The Economics of Welfare

Pigou, Arthur C.
Display paragraphs in this book containing:
First Pub. Date
London: Macmillan and Co.
Pub. Date
4th edition.

Part I, Chapter IV


§ 1. THE issue, deferred from § 6 of the preceding chapter, as to the precise significance of "maintaining capital intact" has now to be taken up. We are debarred by the conventions we have adopted from counting as capital durable goods—other than houses—in consumers' hands, in spite of the fact that, so far as giving employment to labour is concerned, a motor car, for example, belonging to anybody other than an owner-driver, is indistinguishable from one belonging to a hiring establishment. This, however, is a secondary matter. For the present purpose the precise content of capital is immaterial. However we define it, it may be likened to a lake into which a great variety of things, which are the fruit of savings, are continually being projected. These things, having once entered the lake, survive there for various periods, according to their several natures and the fortunes that befall them. Among them are things of long life, like elaborately built factories, things of moderate life, like machinery, and things of very short life, like material designed to be worked up into finished goods for consumption or coal destined to be burned. Length of life in this connection means, of course, length of life as capital in the industrial machine functioning as a going concern, not the length of life which a thing would enjoy if nobody interfered with it. Coal, for example, if left alone, will last without change of form for an indefinite number of years; but, none the less, the "life" enjoyed by coal in the lake of capital, i.e. the period covered between its entrance and its exit, is almost always very short. All things that enter the lake eventually pass out of it again. Some of them pass, so to speak, in their own persons, embodied as material in some finished product, as when cotton yarn emerges as a cloth garment. But exits are not always, or indeed generally, made in the form of a passage outward of the actual elements that originally came in. When coal is burnt in the process of smelting iron, which is to be used eventually in making cutlery, it is the cutlery, embodying the "virtue" of the coal, and not the coal itself, which passes in person out of the lake. In like manner it is, of course, the "virtue" of machines that are worn out in making finished goods, and not the machines themselves, which passes out in person. In one form or another, however, whatever enters also leaves. There is then of necessity always a stream flowing out of the lake so long as it has any contents at all, and in practice there is also always a stream flowing into it. Its contents at any moment consist of everything that has flowed into it in the past minus anything that has flowed out. It is theoretically possible to make an inventory of them and also to evaluate them from day to day. When we speak, in connection with our definition of the national dividend, of the need for "maintaining capital intact," something is implied about the relation between successive inventories or successive evaluations of the contents of the lake we have been describing. It is the task of the present chapter to make clear what precisely that something is.


§2. For our present purpose it is plain that maintenance of capital intact does not require that the money valuation of the contents of the lake shall be held constant. There are certain sorts of change in this valuation to which everybody will agree that, in our reckoning of the national dividend, no attention whatever should be paid. Thus, if in consequence of a contraction in the supply of money in any year, money values all round are substantially reduced, the money value of the stock of capital will contract along with the rest; but nobody would suggest that this should be reckoned when the national dividend is being estimated. Again, Marshall has observed: "The value of the capital already involved in improving land or erecting a building; in making a railway or a machine; is the aggregate discounted value of its estimated future net income."*54 This implies that, if the general rate of interest rises, the money value of the stock of capital will, other things remaining equal, be reduced. Such reductions are irrelevant to the magnitude of the national dividend. When the value of particular items in the capital stock falls because people's taste for the things they help to produce has declined or because foreign competitors offer these things at a diminished price, that fall also ought not, I think, to be deemed relevant; and the same conclusion holds when the creation of a new item of capital equipment diminishes the value of an existing item; e.g. when the construction of an electric lighting plant depreciates a neighbouring gas plant, or when the introduction of a new type of battleship or bootmaking machine renders existing battleships or bootmaking machines obsolete. In fact we may, I think, say quite generally that all contractions in the money value of any parts of the capital stock that remain physically unaltered are irrelevant to the national dividend; and that their occurrence is perfectly compatible with the maintenance of capital intact.


§ 3. It might seem at first sight to follow from what has been said that the maintenance of capital intact must mean the maintenance in an unaltered physical state of the inventory of things lying in the capital lake. Plainly, if this inventory is in no way modified, capital is maintained intact in an absolute sense; and, if somethings fall out of the inventory, it is not maintained intact in an absolute sense. But, for the special purpose of this analysis, maintenance of capital intact does not mean maintenance in an absolute sense. Certain contractions in the physical stock of capital will have to be held compatible with its maintenance intact, as the phrase is here understood. For, it must always be remembered, our concern is to define the national dividend without parting from the phraseology made familiar by Marshall. Thus, suppose that an earthquake or the onslaught of a hostile nation destroys in one year half this country's accumulated stock of wealth. It would be paradoxical and inconvenient to conclude that our national dividend was thereby automatically rendered negative. We must say rather that the loss is a loss on capital account, not on income account. In other words, this sort of loss, though, of course, it is incompatible with the maintenance of capital intact in any literal sense, is not relevant to estimates of the national dividend. Hence for our purpose maintenance of capital intact must be defined to mean maintenance of it intact, not absolutely, but only when this particular type of loss does not occur.


§4. It may perhaps be thought that this opens the way for an endless string of further prevarications; that, for example, the loss of houses by fire or of ships by storm should be put on the same footing as the losses just discussed. That, however, is not so. All disintegrations of capital goods other than catastrophic destructions of the type described in the preceding section are really incidental to the use of them, and are involved in the production of the dividend. This is most obviously true of the ordinary wear and tear which machinery and plant undergo when carrying out their functions. Weathering by lapse of time apart from use is in like case; for a necessary condition of use is subjection to the passage of time. Even the accidents of fire and storm are in like case; for the use of houses implies their subjection to the risk of fire and the use of ships their subjection to that of storm. The national dividend is not truly reckoned until allowance has been made for the replacement of all these types of capital loss. Maintenance of capital intact in our sense is thus equivalent to maintenance in an absolute sense save only that provision must not be made against destruction by "act of God or the King's enemies."


§ 5. We have now reached the conclusion that the maintenance of capital intact for our purpose requires that all ordinary physical deteriorations in the capital stock should be made good. But what exactly do we mean by making good? When a capital stock deteriorates, e.g. through wear and tear, its material components do not disappear from the world, but merely become rearranged in a way that renders them less useful to mankind. Thus what has really disappeared is a physical arrangement embodying a certain sum of values, which we may for convenience measure in money. To make this good there must be added to the capital stock new arrangements of matter embodying a sum of values equal to this sum. Thus, if a machine becomes worn out and has no value at all, we require to add to the capital stock something whose value is equal to that which the machine, had it remained physically intact, would have now. The original cost of the machine, whether in real terms or in money terms, is not relevant. Thus it may have cost a thousand pounds to make; but, if its wearing out reduces its sum of values below what it would otherwise have been, not by £1000 but by £500 or by £1500, the replacing machine must have one or other of these values, not the value of £1000. It follows, inter alia, that if any piece of capital stock, e.g. the equipment of a steel works, has fallen in value in consequence, say, of intensified foreign competition, and if wear and tear depreciates the equipment 10 per cent per annum, to maintain capital intact we need 10 per cent, not of the original cost of the equipment nor of its present replacement cost, but of its present value. If the foreign competition is so strong or if popular taste has turned away from the product made by the equipment so completely that its value has become nil, physical deterioration in it by wear and tear or lapse of time involves no loss of value, and so calls for no replacement. Maintenance of capital intact for our purpose means then, not replacement of all value losses nor yet replacement of all physical losses (not due to acts of God and the King's enemies), but replacement of such value losses as are caused by physical losses other than the above.


§ 6. It is easy to see that different sorts of capital stock undergo losses of this sort—flow out of the capital lake—at different rates. Working capital—materials, coal and so on—usually only has a few months of life in the lake; and fixed capital a life of a number of years, greater or less according to its nature. Hence the annual amount of replacement that is needed to keep £1's worth of working capital intact is much larger than the amount required in respect of a £1's worth of fixed capital. Professor Mitchell quotes some figures which suggest that in the United States the part of fixed capital represented by "movable equipment"—machinery and so on—engaged in industry and agriculture is about equal to the stock of working capital, each being valued at 9000 million pounds.*55 If the normal life of movable equipment be put at ten years and that of working capital at one year, this implies that the annual replacement of fixed capital must, to maintain it intact, amount to 900 and the annual replacement of working capital to 9000 millions—ten times as much. Hence if, in any year, a £100 millions worth of deterioration in fixed capital is made good by adding to the capital stock a new 100 millions worth, not of fixed but of working capital, the value of annual replacement needed in future years to keep the aggregate stock of capital intact will be pro tanto increased. In converse conditions it will be diminished.


§ 7. In conclusion a word may be said about the effect of failure to provide sufficient replacement to maintain capital intact. Let us suppose that we start from a condition of stability; that continually for a long time past savings at the rate of 2 million £ per day have been required for maintenance, and have in fact been forthcoming. Something occurs, as a result of which henceforward only 1 million £ will be forthcoming. It is obvious that the level of the lake must fall. But it will not continue to fall indefinitely. For, as a result of the decline in the inflow, the outflow also must diminish, since the progressive fall in the stock of capital involves at the same time a progressive fall in the daily wastage. Presently the outflow will so far decrease that the reduced inflow of 1 million a day suffices to replace it. The contraction in the capital stock thereupon comes to an end and a new equilibrium is established. The period of time that will need to elapse before this happens will depend on the proportions existing initially between capital items of various lengths of life and on such changes as may take place in these proportions during the course of the decline. If the failure to provide replacements is carried to the point that henceforward none whatever are forthcoming, the stock of capital must, of course, eventually disappear altogether. Items with a short remainder of life will become extinct first; then others and yet others. The out-flowing stream will diminish to a smaller and smaller trickle, until, with the demise of the longest-lived item, it and the lake from which it came alike go dry. In this event, however, humanity will take no interest, for the demise of the last capital item will certainly have been preceded by that of the "last man."

Notes for this chapter

Principles of Economics, p. 593.
Business Cycles (1927), p. 93.

Part I, Chapter V

End of Notes

Return to top