Empirical Studies







 Even if we deny the existence of some general law for accumulation to grow faster than national income, the question posed by Soska will be of great importance to us. If capital intensive technical progress in the years to come becomes reality then it means that our standard of living will grow at a slower rate than if investment saving technical progress could be realized.

From our model we can calculate the rate of accumulation necessary for securing a growth of national income by 8% annually under the following conditions:  

(a) capital intensive technical progress, where the efficiency of capital declines by 2 % per annum, i.e., f = -.02;

neutral technical progress;

(c) capital saving technical progress, where the efficiency of capital grows by 2% per annum, i.e., f =  0.02.


Rate of accumulation necessary for securing 8% growth when the technical progress is10

                                Capital            Capital          Capital
                                intensive          neutral          saving
                0                 20.0                 16.0               12.0

           1                 20.4                 16.0               11.8 
                2                 20.8                 16.0               11.5
                3                 21.2                 16,0               11.3
                4                 21.6                 16.0               11.1 
                5                 22.1                 16.0               10.9

   Soska concludes that “the hypothesis that capital saving technical progress is characteristic of socialism …cannot be considered…as an absolutely valid scientific finding.” This is, of course, a great misunderstanding, which follows from the fact Soska obviously considers the type of technical progress as being objectively given, independently of the decisions in a given economy. If however, we were to consider the type of  technical progress as a result of the so-called “technical choice” 11 then the question could be posed differently – what conditions in the decision-making process determine which type of technical progress will be realized? Certainly, there are many factors influencing the choice of technique but probably the most important among them is “the management model” of the economy.  In this sense we should not ask which type of technical progress is generally characteristic for socialism, but rather, how can we arrange the functioning of a socialist economy so that we can achieve desirable technical progress.   


In his argument  Soska leans almost exclusively on the collection of essays called The Theories of Economic Growth and Present-day Capitalism [11] and focuses primarily on the essay by  Z. Chrupek. Here most of data which he mentions relate to advanced capitalist countries. On the basis of these he arrives at the conclusion that “the analysis of these facts in no way confirms the hypothesis of a fall in investment intensiveness in contemporary economic growth of advanced capitalist countries. “ From this and some further considerations about the situation in Czechoslovakia, he arives at a statement that in Czechoslovakia “obviously factors which increase investment intensiveness will predominate.”

I have many reservations about this procedure. Soska does not provide enough concrete and quantifiable data about development in capitalist countries to be able to prove whether the capital intensiveness in advanced capitalist countries grows or falls. The data about the capital coefficient in the U. S. A. and the U.K. which he takes from the collection of essays of Polish authors, are in direct contradiction to his conclusions. On the contrary a whole series of quantitative analyses, made by Marxist and non-Marxist economists whom we shall mention later), shows that the capital intensiveness in the foremost capitalist countries has been falling  for some time.

Soska does not give any quantifiable data about Czechoslovakia from which one could clearly prove what will be the development tendencies of capital intensiveness. Thus his statement maybe just as hypothetical as the hypothesis which he criticizes. Soska’s hypothesis is proved even less than the others. The quoted  studies of  Sik, Goldmann. and Flek, Nachtigal. Toms and Hajek include quantitative analysis. We may have reservations as to the methods they have used but their conclusions can be disproved only by a more accurate analysis which would prove the contrary. Soska mentions ( page 58) the table taken from the article of H. Flakierski, and then says at once: “The data above testify to a certain oscillation of the capital coefficient in both countries although its trend cannot be evaluated as explicitly falling. On the contrary, up until the years 1928 and 29 this capital coefficient with temporary deviations was growing.”  This, of course, is just making fun of, the reader. From the figures of the table one can observe predominantly falling trend of the capital coefficient from 1909 in the U.K. and from 1919 in the U.S.A. Before this, the capital coefficient either grew or stagnated. 12 This is explained by the fact that industrialization is a change from manual production to machine production and is thus bound to a growth in capital intensiveness. When, however, this process comes to an end, capital saving innovations predominate, because in this way it is easier to lower costs and raise profits.


This “period profile” of the development of the capital coefficient, has been discussed in the wor1d economic literature for a long time. John W. Kendrick13 ascertained on the basis of an extensive analysis that this period profile’ is not only valid for the economy as a whole but also for almost every industry.  He analyzed  33 industry groups. It appeared  that in some industries the capital capital coefficient has been falling (with capital efficiency rising) since 1899, in some since 1909, and in some only since 1919. After 1919, generally, growth of the efficiency of capital prevails; a fall happens only in unique cases. For the whole period 1899 –1953 a fall in the efficiency of capital happened only in three of the 33 analyzed industries, namely in coal, (except anthracite), by an average annual rate - 0. 2%, products of wood by a rate of  - 0.4%, and industrially produced gas by a rate of  - 0.1%. Even within these groups the decline of efficiency  was very small. In the other industries efficiency grew fastest in the electricity industry, where it grew within the period 1899—1953 at an annual average by 4.7%. The efficiency of capital in the production of electricity at that time grew annually by almost 5%. For the whole economy of the U. S. A., Kendrick gives data that are shown in the following left-hand table.  Apart from this he  gives additional data calculated by a different method in an appendix. They also show the similar trend. The falling trend of capital intensiveness of production is also shown  by  the Soviet economists V. Kudrov andG. Shpilko. 15  (The right-hand table) The figures are different, because they use a different method for calculation, but the result is analagous.

The average annual growth rate of
capital efficiency

1899 – 1909


1909 – 1919








1948 – 1953




Capital Intensiveness of the Social Product
in the USA.






































Secondly, Soska bases his argument on considerations of directions in technical development17  and their influence on capital intensiveness. He draws inferences from the character of technical progress in four industries (electric energy, atomic industry, automation, and chemistry) saying that they are basic for the contemporary scientific-technical revolution. But can we be satisfied with this? We must express some doubts about these seemingly persuasive conclusions, based on empirical facts. For example for almost 50 years economic development in the U. S. A. was based, on capital intensive industries such as electricity, engineering and metallurgy yet, nevertheless overall capital intensiveness fell substantially. And how can we reconcile the Soska’s conclusions with the fact that production of electrical energy is a industry where capital intensiveness in the U. S. A. fell fastest?

We must also consider the fact that very fast growing industries can exist which Soska cannot categorize into his structure of the contemporary scientific-technical revolution. Why is electronic left out? Its capital intensiveness is lower than the average, so it influences  his conclusions negatively. It is surprising how boldly he draws his conclusions, when out of the four industries about which he talks, automation raises capital intensiveness only ‘probably’ chemistry lowers it, and “the insignificant share of the atomic industry does not permit formulating conclusions about a pronounced influence of this industry on the capital coefficient in the whole economy.” 18

The influence of the growth of any industry on the average capital coefficient of the whole economy does not only depend on the direct capital intensiveness of this industry but also on the impact which this industry has on the capital intensiveness in other industries linked by input-output relations. Even if production in a capital intensive industry growth quite quickly the average capital intensiveness of the economy may fall  if the use of the products of this industry lowers the capital intensiveness of other industries. Apart from this, the capital intensiveness of the atomic industry and automation is nowadays  high because they are new. One can infer that due to fast growth the capital intensiveness of these industries will  fall sharply.


The idea that technical progress is necessarily related to a growth of capital intensiveness19 has probably its roots in the fact that technical progress in most cases leads to the control of a greater mass of the means of production by human beings in the production process. Thus, the ratio of live labor to the means of production falls. If we consider national income a product of live labor, then it only requires a step to conclude that the relation of national income to capital has to fall concurrently with technical change. But the matter is much more complicated than that, and cannot be grasped by such simple considerations.  The capital coefficient expresses a full (complex) capital intensiveness and not a direct one, so that its magnitude is influenced by complicated feedbacks in the economy. But it is mainly a value relation and not a relation of  physical magnitudes. For this reason the coefficient is influenced by changes in relative prices. Technical progress leads to a relative lowering of prices of investment goods in relation to the price of the labor. 20

In a further argument Soska, claims that the relation between depreciation and accumulation plays a significant role. He shows that the share of depreciation in gross investments grows and this, according to him, “optically” lowers the magnitude of the capital coefficient. But he uses the concept of the capital coefficient in two distinct meanings: sometimes as a incremental (investment) coefficient, and sometimes as an average coefficient. In his tables and considerations  he apparently has in mind the average capital coefficient, i. e., the ratio of national income and total capital. The average capital coefficient does not depend on investments (whether gross or net) and thus, it cannot be distorted by a change of the share of depreciation in gross investments. All quantification of the dynamics of capital intensiveness, based on the average coefficient such as Kendrick, Kudrov and Shpilko, quoted by me, cannot be distorted by this condition and, because of that,  the case where the average capital coefficient falls, while the “real capital intensiveness” (whatever that is) grows, cannot happen. Of course, the situation is different in the case of the incremental or investment coefficient, where  a very significant distortion can happen for two reasons:
First, in a growing economy a certain part of the depreciation fund can be used for net, rather then replacement investments. 

An exact mathematical formulation of this problem was given by E. D. Domar. 21 It follows from his analysis that22 ·    
 a) if the share of depreciation in gross investments is constant (however large)  then it distorts only the magnitude of the investment coefficient and not its dynamics;
b) if the share of  depreciation in gross investments grows, then the growth rate of investment intensiveness is "optically” raised (or the rate of a fall in investment intensiveness is lowered);
 if the share of  depreciation in gross investments falls then the contrary, is true.
Thus we have here the very opposite of what Soska claims. From the table which he shows on page 62, the tendency to raise the share of depreciation in gross investments in the U.S. A. is obvious. It means that in this case a situation cannot arise where the investment coefficient falls, while real investment intensiveness grows.
The second reason for the distortion of the investment coefficient can be an impact of independent technical progress or of so-called noninvestment growth factors.  The distortion occurs because in the investments coefficient the whole increment of national income is related to investment although it was caused by it only partially.  But this further weakens Soska’s case since the greater this distortion is, the more important is the growth of non-investment factors and the less important are investments. This is the very opposite of what he wanted to prove.

In view of the fact that the investment coefficient has these deficiencies, present-day economic science has switched to other instruments of factor growth analysis.  This kind of analysis also confirms the decrease in importance of extensive expansion of capital for economic development.





OK Economics was designed and it is maintained by Oldrich Kyn.
To send me a message, please use one of the following addresses: ---

This website contains the following sections:

General  Economics:

Economic Systems:

Money and Banking:

Past students:

Czech Republic

Kyn’s Publications

 American education

free hit counters
Nutrisystem Diet Coupons