Reconstructing the bilateral American tariff toward
Britain, 1873-19131
By BRIAN D. VARIAN
This article reconstructs the bilateral American tariff toward Britain for 1873-1913, addressing a
number of methodological issues in the process. The resulting bilateral tariff series differs
substantially from the overall tariff of the United States, a finding which inherently challenges the
common practice of associating overall tariffs with bilateral trade flows. Additionally, several
observations about the bilateral American tariff toward Britain are made. The high-water mark for
the bilateral American tariff toward Britain corresponds to the McKinley Tariff of 1890. The
other major piece of protectionist legislation, the Dingley Tariff of 1897, does not induce a similar
increase in the bilateral American tariff toward Britain, since the basket of American imports
from Britain underwent a stronger compositional adjustment following the Dingley Tariff, as
determined through Laspeyres tariff indices.
During the late nineteenth century, the United States was foremost among Britain’s
export markets, having surpassed Germany in 1874.2 Of Britain’s principal export markets,
the United States was further notable for having imposed the highest overall ad valorem
equivalent tariff on imports.3 The overall ad valorem equivalent tariff of the United States
pertains to a basket of imports from many countries, which differed in composition from the
basket of imports exclusively from Britain. Therefore, the overall ad valorem equivalent tariff
is an inferior measure of the actual tariff that British exports to the United States encountered.
This article reconstructs that actual tariff, more technically the bilateral American tariff
toward Britain, on an annual basis for 1873-1913.
In recent literature, Jacks et al. claimed that, during the first era of globalisation,
tariffs were a weaker determinant of trade costs than was distance.4 However, this finding
arose from a country-pair panel regression with an explanatory variable for tariffs expressed
as the log product of overall tariffs, rather than as the log product of the relevant bilateral
tariffs. As Jacks et al. readily acknowledge, they were constrained by the unavailability of
bilateral tariff series. Reconstructing the late nineteenth-century bilateral American tariff
toward Britain is an overdue attempt to remedy, albeit only partially, the general lack of
bilateral tariff series for the first era of globalisation. Moreover, reconstructing the bilateral
American tariff toward Britain could reveal the potential extent of the empirical
inappropriateness of substituting an overall tariff for a bilateral tariff in those situations
calling for the latter.
The bilateral American tariff toward Britain is a particularly insightful bilateral tariff
to reconstruct, given the prominent and changing role that trade played in the nineteenthcentury British economy. As Clark et al. have lately argued, the British economy of the 1850s
was highly dependent on trade, so much so that autarky would have reduced British welfare
1
The author would like to thank Chris Minns and Joan Roses for their most helpful
suggestions. The author would also like to thank Jason Lennard and Alain Naef for commenting on an
earlier draft of this article.
2
Trade of the United Kingdom (1874). In this year, the American share of British exports
amounted to 12%.
3
O’Rourke, ‘Tariffs and growth’, p. 461.
4
Jacks, Meissner, and Novy, ‘Trade costs’, pp. 134-5. Whereas the standardised coefficient
for tariffs was 0.13, the standardised coefficient for distance was 0.40.
1
by 27 per cent in that decade.5 This claim is certainly consistent with the emphasis that earlier
economic historians placed upon trade in the nineteenth-century British economy. Regarding
British exports, Ashworth once wrote, ‘nothing contributed more to the air of prosperity in
mid-Victorian Britain than the expansive condition of so many export markets’.6
However, during the late nineteenth century, several of Britain’s export markets, the
United States and Germany among them, began to industrialise. As Broadberry has shown for
the American case, the relative shift out of the agricultural and into the manufacturing sector,
in which the United States enjoyed a substantial productivity lead over Britain, enabled the
former to overtake the latter in per capita income sometime during the closing decade of the
nineteenth century.7 Additionally, industrialisation overseas meant that domestic production
displaced British exports. Consequently, the impressive growth rate of British exports during
the mid-nineteenth century was not sustained into the late nineteenth century.
Indeed, some scholars have claimed that late nineteenth-century Britain suffered from
export-retarded growth, whereby a decline in the growth rate of exports caused a decline in
the growth rate of industrial output.8 Meyer lent quantitative support to this theory.
According to Meyer, if each major British industry had maintained its export volume growth
rate characteristic of the period 1854-72 into the period 1872-1907, then the growth rate of
British industrial output would have increased from 3.05 (1854-72) to 4.10 (1872-1907) per
cent per annum, instead of having fallen from 3.05 to 1.75 per cent per annum.9 McCloskey
argued that the mid nineteenth-century growth rate of British industrial output could not
possibly have been sustained in the late nineteenth century because the domestic supplies of
labour and capital were far too insufficient to accommodate such growth.10 For McCloskey,
decelerating exports were not an impediment to industrial output. Many other scholars—too
many to discuss here—engaged in the debate over Britain’s export-retarded growth.11 What is
conspicuous about this debate is the absence of anything more than a passing mention of
foreign protective tariffs.
Foreign protective tariffs posed a challenge to Hatton, who was unable to incorporate
them directly into his demand function for late nineteenth-century British exports, citing the
empirical obstacle arising from the great number of tariff-imposing markets for British
exports.12 Granted, Hatton examined the worldwide demand for British exports, and not just
the bilateral demand from a single country. Nevertheless, by reconstructing the bilateral
American tariff toward Britain, this present article begins the process of breaking down an
identifiable empirical barrier.
Clark, O’Rourke, and Taylor, ‘Growing dependence’, p. 119.
Ashworth, Economic history, p. 138.
7
Broadberry, ‘Overtake Britain’, p. 383. Perhaps not coincidentally, the 1890s was also the
most protectionist decade in American history.
8
The decline in the growth rate of exports would cause an even greater decline in the growth
rate of industrial output, per the export multiplier effect. See Clark and Crawford, National income,
pp. 90-103.
9
Meyer, ‘Influence of exports’, p. 17.
10
McCloskey, ‘Victorian Britain’.
11
Other outstanding contributions to this debate include: Aldcroft, ‘Victorian Growth’;
Crafts, ‘Victorian Britain’; Feinstein, ‘British economic growth’; Harley, ‘Globalisation to
fragmentation’.
12
Hatton, ‘British exports’, p. 583. Hatton attempted to determine if foreign protective tariffs
reduced the volume of British exports by testing for a structural break between 1890 and 1891, which
was approximately when several industrialising countries raised their overall tariffs: Italy (1887), the
United States (1890), and France (1892). He found that no such break occurred.
5
6
2
This article proceeds as follows. Section I surveys the literature on historical
reconstructions of tariffs and then considers alternative methods for measuring protection,
specifically those advanced by the Board of Trade, Balassa, and McCloskey. Section II
reconstructs the bilateral American tariff toward Britain in both absolute and relative terms,
using a method best described as a current-year-weighted average of per-industry ad valorem
equivalent tariffs. Section III comments on the level and movement, or lack of movement, in
the absolute and relative bilateral tariff series. This section also constructs Laspeyres tariff
indices in order to gauge the degree of compositional adjustment that occurred in the wake of
the McKinley and Dingley Tariffs. Following the example of Federico and Tena, section III
additionally examines the sensitivity of the bilateral tariff series to certain industries. Section
IV offers some concluding remarks.
I
Several economic historians have reconstructed measurements of American
protection. Irwin reconstructed the overall nominal tariff of the United States for 17901820.13 He accounted for non-dutiable articles of importation, as well as drawbacks paid to
merchants on imports subsequently exported (re-exports), whereas earlier series disregarded
these two items. First Sundararajan and later Hawke calculated the effective tariffs for certain
American industries during the late nineteenth and early twentieth centuries.14 An effective
(as opposed to nominal) tariff for an industry measures the protection extended to the
domestic value added in that industry, after accounting for both the factor share of imported
material inputs in that industry and the tariff applicable to those imported material inputs.15
Making several assumptions about the factor shares of imported material inputs, Sundararajan
reconstructed the annual effective tariffs for pig iron; tinplate and terneplate; and blackplate.
Hawke reconstructed the effective tariffs for a variety of industries, but only for the years
1879, 1889, 1899, and 1904, the years in which the United States conducted censuses of
manufactures, and thus the years for which the factor shares of imported material inputs
could be roughly determined.16
In a recent article, Dedinger reconstructed the late nineteenth-century (nominal)
bilateral French tariff toward Germany, a rare example of an historical reconstruction of a
bilateral tariff.17 In this endeavour, she benefitted from the arrangement of the French trade
statistics, which enabled her to identify, for each product class, the value of French imports
from Germany and the customs revenue collected thereon. Dedinger drew on her bilateral
tariff series to claim that French protectionism was not instrumental in directing the course of
French imports from Germany.18
In 1903, the British Board of Trade actually estimated the (nominal) bilateral
American tariff toward Britain, along with the bilateral tariffs of ten other countries, but for
just the preceding year.19 The Board of Trade estimated that the bilateral American tariff
Irwin, ‘Average tariff’.
Sundararajan, ‘Iron and steel’; Hawke, ‘Industrial protection’.
15
As the ratio of domestic value added to total value increases, the effective tariff decreases,
ceteris paribus. As the tariff on imported material inputs increases, the effective tariff decreases,
ceteris paribus.
16
Hawke, ‘Industrial protection’, p. 88.
17
Dedinger, ‘Franco-German’, p. 1044.
18
Ibid., pp. 1044-45.
19
British and Foreign Trade (P.P. 1903, LXVII), p. 171. The ten other countries and
corresponding bilateral tariffs were as follows: Russia (131%), Austria-Hungary (35%), France
(34%), Italy (27%), Germany (25%), Canada (16%), Belgium (13%), New Zealand (9%), Australian
Commonwealth (6%), and South African Customs Union (6%).
13
14
3
toward Britain was 73 per cent in 1902.20 Although 73 per cent far exceeds this article’s
estimate of the bilateral American tariff toward Britain for that same year, the discrepancy
ought not cause alarm. The Board of Trade calculated bilateral tariffs using a method so
dissimilar to the one undertaken in this article that its estimates represent a fundamentally
different measurement of protection. The Board of Trade summarized the distinguishing
feature of its method as follows: ‘ … the basis of the calculation is not the classes of British
goods which we actually sell to each particular country, but those which we sell to the world
in general’ [emphasis original].21 Thus, the Board of Trade’s estimate of the bilateral
American tariff toward Britain is a weighted average of the ad valorem equivalent tariffs that
the United States imposed upon eleven representative classes of British goods, with the
weights derived not from the composition of British exports to the United States, but instead
from the composition of British exports to all countries.22
The Board of Trade settled on this method for calculating bilateral tariffs because the
resulting estimates are not diminished by the imposition of prohibitive tariffs on classes of
British exports.23 Prohibitive tariffs, which are high enough to block imports entirely, are the
most extreme case of the substitution effect. According to the substitution effect, if the United
States increases its tariff on a particular class of British exports, the value of that class of
British exports to the United States declines, ceteris paribus, and so too does the share of that
class in the overall composition of British exports to the United States.24 By fixing the shares
of the classes of British exports based upon the composition of British exports to all
countries, the Board of Trade attempted to ensure that the substitution effect did not erode its
estimates of bilateral tariffs.
Whereas the Board of Trade’s method for calculating the bilateral American tariff
toward Britain adjusted for the substitution effect, the present author’s does not. There are
two main reasons why. First, not adjusting for the substitution effect preserves the
comparability between my estimates of the bilateral American tariff toward Britain and the
overall tariff of the United States, and hence allows for the computation of the relative
bilateral American tariff toward Britain. Second, the Board of Trade’s method fails to
discriminate between prohibitive tariffs, which block imports, and irrelevant tariffs, which
apply to goods absent from the composition of trade because the formerly importing country
has attained a comparative advantage in the production of these goods.25 The late nineteenthcentury industrialisation of the United States makes this point especially salient.
Balassa, who calculated the overall effective and nominal tariffs of several countries
for the year 1962, was just as assiduous as the Board of Trade in adjusting for the substitution
effect in his measurements of protection.26 Rather than derive weights from the tariffimposing country’s composition of imports, Balassa advocated for weights taken from the
20
Ibid., p. 173.
Ibid., p. 169.
22
Ibid., p. 170. The eleven representative classes of British goods comprised more than threequarters of total British exports.
23
Ibid., p. 169.
24
For a discussion of how higher tariffs reduce imports, see Corden, Theory of protection, pp.
5-11.
25
A prominent debate along these lines addressed the extent to which the American cotton
textile industry was dependent upon protective tariffs. See Harley, ‘International competitiveness’;
Irwin and Temin, ‘Antebellum tariff’; Harley, ‘Different products’. See also Irwin, ‘Iron industry’.
26
Balassa, ‘Tariff protection’. The several countries were Japan, Sweden, the United
Kingdom, the United States, and the countries forming the European Common Market.
21
4
composition of worldwide trade.27 Balassa’s approach to adjusting for the substitution effect,
while suitable for the calculation of overall tariffs, is obviously inadaptable to a bilateral
context.
Another method for measuring protection was put forward by McCloskey, who
sought to compare the level of British protection before and after the mid nineteenth-century
transition to free trade.28 Cognizant of the changing composition of British imports
throughout this period, McCloskey constructed what might be called a Laspeyres tariff index,
calculated as a weighted average of the ad valorem equivalent tariffs that Britain imposed
upon twelve representative articles of importation, with the weights taken from a fixed base
year preceding the transition to free trade, in this case 1841.29 On the basis of the Laspeyres
tariff index, McCloskey showed that the overall British tariff declined from 35 to 13 per cent
between 1841 and 1881.30
McCloskey’s method of constructing a Laspeyres tariff index has the advantage of
being easily adapted to a bilateral context, as is done section III of this article. However,
unlike McCloskey, the present author does not construct Laspeyres tariff indices over broad
intervals, since doing so would ignore longer-term shifts in the relative economic importance
of the index components. Therefore, Laspeyres tariff indices are constructed for briefer
intervals of just several years, with the intent of determining the compositional adjustment
caused by the protectionist McKinley and Dingley Tariffs.
II
The source employed in reconstructing the bilateral American tariff toward Britain is
the Foreign Commerce and Navigation of the United States, a series of reports issued
annually by the United States Treasury Department. The only other potential source, the
Annual Statements of the Trade of the United Kingdom, enumerates British exports to the
United States, but does so in a manner inconsistent with the classification of articles in the
American tariff schedule. Accordingly, this study relies on the American trade statistics. One
of the peculiarities of the Foreign Commerce is that each annual report covers the fiscal year
ending 30 June, rather than the calendar year. For the purposes of this study, 1873/4 means
the year beginning 1 July 1873 and ending 30 June 1874. The bilateral tariff series
reconstructed here spans the forty years from 1873/4 to 1912/3. This period commences with
the beginning of the Great Depression (or Long Depression), when there occurred a secular
decline in British export prices, as well as a precipitous drop in the growth rate of British
export volumes, and ends just prior to the First World War.31
The Foreign Commerce treats dutiable and non-dutiable imports entirely separately.
With regard to the dutiable imports, the two relevant sections of the Foreign Commerce for
reconstructing the bilateral American tariff toward Britain are the article-country
disaggregation and industry-tariff disaggregation sections. The article-country disaggregation
section records, for example, the value of pig iron imported from Britain. The industry-tariff
27
Ibid., p. 579. Since data on the composition of worldwide trade was unavailable, Balassa
compromised and calculated weights based upon the composition of total trade amongst the countries
considered in his analysis.
28
McCloskey, ‘Magnanimous Albion’, pp. 307-8.
29
Ibid.
30
Ibid., 309. She also constructed Laspeyres tariff indices with base years of 1854 and 1881.
All three indices exhibited a similar absolute decline in the overall British tariff.
31
Schlote, Overseas trade, pp. 134-5; Imlah, Pax Britannica, pp. 95-8. Both Schlote and
Imlah calculated that the growth rate of British export volumes approximately halved between the
periods 1842-73 and 1873-1913. For a discussion of the appropriateness of the phrase ‘Great
Depression’, see Beales, ‘Great Depression’.
5
disaggregation section records, for example, the total value of all iron, steel, and
manufactures thereof imported from all countries and the customs revenue collected thereon,
which thus enables the calculation of the per-industry ad valorem equivalent tariff that the
United States imposed upon imports from all countries.
For now, the task remains to construct annual estimates of the ad valorem equivalent
tariff that the United States levied upon the basket of imports from Britain (Tb,y). Before
calculating Tb,y, it is first necessary to calculate the industry-composite bilateral American
tariff toward Britain (Cb,y):
𝑀𝑏,𝑖,𝑦
𝑅
𝐶𝑏,𝑦 = ∑17
)(𝑀𝑎,𝑖,𝑦 )
[1]
𝑖=1(∑17 𝑀
𝑖=1
𝑏,𝑖,𝑦
𝑎,𝑖,𝑦
Here, M represents the value of dutiable imports, while R represents the customs revenue
accruing to the United States from those dutiable imports. The subscripts denote American
imports from Britain (b), American imports from all countries (a), the particular industry (i),
and the year (y).32 Taking Cb,y from Equation 1, it is next possible to calculate the bilateral
American tariff toward Britain (Tb,y):
𝑇𝑏,𝑦 =
(𝐶𝑏,𝑦 )(𝐷𝑏,𝑦 )
𝐷𝑏,𝑦 +𝐹𝑏,𝑦
[2]
In Equation 2, D represents the total value of dutiable imports and F the total value of nondutiable imports. The meanings of the subscripts are retained from Equation 1.
The industry-composite bilateral American tariff toward Britain, represented in
Equation 1, is a weighted average of the per-industry ad valorem equivalent tariffs that the
United States imposed upon dutiable imports from all countries, taken from the industry-tariff
disaggregation section of the Foreign Commerce. The weights, computed from the articlecountry disaggregation section, are the per-industry shares of dutiable imports from Britain
within a composite basket of dutiable imports from Britain spanning seventeen industries:
alkali; automobiles and parts thereof; books; cement; clocks and watches; (bituminous) coal;
cotton manufactures; earthenware and chinaware; flax and manufactures thereof; fur and
manufactures thereof; iron, steel, and manufactures thereof; leather and manufactures thereof;
salt; silk manufactures; tinplate; wool; and wool manufactures. Because the article-country
disaggregation section does not explicitly record the total value of dutiable imports from
Britain for each industry, it is necessary to sum the values of the dutiable articles imported
from Britain for each of the seventeen industries, in order to obtain Mb,i,y. In other words, it is
necessary to sum the values of (dutiable) pig iron, (dutiable) bar iron, and so forth imported
from Britain, in order to obtain the total value of (dutiable) iron, steel, and manufactures
thereof imported from Britain.
The assumption implicit in Equation 1 is that, within each industry, the intra-industry
distribution of dutiable articles that the United States imports from Britain mirrors the intraindustry distribution of dutiable articles that the United States imports from all countries.
Indeed, the danger of this assumption is best conveyed by a simple example. Suppose there is
an industry that includes only two articles, X and Y, which the United States imports in equal
values. The ad valorem equivalent tariff is 20 per cent for article X, 40 per cent for article Y,
and 30 per cent for the industry as a whole. However, the United States imports article X
exclusively from country A and article Y exclusively from country B. In this example, the
32
Prior to 1890/91, the Foreign Commerce does not record imports from Britain as a whole,
but instead from England (including Wales), Scotland, and Ireland. Between 1890/91 and 1908/09,
the Foreign Commerce records imports from Britain as a single country, after which it reverts to the
earlier convention of recording imports from three separate countries. Accordingly, for the years
1873/74-1889/90 and 1909/10-1912/13, the author calculates the total value of dutiable imports per
industry for each of England, Scotland, and Ireland, and then adds these three values together so as to
obtain Mb,i,y.
6
actual bilateral tariff toward country A is 40 per cent for this industry, but the calculation of
Cb,y inappropriately relies on an ad valorem equivalent tariff of 30 per cent.
The tenuous nature of this assumption factors heavily into the selection of the
seventeen industries listed earlier. For each of these seventeen industries, the intra-industry
distribution of dutiable articles imported from Britain broadly approximates the intra-industry
distribution of dutiable articles imported from all countries. Since some industries encompass
many individual articles of importation, especially the industry of iron, steel, and
manufactures thereof, and since the intra-industry distributions of dutiable articles
approximate each other to varying extents, the decision to settle on the seventeen
aforementioned industries is inevitably a discretionary one. The glass industry offers an
example of an industry excluded from the calculations for this reason. In 1889/90, the articlecountry disaggregation section classifies 21 per cent of dutiable glass imports from all
countries as ‘cylinder and crown glass, polished and silvered’, whereas less than 1 per cent of
dutiable glass imports from Britain fall under this classification.33 Because certain industries,
such as the glass industry, are excluded from Equation 1, it is essential to observe that ΣMb,i,y
from Equation 1 is always less than Db,y from Equation 2, as the latter value includes all
dutiable imports from Britain across all industries.
For the industry of flax and manufactures thereof, the intra-industry distributions of
dutiable articles imported from Britain and from all countries are roughly similar until
1883/4, when the Foreign Commerce merges the industries of flax and manufactures thereof;
hemp and manufactures thereof; and jute and manufactures thereof. The amalgamated
industry of flax, hemp, jute, and manufactures thereof encompasses raw hemp and raw jute,
the vast majority of which the United States imported from countries other than Britain.
Consequently, the introduction of this amalgamated industry into the American trade
statistics causes the intra-industry distribution of dutiable articles imported from Britain to
differ considerably from the intra-industry distribution of dutiable articles imported from all
countries. For this reason, the industry of flax and manufactures thereof would ordinarily be
excluded from Equation 1, just as the glass industry is excluded from Equation 1. However,
whereas the glass industry constitutes a relatively minor share of American imports from
Britain, the industry of flax and manufactures thereof constitutes a quite large share; flax and
manufactures thereof accounted for fully 10 per cent of American imports from Britain in
1882/3.34 No truly representative series of the American tariff against Britain can neglect this
important industry.
A third section of the Foreign Commerce, the article-tariff disaggregation section,
provides an acceptable solution to the problem created by the merger of flax and
manufactures thereof; hemp and manufactures thereof; and jute and manufactures thereof.
For the purpose of Equation 1, the author redefines the industry of flax and manufactures
thereof to include just burlaps and linens for the years from 1883/4 to 1889/90. Burlaps and
linens are two dutiable articles of importation listed congruently in the article-country
disaggregation and article-tariff disaggregation sections of the Foreign Commerce.35 For the
years 1883/4 to 1889/90, Equation 1 weights the ad valorem equivalent tariff that the United
States imposed upon burlaps and linens (combined) by the share of burlaps and linens
(combined) within the composite basket of dutiable imports from Britain. It is noteworthy
that, in 1883/4, the value of burlaps and linens imported from Britain was $14.7 million,
while the total value of all flax, hemp, jute, and manufactures thereof imported from Britain
Foreign Commerce (1890). The United States imported almost all of its ‘cylinder and
crown glass, polished and silvered’ from Germany.
34
Foreign Commerce (1883).
35
The classifications of individual articles of importation are not always congruent between
these two sections of the Foreign Commerce.
33
7
was $19.1 million.36 Seen in this light, redefining the industry as just burlaps and linens still
maintains a high degree of representativeness in the calculations.
Yet another classificatory change in the Foreign Commerce requires that the author
once again redefine the industry of flax and manufactures thereof, for the purpose of
Equation 1. Beginning in 1890/91, the article-country disaggregation section shifts linens to
the ubiquitous classification of ‘all other manufactures of flax, hemp, or jute’, a classification
without any equivalent in the article-tariff disaggregation section. Since the article-country
disaggregation and article-tariff disaggregation sections now differentiate between raw and
manufactured flax, hemp, and jute, and since the article-tariff disaggregation section lists an
overall ad valorem equivalent tariff for all dutiable manufactures of flax, hemp, and jute, the
author redefines the industry to include all manufactures of flax, hemp, and jute from 1890/91
until the conclusion of the series. In summary, this industry includes flax and manufactures
thereof for 1873/4-1882/3, burlaps and linens for 1883/4-1889/90, and all manufactures of
flax, hemp, and jute for 1890/1-1912/3.While redefining this industry at two junctures
(1883/4 and 1890/1) introduces a small element of inconsistency to the tariff series being
constructed here, doing so ensures that the intra-industry distributions of dutiable articles
from Britain and from all countries broadly approximately each other, and that the redefined
industry matches an ad valorem equivalent tariff ascertainable from the Foreign Commerce.
The chemical industry, as designated in the American trade statistics, embodies
highly discrepant intra-industry distributions of dutiable articles, with the United States
importing dyestuffs predominantly from Germany and alkali almost exclusively from Britain.
This problem is resolved by employing, in Equation 1, a purposely crafted ‘alkali industry’ in
place of the chemical industry. The alkali industry, as the author has defined it, includes just
three dutiable articles of importation, which are congruently listed in the article-country
disaggregation and article-tariff disaggregation sections: caustic soda, sal soda, and soda ash.
Therefore, Equation 1 weights the ad valorem equivalent tariff that the United States imposed
upon caustic soda, sal soda, and soda ash (combined) by the share of these three articles
(combined) within the composite basket of dutiable imports from Britain.
Starting in 1883/4, the Foreign Commerce incorporates the tinplate industry, formerly
treated as distinct, into the industry of iron, steel, and manufactures thereof. Yet, owing to the
identical recording of tinplate in the article-country disaggregation and article-tariff
disaggregation sections, it is possible to extricate tinplate from iron, steel, and manufactures
thereof, and continue treating tinplate as its own industry for the duration of the series.
Obviously then, Ra,i,y and Ma,i,y for the industry of iron, steel, and manufactures thereof are
adjusted to exclude tinplate between 1883/4 and 1912/3. More than for the sake of
consistency, the rationale for keeping tinplate as a distinct industry lies in the tremendous
value of tinplate that the United States imported from Britain, as well as the atypical
treatment of tinplate in the American tariff legislation, specifically the provision of the
McKinley Tariff of 1890 that delayed an increase in the tariff on tinplate until 1 July 1891.37
In a strict sense, the Foreign Commerce treats wool and manufactures thereof as a
single industry. Yet, the article-country disaggregation section unambiguously notes which
dutiable articles are wool and which dutiable articles are wool manufactures. Likewise, the
industry-tariff disaggregation section deconstructs the ad valorem equivalent tariff for wool
and manufactures thereof into separate ad valorem equivalent tariffs for wool and wool
manufactures. Altogether, there exists no difficulty in treating wool and wool manufactures
as distinct industries in Equation 1. Yet, simplicity itself does not justify this decision; there
are important historical reasons calling for the segregation of these two classes of articles.
36
37
Foreign Commerce (1884).
Tariff Acts Passed (Washington, 1909), p. 469.
8
Table 1. Industry Discontinuities
Year
1883/4
1883/4
1883/4
1890/1
Industry
Cement
Flax and manufactures thereof
Iron, steel, and manufactures thereof
Flax and manufactures thereof
1905/6
1906/7
1909/10
Automobiles and parts thereof
Alkali
Books
Discontinuity
Industry begins
Includes only burlaps and linens
Includes iron ore
Includes all manufactures of flax,
hemp, and jute
Industry begins
Industry ends
Includes paper and manufactures
thereof
Source: See text.
First, wool and wool manufactures, even when treated as separate industries, consistently
rank among the chief American imports from Britain during this period. Second, the United
States levied a much higher tariff on wool manufactures than on wool, in keeping with the
compensating system of duties, whereby American tariff legislation set the tariff on wool
manufactures high enough to offer domestic wool manufacturers both an element of
protection and a ‘compensation’ for the higher price of wool that resulted from there being a
tariff on this material input.38
An important point germane to the (raw) wool industry is that the ephemeral WilsonGorman Tariff of 1894 removed all duties on wool imports. Recall that, in Equation 1, Mb,i,y
represents the value of only the dutiable imports from Britain per industry. Therefore, the
value of Mb,i,y is nil for the wool industry in the years 1895/6 and 1896/7. Since the WilsonGorman Tariff did not become law until 28 Aug. 1894, Mb,i,y takes on a small value for the
wool industry in 1894/5, representing the value of the dutiable wool imported from Britain
during the brief period from 1 July 1894 to 28 Aug. 1894.
Of the seventeen industries covered in Equation 1, several come with a few minor
qualifications. The cement industry is introduced into the calculation of Equation 1 beginning
in 1883/4, when the article-country disaggregation section of the Foreign Commerce first
accords it separate treatment. Likewise, the industry of automobiles and parts thereof is
introduced beginning in 1905/6. In 1906/7, a classificatory change in the article-country
disaggregation section renders impracticable the continued inclusion of the alkali industry in
Equation 1; in this year, the alkali industry is dropped from the calculation.39 Other industries
are characterized by minor internal discontinuities. The industry of iron, steel, and
manufactures thereof includes iron ore starting in 1883/4. From 1909/10 to 1912/3, the book
industry also includes paper and manufactures thereof. Table 1 summarizes the
discontinuities in the various industries.
Moving from Equation 1, the industry-composite bilateral American tariff toward
Britain (Cb,y), to Equation 2, the bilateral American tariff toward Britain (Tb,y) entails the
assumption that the dutiable imports from Britain falling outside of the seventeen industries
are subject to the industry-composite American tariff against Britain. The dutiable imports
unaccounted for in Equation 1 represent anywhere between 11 and 21 per cent of total
38
The mechanics of the compensating system of duties, as applied to wool and woollens, are
detailed in Taussig, Tariff question, pp. 322-4.
39
In 1905/6, American alkali imports from Britain amounted to a paltry $0.2 million. Foreign
Commerce (1906).
9
imports from Britain, depending upon the year.40 There are three categories of dutiable
imports excluded from Equation 1. The first category, already described at length, includes
the dutiable imports of those industries exhibiting highly discrepant intra-industry
distributions, such that the corresponding ad valorem equivalent tariff for that industry would
grossly misrepresent the true bilateral American tariff toward Britain for that industry. The
second category includes dutiable imports from Britain classified in the Foreign Commerce
as ‘all other dutiable articles’.41 The third category includes dutiable imports that the Foreign
Commerce enumerates separately, but that are largely inconsequential, such as artificial
feathers and smokers’ pipes. Equation 2 applies the industry-composite tariff to these three
categories of dutiable imports. Additionally, Equation 2 incorporates the non-dutiable imports
from Britain, with the result being annual estimates of the ad valorem equivalent tariff that
the United States levied upon the whole basket of imports from Britain.
In constructing this series, one final adjustment is necessary. Prior to 1879/80, the
Foreign Commerce follows the arcane convention of recording specie, specifically gold and
silver bullion and coin, as non-dutiable articles of importation in the article-country
disaggregation section. In 1879/80, the American trade statistics cease recording specie as
non-dutiable articles of importation, and begin recording specie flows in an entirely separate
section.42 The consequence of this change in recordkeeping is an inconsistent tariff series,
broken between 1878/9 and 1879/80. To make the tariff series consistent, Equation 2 is
adjusted for the years 1873/4 to 1878/9 by subtracting from the denominator the value of
specie imported from Britain (Sb,y):
𝑇𝑏,𝑦 = 𝐷
(𝐶𝑏,𝑦 )(𝐷𝑏,𝑦 )
𝑏,𝑦 +𝐹𝑏,𝑦 −𝑆𝑏,𝑦
[3]
The first column of Table 2 reports the bilateral American tariff toward Britain, adjusted for
specie flows.
The second column of Table 2 reports the ad valorem equivalent tariff that the United
States levied on all imports from all countries, as recorded in the Foreign Commerce.
However, these figures embody the same inconsistency as described above, namely the
inclusion of specie flows prior to 1879/80. The corrected figures are reported in the third
column. It is worth highlighting that correcting for specie flows causes the ad valorem
equivalent tariff to increase by 1 or 2 per cent for each year between 1873/4 and 1878/9.
Inasmuch as the British share of total American imports ranged between 16 and 34
per cent throughout the four decades covered in this study, the ad valorem equivalent tariff
that the United States levied on all imports from all countries (column 3) is heavily
influenced by imports from Britain. Thus, the fourth column of Table 2 presents the
American tariff toward all countries except Britain. The fifth column presents the relative
bilateral American tariff toward Britain, as determined by dividing the bilateral American
tariff toward Britain (column 1) by the overall American tariff toward all countries except
Britain (column 4).43 The relative bilateral American tariff toward Britain (column 5) is
illustrated in Figure 1.
40
The coverage rate of the tariff series therefore ranges between 79% and 89%. There is no
statistically significant trend in the coverage rate.
41
‘All other dutiable articles’ imported from Britain usually amounted to 1% of total imports
from Britain.
42
A note on p. 557 of the Foreign Commerce (1880) states, ‘This table embraces only
merchandise, specie having been omitted. This fact should be observed in comparisons made with the
data in corresponding tables for previous years, which tables include both merchandise and specie’.
43
The author’s method of calculating a relative bilateral tariff differs from Dedinger’s. She
calculated the relative bilateral French tariff toward Germany by dividing the bilateral French tariff
toward Germany by the overall French tariff toward all countries including Germany.
10
Table 2. Bilateral American tariff toward Britain, 1873/4-1912/3
Year
(1)
(2)
Bilateral
American tariff
toward Britain
(per cent)
Overall
American tariff
(per cent)
33.6
34.3
36.3
34.3
34.9
34.2
35.3
36.0
37.4
35.7
33.9
34.3
35.7
36.8
36.7
37.4
36.6
35.0
42.5
44.1
44.5
30.6
27.7
23.4
35.5
36.0
33.2
31.4
32.8
31.3
32.4
32.0
29.6
26.8
29.8
30.9
26.1
24.0
22.3
20.5
26.9
28.2
30.2
26.7
27.2
29.0
29.1
29.8
30.2
30.0
28.5
30.8
30.4
31.5
30.6
30.0
29.6
25.7
21.6
23.9
20.6
20.4
20.7
21.9
24.8
29.5
27.6
28.9
27.9
27.9
26.3
23.8
24.2
23.3
23.9
23.0
21.1
20.3
18.6
17.7
1873/4
1874/5
1875/6
1876/7
1877/8
1878/9
1879/80
1880/1
1881/2
1882/3
1883/4
1884/5
1885/6
1886/7
1887/8
1888/9
1889/90
1890/1
1891/2
1892/3
1893/4
1894/5
1895/6
1896/7
1897/8
1898/9
1899/1900
1900/1
1901/2
1902/3
1903/4
1904/5
1905/6
1906/7
1907/8
1908/9
1909/10
1910/1
1911/2
1912/3
(3)
Overall
American
tariff,
corrected for
specie
(per cent)
28.3
29.4
31.3
29.2
29.0
30.3
29.1
29.8
30.2
30.0
28.5
30.8
30.4
31.5
30.6
30.0
29.6
25.7
21.6
23.9
20.6
20.4
20.7
21.9
24.8
29.5
27.6
28.9
27.9
27.9
26.3
23.8
24.2
23.3
23.9
23.0
21.1
20.3
18.6
17.7
(4)
(5)
American tariff
toward all
countries
except Britain
(per cent)
25.8
27.3
29.4
27.4
27.1
29.1
26.0
27.5
27.4
28.0
26.8
29.7
28.6
29.8
28.5
27.6
27.3
22.9
16.6
18.2
15.6
17.6
18.6
21.5
22.3
28.1
26.3
28.4
26.8
27.1
25.0
22.2
23.1
22.5
22.8
21.5
20.0
19.5
17.8
17.1
Relative
bilateral
American tariff
toward Britain
1.30
1.26
1.24
1.25
1.29
1.18
1.36
1.31
1.36
1.27
1.27
1.16
1.25
1.23
1.29
1.35
1.34
1.53
2.56
2.42
2.84
1.73
1.49
1.09
1.59
1.28
1.26
1.11
1.22
1.16
1.30
1.44
1.28
1.19
1.31
1.44
1.30
1.23
1.25
1.20
Source: See text.
11
Figure 1. Relative bilateral American tariff toward Britain,
1873/4-1912/3
3.0
2.5
2.0
1.5
1.0
0.5
0.0
III
Neither the Tariff Act of 1875 nor the Mongrel Tariff of 1883 caused any discernible
change in either the absolute or relative bilateral American tariff toward Britain. The Tariff
Act of 1875 raised duties minutely, in an attempt to stem the decline of customs revenues
resulting from the progressively diminishing value of imports after 1872/3.44 The Mongrel
Tariff, however, revised duties in a somewhat more substantial way. The ad valorem
equivalent tariff for earthenware and chinaware, a staple British export industry, increased
from 43 to 56 per cent between 1882/3 and 1883/4.45 Yet, between the same two years, the ad
valorem equivalent tariff for silk manufactures fell from 59 to 50 per cent.46 With regard to
the Mongrel Tariff, the consensus of economic historians applies equally to the bilateral
American tariff toward Britain as it does to the overall American tariff: duties were revised
not inconsiderably, but these revisions produced hardly any net change in the aggregate level
of protection.47
The McKinley Tariff of 1890 represents an abrupt departure from the status quo of
American tariff policy during the 1870s and 1880s, having raised the duties on manufactured
imports across a range of industries. Cotton manufactures, wool manufactures, and tinplate,
all major British exports to the United States, suddenly fell subject to much higher duties. As
for cotton manufactures, the United States imported hardly any of the cheaper grades by the
late nineteenth century, but continued to import the more expensive grades.48 The McKinley
Tariff raised the duties on these more expensive grades of cotton manufactures, causing the
ad valorem equivalent tariff for this industry to increase from 40 to 51 per cent.49 Yet, the
additional protection that the McKinley Tariff extended to cotton manufacturers was not
44
Ashley, Modern tariff, p. 189.
Foreign Commerce (1883, 1884).
46
Ibid.
47
Ashley, Modern tariff, p. 196; Taussig, Tariff history, pp. 249-50.
48
Saul, Overseas trade, p. 145.
49
Foreign Commerce (1890, 1891).
45
12
nearly as great as the additional protection that it extended to wool manufacturers. When the
McKinley Tariff was being crafted in Congress, moderate upward revisions in the duties on
wool were proposed, and were eventually enacted.50 These proposed upward revisions
provided wool manufacturers an occasion to demand greater duties on wool manufactures, in
accordance with the principle of compensating duties, as discussed in the previous section of
this study. However, so generous were the assumptions about the factor proportion of wool
within wool manufactures, and so (purposely) byzantine was the schedule of duties devised
for wool manufactures, that the wool manufacturers ultimately obtained far more than mere
compensation for the higher price of wool.51 By 1891/2, the ad valorem equivalent tariff for
wool manufactures had reached 96 per cent, compared to an already high 69 per cent in
1889/90.52
Given that the McKinley Tariff went into effect on 6 Oct. 1890, the bilateral
American tariff toward Britain ought to register a marked increase between fiscal years
1889/90 and 1890/1, but no such increase is evident. The reason lies in the McKinley Tariff’s
postponement of an increase in the tariff on tinplate until 1 July 1891. In expectation of the
duty on tinplate rising from $0.01 to $0.022 per pound on 1 July 1891, American firms
imported an unusually large amount of tinplate during 1890/1.53 Whereas the United States
imported $20.9 million of British tinplate in 1889/90, she imported $35.6 million of British
tinplate in 1890/1.54 In 1890/1, the ad valorem equivalent tariff for tinplate was, at 29 per
cent, less than the bilateral American tariff toward Britain. Thus, in the calculation of
Equation 1 for 1890/1, increases in the per-industry ad valorem equivalent tariffs for cotton
manufactures and wool manufactures—increases that would have otherwise yielded a higher
bilateral American tariff toward Britain for 1890/91—are counterbalanced by the much
greater weight given to the comparatively low ad valorem equivalent tariff for tinplate.
Because of the delayed increase in the tariff on tinplate, the bilateral American tariff toward
Britain does not reflect the fullness of the McKinley Tariff until 1891/2, when the absolute
bilateral American tariff toward Britain rises from 35 to 43 per cent.
Between 1873/4 and 1889/90, the relative bilateral American tariff toward Britain
fluctuates within the narrow range of 1.2 and 1.4. In 1891/2, the relative bilateral American
tariff toward Britain swells to 2.6, and remains at a similarly elevated level through 1893/4.
This pronounced increase in the relative bilateral American tariff toward Britain is partly
attributable to a higher absolute bilateral American tariff toward Britain, but also attributable
to a reduced American tariff toward all countries except Britain. The McKinley Tariff was
noteworthy for lowering the duties on primary imports, very little of which came from
Britain. The duty on sugar, which regularly comprised over one-tenth of total American
imports, was eliminated altogether.55
The Wilson-Gorman Tariff of 1894 lowered the ad valorem equivalent tariffs for
many industries well represented within the composition of British exports to the United
States. Between 1893/4 and 1894/5, the ad valorem equivalent tariff for cotton manufactures
declined from 56 to 47 per cent; earthenware and chinaware from 58 to 35 per cent; iron,
steel, and manufactures thereof from 50 to 39 per cent; leather and manufactures thereof from
33 to 26 per cent; tinplate from 82 to 57 per cent; and wool manufactures from 97 to 57 per
50
Taussig, Tariff history, pp. 256-9.
Ibid., pp. 259-66. The schedule of duties on wool manufactures was byzantine because
individual articles of importation were subject to both specific and ad valorem duties, which in
tandem often disguised the actual extent of protection.
52
Foreign Commerce (1890, 1892).
53
Ibid., (1891).
54
Ibid. In 1890/1, tinplate accounted for 18% of American imports from Britain.
55
Ibid., (1874-1913).
51
13
cent.56 Moreover, the Wilson-Gorman Tariff removed all duties on wool. Indeed, with respect
to the absolute bilateral American tariff toward Britain, the claim of the early tariff historian
Percy Ashley that the Wilson-Gorman Tariff was one of ‘relatively little change’ simply
cannot apply.57 Moving from 1893/4 to 1894/5, the absolute bilateral American tariff toward
Britain falls from 45 to 31 per cent—a much sharper movement than occurs following the
McKinley Tariff. The relative bilateral American tariff toward Britain gradually returns to its
pre-McKinley level, assisted in this trend by the reimposition of duties on sugar.
With the passage of the Dingley Tariff of 1897, the pendulum swung back in the
direction of protectionism. Duties were once again levied on wool. The ad valorem
equivalent tariff for wool manufactures was compensatorily raised to essentially the same
level as under the McKinley Tariff. As with the McKinley Tariff, cotton manufactures and
tinplate were subjected to higher duties. At the same time, the Dingley Tariff also greatly
increased the ad valorem equivalent tariffs for the industries of alkali (from 31 to 50 per cent)
and earthenware and chinaware (from 34 to 55 per cent).58 Despite these several upward
shifts in the per-industry ad valorem equivalent tariffs, the absolute bilateral American tariff
toward Britain remains substantially below its level during the McKinley years. Furthermore,
the relative bilateral American tariff toward Britain only increases from 1.1 in 1896/7 to 1.6
in 1897/8, compared to an increase from 1.3 in 1889/90 to 2.6 in 1891/2, skipping over the
exceptional year of 1890/1 for reasons already mentioned.
It becomes evident that the McKinley Tariff and Dingley Tariff, both landmark pieces
of protectionist legislation, have fundamentally different effects on the bilateral American
tariff toward Britain, whether measured absolutely or relatively. Following the enactment of
the Dingley Tariff, the bilateral American tariff toward Britain does not reach the high level
of the McKinley years partly because the basket of American imports from Britain underwent
a greater compositional adjustment in the immediate aftermath of the Dingley Tariff. In order
to gauge the degree of compositional adjustment that occurred following the McKinley and
Dingley Tariffs, Laspeyres tariff indices are constructed by following the method advanced
by McCloskey. The industry-composite bilateral American tariff toward Britain, expressed in
Equation 1, is recalculated for the four years following the enactment of both tariff acts
(1890/1-1893/4 and 1897/8-1900/1) using fixed industry weights taken from the years
preceding the acts (1889/90 and 1896/7). Recall that the industry-composite bilateral
American tariff toward Britain pertains only to dutiable imports from Britain. The first
column of Table 3 presents the Laspeyres industry-composite bilateral American tariff
toward Britain. The second column presents the actual industry-composite bilateral American
tariff toward Britain. The third column notes the percentage deviation of the Laspeyres from
the actual industry-composite tariff, which can be interpreted as the extent of compositional
adjustment.
By the second year after the McKinley Tariff, compositional adjustment causes the
actual industry-composite bilateral tariff to deviate from the Laspeyres by 10 per cent.
However, by the second year after the Dingley Tariff, compositional adjustment causes the
actual industry-composite bilateral tariff to deviate from the Laspeyres by 18 per cent, or
nearly twice as much. One of the main reasons for this difference is that the Dingley Tariff
offered additional protection to industries that were already gaining a foothold in the United
States, domestic industries that could easily displace imports, if only given additional
protection. One such industry was alkali. The first major alkali works in the United States, the
Solvay Process Company, was established in 1881, but did not commence production until
56
Ibid., (1894, 1895).
Ashley, Modern tariff, p. 217.
58
Foreign Commerce (1897, 1898).
57
14
Table 3. Laspeyres tariff indices
(1)
Laspeyres industrycomposite bilateral
Year
American tariff toward
Britain (per cent)
1889/90 fixed weights
1890/1
47.9
1891/2
60.6
1892/3
62.5
1893/4
62.9
1896/7 fixed weights
1897/8
1898/9
1899/1900
1900/1
(2)
(3)
Actual industry-composite
bilateral American tariff
toward Britain (per cent)
Deviation between
Laspeyres and actual
tariffs (per cent)
57.1
61.9
59.1
58.4
43.1
55.0
57.8
57.6
10.7
9.8
7.9
8.8
50.7
52.0
49.6
48.7
11.8
17.5
17.4
18.1
Notes: Col. 3 is computed using the mid-point method. For the industry of flax and manufactures
thereof, the 1889/90 fixed weight is based upon burlaps and linens, whereas the per-industry ad
valorem equivalent tariff pertains to all manufactures of flax, hemp, and jute beginning in 1890/1.
This mismatch between the per-industry ad valorem equivalent tariff and its weight does not pose a
serious problem. Because the weight is based upon a more restrictive definition of the industry, the
less-than-average ad valorem equivalent tariff (38% in 1890/1) for the industry of flax and
manufactures thereof is underweighted in the Laspeyres tariff index, causing a slight overstatement of
the Laspeyres tariff index for 1890/1-1893/4. Hence, the mismatch serves as a bias against the finding
of this study.
Source: See text.
1884.59 Throughout the 1890s, output increased at a moderate pace until the passage of the
Dingley Tariff, which enabled the Solvay Process Corporation, along with other smaller
alkali works, to eradicate British alkali exports from the American market swiftly and
conclusively.60 Earthenware and chinaware, an industry that existed in the United States for
some time, also benefitted considerably from the Dingley Tariff, as American consumers
began to substitute domestic earthenware for British earthenware.61
Although both the McKinley and Dingley Tariffs sharply raised the ad valorem
equivalent tariff for wool manufactures, the substitution away from imports of British wool
manufactures was much greater following the latter act. Whereas between 1889/90 and
1890/1, American imports of British wool manufactures decreased from $29.1 to $19.5
million, between 1896/7 and 1897/8, her imports of British wool manufactures decreased
from $23.0 to $7.0 million.62 The post-Dingley falloff in American imports of British wool
manufactures can largely be explained by developments within one particular branch of this
industry: worsteds, which are manufactures of combed wool. American manufacturing of
worsteds grew by leaps and bounds in the 1880s and 1890s, with Clapham noting that, during
these two decades, the number of worsted combs increased by a factor of three, and the
number of worsted spindles by a factor of six.63 Unfortunately, the Foreign Commerce and
59
Haber, Chemical industry, p. 148.
Ibid., p. 216.
61
Saul, Overseas trade, pp. 162-3.
62
Foreign Commerce (1897, 1898).
63
Clapham, Woollen and worsted, p. 253.
60
15
Navigation of the United States does not provide a separate classification for worsted imports
in its article-country disaggregation section. However, the Annual Statements of the Trade of
the United Kingdom do, in fact, distinguish worsted exports in its article-country
disaggregation section. Considering the category of ‘worsted tissues, coatings, broad, all
wool’, the value of British exports to the United States proceeded as follows: £1.1 million
(1896), £1.1 million (1897), £0.2 million (1898), and £0.2 million (1899).64 Here, the impact
of the Dingley Tariff is unmistakable. Due to the expanding capacity of American worsted
factories, especially in the 1890s, the nearly complete substitution away from imports of
British worsteds was possible. The precipitous decline in imports of British worsteds was an
important part of the compositional adjustment that occurred in the immediate aftermath of
the Dingley Tariff.
The tinplate industry provides an interesting contrast to the industries of alkali;
earthenware and chinaware; and wool manufactures. Prior to the McKinley Tariff raising the
duty on tinplate, there was no domestic tinplate industry to speak of.65 The absence of any
pre-existing tinplate industry meant that the United States relied heavily on British exports
for several years after the McKinley Tariff; the degree of compositional adjustment was
accordingly limited. In 1889/90, the year before the passage of the McKinley Tariff,
American imports of British tinplate amounted to $20.1 million.66 In 1892/3, American
imports of British tinplate remained very high at $17.6 million, despite the ad valorem
equivalent tariff for the tinplate industry having more than doubled.67 All in all, the continued
reliance on tinplate imports from Britain hampered the degree of compositional adjustment in
the wake of the McKinley Tariff.
The Dingley Tariff was the longest-governing tariff act in American history,
remaining in effect until the Payne-Aldrich Tariff of 1909. During the first decade of the
twentieth century, a time of stability within the American tariff regime, the absolute bilateral
American tariff toward Britain, expressed as an ad valorem equivalent, slowly diminishes,
primarily owing to a trend of rising import prices. Because many duties were imposed on a
specific basis (e.g. $0.015 per pound of tinplate), rising import prices reduced the ad valorem
equivalent of the specific duties. Between 1897 and 1909, the price index of American
imports rose by 16 per cent, which is hardly insignificant.68 Since the trend of rising prices,
which began in the late 1890s, was a worldwide phenomenon, the American tariff toward all
countries except Britain also diminishes. With both the numerator (the bilateral American
tariff toward Britain) and the denominator (the American tariff toward all countries except
Britain) falling, the relative bilateral American tariff toward Britain remains more or less
constant between 1.1 and 1.4. Finally, there is little to say about the Payne-Aldrich Tariff,
which lacked the decisiveness of preceding tariff acts. The Payne-Aldrich Tariff slightly
reduced the ad valorem equivalent tariffs for certain industries, but these reductions were
minor and, for the most part, inconsequential.
As Federico and Tena have shown for the overall Italian tariff during the late
nineteenth and early twentieth centuries, a measurement of protection is sometimes quite
sensitive to the tariff imposed upon a single article of importation, in their case sugar.69 The
bilateral American tariff toward Britain, however, is generally little influenced by the ad
valorem equivalent tariff for any one industry, much less for any one article of importation.
Trade of the United Kingdom (1896-9). ‘All wool’ indicates that the worsted is composed
solely of wool, as opposed to a mixture of wool and some other textile material.
65
Irwin, ‘Tinplate industry’, p. 349.
66
Foreign Commerce (1890).
67
Ibid., (1893).
68
Lipsey, Foreign trade, p. 146.
69
Federico and Tena, ‘Protectionist country’, p. 81.
64
16
Ad valorem equivalent tariff (per cent)
Figure 2. Industry sensitivities, 1873/4-1912/3
50
45
40
35
30
25
20
15
10
5
0
Bilateral American tariff toward Britain
Bilateral tariff, excluding flax and manufactures thereof
Bilateral tariff, excluding wool manufactures
The sole exception, in this regard, is the tinplate industry. Excluding the tinplate industry
from the calculations in section II reduces the absolute bilateral American tariff toward
Britain from 43 to 39 per cent for 1891/2, thereby diminishing, though not eliminating, the
protectionist tenor of the McKinley Tariff for Britain.
American imports from Britain were greatest in the industries of flax and
manufactures thereof and wool manufactures.70 Yet, as Figure 2 indicates, excluding these
industries from the calculations results in hardly any discernible change in the bilateral tariff.
Indeed, the bilateral American tariff toward Britain emerges from a wide collection of perindustry ad valorem equivalent tariffs, none of which exert an overwhelming influence on the
whole.
IV
The bilateral American tariff toward Britain differs substantially from the overall
tariff of the United States, especially during the years following the McKinley Tariff of 1890.
This finding should alarm those economic historians and economists who, in a variety of
contexts, inappropriately associate overall tariffs with bilateral trade flows. The implication
of doing so is far from trivial. Whereas the coefficient of correlation between the log
differences of the overall American tariff and the value of American imports from Britain is 0.06, the coefficient of correlation between the log differences of the bilateral American tariff
toward Britain and the value of American imports from Britain is an entire order of
magnitude higher at -0.53. Although the author reserves a full analysis of the relationship
between foreign protection and late nineteenth-century British exports for another article, this
present article has demonstrated that any such analysis ought to entail bilateral measures of
foreign protection.
Footnote references
Aldcroft, D. H., ‘McCloskey on Victorian growth: a comment’, Economic History Review,
2nd ser., XXVII (1974), pp. 271-4.
70
Foreign Commerce (1874-1913).
17
Ashley, P., Modern tariff history: Germany—United States—France (3rd edn. 1920).
Ashworth, W., An economic history of England, 1870-1939 (1960).
Balassa, B., ‘Tariff protection in industrial countries: an evaluation’, Journal of Political
Economy, 73 (1965), pp. 573-94.
Beales, H. L., ‘I. The “Great Depression” in industry and trade’, Economic History Review,
2nd ser., V (1934), pp. 65-75.
Broadberry, S. N., ‘How did the United States and Germany overtake Britain? A sectoral
analysis of comparative productivity levels’, Journal of Economic History, 58 (1998),
pp. 375-407.
Clapham, J. H., The woollen and worsted industries (1907).
Clark, C. and Crawford, J. G., The national income of Australia (Sydney, 1938).
Clark, G., O’Rourke, K. H. and Taylor, A. M., ‘The growing dependence of Britain on trade
during the Industrial Revolution’, Scandinavian Economic History Review, 62 (2014),
pp. 109-36.
Corden, W. M., The theory of protection (Oxford, 1971).
Crafts, N. F. R., ‘Victorian Britain did fail’, Economic History Review, 2nd ser., XXXII
(1979), pp. 533-7.
Dedinger, B., ‘The Franco-German trade puzzle: an analysis of the economic consequences
of the Franco-Prussian War’, Economic History Review, 65 (2012), pp. 1029-54.
Federico, G. and Tena, A., ‘Was Italy a protectionist country?’, European Review of
Economic History, 2 (1998), pp. 73-97.
Feinstein, C., ‘Exports and British economic growth, 1850-1914’, in P. Mathias and J. A.
Davis, eds., International trade and British economic growth from the eighteenth
century to the present day (Oxford, 1996), pp. 76-95.
Haber, L. F., The chemical industry during the nineteenth century (Cambridge, 1958).
Harley, C. K., ‘International competitiveness of the antebellum American cotton textile
industry’, Journal of Economic History, 52 (1992), pp. 559-84.
Harley, C. K., ‘The antebellum tariff: different products or competing sources? A comment
on Irwin and Temin’, Journal of Economic History, 61 (2001), pp. 799-805.
Harley, C. K., ‘Trade, 1870-1939: from globalisation to fragmentation’, in R. Floud and P.
Johnson, eds., The Cambridge economic history of modern Britain, II, Economic
maturity, 1860-1939 (Cambridge, 2004), pp. 161-89.
Hatton, T. J., ‘The demand for British exports, 1870-1913’, Economic History Review, 2nd
ser., XLIII (1990), pp. 576-94.
Hawke, G. R., ‘The United States tariff and industrial protection in the late nineteenth
century’, Economic History Review, 2nd ser., XXVIII (1975), pp. 84-99.
Imlah, A. H., Economic elements in the Pax Britannica: studies in British foreign trade in the
nineteenth century (Cambridge, Mass., 1958).
Irwin, D. A., ‘Could the United States iron industry have survived free trade after the Civil
War?’, Explorations in Economic History, 37 (2000), pp. 278-299.
Irwin, D. A., ‘Did late-nineteenth-century U.S. tariffs promote infant industries? Evidence
from the tinplate industry’, Journal of Economic History, 60 (2000), pp. 335-60.
Irwin, D. A. and Temin, P. ‘The antebellum tariff on cotton textiles revisited’, Journal of
Economic History, 61 (2001), pp. 777-98.
Irwin, D. A., ‘New estimates of the average tariff of the United States, 1790-1820’, Journal
of Economic History, 63 (2003), pp. 506-13.
Jacks, D., Meissner, C. M. and Novy, D., ‘Trade costs in the first wave of globalization’,
Explorations in Economic History, 47 (2010), pp. 127-41.
Lipsey, R. E., Price and quantity trends in the foreign trade of the United States (Princeton,
1963).
18
McCloskey, D. N., ‘Did Victorian Britain fail?’, Economic History Review, 2nd ser., XXIII
(1970), pp. 446-59.
McCloskey, D. N., ‘Magnanimous Albion: free trade and British national income, 18411881’, Explorations in Economic History, 17 (1980), pp. 303-20.
Meyer, J. R., ‘An input-output approach to evaluating the influence of exports on British
industrial production in the late 19th century’, Explorations in Entrepreneurial History,
8 (1955), pp. 12-34.
O’Rourke, K. H., ‘Tariffs and growth in the late 19th century’, Economic Journal, 110
(2000), pp. 456-83.
Saul, S. B., Studies in British overseas trade, 1870-1914 (Liverpool, 1960).
Schlote, W., British overseas trade from 1700 to the 1930’s. Translated by W. O. Henderson
and W. H. Chaloner. (Oxford, 1952).
Sundararajan, V., ‘The impact of the tariff on some selected products of the U.S. iron and
steel industry, 1870-1914’, Quarterly Journal of Economics, 84 (1970), pp. 590-610.
Taussig, F. W., Some aspects of the tariff question: an examination of the development of
American industries under protection (Cambridge, Mass., 3rd edn. 1931).
Taussig, F. W., The tariff history of the United States (New York, 8th edn. 1931).
Official publications
Annual Statement of the Trade of the United Kingdom with Foreign Countries and British
Possessions (various issues).
British and Foreign Trade and Industry (P.P. 1903, LXVII).
Foreign Commerce and Navigation of the United States (various issues).
Tariff Acts Passed by the Congress of the United States from 1789 to 1909 (Washington,
1909).
19