Selling the Sewing Machine Around the World: Singer's International Marketing Strategies, 1850-1920

Abstract

The sewing machine was one of the first standardized and mass-marketed complex consumer durables to have been diffused widely around the world before 1920. This global diffusion was almost the sole responsibility of one firm, Singer. Despite its American origins, Singer’s success lay principally overseas. New data provide insight into the company’s international marketing strategies. Although the firm had a reputation for marketing sophistication, Singer did not depend on price discrimination, extensive advertising, or loss-leading expansion of retail networks in its overseas markets. Rather, its success was due to the characteristics of consumer demand for sewing machines, features that combined with its strategic investments in market support services and in its selling organization to create Singer’s enormous competitive advantages in foreign markets.

Full text

266
Selling the Sewing Machine Around
the World: Singer’s International
Marketing Strategies, 1850–1920
ANDREW GODLEY
The sewing machine was one of the first standardized and mass-
marketed complex consumer durables to have been diffused widely
around the world before 1920. This global diffusion was almost the
sole responsibility of one firm, Singer. Despite its American origins,
Singer’s success lay principally overseas. New data provide insight
into the company’s international marketing strategies. Although the
firm had a reputation for marketing sophistication, Singer did not
depend on price discrimination, extensive advertising, or loss-leading
expansion of retail networks in its overseas markets. Rather, its suc-
cess was due to the characteristics of consumer demand for sewing
machines, features that combined with its strategic investments in
market support services and in its selling organization to create
Singer’s enormous competitive advantages in foreign markets.
The sewing machine was one of the first standardized and mass-
marketed complex consumer goods to spread around the world. It
was the nineteenth-century antecedent of the motor car, radio, televi-
sion, and other more recent consumer durables. This global spread
was quite remarkable. By the time of the First World War, sewing
© The Author 2006. Published by Oxford University Press on behalf of the
Business History Conference. All rights reserved. For permissions, please
e-mail: journals.permissions@oxfordjournals.org.
doi:10.1093/es/khj037
Advance Access publication March 31, 2006
ANDREW GODLEY is a Reader in Business History at the Centre for Interna-
tional Business History, University of Reading. Contact information: Centre
for International Business History, University of Reading Business School,
Reading RG6 6AA, UK. E-mail: a.c.godley@reading.ac.uk.
This article has emerged from a very long gestation and owes a great debt to Steve Tol-
liday for his editorial persistence. Intellectual debts have arisen from conversations
over many years with Fred Carstensen, Andrew Gordon, Les Hannah, Geoffrey Jones,
Avner Offer, and Mira Wilkins, although none should be held accountable for any of
the views expressed. For that the author alone has to take full responsibility.

Singer’s International Marketing Strategies 267
machines were already relatively common in households in Russia and
Eastern Europe, parts of Southeast Asia, the Middle East, and Southern
Africa, all far from the industrialized core of the world economy.1 This
global diffusion was predominantly the responsibility of just one firm.
Singer enjoyed a global market share of a little less than one-quarter in
the mid-1870s. However, by 1912 this had increased to “60 percent of
the family sewing machines in America, and probably 90 percent in
foreign markets,” an exceptional degree of market control.2
This ability to corner what was a rapidly growing global demand for
sewing machines meant that Singer had become the seventh-largest firm
in the world in 1912, and it is correspondingly prominent in the general
histories of the emergence of the modern corporation.3 One feature that
marks Singer’s “big business” experience as atypical, however, was the
firm’s extraordinary commitment to international marketing. It began in
the 1850s as a manufacturing firm, but the balance of employment in
Singer began to tilt toward marketing from the 1870s onward, as its retail
network grew. In contrast to all other emerging giant manufacturers of
this era, Singer’s resources came to be focused on overseas sales and not
on domestic sales, far less on production. By 1905 Singer employed
twice as many workers in marketing compared with its production oper-
ations, the overwhelming majority of whom worked outside the United
1. Andrew Godley, “The Global Diffusion of the Sewing Machine, 1850–
1914,” Research in Economic History 20 (2001): 1–46.
2. Robert Davies, Peacefully Working to Conquer the World (New York, 1976),
161. Singer’s history (in chronological order of publication) is covered by Andrew
Jack, “The Channels of Distribution for the Innovation: The Sewing Machine
Industry in America,” Explorations in Entrepreneurial History 9 (Feb. 1957): 113–41;
Mira Wilkins, The Emergence of Multinational Enterprise: American Business
Abroad from the Colonial Era to 1914 (Cambridge, Mass., 1970), chap. 3; Robert
Davies, “‘Peacefully Working to Conquer the World’: The Singer Manufacturing
Company in Foreign Markets, 1854–1889,” Business History Review 43 (Autumn
1969): 299–346; Davies, Peacefully Working; Fred Carstensen, American Enter-
prise in Foreign Markets (Chapel Hill, N.C., 1984); David Hounshell, From the
American System to Mass Production, 1800–1932 (Baltimore, Md., 1984), chaps.
2–3; and Andrew Godley, “Pioneering Foreign Direct Investment in British Manu-
facturing,” Business History Review 73 (Autumn 1999): 394–429. In 1872 Singer’s
total output was 219,758 machines (Godley, “Pioneering Foreign Direct Investment,”
table 2), 26% of the total output (851,236) of all American producers (Godley,
“Global Diffusion,” table 1). If adding in the infant sewing machine industries
around the rest of the world summed to a global output of less than 1 million,
Singer’s share would have been somewhat less than one-quarter. Note that
Friedman’s estimate of Singer’s U.S. market share in 1874 incorrectly conflates
both home and overseas sales; see Walter Friedman, Birth of a Salesman: The
Transformation of Selling in America (Cambridge, Mass., 2004), 97.
3. See Christopher Schmitz, “The World’s Largest Industrial Companies of
1912,” Business History 37 (Oct. 1995): 89; and Alfred D. Chandler, Jr., The Visible
Hand: The Managerial Revolution in American Business (Cambridge, Mass., 1977),
302–5.

268 GODLEY
States.4 Yet despite the firm’s prominence around the world, historians
have hitherto been unable to fully explain why Singer’s commitment to
international markets led to such unparalleled success.
This article aims to explain which features of Singer’s marketing strat-
egy were the most important contributors to this success. Using new data
only recently deposited in the Singer archive, and making comparisons
with the other prominent marketing-led firms of the era, the article con-
siders Singer’s evolving marketing strategies around the world.
The marketing mix (or the “four P’s”) emphasizes the importance
of competing through price, product differentiation, place (or loca-
tion), and promotion strategies in the pursuit of market share.5 But
Singer almost never pursued price competition; indeed, its sewing
machines were always relatively expensive. Nor was Singer promi-
nent in differentiating its machines. It was never a technological
leader, for instance. At different times the machines of Willcox and
Gibbs and Wheeler and Wilson, and later on some German competi-
tors like Pfaff, were supposedly superior.6 And while Singer prided
itself on its high standard of fit and finish, its machines never held a
particular superiority in design. Indeed, because of widespread imi-
tation, its machines were hardly differentiated from competitors at
all. Almost all sewing machines looked very similar around the turn
of the century.7 Moreover, though it is renowned among advertising
4. Singer Sewing Machine Company, A Century of Sewing Service, 1851–1951
(New York, 1951), 7, lists total employment in 1905 (and is presumably the source of
Davies, Peacefully Working, 140). A copy of this pamphlet is held in the Clydebank
District Library Archives, Clydebank, West Dunbartonshire, Scotland. International
Harvester was probably the only other large-scale corporation with a comparable inter-
national focus at this time. See Carstensen, American Enterprise in Foreign Markets.
5. There is an enormous literature on the marketing mix, its strengths, and
weaknesses, which is well summarized in David Jobber, Principles and Practice of
Marketing, 4th ed. (Maidenhead, U.K., 2004), chaps. 1–2.
6. On competitors see Davies, Peacefully Working, 5–15, 54–55; Hounshell,
From the American System to Mass Production, chap. 2; Frederick G. Bourne,
“American Sewing Machines,” in One Hundred Years of American Commerce,
ed. Chauncey M. Depew (1895; New York, 1968), 530; Nicholas Oddy, “A Beauti-
ful Ornament in the Parlour or Boudoir: The Domestication of the Sewing
Machine,” 269–84, and Tim Puttnam, “The Sewing Machine Comes Home,” 285–
302, both in The Culture of Sewing: Gender, Consumption, and Home Dressmak-
ing, ed. Barbara Burman (Oxford, U.K., 1999); and Ross Thomson, The Path to
Mechanized Shoe Production in the United States (Chapel Hill, N.C., 1989).
7. Hounshell, From the American System to Mass Production, 92–99; Oddy,
“Beautiful Ornament,” esp. 288–94. The Singer senior management letterbooks
are full of complaints about “imitators,” especially the German competitors. See
Wilkins, Emergence of Multinational Enterprise, for a discussion; and, for exam-
ple, see McKenzie to Clark, 31 May 1881, Box 94/1, Singer Archives, State Histor-
ical Society of Wisconsin, Madison, Wisconsin. Unless otherwise indicated, all
subsequent references to archival material are to the Singer Archives.

Singer’s International Marketing Strategies 269
historians, Singer’s success in promotion was rather more to do with
its direct selling organization knocking on people’s doors all around
the world.8
Exactly how its promotion strategies enabled Singer to attain its
near monopoly in overseas markets has, however, never been fully
explained. According to Alfred D. Chandler, Jr., Singer’s success sim-
ply came through “aggressive marketing.”9 But other companies
were surely just as aggressive. The Singer company’s principal histo-
rian, Robert Davies, ascribes the company’s success to “the product,
the sales and marketing organization, and the qualities of manage-
ment,” before also going on to emphasize the importance of the ratio-
nal organization, the impact of its advertising strategies, and the
personal qualities of senior management.10 But such inclusiveness
does not explain how the company was able to create its near-
complete monopoly in overseas sewing machine markets in the face
of strong competition. After all, many other companies also had fine
products, sound marketing organizations, excellent advertising strat-
egies, and so on, without going on to enjoy a similar degree of control
in their sectors.
This article supplements the existing body of historical scholar-
ship on Singer with new data from the firm’s internal management
accounts.11 These new sales data have enabled a near-complete
8. Which for “four P’s” aficionados is of course a conflation of “promotion” and
“place” strategies. For a literature on Singer’s role in advertising history, see (in
addition to references already cited) Judith Coffin, “Consumption, Production, and
Gender: The Sewing Machine in Nineteenth-Century France,” in Gender and Class
in Modern Europe, ed. Laura Frader and Sonya Rose (Cornell N.Y., 1996), 111–41;
Coffin, “Credit, Consumption, and Images of Women’s Desires: Selling the Sewing
Machine in Late Nineteenth-Century France,” French Historical Studies 18 (Spring
1994): 749–83; Monique Peyrière, “L’industrie de la machine à coudre en France,
1830–1914,” in La révolution des aiguilles. Habiller les Francais et les Américains,
19e–20e siècles, ed. L’École des Hautes Études en Sciences Sociales (Paris, 1996),
95–114; and Nancy Page Fernandez, “Creating Consumers: Gender, Class, and the
Family Sewing Machine,” in The Culture of Sewing: Gender, Consumption, and
Home Dressmaking, ed. Barbara Burman (Oxford, U.K., 1999), 157–68.
9. Chandler, Visible Hand, 304.
10. Davies, Peacefully Working, 335–37.
11. In the early 1990s the State Historical Society of Wisconsin at Madison,
Wisconsin, received two additional accessions from Singer, consisting of just over
2,000 microfilm reels of material. These are described as “unprocessed acces-
sions,” although the accompanying catalog clearly identifies what the material
contains. Compared with the previously available accession of 90 cubic feet of
archive boxes and 573 reels of microfilm, this recent accession probably repre-
sents at least a doubling of the quantity of available information, with much more
financial data, which is relatively rare in the original collection. Davies, working
with the original accession, states in Peacefully Working (p. x) that he had no
access to financial records, for example.

270 GODLEY
reconstruction of Singer’s growth in its overseas markets. Further-
more, when linked to senior management letterbooks and other qual-
itative sources previously available to scholars, what emerges is first
a confirmation of the extent of Singer’s success in selling machines
to its target markets around the world (and in contrast to its struggles
in its home market), and second a more complete explanation for the
firm’s near monopolization of overseas markets.
Like several previous studies, the article’s principal conclusion is
that Singer’s great innovation was its international selling organiza-
tion, an innovation that actually emerged in Singer’s British sales
subsidiary. In 1882 the company’s London agent referred to Singer’s
sales force as “a living moving army of irresistible power, peacefully
working to conquer the world.”12 But this irresistibility was not sim-
ply from the internationalization of Yankee-style salesmen. It was
rather the combination of Singer’s innovative direct selling model
and its extensive support services with the prevailing characteristics
of consumer demand that enabled the company to secure such a
remarkable stranglehold over the global sewing machine market in
its halcyon years before 1920.13
Taking on the World: The Emergence and Growth of Singer’s
Overseas Markets
For a pioneer in global marketing, Singer’s very first international
operations were hardly auspicious. In 1855 a licensing experiment to
a Parisian manufacturer ended in abject failure. The origins of Singer
are well known and entirely domestic. The French farce was in fact a
distraction from the American focus of the firm in its early years.
An 1850 partnership between the disreputable mechanical genius
Isaac Singer and the shrewd lawyer Edward Clark led to the formation
12. “Report of the Proceedings on the Occasion of Breaking Ground for the
Singer Manufacturing Company’s New Factory (at Kilbowie, near Glasgow,
Scotland),” 18 May 1882, p. 22 [hereinafter, “Breaking Ground”], Clydebank Dis-
trict Library Archives. The source also served as the title for Davies, ‘“Peacefully
Working,’” 319.
13. The disruption to global sales from the First World War, combined with
the sequestration of their Russian assets by the Bolshevik government, meant that
after the global restocking boom of 1919, Singer’s sales never regained their
former prominence. Only in 1929 did machine sales exceed 1919 levels (never
mind 1913), before falling again. See Carstensen, American Enterprise in Foreign
Markets; and Davies, Peacefully Working, esp. 329–30; see also table A2 in the
appendix.

Singer’s International Marketing Strategies 271
of the company I. M. Singer.14 The firm’s key assets were two of the
nine or ten patents critical to the successful manufacture of a sewing
machine. Clark (Singer’s president from 1876 to 1882) was the lead-
ing force behind the creation of the Albany patent pool, where the
holders of these key patents agreed to forgo litigation and to license
their technology to one another, thus allowing the industry to
develop.
After the Paris debacle and for the rest of the 1850s and early
1860s Singer’s senior management remained relatively cautious
when expanding overseas sales, leaving competitors like Wheeler
and Wilson and Willcox and Gibbs to seek out foreign sales.15
Singer’s international expansion was initiated by two energetic
employees, William Woodruff in London and George Neidlinger in
Hamburg.
Woodruff had worked for Singer since 1857, latterly as manager of
the Boston store, before being appointed to the London office in
1864. George Neidlinger took over the Hamburg agency from his
brother in 1865 after having worked as a mechanic in the New York
factory.16 Along with Clark and George McKenzie, a Scottish immi-
grant who was general manager for many years before succeeding
Clark as president in 1882, Woodruff and Neidlinger were responsi-
ble for developing Singer’s overseas markets.17
The timing of their appointments was no coincidence. According
to Clark, the sewing machine industry was just emerging from its
“experimental” stage in the early 1860s, when the American Civil
War disrupted domestic demand.18 But with the dollar depreciating
during the war, American producers exported more to make up some-
what for slack domestic demand. When postwar dollar appreciation
14. Isaac Singer’s career as a (mostly) serial monogamist is well documented
in Ruth Brandon’s biography, A Capitalist Romance: Singer and the Sewing
Machine (Philadelphia, 1977).
15. On Wheeler and Wilson see Davies, Peacefully Working, 28; on Willcox
and Gibbs see Hounshell, From the American System to Mass Production, 79.
16. See Woodruff to Clark, 19 May 1877, Box 93/9, and Neidlinger to Bourne,
16 Sept. 1893, Box 85/1, for their recollections. Summarized in Davies, Peacefully
Working, 39–42, 124.
17. For a more detailed discussion see Godley, “Global Diffusion,” 1–15;
Carstensen, American Enterprise in Foreign Markets, 333–35; and Davies, Peace-
fully Working, 61–62, 120–25. While Woodruff formally retired from the firm in
1885, he was kept on by his successor in London (whom Woodruff had
appointed) as a consultant until 1889. Neidlinger continued to serve in Hamburg
until 1893. See Davies, Peacefully Working, 124; and Neidlinger to Bourne, 16
Sept. 1893, Box 85/1. Their eventual withdrawals from management prompted
the assertion of New York’s control over all overseas markets by the early 1890s
under McKenzie’s successor as president, Frederick Bourne.
18. “Breaking Ground,” 15.

272 GODLEY
threatened sales in these nascent markets, American exporters were
faced with the dilemma of either investing more or withdrawing. Per-
haps because there were several Scots, like McKenzie, in senior man-
agerial positions, Singer alone developed a low-cost manufacturing
base with the first of its Scottish factories opening in 1867.19 Within
a year of its opening, the Glasgow factory supplied almost two-thirds
of London and Hamburg machine shipments, rising to 80 percent of
what were much higher sales by 1871.20
At the time, no doubt Singer’s American competitors thought
that Singer was being foolish. American domestic demand was
booming, with sales of machines increasing twelvefold, from
around 62,000 in 1866 to peak at around 770,000 in 1872. With an
annual growth rate in excess of 50 percent, the principal difficulty
facing American producers was keeping up with domestic
demand, not trying to cultivate sales overseas. But the American
market was transformed by two events in the 1870s. First, American
sales collapsed with the agricultural depression beginning in 1873,
falling by almost half to barely 400,000 in 1874. Demand contin-
ued to stagnate. Total sales across the industry only rose to around
570,000 machines by 1879, still a quarter below the 1872 peak.21
Second, the protection enjoyed by the patent holders ended with
their expiry in 1877. Singer’s response was unequivocal, slashing
retail prices by half to $30 within a week of the patent combina-
tion’s demise. This price war in the U.S. market, Woodruff
thought, “should soon so weed out competitors as practically to
leave the Singer Company alone in the field.” He was right. Within
six years the industry was almost wholly reorganized, with only
Singer and Wheeler and Wilson of the original combination members
surviving.22
From the perspective of New York headquarters, this may have
seemed a pyrrhic victory. For while Singer’s share of its domestic
19. Godley, “Pioneering Foreign Direct Investment,” esp. 413–26. Davies
states that the Scottish factory was 30% cheaper than that in New Jersey and com-
ments on the prevalence of Scottish nationality among senior employees. Davies,
Peacefully Working, 80–81, 125. McKenzie testified to his Scottish roots in
“Breaking Ground.”
20. “Results of American Offices,” Micro 2002, Reel 16, Segment 5, shows that
Glasgow supplied $199,675 worth of machines (at invoice value) out of a total of
$318,175 worth of shipments to London and Hamburg in 1868 (pp. 50, 53) and
$663,434 out of $829,384 in 1871 (pp. 86–87).
21. American domestic sales from Godley, “Global Diffusion,” 3–8.
22. Woodruff to Clark, 19 May 1877, Box 93, Folder 9, London Correspon-
dence Reports, 1870–80. Also see Davies, Peacefully Working, 57–58. New com-
petitors were to become important in the American market later on. Wheeler and
Wilson were eventually taken over by Singer in 1908.

Singer’s International Marketing Strategies 273
market had increased from less than a fifth to around a third by 1879,
this was achieved at a loss that year of almost a quarter million dol-
lars.23 The American business was in trouble. Like its rivals, Singer
had enjoyed the boom years and had used its profits to construct
what was an expensive retail network. But the depressed demand
and lower prices should have then led to a reduced retail coverage in
the American market in the late 1870s, throwing its experiment in
vertical integration into reverse. Instead Singer’s U.S. retail organiza-
tion almost tripled in size, growing from around 200 branches in
1877 to over 550 by December 1879 and going on to employ around
13,000 in the U.S. retail organization alone by 1882.24
With prices having fallen so precipitously in 1877, the costs of
distribution had to be cut. It was a task that proved almost beyond
New York senior management.25 In May 1882 McKenzie grimly
wrote from London, “The American business does not look very
encouraging,” contrasting it with the success of Singer’s Great Britain
agency.26
Table 1 shows the early growth in the London and Hamburg
offices. These overseas sales organizations were by comparison
much leaner than the American one. Together they were able to gen-
erate over $340,000 of business in 1867, rising to over half a million
in 1868, before topping $1 million in 1871 and then $2 million in
1873. Given the context of booming domestic sales, that the two
overseas offices were able to make an inroad into the company’s
23. Singer’s total output in 1872 was 219,758 machines, but 59,048 were
produced in Scotland (see Godley, “Pioneering Foreign Direct Investment,” table 2)
and not sold in the United States. This leaves a maximum of 160,710 that could
have been sold within the United States, but of these several thousand would
have gone to other overseas markets. Godley, “Global Diffusion,” table 1, esti-
mates that 770,037 machines were sold in the United States in 1872. If Singer sold
as many as 150,000 in its home market, that would represent a 19% market share.
See “Results of Business Monthly America and Other Offices, 1879,” P92-9870,
Micro 2002, Reel 17, 27–29, for 1879 profits and losses.
24. On U.S. branches in 1877, see Davies, Peacefully Working, 58. Employ-
ment in the U.S. retail network has been inferred from “Breaking Ground,” 22,
which states that Singer’s total employment in 1882 was 25,000, of whom 6,000
were employed in the European sales force and 2,500 in the Scottish factory. With
the Glasgow factory producing 45% of total output (Godley, “Pioneering Foreign
Direct Investment,” table 2), total factory employment at the two factories (the
main plant was at Elizabethport, New Jersey) was surely no more than 6,000 in
1882, leaving a balance of around 13,000 employed in the U.S. retail network.
25. For the attempts to restructure the expensive American distribution net-
work after the 1877 price reductions, see Carstensen, American Enterprise in For-
eign Markets, 22–23; and Thomas S. Dicke, Franchising in America: The
Development of a Business Method (Chapel Hill, N.C., 1992), 40–42.
26. McKenzie (in London) to Oscar Graham (in New York), 11 May and 27 May
1882, “Travelling letterbooks, McKenzie, 1882, April–June,” Micro 2013, Reel 63.

274 GODLEY
distribution of sales (from less than one-fifth in 1867 to just over one-
quarter by 1873) represented a considerable achievement. Indeed,
already by 1873 the overseas offices were actually more profitable
than the more expensive American retail network.27
Woodruff operated with only thirty-five offices worldwide in
1879, Neidlinger with only fifty-four. By 1882 total employment in
the overseas sales organization was only 6,000, less than half the
number employed in the U.S. retail network.28 Given that overseas
sales through these two offices had by then pulled ahead of Singer’s
American market, the inefficiency of the U.S. retail network was
exposed. It was the profits of well over three-quarters of a million
dollars from Hamburg and London in 1879 that were subsidizing the
loss-leading expansion in the United States.29 By 1879 the U.S. mar-
ket accounted for only 48 percent of Singer’s total sales. By 1902 that
share had fallen to 27 percent and by 1912 to a mere 17 percent.30
Already by the late 1870s, therefore, the center of gravity of
Singer’s business had shifted decisively away from the United States
and moved to encompass Western, Central, and Southern Europe. Its
marketing strategy in the United States had not been able to cope
with a double whammy of a slump in domestic demand and the
demise of price protection. The domestic American market remained
27. “Results of American Offices, for the year ending Dec 31st 1873,” 123,
shows total profits from the American offices of $491,262 and from the two over-
seas offices of $513,900.
28. “Results of Business Monthly America and Other Offices, 1879,” P92-9870,
Micro 2002, Reel 17, pp. 27–29. This may overstate the difference slightly, as the
London and Hamburg agencies were beginning to open branch offices. Employ-
ment in 1882 from “Breaking Ground,” 22.
29. “Results of Business Monthly America and Other Offices, 1879,” P92-9870,
Micro 2002, Reel 17, pp. 27–29.
30. See appendix, table A2.
Table 1 Singer’s United States and Overseas Sales, 1867–1873
Sources: “Results of Business (Branches), Annual 1867–73,” Micro 2002, Reel 16,
Segment 5 (“Gross sales”), Singer Archives, State Historical Society of Wisconsin, Madison,
Wisconsin. Only partial data are available for 1868.
Y
ear United States ($) Overseas ($) Total Sales ($)
U.S. Share of
Total Sales (%)
1867 1,487,717 340,935 1,828,652 81.4
1868 — 521,398 — —
1869 1,783,142 559,039 2,342,181 76.1
1870 2,481,170 504,400 2,985,570 83.1
1871 3,044,301 1,043,523 4,087,824 74.5
1872 4,976,031 1,374,743 6,350,774 78.4
1873 6,004,907 2,050,446 8,055,353 74.6

Singer’s International Marketing Strategies 275
important to Singer, but it never dominated company planning
again.
Figure 1 reports the growth in sales for several key national mar-
kets. It illustrates just how significant the growth was after the period
of these key initial developments in overseas markets. From when
Frederick Bourne succeeded McKenzie as company president in 1889
through to the outbreak of war in 1914, Singer experienced a period of
consolidation in its policies and growth in overseas demand.31
As new markets emerged, Singer widened its geographical cover-
age. Russia emerged as the most important market by far after the
turn of the century, even eclipsing the American market from 1908.
Finally, at the end of the period, demand grew fast in parts of Asia
and Africa. The Philippines, Japan, the Ottoman Empire, and South
Africa, all insignificant markets before 1914, collectively accounted
31. Davies, Peacefully Working, esp. 60–62, 109, 114, 121–28. In the com-
pany’s centenary publication, the Bourne presidency is described as “a period of
consolidation” after the innovations of the Clark and McKenzie eras. Singer
Sewing Machine Company, Century of Sewing Service, 6, Clydebank District
Library Archives. Bourne’s longtime vice president Douglas Alexander succeeded
him as president in 1905. Davies writes, “It is difficult to delineate the policies of
Douglas Alexander from those of his predecessor.” Davies, Peacefully Working, 144.
Figure 1 Singer’s global sales, 1868–1920 (total machines sold and share in major markets).
Source: Appendix, table A2 (U.S. sales for 1879 used for 1880). Market abbreviations in
appendix, table A1. “All Overseas” excludes overseas markets served from New York
headquarters prior to the 1890s, including all North, Central, and South American markets.
UK
RUS
F
ITA
DAH
SP&P
OB
All Overseas
USA
USA
World Sale
s
0
500000
1000000
1500000
2000000
2500000
3000000
1868
1871
1874
1877
1880
1883
1886
1889
1892
1895
1898
1901
1904
1907
1910
1913
1916
1919
machines

276 GODLEY
for almost a quarter million machine sales in 1919, or nearly one-
seventh of the company’s total output (see appendix, table A2).32
The new sales data therefore confirm what previous studies have
shown. Singer had become a truly global company before the First
World War. However, the significance of its overseas developments
is really better understood in the context of the collapse of the American
sewing machine market in the 1870s. For only Singer out of all the
leading American producers had the financial muscle to retain,
indeed to expand, its American domestic presence, despite the losses
incurred. Then, during the 1880s and after, it was Singer alone that
went on to dominate overseas markets.
Singer’s Impact in Overseas Markets: The Diffusion of Singer
Sewing Machines
With almost 85 percent of its sales outside its home market before the
First World War, Singer evidently was committed to overseas markets
to an extent that few, if any, other contemporary American manufac-
turers could match. But Singer’s importance in the development of
global business is not only as a pioneer in many markets but also in
the way it was apparently able to exert its “dominance over the world
markets for sewing machines.”33 Before going on to evaluate the role of
the company’s marketing strategies, it is important to establish just
how successful the company really was in these overseas markets.
Unfortunately, there is simply insufficient data for most of the
individual markets surveyed here to be able to use market share as an
indicator of the company’s overseas success. However, an alternative
method of assessing Singer’s impact in its overseas markets is to
examine the diffusion of its sewing machines around the world. The
sewing machine, as a durable product, was seldom replaced. Once a
family had purchased a sewing machine, they would very rarely pur-
chase another.34 Given that in each market there was a finite number
32. Global manufacturing facilities were also extended but only significantly
so in Russia. On the large Russian factory (established 1905) see Carstensen,
American Enterprise in Foreign Markets; on the small German factory see Davies,
Peacefully Working.
33. Davies, Peacefully Working, 334.
34. Family machines represented 85% of Singer’s output (this comes from figures
for the Scottish factory, compiled by a factory employee, kindly passed on by Fred
Carstensen), so it is reasonable to infer that Singer’s target market was (as its market-
ing literature suggested) families in their homes. Electric machines, the first innova-
tion to prompt product replacement, were first introduced in 1908 and only began to
attain popularity in overseas markets outside the period under consideration here.

Singer’s International Marketing Strategies 277
of families, capturing how successful the firm was in selling to its
potential customers can be done through comparing cumulative
sales over time per household—or the relative diffusion of Singer
sewing machines—across its overseas markets.35 Unfortunately, data
on the number of families and households in each location are also
lacking, so figure 2 shows Singer’s cumulative sales per capita in its
principal markets.
Figure 2 shows how the sewing machine diffused fast and wide in
the United States, where aggregating all sales of sewing machines
from the entire industry suggests that 13 percent of the population
may have had a sewing machine by 1879, or (assuming one per
household) almost two-thirds of all U.S. households; a level of mar-
ket saturation that perhaps helps to explain why sales stagnated
there for several years after 1872.
By comparison in Great Britain 11 percent of the population had
purchased a Singer sewing machine by 1914. The mean British
household size, at 4.4 persons, was slightly smaller than in the United
States, but this still equates to 47 percent of British households
35. In fact, diffusion is overstated in this conventional measurement because
of the typical nonallowance for depreciation. See Sue Bowden and Avner Offer,
“Household Appliances and the Use of Time: The United States and Britain since
the 1920s,” Economic History Review 47 (Nov. 1994): 725–48.
Figure 2 The diffusion of Singer’s sewing machines, 1858–1920 (machines per 000
population). Source: Appendix, table A6. Note that the diffusion figures for the United
States are based on data for all sewing machine sales there.

278 GODLEY
having purchased a Singer machine by then.36 A similar level of satu-
ration had occurred by then in Australia and New Zealand.
Diffusion levels were much lower among the European core. Per
capita diffusion levels reached between 5 and 7 percent of the popu-
lation by 1914 throughout Western Europe, Switzerland, Scandinavia,
and Iberia. In Germany and Austria-Hungary 5 percent of the popula-
tion had a Singer machine. (Although diffusion in Germany was
almost certainly higher than in Austria-Hungary, because of the per-
sistent aggregation of Singer’s sales data for these two national mar-
kets throughout this period, it is impossible to specify either with
any great accuracy.) Data on household size are much scarcer for this
group of nations, but a crude approximation of five people per
household (that is, in between the American and British household
sizes) would lead to the conclusion that somewhere between one-
quarter and one-third of households throughout Europe had bought a
Singer sewing machine by 1914. In the European periphery diffusion
began later but caught up much faster than in the core, with 4 per-
cent of the population in Italy and Russia having bought a sewing
machine by 1914, perhaps equating to somewhere between 15 to 20
percent of households.
The less developed nations outside Europe mostly experienced
much lower diffusion levels throughout, but diffusion gathered pace
in the final years covered in figure 2. The data suggest that by 1920,
5 percent of South Africans, 3 percent of Filipinos, 2.5 percent of the
Ottoman and Balkan population, 0.75 percent of Japanese, but only
0.3 percent of Indians had Singer sewing machines by that time. At
the assumed average size of five persons per household, this trans-
lates into perhaps 25 percent of households in South Africa, 15 per-
cent in the Philippines, 12 percent in the Ottoman region, and just
under 4 percent of households in Japan that had Singers, but only
1 percent in India.
In sum therefore, during the period under consideration, Singer
had been able to sell its sewing machines to (in a rough order of mag-
nitude) half the households in the United Kingdom, Australia, and
New Zealand, somewhere between one-quarter and one-third of
households in the developing European economies and South
Africa, and somewhere between one-in-ten and one-in-five in the
36. Mean household size in the United States was 5.1 (in 1870), and in the
United Kingdom it was 4.4 (in 1911). See Brian R. Mitchell, International Histori-
cal Statistics: Europe, 1750–2000 (Basingstoke, U.K., 2003); Brian R. Mitchell,
International Historical Statistics: Africa, Asia, and Oceania (Basingstoke, 2003);
and more generally Paul Bairoch, Cities and Economic Development (London,
1988).

Singer’s International Marketing Strategies 279
less developed economies of Italy, Russia, the Philippines, and the
Ottoman Empire. Even something approaching one-in-twenty house-
holds in Japan had purchased a Singer machine by 1920, despite tra-
ditional fashion styles there that required far less stitching.
This is not the place for a formal analysis of the relationship
between the diffusion of the sewing machine and its likely determi-
nants (including per capita income, price, household size, fashion
styles, and so on).37 Rather, it is more important here to give some
sort of assessment of Singer’s impact in these overseas markets.
Given the very large disparities in per capita income, it would be
wholly unrealistic to expect similar levels of diffusion in these very
different markets around the world. But comparing Singer’s sales
in these markets with the all-industry diffusion in the United
States in the years up to 1879 gives some sort of benchmark against
which to gauge Singer’s performance. As the nation of invention
and initial commercialization, it would be expected that product
diffusion would occur first in the United States. Once some allow-
ance for this lag is taken into account, by comparing Singer’s sales
with the all-industry sales in the United States, it would appear
that the company was very successful in selling to the Australian,
New Zealand, and British markets up to 1914, for instance. After
all, the diffusion of Singer machines alone there matches that of
the entire industry in the United States in the earlier period. And
while Australian incomes were a little higher, British incomes by
1914 still substantially lagged those prevailing in the United States
in 1879.38 Similar relationships can be observed for the other markets
listed.
Once allowances for the time lag and differences in income levels
are made, the diffusion of Singer machines elsewhere also appears to
be roughly proportionate to the American diffusion for the entire
37. Singer’s pricing strategy is considered below. See Godley, “Global Dif-
fusion,” esp. 23–31, for some tentative comparisons of diffusion with per cap-
ita income levels in these nations. It is noteworthy, for example, that once
differences in income are taken into account, diffusion rates were not posi-
tively correlated with Singer’s factory locations. While diffusion was relatively
advanced of income levels in the United States, it was hardly the case for Great
Britain, Germany, or Russia. There is a great deal of anecdotal and photo-
graphic evidence of fashions changing at the same time that the sewing
machine was spreading, but a systematic analysis of the relationship awaits.
For comments on the diffusion of western dress, see Christopher A. Bayly, The
Birth of the Modern World, 1780–1914: Global Connections and Comparisons
(Oxford, U.K., 2004), 5–15.
38. Jeffrey Williamson, “The Evolution of Global Labor Markets since 1830:
Background Evidence and Hypotheses,” Explorations in Economic History 32
(April 1995): appendix 2.

280 GODLEY
industry.39 This raises an important point: if Singer’s sales were
therefore approximately close to all that could be expected in these
markets, the implication is that there was relatively little additional
demand being met by the company’s competitors. Clearly it is not
possible to infer from these diffusion estimates Singer’s actual share
of its overseas markets with any great precision. But in the absence of
data on competitors’ sales in these markets, these estimates do offer
prima facie evidence to support the view that Singer genuinely did
enjoy market shares somewhere in the order of magnitude of 80 to 90
percent.
Though further analysis of the relationship between income (and
other factors) and sewing machine diffusion obviously remains to be
done, this nevertheless is a noteworthy finding. For there are rela-
tively few occasions in global business history when one firm has
been able to so dominate a global market for such a sustained period
of time. The new sales data therefore do appear to confirm that
Singer enjoyed a remarkable competitive advantage in the sewing
machine industry, an advantage derived from a specific organiza-
tional innovation.
Innovation in Singer’s International Marketing Strategies
Britain, 1864 to the 1880s: The Canvasser-Collector System
Singer began its overseas expansion in the mid-1860s with two agents
already familiar with its retailing operations in the United States.
Accordingly, the early attempts by Woodruff and Neidlinger to develop
British and German sales adopted the American parent’s marketing
strategy. They focused on clothing industry demand, with elaborate,
company-controlled retail outlets in the main clothing industry centers
(smaller centers supplied via wholesalers and independent dealers),
and heavy use of mass advertising to promote the brand more widely.
Compared with its American rivals, the company was unusual in
according even some significance to the family market from almost
39. British real wages in 1913 were around 85% those in the United States in
1876; see ibid. Moreover, in 1913 German, Belgian, Scandinavian, and Irish real
wages were around two-thirds and French, Spanish, Portuguese, Dutch, and Italian
real wages were around half those in the United States in 1876. Süleyman
Özmucur and Sevket Pamuk suggest that real wages in London were 2.7 times
higher than in Istanbul by 1913 (“Real Wages and Standards of Living in the Otto-
man Empire, 1489–1914,” Journal of Economic History 62 [June 2002]: 313), or
perhaps less than a third of those in the United States in the 1870s.

Singer’s International Marketing Strategies 281
the very beginning, with women featuring prominently in its copy
even in the 1850s. For instance, Clark is credited with introducing
installment purchasing into the sewing machine industry in 1856 to
target the family market, perhaps copying the idea from piano sales-
rooms adjacent to Singer’s offices on Broadway. And while Singer
later went on to become “notorious for their aggressive ‘dollar down,
dollar a week’” plans, the family market nevertheless remained
embryonic until after the end of the Civil War. Three-quarters of all
sales were to garment manufacturers in the early 1860s.40
Following the American parent’s lead, Woodruff initially opened
a small branch network in the manufacturing centers only and pur-
sued a mass advertising strategy to reach the British clothing indus-
try in the 1860s. Though the family market took off in the United
States after 1865, it was only in the mid-1870s that Woodruff’s focus
shifted to the family market by, first, introducing the installment sys-
tem into Britain in 1873 and, second, opening more retail outlets in
the population centers from 1875. But the key innovation in reaching
the family market in Britain was to create a cadre of canvassers. Can-
vassers were based in the retail stores, but they approached the pub-
lic directly in their homes, where they demonstrated the sewing
machines and collected the consumers’ signed orders.41 As Woodruff
explained in a letter to Clark in 1877, “In this country ... we must in
all cases keep control of the trade in our own hands. I hold that we
can never make our business solid except by Branches at all the great
centres.... We have long since proved that a small country like this
can be most successfully worked upon the retail prices direct to the
public.”42
Soon a system of checking needed to be introduced. Too many
canvassers were opening new accounts with poor prospects, and the
repossession rate soared. To counter this, Woodruff created a
separate cadre of collectors. The retail branch manager doubled as
the sales manager. Under what was called a “circuit plan,” each
40. Lendol Calder, Financing the American Dream: A Cultural History of
Credit (Princeton, N.J., 1998), 163–65, quotation at p. 165.
41. Davies, Peacefully Working, 39–41; McKenzie to Woodruff, 16 April 1879,
Box 93, Folder 9. On the introduction of the installment system in 1873 and the
extension of the branch network from 1875, see “Early Days,” testimony of long-
serving agent John A Smith, in Singer’s in-house journal in the United Kingdom,
Red ‘S’ Review 1 (1919–20), 20–21. The only extant copies of this journal that I
have been able to locate are in Clydebank Distribt Library Archives. For U.S.
developments in the 1870s see Dicke, Franchising in America; and Friedman,
Birth of a Salesman, 93–97; both based on Jabk, “Channels of Distribution.”
42. Woodruff to Clark, 19 May 1877, Box 93/9. Also see Carstensen, American
Enterprise in Foreign Markets, 19–23.

282 GODLEY
canvasser was allocated a fixed district and expected to knock on the
door of every house in their district every two years, getting to know
every family, offering advice, soliciting introductions, and servicing
existing Singer machines. Once new accounts were opened, collec-
tors then delivered the machines and followed up with the subse-
quent requests for the weekly installments for up to two years
afterward. The canvassers received most of their remuneration as
commission on the accounts they opened, but only once the collec-
tors had begun to receive payments, so they had little incentive to
take dubious orders. This overcame the principal difficulty facing
any sales organization, the risk of agents acting opportunistically.43
“[N]o [canvasser] must be allowed to take an order and complete
arrangements,” stated McKenzie bluntly, so that “it will be almost
impossible to rush in bogus transactions.”44
The retailing structure in the United Kingdom expanded with
alacrity from 26 retail outlets in 1879 to 303 by 1885.45 It was an
enormously successful system, with sales growing in the United
Kingdom from just 30,000 in 1875 to over 90,000 in 1884. Competi-
tors took flight. Singer’s leading American rival, Wheeler and Wilson,
had to withdraw from Britain in 1879 after unsuccessfully trying to
match Singer’s retail investments. A British trade journal stated in
1881 that the output of British sewing machine producers had
“shrunk to very insignificant proportions.”46
In fact, the marketing strategy was too expensive for Singer to
maintain. Flooding the country with shops, canvassers, and collec-
tors was a very expensive method of selling.47 Already in 1882 it was
recognized that the principle of separating canvassing from collect-
ing could be more efficiently pursued by allowing canvassers to both
43. There were additional features, including the role of a guarantee fund to
minimize embezzlement and by the 1890s an audit system of branch offices, but
the key innovation was direct door-to-door selling and collecting. On Woodruff’s
innovation see Carstensen, American Enterprise in Foreign Markets, 19–21; and
Davies, Peacefully Working, 62–66.
44. McKenzie to Woodruff, 16 April 1879, Box 93/9.
45. “Schedule B, Central Office London and Branches, 1879,” P92-8970 Micro
2002 (17), 1: Schedule B, 45; “Ratio of Cash Sales at Offices during 1885,” Box
103/5. On the relative size of Singer’s retail structure in the United Kingdom, see
Scott Fletcher and Andrew Godley, “Foreign Direct Investment in British Retail-
ing, 1850–1962,” Business History 42 (April 2000): 43–62; and Andrew Godley,
“Foreign Multinationals and Innovation in British Retailing: 1850–1962,” Busi-
ness History 45 (Jan. 2003): 80–100.
46. Quoted in Davies, Peacefully Working, 86. On Wheeler and Wilson see
Wilkins, Emergence of Multinational Enterprise, 43–44 (the firm continued to
export).
47. As noted by Friedman, Birth of a Salesman, 97.

Singer’s International Marketing Strategies 283
canvass and collect. A check on the legitimacy of the new accounts
came through the branch manager (or a managing collector) who
supervised the key transaction of delivering the actual machines and
taking the deposit.48 What became known as salesmen-collectors
thus slowly emerged during the following years in place of the sepa-
rate cadres of canvassers and collectors. This revised canvasser-
collector system continued to be hugely successful. After languishing
behind its rivals in Woodruff’s early days in London, Singer’s market
share in the United Kingdom climbed to at least 75 percent by
1890.49
One consequence of developing a new method of direct selling
was to alter the company’s promotion strategy. In Clark’s original
marketing model, advertising was given great prominence; “a large
part of our own success we attribute to our numerous advertisements
and publications,” wrote Clark.50 But as the new selling organization
grew in the United Kingdom, so the public got to hear and see
Singer’s machines directly in their homes from the canvassers.
As table 2 highlights, Singer’s dependence on advertising corre-
spondingly fell. Woodruff’s initial marketing strategy in London fol-
lowed Clark’s American model and was based on quite heavy
advertising expenditures, with between 4 and 6 percent of annual
revenues going to advertising most years between 1867 and 1873.
Indeed, he may even have overextended the intensity of advertising
expenditure at the same time as expanding the canvasser-collector
system. McKenzie railed against Woodruff’s advertising policy in
1879, complaining that “upwards of £10,000 was spent last year on
newspaper advertising alone.... I look upon it as a prodigal waste of
money.” Woodruff was soundly rebuked for losing control of adver-
tising spending in the overseas offices: “The sum spent in France last
year ... was fabulous and averaged 10/- per machine sold! I have
insisted on this being stopped at once.” McKenzie insisted that an
average advertising expenditure of 2/- per machine was sufficient in
all markets.51 With Singer’s New Family sewing machine retailing
for a little over £5, McKenzie was effectively instructing the London
48. See “Circular to Agents,” 25 Sept. 1882, Box 94/4.
49. Andrew Godley, “Singer in Britain: The Diffusion of Sewing Machine
Technology and Its Impact on the Clothing Industry in the United Kingdom,
1860–1905,” Textile History 27 (Spring 1996), 62.
50. Quoted in Pamela Laird, Advertising Progress: American Business and the
Rise of Consumer Marketing (Baltimore, Md., 1998), 36.
51. McKenzie to Clark, 29 March 1879, 14 April 1879, Box 93/9; Woodruff’s
defense, Woodruff to Clark, 14 April 1879, Box 93/9. For an example of
Woodruff’s newspaper advertising material, see Oldham Chronicle, ca. 1880–81,
Box 94/1.

284 GODLEY
office to cut advertising spending from 4 percent (and even more
than 10 percent in France) to less than 2 percent of the retail price in
these key growth markets. As table 2 indicates, Woodruff followed
these new instructions. Shortly thereafter, McKenzie ensured that
New York took full control of advertising expenditure, and so it cor-
respondingly drops out as an expense item of the individual agency
accounts, disappearing from the archival record.52
This decline in the share of revenues going to advertising con-
trasts with other pioneering marketing-led firms of the era, like
Procter and Gamble, Coca-Cola, and Heinz. These firms did not
invest in retail structures and so were more dependent than Singer
on advertising to communicate directly with their target markets.
Evidence on advertising expenditure as a proportion of turnover is
scarce, but Coca-Cola’s advertising expenditure was 13–18 percent of
52. McKenzie to Clark, 29 March 1879, Box 93/9; Davies, Peacefully Working,
98–100. Bourne later formalized McKenzie’s fiat in the New York Advertising
Department, created in 1891. This provided bill posters and pamphlets, the pre-
ferred methods of promotion. Local agencies continued to exercise some discre-
tion over other types of advertising promotional activity, such as newspaper
advertisements and sewing classes, for example, but expenditure on these items was
generally discouraged. On sewing classes see “Business Conditions in Denmark,”
Red ‘S’ Review 2 (1920–21), 362–63; “Business Conditions in Switzerland,”
Red ‘S’ Review 2 (1920–21), 462–63; and “The Wellawatta Sewing School,” Red
‘S’ Review 12 (1930–31), 3347.
Table 2 Singer’s Advertising Expenditure in London Office, 1867–1884
Sources: For 1867–73, “Results of Business (London and Offices together), Annual 1867–73,”
Micro 2002, Reel 16, Segment 5 (gross sales, profits included), Singer Archives, State
Historical Society of Wisconsin, Madison, Wisconsin. For 1882, New York Department,
London, “Abstract of Business ... 1882, Great Britain Agency,” Box 94/6, Singer Archives,
which lists advertising expenditures of £12,625 and total sales of £507,934. For 1883 and 1884,
“Summary of Business ... 1883 and 1884, Great Britain,” Box 95/4, Singer Archives, lists
advertising expenditures of £7,539 and £7,886 and total sales of £548,958 and £586,122,
respectively, all converted at $4.86 to £1.00. No data are available for 1874 to 1881.
Year Expenditure ($) Income ($) Expenditure/Revenues (%)
1867 10,174 166,776 6.1
1868 12,990 263,957 4.9
1869 12,744 243,648 5.2
1870 12,703 245,529 5.2
1871 20,193 685,284 3.0
1872 13,828 900,065 1.5
1873 58,310 1,361,059 4.3
1882 61,538 2,468,559 2.5
1883 36,640 2,667,936 1.4
1884 38,326 2,848,553 1.3

Singer’s International Marketing Strategies 285
revenues in most years before the First World War. Elsewhere,
Wrigley’s $3.5 million spending in 1919 accounted for 13 percent of
total revenues that year.53 In fact, and despite the company’s reputa-
tion among advertising historians, advertising was not so important
to the company’s success in its overseas markets. As Robert Davies
states, “Despite the popularity of these advertising materials they
were always considered ancillary to a ‘wide awake canvasser’ in a
neat ‘Singer wagon, behind a good horse, with a Genuine Singer
Sewing Machine at his back.’”54
International Expansion of the Canvasser-Collector System
From 1877 onward Clark and McKenzie encouraged the extension of
Woodruff’s canvasser-collector system (and subsequently the revised
system, where all salesmen-collectors did their own canvassing)
throughout the rest of the world, first throughout the London agency
overseas branches, then the United States and Neidlinger’s Hamburg
agency. It took at least twenty years to achieve the objective.55 It led
to an enormous expansion in Singer’s worldwide sales force. Already
by 1882 the overseas sales agencies employed a 6,000-man strong
sales force. But by 1905 Singer employed 61,444 people in 4,552
branches throughout its global selling organization.56 This was
wholly exceptional for the time. Compared with the other leading
sales-led organizations in the world, Singer was simply in a different
league.
Henry Heinz, for example, is credited with being a pioneer in the
creation of a nationwide, company-controlled sales force in the
United States and worldwide. But Heinz employed a sales force of
only 450 in 1904. This did rise, but only to 952 in 1919, by which
53. Davis Dyer, Frederick Dalzell, and Rowena Olegario, Rising Tide: Lessons
from 165 Years of Brand Building at Proctor & Gamble (Cambridge, Mass., 2004),
35–40; Daniel Robinson, “Marketing Gum, Making Meanings: Wrigley in North
America, 1890–1930,” Enterprise & Society 5 (March 2004): 24; Richard Tedlow,
New and Improved: The Story of Mass Marketing in America (New York, 1990),
52, table 2-5. There are some outliers; 1897 advertising accounted for 22.9% of
Coca-Cola sales, for example. But for seventeen of the twenty-two years from 1892
to 1913 inclusive, the range was 13–18%.
54. Davies, Peacefully Working, 99–100.
55. McKenzie to Clark, 29 March 1879, Box 93/9; McKenzie to Woodruff, 16
April 1879, Box 93/9; McKenzie to New York, 11 May 1882, 27 May 1882, Micro
2013, Reel 63. See also “Synopsis of New Arrangement of Work in the London
Office,” Box 94/4; and the Agent Pro Forma, Box 94/4. On U.S. developments see
Dicke, Franchising in America, 41–42.
56. For 1882 see “Breaking Ground,” 22. For 1905 see Singer Sewing Machine
Company, Century of Sewing Service, 7.

286 GODLEY
time the company was selling 13 percent of its output outside the
United States. Another pioneer in developing a selling organization,
National Cash Register, increased its sales force from 128 in 1890 to
750 in 1910, with 38 percent of sales going overseas by then as
well.57 By contrast, by 1914 Singer employed 27,486 in its Russian
sales force alone.58 And as was the case in Britain, as the canvasser-
collector system spread, so expenditure on advertising fell.59 From
being the largest sewing machine company in the world in the 1870s,
but with less than a quarter of the global market, Singer grew to dom-
inate the industry during the 1880s. In 1882 McKenzie triumphantly
claimed that Singer was “in every country in the world in the
unquestioned leadership and control of the sewing machine
business.”60
The canvasser-collector selling organization was Singer’s princi-
pal innovation in this period, not any new machine or novel produc-
tion technology. And it was an innovation that competitors simply
failed to match. Even in Germany, where Singer faced the fiercest
competition, the German producers were forced to join together in
the 1880s and lobby for government protection against Singer to
stave off bankruptcy. Frister and Rossmann, the largest German pro-
ducer then, was making enormous losses in the early 1880s. G. M.
Pfaff put the blame squarely on Singer’s “army of canvassers [used
to] flood the German Empire.”61
These canvassing salesmen-collectors were the backbone of the
entire Singer selling organization. In Britain by 1904, for example,
57. Nancy Koehn, “Henry Heinz and Brand Creation in the Late Nineteenth
Century,” Business History Review 73 (Autumn 1999): 385, table 1; Friedman,
Birth of a Salesman, 123, 142.
58. Carstensen, American Enterprise in Foreign Markets, 69 (which is obvi-
ously preferable to Davies, Peacefully Working, 284, which states 25,000). This is
in addition to the 4,894 employees in Singer’s Russian factory in 1914;
Carstensen, American Enterprise in Foreign Markets, 75.
59. In Brussels from £1,582 (relative to total revenues of £38,490) in 1883 to
£1,029 (relative to £40,693) in 1884; in Geneva from £514 (£15,567) to £446
(£18,563); in Milan from £1,431 (£65,906) to £1,201 (£88,185) from 1882 to 1883,
for example. For 1882 figures see New York Department, London, “Abstract of
Business ... 1882,” Box 94/6. For 1883 and 1884 figures see “Summary of Business ...
1883 and 1884,” Box 95/4. For an informative recent synopsis of early advertising
developments in Europe see Véronique Pouillard, “American Advertising in
Europe: J. Walter Thompson’s Belgian Business in the Inter-War Years,” Business
History 47 (Jan. 2005): 44–58.
60. “Breaking Ground,” 22.
61. Press cuttings on the publicity war between Singer and the Kongress Deut-
scher Nähmaschinen Händler German competitors, Box 83/11; Neidlinger to
McKenzie, 19 May 1881, Box 84/1; Neidlinger to Clark, 19 May 1881, on problems
for Grimme and Natalis and for Frister and Rossmann, Box 84/1; Davies, Peace-
fully Working, 221–42 (quotation at p. 235).

Singer’s International Marketing Strategies 287
out of a total of 5,184 employees in the selling organization, 2,817 (or
54 percent) were salesmen and collectors.62 But the nature of the
canvasser-collector system changed yet further when it met new
problems outside the environment of its British birthplace.
As almost all other counties were less urbanized than Britain, the
expense of maintaining any separation along the lines of McKenzie’s
1879 principles could not be justified by sales in more rural areas. As
Singer’s penetration of the European market extended beyond the
cities, so-called country agents emerged as an adaptation to the
British system. These country salesmen took the customers’ deposits
immediately on getting their signed purchase-agreements. Afterward,
it was the same agents who delivered the machines, then subse-
quently collected all the payments, and then closed the accounts.63
By the mid-1890s at the latest, in other words, the underlying princi-
ple of monitoring agent opportunism through separating the creation
of a new account from at least one of the principal financial transac-
tions had been overcome by the imposition of discipline and self-
reporting among the sales force. In Russia, for example, half the
salesmen were country agents. In 1895 the seventy retail outlets
there supported “280 collectors and town travellers [and] 275 coun-
try travellers.”64 In 1914 almost half (8,860) of the 19,638 collectors
were still country collectors.65
Innovation in Pricing
It was not only Singer’s advertising strategy that changed in the face
of the canvasser-collector system. Singer’s pricing strategy changed
markedly from the price competition of the mid-1870s to unques-
tioned price leadership in the 1880s. After the expiry of the patent
pool in 1877, Singer had cut prices in the United States by half.
Similarly, in the leading overseas markets of Britain, France, and
Germany, the company responded to competitor price pressures in
the 1870s. Woodruff was “forced in 1876 to lower the price of
Singer’s basic machine to about $25,” for instance.66 By contrast,
from the 1880s onward Singer began to adopt a near-uniform price
structure in all its markets.
62. “Synopsis of Organisation, 1904,” Box 109/14. Also see “Synopsis of
[U.K.] Organisation, 1906,” Box 109/2; and “List of The Singer Manufacturing
Company’s Branches in GB and Ireland,” 1896 and 1900, Box 198.
63. “Conditions in Norway,” Red ‘S’ Review 4 (1922–23), 411, for example.
64. Maeder to Bourne, June 1895, Box 85/6.
65. Carstensen, American Enterprise in Foreign Markets, 69.
66. Dicke, Franchising in America, 41.

288 GODLEY
Within Europe the cash retail prices of the company’s standard
New Family model varied from a low of $25 in Britain to a high of
$30 in Spain, with France, Italy, Belgium, and others in between.67
The one outlier was Germany. As the other American producers
withdrew from export markets after 1886, Singer’s strongest compet-
itors were the leading German producers, whose prices were typi-
cally 35 percent less than Singer’s normal tariff. Pfaff, for instance,
was able to undercut Singer by 30–40% in Switzerland in the early
1890s, pricing their models under $20.68 Unsurprisingly, in Germany
Singer reduced its New Family price to $22.60, about 10 percent less
than the British price and 25 percent less than in Spain.69 Figure 3
suggests that Singer was eventually successful in moving even German
prices up to the European norm.
Figure 3 divides through the gross income from total sales in each
national market by the total number of sewing machines sold for
each year for which data exist to give a value of the mean income per
machine sold. With income from the sale of other products (twist
67. “Retail Prices in England and Foreign Offices, 31/12/1884,” Box 95/4,
using gold standard exchange rates.
68. On German prices see Davies, Peacefully Working, 229; on American
firms’ leaving export markets after 1886, see Davies, Peacefully Working, 88; on
Swiss pricing see L. Charrière, “Retail Prices in Switzerland,” 8 Jan. 1892, p. 2,
Box 83/5.
69. L. Charrière, “Retail Prices in Switzerland,” 8 Jan. 1892, Box 83/5.
Figure 3 Singer’s mean income per machine sold, 1880–1914 ($ per machine).
Source: Appendix, table A3 divided by table A2.

Singer’s International Marketing Strategies 289
and thread, for example) so small, the result is a reasonable proxy for
the mean price received by the company per machine sold in these
different markets.
Figure 3 shows that the mean income per machine sold across all
these markets was around $30–35 in the 1880s, rising to between $35
and $40 from 1902 to 1914. (Wartime inflation reduces the value of
continuing the comparison after 1914.) With the exception of
Australasia and the United States (where pricing decisions were dis-
torted by tariff barriers), the mean prices received per machine sold
remained fairly closely clustered together across these markets.70
There is perhaps a little more dispersion at the end of the period,
although the number of markets surveyed increases from ten in the
1880s to seventeen after 1907.
There was some product change over the period, most notably
when the Improved Family model was introduced in the mid-1880s
and gradually replaced the New Family, which may account for
some, possibly all, of the upward drift in mean income. The
Improved Family retailed for a cash price of $35–40, somewhat
higher than the $25–30 charged for the New Family machines.
Assuming that the mean income per machine sold across these mar-
kets is a fairly reasonable proxy for the mean price, figure 3 appears
to suggest that model prices were kept broadly constant over time.
This is not surprising, as New York senior management was eager to
establish a rigid price policy in all of Singer’s markets, for this
avoided one of the biggest managerial headaches associated with a
sales force, that of cutting prices to gain sales. Or as Woodruff suc-
cinctly described it, “large discounts mania.”71
Figure 3 also shows the approximate German competitor price
band as a constant $25. It is notable that, apart from the Philippines
for two years, in no other market were Singer machines comparably
priced. Furthermore, the German series suggests that from 1902 at
the latest, there were no longer any price concessions to competitors
there.
Singer was therefore able mostly to set and keep their prices at a
30 to 40 percent premium to the competition and yet, at the same
time, increase their global market share. Once again, this is a note-
worthy finding, for Singer was not targeting a particularly high-
income market but rather ordinary families, who would normally be
assumed to exhibit more than modest price sensitivity. Before trying
to resolve such perplexing consumer behavior, one further piece of
70. On U.S. tariffs see Davies, Peacefully Working, 88, 161–62.
71. Woodruff to Clark, 19 May 1877, Box 93/9 (emphasis in original).

290 GODLEY
evidence is required: the cost of selling sewing machines through the
canvasser-collector system.
The Expense of the Canvasser-Collector Distribution System
Figure 4 shows the retailing expenses from table A4 in the appendix
divided through by the annual sales income from table A3 to show
the cost of distributing a sewing machine as a share of gross income
in these overseas markets.
These retail expenses included the costs incurred by the selling
organization within its market. Simply deducting these retail
expenses from gross income would not, however, provide a straight-
forward profit indicator. They did not include any selling costs
incurred centrally, like advertising from 1885 onward, for example.
Nor did they include the consignment price. From 1882 onward
Singer operated a strict separation between its manufacturing and
selling organization, adopting a “uniform tariff” system. The facto-
ries transferred machines to the selling organizations for a standard
consignment price.72 The retail expenses included therefore cover
the salaries of all the employees involved in the selling organiza-
tions, the costs associated with keeping retail outlets (and associated
warehouses) open, travel costs, and any other incidental expenses.
Regardless of this being only a partial measure, figure 4 confirms
that Singer’s retail network was very expensive, with well over half
of the gross income routinely going to sustain the selling organiza-
tion. In only one year, 1904, was the simple mean of the share of
retail expenses across all these markets less than 50 percent of gross
income, and over the period as a whole the costs of selling accounted
for 58 percent of income. Once centrally funded advertising expendi-
ture is included, the costs of distribution would have summed to
around 60 percent of income over the period. Compared with almost
any other channel of distribution, this was very expensive.73
Senior management was well aware that Singer had a relatively
high cost structure. Woodruff complained that the German competitors
72. “Synopsis of the New Arrangement of Work at the London Office,” 20
April 1882, Box 94/4.
73. James B. Jeffreys, The Distribution of Consumer Goods (Cambridge, U.K.,
1950), 61–84, suggests that in Britain in the 1930s only newspapers and contra-
ceptives were as expensive to distribute. And in trades similarly dominated by
retail chains—tobacco and groceries, or menswear and footwear, for example—
sales costs were much lower. Interestingly, in the early 1960s Tupperware reck-
oned that it cost close to 60% of revenues to distribute their plastic ware through
its direct sales force. See “The Overseas Boom in Door-to-Door Selling,” Dun’s
Review (Nov. 1964), 85.

Singer’s International Marketing Strategies 291
were able to price their machines “at almost our factory prices.”74
Singer’s factory prices were not even that cheap, with the consign-
ment price in the early 1880s close to 50 percent of gross income.75
The implication seems to be that while its factories were profitable,
its selling organizations tended rather to expand to maximize sales
rather than profits.
Further examination of figure 4 suggests that Singer’s overseas
expansion strategy no longer tolerated the loss-leading expansion of
retail markets. This was what had happened for a number of years in
the late 1870s, in the aftermath of the demise of the patent pool,
when profits in European markets were used to subsidize loss-leading
expansion of the American retail network. The strategy ultimately
worked there, as the company’s market share in the United States
subsequently increased to 60 percent. If Singer had tried to squeeze
74. Quoted in Davies, Peacefully Working, 40.
75. Total sales at consignment prices in Britain in 1883 and 1884 were 53.7%
and 53.6% of “the amount at which sold,” respectively. See “Comparative State-
ment of Results of Business at Gt. Britain, 31.12.1884,” Box 95/4. Note that the
factory-cost price of machines sold in Britain in 1883 was £1.79, implying a gross
profit margin of almost 50% out of a mean consignment price of £3.47, or 28% out
of a mean final selling price of £6.46 ($31.40). See “London, New York Depart-
ment, 31.12.1883,” Box 95/1.
Figure 4 Singer’s retail expenses as a proportion of gross income, 1881–1914 (%).
Source: Appendix, table A4 divided by table A3.

292 GODLEY
out competitors elsewhere in this manner, such loss-making activity
would show up on figure 4 as markets where selling expenses
exceeded income.
This did occasionally happen. The opening of the Geneva office
led to relatively high costs in Switzerland in 1881. Australian costs
of distribution were especially high in 1882, when the Sydney office
opened. Similarly, new offices meant Italian and South African costs
of distribution jumped by more than income in 1883 and 1884,
respectively. Each of these occasions was associated with a key event
in developing a new and emerging market. But each of these occa-
sions was isolated. Any overshoot of retail expenses was of limited
duration and seemingly curtailed well before Singer might have been
able to see off competitors. Even when costs in the Philippines
soared relative to income in 1911, or in Japan in 1912, they were
brought back into line the very next year. This was no loss-leading
expansion of retail structures. Indeed in Russia, the emerging market
where Singer succeeded beyond all, the selling organization broke
even throughout.
The more obvious results from figure 4 are rather of a relative uni-
formity in the share of total income spent on retail expenses, and at a
relatively constant level over time. Only one market appears to have
experienced a persistent divergence from the mean, and that was
India, where the costs of selling remained relatively low from 1889 to
1892 and then again from 1902 to 1911 (although not in the early
1880s, nor from 1912).76 None of the more established markets expe-
rienced any persistent below-average retail expenses as a share of
revenues (as might be expected if they had been subsidizing loss-
leading expansion in new and emerging markets). Indeed, with only
a very few exceptions and for a very few years, figure 4 shows some
very considerable clustering, especially after 1890, suggesting that
any variations in local selling practices had little impact on retail
expenses.
Singer’s success in international markets was squarely based on
its selling organization. And in selling, uniformity was one of
Singer’s key competitive advantages. This was because, as Davies
states, “It was an organization whose rules and regulations reduced
the complexities of the sewing machine business to standardized
practices, uniformly and universally applied in an equitable manner
for the employees and customers alike.” It was this uniformity in the
selling organization that Bourne believed to be “our strongest hold
76. The Indian sales agency uniquely adopted a very different selling organi-
zation. See Davies, Peacefully Working, 174–87.

Singer’s International Marketing Strategies 293
commercially.” It was the “simplicity of the rules and regulations ...
[that] enabled thousands of nationals of all cultures, many of whom
were previously unacquainted with American business procedures,
to be thoroughly grounded in the fundamentals of the system.”77
As it evolved from the original canvasser-collector innovation,
Singer’s international selling organization became, in other words, a
triumph of rules and procedures over ad hoc and informal national
practices. There were of course deviations from the norm, as the fig-
ures show. But these were variations on a theme. The attempt to
impose a broad uniformity in pricing and in retail expenses was con-
sistent with the grand strategy of reducing operational autonomy
within the selling organization. But while uniformity of pricing and
in the different national selling organizations helped senior manage-
ment to economize on decision making, what is not so clear from the
evidence is why Singer should have been so successful selling their
high-priced machines through its expensive retail organization.
Singer’s Global Retailing Strategy
Classic marketing theory emphasizes the “four P’s” of product, price,
place, and promotion. As we have seen, Singer’s exceptional success
in international markets was not accounted for by price competition,
product innovation, or through the standard promotional activities
of advertising. Nor was it through creating barriers to entry through
loss-leading expansion of its selling organization. It was rather based
on constructing a high-cost selling organization that sold expensive
sewing machines in vast, vast quantities to consumers who were oth-
erwise (it is reasonable to assume) normally price sensitive. In order
to understand why Singer was so successful in selling its high-priced
machines, full account needs first to be taken of the characteristics of
the late nineteenth-century demand for sewing machines.
The sewing machine was the first complex consumer durable sold
so widely around the world. It held two important characteristics.
First, it was a nonrepeatable purchase, and second, it was a complex
good—not complex by the standards of late nineteenth-century engi-
neering, but it was far more complex than the typical consumer of
sewing machines was able to comprehend. These two features had
important implications for the implicit contract between seller and
buyer.
77. Davies, Peacefully Working, 113 (quotation), 335–36; Bourne quoted in
Davies, Peacefully Working, 114.

294 GODLEY
First, the sewing machine was a nonrepeatable purchase. Unlike
other goods where consumers can try different brands and so make
their own assessments over time about each brand’s qualities, once a
family bought a sewing machine they kept it. For them to be able to
make the right choice depended on their acquiring trustworthy infor-
mation equivalent to that based on personal experience. Above and
beyond word-of-mouth recommendation by friends, for goods like
sewing machines that meant knowledge of the manufacturer’s
reputation.78
Singer went to great lengths to safeguard the company’s reputa-
tion. There are frequent references in the senior management letters
to the company’s standing.79 Because its sales force was the principal
form of contact the consumer had with the firm, senior management
did its utmost in ensuring that the sales organization would always
enhance the company’s reputation. The recruitment of suitable can-
vassers and collectors, their subsequent monitoring and reward exer-
cised the company’s senior management from the 1870s onward.
Indeed, Woodruff’s canvasser-collector reforms were initially con-
ceived of as a device to instill greater discipline in the sales force
rather than generate greater sales.80 If consumers trusted the Singer
brand through their interaction with the sales force, they would be
more likely to purchase its machines, despite the price premium.
Second, the sewing machine was complex. This meant that sales
were dependent on two areas above and beyond product price and
quality: demonstration and after-sales service. To reduce the uncer-
tainty associated with the first purchase of a family sewing machine,
sellers needed to demonstrate to consumers how the machine
worked. Furthermore, given the durable nature of the machine, the
probability of a consumer requiring subsequent, postpurchase dem-
onstration was high. Allied to this was the reasonable probability of a
machine’s requiring servicing and repair at some stage. Moreover,
there was a near certainty of requiring additional thread, oil, and
parts. In many of Singer’s growing markets before 1914, the existing
commercial infrastructure may not have been sufficiently well devel-
oped for these services to be provided by independent retailers. Con-
sumers of this complex durable good were therefore engaging in an
78. The economics of asymmetric information is well developed. This insight
is developed from Oliver Williamson, “The Modern Corporation: Origins, Evolu-
tion, Attributes,” Journal of Economic Literature 19 (Dec. 1981): 1537–68.
79. For example, see the correspondence surrounding the German trademark
dispute, summarized in Davies, Peacefully Working, 106–7; and throughout
“Breaking Ground.”
80. Davies, Peacefully Working, 62.

Singer’s International Marketing Strategies 295
implicit contract with its supplier, requiring before- and after-sales
market support services over a period of time in addition to the
actual product.
Given the durable nature of this implicit contract between buyer and
seller, Singer’s pioneering of mass consumer credit in overseas markets
takes on a more significant light. Contemporaries were amazed, often
appalled, at why working-class families would pay high interest rates
for goods on credit.81 As far as sewing machines were concerned the
answer was simple. A typical repayment period of up to two years
locked the vendor into a relationship with the purchaser, as well as the
other way round. After-sales service was therefore readily available.
Singer’s senior management clearly understood this. The head of
Singer’s U.K. agency explained the benefits of the Singer selling orga-
nization for consumers and compared them with distributing via
independent retailers:
The [independent] Retailer only sells the goods. He rarely has an
organization capable of looking after the article when sold or ren-
dering service in connection with it. This may not matter much
when the article is of a perishable nature or is not a mechanical
nature, but with machinery it is a fatal defect.
The case of a Singer Selling Organization is entirely different. They
deal direct with the Consumer.... It is as a Maintenance Organiza-
tion, however, that the Singer Companies achieve their greatest
value. The Consumer who purchases a Singer Sewing Machine at
the same time becomes entitled to the service of the Singer Mainte-
nance Organizations throughout the world. The fact of ownership
of a Singer Sewing Machine entitles him to that. No matter where
he may go he will be able to get parts for his machine, and needles,
oil, etc. He will find teachers ready to instruct him in the best
methods of using his machine and mechanics to repair and adjust
it. It makes no difference where he may have bought the machine—
in Europe, Asia, Africa, or America.82
The canvassers and collectors, or later the salesmen and sales-
women, were the key people in the retail organization because they
provided the market support services in customers’ homes or took
away and brought back machines needing repair. They represented
81. Paul A. Johnson, Saving and Spending: The Working Class-Economy in
Britain, 1870–1939 (Oxford, U.K., 1985); Calder, Financing the American Dream,
chap. 4.
82. “Commerce,” Red ‘S’ Review 1 (1919–20), 26–27, by “Number One” (a
pseudonym for the head of the London sales agency).

296 GODLEY
the company and so were directly responsible for the company’s rep-
utation with customers. The vast retail organization, with its shops
on every High Street in almost every town in the world, comple-
mented their role. As the leading commercial presence in many parts
of the rapidly developing world, Singer’s stores represented moder-
nity and progress to consumers.
The emphasis on the original canvasser-collector system and its
descendants is warranted because this was the key innovation. It
explains how Singer was able to build its brand and so to tap into a
latent demand for sewing machines around the world. For the over-
whelming majority of sewing machine consumers, women in charge of
family homes, it was the ability of reputable salesmen to demonstrate
and provide after-sales service in the home that proved crucial to
Singer’s success, and so overcame the relative expense of its machines.
In other words, it was the creation of this remarkable market support
system, incorporated within the direct selling organization and grafted
on to the worldwide retail presence, that enabled Singer to gain the
trust of consumers around the world. This allowed the company to
build what was surely the most valuable global brand of the period.
Conclusion
During the first globalization boom, the sewing machine was dif-
fused more widely than any other consumer durable, and far more
than previously realized. It was the antecedent of the motor car, the
radio, the television, and other modern consumer durable goods. As
the new data presented here confirm, one firm, Singer, was able to
capture almost all of this global demand. This was not because of a
competitive price, or product innovation, or advertising campaigns,
or loss-leading expansion of the retail network.83 Rather, Singer’s
extraordinary success in foreign markets was more due to the pecu-
liarities of the demand conditions surrounding the nonrepeatable
purchase of a complex consumer durable like a sewing machine. In
particular, the combination of the installment purchase contract and
the regular presence of salesmen meant that sewing machine con-
sumers, women in their homes, were able to overcome the inherent
uncertainties associated with such a purchase.
The key innovation was the original canvasser-collector system in
Britain in the mid-1870s. The American selling organization, a pioneer
83. Contrast this with Davies, Peacefully Working, 325–27, with his emphasis
on product and advertising, in addition to the selling organization.

Singer’s International Marketing Strategies 297
of vertical integration, was struggling by then to cope with a down-
turn in demand. Indeed, without the profits from European sales sub-
sidizing its domestic sales network, it is surely unlikely that Singer
would have continued with its inefficient U.S. retail organization,
never mind embark on the program of domestic expansion it
launched after 1877. It was the British selling model that allowed the
company to expand overseas.84
Because Singer made the first investments in direct selling organi-
zations in its overseas markets, it was Singer that minimized the
costs to women sampling this complex product before purchasing.
Because Singer made the investments in its selling organization, it
did not need to make the kind of investments in advertising seen in
other cases of marketing pioneers at the time. Because Singer made
strenuous efforts to ensure that its salesmen behaved with utmost
propriety, and that its maintenance organization met customer
expectations, it was Singer that reaped enormous reputational advan-
tages. It was the firm’s investments in these market support services
that enabled it to develop the enormous competitive advantages it
came to enjoy. The company’s reward was to be almost the only
recipients of a fast-growing mass consumer demand around the world.
Competitors were simply unable to match this investment in an
international selling organization. By way of conclusion, the final
irony may be that Singer’s great contribution to the development of
international marketing, the canvasser-collector innovation of the
1870s and the subsequent creation of a worldwide selling organiza-
tion, also contained within it the seeds of its own downfall. Such an
expensive distribution system was possible when Singer’s advan-
tages persisted. But as commercial infrastructures and retailing
developed around the world, Singer no longer held advantages in the
supply of credit or of thread, needles, and parts. Department stores
and other outlets took on the demonstration and after-sales service
function. Without its key advantages, Singer’s expensive selling orga-
nization caused the company’s trading position to deteriorate sharply
in the 1950s, something from which it never really recovered.85
84. Naomi R. Lamoreux, Daniel M. G. Raff, and Peter Temin’s use of Singer as
an example of successful forward integration by a manufacturer into American
retailing is, in other words, factually incorrect. See their “Beyond Markets and
Hierarchies: Toward a New Synthesis of American Business History,” American
Historical Review 108 (April 2003): 418.
85. Michael McDermott, in “Singer: Competition and Closure,” in The History
of Clydebank, comp. John Hood (Glasgow, 1988), 151, 165, states that the “cost of ...
sales-service centres was ... 50%” in the 1950s.

298 GODLEY
Bibliography of Works Cited
Books
Bairoch, Paul. Cities and Economic Development. London, 1988.
Bayly, Christopher A. The Birth of the Modern World, 1780–1914: Global
Connections and Comparisons. Oxford, U.K., 2004.
Brandon, Ruth. A Capitalist Romance: Singer and the Sewing Machine.
Philadelphia, 1977.
Calder, Lendol. Financing the American Dream: A Cultural History of Credit.
Princeton, N.J., 1998.
Carstensen, Fred. American Enterprise in Foreign Markets. Chapel Hill, N.C.,
1984.
Chandler, Alfred D., Jr. The Visible Hand: The Managerial Revolution in
American Business. Cambridge, Mass., 1977.
Davies, Robert. Peacefully Working to Conquer the World. New York,
1976.
Dicke, Thomas S. Franchising in America: The Development of a Business
Method. Chapel Hill, N.C., 1992.
Dyer, Davis, Frederick Dalzell, and Rowena Olegario. Rising Tide: Lessons
from 165 Years of Brand Building at Proctor & Gamble. Cambridge, Mass.,
2004.
Foreman-Peck, James. A History of the World Economy: International
Economic Relations since 1850. Hemel Hempstead, U.K., 1995.
Friedman, Walter. Birth of a Salesman: The Transformation of Selling in
America. Cambridge, Mass., 2004.
Hounshell, David. From the American System to Mass Production, 1800–
1932. Baltimore, Md., 1984.
Jeffreys, James B. The Distribution of Consumer Goods. Cambridge, U.K.,
1950.
Jobber, David. Principles and Practice of Marketing. 4th ed. Maidenhead,
U.K., 2004.
Johnson, Paul A. Saving and Spending: The Working-Class Economy in
Britain, 1870–1939. Oxford, U.K., 1985.
Laird, Pamela. Advertising Progress: American Business and the Rise of
Consumer Marketing. Baltimore, Md., 1998.
Mitchell, Brian R. British Historical Statistics. Cambridge, U.K., 1988.
———. International Historical Statistics: Africa, Asia, and Oceania.
Basingstoke, U.K., 2003.
———. International Historical Statistics: Americas, 1750–2000. Basingstoke,
U.K., 2003.
———. International Historical Statistics: Europe, 1750–2000. Basingstoke,
U.K., 2003.
Tedlow, Richard. New and Improved: The Story of Mass Marketing in America.
New York, 1990.
Thomson, Ross. The Path to Mechanized Shoe Production in the United
States. Chapel Hill, N.C., 1989.
Wilkins, Mira. The Emergence of Multinational Enterprise: American Business
Abroad from the Colonial Era to 1914. Cambridge, Mass., 1970.

Singer’s International Marketing Strategies 299
Articles and Essays
Bourne, Frederick G. “American Sewing Machines.” In One Hundred Years
of American Commerce, ed. Chauncey M. Depew. 1895; New York, 1968,
p. 530.
Bowden, Sue, and Avner Offer. “Household Appliances and the Use of Time:
The United States and Britain since the 1920s.” Economic History Review
47 (Nov. 1994): 725–48.
Coffin, Judith. “Consumption, Production, and Gender: The Sewing
Machine in Nineteenth-Century France.” In Gender and Class in
Modern Europe, ed. Laura Frader and Sonya Rose. Cornell, N.Y., 1996,
pp. 111–41.
———. “Credit, Consumption, and Images of Women’s Desires: Selling the
Sewing Machine in Late Nineteenth-Century France.” French Historical
Studies 18 (Spring 1994): 749–83.
Davies, Robert. “‘Peacefully Working to Conquer the World’: The Singer
Manufacturing Company in Foreign Markets, 1854–1889.” Business His-
tory Review 43 (Autumn 1969): 299–346.
Fletcher, Scott, and Andrew Godley. “Foreign Direct Investment in British
Retailing, 1850–1962.” Business History 42 (April 2000): 43–62.
Godley, Andrew. “Foreign Multinationals and Innovation in British Retail-
ing: 1850–1962.” Business History 45 (Jan. 2003): 80–100.
———. “The Global Diffusion of the Sewing Machine, 1850–1914.” Research
in Economic History 20 (2001): 1–46.
———. “Pioneering Foreign Direct Investment in British Manufacturing.”
Business History Review 73 (Autumn 1999): 394–429.
———. “Singer in Britain: The Diffusion of Sewing Machine Technology and
Its Impact on the Clothing Industry in the United Kingdom, 1860–1905.”
Textile History 27 (Spring 1996): 59–76.
Jack, Andrew. “The Channels of Distribution for the Innovation: The Sewing
Machine Industry in America.” Explorations in Entrepreneurial History 9
(Feb. 1957): 113–41.
Koehn, Nancy. “Henry Heinz and Brand Creation in the Late Nineteenth Cen-
tury.” Business History Review 73 (Autumn 1999): 349–93.
Kuwahara, Tetsuya. “Early Foreign Multinationals in Japan and the Indige-
nous Firms: Singer Sewing Machine and Japanese Enterprises, 1901–
1960s” [in Japanese]. Kokumin Keizai Zasshi [Journal of Economics and
Business Administration] 185 (May 2002): 45–63.
Lamoreux, Naomi R., Daniel M. G. Raff, and Peter Temin. “Beyond Markets
and Hierarchies: Toward a New Synthesis of American Business History.”
American Historical Review 108 (April 2003): 404–33.
McDermott, Michael. “Singer: Competition and Closure.” In The History of
Clydebank, comp. John Hood. Glasgow, 1988, pp. 151–66.
Oddy, Nicholas. “A Beautiful Ornament in the Parlour or Boudoir: The
Domestication of the Sewing Machine.” In The Culture of Sewing: Gender,
Consumption, and Home Dressmaking, ed. Barbara Burman. Oxford, U.K.,
1999, pp. 269–84.
“The Overseas Boom in Door-to-Door Selling.” Dun’s Review (Nov. 1964), 85.

300 GODLEY
Özmucur, Süleyman, and Sevket Pamuk. “Real Wages and Standards of Liv-
ing in the Ottoman Empire, 1489–1914.” Journal of Economic History 62
(June 2002): 293–321.
Page Fernandez, Nancy. “Creating Consumers: Gender, Class, and the
Family Sewing Machine.” In The Culture of Sewing: Gender, Consumption
and Home Dressmaking, ed. Barbara Burman. Oxford, U.K., 1999, pp.
157–68.
Peyrière, Monique. “L’industrie de la machine à coudre en France, 1830–1914.”
In La révolution des aiguilles. Habiller les Francais et les Américains,
19e–20e siècles, ed. L’École des Hautes Études en Sciences Sociales. Paris,
1996, pp. 95–114.
Pouillard, Véronique. “American Advertising Agencies in Europe: J. Walter
Thompson’s Belgian Business in the Inter-War Years.” Business History
47 (Jan. 2005): 44–58.
Puttnam, Tim. “The Sewing Machine Comes Home.” In The Culture of Sewing:
Gender, Consumption, and Home Dressmaking, ed. Barbara Burman.
Oxford, U.K., 1999, pp. 285–302.
Robinson, Daniel. “Marketing Gum, Making Meanings: Wrigley in North
America, 1890–1930.” Enterprise & Society 5 (March 2004): 4–44.
Schmitz, Christopher. “The World’s Largest Industrial Companies of 1912.”
Business History 37 (Oct. 1995): 85–96.
Williamson, Jeffrey. “The Evolution of Global Labor Markets since 1830:
Background Evidence and Hypotheses.” Explorations in Economic His-
tory 32 (April 1995): 141–96.
Williamson, Oliver. “The Modern Corporation: Origins, Evolution,
Attributes.” Journal of Economic Literature 19 (Dec. 1981): 1537–68.
Unpublished Sources
Carstensen, Fred. “American Multinational Corporations in Imperial Russia:
Chapters on Foreign Enterprise and Russian Economic Development.”
Ph.D. diss., Yale University, 1976.
Archival Sources
Clydebank District Library Archives, Clydebank, West Dunbartonshire, Scotland.
Singer Archives, State Historical Society of Wisconsin, Madison, Wisconsin.

Singer’s International Marketing Strategies 301
Appendix
Table A1 List of Abbreviations Used for Singer’s
Overseas Sales Markets
Abbreviation Market
A/NZ Australia and New Zealand
BBelgium
CH Switzerland
DAH Germany and Austria-Hungary
FFrance
GB Great Britain
IND India
IRE Ireland
ITA Italy
JAP Japan
NL The Netherlands
O&B The Ottoman Empire and Balkans
PHI Philippines
PPortugal
RUS Russia
SA South Africa
SCA Scandinavia
SP Spain
SP&P Spain and Portugal
UK United Kingdom
USA United States

Table A2 Singer Sewing Machines Sales in Selected Countries, 1875–1920 (machines per annum)
Year USA GB IRE F B CH NL SP P ITA RUS SCA DAH O&B A/NZ SA IND PHI JAP World Sales
1875 28500 1800 11000 5000 1000 43000 249,852
1876 33650 2150 12000 1000 6000 1500 50000 20 262,316
1877 38800 2500 15000 2000 500 1000 8000 2000 3000 6800 900 50300 2000 300 30 282812
1878 44000 2800 15000 3000 500 1200 12000 2500 5000 12500 1200 53100 3000 500 70 356432
1879 189780 49200 3100 17500 4000 1000 1800 16000 3000 7000 18800 1500 54900 4000 700 130 431167
1880 58502 3750 22500 5064 2102 2000 19205 3961 7500 22300 1800 80200 70 4550 750 250 538609
1881 63588 3829 24842 5239 2700 2650 24160 5521 8981 27262 2650 90100 200 9574 1354 552 561036
1882 75321 4494 26144 5409 2424 5600 18747 5500 10001 31600 5700 107300 400 18136 812 515 603292
1883 81402 4534 26375 5567 2418 6700 20847 4638 12514 35400 6800 114200 1000 18991 760 1088 278 600000
1884 85767 5644 25695 5667 3087 8400 21349 3708 12920 36200 8600 128200 2000 17801 1203 1607 674 500000
1885 83454 5278 28315 6064 3354 8520 17900 3768 12486 41200 9400 85000 2000 13546 1767 1761 494 500000
1886 77277 5155 28900 6200 3457 8640 19285 3894 11900 31200 10200 70000 3000 12316 1950 2012 600 500000
1887 85340 5877 29500 6350 3284 8760 19369 3911 11350 32300 11000 70000 2000 13158 2140 2131 700 500000
1888 92525 6022 30013 6492 3020 8880 21406 4322 10806 52200 11800 70000 1000 13000 2327 2966 782 500000
1889 119925 7341 35096 6765 2965 9000 23927 4831 15726 48100 12600 75000 3400 13000 3170 3578 1220 625000
1890 140870 9138 40091 7807 3328 9120 26590 5065 18010 42000 13400 89300 5400 13000 3627 3763 1195 750000
1891 135573 10328 42891 7470 3704 9240 29165 5562 20001 40700 14200 83600 5800 13000 3473 4145 1074 769000
1892 136179 10382 45263 7185 4116 9360 28213 6257 24592 46100 14922 87000 7000 13000 4362 4824 970 788000
1893 128301 10288 48178 8167 4314 9480 23500 6500 21518 56200 14800 96700 8500 13081 4300 5770 1100 806000
1894 136257 10195 51751 8339 4974 9600 24000 6750 22230 65000 14718 98373 11000 12877 4300 6720 1240 825000
1895 133233 10101 52595 8837 5271 9720 24000 7000 22755 68788 17700 118541 13500 13420 5300 7670 1380 845000
1896 124690 8752 56132 9785 5831 9850 24000 7250 27680 72000 19750 127056 14800 14654 6300 8620 1520 863000
1897 123688 9720 58342 10101 6433 10500 20500 7500 34650 85910 19850 133937 16000 15331 7300 9570 1660 881000
1898 129080 10687 58042 10507 7002 11200 22734 7956 34503 92369 19850 144983 17800 18000 8300 10500 1800 900000
1899 152764 12785 68330 11333 7425 11900 31038 8696 37634 99444 19850 157565 19000 20670 9300 11470 1940 990000
1900 153571 11409 72349 11473 7430 12500 32331 9986 37041 110316 19850 164798 20500 23340 10300 12420 2080 1080000
1901 148926 6432 74987 11642 7440 13209 36083 8965 33944 128249 19998 165394 23448 26010 11300 13352 2220 1170000
1902 321959 142682 8254 82048 11171 7449 13453 45243 10210 40154 132266 20281 172969 30069 28673 12274 13351 2357 1183607
1903 356920 127696 8926 86973 11241 7666 14612 48694 11135 43435 182668 22920 190345 37729 25432 14672 15842 1937 6529 1329886
1904 358601 120692 9598 88936 11842 7602 16095 45092 10864 49970 202592 26319 201986 46042 27558 11196 17781 1350 7619 1399991
1905 395816 132389 10270 93528 12446 8373 17714 39916 10460 51894 310881 29799 224022 46187 26066 12220 18075 1612 9348 1583013
1906 469199 156587 12150 98772 12938 8907 21009 43449 10647 56325 376336 36005 257671 52706 26965 10844 17331 1435 12895 1850489
1907 457216 152498 11478 107514 13675 9264 21124 44313 10705 67219 413905 39935 258749 53900 29593 8662 20478 2067 12421 1933194

Notes and Sources: For the United States—1879: “Results from Branches, 1879, ” P92-8970, Micro 2002 (10); 1902–15: “World Reports, Annual,” Micro AP93-0444, unprocessed
accession, Singer Archives, State Historical Society of Wisconsin, Madison, Wisconsin.
For Great Britain and Ireland, France, Belgium, Switzerland, Spain, Portugal, Italy, Russia, the Ottoman Empire and Balkans, Australia and New Zealand, South Africa, India, and
the Philippines, 1880 to 1914: Andrew Godley, “The Global Diffusion of the Sewing Machine, 1850–1914,” Research in Economic History 20 (2001): 1–46, tables 2–4, and sources
there (with exceptions being amendments to Australia for1881 [from 8,131 revised to 9,574] because of the inadvertent omission of the Sydney office sales in Godley, “Global Diffu-
sion,” and to 1888–92 inclusive [from 13,200 revised to 13,000]). This is another straight-line interpolation but revised down slightly in light of Australia’s macroeconomic crisis from
1887 onwards. See James Foreman-Peck, A History of the World Economy: International Economic Relations since 1850 (Hemel Hempstead, U.K., 1995), 101. For 1915–20: “World
Reports, Annual,” Micro AP93-0444, unprocessed accession, Singer Archives.
For the Hamburg region (less Russia and the Ottoman and Balkans) the estimates have been revised significantly for the years ca. 1883–95. Godley, “Global Diffusion,” 43, states
that the estimates there for 1886–90 for Germany “are a very clear underestimate.” The correction here is based on rereading Neidlinger to McKenzie (8 Dec. 1887, Box 84/4, Singer
Archives), where Neidlinger states, “Requirements of m/es for Germany solely for one year [i.e.1888]” [my italics] are over 70,000. This had previously been interpreted to mean the
entire Hamburg region, which (once the known Russian sales were deducted) left only an improbably small balance of 29,400 and 9,300 machines sold throughout Germany and
Austria-Hungary in 1887 and 1888, respectively. This is now interpreted to mean all of Germany and Austria-Hungary (what became the Hamburg territory from 1894). This more
muted dip in German and Austria-Hungary sales is supported by Pfaff production data for the same years, one of Singer’s leading competitors in Germany and Austria-Hungary. Pfaff
output also fell but only by 50% from 1882 to 1886 and 1887 (9,100 to 4,500 machines per annum). Singer sales in the Hamburg region were 182,000 in 1884 (Neidlinger to McKenzie,
19 Feb. 1884, Box 84/2, Singer Archives). Deducting the Russian total of 36,200 (Maeder to Bourne, 14 June 1895, Box 85/6, Singer Archives) leaves 145,800 for the balance of the
Hamburg region. Plotting Pfaff output onto Singer’s German and Austria-Hungary sales suggests Singer had sales of 128,200 in 1884, meaning that 70,000 in 1887 and 1888 represents
almost 50% of the 1884 estimate. The following table summarizes the changes:
Table A2 (continued)
Year USA GB IRE F B CH NL SP P ITA RUS SCA DAH O&B A/NZ SA IND PHI JAP World Sales
1908 357823 119408 8988 110899 13561 8342 19612 43702 10568 71724 446151 41366 254091 51254 29194 6796 20159 2241 10785 1781590
1909 412700 126805 9544 115129 14862 8477 19048 43918 11367 70444 481796 43327 254048 54463 29369 7781 20873 2642 9232 1901430
1910 421971 121575 9151 112303 15655 8522 19559 52020 11613 78182 505066 45675 262415 66757 31546 10523 24855 4120 14527 2019217
1911 412519 120131 9042 111566 16535 8394 20347 61229 13732 94308 504489 44738 268082 84623 34088 11888 26582 14366 20820 2109600
1912 401172 128018 9636 123833 18330 8790 22693 75413 16491 110792 570371 45921 278909 88509 30527 12944 36481 17714 33894 2326956
1913 132084 9942 126320 19288 7832 23814 72580 16654 110665 675174 47815 275367 78415 29458 14486 44819 27266 24038 2510652
1914 128994 9709 114000 17350 7050 21500 67201 15420 97817 678986 43000 248000 78000 26582 13250 41437 16520 19404 2185104
1915 354656 64844 3177 24883 51300 21500 11510
1916 139329 10487 67406 15103 26886 29987 19203 61296 26397 20976 1725978
1917 143220 10780 35600 30305
1918 160669 12093 44290 21016 21002 24451 23291 43058 44820 50067 1348867
1919 195919 14747 62482 23785 41214 30384 22234 56731 36047 58086 1758353
1920 180410 13579 60806 13658 33477 28659 19479 53732 33656 45491 1507363

Table A2 (continued)
Further plotting Pfaff output trends onto Singer’s German and Austria-Hungary sales suggests sales of 85,000 in 1885, 70,000 in 1886, before climbing up to 75,000 in 1889. For
Pfaff see http://www.ismacs.net/pfaff/pfaffdate.html (accessed 28 Feb. 2006).
Because Godley, “Global Diffusion,” simply estimates Netherlands and Scandinavia sales as shares of this now erroneous Hamburg region balance for these years, another
method is needed to provide an estimate. Their relative share in 1882 is calculated from “Schedule B—Hamburg,” P92-8970, Micro 2002, Reel 17, p.237, Singer Archives. The
Dutch and the Scandinavian sales figures begin from 1892 (“Sales in Sweden and Norway 1892 and 1894,” Box 85/6; “Statement of Fire Risk under Hamburg Office ... 1895/6,”
Box 85/8; both in the Singer Archives). So Scandinavian and Dutch sales from 1882 to 1891–95 are rounded straight-line interpolations of their shares of the Hamburg total.
Because of the large upward revision to the German estimates, it follows that there is a substantial increase for the Dutch and Scandinavian estimates from 1884 to 1891–95.
Otherwise, all figures remain the same.
For Japan: Tetsuya Kuwahara, “Early Foreign Multinationals in Japan and the Indigenous Firms: Singer Sewing Machine and Japanese Enterprises, 1901–1960s” [in Japanese],
Kokumin Keizai Zasshi [Journal of Economics and Business Administration] 185 (May 2002): 50, table 2. My thanks to Professor Andrew Gordon for sending me a copy of the
original source.
For World Sales, 1875–1902: Godley, “Pioneering Foreign Direct Investment in British Manufacturing,” Business History Review 73 (Autumn 1999): 394–429, table 2; and for
1902, “World Reports, Annual,” Micro AP93-0444, unprocessed accession, Singer Archives.
Year Hamburg Old Germany AH Old Hamburg New Germany AH New Pfaff Output NL Old NL New SCA Old SCA New
1883 7000 2000 6700 7200 6800
1884 182000 128200 182000 128200 7500 2650 8400 8900 8600
1885 88700 145000 85000 5500 5600 8520 7800 9400
1886 63200 121000 70000 4500 6000 8640 6500 10200
1887 70700 29400 124000 70000 4500 6700 8760 4500 11000
1888 70000 9300 144000 70000 6500 5300 8880 5000 11800
1889 47800 146000 75000 8000 4100 9000 8500 12600
1890 155403 89300 155403 89300 18500 2500 9120 12700 13400
1891 2500 9240 13000 14200
1892 4200 9360 14922 14922
1893 6000 9480 14800 14800
1894 5800 9600 14718 14718
1895 6300 9720 17700 17700

Table A3 Singer’s Gross Income, by Market (US$), 1880–1914
Year USA UK F B CH NL SP&P ITA RUS SCA DAH O&B A/NZ SA IND PHI JAP
1880 804670 163806 52473 820018 249653 140877 20538 7494
1881 2027485 886969 177798 82975 928114 299152 209451 36946 15503
1882 2447336 937037 180588 74966 806959 320303 633370 26079 17686
1883 2667936 957891 187061 75656 867962 428579 770986 25510 32756 9667
1884 2848558 934797 197768 90216 889613 448704 688584 41495 44435 22769
1885 2847625
1886 2222148 565155
1887 2609543 612141
1888 3152667 1072033 223293 95538 854996 350766 75767 75568 25209
1889 4093991 1272120 233163 95164 928440 492303 99023 86309 34331
1890 4988638 1465541 267990 109025 1002478 615138 114137 91509 34963
1891 5015191 1591209 260786 126109 1103757 684804 117413 104155 31342
1892 5131742 1695045 256244 141336 1128158 811507 153722 121121 28470
1893 4938639 1815805 295503 145226 1010812 708817 708258
1894 5464434 1912231 299329 162129 952945 679454 706051
1895 5680653 1971697 323997 174021 951826 700268 738234
1896 5398853 2147927 371260 194958 967929 885257 809593
1897 5524362 2246566 392102 214735 956779 1119486 2930557 874013
1898 5851655 2260459 406388 236655 1059021 1118589 3226760
1899 6951189 2661735 444797 245965 1389082 1220380 3742917
1900 7025854 2851933 1518416 1210619 4203127
1901 6626600 2956024 1662613 1158368 4966360
1902 14823686 6453798 3250823 447024 275470 354591 2042342 1337150 5268994 586279 6214132 860125 1813629 496741 372161 51699
1903 16205043 5765188 3450611 451551 287676 400407 2227351 1430060 7113892 693183 6906099 1112419 1626263 606596 444758 39197
1904 16399815 5250412 3525282 482219 289450 440064 2111568 1681639 7606920 815120 7351751 1433147 1764464 464283 508042 32609

Notes and Sources: All original sources from the Singer Archives. Blank cells equal no data.
For the United Kingdom—Andrew Godley, “Pioneering Foreign Direct Investment in British Manufacturing,” Business History Review 73 (Autumn 1999): 419–21, table 2, and the sources cited there.
For France (including Algiers, but not Swiss-French customs zone)—1880: estimate derived from the aggregate value of machines etc. sold and net book accounts receivable at Paris Central Office during
1880, “Schedule B: London. Dec. 31, 1880,” P92-8970 Micro 2002, Reel 17, pp. 177–79; 1881: annualized estimate from sales to end of August, “Comparative Statement of Business at Foreign Offices 1881 &
1882,” Box 103/4; 1882: “New York Department, London. Abstract of Business for ... 1882,” Box 94, Folder 6; 1883–84: “Summary of Business at [Foreign] Offices, 1883 & 1884,” Box 95/4; 1888–89: annualized
estimate from sales to end of August, “Comparative Statement of Business ... to 31 Aug. 1888 and 1889,” Box 107/3; 1890–91: “Comparative Statement of Business at Foreign Offices ... to 31 Dec. 1890 and
1891,” Box 98/1; 1890–99: “Paris—End of Year Results, 1890–99,” Micro 703 (1) (note: this source tallies with above); 1900–1904: “Paris—End of Year Results, 1900–04,” P92-8957, Micro 2002, Reel 4, Section
5, p. 126; 1902–13: “World Reports, Annual,” Micro AP93-0444, unprocessed accession.
For the United States—1902–12: same sources as for France.
For Belgium—1880–84: sources as for France; 1888–89: annualized estimate, source as for France; 1890–91: same source as for France; 1890–99: “Brussels—End of Year Results, 1890–99,” Micro 703 (1)
(which tallies with above); 1900: annualized estimate from first eleven months sales, “Etat comparatif ... 1899–1900,” Box 80/3; 1901: annualized estimate from first five months of year, “Etat comparatif ...
1900–1901,” Box 80/3; 1902–13: sources as for France.
For Switzerland—1880–81: annualized estimates from first forty-six weeks of sales, “Summary of Business in Switzerland, 1880 and 1881,” Box 83/4; 1882–89: same sources as for France; 1890–99:
“Geneva—End of Year Results, 1890–99,” Micro 703 (1); 1902–13: sources as for France.
For Spain and Portugal—1880: annualized estimates from “Results at Madrid C.O. to June 30 [1880–] 1881,” Box 106/4; from 1881–97 Spanish and Portuguese results reported together under Madrid Cen-
tral Office; 1882–89: same sources as for France; 1890–97: “Madrid—End of Year Results,” Micro 703 (1); 1898–99: “Spain—End of Year Results, 1898–1899,” and “Portugal—End of Year Results, 1898–1899,”
Micro 703 (1); 1900–1901: “Madrid—Spain only—End of year Results, 1900–1904,” and “Lisbon—Portugal only—End of Year Results, 1900–1904,” P92-8957 Micro 2002, Reel 4, Section 5, pp. 90 and 102;
1902–14: sources as for France (note Spanish figure for 1914 reported).
Table A3 (continued)
Year USA UK F B CH NL SP&P ITA RUS SCA DAH O&B A/NZ SA IND PHI JAP
1905 17523020 5318562 3718423 522623 322138 493965 1941569 1766085 11646870 926838 8196713 1527392 1673047 495319 526499 44099
1906 21605311 6502798 3928056 548596 340754 582069 2124004 1911924 14181573 1138314 9589533 1698926 1747035 437383 515209 39497
1907 21761768 6425622 4267718 588028 359094 590565 2354671 2541080 15824939 1271111 9596877 1838563 1897451 350392 601898 54814
1908 17640985 5037313 4382106 587250 328724 556127 2344488 2749773 17652794 1313374 9515377 1748871 1901366 285971 593587 67825 383435
1909 20815128 5659242 4607084 619072 333439 549708 2473455 2713392 20070494 1385123 9702362 1835384 1935151 336203 625390 76916 570319
1910 21922930 5508180 4563583 659689 335365 565762 2831732 3039136 21640493 1476991 10191693 2269339 2080936 456697 741400 106906 864065
1911 21753449 5281525 4682818 692676 331103 594638 3338267 3637270 22385294 1484132 10674937 2925521 2239449 507635 860109 595721 1372294
1912 21333746 5673117 4948962 758961 353221 667058 4028666 4203578 25640061 1545868 11142328 3023899 2040932 550286 1178816 784803 1027947
1913 5980637 5088064 803256 323735 666569 3916819 4129276 31360709 1615128 11056199 2706084 2034897 611494 1469057 1259581 677168
1914 3778421 1856322 1342590 738833 383435

For Italy—1880–89: estimates and actual results as for France; 1890–99: “Italy—End of Year Results, 1890–1899,” Micro 703 (1); 1900–1901: sources as for Spain and Portugal but p. 114; 1902–14: sources
as for France (note 1914 figure for Italy reported).
For Australasia—1880: estimate as for France; 1881: Australia (Melbourne office only, Sydney opened in 1882) from Thomas to McKenzie, 15 Jan. 1883, Box 76/9, New Zealand annualized estimate as for
France; 1882–84: sources as for France; 1886–87: Australia “Comparative Statement at New South Wales and Melbourne, 1886 and 1887,” Box 78/1 (note: Australian reporting year actually ran from twelve
months to Oct. 31st of reporting year), New Zealand estimated from unweighted mean share of sales relative to Australia for 1881–85; 1893–97: “Comparative Summary of Business, Australasian Branches,
Years 1893–1897,” Box 79/5; 1902–14: sources as for France.
For South Africa—1880–91: estimates and actual results as for France; 1892: annualized estimate from first half, “Comparative Statement of Business at [Foreign] Offices, to end of June 1891 and 1892,”
Box 98/3; 1902–13: sources as for France.
For India—1880–91: estimates and actual figures as for France; 1892: sources as for South Africa; 1902–14: sources as for France (1914 figure reported).
For the Philippines (office opened in 1882)—1882–92: estimates and actual values as for France; 1902–14: sources as for France.
For Russia (including Finland)—1897–99: “St. Petersburg [Russia]—End of Year Sales, 1897–99,” Micro 703 (1) (note 1897 is estimate from second half sales); 1900–1901: “St. Petersburg—End of Year
Sales, 1900–01,” as for France, and confirmed by Fred Carstensen, “American Multinational Corporations in Imperial Russia: Chapters on Foreign Enterprise and Russian Economic Development,” (Ph.D. diss.,
Yale University, 1976), 146, table 4; 1902–13: sources as for France; 1914: Carstensen, “American Multinational Corporations,” 146, table 4.
For Scandinavia—1901: estimate from “Stockholm and Christiana—End of Year Results [from July 1], 1901,” P92-8957 Micro 2002, Reel 4, Section 5, pp.128, 130; 1902–13: sources as for France.
For the Netherlands—1901: estimate based on “G. Neidlinger, Hamburg—End of Year Results, 1901,” P92-8957, Micro 2002, Reel 4, Section 5, p. 50, and deducting Scandinavian and Ottoman estimates
(above and below); 1902–13: sources as for France.
For Germany and Austria-Hungary—1902–13: sources as for France.
For the Ottoman Empire (including Balkans, i.e. Romania, Serbia, Bulgaria, and Greece)—1901: estimate from “Romania and Serbia, Bulgaria and Greece, Turkey and Egypt—End of Year Results [from July
1], 1901,” P92-8957, Micro 2002, Reel 4, Section 5, pp. 51–53; 1902–13: sources as for France.
For Japan—1908–14: sources as for France.

Table A4 Singer’s Retailing Expenses (US$), 1881–1914
Notes and Sources: As for table A3, with the exception of no data for 1880. Blank cells equal no data.
Year USA UK F B CH NL SP&P ITA RUS SCA DAH O&B A/NZ SA IND PHI JAP
1881 862033 458526 94192 47079 385549 150130 95868 16067 5924
1882 986036 496012 100301 45694 434270 175577 238179 16471 7203
1883 1171177 575609 122239 53387 415773 222321 433138 20052 15168 6371
1884 1281154 596458 129796 58262 465311 256866 427928 27896 22735 11708
1885 1259513
1886 1120624
1887 1220234
1888 1616135 621021 129159 58553 430562 248132 29889 32805 13822
1889 1875391 745310 130972 58757 460918 300450 38297 35590 15732
1890 2431584 923373 149147 65048 497634 403607 48177 36450 16898
1891 2524453 935866 153175 78195 586762 407355 52483 42685 16232
1892 2664586 991533 153644 86036 745072 444172 68555 48415 17875
1893 2686885 1080393 162560 87843 898294 434128
1894 2903850 1140081 160310 90252 522371 428798
1895 3128409 1185373 167524 99917 505265 433934
1896 3006010 1317262 187280 112983 552244 506800
1897 2975509 1358587 197655 123125 567015 641380
1898 2972957 1329457 205321 133612 550943 675741 1392402
1899 3413753 1449474 206766 140967 670625 710680 1735366
1900 3609103 1547216 712543 720370 2075239
1901 3775792 1623262 779113 720770 2395708
1902 4940067 3283060 1759611 205536 145207 187590 901615 740032 2617452 327720 1913828 468064 1059131 218959 98957 22215
1903 5417402 3071401 1883210 205510 151895 208382 1006335 724411 3202169 372831 2118886 578504 1003630 284588 118973 19972
1904 5464836 2746642 1909846 213712 152673 227998 1011748 847536 3538130 441526 2453630 730140 1034187 269065 128097 17183
1905 9489707 2684967 1995664 228723 169360 250283 937677 861853 4859987 480131 4080926 797932 940204 295073 140352 22803
1906 10900712 3268671 2083731 237824 169019 290517 1034524 958086 6836211 581169 4825329 851765 941975 287536 150760 22102
1907 11089467 3303547 2253632 251395 181142 311236 1222996 1293067 7743880 647936 5050336 999837 992458 239499 159296 24452
1908 10052466 3075405 2414466 263632 177289 310121 1252888 1386960 8254681 679933 5121353 1049480 1045015 211509 162678 27394
1909 10895455 2971845 2523372 285862 178107 313143 1298807 1408197 8761863 736864 5263862 1075095 1031470 216720 168665 28677 222719
1910 11596858 2942767 2475748 297903 181232 296331 1385124 1572343 9775329 772746 5349467 1187670 1069984 264488 200089 43190 305215
1911 11794243 2974621 2524560 315711 179604 307325 1596724 1802953 10135529 809492 5587096 1408853 1117204 293389 275290 208684 420888
1912 11631855 3041832 2801953 338190 195080 329120 1917502 2058417 10957245 867810 5933485 1652725 1128552 305166 382068 358205 795111
1913 3235049 2861398 358931 184055 332158 1949095 2080953 13911848 866550 6089793 1518995 1144078 325119 500969 520088
1914 2104000 1092138 557410 355347

Table A5 Population Estimates, 1881–1920 (000s)
Year GB IRE F B CH NL SP P ITA RUS SCA DAH O&B A/NZ SA IND PHI JAP
1881 29879 5146 37406 5575 2855 4090 16993 4678 28460 99710 8480 83997 32800 2984 2000 250155
1882 30105 5101 37511 5630 2864 4140 17086 4721 28661 101720 8510 84720 33000 3080 2100 253100
1883 30426 5024 37616 5685 2873 4200 17179 4763 28862 103730 8540 85420 33250 3219 2200 256000 5818
1884 30749 4975 37720 5740 2882 4250 17272 4806 29062 105740 8600 86200 33467 3340 2300 259000 5860
1885 31076 4939 37825 5795 2891 4310 17364 4848 29263 107750 8680 86910 34500 3438 2400 262000 5900
1886 31407 4906 37930 5850 2900 4360 17457 4890 29463 109760 8760 87730 35500 3541 2490 265000 5940
1887 31712 4857 37971 5905 2909 4420 17550 4933 29664 111770 8810 88630 36500 3646 2720 268000 5985
1888 32080 4801 38011 5960 2918 4480 17606 4975 29864 113780 8860 89637 37500 3749 2880 271000 6015
1889 32421 4757 38052 6010 2947 4530 17662 5018 30065 115790 8900 90520 38500 3836 2920 274000 6045
1890 32767 4718 38092 6069 2980 4540 17718 5060 30265 117800 8950 91440 39585 3932 2980 277000 6075
1891 33122 4680 38133 6132 3014 4590 17774 5096 30466 119310 8990 92526 40000 4028 3040 279559 6105
1892 33457 4634 38160 6194 3047 4650 17830 5133 30666 120820 9040 93520 40000 4107 3180 279990 6135
1893 33883 4607 38187 6257 3081 4700 17885 5169 30867 122330 9090 94460 40000 4184 3270 280420 6165
1894 34270 4589 38215 6319 3115 4760 17941 5205 31067 123840 9160 95490 40000 4261 3360 480850 6195
1895 34661 4560 38242 6382 3148 4830 17997 5242 31268 125350 9260 96600 40000 4338 3453 281280 6225
1896 35057 4542 38269 6444 3182 4890 18053 5278 31468 126860 9360 97800 40000 4410 3560 281710 6261
1897 35457 4530 38305 6507 3215 4970 18109 5314 31669 128370 9470 99070 40000 4489 3690 282140 6457
1898 35863 4518 38342 6569 3249 5040 18271 5350 31869 129880 9580 100360 40000 4548 4560 282570 6653
1899 36272 4502 38378 6632 3283 5110 18432 5387 32070 131390 9680 101650 40000 4611 4680 283000 6849
1900 36686 4469 38415 6694 3315 5160 18594 5423 32270 132900 9780 103000 40000 4670 4820 283430 7045

Table A5 (continued)
Year GB IRE F B CH NL SP P ITA RUS SCA DAH O&B A/NZ SA IND PHI JAP
1901 37091 4447 38451 6767 3359 5220 18727 5472 32475 135680 9870 104165 40000 4751 4960 283872 7241 44261
1902 37458 4435 38530 6840 3403 5300 18861 5520 32695 138460 9950 105470 40100 4820 4990 285800 7437 44806
1903 37829 4418 38609 6913 3447 5390 18994 5569 32915 141240 10020 106730 40200 4886 5080 287700 7635 45437
1904 38203 4408 38687 6986 3491 5470 19127 5617 33135 144020 10090 107970 40300 4966 5175 289600 7814 46023
1905 38582 4399 38766 7059 3534 5550 19261 5666 33355 146800 10160 109210 40400 5049 5289 291500 7992 46620
1906 38963 4398 38845 7132 3578 5630 19394 5715 33575 149580 10240 110450 40500 5135 5403 293400 8171 46733
1907 39349 4388 38914 7205 3622 5710 19527 5763 33795 152360 10320 111710 40600 5227 5517 295300 8349 47675
1908 39739 4385 39984 7278 3666 5790 19660 5812 34015 155140 10420 112960 40700 5327 5631 297200 8528 48609
1909 40133 4387 39053 7351 3710 5840 19794 5860 34235 157920 10520 114220 40800 5441 5745 299100 8707 48766
1910 40531 4385 39123 7424 3753 5900 19927 5909 34455 160700 10620 115470 40931 5560 5859 301000 8885 49589
1911 40887 4381 39192 7497 3797 5980 20060 5958 34671 162480 10710 116750 41000 5728 5973 303013 9064 50752
1912 41068 4368 39261 7570 3841 6070 20194 6006 34891 166260 10800 118150 41100 5918 6087 304900 9242 51591
1913 41302 4346 39331 7643 3884 6160 20327 6055 35110 169040 10900 119522 41200 6089 6201 306800 9421 51749
1914 41714 4334 39400 7716 3928 6280 20460 6104 35330 171820 11000 120040 41300 6168 6315 308700 9600 52312
1915 40130 4280 20659 6123 41300 6100 6530 310600 9778 52752
1916 39295 4270 20803 6176 41300 6052 6645 312500 9957 53597
1917 38521 4270 20949 6229 41300 6045 6759 252827 10135 54283
1918 38287 4280 21095 6281 41300 6133 6716 251780 10314 55965
1919 41620 4350 21243 6334 41300 6333 6834 250737 10324 55663
1920 42388 4360 21303 6033 41300 6653 6953 249699 10445 55963

Table A5 (continued)
Notes and Sources: Population data for the United Kingdom come from Brian R. Mitchell, British Historical Statistics (Cambridge, U.K., 1988). Population data for Spain,
Portugal, Italy, France, Switzerland, Belgium, the Netherlands, Germany and Austria-Hungary, Russia (including Finland), and Scandinavia (aggregating Denmark, Norway, and
Sweden) come from Mitchell, International Historical Statistics: Europe, 1750–2000 (Basingstoke, U.K., 2003). Mitchell, International Historical Statistics: Africa, Asia, and
Oceania (Basingstoke, U.K., 2003), gives South Africa (with an adjustment made here to account for the unenumerated Transvaal and Orange Free State population for pre-1904),
India, Australia and New Zealand (with an adjustment made here to account for the unenumerated Australian aboriginal population), the Philippines (Christian population only,
i.e. approximately 85–90% of total), Japan, and Ottoman Empire (1872 boundaries). Ottoman and Austria-Hungary populations supplemented from Jan Lahmeyer’s Populstat web-
site, www.library.uu.nl/wesp/populstat/populframe.html (accessed 28 Feb. 2006). Straight-line interpolations throughout for missing years. Population data for the United States,
1860–80, are summarized in the table that follows:
Source: Brian R. Mitchell, International Historical Statistics: Americas, 1750–2000 (Basingstoke, U.K., 2003).
Year U.S. Population
1860 31443
1865 35631
1866 36469
1867 37306
1868 38144
1869 38981
1870 39818
1871 40852
1872 41886
1873 42919
1874 43953
1875 44987
1876 46021
1877 47055
1878 48089
1879 49123
1880 50156

Table A6 The Diffusion of Singer Sewing Machines, 1881–1920 (cumulative sewing machine sales per 000 population)
Year GB IRE F B CH NL SP P ITA RUS SCA DAH O&B A/NZ SA IND PHI JAP
1881 12.89 4.73 3.69 3.64 2.38 2.11 5.61 4.31 1.11 0.98 0.95 6.65 0.01 7.75 1.80 0.00
1882 15.29 5.65 4.37 4.57 3.22 3.44 6.68 5.44 1.45 1.27 1.62 7.86 0.02 13.40 2.10 0.01
1883 17.81 6.64 5.06 5.50 4.05 4.99 7.86 6.37 1.87 1.59 2.41 9.13 0.05 18.72 2.35 0.01 0.05
1884 20.41 7.84 5.73 6.44 5.11 6.91 9.05 7.08 2.30 1.90 3.39 10.53 0.11 23.37 2.77 0.02 0.16
1885 22.88 8.97 6.46 7.42 6.26 8.79 10.03 7.80 2.71 2.25 4.44 11.42 0.16 26.64 3.39 0.02 0.25
1886 25.10 10.08 7.20 8.41 7.43 10.67 11.08 8.53 3.10 2.49 5.57 12.12 0.24 29.35 4.05 0.03 0.34
1887 27.55 11.39 7.97 9.41 8.53 12.50 12.13 9.24 3.46 2.73 6.78 12.78 0.29 32.11 4.50 0.04 0.46
1888 30.12 12.78 8.75 10.41 9.54 14.32 13.31 10.03 3.80 3.14 8.08 13.42 0.31 34.70 5.06 0.05 0.59
1889 33.50 14.44 9.67 11.45 10.46 16.15 14.62 10.91 4.30 3.51 9.46 14.12 0.39 37.30 6.07 0.06 0.79
1890 37.44 16.49 10.71 12.63 11.46 18.12 16.07 11.82 4.86 3.80 10.90 14.95 0.52 39.69 7.17 0.07 0.98
1891 41.14 18.83 11.82 13.71 12.56 19.94 17.66 12.83 5.49 4.09 12.43 15.68 0.66 41.97 8.17 0.09 1.15
1892 44.79 21.26 13.00 14.74 13.77 21.69 19.19 13.96 6.25 4.43 14.01 16.44 0.83 44.33 9.18 0.11 1.30
1893 48.02 23.62 14.25 15.89 15.02 23.48 20.45 15.12 6.91 4.83 15.56 17.30 1.04 46.64 10.24 0.13 1.47
1894 51.45 25.93 15.60 17.06 16.45 25.20 21.72 16.31 7.58 5.30 17.05 18.15 1.32 48.82 11.25 0.09 1.67
1895 54.71 28.31 16.96 18.27 17.95 26.85 22.99 17.53 8.26 5.78 18.78 19.17 1.66 51.05 12.48 0.18 1.88
1896 57.65 30.35 18.42 19.62 19.59 28.53 24.24 18.78 9.09 6.28 20.69 20.23 2.03 53.54 13.88 0.21 2.11
1897 60.49 32.58 19.92 20.98 21.39 30.19 25.30 20.07 10.12 6.88 22.54 21.32 2.43 56.01 15.36 0.24 2.31
1898 63.41 35.03 21.42 22.38 23.33 31.99 26.32 21.42 11.14 7.51 24.36 22.49 2.87 59.24 14.25 0.28 2.51
1899 66.90 37.99 23.18 23.88 25.35 33.88 27.77 22.89 12.25 8.18 26.16 23.76 3.35 62.92 15.88 0.32 2.72
1900 70.33 40.83 25.04 25.37 27.34 35.97 29.27 24.58 13.32 8.91 27.92 25.05 3.86 67.12 17.55 0.36 2.94
1901 73.58 42.48 26.96 26.82 29.20 38.09 30.99 25.99 14.28 9.68 29.69 26.35 4.45 71.45 19.33 0.41 3.17

Table A6 (continued)
Year GB IRE F B CH NL SP P ITA RUS SCA DAH O&B A/NZ SA IND PHI JAP
1902 76.67 44.45 29.04 28.16 31.01 40.05 33.17 27.62 15.41 10.44 31.49 27.67 5.18 76.38 21.68 0.45 3.40 0.27
1903 79.29 46.64 31.23 29.49 32.84 42.10 35.50 29.37 16.63 11.53 33.56 29.12 6.11 80.55 24.18 0.50 3.57 0.41
1904 81.68 48.93 33.47 30.88 34.60 44.42 37.61 31.06 18.03 12.71 35.93 30.66 7.24 84.80 25.90 0.56 3.66 0.57
1905 84.30 51.36 35.81 32.32 36.55 46.97 39.42 32.64 19.46 14.59 38.62 32.36 8.36 88.57 27.65 0.62 3.78 0.76
1906 87.50 54.14 38.28 33.81 38.59 50.04 41.39 34.22 21.01 16.83 41.83 34.33 9.64 92.34 29.08 0.67 3.87 1.04
1907 90.52 56.87 40.98 35.36 40.68 53.04 43.38 35.79 22.87 19.24 45.38 36.26 10.95 96.37 30.05 0.74 4.03 1.28
1908 92.63 58.96 42.65 36.87 42.47 55.69 45.31 37.31 24.83 21.77 48.91 38.11 12.18 100.04 30.64 0.80 4.21 1.47
1909 94.88 61.11 46.62 38.53 44.25 58.48 47.22 38.94 26.72 24.44 52.57 39.91 13.48 103.35 31.39 0.87 4.43 1.66
1910 96.95 63.23 49.41 40.26 46.01 61.20 49.52 40.58 28.82 27.16 56.37 41.75 15.07 106.81 32.58 0.94 4.80 1.92
1911 99.04 65.35 52.17 42.07 47.69 63.78 52.24 42.55 31.36 29.97 60.07 43.59 17.11 109.63 33.94 1.03 6.29 2.29
1912 101.73 67.75 55.23 44.08 49.43 66.57 55.63 44.96 34.34 32.72 63.83 45.44 19.22 111.26 35.44 1.14 8.09 2.91
1913 104.35 70.38 58.34 46.19 50.90 69.47 58.83 47.35 37.28 36.17 67.63 47.22 21.08 112.98 37.12 1.28 10.83 3.36
1914 106.41 72.81 61.13 48.00 52.13 71.56 61.74 49.49 39.82 39.54 70.92 49.08 22.92 115.84 38.55 1.40 12.35 3.70
1915 110.61 73.73 64.28 49.34 22.99 121.21 37.28 1.56 14.32 3.89
1916 116.51 76.36 67.08 51.36 23.64 127.13 39.52 1.75 16.72 4.22
1917 122.56 78.89 66.61 50.92 23.64 127.27 38.86 2.16 19.94 4.72
1918 127.51 81.53 68.25 53.85 24.15 129.44 42.57 2.34 23.94 5.47
1919 122.01 83.61 70.71 57.15 25.15 130.15 45.09 2.58 27.40 6.55
1920 124.05 86.53 73.37 62.27 25.96 128.19 47.12 2.80 30.31 7.33

Table A6 (continued)
Notes and Sources: Table A2 aggregated and divided by population in table A5. Note that those markets particularly affected by war and its aftermath have been omitted for
1915–20. Even so, the impact of war can be seen on consumption in the Ottoman and Balkans region. Great Britain and Ireland are included as benchmarks. The diffusion of Singer
sewing machines in the United States, 1858–79, is summarized in the table that follows:
Source: The accumulation of industry sales (from Andrew R. Godley, “The Global Diffusion of the Sewing Machine, 1850–1914,” Research in Economic History 20 (2001):
table 1, “Estimated U.S. Sales”) divided through by U.S. population (from notes to table A5).
Year U.S. Diffusion
1858 1.61
1859 3.07
1860 4.20
1861 5.16
1862 6.44
863 7.75
1864 9.39
1865 9.44
1866 10.92
1867 14.87
1868 20.19
1869 27.16
1870 37.21
1871 52.10
1872 69.19
1873 81.20
1874 88.60
1875 95.69
1876 103.33
1877 111.47
1878 120.09
1879 129.17