be desirable to have the basis for capital gains purposes to be the same as the value set up for the purpose of the land and house tax, which in turn must be reasonably close to a value based either on market price or on discounted future return. Accordingly, for land it is suggested that the initial reappraisal value for the purposes of the land and house tax be taken as the basis for capital gains purposes. In the case of farm land, it is further suggested that the period within which this value must be finally determined be extended by one year.

Timber land presents still a further problem, in that its value is in large part due to the investment of outlays for replanting and other care spread over a very long time in the past. It is proper to differentiate this case from that of current expenses in a business, which are embodied in products sold within a relatively brief time. Accordingly in the case of timber land owned by a corporation, outlays for replanting and other care of the timber land should also be revalued by a general price index and added to the land value to get the basis for purposes of computing capital gain. Where timber land is owned by an individual, value declared for the capital levy, or purchase price since, plus subsequent outlays for replanting and care may all be inflated by a price index in determining values for the purpose of computing gain. This will insure, among other things, that owners of timber lands will have enough left after paying the taxes on the proceeds of the sale of their timber to proceed with the replanting of the land.

I. Inflation Losses

For capital losses, however, it will be necessary to be a little more severe. Almost every individual who held any bonds, life insurance policies, yen debts, or even bank deposits or savings accounts during the period of inflation suffered a substantial loss in the real purchasing power of these assets. To a lesser extent this is also true of land. While in principle it would be proper to allow this loss as a deduction if we are at the same time ignoring nominal capital gains, to do this is totally impractical. If such deductions were allowed, nearly every taxpayer would be able to assert a claim to some substantial loss on this score, and the administrative burden would be completely unmanageable. There would also be inequities produced by the varying degree to which taxpayers would realize the extent to which they were entitled to the deduction. And such an allowance would produce a drop in revenue that can hardly be afforded at the present moment.

Accordingly, for purposes of computing a capital loss on non-depreciable assets, a taxpayer would be allowed to use as a basis only an actual market value for a stipulated base period in 1949, established by the taxpayer to the satisfaction of the tax assessor, except that if this is greater than the value obtained by applying the cost of living index increase to the value assessed for capital levy (or to the acquisition cost if acquired since the levy) then this value calculated with the index must be used. However, it may be provided that the capital levy assessment or the acquisition cost, uninflated, may be used if it is greater than the 1949 value. This means that when an asset is disposed of, the amount of the allowable loss shall be either (a) the loss that has occurred after 1949, or (b) the real loss incurred since the capital levy or the acquisition of the asset, whichever of these losses is smaller. However, a net loss in money terms will be deductible even if it did not occur after 1949.

In the case of depreciable assets for which a reappraisal value was determined, net loss will be allowed on the basis of the reappraisal value, as adjusted for subsequent depreciation and other similar charges.

This treatment, by itself, will, of course, involve some discrimination among individuals according to the extent to which they realized their nominal capital gains during the past few years. Those who did sell their assets and reported the nominal gain, paid an additional tax, while those who deferred the realization paid no tax at that time and now will be entirely excused from tax on those nominal profits of the inflation period. However, during this period the tax on capital gains was but scantily enforced, and moreover the tax, even when fully applied, was at specially low rates. This discrimination is, therefore, not so serious as might at first appear.

As a partial offset to this discrimination, and to the absence of any allowance for real losses on savings deposits and similar assets, it is proposed to levy a tax of 6% of the revaluation profit. This tax is to be paid in three annual installments of 3% ,1 1/2% and 1 1/2% [# 3%, 1.5%, 1.5%] in the case of depreciable business assets other than farm assets. On all other assets this tax would become payable at the time the asset is sold, along with the income tax on the capital gain. In effect, capital gains would be divided into tow parts, the nominal gain taxable at 6% and the real gain taxable as a part of the income of the taxpayer. However, where an asset is sold at a price below the established reappraisal value, and part of all of this loss is disallowed because of the provision of the third and second paragraphs preceding, it would be appropriate to abate the tax on the revaluation profit by 6% of the amount of the disallowed loss, in order to avoid in effect levying a tax on a reappraisal gain which did not actually materialize and against which no offsetting loss was allowed.

This treatment of the problem presented by nominal inflation gains is not altogether satisfactory, as it is not as simple as could be desired and as it still leaves a certain amount of arbitrary discrimination uncorrected. Actually this degree of discrimination appears to be considerably less than that obtainable through any other feasible treatment. And the inequities involved are still minute compared with those brought about by the inflation itself, so that the real villain is not the tax system, but the inflationary aftermath in which it operates. It is to be hoped that the inequities and complexities will eventually become a thing of the past and that the drastic inflation that produced them will not recur.

J. Future Price Fluctuations

But while allowance may properly be made in this rough fashion for the recent drastic inflation, it is recommended that no specific allowance be made for further changes in the price level unless such changes should amount to more than 15% per year at least. It is hoped that proper banking policy and adjustment of the budget will be able to keep further price changes within this figure. The inclusion of a moderate amount of inflation profits or deflation losses in the taxable income will on balance, it is believed, be beneficial, as providing some increased degree of built-in flexibility in the tax system. Thus, if prices should start to rise, the extra revenue from inflation profits would to some extent tend to check the rise, and if prices should start to fall, the reduction in revenue resulting from the deduction of losses would likewise tend to check the fall. Attempting to adjust continuously for changes in the general price level would substantially reduce this countercyclical influence of the income tax. And the administrative difficulties introduced by attempting any such adjustment would far outweigh and slight detrimental effect on the economy of the mild net worth tax implied in ignoring a moderate price rise, or of the mild capital implied in ignoring a mild price decline. Thus as the results of the recent inflation wear off, the associated complexities in the tax structures should gradually disappear.

[# end of Appenidx B ]