1 The following calculation shows this : In an income of 97,000 the first
100 is free ; the next 900 will be taxed at 3 per cent, or 27, leaving a
balance of 973 out of the 1000 ; of the second 1000, taxed at 4 per cent.,
the owner will have ^960 ; of the third, ^950 ; and so on, decreasing by 10
for each additional 1000. The maximum income will therefore be .973 plus
the sum of the series

(10+20 + 30 + 40 960) = 973 + 10 (i . 2 . 3 . 4 96) :

973 + I0 - ^ = 973 + 46560 - 47533-
Again, take that mentioned by M. Leroy Beaulieu, in
which the tax triples as the revenue doubles, the rate at
starting being i per cent., and the lowest grade up to 500
francs. This will work out as shown in the following
table 1 :-

Income in Francs. Amount of Tax. Rate per cent.

500 5 4 i

1,000 15 1-50

2,000 45 2-25

4,000 135 3.375

8,000 405 5-06

16,000 1,215 7 - 6

32,000 3,645 1 1. 4

64,000 10,935 17-1

128,000 325805 25-6

256,000 98,4! 5 38-4

512,000 295,245 57-6

1,024,000 885,735 86-5

2,048,000 2,657,205 129.7

In both cases, after a certain point is reached, the total
additional revenue is taken by the tax-collector, and at a
much lower one the rate becomes oppressively high. Actual
examples of progression are, as will appear, much milder.
The highest rate of charge is fixed at a comparatively low
percentage. The fact, however, remains that there is no
self-acting principle by which to determine the scale of
progression. We must perforce agree with M. Say when
he declares that ' progression is naturally arbitrary 2 .' Op-
ponents of the system will naturally hold that the most
moderate form is the least evil, and try to attain that result
(unless they prefer to have an extreme measure in the hope
that its hardships may cause a reaction). Reasonable sup-
porters will recognise that a rapidly increasing rate is both
unjust and economically injurious. But beyond such vague
propositions nothing can be stated. All depends on the
will of the Legislature, i. e. in most modern societies on the
votes of persons who will not directly feel the charges
placed on the higher incomes, and will probably believe
that they will be gainers by them.

(2) A serious obstacle is the danger of evasion. \No
empirical law is better established in Finance than that
which states that high taxation leads to efforts to avoid it.
Duties on luxuries are in part escaped by the smuggler's
aid ; special duties on the better kinds of goods lead to
false declarations ; progressive income and property taxes
cause false returns on the part of the contributors. For
this latter fact there are several reasons. The increased
charge on higher incomes offers a special inducement to
understatement on the part of those liable, as thereby they
obtain a lower rate, a proceeding the more readily excused
to their consciences by the plea that the exaction escaped
is itself unjust. Another reason is the impossibility of em-
ploying effective measures for collection. With a uniform
income tax a great deal of income can be taken at its
source where evasion is impossible ; with progression, as
the rate varies according to the sum of income, the ascer-
tainment of that fact is required for fixing the charge,
though it is undoubtedly very difficult to get a proper
answer to inquiries respecting it. Thus the motives for
evasion would be stronger, and the means of prevention
less effective in the case of a progressive than of a propor-
tional tax. It is the intrusion of the personal and arbitrary
element that raises this difficulty which is accordingly
unavoidable.

(3) A powerful argument against progressive taxation is
derived from its probable effect on the accumulation of
wealth. One of the motives to providence is the desire of
gaining a large fortune, but a system that in its extreme
forms prevents, and in any case hinders the attainment of
this desire must, it is argued, check the growth of capital.
The imposition of special taxation on the larger incomes is,
in fact, a fine on saving, and consequently an impediment
to the supply of one of the auxiliaries of production. If
the legislator is to interfere at all he ought rather to
encourage the formation of new stores of wealth that will,
in the vast majority of cases, be used to assist industry.

The discouragement to capital may operate in two
different ways. There will naturally be a movement of
wealthy persons from a district in which they are subjected
to special penalties. Any existing outflow of wealth will
be increased and the influx of other wealth so far checked.
Such is a very probable and serious danger in a small
district from which movement is easy, and with the modern
tendency to an international movement of capital it may
occur even in large areas. But for countries with a highly
developed system of industries, another effect is more to
be dreaded, viz. the stoppage of saving at an earlier period.
Capital may not emigrate readily from such a country as
England or France, but the annual increment may become
smaller and finally cease. Considering the dependence of
industry on the facilities for obtaining new capital it would
seem that any artificial check to its growth would be a
grave evil and likely to react on the Finances of the State.

In mitigation it may be urged that progressive taxation
is not in fact likely to weaken the disposition to save. It
will only affect those who possess a good deal already, and
such persons save as much from habit as from conscious
motive. There is, too, the further fact that the heavier
taxation on the rich will leave the poor a larger disposable
sum, part of which they may save, and to that extent
increase the store of wealth. But, though in both those
ways, the loss to capital under a moderate progression may
be reduced, it seems clear that some loss there will in-
evitably be, and it is incumbent on the supporters of any
measure tending in this direction to show what compensa-
tion will be gained through fairer distribution.
reach expectation is due partly to the evasions that have
been noticed as incident to the tax, and also to the various
devices not absolutely illegal that are used to escape the
extra pressure. If vigorously collected the tax causes much
capital to emigrate, discretion is therefore very often em-
ployed in enforcing claims, and in either case the revenue
suffers. Another reason is found in the fact that in most
countries large incomes do not form a large proportion of
national revenue. Taxation, to be productive, must draw on
the resources of the middle and working classes. The unpro-
ductiveness of progressive direct taxes is paralleled by the
small yield of taxes on the luxuries of the rich as compared
with duties on articles of general consumption l . To obtain
the funds needed by the State pressure must be placed on
all classes of society, not merely on the well-to-do.

7. The foregoing objections, which may be distinguished
in their order as political, moral, economical, and fiscal, are
so weighty that nothing but a very clear proof of injustice
inflicted by any other system than progression would per-
mit its use. The injustice of proportional or regressive ]
taxation, if established, goes to show that for the realization
of equity progression in some form must be adopted. In
support of this contention, we have nothing but the appeal
to equality of sacrifice as the standard, and the alleged
failure to conform to it by taking equal proportions from
different incomes. The deduction of ,10 from A's income
of 100, and of ; 1 0,000 from B's of ^"100,000 will, it is
maintained, inflict greater suffering on A than on B. Such
is the assumption of the upholders of progression, and their
view accords with popular sentiment. There is, neverthe-
less, room for doubt. Is it really certain that A, whose
income is reduced from ,100 to 90 is worse treated than
B whose ; 1 00,000 is brought down to ,90,000 ? There can
be no dispute as to the wants which the latter will have to
leave unsatisfied being very much slighter than those of A,
ivhen looked at from the same point of view. But the point
of view is not the same, B's system of life on its material
side is so differently constituted from A's that any com-
parison of the kind is absurd 1 . 10 from A's income may
mean the loss of a certain amount of alcoholic drinks ;
B, by having to give up ^"10,000 may lose the chance of
purchasing an estate, or may have to abandon some social
scheme that he could otherwise have carried out. The
economic calculus is not at present competent to deal with
such comparisons. The complexity of the problem is
admittedly great, and not to be solved by simple methods 2 .
The weightiest difficulty that the theoretical advocates of
progression have to meet is the essentially subjective
nature of their standard. Its translation into an objective
rule of taxation can be accomplished only by the aid of
assumptions as to the relations of enjoyment in different
classes that must contain a large element of conjecture.
The modern developments of the theory of utility fail to
supply any definite practical basis on which to frame a
scale of progression.

Progressive taxation has been supported by a very
different line of reasoning in Cohn's brilliant Treatise on
Finance 3 . Proportional taxation is asserted by him to be
the logical result of the * contract/ or assurance theory of
the State. In accordance with that belief, it was fitting
that all should pay the same proportion of income in
exchange for the stipulated services. The modern, or
' higher ' conception of the State, abandons altogether
this theory of the social compact, and therefore its
corollaries in which is included the rule of proportional
taxation. Writers, who like Rau, De Parieu, and Leroy
Beaulieu reject the older view of the State's relation to its
subjects and yet maintain the justice of proportional taxa-
tion are guilty of inconsistency explicable only by their
dread of the often described evils of progressive taxation.

To this ingenious contention, the answer is, that granting
the deduction of proportional taxation from the 'assurance
theory/ the refutation of the latter does not upset the former
since a true conclusion may result from false premisses.
But even this concession need not be made. It has been
argued in the present chapter that the exploded doctrine
of * assurance ' would logically lead not to proportional,
but to what has been called ' regressive,' taxation, i.e. to a
lower percentage on large than on small incomes 1 .

8. Experience of the actual working of progressive
systems might be expected to throw light on the reality of
the evils attributed to them and their real operation. A
large amount of evidence has been collected with this*'object
by very competent inquirers 2 , but, unfortunately, the results
are not decisive. Most of the cases discussed are those of
Swiss cantons or the smaller German States. (The short-
lived income tax of the United States, and the apparent
progressive income tax of Prussia are the only exceptions.)
Now, the financial arrangements of small political bodies
are undoubtedly full of instruction and deserve attentive
study, but they belong to the domain of local rather than
general Finance. The conditions of working are therefore
different, and there is to some extent, room for the use of
a different principle of distribution 3 , since the public ser-
vices rendered by local bodies do often allow of an estima-
tion of their value to individuals, and, besides, have to be
considered in connexion with the taxation of the State.
The peculiar economic conditions under which progressive
taxes have been applied are shown in the discussions re-
specting their merits which are chiefly concerned with the
danger of forcing capital to emigrate and that of undue
discrimination against particular persons. Both are real
and serious in a small area ; within the wider boundaries
of a nation their probability would be smaller. It is hardly
conceivable that the English Chancellor of the Exchequer
should arrange his scheme of taxation with reference to
any small number even of the wealthiest tax-payers ; nor
would the emigration of capital be caused by even a fairly
heavy tax. On the other hand, the facilities for assessment
are much increased by having to deal with a limited district
in which the income and property of each resident can be
ascertained with a close approach to the truth, and as
incomes are in no case very large, there is not the same
room for injustice. Progressive taxation would not be\/
easily applied in national Finance. The forms of wealth
are very numerous, and can be so placed as to escape the
tax-collector's notice, when he has to deal directly with
income as a whole. We have, therefore, no evidence suffi-
cient to modify the unfavourable conclusion reached on
general grounds respecting progressive taxation 1 .

9. The idea of securing equality of sacrifice while es-
caping the dangers of unregulated progression has led to
the adoption of what is known as ' degressive ' taxation ; a
system in which a uniform rate of tax is levied, but incomes
under certain limits are either altogether exempt, or rated
only for a part of their amount. Some of the so-called
progressive taxes in Switzerland are really of this kind :
thus in Zurich, 500 /rancs are free, the excess up to 1,500
francs is rated at only one-fifth, the next 1,500 francs at two-
fifths, the next 3000 at three-fifths, and the next 4000 at four-
fifths, anything beyond being rated at its full amount, e.g.
an income of 12,500 francs (^500) would only pay on 8,300
francs l . By this method the confiscation of the higher
portions of income can never happen, but there is still an
arbitrary power of fixing the several scales which is incon-
venient, while this form of progression is particularly open
to the charge of unproductiveness, and is somewhat hard
to work owing to the minute sub-divisions that are usually
made.

Degressive taxation may, however, like the more moderate
forms of progression, be employed rather to secure, than to
destroy proportionality of taxation, as it affects only one
part of the tax system, and may correct inequalities in other
directions. When the articles consumed by the poorer
classes are heavily taxed they would contribute more than
their share to the maintenance of the State were they not
relieved through the income and property taxes. This is
one of the reasons for the exemption of incomes under
.150 from income tax in the United Kingdom, and the
abatement on those under 400. The duties on tea,
tobacco, and spirits, which chiefly affect the smaller incomes
are thus balanced, and a substantial equality (or what is
believed to be such) attained. The rule of proportionality
is applicable only to the whole tax-system, and it may be
necessary to have several partial inequalities in order to
establish that final equality ' that is one of the principal

/ merits in Finance.

10. Another ground for modifying the rule of propor-

l^tional taxation exists in the doctrine that net income is the
sole available fund for social objects. If certain kinds of
expense be necessary and unavoidable, it seems that any
income which only suffices for meeting them must be
exempt from taxation. On the supposition that the
labourer's wages are just enough to keep him alive, the

1 I.e. 500 Free o

1000 200

1 500 600

3000 1800

4000 | 3200
slightest extra charge will lead to his death unless he is
relieved from some other quarter. Taxation on the mini-
mum of subsistence must, by the nature of the case, be paid
by somebody else. The Physiocrats, as we saw *, extended
this argument to the interest on capital, but their successors
have not accepted this extreme view. However, the doc-
trine known as ' the exemption of the minimum of subsis-
tence' is widely spread. Among its supporters in one
form or other may be reckoned Justi, Bentham, Sismondi,
Hermann and J. S. Mill, and it has received recognition in
the English system of taxation, since there are no duties on
the necessaries of life, and, as mentioned, incomes under
150 per annum are free from direct taxation 2 . The
different interpretations put on the doctrine need to be
distinguished. The primitive and most natural meaning
is that which limits it to the absolute necessaries of exis-
tence, though here there is room for doubt as to the
correctness of including the expense of maintaining a family
under this head. The wider use of the term to cover * the
sum of the means of support which, according to the
standard of a given period, is required for the conduct of an
existence worthy of man 3 ' would extend the exemption far
beyond the limit of physical necessaries, and would almost
approach to the exclusion of whatever expenditure is neces-
sary for the earning of the person's income from the amount
to be taxed 4 . By regarding the outlay requisite for the
support of each grade of income as its expenses of pro-
duction, we might bring the fund available for taxation
down to a very small amount.
Such a construction of the doctrine may be dismissed as
impracticable. The subject's outlay is determined by
himself and is directed for his own advantage. The only
ground for doubt would be the possibility of expenditure on
these ' necessary ' items being curtailed in consequence of
the tax. This effect would be very improbable unless the
rate of taxation were so heavy as to show bad administra-
tion, but even in the limited case of physical necessaries the
argument for remission is not so clear as might be thought.
The danger of relieving the lowest class of labourers from
nearly all the burdens of the State, while it holds pre-
ponderating political power is apparent. Again, there is
much force in the view that public expenses are a part of
necessary expenditure. ' The State,' argues Cohn, ' belongs
as much to the life of every civilized man as his daily food,
or the air ; without the State a civilized existence is not
thinkable. The minimum of every moral existence includes
the blessings of the State. It follows that the minimum of
outlay for existence must also include the necessary ex-
penseNgf the State V Why should not the poorest citizen
pay something towards security as well as purchase the
bread that supports him? The practical side of the
question seems rather to favour the existing English policy.
So far as the argument from ability is concerned it is plain
that those who barely possess the means of subsistence
have little or no ability to contribute. In any country
where legal provision is made for poor relief it would seem
that to tax those at the point of subsistence would be
simply driving them into the ranks of pauperism, and
taking with one hand to give back with the other. The
cost and trouble of raising money by direct taxation from
the poorer classes, added to the foregoing consideration,
strongly supports the method of exemption from direct
taxation of the smaller incomes with the employment of
moderate taxes on the luxuries of the poor. When exemp-
tion is claimed for the minimum it can only be on the
ground that it will be employed in buying necessaries ;
any other application of this amount fairly brings it under
the weight of taxation l .

/ 11. The question of justice may also be raised in
respect of incomes that differ, not in amount but irueiugin.
As usually debated the point is confined to the case of an
income tax, but it is really wider and applies to all forms.
To put the issue in the simplest way let us suppose that of
two persons one, A, obtains by his exertions $oo per
annum, the other, B, obtains the same sum from the rent
of land or interest on capital. Is it just or expedient that
A should pay the same sum in taxes that B does ? The
most natural answer is, perhaps, a negative one, and many
persons have proposed that the ca^itaLj^alues of the two
incomes should be taken as the basis of taxation 2 . A
little reflection will show that under certain conditions
there is nothing unjust in the arrangement. A's income
it is true is less durable, but then so is its chance of
taxation. The permanence of B's receipts involves like-
wise permanence of taxation. So long then as the public
charges are uniform, there is no reason for complaint.
Special occasions will sometimes occur in which extra-
ordinary expenditure actually is, or is deemed to be, ne-
cessary, and then it seems that as there is an extraordinary
call it ought to come from the capital rather than the
income of the community. A convenient mode of realizing
this end would be the imposition of an additional property
tax, which, being met out of the income of the holders,
would accomplish the end of taxing permanent incomes at
a higher rate. Another mode would be to meet the
increased outlay by loans to be repaid in a series of years.