The Online Library of Liberty

A project of Liberty Fund, Inc.

• Preface to the English Edition
• Book I: Utopias and Communistic Experiments
• Chapter I: Plato’s Romance
• Chapter II: The Kingdom of the Incas
• Chapter III: Sir Thomas More’s “utopia” and Its Applications
• Chapter IV: AndreÆ and Campanella
• Chapter V: Paraguay
• Chapter VI: Morelly and the “code De La Nature“
• Chapter VII: Robert Owen and “new Harmony“
• Chapter VIII: Fourier and the American Phalanx
• Chapter IX: The Oneida Community
• Chapter X: Cabet and the American Icarians
• Chapter XI: American Experiments
• Book II: Socialistic Theories
• Chapter I: Saint Simon
• Chapter II: Pierre Leroux and the “circulus“
• Chapter III: Louis Blanc and the Organization of Labour
• Chapter IV: The Labour Conferences At the Luxembourg and the National Workshops
• Chapter V: The Right to Work
• Chapter VI: Proudhon's Theories
• Chapter VII: Proudhon's Proposed Decrees and the Bank of Exchange
• Book III: The Postulates of German Socialism
• Chapter I: “true” Socialism
• Chapter II: The Claims of Marx and Engels
• Chapter III: The Sources of German Socialism
• Chapter IV: Formula B and the “iron Law of Wages“
• Chapter V: Formula A. Work the Measure of Value
• Chapter VI: Karl Marx and FormulÆ A, B, and C
• Chapter VII: The Discoveries of Karl Marx and the Facts
• Chapter VIII: The Two Classes
• Book IV: The Distribution of Capital
• Chapter I: Bernstein and the Concentration of Capital and of Industry
• Chapter II: The Poor Become Poorer
• Chapter III: Financial Feudalism
• Chapter IV: Real and Apparent Income
• Chapter V: The Distribution of Inheritances In France
• Chapter VI: The Distribution of Landed Property In France
• Chapter VII: Marx' Principles and Small Properties
• Chapter VIII: Limited Liability Companies
• Chapter IX: Cartels and Trusts
• Book V: The Distribution of Industries
• Chapter I: Marx' Theory and the Concentration of Industries
• Chapter II: The Distribution of Industries In the United States
• Chapter III: The Distribution of Industries In France
• Chapter IV: The Distribution of Industries In Belgium
• Book VI: The Inconsistencies of Scientific Socialism
• Chapter I: “scientific” Prophecies
• Chapter II: The Prophets of “catastrophes”
• Chapter III: Admissions of the Apostles
• Book VII: Collectivist Organisation
• Chapter I: Collectivist Organisation and Its Economic Conditions
• Chapter II: The Class War and Political Conditions
• Chapter III: The Deflections of Administrative Organs
• Chapter IV: The Impossibility of Collectivism
• Book VIII: The Actual Class War
• Chapter I: Strikes and Trade Unions
• Chapter II: The Sovereignty of the Strikers
• Chapter III: The Nation At the Service of the Strikers
• Chapter IV: The Electricians' Strike
• Chapter V: The Tyranny of Minorities
• Chapter VI: Destruction of Property and Plant and the General Strike
• Chapter VII: Labour Exchanges In France
• Chapter VIII: The American “labor Unions“
• Chapter IX: The Exploitation of Intimidation
• Chapter X: Compulsory Arbitration
• Book IX: Socialism and Democracy
• Chapter I: The Programme of the International Association
• Chapter II: Socialism Versus Democracy
• Chapter III: How Many Are There
• Chapter IV: The Havre Programme and M. JaurÈ' Solutions
• Chapter V: Social and National Policy
• Chapter VI: Positive and Negative Policy
• Chapter VII: Tactics of the Social War
• Chapter VIII: Against the Law
• Chapter IX: Depressing Effect Upon Wealth
• Chapter X: The Impotence of Socialism

BOOK V

THE DISTRIBUTION OF INDUSTRIES

CHAPTER I

Marx' Theory and the Concentration of Industries

karl marx and Engels said, in the “Communist Manifesto” of 1847, which Socialists acclaim as the beginning of a new era, “the whole of society increasingly divides itself into two hostile camps, into two directly opposed classes, the bourgeoisie and the proletariat.”

§18. The middle classes of former times, small employers, business people, and persons of independent means, artisans and peasants—all fall into the ranks of the proletariat. Their small capital succumbs when brought into contact with the great capitalists.

§25. The progress of industry throws considerable sections of the dominant class into the ranks of the proletariat, and at least threatens their existence.

§31. The modern workman, instead of raising himself by the progress of industry, sinks more and more below the level of his own class.

In short, industry and capital become increasingly concentrated in a few hands, while the numbers of the proletariat continually increase, wages decrease, and the hours of labour grow longer. The last assertion has already been refuted; I now proceed to examine whether the phenomenon of the concentration of industry and of capital announced by the “Communist Manifesto” manifests itself in the United States, in France and in Belgium. I have already cited the figures for Germany given by Bernstein according to the industrial census of 1895.

If three establishments, each of them employing one hundred workmen, only form one establishment at the expiration of ten years, we have a concentration, but if each of them continues to exist and to employ a third or a fourth as many workmen again while doing twice the amount of business, this is not concentration but development and expansion of industry.

CHAPTER II

The Distribution of Industries in the United States

• i Difficulties of the industrial census—Manufactures and hand trades—Difficulty of comparison with previous census—Exclusion of undertakings producing less than $500.
• ii Distribution of industrial establishments—Their numbers from 1859 to 1900—Number of existing establishments in 1900 and number founded in 1900—Classification of establishments according to ownership by individuals, firms, and limited companies—Iron and steel industries—Timber trade—Leather—Paper and printing—Metallurgy other than the iron trade—Tobacco—Liquor trade—Chemical industries—Pottery and glass—Coach-makers and wheelwrights—The hand trades.
• iii Seventeen industrial classes in 1850 and in 1900—Increase in number of establishments in twelve classes.
• iv Number of employees and workmen per undertaking—Classification of establishments according to number of wage-earners.
• v Edward Atkinson—Tendency towards individualism Massachusetts—Individual labour—Conclusion.

I

The management of the industrial census of the United States was entrusted to Mr. S. N. D. North, who is now Director-General of the Census, and it was conducted with all possible care. As a man who is convinced that the professional virtue of the statistician, like that of every man who devotes himself to scientific research, consists in the ascertainment of the truth, he loyally points out the difficulties and uncertainties presented by his labour in the important document entitled, “Plan, method and scope of the twelfth census of manufactures.” (vol. vii).

In former efforts the definition of an establishment was left to the discretion of each agent who contributed to the census. The census of 1900 makes a distinction between Manufacturers and Hand Trades. Mr. North sets up the following criterion by which to distinguish them—he treats every establishment which produces uniform types as belonging to the class of Manufacturers, while those in which every object has a special character are classed as Hand Trades. The maker of readymade clothes falls within the class of manufacturers, while the tailor who makes clothes to order falls within that of individual labour or hand trade. The same distinction is made between the manufacturer of wheels, axles and hoods, and the putting them together in small shops at the places where they are used, according to the convenience of the purchaser. The building trades are included in the Hand Trades because what they produce is for local consumption according to the taste of the building-owner, and their operations are distributed among a number of trades. Dentists, to the number of 3,214, who make artificial teeth, had originally been classed by the census of 1900 with manufacturers, but they protested energetically that their occupation was not mechanical but personal, and they claimed to rank among the “professionals” in the category of the “liberal professions.”

The relations between hand trades, shops and workshops, and retail shops are very close. In 1900 the agents of the census were instructed to pay no attention to restaurants and cafés, funeral undertakers, retail druggists, butchers, laundrymen, carpet beaters, dentists, tailors, milliners, dressmakers, hairdressers, etc. Some of the agents conformed to their instructions, others interpreted them in various ways, so that if the number of hand trades referred to in the census is uncertain, it is undoubtedly very much below the actual number.

It is not known how they are dealt with in earlier census. Under their separate classification in the census of 1900, they are taken as numbering 215,800 establishments. For the census of 1840, 1850, 1860 and 1870 the agents were not to refer to any establishment whose total profits did not exceed $500. In 1890 the returns which deal with smaller incomes than this are disregarded, but what accuracy does this limitation carry with it? Every small trader gives the figure he pleases, generally a smaller one than that of his actual income, for fear of the revenue.

In 1900, 127,419 industrial establishments out of the 640,000 referred to in the census lists fell below the 500 dollar limit. In order to preserve a comparison with the earlier census, they are dealt with separately. But the word “establishment” does not represent units of the same degree; a factory employing 7,000 workmen ranks as one “establishment” as well as a workshop employing five.

II

The distribution of industries according to the census of 1900, is as follows:—

Leaving out of account classes (b), (d) and (e) the figures for 1850 to 1900 are:—

But it is obvious that the number of small establishments under the 500 dollar limit and of hand trades is less than it is in fact, and that the difficulty and expenses of including them in the census will cause them to be eliminated, a measure which is formally suggested by Mr. North.

In his analysis of the census, Mr. North says: “It is obviously impossible to determine from the census dates how the actual number of establishments engaged in productive industry in the United States has been affected by the consolidation of industries and the concentration of employment in large mills and factories.” Undoubtedly small establishments are closed; one sees deserted mills on the river banks. Changes in the place of production and of destination involve disturbances to the advantage or disadvantage of particular localities. New establishments spring up in the same industry every day. Many employers, instead of renewing their old establishments, put up entirely new ones. Nevertheless the number of establishments, in fact, increases in every one of the States of the Union.

Table XI. gives us the number of existing establishments and the total number of new ones opened in 1900:—

Mr. North, in putting forward this table, further states that some of the agents of the census did not exercise sufficient care in the collection of their information. Nevertheless, we gather that the new enterprises opened in 1900 represent 8 or 9 per cent. of the total existing enterprises, and that there is an increase without exception in every industry as well as an increase in every State of the Union.

Table XII. gives us the number of establishments and their output, classified according to the nature of their ownership. The 512,254 are distributed as follows:—

The number of establishments owned by individuals represents 72.8 of the total, i.e., nearly three-fourths; of this number 183,500, or nearly one half, were engaged in hand trades. Their output represents 20.6 per cent. of the total, and the average output of each establishment is $7,176. Firms with two or three partners represent 18.9 per cent. of the total, with an output of 19.7 per cent. These two forms of establishment therefore constitute 91.7 per cent. of the total number, with an output of 40.3 per cent. Co-operative societies may be disregarded, their number as well as their output being insignificant.

Limited companies, which represent 8 per cent. of the number of establishments, produce 59.5 per cent. of the output.

The four great industries which are concerned with foodstuffs, textiles, iron and steel, and lumber are primarily represented by limited companies. Nevertheless in the cotton trade 72.8 per cent. of the establishments are owned by individuals or private firms; in the silk trade 27.3 per cent. are owned by individuals and 31.9 by firms, so that only 40.8 per cent. are owned by companies; similarly in the hosiery and lace trade 38.3 per cent. are owned by individuals and 27.4 per cent. by firms.

In the iron and steel industries, only 4,843 establishments out of 13,896, or 34.9 per cent., are owned by companies; but they produce $1,508,493,000 or 84 per cent. of the total output of $1,793,490,000.

In the timber trade, 28,470 establishments are owned by individuals, 13,906 by firms, and only 4,675 by companies, the value of the output of the first two classes being 521 million dollars and that of the third 508,383 millions.

Of the 16,989 establishments in the leather trade, 12,906 are owned by individuals, 2,990 by firms, and 1,091 by companies. The output of the latter is $257,808,000, that of the firms $208,571,000. The census does not state the output of the establishments owned by individuals.

The paper and printing trades comprise 26,747 establishments, of which 16,332 are owned by individuals, 5,682 by firms and only 4,490 by companies. The output of the first two classes is 233 million dollars out of a total of 606 millions, or 38 per cent.

The manufacture of wood pulp is almost entirely in the hands of companies; it is otherwise in the case of printing works and periodical publications.

The liquor trade numbers 7,861 establishments, of which 1,333 are owned by companies, with an output of 305 million dollars, or 81 per cent. of a total of 425 millions.

The number of establishments in the chemical industries includes 2,206 companies out of a total of 5,444, with an output of 450 million dollars, out of a total of 553 millions.

In the pottery and glass trades, the small establishments predominate, 8,760 being owned by individuals, 3,890 by firms and only 2,200 by companies, out of a total of 14,800. The output of the companies is $157,336,000, or 53 per cent. of the total output of $293,564,000.

In the metal trades, other than iron and steel, out of 16,300 establishments there are 10,060 owned by individuals and 4,167 by firms. The census does not state the output of the privately-owned establishments, but the companies produce 578 million dollars out of an estimated total output of 749 millions.

Of 15,520 establishments dealing with tobacco, 12,800 are owned by individuals, 2,080 by firms, and 358 by companies, producing 128 million dollars, or 45 per cent. of a total of 283 millions.

Carriage builders and wheelwrights, including builders of railway waggons, number 10,113 establishments, of which 2,283 are companies producing $430,885,000 out of a total of 508 millions. The astonishing thing is not that these 2,283 companies have an output of 84 per cent. of the total, but that there should still be more than 7,000 establishments owned by individuals or by firms. Anyone possessed by the idea of concentration would imagine that there was only one waggon builder, Pullman, in the United States. It is to be observed that there are competitors.

Shipbuilding comprises 1,116 establishments, of which 151 are companies with an output in 1900 of $55,571,000 out of a total of $74,578,000.

Of the miscellaneous industries, numbering 29,479 establishments, 4,750 are owned by companies with an output of 641 million dollars out of a total of 1,004 millions.

The hand trades only number 2,690 establishments owned by companies out of a total of 215,800, with an output of $100,646,000 out of a total of $1,183,615,000, but, as Mr. North explains, a considerable number of these establishments are unknown, so that their output is still more decidedly an unknown quantity.

Marx' theory of concentration premises a decrease in the number of establishments. Now out of seventeen classes of industry, grouped without refining upon the character of an establishment in 1850 as compared with 1900, we find that there has been a decrease in only five—agricultural implements, boots, tobacco, woollens and worsteds, and, to an inconsiderable degree, cotton. In every other instance there has been a concurrent increase in the number of establishments, and in the output of each, except in the case of worsteds. According to Marx and his followers, all industry is bound to become concentrated in a small number of establishments. The worsted industry presents a phenomenon of a precisely opposite character. The number of establishments increases, but the number of employees per establishment decreases. In 1850 there were only three, each with a capital of 35 per cent. more than the capital of each of the existing establishments, and a staff which is more numerous by 60 per cent.

Mr. North puts forward a table (p. lxxii.) of sixteen industries. I take the two extreme periods, 1850 and 1900. This is what we find.

In the twelve other industrial classes we see the extent of the establishments growing larger, while their capital and the number of their employees increases as well as their output, but so far from establishments, which existed in 1850, having monopolised production, they have stimulated competition, since we find a greater number of establishments in 1900 than in 1850. The industries which employ the largest number of workmen per establishment are those which already employed the greatest number in 1850, such as the woollen trade, the metal trade, and the cotton and cloth factories.

According to Table XXXIV. (p. civ.) the total number of wage-earners is:—

Dividing this total by 640,000, the number of establishments, we have 8.90 per establishment. Deducting the 127,000 establishments with an output of less than $500, we have 5,705,000/513f000 = 11 workmen and clerks for each establishment.

While in the eyes of those who only judge by appearances the whole of the industries of the United States are concentrated in a few gigantic establishments, the average number of wage-earners—clerks and workmen—is 11 per establishment, after eliminating the quite small ones and including the trusts.

The total number of wage-earners—clerks and workmen—is distributed among the different establishments as follows:—

In the 215,814 hand trades, 68,800 employ no hands; 106,000 employ from 1 to 5; 32,000 employ from 5 to 20; and 7,700 employ more than 20. The latter include the building trade and its allied trades. In the manufacturing industries properly so called, there were 41,700 establishments out of 246,000 whose owners employed no workmen at all.

Of the 443 establishments employing more than 1,000 wage-earners, the class of textile industries contains 120, of which the one with the greatest number of workmen is a cotton-mill in New Hampshire, employing 7,268.

The second class of those in which each establishment employs the greatest number of workmen is the metal trades, in which 103 employ more than 1,000. We find one in Ohio with more than 7,400; two in Pennsylvania, which includes Pittsburg, with more than 5,800 and 4,537 respectively; one in Massachusetts with 5,190, and another in Illinois with 5,119.

If we add to these establishments employing more than 1,000 workmen, the 245 others which are distributed among various classes, we find a total of 468 employing more than 1,000 workmen. There are only a few isolated ones which exceed 7,000.

III

So few arguments in favour of the theory of the concentration of industries can be drawn from the industrial census of the United States, that the lamented Mr. Edward Atkinson finds in it a tendency to individualism,1 a result at which he arrived by supplementing the results of the census by the industrial census of the State of Massachusetts. This is the State which contains the greatest number of industrial establishments owned by limited companies. It also contains a much larger proportion of textile establishments than the other States; they are 438 in number, and employ on an average 322 persons. But the total of 250 classes of industry in Massachusetts comprises 29,180 establishments, each of them employing on an average 17 persons, including women and children.

In the State of Pennsylvania, which includes the Baldwin locomotive works with an average number of 18,000 workmen, and Pittsburg with its blast furnaces and steel works, the average is 14.06; in the State of New York the average is only 10.79, and we have seen that for the whole of the United States it is 10.50 for workmen and 11 for workmen and clerks combined.

And yet the emigrants, ordinary workmen who arrive at the rate of a million per annum, present the great industries with a supply of labour ready to hand; and, in fact, Mr. Atkinson pointed out that in a cotton mill in Massachusetts, with which he was familiar, the 2,000 operatives of either sex belonged to sixteen different nationalities. But had they taken away work from the American workman? By no means, for the daughters of the agricultural labourers who had previously been employed in the cotton mills select occupations of a higher order, and leave the mills to the new-comers. They have risen to a higher sphere in the scale of occupations.

Possibly some of the displaced workmen have opened a workshop on their own account, for what is it that nearly all the large establishments supply? They supply manufactured products which have to pass through a workshop before they reach the consumer. The tanner works, not for the public, but for the shoemaker and the saddler; the cloth manufacturer works for the tailor and the upholsterer. The element of individual labour, which requires men and women with an awakened spirit, an observant and accurate eye, and a skilful hand, will not cease to grow in proportion as the tastes of the consumer become refined and his purchasing power increases.

We may therefore conclude that:—

• (1) The industrial census of the United States taken in 1900 does not point to a concentration of industries. If the capital, the personnel and the output of every industrial establishment is greater than they were in the previous census, this is due to their normal development and not to the suppression of competition, since in nearly every branch of industry the number of establishments increases, and has actually increased in each of the states of the Union.
• (2) The census by no means indicates the whole of the share taken by the smaller industries in the economic life of the United States, for it leaves on one side the establishments with an output of less than $500, and the information with regard to hand trades is incomplete.

Despite these gaps, which tend to underrate the apparent importance of the smaller industries, these industries are distributed among a number of establishments which is sufficiently considerable to cause the whole of the establishments, large and small, taken together, to employ an average of only eleven wage-earners, including both clerks and workmen.

IV

Report No. 57 of the American census, entitled “Census of Manufacturers, 1905,” was published in 1907. It deals in fact with the year 1904. Between 1899 and 1904 there was an astonishing explosion of activity in the United States. This is the period which witnessed the genesis of the great trusts. Was the new census of a nature calculated to invalidate the foregoing conclusions? I devoted myself with curiosity to an examination of this question.

To begin with, I gather from Table No. 2 that the number of establishments rose from 512,000 in 1900 to 533,000 in 1905. The number, instead of decreasing, has increased. This first indication is not an indication of concentration. But Mr. S. N. D. North, the Director of the Census, explained in 1900 that the enumeration of small establishments presented considerable difficulties. In 1900 the agents had already been instructed not to include butchers, laundrymen, tailors, milliners, dressmakers, hairdressers, etc., while undertakings whose output was estimated as less than $500 were also excluded. I foresaw that all the small establishments whose inclusion occupied much time and was onerous and not easily controlled, would be omitted, and although the figure of 533,000 occurs in the census of 1905, it really concerns itself with only 216,000 undertakings. (Table I).

Mr. North says that a comparison of these establishments with those of a similar nature included in the census of 1900, yields the result that the number of 207,500 in 1900 has increased to 216,200 in 1905, or an increase of 4.2 per cent.—a fact which is not in accordance with the alleged phenomenon of concentration. Table IX. gives us the position of 14 classes of industrial establishments in 1900 and in 1905. In eight classes the number has increased by 14,500.

In six classes the number has decreased by 5,500.

This shows a total increase of 9,000. We have included shipbuilding among the classes in which the number of establishments has declined; the decrease is only one of ten units, of whose size we are ignorant. The figures, therefore, show that the development of industry in the United States has not contracted them within a small number of establishments from 1900 to 1905 any more than in the preceding period. I may add that Table XVI., dealing with textiles, shows a larger figure for 1905 than for 1900, the manufactures included being cotton, wool, silk, hosiery, linen, hemp, and jute. The figures are 4,312 in 1900 and 4,563 in 1905. Blast furnaces suffered a slight decrease, from 668 to 605, but other metallurgical establishments increased from 215 to 443.

The number of industrial establishments has notdecreased, so that the phenomenon of the concentration of industries has not manifested itself in the United States in the period from 1900 to 1905.

V

The metallurgical industry is one of the most highly concentrated. The figures for blast furnaces and rolling mills are as follows:—

Now, comparing these various elements in 1890 and in 1905, we find that capital has increased 125 per cent. and output 100 per cent. In 1890 output exceeded capital by more than 19 per cent., in 1905 it was 3 per cent. less. The value of raw material as compared with output has remained constant at 68 per cent. In relation to the total output, wages paid to workmen were 18 per cent. in 1890 and 15 per cent. in 1905. The number of workmen has increased by 63 per cent. and the amount of wages by 78 per cent. Compared with the number of workmen, the latter figure shows an increase of 15 per cent.

Taking workmen and clerks together we find that their wages were 19 per cent. in relation to output in 1890 and 17 per cent. in 1905. Their numbers have increased by 70 per cent. and the amount of their wages by 93 per cent.

Despite Mr. Carnegie's United States Steel Corporation, the number of establishments has re mained much about the same. Its promoters claimed that it “controlled” 82 per cent. of the metallurgical output of the United States; it does not at present represent half.

Conclusion.

The number of industrial establishments in the United States has not decreased; the phenomenon of the concentration of industries, according to Karl Marx' formula, has not, therefore, manifested itself in that country.

CHAPTER III

The Distribution of Industries in France

Distribution of the active population—Heads of establishments—Distribution of industrial establishments according to the numbers employed—Number of workmen per establishment—Greater industries in France—Conclusion—Number of patents.

I.

I have already, in my observations upon the distribution of industries in the United States, called attention to the importance to be attached to the method of counting establishments. The statistical results of the census of the population in France1 in 1901 confirm them.

The census of March 24th, 1901, indicates 19,700,000 persons as following some occupation, that is 50.7 per cent. of the total population of 38,961,900. In 1896 the proportion was only 49.3 per cent. The figures for persons of the male sex is 12,911,000 or 65 per cent. in 1901, as compared with 67 per cent. in 1896; and for females the figure is 6,805,000 or 35 per cent., as compared with 33 per cent. The return assumes that this difference arises from the fact that a number of census papers escaped verification in the census of 1896.

Taking the total figures we find the following result:—

The heads of establishments and independent workers would, therefore, number 8,996,900, and represent 45 per cent. of the total. But the figure of 10,655,800 clerks and workmen comprises persons who are included in section 7 (liberal professions) and class 9a (government, departmental, and commercial services). The President of the Republic, senators, deputies, prefects, heads of government offices, ambassadors, magistrates, etc., all go to swell the number of clerks and workmen which is brought into contrast with the number of heads of establishments. This figure includes instructors and teachers, some of whom claim to be simple wage-earners. However, their number cannot be brought into relation with that of the heads of establishments, because their relations are entirely with the State.

It is, therefore, incorrect to say that of 100 persons who followed some occupation at the date of the census, 26 are heads of establishments, 52 clerks or workmen, and 22 independent workers. The figure for the learned professions includes 400,000 persons. The number of independent workers is 36.54 per cent. A medical man, lawyer or artist falls within this category, but I see in the table dealing with the staff of various establishments, that the learned professions are represented by 246,800 persons, of whom 44,500 are heads of establishments, 167,000 are clerks, and 35,000 are workmen. These figures need some explanation. Now I find (vol. iv. p. 124) the following figures:—

Deducting these 1,697,300 individuals from the 10,655,800 workmen and clerks employed in agriculture and industry I find that they are reduced to 8,958,500, that is that they are 38,600 less numerous than the heads of establishments and independent workers. The proportion is destroyed, and instead of being in the minority, they form the majority. If we deduct from the 400,000 individuals belonging to the learned professions the 200,000 who are classed as clerks and workmen, we have a total, in round numbers, of 1,500,000, and we shall then have, on the side of clerks and workmen, 9,155,000, and on the side of heads of establishments and independent workers 9,158,000. We are, therefore, entitled to draw the conclusion that these two large classes are of equal size.

II.

The table on page xix. (vol. iv.) gives the figure of 4,865,000 for heads of establishments; it is with this figure that the number of wage-earners and employees is properly to be compared in order to bring their relative numbers into account. There are not as many establishments as there are heads of establishments. But the economist and the politician require the exact figures for the two classes; they are 4,865,700 heads of establishments on the one hand, and 9,155,000 workmen and clerks on the other. There are, therefore, less than two workmen and clerks to one head of an establishment, the percentage being 65 of the latter to 35 of the former.

In 1896, 2,983,000 establishments were put down in which two or more individuals were working together. In 1901, this figure had been increased to 3,185,000. This would point to the opposite of a concentration of industries, but the report tells us that it proceeds from “new conditions of verification which permitted of the enumeration of a number of family establishments which were not registered in 1896.” Be it so. The report adds: “Leaving on one side establishments conducted solely by husband and wife, or by partners working without assistance, the import of the movement is reversed; the figure for establishments employing at least one workman is reduced to 2,256,000 in 1901, instead of 2,390,000 in 1896, but this decrease has regard to agricultural establishments. In 1896 a large number of children were included in dealing with these, which in 1901 were excluded from the census of the active population. In the industrial sphere, on the contrary, the number of establishments has increased.” But in the table on p. xix. the report disregards all such establishments as employ no outside labour at all. Now in the table on p. xvii., the number of heads of establishments has increased in agriculture as well as in industry:—

Accordingly, there is an increase in each of the three great classes in the number of heads of establishments. This is a phenomenon of diffusion and not of concentration.

The small establishments, which usually include only the members of a household, are too important a factor in production to allow of a clear conception of this phenomenon if they are disregarded, and I call attention here, as in the case of the American census, to the tendency on the part of public statistical departments to eliminate them. It is quite natural that they should do so, the labour involved being too great.

The table on page xix. only contains establishments which employ workmen. It shows a slight falling off in the number of agricultural establishments.

This movement is, therefore, the opposite of a movement towards concentration. The Report states that 573,000 establishments employed from 1 to 20 workmen and other employees in 1896, while for 1901 the figure is 594,000; the number of those employing from 21 to 100 has risen from 15,583 to 17,570, and the number of large establishments employing more than 100 workmen has risen by 600, having risen from 3,668 to 4,268. Out of 100,000 establishments there were more than 619 employing more than 100 workmen in 1896, and 693 in 1901. In trade, the movement is the same; the number of small businesses has increased from 231,000 to 246,000, the moderate businesses from 1953 to 2279, and the larger ones from 143 to 192. Out of 100,000 which employ clerks, 61 employed more than 100 in 1896 and 77 in 1901.

If the number of small establishments had decreased, one might have drawn conclusions in favour of the phenomenon of the concentration of industries in accordance with Marx' formula. But from the time that the number of small establishments, as well as of the greater ones, has been ascertained to increase, we cannot describe the phenomenon as a concentration, but must give it its correct description of a development of industries.

If we now enter into details, we find the following figures (p. 131) relating to agriculture:—

The number of heads of establishments has increased, while the number of independent workers has decreased. We must admit that a number of the latter class have passed into the former. This is the opposite of the constant absorption of the small proprietors by the proletariat which is one of the articles of faith of scientific socialism, so-called.

III.

Volume iv. (p. 191) of the Statistical results of the Census of 1901 contains the figures of the distribution of industrial establishments (exclusive of the carrying trade). These figures were ascertained in the following manner. “The effective personnel of establishments employing more than 5,000 workmen was determined by direct returns. In other cases the number of workmen returned and the estimated numbers do not correspond exactly, the former being 3,606,000 and the latter 3,723,000; the revised total is 3,526,800.” The Report continues: “The figures could not be expected to correspond exactly. In point of fact, it was impossible to classify 3,000 industrial establishments; these are undoubtedly small ones, most probably employing altogether no more than 15,000 workmen.” The first assertion appears to me to be well founded, the second is not based upon any accurate data. “Furthermore,” the Report proceeds, “the figures for industries and for the carrying trade include a large number of common workmen and journeymen who have failed to disclose the establishment which employs them, a proportion of whom are no doubt at work in some industry.” It should be added that all workmen do not work continually in the same establishment. These figures have reference to continually varying phenomena, although they are obliged, from the nature of the case, to appear constant.

But the figure of 3,526,000 wage-earners is lower by 506,000 than the figure of 4,032,000 given in the table on p. 188, in which the figures are carried down to units. The table is preceded by a note which states that the number of workmen in 11,000 establishments is unknown, and it is added that the number of workmen in establishments employing more than 10 has only been obtained by a process of deduction. With these observations I subjoin the table on p. 191.

According to this estimate, 60 per cent. of the workmen are employed in the small or moderatesized industries, and 40 per cent. in the greater ones. In 1896 the percentage was only 36, a difference of 4 per cent. But this difference is insignificant, having regard to the uncertain character presented by these figures. Even if it were strictly accurate, or even greater, the fact remains that in the industrial sphere the number of heads of establishments is 813,000 in 1901, as against 715,000 in 1896. It has, therefore, increased.

IV.

On p. 187 we find a table, the figures in which are not identical with those supplied to the commission (pp. xviii. and xix.); I take those in the table because the following sentence is appended to it: “The average number of workmen per establishment has increased, and this is a primary indication of the concentration of the staff of the various establishments.” But what is the proportion?

Viewed in the light of percentages, the increase in the number of workmen, the primary indication of concentration, averages 30 per 100 establishments (360—330). The maximum increase is in the class of industry and the carrying trade (60 workmen per 100 establishments), while agriculture and the learned professions present the minimum figures (10 and 20 per cent. respectively).

On page 185 the occupations are sub-divided, and the figures of the average number of workmen and clerks per establishment employing more than 1 employee are classified as follows:—

We descend almost directly below 100:—

For the remaining industries the figure is 16 and under. There are, therefore, only five industrial sub-classes with more than 100 workmen per establishment. On p. 186 there is a list of industries showing the average number of workmen employed per establishment for the years 1896 and 1901. It is unfortunate that the number of establishments is not stated side by side with the number of workmen.

If the number of establishments has decreased between the two census, we may say that there has been a concentration. If their number has increased, there has been a development of industry.

As soon as we pass from these five industries we descend to a number of workmen less than 500. We find 18 with more than 200. There are 53 subclasses in all, with more than 100 workmen. Unfortunately, again, the same table does not state the number of establishments represented by them. We only know, from the next table (p. 187) that the number of industrial establishments with more than 100 workmen is distributed as follows:—

The number of all establishments has, therefore, increased, which is evidence, not of concentration, but of development of industry. We perceive at the same time how insignificant a part is played by the larger industries in France. The standard had to be raised to 10,000 in order to obtain the units for comparison, and these were not obtainable for establishments employing more than 2,000 workmen.

As regards small establishments with not more than one workman, these ought, in accordance with Marx' theory, to have disappeared, instead of which they increased from 290,800 to 318,300 or from 4,900 to 5,100 per 10,000. This demonstrates a movement precisely contrary to that of proletarisation. The development of the larger industries has not killed the spirit of enterprise and initiative of the workman who is able to start business on his own account.

In trade (as opposed to manufacturing industries) we find the same phenomenon:—

A table on p. 197 shows the number of wage-earners employed in the different classes of industries properly so-called, with the fraction of this number employed in the smaller establishments. The table only contains six industrial sub-classes in which the workmen employed in establishments with a staff of more than 100 are in the majority:—

This number amounts to 859,000 out of a total of 1,317,0001 workmen employed in establishments which employ more than 100 workmen, or 65 per cent. We see to how small a number of classes the greater industries, in which the majority of workmen are employed in establishments with more than 100 workmen, are limited.

V

A table on page 128 shows the relative importance of the various classes comprising the industrial population, per 10,000 heads of the active population. The proportion in the following industries is:—

These establishments, therefore, representing the admittedly greater industries in France, only include one fifth of the wage-earners, and their relative importance has diminished from 1896 to 1901, because there has been a decrease in the textile industries. But, if we eliminate these, we find the figures to be 784 for 1896, as against 869 for 1901, so that the number employed in these greater industries has only increased by 85 per 10,000, or less than 1 per cent., counting in the total of industries for less than one tenth, viz., 7.84 per cent.

This relative importance has a fictitious as well as a positive side, fictitious because the development of some of these industries, having been called forth by protection, is artificial; and positive, because not only have wants in general continued to develop, but because the metallurgical industry has received a considerable impetus by the Gilchrist method of treating ores, and the rubber industry by the development of the motor trade; and undertakings of these classes require large establishments and a numerous staff.

VI.

All the statistical results of the census of 1901 in France point, not to the phenomenon of concentration to be implied from a decrease in the number of industrial establishments, but to the phenomenon of the development and expansion of industry.

These results are confirmed by the number of patentees. In 1822 there were 955,000. Despite the calling in of patentees by which legislation has benefited minor patentees, the number was 1,660,000 in 1871, 1,862,000 in 1881, 2,005,000 in 1891, 2,154,000 in 1901, and 2,253,000 in 1906. The increase is gradual, but does not suffer any recoil, and the increase is not attributable to any large increase of population.

CHAPTER IV

The Distribution of Industries in Belgium

• i. Definition of an “enterprise”—ii. Number of enterprises and number of employées—iii. Large and small industries.

I.

My information as regards Belgium is drawn from the “Récensement Général des industries et des métiers (analyse des volumes iv. et v.).” Mr. L. March's analysis in 1902 before the French Statistical Society contains the following introduction: “The principal unit for the purposes of the Belgian census is the industrial “enterprise,” but the definition of the enterprise is slightly different from that adopted in France for the establishment (“établissement“) in the census of 1906. In France, the establishment is defined as a group of individuals working in common under a firm name and at a place of business in a particular locality. An establishment may, therefore, comprise, for example, a spinning factory and weaving mill combined, under the direction of the same master, in the same place. The compilers of the Belgian statistics treat two such establishments united in the same building or in continuous buildings as a multiplex or complex enterprise, embracing two divisions of enterprise or two establishments. If an industrial proprietor owns establishments which are not contiguous, in different parts of a town or district, each of these is counted as a separate enterprise.

II.

In October, 1896, there were in Belgium, exclusive of State workshops, 326,089 enterprises in active work, and 11,306 enterprises (or 3.3 per cent.) lying idle. The 326,089 enterprises and divisions of enterprises were distributed as follows:—

• (1) 231,420 or 70.97 per cent. worked by masters or companies.
• (2) 94,334 or 28.93 per cent. consisting of home industries.
• (3) 335 or 0.10 per cent. consisting of “installations” rented by workmen in the public workshops.1

The population engaged in trades and industries numbers 1,102,000 individuals.

Employers manufacturing in their own factories number 232,500; employers who send out work to be manufactured number 5,400; total, 237,900, or 21 per cent. Number of wage-earners 864,200, or 79 per cent.

Persons in receipt of wages or salaries are distributed as follows:—

The enterprises carried on by individuals or partnerships number 324,000; those carried on by limited companies number 2,000. The analysis of the census sets up two categories:—

• (i) Industries properly so-called:—(a) Enterprises carried on by individuals or by actual or nominal partnerships, 229,400 or 99.13 per cent.; (b) Limited or co-operative companies, 2,000 or 0.87 per cent.
• (ii) Home industries or industries carried on in public workshops, individual enterprises or co-operative partnerships, 94,600 or 99.99 per cent.; co-operative societies, 8 or 0.01 per cent.

In the two categories almost the whole are carried on by individuals or partnerships. The number of limited companies is 1,854, but they employ 278,200 wage-earners out of a total of 600,0001 , or 41.90 per cent. of the total number of workmen employed in industry properly so-called. If we deduct the coal-mining industry, in which nearly all the workmen are employed by limited companies, this number falls to 164,000 out of 547,000. The mining industry (underground and surface mines combined) includes 115,800 workmen, of whom 97.48 per cent. are employed by limited companies.

The industries in which enterprises carried on by limited companies employ between 75 and 100 per cent. of the total numbers employed are as follows:—

Then there follow eight industries with less than 2,000 and more than 1,000 workmen, two with more than 500 and six with more than 100.

A great deal is said of Belgian co-operative partnerships. They are 167 in number and only employ 2,100 workmen, of whom 660 are employed in baking and 611 in loading and unloading. The latter are really commercial labour partnerships.

In industry, properly so-called, exclusive of home industries and of industries carried on in the public workshops, 160,400 out of 231,420 enterprises and divisions of enterprises, i.e., 69.32 per cent. or more than two-thirds, belong to the minor industries. In 14,500 one or two masters or heads of establishments work without the assistance of any workmen, members of their family or otherwise. In 17,800 (7.71 per cent.) one master or several masters in partnership work with one or more members of their families, who are very generally children. In the whole of the 231,400 there are only 70,900 or less than one-third, which employ at least one workman properly so-called.

III.

It is difficult to find a standard for the minor industries which is suitable to all branches of manufacture. A flour mill employing 7 or 8 workmen does not fall within the minor industries, while a weaving establishment which only employs ten workmen does.

The directors of the Belgian census take as their empirical standard the figure of four workmen and less; 55,000 enterprises (or 23.76 per cent.) or one-fourth of the total number of those employing at least one workman fall within this standard. They represent a total of 96,000 workmen, or an average of less than two for each enterprise or division of an enterprise. The tailors, dressmakers, shoemakers, joiners, carpenters, bakers, farriers, locksmiths, masons, painters, wheelwrights, slaters, plumbers, seamstresses, milliners, etc., are all included in the minor industries, and a large proportion even in quite the smallest industries.

In Belgium, moderate-sized industries are taken as including establishments employing from 5 to 49 workmen; their number is 13,380, or 6 per cent. of the whole, and they are represented by 173,000 workmen, or 26 per cent. of the total number of workmen. These industries include the businesses of masons, breweries and maltsters, builders, carpenters and joiners, ladies' clothing manufacturers, quarries, foundries, metal workers, etc.

The larger industries, employing from 50 to 499 workmen, include 2,000 establishments, represented by a working population of 295,000, or 146 workmen per enterprise or division of an enterprise. Of a total of 664,000 workmen engaged in industry properly so called, there are therefore 295,000, or 44 per cent. employed in these greater industries. In the coal mining industry, out of 115,800 workmen, there are 86,000 who form part of this group of larger industries. Deducting these, we find 209,000 workmen, representing 142 per enterprise and 38.78 per cent. of the total number of workmen. These large enterprises include the same kind of industries as are found in the United States and in France—the metal trades and constitutional metal works and spinning factories. Of 100 workmen, 44 are employed in the larger, and 15 in the largest, industries—a total of 59 per cent.

By a computation uniting the complex enterprises the Report arrives (p. 23) at the following result:—

This would give 24 per cent., or a quarter of the total number of workmen employed in the largest industries. This is in accordance with the character of Belgian industries—mines, constructional metal works, spinning factories and weaving mills.

This population is classified as follows:—

I have included managers, overseers and clerks in the same class with the heads of establishments, because the workmen look upon them as having interests distinct from their own.

We therefore have on the one hand 71 per cent. of wage-earners as against 29 per cent. of heads of establishments and clerks; that is rather less than three wage-earners for one head of an establishment. The smallest industries are represented by 70 per cent. and the minor industries (4 workmen and less) by 23 per cent. To this must be added the home industries. The greater industries have therefore not stifled the smaller ones in Belgium any more than in the United States and in France. The facts do not confirm the theory of the concentration of industries put forward by Karl Marx in any of these three countries.

I suggested at the sitting of the International Statistical Institute, held at Copenhagen in August, 1907, that the word “concentration” ought not to be employed in the language of statistics except for the purpose of denoting an absolute and a relative decrease in the number of agricultural, industrial, commercial, financial establishments, correlative with an increase in the total activity of the category into which they fall. This suggestion was referred to the Committee which was appointed at this Congress for the purpose of dealing with statistical terminology.

[1]See his “Facts and Figures, the Basis of Economic Science,” 1904.

[1]“Résultats statistiques du Reoensement de la population en France” (1901).

[1]In the Table on p. 191 this figure is given as 1,396,000.

[1]This is the name given to establishments in which men, working either alone or with the members of their families, or with paid workmen, are able to hire a room, and generally motive-power as well.

[1]This figure is slightly less than the figure given above.