The Nature of Rent

by T.R. Malthus

1815

An Inquiry into the Nature and Progress of Rent, and the 
Principles by which it is regulated. by Rev. T.R. Malthus, 
Professor of History and Political Economy In the East India 
College, Hertfordshire London, Printed for John Murray, Albemarle Street 1815. 

Advertisement 
    The following tract contains the substance of some notes on 
rent, which, with others on different subjects relating to 
political economy, I have collected in the course of my 
professional duties at the East India College. It has been my 
intention, at some time or other, to put them in a form for 
publication; and the very near connection of the subject of the 
present inquiry, with the topics immediately under discussion, 
has induced me to hasten its appearance at the present moment. It 
is the duty of those who have any means of contributing to the 
public stock of knowledge, not only to do so, but to do it at the 
time when it is most likely to be useful. If the nature of the 
disquisition should appear to the reader hardly to suit the form 
of a pamphlet, my apology must be, that it was not originally 
intended for so ephemeral a shape. 
 

The rent of land is a portion of the national revenue, which 
has always been considered as of very high importance. 
According to Adam Smith, it is one of the three original 
sources of wealth, on which the three great divisions of society 
are supported. 

By the Economists it is so pre-eminently distinguished, that 
it is considered as exclusively entitled to the name of riches, 
and the sole fund which is capable of supporting the taxes of the 
state, and on which they ultimately fall. 
 
And it has, perhaps, a particular claim to our attention at 
the present moment, on account of the discussions which are going 
on respecting the corn laws, and the effects of rent on the price 
of raw produce, and the progress of agricultural improvement. 
 
The rent of land may be defined to be that portion of the 
value of the whole produce which remains to the owner of the 
land, after all the outgoings belonging to its cultivation, of 
whatever kind, have been paid, including the profits of the 
capital employed, estimated according to the usual and ordinary 
rate of the profits of agricultural stock at the time being. 
 
It sometimes happens, that from accidental and temporary 
circumstances, the farmer pays more, or less, than this; but this 
is the point towards which the actual rents paid are constantly 
gravitating, and which is therefore always referred to when the 
term is used in a general sense. 
 
The immediate cause of rent is obviously the excess of price 
above the cost of production at which raw produce sells in the 
market. 
 
The first object therefore which presents itself for inquiry, 
is the cause or causes of the high price of raw produce. 
 
After very careful and repeated revisions of the subject, I 
do not find myself able to agree entirely in the view taken of 
it, either by Adam Smith, or the Economists; and still less, by 
some more modern writers. 
 
Almost all these writers appear to me to consider rent as too 
nearly resembling in its nature, and the laws by which it is 
governed, the excess of price above the cost of production, which 
is the characteristic of a monopoly. 
 
Adam Smith, though in some parts of the eleventh chapter of 
his first book he contemplates rent quite in its true light,(1*) 
and has interspersed through his work more just observations on 
the subject than any other writer, has not explained the most 
essential cause of the high price of raw produce with sufficient 
distinctness, though he often touches on it; and by applying 
occasionally the term monopoly to the rent of land, without 
stopping to mark its more radical peculiarities, he leaves the 
reader without a definite impression of the real difference 
between the cause of the high price of the necessaries of life, 
and of monopolized commodities. 
 
Some of the views which the Economists have taken of the 
nature of rent appear to me, in like manner, to be quite just; 
but they have mixed them with so much error, and have drawn such 
preposterous and contradictory conclusions from them, that what 
is true in their doctrines, has been obscured and lost in the 
mass of superincumbent error, and has in consequence produced 
little effect. Their great practical conclusion, namely, the 
propriety of taxing exclusively the net rents of the landlords, 
evidently depends upon their considering these rents as 
completely disposable, like that excess of price above the cost 
of production which distinguishes a common monopoly. 
 
M. Say, in his valuable treatise on political economy, in 
which he has explained with great clearness many points which 
have not been sufficiently developed by Adam Smith, has not 
treated the subject of rent in a manner entirely satisfactory. In 
speaking of the different natural agents which, as well as the 
land, co-operate with the labours of man, he observes, 
'Heureusement personne n'a pu dire le vent et le soleil 
m'appartiennent, et le service qu'ils rendent doit m'etre 
paye.'(2*) And, though he acknowledges that, for obvious reasons, 
property in land is necessary, yet he evidently considers rent as 
almost exclusively owing to such appropriation, and to external 
demand. 
 
In the excellent work of M. de Sismondi, De la richesse 
commerciale, he says in a note on the subject of rent, 'Cette 
partie de la rente fonciere est celle que les Economistes ont 
decoree du nom du produit net comme etant le seul fruit du 
travail qui aj outat quelquechose a la richesse nationale. On 
pourrait au contraire soutenir contre eux, que c'est la seule 
partie du produit du travail, dont la valeur soit purement 
nominale, et n'ait rien de reelle: c'est en effet le resultat de 
l'augmentation de prix qu'obtient un vendeur en vertu de son 
privilege, sans que la chose vendue en vaille reellement 
d'avantage.'(3*) The prevailing opinions among the more modern 
writers in our own country, have appeared to me to incline 
towards a similar view of the subject; and, not to multiply 
citations, I shall only add, that in a very respectable edition 
of the Wealth of nations, lately published by Mr Buchanan, of 
Edinburgh, the idea of monopoly is pushed still further. And 
while former writers, though they considered rent as governed by 
the laws of monopoly, were still of opinion that this monopoly in 
the case of land was necessary and useful, Mr Buchanan sometimes 
speaks of it even as prejudicial, and as depriving the consumer 
of what it gives to the landlord. 
 
In treating of productive and unproductive labour in the last 
volume, he observes,(4*) that, 'The net surplus by which the 
Economists estimate the utility of agriculture, plainly arises 
from the high price of its produce, which, however advantageous 
to the landlord who receives it, is surely no advantage to the 
consumer who pays it. Were the produce of agriculture to be sold 
for a lower price, the same net surplus would not remain, after 
defraying the expenses of cultivation; but agriculture would be 
still equally productive to the general stock; and the only 
difference would be, that as the landlord was formerly enriched 
by the high price, at the expense of the community, the community 
would now profit by the low price at the expense of the landlord. 
The high price in which the rent or net surplus originates, while 
it enriches the landlord who has the produce of agriculture to 
sell, diminishes in the same proportion the wealth of those who 
are its purchasers; and on this account it is quite inaccurate to 
consider the landlord's rent as a clear addition to the national 
wealth.' In other parts of his work he uses the same, or even 
stronger language, and in a note on the subject of taxes, he 
speaks of the high price of the produce of land as advantageous 
to those who receive it, it but proportionably injurious to those 
who pay it. 'In this view,' he adds, 'it can form no general 
addition to the stock of the community, as the net surplus in 
question is nothing more than a revenue transferred from one 
class to another, and from the mere circumstance of its thus 
changing hands, it is clear that no fund can arise out of which 
to pay taxes. The revenue which pays for the produce of land 
exists already in the hands of those who purchase that produce; 
and, if the price of subsistence were lower, it would still 
remain in their hands, where it would be just as available for 
taxation, as when by a higher price it is transferred to the 
landed proprietor.'(5*) 
 
That there are some circumstances connected with rent, which 
have an affinity to a natural monopoly, will he readily allowed. 
The extent of the earth itself is limited, and cannot be enlarged 
by human demand. And the inequality of soils occasions, even at 
an early period of society a comparative scarcity of the best 
lands; and so far is undoubtedly one of the causes of rent 
properly so called. On this account, perhaps, the term partial 
monopoly might be fairly applicable. But the scarcity of land, 
thus implied, is by no means alone sufficient to produce the 
effects observed. And a more accurate investigation of the 
subject will show us how essentially different the high price of 
raw produce is, both in its nature and origin, and the laws by 
which it is governed, from the high price of a common monopoly. 
 
The causes of the high price of raw produce may be stated to 
be three. 
 
First, and mainly, that quality of the earth, by which it can 
be made to yield a greater portion of the necessaries of life 
than is required for the maintenance of the persons employed on 
the land. 
   
Secondly, that quality peculiar to the necessaries of life of 
being able to create their own demand, or to raise up a number of 
demanders in proportion to the quantity of necessaries produced. 
 
And, thirdly, the comparative scarcity of the most fertile 
land. 
 
The qualities of the soil and of its products, here noticed 
as the primary causes of the high price of raw produce, are the 
gifts of nature to man. They are quite unconnected with monopoly, 
and yet are so absolutely essential to the existence of rent, 
that without them, no degree of scarcity or monopoly could have 
occasioned that excess of the price of raw produce, above the 
cost of production, which shows itself in this form. 
 
If, for instance, the soil of the earth had been such, that, 
however well directed might have been the industry of man, he 
could not have produced from it more than was barely sufficient 
to maintain those, whose labour and attention were necessary to 
its products; though, in this case, food and raw materials would 
have been evidently scarcer than at present, and the land might 
have been, in the same manner, monopolized by particular owners; 
vet it is quite clear, that neither rent, nor any essential 
surplus produce of the land in the form of high profits, could 
have existed. 
 
It is equally clear, that if the necessaries of life the most 
important products of land - had not the property of creating an 
increase of demand proportioned to their increased quantity, such 
increased quantity would occasion a fall in their exchangeable 
value. However abundant might be the produce of a country, its 
population might remain stationary And this abundance, without a 
proportionate demand, and with a very high corn price of labour, 
which would naturally take place under these circumstances, might 
reduce the price of raw produce, like the price of manufactures, 
to the cost of production. 
 
It has been sometimes argued, that it is mistaking the 
principle of population, to imagine, that the increase of food, 
or of raw produce alone, can occasion a proportionate increase of 
population. This is no doubt true; but it must be allowed, as has 
been justly observed by Adam Smith, that 'when food is provided, 
it is comparatively easy to find the necessary clothing and 
lodging., And it should always be recollected, that land does not 
produce one commodity alone, but in addition to that most 
indispensable of all commodities - food - it produces also the 
materials for the other necessaries of life; and the labour 
required to work up these materials is of course never excluded 
from the consideration.(6*) 
 
It is, therefore, strictly true, that land produces the 
necessaries of life, produces food, materials, and labour, 
produces the means by which, and by which alone, an increase of 
people may be brought into being, and supported. In this respect 
it is fundamentally different from every other kind of machine 
known to man; and it is natural to suppose, that it should be 
attended with some peculiar effects. 
 
If the cotton machinery, in this country, were to go on 
increasing at its present rate, or even much faster; but instead 
of producing one particular sort of substance which may be used 
for some parts of dress and furniture, etc. had the qualities of 
land, and could yield what, with the assistance of a little 
labour, economy, and skill, could furnish food, clothing, and 
lodging, in such proportions as to create an increase of 
population equal to the increased supply of these necessaries; 
the demand for the products of such improved machinery would 
continue in excess above the cost of production, and this excess 
would no longer exclusively belong to the machinery of the 
land.(7*) 
 
There is a radical difference in the cause of a demand for 
those objects which are strictly necessary to the support of 
human life, and a demand for all other commodities. In all other 
commodities the demand is exterior to, and independent of, the 
production itself; and in the case of a monopoly, whether natural 
or artificial, the excess of price is in proportion to the 
smallness of the supply compared with the demand, while this 
demand is comparatively unlimited. In the case of strict 
necessaries, the existence and increase of the demand, or of the 
number of demanders, must depend upon the existence and increase 
of these necessaries themselves; and the excess of their price 
above the cost of their production must depend upon, and is 
permanently limited by, the excess of their quantity above the 
quantity necessary to maintain the labour required to produce 
them; without which excess of quantity no demand could have 
existed, according to the laws of nature, for more than was 
necessary to support the producers. 
 
It has been stated, in the new edition of the Wealth of 
nations, that the cause of the high price of raw produce is, that 
such price is required to proportion the consumption to the 
supply.(8*) This is also true, but it affords no solution of the 
point in question. We still want to know why the consumption and 
supply are such as to make the price so greatly exceed the cost 
of production, and the main cause is evidently the fertility of 
the earth in producing the necessaries of life. Diminish this 
plenty, diminish the fertility of the soil, and the excess will 
diminish; diminish it still further, and it will disappear. The 
cause of the high price of the necessaries of life above the cost 
of production, is to be found in their abundance, rather than 
their scarcity; and is not only essentially different from the 
high price occasioned by artificial monopolies, but from the high 
price of those peculiar products of the earth, not connected with 
food, which may be called natural and necessary monopolies. 
 
The produce of certain vineyards in France, which, from the 
peculiarity of their soil and situation, exclusively yield wine 
of a certain flavour, is sold of course at a price very far 
exceeding the cost of production. And this is owing to the 
greatness of the competition for such wine, compared with the 
scantiness of its supply; which confines the use of it to so 
small a number of persons, that they are able, and rather than go 
without it, willing, to give an excessively high price. But if 
the fertility of these lands were increased, so as very 
considerably to increase the produce, this produce might so fall 
in value as to diminish most essentially the excess of its price 
above the cost of production. While, on the other hand, if the 
vineyards were to become less productive, this excess might 
increase to almost any extent. 
 
The obvious cause of these effects is, that in all 
monopolies, properly so called, whether natural or artificial, 
the demand is exterior to, and independent of, the production 
itself. The number of persons who might have a taste for scarce 
wines, and would be desirous of entering into a competition for 
the purchase of them, might increase almost indefinitely, while 
the produce itself was decreasing; and its price, therefore, 
would have no other limit than the numbers, powers, and caprices, 
of the competitors for it. 
 
In the production of the necessaries of life, on the 
contrary, the demand is dependent upon the produce itself; and 
the effects are, in consequence, widely different. In this case, 
it is physically impossible that the number of demanders should 
increase, while the quantity of produce diminishes, as the 
demanders only exist by means of this produce. The fertility of 
soil, and consequent abundance of produce from a certain quantity 
of land, which, in the former case, diminished the excess of 
price above the cost of production, is, in the present case, the 
specific cause of such excess; and the diminished fertility, 
which in the former case might increase the price to almost any 
excess above the cost of production, may be safely asserted to be 
the sole cause which could permanently maintain the necessaries 
of life at a price not exceeding the cost of production. 
 
Is it, then, possible to consider the price of the 
necessaries of life as regulated upon the principle of a common 
monopoly? Is it possible, with M. de Sismondi, to regard rent as 
the sole produce of labour, which has a value purely nominal, and 
the mere result of that augmentation of price which a seller 
obtains in consequence of a peculiar privilege; or, with Mr 
Buchanan, to consider it as no addition to the national wealth, 
but merely as a transfer of value, advantageous only to the 
landlords, and proportionately injurious to the consumers? 
 
Is it not, on the contrary, a clear indication of a most 
inestimable quality in the soil, which God has bestowed on man - 
the quality of being able to maintain more persons than are 
necessary to work it? Is it not a part, and we shall see further 
on that it is an absolutely necessary part, of that surplus 
produce from the land,(9*) which has been justly stated to be the 
source of all power and enjoyment; and without which, in fact, 
there would be no cities, no military or naval force, no arts, no 
learning, none of the finer manufactures, none of the 
conveniences and luxuries of foreign countries, and none of that 
cultivated and polished society, which not only elevates and 
dignifies individuals, but which extends its beneficial influence 
through the whole mass of the people? 
 
In the early periods of society, or more remarkably perhaps, 
when the knowledge and capital of an old society are employed 
upon fresh and fertile land, this surplus produce, this bountiful 
gift of providence, shows itself chiefly in extraordinary high 
profits, and extraordinary high wages, and appears but little in 
the shape of rent. While fertile land is in abundance, and may be 
had by whoever asks for it, nobody of course will pay a rent to a 
landlord. But it is not consistent with the laws of nature, and 
the limits and quality of the earth, that this state of things 
should continue. Diversities of soil and situation must 
necessarily exist in all countries. All land cannot be the most 
fertile: all situations cannot be the nearest to navigable rivers 
and markets. But the accumulation of capital beyond the means of
employing it on land of the greatest natural fertility, and the 
greatest advantage of situation, must necessarily lower profits; 
while the tendency of population to increase beyond the means of 
subsistence must, after a certain time, lower the wages of 
labour. 
 
The expense of production will thus be diminished, but the 
value of the produce, that is, the quantity of labour, and of the 
other products of labour besides corn, which it can command, 
instead of diminishing, will be increased. There will be an 
increasing number of people demanding subsistence, and ready to 
offer their services in any way in which they can be useful. The 
exchangeable value of food will, therefore, be in excess above 
the cost of production, including in this cost the full profits 
of the stock employed upon the land, according to the actual rate 
of profits, at the time being. And this excess is rent. 
 
Nor is it possible that these rents should permanently remain 
as parts of the profits of stock, or of the wages of labour. If 
such an accumulation were to take place, as decidedly to lower 
the general profits of stock, and, consequently, the expenses of 
cultivation, so as to make it answer to cultivate poorer land; 
the cultivators of the richer land, if they paid no rent, would 
cease to be mere farmers, or persons living upon the profits of 
agricultural stock. They would unite the characters of farmers 
and landlords - a union by no means uncommon; but which does not 
alter, in any degree, the nature of rent, or its essential 
separation from profits. If the general profits of stock were 20 
per cent and particular portions of land would yield 30 per cent 
on the capital employed, 10 per cent of the 30 would obviously be 
rent, by whomsoever received. 
 
It happens, indeed, sometimes, that from bad government, 
extravagant habits, and a faulty constitution of society, the 
accumulation of capital is stopped, while fertile land is in 
considerable plenty, in which case profits may continue 
permanently very high; but even in this case wages must 
necessarily fall, which by reducing the expenses of cultivation 
must occasion rents. There is nothing so absolutely unavoidable 
in the progress of society as the fall of wages, that is such a 
fall as, combined with the habits of the labouring classes, will 
regulate the progress of population according to the means of 
subsistence. And when, from the want of an increase of capital, 
the increase of produce is checked, and the means of subsistence 
come to a stand, the wages of labour must necessarily fall so 
low, as only just to maintain the existing population, and to 
prevent any increase. 
 
We observe in consequence, that in all those countries, such 
as Poland, where, from the want of accumulation, the profits of 
stock remain very high, and the progress of cultivation either 
proceeds very slowly, or is entirely stopped, the wages of labour 
are extremely low. And this cheapness of labour, by diminishing 
the expenses of cultivation, as far as labour is concerned, 
counteracts the effects of the high profits of stock, and 
generally leaves a larger rent to the landlord than in those 
countries, such as America, where, by a rapid accumulation of 
stock, which can still find advantageous employment, and a great 
demand for labour, which is accompanied by an adequate increase 
of produce and population, profits cannot be low, and labour for 
some considerable time remains very high. 
 
It may be laid down, therefore, as an incontrovertible truth, 
that as a nation reaches any considerable degree of wealth, and 
any considerable fullness of population, which of course cannot 
take place without a great fall both in the profits of stock and 
the wages of labour, the separation of rents, as a kind of 
fixture upon lands of a certain quality, is a law as invariable 
as the action of the principle of gravity. And that rents are 
neither a mere nominal value, nor a value unnecessarily and 
injuriously transferred from one set of people to another; but a 
most real and essential part of the whole value of the national 
property, and placed by the laws of nature where they are, on the 
land, by whomsoever possessed, whether the landlord, the crown, 
or the actual cultivator. 
 
Rent then has been traced to the same common nature with that 
general surplus from the land, which is the result of certain 
qualities of the soil and its products; and it has been found to 
commence its separation from profits, as soon as profits and 
wages fall, owing to the comparative scarcity of fertile land in 
the natural progress of a country towards wealth and population. 
 
Having examined the nature and origin of rent, it remains for 
us to consider the laws by which it is governed, and by which its 
increase or decrease is regulated. 
 
When capital has accumulated, and labour fallen on the most 
eligible lands of a country, other lands less favourably 
circumstanced with respect to fertility or situation, may be 
occupied with advantage. The expenses of cultivation, including 
profits, having fallen, poorer land, or land more distant from 
markets, though yielding at first no rent, may fully repay these 
expenses, and fully answer to the cultivator. And again, when 
either the profits of stock or the wages of labour, or both, have 
still further fallen, land still poorer, or still less favourably 
situated, may be taken into cultivation. And, at every step, it 
is clear, that if the price of produce does not fall, the rents 
of land will rise. And the price of produce will not fall, as 
long as the industry and ingenuity of the labouring classes, 
assisted by the capitals of those not employed upon the land, can 
find something to give in exchange to the cultivators and 
landlords, which will stimulate them to continue undiminished 
their agricultural exertions, and maintain their increasing 
excess of produce. 
 
In tracing more particularly the laws which govern the rise 
and fall of rents, the main causes which diminish the expenses of 
cultivation, or reduce the cost of the instruments of production, 
compared with the price of produce, require to be more 
specifically enumerated. The principal of these seem to be four: 
first, such an accumulation of capital as will lower the profits 
of stock; secondly, such an increase of population as will lower 
the wages of labour; thirdly, such agricultural improvements, or 
such increase of exertions, as will diminish the number of 
labourers necessary to produce a given effect; and fourthly, such 
an increase in the price of agricultural produce, from increased 
demand, as without nominally lowering the expense of production, 
will increase the difference between this expense and the price 
of produce. 
 
The operation of the three first causes in lowering the 
expenses of cultivation, compared with the price of produce, are 
quite obvious; the fourth requires a few further observations. 
 
If a great and continued demand should arise among 
surrounding nations for the raw produce of a particular country, 
the price of this produce would of course rise considerably; and 
the expenses of cultivation, rising only slowly and gradually to 
the same proportion, the price of produce might for a long time 
keep so much ahead, as to give a prodigious stimulus to 
improvement, and encourage the employment of much capital in 
bringing fresh land under cultivation, and rendering the old much 
more productive. 
 
Nor would the effect be essentially different in a country 
which continued to feed its own people, if instead of a demand 
for its raw produce, there was the same increasing demand for its 
manufactures. These manufactures, if from such a demand the value 
of their amount in foreign countries was greatly to increase, 
would bring back a great increase of value in return, which 
increase of value could not fail to increase the value of the raw 
produce. The demand for agricultural as well as manufactured 
produce would be augmented; and a considerable stimulus, though 
not perhaps to the same extent as in the last case, would be 
given to every kind of improvement on the land. 
 
A similar effect would be produced by the introduction of new 
machinery, and a more judicious division of labour in 
manufactures. It almost always happens in this case, not only 
that the quantity of manufactures is very greatly increased, but 
that the value of the whole mass is augmented, from the great 
extension of the demand for them, occasioned by their cheapness. 
We see, in consequence, that in all rich manufacturing and 
commercial countries, the value of manufactured and commercial 
products bears a very high proportion to the raw products;(10*) 
whereas, in comparatively poor countries, without much internal 
trade and foreign commerce, the value of their raw produce 
constitutes almost the whole of their wealth. If we suppose the 
wages of labour so to rise with the rise of produce, as to give 
the labourer the same command of the means of subsistence as 
before, yet if he is able to purchase a greater quantity of other 
necessaries and conveniencies, both foreign and domestic, with 
the price of a given quantity of corn, he may be equally well 
fed, clothed, and lodged, and population may be equally 
encouraged, although the wages of labour may not rise so high in 
proportion as the price of produce. 
 
And even when the price of labour does really rise in 
proportion to the price of produce, which is a very rare case, 
and can only happen when the demand for labour precedes, or is at 
least quite contemporary with the demand for produce; it is so 
impossible that all the other outgoings in which capital is 
expended, should rise precisely in the same proportion, and at 
the same time, such as compositions for tithes, parish rates, 
taxes, manure, and the fixed capital accumulated under the former 
low prices, that a period of some continuance can scarcely fail 
to occur, when the difference between the price of produce and 
the cost of production is increased. 
 
In some of these cases, the increase in the price of 
agricultural produce, compared with the cost of the instruments 
of production, appears from what has been said to be only 
temporary; and in these instances it will often give a 
considerable stimulus to cultivation, by an increase of 
agricultural profits, without showing itself much in the shape of 
rent. It hardly ever fails, however, to increase rent ultimately. 
The increased capital, which is employed in consequence of the 
opportunity of making great temporary profits, can seldom if ever 
be entirely removed from the land, at the expiration of the 
current leases; and, on the renewal of these leases, the landlord 
feels the benefit of it in the increase of his rents. 
 
Whenever then, by the operation of the four causes above 
mentioned, the difference between the price of produce and the 
cost of the instruments of production increases, the rents of 
land will rise. 
 
It is, however, not necessary that all these four causes 
should operate at the same time; it is only necessary that the 
difference here mentioned should increase. If, for instance, the 
price of produce were to rise, while the wages of labour, and the 
price of the other branches of capital did not rise in 
proportion, and at the same time improved modes of agriculture 
were coming into general use, it is evident that this difference 
might be increased, although the profits of agricultural stock 
were not only undiminished, but were to rise decidedly higher. 
 
Of the great additional quantity of capital employed upon the 
land in this country, during the last twenty years, by far the 
greater part is supposed to have been generated on the soil, and 
not to have been brought from commerce or manufactures. And it 
was unquestionably the high profits of agricultural stock, 
occasioned by improvements in the modes of agriculture, and by 
the constant rise of prices, followed only slowly by a 
proportionate rise in the different branches of capital, that 
afforded the means of so rapid and so advantageous an 
accumulation. 
 
In this case cultivation has been extended, and rents have 
risen, although one of the instruments of production, capital, 
has been dearer. 
 
In the same manner a fall of profits and improvements in 
agriculture, or even one of them separately, might raise rents, 
notwithstanding a rise of wages. 
 
It may be laid down then as a general truth, that rents 
naturally rise as the difference between the price of produce and 
the cost of the instruments of production increases. 
 
It is further evident, that no fresh land can be taken into 
cultivation till rents have risen, or would allow of a rise upon 
what is already cultivated. 
 
Land of an inferior quality requires a great quantity of 
capital to make it yield a given produce; and, if the actual 
price of this produce be not such as fully to compensate the cost 
of production, including the existing rate of profits, the land 
must remain uncultivated. It matters not whether this 
compensation is effected by an increase in the money price of raw 
produce, without a proportionate increase in the money price of 
the instruments of production, or by a decrease in the price of 
the instruments of production, without a proportionate decrease 
in the price of produce. What is absolutely necessary, is a 
greater relative cheapness of the instruments of production, to 
make up for the quantity of them required to obtain a given 
produce from poor land. 
 
But whenever, by the operation of one or more of the causes 
before mentioned, the instruments of production become cheaper, 
and the difference between the price of produce and the expenses 
of cultivation increases, rents naturally rise. It follows 
therefore as a direct and necessary consequence, that it can 
never answer to take fresh land of a poorer quality into 
cultivation, till rents have risen or would allow of a rise, on 
what is already cultivated. 
 
It is equally true, that without the same tendency to a rise 
of rents, occasioned by the operation of the same causes, it 
cannot answer to lay out fresh capital in the improvement of old 
land - at least upon the supposition, that each farm is already 
furnished with as much capital as can be laid out to advantage, 
according to the actual rate of profits. 
 
It is only necessary to state this proposition to make its 
truth appear. It certainly may happen, and I fear it happens 
frequently, that farmers are not provided with all the capital 
which could be employed upon their farms, at the actual rate of 
agricultural profits. But supposing they are so provided, it 
implies distinctly, that more could not be applied without loss, 
till, by the operation of one or more of the causes above 
enumerated, rents had tended to rise. 
 
It appears then, that the power of extending cultivation and 
increasing produce, both by the cultivation of fresh land and the 
improvement of the old, depends entirely upon the existence of 
such prices, compared with the expense of production, as would 
raise rents in the actual state of cultivation. 
 
But though cultivation cannot be extended, and the produce of 
the country increased, but in such a state of things as would 
allow of a rise of rents, yet it is of importance to remark, that 
this rise of rents will be by no means in proportion to the 
extension of cultivation, or the increase of produce. Every 
relative fall in the price of the instruments of production, may 
allow of the employment of a considerable quantity of additional 
capital; and when either new land is taken into cultivation, or 
the old improved, the increase of produce may be considerable, 
though the increase of rents be trifling. We see, in consequence, 
that in the progress of a country towards a high state of 
cultivation, the quantity of capital employed upon the land, and 
the quantity of produce yielded by it, bears a constantly 
increasing proportion to the amount of rents, unless 
counterbalanced by extraordinary improvements in the modes of 
cultivation.(11*) 
 
According to the returns lately made to the Board of 
Agriculture, the average proportion which rent bears to the value 
of the whole produce, seems not to exceed one fifth;(12*) whereas 
formerly, when there was less capital employed, and less value 
produced, the proportion amounted to one fourth, one third, or 
even two fifths. Still, however, the numerical difference between 
the price of produce and the expenses of cultivation, increases 
with the progress of improvement; and though the landlord has a 
less share of the whole produce, yet this less share, from the 
very great increase of the produce, yields a larger quantity, and 
gives him a greater command of corn and labour. If the produce of 
land be represented by the number six, and the landlord has one 
fourth of it, his share will be represented by one and a half. If 
the produce of land be as ten, and the landlord has one fifth of 
it, his share will be represented by two. In the latter case, 
therefore, though the proportion of the landlord's share to the 
whole produce is greatly diminished, his real rent, independently 
of nominal price, will be increased in the proportion of from 
three to four. And in general, in all cases of increasing 
produce, if the landlord's share of this produce do not diminish 
in the same proportion, which though it often happens during the 
currency of leases, rarely or never happens on the renewal of 
them, the real rents of land must rise. 
 
We see then, that a progressive rise of rents seems to be 
necessarily connected with the progressive cultivation of new 
land, and the progressive improvement of the old: and that this 
rise is the natural and necessary consequence of the operation of 
four causes, which are the most certain indications of increasing 
prosperity and wealth - namely, the accumulation of capital, the 
increase of population, improvements in agriculture, and the high 
price of raw produce, occasioned by the extension of our 
manufactures and commerce. 
 
On the other hand, it will appear, that a fall of rents is as 
necessarily connected with the throwing of inferior land out of 
cultivation, and the continued deterioration of the land of a 
superior quality; and that it is the natural and necessary 
consequence of causes, which are the certain indications of 
poverty and decline, namely, diminished capital, diminished 
population, a bad system of cultivation, and the low price of raw 
produce. 
 
If it be true, that cultivation cannot be extended but under 
such a state of prices, compared with the expenses of production, 
as will allow of an increase of rents, it follows naturally that 
under such a state of relative prices as will occasion a fall of 
rents, cultivation must decline. If the instruments of production 
become dearer, compared with the price of produce, it is a 
certain sign that they are relatively scarce; and in all those 
cases where a large quantity of them is required, as in the 
cultivation of poor land, the means of procuring them will be 
deficient, and the land will be thrown out of employment. 
 
It appeared, that in the progress of cultivation and of 
increasing rents, it was not necessary that all the instruments 
of production should fall in price at the same time; and that the 
difference between the price of produce and the expense of 
cultivation might increase, although either the profits of stock 
or the wages of labour might be higher, instead of lower. 
 
In the same manner, when the produce of a country is 
declining, and rents are falling, it is not necessary that all 
the instruments of production should be dearer. In a declining or 
stationary country, one most important instrument of production 
is always cheap, namely, labour; but this cheapness of labour 
does not counterbalance the disadvantages arising from the 
dearness of capital; a bad system of culture; and, above all, a 
fall in the price of raw produce, greater than in the price of 
the other branches of expenditure, which, in addition to labour, 
are necessary to cultivation. 
 
It has appeared also, that in the progress of cultivation and 
of increasing rents, rent, though greater in positive amount, 
bears a less, and lesser proportion to the quantity of capital 
employed upon the land, and the quantity of produce derived from 
it. According to the same principle, when produce diminishes and 
rents fall, though the amount of rent will always be less, the 
proportion which it bears to capital and produce will always be 
greater. And, as in the former case, the diminished proportion of 
rent was owing to the necessity of yearly taking fresh land of an 
inferior quality into cultivation, and proceeding in the 
improvement of old land, when it would return only the common 
profits of stock, with little or no rent; so, in the latter case, 
the high proportion of rent is owing to the impossibility of 
obtaining produce, whenever a great expenditure is required, and 
the necessity of employing the reduced capital of the country, in 
the exclusive cultivation of its richest lands. 
 
In proportion, therefore, as the relative state of prices is 
such as to occasion a progressive fall of rents, more and more 
lands will be gradually thrown out of cultivation, the remainder 
will be worse cultivated, and the diminution of produce will 
proceed still faster than the diminution of rents. 
 
If the doctrine here laid down, respecting the laws which 
govern the rise and fall of rents, be near the truth, the 
doctrine which maintains that, if the produce of agriculture were 
sold at such a price as to yield less net surplus, agriculture 
would be equally productive to the general stock, must be very 
far from the truth. 
 
With regard to my own conviction, indeed, I feel no sort of 
doubt that if, under the impression that the high price of raw 
produce, which occasions rent, is as injurious to the consumer as 
it is advantageous to the landlord, a rich and improved nation 
were determined by law, to lower the price of produce, till no 
surplus in the shape of rent anywhere remained; it would 
inevitably throw not only all the poor land, but all, except the 
very best land, out of cultivation, and probably reduce its 
produce and population to less than one tenth of their former 
amount. 
 
From the preceding account of the progress of rent, it 
follows, that the actual state of the natural rent of land is 
necessary to the actual produce; and that the price of produce, 
in every progressive country, must be just about equal to the 
cost of production on land of the poorest quality actually in 
use; or to the cost of raising additional produce on old land, 
which yields only the usual returns of agricultural stock with 
little or no rent. 
 
It is quite obvious that the price cannot be less; or such 
land would not be cultivated, nor such capital employed. Nor can 
it ever much exceed this price, because the poor land 
progressively taken into cultivation, yields at first little or 
no rent; and because it will always answer to any farmer who can 
command capital, to lay it out on his land, if the additional 
produce resulting from it will fully repay the profits of his 
stock, although it yields nothing to his landlord. 
 
It follows then, that the price of raw produce, in reference 
to the whole quantity raised, is sold at the natural or necessary 
price, that is, at the price necessary to obtain the actual 
amount of produce, although by far the largest part is sold at a 
price very much above that which is necessary to its production, 
owing to this part being produced at less expense, while its 
exchangeable value remains undiminished. 
 
The difference between the price of corn and the price of 
manufactures, with regard to natural or necessary price, is this; 
that if the price of any manufacture were essentially depressed, 
the whole manufacture would be entirely destroyed; whereas, if 
the price of corn were essentially depressed, the quantity of it 
only would be diminished. There would be some machinery in the 
country still capable of sending the commodity to market at the 
reduced price. 
 
The earth has been sometimes compared to a vast machine, 
presented by nature to man for the production of food and raw 
materials; but, to make the resemblance more just, as far as they 
admit of comparison, we should consider the soil as a present to 
man of a great number of machines, all susceptible of continued 
improvement by the application of capital to them, but yet of 
very different original qualities and powers. 
 
This great inequality in the powers of the machinery employed 
in procuring raw produce, forms one of the most remarkable 
features which distinguishes the machinery of the land from the 
machinery employed in manufactures. 
 
When a machine in manufactures is invented, which will 
produce more finished work with less labour and capital than 
before, if there be no patent, or as soon as the patent is over, 
a sufficient number of such machines may be made to supply the 
whole demand, and to supersede entirely the use of all the old 
machinery. The natural consequence is, that the price is reduced 
to the price of production from the best machinery, and if the 
price were to be depressed lower, the whole of the commodity 
would be withdrawn from the market. 
 
The machines which produce corn and raw materials on the 
contrary, are the gifts of nature, not the works of man; and we 
find, by experience, that these gifts have very different 
qualities and powers. The most fertile lands of a country, those 
which, like the best machinery in manufactures, yield the 
greatest products with the least labour and capital, are never 
found sufficient to supply the effective demand of an increasing 
population. The price of raw produce, therefore, naturally rises 
till it becomes sufficiently high to pay the cost of raising it 
with inferior machines, and by a more expensive process; and, as 
there cannot be two prices for corn of the same quality, all the 
other machines, the working of which requires less capital 
compared with the produce, must yield rents in proportion to 
their goodness. 
 
Every extensive country may thus be considered as possessing 
a gradation of machines for the production of corn and raw 
materials, including in this gradation not only all the various 
qualities of poor land, of which every large territory has 
generally an abundance, but the inferior machinery which may be 
said to be employed when good land is further and further forced 
for additional produce. As the price of raw produce continues to 
rise, these inferior machines are successively called into 
action; and, as the price of raw produce continues to fall, they 
are successively thrown out of action. The illustration here used 
serves to show at once the necessity of the actual price of corn 
to the actual produce, and the different effect which would 
attend a great reduction in the price of any particular 
manufacture, and a great reduction in the price of raw produce. 
 
I hope to be excused for dwelling a little, and presenting to 
the reader in various forms the doctrine, that corn in reference 
to the quantity actually produced is sold at its necessary price 
like manufactures, because I consider it as a truth of the 
highest importance, which has been entirely overlooked by the 
Economists, by Adam Smith, and all those writers who have 
represented raw produce as selling always at a monopoly price. 
 
Adam Smith has very clearly explained in what manner the 
progress of wealth and improvement tends to raise the price of 
cattle, poultry, the materials of clothing and lodging, the most 
useful minerals, etc., etc. compared with corn; but he has not 
entered into the explanation of the natural causes which tend to 
determine the price of corn. He has left the reader, indeed, to 
conclude, that he considers the price of corn as determined only 
by the state of the mines which at the time supply the 
circulating medium of the commercial world. But this is a cause 
obviously inadequate to account for the actual differences in the 
price of grain, observable in countries at no great distance from 
each other, and at nearly the same distance from the mines. 
 
I entirely agree with him, that it is of great use to inquire 
into the causes of high price; as, from the result of such 
inquiry, it may turn out, that the very circumstance of which we 
complain, may be the necessary consequence and the most certain 
sign of increasing wealth and prosperity. But, of all inquiries 
of this kind, none surely can be so important, or so generally 
interesting, as an inquiry into the causes which affect the price 
of corn, and which occasion the differences in this price, so 
observable in different countries. 
 
I have no hesitation in stating that, independently of 
irregularities in the currency of a country,(13*) and other 
temporary and accidental circumstances, the cause of the high 
comparative money price of corn is its high comparative real 
price, or the greater quantity of capital and labour which must 
be employed to produce it: and that the reason why the real price 
of corn is higher and continually rising in countries which are 
already rich, and still advancing in prosperity and population, 
is to be found in the necessity of resorting constantly to poorer 
land - to machines which require a greater expenditure to work 
them - and which consequently occasion each fresh addition to the 
raw produce of the country to be purchased at a greater cost - in 
short, it is to be found in the important truth that corn, in a 
progressive country, is sold at the price necessary to yield the 
actual supply; and that, as this supply becomes more and more 
difficult, the price rises in proportion.(14*) 
 
The price of corn, as determined by these causes, will of 
course be greatly modified by other circumstances; by direct and 
indirect taxation; by improvements in the modes of cultivation; 
by the saving of labour on the land; and particularly by the 
importations of foreign corn. The latter cause, indeed, may do 
away, in a considerable degree, the usual effects of great wealth 
on the price of corn; and this wealth will then show itself in a 
different form. 
 
Let us suppose seven or eight large countries not very 
distant from each other, and not very differently situated with 
regard to the mines. Let us suppose further, that neither their 
soils nor their skill in agriculture are essentially unlike; that 
their currencies are in a natural state; their taxes nothing; and 
that every trade is free, except the trade in corn. Let us now 
suppose one of them very greatly to increase in capital and 
manufacturing skill above the rest, and to become in consequence 
much more rich and populous. I should say, that this great 
comparative increase of riches could not possibly take place, 
without a great comparative advance in the price of raw produce; 
and that such advance of price would, under the circumstances 
supposed, be the natural sign and absolutely necessary 
consequence, of the increased wealth and population of the 
country in question. 
 
Let us now suppose the same countries to have the most 
perfect freedom of intercourse in corn, and the expenses of 
freight, etc. to be quite inconsiderable. And let us still 
suppose one of them to increase very greatly above the rest, in 
manufacturing capital and skill, in wealth and population. I 
should then say, that as the importation of corn would prevent 
any great difference in the price of raw produce, it would 
prevent any great difference in the quantity of capital laid out 
upon the land, and the quantity of corn obtained from it; that, 
consequently, the great increase of wealth could not take place 
without a great dependence on the other nations for corn; and 
that this dependence, under the circumstances supposed, would be 
the natural sign, and absolutely necessary consequence of the 
increased wealth and population of the country in question. 
 
These I consider as the two alternatives necessarily 
belonging to a great comparative increase of wealth; and the 
supposition here made will, with proper restrictions, apply to 
the state of Europe. 
 
In Europe, the expenses attending the carriage of corn are 
often considerable. They form a natural barrier to importation; 
and even the country which habitually depends upon foreign corn, 
must have the price of its raw produce considerably higher than 
the general level. Practically, also, the prices of raw produce, 
in the different countries of Europe, will be variously modified 
by very different soils, very different degrees of taxation, and 
very different degrees of improvement in the science of 
agriculture. Heavy taxation, and a poor soil, may occasion a high 
comparative price of raw produce, or a considerable dependence on 
other countries, without great wealth and population; while great 
improvements in agriculture and a good soil may keep the price of 
produce low, and the country independent of foreign corn, in 
spite of considerable wealth. But the principles laid down are 
the general principles on the subject; and in applying them to 
any particular case, the particular circumstances of such case 
must always be taken into consideration. 
 
With regard to improvements in agriculture, which in similar 
soils is the great cause which retards the advance of price 
compared with the advance of produce; although they are sometimes 
very powerful, they are rarely found sufficient to balance the 
necessity of applying to poorer land, or inferior machines. In 
this respect, raw produce is essentially different from 
manufactures. 
 
The real price of manufactures, the quantity of labour and 
capital necessary to produce a given quantity of them, is almost 
constantly diminishing; while the quantity of labour and capital, 
necessary to procure the last addition that has been made to the 
raw produce of a rich and advancing country, is almost constantly 
increasing. We see in consequence, that in spite of continued 
improvements in agriculture, the money price of corn is ceteris 
paribus the highest in the richest countries, while in spite of 
this high price of corn, and consequent high price of labour, the 
money price of manufactures still continues lower than in poorer 
countries. 
 
I cannot then agree with Adam Smith, in thinking that the low 
value of gold and silver is no proof of the wealth and 
flourishing state of the country, where it takes place. Nothing 
of course can be inferred from it, taken absolutely, except the 
abundance of the mines; but taken relatively, or in comparison 
with the state of other countries, much may be inferred from it. 
If we are to measure the value of the precious metals in 
different countries, and at different periods in the same 
country, by the price of corn and labour, which appears to me to 
be the nearest practical approximation that can be adopted (and 
in fact corn is the measure used by Adam Smith himself), it 
appears to me to follow, that in countries which have a frequent 
commercial intercourse with each other, which are nearly at the 
same distance from the mines, and are not essentially different 
in soil; there is no more certain sign, or more necessary 
consequence of superiority of wealth, than the low value of the 
precious metals, or the high price of raw produce.(15*) 
 
It is of importance to ascertain this point; that we may not 
complain of one of the most certain proofs of the prosperous 
condition of a country. 
 
It is not of course meant to be asserted, that the high price 
of raw produce is, separately taken, advantageous to the 
consumer; but that it is the necessary concomitant of superior 
and increasing wealth, and that one of them cannot be had without 
the other.(16*) 
 
With regard to the labouring classes of society, whose 
interests as consumers may be supposed to be most nearly 
concerned, it is a very short-sighted view of the subject, which 
contemplates, with alarm, the high price of corn as certainly 
injurious to them. The essentials to their well being are their 
own prudential habits, and the increasing demand for labour. And 
I do not scruple distinctly to affirm, that under similar habits, 
and a similar demand for labour, the high price of corn, when it 
has had time to produce its natural effects, so far from being a 
disadvantage to them, is a positive and unquestionable advantage. 
To supply the same demand for labour, the necessary price of 
production must be paid, and they must be able to command the 
same quantities of the necessaries of life, whether they are high 
or low in price.(17*) But if they are able to command the same 
quantity of necessaries, and receive a money price for their 
labour, proportioned to their advanced price, there is no doubt 
that, with regard to all the objects of convenience and comfort, 
which do not rise in proportion to corn (and there are many such 
consumed by the poor), their condition will be most decidedly 
improved. 
 
The reader will observe in what manner I have guarded the 
proposition. I am well aware, and indeed have myself stated in 
another place, that the price of provisions often rises, without 
a proportionate rise of labour: but this cannot possibly happen 
for any length of time, if the demand for labour continues 
increasing at the same rate, and the habits of the labourer are 
not altered, either with regard to prudence, or the quantity of 
work which he is disposed to perform. 
 
The peculiar evil to be apprehended is, that the high money 
price of labour may diminish the demand for it; and that it has 
this tendency will be readily allowed, particularly as it tends 
to increase the prices of exportable commodities. But repeated 
experience has shown us that such tendencies are continually 
counterbalanced, and more than counterbalanced by other 
circumstances. And we have witnessed, in our own country, a 
greater and more rapid extension of foreign commerce, than 
perhaps was ever known, under the apparent disadvantage of a very 
great increase in the price of corn and labour, compared with the 
prices of surrounding countries. 
 
On the other hand, instances everywhere abound of a very low 
money price of labour, totally failing to produce an increasing 
demand for it. And among the labouring classes of different 
countries, none certainly are so wretched as those, where the 
demand for labour, and the population are stationary, and yet the 
prices of provisions extremely low, compared with manufactures 
and foreign commodities. However low they may be, it is certain, 
that under such circumstances, no more will fall to the share of 
the labourer than is necessary just to maintain the actual 
population; and his condition will be depressed, not only by the 
stationary demand for labour, but by the additional evil of being 
able to command but a small portion of manufactures or foreign 
commodities, with the little surplus which he may possess. If, 
for instance, under a stationary population, we suppose, that in 
average families two thirds of the wages estimated in corn are 
spent in necessary provisions, it will make a great difference in 
the condition of the poor, whether the remaining one third will 
command few or many conveniencies and comforts; and almost 
invariably, the higher is the price of corn, the more indulgences 
will a given surplus purchase. 
 
The high or low price of provisions, therefore, in any 
country is evidently a most uncertain criterion of the state of 
the poor in that country. Their condition obviously depends upon 
other more powerful causes; and it is probably true, that it is 
as frequently good. or perhaps more frequently so, in countries 
where corn is high, than where it is low. 
 
At the same time it should be observed, that the high price 
of corn, occasioned by the difficulty of procuring it, may be 
considered as the ultimate check to the indefinite progress of a 
country in wealth and population. And, although the actual 
progress of countries be subject to great variations in their 
rate of movement, both from external and internal causes, and it 
would be rash to say that a state which is well peopled and 
proceeding rather slowly at present, may not proceed rapidly 
forty years hence; yet it must be owned, that the chances of a 
future rapid progress are diminished by the high prices of corn 
and labour, compared with other countries. 
 
It is, therefore, of great importance, that these prices 
should be increased as little as possible artificially, that is, 
by taxation. But every tax which falls upon agricultural capital 
tends to check the application of such capital, to the bringing 
of fresh land under cultivation, and the improvement of the old. 
It was shown, in a former part of this inquiry, that before such 
application of capital could take place, the price of produce, 
compared with the instruments of production, must rise 
sufficiently to pay the farmer. But, if the increasing 
difficulties to be overcome are aggravated by taxation, it is 
necessary, that before the proposed improvements are undertaken, 
the price should rise sufficiently, not only to pay the farmer, 
but also the government. And every tax, which falls on 
agricultural capital, either prevents a proposed improvement, or 
causes it to be purchased at a higher price. 
 
When new leases are let, these taxes are generally thrown off 
upon the landlord. The farmer so makes his bargain, or ought so 
to make it, as to leave himself, after every expense has been 
paid, the average profits of agricultural stock in the actual 
circumstances of the country, whatever they may be, and in 
whatever manner they may have been affected by taxes, 
particularly by so general a one as the property tax. The farmer, 
therefore, by paying a less rent to his landlord on the renewal 
of his lease, is relieved from any peculiar pressure, and may go 
on in the common routine of cultivation with the common profits. 
But his encouragement to lay out fresh capital in improvements is 
by no means restored by his new bargain. This encouragement must 
depend, both with regard to the farmer and the landlord himself,
exclusively on the price of produce, compared with the price of 
the instruments of production; and, if the price of these 
instruments have been raised by taxation, no diminution of rent 
can give relief. It is, in fact, a question, in which rent is not 
concerned. And, with a view to progressive improvements, it may 
be safely asserted, that the total abolition of rents would be 
less effectual than the removal of taxes which fall upon 
agricultural capital. 
 
I believe it to be the prevailing opinion, that the greatest 
expense of growing corn in this country is almost exclusively 
owing to the weight of taxation. Of the tendency of many of our 
taxes to increase the expenses of cultivation and the price of 
corn, I feel no doubt; but the reader will see from the course of 
argument pursued in this inquiry, that I think a part of this 
price, and perhaps no inconsiderable part, arises from a cause 
which lies deeper, and is in fact the necessary result of the 
great superiority of our wealth and population, compared with the 
quality of our natural soil and the extent of our territory. 
 
This is a cause which can only be essentially mitigated by 
the habitual importation of foreign corn, and a diminished 
cultivation of it at home. The policy of such a system has been 
discussed in another place; but, of course, every relief from 
taxation must tend, under any system, to make the price of corn 
less high, and importation less necessary. 
    In the progress of a country towards a high state of 
improvement, the positive wealth of the landlord ought, upon the 
principles which have been laid down, gradually to increase; 
although his relative condition and influence in society will 
probably rather diminish, owing to the increasing number and 
wealth of those who live upon a still more important surplus(18*) 
- the profits of stock. 
 
The progressive fall, with few exceptions, in the value of 
the precious metals throughout Europe; the still greater fall, 
which has occurred in the richest countries, together with the 
increase of produce which has been obtained from the soil, must 
all conduce to make the landlord expect an increase of rents on 
the renewal of his leases. But, in reletting his farms, he is 
liable to fall into two errors, which are almost equally 
prejudicial to his own interests, and to those of his country. 
 
In the first place, he may be induced, by the immediate 
prospect of an exorbitant rent, offered by farmers bidding 
against each other, to let his land to a tenant without 
sufficient capital to cultivate it in the best way, and make the 
necessary improvements upon it. This is undoubtedly a most 
short-sighted policy, the bad effects of which have been strongly 
noticed by the most intelligent land surveyors in the evidence 
lately brought before Parliament; and have been particularly 
remarkable in Ireland, where the imprudence of the landlords in 
this respect, combined, perhaps, with some real difficulty of 
finding substantial tenants, has aggravated the discontents of 
the country, and thrown the most serious obstacles in the way of 
an improved system of cultivation. The consequence of this error 
is the certain loss of all that future source of rent to the 
landlord, and wealth to the country, which arises from increase 
of produce. 
 
The second error to which the landlord is liable, is that of 
mistaking a mere temporary rise of prices, for a rise of 
sufficient duration to warrant an increase of rents. It 
frequently happens, that a scarcity of one or two years, or an 
unusual demand arising from any other cause, may raise the price 
of raw produce to a height, at which it cannot be maintained. And 
the farmers, who take land under the influence of such prices, 
will, in the return of a more natural state of things, probably 
break, and leave their farms in a ruined and exhausted state. 
These short periods of high price are of great importance in 
generating capital upon the land, if the farmers are allowed to 
have the advantage of them; but, if they are grasped at 
prematurely by the landlord, capital is destroyed, instead of 
being accumulated; and both the landlord and the country incur a 
loss, instead of gaining a benefit. 
 
A similar caution is necessary in raising rents, even when 
the rise of prices seems as if it would be permanent. In the 
progress of prices and rents, rent ought always to be a little 
behind; not only to afford the means of ascertaining whether the 
rise be temporary or permanent, but even in the latter case, to 
give a little time for the accumulation of capital on the land, 
of which the landholder is sure to feel the full benefit in the 
end. 
 
There is no just reason to believe, that if the lands were to 
give the whole of their rents to their tenants, corn would be 
more plentiful and cheaper. If the view of the subject, taken in 
the preceding inquiry, be correct, the last additions made to our 
home produce are sold at the cost of production, and the same 
quantity could not be produced from our own soil at a less price, 
even without rent. The effect of transferring all rents to 
tenants, would be merely the turning them into gentlemen, and 
tempting them to cultivate their farms under the superintendence 
of careless and uninterested bailiffs, instead of the vigilant 
eye of a master, who is deterred from carelessness by the fear of 
ruin, and stimulated to exertion by the hope of a competence. The 
most numerous instances of successful industry, and well-directed 
knowledge, have been found among those who have paid a fair rent 
for their lands; who have embarked the whole of their capital in 
their undertaking; and who feel it their duty to watch over it 
with unceasing care, and add to it whenever it is possible. But 
when this laudable spirit prevails among a tenantry, it is of the 
very utmost importance to the progress of riches, and the 
permanent increase of rents, that it should have the power as 
well as the will to accumulate; and an interval of advancing 
prices, not immediately followed by a proportionate rise of 
rents, furnishes the most effective powers of this kind. These 
intervals of advancing prices, when not succeeded by retrograde 
movements, most powerfully contribute to the progress of national 
wealth. And practically I should say, that when once a character 
of industry and economy has been established, temporary high 
profits are a more frequent and powerful source of accumulation, 
than either an increased spirit of saving, or any other cause 
that can be named.(19*) It is the only cause which seems capable 
of accounting for the prodigious accumulation among individuals, 
which must have taken place in this country during the last 
twenty years, and which has left us with a greatly increased 
capital, notwithstanding our vast annual destruction of stock, 
for so long a period. 
 
Among the temporary causes of high price, which may sometimes 
mislead the landlord, it is necessary to notice irregularities in 
the currency. When they are likely to be of short duration, they 
must be treated by the landlord in the same manner as years of 
unusual demand. But when they continue so long as they have done 
in this country, it is impossible for the landlord to do 
otherwise than proportion his rent accordingly, and take the 
chance of being obliged to lessen it again, on the return of the 
currency to its natural state. 
 
The present fall in the price of bullion, and the improved 
state of our exchanges, proves, in my opinion, that a much 
greater part of the difference between gold and paper was owing 
to commercial causes, and a peculiar demand for bullion than was 
supposed by many persons; but they by no means prove that the 
issue of paper did not allow of a higher rise of prices than 
could be permanently maintained. Already a retrograde movement, 
not exclusively occasioned by the importations of corn, has been 
sensibly felt; and it must go somewhat further before we can 
return to payments in specie. Those who let their lands during 
the period of the greatest difference between notes and bullion, 
must probably lower them, whichever system may be adopted with 
regard to the trade in corn. These retrograde movements are 
always unfortunate; and high rents, partly occasioned by causes 
of this kind, greatly embarrass the regular march of prices, and 
confound the calculations both of the farmer and landlord. 
 
With the cautions here noticed in letting farms, the landlord 
may fairly look forward to a gradual and permanent increase of 
rents; and, in general, not only to an increase proportioned to 
the rise in the price of produce, but to a still further 
increase, arising from an increase in the quantity of produce. 
    If in taking rents, which are equally fair for the landlord 
and tenant, it is found that in successive lettings they do not 
rise rather more than in proportion to the price of produce, it 
will generally be owing to heavy taxation. 
 
Though it is by no means true, as stated by the Economists, 
that all taxes fall on the net rents of the landlords, yet it is 
certainly true that they are more frequently taxed both 
indirectly as well as directly, and have less power of relieving 
themselves, than any other order of the state. And as they pay, 
as they certainly do, many of the taxes which fall on the capital 
of the farmer and the wages of the labourer, as well as those 
directly imposed on themselves; they must necessarily feel it in 
the diminution of that portion of the whole produce, which under 
other circumstances would have fallen to their share. But the 
degree in which the different classes of society are affected by 
taxes, is in itself a copious subject, belonging to the general 
principles of taxation, and deserves a separate inquiry. 

NOTES: 

1. I cannot, however, agree with him in thinking that all land 
which yields food must necessarily yield rent. The land which is 
successively taken into cultivation in improving countries, may 
only pay profits and labour. A fair profit on the stock employed, 
including, of course, the payment of labour, will always be a 
sufficient inducement to cultivate. 

2. J.B. Say, Traite d'economie politique, 2nd ed., 2 vols. 
(Paris, 1814) ii, p. 124. Of this work a new and much improved 
edition has lately been published, which is highly worthy the 
attention of all those who take an interest in these subjects. 

3. J.C.L.S. de Sismondi, De la richesse commerciale, 2 vols. 
(Geneva, 1803), i, p. 49. 

4. Adam Smith, An inquiry into the nature and causes of the 
wealth of nations, ed. D. Buchanan, 4 vols. (Edinburgh, 1814) iv, 
p. 134. 

5. Smith, Wealth of nations, iii, p. 212. 

6. It is, however, certain, that if either these materials be 
wanting, or the skill and capital necessary to work them up be 
prevented from forming, owing to the insecurity of property, to 
any other cause, the cultivators will soon slacken in their 
exertions, and the motives to accumulate and to increase their 
produce, will greatly diminish. But in this case there will be a 
very slack demand for labour; and, whatever may be the nominal 
cheapness of provisions, the labourer will not really be able to 
command such a portion of the necessaries of life, including, of 
course, clothing, lodging, etc. as will occasion an increase of 
population. 

7. I have supposed some check to the supply of the cotton 
machinery in this case. If there was no check whatever, the 
effects wold show themselves in excessive profits and excessive 
wages, without an excess above the cost of production. 

8. Smith, Wealth of nations, iv, p. 35. 

9. The more general surplus here alluded to is meant to include 
the profits of the farmer, as well as the rents of the landlord; 
and, therefore, includes the whole fund for the support of those 
who are not directly employed upon the land. Profits are, in 
reality, a surplus, as they are in no respect proportioned (as 
intimated by the Economists) to the wants and necessities of the 
owners of capital. But they take a different course in the 
progress of society from rents, and it is necessary, in general, 
to keep them quite separate. 

10. According to the calculations of Mr Colquhoun, the value of 
our trade, foreign and domestic,and of our manufactures, 
exclusive of raw materials, is nearly equal to the gross value 
derived from the land. In no other large country probably is this 
the case. P. Colquhoun, Treatise on the wealth, power, and 
resources of the British Empire, 2nd ed. (1815), p. 96. The whole 
annual produce is estimated at about 430 millions, and the 
products of agriculture at about 216 millions. 

11. To the honour of Scotch cultivators, it should be observed, 
that they have applied their capitals so very skilfully and 
economically, that at the same time that they have prodigiously 
increased the produce, they have increase the landlord's 
proportion ot it. The difference between the landlord's share of 
the produce in Scotland and in England is quite extraordinary -- 
much greater than can be accounted for, either by the natural 
soil or the absence of tithes and poor's rates. See Sir John 
Sinclair's valuable An account of husbandry in Scotland 
(Edinburgh, 1812) and General Report, 4 vols. (Edinburgh, 1814) 
not long since published -- works replete with the most useful 
and interesting information on agricultural subjects. 

12. See BPP, 1814-5, V, p. 66, evidence before the House of 
Lords, given by Arthur Young. 

13. In all our discussions we should endeavour, as well as we 
can, to separate that part of high price, which arises from 
excess of currency, from that part, which is natural, and arises 
from permanent causes. In the whole course of this argument, it 
is particularly necessary to do this. 

14. It will be observed, that l have said in a progressive 
country; that is, in a country which requires yearly the 
employment of a greater capital on the land, to support an 
increasing population. If there were no question about fresh 
capital, or an increase of people, and all the land were good, it 
would not then be true that corn must be sold at its necessary 
price. The actual price might be diminished; and if the rents of 
land were diminished in proportion. the cultivation might go on 
as before, and the same quantity be produced lt very rarely 
happens, however, that all the lands of a country actually 
occupied are good, and yield a good net rent. And in all cases, a 
fall of prices must destroy agricultural capital during the 
currency of leases; and on their renewal there would not be the 
same power of production. 

15. This conclusion may appear to contradict the doctrine of the 
level of the precious metals. And so it does, if by level be 
meant level of value estimated in the usual way. I consider the 
doctrine, indeed, as quite unsupported by facts, and the 
comparison of the precious metals to water perfectly inaccurate. 
The precious metals are always tending to a state of rest, or 
such a state of things as to make their movement unnecessary. But 
when this state of rest has been nearly attained, and the 
exchanges of all countries are nearly at par, the value of the 
precious metals in different countries, estimated in corn and 
labour, or the mass of commodities, is very far indeed from being 
the same. To be convinced of this, it is only necessary to look 
at England, France, Poland, Russia, and India, when the exchanges 
are at par. That Adam Smith. who proposes labour as the true 
measure of value at all times and in all places, could look 
around him, and vet say that the precious metals were always the 
highest in value in the richest countries, has always appeared to 
me most unlike his usual attention to found his theories on 
facts. 

16. Even upon the system of importation, in the actual state and 
situation of the countries of Europe, higher prices must 
accompany superior and increasing wealth. 

17. We must not be so far deceived by the evidence before 
Parliament, relating to the want of connection between the prices 
of corn and of labour, as to suppose that they are really 
independent of each other. The price of the necessaries of life 
is, in fact, the cost of producing labour. The supply cannot 
proceed, if it be not paid; and though there will always be a 
little latitude, owing to some variations of industry and habits, 
and the distance of time between the encouragement to population 
and the period of the results appearing in the markets: yet it is 
a still greater error, to suppose the price of labour unconnected 
with the price of corn, than to suppose that the price of corn 
immediately and completely regulates it. Corn and labour rarely 
march quite abreast; but there is an obvious limit, beyond which 
they cannot be separated. With regard to the unusual exertions 
made by the labouring classes in periods of dearness, which 
produce the fall of wages noticed in the evidence, they are most 
meritorious in the individuals, and certainly favour the growth 
of capital. But no man of humanity could wish to see them 
constant and unremitted. They are most admirable as a temporary 
relief; but if they were constantly in action, effects of a 
similar kind would result from them, as from the population of a 
country being pushed to the very extreme limits of its food. 
There would be no resources in a scarcity. I own I do not see, 
with pleasure, the great extension of the practice of task work. 
To work really hard during twelve or fourteen hours in the day, 
for any length of time, is too much for a human being. Some 
intervals of ease are necessary to health and happiness: and the 
occasional abuse of such intervals is no valid argument against 
their use. 

18. I have hinted before, in a note, that profits may, without 
impropriety, be called a surplus. But, whether surplus or not, 
they are the most important source of wealth, as they are, beyond 
all question, the main source of accumulation. 

19. Adam Smith notices the bad effects of high profits on the 
habits of the capitalist. They may perhaps sometimes occasion 
extravagance; but generally, I should say, that extravagant 
habits were a more frequent cause of a scarcity of capital and 
high profits, than high profits of extravagant habits.