Alfred & Mary Marshall,
The Economics of Industry
(1879)

Chapter VI.
Influence of Trades Unions on Wages.

§ I. WE have already seen how there is a continual contest between the different industrial classes as to the distribution of the produce of their joint labour; we have now to inquire how this contest is affected by trade combinations. Let us neglect for a time the conflicting interests of different classes of hired labour; and suppose that producers can be divided into one great class of employers and another of employed. And let us inquire whether it would be possible for the employed by combining among themselves to raise wages generally at the expense of profits.

We know that, other things being equal, it will be to the interest of an employer to pay wages equal to the net value of the labourer’s work, if he cannot get a sufficient supply of labour on cheaper terms. If a farmer for instance calculates that the work of an additional labourer would add to the produce of his farm enough to repay with profits the outlay of 14s. a week in wages, it will be to his interest, other things being equal, to offer these wages rather than go without the extra assistance. But other things are very likely not to be equal. If the current rate in the parish is 12s. a week, he could not bid 14s. without incurring odium among his brother farmers, and perhaps tempting the labourers already in his employ to demand 14s. So he will probably offer only 12s., and complain of the scarcity of labour. The price of 12s. will be maintained because competition is not perfectly free; because the labourers have not much choice as to the market in which they sell their labour, and because they cannot bold back their labour at a reserve price equal to the highest wage which the employer can afford to pay.

The disadvantage under which labourers lie in such a case as this, may be seen by considering the position of a shopkeeper in like circumstances. As a rule a shopkeeper fixes the price of his goods; and if the customers who come into his shop on one day refuse to pay that price, he waits till others come who will pay it. But if at any time he were compelled to sell off his goods quickly, taking whatever offers he could get, and not holding back for any reserve price, he might have to sell them at much less than their real value, at all events if he had access to only a few buyers. For these few might not happen to have much occasion for his goods, so that it might not be worth their while to pay him a good price; and they might even combine to take advantage of his necessity, and force him to sell at a lower price than it would have been worth their while to pay.*

* If there are very few buyers in a market. it may make a great difference whether goods are sold by Dutch or English auction. In a Dutch auction the salesman starts with a high price and lowers it till he comes to a price, say 20s, which some one is willing to pay rather than go without the thing. But in an English auction, the salesman would have started with a low price, and raised it till no one would bid any higher: and if there had been only one person willing to pay 18s., and he bid 18s, no bid beyond this would be made, and the thing would have been sold for 2s. less by English than by Dutch auction. See Thornton On Labour, 2nd ed. pp. 56, y.

In the same way, if the labourer has no savings of his own, and does not belong to a trade combination, he may have to sell his labour at whatever price the employers in his neighbourhood may agree to offer; and this price may be considerably less than they would have been willing to pay rather than go without his labour. If these local agreements among employers are common to nearly all places and nearly all trades, profits in all trades alike may be a little higher and wages a little lower, than if they had been determined by that perfectly free competition which is assumed in the theory of Normal value. It is true that profits could not be very much higher than their Normal value, because if they were, the inducement to employers to extend their businesses by hiring more labour at a higher price, would be very strong; the tacit agreements among employers would continually be broken through, and wages would rise and profits would fall nearly to their Normal level. But still it is clear that if the labourers throughout a country have been in the habit of selling their labour without reserve, they may have received among them a smaller share of the Wages-and-profits Fund than they would have got if there had been perfectly free competition among employers, or than they would have got if they had been able to offer their labour at a reserve price.

If then the labourers enter into local trade combinations, and refuse to sell their labour except at a reserve price, it is quite possible that they may increase their share of the Wages-and-profits Fund, and raise wages at the expense of profits. But to what extent scan they do this? Will their action so check the demand for their labour as to cause a reaction in which profits rise at the expense of wages?

§ 2. To answer this question we must start from the fact that wages are labour’s share of the Wagesandprofits Fund, which is the net produce of land, labour and capital, after deducting rent and taxes. Therefore a rise in wages is at all events in some danger of bringing about its own destruction, if it is obtained in such a way as to diminish this Fund. Now the rate of profits is one of many causes that govern the accumulation of capital; and a fall in the rate of profits, unless counteracted by some other cause, tends to diminish the Wages-and-profits Fund, or at least to check its growth.

It is true that this diminution may be of little importance for some years; and meanwhile the advantage which combination gives labourers in bargaining with their employers, may possibly enable them to get so large a share of the diminished Fund as to maintain their wages. But though the effects which the fall in profits exerts on the Wages-and-profits Fund may be small at first, they will increase steadily, unless the rise in wages exercises some compensatory effect. If in one year they cause the Fund to be one per cent. less than it otherwise would have been, this loss will have increased to two per cent. at the end of the second year, to three percent. at the end of the third year, to ten per cent. at the end of the tenth year, and so on.* While this loss increases steadily year by year, there will be no corresponding increase in the advantage which combination gives to labourers in their bargaining; and sooner or later the competition of capital for the aid of labour in production will be diminished; wages will fall, and will probably go on falling until the removal of the causes which checked the growth of the Wages-and-profits Fund.

* The case in the text is understated. Even if the rate of profits remained unchanged without any further fall throughout the ten years, the loss to the Wagesandprofits Fund would increase in geometric, not in arithmetic progression. But further, wages could not be kept at their raised level without throwing a continually increasing burden on profits; and therefore the diminution (or check to the growth) of the Wages-and-profits Fund would be very much greater in the second year than in the first, very much greater in the third year than in the second, and so on. Further, a fall in the rate of interest promotes the use of machinery, and tends to increase Auxiliary-capital at the expense of Wage-capital, and thus to lower wages. An exact treatment of the problem here indicated requires the aid of mathematics.

It is then clear that if a rise of wages is obtained simply at the expense of profits, if it lowers profits without exerting any compensatory effect on the Wages-and-profits Fund, it must be self-destructive in the long run. It must lead in time to such a scarcity of capital and of business power that the total Wages-and-profits Fund will be insufficient to afford high wages to labour, even while capital is getting a low rate of interest, and business power is receiving low Earnings of Management.

§ 3. But it is not necessary that a rise of wages which labourers may obtain by the aid of their combinations, should be self-destructive. For even if it is got at the expense of profits in the first instance, it may be used in such a way as to prevent any diminution of the Wages-and-profits Fund, and to throw no permanent burden on profits.

Firstly, if labourers saved as large a part of their income as capitalists and employers do, a rise of wages at the expense of profits would scarcely affect the accumulation of capital. But the working classes consume nearly the whole of their income on their immediate wants. The wages of hired labour in the United kingdom amount to nearly £500,000,000; or about one half of the total net annual income of the country. But their annual savings, even when allowance is made for the houses and furniture which they buy, and for their contributions to provident societies, form a very small part of the two hundred and forty millions which are added yearly to the wealth of the country. It is true that in those parts of England in which wages have long been high, many artizans own their own houses and a rise in wages leads in the course of time to an increase in the will to save, as well as in the power to save. Still it must be admitted that the immediate effect of a rise of wages at the expense of profits will be to check the growth of Material capital.

Secondly, we have seen that an increase in Time-wages, if it leads to such an increase of efficiency that Task-wages are no higher than before, will not lower profits but raise them. In other words a rise in wages almost always leads to an increase of Personal capital; and the increase of the Wages-and-profits Fund depends on the Personal as much as on the Material capital of the country.

A rise of wages may then to a great extent be devoted to adding to the Material and Personal capital of the working classes, and increasing their efficiency: and if so spent, it will not lead to any diminution of the Wages-and-profits Fund, it will not be self-destructive even though it has been obtained at the expense of profits in the first instance.

It must however be noticed that in supposing the rise of wages to increase the efficiency of labourers, it has been taken for granted that the measures by which they obtain the rise do nut to any great extent diminish their efficiency. It has been taken for granted that they do not insist on such regulations with regard to apprenticeships as tend to prevent the number of skilled labourers from increasing; that they do not oppose improvements in machinery, in the process of manufacture, or in its arrangements; and lastly that they are able to attain their ends without many strikes. Of course the expense which workmen incur in a strike must be deducted from their gains before the real value of the rise can be found. But a further and perhaps more important deduction must be made on account of the indirect injury which strikes and the fear of strikes inflict on production. For production is checked, and therefore the Wages-and-profits Fund is diminished, by everything that hampers the enterprise of employers, or that makes them design timidly and carry out their designs imperfectly the more their energy is diverted from their proper work in production to vexatious controversies with those whom they employ, the less will be the Wages-and-profits Fund.

We conclude then that it is not impossible for trades unions to enable labourers to obtain a general rise of wages which they would not have otherwise got; but that this rise will itself bring into operation causes which will lower wages, and that it cannot be permanent, unless it be obtained by means which do not seriously hinder production, and unless it be used in such a way as to largely increase if not the Material yet the Personal capital of labourers, and to add a great deal to their efficiency.

§ 4. This result helps us to interpret the proposition: “Industry is limited by capitals”* - a proposition which has been often used as a way of stating what has been called the Wages-Fund theory. This theory has been much misunderstood, and has been the occasion of many popular fallacies. But even as explained by its ablest and most careful exponents, it seems to be unsatisfactory; because it rests on the assumption that all wages are paid out of wealth that has already been set by as capital. This was first assumed for the sake of simplicity, rather than with the intention of being made a basis of economic science. But from the habit of using it for convenience, some economists drifted into the habit of thinking and writing as though it were a necessary law of nature that wages should be paid entirely out of wealth that has been set by as capital. Starting from this basis they shewed that the circumstances of the country determine in what proportion capital is divided into the two parts, Auxiliary and Remuneratory. They called the Remuneratory capital in the country its “Wages-Fund;” and they argued that no change could increase this Fund, unless it either increased the total amount of capital in the country, or caused the Remuneratory capital to increase at the expense of the Auxiliary.

* See Book r. ch. iii. § 3.

When therefore trades unions claimed to be able to raise wages at the expense of profits, the upholders of the Wages-Fund theory answered that the action of unions cannot increase, but must rather diminish capital; that they cannot alter the circumstances that determine the ratio in which capital is divided into Auxiliary and Remuneratory*; that as they can do neither of these things, they cannot increase the Remuneratory capital which forms the Wages-Fund; and that therefore any rise that their efforts may obtain in the real wages of one trade, must be compensated by a fall of at least equal amount in the wages of other trades. In fact as Mill says, “In the common” or Wages-Fund “theory, the order of ideas is this. The capitalist’s pecuniary means consist of two parts: his capital, and his profits or income. His capital is what he starts with at the beginning of the year, or when he commences some round of business operations: his income he does not receive until the end of the year, or until the round of operations is completed. His capital, except such part as is fixed in buildings and machinery, or laid out in materials, is what he has got to pay wages with. He cannot pay them out of his income, for he has not yet received it. When he does receive it, he may lay by a portion to add to his capital, and as such it will become part of next year’s wages-fund, but has nothing to do with this year’s.

* We have seen that a fall in the rate of interest increases the use of machinery and other fixed capital, and therefore tends to increase Auxiliary capital relatively to Remuneratory. But the exponents of the Wages-Fund Theory seem generally to have overlooked this argument on their side.

“This distinction, however, between the relation of the capitalist to his capital, and his relation to his income, is wholly imaginary .... His own income ...is advanced from his capital and replaced from the returns, pari passu with the wages he pays. If we choose to call the whole of what he possesses applicable to the payment of wages, the Wages-Fund, that fund is co-extensive with the whole proceeds of his business, after keeping up his machinery, buildings and materials, and feeding his family; and it is expended jointly upon himself and his labourers. The less he expends on the one, the more may be expended on the other, and vice versâ.”

This is the reason which Mill gives for wishing to introduce into the theory of wages contained in his Political Economy “those qualifications and limitations which are necessary to make it admissible.” Instead of holding that there is a certain amount of wealth deliberately set by to be used as Remuneratory capital, he regards wages and profits alike as coming from that net produce of land, labour and capital which, after deducting rent and taxes, we have called the Wages-and-profits Fund: he argues that it is the division of this Fund into the two parts of profits and wages which determines how much of the produce shall become Remuneratory capital.

The difference between the new doctrine and the old can be well illustrated by the case of immigration of labour into a country. According to the old doctrine wages have to be paid out of wealth that has already been set apart as capital: and since the labourers will require some raw material and implements to work with, there must be an increase of Auxiliary capital, and therefore a diminution of Remuneratory; and therefore the total amount of wages got by the larger number of labourers must be less than that which has been got by the smaller. According to the new doctrine this result will not necessarily follow: indeed the opposite result is the more probable. For the increase in the supply of labour will increase the net produce of capital and labour, and therefore the Wages-and-profits Fund. It is true that employers will compete less keenly than before for the hire of labour, partly because there is more labour to be hired; and partly because it will answer their purpose to divert some of their means from hiring labourers to providing more Auxiliary capital; and therefore the rate of wages will fall. But it is not certain, nor even very probable, that the whole share which labour gets of the Wages-and-profits Fund will amount to less than before*.

*Cairnes (Leading Principles, Part it. ch. i.) has argued' that there was no sufficient reason for Mill's changing his position. He has ably restated Mill's old theory and guarded it against some common misinterpretations; but he has not caught the point of Mill's new argument, as is shewn by his going on to contend that "an increase ...in the supply of labour, when it is of a kind to be used in conjunction with Fixed capital and raw material," would cause the Wages-Fund to undergo "diminution as the number who are to share it is increased."

On the other hand Professors Jevons, Clife Leslie, Hearn and Francis Walker, and Mr Shadwell, have all adopted the same general idea that wages are the share of the produce which the laws of supply and demand enable the labourer to secure (see Jevons' Theory of Political Economy, second edition, preface, p. go). Professor Walker has collected some instructive instances in which it is the labourer who lends his labour in advance to his employer, and not the employer who advances his wages to the labourer. See also Thornton On Labour.

The old method of stating the wages problem led working men to regard their wages as paid out of a fund of capital already stored up, the amount of which is, for the time at least, fixed independently of their exertions. The new doctrine shews how their wages depend not only on the capital which others have stored up, but also, and to a greater extent, on the efficiency of their own work.

Chapter VII.
Influence of Trades Unions on Wages (continued)

§ I. WE have seen that it is not impossible for labourers, by forming themselves into trade combinations, to obtain a general rise in wages; and that this rise may be maintained, provided it is so used that the Wages-and-profits Fund is not diminished. The next point to be observed is that when a combination in any trade obtains a rise of wages, this rise is seldom entirely at the expense of profits; the employers are almost always able to shift nearly the whole of the burden on to others. Part of it falls on the consumers of the things produced by the trade, and many of these generally belong to the working classes; part falls on other trades which are directly or indirectly associated with this trade in the process of production.

For instance when carpenters by a strike or a threat of a strike get their wages raised, the master builders seldom bear the chief part of the burden. The rise is generally got when the price of houses is rising; and by checking or threatening to check the supply of building it causes a further rise in the price of building; and this rise in price checks the demand for building. A check in the demand for building checks the demand for the labour of bricklayers, masons, plasterers, painters and other workmen who are directly associated with the carpenters under the same set of employers; and also for the labour of brickmakers, quarrymen and others who, though not working under the same employers with the carpenters, are indirectly associated with them in the work of production. In fact the interests of the carpenters are related to those of the bricklayers and masons, and to those of the brickmakers and quarrymen, in very much the same way as they are to those of the master builders. In matters that affect the demand for building and its price, the interests of all these classes are in harmony; and in questions connected with the way in which this price is shared, the interests of carpenters are very nearly as much in opposition to those of bricklayers and masons, and to those of brickmakers and quarrymen, as they are to those of master builders. Given the price which employers can get for their buildings, the cheaper they buy their bricks and the cheaper they can get them laid, the more building will they undertake, and the greater will be their demand for the labour of carpenters. And vice versâ a fall in the wages of carpenters, other things being equal, will increase the demand for brickmaking and bricklaying.

To take an instance from another branch of production. a fall in the wages of iron-workers may induce iron-masters to take contracts that they would otherwise have refused; and this will tend to prevent a fall in the wages of the coal-miners who raise the coal for the ironworks. On the other hand a fall in the price of iron has to be borne partly by the different classes of employers, partly by the different classes of workmen whose labour contributes to the production of iron; the greater the share that is borne by one, the less is that which has to be borne by each of the others. There is thus a conflict between the interests of the coal-miners and iron-workers which has once or twice brought about an open conflict between their unions; the iron-workers refusing to work until the coal-miners would raise the coal at a cheaper rate. Such feuds between unions are sometimes very bitter and of long standing; so that one of the chief uses of Trades-councils is to “prevent a union from incurring the odium of a strike through some misunderstanding or petty jealousy between two sections of one trade, which, in the end, if it had taken place, would not only have injured both parties to the dispute, but would also have been ruinous to the employer, although he was in no way concerned in the cause of the quarrel*.”

* Howell, ch. x. § 20.

§ 2. But yet this opposition between the interests of associated groups of workmen does not very often attract the attention of the workmen themselves. When one group strikes for higher wages, the others give moral if not material support to the strikers. In each of our great industries the employers are often seen arrayed on the one side, while several classes of employed are arrayed in a strike on the other, and the remaining classes are observing a friendly neutrality towards the strikers.

The causes of this fact are not far to seek. Firstly, the employers are regarded as the representatives of capital. They receive as profits the difference between the price of their goods and their outlay on wages, on raw material, and on keeping up their buildings, machinery, &c. Even though it may be known that they do not own nearly all the capital which they use, the distinction between the interest on it and their Earnings of Management does not obtrude itself; and the large sums which they receive seem to be a plentiful store from which higher wages may be paid to labour. These large sums really include not only interest and Earnings of Management, but Insurance against the loss of capital. When an employer fails, his losses are soon forgotten by others; but success, as long as it lasts, forces itself on every one’s observation. And it is not to be wondered at that in trades in which success is unevenly distributed, some working men should forget the failures which have. kept down the average rate of profits, and have prevented the average Earnings of Management from being more than a fair return for the difficulty and strain of the work: while they look greedily at the few large fortunes which have been amassed, as it were, out of their toil.

Secondly, although there is a well-marked class distinction between skilled and unskilled labourers, and although in some districts more than half the employers have risen from among the employed, yet the great social division of the ranks of industry is that between employers and employed. A working man’s friends and relations are seldom to be found among the employers, they are generally scattered about the trades which are associated with his own; and he himself is much more likely to pass over to one of these trades than to become an employer.

Thirdly, although a rise in the wages of one trade tends to hinder a rise in the wages of other trades that are associated with it in the same process of production, yet the wages of all these trades are continually seen to rise together when the demand for the produce of their joint industry increases, and to fall together when it diminishes. It is true that the profits of the employers are generally rising or falling at the same time; but in this case the correspondence is not quite so close; and as the profits of an employer can only be guessed at, while the wages of a workman are definitely known, it is not nearly so obvious. Again the circulation of labour, which tends to keep the earnings in different occupations in their Normal relations to one another, acts much more quickly between two groups of associated workmen than between them and their employers. If the wages of carpenters are abnormally high relatively to those of masons, the supply of carpenters will soon begin to increase relatively to that of masons, and the inequality will be redressed. There is in fact a belief among the working classes that if the wages of one trade rise, there will be, and in fairness ought to be, a corresponding rise in the wages of others associated with it in the same process of production; and this belief is itself a great force. If carpenters’ wages had risen more rapidly than masons, it would be thought fair and reasonable that the masons should take the first opportunity of demanding a rise, while the carpenters might get but little sympathy from other trades if they demanded a second rise. In physical science we may investigate the laws of nature without making any allowance for the influence of popular opinion; but in Economics and in other moral sciences we cannot do this for in the moral world, the belief that a change ought to occur and will occur, tends to make it occur.

§ 3. We have seen in successive chapters*; (i) what are the general conditions on which the power of a combination to raise prices depends; (ii) what are the constitution and modes of action of trades unions; (iii) what are the limits which the Laws of Normal wages and profits impose on attempts of combinations of labourers to raise wages generally at the expense of profits; and (iv) how a rise in the wages of any trade affects the interests not only of their employers and of the consumers of the commodity which they help to produce, but also of the trades which are associated with them in producing it. We may now collect together the conditions, on which the power of a union to obtain a rise of wages for its members depends.

*Book iii. ch. iv.--vii.

Firstly, with regard to the immediate result of any particular trade dispute.

This depends partly on the number and action of the nonunionist members of the trade and on the difficulty of bringing men from other trades into it; partly on the pecuniary position of the union, on the amount of its reserve fund, and of the contributions which can be levied in support of a strike from those not engaged in, it; partly on the resolution of the unionists, and the belief which they and others have in the justice of their demands. The immediate effect which a. check to the supply of labour in the trade will have upon the price that can be got for it, depends, partly on the strength of any combination there may be among employers; partly on the strength of will and the resources of individual employers.; and partly on the loss which they would incur by suspending work. The amount of this loss depends on the margin which the state of the market affords between the price of the commodity which they produce and the outlay of Circulating capital required for producing it: other things being equal, this margin will be the greater, the lower are the wages of the other groups of labourers whom they employ, and the larger the extent to which the Normal Expenses of production of the commodity consist of permanent charges and of the interest on Fixed capital. In estimating the state of the market for the commodity, account must be taken of the stocks which employers have in hand, and of the competition which they must meet from others who are mot affected by the dispute.

The immediate issue of any trade conflict which a union undertakes may therefore depend upon the character and the personal relations of those who play a leading part in it, and upon a variety of other accidents, much more than upon the action of the Laws of Normal wages and profits.

Secondly, with regard to the permanent effects of the policy of a trades union.

If the commodity which the trade produces can be easily brought from great distances, a single union can seldom succeed in causing a great scarcity of it; and the competition of distant producers in the same or other countries exerts an influence on Task-wages against which the union can make but little way*. But when no such competition is to be feared it is doubtless true that if a union can keep the number of those in the trade permanently less than it otherwise would be, the price of the commodity which they help to produce may be so raised by its scarcity, as to make it worth while for employers to pay. them permanently very high wages.

* See Book iii. ch. iii. § 6.

There are cases in which unions have pursued this end with some partial success, by means of stringent regulations which hinder anyone from joining the trade who has not served an apprenticeship, and which limit the number of apprentices who may be taken. But though such regulations may easily be enforced by a trade whose wages are abnormally low, it is a very difficult thing to enforce them when the wages of the trade are abnormally high relatively to those of others. For then both employers and those workmen who wish to enter the trade have so much to gain by evading them, that no regulations can succeed for a very long time in preventing a large influx of new comers. If its Net Advantages are already up to the level of other trades of equal difficulty in the neighbourhood, it cannot succeed .in obtaining a permanent rise in its wages, unless there is a corresponding rise in the wages of other trades.

It follows then that no course of action can be permanently advantageous to any one trades union, unless the general policy of which it is a part tends to raise the wages of all trades. So that if we consider only ultimate and permanent effects, we may apply to the case of a single union the rule which we obtained in the last chapter. We there saw that labourers cannot succeed in getting a general rise in wages by means of trade combinations except on certain conditions. And we may now conclude that a single trade cannot obtain a permanent rise of wages except on the same conditions; that is, unless the rise is got by measures which do not seriously hinder production, and unless those who get the increased wages use them so as to increase their efficiency and to add largely to the amount, if not of Material, yet of Personal capital in the country. No trades union policy is likely to be permanently successful which prevents people from making the most of their faculties, and doing the best work they can; or which hinders the adoption of improved methods of production; or which much increases the uncertainty of business, or in any other way inflicts great injury on employers and capitalists; or lastly, which leads to a great increase of consumption on those passing enjoyments that leave no permanent good behind them. This is almost the same as saying that it cannot be to the permanent interests of a union to adopt a policy which is injurious to the general wellbeing, and therefore morally wrong.

But unfortunately the pecuniary interests of those who are at any particular time members of a union, are often very different from the ultimate and permanent interests of the union; for they may not live long enough for the full working out in their time of those results which, according to the law of Normal wages and interests, must ultimately follow from their action, If they can artificially limit the numbers in their trade so as to keep up the wages in it abnormally high for several years, it will be to their pecuniary interest to do so; even though the regulations by which they attain their end tend to bring about a fall of wages in the long run. Men cannot be dissuaded from making such regulations by arguments addressed to their prudence, as distinguished from their sense of duty.

§ 4. “Except* on matters of mere detail, there are perhaps no practical questions even among those which approach nearest to the character of pure economic questions which admit of being decided on economic premises alone**” and it is alike unscientific and injurious to the public welfare to attempt to discuss men’s conduct in industrial conflicts without taking account of other motives beside the desire for pecuniary gain. The communists assumed that no one should desire to gain at the expense of an equal loss of happiness to others; but the world is not yet ready for applying in practice principles of so lofty a morality as this. The world is however ready, and working men among others are ready, to endeavour to act up to the principle, that no one should desire a gain which would involve a very much greater loss of happiness to others. Of course the loss of £1 involves much less loss of happiness to a rich than to a poor man. And it would not be reasonable to ask workingmen to abstain from a measure which would give them a net gain of £1 at the expense of a loss of 30s. to profits, unless it could be shewn that this loss would react on wages in the long run. But many of them are willing to admit that no union should adopt a course which will raise its own wages at the expense of a much greater total loss of wages to others; and if this principle be generally adopted as a basis of action, then nearly all the evil that still remains in the policy of unions can be removed by such a study of economic science, as will enable them to discern those remote effects of their action “which are not seen,” as well as those immediate results” which are seen.”

* The remarks on economic method (Book r. ch. i. §§ 2 and g) should be read in connection with this section. ** Preface to Mill’s Political Economy.

Much would be gained if all workmen knew that each trade’s production constitutes a demand for the labour of other trades that since about half of all the things that are produced are bought by, working men, working men as a body have to bear about one-half of the loss that arises from any check to production; and that, though one trade may sometimes gain a higher price for what work it does, by diminishing production, yet one half of this increase in price has generally to be borne by other members of the working classes: that there cannot be general over-production, though there is sometimes a disorganization of industry which looks very like it: that if, under the fear of overproduction, each trade diminished its production by, say, one third, the Real wages of labour, the necessaries, comforts and luxuries which working men are able to purchase, would be diminished by about one-third.

Much would be gained if all working men would reflect that unnecessary hindrances in the way of those who wish to learn a skilled trade are wrong; and that it is almost always wrong for one union to seek to obtain a rise in wages by a line of conduct, the general adoption of which would diminish the production of wealth, and therefore cause in the long run a general fall of wages.

Unfortunately working men are able to say with too much truth that many of their superiors in the commercial world sometimes speak and often act as though nothing more were to be expected in business transactions than that a man should avoid dishonesty and pursue his own interests. It is of course true that employer have sometimes no choice except between limiting their production and incurring very heavy losses; but they are as liable as workmen are to combine to limit production even when there is no such urgent occasion, and thus to seek a small benefit for themselves at the cost of a great loss to the rest of society. And, as working men are sometimes tempted to enter into a contest to get wages much above their Normal level, wages which can only be maintained even temporarily by, harsh regulations and continual conflicts, involving a great diminution of the Wages-and-profits Fund, and inflicting grievous injury on society; so employers are sometimes under a like temptation to strive by arbitrary means for unduly high profits.

The duties of different classes of industry to one another may profitably be discussed from the pulpit, in Social Science Associations, in Chambers of Commerce, in Trades Councils, and in Tradesunion Congresses. And there is much to be gained from all movements which tend to bring employers face to face with their employés, to talk over peacefully the economic and moral grounds of any claims that may be advanced: this is the work of Boards of Conciliation.