If we can have a process of increasing per capita income started somehow, then it is possible to accelerate the process of capital formation by drawing upon a larger and larger proportion of the increment of per capita income over time. Suppose, to start with, that the rate of saving is 9 per cent of national income per minimum resulting, on the assumption of an output-capital ratio of 1:3, in an increment of national income by 3 per cent per minimum. Suppose also that the rate of growth of population is 2 per cent. We have then an increment of per capita income by 1 per cent, which is one-third of the rate of increment of aggregate income. We can thus have our people save as much as 33-J per cent of the increment of aggregate income over the year and yet maintain the same average level of consumption. This is something which is very significant. Let us look at the figures: 9 per cent is the rate of saving, to start with. If the saving out of the increment of income is also 9 per cent, we have a constant average rate of saving and, on the assumption of a constant output-capital ratio, a constant rate of growth also. On the other hand, if the saving out of the increment of income is more than 9 per cent, the average rate of saving is raised, and there are possibilities of a higher rate of capital formation and a higher rate of growth. We can accelerate the process by saving on the average more and more out of the increment of income till it reaches 33| per cent of the increment of aggregate income. It is only when we go beyond this that the average standard of consumption goes down. There are thus two limits between which the saving out of the increment of income (marginal rate of saving, as we may call it) may be permitted to lie one set by the original rate which is the minimum needed for preventing a fall in the rate of growth and the other set by the difference between the rate of growth of national income and the rate of growth of population which is the maximum we can have if the average standard of living is not to fall. So Capital of labour is most important. The more we move towards the maximum point the larger will be the income increment itself, out of which we shall be asked to save and the greater therefore will be the possibility of raising the rate of saving and hence the rate of growth. The advantage of having the marginal rate of saving above the average rate is that it has a cumulative effect on the rate of growth, making it rise progressively. And this is what an under-developed economy needs, at any rate over the period unemployment due to capital shortage still remains a problem. This brings us to a concept which has an important bearing on our present analysis the concept of labour-capital ratio. Given a certain technology, there is a certain quantity of labour that a piece of capital good can accommodate if worked at full capacity. If we know this relation between each type of capital good and the number of labourers that it can absorb at full capacity, we can work out the average labour-capital ratio for the economy as a whole in the same way as the output-capital ratio is worked out. What the labour-capital ratio tells us, in other words, is the amount of capital in value terms that is used in the economy per person employed. Let us fix upon this concept of labour-capital ratio L/C, where L stands for labour and C stands for capital. This ratio has important bearing on the employment aspect of economic development, even as the output-capital ratio (O/C) has on the income aspect. Before we go on to this employment aspect, let us show how the output that we can derive from a unit of labour is given by the ratios that we have already gone into. Let us have a break-down of the ratio of output to labour in the following way The equation tells us that the output that is associated with a unit of currently employed labour is a function of two ratios—the output-capital ratio and the labour-capital ratio. It rises if the value of O/C rises, L/C remaining the same, and falls if the value of L/C rises, O/C remaining the same. Again and this is more realistic it rises if the value of O/C rises more than that of L/C and falls if the value of L/C rises more than that of O/C. All this has an important bearing on the choice of technique in economic development. It is sometimes suggested that in an under-developed economy which suffers from shortage of capital, we should adopt techniques which are associated with high output-capital ratio. This seems on the face of it to be unexceptionable. It seems an easy deduction from Harrod’s equation <? = S. O/C, namely that if S (rate of saving) remains constant, then we raise the value of G (growth rate) by raising the value of O/C. When one considers this, however, along with the equation set up above, one finds that the deduction is a little too facile. If we can raise the value of O/C through technological and organizational improvements which make capital more efficient, well and good. But if we do it by choosing a technique which is just labor-intensive, i.e. by raising the value of L/C, then the procedure becomes dubious. For while we raise the output per unit of capital, we lower the output per unit of labour. So Capital of labour is most important.