This is the fourth chapter of The Causes of Wealth, by Jean Fourastié, translated by Theodore Caplow, 1975.
LEVEL OF LIVING AND THE PRODUCTIVITY OF WORK
THE CONCLUSIONS of the three previous chapters may be summed up in two essential facts:
1) The average level of living of the population of several great nations has been appreciably improved in the course of recent centuries, in spite of reductions in the duration of work and increases in the density of population.
2) This improvement in time has created a disparity in space, due to the fact that the levels of living of the different nations of the world have been raised at very different rates.
Since the disparities in space can be reduced to disparities in time, the essential problem of the level of living is to find out how the improvement in time occurred.
The observed facts allow us to clarify the problem further. During a long sequence of centuries, the level of living seemed to be controlled by the density of the population. Alfred Sauvy has brought to light the fundamental principle of the population optimum. Below a certain density, economic life was or inexistent or at least precarious, but as soon as the population became too dense on a given territory, its level of living decreased because of the decreasing productivity of work. A given square kilometer of land which supports 30 workers and their families with ease, supports 50 less well. As soon as the population reaches a higher figure, let us say 40, life becomes more difficult, even if the 40 men now work harder than the first 30. The least disturbance in the weather may now provoke famine. Famine leads to the struggle for existence. This struggle may first be under the social form of the defense of the rich against the poor, later under the national form of attack by the peoples who have not yet suffered famine upon those who have just suffered it, are going through it, or are about to go through it. The history of the Middle· Ages, and of the ancien regime should be reviewed with this insight into the function of the level of living. Without denying the influence of purely human and political facts, such as inheritances and the quarrels of royal families, the wisdom or the foolishness of kings and ministers, it cannot be doubted that famine has played a crucial part in history, although often unperceived by contemporaries and by historians.
The problem that we face is to discover how this famine mechanism was broken; what were the causes, in the scientific sense of the word, that intervened and stimulated, at least for some nations, a parallel increase of the population and of the level of living, although the two phenomena had hitherto been inconsistent as soon as the population reached a certain density? The same causes must explain how it was possible to raise the average level of living of some populations incomparably higher than the highest of those that have come down to us in records of the past. It is obvious that the search for this factor is of capital interest, not only for economic science but also for political action and, in consequence, for the very life of humanity. If we are able to identify the cause of improvement in the level of living, and if this turns out to be a factor at least partly dependent upon human initiative, we will be able in affecting it, to affect the level of living itself. We shall know how to improve the level of living of any people.
The observed fact that must serve as our point of departure, and that the previous discussion most certainly brought to light, is that the level of living of a population is simply the average per capita production. It is, therefore, by studying production and related factors that we will be able to solve the problem. Taking into consideration the essential element of life, time, leads us from the notion of production to the notion of productivity. We define productivity as production by units of time, or in other words, as the velocity of production.
The study of productivity will turn out to be a satisfactory way of explaining the evolution of the average level of living, the disparities in space, and the differences observed in relation to various forms of consumption. We shall still have to examine the distribution of actual levels of living on both sides of the average. This will lead us to study the function of rents and profits and to show that these rents and profits are themselves results of the phenomena of productivity, in the sense that the countries of low productivity are the countries of high rents. Hence, they are also those with a widely opened fan of incomes.
1. PRODUCTION AND CONSUMPTION
The first theorem of the level of living is that consumption in a closed economic system cannot permanently exceed production. (A closed system is that of any area whose exterior commerce plays a minor part.) This fact is logically obviousand also, more important, it is experimentally demonstrable by the comparisons that can be made for any era and for any country, between total consumption estimated directly or indirectly and total production also estimated directly or indirectly. For example, the average consumption of the masses of the French population around 1725 to 1750 can be estimated from the purchasing power of wages at one and one-half pounds of bread per head and per day. This represents for the kingdoman annual consumption of wheat equal to 50 or 55 million quintals. But the area of all the cultivated land, taking accountof that lying fallow, multiplied by the average output of the time, also gives 50 million quintals. In the same way, agricultural production can be related to effective consumption at any time whatever.
More generally, when we studied the level of living of the French working masses toward 1830, we noted that the French national income at that time was eight billion francs, which gave the sum of 250 francs as the per capita national income, corresponding quite closely, as we have seen, to the subsistence minimum of that period. In order to give each Frenchman, from 1831 onwards, the subsistence minimum of 1939, it would have been necessary to suddenly raise the national product of France from 8 billion to 20 billion francs.
It is quite impossible to raise the average level of living of a people without raising their output. If the effort is limited to a redistribution of income, the most that can be done, - and this with certain difficulties which we shall examine later - is to reduce the dispersion of the observed purchasing powers around the average level; it is not possible to raise the average. For example, it might have been theoretically possible, by a strictly egalitarian distribution, to give to each Frenchman in 1830 the subsistence minimum of Villermé. Even the strictest equalization of income could not have done any more than this. To do better, it would have been necessary to influence production. Leaving for a later section then, the question of the distribution of levels of living on both sides of the average, we shall first study the factors which determine this average.
The second theorem of the level of living is this: Not only does total production determine total consumption, but in the short term the structure of production also determines the structure of consumption. Within a closed economy in the course of a period when stored supplies cannot play an important part, it is impossible to convert one commodity into another rapidly. For example, if 1,000 quintals of wheat have been produced, the consumption of wheat will not be able to exceed 1,000 quintals before the next harvest. Moreover, if more than 1,000 quintals are wanted for consumption the following year, it will be necessary to take advance measures to see that the ground is sown to grain. And as always, the crop will be attendant upon the frivolities of the weather.
In this way, the structure of production and the structure of consumption are closely tied to each other, and can only develop jointly and through their effect on production, a form of action which takes time. More exactly, the problem of the development of the structure of production is a problem of manpower. If we want to increase the production of wheat, we must have more laborers. If we want to increase industrial production, the number of workmen must be increased. To get more refrigerators, more refrigerator-makers will be needed.
In sum, the structure of consumption depends upon the structure of the labor force, and since the structure of consumption changes when the level of living rises, any rise in the level of living of a people presupposes mobility in its labor force, or in other words, is accompanied by changes in occupational distribution.
2. THE NOTION OF PRODUCTIVITY
It appears that the average level of living of a nation cannot be raised unless per capita production itself is increased. Since the structure of consumption changes as the level of living rises, the structure of production must also change as per capita production increases.
These postulates, expressed as functions of time, take the following form: In order for the average level of living of a people to rise, the total annual production per interval of time must increase, and the labor force must change in such a way that the structure of production adapts itself gradually to the changing structure of consumption.
If we use the term productivity for production per unit of time, we see that the level of living cannot rise except by an increase of productivity, unless the duration of work is increased. On the other hand, if the productivity of labor increases in any sector of the economy, while productivity and duration of work in the other sectors remain constant, the level of living must rise because the total annual production has increased. Either the increase of productivity enlarges production in a direction matching the development of the structure of consumption, and then the rise in the level of living takes place without any shift in the labor force, or the supplementary production is not easily consumed. In the latter case, a crisis of over production in this sector will cause a shift of labor to some other unsaturated sector.
The level of living will eventually rise because of the increasing consumption of the goods produced by this other sector. The rise of the level of living, the increase of productivity, and the shifting composition of the labor force are three closely knit phenomena. For each degree of productivity in the different sectors, there is a corresponding value of the average level of living, which correlates to a particular structure of consumption and a particular distribution of the labor force. If, on the other hand, one starts with a profile of average consumption, there will be a corresponding degree of productivity in each of the great sectors of production and a corresponding distribution of the labor force.
Of course, these statements are only applicable to closed economic areas, and indicate only general tendencies. We have noted that differences of custom, climate, and so forth introduce disparities between one country and another. Moreover, the secular trend is greatly influenced in the most backward countries by certain characteristics of the most advanced, so that, for example, although the distribution of the labor force must be about the same in India today as in Europe of 1750, there are already in India a certain number of refrigerators and automobiles, and consequently, though paradoxically, some electricians and some auto mechanics.
The connection between the three phenomena, level of living, labor force, and productivity, being thus recognized, it is not difficult to determine which of them plays the determinant part. Obviously, the level of living can only be a resultant of the two others, and as far as the labor force is concerned, it cannot begin to shift freely from one occupation to another until the time arrives when it is not entirely occupied in the production of food. The mechanism for the improvement of the level of living is then the following: The sciences-first, mathematics, then the physical sciences, finally, the biological sciences and behavioral studies of man at work provide the necessary means. Man uses these means to increase his influence over things. Scientific progress thus engenders technical progress.
Technical progress so defined, that is, in the broadest possible sense of the term, permits an increase in the productivity of work, which is to say, the output obtained per unit of time. As soon as productivity is sufficiently high so that the people are protected from famine, the realignment of the labor force begins. The distribution of labor changes so that the structure of total production responds to the demands of increasing consumption. Historically, this shift of the labor force was accomplished, it should be noted, at the cost of great suffering for the people involved, for in a laissez-faire system, the movement was seemingly imposed by a social situation whose principles were not understood. It was only by financial ruin and by unemployment that men were informed that their production had been refused by the community.
This is not the place to expand upon all of the elements of this scheme which encompasses so much of economic life. It is only necessary to note the relation between the level of living and the level of productivity, by showing that no increase in the average level of living is possible without an increase of productivity. Moreover, empirical observation shows that there is no increase of the purchasing power of the wage earner with reference to commodities that do not benefit from technical progress.
3. PRODUCTIVITY AND PURCHASING POWER
Purchasing power is, as we have seen, the ratio of an income to the price of a consumption item. Up to this point, we have insisted on taking the price of a total schedule of consumption as the denominator of this ratio, for example, the cost of the subsistence minimum, or the typical budget of a working class family in a given country and at a given time. We have already observed that the value of the level of living is closely dependent on the observed structure of consumption, because of important discrepancies in the evolution of the prices of different commodities.
It is now necessary to examine more closely the variations in purchasing power which arise from the nature of the commodities consumed. More exactly, instead of measuring the Level of Living and the Productivity of Work level of living by reference to a total unit of consumption, we must now examine the results obtained by relating income to the price of specific goods and services.
A. Divergent Secular Trends of the Prices of Various Commodities
Classical economic science has always been attached to the study of the general movement of prices. The work of Simiand, of Hauser, of d'Avenel, of Hanauer, as well as the work of Kugner and of other modern economists, is dominated by the idea that a general price level exists. They have searched in vain for the causes underlying this general price level in the movement of precious metals, credit, the issue of paper money, and many another phenomenon. If there were such general price trends, it would be possible to derive an absolute measurement of purchasing power, by relating average earnings to the general price level. The observed facts stand in absolute contradiction to the idea of a general price trend. More exactly, they show that there are as many "general" movements of prices as there are indexes to measure prices. In other words, the usefulness of an index of prices depends entirely on the elements that constitute the index. To really understand the general trend of prices, it would be necessary to set up an index based upon the totality of observed and observable prices in the entire economy. This is either practically impossible and therefore scientifically meaningless, or it results in a calculation of the national income in current prices. By dividing current wages by current national income we do not arrive at an index of level of living. The product of average salary times total employment always gives an approximation of national income.
As a matter of fact, even the studies of the classical economists showed fundamental divergences in the movement of some prices in relation to others. In his work at the École Pratique des Hautes Études, "New and old studies of the general movements of prices from the sixteenth to the nineteenth century," Simiand encountered these divergences at every turn, and it was only by virtue of an extremely powerful preconceived idea that he was able to speak at all of a general price movement. A glance at the diagrams published in this book will show that there is no parallelism at all between the different categories of commodities. His diagram number 5, whose data are reproduced in Table XXIV, shows beyond all possible argument the complete divergence between, for example, the price of spices, with an index of 85, and of lumber, with an index of 910. From the year 1500 to 1875, the price of wood was multiplied by nine, the price of crops by six, while the price of metals was multiplied by only 2.7, that of textile products by 2, and that of spices by 0.85.
How is it possible to avoid the explanation of technical progress for this fanning out in the general trend of prices? The fall in textile prices began with the creation of the first factories by Colbert. The fall in spices is related to the development of navigation and the establishment of overseas colonies. The fall in the price of metals is related to progress in mining and processing of minerals. Technical progress in agriculture, not absent but very feeble until 1789 (the iron axed cart was practically unknown in France before the Revolution), made great strides after 1810. The lumber industry did not benefit from technical progress until about 1900, except for the improvement of transportation.
It is immediately evident that if we measure purchasing power in relation to spices, in relation to manufactured textile products, or in relation to stove wood, we will get results that differ in a ratio of one to ten. The series of Father Hanauer and of Simiand do not include any category of wages, which is a grave omission. It is difficult to see how these authors went about constructing a general index of prices, without taking account of this essential price. If they had included an index of the average earnings of laborers, its value, taking 1450-1500 as 100, would have been around 1,000 for the period 1850-75. Thus the development of the level of living between these two periods exhibits the following values: in relation to meat and other agricultural products, 1.5; in relation to construction materials and textile raw material, 2; in relation to metals, 4; in relation to finished textile products, 5; in relation to spices,12.
These facts confirm the conclusion reached in Chapter Two. The improvement of purchasing power seems to have been extremely variable, depending upon the commodity taken for reference. The improvement is directly related to the effects of technical progress. For convenience of explanation, we shall call those commodities that have benefited from only slight technical improvement in the course of modern history tertiary. We shall term primary or secondary those products which have benefited from great technical progress, reserving the term primary for agricultural products and the term secondary for non-agricultural products. With this terminology, we can formulate the following propositions: The average purchasing power and the average level of living of a people are not improved except in relation to consumption which corresponds to products that have benefited from an improvement in the productivity of labor. Purchasing power grows slowly with regard to the consumption of tertiary goods. It increases to a much greater extent with regard to primary and secondary goods.
These general conclusions are confirmed and elaborated if we attempt to measure the level of living, not only with reference to a group of items of the same type, but also in relation to a good or a service taken alone, as, for example, a quintal of wheat, a square inch of window glass, a haircut, and so forth.
B. The Development of Purchasing Power with Regard to Specific Commodities
In ancient times among the Greeks, an iron cooking utensil was worth more than a woman and more than four cows. In Carolingian times, "a good horse was worth less than his bridle." Thus has technical progress made itself felt since before 1800! A kilogram of pig iron, so unimpressive today, was worth several hundred hours of work in the Middle Ages. The fall of the price of iron in relation to hourly earnings began to be conspicuous after about 1560. However, the long term correlation of the price of wheat and hourly wage rates indicative of a striking causal relationship continued, as we have seen, until about 1800. A great number of products began to decrease sharply in comparison to hourly wages after 1830. This category includes practically all manufactured products. A ton of steel was worth 80 dollars on the world market in 1873. It fell to 20 dollars in 1886 after the introduction of the Bessemer process. A kilo of aluminum was worth 80 gold francs in 1886 and 2.5 in 1910. A meter of merino wool sold at Reims for 16 francs in 1816 and 1.45 francs in 1883. Chilean nitrate brought 187 francs a ton in France when artificial nitrate was offered at 125 francs. A kilogram of sugar was sold at five francs in 1811, the date at which Delessert established the first refineries for beet sugar. It was worth 0.6 francs in 1910. These examples can be multiplied almost indefinitely. They cannot be fully appreciated unless we take into account the slow but continuous increase in hourly wages between 1810 (0.15) and 1910 (0.35), so that there was an improvement of purchasing power in the course of this period not only in relationship to all of the commodities whose price in francs had fallen, but even in relationship to all the commodities whose price had not increased more than two fold.
Instead of piling up examples, we shall examine a little more closely the history of certain items of consumption, some of which have benefited from great technical progress while others have not progressed at all.
Productivity and Selling Price
If it takes 1,000 hours of human labor to produce a commodity, it cannot be sold for long for less than 1,000 times the hourly wage of the workman. Since it still takes a quarter of an hour to cut a man's hair and ten minutes to shave him, the prices of barbers have remained exactly parallel to wage rates for the past 300 years. The price of a haircut varies, according to the luxury of the establishment, between two-thirds and three-thirds of the hourly wage of the journeyman barber.On the other hand, increases in productivity often cause astronomic falls in price.
The price of glass and mirrors
A bead from a necklace of Queen Hatshepsut fixes 1400 B.C. as the oldest date in the history of glass. The glass makers who are represented on the walls of the Hypogeum of Beni Hassoun, blow through pipes of the same model as the glass makers of 1900. Until about 1650, glass panes were made by blowing. It was necessary first to start an enormous bottle. The two ends were then removed and the remaining cylinder was cut along a line and stretched out flat. Towards 1850, important technical progress had been made in handling pieces in process, abolishing the work of those unhappy children whom the workmen called "the fire meat." After the end of the seventeenth century, glass could be made to flow, but it was only in 1902 that the Frenchman Fourcault introduced a workable industrial process for manufacturing window glass by extrusion. Since severe conditions of temperature, of homogeneity, and of transparence were to be met, we can understand how the production of a large piece encountered numerous obstacles before the development of modem techniques. It is quite clear why, until 1900, window glass and mirrors remained very much more expensive, by weight, than goblets and bottles.
In 1702, a single mirror of four square meters required, on the average, 35,000 to 40,000 hours of work. These were the largest which could then be produced. The mirror surface required even more work than the glass pane. It was polished with sand and emery, before being coated. It was not until the twelfth century that the glassmakers discovered how to obtain the necessary temperature conditions to make transparent glass panes or stained glass of a few square centimeters. Before that, no house and no church, however rich it might be, had glass windows. This can be seen by an examination of the paintings of the old masters.
As far as the house was concerned, the fourteenth century was the century of transition. Formerly, there was no glass. The windows were closed only by full shutters, by wooden trellises, or by waxed cloth. The Romans, they say, used horn in their palaces, or mica, a translucent mineral which comes in thin layers. Panels of horn have also been discovered in some French castles. Leaves of oiled parchment were sometimes used. Giotto, Fra Angelico, and all the painters before 1400 show the windows without any covering. The celebrated madonnas Benois and Litta of Leonardo da Vinci in the Museum of Leningrad, and the "Annunciation" of Cima da Conegliano in the same museum, still show windows without any glass, although in very richly furnished rooms. The oldest painting I know that shows a window entirely covered with glass is the "Annunciation" of an unknown old master, at the Museum of Basle, dated by experts at 1470. But this is not a matter of window panes. These are clumsy lumps of glass, translucent but not transparent, such as one still sees in old Alsatian houses.
Glass remained so dear until about 1600 that even the richest houses could only provide panes for the upper part of the windows. The lower part was closed with a full shutter of wood, sometimes lightened by a small opening in the center. This arrangement can be seen in hundreds of paintings of the great Italian, French, and Flemish primitives. One of the most interesting in this respect is the "Annunciation" of the Master of Flemalle, of about 1428, in the Merode collection at Antwerp. One sees very clearly in this picture the glass panes at the top of the window. The lower part is provided with wooden trellis work, while the middle part cannot be closed except by full wooden shutters.
Thanks to the accounts kept by Colbert for the Château of Versailles, it is possible to know the going price of mirrors at the time the palace was constructed. By a privy seal contract executed January 2, 1684, between the royal administration of buildings and Messrs. Pequot and Guymont, agents for the factory recently established at Paris which became the Saint-Gobain Company, the following prices were charged; the first price is regular price and the second : “Special price to King”
14 inch mirrors : 3 livres, 4 sous 2 livres, 15 sous
44-45 inch mirrors 470 livres 352 livres
We see the enormous differences of price occasioned by small differences of surface. The hourly wage of the average workman was then about 18 deniers, equal to 1.5 sous or to 0.075 livres. Louis XIV himself could not have mirrors longer than 45 inches for Versailles. Each of these mirrors was worth 352 livres, or 5,000 hourly wage units of a laborer, or more than 7,000 dollars of our days. In the construction of the fine mansions of the time, the mirrors cost commonly as much as the furniture and the upholstery. For the Hotel d'Avary, in the rue de Grenelle, constructed in 1718, the expenditure for mirrors reached 28,400 livres, as against 16,700 for upholstery and 60,000 for all the woodwork and furniture.
The continued increase of productivity led to a steady decline in the price. The four-square-meter mirror which cost 2,750 livres in 1702 (close to 40,000 hourly wage units), was worth only 1,245 francs in 1845 (6,900 hourly wage units), and 262 francs in 1862 (1,000 hourly wage units). After 1900, the introduction of the turntable method put the bureau with a mirror within reach of the average worker. The four square meter mirror was sold in 1905 for 60 francs, or about 200 hourly wage units. It cost some 20,000 francs in 1955, or 135 hourly wage units. In 1567, the stewards of the Duke of Northumberland still dismounted the glass windows of the castle during the master's absence, but in 1945, window glass had ceased to be a serious item in the budget of a workman.
Thus, measured in four-square-meter mirrors, the purchasing power of a laborer has improved by better than 200 to 1. The Accounts of the Château of Versailles allow us, on the other hand, to measure the evolution of purchasing power with regard to a service that has not benefited from any technical progress until the most recent years-the waxing of wooden floors. To wax ten square meters of floor in 1685 took three sous or about two hourly wage units. To do the same in 1955 by non-mechanical methods cost 300 francs, or still two hourly wage units. The stagnation of purchasing power is empirically observed to be perfect, as theoretically expected.
So we see that economic evolution shows us the progressive reduction of the price of commodities subject to great technical progress in comparison to the price of commodities subject to little or no technical progress. A commodity or a service that has not benefited from any technical progress since 1700 must cost as much today - in hourly wage units - as under Louis XIV.
We are led to ask if the difference that we observe today in the level of living between the different countries of the world does not have its source, and therefore its cause, in differences in the intensity of technical progress achieved in those countries. As soon as it is proved that the evolution of the level of living is caused essentially by technical progress, it is evident that the different countries of the world, not having arrived at the same point along the route of technical progress, must show appreciable disparities of price among themselves. Factually, it is known that in a country like the United States, technical progress has gone much further than in France, while countries like China are near the point where France was in 1850.
If this hypothesis is correct, we ought to find that goods or services not subject to technical progress or at least to very little progress (what we have called tertiary goods and services) cost about the same amount in all countries. Put another way, the price of this given good or this given service, divided by the hourly wage of the place where it is produced, should give the same or similar figures in all countries. By contrast, the price of commodities subject to great technical progress, those which are secondary or primary, should present differences relatively great insofar as the countries compared have arrived at different stages of industrialization.
These expectations can be completely confirmed in the world of today. Towards the end of 1947, the hourly wage was about one dollar in the United States and about 60 francs in France. A frigidaire of 150 cubic decimeters, a secondary product, was worth 150 dollars in the United States, or 150 hourly wage units, and 60 thousand francs in France, or 1,000 hourly wage units. On the other hand, a haircut, a tertiary service, was worth one dollar in the United States, or one hour's salary, and sixty francs in France, or again one hour's salary.
The international money market introduces into these real prices a disturbance which makes them appear anarchic. In effect, the exchange rate is an average rate, intended to permit an approximate equilibrium in the balancing of accounts. This rate of exchange does not serve to equalize the hourly wage rates except between countries that have attained a similar technical level. Between countries of very different technical levels, like France and the United States, the exchange rates make hourly wages appear to be very high in the rich countries and very low in the poor countries. This permits the exchange of products costing three or four hours of work in the poor countries against products that require only one hour of work in the rich countries. At the official rate of exchange, one dollar equals 350 francs. The hourly average wage of a laborer in industry was 450 francs in the United States and 150 francs in France in 1955. With such an exchange rate, France was able to compete on the world market with American producers, with all products for which her productivity was more than one-third of American productivity. An incidental consequence is that tertiary products and services appear very expensive in the United States in relation to their cost in France.
Table XXV below, which compares prices on the basis of the official rate of exchange in the cities of Paris, Cairo, and New York, shows that in countries of little technical progress, tertiary commodities are less expensive than in countries with great technical progress, while on the other hand, secondary products are much more expensive in Cairo than in New York. This is the explanation of the important phenomenon of price disparity in the modern world.
In the same way it follows from the preceding considerations that the purchasing power of the hourly wage of the laborer, expressed in tertiary goods, is the same in all the countries of the world, and is fixed in time. The improvements in the differences of purchasing power are only shown in relationship to primary and secondary goods or services.
The bonds of cause and effect linking technical progress, the productivity of labor, and the average level of living of a people, show themselves clearly. If an object is sold at one time at a price equivalent to 200 hourly wage units and at another time at 50 hourly wage units, it is probably because its cost of production was close to 200 hourly wage units during the first period and later was close to 50 units. Furthermore, if an object is sold in one country for 200 hourly wage units and in another at 50, it is because its manufacture and distribution take about 200 hours of average labor in the first country and about 50 in the second. Here we encounter the problem of cost of production.
Selling price differs from cost of production by variable quantities that may be considerable when positive and may occasionally also appear as negative. These are rents and profits. It is entirely obvious that rents and profits cannot explain differences as large as that of the 200 to 1 which we have noted for the price of mirrors. Nor can they explain the differences of the order of one to five, ten, or twenty shown in Table XXV between the U. S., France, and Egypt. A radio costs about ten to fifteen hourly wage units in the United States, 80 to 120 in France, and 600 hourly wage units in Egypt.
4.THE DISTRIBUTION OF LEVELS OF LIVING ABOVE AND BELOW THE AVERAGE
It was said above that the available documents show an objective trend toward the closing of the fan of salaries in France. This tendency is verified by the observation of backward countries like China and India. It is interesting to see how technical progress does away with traditional rents and profits. The example of land will show this clearly.
Let us take the case of a territory in which internal commerce is not important and outline the history of the formation of rents and of their later disappearance:
1) As long as the country is technically backward, during the traditional period the first occupants of the country will occupy the best lands. Let us suppose that 100,000 inhabitants cultivate 100,000 hectares of very fertile land. They live comfortably, their food is abundant, they proliferate. There is no rent.
2) The youngest sons begin to occupy the less good land adjacent to that of the oldest. Since they cannot cultivate more ground than their brothers in the course of the year, and in fact must cultivate less because it is poorer land, their per capita productivity falls, together with the output per hectare.
The elder brothers enjoy a rent relative to the younger.
3) As the population grows, this is accentuated. The good lands are much sought after. Their proprietors can hire them out and demand rent in cash or kind. Political power grows out of this economic power founded in property and rent, together with social "exploitation" in the Marxist sense.
4) The more densely the country is settled and the more backward its technique, the greater the necessity of cultivating marginal lands. The stony and broken ground of the “Causse” is then cultivated at the same time as the marvelous river bottom of the Lot. The former yields one unit, while the latter gives five. The proprietor of the river bottom can easily rent it out for 3 or 3.5. He becomes a baron and builds the château that keeps the frightened serf respectful of the arrangement. Besides, the serf knows that he cannot find a better income elsewhere. The regime is politically stable.
5) A famine or an epidemic reduces the population. The men of a certain hilly district where the soil yields two or three units, die off. The serfs of the valley desert and occupy the vacant land. The rents of the proprietors of poor land disappear. Those of the proprietors of good land are reduced.
As the population grows again, a new cycle begins.
6) Let us suppose now that by means of technical progress the average output of the land is doubled. If the population does not double, a grave agricultural crisis develops. Schematically, let us suppose that the population is constant. More than half of the fields become useless. The peasant proprietor of poor land, ruined, looks for work in town. Meanwhile the rent of the proprietors of good land falls accordingly.
The limits can be readily seen. If 100,000 hectares produced sufficient food to feed 40 million Frenchmen, only the 100,000 best hectares of France would have any value at all. Moreover, this value would be low, since the output of marginal land (the 100,001st hectare), would be very close to the output of the best hectare.
This model explains land devaluation in technically progressive countries, and can be observed in all the regions of France since 1880. In the course of the nineteenth century, the general movement of the price of land reached a maximum around 1880, which corresponds to the date when the phenomenon technical progress became more important than the phenomenon growth of the population. Today, the real price of good cultivable land in France, that is to say, the price of a hectare expressed in hourly wage units, is four to six times lower than in 1880.
This mechanism also explains the facts observed by Simiand and Colson who were not able to explain them. The income from unbuilt-upon real estate was equal to ten billion hours of work in 1850, compared to 15 billion in 1880, and only five billion in 1925.
The same mechanism accounts for industrial profits. As long as the progress of productivity is restricted to a fraction of the economy, it creates rents for the benefit of those producers whose productivity has advanced. The selling price tends always to be fixed at a point above the cost of the least efficient producer whose production is necessary to satisfy the demand. In the long run, however, technical progress tends to cancel profits by increasing the productivity even of marginal producers, and above all, by increasing the volume of production of the best equipped producers. Every improvement in productivity creates a profit for its initiator, but at the same time destroys the profit of outmoded forms of production.
In the long run, this outcome is inevitable. But in the short run it can be delayed at the insistence of the producers, in the very common situation of imperfect competition. Either the producers combine to maintain the selling price as if no technical improvements had been introduced, or the price maintains itself because of the tacit restrictions of the producers, or even sometimes of the merchants who sell the commodity to the public. In general, it is an economic crisis that forces the sellers to give up their open or covert combination. The crisis is not, as the classic economists believed, the cause of falling prices, but only the occasion. If there had not been an improvement in productivity, there could not be any permanent decrease in selling price, depression or no.
We note that around 1929 the real value of French land (expressed as usual in hourly wage units) was less than a quarter of the value for 1879-81. The houses and castles had fallen by 1924-25 to 70 per cent of their value in 1851, and 40 per cent of their value in 1889. Only the factories and personal property had increased their real value from 1851 to 1925, but even in these categories there had been an appreciable decline after a maximum reached around 1912.
These facts, hardly noticed by the economists, were attributed to "currency devaluation". But they were appreciable before 1912, and could be observed also in the United States.
In our opinion, they are really the direct result of the action of technical progress on rents and profits.
The consumer's budget, the labor force, and the national income. Even if there were as many rich as poor, the level ofliving of the poor could not be raised to that of the middle income group by distributing total income equally among all citizens. The total food consumption of rich and poor together adds to 1,000 plus 360, or 1,360. The average is 680 insteadof the 900, which is the food consumption of the average worker (see Table XXVII). In contrast, the consumption - and thereforealso the production - of miscellaneous items would amount to 6,000 plus 60, or 6,060, for an average of 3,030, althougha middle income budget would require only 900 for miscellaneous expenditures. There would therefore be too many"other" commodities and not enough food.
Thus, to equalize the income of a population composed of two classes equal in number, one rich and the other poor, it would be necessary not only to equalize their money incomes but to modify the structure of production, and especially to obtain a greater volume of food. However, this is not always possible. For example, it was impossible in France before 1800. In the cases where it is possible, it always implies a transfer of manpower which takes time, and then of investments which require accumulation. It also involves the training or retraining of workers.
Economic problems are labor force problems. The average production of rice in a country is 500 grams per capita. If the Maharaja is exiled and his income is divided among the hundred thousand poor Hindus, the nominal monetary income of each one is doubled, but the production of rice does not increase, and so the consumption of rice remains at 500 grams per capita. The price of rice doubles.
In order to raise the level of living of the 100,000 poor Hindus, it is necessary to change the occupation of the individuals to whom the Maharaja gave an opportunity to perform non-agricultural labor by buying their secondary or tertiary production - always supposing that there is enough land to increase the output of rice. In any case, the total national production will be considerably diminished. In order to obtain a very small increase in agricultural production, it will be necessary to virtually ruin secondary and tertiary production, which will ultimately have grave repercussions for primary production as well. Hence the stability of high-rent regimes in very poor countries.
The only practical solution to the problem is an increase of productivity which is a long-term solution. One must have the courage to say that in the short run, there is no solution. By itself, the removal of the Maharaja cannot raise the level of living of the poor Hindu. Table XXIX illustrates the fundamental relationship between the budget structure of the low-income population and the structure of national production and therefore of the labor force.
It is impossible to increase the people's consumption of manufactured products, housing, and so forth, without altering the structure of production. Conversely, the average level of living of a people without exterior commerce can be deduced from the distribution of the labor force alone. The disparities that may be observed between the structure of the worker's budget and the structure of national production are due in part to the consumption of the prosperous classes and even more to savings for productive investment.
However, since the working classes form the mass of the nation, these disparities are necessarily slight. They might even be nil in a socialist society.
The disparities noted between the structure of national income and the structure of the labor force have to do with differences of wage rates and average incomes in different occupations. Agriculture is in general the least favored, while tertiary occupations are the most favored. Here, too, the more a country approaches socialism, the more these disparities diminish.
The fundamental relationship, expressed by Table XXIX, among the level of living, national production, and occupational structure, is mediated by the common element in all of these phenomena-technical progress. Technical progress is the independent variable of economic life. It has as much influence under a liberal as under a dictatorial government, in a capitalist as in a collectivist system. This determinism obviously extends also to the secondary manifestations of economic life: prices, salaries, taxation, and so forth.
The figures of the table show that the problem of purchasing power cannot be resolved by changing the income distribution alone. Suppose a social revolution had taken place in 1830. Even if it had succeeded in equalizing the revenues of the whole population without disorganizing production, it would have had no permanent effect unless it led to a new distribution of the labor force in order to adapt production to a new structure of consumption. It is certain, in any case, that it could not then have provided the level of living that is considered a bare minimum at the present time.
For example, to give 7.3 kilograms of sugar to each inhabitant- the amount now considered as a minimum subsistence ration -, it would have been necessary to produce 237,000 tons of sugar. The actual production was then 75,000. The same may be said of all other products except for the classic commodities of the poor-wheat, potatoes, barley, and rye of which there would have been a small surplus.
Men of the traditional era gradually came to a fairly clear perception of the cause and effect relationship between consumption and production, between the structure of consumption and the structure of production, and finally, the labor force. As a matter of fact, at least eight-tenths of the population did not consume anything but the product of their own work or, rather a part of that product. Technical progress has concealed this determinism by separating work from consumption. The workmen who make automobiles do not eat automobiles, but bread, meat, and so forth. I produce books, courses, lectures, and I consume bread, meat, and many other commodities. The worker is thus led to regard his consumption as independent of his own work, and consumption in general as independent of production in general. His salary alone seems to determine his purchasing power, even though a general revision of salaries only amounts to a change in the monetary unit.
In The Man with the Forty Crowns, Voltaire showed that in his time the essential determinism of the level of living had been perceived. "The Geometer" in the scene is Deparcieux:
THE MAN WITH 40 CROWNS: Do me, I beg you, the favor of telling me how many animals with two hands and two feet there are in France.
THE GEOMETER: There are supposed to be about twenty million, and I should like to adopt this likely figure, at least until it is verified, which would be very easy and which no one has yet done because it is never possible to keep track of everything.
THE MAN WITH 40 CROWNS: How many acres do you suppose are contained in the territory of France?
THE GEOMETER: 130 million, of which half are in roads, in cities, villages, moors, heaths, swamps, dunes, barren land, useless woods, pleasure gardens more agreeable than useful, uncultivated land, bad land badly cultivated. We can reduce the productive land to 75 million square acres, but suppose we reckon 80 million. It is little enough to do for the fatherland.
THE MAN WITH 40 CROWNS: How much do you suppose that each acre produces in a fair year, taking one with another, in wheat, in grains of all kinds, wine, fisheries, wood, metals, cattle, fruit, linen, silk, oil, including all expenses, and without counting taxes?
THE GEOMETER: If they produce 25 livres each, it is a great deal. However, let us say 30 livres so as not to discourage our fellow citizens. There are acres which produce 300 livres consistently. There are others which produce three. The proportional mean between 3 and 300 is 30. As you can see, 3 is to 30 as 30 is to 300. It is true that if there are many three-livre acres and very few at 300 livres, our account will not balance, but once more, I do not wish to haggle.
THE MAN WITH 40 CROWNS: Very well, sir, how much will the 89 million acres produce in money income?
THE GEOMETER: The calculation is already made. They will produce annually two billion, four hundred million livres at current values ...
THE MAN WITH 40 CROWNS: I understand. But you have told me that there are 20 millions of us, men and women, old and young. How much for each one, if you please?
THE GEOMETER: 120 livres, or 40 crowns.
THE MAN WITH 40 CROWNS: You have guessed my income exactly. I have four acres which, counting the idle years together with the productive years, are worth 120 livres. It is very little. Imagine if everyone had an equal portion as in the Golden Age, everyone would have only five golden louis per year?
THE GEOMETER: Not more, according to our estimate, which I have made a little generously. Such is the state of human nature. Life and fortune are strictly limited. Taking one with another, we do not live more than 22 or 23 years in Paris. Taking one with another, we do not have more than 120 livres per year to spend, that is to say, that your food, your clothing, your housing, your furniture, may be represented by the sum of 120 livres [...]
THE MAN WITH 40 CROWNS: ... If we decided to have twice as many children as we do have, so that our population was doubled, and we had 40 million inhabitants in the place of 20, what would happen then?
THE GEOMETER: Either we would each have only 20 crowns on the average to spend, or the soil would have to produce twice as much as it now produces, or there would be twice as many of the poor, or we would have to have twice as much industry and gain twice as much by foreign trade, or send half of the nation to America, or half of the nation might eat the other.
THE MAN WITH 40 CROWNS: Let us then be satisfied with our 20 million people, and our 120 livres per head, distributed as God pleases, but still this situation is sad and your century of iron is a hard one.
THE GEOMETER: No nation is better off; there are many in worse condition. Do you think that they have enough in the North to give the equivalent of 120 livres to each inhabitant? If they had had that much, the Huns and Goths, the Vandals and the Franks, would not have deserted their home lands to go and settle elsewhere, with fire and steel in their hands.
THE MAN WITH 40 CROWNS: If I let you go on, you will soon persuade me that I am happy with my 120 francs.
THE GEOMETER: If you thought yourself happy, then in that case you would be so.
5.THE INFLUENCE OF THE ECONOMIC CRISIS ON THE LEVEL OF LIVING
If we examine the statistics of the level of living since 1830 or 1850, in a period of crisis, that is if we compare, for example, hourly wages with the price of basic commodities, we find only weak effects. The purchasing power of wages varies little during periods of crisis. On the other hand, if, instead of speaking of real wages, we consider the average real income of the whole of the population, we observe an appreciable decline in the average level of living of the nation. On the one hand, wages do not decline more than prices and often less. On the other, the average total income certainly declines during depressions.
Why does income fall more than the average wage rate ? For two important reasons. The first is that crises in the modern period are manifested principally by the fact that a large number of wage earners lose their jobs or are no longer employed full time. It is the fundamental characteristic of crises in the present era to cause very serious unemployment.
The second reason is that the profit of entrepreneurs, and the returns from investment suffer a veritable collapse. This is due to the distress selling of inventories and to the fall of prices generally.
Consequently, the wage earners who do not lose their position maintain a purchasing power that is not very different from what it was previously even if their earnings are reduced, since prices have also been reduced. But the unemployed lose their wages and, in the liberal economies of the nineteenth century, they had no other resources. In the more developed economies of today, the unemployed receive insurance or relief, but their level of living falls considerably nevertheless.
Profits decline more during a crisis than wages. As a matter of fact, it is because of the fall in profits that unemployment occurs. When managers see their businesses threatened and limited, and their sales made with difficulty, they lighten their payrolls and operate on a reduced scale. It may be that the enterprise comes to a complete halt and disappears. Crises can only be remedied by shifts within the labor force.
In conclusion, the facts. that appear to be capital for the scientific study of the causes of improvement in the level of living are the following:
1) There has been an appreciable improvement during the past 150 or 200 years in all of the countries of the world, but according to very different time schedules. As Karl Marx wrote, "The country which is most developed industrially shows to those which follow on the industrial ladder the image of their own futures."
This circumstance gives the economist a research situation that is close to the experimental condition: Wherever the rise in the level of living is most rapid, there the causes of the movement will operate with greater force.
2) The improvement, within a given country, is very irregular when we consider a range of commodities. With respect to certain commodities, no improvement at all can be seen, while for others, the improvement in 150 years has been of the order of 1 to 200.
These two facts enable us to identify the causes of the improvement in the level of living. By comparing very different sequences of development (for example, bicycles and tapestries since 1890), we can discover the determinant factor which affects one case and not the other.
An experimental analysis of this kind makes technical progress appear to be the principal cause of the improvement of the people's level of living and of the purchasing power of the working masses, at least such technical progress as increases the productivity of human labor. It is very easy to observe and to verify empirically that all the products for which there has been improvement in purchasing power have also benefited from an increase in productivity (mirrors, lighting, transportation, bicycles, manufactured and agricultural products in general), while the commodities associated with little or no improvement in purchasing power (the cost of hospital care, institution fees, haircuts, tapestry, book bindings, objects of art, handicraft objects, services in general) have not been subject to any major improvement in productivity.
These experimental facts are easy to understand rationally. Since productivity is the ratio of a volume of production to the direct or indirect labor time necessary to obtain that volume of production, it is simply the reciprocal of the cost of production expressed in hours of labor. If productivity is constant, the cost of production is proportional to wages.
Consequently, with a constant productivity, the improvement of purchasing power cannot be obtained except by a reduction of the difference between the cost of production and the selling price. For all of the important products and services in today's market, this gap turns out to be negligible compared with the gains made possible by the improvement of productivity. On no important product or service, in fact, was the rate of profits ever higher in France than about 20 per cent of the selling price. More often it is of the order of two to Level of Living and the Productivity of Work five per cent. Yet the gains made possible in the past century by the improvement of productivity for these same products are of the order of 100, 200, 300, and even 10,000 per cent. The effects of productivity are more important than the effects of profit, in the middle run as well as in the long run, so far as the improvement of the level of living of the masses is concerned. Among the factors that determine the value of a commodity (labor, scarcity, need, profits, surplus value, etc.), the factor of labor turns out to be predominant for a great number of essential goods.
It would be obviously absurd to conclude from this that there is no problem of profits. In the first place, even when they are small in relative value, they may be large in absolute amount. Again, profits may be slight in comparison to the gains of productivity, but considerable in relation to the cost of production. Finally, for a great number of products, contrary to what occurs for others, the phenomenon of profit is more important than the phenomenon of productivity. It would be extremely useful to make a list of products in rank order of the rate of profit. This does not appear impossible to do. By analyzing the annual accounts of enterprises over a sufficiently long period ‑ thus permitting hidden reserves to appear ‑ we would have at least an approximate list. One could then identify some cases where an effort to improve the level of living would have to be directed principally to the reduction or suppression of profits, along with those cases where the improvement of productivity is called for. It is already certain that many of the important commodities in the worker's budget fall into this latter category. It is also certain that in those cases where profits are important, such profits are more often commercial than industrial or agricultural. The struggle for the reduction of profits, like the effort to improve productivity, is linked to the rational distribution and utilization of the labor force.
Since the level of living is in large measure determined by the productivity of labor, and since this productivity also controls economic development in general-the movement of prices, crises, foreign commerce, government fiscal policy, and the duration of work-it is easy to understand the close relation, in a closed economic system, between the level of living and each of the essential factors of economic life. For example, in a closed economy, with a high level of living, it will always be found that the labor force has shifted towards tertiary occupations and that, nevertheless, tertiary services appear scarce and are very expensive in relation to primary and secondary goods.
 Alfred Sauvy, Théorie générale de la population, Vol. I. 2.
 This important fact reveals the obstacles that make it difficult to raise the level of living of the masses by the suppression of high incomes alone. We have seen that spending habits vary a great deal with income. The poor spend from 60 to 90 per cent for food. The upper income groups, on the contrary, spend 60 to 80 per cent for services and luxuries. If the highest incomes were divided among the poor, the latter would not buy services or luxuries, but food. Therefore, food would not be more abundant because high incomes were suppressed. It would be necessary that the men who provide services change their work.
For example, if, at the present time, we wanted to raise the level of living of the population of India by suppressing the high incomes of the local princes, the nominal income of each worker could probably be doubled. They would then try to buy double the amount of rice they now consume. Since the production of rice would not have been ipso facto changed, the price of rice would double, and there would be no improvement in real buying power. There could only be an improvement by returning to work on the land the labor force responsible for the services and luxuries of the Maharaja. But since the land capable of growing rice is already under cultivation, and since the law of diminishing returns would come into play, such action would be rather ineffective. It must be realized, too, that this transfer of population would not be easy, and that the new arrivals would be ill-adapted to agricultural labor. The controls and the police necessary for this change would absorb an important fraction of the labor force. Furthermore, the average level of living being insufficient to provide for education, it would be necessary to reconstitute a privileged class entitled to education. This example, while completely theoretical, suggests the necessary conditions for efficacious political reform. In brief, it is not taking away Mr. Rockefeller's box at the opera which provides more steak for the poor, but rather the production of more steak.
These facts give some insight into the difficulty of solving the problems of a technically backward country. They evoke the essential problems of present day China and India.
 J. Fourastié, "Indirect Measurement of Productivity," Productivity, Prices, and Salaries (OEEC, 1957).
 It is instructive under these conditions to see how Simiand, dominated by his thesis, neglected these differences. A different frame of reference would have made him see them in a different light and helped to explain them. A scientific attitude, sincerely believed to be "objective," is in reality influenced by general ideas and even more strongly so because they are latent. Objective attitude? But what objects should be observed? The mind must choose. It is this opposition between the limited mental process and the multiplicity of the external world that sets the limits of science. The criterion of success is not to take in all observable facts (which is still impossible), but first to identify and then to relate those that are important for the proposed action.
One must always be aware of the subjective quality of objectivity. It is indispensable to scientific research. If Simiand had been aware of this, he would have been guided by facts other than those which captured his attention. He would have seen that in the matter of the evolution of prices, disparity was more important than parallelism. The observation of disparities would have revealed the causes of parallelisms to him and led to a solution of his problem. It is interesting to follow the development of his thoughts in his book. At times he was so close to the solution that another step would have solved the problem.
Thus when he wrote: We have, on the other hand, price changes of industrial materials, or even of industrial products, which would indicate a relatively very low level: Is this not due to a grouping of economic conditions, evidently of great importance for industrial exploitation, and which might have a connection with the changes leading to the diminution of cost of production change in the organization of work, change in the utilization of other means of production, or in other words, the group of factors which we call "the industrial revolution"? Unfortunately Simiand stops here; he does not develop this idea. He returns immediately to the general consideration of modifications in income and relative levels, of the economic conditions of consumption and distribution, etc. (Op. cit., p. 198).
 It might seem more logical to use only two adjectives here, one for technically progressive products and one for technically unprogressive products. But it will be seen further on that among the technically progressive products it is useful to distinguish between agricultural and non-agricultural items. The distinction between degrees of progress is itself somewhat arbitrary, since we are dealing with a continuum. The reader will recognize the terms "primary," "secondary," and "tertiary" as those of Allen B. Fisher. Although I give these words quite a different meaning from his, I hope that Mr. Fisher will think it legitimate.
 The fundamental question of window glass will be taken up later. We can speak here of the "civilization of glass." Among the paintings showing the history of glass, the following are particularly representative: a) Paintings representing windows with no system of closing: Giotto (1276-1336), Announcement Made to St. Anne (Chapel of Sacro Vegni). Leonardo da Vinci (1452-1519). The Benois Madonna, The Litta Madonna (Leningrad Museum); The Last Supper (Milan). Botticelli (1441-1510), Portrait of St. Vespucci (Berlin, Private collection). Metzys (1466-1530), Virgin with Cherries (Cincinnati, Private collection). Schaufelein (1480-1540), Portrait of a Man (Basle Museum). Maître de Moulins (1498), Peter II with a Patron Saint. b) Pictures representing richly furnished interiors, with the windows partly covered with small translucent squares of glass: Maître de Flemalle (1428), The Annunciation (Merode Collection at Antwerp.) [The painted window in this picture is typical; the top part is of stationary stained glass panes, the middle of wooden shutters, and the bottom of wood lattice.] Carpaccio (1456-1525), The Celestial Apparition to St. Ursula (Academy of Venice). Van der Weyden (1399-1464), Annunciation (Louvre). Cima da Conegliano (1459-1517), Annunciation (Leningrad Museum). Froment (1484), The Resurrection of Lazarus (Uffizi, Florence). École d'Avignon (XIV), The Altarpiece of Thouzou (Petit Palais, Paris). Bouts (1420-75), Annunciation (Leningrad Museum). Maître de Moulins (1490), Portrait of Margaret of Austria, fiancée of Charles VII (Lehmann collection, New York). These pictures prove that in this era even the richest men could not completely cover their windows with glass. c) Paintings representing casement windows made entirely of translucent glass. Van der Weyden (1400-64), St. Luke drawing the portrait of the Virgin (Leningrad Museum). Also, his Virgin and Child (Royal Museum, Brussels). Bouts (1420-75), Altarpiece (St. Peter's Church, Louvain). Memling (1435-94), Annunciation (Lehmann collection, New York). The Virgin of Martin van Nieuwenhoven (St. John's Hospital, Bruges). Metzys (1466-1530), The Banker and His Wife (Louvre). French School (XVI), Gabrielle d'Estree (Condé Museum, Chantilly). Van Eyck (1380-1440), Virgin of Autun (Louvre). Tintoretto (1518-94), The Birth of St. John the Baptist (Leningrad Museum). From the seventeenth century on, paintings of rich interiors almost always showed windows with glass that could be opened. d) Pictures representing peasants houses without glass windows. There are many such pictures up to about 1800. Often there is just a wooden lattice: Brueghel (1525-69), Children at Play (Vienna).
 This fundamental fact was demonstrated in Le Grand Espoir du XXe siècle.
 On the situation in ancient Rome, cf. J. Carcopino, La vie quotidienne à Rome à l'apogée de l'Empire, p. 36 ff.
 This explains the phenomenon observed by Labrousse: from 1730 to 1789, rents went up while real wages went down. The income of the agricultural worker is related to the productivity of marginal land. It decreases as the cultivated area increases. On the other hand, rent goes up when the countryside is prosperous and the peasants are poor.
 See Le grand espoir du XXe siècle, Chapter V, and Recherches sur l'évolution des prix en période de progrès technique (1st Series), pp. 30-33.
 Space precludes the treatment of this important matter of investments in this book. See J. Fourastie, "La Prévision Économique au Service de l'Entreprise et de la Nation."
 If economic factors acted equally on all forms of consumption and production, the improvement in the level of living could be seen in haircuts and in bread as well as in window glass. The number of haircuts would be greater in one country than another, according to the political and social regime.
 Since the selling price is the price of a unit of production.
 For example, the profit of the first manufactured glass was about 20 per cent. In suppressing this profit (and supposing that this had no effect on productivity), a piece of glass four meters square might have been sold for 32,000 hourly wage units instead of 200 as at present. The same principles apply to many other commodities, none of which would modify the level of living of the people to any extent. This explains why a political revolution, even a radical one, as in China, does not improve buying power. It also shows why, as Lenin said, the true revolution would be the electrification of the country, and why the nationalization in France of various industries did not lower their prices. The purchasing power of the worker's hourly wage in the United States has increased as follows in the past 150 years despite the system of private profit.
Wheat 1000 per cent