Dynamic Neighborhood Taxonomy Project Insights

Dynamic Neighborhood
Taxonomy
A Project of
LIVING CITIES
By
RW Ventures, LLC
MacArthur Foundation
June 2, 2008

Agenda
Background: The DNT Project
The Nature of Neighborhood Change
Digging Deeper: Specialized Drivers By Factors,
Types of Neighborhood and Patterns of Change
Implications 1.0: Dynamic, Specialized
Neighborhoods
I
Implications 2.0: Specialized Tools - From
Diagnostics to Investment
II
III
IV
V
Discussion: What Next?VI

Agenda
I Background: The DNT Project

About Living Cities
“A partnership of financial institutions, national foundations and federal
government agencies that invest capital, time and organizational
leadership to advance America’s urban neighborhoods.”
AXA Community Investment Program
Bank of America
The Annie E. Casey Foundation
J.P. Morgan Chase & Company
Deutsche Bank
Ford Foundation
Bill & Melinda Gates Foundation
Robert Wood Johnson Foundation
The Kresge Foundation
John S. and James L. Knight Foundation
John D. and Catherine T. MacArthur Foundation
The McKnight Foundation
MetLife, Inc.
Prudential Financial
The Rockefeller Foundation
United States Department of Housing &
Urban Development
LIVING CITIES PARTNERS:

Partners and Advisors
The Urban Institute
Participating Cities: Chicago, Cleveland, Dallas and Seattle
… And Over 70 Advisors including Practitioners, Researchers,
Funders, Civic Leaders and Government Officials

We Know Where We Want to Go...
Common Goal:
Building Healthier Communities

The Challenge:
Scarce Resources, Many Options
 Community-Based Organizations: select
interventions, identify assets and attract
investment
 Governments: tailor policy and interventions
 Businesses: identify untapped
neighborhood markets
 Foundations: target interventions,
evaluate impacts

Comprehensive Neighborhood Taxonomy
Business People
Real Estate Amenities
Social Environment
 Improvement or
deterioration within
type
 Gradual vs. Tipping
point
 From one type to
another
 Port of entry
 Bohemian
 Retirement
 Urban
commercialized
 Employment
 Education
 Crime
 Housing stock
 Investment
activity

Agenda
The Nature of Neighborhood ChangeII

Agenda
The Nature of Neighborhood ChangeII
a. Measuring Change: the RSI
b. Overall Patterns
c. Degree and Pace of Change
d. Neighborhoods and Regions
e. Drivers of Change

Agenda
IIa Measuring Change: the RSI

PHYSICAL:
Distance from CBD, vacancies, rehab activity, …
Drivers Model and Data
TRANSPORTATION:
Transit options, distance to jobs, …
CONSUMPTION:
Retail, services, entertainment, …
PUBLIC SERVICES:
Quality of schools, police and fire, …
SOCIAL INTERACTIONS:
Demographics, crime rates, social capital…

Theoretical Framework
 Use Demand for Housing as Proxy for Neighborhood Health
 Look at Quality Adjusted Housing Values to Capture
Neighborhood Amenities
 Look at Change in Quantity of Housing to Account for
Supply Effects
Amenities
Structure Rent
Housing
Price

The Challenge: Finding a Metric that Works
Issues:
 Measure change in prices controlling for change in quality of the
housing stock
 Estimate at very small level of geography
 Track continuous change over time
Solutions:
 Repeat Sales to Control for Changes in Neighborhood Housing
Stock
 Spatial Smoothing: Locally Weighted Regression to account for
“fluid” neighborhood boundaries and address sample size
 Temporal Smoothing: Fourier expansions to track change over
time

IIa
Developing the Repeat Sales Index (RSI):
Spatial and Temporal Smoothing
Correlations between different RSI Versions
p01 p01i p01s p01c p05 p05i p05s p05c p10 p10i p10s p10c nb
p01
p01i 0.96
p01s 0.82 0.89
p01c 0.53 0.58 0.82
p05 0.94 0.96 0.79 0.49
p05i 0.94 0.97 0.82 0.52 0.99
p05s 0.83 0.90 0.99 0.77 0.84 0.87
p05c 0.53 0.59 0.82 1.00 0.50 0.53 0.79
p10 0.92 0.94 0.76 0.47 0.99 0.99 0.82 0.49
p10i 0.92 0.95 0.79 0.50 0.98 0.99 0.85 0.51 0.99
p10s 0.84 0.90 0.97 0.75 0.85 0.88 1.00 0.77 0.83 0.86
p10c 0.53 0.59 0.82 0.99 0.51 0.54 0.79 1.00 0.49 0.51 0.77
nb 0.11 0.13 0.11 0.07 0.13 0.13 0.12 0.07 0.13 0.13 0.12 0.07
Number of Fourier Expansions
Measuring Change: the RSI
Optimizing sample size and fluid boundaries through
extensive modeling and cross-validation procedures

Final Product: The DNT RSI
RSI Estimation Coverage Using Case/Shiller Method
Time Period: 2000 - 2006
RSI Estimation Coverage Using DNT RSI Method
Time Period: 2000 - 2006
Improving upon traditional repeat sales indices, the DNT RSI can be estimated
for very small levels of geography, and is more accurate, more robust and less
volatile.

Looking at Particular Tracts:
Appreciation in Greater Grand Crossing

Contour Model of 2004-06 Prices around Mt. Prospect Property
Spatial Distribution of Housing Values –
Mount Prospect
1450 S. Busse
Sales Transactions
Tract Boundaries

Chicago Neighborhood Change, 1990-2006
1990
IIb Overall Patterns

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

Initial Conditions and Appreciation
APPRECIATION
14.9
-0.17
MEDIAN HOUSING PRICES 1990
Up to $38,000
$38,001 - $63,750
$63,751 - $107,500
Over $107,500
Up to 52,600
$52,601 – $72,125
$72,126 – $108,175
Over $108,175

Change in Price: Poor Neighborhoods Present
the Most Opportunities for Investment
Many of the poorest neighborhoods are
the ones that grew the most,
outperforming wealthier communities
in each of the four sample cities.

Partly Due to Lack of Information,
These Areas Are Also the Most Volatile
TEMPORAL VOLATILITY OF INDEX
0.14 - 0.93
0.94 - 1.33
1.34 - 2.54
2.55 and above
APPRECIATION
14.9
-0.17
By increasing the availability of
information on these markets, we could
reduce risk, increase market activity,
and help stabilize these communities,
further strengthening their
performance.
0.25 – 0.53
0.53 – 0.60
0.60 – 0.79
0.79 – 11.2

Applications and New Capacities
Develop the “S&P/Case Shiller Home Price Index”
Equivalent for Neighborhood Markets
 Maintain DNT RSI and Expand it to New Markets
 Create a Real Estate Information Company to Market
and Distribute the RSI
 Specialize in Guiding Investment in Workforce
Housing

Beyond Change in Price:
Relationship of Price and Quantity
Tract 016501: 17% Developable Parcels in 1990Tract 016501: 17% Developable Parcels in 1990
Tract 002900: 0.05% Developable Parcels in 1990Tract 002900: 0.05% Developable Parcels in 1990
Development of new housing in a neighborhood can keep prices more affordable, as
places with greater constraints on new construction experience faster appreciation.

Neighborhood Change Is a Slow Process
Neighborhood Mobility by Time Interval
5 Years
10 Years
15 Years
No Change 1 Quintile 2 or More Quintiles
58%
33%
8%
64%
30%
6%
71%
25%
4%
IId Degree and Pace of Change
Even over 15 years, most neighborhoods do not change their
position relative to other neighborhoods in the region.

Yet Substantial Change Occurs
in Select Neighborhoods
Median Sales Price Transition Matrix
Cleveland, 1990-2004
Final Quintile
Initial Quintile 1 2 3 4 5
1 76.9% 15.4% 7.7% 0.0% 0.0%
2 5.1% 51.3% 25.6% 15.4% 2.6%
3 2.6% 26.3% 26.3% 39.5% 5.3%
4 7.7% 2.6% 28.2% 23.1% 38.5%
5 7.7% 5.1% 10.3% 23.1% 53.8%
IId Degree and Pace of Change
In Cleveland, 13% of all the tracts at the bottom of the distribution
in 1990 moved up to the top 2 quintiles 15 years later.

Neighborhoods Tend to Move with their Regions
IIe Neighborhoods and Regions
Most neighborhoods follow their region closely, but there are important exceptions.

Regional Effects Vary by Area
Across Cities, 35% of Neighborhood
Change is Accounted for by Regional Shifts
R Squared from Regression Models of Tract RSI on Region
Cleveland Chicago
Dallas
Seattle

Regional Effects Vary by Area
Across Cities, 35% of Neighborhood
Change is Accounted for by Regional Shifts
R Squared from Regression Models of Tract RSI on Region
Cleveland Chicago
Dallas
Seattle
81%
57%
28%
7%
Neighborhood performance depends upon
neighborhood characteristics, regional trends,
and linkages between the two

Drivers of Change: The Big Picture

Drivers of Change: Modeling Strategy
 Two Dependent Variables:
– Change in Quantity of Residential Units
– Quality-Adjusted Change in Price (RSI)
 Three Sets of Models:
– 1990-2000 Decennial Model
– 1994-2004 Time Series
– 1999-2004 Time Series
 Three Specifications:
– Regress Change on Initial Conditions
– Random Effects (Time Series Including both Initial
Conditions and Change for Independent Variables)
– Fixed Effects (Time Series with Time-Variant Independent
Variables Only)
 Models Across Four Cities Supplemented with City-Specific
Models for Chicago
IIe Drivers of Change

The Big Picture:
Urban Neighborhoods Are Coming Back
Chicago Cleveland Dallas Seattle
0.00%
1.00%
2.00%
3.00%
4.00%
5.00%
6.00%
7.00%
8.00%
9.00%
Average Tract Appreciation Rate (1990-2004)
Suburbs
City
Over the past 15 years, tracts in the central city
grew faster than tracts in nearby suburbs.

The Big Picture:
Neighborhood Change = Changing Neighbors?
Ratio of HMDA Borrower Income (2000-2005) to Census Income (2000)
The primary mechanism of change overall appears to be the movement of people.

Exploring the Relative Importance of Different
Drivers of Change
(1994-2004 Random Effects Model, Standardized Coefficients with 95% Confidence Interval)

The Big Picture:
Drivers of Neighborhood Change
 Mobility is the key mechanism of change
 Movers are attracted to areas with undervalued housing
but sound economic fundamentals (employment, income,
education, young adults)
 Being connected is important: proximity to job centers,
access to transit, lower commuting times are positive
 Cultural and Recreational Amenities (art galleries, bars
and restaurants) help, but are not the main event
“The Goldilocks Theory” …
… Neighborhoods of Opportunity are “Just Right.”

The Big Picture:
Drivers of Neighborhood Change
 Race is still a factor: even controlling for income, influx
of minorities in a neighborhood leads to lower
appreciation – but there are important exceptions
 Neighborhood Spillovers are important: what happens in
your neighborhood reflects what happens in the
neighborhoods around you
 Context matters: substantial variation by type and stage;
current conditions have large effects on degrees and
patterns of change

DETROIT—Notorious for its abandoned buildings,
industrial warehouses, and gray, dilapidated roads,
Detroit's Warrendale neighborhood was miraculously
revitalized this week by the installation of a single, three-
by-four-foot plot of green space.
The green space, a rectangular patch of crabgrass
located on a busy median divider, has by all accounts
turned what was once a rundown community into a
thriving, picturesque oasis, filled with charming shops,
luxury condominiums, and, for the first time ever, hope.
The Johansens, who just moved to Warrendale, enjoy
some outdoor time.
3'-By-4' Plot Of Green Space Rejuvenates Neighborhood
FEBRUARY 11, 2008 | ISSUE 44•07
The “Little” Picture: Few Silver Bullets

The “Little” Picture: Few Silver Bullets
What matters varies a great deal by type of neighborhood.
Preliminary – For Illustration Purposes Only Preliminary – For Illustration Purposes Only

The Effect of LIHTC Projects:
Good in Moderation
[Overall Model Result: Non Significant]
Specialized Drivers: Looking at Particular FactorsIII
LIHTC projects help, up to a point: as the concentration of LIHTC units exceeds
a certain threshold, the effect becomes less positive
EffectonDNTRSI,PartialResiduals
(GAM Output based on 1994-2004 Fixed Effects Model, All Cities)

The Effect of Sub-Prime Lending:
A Four Year Fallout
-14
-12
-10
-8
-6
-4
-2
0
2
4
6
1 2 3 4 5
Year
Effect of Subprime Lending on Neighborhood Appreciation
RegressionCoefficient,DNTRepeatSalesIndex
[Overall Model Result: Positive and Significant]
(1994-2004RandomEffectsModel,AllCities)
By monitoring the levels of sub-prime activity over the past few years, practitioners can have a
good sense of how the effects will play out in the years to come.

 Detailed Analysis of Particular Factors of Interest:
e.g. Access to and Use of Credit, Different Types of Business
Establishments, Arts and Cultural Centers, etc.
 Further Investigation of the Relationships
between Drivers of Change: e.g. Examine Interactions
between Different Drivers (e.g. Crime and Transit)
 Return on Investment Analysis: Tie Magnitude of Effect
to Cost of Interventions to Assess which Strategies are Most
Cost Effective

Specialized Drivers: Looking at Particular Neighborhood SegmentsIII

Neighborhood Convergence
Beta Convergence in Cook County, 1990 - 2005
Neighborhoods Now Tend to Naturally Catch Up to Each Other

But not All Neighborhoods Follow this Trend...
Neighborhood Convergence
Beta Convergence in Cook County, 1990 - 2005

Characteristics of Poorer Neighborhoods
that Tend to Converge
 Closer to the Central Business District
 In neighborhoods with more turnover and potential
for redevelopment
– More apartment buildings; lower homeownership; less
residential stability
 Near Neighborhoods with more amenities and social
capital
– Proximity to Transit, Grocery Stores, Art Galleries and
Eating Establishments

Improvement With High and Low Turnover

Improvement With High and Low Turnover
What characterizes the neighborhoods that improved with less displacement?

Drivers of “Improvement in Place”
Improvement with Low Turnover Is Associated With:
 High Home Ownership Rates
 Low Vacancy Rates
 Access to Transit
 Reduction in Unemployment
 Presence of Employment Services
 High Social Capital
 High Percentage of Young Adults

Targeting Interventions:
 Identify neighborhoods that are more likely to converge,
prioritize affordability preservation efforts
 Focus development interventions to places that are less
likely to be “rediscovered” by the market.
Further Investigation of Specific Neighborhood
Segments:
 Analysis of Stable Mixed-Income Communities
 Analysis of Drivers of Improvement in Place for Specific
Subsets of Neighborhoods (e.g. Immigrant, Startup
Family, etc.)

Overall Summary  Implications
 Neighborhoods are highly specialized
 Neighborhood change is a function of people and money
moving in and out
 This in turn varies based on neighborhood type and stage
of development – different people and investors choose
different neighborhoods in the context of larger markets
and systems
 As a result, what matters varies by place. For any given
neighborhood, the goal could be continuity or change in
type, and implementation entails understanding who you
want to stay or move in, and what factors matter
to them

Summary  Implications
Two major implications:
1. We need a framework for understanding
neighborhoods as dynamic, specialized, and
nested in larger systems
2. We need much better tools for customized
analysis of local economies

Agenda
Dynamic, Specialized Neighborhoods (with Special Thanks to Andy Mooney!)IV

Neighborhoods are Complex
Dynamic, Specialized Neighborhoods (with Special Thanks to Andy Mooney!)IV

Neighborhoods are Complex
Dynamic, Specialized NeighborhoodsIV

Neighborhoods are Dynamic

Neighborhoods are Nested in Larger Systems
Which Drive the Flows of People and Capital

Neighborhoods are Nested in Larger Systems
which drive the Flow of Capital and People

Functioning Neighborhoods Connect Residents
and Assets to Larger Systems
Poverty Productivity
Connectedness
Isolation

Functioning Neighborhoods Connect Residents
and Assets to Larger Systems
 Employment networks
 Entrepreneurial opportunities
 Business, real estate investment
 Expanded products and services
 Productive, healthy
communities
 Undervalued,
underutilized assets
Poverty Productivity
Connectedness
Isolation

Applying the Framework
STEP 1A: What type of neighborhood do you want to be?

Starter Home Community

STEP 1B: What drivers will get you there?

STEP 1B: What drivers will get you there?
• Specific Retail Amenities
• Child Care
• Schools
• Safety
• Affordability

ECONOMIC SYSTEM
STEP 2: Identify Relevant System
Retail
Markets

STEP 3: Identify Change Levers Within System
ECONOMIC SYSTEM
Retail
Markets

STEP 3: Identify Change Levers Within System
Commercial Land Assembly
(production – costs)
Specialized Market Data
(exchange – finding costs)
ECONOMIC SYSTEM
Retail
Markets

STEP 4: Specify Interventions
Commercial Land Assembly
(production – costs)
Specialized Market Data
(exchange – finding costs)
ECONOMIC SYSTEM
Retail
Markets

Developing New Tools for the Field
Question/Goal Tool
How will a specific intervention affect its surrounding area? Impact Estimator
Enabling investment in inner city real estate markets RSI  REIT
What drivers differentiate neighborhoods with respect to a
specific outcome of interest?
CART
Identify comparable neighborhoods based on drivers of
change and other key characteristics
Neighborhood Typology
What neighborhoods are similar along particular factors of
interest?
Custom Typologies
How does the impact of an intervention vary in different
places?
Geographically Weighted Regression
Track affordability and neighborhood housing mix Housing Diversity Metric
Anticipate and manage neighborhood change Pattern Search Engine
Identify “true” neighborhood boundaries Locally Weighted Regression
Specialized Tools - From Diagnostics to InvestmentV

Impact Estimator
What It Does:
 Estimate impact of an intervention on
surrounding housing values (or on other outcome,
e.g. crime)
Possible Applications:
 Evaluate the impact of a development policy
 Choose among alternative interventions based on
estimated benefits to the surrounding community
 Advocate for a specific intervention

Example: What is the effect over time and
space of LIHTC housing?
PRELIMINARY – FOR ILLUSTRATION PURPOSES ONLY
Monte Carlo Simulation to Estimate Impact Variation with Distance
Homes within 1000 ft of an LIHTC site appreciate at a 4%
higher rate than homes between 1000 ft and 2000 ft.
Homes within 1000 ft of an LIHTC site appreciate at a 4%
higher rate than homes between 1000 ft and 2000 ft.

Impact of LIHTC on Surrounding Properties
Estimated Distance Decay Function – LIHTC ProjectsEstimated Distance Decay Function – LIHTC Projects
Distance from Project Location (in Miles)Distance from Project Location (in Miles)
DNTRepeatSalesIndex,1=FurthestAwayDNTRepeatSalesIndex,1=FurthestAway

Applying the Estimator to a Specific Project:
New Shopping Center in Chicago
New Shopping CenterNew Shopping Center
Estimated benefits to the community: $29 million in increased
property values, or an average of $1,300 per home owner.

Classification and Regression Tree (CART)
What It Does:
 Identify similar neighborhoods with respect to an
outcome of interest and its drivers
Applications:
 Identify leverage points to affect the desired
outcome
 Meaningful comparison of trends and best
practices across neighborhoods

Sample CART: Foreclosures
40 Variables Tested
What Neighborhoods Have Similar Numbers of Foreclosures...

Sample CART: Foreclosures
40 Variables Tested
Outcome:
Number of
Foreclosures (2004)
Drivers:
% Subprime Loans in
Previous Years
Mean Loan Applicant
Income
% FHA Loans
% Black Borrowers
... And Why

CART Output: Chicago Segments

Cluster 8: Defining Traits and Risk Factors
Segment Profile:
 Isolated, underserved, predominantly African American
communities. High rates of unemployment and sub-prime
lending activity.
Primary Risk Factor:
 Percentage of sub-prime
loans (primary driver of
foreclosures) is at its
highest and still on
the rise
Percentage of Sub-prime Loans by Year

DNT Neighborhood Typology
What It Does:
 Identifies distinct neighborhood types based on
drivers of change and other characteristics
Possible Applications:
 Tailor strategies to specific neighborhood types
 Benchmark neighborhood performance
 Peer analysis and relevant best practices
 Facilitate impact analysis

A Preliminary Taxonomy of Neighborhoods
Key Dimensions:
•People
•Income
•Age
•Foreign Born
•Place
•Land Use
•Housing Stock
•Business Types
Variables

Key Dimensions:
•People
•Income
•Age
•Foreign Born
•Place
•Land Use
•Housing Stock
•Business Types
Variables
Type 7: “Homeowners”Type 7: “Homeowners”Type 7: “Homeowners”Type 7: “Homeowners”

Key Dimensions:
•People
•Income
•Age
•Foreign Born
•Place
•Land Use
•Housing Stock
•Business Types
Variables
Type 8:Type 8:
““Young Professionals”Young Professionals”
Type 8:Type 8:
““Young Professionals”Young Professionals”

Taxonomy Structure: “Genus, Phylum, Species”

Chicago Neighborhood Types

Example: Type 3, “Low Income Stable”
Cluster Profile:
 These are lower income neighborhoods with a high
percentage of single family homes and a stable resident
base. These areas are also characterized by a significant
presence of vacant land and very few retail and service
establishments. Unemployment rates are lower than in
other low income neighborhoods, and neighborhood
residents tend to be employed in sales, service and
production occupations.
Other Characteristics:
 Neighborhoods in this cluster had the highest foreclosure
rates.
 Stable type, though 4% transitioned to type 7
“Homeowners”

Example: Type 3, “Low Income Stable”
• 177 Tracts in Chicago
• 20% of Total

Moving Down the Taxonomy:
From “Phylum” to “Species”

Applying the Taxonomy...

Tract 680800Tract 680800
Type 3-B,Type 3-B,
“Vacancies and“Vacancies and
Social Capital”Social Capital”
Change in Value 6%6% 125%125%
Median Income $26,319 $21,600
Vacant Units 18.5% 14.9%
Social Capital 1.5 4.2
Unemployment Rate 46.5% 19.3%
Turnover (% Moved in
Last Five Years) 47.4% 37.4%
Educational Attainment
– More than High School 25% 29.8%
Comparing Within Type:

Tract 680800Tract 680800
Type 3-B,Type 3-B,
“Vacancies and“Vacancies and
Social Capital”Social Capital”
Change in Value 6%6% 125%125%
Median Income $26,319 $21,600
Vacant Units 18.5% 14.9%
Social Capital 1.5 4.2
Unemployment Rate 46.5% 19.3%
Turnover (% Moved in
Last Five Years) 47.4% 37.4%
Educational Attainment
– More than High School 25% 29.8%
Comparing Within Type:
Identify Factors
Affecting
Neighborhood
Performance
Compared to Peers

Type 6-CType 6-C
New DevelopmentNew Development
Cluster 8Cluster 8
YoungYoung
ProfessionalsProfessionals
Cluster 7Cluster 7
HomeownersHomeowners
Age 19-34 35% 49% 24%
Stability (Less than
5) 69% 71% 39%
Single Family
Housing 27% 11% 71%
Commercial Land
Use 8.4% 7.7% 3.9%
School Quality
(reading test scores) 76 55 60
Homeownership 33% 34% 69%
Comparing Across Types:

Type 6-CType 6-C
New DevelopmentNew Development
Cluster 8Cluster 8
YoungYoung
ProfessionalsProfessionals
Cluster 7Cluster 7
HomeownersHomeowners
Age 19-34 35% 49% 24%
Stability (Less than
5) 69% 71% 39%
Single Family
Housing 27% 11% 71%
Commercial Land
Use 8.4% 7.7% 3.9%
School Quality
(reading test scores) 76 55 60
Homeownership 33% 34% 69%
Comparing Across Types:
What Would it Take
to Become a
Different Type of
Neighborhood?

Housing Diversity Metric
What It Does:
 Tracks the affordability and mix of the housing
stock (distribution, not just median)
 Applications:
 Enables tracking the range of housing available in
the neighborhood
 Better indicator of possible displacement than
median prices alone

Example: Tracking the Price Mix
Strong Overall Appreciation,Strong Overall Appreciation,
Range of Housing Options Is NarrowingRange of Housing Options Is Narrowing

Example: Tracking the Price Mix
Strong Overall Appreciation,Strong Overall Appreciation,
Range of Housing Options Is NarrowingRange of Housing Options Is Narrowing
Strong Overall Appreciation, butStrong Overall Appreciation, but
Range of Housing Options Is Still WideRange of Housing Options Is Still Wide

Percentage of Affordable Homes, 1990-2006
19901990

19911991

19921992

19931993

19941994

19951995

19961996

19971997

19981998

19991999

20002000

20012001

20022002

20032003

20042004

20052005

20062006

Sample “Affordability Reports”

Pattern Search
What It Does:
 For identified patterns of change, finds other
neighborhoods that have been through or are
going through the same pattern
Applications:
 Enables identifying comparable neighborhoods
with respect to particular patterns of change in
order to identify key factors and effects
 Enables anticipating and managing particular
patterns of change

Pattern Search Example:
Gentrification in Chicago
 Goal: Anticipating Neighborhood Change
 How it Works: Define a Pattern and Find Matching Cases
 Example: Possible Gentrification Pattern Defined Based
on a Neighborhood in Chicago

Zooming In: Wicker Park Area

Pattern “Spreading”
to Nearby Tracts

Possible Application:
Anticipating and Managing Gentrification
Different Appreciation Patterns Found in the DNT RSI

“LWR+” (Locally Weighted Regression +)
What It Does:
 Much more granular examination of neighborhood
dynamics
 Showing what areas share common trends;
identify “true” neighborhood boundaries
Applications:
 Better tailor interventions to areas undergoing
different kinds of change
 Assess appropriate geographic scope of
interventions

Example: Logan Square
“Standard” Look at a Community Area:
Overall Increase in Median Home Values

“Dissecting” Community Trends: Using LWR to
Uncover Local Dynamics
Rapid change in Southeast section, which went from
cheapest to most expensive area in the neighborhood

One Neighborhood... Or Three?

One Neighborhood... Or Three?
Distinct Patterns of Change in Different Parts of the Neighborhood

Filling In the Picture: Real Estate Dynamics
Higher Levels of Rehab and New Construction
in Rapidly Appreciating Section

Filling In the Picture: Demographic Shifts
-25%
-20%
-15%
-10%
-5%
0%
5%
10%
Overall Northwest Central Southeast
Percent Change in Hispanic Population, 1990-2000
0%
20%
40%
60%
80%
100%
120%
140%
Overall Northwest Central Southeast
Percent Change in Household Income, 1990-2000
Higher income, White households moving into Southeast area
possibly pushing original Hispanic population Northwest

Filling In the Picture: Business Patterns

Filling In the Picture: Business Patterns
Number of bars and restaurants doubled in Southeast
section, while it remained the same in the rest of the
neighborhood

Summary: Applying the Findings and Tools
Quick Case Study – using the tools for
neighborhood development stategies
Other Applications – using the tools to
target interventions

Applying Metrics, Findings and Tools:
Auburn Gresham

Step One: Figure Out Who You Are
(Challenges and Opportunities)
 Type 3-A, “Single Parents:” Low income, single family homes, stable
resident base. More children and single parent households.
 Underperforming relative to peers: 61% growth in RSI compared to
129% for peer group
 “Red Flags”:
– Population Loss: 6% decline between 1990 and 2000
– Business Attrition: Lost over 50% of its retail and service
establishments between 1990 and 2006
– Financial Distress: high percentage of credit past due, high
foreclosure rates
 Positive Signs:
– Decline in Unemployment: from 20% to 14% between 1990 and
2000
– Low Crime: crime rates are half of peer group’s

Step Two: What to Expect and Where You Can Go
 What to Expect Based on Model Findings:
– Unlikely to Converge: Targeted Development
Interventions Needed to Spearhead Change
– High Foreclosure and Sub-Prime Lending Rates, on the
Rise Through 2005: Negative Effects Should “Bottom Out”
around 2009-2010
– (Apply Pattern Search Tool: Identify Other Neighborhoods
with Similar Patterns of Change, see what happened next.)
 Where You Can Go (Based on Typology): Type 3 neighborhoods
tend to remain the same type, but can transition to Type 7,
“Homeowners” (stable neighborhood, similar housing stock, but
higher income).
 What to Aim For: Housing Stock and Demographics Make This
Neighborhood a Good Candidate for Improvement in Place (Which
Would Help Transition to Type 7)

Step Three: How Can You Get There?
 Key Areas of Focus (Based on Model Findings):
– Homeownership
– Build on Positive Employment Trends
– Improve Connection to Jobs
– Address Population Loss and Vacancies (possibly
related to foreclosure issues)
– (Social Capital and Retail)
 Possible Interventions:
– Perfect Location for Center for Working Families?
 Possible Additional Steps:
– Apply Model Findings to Assess Cost-Benefit of Alternative
Interventions
– Apply Impact Analyst to Evaluate Sites and Programs
– Apply LWR to identify which neighboring communities you
particularly need to work with

Other Applications:
Using CART to Improve Child Care Interventions
Outcome:Outcome:
• Number of Child CareNumber of Child Care
Slots per ChildSlots per Child
Drivers:Drivers:
• Distance to Downtown (+)Distance to Downtown (+)
• Percent of Foreign Born andPercent of Foreign Born and
Hispanics (-)Hispanics (-)
• Educational Attainment (+)Educational Attainment (+)
• Home Ownership (-)Home Ownership (-)
Segmentation Based on Child Care Capacity
0.49
0.27
0.25
0.19
0.18
0.11
Segment Child Care Slots per Child

Other Applications: Selecting Target Areas
Convergence Model
Results Can be Used
to Identify Areas that
Are in Greater Need
of Interventions

Other Applications:
Using Affordability Reports to Prioritize Housing Work
0
-94

Other Applications:
Using Affordability Reports to Prioritize Housing Work
0
-94
0%
-100%

Open Source: Multiple Parties Are Already
Interested in Moving the Work Forward
 LISC – Analysis of Impact of LIHTC Projects
 UMI/NNIP – Embed Tools in Existing Web Platforms
 Preservation Compact – Affordability Reports and Real Estate Metrics
 Chicago Department of Children and Youth Services – Targeting
Location of Youth Programming
 Metropolitan Mayor’s Caucus/Chicago Metropolis 2020 – Affordability
Reports, Real Estate Metrics, Patterns of Change Analysis
 MCIC – RSI and Other Housing Values Indicators
 MDRC – Analysis of Impact of New Communities Program
 Case Western – Analysis of Mixed-Income Communities
 Ansonia Properties, LLC and CityView – Apply Tools to Guide
Investment in Workforce Housing
What Next?VI

Building on the DNT Foundation:
Possible Next Steps
 Make the tools available to practitioners and investors
(e.g. by embedding them in existing web-based data
platforms)
 Apply the work to particular neighborhoods and
interventions (and improve the analysis and tools based
on repeated application)
 Maintain and update core metrics (particularly RSI) going
forward
 Extend the capacity to rest of the region and other cities:
expand data, metrics, models and tools
 Create a “learning network”?
What Next?VI

Building on the DNT Foundation
Additional Possible Directions of Further Work:
 Build on Typology Results to Identify Drivers by Type
 Analyze the Impact of Specific Development Interventions (from
employment centers to safety initiatives to foreclosure prevention
and remediation)
 Develop a Gentrification Early Warning System
 Identify What Amenities Attract Different Demographics
 Identify Opportunities for Workforce Housing
 Business Evolution: What are the Stages of Business Development in
Neighborhoods?
 ...
What Next?VI

Discussion
 General Comments and Questions
 What are You Trying to Better Understand About
Neighborhoods?
 What Impacts are You Trying to Achieve and Measure?
 What Tools and Applications Would Be Most Useful?
 Partners: Corollary Research, Tool Development and
Testing, Other?
DiscussionVI

Core Project Team
Christopher Berry, The University of Chicago
Graeme Blair, Econsult Corporation
Riccardo Bodini, RW Ventures, LLC
Michael He, RW Ventures, LLC
Richard Voith, Econsult Corporation
Robert Weissbourd, RW Ventures, LLC

Dynamic Neighborhood Taxonomy Project Insights

Recomendados

Recomendados

Mais conteúdo relacionado

Semelhante a Dynamic Neighborhood Taxonomy Project Insights

Semelhante a Dynamic Neighborhood Taxonomy Project Insights (20)

Mais de RWVentures

Mais de RWVentures (20)

Último

Último (20)

Dynamic Neighborhood Taxonomy Project Insights

Notas do Editor