kEmployment Segregation in the Early Twentieth Century
Cliometric Sessions at 1990 ASSA Meetings--December 28, 10:15 AM

1

Getting Started in the Auto Industry:
Black Workers at the Ford Motor Company, 1918-1947

Warren Whatley
University of Michigan
and
Gavin Wright
Stanford University

1. Introduction

The Ford Motor Company has a unique place in the economic history of black Americans. On the eve of World War 11, Ford employed more than 10,000 black workers, a far larger number and a far greater representation than at any other firm in the automobile industry. Whereas other auto companies confined blacks to menial positions, only Ford had black workers on the assembly line and in other important production jobs. Because the wage levels in the auto industry were high even by the standards of other high-wage northern industries, Ford's black workers earned incomes which were well outside the normal range of black opportunities. Though Ford's unusual racial history figures prominently in accounts of race relations and unionization (e.g., Meier and Rudwick 1979), the labor market dimensions of Ford's policies have yet to be analyzed. This paper offers a preliminary assessment, drawing on newly developed evidence from the company's personnel records.

The central issue is this: Did Ford pursue a nondiscriminatory policy, and thereby gain a competitive advantage over its rivals (as many economic theories hold)? Or did Ford itself discriminate against blacks, and if so how: in wages, job assignments, promotion opportunities, or in some other way? Inevitably, questions about the character of company policies will lead to further questions about the characteristics of black labor supply: If blacks were treated differently, were these distinctions mere race prejudice, or were they attributable to differences in qualifications, such as lower educational attainment? Or, did the company make racial distinctions which might be attributed to "statistical discrimination," i.e., based on the relatively poor performance of the average black worker? These are the sorts of questions which our evidence allows us to address.

These historical issues are relevant to current debates over black economic progress, and over the efficacy of governmental programs to ameliorate racial inequality. It is now widely believed that that the major governmental initiatives dating from the 1960s have been ineffective; and that long-term relative black progress is largely attributable to long-term progress in education and skill levels (Smith 1984; Smith and Welch 1986). Of course, migration from the South plays an important role even in this simple supply-driven story. But that role is chiefly seen as liberation from segregation: from inferior black schools in the South, and perhaps also from the overt restrictions on black occupational status in that backward region. Once blacks began to enter the fast-moving, cosmopolitan, competitive environment of modern industrial America (it would seem), their progress was in their own hands. Is this scenario accurate? In fact, remarkably little is known about the experience of black Americans in the cities and industries of the northeast and midwest between World War I and the 1950s. It is difficult to see how we can judge the effectiveness of the policies of the 1960s, without knowing the operation of the labor market for black workers in earlier times. A study by Heckman and Payner (1989) of desegregation in the textile industry during the mid- 1960s has had a big impact on the ongoing discussion, by taking up a specific example in detail, in contrast to the aggregated correlations so often presented. We hope to follow their lead for northern employers, beginning with the Ford Motor Company.

II. The Historical Context: Trends in Black Employment

Between 1890 and World War I, American industry surged into a position of global pre-eminence, but for the most part this feat was accomplished without black labor. The bulk of the black population stayed in the South after emancipation, and prior to World War 1, black representation was negligible in the major industries of the northeast and midwest. In 1910, blacks comprised 1 percent or less of the industrial workers in New York, Cleveland, and Chicago; and less than 4 percent in Pittsburgh and Philadelphia. Detroit's black representation was among the lowest in the country, at one-half of one percent of industrial workers. A sample of male workers in northern cities in 1910 shows that blacks made up about 4 percent of overall employment, and most of this relatively small number were concentrated in such low-paying occupations as domestic and personal service, or in eating/drinking establishments or hotels (Table 1).

Because of the labor shortages of World War 1, blacks from the South were heavily recruited by northern employers in a wide range of industries. Recruiters flooded the entrepots of the South, such as Birmingham, Alabama, offering "justice tickets" which allowed migrants to use their belongings as collateral in exchange for fare north and a percentage of their first year's wages. After this initiation, the migration developed a momentum sustained by letters and assistance from those already established in the north (U.S. Department of Labor 1917). The heavy volume of black migration continued during the 1920s, despite the drastic slowdown in northern industrial employment growth. It does not seem that the ongoing migration after 1920 was chiefly driven by the pull of demand for black workers, but by the push of the boll weevil and declining farm wages in the South, coupled with the opportunities opened up by new lines of communication, and by the new black presence in northern industry. Between 1910 and 1940, there were disproportionate increases in the black share of employment in industries like steel, railroads, and meatpacking (Table 1). Probably the largest black increase in percentage terms was that of the auto industry, overwhelmingly the result of policies of the Ford Motor Company.

The experience of 1915-1940 was a dramatic break with the past, but it is equally important to keep several things in perspective. First, Table 1 shows that blacks remained in 1940 a distinctly small minority in northern cities and industries, a much smaller share of employment than they would later become. Secondly, the influx of black population was considerably faster than the growth of black industrial employment. This may be seen in Table 1 from the observation that easily the largest black growth sector in absolute terms was "Construction;" but most of that growth disappears when one excludes the self-employed, the unemployed, and government employees (which includes those on federal work-relief projects). There were also large absolute increases in domestic and personal service, and laundry work. Thus, for most black households, auto industry jobs were prime opportunities. Black auto workers' average annual earnings in 1939 exceeded the average for all workers in all manufacturing industries, and were nearly twice the average for all black manufacturing workers (Bailer 1943, 419-420).

The representation of blacks was clearly not uniform among industries, among firms and plants, and among occupations within firms and plants. There is widespread testimony that blacks in many industries were assigned only to the most unpleasant, unskilled jobs, and that they met with severe discrimination in promotion opportunities (Gottlieb 1987, 89, 101; Chicago Race Commission 1922, 365-66, 385-91). In the auto industry, the number of black workers at Ford was far larger than at any other firm, and the black percentage was three to five times that of the other large companies (Table 2). Before World War 11, the Fisher Body plants of General Motors were wellknown for their "lily-white" policies, as were many others (Bailer 1943, 420). Indeed, it was reported that before the war blacks did not work on the assembly line anywhere except at Ford, though there may have been scattered exceptions (Northrup 1968, 10). Though they were a small minority, it is clear that blacks could not simply blend invisibly into the employment structure of American industry. The high sensitivity to race among white workers was well demonstrated during World War 11, by numerous incidents, walkouts and protests when blacks first entered what had previously been all-white departments and plants (Bailer 1944, 566-69; Denby 1978, 98).

Even at Ford, policies could hardly be called color-blind. Henry Ford's willingness to hire black workers was said to have been a response to appeals from black ministers during the depression of 1920-21. Thereafter, when the giant River Rouge plant opened in the 1920s, the black employment percentage held firm. But those blacks who had been employed at Highland Park were moved to the Rouge Plant, and few blacks were employed elsewhere (Table 3). It was widely reported that blacks were disproportionately assigned to the least-desirable jobs such as the foundry (Peterson 1987, 25). A frequent complaint was that opportunities for promotion were unequal. Seniority rules were applied asymmetrically, being used against blacks when layoffs were required, but being qualified by "ability" requirements when openings had to be filled. "Ability" could be defined so as to include "acceptability" to others in a department (Bailer 1944, 565). Blacks were denied access to apprenticeships and training which would qualify them for skilled positions. As late as 1937, the Wilbur Wright Trade School in Detroit refused admission to blacks, on the grounds that blacks would never be accepted as skilled workers (Peterson 1987, 28).

It is not simple to characterize Ford's philosophy or economic strategy with respect to employment, wages, or racial equity. From the time of the five-dollar day in 1914, Ford has been seen as the epitome of the high-wage American employer (Meyer 1981). The sociological department established at the same time seems to presage the adoption of internal-labor-market strategies by many other employers, implying more careful screening and more lasting attachments between firms and workers. Yet Ford itself ceased to be "Fordist" after 1921, abolishing the sociological department and instituting a regime of speedup and continuous insecurity (Lewchuk 1987, 63; Klug 1989, 63). Wages at other companies had caught up to those at Ford, and Ford's reputation as an employer was dismal. It is plausible that the company could have exploited what was in effect monopoly power over black workers, the result of the refusal by other high-wage employers to take them. Herbert Northrup writes that before World War 11, "Ford could take its pick of one of every four Negroes who applied" (Northrup 1968, 20). Until now, however, we have not had the detailed records of black workers' experience at Ford to determine whether this conjecture is true.

III. The Ford Motor Company Employee Records

One of the legacies of the Five-Dollar Day era at Ford that was not abolished during the 1920s was the personnel department's detailed records on every Ford employee. Several years ago the company transferred over 1.5 million employee records from the River Rouge plant to the Inactive Records department at their industrial archive in Redford, Michigan. After two years of correspondence, we obtained permission to sample these records. Four different samples containing 4305 employee records have been collected. The first is a random sample from the entire collection (N=2179), and the second is a random sample of all black workers (N=1064). These have been supplemented by two intensive samples from the period after June 1943 (N=532) and the period between April 1922 and March 1926 (N=530), the two periods when the EDUCATION category was most frequently filled out. The coded information includes the following:

PLANT (River Rouge, Highland Park, Fordson, Lincoln, Willow Run, and several smaller plants)
AGE
SEX
ENGLISH (language)
NATIONALITY
EDUCATION
PERFORMANCE
TERMINATION reason for leaving Ford)
BIRTHPLACE
PREVIOUS EMPLOYMENT
WAGE
CODE (designates change in occupation, wage or department)

Cleaning of these data has only recently progressed to the point that they are available for analysis. On a preliminary basis, the samples seem consistent with other available information about black workers at Ford. When the individual employee histories are converted into labor force "stocks" as of the beginning of each year, the estimated black shares for the late 1930s are very close to those reported by Bailer for that period. Overall, 7.15 percent of Sample 1 is black, and 31.3 percent of the combined samples is black.

Figures 1-3 show that historical periods are represented unevenly in the Ford samples. because of the extreme fluctuations in the rate of hiring. Both for blacks and whites, the sample is dominated by the cohorts of the 1920s; the peaks in hiring during World War I were exceeded in four separate years during this decade (1922, 1923, 1925, and 1928). More than half of the cumulative total of employees were hired before 1929 (Figure 2). The estimated black share of new hires fluctuated roughly between six and ten percent from 1918 to 1943, before jumping to much higher levels in the 1940s (Figure 3). The black share of total hourly employment averaged around ten percent, primarily at the Rouge Plant after the mid-1920s (Figure 4).

Because the economic experience of entering workers depended in obvious ways on the state of the firm and the economy in the years of their employment, most of the interracial comparisons presented here are restricted to cohorts with the same year of first hire. We have looked for differences between white and black workers in such dimensions as entry level wages; attributes at time of hire (age, marital status, years of education); length of job tenure; and the rate of wage growth during job tenure. Fundamentally, we want to know whether Ford's economic treatment of black workers was systematically different from the treatment of whites; if it was different, whether such discrimination could perhaps be "justified" by differences in observed worker attributes or behavior (such as a tendency to high departure rates); and whether these patterns show any significant changes over time. Concerns about unobservable heterogeneity in labor quality, which commonly plague studies of this type, are greatly minimized here by the simple consideration that our data come from the very records the company itself chose to collect and utilize for exactly these purposes.

IV. Black and White Workers at Ford: Preliminary Results

The first question we address is whether blacks were hired by Ford at the same initial wages as white workers. Previous studies have indicated that racial discrimination was mainly reflected in job assignments, and not in lower wages for the same job (Bailer 1943, 419; Peterson 1987, 27). This characterization is consistent with more broadly-based evidence on entry- level wages in northern industries, which shows virtually no racial difference, in contrast to the situation in the South at that time (Table 4). Note that this evidence goes beyond the mere assertion that black and white workers were not given identical job assignments at different wage rates. The data in Table 4 are also inconsistent prima facie with discriminatory assignment to lower-paying initial jobs on the basis of race. Contrary to expectations, the 11.1 percent advantage for blacks in Michigan is not a direct reflection of Ford's employment policies, because the automobile industry was not included in the survey. But the margins shown in favor of blacks in Michigan and several other northern states are almost surely compositional; the survey reported that 90 percent of the northern firms (of those hiring both blacks and whites) paid the same entry wage to both races (Perlman and Frazier 1937, 1498). Did this pattern hold true at Ford, whose heavy employment of blacks was so different from other firms in the industry? Table 5 presents mean entry-level wages for blacks and whites at the four largest Ford plants: River Rouge, Highland Park, Lincoln, and Willow Run. It is evident that there was no systematic differential. There are occasional gaps in particular years (mainly years where little hiring occurred), but these are as likely to favor blacks as whites. In almost all of the years of heavy hiring, the entry wages for blacks and whites were within a point or two of each other. It appears, therefore, that there was no racial differentiation in initial wage levels, neither explicitly nor implicitly in the form of assignments to lower-wage Jobs.

It remains possible that entry-level wage discrimination might have taken another implicit form. If the qualifications for employment were in effect higher for blacks than for whites, then equal-initial-wages could be considered discriminatory in terms of payment relative to labor quality. Table 6 presents evidence on worker attributes by race, for three traits that a reasonable employer might consider relevant: age, marital status, and education. The close correspondence of age at first hire for blacks and for whites is striking. Workers of both races took their first Ford jobs at about age 28 to 30, and this level shows virtually no change over the thirty years recorded here. It is interesting, however, that a higher percentage of new black workers were married (about 2/3, as opposed to a little over 1/2 for whites). The racial "marriage gap" persisted over the thirty-year period, suggesting that it reflects either an underlying behavioral difference or an employment policy. At the time of the Five- Dollar day, Ford certainly did take an active interest in the household stability of its workers, as an indication of the likely duration of employment. It is plausible that the company might have been more concerned with turnover for blacks than for whites, and hence gave greater relative preference to black married men. But it is also possible that the gap arises from differential self-selection on the part of black and white workers: married white men may have found work at Ford unattractive relative to other opportunities, whereas married black men had fewer alternatives. Finally, the difference may merely reflect an earlier age at marriage for blacks in the relevant labor pool. The most surprising columns in Table 6 are those for education. Though the figures must be viewed with caution because the question was not always answered (and the probability of response may not have been random), it is striking that racial differences in years of education were small. During 1918-20 blacks had just as many years of education as whites, and during the heavy-hiring years of 1923-25 the average difference was less than a year. In 1928 new black hires had more years of education than whites. It appears that entry requirements were essentially fixed at an eighth-grade standard, which had perhaps drifted up to 9th grade by the 1940s; but these standards were roughly the same for blacks as for whites. One can imagine possible biases in these simple racial comparisons, such as selective responses or relative educational quality. But nothing in the evidence that the company itself collected implied that black workers were any less attractive than whites. Black workers at Ford did not constitute a grossly inferior, uneducated, underpaid segment of the work force.

One may be tempted to wave away this entry-level evidence, on the grounds that terms of hiring essentially reflected conditions in the open market for unskilled labor, which was more-or-less a spot market in character. In such a market unjustified racial distinctions are difficult to maintain. A better test of Ford's racial policies would focus instead on the post-entry experience of blacks compared to whites. Table 7 presents data on mean job tenure by race, for workers grouped as before according to year of first hire. In an effort to improve comparability, two additionally refined breakdowns are presented: mean tenure for only those workers who stayed long enough to get a wage increase; and for only those workers who entered the relevant modal wage. The latter restriction tries to pull out those who entered at below-normal wages as apprentices or at above- normal wages as skilled workers. The former restriction hopes to exclude the large number of short-term workers of both races, who would never have been candidates for promotion because they did not stay.

These refinements do not seem to have been necessary for our purpose, because Table 7 shows that whichever pair of columns one chooses, there was no consistent racial difference in turnover or average job tenure. There are large differences in occasional years, but they go in both directions. They may merely reflect unevenness in the timing of hires by race within the year. Mean tenure is a crude measure, since it does not distinguish voluntary quits (which might be the basis for "statistical discrimination" in promotions) from layoffs (which would themselves be reflections of company policies). But the Table 7 does not suggest that there were significant racial distinctions of either type. Figure 5 displays black-white ratios graphically for all of the indicators discussed thus far. There are year-to-year movements in both directions, but for every measure except percent married, the graph oscillates around a ratio of unity. On the basis of this information, an employer would have a difficult time justifying discriminatory policies on the grounds that blacks were less reliable or less committed to industrial work than whites.

V. Regression Analysis

Like most other northern employers, Ford evidently did not have an explicit or institutionalized wage differential by race. But this fact does not necessarily mean that Ford did not practice wage discrimination. Following the usage conventional in labor economics, wage discrimination occurs when workers who are otherwise comparable are paid unequally on the basis of race. To test for discrimination of this type, we have run regressions of the entry wage as a function of worker characteristics, including age at first hire, marital status, gender, education, participation in an apprenticeship program, and race. We also test for the interaction of race with each of the other variables. The results are presented in Tables 8, with and without the education variable (which is available on only about 40 percent of the observations). In both regressions, all of the standard variables have the expected signs and statistically significant coefficients. The regressions also included a series of plant and wage- regime dummies, whose coefficients are not displayed in the table.

Surprisingly, the coefficient on race is significantly positive, suggesting at first glance that Ford actually paid a positive wage premium to blacks, ceteris paribus. This is not the case, however, because the coefficients of the interactive variables (race interacted with age, marital status, and education) are all negative. The largest of these negative effects is race*age-at-hire. Combining the positive race intercept with the negative interactions and evaluating the wage at the means, we find that black workers of "average" characteristics earned a beginning wage which was lower than that of comparable white workers by 2.6 percent according to regression 1, or 1.6 percent according to regression 2.

Discrimination rates in the neighborhood of 2 to 3 percent may seem inconsequential. The results are relevant, however, not so much because of the magnitude of the implicit discrimination, but because of what they say about Ford's evaluation of black applicants. Attributes which warranted higher pay for whites (age, marriage, education) received little or no weight when blacks were involved. The regressions actually imply that there was no positive relationship at all between the black wage and age or marital status, though these coefficients were strongly positive for whites. The coefficient for black education is positive, but only half the level of the white education coefficient. For both races, effect of education on the wage was small, and differences in average education levels explain virtually none of the racial difference in the wage paid to the typical worker.

These results suggest that black job applicants were stereotyped by race, even at Ford. At the same company that challenged prevailing racial employment practices, blacks were still judged by their race and not by their individual attributes. This statement is another way of looking at the evidence in the previous section. At a wage no higher than that which they paid to beginning white workers, auto companies could readily hire black workers who were mature, married, stable, educated to about the same extent as white applicants and with an expected job tenure that was no lower than that of whites. Either Ford was simply prejudiced. or else the company visualized different types of work assignments for blacks, for which these individual attributes made less difference.

Vl. Turning the River Rouge Foundry into a Black Department

There is another sense in which wage comparisons understate racial discrimination, even when adjusted for worker characteristics. To a disproportionate extent, blacks were placed in working conditions which were hot, dirty, heavy and disagreeable. By conventional criteria, blacks should have received "compensating" wage differentials for accepting these disagreeable jobs, but evidently they did not. We have no quantitative measure of job quality. But the most striking concentration of black workers was in the foundry at River Rouge, widely reputed to be the least attractive workplace in the firm. As one Ford employment official stated in the late 1920s: "Many of the Negroes are employed in the foundry and do work that nobody else would do" (Dunn 1929, 69, emphasis added). Between 1918 and 1927, more than two-thirds of new black hires at River Rouge began work in the foundry, while over the same period only 25 percent of new white hires began in that department. Though blacks constituted less than 10 percent of new Rouge hires prior to World War 11, by the 1930s half of the new workers assigned to the foundry were black, and this share rose above three-quarters during the 1940s (Table 9a). What motivations might account for this development? Though racial stereotypes were undoubtedly involved, it was not the case that all black workers were assigned to the foundry, nor that these initial assignments were always permanent. Hundreds of black workers were employed in production jobs in every major Rouge department (Bailer 1943, 418), and a non-trivial number were foremen (Meier and Rudwick 1979, 9). A more plausible interpretation is the one suggested by the Ford official just quoted, namely that assignment to the foundry was merely a form of implicit wage discrimination, correlated only coincidentally with race. Workers whose reservation wage was low (i.e., those with limited employment opportunities) were willing to accept unpleasant working conditions because the pay at Ford was so much better than their alternatives. By this reading, the implicit racial discrimination at Ford was largely a byproduct of the far more severe policies of racial discrimination prevailing elsewhere.

A closer examination of the evidence suggests, however, that a more complex internal dynamic was involved. Table 9b presents data on worker characteristics, quit rates, and job tenure for entering workers at Rouge, distinguishing blacks and whites, foundry and non-foundry, all hires from "stayers" (those who remained at Ford for 60 days or more). There do not appear to have been major differences in observable worker characteristics between foundry and non-foundry workers, beyond the general racial comparisons discussed earlier. During the years 1918-1927, when nearly one-quarter of new white hires began work in the foundry, quit rates for whites in the foundry were actually somewhat lower than those elsewhere, while average job tenures were higher. After 1927, however, when the foundry became predominantly black, those whites who worked there did not stay long. In contrast, blacks who entered the foundry quit less and stayed longer than workers in any other category. The black foundry quit rate actually declined over time, while the opposite was true for whites. Evidently a tipping point was reached, after which the foundry became established as a "black department."

Among the many possible interpretations of this development, two main processes come to mind. The first is an aversion to racial contact of the "Becker-type," in which whites object to working in a department with large numbers of blacks. Schelling (1978) shows that even moderate preferences of this type can give rise to dynamic "unraveling" effects, with segregation a likely outcome. The second possibility (not mutually exclusive) has more economic contact, building on the proposition that elasticities of labor supply to Ford must surely have been quite different for the black and white populations. As the foundry came to be majority- black, the company may have adapted the package of wages, working conditions, and promotion possibilities to fit the labor supply properties of blacks rather than that of the general population. By this argument, white aversion to the foundry may have been less to association with blacks than to job characteristics that were attractive only to blacks. Further study of the Ford records may indicate which of these effects prevailed, or how they interacted. Contemporary testimony suggests a blend of the two, as in the 1940 statement by a Packard official to an interviewer: "White and colored get along all right in the foundry because the average white worker doesn't want a foundry job anyway. White foundry workers are foreigners" (quoted in Bailer 1943, 421). Supposedly the Rouge plant had no racial tension on issues of this sort, but even at Rouge a substantial degree of segregation prevailed.

Vll. Conclusions

The most important implications of this preliminary study of Ford's experience with black workers probably do not concern that company at all. Ford hired blacks extensively, and the evidence shows that these workers were mature, stable, productive employees, with levels of education comparable to those of whites hired for similar positions. Despite this success, other major companies refused to hire blacks for anything but menial tasks, prior to World War II. We believe this evidence shows the importance of demand-side factors in black economic history. Since other companies had access to the same pool as Ford, it is difficult to see how their exclusion of blacks can be justified on the basis of inferior educational backgrounds or inadequate performance on the job. As late as the 1960s, the legacy of Ford's early experience with black employment could be seen in that company's higher black representation among skilled and white-collar workers (Northrup 1968).

Yet the evidence also shows that even at Ford, race played a significant role in hiring and job assignments. Conventional quantitative measures of racial discrimination at Ford do not seem alarmingly large. They show, however, that the company made fewer distinctions among blacks than among whites, ignoring potentially productive characteristics such as age, marital status, education, and expected job tenure where blacks were involved. Either for reasons of prejudice or because of economic incentives to treat blacks differently, black workers became increasingly concentrated in departments like the foundry at River Rouge, even while some black workers were making their way into more skilled and better-paying positions. The origin and character of such tendencies toward segregation deserve much closer scrutiny.

Table 1. Black Workers in Northern Cities, 1910-1950
(18 cities, men aged 14-64)

A. 1/250 Sample
			1910					1940		
		1950
		No.		%Black			No.		%Black	
	No.		%Black
	
All Occupations		847		3.82			3677	6.09	
		6621		9.36		.		
Construction		80		3.51			606		11.69
			517		9.90
Steel		1 0		2.53			77		5.76			231
		12.81
Autos		1		0.53			77		5.33		
	330		14.77
Railroad		44		4.30			136		8.76		
	269		16.20
Meat Products		7		4.35			43		11.56
			89		25.28
Domestic & Pers.	68		14.44			231		20.09
			141		18.24
Eat&Drink,Hotel	110		13.21			172		9.38	
		290		13.98
Laundry, etc.		2		1.92			51		10.78
			112		19.18
B. Exclude Self-Employed, Unemployed, Government Employees
			1910					1940		
			1950
		No.		%Black		No.		%Black		
	No.		%Black
All Occupations		588		4.09			155	6	5.03	
		378	 	28.74
Construction		68		4.09			105		5.75	
		381		10.27
Steel		1 0		2.75			59		5.06			221
		12.84
Autos		1		0.56			67		5.24		
	291   		14.13
Railroads		38		4.06			121		8.77		
	230		14.64
Meat Products		7		4.79			36		10.68
			80		24.54
Domestic & Pers.	55		20.30			166		24.30
			123		17.98		
Eat & Drink, Hotel	92		15.26			116	
	10.97			233		16.28
Laundry, etc.		2		1.92			51		10.78
			112		19.18
   SOURCE: Public Use Census Tapes for 1910, 1940, 1950
Table 2. Large Employers of Black Auto Labor, 1939-40

	Total Hourly			Black				Percent 
Black
	Workers			    Workers
Ford Motor CO.a	90,000			11,000			
	12.22		
General Motorsb	100,000			2,500				
	2.50
Chrysler Corp.a	50,000			2,500				
	4.00
Briggs Mfg. Corp.a	14,000			1,300			
		9.29
Midland Steel                4,	100			1,250			
		30.49		
Bohn Alum& Brass	2,798			668				
	23.87	
Packard Motor	16,000			600					3.75
Kelsey-Hayes	3,050			365					11.97
Murray Corp.	7,000			350					5.00
Hudson Motor	12,200			225					1.84
SOURCE: Lloyd C. Bailer, "The Negro Automobile Worker," Journal of
Political Economy 51 (October 1943), p. 416.
aMichigan plants only   bMichigan and Indiana plants only
Table 3. Black Workers in Ford Michigan Plants, 1939-40

	Total Hourly					Black		
	Percent Black
	Workers					Workers

River Rouge	84,096				9,825			
		11.7
Lincoln	2,332				31					1.3
Highland Park	992				16					1.6
			
Ypsilanti	805				9					1.1
Flat Rock	548				1					0.2

SOURCE: Herbert R. Northrup, The Negro in the Automobile Industry, The 
Racial Policies of American Industry, Report No. 1, (University of 
Pennsylvania, 1968), p. 13. From Lloyd C. Bailer, 1940 unpublished MS
Table 4
Average Hourly Rates for Entry-Level Workers in 20 Industries, 
July 1937
	White		Black		%Difference			
		
United States				$.534		$.420		27.1
North				.552			.	566			-2.5
	Pennsylvania			.558			.584			-4.7
	Ohio				.555			.563			-1.4
	Indiana				.	560			.552		
	1.4
	Illinois				.595			.591			0.7
	Michigan				.536			.596			-11.1
South				.434			.345			25.8
	North Carolina			.	331			.258		
	28.3	
	South Carolina			.298			.228			30.7
	Georgia				.305			.260			17.3
	Alabama				.415			.389			6.7
	Mississippi				.365			.302		
	20.9
	Louisiana				.369			.305			21.0
Southern Firms Hiring
	Whites Only			.444-	
	Blacks Only					-			.296
	Blacks & Whites		.428				.358			19.6

Source: Jacob Perlman and Edward K. Frazier, "Entrance Rates of Common 
Laborers in 20 Industries, July 1937," Monthly Labor Review 45 
(December 1937), PP. 1494, 1497.
** Tables 5-8 and Figures 1-4 could not be included in this summary but 
are available from the authors.