[Note]

 

See accompanying letter

I have been considering your "Relation" further and would like to suggest the following points. I am not sure how important they are in fact.

There are three possible cases:--

Case I. There is unemployed labour--(or a growing population of workers, the important point being that there is labour to be absorbed)--and unemployed capital, (i.e. firms are closed down or working at very low capacity.)

Case II. There is no unemployed labour--(& a stationary population of workers)--and no unemployed capital.

Case III. There is unemployed labour--(or a growing population of workers)--but no unemployed capital.

(For formal purposes there is a Case IV of no unemployed labour and <some> unemployed capital; but this is of no practical importance.)

I suggest that it is only in Case III that the difficulties due to your Relation are present in their pure or acute form, assuming throughout constant technique and tastes, i.e. no inventions. In Case I if consumption is increasing, labour and capital are being absorbed into production and there is no change in the proportion of labour to capital. In this case your Relation does not hold good, since the proportionate rate of increase in the capital stock in existence need not be equal to the proportionate rate of increase of consumption. If net investment is taking place at an increasing rate in order (because of the multiplier) to cause consumption to increase, then investment is only taking place to substitute capital for labour by the construction of new types of machinery, i.e. because the rate of interest is falling or the price of machines is rising less quickly than the wage-rate of labour. When therefore the point is reached at which all labour and all capital are absorbed, there need be no crisis; for the demand for new machinery does not now fall off because the rate of increase in consumption falls, since the construction of new machinery was only taking place to substitute capital for labour, while the rate of increase of consumption was being met out of the stock of unused capital. To prevent a crisis all that is necessary is that the rate of interest should go on falling as before to cause a continuing substitution of capital for labour. If at the start the amount of unemployed labour bears a smaller (or greater) ratio to the stock of unemployed capital than the amount of employed labour bears to the stock of employed capital, we shall reach a point at which all labour is employed while some capital is still unemployed (or all capital is employed and some labour still unemployed). i e. Case I becomes case [b] IV or case III.

In case II, since there is no unemployed labour to absorb, any increase in consumption must be due to the substitution of capital for labour, i.e. investment must be taking place because the rate of interest or the price of machines in relation to the wage of labour is falling and capital is being substituted for labour. In this case to avoid a crisis all that is necessary is to preserve the rate of fall of the rate of interest and the rate of substitution of labour for capital.

In case III your "Relation" does prevent the most devastating problem, which I had never seen before. Here at first there need be no substitution of capital for labour and no fall in the rate of interest or fall in the price of machines in relation to the wage rate of labour. The proportionate rate of increase in the capital stock will in these circumstances = the proportionate rate of increase in consumption = the proportionate rate of increase in employment. But as soon as all unemployment is absorbed--(or starting with full employment, as soon as the population of workers grows at a smaller proportionate rate of increase)--the rate of increase of consumption will fall and the absolute level of investment will therefore fall with all the consequences, which you argue. To preserve full employment in these circumstances it is necessary that suddenly, at the point at which all labour is absorbed, a demand for capital for purposes of substitution for labour should develope as great as the demand for capital which did exist to keep the proportion between capital and labour constant, while unemployed labour was still there to absorb. I agree that it is quite fantastic to assume that the rate of interest can suddenly begin to fall in fact at a rate, which will immediately have this effect.

(Case IV is in no essential particular different from Case II. Here again, any investment that is taking place must be for purposes of substitution of new forms of capital for labour, and there is no question of suddenly developing a "substitution" demand for capital to take the place of a "non-substitution" demand.)

I conclude from this that your "Relation" has shown a very important difficulty, which exists when attempts are made to cure a situation in which there is much unemployed labour and little unemployed capital, (i.e. Case III or Case I turning into Case III.) But when there is either no unemployed labour or as much unemployed capital as unemployed labour, I think your "Relation" very much exaggerates the difficulties of maintaining full employment. I hasten to add that it is extremely likely (a) when there has been prolonged unemployment of labour, so that the existing plant will not have been constructed in a form to employ all the labour available even if it is worked near capacity and (b) when there is a low rate of increase of working population, Case III is very likely to be met.

Would you agree with these remarks?

  1. 1. Harrod, The Trade Cycle ( 1936:8 ), pp. 53-65.

    2. Meade was actually working on a typescript: this is specified by Hubert Henderson, who was sent the same materials--pagination, in fact, corresponds--(letter 524 , [jump to page] ), and by Allen, to whom the copy commented upon by Meade was passed on (letter 567 , [jump to page] ).

    1. a. ALS, two pages on one leaf, in HP IV-745-767/5. The untitled autograph note, six pages on six numbered leaves, is annexed to folder HP IV-745-767/5.

      b. Capitalization inconsistent in original.


Welcome page

top of page

Return to index of this section

Go to previous page

Go to next page