Main NI Index | Main Newspaper Index

Encyclopedia of Trotskyism | Marxists’ Internet Archive

The New International, May 1938


David Cowles

Crisis and Reform Labor Politics


From New International, Vol.4 No.5, May 1938, pp.133-136.
(Corrected in line with the erratum in Vol.4 No.6)
Transcribed & marked up by Einde O’Callaghan for ETOL.


THE TRADE UNIONS ARE in politics and they are there to stay. Along with this there is a growing sentiment within the labor movement for it to continue and spread out. Unfortunately the sentiment is too often accompanied by little knowledge of politics and less knowledge of labor politics. Often the spurs to action are high hopes and vague promises. At such time it becomes imperative to take stock, to see concretely the purpose of politics and to define the scope and limitations of labor politics specifically.

Politics under any system is a struggle by conflicting groups or classes for control of the state apparatus. Under capitalism, more than under any other system, the motive force of struggle is the endeavor to redistribute the wealth and income in closer accord with the demands of the victors. The struggle has a double aspect. One aspect looks toward the redistribution of national wealth and income within the limits of developing capitalism. Within these limits, after the Civil War, the northern industrialists fought party battles with the planter South, the western farmers fought their battles with the northern industrialists, and labor parties rose and fell. The assumption in all these struggles was that whoever won the state apparatus could distribute economic gains for their class without disturbing class-political relations. The second aspect looks toward a redistribution of wealth and income which is incompatible with the growth or maintenance of capitalism. This can no longer be settled by mere parliamentary victory.

The Civil War is a classic example. For decades before the war the growth of capitalism in the North and its extension West and South was becoming more and more incompatible with the expansion of the southern slave economy. The destruction of the class-political dominance of the cotton planters and the capture of the state apparatus by the northern industrialists were the preconditions of the further development of capitalism. At the same time, their victory meant the economic and political subordination of the South to the needs of northern industrial development. Both sides saw the full meaning of the conflict with increasing clarity. Which economic system shall prevail? Which social-economic class shall rule? The questions were posed in heated debates and parliamentary struggles. They were answered and settled by the roar of cannon and the smoke and battle of Civil War.

The entry of the trade unions into politics does not change the essence of politics. In a vague way, the rank and file union member feels that labor politics will enable the workers to get hold of the government and permit them to use it to strengthen the labor movement, to give the unemployed more relief and decent jobs, and force the capitalists to redistribute a bigger share of the national wealth and income to the working class generally. In the same vague way, they feel that this can be done within the limits of capitalism and within the bounds of City Halls, state legislatures and Congressional corridors. The feeling is strengthened by the speeches of well-meaning reformers and the deceptions of the Stalinists. For labor reform politics, professional reformism, Stalinist opportunism and the vague sentiments of the untutored worker all agree on this: They all feel or believe or try to make the workers believe that substantial concessions can be won by labor politics fighting a parliamentary battle within the limits of capitalism, a battle that leaves undisturbed the class control of the state.

Those who take seriously their responsibility to the workers will not be satisfied with just proclaiming their beliefs or mouthing sentiments that gain fleeting favor. They will test their beliefs before they proclaim them. They will face the basic questions: Can capitalism grant substantial economic concessions to the workers? Can they be won in parliamentary struggles? Can they be won without disturbing the class control of the state? Can a reformist labor politics, as it is today and is developing into tomorrow, win and retain for the workers substantial economic concessions within capitalism? If capitalism cannot grant substantial concessions, such labor politics is built on quicksand. If it can grant them but will not so long as the struggle is a parliamentary one, then reformist politics is self-imposed blindness. If its class control of the state can nullify any parliamentary victories of the workers and labor politics leaves class-political relations undisturbed, then this politics is the politics of defeat. If the very nature of labor reform politics makes it incapable of winning or, if it can win, of holding on to the concessions it has gained for the workers, then reform labor politics is false to labor. But whether false or true is a question of fact. Let us consider the facts.

The Basic Question Posed

The first question we must consider is: Can the workers gain substantial economic concessions within the limits of capitalism? This is the basic question, the answer to which determines the whole approach to labor party politics. The reason why is simple: Reformist politics operate within self-imposed limits, the limits of capitalism and its class relations. If capitalism is progressing and increasing production, profits and employment, reformist labor politics have room for effective action. In fact, such politics can benefit the workers substantially only in a period of progressing capitalism. But when capitalism is declining, and the capitalists are tearing down the concessions they had granted in the past, the limits within which such labor politics can function disappear. The concessions which it can win are mythical because capitalist decline is real. Reformist labor politics, which arise from the economic problems of the workers, are helpless to solve those problems.

There can be only two answers to this question – “yes” and “no”. Those who say “yes” believe that capitalism will continue into an indefinite future, that it has within itself the elements of progress and growth, and that the workers will be able better to share in the profits of growth by means of reformist labor politics. Those who say “no” believe that capitalism is in decline and that its profits are declining. The concessions it can give to the workers are diminishing. The slight gains that such labor party politics can get will be distributed to a small, favored section of the working class. But for the workers as a whole, substantial concessions are impossible. They are incompatible with the continued existence of capitalism. The first is the answer of reformism. The second is that of Marxism. Reformist labor politics assume the first. Which do the facts support?

Whether American capitalism is progressing or declining can be determined easily by comparing two periods of economic activity. The appropriate years for comparison are 1929 and 1937. Both are peaks of economic activity, following years of depression. Both are turning points into depression. In all previous history the latest peak of economic activity was always higher than the one preceding. The trend was upward. How does 1937 compare with 1929? What is the trend here?

The trend here is plainly downward. According to all the most general indices of economic activity, 1937 was much lower than 1929. According to the comprehensive index of activity compiled by Business Week, business activity in 1937 was fully 30 percent less than in 1929. Much of the fall is due to the precipitate decline of financial expansion and stock exchange activity. The physical volume of production is much more significant. For production is the precondition of consumption and the sustaining force of society. What happened to production? It, too, declined although not so sharply. According to the Federal Reserve System, the averages of industrial production as a whole, which includes manufacturing and minerals, were:








The drop in industrial production is nine percent. However, this does not take into consideration the fact that the population increased by millions. If we take account of the population increase, the decline between 1929 and 1937 is not nine but almost 15 percent. (The Monthly Labor Review of November 1937 estimated the loss between 1929 and 1936 as being sixteen percent. Due to further increase in population at the same tune that there was an increase in production, pretty much the same loss held for 1937.)

The class significance of the fall in production comes out more clearly when we divide industrial production into capital goods production and consumption goods production. From the point of view of the health of capitalism, the production of an increasing volume of capital goods is essential. Capital goods increase claims on income and increase the extraction of surplus value, thus increasing the rate or mass of profits, or increasing both rate and mass. From the point of view of consumption and the standard of living of the workers, an increasing volume of capital goods production, if not diverted to armaments, means a greater supply of the means of production to increase the plenty of consumers’ goods. From either standpoint, a decline in the production of capital goods indicates a decline in capitalism.

And, certainly, the indices of capital goods activity paint a vivid and unmistakeable picture of the decline of American capitalism. According to Standard Statistics, one of the best known agencies selling information to business firms and stock speculators, capital goods activity was:








Here is a loss of capital goods activity equalling 20 percent. This is a fact of profound significance. Viewed in isolation, it means that within nine short years one-fifth of the production of capital goods has been destroyed.

This is important in itself as a sign of economic decline. But it is even more important when seen against the background of economic trends and when it is looked at within the matrix of capitalist prosperity. In all previous business cycles, each successive peak of capital goods production was higher than the previous one. But capital goods production in 1937 did not exceed the previous peak. Not only did it not exceed, it did not equal it. And not only did it neither exceed nor equal but it remained stunted in its upswing twenty percent below the 1929 peak and then relapsed into the sharpest drop in economic annals. This is even more significant for prosperity under capitalism. Prosperity in the past was especially due, and mainly due, to the increasing output and absorption of capital goods. This stimulated prosperity. This sustained prosperity. As capital goods output increased, so did prosperity. The twenty percent drop in capital gods output has destroyed, within nine years, one-fifth of the economic foundations upon which American capitalism and its prosperity rest.

The dreary picture of widespread decline which this twenty percent drop sums up, does not show how unevenly distributed it was between specific industries, and within what a wide range the distribution took place. The fact is that in the nine years, 1929-1937, some industries fell as much as fifty, sixty and seventy percent. This was especially true of those industries which supplied the railroads. Among those capital goods industries which fell between forty and seventy percent were:


Percent Loss Between



Railroad passenger cars


Freight cars


Still other industries, especially those depending upon building construction, fell between twenty and forty percent between 1929 and 1937. These were:


Percent Loss
Between 1929-1937



Fabricated Steel






Among those industries which declined between ten and twenty percent were:


Percent Loss
Between 1929-1937

Bituminous Coal


Pig Iron


Electrical Equipment
     (new orders)


And what is important in all instances is that the drastic declines occurred in industries which are the basis of industrial production.

Even the gains that were made in certain industries only emphasized the general decline. The machine tool industry produced twenty percent more in 1937 than in 1929. However, this increase was not due to domestic demand but “was largely the result of a pronounced rise in foreign buying” (Survey of Current Business, March 1938). When the rise subsided due to world depression, machine tool production fell precipitately. Electric power production rose 24 percent. However, this was not accompanied by greater industrial production but by intensification and displacement of labor. Truck production rose 16 percent. But this only indicated that small business men were increasing in number due to the efforts of unemployed workers to escape unemployment by going into business. The proof is that the output of large trucks, which are used by big firms, did not account for the rise in truck production.

“The light commercial truck continued to account for most of the increase in total output.” (Survey of Current Business, March 1938.)

And compared with the sharp and widespread drops in capital goods output in other and basic industries, these increases were insignificant.

Excepting those employed in them, the decline in these industries did not affect the living standards of the workers immediately and directly. These are capital goods industries and their decline is directly felt either by the capitalists who cannot produce, or those who cannot absorb, as much capital goods as before. The building construction industry bridges the gap between capitalists and workers, and its activity affects both classes directly and generally. For the capitalists, it is a great absorber of capital goods and a strategic factor in prosperity. For the workers, it means shelter, housing, an important item in their cost of living. Increasing building constructions aids the capitalists by absorbing capital goods. It aids the workers by causing greater competition between landlords, resulting in an easing up of rents, thus leaving greater purchasing power among the workers for other goods. Decreased building construction not only destroys a great market for capital output but it also leaves dilapidated houses and lowers the living standards of the masses by forcing up their rents. Yet this very important industry declined 54 percent between 1928 and 1937.

What affects the workers even more directly and substantially than housing is the output of consumption goods. Consumption goods output sustains life and determines the standard of living of the masses. The greater the volume of consumption goods output, the greater is the objective plenty which, if distributed, will lift the standard of living of the workers. Under capitalism, a fall in output accompanies a fall in mass purchasing power. Output is therefore a rough measure of general living standard. What happened to consumption goods output between 1929 and 1937?

Output of consumption goods dropped, although not nearly as much as in the capital goods industries. The index of consumption goods activity compiled by Standard Statistics shows a loss of 2.3 percent between 1929 and 1937. However, if we take into account the population growth which required a proportionate growth in consumption goods output, the actual decline was 8 to 9 percent.

What is more important, the Standard Statistics index does not show that the greatest losses in consumption goods output were in basic food commodities, such as meats, wheat flour, and sugar. Output in these commodities fell between 12 and 17 percent in 1937 as compared with 1929. Standing by themselves, these figures indicate a substantial enough loss in the living standards of the workers in whose food budget these are major items. But taken in conjunction with the increased population, the increased number of mouths which this falling production was to feed, the fall in living standards was even greater. Moreover, passenger car production which is an index of the purchasing power of the better paid workers and the middle class, was 18 percent lower in 1937 than in 1929.

Where there were gains in specific consumption goods industries, the gains were small. Where the percentage gain was large, it was because the industry was new and growing, and its commodities did not depend for their sale upon the wide masses of workers and lower middle class. Examples of such industries are electric refrigerators and electric washing machines. The former increased 182 percent between 1929 and 1937, and the latter gained 55 percent. In neither case were they large enough to make appreciable demands for capital goods, or contribute much to economic recovery.

The Workers Share Capitalism’s Decay, Not Its Profits

This widespread decline in both capital and consumption goods industries brings to a sharp focus the basic contradiction of capitalism – the contradiction between production and consumption. Capitalism does not produce unless ultimately it has consumers to whom it can sell its goods at a profit. But if it sells at a profit, it redistributes wealth and income and undermines and destroys future consumption. At the same time, the plight of the workers is that they cannot buy goods unless they are employed in production, getting in this way the wages and purchasing power which make them the greatest class of consumers. Their purchasing power is at once a by-product of production and the ultimate sustaining force of production. The declines in production first destroy employment and the purchasing power which employment gives the workers. But in destroying the workers’ purchasing power, falling production destroys also the ultimate force that can alone sustain production – consumption. Capitalist production, which first destroys employment and purchasing power, ultimately destroys itself.

The United States is the greatest market for its producers. Nine-tenths of all its production is sold in the United States. Among the consumers upon which all this production ultimately depends, the workers are by far the greatest class. They form seven-tenths of the working population. They represent the largest section of the whole population. They have no source of purchasing power outside of production. Their ability to consume the output of production, and thus sustain production, is itself a by-product of production. How have the production declines between 1929 and 1937 affected their employment and wages – their sole source of purchasing power?

The widespread declines in industrial production were accompanied by falling employment and even more sharply falling payrolls. Manufacturing, which employs about one-fourth of all workers, provided one-tenth less jobs and one-eighth less wages in 1937 than in 1929. The composite indexes of employment and payrolls fell 10 percent for employment and 13 percent for payrolls. [1] What this means becomes clearer when we separate the indexes into their component parts of durable and non-durable goods.

Employment and payrolls in the durable goods industries are especially significant because they contribute more employment and greater payrolls for each dollar of value produced than in other industries. But between 1929 and 1937 durable goods indexes fell 13 percent in employment and 14 percent in payrolls. It would have fallen further if it were not sustained by the auto industry, where the CIO unionization drive forced up employment by 14 percent and payrolls by 13 percent. Shipbuilding which returned to its condition of 1929 stood at the transition point between this one sign of increased employment and the rest of the industries which differed between themselves only in the sharpness of decline. The machinery industries group fell off 5 percent in employment and 10 percent in payrolls. Steel lost 9 percent in employment and 8 percent in payrolls. Railroad repair shops, and lumber and allied products, fared much worse. The first lost 23 percent in employment and 26 percent in payrolls and in the second employment fell 36 percent and payrolls 42 percent.

Even where an industry did exceed its 1929 production, the workers got no benefits. The machine tools industry, whose output of 1937 was 20 percent higher than it was in 1929, celebrated this increase by cutting employment 18 percent and slashing payrolls by 23 percent.

The non-durable goods industries produce consumption goods. The growing population should have acted as a stimulus to increased output of consumption goods and an increased amount of employment and payrolls. Despite this, production declined and the indexes of employment and payrolls fell, 5 percent in employment and twice as much in payrolls. The disproportionate drop in payrolls was due to several reasons: the fall in employment was not so sharp because increased population sustained and increased the demand for output. However, the effective demand, the demand backed up by purchasing power, fell considerably due primarily to the sharp decline in employment and payrolls in the durable goods industries. The workers had less money with which to buy food and clothing. The competition between capitalists in consumption goods industries for the workers’ purchasing power drove prices downward. Falling prices reduced profits. The capitalists passed on a substantial part of their decay to the workers in the form of falling payrolls. The absence of a strong labor movement in the consumption goods industries made it all the easier to do this.

A cursory glance at the individual groups of industries shows both the downward pressure of economic decline on employment and payrolls and the upward pressure of union organization. Taken as a whole, textiles and its products fell off 5 percent in employment and 16 percent in payrolls between 1929 and 1937. But the cotton goods industry, the most important one in the group, gave out 5 percent more employment and 8 percent more payrolls because it was wrested from the capitalists by the organization drive of the CIO. Food and kindred products, which are comparatively well organized, fell off in employment by 5 percent but the loss in payrolls was less than one percent. Chemicals and allied products gained 5 percent in employment and 5 percent in payrolls. However, these few bright spots were put completely into the shade as leather and its manufactures dropped off 10 percent in employment and almost 20 percent in payrolls; rubber products saw 19 percent of the jobs and 23 percent of the payrolls vanish into thin air; and tobacco manufactures experienced a 30 percent fall in employment and a 30 percent cut in payrolls.

Employment and payrolls fell sharply enough in the manufacturing industries. However, an occasional increase broke the monotony of decline. This is not true of the non-manufacturing industries. Here the decay of American capitalism reigns supreme and spreads decline with unvarying monotony. The declines differ only in sharpness, and on this basis we can divide the industries roughly into two groups. In the first group employment fell between 5 and 10 percent; in the other group it dropped between 20 and 50 percent.

In the first group are three industries:


Percent Decline
in Employment Industry


Percent Decline
in Payrolls


Electric Light and Power
 and Manufactured Gas


Less than 1%

Wholesale Trade



Retail Trade



The fall in employment in the first of the three industries, like that which occurred in machine tools, took place despite the fact that its output in 1937 was fully 16 percent greater than in 1929. But the decline in employment in the other two industries reflect the decline in business activity and consumption. The reduction in wages amounting to three times as much as in employment shows vividly that here, where unions practically do not exist, the capitalists were able to place big chunks of their own decline on the backs of the workers. In short, where industries did enjoy greater activity and profits, the capitalists alone benefitted. Where they suffered decline, the capitalists shifted the burden on the backs of the workers.

In the second group are industries that have been the very backbone of American economic development and the very sustaining forces of capitalist upswing. But just as they rose most buoyantly in the days of prosperity and progress, so now they crashed most precipitately, carrying downwards with them both employment and payrolls. In the order of falling employment, the industries are:


Percent Decline
in Employment



Percent Decline
in Payrolls


Bituminous Coal



Telephone and Telegraph



Crude Petroleum Products



Electric Railroad and Motorbus
   Operation and Maintenance



Quarrying and Non-metallic Mining






In addition, employment in class I steam railroads, which means the largest railroad systems in the country, fell 33 percent between 1929 and 1937 and construction lost about 35 percent of its employment.

This, then, is the picture of American capitalism. It is declining sharply and, in its decline, it is spreading destruction everywhere – destruction of whole industries, destruction of employment, destruction of purchasing power, destruction of the standard of living of the American workers and farmers. Most of all, it destroys the myth that capitalism is progressive, and that it can give the workers substantial economic concessions without destroying itself entirely. And in doing this declining capitalism smashes the very foundation upon which reformist labor politics rest.

Subsequent articles will deal with the class political significance of reform labor politics in this period of economic decline.



1. These and subsequent computations, are based on the figures given by Standard Statistics in their book of basic statistics. Their source is the Bureau of Labor Statistics. There is some discrepancy between these figures and the figures given by the Survey of Current Business, which also says it draws its estimates from the Bureau of Labor Statistics. The discrepancy between them is often considerable. This may be due to the fact that both are using different indexes published by the same Bureau of Labor Statistics. Since a summary bulletin of employment and payrolls for 1937 has not been published at the time of writing, the choice was between using the Survey of Current Business, published by the Department of Commerce, and using Standard Statistics. I have disregarded the figures of Survey of Current Business, first, because they are based on estimates which minimize and hide the plight of the workers; second, because the periodical uses its official position to spread Chamber of Commerce propaganda about labor. It is therefore not the most reliable source of labor statistics. Standard Statistics, which is an outright capitalist agency selling information to clients and responsible to them for its exactness, is more likely to publish unvarnished facts, without mincing or minimizing.

Why there are two index series, a new one which is for public consumption and the old one which is to be obtained on request, may puzzle those who believe in the glories of democracy, “ours” included, and the impartiality of its statistics.

That does not trouble the Bureau of Labor Statistics which publishes the two series in order to minimize and hide the extent of the declines in employment and therefore, payrolls. Even when it does publish its annual summary of 1937 the figures will have to be used with care. The extent to which it minimizes the true situation can be seen by taking two instances. In its monthly release of Employment and Payrolls of January, 1938, the Bureau of Labor Statistics gives its new series for anthracite and bituminous coal mining, with 1929 as 100. National Income in the United States, 1929-1935 also gives indexes for the two industries and also takes 1929 as 100. That the new series minimize the depression is obvious from the figures:


Indexes for 1934

New Series

Series in National
Income in the US

Employment in Anthracite



Employment in Bituminous



Payrolls in Anthracite



Payrolls in Bituminous



(Sources: Emp. and Payr., Jan, 1938, pp.23ff.;
Nat. Inc. in the US, 1929-1935, pp.83, 85.)

Top of page

Main NI Index | Main Newspaper Index

Encyclopedia of Trotskyism | Marxists’ Internet Archive

Last updated on 4.8.2006