Diana Gonzalez_ Final Blog

Evaluating the Concept of a Jobs-Housing Balance through LA County

The concept of a jobs-housing balance is seen as a solution to the problems of congestion, commutes, and pollution that plague American urban areas. The basic idea behind this concept is that if people can both live and work in a given area (and thus that area is in balance), then there would be no need for a commute to work. No commutes reduces the congestion and pollution that result from automobile use. Living in an area with more housing then jobs creates commutes since people have to travel outside of the area to work. In this GIS project I use Los Angeles County to explore the validity of this concept. I ask whether this concept can really be used to reduce congestion, commutes, and pollution.

I begin by breaking down the county by neighborhoods and calculating the jobs- housing ratio for each. I classify each neighborhood as balanced, housing rich, and jobs rich. I then explore the relationship between a neighborhood’s ratio and commute times. My maps reveal an apparent relationship between the jobs-housing ratio and commute times. Despite this relationship I argue, through my maps, that the concept of a jobs-housing ratio cannot be used to reduce congestion, commutes, and pollution. I reveal 5 flaws in the concept that make its use quite problematic.

Layout #1: LA County Jobs Housing Ratio by Neighborhoods

This layout reveals that the majority of LA County is housing rich with very few balanced areas. The job rich areas are mainly along the coast, West LA, South Bay, Downtown, and in the industrial parts of the county.

Jobs data from the U.S. Census 2009 County Business Patterns Survey and housing data from the U.S. Census 2009 American Community Survey (ACS) was used to create the jobs-housing ratios. Classification of what areas are balanced, housing rich, or jobs rich was taken from the Southern California Association of Government’s (SCAG) RTP 2012 Growth Forecast report. This report defines an area with more housing then jobs as housing rich with a ratio less than 1, an area in balance with a ratio between 1 and 1.29, and an area with more jobs than housing as job rich with a ratio greater than 1.29. The skills used to make this layout include inset map, geoprocessing (clip and merge to create the neighborhoods), aggregating attribute fields (created a new field in the attribute table and used field calculator to divide the number of jobs in a given neighborhood by the number of housing units in that neighborhood), attribute sub-sets selections (freeways taken out of a larger road file by selecting freeways in the attribute table and making a new shapefile), and charts.

Layout #2: Average Commute Time Comparison

This layout seems to reveal that a relationship does exist between a neighborhood’s ratio and its average commute time, especially for housing rich areas. Those areas that are housing rich, with a low ratio, generally have higher than average travel times (the average being 29 minutes, the mean travel time for Los Angeles County). This is particularly noticeable in the housing rich areas of Northern LA County, South Central Los Angeles, and East San Gabriel Valley.

Mean travel times by neighborhood was calculated using data from the 2009 U.S. Census 2009 American Community Survey (ACS). I summed up the aggregate travel time in minutes for every tract in a given neighborhood which gave me the total minutes that workers in that neighborhood spent commuting. I then added up the total number of workers in each tract within that neighborhood. I calculated average commute time by aggregating attribute fields (created a new field in the attribute table and used field calculator to divide total minutes spent commuting in a neighborhood by total number of workers in neighborhood to get mean travel times).

Layout #3: Comparing Two Different Neighborhoods: Beverly Hills and Vermont Square

In this layout I continue to explore the apparent relationship between jobs housing ratio and commute times by looking at two specific neighborhoods. The first neighborhood I selected was Beverly Hills in West LA which is jobs rich (with a ratio of 2.42) and has a lower than average commute time (22 minutes). The second neighborhood I chose was Vermont Square in South Central Los Angeles. This neighborhood is housing rich (with a ratio of .23) with a higher than average commute time (33 minutes). Both neighborhood support the idea of the lower the ratio the longer people have to commute and the higher the ratio the less the commute.

The same data from layer one was used to create this layout. The skills used include inset map and graduated symbols.

Layout #4: Why the Difference in Ratio? Comparing BH and VS Land Parcel Use

I wanted to further explore why the difference in ratio between the two neighborhoods so I decided to use land assessor parcel data to see how land use in each neighborhood differed. I looked at only one zip code per neighborhood and noticed that the difference does not seem to stem from greater commercial activity in Beverly Hills. The difference seems to be because there are more apartment buildings in Vermont Square and thus more housing units. The greater the number of housing units, the lower the ratio.

As already stated I used land assessor parcel files for this layout. Skills used include inset map and original data (base map was created using georefrencing).

Layout #5: Comparing BH and VS Average Commute Distance

Problem #1: Why are Job Rich Residents Commuting Outside Job Rich Areas

Beverly Hills residents have an average commute time of 22 minutes while Vermont Square Residents have an average commute time of 33 minutes. Using network analyst I was able to create a buffer showing the average commute distance of residents in each neighborhood. Vermont Square’s small jobs-housing ratio seems to help explain its larger commute distance. But what about Beverly Hills? If Beverly Hills is so jobs rich why does it have a commute that goes beyond the neighborhood boundaries? According to the concept of a jobs housing balance, people are supposed to live and work in the same area if it is balanced. The same rule applies if the area is job rich since there are enough jobs for residents. This is not the case in Beverly Hills though.

This map was created using U.S. Census 2009 American Community Survey and UCLA Map Shares LA County roads file. Skills used where network analysis which fulfills the measurement/analysis requirement.

Layout #6: % of Workers Working Outside their Balanced Neighborhoods

Problem #2: Why are residents of balanced neighborhoods working outside?

According to the concept of a jobs-housing balance, reaching a balance should eliminate commutes since people can both live and work in their balanced neighborhood. If this is the case then why is it that there are residents in every balanced neighborhood in Los Angeles County commuting outside their neighborhood? This map highlights all the balanced neighborhoods in LA County and gives examples of the percent of worker in some neighborhoods that commute to a job outside that neighborhood.

Data on percent of workers who work outside their place of residence from the U.S. Census 2009 American Community Survey.

Layout #7: Multiple Workers per Household

Problem #3: What does living near work mean for households with more than one worker?

One of the basic ideas behind the jobs-housing ratio is that workers will try to live as near as they can to their jobs. This is problematic in today’s world where the increasing majority of household have more than one worker. What does living near work mean for a two worker household? For instance living by one person’s job may mean living far away for the other person’s job. As shown through this layout, the majority of neighborhoods in LA County average more than one worker per household.

U.S. Census 2009 American Community Survey data was used for this layout. Workers per household was calculated by aggregating attribute fields (created a new field in the attribute table and used field calculator to divide total number of workers in a neighborhood by total housing units in a neighborhood to calculate average number of workers per household).

Layout #8: Changing the Geographic Scale to Ideal Commute Area

Problem #4: What is the appropriate area at which to calculate the ratio?

For this project I used neighborhoods as the scale in which to measure the ratio but there is no consensus as to what the appropriate scale actually is. Is it by census tract? Is it by county? The problem is that the classification (housing rich, balanced, or job rich) of the area a resident may live in can differ depending on the scale at which the ratio is calculated. In this layout I calculate the ratio that residents of three different neighborhoods live in (Encino, Lynwood, and Azusa) based on ideal commute area instead of neighborhood boundaries. Ratio by neighborhood indicate residents of Encino living in a job rich environment while residents of Lynwood live in a housing rich environment. Residents of Azusa live in a balanced neighborhood. By calculating the ratio by ideal commute area Encino and Lynwood residents now live in a balanced neighborhood together with Azusa residents. Thus, the issue of what geographic scale is the appropriate scale makes using this concept problematic.

I used U.S. Census 2009 American Community Survey jobs data and U.S. Census 2009 County Business Patterns Survey housing data. In order to calculate ideal commute area I needed the average speed at which LA County residents travel. This is 30 mph according to LA Metro. I also needed an estimate of ideal commute time. According to Redmond and Mokhtarian in their article “The Positive Utility of the Commutes Modeling Ideal Commute Time and Relative Desired Commute Amount” ideal commute times fall between 15 and 19 minutes. I used 17 minutes. By multiplying 17 minutes by 30 mph I got an ideal commute area of 8.5 miles. Skills used for this map include buffering, extracting data from a buffer (calculating the new ratio required extracting and dissolving the jobs and housing data from all neighborhoods that fell within the 8.5 mile buffer around the centers of Encino, Lynwood, and Azusa), modeling, and measurement/analysis.

Layout #9: Changing the Definition of Balance to 1.5

Problem #5: What is the appropriate balance ratio?

For this project I used SCAG’s definition of balanced areas as having a ratio between 1 and 1.29. Much like with geographic scale there is also little consensus on what ratio means an area is balanced. Cervero in his article “Jobs Housing Balancing and Regional Mobility” puts balance at a ratio of 1.5. In this layout I highlight those neighborhoods whose status changed when balanced is defined as 1.5. About 40 neighborhoods see their status change.

Sources for this layout include the U.S. Census 2009 American Community Survey, the U.S. Census 2009 County Business Patterns Survey, and the article by Cervero. A chart was used to display the number of neighborhoods whose status changed.


There does seems to be a connection between a neighborhood’s job-housing ratio and average commute time, as exemplified by Vermont Square and Beverly Hills. Nevertheless this concept cannot be used to reduce congestion, commutes, and pollution because the concept itself is quite problematic. The following were the problems discussed in this project:

– Workers in balanced and jobs rich neighborhoods working outside their neighborhood

– Multiple workers per household

– Lack of consensus on the appropriate geographic scale to calculate the ratio

-Lack of consensus on what ratio means balance

A concept that is so problematic should not be used to tackle the issues of congestion, commutes, and pollution. These two issues should be tackled directly and not indirectly through this concept of a jobs-housing balance. I believe GIS has proven to be very useful to address this type of issue. Visualizing neighborhood data for all of Los Angeles County is nearly impossible to do without maps. GIS helped enhance the understanding and clarity of this issue.

List of Skills Used

1. Original Data

2. Measurement/Analysis

3. Modeling

4. Inset Map

5. Geoprocessing

6. Aggregating Attribute Fields

7. Graduated Symbols

8. Network Analyst

9. Extracting Data from a Buffer

10. Buffering

11. Charts and Images

12. Attribute Sub Selections



My biggest challenges where creating a neighborhood shapefile and manipulating my data for neighborhoods since all the information was in census tracts or zip codes. I could not find a shapefile of Los Angeles County neighborhoods. I thus had to make my own shapefile of LA county neighborhoods modeled after the LA Times neighborhood maps. I downloaded a census tract shapefile and joined it with a excel sheet containing all my data on jobs and housing by tract. Using the union tool under editor and the merge tool under geoprocessing I went through every single census tract in the county and placed them in their corresponding neighborhood. Another challenge was getting the jobs data. All my data was from the 2009 Census American Community Survey (ACS) and by census tracts except for the jobs data. I could only find jobs data in the 2009 U.S. Census County Business Patterns Survey. The data was not provided in census tracts but by zip code. So I had to figure out what tracts belonged in which zip codes, and split up the number of jobs in a given zipcode among tracts according the percentage of land a given tract held within that zipcode.


(2008). Rtp 2012 growth forecast. Los Angeles: Southern California Association of Governments. Retrieved from http://www.scag.ca.gov/forecast/downloads/excel/RTP2012-GROWTH- FORECAST.xls

(2012). Los Angeles County Parcels. Los Angeles: Office of Assessor. Retrieved from http://gis.ats.ucla.edu//Mapshare/Default.cfm#

Cervero, R. (1989). Jobs-housing balancing and regional mobility. Journal of the American Planning Association, 55(2), 136-150. Retrieved from www.uctc.net/papers/050.pdf

Metro. (2012). I-405 Sepulveda Pass Improvement Projects. Retrieved February 7, 2012 from http://www.metro.net/projects_studies/I405/images/I405-Project-Benefits-Fact-Sheet.pdf

Redmond, L. and Mokhtarian, P. (2001 ). The positive utility of the commutes modeling ideal commute time and relative desired commute amount. University of California Transportation Center. Retrieved from http://uctc.net/research/papers/526.pdf

U.S. Census Bureau. (2009). Aggregate travel time to work (in minutes) of workers by sex. Retrieved February 6, 2012, from http://factfinder2.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ACS_09_5YR_B08013&prodType=table

U.S. Census Bureau. (2009). Commuting Characteristics by Sex. Retrieved February 6, 2012, from http://factfinder2.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ACS_09_5YR_ S0801&prodType=table

U.S. Census Bureau. (2009). County business patterns: zip code business statistics: total for zip code. Retrieved January 23, 2012, from http://factfinder2.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=BP_2009_00CZ1&prodType=table

U.S. Census Bureau. (2009). Sex of workers by place of work-place level .Retrieved March 14, 2012, from http://factfinder2.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ACS_09_5YR_B08008&prodType=table

U.S. Census Bureau. (2009). Universe: Housing units. Retrieved January 25, 2012, from http://factfinder2.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ACS_09_5YR_B25001&prodType=table