Principles of Political Economy with some of their Applications to Social Philosophy
By John Stuart Mill
John Stuart Mill (1806-1873) originally wrote the
Principles of Political Economy, with some of their Applications to Social Philosophy very quickly, having studied economics under the rigorous tutelage of his father, James, since his youth. It was published in 1848 (London: John W. Parker, West Strand) and was republished with changes and updates a total of seven times in Mill’s lifetime.The edition presented here is that prepared by W. J. Ashley in 1909, based on Mill’s 7th edition, 1870. Ashley followed the 7th edition with great care, noting changes in the editions in footnotes and in occasional square brackets within the text. The text provides English translations to several lengthy quotations originally quoted by Mill in French. Ashley selected these from an 1865 “People’s Edition” of the Principles, but left in those quotations that had been omitted in that edition. He also prepared a useful Bibliographical Appendix, with additional readings and excerpts from some of Mill’s later writings, which we also include in this Econlib Edition. More on Mill’s life and works, as well as details of Ashley’s procedure, can be found in his Introduction.A few corrections of obvious typos were made for this website edition. However, because the original edition was so internally consistent and carefully proofread, we have erred on the side of caution, allowing some typos to remain lest someone doing academic research wishes to follow up. We have changed small caps to full caps for ease of using search engines.Internal references by page numbers have been replaced by linked paragraph reference numbers appropriate for this online edition. Paragraph references typically have three parts: the book, chapter, and paragraph. E.g.,
I.XI.15 refers to Book I, Chapter XI, paragraph 15.
William J. Ashley, ed.
First Pub. Date
London; Longmans, Green and Co.
The text of this edition is in the public domain. Picture of John Stuart Mill courtesy of The Warren J. Samuels Portrait Collection at Duke University.
- Preliminary Remarks
- Bibliographical Appendix
Of Direct Taxes
Book V, Chapter III
§1. Taxes are either direct or indirect. A direct tax is one which is demanded from the very persons who, it is intended or desired, should pay it. Indirect taxes are those which are demanded from one person in the expectation and intention that he shall indemnify himself at the expense of another: such as the excise or customs. The producer or importer of a commodity is called upon to pay a tax on it, not with the intention to levy a peculiar contribution upon him, but to tax through him the consumers of the commodity, from whom it is supposed that he will recover the amount by means of an advance in price.
Direct taxes are either on income, or on expenditure. Most taxes on expenditure are indirect, but some are direct, being imposed not on the producer or seller of an article, but immediately on the consumer. A house-tax, for example, is a direct tax on expenditure, if levied, as it usually is, on the occupier of the house. If levied on the builder or owner, it would be an indirect tax. A window-tax is a direct tax on expenditure; so are the taxes on horses and carriages, and the rest of what are called the assessed taxes.
The sources of income are rent, profits, and wages. This includes every sort of income, except gift or plunder. Taxes may be laid on any one of the three kinds of income, or an uniform tax on all of them. We will consider these in their order.
§2. A tax on rent falls wholly on the landlord. There are no means by which he can shift the burthen upon any one else. It does not affect the value or price of agricultural produce, for this is determined by the cost of production in the most unfavourable circumstances, and in those circumstances, as we have so often demonstrated, no rent is paid. A tax on rent, therefore, has no effect, other than its obvious one. It merely takes so much from the landlord, and transfers it to the state.
This, however, is, in strict exactness, only true of the rent which is the result either of natural causes, or of improvements made by tenants. When the landlord makes improvements which increase the productive power of his land, he is remunerated for them by an extra payment from the tenant; and this payment, which to the landlord is properly a profit on capital, is blended and confounded with rent; which indeed it really is, to the tenant, and in respect of the economical laws which determine its amount. A tax on rent, if extending to this portion of it, would discourage landlords from making improvements: but it does not follow that it would raise the price of agricultural produce. The same improvements might be made with the tenant’s capital, or even with the landlord’s if lent by him to the tenant; provided he is willing to give the tenant so long a lease as will enable him to indemnify himself before it expires. But whatever hinders improvements from being made in the manner in which people prefer to make them, will often prevent them from being made at all: and on this account a tax on rent would be inexpedient, unless some means could be devised of excluding from its operation that portion of the nominal rent which may be regarded as landlord’s profit. This argument, however, is not needed for the condemnation of such a tax. A peculiar tax on the income of any class, not balanced by taxes on other classes, is a violation of justice, and amounts to a partial confiscation. I have already shown grounds for excepting from this censure a tax which, sparing existing rents, should content itself with appropriating a portion of any future increase arising from the mere action of natural causes. But even this could not be justly done, without offering as an alternative the market price of the land. In the case of a tax on rent which is not peculiar, but accompanied by an equivalent tax on other incomes, the objection grounded on its reaching the profit arising from improvements is less applicable: since, profits being taxed as well as rent, the profit which assumes the form of rent is liable to its share in common with other profits;
*30but since profits altogether ought, for reasons formerly stated, to be taxed somewhat lower than rent properly so called, the objection is only diminished, not removed.
§3. A tax on profits, like a tax on rent, must, at least in its immediate operation, fall wholly on the payer. All profits being alike affected, no relief can be obtained by a change of employment. If a tax were laid on the profits of any one branch of productive employment, the tax would be virtually an increase of the cost of production, and the value and price of the article would rise accordingly; by which the tax would be thrown upon the consumers of the commodity, and would not affect profits. But a general and equal tax on all profits would not affect general prices, and would fall, at least in the first instance, on capitalists alone.
There is, however, an ulterior effect, which, in a rich and prosperous country, requires to be taken into account. When the capital accumulated is so great and the rate of annual accumulation so rapid, that the country is only kept from attaining the stationary state by the emigration of capital, or by continual improvements in production; any circumstance which virtually lowers the rate of profit cannot be without a decided influence on these phenomena. It may operate in different ways. The curtailment of profit, and the consequent increased difficulty in making a fortune or obtaining a subsistence by the employment of capital, may act as a stimulus to inventions, and to the use of them when made. If improvements in production are much accelerated, and if these improvements cheapen, directly or indirectly, any of the things habitually consumed by the labourer, profits may rise, and rise sufficiently to make up for all that is taken from them by the tax. In that case the tax will have been realized without loss to any one, the produce of the country being increased by an equal, or what would in that case be a far greater, amount. The tax, however, must even in this case be considered as paid from profits, because the receivers of profits are those who would be benefited if it were taken off.
But though the artificial abstraction of a portion of profits would have a real tendency to accelerate improvements in production, no considerable improvements might actually result, or only of such a kind as not to raise general profits at all, or not to raise them so much as the tax had diminished them. If so, the rate of profit would be brought closer to that practical minimum to which it is constantly approaching: and this diminished return to capital would either give a decided check to further accumulation, or would cause a greater proportion than before of the annual increase to be sent abroad, or wasted in unprofitable speculations. At its first imposition the tax falls wholly on profits: but the amount of increase of capital, which the tax prevents, would, if it had been allowed to continue, have tended to reduce profits to the same level; and at every period of ten or twenty years there will be found less difference between profits as they are, and profits as they would in that case have been: until at last there is no difference, and the tax is thrown either upon the labourer or upon the landlord. The real effect of a tax on profits is to make the country possess at any given period, a smaller capital and a smaller aggregate production, and to make the stationary state be attained earlier, and with a smaller sum of national wealth. It is possible that a tax on profits might even diminish the existing capital of the country. If the rate of profit is already at the practical minimum, that is, at the point at which all that portion of the annual increment which would tend to reduce profits is carried off either by exportation or by speculation; then if a tax is imposed which reduces profits still lower, the same causes which previously carried off the increase would probably carry off a portion of the existing capital. A tax on profits is thus, in a state of capital and accumulation like that in England, extremely detrimental to the national wealth. And this effect is not confined to the case of a peculiar, and therefore intrinsically unjust, tax on profits. The mere fact that profits have to bear their share of a heavy general taxation, tends, in the same manner as a peculiar tax, to drive capital abroad, to stimulate imprudent speculations by diminishing safe gains, to discourage further accumulation, and to accelerate the attainment of the stationary state. This is thought to have been the principal cause of the decline of Holland, or rather of her having ceased to make progress.
Even in countries which do not accumulate so fast as to be always within a short interval of the stationary state, it seems impossible that, if capital is accumulating at all, its accumulation should not be in some degree retarded by the abstraction of a portion of its profit; and unless the effect in stimulating improvements be a full counter-balance, it is inevitable that a part of the burthen will be thrown off the capitalist, upon the labourer or the landlord. One or other of these is always the loser by a diminished rate of accumulation. If population continues to increase as before, the labourer suffers: if not, cultivation is checked in its advance, and the landlords lose the accession of rent which would have accrued to them. The only countries in which a tax on profits seems likely to be permanently a burthen on capitalists exclusively, are those in which capital is stationary, because there is no new accumulation. In such countries the tax might not prevent the old capital from being kept up through habit, or from unwillingness to submit to impoverishment, and so the capitalist might continue to bear the whole of the tax. It is seen from these considerations that the effects of a tax on profits are much more complex, more various, and in some points more uncertain, than writers on this subject have commonly supposed.
§4. We now turn to Taxes on Wages. The incidence of these is very different, according as the wages taxed are those of ordinary unskilled labour, or are the remuneration of such skilled or privileged employments, whether manual or intellectual, as are taken out of the sphere of competition by a natural or conferred monopoly.
I have already remarked, that in the present low state of popular education, all the higher grades of mental or educated labour are at a monopoly price; exceeding the wages of common workmen in a degree very far beyond that which is due to the expense, trouble, and loss of time required in qualifying for the employment. Any tax levied on these gains, which still leaves them above (or not below) their just proportion, falls on those who pay it; they have no means of relieving themselves at the expense of any other class. The same thing is true of ordinary wages, in cases like that of the United States, or of a new colony, where, capital increasing as rapidly as population can increase, wages are kept up by the increase of capital, and not by the adherence of the labourers to a fixed standard of comforts. In such a case some deterioration of their condition, whether by a tax or otherwise, might possibly take place without checking the increase of population. The tax would in that case fall on the labourers themselves, and would reduce them prematurely to that lower state to which, on the same supposition with regard to their habits, they would in any case have been reduced ultimately, by the inevitable diminution in the rate of increase of capital, through the occupation of all the fertile land.
Some will object that, even in this case, a tax on wages cannot be detrimental to the labourers, since the money raised by it, being expended in the country, comes back to the labourers again through the demand for labour. The fallacy, however, of this doctrine has been so completely exhibited in the First Book,
*31 that I need do little more than refer to that exposition. It was there shown that funds expended unproductively have no tendency to raise or keep up wages, unless when expended in the direct purchase of labour. If the government took a tax of a shilling a week from every labourer, and laid it all out in hiring labourers for military service, public works, or the like, it would, no doubt, indemnify the labourers as a class for all that the tax took from them. That would really be “spending the money among the people.” But if it expended the whole in buying goods, or in adding to the salaries of employés who bought goods with it, this would not increase the demand for labour, or tend to raise wages. Without, however, reverting to general principles, we may rely on an obvious
reductio ad absurdum. If to take money from the labourers and spend it in commodities is giving it back to the labourers, then, to take money from other classes, and spend it in the same manner, must be giving it to the labourers; consequently, the more a government takes in taxes, the greater will be the demand for labour, and the more opulent the condition of the labourers. A proposition the absurdity of which no one can fail to see.
In the condition of most communities, wages are regulated by the habitual standard of living to which the labourers adhere, and on less than which they will not multiply. Where there exists such a standard, a tax on wages will indeed for a time be borne by the labourers themselves; but unless this temporary depression has the effect of lowering the standard itself, the increase of population will receive a check, which will raise wages, and restore the labourers to their previous condition. On whom, in this case, will the tax fall? According to Adam Smith, on the community generally, in their character of consumers; since the rise of wages, he thought, would raise general prices. We have seen, however, that general prices depend on other causes, and are never raised by any circumstance which affects all kinds of productive employment in the same manner and degree. A rise of wages occasioned by a tax, must, like any other increase of the cost of labour, be defrayed from profits. To attempt to tax day-labourers, in an old country, is merely to impose an extra tax upon all employers of common labour; unless the tax has the much worse effect of permanently lowering the standard of comfortable subsistence in the minds of the poorest class.
We find in the preceding considerations an additional argument for the opinion already expressed, that direct taxation should stop short of the class of incomes which do not exceed what is necessary for healthful existence. These very small incomes are mostly derived from manual labour; and, as we now see, any tax imposed on these, either permanently degrades the habits of the labouring class, or falls on profits, and burthens capitalists with an indirect tax, in addition to their share of the direct taxes; which is doubly objectionable, both as a violation of the fundamental rule of equality, and for the reasons which, as already shown, render a peculiar tax on profits detrimental to the public wealth, and consequently to the means which society possesses of paying any taxes whatever.
§5. We now pass, from taxes on the separate kinds of income, to a tax attempted to be assessed fairly upon all kinds; in other words, an Income Tax. The discussion of the conditions necessary for making this tax consistent with justice, has been anticipated in the last chapter. We shall suppose, therefore, that these conditions are complied with. They are, first, that incomes below a certain amount should be altogether untaxed. This minimum should not be higher than the amount which suffices for the necessaries of the existing population. The exemption from the present  income tax, of all incomes under 100
l., a year, and the lower percentage formerly levied on those between 100
l., and 150
l., are only defensible on the ground that almost all the indirect taxes press more heavily on incomes between 50
l., and 150
l., than on any others whatever.
*32 The second condition is, that incomes above the limit should be taxed only in proportion to the surplus by which they exceed the limit.
*33Thirdly, that all sums saved from income and invested, should be exempt from the tax: or if this be found impracticable, that life incomes, and incomes from business and professions, should be less heavily taxed than inheritable incomes, in a degree as nearly as possible equivalent to the increased need of economy arising from their terminable character: allowance being also made, in the case of variable incomes, for their precariousness.
An income-tax, fairly assessed on these principles, would be, in point of justice, the least exceptionable of all taxes. The objection to it, in the present low state of public morality,
*34 is the impossibility of ascertaining the real incomes of the contributors. The supposed hardship of compelling people to disclose the amount of their incomes, ought not, in my opinion, to count for much. One of the social evils of this country is the practice, amounting to a custom, of maintaining, or attempting to maintain, the appearance to the world of a larger income than is possessed; and it would be far better for the interest of those who yield to this weakness, if the extent of their means were universally and exactly known, and the temptation removed to expending more than they can afford, stinting real wants in order to make a false show externally. At the same time, the reason of the case, even on this point, is not so exclusively on one side of the argument as is sometimes supposed. So long as the vulgar of any country are in the debased state of mind which this national habit presupposes—so long as their respect (if such a word can be applied to it) is proportioned to what they suppose to be each person’s pecuniary means—it may be doubted whether anything which would remove all uncertainty as to that point, would not considerably increase the presumption and arrogance of the vulgar rich, and their insolence towards those above them in mind and character, but below them in fortune.
Notwithstanding, too, what is called the inquisitorial nature of the tax, no amount of inquisitorial power which would be tolerated by a people the most disposed to submit to it, could enable the revenue officers to assess the tax from actual knowledge of the circumstances of contributors. Rent, salaries, annuities, and all fixed incomes, can be exactly ascertained. But the variable gains of professions, and still more the profits of business, which the person interested cannot always himself exactly ascertain, can still less be estimated with any approach to fairness by a tax-collector. The main reliance must be placed, and always has been placed, on the returns made by the person himself. No production of accounts is of much avail, except against the more flagrant cases of falsehood; and even against these the check is very imperfect, for if fraud is intended, false accounts can generally be framed which it will baffle any means of inquiry possessed by the revenue officers to detect: the easy resource of omitting entries on the credit side being often sufficient without the aid of fictitious debts or disbursements. The tax, therefore, on whatever principles of equality it may be imposed, is in practice unequal in one of the worst ways, falling heaviest on the most conscientious. The unscrupulous succeed in evading a great proportion of what they should pay; even persons of integrity in their ordinary transactions are tempted to palter with their consciences, at least to the extent of deciding in their own favour all points on which the smallest doubt or discussion could arise: while the strictly veracious may be made to pay more than the state intended, by the powers of arbitrary assessment necessarily intrusted to the Commissioners, as the last defence against the tax-payer’s power of concealment.
It is to be feared, therefore, that the fairness which belongs to the principle of an income tax, cannot
*35 be made to attach to it in practice: and that this tax, while apparently the most just of all modes of raising a revenue, is in effect more unjust than many others which are
primâ facie more objectionable. This consideration would lead us to concur in the opinion which, until of late, has usually prevailed—that direct taxes on income should be reserved as an extraordinary resource for great national emergencies, in which the necessity of a large additional revenue overrules all objections.
The difficulties of a fair income tax have elicited a proposition for a direct tax of so much per cent, not on income, but on expenditure; the aggregate amount of each person’s expenditure being ascertained, as the amount of income now is, from statements furnished by the contributors themselves. The author of this suggestion, Mr. Revans, in a clever pamphlet on the subject,
*36 contends that the returns which persons would furnish of their expenditure would be more trustworthy than those which they now make of their income, inasmuch as expenditure is in its own nature more public than income, and false representations of it more easily detected. He cannot, I think, have sufficiently considered, how few of the items in the annual expenditure of most families can be judged of with any approximation to correctness from the external signs. The only security would still be the veracity of individuals, and there is no reason for supposing that their statements would be more trustworthy on the subject of their expenses than that of their revenues; especially as, the expenditure of most persons being composed of many more items than their income, there would be more scope for concealment and suppression in the detail of expenses than even of receipts.
The taxes on expenditure at present in force, either in this or in other countries, fall only on particular kinds of expenditure, and differ no otherwise from taxes on commodities than in being paid directly by the person who consumes or uses the article, instead of being advanced by the producer or seller, and reimbursed in the price. The taxes on horses and carriages, on dogs, on servants, are all of this nature. They evidently fall on the persons from whom they are levied—those who use the commodity taxed. A tax of a similar description, and more important, is a house-tax; which must be considered at somewhat greater length.
§6. The rent of a house consists of two parts, the ground-rent, and what Adam Smith calls the building-rent. The first is determined by the ordinary principles of rent. It is the remuneration given for the use of the portion of land occupied by the house and its appurtenances; and varies from a mere equivalent for the rent which the ground would afford in agriculture to the monopoly rents paid for advantageous situations in populous thoroughfares. The rent of the house itself, as distinguished from the ground, is the equivalent given for the labour and capital expended on the building. The fact of its being received in quarterly or half-yearly payments, makes no difference in the principles by which it is regulated. It comprises the ordinary profit on the builder’s capital, and an annuity, sufficient at the current rate of interest, after paying for all repairs chargeable on the proprietor, to replace the original capital by the time the house is worn out, or by the expiration of the usual term of a building lease.
A tax of so much per cent on the gross rent, falls on both those portions alike. The more highly a house is rented, the more it pays to the tax, whether the quality of the situation or that of the house itself is the cause. The incidence, however, of these two portions of the tax must be considered separately.
As much of it as is a tax on building-rent, must ultimately fall on the consumer, in other words the occupier. For as the profits of building are already not above the ordinary rate, they would, if the tax fell on the owner and not on the occupier, become lower than the profits of untaxed employments, and houses would not be built. It is probable however that for some time after the tax was first imposed, a great part of it would fall, not on the renter, but on the owner of the house. A large proportion of the consumers either could not afford, or would not choose, to pay their former rent with the tax in addition, but would content themselves with a lower scale of accommodation. Houses therefore would be for a time in excess of the demand. The consequence of such excess, in the case of most other articles, would be an almost immediate diminution of the supply: but so durable a commodity as houses does not rapidly diminish in amount. New buildings indeed, of the class for which the demand had decreased, would cease to be erected, except for special reasons; but in the meantime the temporary superfluity would lower rents, and the consumers would obtain perhaps nearly the same accommodation as formerly for the same aggregate payment, rent and tax together. By degrees, however, as the existing houses wore out, or as increase of population demanded a greater supply, rents would again rise; until it became profitable to recommence building, which would not be until the tax was wholly transferred to the occupier. In the end, therefore, the occupier bears that portion of a tax on rent, which falls on the payment made for the house itself, exclusively of the ground it stands on.
The case is partly different with the portion which is a tax on ground-rent. As taxes on rent, properly so called, fall on the landlord, a tax on ground-rent, one would suppose, must fall on the ground-landlord, at least after the expiration of the building lease. It will not, however, fall wholly on the landlord, unless with the tax on ground-rent there is combined an equivalent tax on agricultural rent. The lowest rent of land let for building is very little above the rent which the same ground would yield in agriculture: since it is reasonable to suppose that land, unless in case of exceptional circumstances, is let or sold for building as soon as it is decidedly worth more for that purpose than for cultivation.
*37If, therefore, a tax were laid on ground-rents without being also laid on agricultural rents, it would, unless of trifling amount, reduce the return from the lowest ground-rents below the ordinary return from land, and would check further building quite as effectually as if it were a tax on building-rents, until either the increased demand of a growing population, or a diminution of supply by the ordinary causes of destruction, had raised the rent by a full equivalent for the tax. But whatever raises the lowest ground-rents, raises all others, since each exceeds the lowest by the market value of its peculiar advantages. If, therefore, the tax on ground-rents were a fixed sum per square foot, the more valuable situations paying no more than those least in request, this fixed payment would ultimately fall on the occupier. Suppose the lowest ground-rent to be 10
l. per acre, and the highest 1000
l., a tax of 1
l. per acre on ground-rents would ultimately raise the former to 11
l., and the latter consequently to 1001
l., since the difference of value between the two situations would be exactly what it was before: the annual pound, therefore, would be paid by the occupier. But a tax on ground-rent is supposed to be a portion of a house-tax, which is not a fixed payment, but a percentage on the rent. The cheapest site, therefore, being supposed as before to pay 1
l., the dearest would pay 100
l., of which only the 1
l. could be thrown upon the occupier, since the rent would still be only raised to 100
l. Consequently, 99
l. of the 100
l. levied from the expensive site, would fall on the ground-landlord. A house-tax thus requires to be considered in a double aspect, as a tax on all occupiers of houses, and a tax on ground-rents.
In the vast majority of houses, the ground-rent forms but a small proportion of the annual payment made for the house, and nearly all the tax falls on the occupier. It is only in exceptional cases, like that of the favourite situations in large towns, that the predominant element in the rent of the house is the ground-rent; and among the very few kinds of income which are fit subjects for peculiar taxation, these ground-rents hold the principal place, being the most gigantic example extant of enormous accessions of riches acquired rapidly, and in many cases unexpectedly, by a few families, from the mere accident of their possessing certain tracts of land, without their having themselves aided in the acquisition by the smallest exertion, outlay, or risk. So far therefore as a house-tax falls on the ground-landlord, it is liable to no valid objection.
In so far as it falls on the occupier, if justly proportioned to the value of the house, it is one of the fairest and most unobjectionable of all taxes. No part of a person’s expenditure is a better criterion of his means, or bears, on the whole, more nearly the same proportion to them. A house-tax is a nearer approach to a fair income tax, than a direct assessment on income can easily be; having the great advantage, that it makes spontaneously all the allowances which it is so difficult to make, and so impracticable to make exactly, in assessing an income tax: for if what a person pays in house-rent is a test of anything, it is a test not of what he possesses, but of what he thinks he can afford to spend. The equality of this tax can only be seriously questioned on two grounds. The first is, that a miser may escape it. This objection applies to all taxes on expenditure: nothing but a direct tax on income can reach a miser. But as misers do not now hoard their treasure, but invest it in productive employments, it not only adds to the national wealth, and consequently to the general means of paying taxes, but the payment claimable from itself is only transferred from the principal sum to the income afterwards derived from it, which pays taxes as soon as it comes to be expended. The second objection is, that a person may require a larger and more expensive house, not from having greater means, but from having a larger family. Of this, however, he is not entitled to complain; since having a large family is at a person’s own choice: and, so far as concerns the public interest, is a thing rather to be discouraged than promoted.
A large portion of the taxation of this country is raised by a house-tax. The parochial taxation of the towns entirely, and of the rural districts partially, consists of an assessment on house-rent. The window-tax, which was also a house-tax, but of a bad kind, operating as a tax on light, and a cause of deformity in building, was exchanged in 1851 for a house-tax properly so called, but on a much lower scale than that which existed previously to 1834. It is to be lamented that the new tax retains the unjust principle on which the old house-tax was assessed, and which contributed quite as much as the selfishness of the middle classes to produce the outcry against the tax. The public were justly scandalized on learning that residences like Chatsworth or Belvoir were only rated on an imaginary rent of perhaps 200
l. a year, under the pretext that owing to the great expense of keeping them up, they could not be let for more. Probably, indeed, they could not be let even for that, and if the argument were a fair one, they ought not to have been taxed at all. But a house-tax is not intended as a tax on incomes derived from houses, but on expenditure incurred for them. The thing which it is wished to ascertain is what a house costs to the person who lives in it, not what it would bring in if let to some one else. When the occupier is not the owner, and does not hold on a repairing lease, the rent he pays is the measure of what the house costs him: but when he is the owner, some other measure must be sought. A valuation should be made of the house, not at what it would sell for, but at what would be the cost of rebuilding it, and this valuation might be periodically corrected by an allowance for what it had lost in value by time, or gained by repairs and improvements. The amount of the amended valuation would form a principal sum, the interest of which, at the current price of the public funds, would form the annual value at which the building should be assessed to the tax.
As incomes below a certain amount ought to be exempt from income tax, so ought houses below a certain value from house-tax, on the universal principle of sparing from all taxation the absolute necessaries of healthful existence. In order that the occupiers of lodgings, as well as of houses, might benefit, as in justice they ought, by this exemption, it might be optional with the owners to have every portion of a house which is occupied by a separate tenant, valued and assessed separately, as is now usually the case with chambers.
Book V. Chapter III. Section 4
Book V. Chapter III. Section 5
Book V. Chapter III. Section 6
 It has been also objected that house-rent in the rural districts is much lower than in towns, and lower in some towns and in some rural districts than in others: so that a tax proportioned to it would have a corresponding inequality of pressure. To this, however, it may be answered, that in places where house-rent is low persons of the same amount of income usually live in larger and better houses, and thus expend in house-rent more nearly the same proportion of their incomes than might at first sight appear. Or if not, the probability will be, that many of them live in those places precisely because they are too poor to live elsewhere, and have therefore the strongest claim to be taxed lightly. In some cases, it is precisely because the people are poor that house-rent remains low.
Book V. Chapter IV. Section 2