Thursday, December 1, 2011

Michigan: A Demographic and Political Analysis in Three Parts

Part I: Introduction and General Population Trends, 1960-2010

Easily lost in all the hand wringing following last February’s announcement that Michigan was the only state in the Union to lose population over the past decade was the broader implications of the 2010 decennial census data. Michigan has experienced significant population transition within its borders over the past decades, and to better understand the impact of this shift I spent some time over the past few months reviewing census data for units of governments on the county subdivision level between 1960 and 2010.

I chose to analyze county subdivision level data for a number of reasons. First, county subdivisions, which are defined in Michigan as municipalities or townships, have largely had stable boundaries since 1960, thanks to the 1963 state constitution which limited physical growth by municipalities by placing high barriers against annexation by granting townships the ability to become charter townships. A township must have a population 2,000 to become a charter township, and under state statue charter townships that have a population density of greater than 150 people per square mile (along with a few other conditions) are protected from annexation by other municipalities. With the exception of Oakland County, there have been relatively few border changes and mergers of county subdivisions in the past 50 years. Secondly, census demographic and economic data is readily available starting with the 1970 census, and there is some racial data available from the 1960 census. This makes a longitudinal study of demographic changes in Michigan’s communities possible, and sheds light on changes that occurred following 1960. Finally, partisan electoral data that is available for county subdivisions shows the impact of these population and demographic changes.

Just a small note: some of the population numbers and demographic data to not add up to 100 percent, particularly for the 1980 census data. For example, in Burton City, the 1980 total population was 29,976, while the population total for the racial identification question was 29,929, a sum larger by 47 people. In this sort of instance I used the racial data to provide a demographic percentage for analysis, but kept the total population number given the 1980 census. All total population data from the 1960 to 1990 censuses were obtained from the Michigan Department of Technology, Management & Budget, while racial data for this time period were obtained from the National Historical Geographic Information System. Total population and racial data from the 2000 and 2010 census were obtained from the US Census website. Some municipalities are split between two counties, and tabulated as separate entities

General Population Trends, 1960-2010

Michigan’s population increased rapidly between 1960 and 1970, but grew at a much slower rate in the forty years that followed as shown in Figure 1. 7,823,194 residents lived in Michigan in 1960, a figure that climbed by more than a million ten years later to 8,875,083, a 13.4% increase. By 1980 the population increased to 9,262,078, a gain of 386,995 more residents. A modest growth in population followed in 1990 giving Michigan a population of 9,295,297 residents, and the economic growth during the following decade resulted in a total population of 9,938,444 and gain of 643,147 people. The economic malaise over the most recent decade dropped Michigan’s population by 54,804 residents.

As Figure 1 shows, Michigan’s population has become more diverse over the past fifty years. Racial and ethnic data gathered in 1960 asked respondents whether they were White or non-White, the later numbered 737,329, constituting 9.4% of the total population. By 2010 Michigan’s non-White population had more than doubled, reaching 21.1%. Michigan’s 2010 White population of 7,803,120 is smaller than its 1970 population (7,833,473), while the state’s Black population increased from 991,067 to 1,400,362 during the same time period. While the state’s Black and White population stagnated over the past decade, the number of Hispanic residents increased significantly, rising from 323,877 to 436,358. Asians also account for a growing share of Michigan’s population, growing from 175,311 to 238,199 over the past decade.

While Michigan’s was becoming more diverse, it was also becoming more dispersed. In 1960, Michigan’s 20 largest communities were home to more than 3,309,313 people, or 42% of the state’s total population, but 2010 these same 20 communities only were residences to 22.3% or 2,208,322. Figure 2 shows just how drastic the population decline was. Detroit accounts for 86.8% of the total population loss, but only six of the 20 subdivisions gained residents and of these only Ann Arbor and Warren added more than 40,000 residents. With the exception of Detroit and Flint, communities that lost residents overwhelmingly did so between 1960 and 1980, and stabilized somewhat in the thirty years that followed.

Communities that gained the most residents between 1960 and 2010 tended to be communities in metropolitan southeast Michigan that captured Detroit’s fleeing residents. Only two communities outside of metropolitan Detroit were among the top twenty (Georgetown Township in Ottawa County and Kentwood in Kent County).

Michigan’s population dispersal has reduced the state’s population density. While Michigan’s overall population density increased slightly from 138 persons per square mile in 1960 to 174 by 2010, the addition of 2,115,250 new residents during the same time period meant that much of the population growth went into new development on the outskirts of the urban fringe. The table below shows communities with the highest population density back in 1960. When viewed next to the 2010 population density figures, you can see what a beating the urban core of Michigan took in the past fifty years. While the Detroit metropolitan region could arguably supported a dense commuter rail network in 1960, as numerous communities had population densities greater than 4,000 people per square mile that is thought as the minimal density needed for effective mass transit, the de-densification of communities such as Detroit and Highland Park makes the implementation of mass transit much less cost-effective.

Of course the movement of people from central cities to suburbs is nothing new in Michigan, let alone the United States. However, the dispersal of Michigan’s White population from urban areas was matched by two smaller-sized migrations of Black residents. The first relatively minuscule migration was from historic rural Black areas of western Michigan (such as Lake, Van Buren, and Cass Counties) into other urban centers in Michigan, especially Benton Harbor, Flint, and Detroit. The second and larger migration was the movement away from core urban centers, especially in Detroit and Flint, to the surrounding suburbs. Most of the Black population movement has flown to working class suburbs on the periphery of established urban centers. Suburbs such as Harper Woods and Eastpointe literally changed overnight, while other communities such as Oak Park and Lathrup Village have steadily attracted new Black residents over decades.

In contrast to the movement of Black Michiganders, Hispanics have concentrated outside of Southeast Michigan and are spread throughout Michigan. Only 10% of the total Hispanic population resides Detroit and only make up 6.8% of the city’s total population. While county subdivisions with the greatest increase in the Hispanic share of the total population are listed below. Interestingly, rural communities in western Michigan are home to large sizable Hispanic populations, largely due to the reliance farming communities have on migrant workers who have historically been Hispanic. However, large Hispanic communities have moved to urban centers and suburbs in western Michigan, including Grand Rapids and Holland. Kent County in particular has a large Hispanic population that is just under 10% of the total population.