Capital Gains and Losses

• Author: Vern Krishna
• Profession: Professor of Common Law, University of Ottawa Barrister at Law
• Pages: 257-296

257

CHA PTER 9

capITaL gaInS

andLoSSeS

A. gENER AL COMMENT

Capital gains are a sepa rate and distinct source of income. A taxpayer

must include 50 percent of capital gains in income and may deduct there-

from 50 percent of capital losses.1 There is no separate tax on capita l gains.

Capital gains and losses merely expand or contract the tax able base to

which we apply the normal tax rates. Thus, capital gains and losses are a

separate source of realized i ncome subject to lower effective tax rates.

The “preference” for capital gains adds enormous complexity to

the Act. Quite apart from the underst andable preference of taxpayers

to characterize i ncome as capital gains, the Act furt her complicates

the structure by creat ing numerous subcategories of capital properties,

each of which has specia l rules and exemptions.

Our tax law ha s always “preferred” capital gains over other forms

of income. The capital gains preference debate is now long in the tooth,

and we are no closer to resolving the i ssue today than we were when

capital gains were f‌irst e xcluded from income.

Why should we treat capital gains preferential ly? First, there is an

intuitive notion that the appreciation of capital i s not what we normal ly

consider as “income.” There is a clear split between the courts and

policymakers on this issue. For example, in 1872, long before the de-

velopment of the Haig-Simons formulation of income, the United States

1 Income Tax Act, RSC 1985, c 1 (5th Supp) [ITA], s 38.

Income Tax L aw258

Supreme Court stated that “the mere fact that property has advanced

in value between the date of its acquisit ion and sale does not authorize

the imposition of the tax on the amount of the advance. Mere advance

in value in no sense constitutes t he gains, prof‌its, or income . . . . It

constitutes and can be t reated merely as increase of capital.”2

Similarly, in 1923, United States Treasury Secretary, Andrew Mellon,

said he “believed it would be sounder taxation policy generally not to

recognize either capital gain or capital loss for purposes of income tax.”

The Carter Commission (1967) clearly favoured the theory that “a

buck is a buck” regardless of its source. In Can ada today one can say

that there are no clearly def‌ined objectives of capital gains taxat ion, and

that our policies, both of tax ation and preference, are an amalgam of

intuition, politica l ideology, and compet itive economic considerations.3

A capital gain derives f rom an increase in the capital value of an asset.

This raises t wo distinct issues: (1) when should we tax the incre ase? and

(2) how much of the increase should we tax? An increase in the value of

capital ref‌lects the increase in the discounted future cash income from

the underlying investment. In t he case of stocks, for example, the incre-

ment in stock values is either the enh ancement in anticipated future cash

f‌lows or a reduction in the discount factor. Since we tax dividends from

earnings, we must al so tax capital gains to the extent th at they represent

undistributed ear nings. However, although there are some similarities

between dividends and capital gains, there are also sig nif‌icant differen-

ces with respect to timing, bunching and inf‌l ation.

Second, we can justify t he capital gains preference as relief from the

“bunching” of accrued property appreciation. Capital gains usually result

from appreciation that builds up over an extended per iod. The gain one

realizes usually extends beyond one f‌iscal year. The realization principle

of income recognition is a rule of admini strative convenience. For ex-

ample, if one buys shares in Year 1 for $20 and sells the shares in Year 5 for

$120, the realized gain of $100 ref‌lects the unrealized accr ual of gains over

f‌ive years. Since the personal t ax rate structure is progre ssive,4 we would

penalize t he investor with a higher marginal ta x rate if we taxed the entire

gain in his income in Year 5. This might be unfa ir to the investor.

Taxing only one-half of a gain is a rough-and-ready (palm tree)

way of mitigating the effect of progressive rates on “bunched” income.

2 Gray v Darlingt on, 82 US 63 at 66 (1872). This position was a ltered by the pas-

sage of the 16th Amendme nt in 1913.

3 See Stanley Sur rey, “Def‌initional Problems in Capit al Gain Taxation” (1956) 69

Harv L Rev 985; Walter Blum, “A Handy Sum mary of the Capital Ga ins Argu-

ments” (1957) 35 Taxes 247.

4 See Chapter 1.

Capital Gai ns and Losses 259

The United States Supreme Court recognized thi s in Burnet v Har mel,5

where it held that the purpose of the capita l gains preference was to

“relieve the taxpayer from . . . excessive tax burdens on gains result ing

from a conversion of capital investments, and to remove the deterrent

effect of those burdens on such conversions.”

Of course, bunching has no effect at all on an investor whose mar-

ginal rate of tax in each of the f‌ive preceding years would have been

in the top tax bracket even without the capital ga in. Such an investor

derives a windfa ll from the preference. Also, one can always relieve

against the bunching effect by allowing the investor to spread back the

tax on his entire ga in at the average rate that would have applied had

the investor accrued the gain an nually over the f‌ive-year period. This

is a complicated averaging mechani sm to the bunching problem. In the

end, averaging may, accidentally, yield the same approximate result as

simply reducing the overall effective rate to 50 percent.

Third, the preference mitigates the “lock-in” effect of the realization

rule, and makes it less costly for investors to switch investments when

it is in their economic interest to do so. The “lock-in” effect stems from

the realizat ion principle. In the above example, the investor with an

unrealized ga in of $100 in Year 5 may identify a better economic invest-

ment with a higher potential y ield. If she sells the initial investment,

however, the investor will trigger ta x of, say, 40 percent, which would

leave her with only $60 to reinvest. This effectively reduces the net rate

of return on the new investment, and makes it less desirable. Thus, the

investor might choose not to sell the origin al investment and defer the

tax that would otherw ise be payable. Usually, she could defer the tax

until the later of her or her spouse’s death.6 If the investor can defer the

tax for 30 years, and the interest rate is 8 percent, the future value of

the $40 tax that would be payable if she sells today is $403.7 Thus, the

investor can multiply her tax sav ing ten times simply by not selling and

locking herself into the original investment.

Lock-in restricts t he mobility of c apital a nd reduces its eff‌iciency. To

liquidate a poorly performing invest ment and reinvest in another ven-

ture, the return on the new invest ment must be suff‌icient to pay for the

capital gains tax bite on the old investment. The investor will liquidate

the initial investment only if the return on the new opportunity is suf-

f‌iciently higher to offset the tax bite on the old as set. Otherwise, it is

better to lock in to the initia l investment. The capital gains preference

5 2 87 US 103 at 106 (1932).

6 ITA , above note 1, subss 70(5) & (6).

7 Future value = $40 × 10.063 (see Append ix A).