Labor-pain and value vary together only when we are comparing goods produced by laborers within a competing group. Laborers in one group do not compete with laborers in another group. There is perfect competition in the capital market, however, and so capital costs ("abstinence") are perfectly correlated with value, to the extent that capital enters. Cairnes seems to think that the whole difficulty with his real cost doctrine comes from the failure of competition. In fact, however, it comes also from the inequalities in wealth. And even in his highly competitive capital market it is equally true that abstinence, or even marginal abstinence (a term which Cairnes does not use) has no constant relation to amount of capital accumulated, value produced, or interest received. The cost theory breaks down at every point when it runs in labor-abstinence-risk terms. So generally has this been recognized, that the cost theory has generally given way to the utility theory, and cost doctrine when it appears in modern economics is either the very superficial money-outlay notion of Mill, or else the Austrian cost doctrine, later to be discussed, which is still a pecuniary concept. I have elsewhere undertaken to show (_Social Value_, chs. 3-7, and the ch. on "Marginal Utility," _infra_) that these defects of the "real-cost" theory, are just as much in evidence in the utility theory. The failure of the real cost theory of value is by no means a vindication of the utility theory. Both have the same vice--the effort to combine into a homogeneous sum a lot of individual psychological magnitudes measured in money, when the money-measure has a different psychological significance for each individual, and so comparison and addition are impossible. But in any case, the real cost doctrine of the Classical School has failed, and so cannot serve as the basis of the theory of the value of money. Obviously the money-outlay cost theory of Mill cannot explain the value of money itself. The marginal cost of producing twenty-three and twenty-two hundredths grains of gold will always be a dollar, however the dollar may vary in value. Indeed, in general, the assumption of a constant value of the money-unit is implied in the monetary cost concept. Cost curves are _supply_-curves and the reasoning already given as to the need for assuming constant value for money in the supply and demand concept will apply here. Costs function in value-determination only by checking supply. Rising costs tend to mean a lessened supply. But if the cost-curve is rising _because_ of a fall in the value of money, then the demand-curve will be rising also, and production will not be checked. The general law as to the relation of cost to demand and supply assumes a fixed value of the unit of cost, the dollar.