Blog Archive

November 2014

City of Change – Occupancy Density in Detroit’s Residential Neighborhoods

City of Change – Occupancy Density in Detroit’s Residential Neighborhoods

City of Change is a Data Driven Detroit (D3) blog series analyzing changes in Detroit’s residential neighborhoods from 2009 through 2014. This series is a collaborative effort between Noah Urban at D3 and Gary Sands, professor emeritus of urban planning at Wayne State University. 

**Note:  This blog post will make several references to Detroit’s Master Plan Neighborhoods. If you would like to view a reference map of these neighborhoods to help orient yourself, please click here.

Overview

This edition of City of Change analyzes the density of occupied residential structures (the number of residential structures per square mile that are occupied) and uses this indicator as a proxy for population.  Many neighborhoods in Detroit have endured increasing emptiness since 1960, though some areas of considerable density do remain. If these neighborhoods suffer decline, the potential impact on the city in terms of lost population and revenue is far greater than decline in areas of lower density.

As in the previous installment of City of Change, the data have been summarized for 840 census block groups in the city of Detroit. To account for differences in the size of the block groups, the analysis measures the number of occupied residential structures per square mile. The occupancy rates are based on data collected by the Detroit Residential Parcel Survey (DRPS) in 2009 and Motor City Mapping in 2014.

Occupancy in 2009

Detroit had an average of just over 2,100 occupied residential structures per square mile in 2009. This figure represents a substantial decline (almost 43 percent) from a peak of 3,675 structures per square mile in 1960 (based on archival Census data). The declines over that half-century have been severe across the city — just 18 block groups in 2009, out of 840 total, had occupancy densities greater than the 1960 citywide average.

There was considerable variation evident across block groups, with densities ranging from 51 to more than 4,500 occupied residential structures per square mile. It is important to note that some block groups with low residential densities contain large parcels with industrial, institutional or commercial land uses (parks, cemeteries, factories, etc.). In these areas, residential density is below average, regardless of the vacancy rate. In block groups that are predominantly residential, however, a low occupancy rate can reflect both a predominance of vacant lots and a high number of vacant residential structures.

The highest occupancy densities were generally found on the Far East Side (particularly in the Finney and Denby neighborhoods), the Northwest Side, and in neighborhoods close to Dearborn (particularly the Cody Rouge area). The relatively high density in these latter block groups was likely affected by the expansion of the growing immigrant population in Dearborn. Although many of the areas closest to downtown were not included in the DRPS project boundaries, areas closest to the city center possessed some of the lowest densities. Even along the east riverfront, occupancy levels were primarily in the bottom two ranges.

Figure 1:

MCI_DRPS_OccupiedStructures

This map shows the density of occupied residential structures per square mile in Detroit based on data from the 2009 Detroit Residential Parcel Survey. Note that the Detroit Residential Parcel Survey only collected information for 1-4 unit residential structures.

Occupancy in 2014

Continued population decline over the past five years, along with increased demolition activity, brought the average density of occupied residential structures down to 1,860 in 2014, almost 12 percent below the 2009 figure and less than half the peak density of 1960. As indicated in Table 1, the number of block groups in the highest-density range decreased by 77, or more than 35 percent. The lowest-density block group had just 18 occupied residential structures per square mile.

Table 1: Change in Density of Occupied Structures

 Number of Block Groups
Occupied Structures per Square Mile20092014Change
2,850 or Higher210133-77
2,166 to 2,849210197-13
1,411 to 2,165210216+6
1,410 or Lower210294+84

The decline in occupancy occurred across the city. While the same general areas continued to have the highest occupancy densities, many of the block groups shifted from the highest to the second-highest category, indicating that even denser neighborhoods are experiencing depopulation. Declines are particularly noticeable on the East Side, including much of the Osborn area (identified as a portion of the Mt. Olivet Master Planning neighborhood). Densities remained high in areas with large immigrant populations, including the area just north of Hamtramck. There are even fewer block groups than in 2009 that are close to the city center and have residential occupancy densities higher than the lowest range.

Figure 2:

MCI_MCM_OccupiedStructures

This map shows occupied structures per square mile in 2014, using the same ranges as the map from 2009. Note the considerable decrease in the number of block groups in top range. This decrease is even more notable when considering that the Motor City Mapping survey expanded upon the DRPS to encompass all residential structures.

Absolute Changes in Residential Density Between 2009 and 2014

Between 2009 and 2014, only 90 block groups recorded an increase in residential structures per square mile; 10 recorded no change, and 740 recorded a decline. Thirteen percent of all block groups reported declines of more than 600 units per square mile in only five years. Areas with increasing occupancy densities are found along the Woodward Corridor and the Near West Side (Woodbridge and Corktown areas), which have been the focal points of several residential investment initiatives. Elsewhere in the city, some of the higher-density block groups in 2009 did observe an increase in residential occupancy density, but these areas are somewhat randomly distributed.

Figure 3:

MCI_OS_Change

This map shows block groups that saw either an increase or a decrease/no change in occupied structures per square mile from 2009 to 2014. The block groups in the latter category accounted for nearly 90% of all block groups studied in this analysis.

To a large extent, the decline in occupancy density is a result of a rise in vacancies, rather than a decline in the number of residential structures. In 2009, the average residential occupancy density was 82 percent of its maximum potential (that is, if every residential structure in the block group were occupied). In 2014, the average was just 73 percent of the maximum potential. The number of block groups where the occupancy density was less than half of the potential increased from just five in 2009 to 44 in 2014.

Conclusions

The past five-year period has seen a substantial decrease in the density of occupied residential structures in Detroit. Since the number of new homes built during this period was relatively low, the lower densities are the result of higher vacancy rates in the existing housing stock. A net increase in the number of households occurred in just 10.6 percent of the block groups. These trends seem to suggest that, aside from scattered pockets throughout the city, the population declines that have characterized the past fifty years in Detroit – culminating in the 25 percent decline in population from 2000 to 2010 – are continuing. Future policies should recognize this trend and, in addition to aiming to reverse the decline, should address the potential that it may continue in the future.

*             *             *

City of Change will return in two weeks with an in-depth examination of mortgage deeds and residential market health in Detroit’s neighborhoods over the past five years.

*             *             *

The Motor City Mapping data referenced in this article are available in their raw format on the Data Driven Detroit (D3) Open Data Portal – http://portal.datadrivendetroit.org.

City of Change – Evolution in the Condition of Detroit’s Housing Stock

City of Change – Evolution in the Condition of Detroit’s Housing Stock

City of Change is a Data Driven Detroit blog series analyzing changes in Detroit’s residential neighborhoods from 2009 through 2014.  This series is a collaborative effort between Noah Urban at D3 and Gary Sands, Professor Emeritus of Urban Planning at Wayne State University.  This week, the series examines variation and change in Detroit’s residential structure condition.

**Note:  This blog post will make several references to Detroit’s Master Plan Neighborhoods.  If you would like to view a reference map of these neighborhoods, please click here.

What is Average Residential Structure Condition?

The Motor City Mapping project is a survey of Detroit properties that was completed in early 2014, providing information on some 380,000 individual parcels in the city.  This survey recorded whether each parcel had a structure on it, the condition of the structure, and whether or not the building was occupied. This information has also been summarized for the roughly 880 census block groups in Detroit (each block group includes several contiguous blocks).

Much of the data from Motor City Mapping can be compared to the 2009 Detroit Residential Parcel Survey (DRPS) conducted by Data Driven Detroit and its partners.  Though the earlier survey was limited to 1-4 unit residential properties, it is possible to compare the structural condition data for about 840 block groups at two points in time, from 2009 to 2014.  For every structure surveyed, the property’s condition was assigned a value from 1 to 4 with 1 indicating the best possible condition.  To account for differences in size of block groups, we created a weighted average of block group condition by summing condition values for all residential structures in the block group and then dividing the total by the total number of residential structures in the block group.  Values closer to 1 indicate stronger average condition, while a value closer to 2 (or even below 2) indicates an area with much poorer structural condition.

Average Residential Structure Condition in 2009

In general, the condition of residential structures is related to their age.  Many block groups with the strongest average condition were found at the outer edges of the city, while many of those with the poorest-condition housing stock were located in older neighborhoods just outside of the core of the city (Figure 1). An exception is evident in northwest Detroit, where the Brightmoor neighborhood stands out as a pocket of blight located between more stable areas.  This may be due to the wood-frame, 1950’s housing construction that once abandoned, has decayed at a faster rate than in other areas of the city where the construction typically incorporates greater amounts of stone or brick.

Figure 1:

MCI_DRPS_AvgCond

This map shows Detroit’s residential average housing condition based on data from the 2009 Detroit Residential Parcel Survey. Note the clustering of stronger-condition neighborhoods on the edges of the city.

With limited exceptions, the area south of the Ford Freeway (Interstate 94) included very few block groups in the highest two ranges.  The few such neighborhoods with better average condition ratings included the Far East Side (particular the Finney Master Plan Neighborhood), some areas of the East Riverside and Indian Village neighborhoods, and portions of Corktown, Hubbard Richard, Vernor/Junction, and Springwells in Southwest Detroit.

There was typically a gradual transition between the average condition ratings in adjacent neighborhoods.  That is, neighborhoods in the top range were typically bordered by neighborhoods in the first or second categories.  There were, however, a few instances where block groups with the strongest average condition were adjacent to those with the weakest average condition.  Particularly prominent examples existed on either side of Woodward north of 7 Mile, as well as some of the block groups on the borders of the Brightmoor and Rosedale neighborhoods.

Average Residential Structure Condition in 2014, Compared to 2009

By the time that the Motor City Mapping Survey took place, conditions had changed considerably in many portions of the city.  The average condition rating for each of the block groups in 2014 is shown in Figure 2.   While the overall geographic concentrations of good and poor quality housing are similar to 2009, there are some important differences.  Southwest Detroit has seen a decline in housing condition, and the blight that was evident around Brightmoor in northwest Detroit and the Gratiot-McNichols area in northeast Detroit seems to be spreading.  The Rosedale and Cerveny/Grandmont Master Plan Neighborhoods (which contain Grandmont-Rosedale) are no longer solidly in the top range of block groups, and similar declines can be seen in the Finney Master Plan Neighborhood, which includes East English Village.  Both of these areas have resisted the encroachment of nearby blight, but these data indicate that even traditionally-stable neighborhoods are experiencing some degree of erosion.

Figure 2:

MCI_MCM_AvgCond

This map shows average residential structure condition in 2014, using the same categories as the map from 2009. Note the significant decline in average condition that is visible on the far western side of the city, particularly in the Cody Rouge area (west of M-39, south of I-96).

There was substantial movement between the categories defined in 2009 and 2014, as shown in Table 1.  The top two ranges contained a net total of 79 fewer areas, a decrease of nearly 19%.  The growth in the range with the weakest average condition – an increase of 70 block groups – is particularly concerning, and indicates that an increasing number of neighborhoods across the city may be entering steeper spirals of structural decline.

Table 1:  Change in Residential Structure Condition

 Number of Block Groups
Average Condition Rating20092014Change
1.05 or lower (strongest)210150-60
1.06-1.15210191-19
1.16-1.35210219+9
1.36 or higher (weakest)210280+70

Absolute Changes in Residential Condition from 2009 to 2014

The two surveys recorded a small overall decline in the average residential condition rating for the city of Detroit, from an average rating of 1.23 to 1.27 in 2014.  As shown in Figure 3, more than 500 block groups observed declines in average condition, while 315 saw an improvement or no change. Much of the improvement occurred on the East Side, including some of the neighborhoods with the poorest structural conditions in 2009.  Considering that many of these areas have been identified by city planning processes as high-vacancy, it is likely that much of the observed increase is due to demolition of blighted structures, rather than as a result of new construction or rehabilitation.

Figure 3:

MCI_AvgCond_Change

This map shows block groups that had an increase/no change or a decrease in average residential structure condition from 2009 to 2014. Note that many of the improvements were in the areas that had the weakest average condition in 2009 and 2014, indicating that these trends may be due more to demolition activity than new construction or improving physical condition.

Studying the data based on the ranges defined in 2009 reveals additional insights.  Most of the block groups in that were in the top category in 2009 experienced a decline in average condition rating; only one in eight showed improvement.  In contrast, over 60% of the neighborhoods in the weakest category in 2009 observed an improvement in average condition.  In 2009, 25% of block groups had 95%+ of structures rated in good condition.  By 2014, this number had declined by 20%, and the number of block groups where less than two-thirds of all structures received a good rating increased from 202 to 269.  The average score in the top quarter of block groups declined by 0.042 while the average in the bottom quarter showed an improvement of 0.036.

Conclusions

Although five years is a relatively short time in the life of a city, there has been a noticeable decline in the average condition of residential structures since 2009.  The areas where the best-quality housing predominates are shrinking, while the pockets of blight are growing. While there are large areas where the average structural condition rating has improved, this appears to have most often been the result of the demolition of the poorest-condition homes.   In general, the changing landscape of average residential structure condition illustrates a concerning trend in the city that must be reversed if Detroit’s neighborhoods are to have any chance at recovery.

*        *        *

Next week, City of Change will examine occupancy in Detroit’s neighborhoods, and the changing patterns of where the city’s residents call home.

*        *        *

The Motor City Mapping data are available in their raw format on the Data Driven Detroit (D3) Open Data Portal – http://portal.datadrivendetroit.org.

Introducing City of Change – A window into Detroit’s residential neighborhoods from 2009 to 2014

Introducing City of Change – A window into Detroit’s residential neighborhoods from 2009 to 2014

In 2009, Data Driven Detroit (D3) participated in the Detroit Residential Parcel Survey (DRPS), collecting data on roughly 350,000 structures and vacant lots in the city of Detroit.  The survey captured information on the physical condition of Detroit’s residential neighborhoods and empty lots.  The DRPS provided a snapshot of Detroit’s urban fabric that has formed the backbone of policy efforts across Detroit, ranging from the high-profile Detroit Future City project to neighborhood-oriented mortgage market studies and local community agriculture endeavors.

Eventually, passage of time rendered the DRPS dataset less representative of current conditions in the city and thus less useful for decision-makers.  In the winter of 2013, the Motor City Mapping project once again undertook the collection of parcel-level data in Detroit.  On this project, D3 worked with multiple partners, including the Michigan Nonprofit Association, LOVELAND Technologies, and Rock Ventures, to survey every property in the city, regardless of use.  Using teams of resident surveyors and volunteer drivers, Motor City Mapping covered nearly 380,000 parcels in only six weeks of field work, providing information on structural condition and occupancy.  Many of the definitions used in Motor City Mapping were adopted from the DRPS.  In addition, D3 incorporated more than twenty other datasets into Motor City Mapping, creating the most comprehensive property database ever for Detroit.

The Motor City Mapping project provides a new benchmark dataset for policymakers and analysts.  With Motor City Mapping and DRPS combined, it is now possible to compare data across time with an unprecedented level of granularity, illuminating how Detroit’s neighborhoods have changed from 2009 to 2014.  The observed changes, based on two data points just five years apart, illustrate a small slice of a constantly-evolving environment.

City of Change is a new weekly D3 blog series dedicated to using these newly available data to explore how Detroit has changed over the past five years.  We assembled indicators that tell the city’s story from a number of different perspectives.  We then mapped these indicators at the census block group level, comparing 840 separate geographies between 2009 and 2014.  The insights offered by these comparisons are striking, frequently shocking, and occasionally hopeful.  They reinforce some of the trends that have been well-documented over the past five years, and shed new light on others.  They paint a picture of both tremendous decline and overwhelming potential.  They highlight neighborhoods that have faced tremendous stresses over the past five years, as well as areas that have endured the city’s continued crises, and even some areas where nascent turnarounds may be starting to become more entrenched.

This series is organized around two main themes.  The first part of the series will evaluate the various changes that have taken place over the preceding half-decade at a city-wide level, examining trends in Detroit’s population, housing, and markets.  The second half of the series will examine several high-profile, geographically-concentrated investment initiatives, with a particular focus on the changes within these various areas compared to the rest of the city over this five-year period.

Next week, we’ll take a deeper dive into Detroit’s structural climate – where buildings in the best condition in 2009 were located, which areas appear to have improved given the updated data from Motor City Mapping, and which areas appear to be facing the greatest threat from declining structural condition.

*             *             *

For more information about the Motor City Mapping project, please visit www.motorcitymapping.org.  You can download the full, parcel-level survey results from D3’s Open Data Portal, http://portal.datadrivendetroit.org.  We’ll be posting many additional datasets from the Motor City Mapping comprehensive property database, so be sure to check the Open Data Portal regularly in the coming weeks!