CHAPTER II - DEMOGRAPHICS
A. EXISTING POPULATION CHARACTERISTICS
1. Population Trends
In 1990, 80,277 individuals resided in the City of Kalamazoo, accounting for 35.9 percent of Kalamazoo County's total population (Table II-1). The City's share of the County's total population declined during the 1970's and 1980's, as population growth in the County's suburban communities outpaced that of the City. From 1980 to 1990, the City's population increased 0.7 percent, while County population outside of the City grew 7.9 percent. Yet, the decade of the 1980's contrasts the decade of the 1970's when the City's population declined 6.8 percent while the County's population outside the City grew 14.4 percent.
Along with the modest decline in the City's overall share of the County's population, its share of the County's African American, Hispanic and elderly population also declined during the 1980's (See Table II-2 and II-3). In 1980, 78.6 percent of the County's 19,919 African Americans lived in the City. By 1990, this percentage had fallen slightly to 75.6 percent. The City's share of the County's Hispanic population also fell from 61.1 percent in 1980 to 59.9 percent in 1990.
Although a greater number of the County's minority population were residing outside the City in 1990, compared to 1980, the majority of the County's African Americans and Hispanic residents still lived within the City's boundaries.
The 1990 median age of Kalamazoo City residents, 25.6 years, was below the median for the County, 29.9 years, due to the student population at Western Michigan University and Kalamazoo College. In comparison to the age composition of County residents, who lived outside the City, the City housed a greater share of both individuals between the ages of 15 and 24 (e.g., college students), and individuals 85 years or older. In the City, 27.7 percent of the residents were between the ages of 15 and 24 years of age, compared to only 14.6 percent of the County's population living outside of the City (Table II-4).
3. Household Types of Relationships
The number of persons per household in the City has declined during the past twenty years.In 1990, the average household had 2.41 persons, compared to 2.84 persons in 1970. The average number of individuals living in families also declined slightly from 1980 to 1990, from 3.09 to 3.03. In 1990, households in the County, but outside of the City, were larger than City households. On average, County households located outside the City numbered 2.62 persons.
In 1990, 9,461 individuals, who resided in the City, lived in group quarters. (The balance of the County had only 1,060 people in group quarters in 1990.) Of this total, most (about 7,258 persons) lived in college dormitories. Another 237 persons lived in emergency shelters for the homeless. The balance of 1,966 individuals were generally lodged in institutions under formal authorized supervised care or custody (such as nursing homes and correctional facilities). In 1990, 65.5 percent all persons living in County in supervised institutions were located in the City of Kalamazoo.
In 1990, nearly 80 percent of all City residents, 25 years or older, earned at least a high school diploma, and nearly 30 percent earned at least a four-year college degree (Table II-2).In comparison to the County, as a whole, a lower percentage of the City's residents have graduated high school; however, a higher percentage have completed a four-year degree. Among the City's African-American residents, only 66.6 percent have completed high school, and 9.1 percent have earned a four-year college degree.
5. Income Characteristics
The City's 1989 median household income stood at $23,207, which was only 74.7 percent of the County's overall 1989 median income of $31,060. Although the City accounted for 35.1 percent of the County's overall number of households in 1990, it housed 53.1 percent of all of the County's households making less than $15,000 in 1989.
In 1989, 19.3 percent of all families residing in the City lived in poverty, an increase from 13.7 percent in 1979. During the 1980s, the County's families living in poverty became more concentrated in the City. According to the 1980 census, 59.8 percent of all County families living in poverty resided in the City. By the 1990 census, this percentage rose to 64.3 percent.
One-third of the City's children live below the poverty level in 1989, compared to 24.4 percent in 1979. Moreover, 67.9 percent of the County's children living in poverty resided in the City in 1989.
While, all low-income families have become more concentrated in the City of Kalamazoo, it is especially true for African Americans. According to the 1990 census, 91.3 percent of all African Americans living in poverty in the County resided in the City.Overall, 41.3 percent of the City's African-American families lived below the poverty level in 1989.
The decline in household size over the past decades has been accompanied by an increase in households and housing units that outpaced population growth. Between 1980 and 1990, there was a 3.7 percent increase in the number of occupied housing units to 29,409. Citywide, 47.4 percent of all occupied housing units were owner-occupied in 1990 which was down from 50.3 percent in 1980 and 57.4 percent in 1970. For the County area outside of the City, owner-occupied housing units accounted for 73.6 percent of all occupied dwellings in 1990. In the City, 2,079 housing units (6.6 percent of all housing units) were vacant in 1990. Of these vacant units, 49.5 percent were for rent and 14.3 percent were for sale. Similar statistics hold true for the area in the County but outside the City: 5.5 percent of all housing units were vacant, 48.9 percent of the vacant dwelling units were for rent, and 13.9 percent were up for sale.
The self-reported 1990 median value of owner-occupied housing units in the City was $48,600. County wide, the median value of owner-occupied units, as estimated by their owners, was $72,800. The City's median contract rent was $357 per month, which was also below the overall County median rent of $370 per month.
Housing affordability is a key concern for all residents. For City households earning less than $20,000 in 1989 (about 86 percent of the median income) and owning their own home, the median ratio of monthly owner cost of housing to household income stood at 30.0 percent (Table II-5). For homeowners earning between $20,000 and 34,999, the ratio was 18.3 percent. For homeowners earning between $35,000 and $49,999, the ratio of monthly owners' cost of housing to household income was an even lower 16.1 percent, and for homeowners earning $50,000 or more, the ratio stood at 11.4 percent. Across all income groups the median monthly owners cost of housing to income for City residents were lower than County-wide, except for individual earning less than $20,000.
For renters, affordability is a more serious issue, however. For households earning less than $10,000, the median ratio of gross rent to income was above 50 percent in the City, as well as in the County. For City households earning between $10,000 and $19,999, the median ratio of gross rent to income stood at 34.0 percent compared to 35.4 percent County wide.
In summary, housing affordability is a key concern for City low-income households especially those who rent, but it is an issue that is shared by all County residents.
7.Census Tract Analysis
In Appendix A of this Chapter is a detailed analysis of social and economic trends for each of the Census Tracts in the City from 1970 to 1990. (See Figure II-1 for Census Tract boundaries.) The U.S. Census Bureau defined the boundaries for the Census Tracts in the City and, unfortunately, they do not coincide with Neighborhood boundaries determined by the area residents (Figure II-1). Still, as can be seen in the appendix, the Neighborhood(s) which the Census Tract contains are listed. Because of changes in Census Tract boundaries during this time period, in some instances, some tracts had to be combined into larger units as explained in the technical notes in the Appendix.
The Census Tract analysis is provided in two tables in Appendix A. Table A.1 shows social and economic trends, while Table A.2 provides information on housing conditions by Census Tract.
8. Neighborhood Analysis
As stated above, Census Tract boundaries do not coincide with established Neighborhood boundaries (Figure II-1). Using Block Group (BG) statistics, which are the smallest geographic areas for which sample Census data are available, information is provided on the City Neighborhoods and, in some cases, information on the sub-neighborhood level for 1990 and 1980. (See Table A.3 of Appendix A.) Please note, however, that due to changes in the geographic boundaries of some block groups, strict comparisons between 1980 and 1990 are not available for all neighborhoods. Working in cooperation with Community Information System at Western Michigan University, comparable areas for 1980 and 1990 were constructed as close as possible. Nevertheless, it is possible, in some instances, that measured changes between 1980 and 1990 could be the result of changes in geography, as well as, demographic variations.
9. Travel Analysis Zones
For the purposes of examining the adequacy of the transportation system, the community is divided into geographic areas known as Travel Analysis Zones (TAZs). (See Figure II-1.) Existing and projected households and employment within the TAZ are translated into trip productions and attractions, and assigned to the roadway network to evaluate the ability of the existing roadway network and proposed improvements to accommodate travel. The TAZ boundaries are usually defined by the collector and arterial roadway system. Thus, TAZs are smaller than Census Tracts, tend to be comparable to Census Block Groups, but do not respect jurisdictional (or Census Tract) boundaries which fail to coincide with the arterial or collector roadway system, railroads or major rivers.
The correlation between Census Tracts and Travel Analysis Zones appears in Appendix B. This provides information by place of work that is not found in Census information.
10. Comparison Analysis with Other Similar-Sized Metropolitan Areas in Michigan
In Tables II-6 through II-11, the City of Kalamazoo is compared to similar-sized central cities in the state in order to identify its relative performance to cities in similar economic and social situations. As shown in Table II-6, Kalamazoo's population growth from 1980 to 1990 was close to the average for the comparison group. Excluding Battle Creek which grew 49.9 percent during the period due to the merger of the city with Battle Creek Township in 1983, Kalamazoo outperformed Saginaw and Jackson, but fell behind Grand Rapids and Ann Arbor. During the more recent 1990 to 1994 period, the City's population grew by 1.7 percent, which again put it in the middle of the pack. Battle Creek and Jackson both achieved average annual growth rates of above 2.0 percent. On the other hand, Ann Arbor lost population.
Figure II-1: Census Tract, Neighborhood and Travel Analysis Zone Boundaries
Not surprising, a greater portion of Ann Arbor's residents earned a bachelor's degree or higher in 1990 and 1980 compared to the other areas (Table II-8). In 1990, 64.2 percent of Ann Arbor's residents earned a Bachelors Degree or higher. Kalamazoo came in a distant second with 29.8 percent. The percent of the cities' population with a four-year college degree or better increased in all of the cities except Jackson during the ten-year period. In 1990, Ann Arbor was the preferred location for highly educated individuals in the County.Only 29.0 percent of Washtenaw County's residents outside of Ann Arbor held bachelor degrees compared to 64.2 percent inside the city's borders.In Kalamazoo County, 25.9 percentof its residents, who selected to live outside of Kalamazoo City, earned a four-year degrees or higher, which was only slightly below the City's 29.8 percent.
Kalamazoo ranked fourth among the six cities in terms of median 1989 household income (Table II-9). The City's median household income was a full $10,000 below that for Ann Arbor. More disturbing, Kalamazoo City ranked fifth among the six cities in regards to the percent of its residents living in poverty in 1989. As shown on Table II-10, 26.2 percent of its residents lived in poverty in 1989. Moreover, the percent of Kalamazoo's residents living in poverty increased from 17.7 percent to 26.2 percent during the ten-year period ending in 1989. It declined in Ann Arbor and Grand Rapids during the same time period. The percent of individuals in poverty also fell in Battle Creek; however, this was due, in large part, to the city merging with Battle Creek Township in 1983.
The total number of occupied dwelling units increased 3.6 percent in Kalamazoo City from 1980 to 1990 while they declined in Jackson and Saginaw (Table II-11). However, the number of owner-occupied households fell 2.5 percent Owner-occupied housing increased 11.0 percent in Ann Arbor and inched up a scant 0.1 percent in Grand Rapids. Again, Battle Creek's robust 62.3 percent increase in owner-occupied dwellings was due primarily to its 1983 merger with Battle Creek Township.
Overall, the above comparison analysis suggests that the City of Kalamazoo is performing no better than average when compared to similar cities in the state
3.Future Age Distribution
Due to the student population at Western Michigan University and Kalamazoo College, the City will continue to house a greater number of individuals in the 15 to 24 age range than the balance of the County in the year 2015 (referring to Tables II-12 and II-13). Between 1990 and 2015, the proportion of the population age 65 and older will decline from 10.9 percent to 7.5 percent in the City in contrast to the growth in the age 65 and older population in the balance of the County from 10.4 percent in 1990 to 16.1 percent in 2015.
4. Future Group Quarters Population
Referring to Table II-1, the projected population in group quarters was held relatively constant for the City and County to the year 2015. Nearly 90 percent of the group quarters population of the County resides in the City, and people in college dormitories accounted for 77 percent of the group quarters population in the City. People in nursing homes account for most of the population in group quarters outside the City. If a correlation is made between the number of persons in nursing homes and the population age 65 and older, the ratio is 8.2 percent for the City and 7.2 percent for the County in 1990. Based on the population age 65 and older in the year 2015, there would be a decline of 231 persons in nursing homes in the City and an increase of 639 persons in nursing homes County wide. Because the population in nursing homes in the City accounts for only 7.6 percent of the persons in group quarters, the correctional institution population is of comparable size and the college dormitory population is such a large component, a small shift in the future elderly population in the City was not considered significant to alter the assumption of a constant percentage of the total population in group quarters in the City. In the case of the County as a whole, the increase in elderly population in future years argues for an increase in nursing home component of the group quarters population; however, because the percentage of population in group quarters outside the City dropped over the past decade, the absolute number of persons in group quarters was held constant for future years to create a modest decline (0.4 percent) between 1990 and 2015 of the percentage of the population in group quarters.
5. Future Household Size
Although the size of households has declined for the past thirty years, the rate of household size decline decreased from the decade of the 1970's to the decade of the 1980's (Table II-1). In contrast, the Woods & Poole forecasts assume a constant rate of household size decline. Assuming household size will continue to decline but at a slower rate in future years, the County household size is forecasted to drop from 2.54 persons per household in the year 1990 to 2.35 persons per household in the year 2015. Maintaining the City to County relationship, the City household size declines from 2.41 persons per household in the year 1990 to 2.22 persons per household in the year 2015.
6. Future Households and Ownership
The number of future households and number of future occupied housing units are synonymous (Table II-1). The number of future households is a derivative of group quarters and household size assumptions within the overall population projections for the City and County. For the City, the number of households is forecasted to increase by 2,221 households between the years 1990 and 2015 despite a slight total population decline.
The percent of owner-occupied housing is forecasted to decline in both the City and County at constant rate reflecting the past decade. However, the rate of home ownership decline in the City is more than twice that of the County as a whole because of the modest household growth of only 7.6 percent between the years 1990 and 2015, the conversion of existing housing units from owner to renter status due to continued outward migration of middle and upper income households, and the concentration of new housing units over the past sixteen years (1980 through 1995) on multi-family housing (89 percent of all new housing units).
7. Future Housing Units
The forecast of future housing units is derived from the projection of households and assumptions concerning the housing vacancy rate. With new residential construction comprising about 15 percent of the total housing units between 1990 and 2015, the average structural age of housing will continue to decline in the City. Thus, rate of increased housing unit vacancy for the decade of the 1980's was continued into the future at 0.6 percent per decade. This results in 8.0 percent of the housing stock being vacant in the year 2015 compared to 6.6 percent in the year 1990. For the County, the housing vacancy rate was held constant for the balance of the County outside the City; therefore, the slight increase in the vacancy rate for housing County wide reflects the impact of the City's increased vacancy rate.
Referring to Table II-1, the number of housing units increases 2,886 units in the City and 17,238 units County wide between the years 1990 and 2015. To check the reasonableness of the housing unit projections, an examination was made of housing construction activity (see Table V-2). In the decade of the 1980's, the 2,800 new housing units in the City translated to an increase of 1,300 units in the total housing stock (46.4 percent). Thus, the average annual rate of new housing construction over the past six years (at 100 units per annum) translates to an increase of only 1,160 units in the housing stock between the years 1990 and 2015. On the other hand, if the average annual rate of new housing construction over the past sixteen years (at 212 units per year) is assumed through the year 2015, the housing stock would increase 2,459 units between the years 1990 and 2015. Finally, if the average annual rate of new housing starts in the decade of the 1980's (280 units per year) is assumed, the housing stock would increase 3,248 units. Accordingly, the forecast of housing units in the future falls between the average annual rate of the 1980's and the 16-year annual average -- roughly 250 units per year. In light of the six-year annual average rate of new housing units at only 100 units per year, the new housing market must be stimulated or the rate of existing housing cost reduced to achieve the forecasted number of housing units in the year 2015.
In the case of the County, new housing construction in the 1980's at 11,045 units translated to an increase of 9,643 units in the County wide housing stock (about 87.3 percent). [If the balance of the County were examined, each new building permit translates directly to an increase in the housing stock.] Thus, six-year annual average of new housing at 943 permits per year translates to an increase in the housing stock of 20,581 units. The five-year annual average (1990 through 1994) rate (at 858 permits per year) translates to an increase of 18,726 units in the housing stock between the years 1990 and 2015 because the forecasted housing unit increase between 1990 and 2015 reflects about 790 units per year in new housing construction, no special effort is needed to stimulate the housing market outside the City to achieve the forecast.
8. Future Travel Analysis Zones
The correlation between Travel Analysis Zones and Census Tracts for population, employment appears in Appendix B. The population, household and employment projections by the Kalamazoo Area Transportation Study (KATS) served as control totals. Please note that KATS assumed a constant household size through the year 2015 and no loss of households in older areas. Thus the TAZs generally overestimate population for the City. Because trip generation per household has been increasing, absolute number of trips for built out TAZs remains constant for the present and future assuming both household size and trip generation rates are held constant over time for travel model purposes.
Appendix A: Census Tract Analysis
Appendix B: Census Tract and TAZ Comparison