Jump to ContentJump to Main Navigation
Soft Coal, Hard ChoicesThe Economic Welfare of Bituminous Coal Miners, 1890-1930$

Price V. Fishback

Print publication date: 1992

Print ISBN-13: 9780195067255

Published to Oxford Scholarship Online: October 2011

DOI: 10.1093/acprof:oso/9780195067255.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 17 January 2017

(p.234) Appendix B Sources of Data for Panel of Twenty-three Coal States from 1901 to 1930

(p.234) Appendix B Sources of Data for Panel of Twenty-three Coal States from 1901 to 1930

Source:
Soft Coal, Hard Choices
Publisher:
Oxford University Press

Throughout the book there are discussions of regressions based on the panel data set for twenty-three coal states from 1901 to 1930. The coal states include Alabama, Arkansas, Colorado, Illinois, Indiana, Iowa, Kansas, Kentucky, Maryland, Michigan, Missouri, Montana, New Mexico, Ohio, Oklahoma, Pennsylvania, Tennessee, Texas, Utah, Virginia, Washington, West Virginia, and Wyoming.

Accident Rates

Accident rates were calculated by dividing the measure of fatalities by a measure of hours worked. There were three measures of fatalities proposed in the text: underground deaths, underground deaths minus accidents killing five or more workers, and roof falls. The data for accidents killing five or more workers came from U.S. Bureau of Mines (BuMines), Bulletin 355, “Coal-Mine Accidents in the United States, 1930,” by W. W. Adams, L. E. Geyer, and L. Chenoweth (Washington, DC: U.S. Government Printing Office, 1932), pp. 94–100. Data on the remaining fatalities come from the following sources: for 1930, BuMines Bulletin 355, pp. 8–9; for 1925–1929, U.S. Bureau of Mines, Bulletin 341, “Coal-Mine Fatalities in the United States, 1929,” by W. W. Adams (Washington, DC: U.S. Government Printing Office, 1931), pp. 68–72; for 1922–1924, U.S. Bureau of Mines, Bulletin 283, “Coal-Mine Fatalities in the United States, 1926,” W. W. Adams (Washington, DC: U.S. Government Printing Office, 1927), pp. 58–62; for 1919–1921, U.S. Bureau of Mines, Bulletin 241, “Coal-Mine Fatalities in the United States, 1923,” by W. W. Adams (Washington, DC: U.S. Government Printing Office, 1924), pp. 31–36; for 1914–1918, U.S. Bureau of Mines, Bulletin 196, “Coal-Mine Fatalities in the United States, 1919,” by Albert H. Fay (Washington, DC: U.S. Government Printing Office, 1920), pp. 42–47; for 1901–1913, U.S. Bureau of Mines, Bulletin 115, “Coal-Mine Fatalities in the United States, 1870–1914,” by Albert H. Fay (Washington, DC: U.S. Government Printing Office, 1916), pp. 142–355.

Underground hours worked are calculated as the product of total average underground workers and days worked and average hours worked per day. Days worked is defined as the average number of days the mines were open in each state in the following sources: for 1921–1930, BuMines Bulletin 355, p. 82; for 1913–1922, BuMines Bulletin 241, p. 52; for 1901–1914, (p.235) BuMines Bulletin 115, pp. 142–355. The Bureau of Mines did not report the average number of days the mines were open in 1909.1 interpolated values for 1909 by calculating the ratio of net tonnage to total days worked (the product of average days worked and total employment) for 1908 and 1910 and then averaging them. I then used the average ratio to get an estimate of days worked for 1909. This technique gave odd estimates for Montana and North Dakota; so, missing values were left for those states. Average hours worked per day is calculated by using a weighted average of hours worked. The weights are the number of workers working that particular number of hours per day as a percentage of workers for whom hours per day were reported. The sources for these data include the following: for 1930, U.S. Bureau of Mines, Mineral Resources of the United States, 1930, Part IINonmetals (Washington, DC: U.S. Government Printing Office, 1932), (henceforth, to be referred to as BuMines Nonmetals, 1930), p. 654; for 1922–1929, BuMines Bulletin 341, pp.76–79; for 1919–22, BuMines Bulletin 241, pp. 44–46; for 1914–18, BuMines Bulletin 196, pp. 48–51; for 1903–30, BuMines Bulletin 115, pp 142–355.

The Bureau of Mines reported the number of underground workers directly for the years 1913 to 1930 in the following sources: for 1921–30, BuMines Bulletin 355, p. 82; for 1913–20, BuMines Bulletin 241, p. 53. The number of underground workers for 1901 to 1912 were interpolated using the total number of workers and benchmarks for the ratio of underground to total workers from 1902, 1909, and 1913. Ratios of underground workers to total workers were calculated for 1902 from Bureau of the Census, Special Report: Mines and Quarries, 1902 (Washington, DC: U.S. Government Printing Office, 1905) pp. 709–11 using the total average number of wage earners underground and above ground; for 1909 from Bureau of the Census, Thirteenth Census of the United States, Volume XI, Mines and Quarries, 1909, pp. 232–33; and for 1913 from BuMines Bulletin 115, pp. 142–355. The ratios for intervening years were determined by a straight-line interpolation between 1902 and 1909 and then between 1909 and 1913. The interpolated ratios were then multiplied by the average total number of coal workers, also found in BuMines Bulletin 115.

Coal Prices, Technological Variables, Strikes, Union Strength, and Mine Size

Coal prices are defined as the average coal price received at the mine per ton. They were calculated by the Bureau of Mines as the total revenue at the mine divided by total coal production. The sources include the following: for 1925–30, BuMines, Nonmetals, 1930, p. 648); for 1915–24, BuMines Bulletin 275, p. 55; and for 1901–14, BuMines Bulletin 115, pp. 142–355. The coal prices were turned into real coal prices by deflating by the Consumer Price Index for the United States (1967 = 100), series El35 in U.S. Bureau of the Census, Historical Statistics of the United States, Colonial Times to 1970 (Washington, DC: U.S. Government Printing Office, 1975), p. 211.

The percentage of tonnage mined with the machine cutter is reported directly in the document for 1901 to 1911. For 1911 and later, percent machine mined is the percent machine mined as a percentage of the total percentage of mines for which the technological information was reported. The sources for this information include for 1930, BuMines, Nonmetals, 1930, p. 661; for 1922–29, BuMines Bulletin 341, pp. 73–75; for 1919–21, BuMines Bulletin 241, pp. 37–38; for 1914–18, BuMines Bulletin 196, pp. 37–41; for 1901–14, BuMines Bulletin 115, pp. 142–355.

Potential strike variables include the ratio of men on strike to the average number of men employed and days idle on strike per employee. Information on the average number of men employed is discussed in the section on accident rates. Information on total days idle on strike and the number of men idle during strikes comes from the following sources: for the years (p.236) 1923–30, BuMines, Nonmetals for the following years 1930 (p.659), 1929 (p. 748), 1928 (p. 480), 1927 (p. 372), 1926 (p. 483), 1925 (p. 439); for 1919–22, U.S. Geological Survey, Mineral Resources of the United States, 1922, Part IINonmetals (Washington, DC: U.S. Government Printing Office, 1925), p. 517 (henceforth referred to as U.S. Geo. Survey, Nonmetals); U.S. Geo. Survey, Nonmetals, 1921, p. 505; U.S. Geo. Survey, Nonmetals, 1918, p. 723; for 1914–18, BuMines Bulletin 196, pp. 51–52; and for 1901–13, BuMines Bulletin 115, pp. 142–355. To calculate days idle on strike per employee in the rest of the United States for state i, I subtracted state i’s information for days idle (and employment) from the U.S. total and then calculated the ratio.

The relative strength of labor unions is based on data from the U.S. Coal Commission, Report, Part III (Washington, DC: U.S. Government Printing Office, 1925), p. 1052. The measure reported there is the ratio of paid up membership in the United Mine Workers of America as of November 30 to the average number of employees in each area as reported by the U. S. Geological Survey. The coverage of the variable is incomplete. It covers only the years 1902, 1905, 1908, 1912, 1915, 1918, 1921, and May 1923. Data is inserted for the remaining years prior to 1923 using straight-line interpolations between each pair of years for which data is reported. Several states were also combined in the reporting process, so the same value was inserted for each of the combined states. The combined states were: Arkansas, Oklahoma, and Texas; Colorado and New Mexico; Kentucky, Tennessee, and Virginia; Maryland and West Virginia; and Wyoming and Utah. There are missing data for the following states: North Dakota is not listed at all; Montana is not listed for 1902, 1905, and 1908; Washington for 1902; and Wyoming and Utah for 1902. The Coal Commission calculated the ratio for May 1923 using the number of employees for 1921. I readjusted the ratio so that the 1923 measure in the data base is the ratio based on the number of employees for 1923.

The mine size variable is the percentage of coal produced by mines producing less than 100,000 tons of coal per year. These figures are for commercial mines only and do not include wagon mines (small mines that deliver coal by wagon to the railroads). Most wagon mines were not regulated by state agencies. Mine size data was not available prior to 1909 and for the years 1915, 1916, and 1918. The remaining data come from the following sources: for 1922–30, BuMines, Nonmetals for the following years (pages in parentheses) 1930 (641–42), 1929 (713–14), 1928 (464–65), 1927 (396–97), 1926 (491–92), 1925 (461–64), 1923 (595–96); for 1909–14, 1917, and 1919–22, U.S. Geo. Survey, Nonmetals for the years 1922 (531–34), 1921 (520–23), 1917 (947–48), 1914 (610–13), 1912 (39–42), 1910 (36–39).

Wage Rates

The measure of real wages (listed in Table B-1) is the average hourly rate paid to daymen in each state deflated by the Consumer Price Index for the United States (1967 = 100), series E135 in U.S. Bureau of the Census, Historical Statistics. The wage rate data comes from two sources, Waldo Fisher and Anne Bezanson, Wage Rates and Working Time in the Bituminous Coal Industry 1912–1922 (Philadelphia: University of Pennsylvania Press, 1932) pp. 248–53 and 338–43 and U.S. Bureau of Labor Statistics, “Hours and Earnings in Bituminous Coal Mining, 1929” Bulletin No. 516, (Washington, DC: U.S. Government Printing Office, 1930, pp. 31–39. The Bezanson and Fisher study is based on data collected by the U.S. Coal Commission from fifty-two coal districts located in twenty-three states at six month intervals beginning January 1, 1912 and ending January 1 1923. The Coal Commission collected data on pick mining piece rates, machine mining piece rates, and daily wages for various occupations inside the mine. Bezanson and Fisher then undertook the task of aggregating the data to (p.237) the national level for piece rate workers and for day workers. The wage series used here for individual states are those for all inside day workers. Using the data for inside day workers allowed coverage of all twenty-three states in the sample. Furthermore, it reduced problems with comparing wage rates across states. As described in Chapter 5, piece rates were often adjusted to offset differences in the natural conditions of the mine and sometimes did not reflect all of the payments made to the miner because the miner might get paid extra for “deadwork.” The daily pay to those who were paid on a time basis in most cases tended to move in the same directions as the average hourly earnings of workers paid by the piece; therefore, the day wages seem to be a reasonable approximation of the direction in the pay of piece rate workers. Piece rate workers and daymen were hired in the same labor market. Correlations of daymen’s hourly wages and hourly earnings for piece rate workers were 0.902 for cross-state data from 1922 and 0.895 for pooled data for 1922 and 1924. The correlations are based on evidence from U.S. Bureau of Labor Statistics, “Hours and Earnings in Bituminous Coal Mining, 1929,” Bulletin No. 516, (Washington, DC: U.S. Government Printing Office, 1930), pp. 27–35.

Bezanson and Fisher reported index values for the wages at the district level and at the national level but not aggregated to the state level (the level of aggregation for the remaining variables in the panel). For states containing several districts, I aggregated the values from the districts to provide a state wage. The first step was to calculate the average wage index for each year for each district, which was the average of the wages for December 31 the previous year, June 30 of the current year, and December 31 of the current year. These indexes were found on pp. 248–53 in Bezanson and Fisher. The indexes were then multiplied by the base wage rate for December 31, 1920. The average wage for the state was then calculated as a weighted average of the district wages with the weights determined by the district’s share of bituminous coal produced that year (see pp. 338–43). This was done for all wages from 1912 to 1922. The wage for 1923 is based on the index from January 1 of that year.

Although the analysis of wages in the text was limited to the data for 1912 to 1923 above, the wage series can be extended to 1924, 1926, and 1929 using BLS data for the following states: Alabama, Colorado, Illinois, Indiana, Kentucky, Ohio, Pennsylvania, and West Virginia. For each state and year (1922, 1924, 1926, and 1929) the average wage for daymen is created as the weighted average of the wages across ten underground dayman occupations: brakeman, bratticeman and timberman, eager, driver, laborer, motorman, pumpman, trackman, trapper, and other. The weights were based on the number of workers in each occupation as a share of workers in all dayman occupations underground in that state and year. The two-wage series can then be spliced together using the ratio of the 1922 BLS wage (based on surveys taken between December 1921 and January 1922) to the December 1921 wage calculated from the data provided by Bezanson and Fisher.

Workers’ Compensation Legislation

The Worker’s Compensation law variable was compiled primarily from the volumes in the Bureau of Labor Statistics’ Workmen’s Insurance and Compensation Series. The law was assumed to go into effect at the beginning of the year after passage. The BLS volumes in the Workmen’s Compensation Series were not published every year, so I treated them as benchmark volumes. If there was no change in the status of the compensation law in the benchmark volumes, the variable was given the same value for the years between the benchmarks. If there was a difference in the law in the benchmark volumes, I examined the BLS’s annual descriptions of changes in labor law in particular volumes devoted to labor legislation in a particular year or in volumes of the Monthly Labor Review until I found the (p.238) change in the law. The following Bulletins of the Bureau of Labor Statistics in the Workmen’s Insurance and Compensation Series (date in parentheses) were used as benchmarks for the compilation: 92 (1911), 126 (December 1913), 189 (October 1915), 243 (September 1918), 272 (January 1921), 301 (April 1922), 332 (June 1923), 379 (January 1925), 423 (July 1926), and 496 (January 1929). The final benchmark volume was Bureau of Labor Statistics, “Handbook of Labor Statistics, 1931,” Bulletin No. 541 (Washington, DC: U.S. Government Printing Office, 1931), pp. 891–909.

The Employer Liability Law variable was constructed using information from Lindley D. Clark, “The Legal Liability of Employers for Injuries to Their Employees in the United States,” U.S. Department of Labor Bulletin No. 74, Vol. 16 (Washington, DC: U.S. Government Printing Office, 1908). Additional dating of changes in the employer liability and workers’ compensation laws came from examination of the annual or biennial state law volumes (for example, Laws of Indiana) and the compilations of statutes (for example, Burn’s Annotated Indiana Statutes).

State Mining Legislation and Enforcement

The information on inspectors, inspectors’ salaries, and the state laws were compiled from a variety of sources. The initial sources used as benchmarks included the following: U.S. Commissioner of Labor, “Labor Laws of the United States 1907,” 22 nd Annual Report of the Commissioner of Labor (Washington, DC: U.S. Government Printing Office, 1908); U.S. Bureau of Labor Statistics, “Labor Laws of the United States with Decisions of the Courts Relating Thereto, April 10, 1914,” Bulletin No. 148 (Washington, DC: U.S. Government Printing Office, 1914); and U.S. Bureau of Labor Statistics, “Labor Laws of the United States with Decisions of the Courts Relating Thereto, May 1925,” Bulletin No. 370 (Washington, DC: U.S. Government Printing Office, 1925). From there the intervening years were filled in by examining the annual or biannual state statute volumes and the law compilations for various years from all of the states in the study. As with workers’ compensation legislation, if there was a change in the law in a particular year, the change is not reflected in the data until the following year. Prior to statehood, Oklahoma (1907) and New Mexico (1912) came under the jurisdiction of U.S. territorial mining law.

Quantifying the laws was relatively straightforward once the definition of the law was established. There were relatively few problems with the laws concerning dusting the mines, requirements for fire bosses, provision of timbers by the operator, insulation of electric wires, riding on cars, use of permissible explosives, tests for inspectors, inspectors’ ability to close the mines without a court order, police power, number of foremen’s visits, licensing of foreman, licensing of miners by a state board, and licensing of miners by the foreman.

Compiling the number of inspectors and their salaries and therefore the budget for inspectors’ salaries described in the text was considerably more complex. Typically, the state mining law established the number of inspection districts, the number of inspectors, their annual salaries, and the role of the chief of the mine department, which varied across states. Changes in salaries and the number of inspectors were usually legislated as amendments to the mining law. In many of the states the appropriations legislation also contains information on the budget for salaries of the mine inspectors. In most states both sources agree. Where possible, appropriations information is used as the basis for the salary budgets. Since the accident data is based on the calendar year and the states had a variety of fiscal years, the salary budgets are adjusted to the calendar year using appropriate weights.

The following states contained complete appropriations information: Alabama, Arkansas, Missouri, New Mexico, Virginia, West Virginia, and Wyoming. The Missouri appropriations arc adjusted to reflect the fact that only part of the resources appropriated were devoted to (p.239) coal mining inspections. These adjustments are based on the explicit allocations described in the mining statutes. In Virginia the state mining statutes actually provided for a mine inspector in 1912, but the appropriations legislation shows that the actual funding establishing the inspector’s position did not occur until 1914.

In several states the appropriations were made in the mining legislation itself or no appropriations were listed. In those states the information is based purely on the statutes. New legislation on salaries or numbers of inspectors was treated as if it went into effect in the beginning of the fiscal year after the date of the legislation. For example, legislation approved in Illinois on June 5, 1911, was treated as going into effect at the start of the following fiscal year, beginning July, 1, 1911, and ending June 30, 1912. An exception was made for legislation that passed within a month after the beginning of the fiscal year. In all these cases it is clear that the change was designed to take place immediately. A law in Tennessee that passed on April 15, 1907, was treated as going into effect at the beginning of the fiscal year starting March 19, 1907, and ending March 18, 1908. Only statute information was used for each of the following states: Illinois, Indiana, Kentucky, and Michigan. In Indiana the appropriations and the statute information are dissimilar; data on disbursements are consistent with the statute information and are therefore used as the basis for inspectors’ salary budgets.

In several states appropriations data were available for only part of the period; therefore, a mixture of statute and appropriations data was used. In those cases the rules above for the date the statute went into effect are followed. The mixture was used in the following states (with the period for which appropriations were used); Colorado (appropriations through March 31, 1913), Iowa (appropriations after July 1, 1919), Kansas (appropriations throughout except mine inspectors’s salaries up to June 30, 1907; expense payments and salaries were not separated in the appropriations, but a daily salary was stated in the statutes; inspectors were treated as working 365 days a year), Maryland (appropriations to September 30, 1922), Montana (appropriations for State Coal Mine Inspector through February 28, 1916), Ohio (appropriations except for 1919, when a statute change took precedent over the appropriations legislation), Oklahoma (appropriations through June 30, 1929), Tennessee (appropriations after March 19, 1917), Texas (appropriations after August 31, 1909), Utah (appropriations through 1916 but missing values thereafter because mine inspection came under the Industrial Commission, which provided no detailed information after that date), and Washington (appropriations through March 31, 1919). In Pennsylvania the appropriations for mine inspectors were confusing because the mining department inspected both anthracite and bituminous mines. Statute information was used for the inspectors’ salaries, while information on the number of bituminous districts and inspectors comes from the Report of the Department of Mines of Pennsylvania, Part 11Bituminous, 1929–1930, p. 53. The years listed on that page were treated as ending May 31 of the year listed, which was the end of Pennsylvania’s fiscal year.

Several other factors deserve note. Both Indiana and Illinois state law allowed for appointments of county inspectors beyond the district inspectors. I found no appropriations for these inspectors in the state acts and no mention of them in the annual reports of the bureaus of mines in these states, so 1 left the county inspectors out of the calculations. In Kansas the number of inspectors was not listed directly in the state mining law. Instead they were included in laws for a mine safety association and then under the aegis of the commissioner of labor and industry. In later years the assistant commissioner of labor and industry was the ex officio chief inspector of mines. After 1917 the Montana industrial board oversaw mine inspection; as far as could be told Montana had only one inspector of coal mines at a constant salary from that time forward. Determining salaries in some cases was also complex. In some states travel expenses were mentioned as part of the mining law. However, the coverage was so scattered that I was unable to obtain a consistent series for such expenses and left them out.

In determining the number of inspectors, the role of the chief inspector also became an issue. I constructed an alternative budget for inspectors’ salaries across states in which chief inspectors whose job descriptions did not include direct inspections of mines were excluded from the budget. However, the correlation between the alternative budget and the budget in the text was above 0.9, so I did not pursue this line of inquiry further.

(p.240)

Table B-1. Average Hourly Wage Rates by State for Inside Daymen (in 1967$)

State

1912

1913

1914

1915

1916

1917

1918

1919

1920

1921

1922

Alabama

0.65

0.66

0.66

0.65

0.66

0.66

0.88

0.86

0.86

0.90

0.73

Arkansas

1.18

1.17

1.15

1.14

1.08

1.21

1.36

1.23

1.29

1.72

1.83

Colorado

1.03

1.21

1.26

1.24

1.17

1.29

1.42

1.28

1.34

1.63

1.44

Illinois

1.16

1.15

1.14

1.13

1.09

1.20

1.35

1.21

1.28

1.70

1.82

Indiana

1.19

1.18

1.16

1.15

1.11

1.22

1.36

1.23

1.29

1.72

1.84

Iowa

0.92

0.91

0.90

0.89

0.85

0.94

1.06

0.96

1.01

1.35

1.44

Kansas

1.20

1.19

1.18

1.16

1.09

1.23

1.38

1.24

1.30

1.73

1.85

Kentucky

0.85

0.83

0.81

0.80

0.99

0.90

1.08

0.98

1.12

1.34

1.32

Maryland

0.81

0.79

0.78

0.78

0.76

1.12

1.33

1.20

1.27

1.71

1.82

Michigan

1.19

1.18

1.16

1.15

1.11

1.22

1.37

1.25

1.32

1.76

1.88

Missouri

1.20

1.19

1.17

1.16

1.09

1.23

1.38

1.24

1.31

1.74

1.86

Montana

1.52

1.50

1.48

1.46

1.38

1.45

1.56

1.40

1.44

1.88

2.01

New Mexico

1.32

1.29

1.29

1.32

1.23

1.30

1.44

1.30

1.43

1.71

1.43

Ohio

1.11

1.10

1.08

1.07

1.04

1.15

1.30

1.17

1.23

1.65

1.76

Oklahoma

1.16

1.14

1.13

1.12

1.05

1.198

1.33

1.20

1.26

1.67

1.75

Pennsylvania

1.03

1.01

1.00

0.99

0.98

1.15

1.34

1.21

1.27

1.59

1.62

Tennessee

0.66

0.64

0.63

0.63

0.59

0.70

0.91

0.84

0.97

1.22

1.01

Texas

1.20

1.19

1.17

1.16

1.09

1.22

1.37

1.23

1.30

1.73

1.73

Utah

1.38

1.34

1.33

1.31

1.23

1.33

1.50

1.36

1.41

1.84

1.96

Virginia

0.68

0.69

0.69

0.69

0.67

0.76

0.99

0.96

1.12

1.39

1.13

Washington

1.42

1.39

1.37

1.35

1.28

1.35

1.49

1.34

1.36

1.73

1.64

West Virginia

0.74

0.74

0.75

0.74

0.74

0.92

1.11

1.02

1.17

1.53

1.39

Wyoming

1.45

1.43

1.41

1.40

1.31

1.36

1.49

1.35

1.39

1.83

1.97

United States

1.02

1.00

0.99

0.98

0.97

1.09

1.26

1.14

1.22

1.57

1.56

Source: See text of wage rate section of Appendix B. The deflator does not adjust for differences in the cost of living across states.

Another problem in calculating the budget for salaries of inspectors arose in determining the role of the chief inspector and deputy chief inspector of the Department of Mines in Pennsylvania, where they oversaw both bituminous and anthracite mine inspections. In this case I multiplied their combined salaries by the number of bituminous districts as a percentage of the number of all mining districts in Pennsylvania and then added this figure to the budget. A similar situation arose in Missouri, where the chief of the Department of Mines oversaw two coal mine inspectors and four inspectors of other mines. I allocated one-third of his salary to the coal mining inspection budget.

In all the states except Oklahoma prior to 1929 and Tennessee the mining laws specifically established the number of coal mine inspectors. In these two states the mining laws were more general. To establish what share of the inspection budget should be allocated to coal mining in these states, I examined the number of coal mines as a percentage of all mines reported in the United States Mining Censuses of 1902, 1909, 1919, and 1929. This examination showed that the coal mine share of all mines in Tennessee was 76 percent in 1902, 73 percent in 1909, 82 percent in 1919, and 87 percent in 1929. On this basis I assumed that 80 (p.241) percent of the Tennessee mine inspection budget was spent inspecting coal mines between 1901 and 1930. For Oklahoma the percentages were 65 percent in 1909, 52 percent in 1919, and 55 percent in 1929, and I assumed that 60 percent of the Oklahoma mine inspection budget was expended on inspecting coal mines between 1907 and July 1, 1929.