Technological Innovation And The Great Depression

Richard Szostak

• Publisher: Routledge
• Year: 2019
• ISBN:
  9781000314052
• Edition: 1
• Topic: History
• Subtopic: World History
• Index: History

This volume takes an innovative approach toward analyzing the Great Depression of the 1930s. Exploring the technological and employment experience of specific sectors, it looks at trends in income distribution and population and other factors that created the ultimate economic depression.

Part One
Theoretical Considerations

1
The Great Depression Revisited

Most economists would agree that we have as yet no widely accepted explanation of the Great Depression. Various theories have been put forward, and some are felt to be capable of explaining some, though far from all, of the calamitous events of the 1930s (as we will see in Chapter 2). The situation is ripe for a new approach. The purpose of this work is to proffer a technological explanation of much of the Depression experience. While Schumpeter (1939) and others, especially at the time of the Depression, have suggested that technology played a substantial role in causing the Depression, the approach pursued here is quite novel. I will detail in the next five chapters a theoretical explanation of how an abundance of labor-saving production technology coupled with a virtual absence of new product innovation could affect consumption, investment, and the functioning of the labor market in such a way that a large and sustained contraction in employment would result. I will also describe how such a technological situation could have arisen historically. In the succeeding several chapters, I will provide evidence at the industry level that these technological forces were at work, and that they were powerful enough to generate most of the observed unemployment in the 1930s.

The fundamental premise of this work is that the key to understanding the Great Depression is to be found through analysis at the industry level rather than through the highly aggregated research that has dominated the field in the past. This does not mean, of course, that we can turn our back on the concepts of macroeconomics. We must, indeed, couch our analysis carefully in a general equilibrium framework. It is too easy to describe why some sectors of the economy went into decline in or near 1929. This may provide limited insight into why the Depression began, but can tell us little about why recovery was so slow. We must understand why the resources released by some sectors did not — indeed, could not — flow smoothly into other sectors. Why didn’t the labor market clear? Once we remove some simple misconceptions about the functioning of the labor market, and the determinants of consumption demand, we will find that the answer to this question is also to be found at the industry level.

We can already see that there is not just one Depression question. Why was the initial turndown so severe? Why was recovery so slow? We can add others.¹ Why was the recovery to 1936 so feeble and so overwhelmed by the downturn of 1937? How was the Depression transmitted around the world at a time when links between nations were much weaker than today? On the other hand, why did some countries, such as Britain, fare so much better than others in the 1930s? Most previous works have tended to focus on only one of these questions. There is, naturally, no logical necessity that the same factors provide the answer to all. It would, though, be an awful coincidence for a number of horrible events to occur in a short time period for quite different reasons. I will, in what follows, outline an approach which can deal with all of these questions.

This does not mean that I am pursuing some simple monocausal explanation of the entire Depression experience (though I must inevitably focus on my own theoretical approach to the exclusion of others in much of what follows). As Scitovsky (1986) so aptly put it, “Nothing is ever so simple that a single explanation will adequately explain it.” Certainly the Great Depression is not so simple. As noted above, the existing literature does adequately explain some of what happened, but is unconvincing in its attempt to explain the entire Depression. My purpose is to fill much of that gap.

A Synopsis of the Theory Employed in This Work

What is the connection between new product innovation and economic performance? Let us start by asking a question which never arises in standard attempts to comprehend deficient aggregate demand; what goods or services should the mass of unemployed have produced in the 1930s? Full employment would only have been possible if the goods and services produced could be sold profitably. It would seem natural to look ahead to the next period of peacetime full employment and see what workers were producing during the phenomenal postwar economic expansion.

The sectors which provided employment growth to power postwar economic expansion could not have done so in the 1930s. In most cases, this was due to the timing of technological innovation. Television, the modern airplane (the DC-3 of 1935), sulfanomides and vitamins, synthetic fibres, and improved plastics were developed only in the late 1930s. With the exception of the electric refrigerator — where sales expanded steadily through the 1930s — there was quite simply a dearth of new product technology in 1929. Moreover, the “new” products of the 1920s, notably the automobile and the radio, had largely saturated their markets, and thus investment and employment in these sectors were destined to fall in the early 1930s. Sluggish sales during the 1930s and World War Two, along with product improvements, provided the auto industry with much room to expand postwar.

Is it simply a coincidence that none of the sectors which powered post-war growth could possibly have done so in 1929 for technological reasons? To be more modest, we could pose a counterfactual; if this technology had been available in 1929, could the Depression have been avoided? The answer must surely be that if TV, the DC-3, and vitamins were ready to go in 1929, investment and consumption would have been significantly higher, and the Depression much less severe. To recognize that a shortage of growing sectors may be responsible for the depth and length of the Depression is to suggest that perhaps declining sectors — in the absence of counteracting growth potential — may have caused the downturn. The Keynesian multiplier would have time to operate at full force.

What caused consumption expenditure to fall from 1929? To answer this question, we need to take a longer view than is common in economic theory. Our propensity to consume naturally depends on the range of goods available for consumption. Much of the increase in incomes over the last century has been spent on goods and services our great-grandparents had never heard of or could not dream of affording: TVs, VCRs, cars, pharmaceuticals, radios, movies etc. It is surprising, then, that the discussion of particular episodes in the behavior of aggregate consumption demand is never related to the particular bundle of goods available at that time and place. The simple fact is that, in the absence of the creation of new products, aggregate consumption demand can be highly inelastic. Increases in incomes will result in higher savings rates. Moreover, as we enter a world of consumer durables, it not only becomes potentially difficult to extricate an economy from a situation of deficient demand, but quite straightforward to enter such a world. Purchases tend naturally to bunch as a new product is introduced. Automakers and other durable producers, fighting for market share, could not even out the time path of production. Consumption expenditure thus contracted in 1929.

Why did investment not absorb the increased savings, but rather fall sharply from 1929? We can not answer this question if we examine aggregate data alone. Trends in investment can only be understood at the sectoral level. Demographic variables limited the scope for expansion within the traditional necessities. Saturation in durables naturally discouraged investment there. While the rise and decline of industries is a natural characteristic of modern economies, in the late 1920s there were no new products to pick up the slack. Whatever limited investment potential there was in “old” industries had largely been achieved earlier in the decade. An automobile-induced construction boom led to overbuilding of all types of structures in the mid-1920s. A new generation of machine tools appeared at the beginning of that decade which would not be rendered obsolete for decades.

Why was unemployment so high? As Keynesians have long argued, the fall in consumption and investment induced a dramatic fall in output and employment through the multiplier effect (though they could not explain why consumption and/or investment had fallen). A third source of unemployment was labor-saving technology. At the same time that no new products were being developed, it became possible to produce the existing range of goods with a much smaller labor force. Electrification, assembly lines and continuous processing induced the largest decadal increase in labor productivity the country had ever seen in the 1920s. This had contributed to chronic unemployment through that decade. However, many redundant workers were only let go in the downturn of the 1930s. This dishoarding of labor swelled the unemployment rolls.

The idea that labor-saving technology must reduce employment has been around since at least the Industrial Revolution. At the industry level, depending on the division of gains between employer, employee, and consumer, and on the elasticity of demand, this may be true. At the economy-wide level and in the long run we would expect displaced workers to be absorbed elsewhere. This, though, depends on there being expanding sectors. The phrase “technological unemployment” emerged in the 1920s, not because the problem was new but because it involved much greater numbers than before.

In the absence of growing sectors, the multiplier mechanism was able to operate at full force. Those who lost their jobs reduced their expenditures and induced further unemployment. The unemployed were, unless coupled with capital, wholesale raw materials, and access to wide markets, inherently unproductive. For full employment to be restored, entrepreneurs had to come forth who saw profitable opportunities in particular product lines. The multiplier mechanism might then be put to work in the opposite direction. Aggregate expectations remained high until 1932. Never having seen a depression so severe, the natural supposition was that recovery must be imminent. Market saturation and a lack of new products prevented such expectations from being translated into investment and job creation.

Keynesian explanations of the Depression have been justly criticized for relying on coincidental causes of declines in consumption and/or investment. If a surfeit of labor-saving process innovation and deficiency of new product innovation are to be credited with causing the Depression, the timing of innovation must be explained. Economic historians often speak of a Second Industrial Revolution of the late nineteenth century, and identify three main elements: the intemal-combustion engine, economical electric power production, and a revolution in chemical understanding and application. It is widely recognized that much of the technological — and economic — history of the twentieth century is derivative of these three breakthroughs. A perusal of a list of key innovations of the first half of the twentieth century, indeed, indicates that the most important innovation which can not be traced to any of these three is the zipper.

If these three new sources of technological potential had spawned a steady stream of new products over the next decades, economic stability might have been ensured. This was not to be the case. The automobile was a (temporarily) spent force in 1929, but the modern aircraft was not yet ready. The most far-reaching product breakthroughs in chemistry — in plastics, synthetic fibers, and pharmaceuticals — occurred in the 1930s. The radio expanded and declined somewhat less spectacularly than the automobile in the 1920s; the television was not perfected for another decade. Each of the three encouraged widespread process innovation in the 1920s. The automobile spawned the assembly line in 1913. The counterpart to the assembly line for homogeneous output is continuous processing; this was developed in chemicals and spread through such industries as food processing and oil refining in the 1920s. Electrification replaced steam engine belting, allowed improved factory layout, facilitated flexibility in machine operation, and encouraged a 50% increase in horsepower per worker in that one decade. Manufacturing output rose 64% in the 1920s, with virtually no change in employment.

The time-path of innovation is a function of both technical and social considerations. The unfolding of the potential of the Second Industrial Revolution could not be unaffected by the rise of the industrial research laboratory (itself largely a creature at first of the chemical and electrical industries). These, emerging shortly after the turn of the century in the United States, emphasized process improvement and minor product development almost exclusively in their first years of existence. While these have remained their foci, a greater role has been accorded radical product innovation in later decades. Since these labs attracted scarce human resources from other settings and competed with independent innovators in both marketplace and patent office they could skew inventive effort away from new product development through the first decades of this century.

Major innovations generally evolve over a period of decades. It is incorrect, therefore, to think that innovation can quickly respond to economic conditions. Schumpeterian entrepreneurs do not, in fact, have access to a pool of dormant product technology when they feel the time is right. While the Depression itself might encourage some types of research, such investigations could only build on the existing technical base.

Can secular (i.e. slowly acting) forces explain the Great Depression? Gordon (1961: 449) raised this question, and replied that, “It is not sufficient to reply that secular stagnation was also operating in the 1920s but was temporarily offset by a speculative boom and by the investment opportunities created by the spread of the automobile and electric power.” Yet surely he overstates the case. Fleisig (1974) is one who has argued that secular forces could be causes of the Depression. Boulding has warned that we must be aware of the great discontinuities in the world; we live on the tablelands and must not fall off the cliffs. Fisher (1933) characterized the economy as a ship on the ocean; normally it naturally rights itself but in the Depression it tipped too far and thus disequilibrating forces came to dominate. Depressing influences can thus exist for some time largely unnoticed, and then be triggered to full effect. We argue, for example, that much redundant labor was hoarded in the 1920s only to be summarily dishoarded in the 1930s. Note also that, as we will see below, the 1920s were not a decade of prosperity for all, and thus the Depression begins from a position of weakness.

Still, the very notion of a trigger mechanism implies that secular forces must be coupled with some that act more suddenly. Our approach contains many of these. We could note in particular that without the spectacular rise and fall of the automobile industry the American interwar period would have looked like the British: no spectacular decline but two decades of lackluster performance.

We should briefly discuss how this theoretical approach relates to the existing literature on the Great Depression (the subject of Chapter 2). Much of that literature has been cast ill terms of a debate between two macroeconomic views: the monetary approach and the Keynesian approach. Whereas decades ago adherents of each may have claimed that they had the answer, it is now widely recognized that no one theory adequately explains the entire Depression experience. The major problems with the Keynesian approach are that it has not been able to tell us why consumption and/or investment should have fallen so precipitously from 1929, and that it can not explain why recovery was so sluggish. As this book’s approach can answer these questions, it can be viewed as complementary to the Keynesian position. Our discussion of labor markets in Chapter 5 and of structural elements throughout should firmly establish, though, that this book does not strictly adhere to a Keynesian view.

Some Keynesians have argued that the cause of the downturn is to be found in expectations; people became pessimistic about the future and thus postponed investment (and consumption) decisions. It is always difficult to ascertain what people were thinking at any point in time. The evidence we have, though, suggests strongly that expectations were high well into the 1930...