FS IV 92 - 16
Small Business in Industrial Economics:
The New Learning
David B. Audretsch
September 1992
ISSN N r. 0722 - 6748
F orschungssch werpunkt Marktprozeß und Unter
nehmensentwicklung (IIMV) Research Unit
Market Processes and
Corporate Development (DM)
Small Business in Industrial Economics: The New Learning, Discussion Paper FS IV 92 - 16, Wissenschaftszentrum Berlin, 1992.
Wissenschaftszentrum Berlin für Sozialforschung gGmbH, Reichpietschufer 50, W-1000 Berlin 30, Tel. (030) 2 54 91 - 0
The New Learning
The focus o f analysis in industrial organization has traditionally been on the largest firms comprising industries. However, a recent wave of studies has not only identified that small firms are playing an increased role in industrial markets, but that their economic function is distinct from that o f larger enterprises. The purpose o f this paper is to weave together these studies which, combined together, constitute a new learning with respect to the role of small firms in industrial markets. In particular, the role o f small firms under static analysis in the industrial organization literature is contrasted to the dynamic role that emerges when viewed through a more evolutionary lens.
ZUSAMMENFASSUNG
Kleinunternehmen in der Industrieökonomik:
Ein neuer Ansatz
In der Industrieökonomik lag der Schwerpunkt des Interesses traditionell auf den Großunternehmen. Eine Reihe aktueller Arbeiten hat jedoch nachgewiesen, daß kleinen Unternehmen eine entscheidende Rolle beim wirtschaftlichen Geschehen zukommt, wobei sich ihre Funktion jedoch grundlegend von der größerer Unternehmen unterscheidet. Die vorliegende Arbeit beabsichtigt, die erwähnten aktuellen Arbeiten auf diesem Gebiet miteinander in Verbindung zu setzen. - Als ein Ganzes gesehen, stellen diese Arbeiten einen neuen Ansatz bei der Bewertung der Rolle kleiner Unternehmen dar. Besondere Aufmerksamkeit erfahrt in diesem Zusammenhang der Unterschied zwischen der herkömmlichen statischen Analyse der Industrieökonomik und der dynamischen Betrachtung, die im M ittelpunkt des evolutionären Forschungsansatzes steht.
I. Firm Size and Industrial Organization
In 1961, Soviet Prem ier Nikita Kruschev banged his shoe on the negotiating table of the United Nations and challenged President John F. Kennedy, "We will bury you".
The West took K ruschev's threat seriously. At the heart o f K ruschev's challenge was not necessarily a military threat, but rather an economic threat. A fter all, not only had the Soviets beaten the Americans in the space race with the launching o f the Sputnick ju st several years earlier, but perhaps even more disconcerting was the growth in Soviet productivity, which at least at that time appeared to greatly exceed that in the West during the 1950s.1
Thus, by the early 1960s there was little doubt among politicians, intellectuals, and economists about the credibility of the threat from the East. Even as late as 1966, the Joint Economic Committee of the United States Congress warned o f a "...planned average annual increase in industrial output of 8.0-8.4 percent during 1966-70" in the Soviet Union (Noren, 1966, p. 301). After all, the nations o f Eastern Europe, and the Soviet Union in particular, had a "luxury" inherent in their systems o f centralized planning -- a concentration of economic assets at a scale beyond anything imaginable in the West. F o r example, before the Berlin Wall fell, the East German economy consisted o f 224 firms -- Kombinate, or combines -- of which around 180 were in manufacturing (Bannasch, 1991). There was essentially one firm, and one firm only, for each m ajor manufacturing industry. This degree o f concentration and centralization was the rule and not the exception in Eastern Europe. By contrast, in the West, the commitment to democracy represented a concomm itant com m itm ent to political and economic decentralization. Thus, the Soviet Union and its Eastern European partners were free to concentrate economic assets and exploit economies o f scale in a m anner that could never be acceptable in the West.
Inherent in the M arxist system was what was termed by the Soviet Central Planners
"giantism", and which Nathan Rosenberg (1992, p. 197) has characterized as " ...a n unabaiding faith in, and commitment to, large-scale production." The intellectual antecedent for this giantism undoubtedly lies in Karl M arx's (1912) admiration for the large-scale technologies o f the Britsch Industrial Revolution. This was reflected by M arx's forceful articulation of the view that, in the competitive process, the large- scale capitalist always beats out his smaller counterpart. M arx, in fact, had written,
"The battle o f competition is fought by the cheapening o f commodities. The
t See Moore (1992) for a recent documentation of the "...view held widely at the time that Soviet central planning would produce persistently high growth rates into the foreseeable future” (1992, p. 72).
The fear in the West was not only that the massing o f economic assets would lead to unprecedented productivity growth in the Soviet Union and Eastern Europe, but perhaps what was o f paramount concern was the assumed leaps and bounds in techniological progress that would emerge from the huge and concentrated research and development programs that were being assembled. From the vantage point o f the late 1950s and early 1960s, the West seemed not only on the verge o f losing the space race, but perhaps even more importantly, the economic growth race.
W hile there may have been considerable debate about what to do about the perceived Soviet threat some three decades ago, there was little doubt at that time that firm size mattered. And even more striking, when one reviews the literature o f the day, there seemed to be near unanimity about the way in which firm size mattered. It mattered in terms o f the ability of firms to exhaust scale economies, to generate innovative activity, and ultimately to spur economic growth.
It is no doubt an irony o f history that a remarkably similar version o f the giantism inbedded in the Soviet Doctrine, which was fueled by the writings o f M arx and ultimately implemented by the iron fist o f Stalin, was also prevalent throughout the West. It must not be forgotten that the 1950s and 1960s represented the pinnacle o f what Michael Piore and Charles Sabel (1984) have termed the era o f mass-production.
During this era, economies of scale seemed to be the decisive factor in dictating efficiency. Why did the United States dominate world trade in industries such as automobiles and steel during this era? Presumably because the greatest endowment o f physical capital was to be found in the U .S ., enabling its firms and industries to most fully exploit scale economies and enjoy the highest levels o f productivity growth in the post-W orld W ar II West. This was the world so colorfully described by John Kenneth Galbraith (1956) in his theory o f counterveiling power, where the pow er o f big business was held in check by big labor -- that is, unions — and by big government. This was the era o f the "man in the gray flannel suit" and the
"organization man"3, where virtually every major social and economic institution acted to reinforce the stability and predictability needed for mass-production.
N ot only was the large corporation thought to have superior productive efficiency but also to be the engine o f technological change and innovative activity. A fter all, Schumpeter wrote in 1950, "What we have got to accept is that the large-scale establishment has come to be the most powerful engine o f progress" (1950, p. 106).
A few years later, Galbraith (1956, p. 86) echoed Schum peter's sentiment when he cheapness o f commodities depends, ceteris paribus, on the productiveness o f labor, and this again on the scale o f production. Therefore, the large capitals beat the smaller. "2
3
Quoted from Rosenberg (1992, p. 197).
For a description of these see Whyte (1960).
lamented, "There is no more pleasant fiction than that technological change is the product of the matchless ingenuity o f the small man forced by competition to employ his wits to better his neighbor. Unhappily, it is a fiction."
Thus, regarding East-West relations, optimists, such as Galbraith, spoke o f a convergence between the communist systems o f Eastern Europe and the W estern style of "managed" capitalism. It seemed that both the East and W est were converging towards economies dominated by ju st a handful o f powerful enterprises, constrained only by the counterveiling powers o f the state and workers. The only "trivial"
difference would be the ownership.
The pessimists, on the other hand, were alarmed into thinking that perhaps the West would, after all, be buried by the East - productivity gains and a surge o f economic emanating from the overpowering Soviet Combines would simply overwhelm the outdated and out-scaled firms in the West, burdened with antiquated constraints, such as antitrust laws.
In fact, as has been made all too clear by the events throughout the Soviet Union and Eastern Europe in the last three years, neither the pessimists nor the optimists in the West were correct. That is, neither did the two economic systems converge in the manner that economists like Galbraith had predicted, nor was the W est buried by an avalanche o f productivity growth and wave o f innovative activity from the East.
What happened? What went wrong? A paradox seems to be that the industrial organization of Eastern Europe, which was structured around the principle o f giantism and which placed such a high premium on economic growth, resulted in exactly the opposite - stagnation, both in terms o f production and in terms o f technological change.
Marx had warned that because "One capitalist kills many", that ultimately,
"Capitalism bears the seeds o f its own self-destruction." In fact, in what must be one of the greater ironies o f history, the mature capitalist countries o f the W est have been going through not a process o f concentration and centralization during these past two decades, as Marx had predicted, but rather a process o f deconcentration and decentralization. Consider the example o f the United States. Between 1947 and 1980, real gross national product (GNP) per firm rose by nearly two-thirds, from $150,000 to $250,000 (Brock and Evans, 1989). Curiously, however, ju st within the following six years, it dropped sharply, by 14 percent to $210,000.4
Such a drop in the average firm size, is consistent with the startling findings by David Birch (1981) from his long-term study o f U .S. job generation. Despite the prevailing conventional wisdom at the time, Birch (1981, p. 8) reported that "..w hatever else
4 For a careful documenatation of how the firm-size distribution has shifted throughout developed nations, see the individual country studies contained in Acs and Audretsch (1993), and Loveman and Sengenberger (1991).
The recent shift in economic activity towards small-scale firms has not escaped the attention o f the popular press. For example, The Economist reports: "Despite ever- larger and noisier mergers, the biggest change coming over the world o f business is that firms are getting smaller. The trend o f a century is being reversed. Until the mid-1970s, the size o f firms everywhere grew; the numbers o f self-employed fell.
Ford and General Motors replaced the carriage-m aker's atelier; M cD onald's, Safeway and W .H . Smith supplanted the com er shop. No longer. Now it is the big firms that are shrinking and small ones that are on the rise. The trend is unmistakable -- and businessmen and policy-makers will ignore it at their p eril."6
O f course, there is a temptation to attribute this relative increase in the num ber of small firms to the obvious long-term transition of employment into services and out of the manfuacturing sector. However, the shift in economic activity towards smaller- scale enterprises has actually been more pronounced in manufacturing than in services. In 1976 firms with fewer than 500 employees (what the U .S. Small Business Administration defines as constituting a "small enterprise") accounted for about twenty percent of all manufacturing sales in the United States. By 1986 the small-firm sales share had risen to around twenty-five percent (Acs and Audretsch, 1993).
This shift in manufacturing activity away from large corporations towards small firms has not been restricted just to the United States. Rather, as Table 1 shows, a substantial shift in economic activity away from large corporations and towards small enterprises has been experienced thoughout the developed western nations. In order to facilitate international comparisons, Table 1 focuses on one measure -- the share o f employment accounted for by small firms - and tracks it between the 1970s and 1980s.7 W hile the magnitude o f the shift varies considerably across nations, the they are doing, large firms are no longer the major providers o f jobs for Americans."
Rather, Birch claimed to have discovered that most new jobs emanated from small firm s.5
The exact methodology, application, and interpretation of the underlying data used to make inferences in the job studies have been called into question (Armington and Odle,
1982; Storey and Johnson, 1987; FitzRoy, 1989; and Brown, Hamilton and Medoff, 1990). As Brown, Hamilton and Medoff (1990) point out, job generation may be a deceptive measure because many of the newly generated jobs subsequently disappear.
That is, without consideration of the number of job disappearances, focusing solely on the amount of job generation emanating from small firms is misleading and results in an overstatement of the amount of economic activity actually stemming from small firms (Storey, 1990).
"The Rise and Rise of America's Small Firms," The Economist, 21 January, 1989, pp.
173-74.
A similar shift away from large manufacturing firms towards small enterprises has also been observed for Japan (Sato, 1989).
Table 1: Summary fron Country Studies of Small—Firm Share of Wannfantiri ng Employment and Shift Over Time a
Country Year
Small-Firm Emoloyment
Share (%) Year
Small-Firm Emoloyment
Share (%) Change
United Kingdom 1986 39.9 1979 30.1 + 9.8
Federal Republic
of Germany 1987 57.9 1970 54.8 + 3.1
United States 1987 35.2 1976 33.4 +1.9
Netherlands 1986 39.9 1978 36.1 + 3.8
Portugal 1936 71.3 1932 68.3 + 3.5
Italy c
North 1987 55.2 1931 44.3 + 10.9
South 1987 68.4 1981 61.4 + 7.0
Czechoslovakia 1988 1.4 1954 13.0 -11.6
East Germany 1936 1.1 — — —
Poland b 1985 10.00 1937 33.0 -23.0
a A small firm is defined as an enterprise with fewer than 500 employees, unless designated otherwise.
k A small firm is defined as an enterprise with fewer than 100 employees.
c A small firm is measured as an enterprise with fewer than 200 employees.
Source: Acs and Audretsch (1993)
direction does not. The shift in the firm-size distribution ranged from an increase o f 1.9 percentage points in the small-firm employment share in the United States between 1976 and 1986 (Acs and Audretsch, 1993) to an increase o f 10.9 percentage points between 1981 and 1987 in the North of Italy (Invemizzi and Revelli, 1993), and to an increase o f 9.8 percentage points between 1976 and 986 in the United Kingdom (Hughes, 1993).
The strength and prevalence o f this shift across a broad range o f industrialized nations confronts industrial organization economists with three fundamental questions:
(1) What is the economic role o f small firms in industrial markets? That is, why do they exist and how are they able to exist?
(2) Why has the organization o f industries shifted so dramatically towards smaller enterprises over the last two decades? and
(3) Is this shift desireable or undesireable (the welfare question)?
n . Why Do Small Firms Exist?
Jakob Viner, in his classic 1932 article predicted that, based on the assumption that individual firms have identical U-shaped long-run average cost functions, a unique firm-size distribution -- that is, a singular firm size — would be found in every industry. Two years later, Nickolas Kaldor (1934) similarly argued that a singular optimal firm size would result from the fixed input o f entrepreneurial talent into the firm production function.
That a clear rift existed for decades between these pioneering theoretical predictions o f a singular optimal firm size and the obvious empirical observation o f a vast spectrum o f firm sizes does probably not need to be pointed out here. It was not until some forty-five years later that Robert Lucas (1978) proposed the first major theory explaining the simultaneous existence o f large and small firms in a given industry.
Lucas assumed that all individuals are identical as workers but are endowed with varying amounts o f entrepreneurial ability, where an entrepreneur is defined as someone who combines labor and capital inputs into output through some production process. Given a distribution o f entrepreneurial ability, each individual relates his/her expected profits from undertaking entrepreneurial activity to the com petitive wage and decides whether to become an entrepreneur or remain a w orker.8 But those individuals who decide to become an entrepreneur also possess varying amounts o f managerial com petence.9 Those managers who are endowed with a greater amount o f
Models of entrepreneurial choice have been developed further by Evans and Jovanovic (1989), Evans and Leighton (1989, 1990a and 1990b), de Wit and van Winden (1989 and 1991) and Blanchflower and Meyer (1993).
9 For an extension of the Lucas model, see Jovanovic (1993).
competence are able to manage larger enterprises. Those endowed with only a more modest amount o f ability are restricted in the size o f the firm. The resulting distribution o f intra-industry firm sizes is therefore attributable to the given talent distribution o f entrepreneurs.
The major accomplishment o f Lucas was to resolve the paradox posed between the microeconomic theory of Viner and Kaldor and the very real world observation o f a tenaciously persistent asymmetric firm size distribution over time. But why should that firm-size distribution vary so much across industries? Here Lucas provided us with no clues.
Lucas' model also provides no insight as to why entrepreneurs should choose to enter different industries at different rates. That is, as has now been well established in the industrial organization literature, the entry o f firms is anything but constant across industries.10 In fact, a different strand o f industrial organization literature argues that variations in entry occur accross industries because the degree that profits in excess of the long-run equilibrium profit rate also varies across markets. This model makes good sense in many situations. For example, when the barrel price o f crude oil skyrocketed from $3.00 to $12.00 in 1973, the subsequent response to the unprecedented profitability enjoyed by the incumbent firms was a wave o f entry that lasted at least throughout the remainder o f the decade.
But what about, say, the American steel market between the mid-1970s and mid- 1980s? The traditional model in industrial organiztion would predict that negative net entry, that is exit, would occur in this industry during this time. In fact, while some of the large integrated steel producers did eventually exit from the market, the industry experienced a wave o f new entrants, mostly in the form o f the "mini-mills".
A similar wave o f new-firm start-ups occurred throughout the mature and declining U.S. industries during the late 1970s and early 1980s.
Why did, for example, Chaperal Steel, a mini-mill in Texas enter the industry?
Probably not in spite o f the negative profits prevalent throughout the steel industry, but rather because o f them. That is, Chaperal Steel was essentially serving as an agent o f change in the market. Like virtually all o f the mini-millls, it used different inputs, produced different products, and applied different managerial techniques than
For a recent compilation of country studies identifying the determinants of entry across industries, see Geroski and Schwalbach (1991).
those practiced by the large integrated steel m ills.11 The traditional model o f entry in industrial organization assumes a constant product, production technology, inputs, and input prices. According to this model, entry is about business as usual — it is ju st that the entrants provide more o f it.
In fact, much of entry may not be at all about business as usual. Rather, the entrant is betting that (s)fie can do better than the incumbent firms by doing something different. Thus, there are three fundamental functions o f entry in industrial organization:
(1) As the traditional model implies, entry serves to erode excess profits by increasing the amount of output in the market,
(2) as Stigler (1964) pointed out, entry also serves to raise the cost o f reaching a collusive agreemeent and enforcing that agreement, as well as lowers the benefits accruing from such agreements, and
(3) as an agent o f change.
W hile the first two functions are essentially equilibrating in nature, in that they restore the market back to the long-run equilibrium profit rate and price, the third function is fundamentally disequilibrating in nature.
HI. Technological Change
The notion o f entry as an agent o f change, that is, in engaging in innovative activity invites a certain paradox. As Cohen and Levin (1989) point out, virtually every study has found that innovative output is the product o f knowledge-generating inputs, of which research and development (R&D) is the most crucial. In addition, there are substantial scale economies associated with large R&D laboratories (Cohen and Klepper, 1991 and 1992) and large corporations are better situated to appropriate the rents accruing from investments in R&D (Cohen et al., 1987; and Levin et a l., 1985).
Thus, as Scherer (1991) observes, about ninety percent o f private R&D in the United States is undertaken by ju st four hundred large corporations. Not surprisingly, as Table 2 shows, the most innovative American firms rank not only among the largest industrial corporations, but also have invested in substantial R&D laboratories. The measure o f innovative activity used in Table 2 was introduced by Acs and Audretsch
11 For example, The Wall Street Journal (10 May, 1991, p. 1) reports that "Wall Street has been in love with Nucor Corp.", which has become the seventh largest steel company through its fifteen mini-mill plants. Nucor has pursued a strategy not only of
"...declaring war on corporate hierarchy", but also by being "...terribly efficient, aggresively non-union and quite profitable. Most of its fifteen mini-mills and steel- fabrication operations are situated in small towns, where they have trained all sorts of people who never thought they'd make so much money. And Nucor has developed a revolutionary new plant that spins gleaming sheet steel out of scrapped cars and refrigerators."
Most Innovative Firms Table 2
Firm
Number of Innovations
Sales (millions)
R&D Expenditure
(millions)
R&D/Sales (%)
Hewlett Packard 55 981.0 89.6 9.1
Minnesota Mining & Mfg. 40 3127.0 143.4 4.6
General Electric 36 13399.0 357.1 2.7
General Signal 29 548.0 21.2 3.9
National Semiconductor 27 235.0 20.7 8.8
Xerox 25 4054.0 198.6 4.9
Texas Instruments 24 1368.0 51.0 3.7
Pitney Bowes 22 461.0 10.5 2.3
RCA 21 4790.0 113.6 2.4
IBM 21 14437.0 946.0 6.6
Digital Equipment 21 534.0 43.5 9.1
Gould 20 773.0 23.1 3.0
Motorola 19 1312.0 98.5 7.5
Wheelabrator Frye 19 332.0 2.0 0.6
United Technologies 18 3878.0 323.7 ' 8.3
Hoover 18 594.0 4.3 0.7
Honeywell 18 2760.0 164.2 5.9
Rockwell International 17 4943.0 31.0 0.6
Johnson & Johnson 17 2225.0 97.9 4.4
Eastman Kodak 17 4959.0 312.9 6.3
Data General 17 108.0 11.6 10.8
Exxon 16 44865.0 187.0 0.4
Du Pont 16 7222.0 335.7 4.6
Stanley Works 15 464.0 3.5 0.7
Sperry Rand 15 3041.0 163.5 5.4
Pennwalt 15 714.0 15.7 2.2
North American Philips 14 1410.0 22.5 1.6
Harris 14 479.0 21.1 4.4
General Motors 14 35725.0 1113.9 3.1
Becron, Dickinson 14 456.0 17.8 3.9
Source: Acs and Audretsch (1991)
(1987, 1988, and 1990) and contains, in principle, each new product and process innovation made in 1982. An innovation is defined as, " ...a process that begins with an invention, proceeds with the development o f the invention, and results in the introduction of a new product, process or service to the marketplace" (Edwards and Gordon, 1984). Because the innovations recoreded in 1982 were the result o f inventions made, on average, 4.2 years earlier, and allowing for a two-year lag between the expenditures on R&D and the invention, 1975 firm R&D and sales data are presented along with the innovation measure.12
Not only does R&D constitute the crucial input in generating innovative output, but, as M ueller and Tilton (1969, p. 578) argue, it also constitutes a barrier to the entry o f new firms because, "The chief component of these barriers generally is the extent o f economies o f scale in the R&D process. The second major factor contributing to R&D entry barriers is the accumulation o f patents and know-how on the part o f incumbent firms." From the theoretical arguments and empirical evidence, it would seem that highly innovative industries, which clearly require a substantial R&D input, are not only not particularly hospitable to the start-up o f new firms, but in fact represent a singularly hostile environm ent.13
But what kind o f firms actually comprise high-technolgoy industries? One does not have to drive through Silicon Valley in California, on route 128 around Boston, or in Cambridge, England, to conclude that the answer is frequently, "Firm s that most o f us have never heard of." As Kenney (1986) emphasizes, the biotechnology industry is comprised virtually entirely o f small firms. Despite repeated efforts, the large drug companies have failed at entering the biotechnology industry.
So the paradox is that R&D apparently represents a crucial input into generating innovations, therefore creating not only substantial scale economies in R&D laboratories, but also significant barriers to entry in R&D intensive industries. Yet new entrants and small firms, which are seemingly without the advantages o f large R&D laboratories,14 contribute to at least some o f that innovative activity, especially in high-technolgoy indsutries.
A t least some resolution to this paradox was provided by Sidney W inter (1984), who hypothesized about the existence o f what was termed as two distinct "technological
12 The R&D data are taken from the 1975 Business Week survey of the corporations spending the most on R&D.
13 The hypothesis that R&D intensity impedes entry has at least some empirical support.
Orr (1974) found that Canadian net entry was adversely affected by R&D intensity, and Baldwin and Gorecki (1987) found that entry via plant creation is negatively related to R&D.
14 Kleinknecht (1987 and 1989) and Kleinknecht et al. (1991) have found that official measures of firm expenditures on R&D tend to underestimate the extent of R&D in small firms, which is generally performed outside of formal R&D laboratories.
regimes". According to W inter (1984, p. 297), "An entrepreneurial regime is one that is favorable to innovative entry, and unfavorable to innovative activity by established firms; a routinized regime is one in which the conditions are the other way around."
Although never rigorously tested, there is at least some evidence consistent with the notion o f separate technological regimes which are alternatively conducive to innovative activity by incumbent and new firms. Table 3 shows, the most innovative U.S. industries (at the four-digit standard industrial classification level). In some o f the most innovative industries, the large firms, which are defined as having at least 500 employees, clearly contributed most o f the innovations. F or excample, in the aircraft industry, the large corporations made 31 o f the 32 innovations. Similarly, in the pharmaceutical preparations industry, the large enterprises proved to be about fifty percent more innovative, and in the photographic equipment industry, around ninety percent of the innovations emanated from large corporations.
By contrast, in other industries the small firms clearly proved to be more innovative.
For example, 45 of the innovations in the measuring and controlling devices indsutry were made by small enterprises, while only 3 came from large corporations.
Similarly, in the computer industry, the small firms contributed about fifty percent more innovations than did their larger counterparts. And in optical instruments and lenses, 21 o f the 34 innovations emenated from small enterprises. In fact, Acs and Audretsch (1987, 1988 and 1990) identified a clear systematic tendency for the innovative activity o f small firms to be promoted in a considerably different economic environment than that o f large enterprises.
The new learning has raised a number o f explanations why smaller enterprises may, in fact, tend to have the observed innovative advantage, at least in certain industries.15 Rothwell (1989) suggests that the factors yielding small firms with the innovative advantage generally emanate from the difference in management structures between large and small firms. Similarly, Scherer (1991) argues that the bureaucratic organization o f large firms is not conducive to undertaking risky R&D. The decision to innovate must survive layers o f bureaucratic resistence, where an inertia regarding risk results in a bias against undertaking new projects. However, in the small firm the decision to innovate is made by relatively few people.
Second, innovative activity may flourish the most in environments free o f bureaucratic constraints (Link and Bozeman, 1991). That is, a num ber o f small-firm ventures have benefited from the exodus o f researchers who felt thwarted by the managerial restraints in a larger firm. Finally, it has been argued that while the larger firms reward the best researchers by promoting them out o f research to management positions, the smaller firms place innovative activity at the center o f their competitive
15 See the statements before the U.S. Congress by Scherer (1988) and Acs and Audretsch (1988).
Table 3: dumber of Significant Innovations for Large and
Small Firms in tlie Most Innovative Industries, 1982*
Industry Total
Innovations
Large-Firm Innovations
Small-Firm Innovations
Electronics computing equipment 395 158 227
Process control instruments 165 63 93
Radio and TV communication equipment 157 83 72
Pharmaceutical preparations 133 120 72
Electronic components 128 54 73
Engineering and scientific instruments 126 43 83
Semiconductors 122 91 29
Plastics products 107 22 82
Photographic equipment 88 79 9
Office machinery 77 67 10
Instruments to measure electricity 77 28 47
Surgical appliances and supplies 67 54 13
Surgical and medical instrument 66 30 36
Special industry machinery 64 43 21
Industrial controls 61 15 46
Toilet preparations 59 41 13
Valves and pipe fittings 54 20 . 33
Electric housewares and fans 53 47 6
Measuring and controlling devices 52 3 45
Food products machinery 50 37 12
Motors and generators 49 39 10
Plastic materials and resins 45 30 15
Industrial inorganic chemicals 40 32 8
Radio and TV receiving sets 40 35 4
Hand and edge tools 39 27 11
Fabricated platework 38 29 9
Fabricated metal products 35 12 17
Pumps and pumping equipment 34 18 16
Optical instruments and lenses 34 12 21
Polishes and sanitation goods 33 13 19
Industrial trucks and tractors 33 13 20
Medicinals and botanicals 32 27 5
Aircraft 32 31 1
Environmental controls 32 22 10
* Large and small-firra innovations do not always sum to total innovations, because several innovations could not be classified according to firm size.
Source: Acs and Audretsch (1990)
strategy (Scherer, 1991). According to Scherer (1988, pp. 4-5), "Smaller enterprises make their impressive contributions to innvoation because o f several advantages they possess compared to large-size corporations. One important strength is that they are less bureaucrateic, without layers o f 'abominable no-men' who block daring ventures in a more highly structured organization. Second, and something that is often overlooked, many advances in technology accumulate upon a myriad o f detailed inventions involving individual components, materials, and fabrication techniques.
The sales possibilities for making such narrow, detailed advances are often too modest to interest giant corporations. An individual entrepreneur's juices will flow over a new product or process with sales prospects in the millions o f dollars per year, whereas few large corporations can work up much excitement over such small fish, nor can they accomodate small ventures easily into their organizational structures.
Third, it is easier to sustain a fever pitch o f excitement in small organizations, where the links between challenges, staff, and potential rewards are tight. 'A ll-nighters' through which tough technical problems are solved expeditiously are common."
The research laboratories o f universities provide a source o f innovation-generating knowledge that is available to private enterprises for commercial exploitation. Jaffe (1989) and Acs, Audretsch and Feldman (1992), for example, found that the knowledge created in university laboratories "spills over" to contribute to the generation o f commercial innovations by private enterprises. Acs, Audretsch, and Feldman (1993) found considerable evidence that the spillovers from university research contribute more to the innovative activity o f small firms than to the innovative activity o f large corporations. Similarly, Link and Rees (1990) surveyed 209 innovating firms to examine the link between firm size and university research.
They found that large firms are more active in university-based research. However, small- and medium-sized enterprises apparently are better able to exploit their university-based associations in generating innovative activity. Link and Rees (1990) conclude that, contrary to the conventional wisdom, diseconomies o f scale in producing innovations exist in large firm s.16 They attribute these diseconomies o f scale to the "inherent bureaucratization process which inhibits both innovative activity and the speed with which new inventions move through the corporate system towards the market" (Link and Rees, 1990, p. 25).
IV. Firm Selection and Industry Evolution
The important contribution that small firms make in technological change can help to reconcile two empirical results that have emerged consistently in the literature and that pose something o f a puzzle to industrial organization economists. The first, which has received considerable attention at least since the seminal study by Simon and Bonini (1958) more than three decades ago, is the persistence o f an asymmetric firm-
For more specific evidence that diseconomies of scale exist in the generation of innovative activity with respect to firm size and R&D inputs, see Acs and Audretsch (1991), and Audretsch and Acs (1991).
size distribution predominated by small enterprises. Such an asymmetric firm size distribution has been observed to not only exist in virtually every manufacturing industry (Acs and Audretsch, 1990, chapter four) o f every developed industrial nation, but also to persist with remarkable stability over decades. This is consistent with the observation by Weiss (1964 and 1976), Scherer (1973), and Pratten (1971) that the bulk of firms in most industries is operting at a suboptimal scale o f output.17 The second puzzle, which is somehwat more tentative, is the confirmation by a number o f studies that the entry o f new firms into an industry is apparently not substantially deterred in industries where scale economies play an important ro le.18 These findings raise two troubling questions to industrial organization economists,
"Why is it that most enterprises in virtually every manufaturing industry o f every industrialized nation are small, and how are they able to remain viable when confronted by an inherent scale disadvantage?", and "Why are entrepreneurs not more noticeably deterred from entering industries characterized by substantial scale economies?"
The resolution of this paradox lies in a new learning regarding the dynamic role that small firms in the process o f firm selection and industry evolution. Jovanovic (1982) proposed a model where new entrants face random costs which differ across firms. A central feature o f his model is the assumption that a new firm does not know what its costs, i.e. its relative efficiency, will be, but rather discovers this through the process of learning from its actual post-entry performance. In particular, Jovanovic (1982) assumed that entrepreneurs are unsure about their ability to manage a new start-up and therefore their prospects for success. While entrepreneurs may establish a new firm based on a vague sense of expected post-entry performance, they only discover their true ability to manage in the given environment once their business is established.
Those entrepreneurs who discover that their ability exceeds their expectations expand the scale o f their business, while those discovering that they have less inherent talent than expected contract the scale o f output or possibly even exit from the industry.
Thus, Jovanovic's (1982) model is a theory o f "noisy" selection, where efficient firms grow and survive and inefficient firms decline and fail.
Jovanovic's (1982) theory o f the post-entry perform ance o f new firms is particularly appealing in view o f the rather startling small size o f most new firms. F or example, Audretsch (1991 and 1993) shows that the mean size o f more than 11,000 new start
ups in U .S. manufacturing in 1976 was less than eight workers per firm. W hile the
17 For example, Weiss (1991, p. xiv) observed that, "In most industries the great majority of firms is suboptimal. In a typical industry there are, let's say, one hundred firms.
Typically only about five to ten of them will be operating at the MES (minimum efficient scale) level of output, or anything like it."
18 See for examples, Acs and Audretsch (1989 and 1990), Evans and Siegfried (1992), Austin and Rosenbaum (1991), and the country studies in Geroski and Schwalbach (1992).
MES varies substantially across industries, and even varies to some degree across various product classes within any given industry, the observed size o f most new start-ups is sufficiently small to ensure that the bulk o f new firms will be operating at a suboptimal scale of ooutput. Why would an entrepreneur start a new firm that would immediately be confronted by scale disadvantages? An implication of Jovanovic's (1982) theory is that firms may begin at a small and even suboptimal scale o f output and then, if merited by subsequent performance, expand. Those firms which are successful grow, while those that are not successful remain small and may be forced to exit from the industry if they are operating at a suboptimal scale of output.
An important ingredient in determining whether or not a firm is successful may be its ability not ju st to replicate the exact products and production methods o f the incumbent enterprises, but to fulfil the role discussed in the previous section as and
"agent o f change". This is consistent with the extension o f Jovanovic's (1982) theory by Pakes and Erikson (1987), who argue that firms can actively accelerate the learning process by investing in knowledge-creating activities, such as R & D .19
In fact, studies by Mansfield (1962), Dunne, Roberts, and Samuelson (1988 and 1989), Evans (1987a and 1987b), Phillips and Kirchhoff (1989), Audretsch (1991 and 1993), and Hall (1987) have established that (1) the likelihood o f survival is positively related to firm size and age, and (2) firm growth tends to be negatively related to firm size. But none o f these studies focused on why new-firm survival and growth should vary so much across different manufacturing industries. Based on a longitudinal data base tracking the post-entry perform ance o f over 11,000 new U .S. firm s established in 1976 over the subsequent ten-year period, Audretsch and M ahmood (1992) are abale to shed considerable light on the dynamic role that small firm s play in the process o f firm selection and industry evolution. They find that the post-entry growth o f surviving firms, at least in the first few years, tends to be spurred by the extent to which there is a gap between the MES level o f output and the size o f the firms. At the same time, the likelihood o f any new firm surviving tends to decrease as this gap increases. Firm s entering industries characterized by a high degree o f innovative activity face a greater prospect o f growth, but they are burdened with a lower likelihood o f survival. In general, the impact o f firm size, scale economies, and innovative activity tends to be asymmetrical with respect to new -firm growth and survival. Factors which promote growth tend to reduce the probability o f survival, and vice versa.
Audretsch and M ahmood (1992) find that firms are more likely to be operating at a suboptimal scale o f output if the underlying technological conditions are such that there is a greater chance o f making an innovation and subsequently growing. If firms successfully learn and adapt, or are ju st plain lucky, they grow into viably-sized
Gort and Klepper (1982) found that innovation plays an important role as the mechanism by which new firms can enter an industry.
enterprises. If not, they stagnate and may ultimately be forced to exit from the industry. This suggests, just as some recent empirical studies have found, that entry and the start-up o f new firms is not greatly deterred in the presence o f scale economies. As long as entrepreneurs perceive that there is some prospect for growth and ultimately survival, such entry will occur. Thus, in industries where the MES is high, the growth rate o f the surviving firms tends to be high. At the same time, new firms not able to grow and attain the MES level o f output are forced to exit from the industry, resulting in a relatively low likelihood o f survival. By contrast, in industries characterized by a low MES, neither the need for growth, nor the consequences o f its absence are as severe, so relatively lower growth rates but higher survival rates tend to be observed. Similarly, in industries where the probability o f innovating is greater, more entrepreneurs take a chance that they will succeed by growing into a viably sized enterprise. In such industries, the growth o f successful enterprises tends to be greater, but the likelihood o f survival is lower.
While Lucas (1978) attempted to explain the pervasiveness o f small and suboptimal scale enterprises in the firm-size distribution with a static theory, viewed through a dynamic lens the persistent asymmetric size distribution o f firms becomes more understandable. According to this view, the answer to the question, "How are such small and suboptimal enterprises able to be viable," is, "They are not — at least not by remaining small and suboptimal." Rather, such new small-scale firms are engaged in the selection process, whereby the successful enterprises grow and ultimately approach or attain the MES level o f output, while the remainder stagnate and tend to be forced to exit from the industry. Thus, the persistence o f an asymmetric firm-size distribution skewed towards small enterprises and the large presence o f suboptimal scale firms presumably reflects a continuing process o f entry into industries and not necessarily the survival of those small and suboptimal enterprises unable to grow over a long period o f time. That is, although the skewed size distribution o f firms persists with remarkable stability over time, it does not appear to be a constant set o f small and suboptimal scale firms responsible for this skewed firm-size distribution.
The industrial organization o f virtually every manufacturing industry can be therefore viewed as a conical revolving door, where the top part — which represents the largest enterprises in the industry -- revolves much more slowly than the low er part -- which represents the small enterprises in the industry. Further, the extent to which there are barriers to survival, that is the gap between the typical start-up size and the MES level of output, will determine how fast the lower part o f the door revolves. Industries with a high MES and with considerable innovative activity tend to have a faster revolving door than those with a low MES and with only sluggish technological change. This explains why, given the persistence o f an asymm etric firm-size distribution, entry has to be virtually higher in high-MES and innovative industries -- to replace those firms that entered but were unable to adequately grow and subsequently failed.
V. Conclusions
The traditional view in industrial organization is that small firms and enterprises operating at a sub-optimal scale represent a net loss in economic efficiency. For example, Leonard Weiss (1979, p. 1137) advocated any public policy which
"...creates social gains in the form o f less sub-optimal capacity." W hile this may be true in a static sense, it clearly does not hold when the role of'sm all firms is viewed through a dynamic lens. That is, while some and even most o f the sub-optimal firms o f today will be gone tomorrow, others will have evolved into the incumbent firms o f the future. And the mechanism by which they succeed and evolve is, at least in some cases, through the introduction o f new products and processes - innovative activity.
Thus, what appears to be an inefficient organization o f industry in a static model may prove to be optimal in a dynamic context.
References
Acs, Zoltan J. and David B. Audretsch, 1991, "R&D, Firm Size and Innovative Activity," in Zoltan J. Acs and David B. Audretsch (eds.), Innovation and Technological Change: An International Comparison. Ann Arbor:
University o f Michigan Press, 39-59.
Acs, Zoltan J. and David B. Audretsch, 1990, Innovation and Small Firm s.
Cambridge, MA: M IT Press.
Acs, Zoltan J. and David B. Audretsch, 1989a, "Small-Firm Entry in U .S.
M anufacturing," Economica. 56(2), May, 255-265.
Acs, Zoltan J. and David B. Audretsch, 1989b, "Births and Firm Size," Southern Economic Journal. 56(2), October, 467-475.
Acs, Zoltan J. and David B. Audretsch, 1988a, "R&D and Small Firm s," Testimony before the Subcommittee on Monopolies and Commercial Law,
Committee on the Judiciary, U.S. House o f Representatives, February 24.
Acs, Zoltan J. and David B. Audretsch, 1988b, "Innovation in Large and Small Firm s: An Empirical Analysis," American Economic Review . 78(4), September, 678-690.
Acs, Zoltan J. and David B. Audretsch, 1987, "Innovation, M arket Structure and Firm Size," Review of Economics and Statistics. 69(4), November, 567- 575.
Acs, Zoltan J ., David B. Audretsch and Maryann P. Feldman, 1993, "R&D
Spillovers and Recipient Firm Size," Review o f Economics and Statistics.
99.
Acs, Zoltan J ., David B. Audretsch, and M aryann P. Feldman, 1992, "Real Effects o f Academic Research: Comment," American Economic Review. 82(1), M arch, 363-367.
Armington, Catherine and M aijorie Odle, 1982, "Small Business - How Many Jobs?”
The Brookings Review. 1, W inter, 14-17.
Audretsch, David B., 1993, "Start-Up Size and Establishment Exit," Review of Economics and Statistics.
Audretsch, David B., 1992, "The Technological Regime and M arket Evolution: The New Learning," Economics o f Innovation and New Technology. 2.
Audretsch, David B., 1991, "New-Firm Survival and the Technological Regime,"
Review o f Economics and Statistics. 60(3), August, 441-450.
Audretsch, David B., 1989, The M arket and the State: Government Policies Towards Business in Europe, Japan and the U .S .. New York: New York University Press.
Audretsch, David B. and Zoltan J. Acs, 1991, "Innovation and Size at the Firm Level," Southern Economic Journal. 57(3), January, 739-744.
Audretsch, David B. and Talat Mahmood, 1991, "The Rate of Hazard Confronting New Firms and Plants in U.S. M anufacturing," discussion paper FS-IV 91-7, Wissenschaftszentrum Berlin für Sozialforschung.
Austin, John S. and David I. Rosenbaum, 1990, "The Determinants o f Entry and Exit Rates into U.S. Manufacturing Industries," Review o f Industrial
Organization. 5(2), 211-223.
Baldwin, John R. and Paul K. Gorecki, 1987, "Plant Creation Versus Plant
Acquisition: The Entry Process in Canadian M anufacturing," International Journal of Industrial Organization. 5, M arch, 27-42.
Bannasch, Hans-Gerd, 1993, "The Evolution o f Small Firm s in East Germ any," in Zoltan J. Acs and David B. Audretsch (eds.), Small Firm s and
Entrepreneurship: An East-West Perspective. Cambridge, England:
Cambridge University Press.
Bannasch, Hans-Gerd, 1990, "The Role o f Small Firm s in East G erm any", Small Business Economics. 2(4), 307-312.
Birch, David L ., 1981, "Who Creates Jobs?," The Public Interest 65, Fall, 3-14.
Blair, John M ., 1948, "Technology and Size," American Economic R eview . 38(2), M ay, 121-152.
Blanchflower, D. and B. M eyer, 1993, "A Longitudinal Analysis o f Young Entrepreneurs in Australia and the United States," Small Business Economics. 5(4).
Blanchflower, D. and A. Oswald, 1990, "What Makes an Entrepreneur?," NBER W orking Paper 3252, September.
Brock, William A. and David S. Evans, 1989, "Small Business Econom ics," Small Business Economics. 1(1), 7-20.
Brown, Charles, Judith Connor, Steven Heeringa, and John Jackson, 1990, "Studying (Small) Business with the Michigan Employment Security Commission Longitudinal Data B ase,” Small Business Economics. 2(4), 261-278.
Brown, Charles, James Hamilton, and James M edoff, 1990, Em ployers Large and Small. Cambridge, MA: Harvard University Press.
Brown, Charles, and James M edoff, 1989, "The Em ployer Size W age Effect,"
Journal o f Political Econom y. 97(4), October, 1027-1059.
Cohen, Wesley M. and Steven Klepper, 1992, "The Tradeoff between Firm Size and Diversity in the Pursuit o f Technological Progress," Small Business Economics. 4(1), M arch, 1-14.
Cohen, Wesley M. and Steven Klepper, 1991, "Firm Size Versus Diversity in the Achievement of Technological Advance," in Z. J. Acs and D. B.
Audretsch (eds), Innovation and Technological Change: An International Comparison. Ann Arbor, University o f Michigan Press, 183-203.
Cohen, Wesley M. and Richard C. Levin, 1989, "Empirical Studies o f Innovation and Market Structure," in Richard Schmalensee and Robert Willig (eds), Handbook of Industrial Organization. Volume II, Amsterdam: North- Holland, 1059-1107.
Cohen, Wesley M ., Richard C. Levin, and David C. M owery, 1987, "Firm Size and R&D Intensity: A Re-examination," Journal o f Industrial Economics. 35, June, 543-565.
De Wit, Gerrit and Frans A .A .M . van Winden, 1991, "An M-Sector,
N-Group Behavioral Model o f Self-Employment," Small Business Econom ics. 3(1) M arch, 49-66.
De Wit, Gerrit and Frans A .A .M . Van Winden, 1989, "An Empirical Analysis o f Self-Employment in the Netherlands," Small Business Economics. 1(4), 263-272.
Dosi, Giovanni, 1988, "Sources, Procedures, and Microeconomic Effects o f Innovation," Journal o f Economic Literature. 26(3), September, 1120-
1171.
Dunne, Timothy, M ark J. Roberts, and Larry Samuelson, 1989, "The Growth and Failure o f U.S. Manufacturing Plants," Quarterly Journal o f Economics.
104(4), November, 671-698.
Dunne, Timothy, M ark J. Roberts, and Larry Samuelson, 1988, "Patterns o f Firm Entry and Exit in U .S. Manufacturing Industries," Rand Journal o f Econom ics. 19(4), W inter, 495-515.
Edwards, Keith L. und Theodore J. Gordon, 1984, "Characterization o f Innovations Introduced on the U .S. M arket in 1982," The Futures Group, prepared for the U .S. Small Business Administration under Contract No. SBA-6050- OA-82, March.
Evans, David S ., 1987a, "Tests o f Alternative Theories o f Firm G rowth," Journal o f Political Economy. 95(4), August, 657-674.
Evans, David S., 1987b, "The Relationship Between Firm Growth, Size and Age:
Estimates for 100 Manufacturing Industries," Journal o f Industrial Econom ics. 35(2), June, 567-581.
Evans, David S. and Linda S. Leighton, 1990a, "Small Business Form ation by
Unemploymed and Employed W orkers," Small Business Economics. 2(4), 319-330.
Evans, David S. and Linda S. Leighton, 1990b, "Some Empirical Aspects o f
Entrepreneurship," in Zoltan J. Acs and David B. Audretsch (eds.), The Economics o f Small Firms: A European Challenge. Boston, Kluwer Academic Publishers, 79-97.
Evans, David S. and Linda S. Leighton, 1989, "The Determinants o f Changes in U .S. Self-Employment, 1968-1987," Small Business Economics. 1(2),
111-120.
Evans, David and Boyan Jovanovic, 1989, "Estimates o f a Model o f Entrepreneurial Choice under Liquidity Constraints," Journal o f Political Econom y. 95, August, 657-674.
Evans, Laurie Beth, and John J. Siegfried, 1992, "Entry and Exit in United States Manufacturing Industries from 1977 to 1982," in David B. Audretsch and John J. Siegfried (eds.), Empirical Studies in Industrial Organization:
Essays in Honor o f Leonard W. W eiss. Boston: Kluwer Academic Publishers.
Fitzroy, Felix R ., 1989, "Firm Size, Efficiency and Employment: A Review A rticle,"
Small Business Econom ics. 1(1), 75-80.
Galbraith, John K ., 1956, American Capitalism: The Concept o f Coutervailing Pow er, revised edition, Boston, MA: Houghton M ifflin.
Geroski, Paul and Joachim Schwalbach (eds.), 1991, Entry and M arket
Contestability: An International Comparison. Oxford: Basil Blackwell.
Gort, Michael and Steven Klepper, 1982, "Time Paths in the Diffusion o f Product Innovations," Economic Journal. 92(3) September, 630-653.
Hall, Bronwyn H ., 1987, "The Relationship Between Firm Size and Firm Growth in the U .S. M anufacturing Sector," Journal o f Industrial Econom ics. 35, June, 583-605.
Hughes, Alan, 1993, "Industrial Concentration and the Small Business Sector in the U .K .: The 1980's in Historical Perspective," in Zoltan J.A cs and David B. Audretsch (eds.), Small Firms and Entrepreneurship: An East-W est Perspective. Cambridge, England: Cambridge University Press.
Invem izzi, B. and R. Revelli, 1993, "Small Firms in the Italian Economy: Structural Changes and Evidence o f Turbulence," in Zoltan J. Acs and David B.
Audretsch (eds.), Small Firm s and Entrepreneurship: An East-W est Perspective. Cambridge, England: Cambridge University Press.
Jaffe, Adam B ., 1989, "Real Effects o f Academic Research," American Economic Review . 79(5), December, 957-970.
Jaffe, Adam B ., 1986, "Technological Opportunity and Spillovers o f R&D: Evidence from Firm s' Patents, Profits and M arket V alue," American Economic Review. 76, December, 984-1001.
Jovanovic, Boyan, 1993, "Entrepreneurial Choice W hen People D iffer in Their Management and Labor Skills," Small Business Econom ics. 5(4).
Jovanovic, Boyan, 1982, "Selection and Evolution o f Industry," Econom etrica. 50(2) M ay, 649-670.
Kaldor, Nickolas, 1934, "The Equilibrium o f the Firm ," Economic Journal. 44 M arch, 60-76.
Kenney, M ., 1986, Biotechnology: The University-Industry Com plex, New Haven:
Yale University Press.
Kleinknecht, Alfred, 1991, "Firm Size and Innovation: Reply to Scheirer," Small Business Economics. 3(2), 157-158.
Kleinknecht, Alfred, 1989, "Firm Size and Innovation: Observations in Dutch Manufacturing Industry," Small Business Economics. 1(1), 215-222.
Kleinknecht, Alfred, 1987, "Measuring R&D in Small Firms: How Much Are W e Missing?," Journal o f Industrial Economics. 36(2), December, 253-256.
Kleinknecht, Alfred, Tom P. Poot and Jeroen O .N . Reijnen, 1991, "Technical Performance and Firm Size: Survey Results from the Netherlands," in Zoltan J. Acs and David B. Audretsch (eds), Innovation and
Technological Change: An International Com parison. Ann Arbor:
University of Michigan Press, 84-108.
Levin, Richard C ., Wesley Cohen, and David C. M owery, 1985, "R&D
Appropriability Opportunity and M arket Structure: New Evidence on the Schumpeterian Hypothesis," American Economic Review. 15, M ay, 20- 24.
Link, Albert N. and Barry Bozeman, 1991, "Innovative Behavior in Small-Sized Firm s," Small Business Economics. 3(3), September, 179-184.
Loveman, Gary and W erner Sengenberger, 1991, "The Re-emergence o f Small-Scale Production: An International Perspective", Small Business Econom ics.
3(1), March, 1-38.
Lucas, Robert E., Jr., 1978, "On the Size Distribution o f Business Firm s," Bell Journal of Economics. 9, Autumn, 508-523.
Mansfield, Edwin, 1962, "Entry, G ibrat's Law, Innovation, and the Growth o f Firm s," American Economic Review. 52(5) December, 1023-1051.
Marx, Karl, 1912, Capital, translated by Ernest Untermann, vol. 1, Chicago, IL:
Kerr.
M oore, John H ., 1992, "Measuring Soviet Economic Growth: Old Problem s and New Complications," Journal o f Institutional and Theoretical Econom ics.
148(1), March, 72-92.
M ueller, Dennis C. and J. Tilton, 1969, "Research and Development Costs as a Barrier to Entry," Canadian Journal o f Economics. 56(4), November, 570-579.
Noren, J .H ., 1966, "Soviet Industry Trands in Output, Inputs, and Productivity," in U .S. Congress, Joint Economic Committee, New Directions in the Soviet U nion. Part II-A, W ashington, D .C .: Government Printing Office, 271-
Orr, Dale, 1974, "The Determinants o f Entry: A Study o f the Canadian
M anufacturing Industries," Review o f Economics and Statistics. 56(1), February, 58-66.
Pakes, A. and R. Ericson, 1987, "Empirical Implications o f A lternative Models o f Firm Dynamics," manuscript, Dept. of Economics, University o f Wisconsin, Madison.
Phillips, Bruce D. and Bruce A. Kirchhoff, 1989, "Form ation, Growth and Survival:
Small Firm Dynamics in the U.S. Economy," Small Business Econom ics.
1(1), 65-74.
Piore, Michael J. and Charles F. Sabel, 1983, "Italian Small Business Development:
Lessons for U.S. Industrial Policy," in John Zysman and Laura Tyson (eds.), American Industry in International Com petition. Ithaca, NY:
Cornell Unversity Press, 391-421.
Pratten, C .F ., 1971, Economies o f Scale in Manufacturing Industry. Cambridge, England: Cambridge University Press.
Rosenberg, Nathan, 1992, "Economic Experim ents." Industrial and Corporate Change. 1(1), 181-204.
Rothwell, Roy, 1989, "Small Firm s, Innovation and Industrial Change," Small Business Economics. 1(1), January, 51-64.
Sato, Yoshio, 1989, "Small Business in Japan: A Historical Perspective," Small Business Economics. 1(2), 121-128.
Scherer, F .M ., 1991, "Changing Perspectives on the Firm Size Problem ," in Zoltan J. Acs and David B. Audretsch (eds.), Innovation and Technological Change: An International Com parison. Ann Arbor: University of Michigan Press, 24-38.
Scherer, F .M ., 1988, "Testimony before the Subcommittee on M onopolies and Commercial Law", Committee on the Judiciary, U .S. House of Representatives, February 24.
Scherer, F .M ., 1973, "The Determinants o f Industry Plant Sizes in Six Nations,"
Review o f Economics and Statistics. 55(2) May, 135-145.
Schumpeter, Joseph A ., 1911, Theorie der wirtschaftlichen Entwicklung: Eine Untersuchung über Untemehmensgewinn. Kapital. K redit und den Koniunkturzyklus. (5th edition, 1952), Berlin: D uncker & Humblot.
Schumpeter, Joseph A ., 1950, Capitalism. Socialism and D em ocracy. 3rd edition, New York, NY: H arper and Row.
Schwalbach, Joachim, 1989, "Small Business in German M anufacturing," Small Business Economics. 1(2), 129-136.
Simon, H erbert A. and Charles P. Bonini, 1958, "The Size Distribution o f Business Firm s," American Economic Review. 48(4) September, 607-617.
Storey, David J ., 1990, "Firm Perform ance and Size: Explanations from the Small Firm Sectors," in Zoltan J. Acs and David B. Audretsch, (eds.), The Economics o f Small Firm s: A European Challenge. Boston: Kluwer Academic Publishers, 43-50.
