Spring / 2004 Caleb Williams

Dana Lockhart

Social Ethics--PHIL 401

“Color Tax”: A Reality in Bowling Green /Warren County?

Hypothesis:

The search for affordable housing embodies one of many arduous tasks that confront Bowling Green-Warren County’s middle to lower income residents with unpleasant regularity. The fight to balance housing costs against the economic nurturing equally required of all other life necessities is a prevalent struggle that crosses even ethnic boundaries within the middle to lower income bracket.

On the city-county, state, and national level the last twenty-to-thirty years has offered a milieu of somewhat progressive methods for alleviating this difficulty through legislative means. Most of these solutions have been aimed at providing underprivileged groups with the economic means to afford adequate housing. From Section 8 vouchers--which directly provide the capital to afford adequate housing through subsidies--to job skills training programs that aim to provide this economic sector with a higher earning power, the mantra of solvency has been “give them capital or the ability to make more capital and the discrepancy will end.” It would be an obvious disservice to claim that these currently instituted solutions have not at all helped alleviate fair housing problems. However, an obstacle may currentlyexist within the Bowling Green-Warren County area that hinders the full capacity of these solutions to effectively reduce the affordable housing problem.

Anderson and Pickering’s Confronting the Color Line: The Broken Promise of the Civil Rights Movement in Chicago[1] outlines a specific procedure put to use during the Chicago Movement to determine whether or not minority residents--specifically African Americans--were subject to a “color tax.” The “color tax” is a phenomenon coined by William H. Moyer, a Civil Rights activist involved in research at the American Friends Service Committee, following 1951 (Anderson, 1986). Moyer’s “color tax,” as outlined by Anderson and Pickering proposed that African Americans in Chicago paid a de facto tax for housing that whites did not. According to Moyer, a “color tax” is present within an area if non-whites are getting less housing for their dollar than are whites. To identify this disparity the examination of eight census variables within a community is necessary:

  1. Median white monthly rent
  2. Median White monthly income
  3. Average number of occupants in median white household
  4. Number of rooms in median white household
  5. Median non-white monthly rent
  6. Median non-white monthly income
  7. Average number of occupants in median non-white household
  8. Number of rooms in median non-white household

Examining these variables in the 1960s census made it possible for Moyer to conclude that “blacks [at the time] got less housing for their dollar than whites (Anderson, 1986).” This conclusion provided a striking artifact for the Chicago movement. The “color tax” helped prove (statistically) that racially discriminative procedures and practices were a catalyst to housing disparities.

While Moyer’s analysis took place well over thirty years ago, the age of this model does not negate its applicability to the Bowling Green-Warren County area. In fact, despite legislative action, the Bowling Green-Warren County area continues to experience affordable housing disparities. For this reason, fair ground exists to pose the following question: “Is the color tax a reality in the Bowling Green-Warren County area?” Discovering a “color tax” in Bowling Green-Warren County would provide two clear implications. First, despite the fact that all middle-lower income residents have difficulty locating affordable housing this task proves most difficult for non-white Bowling Green-Warren County residents. Second, a “color tax” would prove the “give them capital” mantra inadequate for solving affordable housing problems among the housing markets’ most vulnerable members: African Americans. A “color tax” would show that money (government-provided or not) is worth less in the hands of a non-white homebuyer than in the hands of a white homebuyer. These two implications would help provide a startling artifact that could be used to inspire a more critical examination of the housing problem in the Bowling Green-Warren County area, as well as afford possible solutions.[2]

Complications:

Moyer, in his original “color tax” analysis, used a study completed by Otis Duncan from the University of Chicago to obtain population statistics.[3] The analysis at hand relied on the (free) population statistics provided by the U.S. Census Bureau. It is necessary to recognize that the statistical information Moyer used in his mid-1960 Chicago analysis contains obvious differences from the available Bowling Green-Warren County census data. The following differences, while not devastating to this study’s process, did provide several inherent complications that are necessary to address:

  1. The Bowling Green-Warren County information, which is broken down into block groups, is limited. Financial information and information pertaining to the racial makeup of block groups was unavailable outside of the concentrated minority areas. Because Bowling Green’s census tracts are large and non-homogeneous in nature, block groups provide a much more accurate sample of the community. In any community comparable to the size of Bowling Green-Warren County, an analysis that uses tract data will be skewed towards the majority group.
  2. The lack of available information from 100 percent data sets required the use of sample data sets. Sample data sets are projections based on sampled data. Simply put, this study estimates from Census Bureau projections.
  3. Whereas Moyer’s study was able to use income data obtained from 726,389 Chicago families living in rent housing, the study at hand used a much smaller population. Therefore, this study drew its conclusion based on relatively small differences within relatively small communities. Particular circumstances of individuals within the population, both positive and negative, could have had a major impact on the outcomes of this study.
  4. The census data used did not allow the differentiation between homeowners and home renters in certain categories. For example, information on median household income uses the median of both homeowners and home renters. Even though Bowling Green-Warren County has an unusually high rate of renters, inclusion of homeowners in renter statistics may have skewed the results of this study.

Process:

Moyer’s eight variables were extracted from a sample of the Bowling Green-Warren County census tracts with the following modifications: (Table 1)

  1. Information on median number of rooms was not available broken down by race. Instead, the median number of all households was used. Because census tract 102 has the highest percentage of African Americans, its numbers were used to substitute for the median among African Americans when analyzing other census tracts.
  2. To account for the small number of minority households outside of the central city, the total number of white households and African American households was approximated and given. This was done by using the total number of households reported in the sample data.

Results:

Results of this first step proved inconclusive in establishing a “color tax” for the following reasons:

  1. Average rent in lower economic tracts was shown on average to be significantly lower than rent in higher economic areas. This negates the required existence of artificially inflated rents.
  2. Median rent for black households was lower than median rent for white households in the majority of census tracts surveyed, negating yet another “color tax” criterion.
  3. The two instances where African American rent was not shown to be lower than white rent were all located in census tracts classified as upper-middle class. In census tract 119 there was insufficient data for the U.S. Census Bureau to produce a median rent figure.
  4. While median black household income was lower than that for whites in the majority of census tracts in the two instances where median black income was higher, rent continued to be lower for median black households (results from tract 104 were ignored in this instance). Furthermore, African American households constituted less then 10 percent of households in tracts with significant differences in average household income. Black households constituted less then 5 percent of households in instances where both the median income was lower and the median rent was higher. The exception here is tract 103, which preliminary research indicated was the one tract where these conditions for a color tax could be verified.
  5. With the exception of one census tract (119), all tracts reported a higher median number of occupants for black households. However, this rate was only significantly different in four tracts (with 104 again being discounted). Without being able to compare the median number of rooms, this result is limited in its ability to help establish a color tax.
  6. It is unclear whether the artificial factors necessary for maintaining inflated rents in minority areas are present in Bowling Green. Simply put, many items that seem to indicate a color tax could also be explained by a lack of affordable housing and the concentration of the affordable housing stock in the downtown area.
  7. Without separate information for homeowners and renters, the economic data at hand is suspect. Combined with the lack of information available in block groups, this lead to the conclusion that this test could not independently verify the existence of a color tax. However, data from census tract 103 indicated that many of the conditions laid out by Moyer’s study are present in that area of Bowling Green.

Conclusions:

While the “Color Tax” hypothesis proved inconclusive, several attempts were made to apply Moyer’s concepts to other social classifications (“Poor Tax” and “Track Tax”). Attached to this study are tables created to study these attempts. An initial review of their results also proved to be inconclusive. However, they may provide useful fodder for future research. Included within the addendum is an area of further analysis that can help to guide any further study that may be undertaken using the included charts. Our hope is that while this project did not produce tangible results, it does contain valuable research that will help advance the goals of the ongoing Social Ethics Research Seminar.

[1] Alan B. Anderson and George W. Pickering, Confronting the Color Line: The Broken Promise of the Civil Rights Movement in Chicago. (Athens, GA: University of Chicago Press, 1986)

[2] The 2000 U.S. Census Report was the source used for all Bowling Green/ Warren County statistics

[3] Moyer, Bill. An Analysis of the System of Housing Negroes in Chicago. Document released by The American Friends Service Committee in 1966.