Page images

The gain of the class immediately above 900 marks in the ten years covered by the above statistics is very notable. The limit thus set for taxation is so low as to include not merely the salaried, but also some of the wage-labor class, which is also considerably increasing its income. Relative to all taxed incomes those of the middle classes are losing ground, despite the rapid increase of the salaried class. These statements are in terms of absolute income classes. But the average income is slightly lower at the later than at the earlier date, which makes a relative test not indispensable.* A comparison of the logarithms of the quantities involved of course rather strengthens the conclusion.t

The incomes taxed are unhomogeneous in character. It is impossible to separate incomes from labor from those from property, so as properly to weight the latter. The most exacting attention to the requirements of method cannot make up for these defects in the material. We can only refer to the known tendency towards an increase of the salaried classes, marked in Germany as elsewhere, and also, for these statistics, to the tendency towards higher money wages, as strengthening

* The facts that the average of incomes included is somewhat lower at the later than at the earlier date, and that the number included is a larger proportion of the population, suggest an increase of administrative efficiency. Which way this would affect the indications of the statistics as regards concentration, and how much, is a question. It looks as if there is no neglect of the smaller incomes. If increased adequacy is greatest here, the inference as to concentration is strengthened.

† Logarithms of numbers in income classes of 900 marks and above are:

[merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small]

the inference from the greater increase of the large incomes that there is concentration of wealth. It is these large incomes that are incomes from property. The actual fortunes are of course, if the interest rate is falling, increasing at a faster rate.

We should expect to find a reflection of the effects of the rapid industrialization of Germany in just these contemporary Prussian income statistics. The Berlin and Rhineland districts show the most pronounced tendency to concentration.* As regards the symptoms observable in other countries, pointing to a tendency toward a rehabilitation of the small-propertied class, such a tendency could hardly affect these statistics noticeably, and is perhaps less to be expected to show itself in Germany as yet on account of that country's being in an earlier stage of the abstract-property development.f

Only the most satisfactory bases of comparison and tests for the growth of fortunes have been used in the foregoing. Aside from lack of space, other methods, for example based upon statistics resulting from general property and house taxes and the like, are less accurate, as the data are even more impeachable than the above.

One point of general evidence, not of a statistical nature, may be added. Prices paid for curios and articles of luxury by the rich have enormously increased in the last two generations. Fine pictures and fine furniture, fine building sites and fine houses, have, wherever scarcity checks production and stimulates ambition, been bid up to astonishingly high prices. The buying of paintings of “old masters” and others by American

* See the article cited, Bul. Inst. Internat.

† British income statistics for Schedule D (that is, “profits from businesses, concerns, professions, employments,” etc., classified by size) are often used as evidence of decreasing concentration of wealth (especially by Giffen and Goschen and by others following them). They show gains for middling incomes of the professional and salaried class, independent entrepreneurs, and the like. But they are not highly satisfactory material, since they are fractions of incomes, not the whole incomes of the persons entered (Report of Commissioners of Inland Revenue for 1902-03, p. 207), since they are also from the most unreliable of the schedules for accuracy (Ibid., p. 173), and since they are affected by transfer into other classes (Jour. Roy. Stat. Soc., 1888, p. 640). They are quite unsuited for our purposes, moreover, because the major portion of income from property, and especially abstract-property income, is entered under the other schedules.

millionaires in the last few years, at prices which competing European millionaires and princes could not afford to pay, is highly significant both of gratifying improvement in the ambitions of some of our multi-millionaires and of the fact that we possess the multi-millionaires.

The conclusion of this examination of the evidence relating to concentration of riches is: there has been, on the whole, in the last half-century or so, a tendency to concentration in the leading countries of the Occidental world. There appears to be more difference of opinion on this question than the evidence justifies, because the question of concentration of wealth is confused with that of concentration in the distribution of incomes. The poor are not getting poorer in the sense of receiving smaller incomes. Wages have demonstrably risen in the last half-century. But the concentration of wealth is a different question. Even in property the poor have, on the whole, probably gained absolutely. The question whether there has been enough rise in wages and salaries and enough absolute gain in property, both combined, to compensate or more than compensate for the relatively greater gains in property by the rich is irrelevant to our present purpose.

The situation, even when strictly confined to the question of concentration of wealth, is not without its encouraging features. The tendency is not so decisive, not even so unequivocal, as some would claim. Lately the small-propertied class may have been gaining in weight, relatively as well absolutely. Their saving power is such, and most recently their opportunities of investment have so improved, that their set-back is perhaps only temporary. In manufactures and the like, where the critical field is, industrial organization is adapting itself to subdivided ownership.

That there has been a general tendency to a disproportionate growth of large fortunes in the last half-century is clearly established. This we should expect as the primary though not necessarily the ultimate effect of the modern régime in industry.



Any massing of living organisms into local aggregates interferes in a degree with the freedom of action of each individual organism. The aggregation of human beings in communities occasions a relatively high degree of interference, because merely physical obstruction is supplemented by the restrictions of liberty that are imposed by psychological evolution. The thought and the will of each affects the conduct of all: the concerted will of many often restrains or coerces the behavior of any one or of a few.

Let us call the restraining or compelling power which the social aggregate imposes upon its units a social pressure. Is it possible to investigate social pressure by quantitative methods? Can we measure it? The purpose of this brief paper is to describe a method of obtaining index numbers of social pressure. I have been led to believe that it is quite as possible to determine relative intensities and changes of social pressure in different nations, commonwealths, or minor communities, as it is by the same general method to determine changes in the purchasing power of gold from time to time or from place to place.

In modern communities the social will finds definite expression in legislation, and social pressure is therefore most extensively and precisely exerted through the forms of law. Of these forms the statutes enacted by legislatures are relatively explicit, they admit of classification, and they can be counted. Penalties for non-observance also admit of numerical statement. With relatively few exceptions, they are fines, ranging from one dollar up, or terms of imprisonment for stated periods.

As expressions of social pressure, statutes fall naturally into the following classification. Some prohibit or deny, as, for example, an enactment forbidding a specified form of gambling

or forbidding the sale of intoxicating drinks. Some restrict, by establishing a State monopoly or otherwise, as, for example, the federal statutes creating a postal system. Some regulate, as, for example, statutes governing banking, insurance, or railway transportation. Under the first of these categories we may conveniently make as many as five subdivisions according to penalty. Under each of the other categories we may make the subdivisions “generally," and "local only." Adding to our categories the class “Unrestricted,"—that is to say, all acts or kinds of behavior that the State does not prohibit, restrict, or regulate, save in that minor degree which is described as “keeping the peace,”—we have a gradation from Prohibited” down to “Unrestricted” of ten degrees, which may be designated by the numerals from ten down to one. In like manner the law may make certain acts or relations compulsory; it may create an agent or commission as an organ of government with power; or it may merely maintain or aid an enterprise at public cost. All remaining acts and relations may be classed as “Voluntary.” The category “Compulsory” may be subdivided into five grades according to penalty, and the category “Established as an Organ of Government” may be subdivided into the grades "with maximum," "with medium,” and “with minimum power.” We may thus again get ten degrees of social pressure, which may be designated by numerals from ten down to one.

Using these gradation categories as column headings, we may then, as in the accompanying form, construct a table of legal provisions in force in any commonwealth, so as to show their significance as manifestations of social pressure.

It will be observed that the number entered against any item is the number that stands at the head of the column to which as a fact of classification the provision in question is properly assigned. The investigator, therefore, does not mark according to any subjective standard. The values are assigned in advance. He has only to be careful to avoid errors of classification, and his decisions can be checked by other investigators or by his readers, his data always being accessible.

« PreviousContinue »