Printer Friendly

Taxing the consumption of capital gains.

This piece traces the history of the capital gain concept, and then the policy arguments for a lower rate on capital gains, to argue for an express requirement that the lower rate for capital gains be available only if the gains are reinvested. Consumed capital gain, the Article argues, is not real capital gain within the original understanding or tax policy arguments. Indeed since capital gain is available for consumption, a viable alternative is to repeal the capital gain preference and engineer investment tax incentives to the front end of investment.


 A. American Capital Gain: Merchants' Loan & Trust Co.

 v. Smietanka 481

 1. Dubious Macomber 483

 2. Realized Gains are Still Trust Capital 485

 3. Congress Mostly Reverses Smietanka 487

 B. Why Did the British Exclude Capital Cains from Tax? 488

 1. Parliamentary Adoption of the Income Tax 488

 2. The Strict Settlement Model 490

 3. Distributed Amounts Are Distinguished 494

 4. Review and Retention, then Revocation 496

II. The Reinvestment Presumption in Tax Policy 498

 A. The Reinvestment Presumption 499

 1. Lock-in 499

 2. Amnesty at Death 502

 B. Arguments Not Depending on Reinvestment 505

 1. Bunching 505

 2. Inflation 506

 3. Savings Incentives 507

 4. Shareholder Gain 511

 C. Like-Kind Exchanges and Reorganizations 512

 D. Capturing Standard of Living 514


 A. Presumptions of Investment or Consumption 515

 1. The Efficient Market Thesis 515

 2. Diversification 518

 3. Life-Cycle Model 519

 B. Accounting for Consumption and Reinvestment 522

 C. Ordinary on the Ground of Potential for Consumption 527

The tax law gives a lower tax rate to capital gains, under the unstated assumption that capital gains will be reinvested. Capital gains are the gains from sale of investment property held for the required amount of time. (1) When first adopted in 1921, the lower capital gain rate was 12 1/2% at a time when ordinary income was taxed at up to a 54% rate. (2) Under current law, capital gains are taxed at most at 15%, whereas ordinary income is taxed at up to a 35% tax rate. (3)

Capital gain is defined by an intellectual tradition going back to feudal land tenure. The King, it is said, enfeoffed his nobles with possession of land in return for a pledge of knightly service. The living, while in possession of the land, held the land for the benefit of a male heir, so that the land would always support the warrior knights who would serve the king at his call. The house and manor belonged to the heir, no matter what they were worth. If the house and manor were sold, the proceeds had to be preserved for benefit of the male heir. Living possessors lived off the harvests of the land, but did not own the land itself and had no access to its gain in value. Under current tax law, the harvest of the land and things that are like the harvest are income from capital, taxed at ordinary rates, and gains that are like gains on the sale of the castle and manor are capital gains eligible for the preferential rate.

In adopting a preferential treatment for capital gain, both the British Parliament and the American Congress were thinking of capital gain within the framework of trusts that imitated the feudal estates. Capital gain, within the trusts, belonged to the corpus or remainder interests. Income beneficiaries would receive "income," but not the capital or the capital gain. Only the "income," that is, the amount distributable to the income interest could be consumed. Congress adopted the preferential rate for capital gain to imitate the British and also to reverse the Supreme Court, which had just held that capital gain that had to be returned to trust corpus or principal would be taxed at ordinary rates. Under the implicit assumption of what capital gains were about, capital gains are "capital," meaning amounts "not consumed immediately," but "contribut[ing] a quota to the national wealth". (4) Under traditional usage of the term "capital gain," "reinvested capital gain" is a redundancy. "Consumed capital gain" is an internal inconsistency. Consumption of capital gains is violation of the assumed norm, much like eating seed corn.

This Article argues that the implicit assumption that capital gains will be reinvested should be made into a requirement. The best policy arguments in favor of a lower rate for capital gain presume reinvestment. Lower rates are said to be needed to unlock capital, for instance, so that the capital can be pulled from less productive investment and put into more productive investment. If the capital is to be liquidated, however, there is no advantage to unlocking it. Other rationales for preferential rate for capital gain that do not depend on reinvestment do not work so well.

Consumed capital gain, by contrast, should be subject to ordinary tax rates because the tax brackets applied to ordinary income are better calibrated to the taxpayer's standard of living. The bracket system gives low rates for tax payers just above the level of subsistence, and high rates, up to 35% under current law, for amounts used to support a high, even profligate standard of living. A 15% tax rate is the rate judged appropriate at the middle of the range of standards of living. The 15% tax rate now given to capital gain is too low for dollars consumed to support the country's highest standards of living.

A side effect of the argument here is that the lower tax rates for capital gain seem to apply appropriately to like-kind exchanges and reorganizations, where capital remains invested. Like-kind exchanges and reorganizations are now tax exempt, but the arguments supporting tax exemption are the same arguments that underlie the preferential rate for capital gain, that is, that the amounts remain as capital.

The determination of whether capital gain has been reinvested should be done by looking to a taxpayer's overall position. If, for instance, a taxpayer invests the proceeds of the capital gain sale, but reduces other investments so that investments go down by the end of the year, it is not appropriate to view the capital gain as reinvested. The necessary accounting is a nonsimplistic combination of cash flow and basis accounting. Presumptions might replace the global accounting, but the presumptions yield rough justice.

The difficulty or roughness of determining whether capital gains have been reinvested invites considering the alternative of repealing the capital gain preference, and giving tax relief for capital only at the front end of investment. Incentives to investment, in fairness and efficacy, should be given to all investing. The benefit of the capital gain preference for investment is confined to existing capital held largely by existing rich. The capital gain preference is inconsistent with the more effective and fairer front-end incentives to investment. The British repealed their exemption for capital gain when they saw that capital gains might be consumed; there is wisdom in that rationale for the United States as well.

This piece traces the history of the capital gain concept, and then the policy arguments for a lower rate to argue for an express requirement that the lower rate for capital gains be available only if the gains are reinvested. Consumed capital gain, this Article argues, is not real capital gain within the original understanding. Indeed since capital gain is available for consumption, it seems wise to repeal the capital gain preference and direct investment tax incentives to the front end of investment.


A. American Capital Gain: Merchants' Loan & Trust Co. v.


Congress first passed lower rates for capital gains in reaction to the Supreme Court's refusal to find that capital gains were tax exempt by constitutional mandate, even though the capital gains in the case had to be reinvested. In March, 1921, in Merchants' Loan & Trust Co. v. Smietanka, (5) the Supreme Court held that Congress could constitutionally tax capital gains under the federal income tax then in place. Within months, Congress reacted by dropping the maximum tax rate on capital gains to 12 1/2%, at a time when ordinary income was taxed at rates as high as 54%. (6) As the Senate Finance Committee described it, Congress took an "intermediate position between the extreme views embodied, respectively in American laws[, which under Smietanka would have taxed capital gains in full at ordinary rates] and British laws, [which at the time exempted capital gains from income tax in full.]" (7) Congress seems to have been more impressed by the British exemption for capital gain than by the full tax applied in Smietanka, because the rate chosen, 12 1/2%, is closer to zero than to the 54% ordinary tax rate.

Smietanka involved capital gains realized by a trust that were allocated to the remainder interest and which were not available to the income beneficiaries. Within a system that claimed that "capital" was not subject to an income tax, Smietanka presented a best case for the taxpayer. In the specific facts of the case, one Arthur Ryerson died in 1912, before there was an individual income tax, going down on the Titanic. He left shares of his family corporation in trust with income to his wife and remainder at her death to surviving children. The trustee bank. Merchant's Loan and Trust Co., sold shares in 1917, realizing $700,000 gain - about $11 million in current dollars. (8) The trustee reinvested all the proceeds back into capital for the benefit of the corpus or remainder interest, as required by the trust documents. The income beneficiary received income from capital, and the income was enhanced because the capital had grown, but she was not entitled to any interest in or distributions reflecting the capital profit from the sale of the stock.

The taxpayer-trustee argued in Smietanka:
 If there were only one case where a conversion of capital assets does
 not produce "'income' ..., it must be the one where the conversion

 is by a trustee ... where the result of the conversion is merely a
 reinvestment of the proceeds as capital of the trust estate. ... The
 [trust] capital in question was not merely the particular stock and
 bonds held by the trustee, but was the principal of the trust estate
 in whatever form it might be during the continuance of the trusts."

The capital of the trust belonged to the remainderman even when the specific assets making up the original capital were replaced by other assets.

It was "universally recognized," the trust argued, "that the increase in value, ascertained by [sale], primarily belong[ed] to the capital of the trust estate as between life tenant and remainderman," unless the instrument otherwise provided. (10) For the trust, the sale was merely an ascertaining of an accretion to capital. To this day, in absence of contrary instructions, enhancements in the value of an estate or trust property, manifested by sale, are corpus distributed eventually to the remainderman, and do not belong to the income beneficiaries. (11) Arthur Ryerson's will, moreover, expressly provided that what he called, "accretions of selling values," would "be considered principal and not income." (12)

In Eisner v. Macomber, (13) decided the year before Smietanka, the Supreme Court had held that the 16th Amendment allowed Congress to tax income, but did not allow it to tax capital. Specifically the Macomber Court held that a prorata stock dividend had to be considered capital and not income, as a constitutional matter. The Macomber rationale was that income required that amounts be severed from capital, for the taxpayer's separate use, (14) and that rationale meant that any unrealized appreciation in the value of a taxpayer's assets are not taxable, at least until the assets are sold. Macomber was wrongly decided, 1 will argue next, but it was settled law in 1921 and prestigious beyond the borders of its facts.

1. Dubious Macomber

Macomber was probably wrong when decided on both the general constitutional exemption for capital and on its categorization of a stock dividend as not income. The tax exemption for capital found by the Macomber court, first, rested on Pollock v. Farmer's Loan & Trust (15) which had denied Congress the constitutional power to tax income, unless it apportioned the tax, pursuant to the Constitutional requirement that "direct taxes" be apportioned among the states according to their population. As I have argued elsewhere, Pollock was wrongly decided. (16) The pre-existing law, going back to the Founders, had correctly held that apportionability was part of the assumed definition of "direct tax." (17) A direct tax is a requisition upon the states except that the Congress and not the states decides the objects of tax. The original point of the clause was not to require apportionment but rather to say that if Congress did apportion a tax as in a requisition or direct tax, it had to follow the deal between the slave and nonslave states that slaves would be counted, in determining the wealth of a state by looking to population, by counting slaves as only three-fifths of a person. A tax on capital is not a direct tax. unless Congress decides to apportion it, because a tax on capital cannot reasonably be apportioned. Without the error in Pollock, there is no constitutional objection to Congress taxing capital.

Categorizing the stock dividend as capital rather than income, moreover, was inconsistent with New York law at the time, and the taxpayer, Mrs. Macomber, was a New York domiciliary. Under New York law, a stock dividend was allocated to the income beneficiaries of a trust, and not to principal or capital interest. (18) The rationale behind New York law was that the income beneficiary had an equitable right to corporate earnings as they arose. A Board of Directors of a corporation could not. by any form or procedure whatever, deprive the owners of the earnings, and give it to others not entitled. (19) A stock dividend was sufficiently like the distribution of the earnings to the rightful owner, because the end result of a stock dividend was as if the corporation had paid out cash and then collected the same amount back by selling more shares. (20) The New York rule, if anything, was even more pro-income than the usual cases that considered a stock dividend to be income because its rule was that "ordinary" stock dividends would be treated just like a cash dividend and income without regard to when the corporate earnings occurred. Other states, following the tradition that stock dividends were income, allocated stock dividends in part to corpus, if the corporate earnings represented by the stock dividend were earned before the contribution of the stock to the trust. (21)

The Supreme Court in Macomber decided that the stock dividend was in the nature of capital, relying on trust decisions in jurisdictions other than New York, which indeed characterized stock dividends as belonging to principal. (22) The Court did not consider the New York law that should have governed the character. Capital gains originated as a trust concept. In 1920, capital gains had no tax meaning. Thus the New York decisions on allocation of stock dividends to income beneficiaries of a trust should have been sufficient to allow the government to take its share. The United States was an income beneficiary given its tax on income, and under New York law income beneficiaries could not be deprived of their share of the earnings by a Board decision as to whether to distribute or accumulate the earnings. (23) In 1921, however, Macomber was stare decisis and prestigious beyond its facts.

2. Realized Gains are Still Trust Capital

The taxpayer's counsel in Smietanka tried to make Macomber binding on realized gain, as well as unrealized accretions: "The present argument is addressed to a court which ... is committed to the proposition that a mere increase in the value of capital assets, unascertained by conversion, is not income within the Sixteenth Amendment." (24)

The taxpayer trust argued that the sale and reinvestment did not make Merchant's Loan different from Macomber. "The increase in the value of the capital assets is just as much a mere increase in such value after it is ascertained by conversion in dollars and cents as it was when ascertained by calculation prior to conversion." (25) Sale was merely an ascertaining of accretion to capital: "As between life tenant and remainderman, the increase in value of capital assets of a trust estate ascertained by conversion, continues to be capital of the trust estate and not income." (26) "A conversion ... leaves the proceeds of the capital assets just as much capital assets as was the property prior to conversion." (27)

On that point, the Supreme Court rejected the taxpayer argument. The Court read Macomber as requiring that taxable amounts had to be severed from capital, and the sale in Merchant's Loan was a sufficient severance. The gain, the Court said, was '"produced by' or 'derived from' that investment, and that it 'proceeded,' and was 'severed' or rendered severable, from it, by the sale for cash, and thereby became that 'realized gain' which ha[d] repeatedly been declared to be taxable income." (28)

As to the trust in Smietanka, the sale replaced stock with cash. The trust was the taxable entity in the case. The beneficiaries were not before the court and were not asked to pay tax. The Smietanka gain might have been capital within the trust allocation, but it was a severance from the Ryerson stock as to the trust itself.

The trust in Merchant's Loan also argued that the gain was capital even if distributed to income beneficiaries:
 Such a rule is to be supported (if at all) not on the ground that the
 gain is income, but in spite of the fact that it is capital--the
 increment of value being distributed to the life tenant because of an
 expressed intent of the creator of the trusts, based upon surrounding
 circumstances, or upon as assumed intent founded upon equitable
 considerations. (29)

The argument is at least uncomfortable. The terms "income" and "principal" originally signified who is to get the money as a matter of law. Money belongs to the income beneficiary because it is income. For something to go to the remainderman, it must be capital. Both capital gains and income from capital are the products of capital invested, but only the former is principal and income from capital is distributable to the income interest, and it is just income. Under the facts of the case, the gains were not allocated to or distributable to income beneficiaries. The Court did not--and did not have to--decide whether allocation of the gain made any difference.

The Smietanka decision to tax the reinvested capital gain is perfectly defensible under an income tax. Income is now defined, in the standard definition, by where it goes, that is, either to consumption or to investment. (30) Trust receipts that went to the income beneficiaries were at least capable of consumption and trust receipts returned to corpus were investments. Still "income" should encompass both consumption and investment. The trust in receiving the cash had had the severance required by Eisner v. Macomber. The gain, equivalent to what is now $11 million, was gain under any reasonable accounting measure. What the trust did with the money did not undermine the gain. (31)

3. Congress Mostly Reverses Smietanka

Smietanka was decided in March, 1921 and within months, Congress mostly rejected it. In the Revenue Act of 1921, (32) Congress gave capital gain the benefit of a maximum 12 1/2% tax. when ordinary income bore tax rates as high as 54% (33) Cases won by the government in the Supreme Court are commonly contested immediately in Congress with varying success. (34) The only relevant support in the legislative history of the 1921 Act was testimony by a Frederick Kellogg of the New York Bar who argued before the Senate Finance Committee in May, 1921, that the tax on capital gains was a "full stop to business and not ... a revenue producer." (35) He argued ingeniously that tax on capital gain was unfair to the government because it prevented sales and deprived the government of revenue. (36) The United States should adopt the British tax law, he testified, "and cancel all taxation on such profit." (37)

As a matter of logic, one should be skeptical that cancelling all tax from capital gain would increase revenue. Still Kellogg's single-handed argument urging the British model was largely victorious. The Senate Finance Committee said it would adopt a position between the extreme positions of full tax (as allowed by Smietanka) and no tax (under English tax law). (38) The Revenue Act of 1921 was the first time that capital gains were taxed at a rate lower than ordinary income under American tax law. To be eligible for the lower rate, the taxpayer had to have held the property for two years. (39) The rate chosen for capital gain, 12 1/2%, is much closer to the British zero rate than to the 54% maximum rate Smietanka applied to capital gain, so it is the British example that was more influential. It is thus to Britain that one must go to understand the rationale.

B. Why Did the British Exclude Capital Gains from Tax?

Capital gain was originally excluded from the British income tax when the first general income tax was adopted in 1799 under the presumption that capital gains were allocated to capital or investment and were not available for consumption. The exclusion was reviewed in 1920 and 1957 and renewed on the explicit ground that capital gain was considered to be just an adjustment of capital. The exclusion was ended in 1965, moreover, when capital gains were perceived to be available for consumption. In 1965, Parliament said it must ensure that those who "spend money, [whether from realized gains or some other source,] spend it out of a taxed fund." (40)

1. Parliamentary Adoption of the Income Tax

The first general income tax in Great Britain was adopted by the British Parliament in 1799 under the sponsorship of Prime Minister William Pitt, the Younger, to raise revenue to fight Napoleon. There had been taxes on salaries in England, going back into the seventeenth century, that were said to fix English opinion that "income [was] the standard for all who paid direct taxes." (41) The 1799 income tax, however, created a more general tax on income, replacing the disjointed "triple assessments" of prior law. (42) The tax dramatically increased the revenue yield to meet the needs of the war.

Capital gains were excluded from the 1799 income tax for reasons that were under-theorized. The Parliamentary debates in the House of Commons focused on a proposal that Parliament should collect more of the needed revenue from a tax on value of capital and less from its tax on salary and earnings, and also on the proposal that the Parliament should borrow more and tax less. (43) In the House of Lords, the primary opposition argument was that a tax on income would make it difficult for the landed aristocracy to meet their obligations. (44)

A widely cited explanation for the exclusion of capital gains was that "[i]ncome tax ... is a tax on income. It is not meant to be a tax on anything else." (45) The explanation leaves something to be desired for those looking for an explanation of why money from capital gains was not income. The House of Lords later found an implied exclusion for capital gains, notwithstanding the literal wording of the income tax act, on the ground that Parliament "never ... intended to tax capital." (46) Gain from capital was just considered to be capital for reasons that seem to have been hard to articulate. "Capital" is defined as amounts "not 'consumed' immediately," but "contribut[ing] a quota to the national wealth." (47) Capital gains, within the British thinking, were considered to be still investment or wealth.

2. The Strict Settlement Model

It is possible to reconstruct the working presumption that capital gains were still wealth under the trust restrictions of the "strict settlement" of land. In 1799 when Parliament adopted the income tax, most land in England was held under a trust arrangement called the "strict settlement" that prevented the current generation from getting access to capital gain from sale of the land. When the eldest son of the family became of age or married, his father would convey the family lands to himself in trust for life, to his son for life, with remainder over to unborn heirs. After the conveyance, neither father nor son could sell the underlying property. Only at the death of the eldest son would the lands be freed up to be sold and even then, the expectation was the grandson would repeat the arrangement and continue the restrictions. (48)

The function of the strict settlement was to keep the large estate intact to be managed by the eldest sons in succession. Powerful peerages needed land to back them up. The landed gentry largely imitated the peers. All daughters, wives and younger sons were given an income interest for life, backed by the rents or harvests from the land, but no power to make the decisions of management of the estate. The terms for the strict settlement were commonly negotiated with the family of the prospective bride as a part of the marriage arrangements for the first son. (49)

The strict settlement was a trust arrangement because the courts and Parliament had disallowed similar restrictions under deeds of land that were not in trust. Entailments to keep a family estate together had to be contingent on survival because in an age of high mortality, no could be sure that a competent eldest son would survive. Contingent arrangements meant that no living person could sell the property, because the interest of unknown final beneficiaries could not be determined, paid for and cut off. The courts and Parliament voided the long term contingent restrictions, to quote a 1796 decision, because the "great policy of the common law [was] that alienations should be encouraged; for it is the greatest preserver and promoter of industry, trade, arms, and study. ..." (50) Under the parliamentary and common law, contingent arrangements could not remain contingent for more than a life or lives in being. (51) With high mortality, the contingent remainder was both necessary if a grand estate was to be given to a single heir and impossible at common law for legal (nontrust) estates.

The strict settlement system was not universal. Merchants never adopted it. A merchant's wealth was in money that fluctuated in amount. A merchant did not think of himself as the temporary custodian of the wealth in the same way that the landed rich thought of themselves as mere custodians for ancestral land. A merchant might leave his business to his eldest son with provisions for a surviving wife, daughters and younger sons, but he would not tie up the use or disposal of capital for future generations. (52) Even as to the landed aristocracy, the strict settlement was a privately drafted trust, with provisions that varied according to the settlor's situation and desires. A great family might, for instance, keep outlying lands free for possible sale, while keeping the family's traditional land and manor under the strict settlement. (53) Individuals who opposed the idea sometimes could keep a great estate out of the restrictions of the strict settlement for a generation or two. Even with the restrictions of the strict settlement, lenders were willing to lend on the security of the underlying land even when they might need to wait for two generations to get access to the property. Sales of land in payment of accumulated debt were, in any event, common. (54)

In 1882, Parliament provided that the current life beneficiary under a strict settlement could sell the underlying land and the purchaser would have good title in fee without the restrictions of the strict settlement. (55) The purpose of the 1882 Act was to "release the land from the fetters of the settlement--to render it a marketable article notwithstanding the settlement." (56) The selling life beneficiary, however, had to reinvest the proceeds of the sale in trust with the income going to the various life beneficiaries as provided under the terms strict settlement of the sold land and with the corpus of the trust distributed as the strict settlement had provided. After 1882, the strict settlement would no longer could keep the ancestral lands intact, but even after the 1882 Act, the capital gain from the settled land was returned to corpus of the trust, and not consumable by life beneficiaries.

The strict settlement was a trust drafted in private and it was "not in any essential aspect a creature of the state." (57) It nonetheless dominated thinking: "But throughout the two centuries after 1660, so long as a family owned a mansion and drew a substantial part of its income from rents, it was likely to retain the strong ambition to entail the estate for the benefit of the eldest son and use the strict settlement for the purpose." (58) The courtships described by Jane Austen, not long after the adoption of the income tax, take place within a framework in which the underlying land was held in trust. The fortune of a "single man of a good fortune" who, according to Austen, "must be in want of a wife" (59) was measured by his income interest, not by the value of underlying lands. (60) Elizabeth Bennett's situation in Pride and Prejudice was precarious because her parents held their estate for their lives only, with remainder to a remote relative, and they could not leave an estate for Elizabeth or her four unmarried sisters. (61)

When land was held under the strict settlement, the view that capital gain was not income also arose from identifying capital or corpus with a physical property, rather than a sum of money. The original capital was primarily land, a physical thing. Capital in the concrete sense refers to a thing. (62) The castle and manor belonged to the heir no matter what it was worth. "The tenant for life [by contrast]," as one treatise put it, "takes [only] 'the annual produce, the grass, the apples, and things of that sort." (63) Income was consumable and capital was dirt. Strictly speaking, of course, the gain from sale of the land was no longer dirt, but money, but even after the 1882 Act allowing purchasers to get title in fee for property formerly held under strict settlement, the cash from the gains from property held in strict settlement had to be reinvested for the corpus beneficiaries.

Sources contemporaneous with the 1799 adoption of the income tax describe the strict settlement restrictions as a vestige of the feudal military system. The principle underlying feudal obligations was said to be that the king had once owned all land and had let it out to his nobles in return for services, primarily military services of the noble and his knights. The restrictions of primogeniture were said to be necessary to maintain support for warrior knights with upper body strength who would serve the king at the muster. Laws "last longer than their causes," wrote Dr. Samuel Johnson, "[and] the limitation of feudal succession to the male arose from the obligation of the tenant to attend the chief at war." (64) A "fundamental maxim and necessary principle (though in reality a mere fiction) of our English tenures," Blackstone said, is that "the king is the universal lord and original proprietor of all the lands in the kingdom and that no man doth and possess any part of it but what has ... been derived as a gift from him, to be held upon feodal services." (65) The English ancestors, Blackstone said, were trying to maintain a military system. "[T]he reason of conferring the feud being the personal ability of the feudatory to serve in war," Blackstone explained, the current holder of the land should not be "at liberty to transfer this gift, either from himself, or from his posterity who were presumed to inherit his valour, to others who might prove less able." (66) The life tenants maintained their standard of living solely from the harvest or income from the land. The land, under feudal tenures, belonged to the next warriors.

One should be skeptical, notwithstanding Dr. Johnson and Blackstone, about the continuity of restrictions on sale of land from feudal knights' service to strict settlement. By the end of the thirteenth century, a noble could satisfy his military obligations by paying scutage, or fees in lieu of knight's service. The crown found professional mercenaries provided a better military than aging noblemen and wanted the scutage money instead of the service. (67) Actual military service seems to have been largely replaced as a functioning system by supplemental tax by the late 13th century. (68) In 1290, Parliament allowed the holder of a tenure to sell to a stranger, so long as the purchaser undertook the feudal obligations to the lord. (69) The purchaser, of course, did not necessarily have the military prowess of his selling predecessor. There is a thus a more than 500 year gap between the restrictions on sale to maintain knights' service and the 1799 income tax. As noted, moreover, the voluntary, privately drafted arrangements of the strict settlement trust were necessary because the common law defeated contingent interests at law that prevented sale of the land. (70) Reinvestment of capital gains under the strict settlement was entirely a trust concept drafted without participation of law or state. America, of course, never had knights' service, or knights, so that the concept of mandatory reinvestment of capital gain to maintain knights' service comes only by borrowing from Great Britain.

3. Distributed Amounts Are Distinguished

Both income from capital and capital gain are the product of the investment, but only the latter was exempt because only the latter had to be reinvested. The harvest, and things like the harvest, belonged to current income beneficiaries even though it was a product of capital. Amounts available for distribution income beneficiaries were not capital gain.

Within some branches of the intellectual family of capital and income, mere distribution of the item changed the character from capital into income. A corporation is entitled to make dividend distributions to its shareholders only out of the earnings of the corporation. The corporation must keep its capital intact: "The capital of the corporation is a trust fund for the benefit of every one interested therein," it was said, "and, if it is misappropriated, those responsible for the diversion are liable jointly and severally for such payments as are made with their sanction." (71) The preservation of corporate capital seems to come from the same set of ideas as shaped the mandate for preservation of trust capital first in feudal estates and then the trusts. Under the corporate dividend restrictions, however, capital gain realized by a corporation justifies dividends to shareholders. In the 1892 British case of Lubbock v. British Bank of South America, (72) for instance, a corporation sold its Brazilian banking business, which seems to have been the bulk of the corporate activities, and distributed the profits. The chancellor in equity held that the dividend was permissible, justified by the corporate "earnings," that is, profits from sale, on the rationale that "[a]ccretions to ... capital may be realized and turned into money, which may be divided amongst the shareholders." (73) In American tax law, consistently, capital gain on the corporate level becomes ordinary income to shareholders when distributed to the shareholders. (74) There has been some expressed unhappiness with the rule that a corporation can distribute its capital gains. (75) Corporate capital gains that are distributed, however, and can be consumed by shareholders cease to be capital and cease to be capital gains.

While a corporation's distributions of corporate capital gain will be ordinary income to shareholders, distributions by a trust of trust capital gain are now capital gain to beneficiaries, including income beneficiaries. A trust is a pass-through entity, a semi-transparent tax entity under American tax law, and amounts distributed "have the same character in the hands of the beneficiary as [they had] in the hands of the trust." (76) This is a case in which the pass-through tax model has come to dominate the original function of income or principal characterization used to determine whether the current beneficiaries get a distribution of the item. Corporations, unlike trusts, are not pass-through entities for tax. Nonetheless the original purpose of principal and income was to determine whether amounts are distributable and that idea survives at least as a worry. The Bogert treatise on trusts, for example, has worried that if capital gains of a certain kind were allocated to income, that the allocation for trust purposes will decide tax character. "It may be argued," says Bogert "that if capital gains on shares held in trust go to trust income, either by court decision or federal tax laws, it will be provided the these distributions are not capital gains but rather ordinary in income for tax purposes." (77) Capital gains, in the original concept, meant amounts allocated to principal and not available for distribution. The yield from capital that is available for distribution to income beneficiaries were just "income."

4. Review and Retention, then Revocation

Great Britain continued the exclusion of capital gains from tax in 1920 and again in 1955 on the argument that capital gains represented a mere change of investments. In 1920, not long before Smietanka, a Royal Commission recommended lightening up the income tax to bring merchants' trade profits within the income tax even if they were not "annual" or recurring. (78) As noted, merchants never had much use for the strict settlement tying-up of capital for future generations. Merchants' gains were apparently not what the Parliament was thinking of when it exempted capital gains from a tax. The 1920 Royal Commission, however, recommended continuation of the exclusion for capital gains from investment saying that "profits that arise from ordinary changes of investments should normally remain outside the scope of tax." (79) Capital gains were by definition in 1920 a mere change of the assets of investment which preserve capital through the sale and reinvestment.

Similarly, a 1955 Royal Commission concluded that it was "much better to assume that, on the whole, capital gains will be saved," (80) and it did not think it was appropriate to tax "merely a change of investment." (81)

In 1965, however, Great Britain ended its tax exclusion for capital gains on the argument that the seller could spend the gains, and that spending from whatever source should be done from a taxed fund:
 The justification for the Capital Gains Tax is that of equity and
 justice. Most people in this country, whatever they spend, spend out
 of a fund of taxed income. There are a select few who, when they
 spend, spend money out of an untaxed fun. It is a as simple as that.
 and what the Capital Gains Tax does is to ensure that those who spend
 money, be realising gains on their capital, spend it out of a taxed
 fund. We believe that that it is wholly unjust and unfair that the
 small man who has to pay tax on every penny which reaches his pocket
 before he spends it, should be in a different position from his
 wealthier neighbour who has not had to pay tax on his capital gains.

The only change between the period from 1799 through the review and retention in 1920 and 1955 and the repeal of the exclusion in 1965 was that in the early decisions capital gain gains were conceived of as mere adjustments in investment that preserved capital, whereas in 1965, the capital gains were perceived to be available for spending or consumption. Once Parliament perceived that capital gains could be consumed, it ended the exemption.


As a matter of tax policy, as well of tax history, the most influential arguments that explain the reduced tax rate on capital gains presume that the proceeds from a sale qualifying for lower capital gains rates will be reinvested. Lower rates on capital gain are a legacy of the argument that capital gains must be reinvested and preserved, for instance, when it is argued that low rates are necessary to ease lock-ins of capital that prevent a holder with built-in gain from moving from a current to a better investment. Capital gains are also advocated, without necessarily assuming reinvestment, to offset 'bunching" of income, inflation, and penalties on savings, but capital gain is an imperfect remedy for those complaints and often gives the tax relief in the wrong place. When capital gains are not preserved but consumed, the tax-rate brackets for ordinary income better capture the appropriate rate for the taxpayer's standard of living.

A. The Reinvestment Presumption

1. Lock-in

Supporters of a preferential capital gains rate have argued that taxes imposed on the sale of property "lock-in" capital, preventing the capital from shifting from old to better investments. (83) Capital gains taxes are a toll charge on moving from an old investment to new better investments. An investor will sell an old investment to buy the new investment only if the new investment gives a return that exceeds the return from the old by enough to pay the toll charge. How much higher the return on the new investment must be depends, not just on the capital gain rate, but also on how much gain has built up on the old investment. (84) Empirically, real taxpayers do not behave quite so sharply as rationality implies, but tax does suppress sales, unsurprisingly, by comparison to behavior in no-tax environments. (85) If the toll charge were lowered, the argument goes, capital would shift from old to the better investment more readily. Even opponents of preferential rates have called the lock-in argument the most persuasive of the traditional arguments for preferential rates on capital gains. (86) If capital now locked in were going to be liquidated in consumption, by contrast, then we would get more productive capital with lock-in than with sales.

In the debates on what rates capital gains should bear, capital gains taxes are typically described as taxes on reinvested capital by definition, sometimes in the face of evidence cited to the contrary. (87) Staff economists for the congressional staffs that review tax policy, including the Joint Committee on taxation, (88) the Library of Congress Congressional Research Service, (89)and the Congressional Budget Office (90) have all presumed that capital from capital gains sales will be reinvested, without regard to evidence. Reinvestment is an assumption in the arguments, part of the definition of the issue they are looking at.

One should be skeptical, in the first place, that lock-in needs a remedy. Lock-in is not likely to damage allocation of capital in the economy as a whole because there is enough capital not affected by lock-in to ensure that prices will reach their correct reflection of the expected returns, without the locked in capital. For example, pension funds can move to the more attractive return investments without being affected by tax. New savings from corporate, foreign or individual earnings will go to the most attractive available return, without paying a capital gain tax toll charge on the way. Opportunities not taken by taxpayers who have existing assets with gain they do not want to realize will be taken up at their correct relative price by competing bids coming from pension funds or new savings. The alternative sources of capital will maintain correct relative prices and sufficient capital for the new attractive investments. The argument that the capital gain toll charge is preventing better investments from getting capital may well be the best argument in favor of the lower rate, but with sufficient alternative bidders, lock-in is not of public economic concern. (91)

For the holder of old capital with material built-in gain, the lock-in deprives the owner of the opportunity to make a higher return that would be available if the owner could avoid tax. Still, the equitable case in favor of lower tax on reinvested capital gain is not very strong. In an income tax, investments are made only out of post-tax "take-home" pay, so that there is ordinary income tax to pay before entry into an investment. Reinvestment of capital does not seem any more worthy of better treatment than first time investment, so perhaps reinvestment should also pay ordinary tax before investment. If there are going to be incentives or reduced penalties on investment, then the new investment and the reinvestment should benefit equally. Capital gains arise not for every investment but only for capital that has been invested for some time, and people with the most old capital that has appreciated over time are called rich.

If the holder of the asset is not reinvesting, but rather consuming it, then there is nothing to the lock-in argument even under its own terms. The purpose of reducing lock-in is to allow capital to move to a better return investment, but if the capital is consumed, there is no investment at all. Both lock-in and the underlying concepts of returning capital gains to corpus or principle are built around the goal of preserving capital for future heirs. If the capital freed up by a lower toll charge is consumed, however, the capital is not preserved but destroyed.

2. Amnesty at Death

The primary cause of lock-in, in any event, is the amnesty given to capital gains tax at death. Under current law and since the adoption of the income tax, successors receiving assets by bequest upon death of the prior owner get to treat their cost basis as if they had purchased the asset for its fair market value at the death of the prior owner. (92) The capital gain that arose in the hands of the successor before his or her death is never taxed.

The reason for the step up is another under-theorized aspect of capital gains taxation. (93) Congress's stated reason for the step up in basis at death was that it was just conforming the statute to the existing Treasury rule. (94) Basis, in general, is a ledger account describing amounts the taxpayer or some successor has already paid tax on. We have gotten used to the concept of carry over basis under which a taxpayer steps into the shoes of his or her predecessor as to basis. (95) As the Supreme Court said upholding the carryover of basis in what is now section 1015, "[i]n truth the stock represented only a single investment of capital--that made by the donor." (96) Within an income tax, capital is taxed before it has been invested, so that treating capital gains as capital is a better reason for subjecting the gains to tax than for exempting them.

The best explanation for a step up in basis in the hands of a transferor without prior tax seems to be in the tradition going back to feudal tenures viewing capital as primarily a physical thing. (97) The castle and manor belonged to the next knight, under the tradition, no matter what they were worth.

The step up in basis also seems to assume that capital is perpetual, and will not be consumed. If heirs are going to consume the proceeds of a sale, then it is difficult to see why the gain should not be run through a bracket system that taxes subsistence and modest standard of living money at a lower rate than luxuries and high standard of living money. To put a certain spin on the question, if the ancestral lands are going to be sold and spent by wastrel heirs on bon-bons and character destruction, why should the real gain not be reached by the ordinary income tax brackets system?

Given the forgiveness for capital gains, moreover, it is difficult to maintain a high tax on voluntary sales during life. (98) It is hard to maintain any pressure in a pressure chamber if a door is wide open down a short corridor from the core. Taxpayers live off of income, where realization is discretionary, or by selling investments with a loss or little gain. The availability of tax forgiveness at death acts as a powerful suction. Forgiveness of capital gain tax at death absolves between 54% and 90% of capital gain from tax. (99)

As a practical political matter, the primary attractiveness of a reduced tax on capital gain is the possibility that the Treasury revenue might increase in reaction to a rate cut. Starting from a premise that amnesty on death must be available, then reducing capital gain tax on sales during life to catch up with the amnesty might increase neutrality of tax. Reductions in capital gains tax on sales during life can increase realizations by enough to make a rate reduction a revenue gain. A cut in rates that increases Treasury revenue is a gain both to investors and to the Treasury. Everyone would be better off, and no one will be worse off. (100)

Increase in revenue from a tax cut is possible, however, only from the questionable baseline that tax must be forgiven at death. Increased realizations of the capital gain can generate permanent revenue increases only if taxpayers give up their tax amnesty at death and pay tax instead. (101) If capital gain tax is just an early realization of gain that would be taxed early, then any amount realized this year will reduce the gain left to be realized next year. Starting from a premise that the gains will be taxed eventually, a rate cut does not increase revenue. (102)

If lock-in were a serious problem in economics or equity, it is possible to engineer a better remedy for the lock-in and revenue loss arising from step up in basis at death, by taxing the built in gain on taxpayer assets as if the taxpayer had sold the property at the time of his death. If taxpayers faced a rate of tax on death higher than the capital gains rates, there would be an incentive to realize the gain in voluntary sale early to take advantage of the lower capital gains rates. Lowered capital gains rates would be saved for the voluntary sales and not for the inevitable realization of gain on death. It is impossible to keep up with a zero rate and still collect revenue, but it would be possible to have voluntary realizations if the gain was inevitable taxed anyway at death at a rate higher than the capital gain rate.

B. Arguments Not Depending on Reinvestment

There are also arguments in favor of lower tax on capital gain that do not depend on a reinvestment of the capital. Lower tax rates for capital gain are said to be necessary to counteract bunching, inflation, and to reduce disincentives to saving. The literature is extensive enough on these arguments that it is impossible to do more than a cursory review here. At the very least however, the non-reinvestment based arguments for capital gain are imprecise and give tax relief where none is merited.

1. Bunching

It is sometimes argued that the lower tax on capital gains is necessary to prevent the inequity of a bunching of many years of income into a single year. Assume a farmer who is normally in a 15% tax bracket sells a farm upon his retirement after many years of holding the farm. The gain reported in the year of sale might well be large enough to be taxed in the highest tax bracket. If the gain had been taxed as it arose, the farmer might have still been in a 15% tax bracket for all the gain. Moreover, if the farmer will consume the gain over the next thirty years of retirement, the 15% tax bracket might well describe the tax rate appropriate to his standard of living. (103)

The right remedy for a bunching argument, however, is to average income over the period of holding of the asset and not to give an automatically low rate. (104) Typically taxpayers with the most capital are appropriately in the highest tax bracket, perpetually. If one averaged the gains over a lengthy time, the appropriate rate for some of the rich, according to their level of consumption, would still be 35%. If gains were taxed as they arose, moreover, the tax would have been due earlier. Under time value of money principles, the deferral of tax until sale is itself a reduction of the economic impact of the tax. (105) Offsetting the advantage of deferral may imply a tax on holding gains that is equitably higher than ordinary tax rates, not lower.

2. Inflation

Capital gain is sometimes advocated to offset the impact of inflation. Tax (and nontax) accounting treats all dollars as profit ignoring the impact of inflation. Ignoring inflation creates fool's gain, measured in dollars that do not reflect a true improvement of standard of living. If investor buys stock for $100 and sells it for $120 after there has been 10% inflation, for instance, the first $110 of the sale price is not gain at all, but just the number of dollars necessary to match the original investment, measuring the original investment in terms on what the original $100 could buy. A 25% tax on the measured $20 gain is economically a 50% tax on the $10 real gain taking out the inflation. If the investor sells for $103, under the same inflation, the investor has a real $7 loss in standard-of-living terms and also a measured, but artificial $3 gain. In inflationary times, the tax can commonly exceed any gain.

Fixing the taxation of fool's gains would require a symmetrical remedy, disallowing deduction of interest that just offsets inflation. Congress's unwillingness to fix interest, for instance, on home mortgages will probably block any possibility of fixing the taxation of inflationary gains. Still, to the extent that the investment is not debt-financed, ignoring inflation overstates the taxpayer's income.

For capital gain held for a long time, the deferral of tax from the time the gain accrues until the time of sale is an offsetting benefit. Income from capital that is taxed currently does not benefit from the deferral. Given the general availability of real returns greater than inflation, deferral will eventually be more beneficial to the holder in the usual case than inflation is a detriment. (106) The forgiveness of capital gain tax at death makes 50% to 90% of the capital gain go away. The combination of deferral of tax and forgiveness at death mean that even with the taxation of fool's profits, the expected effective capital gains rates can be lower than the tax on other capital income. Lower rates for capital gain are available even when deferral and forgiveness at death already reduce the tax rate below the norm applied to other capital income.

The best case for inflation offsets is for the ordinary income from capital rather than for the capital gain. An accounting system overstates income when it ignores inflation in accounting for rents, interest, harvests, and business incomes derived from the investment of capital. Ordinary income from capital is taxed currently, so there is no benefit of deferring tax and not much of a chance of the step up in basis at death giving it any benefit. Inflation, indeed, induces investors to avoid the income from capital that is over taxed currently and plan instead for investments that can delay or avoid realization of income. Reducing the tax rate on the capital gain but not other capital exacerbates the bias. (107) However, it is the current income and not the holding gains that should be the first beneficiary of an inflation remedy.

3. Savings Incentives

A cut in capital gain is often advocated as necessary to give investors better returns on their investment after tax in order to induce more savings. (108) However, incentives to more savings are plausibly ineffective and might even work in the wrong direction. Reduced tax on capital gain misdirects the remedy to old capital already formed, rather than to new capital. The capital gain remedy, moreover, gets in the way of incentives at the front end for capital being formed.

One might be skeptical that the world needs more capital right now. The globe is described as having a glut in capital (109) that drives down the returns that are available to capital. A global shortage of oil, relative to increased demand, meant that there were dramatic increases in the price of oil. (110) There have been no similar increases in the price paid for capital. Real return rates are stagnant, even falling, because the supply of capital is adequate.

One should also be skeptical that reducing tax rates will push savings in the right direction. Much of real savings is target savings trying to reach a real goal, and target savings drop when tax rates on returns to savings go down. Raising taxes on capital, by contrast, increases target savings. Assume, for instance, a particular parent needs $500,000 in eighteen years to send a child to four years of college. If return rates are 4% after tax, the parent must set aside $246,000 now to reach the target. (111) If tax is reduced to increase aftertax returns to 7%, then the same target can be reached by setting aside only $118,000. (112) Increasing the return will allow the parent to spend the difference or reduce savings now by $98,000. Savings to reach a down payment, basic retirement or, seed money for an enterprise also behave as target savings. Decreasing the tax on returns will decrease target savings. Overall, target savings seem to dominate in our economy. When interest rates go down, savings can go up, apparently because of the reaction of target savers. (113) Over time and differing circumstances the incentive effects of lower tax and the target savings phenomenon wrestle for dominance over savings with no clear winner." Still, the incentive effect is never a clear winner. The "[s]upply side tax policy [in the Tax Reform Act of 1986, for instance] conspicuously failed to encourage private savings."(115)

Even if subsidy to capital were justified, capital gains cuts direct more benefit to existing capital. Only a very small fraction of capital every year is new capita--as savings rates have plummeted, the new capital represents an ever smaller fraction of the whole. Capital gains arise slowly over time so that the older the capital is, the larger the fraction it is of the total capital gains. A new saver cannot expect to get any material benefit from capital gain for many years. Capital gains cuts spend most of their revenue lost on savings that were made years before the enactment of the cut, before anyone could know what the capital gains rates were going to be. One cannot make retroactive incentives to have any effect to behavior before the incentives were known. (116) Money spent to give an incentive to things that have already happened is pure waste. If subsidies were to be given to encourage the formation of capital, the incentives should be delivered to new savings. Old savings do not even need to participate. Indeed lower rates on capital gains in general are giving up on the occasion of liquidating the savings, not in creating it. A reward or incentive to sell and liquidate savings will decrease capital when the capital is consumed.

There is a considerable literature in favor of a consumption tax to end double tax on capital. (117) An income tax reduces investment both when the investment is made and on investment profits. If taxpayer receives a four-acre apple orchard as income, for example, under a 50% income tax, two acres are lost when the orchard is received and tax of the harvest takes half the apples that are left. The result is that a pre-tax four acres worth of apples are reduced to one acre worth of apples under a 50% income tax. Consumption tax advocates call for either exemption of the receipt of the orchard or exemption of the apples.

The right place to put tax relief under consumption tax norms is on the front - or orchard - end of the investments: (1) If the investment is like a double or nothing bet, only the tax of the results of the bet will distinguish the destitution of loser from the lap-of-luxury consumption of the winner; (118) (2) Greater than normal returns must often be explained as labor, skill or entrepreneurial input, which can be reached only after the effort by taxing outcomes;(119) (3) Front-end tax of high rates of return skill leave the taxpayer in the same position as back, or apple, end taxation of the investment, but front end tax leaves the government worse off because it does not share in the extraordinary return. (120) Capital gain is a form of back - or apple - end tax relief and it is inferior at taxing the standard of living from extraordinary returns. The best place for relief from the double tax is at the front end.

Lower capital gains rates are inconsistent with front or orchard end reductions in tax. For example, when an investment benefits from both a deduction of the investment on the front end and a capital gain at the liquidation of the investment, the combination of the two yields a negative tax or subsidy in which the return is better after tax than before tax. The subsidy, under one set of assumptions, turns a 10% rate of return from an investment before tax into a 25% rate of return after tax. (121) There are numerous opportunities to deduct investments in our income tax. including, e.g., investments in business intangibles such as development of pharmaceuticals or software, trademarks, and workforce, advertising, research and development, oil drilling, and mine development. (122) Negative tax subsidies allow investments to go forward that would never be made in absence of tax because they cannot carry the going cost of capital, and that wastes capital. Greater incentives on the front end of investment will require repeal of capital gain.

4. Shareholder Gain

A final nonreinvestment rationale for lower taxes on capital gains is to moderate the double tax on corporate income. If a corporation pays 35% tax on $100 of income, the $65 that is left is subject to tax again on the shareholder level. If shareholders pay 35% on the remaining $65. there is only $42.25 left after tax. Debt capital and partnership capital would yield $65 on the same economics of $100 pretax income. The extra shareholder tax discourages public corporations, notwithstanding their virtue in amassing great amounts of capital, and encourages debt capital, notwithstanding the greater fragility it adds to the capital system.

Internal Revenue Code (Code) section l(h)(11) now gives dividends the benefit of the 15% capital gain, under the rationale that dividend income has already borne a corporate tax. (123) Assuming, arguendo, that a corporation really paid 35% tax on its income, the 15% capital gain rate would improve the after tax income from $42.25 to $55.25, which is more in line with the $65 after tax available from debt or from partnerships. Capital gain on corporate stock also reflects past and future earnings. Therefore, if past and future earnings bore the 35% then there would be merit for the 15% tax on both dividends and on capital gain from sale of corporate stock.

The primary difficulty with the double tax argument under current law is that the 35% corporate tax is fiction. Taxpayers display the real or effective rate of tax they pay on investments by their willingness to pay for tax exemption on municipal bonds by accepting lower rates of interest. Municipal bonds compete at the margin with all other investments and the discount on interest they offer displays the tax rates on any investment. The discount on interest on long-term bonds is under 10% because the market gives opportunities to reduce tax to under 10% real or effective rate on other investments. (124) A 15% shareholder tax combined with a 10% effective tax rate at the corporate level yields $76.50 after tax. (125) The maximum statutory tax rate says that 35% tax or $65 after tax is the appropriate level to impose on money going to support the highest standard of living consumption, so that under current conditions the capital gain tax is too low even as a double tax. Capital gain on corporate stock, but not on partnership or LLC entities would make sense within a system that did a better job of taxing corporate income. (126)

C. Like-Kind Exchanges and Reorganizations

A side effect to the conclusion that reinvestment of capital is an assumption in both the history and policy of capital gain, is the principle that like-kind exchanges and reorganizations have no better case for lower tax than capital gain does. Under current law, no tax is imposed on certain nonrecognition exchanges on the "[t]he underlying assumption ... that the new property is substantially a continuation of the old investment still unliquidated." (127) In a like-kind exchange, "the taxpayer's money is tied up in the same kind of property as that in which was originally invested." (128) In a corporate reorganization, the amount realized in the exchange is not taxed because they represent "only a readjustment of continuing interest in property under modified corporate form. (129) The transferring taxpayer in the nonrecognition transactions puts the realized gain into new property. In both like-kind exchanges and reorganizations, any cash withdrawn from the transaction in the exchange is taxed as boot; that is. the gain is taxed to the extent of cash received. (130) The rationale given for nonrecognition is that gains remain invested after the exchange. The rationale is neither different from nor stronger than the rationale for the lower tax on capital gain. It would be appropriate, then, to impose the lower, now 15% tax rate on like kind exchanges, reorganizations and similar transactions.

There are generally no serious valuation problems in the exchanges now qualifying for nonrecognition. If there are any boot or debt assumptions or new purchases in connection with a like-kind exchange, both sides to the transaction know to the penny the value of what they have given up and the value of what they have received. For reorganizations, the public market usually sets the value without much controversy. Valuation problems that remain can be handled by fair estimates. Certain barters might be exempted from the gain requirement, but only if they are small, informal and not worth mentioning. (131) Under this proposal, the gain calculated from the fair market value of the property received would be capital gain, even if the transaction qualifies for like-kind exchange or as a reorganization under current law.

D. Capturing Standard of Living

It is proposed that realized capital gain that is not reinvested should be subject to ordinary tax rates. The best reason is that the tax rates of the brackets for ordinary income are a better measure of the appropriate level of tax calibrated to the taxpayer's standard of living than is the capital gain rate.

Section 1 of the Code provides for a bracket system under which the rate of tax depends upon what income the taxpayer has. (132) A bracket system attempts to adjust the tax rate to the level of the standard of living of the recipient. The rationale for the bracket system is that there is a varying value for dollars depending upon the economic situation of the recipient. Every consumer spends first for the most critical needs and then spends for less desperate needs as income rises. The pattern is continuous from low income to high. Dollars received, for example, by the Little Match Girl for subsistence are too valuable to her to be taken away, but dollars received, e.g., by Uncle Scrooge McDuck, are added to big vaults of money and do not themselves add much to his utility. The dollars are most desperately needed at low levels of income and less desperately needed as income rises. The bracket system adjusts to the value of the dollars. Recipients of income get zero tax rates for amounts necessary to keep them alive, and modest rates for modest living. High tax rates are applied to the highest standard of living. The brackets reflect the reasonable judgment that the government will do less harm in tax if it pulls its tax from a standard of living where the loss will be less. The capital gain rate, now at 15%, is too low a tax rate to apply to money consumed for the highest standard of living. The duly enacted statutory judgment is that 15% is the appropriate tax rate for households below or just at the median level of income. (133) Money spent for consumption by our richest households is now supposed to bear tax at 35%. If capital gain is spent for luxury, or a high standard of living, it should bear the highest ordinary tax rate.


It is proposed here that the lower capital gains rates be available only for capital gains that remain part of capital. The models that Parliament and then Congress used to give lower rates were trusts where capital gain had to be reinvested, and the best policy arguments for lower rates presume reinvestment. Reinvestment of capital gains is not explicitly required under current law. If capital gain is consumed, however, the tax brackets for ordinary income better capture the rate appropriate to the taxpayer's standard of living than does the lower, now 15%, capital gain tax rate.

This section looks first at presumptions as to whether amounts are saved or consumed and then at the accounting that would be needed to keep track of an individual's savings and consumption. After reviewing the serious difficulty of distinguishing consumed and saved amounts, this section considers the alternative of taxing capital gains in full because they are available for consumption.

A. Presumptions of Investment or Consumption

Because dollars are fungible, it is often difficult to determine whether a taxpayer has in fact consumed or reinvested capital gain. To avoid the problems of accounting for a taxpayer's actual consumption and reinvestment, it might be feasible to rely on rules of thumb derived from overall patterns of behavior, and ignore what actually happens.

1. The Efficient Market Thesis

The efficient market thesis implies that it is irrational to sell an investment in order to reinvest the proceeds in another investment. The efficient market creates a filter. Sales that pass through the filter are more likely to be sales made for the purpose of consumption.

In an efficient market, prices move quickly to reflect all publicly available information. Once a smart or efficient market has adjusted the price to reflect its true value, there is no longer a profit available from a sale to reinvest. One cannot find a bargain in high quality goods or investments because the self-serving sellers of such quality investments are trying to extract the maximum price and competing bidders have bid up the price to reflect the value. One cannot profit from selling a low quality asset to invest in something better because of the skepticism of self-serving buyers. Trading low quality assets for high quality assets will be unprofitable when the low price for the sold assets and the high price for the purchased asset already reflect the relative quality. If the market is reasonably efficient, no investor can be confident enough that a sale to reinvest will be profitable enough to bear any substantial transaction costs. In a perfectly efficient market, even modest transaction costs will drive out a sale to reinvest from having any expected profit. Even in a market only imperfectly efficient, price adjustments that occur filter out sales to reinvest and leave the sample of sales that do happen enriched with sales made for consumption.

The stock market is said to be efficient because its prices are the result of thousands of active bidders and hundreds of thousands of dollars of research by sophisticated investors bidding in their self-interest. (134) There is no other proposition in economics, it is said, "which has more empirical research supporting it than the efficient markets hypothesis." (135) Bidders who consider a stock underpriced given its real future returns will bid up the stock price quickly; owners and short sellers who consider a stock to overvalued by price sell and drive down the price. (136)

If anything, the stock markets are getting faster in their speed of adjustment. Kenneth Froot and Andre Perold, (137) for instance, looked at changes in price over fifteen-minute intervals in the price of marketed stock over the period from 1983 to 1989. In the early years of their study, they found that stock prices exhibited trend lines. The change in price over a prior fifteen minutes was a good predictor for the change in price that would occur in the next fifteen minutes. By the late 1980s, however, the trend lines were weak or nonexistent and the correlation of one change to the last was nearly random. The existence of hedge funds and other financial innovations has meant that news is disseminated to and digested by the market even more rapidly. Stock market prices seem to adjust to news and reach randomness in some period of less than fifteen minutes. (138) As the price of acquiring various kinds of information declines, the market becomes more efficient. (139) The wider application of sophisticated statistical tools to identify anomalies in pricing has plausibly increased the efficiency of the markets, even since the Froot and Perold study. (140)

There are limits on how efficient the markets can be. Research that keeps price in tune with fundamental value is expensive. Institutional investors making large trades conduct the best research, as they can amortize high costs over large blocks of stock. The costs of research to the largest investors put a limit on how efficient the market can be. (141) Only public information, that is, information accessible to the largest investors for a price they can carry, can be expected to affect price. Prices are also affected by irrationality, irrational exuberance and panicky sales. (142) The market can be fooled by company frauds, at least until the fraud is known to the public. (143) Stock value depends upon predictions about future events and opinions can and do differ about the probabilities of future events. Still a market need not be perfectly efficient to suppress sales for reinvestment. No individual investor can be assured that price differs from value and in what direction because if the departure is generally known, informed institutional bidders in the market quickly end the disparity.

Under the efficient market thesis, all sales realizing a substantial gain should be presumed to be sales made for the purpose of consumption.

2. Diversification

Sales are rational even in a smart market to maintain diversification of investments. Modern portfolio theory mandates that an investor maintain a diverse portfolio of investments, so that losses on one investment or kind of investment will be offset by gains in other investments. Offsetting patterns from diverse investments smoothes the roller coaster ride of a single stock. (144) A portfolio of investments that started with adequate diversification will became unbalanced over time with volatile investments because the usual pattern is that a few stocks have large appreciations, and many stocks have modest growth or decline. (145) The few dramatically appreciating stocks will come to dominate the portfolio just because of their appreciation, and because the loss stocks shrink as a proportion of the whole. The stock that dominates the portfolio needs to be sold to retain diversification of the overall portfolio even when the reason for the imbalance is past success.

When diversification is most needed, however, there will also be tax losses that make it possible to diversify without any tax, even a capital gains tax. Section 1211 of the Code, with a small exception, allows capital losses to be used only against capital gains. (146) Losses tend to be present when diversification is most necessary. A portfolio of low volatility investments can be expected to rise slowly without dramatic gains or losses. Without the gains or losses, an initially diversified portfolio will not become imbalanced. The imbalances arise on a portfolio of volatile investments. A volatile portfolio, however, will generate losses as well as the extraordinary gains, leaving the taxpayer on net with a portfolio that appreciates by a more stable rate. When diversification is most necessary, in sum, there tends to be losses that arise to allow sales of the gains, without any tax.

Given the tax-free outlet for sales made to diversify investments, and the suppression of sales to reinvest by the pricing of the efficient market, it is fair to presume that where there is taxable capital gain, the sale was made to reinvest.

3. Life-Cycle Model

A different set of presumptions would arise from the life-cycle model of investing, which implies that early capital gain might be reinvested but that late in life capital gains are consumed. The life-cycle model arises from the general rule that one gets the most value out of dollars if one consumes at a smooth, steady rate. The rationality of steady consumption is a variety of the same diminishing marginal utility of dollars that supports a tax bracket system. Everyone spends for the most essential goods first, and then spends for ever less necessary items as more is spent. Spending according to a pattern of feast and famine means that an opportunity for better value is lost. The big consumptions in the feast (high-income) years would give more value for the same dollars if spent in the famine (low-income) years. A consumer rationally moves his money from high-income years for consumption in low-income years until the consumer has level consumption in all the years.

Over the course of a lifetime, the smoothing of consumption should lead to a predictable pattern called the life-cycle model. (147) One rationally borrows in early low-income years, to pay for tuition and buy a home. When an earner reaches years in which pay is better than the lifetime average, then the earner needs to pay off earlier debts and save for retirement. Both the early borrowing and the saving for retirement allow the earner to carry income from high-income years to lower income years and to maintain a steady consumption.

In national averages, peak years of earning for males with high capital gains are from the age of forty-five to fifty-four, and income drops below the average for one's entire lifetime by age sixty-five. (148) Rationally, once earnings drop below the lifetime average, one should be drawing down savings to maintain a steady level of consumption. Under the life cycle model, tax law might treat capital gains sales after sixty-five as sales made for consumption, and all sales made before sixty-five as sales within an overall pattern of increasing savings and hence presume the gains are reinvested. (149)

Post-sixty-five aged taxpayers are especially vulnerable to lock-in under current law because they are the ones for whom the step up in basis at death is nearest and most cognizable. Lock-in does not do any general economic harm when there are other actors in the market to bid for the better investments. Lock-in could also be avoided as to post-sixty-five taxpayers by tax law changes making death a constructive sale. If all built-in gains would be taxed in the final accounting, then the post-sixty-five year olds would realize that delaying a real sale was not much of a delay in impact of the tax.

There are other patterns within the overall life cycle pattern. An earner rationally saves to provide a cushion for unwelcome drops in income before retirement, for instance, to provide steady consumption if the earner is fired or disabled. An earner might also just quit a job to choose leisure. Within the smaller patterns, the tax system might treat gains recognized while unemployed as consumed gains. Any time a taxpayer's income drops below his working life average, we should presume that the gains are realized for the purpose of consumption. When income drops, then under a tax bracket system, the tax rate drops as well, but the ordinary income tax might still be higher than the 15% rate now applied to capital gains.

Empirically, earners do not act consistently with the rational behavior that the life-cycle model predicts. There is a strong "mail box" effect, under which checks in the mail increase consumption, even though they represent just a conversion of prior wealth from noncash to cash form. Under the life-cycle hypothesis, cashing in on wealth without a change in wealth should not increase consumption because the rational investor has already adjusted consumption to reach a smooth lifetime average in every period. Nonetheless, cash from a sale increases current consumption. (150) James Poterba has estimated that shareholders who received a large involuntary buyout spent between forty and fifty-nine cents of every dollar that they received for consumption. Almost all of that consumption was in the form of the purchase of consumer durables. (151) Consistently, Professors Campbell and Mankiw have estimated that between 40% and 50% percent of income is received by individuals who consume according to current income rather than lifetime average income. (152) Consumers will set aside a cushion for unexpected drops in income, but not enough to maintain their average standard of living into their retirement years. Consumers apparently consume more of their capital gains in early years than the lifecycle-cycle model would imply they should.

The rational life-cycle model might lead to a presumption for tax purposes that capital gains while employed are reinvested, unless the taxpayer drops in income, for instance, because of unemployment. Capital gains after retirement or after age 65 or when income drops would be presumed to be for consumption. If we adjust the life-cycle model to reflect real behavior which tends to be short sighted from a life-cycle perspective, then half of the gains realized early might also be treated as consumed, even though the life-cycle model says that it rationally should be reinvested.

Relying on economic patterns rather than the accounting for taxpayers' actual reinvestment will mean some injustice. Relying on the efficient market pricing, for instance, would say that all gains are for consumption, even when the facts of the taxpayer's own situation show investment elsewhere increasing by the realized gains. Relying on the life-cycle pattern would mean that retirees would pay ordinary income rates on their capital gains, even if it turns out in the specific case the retiree or employed taxpayer reinvested the gains. The presumptions might go the other way. Relying on the rational life-cycle model would also give a boon to younger sellers who would get away with a 15% capital gain tax, even when they are consuming bonbons we would want to tax at the highest rates.

The alternative to reliance on rough-justice presumption is to require and allow a taxpayer to show whether the gains are allocated to reinvestment or consumption, and, as explained next, that is hard to do.

B. Accounting for Consumption and Reinvestment

It might be possible to construct an accounting system to tax reinvested capital gains at preferential rates, but also tax consumed capital gains through the ordinary tax brackets. There are a number of accounting decisions that need to be settled and while each accounting decision is discrete, there are enough of them that the resulting system has considerable accounting complexity. Money, moreover, is fungible. That means the sales proceeds and other sources of cash should be viewed as going into a common pot and investment and consumption should be viewed as coming from a common pot, without regard to any tracing of dollars. A proper accounting should track the taxpayer's global investments. The following issues are only a start for separating consumed and reinvested capital gain. Even the bare bones framework indicates that the accounting will not be simple. (153) Readers convinced early that it is not feasible to determine whether gains arc reinvested may skip the remaining paragraphs of the description.

1. Reinvestment required by the transaction. Sometimes the transaction itself entails that the gains are reinvested. Property held under a trust that required that capital gains be returned to principal was the model case for the capital gains preference for both the British Parliament (strict settlements) and American Congress (Smietanka). Similarly, like-kind exchanges and reorganizations are transactions in which the gain given up in the exchange is reinvested in the property received by nature of the transaction.

2. Rollovers. Capital gain rates should be available if the proceeds of a cash sale are reinvested in new property. There need not be a direct exchange of new property for old. Section 1033, by analogy, currently allows nonrecognition for cash received in an "involuntary conversion," that is, a sale under eminent domain, or recoveries from a property tort or insurance policy, provided the cash received is reinvested in similar property. So a reinvestment of the proceeds of any sale should be considered reinvestment, which means that the gain would qualify as capital gain. If the proceeds of sale were not reinvested in a new investment, the gain would be ordinary income.

3. Partial Reinvestments. If only part of the proceeds of a sale are reinvested, the gain would be ordinary income to the extent of the cash withdrawn. The ordinary income would be the lesser of the cash not reinvested or the gain from the sale of the old property. (154) A single sale could generate both capital gain and ordinary gain. The "boot" rule, applied here, would tax gains as ordinary income to the extent that cash is withdrawn from investment.

Assume that taxpayer has basis in stock of $100 and gives up the stock in a merger for cash of $50 and stock of the acquiring company worth $250. The taxpayer has realized a total of $300 and gain of $200 and reinvested $150 of the gain, which would qualify as capital gain. Assume that the $50 cash is consumed and not reinvested elsewhere. There would thus be $50 ordinary gain from the reorganization. No more than $200 would be capital gain or ordinary gain, however. If taxpayer received $250 cash and stock worth $50 in the merger, the gain was only $200, all of which would be ordinary gain. The extra $50 cash received in excess of gain would be a recovery of basis or investment and would not be not taxed at all.

4. Borrowing. Borrowed cash and promises to pay for property are part of basis under the income tax, but borrowing needs to be subtracted from basis to calculate the savings or reinvested amount. Borrowing is the mirror image of investment--a kind of anti-investment or reduction of investment. If a taxpayer increased savings by $300 and also increased global borrowing by $300, the taxpayer has no net savings increase, and all the gain recognized in the year should be ordinary gain, thus if a taxpayer sells property for $300 and buys property with debt of $200, and a $100 cash down payment, the taxpayer has $200 of ordinary gain if there is at least $200 gain in the sold property.

Borrowing would reduce the amount considered to be invested, and that will mean that some gain will be treated as ordinary because borrowing reduced the amount considered to be reinvested. But borrowing alone presumably would not be treated as taxable income. Because banks charge interest, borrowing is not ordinarily an economic gain. Rather the borrowing would create a negative savings account that would absorb future investment and future gains, so that future gains might be treated as ordinary.

5. Negative savings account from borrowing. Repayment of borrowing would be treated as savings, except when it makes up a negative account created by borrowing. Repayments of borrowing in a negative savings account would not be treated as reinvestment, on the ground that in some prior period that there was borrowing that did not increase savings, i.e. consumption that needs to be taxed.

Suppose, for example, the taxpayer borrows $300, spends it on a vacation, and then repays the debt in the following year from the proceeds of a sale. Repayment of borrowing is kind of savings. It increases global net worth. But in this case, there was a savings of negative $300 when the vacation money was borrowed, and the repayment would reduce the negative but did not increase the savings accounting enough to treat the gain as reinvested. The vacation money would not be taxed when borrowed, but repayment of the vacation money loan would not qualify as a savings, so that the gains used to repay the vacation borrowing be treated as ordinary gain.

There is no need to trace whether the original borrowing was for vacation or investment. Any borrowed amount would reduce basis of savings, and borrowing would create a negative savings account only if borrowing exceeded basis.

6. Ignore untaxed gain. A taxpayer might well have an increase in savings because of appreciation on assets. But untaxed appreciation cannot count. There has to be an increase in saved and taxed cash, that is, in basis ignoring debt.

7. Credit only. Capital gain tax paid would generate a credit for tax, in the amount determined by capital gains rates, but not an increase in basis. It would be inappropriate to allow the low capital gains tax to create basis that reduces ordinary income. For example, assume a taxpayer with $100 basis in stock took stock in a merger worth $300. The $200 is reinvested and qualifies for the capital gains rate, so that there is $30 tax due at the current 15% capital gains rate. Now assume the taxpayer sells the acquiring corporation stock for $500 the following year, and does not replace the investment. Current law provides that recognition of gain increases basis to prevent double tax, (155) but a 15% tax on reinvestment should not replace a 35% tax on the withdrawal of $500 cash, as determined by the tax brackets according to the standard of living. The proper accounting is that the taxpayer has $300 of gain from the sale at $500, a tax of $105, under the assumed 35% tax bracket, and gets a credit for $30 of 15% capital gain tax paid earlier on the merger exchange.

The credit could be used against subsequent capital gain. Assume for example there is a subsequent merger. Assume the taxpayer starts with stock at basis of $100 and receives stock of $300 in the first merger. The taxpayer then gives up the merger stock for stock of yet another acquirer worth $500. The taxpayer would have capital gain of $400 in the second merger, but the capital gain tax paid on the first merger would be a credit to reduce the capital gain tax due on the second merger.

8. New gain before credit. If a taxpayer has both untaxed gain, and capital gain credit, then the first withdrawals would be ordinary income without use of the credit. Assume for example the two mergers above in which taxpayer started with $100 basis, realized $200 gain, all capital, in the first merger, and then in the second merger realized another $200 gain, but received $50 boot and $450 stock. The $50 boot would be ordinary gain (if it is not reinvested elsewhere) without use of the tax credit or basis. Only after all previously untaxed gain is taxed would the taxpayer be allowed some of the capital gain credit. A taxpayer who has some unrealized appreciation at the end of the year cannot use the credit for prior capital gain. The situation is analogous to the boot rule more generally: no basis may be recovered until gain is taxed in full, and the credit is like a basis rule used to prevent games with rate differences.

9. Consumer durables. Purchases of consumer durables like the principle or a second home, a car, a boat, or a refrigerator are not strictly an immediate consumption, but are rather a kind of saving that are withdrawn and consumed over the course of the life of the durable. Accounting for consumption can follow that pattern, and assign a depreciation schedule to determine how the durable should be assumed to be drawn down. Some arbitrary schedule needs to be used to preclude estimate controversies over how a durable is consumed in fact. The depreciation schedule can be fast and adverse to the taxpayer because consumer durables, if investment, are the kind of investment that does, say, invent new products or improve business productivity.

Depreciation of the durable good would be treated as if the taxpayer withdrew cash. A purchase of a second home for $5 million might be treated as consumption over twenty-five years so that $200,000 would be treated as gain withdrawn from investment. Tuition payments for children are, strictly speaking, gifts to children and not investments of any kind. If we treat the taxpayer and children as part of the same household or economic unit, however, then we can treat tuition as a long-term investment much like a consumer durable. Tuition might, for instance, be treated as giving out consumable values over the next fifty years after payment.

10. Bank deposits. Bank deposits can be treated as reinvestment. A taxpayer might well want to deposit money in a readily accessible account deciding whether to invest the money in stock or a second house. Deposit of gains into an ordinary bank or checking account can be treated as reinvestment, but then when the bank account shrinks, there is a withdrawal of previous capital gain that is both taxable as ordinary income and gets a credit for capital gain.

11. One big pool. Consumption or capital gain would be made from a global balance sheet approach. It should neither be necessary nor permissible to trace dollars received from a sale to a specific use. Dollars are fungible, and all dollars are in the same pool and perfect replacements for each other. A taxpayer has reinvested capital gain in economics only if the amount saved over all goes up by the full amount of the sales proceeds, even if the taxpayer can trace the use of the sale proceeds to some investment. Relying on tracing allows the taxpayer to manipulate the question of reinvestment, too easily, without actually increasing investment. Indeed, conversely, a taxpayer who thinks he has consumed his gains when we trace the dollars to some consumption item, but who ends up at year end with more than enough increase in savings to cover the gains realized has in fact reinvested his gains as a matter of economics no matter what his subjective understanding at the time, and should be treated as one whose gains were reinvested. It is illogical to try to tag some dollars when all dollars are the same.

Under a global approach, the taxpayer has only one big overall asset. After a global balance sheet system has been in effect for a while, the taxpayer has no idea of the basis of any given portfolio property because reinvested cash is not traced to any specific property, but treated just a increase in the one overall investment. If the taxpayer has any gain built into any asset, the cash withdrawn would ordinary gain, without use of the credit for capital gain, up to the extent of that gain. Cash withdrawn from the asset is taxed except when the global asset basis increases by the amount of the cash, disregarding borrowing. A taxpayer would have ordinary gain if there were any gain anywhere in the taxpayer's portfolio. The taxpayer would be able to use the credit for capital gain and then recovery of basis only after depleting all built-in gain across the system. If we go to a truly global perspective, the taxpayer must know the cost of all his investments and their value just to understand the taxation of any one sale.

It would be simpler just to give a 20% tax credit for all new basis upon investment. To say that a taxpayer will get a 15% tax rate rather than a 35% rate on gain provided the gain is reinvested is equivalent to saying that all gain is ordinary, but the tax payer gets the difference, now 20% as a credit against the ordinary income if savings are saved. The credit, however, is available only from old capital that already exists. People with a lot of capital are called rich. Why not extend the 20% credit to all those who increase their investment? Thus all gain would be ordinary, but all investment including reinvestment of newly recognized gain would get the benefit of the credit. The credit would not be given to investments treated as expenses such as intangible business development, research, and advertizing, and it would be in lieu of accelerated depreciation.

C. Ordinary on the Ground of Potential for Consumption

A final alternative that needs to be considered would be to tax capital gains at ordinary rates because they might be consumed. The global tracking of the taxpayer's total assets are complicated enough and reliance on the presumptions are rough enough that the benefits of determining whether capital gain is consumed or reinvested are not worth the effort.

When Great Britain ended the tax exclusion for capital gain in 1965, it did so, it said, to ensure that those who spend or consume capital gains should spend it out of a taxed fund, (156) and that people should be taxed according to their means and irrespective of the origin of the means, whether it be capital appreciation or income. (157) Capital gains are income, increasing the taxpayer's claim to resources, just like any other income, and the gains can be used as the taxpayer decides. In the 1965 British decision, capital gains were treated as income as much for the potential that they would be spent, meaning consumption, as for the actual use of the gains. If capital gains are consumable the ordinary tax brackets capture the appropriate level of tax for the taxpayer's standard of living.

Capital gains can be consumed, just like other income, even if they are not in fact consumed. Even the trusts--like the Smietanka trust and the strict settlement that were the models for the American and British decisions to give a lower rate to capital gain--were private arrangements. The arrangements were by the father or the grandfather of the beneficiaries, but the restriction when imposed was entirely private and voluntary. A trust can be drafted and often is drafted to give capital gains to income beneficiaries. (158) Now that capital gains are thought of as sources of current spending, we should expect more trusts to give their capital income, including capital gains, to the income beneficiary. The trust restrictions requiring that capital gain be returned to principal, are no longer part of a public purpose: It has after all been over 700 years since the reinvestment of capital gain was mandated by law as part of a military system to support knights' service.

The overall argument in favor of lower rates for capital gain even for reinvested amounts may not be strong enough to justify the effort of determining whether capital gain is consumed or reinvestment. Lock-in is not a public problem that needs to be remedied, as long as there is an adequate supply of capital from new savings and pension funds to keep the price at the right level to allocate capital according to fundamental values and to provide the equity capital for the better investments. Lock-in is not an equitable problem in a world in which competing investors pay ordinary income tax on their income before having it available for investing. If lock-in were a problem, ending the tax amnesty for capital gain to be consumed by wastrel heirs would be a better remedy for the problem. Incentives to capital need to be given to new capital and not to old capital of the old rich. Capital gain is inconsistent with front end incentives to capital and incentives for capital should be directed to the front end of investing and all investors, rather than just to existing capital. Repeal of the capital gain preference would strip the field and allow us to redesign better investment incentives from scratch.

The function of a lower rate on capital gain to nurture capital is better performed by other mechanisms. Once the capital gain preference is limited to gains that are reinvested, the lower rate can be thought of as equivalent to a credit for investment of realized gain. A system that gives taxpayers a 15% tax rate on reinvested capital gains, but a 35% tax rate on consumed or withdrawn capital gains, is strictly equivalent to a system that taxed all gain at 35%, but then gave a 20% tax credit to the act of investing. A rate cut contingent on reinvestment is a reward for reinvesting. In that format, however, it becomes obvious that the credit is limited to those who have existing capital. Capital gain arises as an appreciation on capital and those who have capital are called rich. The longer the capital is held, the more appreciation there will be, all other things being equal, so that people with capital gains are disproportionately old rich. Lower rates on capital gain are an incentive to old capital only. If the point is to encourage capital investment, why not extend the credit to new savings? Is not the money of the new saver as good as the money of the old?

As noted, capital gain gets in the way of incentives given to new savings and investment. The combination of expensing and capital gain acts as a negative tax or subsidy to an investment, and it favors inferior investments that could not pay the cost of capital. A tax system should not make inferior investments rational.

Given the dubious case for capital gains preference in the first place, taxing capital gains because they might be consumed would be a wise and simple remedy, avoiding both the accounting calculations of an individual's overall savings and the presumptions arising from what most people or a rational person would do.

* University of Texas Law School. The author wishes to thank Professors Mark Ascher. Charles Davenport, Mark Gergen, and Michael Sturley for helpful comments to a prior draft. If there is error that remains, they didn't do it.

(1) I.R.C. [section] 1222(3) now defines "long term capital gain" as "gain from the sale or exchange of a capital asset held for more than 1 year ." When first adopted the requisite holding period was two years. Revenue Act of 1921, Pub. L. No. 67-98, [section] 206(a)(6), 42 Stat. 227, 233. The requisite holding has been as short as six months. See Tax Reform Act of 1976. Pub. L. No. 94-455, [section]1402(a)(l), 90 Stat. 1520. 1731 (supplanting a six month holding period that had been in existence since 1942).

(2) Revenue Act of 1921. Pub. L. No. 67-98. 42 Stat. 227, 233. [section] 210 (4% normal tax), 42 Stat. 227. 235. [section] 211(a)(2) (50% surtax). 42 Stat. 227. 233, [section] 206(b) (12 1/2% tax on capital gain).

(3) I.R.C. [section] 1 (a)-(e) (35% maximum ordinary rate), [section] 1(h)(1)(C) (maximum 15% capital gains rate).

(4) 2 OXFORD ENGLISH DICTIONARY 863 (2d ed. 1989) (quoting TIMES TRADE & ENGIN. SUPPL. (London), Jan. 24, 1931, at 430).

(5) Merchants' Loan & Trust Co. v. Smietanka, 255 U.S. 509 (1921).

(6) Revenue Act of 1921, Pub. L. No. 67-98, 42 Stat. 227, 233, [section] 210 (4% normal tax), 42 Stat. 227, 235. [section] 211(a)(2) (50% surtax), 42 Stat. 227. 233, [section] 206(b) (12 1/2% tax on capital gain).

(7) S. REP. No. 67-275, at 13 (1921) (emphasis added).

(8) Samuel H. Williamson, Measuring Worth, Six Ways to Compute the Relative Value of a U.S. Dollar Amount, 1790 to Present (2008), (using the Consumer Price Index, 1917 to 2006).

(9) Brief and Argument on Behalf of Plaintiff in Error at 27, Merchants' Loan & Trust Co. v. Smietanka, 255 U.S. 509 (1921).

(10) Id. at 38 (citing, inter alia, In re Gerry, 103 N.Y. 445 (1886) (holding that bonds that increased in value from natural causes were capital not income when sold, where the will did not contemplate any traffic in securities by the trustee, but a permanent investment in interest-bearing obligations); In re Graham's Estate, 198 Pa. 216 (1901) (holding that increase in value of real estate held as an investment is principal, and goes to the remainderman); see also In re Mark's Estate, 38 Pa. D. & C. 489, 493 (Pa. Orph. 1940) (holding that no part of the gain in value of unproductive real estate was available to the income beneficiary because "[u]nder no guise is corpus ever depleted merely to aid necessitous life tenants") (quoting In re Levy, 34 Pa. D. & C. 312, 319 (1938) (Stearne, J., dissenting)).

(11) UNIF. PRINCIPAL AND INCOME ACT OF 1997 [section] 404(2) ( amended 2000), 7A U.L.A 497; cf. W.W. Allen, Rights of Life Tenant and Remaindermen Inter Se Respecting Increase, Gains, and Enhanced Values of the Estate, 76 A.L.R.2d 162 (1961).

(12) Smietanka, 255 U.S. at 515.

(13) 252 U.S. 189(1920).

(14) Id. at 207.

(15) 157 U.S. 429; 158 U.S. 601 (1895).

(16) The following paragraphs are drawn from Calvin Johnson, The Four Good Dissenters in Pollock, 32 J. SUPR. CT. HIST. 162 (2007) and Calvin Johnson, Fixing the Constitutional Absurdity of the Apportionment of Direct Tax, 21 CONST. COMMENT. 295 (2004), which give more details and sources.

(17) Hylton v. United Slates, 3 U.S. 171 (1796).

(18) McLouth v. Hunt, 154 N.Y. 179. 198 (1897). The New York courts would bother to apportion a stock dividend between earnings before contribution to the trust (which would go to capital) and earning after contribution (which would go to income beneficiaries) only in the case of an extraordinary stock dividend. In re Osborne, 209 N.Y. 450 (1913).

(19) See, e.g., Appeal of Earp, 28 Pa. 368. 374 (1857): Pritchelt v. Nashville Trust Co.. 96 Tenn. 472 (1896); EDWIN A. HOWES. AMERICAN LAW RELATING TO INCOME AND PRINCIPAL 29 (1905).

(20) Trefry v. Putnam, 227 Mass. 522, 535 (1917).

(21) EDWIN A. HOWES, supra note 19, at 220-21, 228-31 (describing Pennsylvania law that apportioned stock dividend between income and principal according to whether earnings of the corporation occurred before or after trust acquired the stock).

(22) See, e.g., Gibbons V. Mahon, 136 U.S. 549 (1890); De Koven v. Alsop, 205 III. 309, 314-16 (1903); Minot v. Paine, 99 Mass. 101, 107 (1868). See generally Edwin A. HOWES, supra note 19. at 26 (describing the strands of American law including the line that treated stock dividends as capital). Macomber was decided before Erie v. Tompkins, 304 U.S. 64 (1938), which mandated that the courts not look to brooding ominipresent law but to state law at issue to determine rights.

(23) There is no question that federal tax law can characterize state-law-created rights in a different way than state law characterizes them. A state cannot and should not be able to undercut federal collections of revenue with a tax advantageous characterization under local law. Still capital gains in tax was derived from the trust concept of capital gains, even in the Macomber decision itself, and New York's position on stock dividends gave income that income beneficiaries, including presumable the federal tax man, had claims on the stock dividends should have been sufficient for the government to win the case.

(24) Brief and Argument on Behalf of Plaintiff in Error at 26, Merchants' Loan & Trust Co. v. Smietanka. 255 U.S. 509 (1921).

(25) Id. at 52.

(26) Id. at 5.

(27) Id. at 14.52.

(28) Smietanka, 255 U.S. at 520.

(29) Brief and Argument on Behalt of Plaintiff in Error at 10, Merchants' Loan & Trust Co. v. Smietanka. 255 U.S. 509 (1921).

(30) HENRY SIMON, PERSONAL INCOME TAXATION 50 (1938) (famously defining income as "the algebraic sum of (1) market value of rights exercised in consumption and (2) the change in value in the store of property between the beginning and end of the period in question").

(31) Under UNIF. PRINCIPAL AND INCOME TAX ACT OF 1997 [section] 505(b) (amended 2000), 7A U.L.A 539, the capital gain tax reduces the principal interest.

(32) Revenue Act of 1921, Pub. L. No. 67-98, [section] 206, 42 Stat. 227. 232 (1921).

(33) Id. [section] 210 (4% normal tax), [section] 211 (54% surtax).

(34) See, e.g., Calvin H. Johnson, The Thor Power Tool Decision and Unrealized Inventory Losses, 11 TAX Notes 1259 (Dec. 29, 1980) (discussing unsuccessful proposals to reverse Thor Power Tool Co. v. Commissioner, 439 U.S. 522 (1979)); S. REP. NO.77-1631, at 78-79 (describing the successful overruling of Helvering v.Bruun, 309 U.S. 461 (1940), which taxed landlords on receipt of tenant improvements).

(35) Revenue Act of 1921: Hearing on H.R. 8245 Before the S. Comm. on Finance, 67th Cong. 537(1921).

(36) Id. at 535-36

(37) Id. at 544. Mr. Kellogg had also testified asking to follow English law in 1920. Id

(38) S. REP. No. 67-275, at 12-13 (1921).

(39) Revenue Act of 1921, Pub. L. No. 67-98, [section] 206(a)(6), 42 Stat. 227. 233 (2 year holding period requirement).

(40) 716 PARL. DEB., H.C. (5th ser.) (1965) 794.

(41) WILLIAM KENNEDY, ENGLISH TAXATION: 1640-1799, at 38 (1913).


(43) 34 PARL. HIST. ENG. (1798) 79-104.

(44) 34 PARL. HIST. ENG. (1799) 180-206.

(45) Attorney-General v. London Council, [1901] A.C. 26 (H.L), noted in R. Lachs, Income Tax on Capital Profits, 6 MOD. L. REV. 148 (1943).

(46) Scoble v. Secretary of State for India, [1903] A.C. 299 (H.L.). In Scoble, the Government of India purchased all of the stock of an Indian Railway from the taxpayers with an annuity, payable over the next fifty years. The statute at the time literally included "profits from annuities" within taxed income. The House of Lords, however, decided that profits realized by the sale were an installment purchase of "payments of money due as capital" and that the Act did not intend to tax capital. Id. at 621. The interest on the purchase price paid out over fifty years, as calculated by an actuary, was income, the Lords said, but the profits the shareholders made by constructing and selling the railroad were capital profits, even though called an annuity, and the capital gains were exempt from tax.

(47) 2 OXFORD ENGLISH DICTIONARY 863 (2d ed. 1989) (quoting TIMES TRADE & ENGIN. SUPPL. (London), Jan. 24, 1931, at 430).


(49) Id. LLOYD BONFIELD, MARRIAGE SETTLEMENTS, 1601-1740 (1983) argues that the strict settlement was a part of the marriage arrangements for the first son, but Habakkuk found many arrangements made when the eldest son reached twenty-one or at some other point.

(50) Scattergood v. Edge, (1976) 88 Eng. Rep. 1320.1326 ( K.B.).

(51) Id.

(52) HABAKKUK, supra note 48, at 4.

(53) Id. at 36.

(54) Id. at 361.

(55) Settled Land Act, 1882. 45 & 46 Vict., c. 38. [section] 3 (Eng.).

(56) Bruce v. Ailesbury, [1892] A.C. 361. (H.L.).

(57) HABAKKUK, supra note 48 at 18.

(58) Id. at 36. Estimates of the proportion of agricultural land held in strict settlement range from half to "[n]early all." Id. at 47.

(59) JANE AUSTEN, PRIDE AND PREJUDICE 27 (Dell Publishing 1959) (1813).

(60) See, e.g., id. at 29 (Mr. Bennett saying that, for the sake of his unmarried daughters, "I hope you will ... live to see many young men of four thousand a year come into the neighborhood.").

(61) Id. at 50.

(62) Walter Strachan. Capital and Income Under the Income Tax Acts, 29 L.Q. REV. 163, 174 (1913).


(64) Letter of Dr. Samuel Johnson to James Boswell (Feb. 3, 1776), reprinted in JAMES BOSWELL, THE LIFE OF SAMUEL JOHNSON 267 (William P. Nimmo & Co. 1882) (1791).


(66) Id. at *57.

(67) RICHARD FITZNIGEL, DIALOGUE OF THE EXCHEQUER 81 (Emilie Amt trans., Clarendon Press 2007) (1177) ("For the prince prefers to expose mercenaries, rather than his own people, to the hazards of war. This sum, because it is paid for each knight's shield, is called scutage.");C. Warren Hollister. The Significance of Scutage Rates in Eleventh- and Twelfth-Century England, 75 ENG. HIST. REV. 577, 579 (1960) (saying that in the century after the Norman Conquest, scutage tax was the cost of substitutes times the customary term of wartime military service); see also I.J. SANDERS, FEUDAL MILITARY SERVICE IN ENGLAND 92-93 (1956) (noting that by 1282 a noble could hire a substitute to satisfy a feudal muster).

(68) N. B. Lewis, The Last Medieval Summons of the English Feudal Levy, 13 June 1385, 73 ENG. HIST. REV. 1, 1 (1958).

(69) FREDERICK POLLOCK & FREDERIC MAITLAND, THE HISTORY OF ENGLISH LAW BEFORE THE TIME OF EDWARD I, at 337 (2d ed. 1899) (describing the Quia Emptores Terrarum statute).

(70) HABAKKUK, supra note 48, at 18

(71) Cochran v. Shelter, 133 A. 232, 234 (Pa. 1926); see also Joseph L. Weiner & James C. Bonbright, Theory of Anglo-American Dividend Law: Surplus and Profits, 30 COLUM. L. REV. 358 (1930) (noting the rule that dividends must keep capital intact). Giving dividends out of capital was also said to be misleading to investors because dividends are not a true sign of prosperity unless they come from profits. SEYMOUR D. THOMPSON & JOSEPH W. THOMPSON, 7 COMMENTARIES ON THE LAW OF CORPORATIONS [section] 5290 (3d ed. 1927). Control of the law of dividends was also said to be a tool to channel money into investment or consumption. Henry Ballantine & George S. Hills, Corporate Capital and Restrictions Upon Dividends Under Modern Corporation Laws; 23 Cal. L. Rev. 229. 229-30 (1935) (advocating loosening dividend restrictions to channel corporate funds away from over-investment and into consumption).

(72) [1892] 2 Ch. 198.

(73) Verner v. General Trust, [1894] 2 Ch. 239. 265-66 (citing Lubbock, [1892] 2 Ch. 198).

(74) I.R.C. [section] 316 (defining "dividend" taxed as ordinary income as distributions from earnings and profits of the corporation); Treas. Reg. [section] 1.312-6(b) (1960) (defining earnings and profits to include corporate capital gains). By contrast, corporate distributions out of capital of the corporation are, first, tax-exempt recoveries of capital to shareholders. t.R.C. [section] 301(c)(2). If the distribution exceeds the shareholder's basis, the distribution produces capital gain. I.R.C. [section] 301(c)(3).

(75) See, e.g., Life Tenant and Remainder II, 33 ALB. L. J. 424, 427 (1886) (arguing that if a company "declares what purports to be a dividend of its net profits, but which in fact is a dividend of money realized from the sale of some of its corporate property," then the gain should go to the remainder, otherwise the corporation "might seriously impair the value of [the] interest in remainder by reducing the amount of corporate property"); Ross G. Walker, The Base-Stock Principle in Income Accounting, 5 HARV. BUS. REV. 76, 85 (1936) (saying that a corporation should define its income to make a dividend only of that appropriately consumable after keeping "the basic income-producing estate intact").

(76) I.R.C. [section] 652(b). I.R.C. [section] 662(b) has almost the same language.


(78) REPORT OF THE ROYAL COMMISSION ON THE INCOME TAX 20, [paragraph] 90 (1920) (saying that "on the score of equity, practically nothing can be said for the present exemption of [nonrecurring business or trade] profits."). Gray v. Darlington, 82 U.S. 63 (1872) had decided that the U.S. Civil War income tax did not apply to realized capital gains because the gains were not "annual."

(79) REPORT OF THE ROYAL COMMISSION, supra note 78, at 20,[paragraph] 90.

(80) Royal Commission on the Taxation of Profits and Income 37 (1955).

(81) Id. at 30. The Commission also assumed it was talking about reinvested capital gains in its illustration of a taxpayer who realized gain but maintained the same income:
 Many of these gains on realisation occur to individuals or
 institutions for whom the act of realisation is merely a change of
 investment: the proceeds of sale are spent on acquiring a new source
 of income. Yet it by no means follows that a larger income in money
 terms results form a gain. To take the field of gilt-edged
 investments alone, appreciation or depreciation in value of the
 holding is mainly attributable to changes in the pure rate of
 interest occurring during the life of the security, omitting for the
 moment the more specialised question of maturing redemptions.
 Consequently, if appreciation takes place through a fall in the rate
 of interest, a reinvestment of sale proceeds has to be made upon the
 terms that the investor must accept a lower yield upon his money and
 he needs therefore all or most of his gain to maintain the same
 nominal income.

Id. at 30-31.

Of course if the taxpayer took his gains and consumed them immediately, there would be no further income, whether constant or not. Also, capital gains resulting from greater demand for the product of the capital asset should leave the seller able to command other kinds of commodities at the now relatively advantageous purchase price.

(82) 716 PARL. DEB., H.C. (5th ser.) (1965) 794-95; see also id. at 920 (saying that the "basic principle ... is that it means people will now be taxed on a basis according to their means and irrespective of the origin of those means, whether it be capital appreciation or income.").

(83) See, e.g, Jonathan A. Brown, THE LOCKED-IN PROBLEM, IN FEDERAL TAX POLICY FOR ECONOMIC GROWTH AND STABILITY: PAPERS SUBMITTED BY PANELISTS APPEARING BEFORE THE SUBCOMMITTEE ON TAX POLICY 367, 381 (Joint Comm. Print 1955); RICHARD GOODE, THE INDIVIDUAL INCOME TAX 197-208 (rev. ed. 1976); James Wetzler. Capital Gains and Losses, in COMPREHENSIVE INCOME TAXATION 115, 135-138 (Pechman, J. ed., 1977); Cynthia Blum, Rollover, An Alternative Treatment of Capital Gains, 41 TAX L. REV. 383, 387-388 (1986); Charles C. Holt & John P. Shelton, The Lock-In Effect of the Capital Gains Tax, 15 NAT'LTAX J. 337 (1962); Richard Schmalbeck, The Uneasy Case for a Lower Capital Gains Tax: Why Not the Second Best, 48 TAX NOTES 195, 200 (July 9, 1990); Joel Slemrod & Martin Feldstein, The Lock-In Effect of the Capital Gains Tax: Some Time Series Evidence, 1 Tax Notes 134 (Aug. 7, 1978).

(84) The following table shows how much better the return on the new investment must be than the return on the existing investment if the investor is going to sell the old asset to buy a new one. The required new return is stated as a multiple of the return on the old investment. The assumption is that the individual investor will hold a perpetuity until death, avoiding all capital gain tax. unless the return from a competing perpetuity is high enough to justify paying the capital gain. The tax on a sale depends upon tax rate and also upon what fraction of the sales proceeds is taxable gain, so that the required multiple varies with both tax rate and the traction of the sales proceeds which taxable gain represents.
Table 1: How much higher does the new return rate have to be to cover a
capital gain toll charge?

(Breakeven new return rate as multiple of old return rate)

Gain as a fraction of sale price 10% 30% 50% 70% 90%

Tax rate on gain

10% 1.01 1.03 1.05 1.075 1.10

15% 1.05 1.05 1.08 1.12 1.16

20% 1.02 1.06 1.11 1.16 1.22

35% 1.04 1.12 1.21 1.32 1.46

The multiple of the old rate of return that the new rate must meet is calculated as the ratio, g/I-t, where g is the gain as a fraction of the whole value, and t is the capital gain rate. The logic is from Richard Goode, supra note 83, at 200.

(85) See, e.g., Zoran Ivkovic, James Poterba & Scott Weisbenner, Tax Motivated Trading by Individual Investors. 95 AM. ECON. REV. 1605 (2005). Stock investors apparently display some behavioral economics in selling gains and avoiding admitting their losses, even though sale of gains is penalized and sale of losses is subsidized by tax. See also Benjamin Ayers, Craig Lefanowicz & John Robinson, Capital Gains Taxes and Acquisition Activity: Evidence of the Lock-In Effect, 24 CONTEMP. ACCT. Res. 315 (2007) (finding significant lock-in affect on acquisitions with high capital gains tax, notwithstanding availability of tax free mergers); Janet Meade, The Impact of Different Capital Gains Tax Regimes on Lock-In Effect and New Risky Investments Decisions, 65 Acct. Rev. 406 (1990) (finding lock-in in game simulations).

(86) Joseph Dodge, Restoring Preferential Capial Gain Treatment under a Flat Rate Income Tax, 44 TAX NOTES 1133. 1137 (Sept. 4. 1989); see also Walter Blum, A Handy Summary of the Capital Gains Arguments, 35 Taxes 247, 257 (1957) (characterizing the lock-in argument as a "formidable indictment").

(87) Capital Gains and Losses: Hearing Before the House Committee on Ways & Means, 93rd Cong. 276 (1973) (statement of Henry C. Wallich, Professor of Economics, Yale ) (asserting that "the capital gains tax reduces the supply of savings because the gains upon which it falls tend to be predominantly saved." but conceding that "[s]tatistical tests ... increasingly find evidence of spending out of capital gains").

(88) Alan J. Auerbach & Kenneth Hassett. Corporate Savings and Shareholder Consumption, in NATIONAL SAVINGS AND ECONOMIC PERFORMANCE 75 (B. Bernhein & J. Shoven eds., 1991) (arguing that consumption rationally should not increase just because shareholders have involuntary cash realization of their wealth).

(89) Donald Kiefer, Lock-In Effect Within a Simple Model of Corporate Stock Trading, 43 NAT.NAT'L TAX J. 75, 76 (1990) (excluding sales for consumption from the model).



(92) I.R.C. [section] 1014.

(93) Calvin H. Johnson. The Legitimacy of Basis from a Corporation's Own Stock, 9 AM. J. TAX POL'Y 155. 166-77 (1991) (inventories the cases in which basis is considered to be "initial value" to the taxpayer while being critical of the result in giving basis for untaxed capital).

(94) S. REP. NO. 67-275, at 10-11 (1921).

(95) I.R.C. [section][section] 362, 723, 1015.

(96) Taft v. Bowers, 278 U.S. 470,482 (1929).

(97) Walter Strachan, Capital and Income Under the Income Tax Acts, 29 L. Q. REV. 163, 174 (1913) (arguing that capital in the concrete sense refers to a physical thing).

(98) Lock-in would also disappear if the tax law taxed gains as they happened economically without waiting for a sale. Nontax financing accounting has now moved over to taking account of gains on marketable stock as the stock increases in value, without need for a sale. FINANCIAL ACCOUNTING STANDARDS BOARD, STATEMENT OF FINANCIAL STANDARDS NO. 115: ACCOUNTING FOR CERTAIN INVESTMENTS IN DEBT AND EQUITY SECURITIES (1993).

(99) Jane C. Gravelle, Limit to Capital Gains Feedback Effects, at 4-6 (Cong. Res. Serv., Rep. No. 91-250RCO, 1991) (finding that 30% of accrued gains are realized, but excluding timber, housing and debt, to find that 46% of accrued gains are taxed and 54% excluded). An earlier work, Jane G. Gravelle & Lawrence B. Lindsey, Capital Gain, 38 Tax Notes 397, 400 (Jan. 25, 1988), found that 76% of capital gains are held until death. Kotlikoff and Summers argue amounts once invested are not commonly disinvested, but are largely held until death. Only 20% of individual wealth is consumed by the household later in life, they find, and that 80% of wealth is transferred to the next generation. Laurence Kotlikoff, Intergenerational Transfers and Savings, 2 J. ECON. PERSP. 41, 43 (1988); Laurence Kotlikoff & Lawrence Summers, The Role of Iniergenerational Transfers in Aggregate Capital Accumulations, 89 J. POL. ECON. 706 (1981). The estimate is controversial, although possible. For a sample of the debate, see Alan Blinder, Comments on Chapter 1 & Chapter 2, in MODELLING THE ACCUMULATION AND DISTRIBUTION OF WEALTH 68 (Denis Kessler & Andre Masson eds., 1988); Michael Hurd & Gabriela Mundaca, The Importance of Gifts and Inheritances Among the Affluent, in THE MEASUREMENT OF SAVING, INVESTMENT AND WEALTH 736 (NBER Studies in Income and Wealth vol. 52, Robert E. Lipsey & Helen Stone Tice. eds., 1989) [hereinafter THE MEASUREMENT OF SAVING]; Denis Kessler, Comment, in THE MEASUREMENT OF SAVING, supra, at 758; Denis Kessler & Andre Masson, Bequests and Wealth Accumulation: Are Some Pieces of the Puzzle Missing?, 3 J. ECON. PERSP. 141 (1989); Franco Modigliani, The Role of Intergenerational Transfers and Life Cycle Saving in Accumulation of Wealth, 2 J. Econ. PERSP. 15 (1988). Kotlikoff and Summers do not discriminate between high-basis and low-basis wealth. If 80% of all wealth is held until death, then we should expect that wealth to be especially rich in wealth with unrealized gain, given the incentives to hold high gain property and rely on loss or low-gain property for standard of living.

(100) Alan Auerbach, Capital Gains Taxation and Tax Reform, 42 Nat'l Tax J. 391, 391 (1989).

(101) E. Auten & Joseph J. Cordes, Policy Watch: Cutting Capital Gains Taxes, 5 J. ECON. PERSP. 181, 184 (1991) (arguing that for gain to be significantly higher in the long run, additional realizations would have to come primarily from gains otherwise not taxed at death).

(102) Gravelle, supra note 99, at 3 (also noting that in the long term realizations cannot exceed accruals and thus criticizing high elasticities that implied realizations exceeding gross national product).


(104) See, e.g., RICHARD GOODE, THE INDIVIDUAL INCOME TAX 234-38 (1964); cf. Stanley S. Surrey, Definitional Problems in Capital Gains Taxation, 69 HARV. L. Rev. 999, 1018 (1956) (recommending averaging over three years as capturing most of the problems and simpler). But see Neil H. Buchanan, Case Against Income Averaging, 25 VA. TAX REV. 1151 (2006) (arguing that income averaging is not an equitable base line).

(105) If, for example, the gain accrued thirty years ago and the ordinary tax would have been $10,000, then deferring the tax is like paying $2,300 tax, dropping the real impact of the tax to a quarter of the nominal rate, assuming a 5% discount rate, because $2,300 will grow large enough to pay the $10,000 tax by the time it is due. Calvin Johnson. The Undertaxation of Holding Gains, 55 TAX NOTES 807 (May 11, 1992) (calculating dropping effective tax rate on appreciation as the property is held).

(106) Id. at 815 (graphing the effective rate on gains with inflation and deferral of tax).

(107) Id.

(108) See. e.g., Savings and Investment Provisions in the Administration's Fiscal 1998 Budget Proposal: Hearing Before the H. Comm. on Ways and Means. 105th Cong. 149 (1997) (statement of James F. Higgins. Chairman, Securities Industry Association); Stephen Moore & Phil Kerpen, A CAPITAL GAINS TAX CUT: THE KEY TO ECONOMIC RECOVERY (2001). available at IPI%20-%20CapGainsKey.pdf.

(109) R. Glenn Hubbard. A Paradox of Interest, WALL STREET J., June 23, 2005, at A12; Paul Krugman, Don't Cry for Me, America, N.Y. TIMES, Jan. 18. 2008, at A21; Ben S. Bernanke, Governor, Fed. Reserve, The Global Saving Glut and the U.S. Current Account, Remarks at the Sandridge Lecture, Virginia Association of Economics. (Mar. 10, 2005). available at

(110) See, e.g., Paul Krugman, Dealing With the Dragon, N.Y. TIMES, Jan. 4, 2008, at A19 (discussing oil prices hitting $100 a barrel because of sluggish supply); Noureddine Krichene, Recent Dynamics of Crude Oil Prices, (Int'l Monetary Fund Working Paper No. 06/299, 2007) (citing limited supply and higher demand as reasons for price increases).

(111) $123,000*[(l+4%).sup.15] = $250,000.

(112) $74,000*[(1+7%).sup.15] = $250,000.

(113) See, e.g., A. Lars Bovenberg, Tax Policy and National Saving in the United States: A Survey, 42 NAT'L TAX J. 123, 128 (1989) (reviewing literature that found high interest rates decrease savings).

(114) See E. Philip Howrey & Saul Hymans, The Measurement and Determination of Loanable-Funds Savings, in WHAT SHOULD BE TAXED: INCOME OR EXPENDITURE? 1, 30-31 (J. Pechman ed., 1980) (finding no response to increased interest); Robert Hall, Intertemporal Substitution in Consumption, 96 J.POL. ECON. 339 (1988) (finding no savings response to increased interest returns and explaining away apparent findings that savings respond to increased interest); Barry Bosworth, Gary Burtless & John Sabelhaus, The Decline in Saving: Evidence from Household Surveys, in BROOKINGS PAPERS ON ECONOMIC ACTIVITY 183 (vol. 1 1991); Jonathan Skinner & Daniel Feenberg. The Impact of the 1986 Tax Reform Act on Personal Savings 17 (Nat'l Bureau of Econ. Research. Working Paper No. 3257. 1990) (describing a consensus in the economic literature that any positive response of savings to interest rate increase is "fragile and fleeting").

(115) Barry Bosworth & Gary Burtless, Effects of Tax Reform on Labor Supply, Investment, and Saving, 6 J. ECON. PERSP. 3,4 (1992).

(116) See, e.g., Laurence Kotlikoff. The Crisis in U.S. Saving and Proposals to Address the Crisis, 43 NAT'L TAX J. 233, 239 (1990) (arguing that capital gain cut is poorly designed because of windfall benefit to those who accrue gains before the act).

(117) See, e.g., William Andrews, A Consumption-Type or Cash Flow Personal Income Tax, 87 HARV. L. REV. 1113 (1974); Joseph Bankman & David Weisbach, The Superiority of an Ideal Consumption Tax Over an Ideal Income Tax, 58 STAN. L. REV. 1413(2006).

(118) Calvin Johnson, Consumption vs. Wage Tax. 67 TAX NOTES 1382 (June 5, 1995) (arguing that if Jacob receives the blessing of Isaac and prospers, and Esau does not and loses all, the tax collected from each should be different - both Esau and Jacob had the same expected amount and investment at the Start, but their outcome was quite different).

(119) See. e.g., Alvin C. Warren, Jr., How Much Capital Income Taxed Under an Income Tax is Exempt Under a Cash Flow Tax?. 52 TAX L. REV. 1 (1996).

(120) David Elkins and Christopher Hanna. Taxation of Supernormal Returns. 62 TAX LAWYER (forthcoming 2009). Stated algebraically, if a taxpayer pays tax at the front end. but none at the back end. the terminal value of the investment after tax is $l00 *(l-t) *(l+R)^n where $100 is a unit of income, t is tax rate, R is the extraordinary tax rate and n is the number of years of compound growth. If a taxpayer pays tax at the back end, but not the front end, the terminal value after tax is $100*(l+R)^n*(l-t). Since the order of multiplied terms does not matter, the terminal values are the same, if t is constant. But the government's terminal value of t collected at the start is $100*t*(l+i)^n where i is the government's low investment rate, and the government collects $/00*(l+R)^n*t at the end. Since R>i, the government's position is better with the back end cash flow tax, even though the taxpayer's position is the same in both. The difference can be explained that the tax at the front end does not participate in R, but the tax at the back end does.

(121) Assume for example an investment that gives, pretax, a 10% internal rate of return, with the following cash flows in the specified year:
Pretax Yr 0 Yr 1 Yr 2 Yr 3

Cash flow ($10) ($10) ($10) $34.41

The cash flows have a 10% internal rate of return ("i") because the S10 amountswill grow to equal $34.41 under the formula for the terminal value of annuities where i is 10%: $10*[[(l+i).sup.3]-l)/i]=$34.41.

Assume that the $10 inputs are deductible when made at a 35% tax rate and the $36.41 is capital gain subject to 15% tax. The after-tax cash flows are as follows:
After tax Yr 0 Yr 1 Yr 2 Yr 3

Cash flow ($6.50) ($6.50) ($6.50) $30.95

The after-tax cash flows shown above have a 25% internal rate of return ("i") because the $6.50 amounts will grow to equal $30.95 under the formula for the terminal value of annuities where i is 10%: $6.50*[[(1+i).sup.3]-l)/i]=$30.95. The example is from Calvin H. Johnson. Sale of Goodwill and Other Intangibles as Ordinary Income, 118 TAX NOTES 321, 323 (Jan. 14.2008).

(122) See, e.g., I.R.C. [section][section] 174 (allowing expensing of research and development), 263(c) (allowing expensing of oil drilling); Rev. Proc. 69-21, 1969-2 C.B. 303 (allowing expensing of cost of developing software).

(123) I.R.C [section] l(h)(11).

(124) See Calvin Johnson, A Thermometer for the Tax System: The Overall Health of the Tax System as Measured by Implicit Tax, 56 SMUL. REV.13 (2003).

(125) $100 - $100*10% = $90, $90*15% = $13.50, and $90 - $13.50 = $76.50.

(126) See Calvin Johnson, The Bush 35 Percent Flat Tax on Distributions from Public Corporations, 98 TAX NOTES 1881 (Mar. 24, 2003) (proposing a precompte or 35% prepayment of corporate tax if a corporation distributes dividends without having paid 35% tax on the amount distributed).

(127) Treas. Reg. [section] 1.10021(c) (1960).

(128) H.R. REP. No. 73-704 (1934), reprinted in 1939-1 pt. 2 C.B. 554, 564.

(129) Treas. Reg. [section] 1.368-1 (b) (2007).

(130) I.R.C. [section][section] 356. 1031(b). 1033(a)(2).

(131) In 1934 Congress continued the exemption for like-kind exchanges to prevent taxation of "thousands of horse trades and similar barter transactions." H.R. REP. NO. 73-704 (1934), reprinted in 1939-1 pt. 2 C.B. 554, 564, Like-kind exchanges today rarely fit the model of informal barters. There might be allowed a continuation of the exemption for like-kind exchanges for property not listed on a regular market, worth less than $20,000 in the offers to sell in classified ads or similar on-line services, where no cash, debt assumptions or purchases for the purpose of exchange take place. Those are the informal barters in which doubts about valuation might well make the taxation not worth the effort.

(132) I.R.C. [section] l(a)-(d), (f)(l)-(2), (i). Because the dividing line between brackets is adjusted for inflation annually, the actual brackets are not available in the Code, but are published annually. See, e.g.. Rev. Proc. 2006-53, [section] 3.01, 2006-2 C.B. 996 (publishing brackets for taxable years beginning in 2007).

(133) For 2007, the 15% bracket applies to taxable income between $15,650 and $63,700. Rev. Proc. 2006-53, [section] 3.01. 2006-2 C.B. 996. The median family income in 2005 was $55,832 which, adjusted by two years of inflation, would be over $60,000. U.S. CENSUS BUREAU, 2008 STATISTICAL ABSTRACT OF THE UNITED STATES 455 tbl.685 (2007), available at

(134) See, e.g., Burton Malkiel, Is the Stock Market Efficient?, 243 SCIENCE 1313, 1317 (1989): see also Eugene Fama, Efficient Capital Markets: A Review of Theory and Empirical Work, 25 J. FIN. 383 (1970); Christopher Paul Saari, Note, The Efficient Capital Market Hypothesis, Economic Theory and the Regulation of the Securities Industry, 29 STAN L. REV. 1031 (1977). Measurements of the efficiency of the market simultaneously test the pricing of stock according to discounted value of cash, that is, stock prices, under the efficient market thesis, reflect what is known publicly about future distributions, and appropriate discount rates. See, e.g., GEORGE FOSTER. FINANCIAL STATEMENT ANALYSIS 363 (1978).

(135) Michael C. Jensen, Some Anomalous Evidence Regarding Market Efficiency, 6 J. FIN. ECON. 95, 95 (1978); see also id. at 96 ("In the literature of finance, accounting, and the economics of uncertainty, the Efficient Market Hypothesis is accepted as a fact of life, and a scholar who purports to model behavior in a manner which violates it faces a difficult task of justification."). The evidence that the stock markets are efficient includes evidence that future prices of stocks are independent of past prices, that prices adjust rapidly to newly disclosed information, and sometimes anticipate (or incorporate leaked) information, that changes in accounting methods do not affect stock prices, and that market professionals cannot consistently perform better, after research expenses, than randomly selected diversified portfolios do.

(136) See Robert E. Verrecchia, Consensus Beliefs, Information Acquisition and Market Information Efficiency, 70 AM. ECON. REV. 874 (1980) (arguing that the weighted average reflected in the market will be better than any of the participants' estimates about the future).

(137) Kenneth Froot & Andre Perold, New Trading Practices and Short-Run Market Efficiency (Nat'l Bureau of Econ. Research, Working Paper No. 3498, 1990).

(138) The absence of a negative correlation between fifteen-minute price changes indicates the market was also not over-reacting and then correcting itself. Id. at 1.

(139) Ronald Gilson & Reinier Kraakman, The Mechanisms of Market Efficiency, 70 Va. L. REV. 549, 610 (1984).

(140) Hedge funds first arose with sophisticated statistical tools to identify price anomalies and then were unable to find significant profits with their tools. The application of the sophisticated statistical tools and the diminution of the pure profits from risk free trades both contributed to the smartness of the market and are evidence for it. One scholarly assessment of the hedge funds finds that they at least have done no measurable harm. William Fung & David A. Hsieh, Measuring the Market Impact of Hedge Funds, 7 J. EMPIRICAL FIN. 1 (2000) (finding little or no evidence that hedge funds caused prices to diverge from market fundamentals).

(141) See Frank Easterbrook & Daniel Fischel, The Proper Role of a Target's Management in Responding to a Tender Offer, 94 HARV. L. REV. 1161, 1166-67 (1981); Sanford Grossman & Joseph Stiglitz, On the Impossibility of Informationally Efficient Markets, 70 Am. ECON. REV. 393 (1980) (arguing that the markets reach not perfect knowledge, but an "equilibrium degree of disequilibrium").

(142) For an interesting attempt to assess the excess volatility of market prices not explained by market fundamentals or new information, see Allen Ferrell, If We Understand the Mechanisms, Why Don't We Understand the Output?, (Harvard John M. Olin Discussion Paper Series No. 414, 2003), available at

(143) Ronald Gilson & Reinier Kraakman, MOME in Hindsight, 26 REG. 64, 70 (2004), (expressing pessimism that the market can overcome the corporate frauds, such as those in the Enron cohort. An individual investor would react to the frauds with a generalized paranoia about all corporate stock, but would not be able to identify that it is time to sell a specific stock to invest in some other stock).

(144) The seminal theoretical work is Harry Markowitz, Portfolio Selection, 7 J. FIN. 77 (1952); see also Harry Markowitz, Portfolio Selection: Efficient Diversification of Investments (1959).

(145) See, e.g., LESTER THUROW, DANGEROUS CURRENTS: THE STATE OF ECONOMICS 152 (1983) (arguing that price reflects a wide distribution of possibilities with a long tail such that, for instance, investors who owned IBM stock in the 1930s would be very wealthy by 1983).

(146) There is a "dribble out" rule under which $3000 of capital losses, not offset by capital gains, may be deducted every year. I.R.C. [section]1211. If the losses are not exclusively usable against gains, then the gains will produce tax upon realization. Still, the $3000 allowance is not a material fraction of large losses.

(147) Franco Modigliani, The Life-Cycle Hypothesis of Saving, the Demand for Wealth and the Supply of Capital, 33 SOC. RES. 160 (1966); see also Alan Auerbach & Kenneth Hassett, Corporate Savings and Shareholder Consumption, in National Savings and Economic Performance 75 (B. Douglas Bernheim & John B. Shoven eds., 1991) (unexpected spike in cash received is not a rational reason to increase consumption, by more than life time average); Robert Merton, In Honor of Franco Modigliani, 1 J. Econ. Persp. 145,147-49 (1987) (applauding the elegance of the Life Cycle theory on the occasion of Modigliani's receipt of the Nobel Prize in Economics).

(148) U.S. Census Bureau, 2008 Statistical Abstract, supra note 133, at 146 tbl.220.

(149) The author will be sixty-five years old on his next birthday.

(150) See Bosworth et at., supra note 114, at 226 (summarizing the research); Marjorie Flavin, The Adjustment of Consumption to Changing Expectations About Future Income, 89 J. OF Pol. ECON. 974, 1006 (1981); Jonathan Skinner, Risky Income, Life Cycle Consumption and Precautionary Savings, 22 J. Monetary ECON. 237 (1988) (saving against income drops explains preponderance of savings); Chris Carroll & Lawrence H. Summers, Consumption Growth Parallels Income Growth: Some New Evidence (Nat'l Bureau of Econ. Research, Working Paper No. 3090. 1989) (data not consistent with lifetime averaging).

(151) James Poterba, Dividends, Capital Gain and the Corporate Veil: Evidence from Britain, Canada and the United States, in National Savings and Economic Performance 49, 63 (B. Douglas Bernheim & John B. Shoven eds., 1991).

(152) John Y. Campbell & N. Gregory Mankiw, Permanent Income, Current Income and Consumption (Nat'l Bureau of Econ. Research, Working Paper No. 2436, 1987).

(153) I am, candidly, not sure that the system would work.

(154) By analogy, under the "boot rule" for nonrecognition exchanges, gain is recognized to the extent of cash withdrawn from the transaction. I.R.C. [section] [section] 358(a)(l)(B)(ii), 1031(b).

(155) I.R.C. [section] [section] 358(a)(l)(B)(ii), 1031(d).

(156) 716 PARL. DEB., H.C. (5th Ser.) (1965) 795; see also text accompanying supra note 75.

(157) Id. at 920.

(158) See, e.g., In re Gardiner's Estate, 425 P.2d 427, 5 Ariz. App. 239, (Ariz. Ct. App. 1967) (finding that income beneficiaries were entitled to capital gain, under a trust given to give trustee administrative powers to allocate income).
COPYRIGHT 2009 Virginia Tax Review
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Johnson, Calvin H.
Publication:Virginia Tax Review
Date:Jan 1, 2009
Previous Article:The corporation as administrative agency: tax expenditures and institutional design.
Next Article:On discounted partnership interests and adequate consideration.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters