Jump to ContentJump to Main Navigation
Enterprising AmericaBusinesses, Banks, and Credit Markets in Historical Perspective$

William J. Collins and Robert A. Margo

Print publication date: 2015

Print ISBN-13: 9780226261621

Published to Chicago Scholarship Online: May 2016

DOI: 10.7208/chicago/9780226261768.001.0001

Show Summary Details
Page of

PRINTED FROM CHICAGO SCHOLARSHIP ONLINE (www.chicago.universitypressscholarship.com). (c) Copyright University of Chicago Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in CHSO for personal use (for details see http://www.chicago.universitypressscholarship.com/page/privacy-policy). Subscriber: null; date: 17 July 2018

Corporate Governance and the Development of Manufacturing Enterprisesin Nineteenth-Century Massachusetts

Corporate Governance and the Development of Manufacturing Enterprisesin Nineteenth-Century Massachusetts

Chapter:
(p.73) 2 Corporate Governance and the Development of Manufacturing Enterprisesin Nineteenth-Century Massachusetts
Source:
Enterprising America
Author(s):

Eric Hilt

Publisher:
University of Chicago Press
DOI:10.7208/chicago/9780226261768.003.0003

Abstract and Keywords

This chapter analyzes the use of the corporate form among nineteenth century manufacturing firms in Massachusetts, from newly collected data from 1875. An analysis of incorporation rates across industries reveals that corporations were formed at higher rates among industries in which firm size was larger. But conditional on firm size, the industries in which production was conducted in factories, rather than artisanal shops, saw more frequent use of the corporate form. On average, the ownership of the corporations was quite concentrated, with the directors holding 45 percent of the shares. However, the corporations whose shares were quoted on the Boston Stock Exchange were ‘widely held’ at rates comparable to modern American public companies. The production methods utilized in in different industries also influenced firms’ ownership structures. In many early factories, steam power was combined with unskilled labor, and managers likely performed a complex supervisory role that was critical to the success of the firm. Consistent with the notion that monitoring management was especially important among such firms, corporations in industries that made greater use of steam power and unskilled labor had more concentrated ownership, higher levels of managerial ownership, and smaller boards of directors.

Keywords:   corporate governance, manufacturing, corporate ownership, investor protections, corporation law

2.1 Introduction

The ownership structure of public companies in the United States is unique. Relative to those of other countries, American corporations are less likely to have a parent company or be part of a family business group, and more likely to be widely held (LaPorta et al. 1999). The historical origins of the distinctive patterns of corporate ownership in the United States are the subject of some debate, and a number of competing theories have been offered, ranging from populist politics (Roe 2004) to the protections of investors resulting from the early origins of the American legal system (LaPorta et al. 1998). But the historical evolution of corporate ownership in the United States is poorly documented, and the timing of the emergence of the distinctively American style of corporate ownership, along with the factors that may have been responsible for its emergence, are not well understood.

Most accounts of the history of American corporate ownership tend to echo the influential work of Berle and Means (1932), who argue that prior to the rise of “big business” around the turn of the twentieth century, American industrial corporations were owned by limited numbers of investors who participated actively in the governance of their (p.74) firms.1 More recent work has challenged this view, arguing from early nineteenth-century data that the separation of ownership from control occurred much earlier (Hilt 2008). The contradictory and fragmentary findings of this literature reflect the frustrating scarcity of data on early American corporations, particularly industrial firms, which were subject to relatively few disclosure requirements in most states’ corporation laws.

This chapter presents new, comprehensive data on the ownership and governance of industrial corporations from the third quarter of the nineteenth century—before the rise of “big business”—from the state of Massachusetts. At the time, Massachusetts was among the most heavily industrialized states. Entrepreneurs in Massachusetts were pioneers in the textile industry, and developed the first “integrated” cotton textile mills in the United States in the early nineteenth century (see Ware 1931). Dozens of enormous textile corporations were eventually founded in the state, and the shares of many of these enterprises were traded on the Boston Stock Exchange, which was then the premier market for American industrial securities (Atack and Rousseau 1999). In the second half of the nineteenth century, a number of new industries developed in Massachusetts, including chemicals, fabricated metals, and machinery, and entrepreneurs in these industries also made heavy use of the corporate form.

Relative to those of nearly all other states, Massachusetts’s corporation law included an unusually strict disclosure requirement that mandated that several different classes of firms submit “certificates of condition” to the state government, which listed the names of their directors and stockholders.2 In this chapter, I use the certificates filed for the year 1875 to construct a comprehensive data set of operating manufacturing corporations in the state, which includes detailed ownership information. I then classify each corporation by industry and match them to data from the state’s 1875 manufacturing census, which recorded information on production methods, total numbers of firms and employees, and wages in different industries. These data present a detailed and comprehensive picture of corporate ownership among manufacturing firms in the third quarter of the nineteenth century. With the matched data I am able to analyze the variation in incorporation rates as well as the differences in corporate ownership structures across industries. In particular, I can document the extent of the separation of ownership from control among the prominent textile corporations in the state, and compare them to corporations operating in other industries.

(p.75) The analysis proceeds in three steps. First, I study the adoption of the corporate form across industries. The data reveal substantial variation in incorporation rates, ranging from a high of more than 60 percent among producers of chemicals and of glass, to zero among producers of carriages and wagons. Perhaps unsurprisingly, an analysis of the determinants of incorporation rates reveals that they were higher among industries where average firm size, measured by either capital or total employees, was large. However, conditional on firm size, industries that made greater use of steam power and unskilled labor, and industries in which a relatively large proportion of firms’ capital was accounted for by fixed assets, were incorporated at higher rates. This is consistent with the notion that the corporate form found heaviest use among those industries in which production was undertaken within factories, rather than artisanal shops.3 The data suggest that the corporation was used not only to achieve greater scale, but also to adopt mechanized, factory-based production methods.

In the second step I analyze the ownership of the corporations, and calculate a variety of statistics comparable to those commonly used in the analysis of modern corporate governance. The data indicate that the degree of ownership dispersion in general, and managerial ownership in particular, varied widely across industries. The great textile corporations, whose shares were traded on the Boston Stock Exchange, were “widely held” at even higher rates than those of modern American publicly traded corporations.4 The degree of the separation of ownership from control among those firms was comparable to large modern publicly traded firms, and Berle and Means (1932) would have characterized nearly all of them as subject to “management control.” The ownership structures of those firms certainly contradict any simplistic narrative of the evolution of American corporate ownership in which the separation of ownership from control suddenly occurs at the turn of the twentieth century.

However, the data also indicate that the great textile mills were rather atypical of the state’s industrial corporations: they were larger, had greater numbers of shareholders, and lower degrees of managerial ownership. Entrepreneurs were able to use the flexibility of Massachusetts’s corporation law to configure their enterprises in a variety of ways, according to their needs and circumstances. Most corporations had relatively few shareholders and high levels of managerial ownership, but the degree of ownership concentration observed across industries, and also among firms within particular industries, varied significantly.

(p.76) Finally I analyze the determinants of ownership concentration and managerial ownership. Unsurprisingly, larger firms were more widely held. But the results indicate that, conditional on firm size, corporations that made greater use of steam power and unskilled labor had more concentrated ownership. That is, relative to corporations of similar sizes in industries that relied more on skilled labor and less on steam power, those in industries that organized production within factories, rather than large artisanal shops, had fewer shareholders, and a greater proportion of the shares were held by the directors. Investors responded to the complex role performed by the managers of those firms by ensuring that there were adequate ownership incentives to monitor and supervise management.

This chapter contributes to three interrelated lines of research. First, it complements the large and growing literature on the development of manufacturing in New England.5 Much of this work has focused narrowly on textiles and the very large corporations that were formed within that industry (for example, McGouldrick 1968). This chapter complements that work by presenting an account of how the corporate form was utilized within manufacturing. The data reveal that in the third quarter of the nineteenth century, the corporation had proliferated well beyond textiles into a broad range of industries, and that across those industries somewhat different “styles” of ownership were adopted.

Second, the chapter contributes to a literature that documents the history of corporate ownership, both in the United States and elsewhere.6 This chapter complements that literature by presenting comprehensive statistics on the ownership of nineteenth-century industrial corporations, prior to the rise of “big business.”

Finally, the chapter also contributes to the literature on the adaptability of the corporate form to the needs of small- and medium-sized enterprises (SMEs). The early corporation laws of many American states were quite rigid, and regulated firms’ internal governance institutions in ways that may have been unattractive to SMEs.7 Some contributions to this literature have argued that innovations in the menu of organizational forms available to American firms in the later twentieth century, such as what have been termed the private limited liability company (PLLC), created alternatives that were superior to the corporation for the needs of SMEs (Guinnane et al. 2007). The results of this chapter show that Massachusetts was somewhat of an (p.77) exception. The laws of Massachusetts were unusually flexible with regard to the internal governance of corporations, and entrepreneurs were able to successfully adapt the form into a wide range of enterprises, including small closely held firms.

2.2 The Massachusetts Legislature and Corporation Law, 1790–1850

As in most American states, during the first half of the nineteenth century the corporate form was not freely available to entrepreneurs in Massachusetts. Instead, incorporation was only possible if the state legislature passed a law granting a charter to a business. These “special act charters” were probably not accessible to entrepreneurs who lacked a fair measure of legal sophistication and financial resources. Nonetheless, over the first half of the nineteenth century, Massachusetts granted charters to nearly 550 manufacturing firms. The terms of these charters were initially restrictive in some respects, but they quickly became quite liberal, particularly with regard to the internal governance of the firms they created. This flexibility was unusual, relative to the terms of other states’ corporation laws, and may have contributed to the heavy use of the corporate form in Massachusetts.

Beginning in the early national period, the state government actively used the law to promote economic development, offering public support to private enterprises that would furnish transportation infrastructure or develop the capacity for manufacturing (see Handlin and Handlin 1974). When entrepreneurs sought charters to incorporate manufacturing businesses, they were generally accommodated. As the state industrialized and new companies proliferated, demand for corporate charters grew rapidly, and the state showed a clear willingness to meet that demand. In the first half of the nineteenth century, Massachusetts granted the highest number of corporate charters of all the American states and territories. Figure 2.1 presents the charters granted by the state in a comparative perspective. By 1850 Massachusetts had granted more than twice the number of corporate charters relative to its population than the national average.

Especially in the period before 1830, however, these charters often did not contain all the terms sought by entrepreneurs. For example, the petition for the Boston Manufacturing Corporation, the firm that would become the first to create an integrated cotton mill, sought banking powers for their enterprise, which were refused (McGouldrick 1968). The great success of that firm and the other Waltham-Lowell mills that followed demonstrated quite clearly that banking powers were unnecessary. Yet those firms’ charters lacked another important power that was routinely granted to manufacturing corporations in other states: limited liability for the shareholders. The state refused to grant limited liability to any manufacturing corporation in the 1810s and 1820s. All charters granted to such enterprises explicitly made shareholders subject to an 1809 statute, which made them personally (p.78)

Corporate Governance and the Development of Manufacturing Enterprisesin Nineteenth-Century Massachusetts

Fig. 2.1 Cumulative corporate charters per 1,000 persons

Sources: Corporate charters from Sylla and Wright (2013); population figures from the decennial federal census.

Note: The data do not include corporations created through general acts.

liable for their firms’ debts.8 When it was later objected that shareholders could circumvent this provision by selling their shares to “men of straw,” the legislature strengthened its requirement of individual liability by passing legislation that made shareholders liable for any debts incurred while they were shareholders, even if they subsequently sold their shares.9 Yet in spite of this restriction, manufacturing enterprises sought to incorporate in Massachusetts at very high rates; from 1800 to 1809, fifteen charters were granted to manufacturing enterprises, and from 1810 to 1819, 133 were granted. In the 1820s, another 146 were granted.10

The burden of unlimited liability for shareholders ultimately came to be perceived by many to be limiting economic development. In 1829, a year of high numbers of business failures, the personal liability of many households owning corporate stock led to “wide spreading and irretrievable ruin (p.79) to individuals.”11 Ultimately the governor, Levi Lincoln, took up the cause and despite the vigorous opposition of some influential merchants, a new law granting limited liability to manufacturing enterprises was passed in 1830.12 This law, a “general regulating act” that dictated virtually all of the terms of subsequent manufacturing charters, stated that the shareholders of these enterprises would have unlimited liability for debts until their capital subscriptions were fully paid in, at which point their liability would be limited to the amount of their shares. The act included various other safeguards for the creditors of corporations; for example, limiting total indebtedness and prohibiting the payment of dividends from the capital stock or loans to stockholders.

The 1830 act did not, however, include any terms relative to the governance of the corporations subsequently created, other than imposing the requirement that each corporation have a president, a clerk, a treasurer, and at least three directors. The voting rights of the shareholders, and their method for choosing these officers, were left to the corporations themselves to decide. The silence of the law on these issues was quite unusual; most states’ early corporation laws strictly regulated director elections and shareholder voting rights (Hilt 2014). All subsequent manufacturing charters were quite brief, stating only the name of the firm, the nature of its operations, and the size of its capital stock, and then simply indicating that the firm was subject to the 1830 act. In the twenty years between 1830 and 1850, Massachusetts granted more than 400 charters to manufacturing firms.13

Finally, in 1851 Massachusetts took the important step of passing a general incorporation act for manufacturing enterprises.14 Rather than applying to the legislature for a charter, the act provided that any three or more people could form a corporation, in virtually any manufacturing or mining industry, by simply filing the certificates required in the act with the Secretary of the Commonwealth and with their county.15 Following the precedent of the 1830 general regulating act, the 1851 general incorporation act was mostly silent regarding the internal governance of the corporations, beyond requiring that the firms have a president and a treasurer—the act did not even specify a minimum or maximum number of directors. The act required that corporations created through its terms have a minimum of $5,000 in capital, and also imposed a maximum of $200,000, which was far smaller than the capital of many chartered corporations. The statute therefore served as an alternative route to incorporation for small firms, while large firms were still required to (p.80) seek charters from the legislature. In 1855, the maximum capital permitted for corporations formed under the general act was raised to $500,000, but the legislature continued to retain control over access to the corporate form for the largest enterprises.16

The 1851 general act imposed one significant burden on the corporations created through its terms that chartered corporations were not subject to, and that was an annual report known as a certificate of condition, which stated the names of the officers and shareholders, and provided financial information such as the amount of paid-in capital and the total indebtedness.17 In 1870 the state formally imposed a requirement that a more detailed certificate of condition be submitted annually by all industrial corporations in the state, whether they were chartered or incorporated through the general act. These certificates of condition form the basis for the data analyzed in this chapter.

2.3 The Adoption of the Corporate Form in Massachusetts

The corporate form was utilized with great frequency, but many multiowner firms remained unincorporated, effectively choosing the partnership form. The privileges of incorporation should have been most attractive to firms seeking to raise relatively large amounts of capital from investors. For example, the transferability of shares, the governance structure of a board of directors to whom control over day-to-day management would be delegated, and the limitation of personal liability for shareholders would all seem to be well suited to the needs of passive, outside investors. For a firm with a small number of owners, who were perhaps from the same family or had been in business together previously, the formalities of an annual meeting and director elections and the requirement of detailed annual disclosures probably represented a substantial nuisance. On the other hand, the corporation laws of Massachusetts were relatively flexible, and effectively permitted incorporators to configure their enterprises’ voting rights and decision-making procedures as they wished. Did small firms, or firms with small numbers of owners, actually incorporate?

One way to address these questions is to examine the industries in which incorporation rates were relatively high, and compare them to industries in which incorporation was uncommon. Massachusetts’s manufacturing censuses reported detailed information on the total numbers of establishments, their capital, and their employees, by industry. These records can be compared to the filings of manufacturing corporations, whose certificates of condition stated their capital and other information. The certificates unfortunately do not specify the industry of the corporation or its products or (p.81) revenue. However, the corporate names (e.g., “Bay State Faucet and Valve Company”) often provide a relatively clear indication of the firm’s industry. For those with names that do not provide identifying industrial information (e.g., “Paul Whitin Manufacturing Company”) contemporary business directories were used to classify most corporations into the categories of the state census.18 The earliest year for which totally comprehensive corporation records are available, and a manufacturing census is available, is 1875.19 In that year, the manufacturing census listed more than 10,000 manufacturing establishments in Massachusetts, and the certificates of condition of 601 corporations could be classified into the industrial categories of the census.20

The resulting data are presented in table 2.1.21 The data in the table show quite clearly that incorporation rates differed significantly across industries. Several of the state’s largest industries (measured by the number of establishments), such as boots and shoes, clothing, food preparations, and printing and publishing, had very few incorporated firms at all, and vanishingly low incorporation rates. At the other end of the spectrum, there were smaller industries with relatively small numbers of establishments, such as chemicals, glass, jute baggings, and textile printing (“print works”), where the corporate form was quite dominant. The various categories within the textile industry, as expected, had large numbers of corporations and relatively high incorporation rates. But there were also relatively large numbers of incorporated firms producing machinery, metallic goods, paper, and brick and stone.

The data in the table also seem consistent with the notion that incorporation rates were higher in industries with higher average capital per firm. The industries with the smallest capital, such as tobacco, lumber, vessels, and carriages and wagons, all had incorporation rates of 2 percent or less, (p.82)

Table 2.1 Establishment size and incorporation rates: Industry averages, Massachusetts (1875)

All establishments

Corporations

Incorporation rate

N

Average capital

N

Average capital

(1)

(2)

(3)

(4)

(5)

Clothing

   Boots and shoes

1,461

12,795

12

125,707

0.01

   Other clothing

1,088

8,442

23

202,174

0.02

Food and tobacco

   Food preparations

783

12,580

16

175,875

0.02

   Liquors and beverages

155

26,802

1

150,000

0.01

   Tobacco

264

3,076

4

14,088

0.02

Instruments

   Clocks and watches

14

132,425

3

588,533

0.21

   Scientifc instruments and appliances

52

8,244

7

107,382

0.13

   Musical instruments and materials

71

54,163

8

122,363

0.11

Metals, metallic goods, and machinery

   Agricultural implements

38

30,118

6

190,833

0.16

   Arms and ammunition

20

48,215

1

9,398

0.05

   Artisans’ tools

124

17,956

12

118,133

0.10

   Machines and machinery

311

44,565

69

157,666

0.22

   Other metals and metallic goods

768

28,526

87

171,375

0.11

Oils and chemicals

   Chemical preparations

9

34,644

6

106,935

0.67

   Fertilizers

9

136,722

2

218,000

0.22

   Oils and illuminating fluids

33

69,311

7

112,929

0.21

   Paints and colors

20

55,790

2

35,500

0.10

Paper and paper goods

   Paper

120

90,502

38

119,314

0.32

   Printing and publishing

533

12,033

11

69,755

0.02

Textiles

   Carpetings

24

160,665

6

520,567

0.25

   Cotton goods

220

290,203

107

449,478

0.49

   Linen

5

184,800

2

550,000

0.40

   Print works

9

285,556

5

185,200

0.56

   Silk

6

81,333

1

120,000

0.17

   Woolen goods

183

94,044

32

198,005

0.17

   Other textiles

28

169,700

15

140,173

0.54

Vessels and carriages

   Carriages and wagons

356

6,777

1

84,000

0.00

   Vessels

163

5,733

1

350,000

0.01

Wooden goods

   Furniture

294

16,836

6

237,807

0.02

   Lumber

579

4,697

7

35,971

0.01

   Other wooden goods

460

9,728

10

67,975

0.02

Other industries

   Bricks

104

15,939

9

186,222

0.09

   Glass

13

119,615

8

247,963

0.62

   Jute

6

72,833

3

119,000

0.50

   Leather

495

16,969

10

164,110

0.02

   Rubber

23

151,509

4

115,000

0.17

   Stone

151

11,020

21

90,468

0.14

Miscellaneous

   Miscellaneous manufactures

1,250

21,396

37

113,103

0.03

(p.83) whereas those with the highest firm capital, such as cotton goods, textile printing, linen, and “other textiles,” all had incorporation rates of more than 40 percent. The data in table 2.1 also indicate that in all but one industry the average capital of corporations was larger than the average capital of all establishments, sometimes by as much as a factor of ten. In at least a few cases it seems very likely that the corporations were effectively operating in a slightly different industry, even though according to the rough classification system of the census they were grouped into the same category.

How consistent is the relationship between average firm size and incorporation rates at the industry level? An analysis of this relationship is presented in figure 2.2. The scatterplot in the figure indicates that industries with higher levels of capital per firm (in logs) indeed had higher incorporation rates. The regression line included in the figure illustrates the strong tendency toward higher incorporation rates among firms in industries with higher than average capital. However, the residuals of many industries are also high, and in particular, there are several industries with relatively high incorporation rates and relatively low levels of average capital. The lower panel illustrates the same pattern, using the log of total employees, rather than capital, as a measure of firm size.

The census recorded detailed data on certain elements of manufacturing firms’ capital, workforce, and operations. These data, which are summarized in table 2.2, provide some insight into the production methods utilized by firms in different industries. For example, some industries appear to have been dominated by artisanal shops, with relatively small numbers of workers and little fixed capital. Producers of tobacco products, scientific instruments, and food preparations on average had fewer than six employees, and less than $10,000 in capital. Firms in each of these industries also made relatively (p.84)

Corporate Governance and the Development of Manufacturing Enterprisesin Nineteenth-Century Massachusetts

Fig. 2.2 Incorporation rates and firm characteristics by industry

(p.85)

Table 2.2 Industry characteristics

Mean

SD

Min.

Max.

Incorporation rate

0.178

0.193

0

0.670

Capital

   Log capital

10.340

1.270

8.031

12.578

   Capital-output ratio

0.618

0.305

0.171

1.497

   Steam engines per establishment

0.342

0.253

0

0.889

   Fraction assets machines and buildings

0.589

0.133

0.388

0.933

Labor

   Log total employees

3.286

1.208

1.220

5.873

   Fraction workforce female

0.215

0.238

0

0.815

   Log annual wage, all employees

6.273

0.273

5.521

6.764

   Log annual wage, adult male employees

6.381

0.211

5.727

6.782

Growth rate of output, 1865–1875

0.433

0.699

–0.849

2.020

little use of steam power, with around 0.10 steam engines per establishment or less. In contrast, print works, as well as producers of cotton goods or of linens, typically had well over 200 employees and $200,000 or more in capital, and in these industries there were 0.8 steam engines per establishment. Establishments in these industries are best characterized as factories.

Likewise the workforce of some industries consisted of a relatively large proportion of skilled labor, whereas others relied heavily on unskilled labor. The table presents proxies for the share of skilled labor, based on wages and the gender composition of the labor force. A higher wage is consistent with a greater level of skill; the fraction of the labor force that was female is an indicator for the proportion that was unskilled. The table presents wages for all workers, which reflects the mix of skilled and unskilled labor (in the form of women and children) and also the wage for adult male workers only. The industries with the highest levels of wages were musical instruments and clocks and watches, whose adult male workers earned $876 and $882 per year, respectively. The lowest wages of adult males were in jute baggings ($307) and linen ($390), and in these industries female workers represented more than 50 percent of total employees.

We can gain some insight into the purposes for which the corporate form was utilized by analyzing how incorporation rates varied with these industry characteristics. In addition to achieving a greater scale, incorporation facilitated the creation of firms that utilized particular kinds of assets or production processes, and the variation in incorporation rates across industries reflected these patterns. Table 2.3 presents regressions of incorporation rates on industry characteristics. Panel A presents univariate regressions, and since many of these characteristics may be correlated with firm scale, in panel B the regressions are repeated with a measure of scale included—the log of the number of employees per establishment.

(p.86)

Table 2.3 Determinants of industry incorporation rates

(1)

(2)

(3)

(4)

(5)

(6)

(7)

A. Univariate regressions

Capital-output ratio

0.236**

(0.086)

Steam engines per establishment

0.492**

(0.116)

Fraction assets machines and bldgs.

0.596*

(0.235)

Percent workforce female

0.230

(0.137)

Log wage, all employees

−0.322**

(0.086)

Log wage, adult male employees

−0.368**

(0.123)

Growth rate of output, 1865–1875

−0.083+

(0.046)

Constant

0.0364

−0.0065

−0.166

0.137**

2.196**

2.521**

0.195**

(0.0542)

(0.0324)

(0.136)

(0.0433)

(0.540)

(0.792)

(0.0409)

R- squared

0.138

0.475

0.156

0.077

0.202

0.158

0.113

B. Controlling for average firm scale

Capital-output ratio

0.090

(0.069)

Steam engines per establishment

0.368*

(0.146)

Fraction assets machines and bldgs.

0.373+

(0.213)

Percent workforce female

−0.058

(0.120)

Log wage, all employees

−0.140

(0.099)

Log wage, adult male employees

−0.214*

(0.099)

Growth rate of output, 1865–1875

−0.018

(0.028)

Log total employees

0.098**

0.053*

0.099**

0.112**

0.096**

0.100**

0.103**

(0.016)

(0.021)

(0.020)

(0.018)

(0.020)

(0.018)

(0.015)

Constant

−0.199**

−0.132**

−0.362**

−0.174**

0.747

1.223+

−0.170**

(0.055)

(0.042)

(0.088)

(0.048)

(0.648)

(0.639)

(0.045)

R- squared

0.459

0.561

0.504

0.444

0.450

0.469

0.574

Observations

38

35

35

36

33

33

34

Note: Robust standard errors in parentheses.

(**) Significant at the 1 percent level.

(*) Significant at the 5 percent level.

(+) Significant at the 10 percent level.

(p.87)

(p.88) The estimates in panel A indicate that incorporation rates were higher in industries in which production took place primarily within factories, rather than artisanal shops. The capital-output ratio, the use of steam power, and the use of unskilled labor (reflected in both lower average wages and a higher percentage of female labor, although the latter estimate is not statistically significant) were all substantially higher among industries with higher incorporation rates. The incorporation rate also varied with the composition of firms’ capital. Column (3) presents a regression of the percentage of firms’ assets represented by “fixed assets”—buildings and machinery—rather than stock on hand. This measure was also positively correlated with incorporation rates, which likely reflects the fact that the capital of incorporated firms was used to finance fixed investments.22

Finally, column (7) regresses the growth rate of industry output in Massachusetts between the 1865 and 1875 censuses on incorporation rates. If the corporate form facilitated investments in rapidly growing, new, or innovative industries, then one would expect that this measure would be positively correlated with incorporation rates. However, the estimated correlation is negative, and even significant at the 10 percent level. This is likely due to the negative correlation between industry growth rates and average firm size—industries with larger firms grew more slowly.23

In panel B of table 2.3, the same regressions are estimated, but with log total employees included as a measure of scale. That is, in these regressions the relationship between industry characteristics and incorporation rates are analyzed, conditional on average firm size. Some of the estimated relationships change substantially, implying that the correlations in panel A were simply due to the greater scale in industries that had higher incorporation rates. In particular, the estimated correlations with the capital-output ratio (column [1]), the percentage of the labor force that was female (column [4]), the log wage for all employees (column [5]), and the growth rate of industry output following 1865 (column [7]) all show substantial decreases in absolute magnitude and statistical significance. Evidently these relationships were driven by scale. Interestingly, the log average wage paid to adult male employees—which should reflect the degree of skill among those workers, rather than the overall mix of skilled and unskilled labor—continues to (p.89) indicate a strong negative relationship, although the magnitude of the estimate is diminished. Likewise the estimated effects of the use of steam power and the fraction of firm assets represented by fixed capital, both of which were likely correlated with factory production, were robustly correlated with incorporation rates, even conditional on firm scale.

Unfortunately it is impossible to infer from these data whether access to the corporate form facilitated the creation of firms that could not otherwise have existed, or if it enabled firms to achieve a greater scale or adopt different production methods than would have been possible otherwise. It is worth noting that if the corporate form enabled firms to increase their scale or adopt steam power, relative to what was attainable as a partnership, then these results imply that the corporate form increased productivity. Using data from the federal census, Atack, Bateman, and Margo (2008) find strong productivity gains associated with the adoption of steam power, and that these gains were increasing in firm size.

But overall these results indicate that corporations were formed in industries in which establishments resembled factories more than artisanal shops. Even conditional on average firm size, industries with high incorporation rates were more likely to utilize steam power and relied on less skilled male workers, relative to their peers. In what follows, I analyze the ownership of corporations and how their governance may have responded to these industry characteristics.

2.4 Ownership and Governance of Massachusetts’s Corporations

Concentrated corporate ownership imposes costs, and also presents some benefits. The most obvious cost is that the wealth of investors holding large blocks of stock will be illiquid and poorly diversified. In cases where a shareholder holds a controlling stake, an additional cost may arise, namely that the controlling shareholder may engage in self-dealing or other actions that benefit himself at the cost of the other owners, and those other owners may have little recourse. On the other hand, concentrated ownership creates incentives for investors to monitor the management of the firm. In the nineteenth century many corporate directors likely participated directly in management (they were the managers), implying that large stakes held by the board would create strong ownership incentives for diligent effort. One of the main arguments of Berle and Means (1932) is that when there are no large blockholders, control falls into the hands of management, who become unaccountable to the shareholders, except in rare circumstances.

The historical record indicates that among the very large textile corporations in the state, there was an absence of substantial owners who would have had strong incentives to monitor management, and this appears to have been a source of some concern. Sophisticated merchants often expressed skepticism that the managers of early corporations would be capable of performing (p.90) as well as those who operated on “an individual basis.” For example, Henry Lee, a Boston merchant, complained in his correspondence that many major textile corporations were “in danger of being ruined by extreme salaries and high wages in all the departments,” a problem he attributed to weak performance incentives for managers (Porter 1937, 125). But the most clear and direct evidence of such managerial opportunism is found in the early 1860s, when an activist investor named J. C. Ayer initiated a campaign to reform the governance institutions of the major textile corporations. He produced a pamphlet, On the Usages and Abuses in the Management of Our Manufacturing Corporations (1863) that argued that opportunism by directors was rampant: they engaged in self-dealing in their transactions with firms to whom the purchase of raw materials or the sale of finished products were delegated, and paid excessive fees; they hired their relatives for important supervisory positions; they drew excessively high salaries; and they concealed the effects of these practices from the shareholders. Ayer specifically argued that “relations of owners and managers” had changed since the founding of the companies. Unlike the original investors, the existing owners decades later were completely passive, and bought their shares “in the hope that somebody interested in it can and will take care of it.” He also argued that the directors perpetuated their control over their firms by soliciting proxy votes from the shareholders through duplicitous means, and, where necessary, by holding the annual meetings of companies with many shareholders in common simultaneously, thereby preventing the larger shareholders from participating in more than one.

Although it is impossible to verify many of Ayer’s claims, it is possible to discern the level of shareholder participation in annual meetings for at least a handful of companies and thereby assess whether or not the scope for managerial opportunism was as broad as Ayer claimed. Indeed, it does appear to be the case that stockholders participated in annual meetings only infrequently.24 The Massachusetts legislature responded to the complaints of Ayer and other stockholders by enacting a statute in 1865 intended to limit the power of directors to utilize proxy votes to perpetuate their control.25 In particular, the statute limited the number of proxy votes that a sitting director could exercise to twenty, a very small fraction of the total of around 1,000 shares that were typically outstanding.

2.4.1 Ownership Structures of Manufacturing Companies

Whether or not these problems were representative of those faced by investors in early manufacturing corporations generally depends, at least in (p.91)

Table 2.4 Ownership of manufacturing corporations

Mean

Median

SD

Min.

Max.

A. All manufacturing corporations

Total paid-in capital

210,638

100,000

323,753

1,000

2,500,000

Total shareholders

47

18

87

2

730

Board size

4.06

4

1.44

2

13

Percent owned by directors

0.45

0.44

0.29

0.01

1

Percent held by largest shareholder

0.28

0.24

0.21

0.01

0.99

Widely held

0.42

0

0.49

0

1

B. Manufacturing corporations traded on Boston Stock Exchange

Total paid-in capital

912,742

750,000

589,363

100,000

2,500,000

Total shareholders

261

237

182

60

730

Board size

4.90

5

1.08

3

7

Percent owned by directors

0.10

0.07

0.06

0.02

0.26

Percent held by largest shareholder

0.07

0.05

0.07

0.01

0.36

Widely held

0.97

1

0.18

0

1

part, on how unusual the ownership structures of the great textile corporations were, relative to other firms. Panel A of table 2.4 presents data on the degree of ownership concentration among all manufacturing corporations. The average manufacturing corporation had around $210,000 in paid-in capital and forty-seven shareholders. It had a relatively small board consisting of four directors, who owned around 45 percent of the shares. Its ownership was relatively concentrated by the standards of modern public companies, with the largest investor holding 28 percent of the shares. By the definition of La Porta, Lopez-de-Silanes, and Shleifer (1999), only about 42 percent of Massachusetts’s corporations were “widely held,” in the sense of not having a 20 percent owner. For the average firm, the complaints of Ayer seem unlikely to have been relevant.

In contrast, panel B of the table presents the same statistics for the thirty-one manufacturing corporations in the sample whose shares were traded on the Boston Stock Exchange.26 Those firms included most of the great Waltham-Lowell textile mills, as well as a few other major industrial firms from other regions in the state. The data reveal that the Boston Stock Exchange firms were quite unusual. Their capital was more than fourfold greater than (p.92) average, and their numbers of shareholders were fivefold greater. Ownership by management was less than one-fourth that of the average corporation, as was the size of the largest stake held. These were huge corporations with an extraordinary degree of diffusion in their ownership. The problems faced by the shareholders of these firms were likely unique.

A more detailed portrait of the structure of ownership of manufacturing corporations is presented in table 2.5, which shows averages for each industry group where there was more than one operating corporation. The data in the table indicate that in nearly every industry managerial ownership was on average quite significant, and typically the largest investor owned more than 20 percent of the shares. Concentrated ownership was the norm, and the managers’ own stakes were likely sufficiently large so that they would at least partly internalize the costs associated with shirking or taking other actions harmful to the performance of the firm. On the other hand, their stakes were often so large that they held majority control and could not be removed from their positions by the other shareholders. This suggests that oppression of minority shareholders by dominant owners was likely to have been a potential problem among a substantial portion of the corporations.27

What explains the variation in ownership structures across industries? Why did some firms have much larger managerial ownership and smaller numbers of outside shareholders than others? The data in table 2.5 suggest that scale played a role: in the industries with the largest average capital, the degree of ownership concentration appears to be lower. This was likely driven, at least in part, by the constraints of raising large sums of money—it was probably necessary in the case of very large firms for a group of founding investors to seek investments from large numbers of outsiders.

The relationship between average firm scale and ownership across industries is explored more systematically in figure 2.3. The scatter plots in the figure clearly indicate that the number of shareholders was increasing, and the degree of managerial ownership was decreasing, in the average scale of the firm. However, scale was not the only driving force behind the variation in ownership concentration. Across industries and firms, ownership was likely more concentrated where the resulting benefits were greater. The next section analyzes the variation in ownership concentration and investigates the circumstances in which there may have been greater benefits to more concentrated ownership.

2.4.2 Ownership Structures and Production Methods

Incorporators chose the governance institutions of their firms endogenously, which renders any analysis of the relationship between those governance institutions and firm performance empirically problematic. But the (p.93)

Table 2.5 Corporate ownership: Industry averages, Massachusetts (1875)

Total capital

Total shareholders

Share owned by directors

Largest stake held

Share widely held

Clothing

   Boots and shoes

125,707

25

0.48

0.28

0.44

   Other clothing

202,174

64

0.42

0.24

0.43

Food and tobacco

   Food preparations

175,875

32

0.32

0.17

0.62

   Tobacco

14,088

8

0.66

0.15

1.00

Instruments

   Scientifc instruments and appliances

107,382

25

0.60

0.37

0.00

   Musical instruments and materials

122,363

10

0.56

0.29

0.20

Metals and metallic goods

   Agricultural implements

190,833

54

0.43

0.22

0.40

   Arms and ammunition

9,398

7

0.71

0.63

0.00

   Artisans’ tools

118,133

68

0.40

0.31

0.29

   Machines and machinery

157,666

26

0.52

0.28

0.27

   Other metals and metallic goods

171,375

25

0.45

0.29

0.40

Oils and chemicals

   Chemical preparations

89,076

25

0.40

0.28

0.25

   Fertilizers

218,000

30

0.41

0.41

0.50

   Oils and illuminating fluids

112,929

27

0.52

0.21

0.40

   Paints and colors

35,500

16

0.27

0.27

0.25

Paper and paper goods

   Paper

119,314

22

0.58

0.37

0.22

   Print works

185,200

12

0.66

0.48

0.25

   Printing and publishing

69,755

18

0.72

0.42

0.17

Textiles

   Carpetings

520,567

113

0.41

0.26

0.50

   Cotton goods

449,478

100

0.41

0.27

0.53

   Linen

550,000

36

0.55

0.41

0.00

   Woolen goods

198,005

39

0.54

0.31

0.28

   Other textiles

140,173

17

0.51

0.32

0.31

Vessels and carriages

   Carriages and wagons

84,000

6

0.91

0.39

0.00

   Vessels

350,000

42

0.45

0.23

0.50

Wooden goods

   Furniture

237,807

15

0.59

0.43

0.00

   Lumber

35,971

13

0.56

0.30

0.50

   Other wooden goods

67,975

35

0.56

0.26

0.50

Other industries

   Brick

186,222

35

0.44

0.27

0.50

   Glass

247,963

109

0.25

0.22

0.57

   Jute

119,000

14

0.49

0.28

0.33

   Leather

164,110

41

0.47

0.33

0.25

   Rubber

115,000

19

0.35

0.29

0.00

   Stone

119,194

31

0.44

0.31

0.43

Miscellaneous

   Miscellaneous manufactures

113,103

29

0.36

0.20

0.61

(p.94)
Corporate Governance and the Development of Manufacturing Enterprisesin Nineteenth-Century Massachusetts

Fig. 2.3 Firm size and ownership structure by industry

(p.95) corporate form was adapted into a broad range of industries, and if industry characteristics can be taken as exogenous for individual incorporators, then they can be used to empirically analyze some of the determinants of firms’ governance institutions.

In what follows, I analyze how the degree of ownership concentration, which is likely to be correlated with owners’ incentives to monitor management, varied with the production methods used across industries. The corporate form was sometimes adapted into industries dominated by establishments resembling artisanal shops, but it was much more commonly chosen within industries dominated by factories. Factories made greater use of steam power and unskilled labor, and their production processes likely created a greater division of labor and specialization of tasks among employees (see the discussion in Katz and Margo [2013]). Among these firms, managers likely performed a complex supervisory role that was critically important to the success of the firm. Although the organizational structure of most manufacturing corporations in 1875 was rather simple, some of the larger and more complex firms may have begun to develop new managerial systems and organizational structures (Chandler 1977). The complexity and importance of the role of managers within these firms, and the potential difficulties outsiders may have faced in monitoring and evaluating their performance, likely presented a challenge for their governance. Concentrated ownership may have been a solution.

From the certificates of condition, we cannot observe the production process or managerial structure of the corporations. However, the industry averages available from the census, which reflect the typical production methods used among all firms, can be used as a proxy for the firms’ own process. To the extent that they reflect the available choices of production methods given the state of technology and knowledge of management techniques, these industry averages can be taken as exogenous determinants of the governance institutions of individual corporations within each industry. In order to analyze this relationship, I estimate the following model:

yijc=α+δc+γprodj+θlogki+Xβ+εi,

where yijc is the governance measure of interest for firm i in industry j in county c, c is a county fixed effect, prodj is the average production method (such as the rate of use of steam power) in industry j, logki is the log level of capital of firm i, and X is a vector of other characteristics observed at the firm level. Thus the model estimates the relationship between industry-average production methods and the governance of individual firms, conditional on firm scale. Controls for geographical locations, in the form of county fixed effects, are included in order to address the possibility that different production techniques (again, such as the use of steam power) may have shifted the location of production into places that may have independently influenced firms’ governance structures. In order to address potential problems associated (p.96)

Table 2.6 Determinants of ownership structures

Dependent variable

Log total shareholders

Fraction owned by directors

Board size

Largest stake held

(1)

(2)

(3)

(4)

Steam engines per establishment

–0.581*

0.105+

–0.616+

0.0860+

(0.249)

(0.0526)

(0.313)

(0.0459)

Log paid-in capital

0.449**

–0.0558**

0.224**

–0.0257*

(0.0859)

(0.0163)

(0.0592)

(0.00945)

Constant

–1.969*

1.065**

1.701**

0.544**

(0.949)

(0.187)

(0.597)

(0.113)

Observations

459

461

461

460

R-squared

0.291

0.142

0.093

0.105

Log wage, adult male employees

0.689*

–0.135*

0.583+

–0.123**

(0.269)

(0.0511)

(0.334)

(0.0357)

Log paid-in capital

0.467**

–0.0579**

0.231**

–0.0267*

(0.0918)

(0.0182)

(0.0639)

(0.0108)

Constant

–6.829**

1.997**

–2.375

1.372**

(2.161)

(0.425)

(2.563)

(0.279)

Observations

453

455

455

454

R-squared

0.289

0.138

0.086

0.106

Fraction workforce female

–0.163

0.0945*

–0.224

0.0715+

(0.238)

(0.0398)

(0.297)

(0.0371)

Log paid-in capital

0.452**

–0.0589**

0.216**

–0.0266*

(0.0884)

(0.0176)

(0.0651)

(0.0109)

Constant

–2.258*

1.127**

1.549*

0.578**

(0.989)

(0.203)

(0.687)

(0.127)

Observations

464

466

466

465

R-squared

0.282

0.136

0.079

0.096

Note: Robust standard errors, adjusted for clustering by firm county, in parentheses. All specifications include county fixed effects.

(**) Significant at the 1 percent level.

(*) Significant at the 5 percent level.

(+) Significant at the 10 percent level.

with the use of industry-averages for the production method variables used in the regression, the standard errors will be adjusted for clustering by industry. The governance characteristics used as dependent variables in these regressions are the log number of shareholders, the fraction of the firm owned by the directors, the size of the board, and the size of the largest stake held.

Table 2.6 presents the results for specifications using steam power, the log annual wage paid to adult male employees, and the fraction of the workforce that was female—the first an indication of factory production; the second an indication of a greater degree of skilled labor, less consistent with factory (p.97) production; and the third a measure of the use of unskilled labor, consistent with factory production. The results of all specifications are clear and broadly consistent. The use of factory production methods, conditional on firm scale, was associated with more concentrated ownership and stronger governance by shareholders. In particular, the upper panel of table 2.6 indicates that in industries that relied more heavily on steam power, the corporations had fewer shareholders, were owned to a larger extent by their directors, had smaller board sizes, and the size of the largest stake held in their equity was greater.28 In contrast, corporations in industries where the wages paid to adult male workers were higher, which likely utilized more skilled labor in their production, had larger numbers of shareholders, lower ownership by directors, larger boards, and smaller maximum ownership stakes. The estimates for the share of the workforce that was female, although smaller and less statistically precise, are of the same sign as those for the use of steam power.

One potential source of concern with these results could be that an omitted variable, firm age, may be partly responsible for the observed correlations. If newer corporations tended to be formed in industries that utilized factory production methods, and newer corporations also had more concentrated ownership structures, as their founders had not yet sold off parts of their stakes (and the shares generally had less time to diffuse among investors), then the observed correlations between production methods and firm governance could be driven by firm age. In order to address this possibility, the date of incorporation of each firm was obtained from the Massachusetts Tax Commissioner’s Annual Report, and from it the log age of each firm was calculated.

Table 2.7 presents the results of regressions of the same specifications as those of table 2.6, but with log firm age included as an additional control. As expected, firm age is positively and strongly correlated with the number of shareholders, although not with other measures of ownership concentration. However, the inclusion of this variable does not substantially change the estimated effects of the production methods of the corporations on their governance. We can conclude that the results are not driven by firm age.

These results are somewhat speculative, and await confirmation in future work using more detailed data in which production methods can be observed for individual firms, rather than at the industry level. Nonetheless, they suggest that one of the ways in which the corporate form was successfully adapted into different industrial contexts was that ownership structures and governance institutions were varied endogenously. In particular, in corporations that operated in industries that relied heavily on factory production using steam power and a high proportion of unskilled labor, there was more concentrated ownership relative to corporations of similar sizes in industries (p.98)

Table 2.7 Determinants of ownership structures, conditional on firm age

Dependent variable

Log total shareholders

Fraction owned by directors

Board size

Largest stake held

(1)

(2)

(3)

(4)

Steam engines per establishment

–0.650*

0.106+

–0.606+

0.0860+

(0.260)

(0.0540)

(0.313)

(0.0463)

Log paid-in capital

0.416**

–0.0512**

0.244**

–0.0252*

(0.0815)

(0.0159)

(0.0642)

(0.0100)

Log firm age

0.138**

–0.0142

–0.0706

–0.000689

(0.0432)

(0.0113)

(0.0587)

(0.00677)

Constant

–1.818+

1.039**

1.610*

0.539**

(0.904)

(0.183)

(0.624)

(0.116)

Observations

455

457

457

456

R-squared

0.308

0.143

0.097

0.103

Log wage, adult male employees

0.815**

–0.156**

0.482

–0.131**

(0.283)

(0.0526)

(0.359)

(0.0379)

Log paid-in capital

0.438**

–0.0539**

0.250**

–0.0260*

(0.0883)

(0.0180)

(0.0699)

(0.0113)

Log firm age

0.132**

–0.0147

–0.0733

–0.00281

(0.0441)

(0.0107)

(0.0585)

(0.00691)

Constant

–7.533**

2.109**

–1.797

1.426**

(2.199)

(0.430)

(2.711)

(0.283)

Observations

449

451

451

450

R-squared

0.304

0.140

0.089

0.105

Fraction workforce female

–0.187

0.0972*

–0.202

0.0727+

(0.251)

(0.0409)

(0.297)

(0.0380)

Log paid-in capital

0.421**

–0.0545**

0.239**

–0.0260*

(0.0856)

(0.0174)

(0.0722)

(0.0115)

Log firm age

0.120**

–0.0132

–0.0802

–0.00128

(0.0424)

(0.0103)

(0.0584)

(0.00639)

Constant

–2.120*

1.101**

1.435+

0.573**

(0.957)

(0.200)

(0.722)

(0.130)

Observations

460

462

462

461

R-squared

0.296

0.138

0.083

0.095

Note: Robust standard errors, adjusted for clustering by firm county, in parentheses. All specifications include county fixed effects.

(**) Significant at the 1 percent level.

(*) Significant at the 5 percent level.

(+) Significant at the 10 percent level.

that relied more on skilled labor and less on steam power. This is consistent with the notion that incorporators and investors responded to the challenges posed by the complex role performed by managers in those firms by ensuring that there was adequate ownership incentives to monitor and supervise management.

(p.99) 2.5 Conclusion and Epilogue

Over the course of the nineteenth century the corporate form was adopted at high rates by manufacturing firms, particularly in Massachusetts. This chapter has analyzed the variation in incorporation rates, and the ways that corporations were owned and configured, across industries in 1875.

One important finding of the chapter is that among the large textile corporations of the state—the so-called Waltham-Lowell mills, whose shares were traded on the Boston Stock Exchange—the degree of ownership by the board was extraordinarily low, and there were very few if any large block-holders. The degree of separation of ownership from control among these firms was in fact typical of widely held modern American firms. However, the data presented in the chapter indicate that these great textile corporations were quite unusual. Most nineteenth-century manufacturing corporations were smaller, had fewer owners, and a high degree of ownership by their managers. Some were indeed extremely small—around 10 percent had four shareholders or fewer. Many of these firms adapted the corporate form to their needs by creating extremely small boards of directors; 10 percent had boards of two or fewer people. Most Massachusetts corporations were in fact controlled and operated by the men who owned them.

Unsurprisingly, the corporate form was adopted more frequently among firms in industries where average establishment size was higher. However, conditional on firm size, industries that made greater use of steam power and unskilled labor were incorporated at higher rates. This implies that the corporate form found heaviest use among those industries in which production was undertaken within factories, rather than artisanal shops. Evidently the corporation was used not only to achieve greater scale, but also to adopt mechanized, factory-based production methods.

The results also indicate that, conditional on firm size, corporations operating in industries that made greater use of steam power and unskilled labor had more concentrated ownership. That is, relative to corporations of similar sizes, those in industries that organized production within factories, rather than artisanal shops, had fewer shareholders, and a greater proportion of the shares were held by the directors. This is consistent with the notion that incorporators and investors responded to the challenges posed by the complex role performed by managers in those firms by ensuring that there was adequate ownership incentives to monitor and supervise management.

What happened after 1875? In the 1890s, several states, beginning with New Jersey, substantially liberalized their corporation laws, permitting businesses located in other states to incorporate within their borders, eliminating many restrictions on capital contributions, and enabling the formation of holding companies (see Larcom 1937; Grandy 1989). For a brief period, Massachusetts’s corporation laws, with their detailed annual disclosure requirements and strict limitations on corporate powers, were among the (p.100) most conservative in the United States.29 Ultimately in 1903, Massachusetts substantially liberalized its laws to reflect the “modern view that the State owes no duty to investors to look after the solvency of corporations” (Hall 1908). With this change, the detailed data on business corporations utilized for this study ceased to be collected.

Nevertheless, it is possible to follow the evolution of the use of the business corporation in Massachusetts into the early twentieth century using data generated by the imposition and collection of the federal corporate income tax in 1909. In 1909, there were 3,637 operating manufacturing corporations in the state, with $1.013 billion in capital. This was equivalent to 1.08 manufacturing corporations and $598 in capital per 1,000 persons, relative to a national average of 0.97 manufacturing corporations and $234 in capital nationally.30 Thus Massachusetts remained a prominent center of manufacturing corporations, but it was no longer as unusual in the extent to which the corporate form was utilized as it had been in the mid-nineteenth century. As new industries and new centers of innovation emerged, and as many states revised and liberalized their corporation laws, Massachusetts was eclipsed by other states.

References

Bibliography references:

Acheson, Graeme, Gareth Campbell, John D. Turner, and Nadia Vanteeva. 2014. “Corporate Ownership and Control in Victorian Britain.” QUCEH Working Paper Series no. 14–01. Queen’s University Centre for Economic History, Queen’s University Belfast.

Atack, Jeremy, Fred Bateman, and Robert A. Margo. 2008. “Steam Power, Establishment Size, and Labor Productivity Growth in Nineteenth-Century American Manufacturing.” Explorations in Economic History 45:185–98.

Atack, Jeremy, and Peter Rousseau. 1999. “Business Activity and the Boston Stock Market, 1835–69.” Explorations in Economic History 36:144–79.

Ayer, J. C. 1863. Some of the Usages and Abuses in the Management of Our Manufacturing Corporations. Lowell, MA: C. M. Langley & Co.

Becht, Marco, and J. Bradford Delong. 2005. “Why Has There Been So Little Blockholding in America?” In A History of Corporate Governance around the World, edited by R. Morck. Chicago: University of Chicago Press.

(p.101) Berle, Adolf, and Gardiner Means. 1932. The Modern Corporation and Private Property. New York: Macmillan.

Bodenhorn, Howard. 2012. “Voting Rights, Share Concentration and Leverage in Nineteenth-Century US Banks.” NBER Working Paper no. 17808, Cambridge, MA.

———. 2013. “Large Block Shareholders, Institutional Investors, Boards of Directors and Bank Value in the Nineteenth Century.” NBER Working Paper no. 18955, Cambridge, MA.

Cantillo Simon, Miguel. 1998. “The Rise and Fall of Bank Control in the United States: 1890–1939.” American Economic Review 88 (5): 1077–93.

Chandler, Alfred D. 1977. The Visible Hand. Cambridge, MA: Harvard University Press.

Cheffins, B. R., D. Chambers, and D. K. Koustas. 2013. “Ownership Dispersion and the London Stock Exchange’s ‘Two-thirds Rule’: An Empirical Test.” Business History 55 (2013): 670–93.

Coffee, John C. 2001. “The Rise of Dispersed Ownership: The Roles of Law and the State in the Separation of Ownership and Control.” Yale Law Journal 111 (1): 1–82.

Davis, Joseph S. 1917. Essays in the Earlier History of Corporations. Cambridge, MA: Harvard University Press.

Davis, Lance E. 1958. “Stock Ownership in the Early New England Textile Industry.” Business History Review 32:204–22.

Dodd, Edwin M. 1938. Lectures on the Growth of Corporate Structure in the United States with Special Reference to Governmental Regulation. Cleveland, OH: Cleveland Bar Association.

Freeman, Mark, Robin Pearson, and James Taylor. 2011. Shareholder Democracies? Corporate Governance in Britain and Ireland Before 1850. Chicago: University of Chicago.

Grandy, Christopher. 1989. “New Jersey Corporate Chartermongering, 1875–1929.” Journal of Economic History 49 (3): 677–92.

Guinnane, Timothy W., Ron Harris, Naomi Lamoreaux, and Jean-Laurent Rosenthal. 2007. “Putting the Corporation In Its Place.” Enterprise and Society 8:687–729.

Hall, Prescott F. 1908. The Massachusetts Business Corporation Law of 1903. Boston: William J. Nagel.

Handlin, Oscar, and Mary Flug Handlin. 1974. Commonwealth: A Study of the Rule of Government in the American Economy, Massachusetts, 1774–1861. Cambridge, MA: Harvard University Press.

Hannah, Leslie. 2007. “The Divorce of Ownership from Control from 1900: Re-Calibrating Imagined Global Historical Trends.” Business History 49:404–38.

Hilt, Eric. 2008. “When Did Ownership Separate from Control? Corporate Governance in the Early Nineteenth Century.” Journal of Economic History 68:645–85.

———. 2014. “General Incorporation and the Shift toward Open Access in the Nineteenth-Century United States.” Working Paper, Wellesley College.

Hilt, Eric, and Jacqueline Valentine. 2012. “Democratic Dividends: Stockholding, Wealth, and Politics in New York, 1791–1826.” Journal of Economic History 72 (2): 332–63.

Hovenkamp, Herbert. 1991. Enterprise and American Law, 1836–1937. Cambridge, MA: Harvard University Press.

Hurst, James W. 1970. The Legitimacy of the Business Corporation in the Law of the United States, 1780–1970. Charlottesville: University of Virginia Press.

Katz, Lawrence, and Robert Margo. 2013. “Technical Change and the Relative Demand for Skilled Labor: The United States in Historical Perspective.” NBER Working Paper no. 18752, Cambridge, MA.

Knowlton, Evelyn H. 1948. Pepperell’s Progress: History of A Cotton Textile Company, 1844–1945. Cambridge, MA: Harvard University Press.

(p.102) Lamoreaux, Naomi. 2014. “Revisiting American Exceptionalism: Democracy and the Regulation of Corporate Governance in Nineteenth-Century Pennsylvania.” NBER Working Paper no. 20231, Cambridge, MA.

La Porta, Rafael, Florencio Lopez-de-Silanes, and Andrei Shleifer. 1999. “Corporate Ownership around the World.” Journal of Finance 54 (2): 471–517.

Larcom, Russell C. 1937. The Delaware Corporation. Baltimore: Johns Hopkins Press.

Majewski, John. 2006. “Toward a Social History of the Corporation: Shareholding in Pennsylvania, 1800–1840.” In The Economy of Early America: Historical Perspectives and New Directions, edited by C. Matson, 294–316. Philadelphia: University of Pennsylvania.

Martin, Joseph G. 1886. Martin’s Boston Stock Market. Boston: Published by the author.

McGouldrick, Paul F. 1968. New England Textiles in the Nineteenth Century. Cambridge, MA: Harvard University Press.

Porter, Kenneth W. 1937. The Jackson and the Lees. Cambridge, MA: Harvard University Press.

Roosevelt, Theodore. 1901. Message of the President of the United States Communicated to the Two Houses of Congress at the Beginning of the First Session of the Fifty-Seventh Congress. Washington, DC: Government Printing Office.

Sylla, Richard, and Robert E. Wright. 2013. “Corporation Formation in the Antebellum United States in Comparative Context.” Business History 55 (4): 653–69.

Temin, Peter. 2000. “The Industrialization of New England, 1830–1880.” In Engines of Enterprise: An Economic History of New England, edited by P. Temin, 109–52. Cambridge, MA: Harvard University Press.

US Treasury Department. 1910. Annual Report of the Commissioner of Internal Revenue for the Fiscal Year Ended June 30 1910. House of Representatives, Document 1045. Washington, DC: Government Printing Office.

Ware, Caroline F. 1931. The Early New England Cotton Manufacture. New York: Houghton Mifflin.

Werner, Walter. 1986. “Corporate Law In Search of Its Future.” Columbia Law Review 81:1610–66.

Wright, Carroll D. 1877. A Compendium of the Census of Massachusetts: 1875. Boston: Albert J. Wright.

Wright, Robert E. 1999. “Bank Ownership and Lending Patterns in New York and Pennsylvania, 1781–1831.” Business History Review 73:40–60.

Yermack, David. 1996. “Higher Market Valuation of Companies with a Small Board of Directors.” Journal of Financial Economics 40 (2): 185–211.

Comment

Eric Hilt’s chapter focuses on late nineteenth-century Massachusetts, a state at the forefront of America’s industrial revolution. In particular, Hilt writes about the expansion of the corporate form among manufacturing firms that (p.103) went far beyond the well-known cases of the large textile mills in Waltham and Lowell. These mills were at the very start of America’s corporate ownership, but whether they were representative or exceptional in the set of firms adopting the corporate form is unclear in the current literature. Also unclear thus far is the organization of other manufacturing firms that adopted the corporate form. This chapter is therefore a welcome addition to the literature on organizational innovation that has lagged behind that on technological innovation.

Corporation data are available for Massachusetts since the early 1870s, when the state required all firms that had ever adopted the corporate form to file corporate certificates listing all managers and owners. Hilt collects information on all corporate certificates for 1875, matching 601 of these firms to business directories (for the business category), and also to the state manufacturing censuses (for firm size and incorporation date). The findings in table 2.1 show that textile firms were not typical among Massachusetts’s corporations, due to their unusually large size and dispersed ownership structure. Furthermore, there was a wide range of corporations in a variety of other sectors that adapted the corporate form to their particular needs, maintaining a small number of directors and stockholders with a high degree of manager ownership.

The diverse set of corporations in the Massachusetts manufacturing sector could be illustrated with further analysis of the available information in table 2.1. Incorporation rates are higher in more capital-intensive industries, as seen in figure 2.2, but incorporation rates could also have varied with respect to the degree of competition in a given industry. In a sector with few firms, adopting the corporate form may have been the only way to compete effectively, whereas in the presence of a large number of competitors such strategic behavior may not have paid off.

It is unclear where textiles fit in this picture. Even though the average capital in all textile establishments (as well as textile corporations) was higher than in other sectors, the incorporation rates did not take extreme values and varied between 17 and 56 percent. Nevertheless, these were higher incorporation rates than the nontextile average, which is perhaps telling of how special this sector was. Since one of the main contributions of the chapter is to highlight the importance of Massachusetts’s manufacturing firms beyond textiles, it would be useful to evaluate the overall impact of the textile sector. Do the average variables of interest (for example adoption of the corporate form, capital concentration) vary substantially if textiles are included? There certainly is value in the characterization of Massachusetts’s manufacturing industry beyond textiles. But the interested reader would also like to know how much the current literature missed of the early industrial revolution in the United States by focusing primarily on textiles.

Other than characterizing Massachusetts’s manufacturing firms, the chapter also evaluates two potential challenges of the adoption of the corporate (p.104) form. The first is managerial opportunism occurring when a firm’s ownership is so widespread that managers were virtually unaccountable to shareholders. The second is minority oppression present when managers use their privileged position to pursue their own interest, or that of large shareholders, at the expense of small shareholders.

To investigate the prevalence of these problems Hilt extracts ownership and governance data from the 1875 corporate certificates for the 601 matched firms in table 2.4. These firms show high degrees of ownership concentration relative to modern standards. The standard deviations of total paid-in capital and total shareholders are higher than the corresponding averages, which in turn are higher than the median values. This pattern signals a highly skewed distribution where very large firms distort average and median results, an indicator of minority oppression. Once the analysis is restricted to the very large firms traded on the Boston Stock Exchange (thirty-one in total) ownership concentration was considerably lower. This evidence suggests the prevalence of managerial opportunism for which there is indeed some anecdotal evidence in the large textile corporations (Ayer 1863; Porter 1937).

On this note, it would be useful to explore whether there was a trade-off between managerial opportunism and minority oppression as ownership concentration varied. In theory, higher managerial ownership should lead to minority oppression but there seems to be no anecdotal evidence for this, potentially due to three reasons: (a) higher degrees of ownership tend to occur in smaller firms, which do not make it to the newspapers or pamphlets and, in general, have less information available in the surviving records; (b) even if such information exists, minority oppression may have been easier to hide since it likely referred to small numbers of minority shareholders; and (c) minority oppression may also have been a smaller problem as minority shareholders in smaller corporations could have been more passive, otherwise they would probably be in management. The available data may not allow the analysis of all these issues, but some exploration of the trade-off would be welcome.

Among the thirty-one firms that were publically traded in 1875 it would be good to know the exact contribution of the large textile mills and if there was any considerable portion of those mills out of this restricted sample. Would excluding the other few major industrial firms from this set change in any way the comparison with the full sample? If the contribution of the remaining firms was substantial and that of unincorporated textile firms was negligible, it could be the case that only firms above a certain scale (eventually just in textiles) could take full advantage of the corporate form. Small firm owners (in textiles or not), therefore, could just have experimented with the new organizational form to surpass original resource constraints but not to considerably expand their production and become big firms.

Ultimately, the objective would be to evaluate the backbone of the Massachusetts manufacturing sector and whether the organizational innovation (p.105) helped it thrive. Were just a few large corporations pulling all the weight, or was this the role of many tiny firms whether or not they had adopted the corporate form? A clear answer to this question would allow for two additional avenues of research: (a) a within-industry comparison of adopters and nonadopters of the new organizational form to possibly identify its differential effect, and (b) Massachusetts’s changing role in American industrialization from an organizational perspective.

The epilogue alludes briefly to the stagnant status of corporate law in Massachusetts after 1875 when other states (New Jersey, for example) embarked in a liberalization trend, which made Massachusetts relatively less competitive. The natural question to ask is, therefore, whether Massachusetts was also falling behind in technology in the last quarter of the nineteenth century, or if textiles were no longer contributing much to the country’s industrial production. In either case, if the more competitive sectors were underrepresented in the state of Massachusetts there was little scope to compete with more organizationally advanced states.

Regarding the origin of Massachusetts’s organizational advantage, figure 2.1 shows a rapid rise of corporation charters in the state, relative to New England and also to the entire United States, until 1850. The data analyzed later in the chapter, however, comes from 1875 when Massachusetts may or may not have been dominant in the adoption of the corporate form. After the Civil War, Massachusetts’s organizational advantage could have changed via industry relocation due to the temporary disruption of cotton prices (Hanlon 2013) or the rising availability of cheap labor in the South in the last quarter of the nineteenth century (Carlson 1981; Wright 1979). Both these events were occurring under continuous innovation in the power technology sector, which itself could have affected Massachusetts’s competitive edge (Hekman 1980). Were figure 2.1 extended another fifty years we would be able to observe not only the rise of Massachusetts’s corporate advantage, but also its erosion. The latter was evident by 1901 when the state was considered one of the most conservative in terms of corporate law.

An ideal figure would provide a direct comparison of Massachusetts’s corporate and industrial advantages between 1790 and 1900. There would be two sets of lines corresponding to two different scales: the first would illustrate the number of corporations per 1,000 persons relative to New England and the United States (as already seen in figure 2.1) and the second set of lines would show industrial capacity per 1,000 persons (or any other metric) also relative to New England and the United States. We already know from figure 2.1 that Massachusetts outperformed New England and the United States until 1850 and supposedly the trend in industrial capacity was no different during that time period. We also know that at some time between 1850 and 1901 the state’s organizational advantage vanished, but this was a very eventful half-century in United States history so more discussion is needed on this matter.

(p.106) If the industrial advantage was still favoring Massachusetts in the early twentieth century, then the chapter could potentially say that the organizational edge was a leading indicator of future economic activity. In particular, it would be relevant to investigate whether the textile sector (despite not being representative of Massachusetts’s manufactures) was still at the technological forefront once the corporate edge was gone. If, on the contrary, the two types of advantages showed similar progress, organization may be more connected to technology than previously thought and there may be lessons to take from that as well.

In sum, Eric Hilt’s chapter sheds much-needed light in the understudied field of organizational change during the industrialization period and raises important questions about the contribution of the corporate form to American enterprise.

References

Bibliography references:

Ayer, J. C. 1863. Some of the Usages and Abuses in the Management of Our Manufacturing Corporations. Lowell, MA: C. M. Langley & Co.

Carlson, Leonard A. 1981. “Labor Supply, the Acquisition of Skills, and the Location of Southern Textile Mills, 1880–1900.” Journal of Economic History 41 (1): 65–71.

Hanlon, Walker. 2013. “Industry Connections and the Geographic Location of Economic Activity.” Unpublished Manuscript, Department of Economics, University of California, Los Angeles.

Hekman, John S. 1980. “The Product Cycle and New England Textiles.” Quarterly Journal of Economics 94 (4): 697–717.

Porter, Kenneth W. 1937. The Jacksons and the Lees. Cambridge, MA: Harvard University Press.

Wright, Gavin. 1979. “Cheap Labor and Southern Textiles before 1880.” Journal of Economic History 39 (3): 655–80.

Notes:

Eric Hilt is associate professor of economics at Wellesley College and a research associate of the National Bureau of Economic Research.

I would like to thank Jeremy Atack, Carola Frydman, Robert Margo, Claudia Rei, and William Collins for helpful comments. For acknowledgments, sources of research support, and disclosure of the author’s material financial relationships, if any, please see http://www.nber.org/chapters/c13132.ack.

(1.) For example, Becht and Delong (2005, 614) argue that the unusually diffuse ownership of American corporations “is not a long-standing historical tradition.” Likewise, Cantillo Simon (1998) argues that prior to 1890 American stock markets did not function actively, since corporate ownership was so concentrated in the hands of company founders. See also Dodd (1938), Hovenkamp (1991), Hurst (1970) and Coffee (2001). An important exception is Werner (1986).

(2.) The corporations subject to this requirement included manufacturing firms, utilities, and cooperatives. Banks, insurance companies, and railroads were subject to different regulations.

(3.) On the distinction between the two, see Katz and Margo (2013) and the references cited therein. On the significance of steam power for productivity, see Atack, Bateman, and Margo (2008).

(4.) La Porta, Lopez-de-Silanes, and Shleifer (1999) document that 80 percent of their sample of large modern American public companies are widely held in the sense that they do not have an owner holding 20 percent or more of their stock. In contrast, 98 percent of the manufacturing corporations traded on the Boston Stock Exchange were widely held by that definition.

(5.) See, for example, Temin (2000) and Handlin and Handlin (1974).

(6.) Recent work on British corporations, for example, includes Acheson et al. (2014); Cheffins, Chambers, and Koustas (2013); Freeman, Pearson, and Taylor (2011); and Hannah (2007). Work on the historical ownership of American corporations includes Bodenhorn (2012, 2013), Davis (1958), Hilt and Valentine (2012), Majewski (2006), and Wright (1999).

(7.) Lamoreaux (2014) details the origins and consequences of the restrictive elements of Pennsylvania’s corporation laws. Hilt (2014) presents detailed tabulations of the terms of the American states’ general incorporation acts.

(8.) This general regulating act for manufacturing companies to some extent standardized many of the terms in subsequent charters, and helped reduce the scope for special privileges to be granted in particular charters (Massachusetts Laws, 1809, ch. 65). The special privileges granted to some very early Massachusetts manufacturing corporations, ranging from lottery tickets to grants of land, are described in Davis (1917).

(9.) Massachusetts Laws (1822, ch. 38).

(10.) Massachusetts Senate Documents (1836, no. 90).

(11.) Governor’s message, January 1830, in Massachusetts Resolves, 1830.

(12.) Massachusetts Laws (1830, ch. 53).

(13.) Author’s calculations from the charters themselves, obtained from Massachusetts Laws (1830–1850).

(14.) Massachusetts Laws (1851, ch. 133).

(15.) Massachusetts was relatively late to adopt a general incorporation act; see Hilt (2014) for a comprehensive tabulation of general incorporation acts for manufacturing firms.

(16.) Massachusetts Laws (1855, ch. 68).

(17.) Unfortunately, the certificates of condition submitted prior to 1870 do not survive.

(18.) In particular, the Massachusetts Register and Business Directory (1878) and the New England Business Directory and Gazetteer (1877) were consulted, along with directories of individual towns. The industries of eleven of corporations could not be identified and were excluded from the analysis.

(19.) The collection and analysis of the data for the 1875 census was overseen by the chief of the Massachusetts Bureau of Statistics of Labor, Carroll D. Wright, who would later become the US Commissioner of Labor and oversee the 1890 Federal Census. The 1875 Massachusetts Census was designed and implemented using relatively sophisticated methods, and represented a substantial improvement over earlier state censuses. See Wright (1877).

(20.) The state did not require a minimum for revenues or size for establishments to be included in the census (Wright 1877, 103). However, excluded from these data are around 11,000 firms engaged in what the census categorized as “occupations,” rather than manufacturing. These occupations included blacksmithing, coopering, butchering, painting, sewing machine repairing, fish curing, cobbling, tinsmithing, roofing, plumbing, and related tasks. These firms had been classified as engaged in manufacturing in earlier state censuses.

(21.) The average capital of all establishments, column (2) in the table, is calculated by dividing total capital in the industry by the number of establishments. There is not sufficient data to calculate median capital from the census data. The table excludes industry categories with very few firms, and industry categories where the census reports did not present data on firm characteristics.

(22.) In contrast, working capital (“stock on hand’) was likely financed by commercial credit. All else equal, borrowing terms for unincorporated firms, whose owners faced unlimited liability, may have been more favorable than those faced by incorporated firms, and this may have contributed to the lower incorporation rates among firms in which stock on hand represented a substantial portion of total assets.

(23.) The correlation between the industry growth rate and log total employees was –0.36. It should also be noted that over the 1865–1875 period many industries in Massachusetts contracted significantly, and this contraction may have been related to the end of the Civil War. In addition to the arms and ammunition industry, which contracted by more than 70 percent, the oils and illuminating fluids, glass, and woolen goods industries, all of which had relatively high incorporation rates, saw substantial contractions in their output.

(24.) For example, between 1850 and 1875, the number of stockholders in the Pepperell Manufacturing Company grew from 117 to 321. At the annual meetings during that period, the number of stockholders present generally ranged from ten to twenty-five (Knowlton 1948, 16). See also McGouldrick (1968).

(25.) Massachusetts Laws (1865, ch. 236).

(26.) Martin’s Boston Stock Market indicates that in 1875, the stocks of around forty-four New England manufacturing companies were traded regularly on the Boston Stock Exchange. Among those forty-four, at least eleven were located in states other than Massachusetts. See Atack and Rousseau (1999) on the performance of Boston Stock Exchange traded shares during this period.

(27.) Hilt (2008) argues that early corporate governance institutions were often focused on addressing this problem.

(28.) Larger boards of directors have been found to hamper performance across a variety of measures among modern corporations; see Yermack (1996) and the references cited therein.

(29.) When Theodore Roosevelt became president, his first message to Congress signaled his intention to impose federal corporation laws, but reassure the business community that “supervision of corporations by the National Government need not go so far as is now the case with the supervision exercised over them by so conservative a State as Massachusetts” (Roosevelt 1901).

(30.) Author’s calculations from the US Treasury Department (1910) and the federal census. It is important to note that these data are not directly comparable to those of figure 2.1, which presents the total number of corporations created in all sectors, rather than manufacturing corporations that were actually in operation.

Claudia Rei is assistant professor of economics and history at Vanderbilt University.

For acknowledgments, sources of research support, and disclosure of the author’s material financial relationships, if any, please see http://www.nber.org/chapters/c13352.ack.