\documentclass[10pt]{article}
\usepackage{fullpage}
\usepackage{setspace}
\usepackage{parskip}
\usepackage{titlesec}
\usepackage[section]{placeins}
\usepackage{xcolor}
\usepackage{breakcites}
\usepackage{lineno}
\usepackage{hyphenat}
\PassOptionsToPackage{hyphens}{url}
\usepackage[colorlinks = true,
linkcolor = blue,
urlcolor = blue,
citecolor = blue,
anchorcolor = blue]{hyperref}
\usepackage{etoolbox}
\makeatletter
\patchcmd\@combinedblfloats{\box\@outputbox}{\unvbox\@outputbox}{}{%
\errmessage{\noexpand\@combinedblfloats could not be patched}%
}%
\makeatother
\usepackage[round]{natbib}
\let\cite\citep
\renewenvironment{abstract}
{{\bfseries\noindent{\abstractname}\par\nobreak}\footnotesize}
{\bigskip}
\titlespacing{\section}{0pt}{*3}{*1}
\titlespacing{\subsection}{0pt}{*2}{*0.5}
\titlespacing{\subsubsection}{0pt}{*1.5}{0pt}
\usepackage{authblk}
\usepackage{graphicx}
\usepackage[space]{grffile}
\usepackage{latexsym}
\usepackage{textcomp}
\usepackage{longtable}
\usepackage{tabulary}
\usepackage{booktabs,array,multirow}
\usepackage{amsfonts,amsmath,amssymb}
\providecommand\citet{\cite}
\providecommand\citep{\cite}
\providecommand\citealt{\cite}
% You can conditionalize code for latexml or normal latex using this.
\newif\iflatexml\latexmlfalse
\providecommand{\tightlist}{\setlength{\itemsep}{0pt}\setlength{\parskip}{0pt}}%
\AtBeginDocument{\DeclareGraphicsExtensions{.pdf,.PDF,.eps,.EPS,.png,.PNG,.tif,.TIF,.jpg,.JPG,.jpeg,.JPEG}}
\usepackage[utf8]{inputenc}
\usepackage[greek,english]{babel}
\begin{document}
\title{Civic HW\_Analysis part}
\author[1]{sunglyoung Kim}%
\affil[1]{NYU Center for Urban Science \& Progress}%
\vspace{-1em}
\date{\today}
\begingroup
\let\center\flushleft
\let\endcenter\endflushleft
\maketitle
\endgroup
\sloppy
\textbf{Income integration and Housing market stability}
\textbf{Social Impact Project // Methods and descriptive Analysis}
\textbf{by: Yuwei Lin, Sunglyoung Kim, Fangshu Lin and Dana Chermesh
Reshef}
\textbf{Instructor: Prof. Constantine E. Kontokosta}
\textbf{Submission date: Nov 17th 2017}
\textbf{NYU CUSP, Center for Urban Science and Progress, New York}
\textbf{Overview}
\textbf{New York City is well known for its unaffordable housing market,
its increasing gap between rich and poor and the unfortunate process of
displacement that is related to rapidly gentrified areas (Freeman \&
Braconi, 2004). Economic segregation is strongly correlated to income
inequality (Watson, 2009). Economic segregation has increased during the
past three decades across the United States and in 27 of the nation's 30
largest major metropolitan areas (Fry \& Taylor, 2012; Pendall, \&
Carruthers, 2010).}
\textbf{This study aims to assess the correlation between income
integration level and housing market stability. In order to do so, there
were several approaches taken in the analysis process. First, we
included descriptive statistics of the data, Then linear regression
models were developed to analyze the relationships between income
integration level and income, rent burden (\%) and rent growth
separately. All data were in use for the analysis is CENSUS data for
years 1990, 2000, 2010.}
\textbf{What is housing stability? How we can define that. - Rent
burden}
\textbf{}
\textbf{Data Inventory}
\textbf{CENSUS Data}
\textbf{The CENSUS data were collected from the Geolytics Neighborhood
Change Database (NCDB). This dataset includes the Census data for 1970,
1980, 1990 and 2000 at the census tract level. Data of population,
income levels and rents were selected from the years 1990, 2000, 2010.
The data of 1990 and 2000 have been recalculated and normalized
according to the 2010 tract ID, in order to conduct comparisons of
historic data by the exact same tract boundary definitions. Since 2010,
the Decennial Census stopped using long form survey and only includes
some basic demographic and housing tenure information. Therefore,
income, house value and rent data for this time period were collected
from the American Community Survey (ACS) of 2006-2010 instead. Data were
cleaned and merged to overcome the different income levels for 1990 and
2000 census data.}
\textbf{The dataset includes both household and family numbers of
different income levels. Since there is no household income distribution
data published in 1990 census data, we chose to use family income
distribution data for analysis. Income boundaries following the NCDB
income distribution data (14 groups for 1990 and 16 groups for
2000/2010) were in use. As Galster (2008) mentioned in his work, the HUD
income boundaries measured in AMI do not match the NCDB income groups.
Therefore Gaster (2008) used interpolation method to divide the NCDB
income distribution data into six income groups. To avoid controversy,
we didn't manipulate the NCDB income distribution data which could be
done in future work.}
\textbf{Public Use Microdata Areas (PUMA) data}
\textbf{We choose PUMA as the main unit for our neighborhood level
analysis, given the data availability. PUMAs are the Census statistical
geographies, created by aggregating census tracts and designed to cover
100,000 residents per PUMA, revealing about 40 tracts. Aggregating the
2168 census tracts data into PUMAs was a straightforward task, with no
boundaries errors when merging data due to the fact that both geographic
units are used by US Census Bureau. Additionally, the 55 PUMAs in NYC
are almost exactly equivalent to the city's 59 Community districts;
except for 8 districts in Manhattan and Bronx that are combined into 4
PUMAs due to their small population with each PUMA comprising two
districts. Every Community district corresponds to a community board,
the local representative body, and is in accordance with the groups of
neighborhoods in NYC. Hence, it was reasonable to use PUMA data for
neighborhoods' level analysis.}
\textbf{Data were cleaned and arranged in a final dataset to be
analyzed. The final dataset contains the following information of 55
PUMAs:}
\begin{enumerate}
\tightlist
\item
\textbf{Median Family Income (1990, 2000, 2010) // Annually, divided
by income level default groups.}
\item
\textbf{Median Rent (1990, 2000, 2010) // Monthly}
\item
\textbf{Rent burden (\%); 1990, 2000, 2010 // Calculated by dividing
Median Rent (multiplied by 12, for annual median rent) by Median
Income.}
\item
\textbf{Income integration level (Range of 0-1); 1990, 2000, 2010 //
Calculated by Entropy Index as explained in the following methods
section.}
\item
\textbf{Rent growth (\%); 1990-2000, 2000-2010}
\item
\textbf{Income integration level change (\%); 1990-2000, 2000-2010}
\end{enumerate}
\textbf{Methods}
\textbf{Entropy index calculation}
\textbf{Entropy index is a commonly used method for segregation
measurement (Galster, Booza \& Cutsinger, 2008, Kontokosta, 2014). In
order to obtain the income integration level of each PUMA in the two
decades we used the Entropy Index formula, returning a score of 0-1
range, when 0 is fully segregated (only one group of income is
represented) and 1 is fully integrated (all income groups represented
evenly):}
\selectlanguage{greek}\textbf{πim = the percentage of the number of tract i consisting of
families from group m.}\selectlanguage{english}
\textbf{m = 1, 2, \ldots{}, M.}
\textbf{M = Number of income level groups.}
\textbf{Descriptive statistics}
\textbf{After organizing and calculating the data for the analysis, we
visualize the primary variables to observe their magnitude and trends
over the decades analyzed. First were the change of median rent and
median income, as shown in Fig.1.}
\textbf{}
\textbf{}
\textbf{Fig.1 Growth of Median Census Tracts' Income and Median Rent
between 1990-2010}
\textbf{Viewing a steeper slope of the median rent (red line) in
comparison to the median family income (blue line), it is clear that the
growth of median rent was more rapid than the growth of median income
between 1990-2010, which can explain the housing crisis and the city's
increasing rent burden. The rent burden is defined~}
\textbf{}
$\Rent Burden = \frac{\Medianrent \ Medianhouseincome}$
$\gamma_{i}(t)\equiv P(X_{t} = i \vert Y,\theta) = \frac{\alpha_{i} (t) \beta_{i} (t)} {\sum\limits_{j=1}^{N} \alpha_{j} (t) \beta_{j}(t)}$
\$\$ /frac\{Median rent / Median family income\} \$\selectlanguage{english}
\begin{figure}[h!]
\begin{center}
\includegraphics[width=0.70\columnwidth]{figures/RentBurden/RentBurden}
\caption{{This is a caption
{\label{427882}}%
}}
\end{center}
\end{figure}
\textbf{Fig.2 Rent burden distribution and median, 1990, 2000, 2010}
\textbf{Over the years the rent burden distribution's variance are
increased, as the median of the city's rent burden has consistently
increased. Also, clearly the distribution has long tail therefore it is
not a gaussian distribution. It informs that extreme rent burdened PUMA
exists and will need to be cleaned for analysis. ~~}
\par\null
\textbf{Fig.3 contains maps of NYC PUMAs' Rent Burden in 1990, 2000 and
2010.}
\textbf{Fig.4 Rent Burden by PUMA, 1990, 2000, 2010}
\par\null\selectlanguage{english}
\begin{figure}[h!]
\begin{center}
\includegraphics[width=0.70\columnwidth]{figures/Rentburden-change-1990-2010/Rentburden-change-1990-2010}
\caption{{This is a caption
{\label{227574}}%
}}
\end{center}
\end{figure}
\textbf{Fig.5 Percentage Change of Rent Burden by PUMA, 1990, 2000,
2010}
\textbf{Viewing the city's rent burden spatial distribution over the
years one can identify the PUMAs that are at risk for displacement due
to increasing rent burden. Also Figure 4 shows that the percentage
difference between rent burden 2010 and 1990. The map shows northern
brooklyn, northern east Queens, northern the Bronx, and Staten Island
has higher rent burden growth than other NYC neighborhoods.~The fig.5
shows dynamics of income integration level across NYC in PUMA. Overall,
most of the neighborhoods have become more integrated in two decades,
except top 6 median income level and Brooklyn Height. 5 out of 7
neighbors are in Manhattan and the others are located in Brooklyn.
~Fig.6 shows three maps of each year's spatial distribution of the
Income Integration level (entropy index).}\selectlanguage{english}
\begin{figure}[h!]
\begin{center}
\includegraphics[width=0.70\columnwidth]{figures/IncomeChange-1990-2010/IncomeChange-1990-2010}
\caption{{This is a caption
{\label{113379}}%
}}
\end{center}
\end{figure}
\textbf{Fig.6 Change of Income Integration level of 55 PUMAs in NYC,
1990-2010, sorted by median income}
\par\null
\textbf{Fig.6 Income Integration level (entropy index), 1990-2000-2010}
\textbf{}
\par\null\selectlanguage{english}
\begin{figure}[h!]
\begin{center}
\includegraphics[width=0.70\columnwidth]{figures/Income-Ite-sort/Income-Ite-sort}
\caption{{This is a caption
{\label{892918}}%
}}
\end{center}
\end{figure}
\textbf{}
\textbf{Fig.7 Percentage Change of Income Integration in PUMA level
district from 1990 to 2010}
\textbf{Having the red colors stand for income segregation and the green
colors revealing higher integration by income level, we can track trends
in the NYC's PUMAs. An example for PUMA that became more integrated is
Chinatown and Lower East Side. Park Slope, Carroll Gardens and Red Hook,
on the other hand, had higher income integration level in 1990 and
became more and more segregated by 2010. Most of Manhattan's PUMAs got
less income integrated, so as Downtown Brooklyn and Dumbo; while Harlem,
which we know is facing gentrification in the last few years, indeed
increases its income integration level.}
\textbf{OLS linear Regression}
\textbf{First, rent burden is defined by neighborhood median rent value
over ~neighborhood median income. The housing stability could be tracked
by the ratio because the higher ratio means renters are suffered by
paying the rent. The figure 8 shows that how the rent burden changes}
\par\null
\textbf{Fig.8 Rent Burden and Income Integration level, 1990-2000-2010;
observations + fitted linear regression model line for all years}
\textbf{Rent Burden over Income Integration level (entropy index)}
\textbf{The linear regression models did not seem to describe / predict
the data sufficiently, and in accordance their R-squared are low. The
differences between the years are interesting: First, neighborhoods
become more integrated by income, though the trend is polarized as the
rich becoming more segregated. Secondly, the correlation, although not
significant, changed its direction; from a negative correlation to a
positive correlation, which means that in 1990 the more integrated a
neighborhood was the lower was its rent burden; though in 2010, the more
integrated the higher the rent burden is.}
\textbf{That change depicts the gentrification process and the
consistently deepening economic gap. With the majority of neighborhoods
in the higher side of the integration level, more neighborhoods cross
the line of 30\% rent burden.}
\textbf{From there we need to look up income integration and income
level. How NYC is diversified by income}
\textbf{Fig.9-11 Income VS Income Integration level with linear
regression, 1990, 2000, and 2010}
\par\null
\textbf{In order to measure the magnitude and direction of the
correlations between the examined variables we run simple linear
regression models. The linear model's dependent variable is median
income of each year in PUMA level (1990, 2000, and 2010) and independent
variable is income entropy index (income integration level). This result
corresponds to Figure 7, 8 and 9. Dash lines in the figures indicate
median income level of NYC and it reveals neighborhood which is located
nearby the median income line has higher income integration level. One
interesting trend is revealed that neighborhoods that has higher than
median NYC income didn't change a lot in integration level over two
decades. By contrast, neighborhoods that has lower than median NYC
income have been clustered to higher integration level.}
\textbf{Observing the change of Median Income over Income integration
through the 20 years we understand that the trend in NYC, unlike the
deepening of economic segregation in the country (Fry \& Taylor, 2012;
Pendall, \& Carruthers, 2010), is towards more and more integrated by
income neighborhoods. Though we do not have the consequences of the
situation in the US, here we see the neighborhoods' polarization to
highly integrated and highly segregated, and the economic gap that
enlarged along with this process.}
\par\null
\textbf{Rent Growth over Income Integration}
\textbf{The last part of the regression models' analysis was to assess
the correlation between Rent growth and Income Integration level. The
models demonstrated low correlations, which made us define the richest
PUMAs as outliers that were about to be excluded from the analysis.}
\textbf{The outliers were defined by exceeding two standard deviation
from the mean of the Median Family Income. The reason for excluding only
one-tail outliers was that these neighborhoods are too rich to be
considered as taking part in the city's urban renewal processes. Figures
11 and 12 shows the change in median rent per PUMA, between 1990 and
2000 and between 2000 and 2010. The level of income is revealed as four
groups varied by color: red for the poorest PUMAs, light-red for poor
PUMAs, light-blue for relatively rich and blue for the richest PUMAs.}
\par\null
\textbf{Fig.11-12 Rent Growth Income Integration level excluding
outliers, 1990-2000-2010}
\textbf{The linear models, although excluding outliers of very rich
PUMAs, were insufficient with explaining the data of rent growth over
income integration level, with both R-squared being very low. The fitted
lines do express a change in the correlation direction between the
years: if between 1990 and 2000 the more integrated the PUMA was the
more significantly increased was its median rent, it seems that at the
later 10 years period the correlation, although weak, was opposite.
Overall, those are inadvisable models for this analysis.}
\par\null
\textbf{Results}
\textbf{The information emerged from the data analysis was clear in the
sense of revealing the urban renewal processes occurring in NYC in the
past few decades, including gentrification, increasing rent burden, and
the risk of displacement. New York City is becoming more and more Income
Integrated, and more and more polarized in its neighborhoods' income
integration level, losing its middle class neighborhoods to integrated
neighborhoods on one hand and poor and rich neighborhoods on the other
hand. The meaning of this trend, though, remained ambiguous. One side of
it is more opportunities to social and economic mobility, while the flip
side is the increasing rent burden that risks the poorer households with
displacement.}
\textbf{The data showed a good chance of clustering, meaning the richer
neighborhoods are going through a different kind of income integration
process from the less affluent neighborhoods, and that each group should
get a individual analysis to better understand the trend and its
consequences.}
\par\null
-\textgreater{} housing stability.
-\textgreater{} rent burden -\textgreater{} in our definition
-\textgreater{} median rent price / median income at PUMA level
-\textgreater{} change of rent burden from 1990 to 2010 shows there is a
pattern with certain neighborhoods.
-\textgreater{} we wanna tracking the rent burden because it's related
to housing stability.
-\textgreater{} is it spatial correlated to the neighborhood?
-\textgreater{} Global Moran'S I - -\textgreater{}~ Yes!
What is characteristic of the neighborhood?
income segregation?
Look up High High and High Low in NYC~
Correaltion~rent burdern and income integration level of HH and HL
It doesn't tell anything~to us but we believe we can find other
charateristic~on the HH and HL area on further research.
\par\null\par\null\par\null
\textbf{References}
\textbf{Freeman, L., \& Braconi, F. (2004). Gentrification and
Displacement New York City in the 1990s.}
\textbf{Journal of the American Planning Association,70(1), 3952.}
\textbf{Fry, R., \& Taylor, P. (2012). The rise of residential
segregation by income. Pew Research Center.}
\textbf{Glass, R. (1964). London : aspects of change . London:
MacGibbon.}
\textbf{Kontokosta, C. E. (2014). Mixedincome housing and neighborhood
integration: Evidence from}
\textbf{Inclusionary Zoning programs. Journal Of Urban Affairs, 36(4),
716741.}
\selectlanguage{english}
\FloatBarrier
\end{document}