adela hernandez, et al. v winstar properties, et al. - …... et al. v winstar properties, et al....

Doherty Report – 1

Report of Dr. Joseph W. Doherty

Adela Hernandez, et al. v

Winstar Properties, et al.

Case No. 2-16-cv-04697-ODW-KS

Qualifications

1. My name is Joseph W. Doherty. I earned a Ph.D. in Political Science from the

University of California, Los Angeles, in 2006. My areas of concentration in the

doctoral program included research methodology. My graduate coursework included

two years (seven classes) of statistics, data analysis, and policy analysis. I am the

President of JDVR Associates, Inc., a private statistical consulting firm. From 2006 to

2016 I was the Director of the Empirical Research Group at the UCLA School of Law,

where I designed and executed empirical research with and for the law faculty. I was

also the co-Director of the UCLA-RAND Center for Law and Public Policy, which

merges econometrics with legal scholarship to produce legal research that is relevant to

policymakers. A copy of my CV is attached as Appendix A.

2. I have worked with hundreds of large datasets, including data from public opinion

surveys, employment and housing discrimination complaints from the California

Department of Fair Employment and Housing, campaign finance disclosure from the

Federal Election Commission and the California Secretary of State, judicial decision

making data from the Administrative Office of the Courts, crime and police activity data

(including use of force) from the Seattle Police Department, and U.S. Census data (at the

tract, block, and individual level). I have also designed data collection instruments for

surveys, court files, experiments, observations, and interviews.

Case 2:16-cv-04697-ODW-KS Document 54-1 Filed 06/26/17 of 57 Page ID #:483


3. I was an adjunct professor at UCLA School of Law where I taught statistics, research

design, and composition to law students. I have co-authored numerous papers and

articles on a broad range of topics, including bankruptcy, bargaining strategy, the living

wage, employment discrimination, international criminal courts, and civil justice. In

each of those manuscripts I was primarily responsible for the data analysis and

interpretation. Many of the articles and reports have been published in peer- reviewed

journals, and I have been the peer-reviewer of articles involving data analysis in

refereed journals. I have testified before the California Legislature about my research

on employment discrimination enforcement, and my co-authors Jody Freeman and Lynn

LoPucki have testified separately before Congress about research on which I

collaborated. I am currently serving as a member of the Monitoring Team of the Seattle

Police Department Consent Decree.

Compensation

4. Los Angeles Center for Community Law and Action (LACCLA), counsel for plaintiffs,

has retained me as an expert in this action. My hourly rate is $250 per hour for research

and reporting, and $350 per hour for testimony. My compensation is not dependent on

the outcome of this case.

Assignment 5. Counsel has asked my opinion about the rental policies of the defendants Winstar

Properties, Inc., and Manhattan Manor, LLC, as they relate to disparate impact

provisions of the Federal Fair Housing Act. Specifically, counsel has asked my opinion

on the following two questions:

A. Do the rental and rent increase policies of the defendants result in a disparate



impact on a group of persons because of race or national origin?

B. If the rental and rent increase policies of the defendants have a disparate impact,

what is the likely impact of those policies on the group that is affected by the

policy?

6. In forming my opinions I considered the following facts and data.

A. American Community Survey 2015 5-year and 1-year Public Use Microdata

Sample Files (PUMS), available for public download at Census.gov. These files

are a demographically granular snapshot of Los Angeles County in the year

preceding the purchase of the property in January 2016. Information in these files

are reported at the level of the individual household and person, with a geographic

coverage of the Census Public Use Microdata Area (PUMA).1 The address of the

Manhattan Manor, LLC and Winstar Properties, Inc property at issue in this case

is 4330 City Terrace Drive, Los Angeles, California (the “Property”). This

address is located near the northern border of PUMA 3743.2 There are seven

PUMAs that share a border with PUMA 3743: 3735, 3736, 3737, 3740, 3741,

3742, and 3744.3 My opinion is based upon analyses conducted on data from

PUMA 3743 and on a larger dataset that includes all of the surrounding PUMAs.

B. Information contained in the First Amended Complaint, filed 07/07/2016.

C. Information provided to me by counsel LACCLA about tenants of the Property at 1 “Public Use Microdata Areas (PUMAs) are geographic areas for which the Census Bureau provides selected extracts of raw data from a small sample of census records that are screened to protect confidentiality. These extracts are referred to as public use microdata sample (PUMS) files. “For the 2010 Census, each state, the District of Columbia, Puerto Rico, and some Island Area participants delineated PUMAs for use in presenting PUMS data based on a 5 percent sample of decennial census or American Community Survey data. These areas are required to contain at least 100,000 people.” https://www.census.gov/geo/reference/gtc/gtc_pumas.html, last viewed on 6/13/2017. 2 This address was geocoded using the online TIGERWeb application published by the Census. https://tigerweb.geo.census.gov/TIGERweb2010/ 3 A map copied from TigerWeb identifying the relevant PUMAs is attached as Appendix C.



the time it was purchased in January 2016.

Analysis

7. I have a four-part strategy for addressing the questions posed to me by counsel, and for

answering at least one possible rebuttal argument of the defendants.

A. Describe the housing market in the area of the subject property. This descriptive

section puts the findings in context, and is a guide for the simulation described

below.

B. Answer the question: Do the policies of the defendant result in a disparate impact?

For the purpose of answering this question I limit my analysis to the type of

housing unit occupied by the plaintiffs -- a two-bedroom apartment.4 Using

Census data I identified those units occupied by Hispanic immigrants,5 and

compared the rents paid for two-bedroom apartments by that group to rents paid

by others. I also disaggregated the categories of race and national origin to

identify those units occupied by other groups (including Hispanic non-immigrants)

to test whether disparate impact is specifically applied to immigrants or whether it

applies to Hispanics as a group. I then applied tests to determine whether the

comparisons result in differences that are statistically significant by conventional

standards.

C. If the policies have a likely disparate impact, what would be the predicted effect of

those policies if they were adopted broadly? I limited my analysis again to the

4 See First Amended Complaint, Page 6, Paragraph 24. 5 This report uses the terms “Hispanic” and “Latino” interchangeably, in conformance with the usage of the U.S. Census (https://www.census.gov/content/dam/Census/library/publications/2011/dec/c2010br‐02.pdf). See also, “Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity,” issued by the OMB, Federal Register Vol. 62, No. 210, 97‐28653 (October 30, 1997) https://www.gpo.gov/fdsys/pkg/FR‐1997‐10‐30/pdf/97‐28653.pdf. Nearly all of the Hispanic/Latino persons in the dataset are from Mexico (86%), Central America (10%), South America (1%), and the Caribbean (1%).



same type of unit (two-bedroom apartment), and projected onto the broader

population the effect of raising rents to levels proposed by defendants ($2,000 per

month for a two-bedroom unit6). This simulation assumes that the ratio of rent-to-

income is a significant limit on housing affordability, and uses Census data to

project the racial and national origin profile of the area if a higher income is

required.

D. One rebuttal argument that the defendants could make is that the proposed rent

increase is an adjustment to bring these units into line with the market rate for

rentals in this area. I analyzed Census data to test whether $2,000 per month is a

typical rent for a two-bedroom apartment in the area, and if not typical,

approximately how atypical that rent would be. This analysis was conducted for

all two-bedroom units, and for two-bedroom units that have been occupied for

four or fewer years. A finding that monthly rent of $2,000 for a two-bedroom

apartment in the area is atypical would weaken a defense of business necessity.

8. These analyses are conducted in two geographic areas, one nested inside the other. The

smaller area is designated PUMA 3743, which contains the property in question. The

larger area includes PUMA 3743, as well as the surrounding PUMAs that are

contiguous with PUMA 3743. I selected two areas for this analysis to test the

sensitivity of the findings across neighborhoods, and because the likely market for these

units is broader than a single PUMA.

9. I conducted my analysis using the statistical software Stata. Stata is a powerful data

analysis program that is widely used in academic research. It is appropriate for the

analysis done in this report. The Stata log file associated with this report is attached as 6 See First Amended Complaint, Page 7, Paragraph 26.



Appendix B.

Findings

10. Approximately two-thirds (65%) of the housing units in the area of Manhattan Manor

(PUMA 3743) are rentals.7 More than one-half of these (62%) are one-family homes,

and the rest of the rental units (38%) are located in apartment buildings of two or more

units. Overall, one-quarter (25%) of all housing units in the area of the Property are

rental apartments and the subset of two-bedroom apartments comprise approximately

ten percent of all housing units in this area. The median rent for a two-bedroom

apartment in this area is $1,000, and ninety-five percent of all two-bedroom apartments

in this area rented for $1,300 or less. The rents charged to the tenants prior to purchase

of this building by Manhattan Manor, LLC, $1,250,8 were similar to rents for nearly all

two-bedroom units in this area. Winstar’s increase to $2,000 was more expensive than

ninety-nine percent of two-bedroom units in the area.

11. In the larger geographic area of contiguous PUMAs surrounding the Property these

proportions are similar. As above a large proportion of the housing units are rentals

(62%), but a larger proportion of them are located in apartment buildings (58%,

compared to 38% above). In this larger area, over one-third (36%) of all housing units

are rental apartment units (compared to 25% above). Two-bedroom apartments

comprise approximately thirteen percent of all housing units in this area. The median

rent for a two-bedroom apartment in this area is $1,100, and ninety percent of all two-

bedroom apartments in this area rented for $1,800 or less.

7 This figure excludes group quarters. 8 See First Amended Complaint, , Paragraphs 21‐22.

Case 2:16-cv-04697-ODW-KS Document 54-1 Filed 06/26/17 Page 6 of 57 Page ID #:488


12. Hispanic immigrants9 comprise approximately forty percent (42%) of the population in

the area of the Property (PUMA 3743). This proportion includes all ages and all

positions in the household (parents, children, etc.). This proportion does not include

persons born in the U.S. who live in a household with Hispanic immigrants. A much

higher proportion of households (65%) are headed10 by a Hispanic immigrant. In the

larger geographic area, Hispanic immigrants comprise one-third (30%) of the

population, and head forty-one percent (41%) of households.

13. Hispanic immigrants occupy two-bedroom apartments at roughly the same rate that they

appear in the population. Of the population living in two-bedroom units in PUMA

3743, forty-one percent are Hispanic immigrants. Of those living in the larger

geographic area, thirty percent are Hispanic immigrants.

14. Hispanic immigrants pay the same monthly rent as all others for two bedroom

apartments in the area where the Property is located (PUMA 3743). The median, 25th,

and 75th percentiles are not statistically significantly different. Across the larger

geographic area, however, Hispanic immigrants pay monthly rents that are significantly

less than those paid by others. Households headed by Hispanic immigrants pay a

median rent of $1,000 for a two-bedroom apartment, while those headed by others pay a

median rent of $1,200. At the 90th percentile the difference between the two groups is

$500 per month ($1,300 for Hispanic immigrants, $1,800 for others). This difference is

statistically significant.11 The 95th percentile of rents paid by Hispanic immigrant

9 Immigrants are defined as residents who were not born in the U.S. and were not born abroad to U.S. citizen parents. 10 I am imputing “head of household” to the person who filled out the American Community Survey questionnaire. That person is listed as Person 1 in the dataset. 11 A difference based on a sample is statistically significant if there is a low probability that we would observe it if in fact there is no difference in the population from which it was sampled. The method I used to determine the



households is $1,400, compared to $2,200 for other households. Hispanic immigrants

rely on lower-cost housing than many of those in the other category. Raising the rent on

all two-bedroom units would have a disparate impact on Hispanic immigrants.

15. The differences are illustrated in Figure 1. The solid black line represents the

distribution of rents paid by Hispanic immigrant households, and the dotted line

represents the distribution of rents paid for other households. If there was no difference

statistical significance of these percentiles was to estimate the 95% Confidence Intervals (95% CI) for each one and compare them across the groups. A 95% CI can be thought of as the margin of error. If a value is outside of the 95% CI, we can say that value is rejected at the .05 level. In this case, both values are estimated, and so we need to calculate and compare the confidence intervals. In this particular instance, at the 90th percentile Hispanic immigrant households paid rents of $1,300 per month [95% CI: $1,251 – $1,349], and Other households paid rents of $1,800 per month [95% CI: $1,702 – $1,897]. More sophisticated analyses could be done to calculate the precise p‐value associated with this difference, but the complete lack of overlap between the two confidence intervals indicates that the probability that Hispanic immigrant households pay the same rents as other households is less than 5%, which is the standard for conventional statistical significance.



between the two groups the lines would trail each other, and there would be no

systematic gaps between them. What I found is that the modal Hispanic immigrant

household is paying somewhat less than $1,000 and the modal other household paying

somewhat more than $1,000. The gap between the lines on the left side of the graph

indicates that more Hispanic immigrant households live in lower-rent apartments than

do other households. Additionally, the gap between the lines in the middle of the graph

indicates that fewer Hispanic immigrant households pay higher rents (starting at about

$1,400) than other households. A higher proportion of other households pay rents of

$1,400 or more than Hispanic immigrant households. As noted above, this difference is

statistically significant by conventional standardsand is evidence of disparate impact

based on race or national origin.

16. It is also apparent from Figure 1 that very few households living in two-bedroom

apartments pay $2,000 per month in rent. The solid flat line at the bottom of the graph

above the $2,000 tick mark illustrates that virtually no Hispanic immigrant households

pay that much or more in rent. The area between the solid line and the dotted line

above it illustrates the difference between Hispanic immigrant households and the other

households. Nearly all of the households that pay $2,000 or more in rent are non-

immigrant households.

17. To further test whether these differences constitute disparate impact based upon race or

national origin I disaggregated the “other” category into multiple race and national

origin groups. In the broad geographic area that includes the rental market for this

property, Hispanic immigrants and Hispanic non-immigrants pay similar amounts in

rent for two-bedroom apartments (Table 1). The median rents are $1,000 and $1,100,



respectively, and the 75th, 90th, and 95th percentiles are similarly close and statistically

indistinguishable. Non-Hispanic White and Mixed Race households spend a great deal

more on rent than Hispanics. Whites are paying higher rents than Hispanics at a

statistically significant level at all percentiles. No more than five percent of Hispanic

immigrant households spend $1,400 or more for monthly rent, compared to fifty percent

of Non-Hispanic White households. The comparison of Hispanic Non-Immigrant

households to Non-Hispanic White households is similar, and the differences between

the two groups are statistically significant.

Table 1. Distribution of Monthly Rent for a 2 Bedroom Apartment, by Race/National Origin

Rent Percentile for Each Group

Median 75th 90th 95th

Hispanic Immigrant

$1000 $1200 $1300 $1400

Hispanic Non‐Immigrant

$1100 $1300 $1500 $1600

Non‐Hisp White $1400 $1900 $2300 $2600

Black $1100 $1400 $1500 $1700

Asian $1200 $1400 $1700 $2300

Other $980 $1200 $1300 $1700

Mixed Race $1400 $2100 $2600 $2600

Source: American Community Survey 5 Year PUMS File, 2011‐2015.

18. This point is further illustrated in Figure 2. Hispanic (immigrant and non-immigrant)

households pay monthly rents for two-bedroom apartments that are significantly lower

than those paid by Non-Hispanic White households. As indicated in Table 1, nearly all

(95%) of Hispanic households rent units for $1,600 per month or less. There is almost

no market in the Hispanic community for a two-bedroom apartment that costs $2,000

per month. There is a market for such units in the Non-Hispanic White community.



19. My conclusions follow directly from the data. Increasing the rent on two-bedroom

apartments to $2,000 or more in the subject area will have a disparate impact on

Hispanic households, particularly Hispanic immigrant households. The implications of

this are uncomplicated. Among those renting two-bedroom households in the larger

geographic area, Hispanic and Non-Hispanic White households pay similar proportions

of their income for rent (Figure 3), but Non-Hispanic White households have

significantly higher household incomes than Hispanic households (Figure 4). Assuming

that incomes are relatively fixed and that the percentage of a household’s income that

can be spent on housing is also relatively fixed, the effect of charging rents that are

significantly higher is likely to be the displacement of Hispanic households in the area.



20. I simulated the magnitude of the displacement of Hispanic households by estimating the

capacity of each household to absorb a rent increase from $1,250 to $2,000. I

concluded above that a change in the price of rental apartments will have a disparate

impact on Hispanic households. The purpose of this simulation is to estimate the

composition of the market for whom these apartments are affordable at both prices. I

calculated that the median household in a two-bedroom apartment spent almost one-

third of its income on rent (the rent-to-income ratio is 0.31), and for the purposes of this

simulation I assumed this is true for all households.

Table 2. Simulated Change in Affordability for a 2 Bedroom Apartment, by Race/National Origin

Affordable Rent Simulated Change $1,250 $2,000

Hispanic Immigrant 32% 20% ‐37%

Hispanic Non‐Immigrant 26% 26% 0%

Non‐Hisp White 14% 21% +50%

Black 2% 2% 0%

Asian 24% 27% 13%

Other 1% 1% 0%

Mixed Race 2% 3% 50%


21. The market for two-bedroom apartments at a rent price of $1,250 per month is 58%

Hispanic, with the majority of that group (32%) being Hispanic Immigrants and 26%

Hispanic Non-Immigrant (Table 2). It is also 14% Non-Hispanic White and 24% Asian.

The market for two-bedroom apartments at a rent price of $2,000 is significantly less

Hispanic (46%), and significantly more Non-Hispanic White (21%). Within the

Hispanic community the simulation suggests that the impact would be borne entirely by

immigrant households, who comprised 32% of the market at $1,250/month and only

20% of the market at $2,000/month. The likely effect of a rent increase is to displace



Hispanic households and increase the proportion of households that are Non-Hispanic

White.

22. Defendants may offer rebuttal testimony that the disparate impact of their rental

increase is justified by business necessity. They may argue that it reflects local market

prices for two-bedroom units in the region of the Property, and that the rents charged to

tenants before January 2016 were well-below market rates. As discussed above, the

rents charged to the tenants prior to purchase of this building by Manhattan Manor, LLC

were similar to rents for nearly all two-bedroom units in the larger geographic area. The

median rent for a two-bedroom apartment in this area was $1,100, and ninety percent of

all two-bedroom apartments in this area rented for $1,800 or less. Defendants’ rent of

$2,000 was more expensive than ninety-five percent of two-bedroom units in the area.

23. Defendants may also argue that the rents it charged were typical of rents paid for

recently occupied two-bedroom apartments, as opposed to units in which the tenants

had been resident for many years. To test this argument I relied on a more recent

dataset, the 2015 PUMS 1 Year File, which includes only survey responses from the

year before Manhattan Manor, LLC purchased the property in question in January 2016.

For the purpose of this analysis I have defined “recently occupied” as two-bedroom

apartments in which the tenants have been in residence for two years or fewer, who took

up residency between 2013 and 2015.

24. The median rent for recently occupied two-bedroom apartments in the larger geographic

area was $1,300.12 In other words, between 2013 and 2015, more than one-half of the

recently-occupied two-bedroom apartments rented for $1,300 or less. More than 75%

12 [95% CI: $1,202 ‐ $1,398]



of the same type of apartments rented for $1,60013 or less, and only 15% of the recently

occupied two-bedroom apartments in the larger geographic region had rents of $2,00014

or more. From these findings I conclude that Defendants priced the units at the

Property well above the market rate for a typical two-bedroom apartment in the area.

Conclusion

25. The area of Los Angeles County within which the Property is sited has a large Hispanic

community. A large percentage of this community is comprised of immigrants for

whom housing in this area is affordable. A change in the rent of the magnitude

proposed by Defendants would likely have a disparate impact on Hispanic households

generally, and on Hispanic immigrant households specifically. This impact would lead

to a decrease in the percentage of Hispanic households in the market that can afford

two-bedroom apartments in the area, and increase the percentage of Non-Hispanic

White households in that market. And this increase would not reflect typical market

rates for recently occupied two-bedroom apartments in the area. Two-bedroom

apartments in the area renting for $2,000 apiece would be outliers in the market.

Dated: June 26, 2017

Respectfully submitted,

By:

Joseph W. Doherty, Ph.D. 3641 S Bentley Avenue Los Angeles, CA 90034 310-500-5696

13 [95% CI: $1,355 – $1,845] 14 [95% CI: $1,706 ‐ $2,294]


Appendix A

CV of Joseph W Doherty, Ph.D.


JOSEPH W. DOHERTY, PH.D. 3641 S. Bentley Avenue, Los Angeles, California 90034 Telephone: 310-500-5696 E-mail: [email protected] Education

University of California, Los Angeles Ph.D. Political Science 2006 American Politics, Methodology Dissertation: The Candidate-Consultant Network in California Legislative Campaigns: A Social Network Analysis of Informal Party Organizations. Dissertation co-chairs: Barbara Sinclair, John Petrocik.

University of California, Los Angeles B.A. Political Science 1992 Summa Cum Laude, College Honors, Departmental High Honors

Positions 2016 to present

President

JDVR Associates, Inc. Los Angeles, California

2009 to 2016

Assistant Adjunct Professor UCLA School of Law

2008 to 2016 Co-Director UCLA-RAND Center for Law and Public Policy UCLA School of Law

2006 to 2016 1999-2006

Director Associate Director

Empirical Research Group UCLA School of Law

2002-2006 Project Director The Campaign Disclosure Project A collaboration of the UCLA School of Law, the Center for Governmental Studies and the California Voter Foundation to improve campaign disclosure in the states Funded by The Pew Charitable Trusts

1996-2002 Consultant/Analyst Fairbank, Maslin, Maullin & Associates Public Opinion and Policy Research Santa Monica, California


JOSEPH W. DOHERTY, PH.D

Page 2 of 8

1990-1993 Clerk/Investigator City of Santa Monica, Office of the City Attorney, Criminal Division Santa Monica, California

1981-1990 Joe Doherty Photography Los Angeles, California

Publications – Books & Monographs

2012 Confidentiality, Transparency, and the U.S. Civil Justice System. Edited by Joseph W. Doherty, Robert T. Reville and Laura Zakaras, Oxford University Press (2012).

2011 Controlling Professional Fees in Corporate Bankruptcies: Data, Analysis, and Evaluation. With Lynn M. LoPucki. Oxford University Press (2011).

2010 California Employment Discrimination Law and Its Enforcement: The Fair Employment and Housing Act at 50. With Gary Blasi. Los Angeles, CA: UCLA-RAND Center for Law and Public Policy (2010).

2003-2007 Grading State Disclosure. The Campaign Disclosure Project (2003, 2004, 2005, 2007).

2002 The Economic and Distributional Consequences of the Santa Monica Minimum Wage Ordinance. With Richard Sander and E. Douglass Williams. Employment Policy Institute, Washington DC (2002).

Publications – Journal Articles & Book chapters

2016 2015

Punishment and Policy in International Criminal Sentencing: An Empirical Study. With Richard H. Steinberg. 110 American Journal of International Law 49 (2016)

Bankruptcy Survival.

With Lynn M. LoPucki. 62 UCLA Law Review 970 (2015)

2012 One Client, Different Races: Estimating Disparity in Chapter Choice Using Matched Pairs of Debtors. 20 American Bankruptcy Institute Law Review 651 (2012)



Page 3 of 8

2012 Expectations, Outcomes and Fairness: Lessons from the Civil Justice Reform Act Evaluation. With Stephen J. Carroll. In Confidentiality, Transparency, and the U.S. Civil Justice System. Edited by Joseph W. Doherty, Robert T. Reville and Laura Zakaras, Oxford University Press (2012).

2011 Managerial Judging Goes International, but Its Promise Remains Unfulfilled: An Empirical Assessment of the ICTY Reforms. With Maximo Langer. 36 Yale Journal of International Law 241 (2011).

Routine Illegality Redux. With Lynn M. LoPucki. 85 American Bankruptcy Law Journal 35 (2011).

2009 Routine Illegality in Big-Case Bankruptcy Court Fee Practices. With Lynn M. LoPucki. American Bankruptcy Law Journal 83.3: 425-482 (2009).

Who Wins in Settlement Negotiations? With Russell Korobkin. American Law and Economics Review 11: 162-208 (2009).

2008 Professional Overcharging in Large Bankruptcy Reorganization Cases. With Lynn M. LoPucki. Journal of Empirical Legal Studies 5.4: 983-1017 (2008).

Rise of the Financial Advisors: An Empirical Study of the Division of Professional Fees in Large Bankruptcies. With Lynn M. LoPucki. American Bankruptcy Law Journal 82.1: 141-174 (2008).

Bankruptcy Verite (response to White). With Lynn M. LoPucki. Michigan Law Review 106.4: 721-743 (2008).

2007 Bankruptcy Fire Sales With Lynn M. LoPucki. Michigan Law Review 106.1: 1-59 (2007)

2006 Delaware Bankruptcy: Failure in the Ascendancy With Lynn M. LoPucki. Chicago Law Review 73.4: 1387-1419 (2006).

2005 Campaign Disclosure Project Symposium Part One: Introduction. With Daniel Hays Lowenstein. Election Law Journal 4.4: 279-281 (2005).

2004 The Determinants of Professional Fees in Large Bankruptcy Reorganization Cases. With Lynn M. LoPucki. Journal of Empirical Legal Studies 1.1: 111-142 (2004).

2002 Why are Delaware and New York Bankruptcy Reorganizations Failing? With Lynn M. LoPucki. Vanderbilt Law Review 55.6: 1933-1985 (2002).



Page 4 of 8

2001 Outside Money in the California 2000 Presidential Primary. With Diana Dwyre, Bruce Cain, and Ray La Raja. PS: Political Science and Politics (June, 2001).

2000 Outside Money in the California 2000 Presidential Primary. With Diana Dwyre, Bruce Cain, Ray La Raja and Samuel Kernell. In Getting Inside the Outside Campaign: Issue Advocacy in the 2000 Presidential Primaries. David B. Magleby editor. Center for the Study of Elections and Democracy, Brigham Young University. Report of a grant funded by the Pew Charitable Trust (2000).

1996 The Road to Divided Government: Paved Without Intention. With John R. Petrocik. In Divided Government: Change, Uncertainty, and the Constitutional Order. Edited by Peter Galderisi, Roberta Q. Herzberg and Peter McNamara. Rowman and Littlefield (1996).

Conferences organized

2015

2012-2016

Discount Justice: State Court Budgeting in an Era of Fiscal Austerity. UCLA-RAND Center for Law & Public Policy.

Western Empirical Legal Studies. A conference organized for law and graduate students conducting original empirical legal research. UCLA School of Law

2013 Bankruptcy Success Modeling. UCLA School of Law

2009 Third Party Litigation Funding. UCLA-RAND Center for Law & Public Policy.

2007 Transparency In the Civil Justice System. UCLA-RAND Center for Law & Public Policy

Research, Presentations & Conference Papers

2014 Bankruptcy Survival. With Lynn LoPucki. Prepared for presentation at the 9th Annual Conference on Empirical Legal Studies. Berkeley Law, California (7-8 November 2014).

2014 The Organized Hypocrisy of International Criminal Sentencing: An Empirical Study of Doctrine versus Practice.

With Richard H. Steinberg. Prepared for presentation at the Institute of International Studies, UC Berkeley, California (10 October 2014).



Page 5 of 8

2012 One Client, Different Races: Estimating Disparity Using Matched Pairs of Debtors. Invited paper. Prepared for presentation at the American Bankruptcy Institute Law Review Fall 2012 Symposium, “Bankruptcy and Race: Is There a Relation?” St. John’s School of Law, Queens, NY (2012).

2012 Do Antidiscrimination Regimes Discriminate? With Gary Blasi. Prepared for presentation at the International Law & Society Conference. Waikiki Hilton, Hawaii. (2012).

2010 Do Antidiscrimination Regimes Discriminate? Processing Claims Through Administrative and Legal 'Pyramids' and the Role of the Plaintiffs’ Bar: A California Case Study. With Gary Blasi. Prepared for presentation at the 5th Annual Conference on Empirical Legal Studies, Yale University Law School (2010).

2009 An Empirical Study of ICTY and ICTR Sentencing: Doctrine Versus Practice. With Richard H. Steinberg. Prepared for presentation at the 4th Annual Conference on Empirical Legal Studies, University of Southern California (Gould) Law School (2009).

2009 The Expectations Gap and Perceptions of Fairness in Civil Litigation. With Stephen J. Carroll. Presented at the Annual Conference of the Midwest Political Science Association (2009).

2008 Why Did the Managerial Reforms at the International Criminal Tribunal for the Former Yugoslavia Not Achieve Their Purpose? The Bureaucratization Effect in Procedural and Judicial Reform. With Maximo Langer. Presented at the 3rd Annual Conference on Empirical Legal Studies, Cornell University Law School (2008).

2007 Who Wins in Settlement Negotiations? With Russell Korobkin. Presented at the 2nd Annual Conference on Empirical Legal Studies. New York University School of Law (2007).

2007 The Campaign Disclosure Project: Lessons from the United States. Invited Paper. Presented at The Crinis Conference, Berlin, Germany. Sponsored by Transparency International and The Carter Center. July 12th, 2007.

2006 The Determinants of Professional Fees in Large Bankruptcy Reorganization Cases, Revisited. With Lynn M. LoPucki (2006)

2006 The Candidate-Consultant Network in California Legislative Campaigns: A Social Network Analysis of Informal Party Organizations. Dissertation in Political Science at UCLA (2006). Barbara Sinclair and John R. Petrocik, co-chairs.



Page 6 of 8

2006 Judicial Review of Agency Rulemaking: An Empirical Analysis of Circuit Court Decisions, 1995-2004. With Jody Freeman. Presented at the Congressional Research Service symposium, “Presidential, Congressional and Judicial Control of Rulemaking.” Washington, DC (September 11, 2006).

2006 Rise of the Financial Advisors: An Empirical Study of the Division of Professional Fees in Large Bankruptcies. With Lynn M. LoPucki. Presented at the CELS 2006 1st Annual Conference on Empirical Legal Studies in Austin, TX (2006).

2006 The Congressional Campaign Network: Candidate-Consultant Linkage in House Races, 1996-2004. Presented at the American Political Science Association Annual Conference in Philadelphia, PA (2006).

2005 Campaign Disclosure Website Usability. Presented at the Council on Governmental Ethics Laws (COGEL) Annual Conference in Boston, MA (2005).

2005 Organized by Competition: Candidate-Consultant Networks in California Legislative Campaigns. Presented at the American Political Science Association Annual Conference in Washington, DC (2005).

2004 Campaign Networks: Using Network Analysis to Study Patterns in Campaign Disclosure Reports. Presented at the Western Political Science Association Annual Conference in Portland, OR (2004).

2003 The Strength of Campaign Ties: Political Networks as Political Parties. Presented at the Northeastern Political Science Association Annual Conference in Philadelphia, PA (2003).

2001 Evaluation of the Van Nuys Legal Self-Help Center, Final Report. Empirical Research Group, UCLA School of Law (2001).

2000 A Computational Model of Housing Segregation. With Richard H. Sander and Darren Schreiber. Presented at the Western Political Science Association Annual Conference in San Jose, CA (2000)

1998 A Broader Conception of Parties-in-Elections. Presented at the American Political Science Association Annual Conference in Boston, MA(1998).



Page 7 of 8

1998 Voter Disloyalty in French Elections: Two-Round Ballots and Spatial Competition. Presented at the Midwest Political Science Association Annual Conference in Chicago, IL (1998).

1996 Amateur Politicians, Landslide Elections, and Changes in the Party System. Presented at the Midwest Political Science Association Annual Conference in Chicago, IL (1996).

1995 If You Can’t Join ‘em, Divide ‘em: The Partisan Basis of the Divided Government Response. Masters thesis on the role of partisanship and political environment in public attitudes toward split party control of government (1995).

1995 The Road to Divided Government is Paved Without Intention. With John R. Petrocik. Presented at the Midwest Political Science Association Annual Conference in Chicago, IL (1995).

Book Reviews

2007 Myopia and the “Insiders’ View” of Presidential Campaigns. Election Law Journal 6.4: 413-416 (2007) Reviewing Electing the President 2004: The Insiders’ View. Kathleen Hall Jamieson, ed. Philadelphia, PA: U Pennsylvania Press (2006).

Courses Taught

Law 279 – Empirical Legal Studies: Theory and Methods

Law 579 – Empirical Legal Studies: Research and Composition

Law 379 – Election Law

Research methodology workshop for UCLA Law Faculty

University Service

May 2009 to June 2016

UCLA Conflict of Interest Review Committee

2012 to 2013 UCLA Research Informatics Strategic Planning Committee (Chair, Business, Law, and Public Affairs Subcommittee,)



Page 8 of 8

Expert Witness Reports & Testimony

2016 Edward Michael York, et al v. City of Fullerton, et al. Case No. 30-2013-0067291-CU-WM-CJC. California Superior Court, Orange County. Deposition October 11, 2016.

2015 Leo Rampersad, et al v. City of Thousand Oaks. Case No. 56-2011-00408151-CU-WM-SIM. California Superior Court, Ventura County. Declaration in Opposition to Defendant’s Motion for Summary Judgment or Summary Adjudication.

2015

Fair Housing Council of Riverside County, Inc., et al. v. City of Riverside. Case No. 5:14-cv-01391-JGB-SP. U.S. District Court, Central District of California.

2014

Application for Closure of Buena Vista Mobile Home Park, Pursuant to Palo Alto Municipal Code Chapter 9.76. Hearing May 12-14, 2014.

2014

California Department of Fair Employment and Housing vs. Morrison, et al., Case No. SCV0032487, California Superior Court, Placer County.

2014 Bhayani, et al. vs. The JAM Limited Partnership, et al., Case No. 110CV167490, California Superior Court, Santa Clara County. Testimony on April 2, 2014. Deposition on February 17, 2014.

2012 Blackington vs Quiogue Family Trust, Case No. 11CV1670 W WVG, U.S. District Court, Southern District of California. Deposition on September 14, 2012.

2012 In re: VITRO, S.A.B. de C.V., Case No. 11-33335-hdh15, U.S. Bankruptcy Court for the Northern District of Texas, Dallas Division. Testimony on June 5, 2012. Deposition on May 19, 2012.

2011 Fair Housing of the Dakotas, Inc. vs. Goldmark Property Management, 778 F. Supp. 2d 1028 (2011).

2010 California Employment Discrimination Law and Its Enforcement: The Fair Employment and Housing Act at 50. Testimony before a Joint Hearing of the California Senate and Assembly Judiciary Committees. February 23, 2010. Sacramento, California.

Memberships

2007 – 2016 2011 – 2013

Society for Empirical Legal Studies Law and Society Association


Appendix B

Stata Log File


name: <unnamed> log: D:\Dropbox\Expert\Winstar\log\WinstarAnalysis.smcl log type: smcl opened on: 26 Jun 2017, 11:09:07

1 . use "data\ss15hca.dta", clear

2 . 3 . merge 1:m serialno using "data\ss15pca.dta" , gen(lamerge)

Result # of obs. not matched 657 from master 657 (lamerge==1) from using 0 (lamerge==2)

matched 50,373 (lamerge==3)

4 . drop if lamerge==1 /* drops unoccupied housing units */(657 observations deleted)

5 . 6 . lab var rntp "Monthly Rent"

7 . 8 . /* Identify Core PUMA */9 . gen p3743 = puma10 == 3743

10 . 11 . /*HOUSING*/12 . /* Label tenure variable */13 . label define ten 1 "Owned with mortgage" 2"Owned free and clear" 3"Rented" 4"Occupied without rent

> "

14 . lab val ten ten

15 . tab ten [fw=wgtp]

TEN Freq. Percent Cum.

Owned with mortgage 235,687 28.67 28.67 Owned free and clear 82,009 9.98 38.65 Rented 494,954 60.21 98.86Occupied without rent 9,394 1.14 100.00

Total 822,044 100.00

16 . 17 . /* Identify rental units */18 . gen rentalunit = ten==3

19 . 20 . /* Create apartment binary variable */


21 . gen apt = bld>=4 & bld<=9

22 . lab var apt "Apartment 2+ units"

23 . 24 . /* Create variable of one-family homes, apts, and other */25 . recode bld (2 3 =1)(4/9=2)(1 10=3), gen(bld3)

(48842 differences between bld and bld3)

26 . lab def bld3 1"One family home" 2"Apt building" 3"Other"

27 . lab val bld3 bld3

28 . 29 . /* Create variable for comparable two bedroom rental units */30 . gen twobedroomrental = bdsp==2 & rentalunit == 1 & apt==1

31 . 32 . /* Create a subpop for two-bedrooms apts in PUMA 3743 */33 . gen puma2bdrm = twobedroomrental * p3743

34 . 35 . /* Amount of monthly rent */36 . gen monthlyrent = rntp

(23,440 missing values generated)

37 . 38 . /* Rent/HH Income ratio */39 . gen rentinc = (rntp * 12)/hincp


40 . lab var rentinc "Rent-to-Income Ratio"

41 . 42 . /* Create a logarithm variable for household income */43 . gen loghincp = log10(hincp)


44 . lab var loghincp "Annual Household Income"

45 . 46 . /* Create variables with assumptions about HHincome and rent-to-income ratio

> and test the percentage of each in race/nat origin categories */47 . /* get the distribution of rentinc and use the 25th, 50th, and 75th quartiles */48 . sum rentinc if twobedroomrental==1, det

Rent-to-Income Ratio

Percentiles Smallest 1% .0664723 .0058252 5% .1083871 .005825210% .137931 .0058252 Obs 5,80325% .201286 .0072 Sum of Wgt. 5,803

50% .3122677 Mean 1.10152 Largest Std. Dev. 19.2344275% .5142857 72090% .8386277 720 Variance 369.962895% 1.248 720 Skewness 36.2309599% 4.615385 720 Kurtosis 1347.287


49 . /* use those figures to calculate monthly rent capability of each household */50 . gen r50p = (hincp * .312) / 12


51 . 52 . /* Create categorical variables for rents of $1250 and $2000 */53 . gen r50b1250 = r50p>= 1250

54 . gen r50b2000 = r50p>= 2000

55 . 56 . 57 . /* Create immigrant variable

> Not (born in U.S. or born abroad of American parents) */58 . gen immig = inlist(cit,2,4,5)

59 . 60 . /* Label Hispanic variable */61 . run do\HispanicLabels.do

62 . 63 . /* Create Hispanic/Latino variable */64 . gen hispanic = hisp >1

65 . 66 . /* Identify Hispanic/Latino Immigrants */67 . gen hispimmig = immig==1 & hispanic==1

68 . 69 . /* Create a 5 category variable for race/national origin */70 . recode rac1p (1=1 "White")(2=2 "Black") (6=3 "Asian")(3 4 5 7 8 =4 "Other") ///

> (9=5 "Mixed Race"), gen(race5)(23611 differences between rac1p and race5)

71 . replace race5 = 6 if hispanic==1(34,964 real changes made)

72 . replace race5 = 7 if hispimmig==1(14,810 real changes made)

73 . lab def race5 6"Hispanic non-immigrant" 7"Hispanic immigrant", modify

74 . 75 . /* Create a hispanic/white variable */76 . gen hispwhite = hispanic

77 . recode hispwhite (0=.) if race5!=1(hispwhite: 11044 changes made)

78 . lab def hispwhite 0"Non-Hispanic White" 1"Hispanic"

79 . lab val hispwhite hispwhite

80 . lab var hispwhite "Hispanic v White"


81 . 82 . 83 . /*HOUSING UNIT ANALYSIS*/84 . /* housing weights */85 . svyset [pw=wgtp], sdr(wgtp1 - wgtp80) vce(sdr)

pweight: wgtp VCE: sdr MSE: off sdrweight: wgtp1 wgtp2 wgtp3 wgtp4 wgtp5 wgtp6 wgtp7 wgtp8 wgtp9 wgtp10 wgtp11 wgtp12 wgtp13 wgtp14 wgtp15 wgtp16 wgtp17 wgtp18 wgtp19 wgtp20 wgtp21 wgtp22 wgtp23 wgtp24 wgtp25 wgtp26 wgtp27 wgtp28 wgtp29 wgtp30 wgtp31 wgtp32 wgtp33 wgtp34 wgtp35 wgtp36 wgtp37 wgtp38 wgtp39 wgtp40 wgtp41 wgtp42 wgtp43 wgtp44 wgtp45 wgtp46 wgtp47 wgtp48 wgtp49 wgtp50 wgtp51 wgtp52 wgtp53 wgtp54 wgtp55 wgtp56 wgtp57 wgtp58 wgtp59 wgtp60 wgtp61 wgtp62 wgtp63 wgtp64 wgtp65 wgtp66 wgtp67 wgtp68 wgtp69 wgtp70 wgtp71 wgtp72 wgtp73 wgtp74 wgtp75 wgtp76 wgtp77 wgtp78 wgtp79 wgtp80 Single unit: missing Strata 1: <one> SU 1: <observations> FPC 1: <zero>

86 . 87 . /* Describe housing */88 . svy, subpop(p3743): tab ten if sp==1

(running tabulate on estimation sample)

Number of obs = 17,217 Population size = 266,050 Subpop. no. obs = 1,393 Subpop. size = 24,392 Replications = 80

TEN proportion

Owned wi .2267 Owned fr .11 Rented .6542 Occupied .0091 Total 1

Key: proportion = cell proportion

89 . svy, subpop(p3743): tab bld3 ten if sp==1, col(running tabulate on estimation sample)


RECODE of TEN bld (BLD) Owned wi Owned fr Rented Occupied Total

One fami .956 .9072 .6177 .8829 .7286 Apt buil .044 .0801 .3819 .036 .2689 Other 0 .0127 4.4e-04 .0811 .0024 Total 1 1 1 1 1

Key: column proportion


Pearson: Uncorrected chi2( 6) = 2617.5428

90 . svy, subpop(p3743): tab bld3 ten if sp==1(running tabulate on estimation sample)



One fami .2167 .0998 .4041 .008 .7286 Apt buil .01 .0088 .2498 3.3e-04 .2689 Other 0 .0014 2.9e-04 7.4e-04 .0024 Total .2267 .11 .6542 .0091 1

Key: cell proportion


91 . svy, subpop(p3743): tab twobedroomrental if sp==1(running tabulate on estimation sample)


twobedroo mrental proportion

0 .9072 1 .0928 Total 1


92 . 93 . svy: tab ten if sp==1


Number of obs = 15,715 Population size = 266,050 Replications = 80

TEN proportion

Owned wi .253 Owned fr .1162 Rented .6155 Occupied .0152 Total 1



94 . svy: tab bld3 ten if sp==1, col(running tabulate on estimation sample)



One fami .9058 .8851 .4118 .7596 .5971 Apt buil .0901 .0888 .5797 .2301 .3934 Other .0041 .0261 .0085 .0104 .0095 Total 1 1 1 1 1



95 . svy: tab bld3 ten if sp==1(running tabulate on estimation sample)



One fami .2292 .1029 .2535 .0116 .5971 Apt buil .0228 .0103 .3568 .0035 .3934 Other .001 .003 .0052 1.6e-04 .0095 Total .253 .1162 .6155 .0152 1

Key: cell proportion


96 . svy: tab twobedroomrental if sp==1(running tabulate on estimation sample)



0 .8723 1 .1277 Total 1



97 . 98 . epctile monthlyrent if sp==1, p(10 25 50 75 95 99) svy over(twobedroomrental p3743)

Warning: if and in conditions should be implemented via svy, subpop option(running mean on estimation sample)

SDR replications ( 80) 1 2 3 4 5

.................................................. 50

..............................

Survey: Mean estimation Number of obs = 8,818 Population size = 163,754 Replications = 80

Over: twobedroomrental p3743 _subpop_1: 0 0 _subpop_2: 0 1 _subpop_3: 1 0 _subpop_4: 1 1

SDR Over Mean Std. Err. [95% Conf. Interval]

__000006 _subpop_1 -.0008408 .0042866 -.0092424 .0075607 _subpop_2 -.0044907 .0115883 -.0272033 .0182219 _subpop_3 -.0057377 .0072516 -.0199506 .0084752 _subpop_4 -.0244366 .0223744 -.0682897 .0194165

__000007 _subpop_1 -.0272322 .005224 -.037471 -.0169933 _subpop_2 -.000858 .018358 -.036839 .035123 _subpop_3 -.0352144 .0110213 -.0568157 -.013613 _subpop_4 -.0144719 .0440764 -.10086 .0719161

__000008 _subpop_1 -.0011027 .0080191 -.0168198 .0146144 _subpop_2 -.0259584 .023202 -.0714334 .0195167 _subpop_3 -.0760088 .0134244 -.1023202 -.0496975 _subpop_4 -.0050817 .0470522 -.0973024 .0871389

__000009 _subpop_1 -.0190909 .0061971 -.031237 -.0069449 _subpop_2 -.0504746 .0198252 -.0893313 -.0116179 _subpop_3 -.0355927 .0111082 -.0573644 -.013821 _subpop_4 -.0098321 .0455206 -.0990509 .0793868

__00000A _subpop_1 -.0049468 .0034432 -.0116953 .0018018 _subpop_2 -.0126506 .0104799 -.0331909 .0078897 _subpop_3 -.0034048 .005704 -.0145844 .0077748 _subpop_4 -.0445647 .0235801 -.0907809 .0016514

__00000B _subpop_1 -.0000451 .0014259 -.0028399 .0027496 _subpop_2 -.0047499 .0054452 -.0154223 .0059225 _subpop_3 -.0006557 .0028359 -.0062141 .0049026 _subpop_4 -.0173973 .0173288 -.051361 .0165665


Percentile estimation

SDR monthlyrent Coef. Std. Err. z P>|z| [95% Conf. Interval]

p10 0_0 450 15 30.00 0.000 420.6005 479.3995 0_1 550 25 22.00 0.000 501.0009 598.9991 1_0 700 20 35.00 0.000 660.8007 739.1993 1_1 600 50 12.00 0.000 502.0018 697.9982

p25 0_0 700 5 140.00 0.000 690.2002 709.7998 0_1 720 12.5 57.60 0.000 695.5005 744.4995 1_0 900 12.5 72.00 0.000 875.5005 924.4995 1_1 800 47.5 16.84 0.000 706.9017 893.0983

p50 0_0 920 12.5 73.60 0.000 895.5005 944.4995 0_1 900 20 45.00 0.000 860.8007 939.1993 1_0 1100 . . . . . 1_1 1000 25 40.00 0.000 951.0009 1048.999

p75 0_0 1300 . . . . . 0_1 1100 25 44.00 0.000 1051.001 1148.999 1_0 1300 . . . . . 1_1 1200 25 48.00 0.000 1151.001 1248.999

p95 0_0 2000 25 80.00 0.000 1951.001 2048.999 0_1 1500 50 30.00 0.000 1402.002 1597.998 1_0 1900 75 25.33 0.000 1753.003 2046.997 1_1 1300 150 8.67 0.000 1006.005 1593.995

p99 0_0 2700 75 36.00 0.000 2553.003 2846.997 0_1 1800 50 36.00 0.000 1702.002 1897.998 1_0 2800 125 22.40 0.000 2555.005 3044.995 1_1 1500 100 15.00 0.000 1304.004 1695.996

99 . epctile monthlyrent if sp==1, p(10 25 50 75 95 96 97 98 99) svy over(twobedroomrental)Warning: if and in conditions should be implemented via svy, subpop option(running mean on estimation sample)


.................................................. 50

..............................


0: twobedroomrental = 0 1: twobedroomrental = 1



__000006 0 -.0002705 .0041295 -.0083642 .0078233 1 -.0024218 .0072571 -.0166454 .0118018

__000007 0 -.0295116 .0050059 -.039323 -.0197002 1 -.0266236 .0105395 -.0472806 -.0059666

__000008 0 -.0523268 .0079211 -.0678518 -.0368018 1 -.0631051 .0125505 -.0877037 -.0385065

__000009 0 -.0064286 .0055908 -.0173864 .0045292 1 -.0228717 .0102362 -.0429343 -.0028092

__00000A 0 -.0114775 .0031186 -.0175899 -.0053651 1 -.0092944 .0060654 -.0211822 .0025935

__00000B 0 -.0098802 .0030464 -.0158509 -.0039094 1 -.0023741 .0049636 -.0121025 .0073543

__00000C 0 -.0045069 .0025498 -.0095045 .0004906 1 -.0033108 .004377 -.0118896 .0052681

__00000D 0 -.0029327 .0020563 -.006963 .0010976 1 -.0008339 .0040234 -.0087198 .0070519

__00000E 0 -.0009192 .0015328 -.0039235 .002085 1 -.0007112 .0027881 -.0061759 .0047534



p10 0 460 17.5 26.29 0.000 425.7006 494.2994 1 700 15 46.67 0.000 670.6005 729.3995

p25 0 700 5 140.00 0.000 690.2002 709.7998 1 900 7.5 120.00 0.000 885.3003 914.6997

p50 0 900 10 90.00 0.000 880.4004 919.5996 1 1100 . . . . .

p75 0 1300 25 52.00 0.000 1251.001 1348.999 1 1300 . . . . .

p95 0 1900 25 76.00 0.000 1851.001 1948.999 1 1800 100 18.00 0.000 1604.004 1995.996

p96


0 2000 25 80.00 0.000 1951.001 2048.999 1 2000 100 20.00 0.000 1804.004 2195.996

p97 0 2100 50 42.00 0.000 2002.002 2197.998 1 2200 75 29.33 0.000 2053.003 2346.997

p98 0 2300 50 46.00 0.000 2202.002 2397.998 1 2400 100 24.00 0.000 2204.004 2595.996

p99 0 2600 75 34.67 0.000 2453.003 2746.997 1 2700 125 21.60 0.000 2455.005 2944.995

100 . 101 . svy, subpop(p3743): tab hispimmig if sp==1



hispimmig proportion

0 .3527 1 .6473 Total 1


102 . svy: tab hispimmig if sp==1(running tabulate on estimation sample)



0 .5925 1 .4075 Total 1



103 . 104 . svy, subpop(twobedroomrental): mean rntp if sp==1, over(hispimmig)

(running mean on estimation sample)


.................................................. 50

..............................


0: hispimmig = 0 1: hispimmig = 1


rntp 0 1229.962 16.25733 1198.098 1261.825 1 989.208 11.72156 966.2342 1012.182

105 . svy, subpop(puma2bdrm): mean rntp if sp==1, over(hispimmig)(running mean on estimation sample)


.................................................. 50

..............................


0: hispimmig = 0 1: hispimmig = 1


rntp 0 967.5815 34.70866 899.5538 1035.609 1 956.3982 30.76891 896.0922 1016.704

106 . 107 . epctile monthlyrent if sp==1, p(10 25 50 75 90 95 ) svy over(puma2bdrm hispimmig)



.................................................. 50

..............................



Over: puma2bdrm hispimmig _subpop_1: 0 0 _subpop_2: 0 1 _subpop_3: 1 0 _subpop_4: 1 1



__000007 _subpop_1 -.0161495 .006437 -.0287658 -.0035331 _subpop_2 -.0042777 .0073924 -.0187666 .0102111 _subpop_3 -.0611413 .0606909 -.1800933 .0578107 _subpop_4 -.0365095 .0496854 -.133891 .060872

__000008 _subpop_1 -.0467387 .0091388 -.0646504 -.028827 _subpop_2 -.0409791 .0092151 -.0590403 -.0229179 _subpop_3 -.0203804 .0827615 -.1825901 .1418292 _subpop_4 -.0193189 .0551873 -.1274841 .0888462

__000009 _subpop_1 -.0227752 .0074444 -.037366 -.0081844 _subpop_2 -.0090984 .007701 -.024192 .0059952 _subpop_3 -.1915761 .0816827 -.3516711 -.031481 _subpop_4 -.0171906 .0537764 -.1225903 .0882092

__00000A _subpop_1 -.0196439 .0049882 -.0294206 -.0098673 _subpop_2 -.0192232 .0056618 -.03032 -.0081263 _subpop_3 -.0168478 .0527304 -.1201975 .0865019 _subpop_4 -.1671906 .0537764 -.2725903 -.0617908

__00000B _subpop_1 -.009239 .0038858 -.0168551 -.0016229 _subpop_2 -.0014573 .0045915 -.0104564 .0075419 _subpop_3 -.0668478 .0527304 -.1701975 .0365019 _subpop_4 -.0338245 .0274747 -.087674 .020025



p10 0_0 430 17.5 24.57 0.000 395.7006 464.2994 0_1 550 10 55.00 0.000 530.4004 569.5996 1_0 670 62.5 10.72 0.000 547.5023 792.4977 1_1 600 45 13.33 0.000 511.8016 688.1984

p25 0_0 750 7.5 100.00 0.000 735.3003 764.6997 0_1 740 7.5 98.67 0.000 725.3003 754.6997 1_0 800 50 16.00 0.000 702.0018 897.9982 1_1 750 55 13.64 0.000 642.202 857.798

p50 0_0 1000 25 40.00 0.000 951.0009 1048.999 0_1 900 2.5 360.00 0.000 895.1001 904.8999


1_0 1000 67.5 14.81 0.000 867.7024 1132.298 1_1 990 25 39.60 0.000 941.0009 1038.999

p75 0_0 1400 . . . . . 0_1 1200 25 48.00 0.000 1151.001 1248.999 1_0 1100 50 22.00 0.000 1002.002 1197.998 1_1 1200 50 24.00 0.000 1102.002 1297.998

p90 0_0 1800 25 72.00 0.000 1751.001 1848.999 0_1 1400 . . . . . 1_0 1300 50 26.00 0.000 1202.002 1397.998 1_1 1200 . . . . .

p95 0_0 2100 25 84.00 0.000 2051.001 2148.999 0_1 1600 25 64.00 0.000 1551.001 1648.999 1_0 1300 50 26.00 0.000 1202.002 1397.998 1_1 1300 50 26.00 0.000 1202.002 1397.998

108 . epctile monthlyrent if sp==1, p(10 25 50 75 90 95 ) svy over(twobedroomrental hispimmig)Warning: if and in conditions should be implemented via svy, subpop option(running mean on estimation sample)


.................................................. 50

..............................


Over: twobedroomrental hispimmig _subpop_1: 0 0 _subpop_2: 0 1 _subpop_3: 1 0 _subpop_4: 1 1



__000007 _subpop_1 -.0173961 .0072761 -.031657 -.0031352 _subpop_2 -.0439888 .0072076 -.0581154 -.0298622 _subpop_3 -.0022555 .0146919 -.0310512 .0265402 _subpop_4 -.0190685 .0180362 -.0544189 .0162819

__000008 _subpop_1 -.0093279 .0099717 -.0288722 .0102163 _subpop_2 -.0003636 .0106107 -.0211603 .020433 _subpop_3 -.0287255 .0179782 -.0639622 .0065112 _subpop_4 -.057869 .0210006 -.0990293 -.0167086

__000009 _subpop_1 -.0159099 .0082273 -.032035 .0002153


_subpop_2 -.0062624 .008412 -.0227497 .0102249 _subpop_3 -.038292 .0151442 -.0679741 -.0086099 _subpop_4 -.0213479 .0179951 -.0566177 .0139219

__00000A _subpop_1 -.0233722 .0061237 -.0353745 -.0113699 _subpop_2 -.0314378 .0061217 -.0434362 -.0194395 _subpop_3 -.0014207 .0102888 -.0215864 .0187449 _subpop_4 -.0206369 .0129239 -.0459673 .0046935

__00000B _subpop_1 -.0070151 .0044418 -.0157209 .0016908 _subpop_2 -.0077779 .0054399 -.0184399 .0028842 _subpop_3 -.0068806 .0076578 -.0218897 .0081284 _subpop_4 -.0128019 .0108768 -.0341201 .0085163



p10 0_0 380 20 19.00 0.000 340.8007 419.1993 0_1 540 15 36.00 0.000 510.6005 569.3995 1_0 730 27.5 26.55 0.000 676.101 783.899 1_1 600 35 17.14 0.000 531.4013 668.5987

p25 0_0 700 7.5 93.33 0.000 685.3003 714.6997 0_1 700 7.5 93.33 0.000 685.3003 714.6997 1_0 990 12.5 79.20 0.000 965.5005 1014.5 1_1 850 22.5 37.78 0.000 805.9008 894.0992

p50 0_0 980 12.5 78.40 0.000 955.5005 1004.5 0_1 900 7.5 120.00 0.000 885.3003 914.6997 1_0 1200 25 48.00 0.000 1151.001 1248.999 1_1 1000 . . . . .

p75 0_0 1400 25 56.00 0.000 1351.001 1448.999 0_1 1200 25 48.00 0.000 1151.001 1248.999 1_0 1400 . . . . . 1_1 1200 25 48.00 0.000 1151.001 1248.999

p90 0_0 1800 25 72.00 0.000 1751.001 1848.999 0_1 1400 25 56.00 0.000 1351.001 1448.999 1_0 1800 50 36.00 0.000 1702.002 1897.998 1_1 1300 25 52.00 0.000 1251.001 1348.999

p95 0_0 2100 50 42.00 0.000 2002.002 2197.998 0_1 1600 50 32.00 0.000 1502.002 1697.998 1_0 2200 100 22.00 0.000 2004.004 2395.996 1_1 1400 50 28.00 0.000 1302.002 1497.998


109 . 110 . kdensity rntp if hispimmig==1 & twobedroomrental ==1 & sp==1 [aw=wgtp], ///

> addplot(kdensity rntp if hispimmig==0 & twobedroomrental ==1 & sp==1 [aw=wgtp]) ///> title("Figure 1. Distribution of Monthly Rent for" ///> "Two-Bedroom Apartments by National Origin") ///> sch(s2mono) legend(label(1 "Hispanic Immigrant HH") lab(2 "Other HH")) ///> note("") name(Figure1a, replace)

111 . *graph export fig\Figure1a.png, replace112 . 113 . kdensity rntp if hispwhite==1 & twobedroomrental ==1 & sp==1 [aw=wgtp], ///

> addplot(kdensity rntp if hispwhite==0 & twobedroomrental ==1 & sp==1 [aw=wgtp]) ///> title("Figure 2. Distribution of Monthly Rent for" ///> "Two-Bedroom Apartments by National Origin") ///> sch(s2mono) legend(label(1 "Hispanic HH") lab(2 "Non-Hispanic White HH")) ///> note("") name(Figure2a, replace)

114 . *graph export fig\Figure2.png, replace115 . 116 . 117 . epctile monthlyrent if sp==1 & twobedroomrental==1, p(50 75 90 95 99 ) svy over(race5)



.................................................. 50

..............................


1: race5 = 1 2: race5 = 2 3: race5 = 3 4: race5 = 4 5: race5 = 5 6: race5 = 6 7: race5 = 7


__000006 1 -.0507941 .0399334 -.1290622 .027474 2 -.0149051 .0952325 -.2015575 .1717472 3 -.0284187 .0292325 -.0857132 .0288759 4 -.0411765 .2067827 -.4464631 .3641102 5 -.0506849 .1270317 -.2996624 .1982925 6 -.120874 .026581 -.1729717 -.0687763 7 -.057869 .0210006 -.0990293 -.0167086

__000007 1 -.0056188 .0351791 -.0745686 .0633309 2 -.0372629 .0861238 -.2060624 .1315366 3 -.0212437 .0238673 -.0680228 .0255355 4 -.004902 .1322095 -.2640279 .254224 5 -.0020548 .1029828 -.2038975 .1997879 6 -.0419952 .0278711 -.0966216 .0126312 7 -.0213479 .0179951 -.0566177 .0139219

__000008 1 -.0003896 .0284792 -.0562078 .0554286 2 -.0233062 .0744185 -.1691639 .1225514 3 -.0283061 .0171673 -.0619535 .0053412


4 -.0058823 .0913459 -.184917 .1731523 5 -.0506849 .1022232 -.2510386 .1496688 6 -.006909 .0197782 -.0456735 .0318555 7 -.0206369 .0129239 -.0459673 .0046935

__000009 1 -.0000449 .0198712 -.0389918 .0389019 2 -.0394309 .0733536 -.1832014 .1043396 3 -.0013506 .0146735 -.0301101 .0274089 4 -.0166667 .0717419 -.1572782 .1239448 5 -.1006849 .1022232 -.3010386 .0996688 6 -.0156154 .0169452 -.0488274 .0175966 7 -.0128019 .0108768 -.0341201 .0085163

__00000A 1 -.0058825 .0110129 -.0274673 .0157023 2 -.0008401 .0123946 -.025133 .0234528 3 -.0113844 .0102287 -.0314323 .0086635 4 -.0566667 .0717419 -.1972782 .0839448 5 -.1406849 .1022232 -.3410387 .0596688 6 -.0037645 .0067314 -.0169578 .0094288 7 -.0089835 .0073916 -.0234708 .0055039



p50 1 1400 75 18.67 0.000 1253.003 1546.997 2 1100 87.5 12.57 0.000 928.5032 1271.497 3 1200 25 48.00 0.000 1151.001 1248.999 4 980 212.5 4.61 0.000 563.5077 1396.492 5 1400 250 5.60 0.000 910.009 1889.991 6 1100 . . . . . 7 1000 . . . . .

p75 1 1900 100 19.00 0.000 1704.004 2095.996 2 1400 125 11.20 0.000 1155.005 1644.995 3 1400 25 56.00 0.000 1351.001 1448.999 4 1200 110 10.91 0.000 984.404 1415.596 5 2100 300 7.00 0.000 1512.011 2687.989 6 1300 25 52.00 0.000 1251.001 1348.999 7 1200 25 48.00 0.000 1151.001 1248.999

p90 1 2300 175 13.14 0.000 1957.006 2642.994 2 1500 50 30.00 0.000 1402.002 1597.998 3 1700 75 22.67 0.000 1553.003 1846.997 4 1300 150 8.67 0.000 1006.005 1593.995 5 2600 350 7.43 0.000 1914.013 3285.987 6 1500 50 30.00 0.000 1402.002 1597.998 7 1300 25 52.00 0.000 1251.001 1348.999

p95 1 2600 325 8.00 0.000 1963.012 3236.988 2 1700 150 11.33 0.000 1406.005 1993.995 3 2300 300 7.67 0.000 1712.011 2887.989 4 1700 250 6.80 0.000 1210.009 2189.991 5 2600 350 7.43 0.000 1914.013 3285.987 6 1600 75 21.33 0.000 1453.003 1746.997 7 1400 50 28.00 0.000 1302.002 1497.998

p99 1 3600 400 9.00 0.000 2816.014 4383.986


2 1900 100 19.00 0.000 1704.004 2095.996 3 3000 200 15.00 0.000 2608.007 3391.993 4 1700 250 6.80 0.000 1210.009 2189.991 5 2600 250 10.40 0.000 2110.009 3089.991 6 2200 200 11.00 0.000 1808.007 2591.993 7 1700 100 17.00 0.000 1504.004 1895.996

118 . 119 . epctile hincp if sp==1 & twobedroomrental==1, p(50 75 90 95 99 ) svy over(race5)



.................................................. 50

..............................


1: race5 = 1 2: race5 = 2 3: race5 = 3 4: race5 = 4 5: race5 = 5 6: race5 = 6 7: race5 = 7


__000006 1 -.0025472 .0445801 -.0899226 .0848282 2 -.0189702 .0846016 -.1847863 .146846 3 -.0011255 .0276776 -.0553726 .0531216 4 -.154902 .1384729 -.426304 .1165 5 -.0315068 .1501568 -.3258088 .2627951 6 -.0014032 .0316219 -.063381 .0605747 7 -.0058175 .0196445 -.04432 .032685

__000007 1 -.0059185 .0352658 -.0750382 .0632012 2 -.0413279 .0849696 -.2078653 .1252094 3 -.0057681 .0248318 -.0544376 .0429013 4 -.0558824 .1708715 -.3907843 .2790196 5 -.0075342 .1031709 -.2097455 .194677 6 -.003107 .0266679 -.0553751 .049161 7 -.001616 .0160308 -.0330358 .0298039

__000008 1 -.0030866 .0248558 -.051803 .0456298 2 -.0680217 .0818879 -.2285191 .0924757 3 -.0055149 .017129 -.0390871 .0280573 4 -.072549 .1303107 -.3279534 .1828553 5 -.1191781 .0986675 -.3125628 .0742066 6 -.0028999 .0221764 -.0463648 .040565 7 -.0006328 .0125953 -.0253192 .0240536

__000009 1 -.0030417 .0192023 -.0406774 .0345941 2 -.0055556 .0362021 -.0765103 .0653992 3 -.0041643 .0157178 -.0349706 .026642


4 -.122549 .1303107 -.3779534 .1328553 5 -.0239726 .0548653 -.1315066 .0835614 6 -.0006481 .0146687 -.0293983 .0281021 7 -.0004185 .0077677 -.0156428 .0148059

__00000A 1 -.0124753 .01457 -.041032 .0160814 2 -.0171003 .0274573 -.0709157 .0367151 3 -.0004108 .0060285 -.0122264 .0114048 4 -.162549 .1303107 -.4179534 .0928553 5 -.0228767 .0354203 -.0922992 .0465457 6 -.0013591 .0062005 -.0135118 .0107936 7 -.0005464 .0033748 -.0071608 .0060681


SDR hincp Coef. Std. Err. z P>|z| [95% Conf. Interval]

p50 1 52300 6900 7.58 0.000 38776.25 65823.75 2 29000 5225 5.55 0.000 18759.19 39240.81 3 39400 2500 15.76 0.000 34500.09 44299.91 4 16800 12825 1.31 0.190 -8336.538 41936.54 5 77000 32900 2.34 0.019 12517.18 141482.8 6 42300 2600 16.27 0.000 37204.09 47395.91 7 32000 1250 25.60 0.000 29550.05 34449.95

p75 1 115000 13462.5 8.54 0.000 88613.98 141386 2 49000 11375 4.31 0.000 26705.41 71294.59 3 74100 7325 10.12 0.000 59743.26 88456.74 4 49000 16100 3.04 0.002 17444.58 80555.42 5 115160 15150 7.60 0.000 85466.55 144853.5 6 72000 4300 16.74 0.000 63572.15 80427.85 7 51000 1500 34.00 0.000 48060.05 53939.95

p90 1 189000 17750 10.65 0.000 154210.6 223789.4 2 79100 15050 5.26 0.000 49602.54 108597.5 3 117100 8500 13.78 0.000 100440.3 133759.7 4 98900 27895 3.55 0.000 44226.8 153573.2 5 140000 17500 8.00 0.000 105700.6 174299.4 6 95500 6825 13.99 0.000 82123.25 108876.8 7 73000 3700 19.73 0.000 65748.13 80251.87

p95 1 231000 37500 6.16 0.000 157501.4 304498.6 2 119100 20000 5.95 0.000 79900.72 158299.3 3 166900 24767.5 6.74 0.000 118356.6 215443.4 4 98900 27895 3.55 0.000 44226.8 153573.2 5 145600 2800 52.00 0.000 140112.1 151087.9 6 124000 8500 14.59 0.000 107340.3 140659.7 7 90200 4925 18.31 0.000 80547.18 99852.82

p99 1 350000 10500 33.33 0.000 329420.4 370579.6 2 175000 29000 6.03 0.000 118161 231839 3 242000 11000 22.00 0.000 220440.4 263559.6 4 98900 24950 3.96 0.000 49998.9 147801.1 5 160000 10000 16.00 0.000 140400.4 179599.6 6 155000 7500 20.67 0.000 140300.3 169699.7 7 131100 17700 7.41 0.000 96408.64 165791.4


120 . 121 . 122 . 123 . /* Rent to income ratio */124 . preserve

125 . keep if rentinc <=1(25,512 observations deleted)

126 . kdensity rentinc if hispwhite==1 & twobedroomrental ==1 & sp==1 [aw=wgtp], ///> addplot(kdensity rentinc if hispwhite==0 & twobedroomrental ==1 & sp==1 [aw=wgtp]) ///> title("Figure 3. Distribution of Rent-to-Income Ratio for" ///> "Two-Bedroom Apartments by National Origin") ///> sch(s2mono) legend(label(1 "Hispanic HH") lab(2 "Non-White Hispanic HH")) ///> note("") name(Figure3a, replace)

127 . restore

128 . *graph export fig\Figure3.png, replace129 . 130 . preserve

131 . keep if loghincp >3.5(385 observations deleted)

132 . kdensity loghincp if hispwhite==1 & twobedroomrental ==1 & sp==1 [aw=wgtp], ///> addplot(kdensity loghincp if hispwhite==0 & twobedroomrental ==1 & sp==1 [aw=wgtp]) ///> title("Figure 4. Distribution of Household Income for" ///> "Tenants of Two-Bedroom Apartments by National Origin") ///> sch(s2mono) legend(label(1 "Hispanic HH") lab(2 "Non-White Hispanic HH")) ///> note("") name(Figure4a, replace) ///> xlab(4 "$10k" 4.4 "$25k" 4.7 "$50k" 5.2 "$150k")

133 . restore

134 . 135 . /* What percentage of each group can afford an apt at $1250? at $2000?

> This analysis uses the population rent-to-income of two-bedroom > rental households */

136 . svy, subpop(twobedroomrental): tab race5 r50b1250 if sp==1, col(running tabulate on estimation sample)


RECODE of rac1p r50b1250 (RAC1P) 0 1 Total

White .0727 .1415 .0982 Black .0244 .0171 .0217 Asian .1933 .2362 .2092 Other .0083 .0062 .0075 Mixed Ra .0064 .0181 .0107 Hispanic .1969 .2599 .2202 Hispanic .498 .321 .4325 Total 1 1 1




137 . svy, subpop(twobedroomrental): tab race5 r50b2000 if sp==1, col(running tabulate on estimation sample)


RECODE of rac1p r50b2000 (RAC1P) 0 1 Total

White .0736 .2089 .0982 Black .0221 .0201 .0217 Asian .195 .2731 .2092 Other .0072 .0088 .0075 Mixed Ra .0061 .0314 .0107 Hispanic .2119 .2575 .2202 Hispanic .484 .2002 .4325 Total 1 1 1



138 . 139 . 140 . /*PERSON LEVEL ANALYSIS*/141 . /* person weights */142 . svyset [pw=pwgtp], sdr(pwgtp1 - pwgtp80) vce(sdr)

pweight: pwgtp VCE: sdr MSE: off sdrweight: pwgtp1 pwgtp2 pwgtp3 pwgtp4 pwgtp5 pwgtp6 pwgtp7 pwgtp8 pwgtp9 pwgtp10 pwgtp11 pwgtp12 pwgtp13 pwgtp14 pwgtp15 pwgtp16 pwgtp17 pwgtp18 pwgtp19 pwgtp20 pwgtp21 pwgtp22 pwgtp23 pwgtp24 pwgtp25 pwgtp26 pwgtp27 pwgtp28 pwgtp29 pwgtp30 pwgtp31 pwgtp32 pwgtp33 pwgtp34 pwgtp35 pwgtp36 pwgtp37 pwgtp38 pwgtp39 pwgtp40 pwgtp41 pwgtp42 pwgtp43 pwgtp44 pwgtp45 pwgtp46 pwgtp47 pwgtp48 pwgtp49 pwgtp50 pwgtp51 pwgtp52 pwgtp53 pwgtp54 pwgtp55 pwgtp56 pwgtp57 pwgtp58 pwgtp59 pwgtp60 pwgtp61 pwgtp62 pwgtp63 pwgtp64 pwgtp65 pwgtp66 pwgtp67 pwgtp68 pwgtp69 pwgtp70 pwgtp71 pwgtp72 pwgtp73 pwgtp74 pwgtp75 pwgtp76 pwgtp77 pwgtp78 pwgtp79 pwgtp80 Single unit: missing Strata 1: <one> SU 1: <observations> FPC 1: <zero>

143 .


144 . /* What is the national origin of those in the sample? */145 . svy, subpop(hispanic): tab hisp6



RECODE of hisp (Hispanic or Latino Origin) proportion

1 .8568 2 .011 3 .1047 4 .0125 5 .0029 6 .012 Total 1


146 . svy, subpop(hispimmig): tab hisp6(running tabulate on estimation sample)


RECODE of hisp (Hispanic or Latino Origin) proportion

1 .8104 2 .0116 3 .1508 4 .0184 5 .0017 6 .007 Total 1



147 . 148 . 149 . svy, subpop(p3743): tab hispimmig




0 .5812 1 .4188 Total 1


150 . svy: tab hispimmig(running tabulate on estimation sample)



0 .7035 1 .2965 Total 1


151 . 152 . svy, subpop(puma2bdrm): tab hispimmig


Number of obs = 50,373 Population size = 917,141 Subpop. no. obs = 416 Subpop. size = 8,875 Replications = 80


0 .5888 1 .4112 Total 1



153 . svy, subpop(twobedroomrental): tab hispimmig(running tabulate on estimation sample)



0 .6954 1 .3046 Total 1


154 . 155 . svy, subpop(hispimmig): tab twobedroomrental




0 .8686 1 .1314 Total 1


156 . 157 . svy, subpop(hispimmig): tab bld3



RECODE of bld (BLD) proportion

One fami .6636 Apt buil .3249 Other .0114 Total 1



158 . 159 . 160 . /*******************************************************/161 . /*** 1-year analysis for rent paid by recent movers ****/162 . cd $dropbox

D:\Dropbox

163 . cd Expert\WinstarD:\Dropbox\Expert\Winstar

164 . 165 . use data\ss15hca-1year.dta, clear

166 . merge 1:m serialno using "data\ss15pca-1year.dta" , gen(lamerge)

Result # of obs. not matched 154 from master 154 (lamerge==1) from using 0 (lamerge==2)

matched 12,464 (lamerge==3)

167 . drop if lamerge==1 /* drops unoccupied housing units */(154 observations deleted)

168 . 169 . lab var rntp "Monthly Rent"

170 . 171 . /* Identify rental units */172 . gen rentalunit = ten==3

173 . 174 . /* Create apartment binary variable */175 . gen apt = bld>=4 & bld<=9

176 . lab var apt "Apartment 2+ units"

177 . 178 . /* Create variable for comparable two bedroom rental units */179 . gen twobedroomrental = bdsp==2 & rentalunit == 1 & apt==1

180 . 181 . /* Amount of monthly rent */182 . gen monthlyrent = rntp


183 . 184 . /* Create 2-category duration of tenure variable */185 . gen years2 = inlist(mv,1,2)

186 . lab var years2 "Lived in unit 2 or fewer years"


187 . 188 . /*HOUSING UNIT ANALYSIS*/189 . /* housing weights */190 . svyset [pw=wgtp], sdr(wgtp1 - wgtp80) vce(sdr)

pweight: wgtp VCE: sdr MSE: off sdrweight: wgtp1 wgtp2 wgtp3 wgtp4 wgtp5 wgtp6 wgtp7 wgtp8 wgtp9 wgtp10 wgtp11 wgtp12 wgtp13 wgtp14 wgtp15 wgtp16 wgtp17 wgtp18 wgtp19 wgtp20 wgtp21 wgtp22 wgtp23 wgtp24 wgtp25 wgtp26 wgtp27 wgtp28 wgtp29 wgtp30 wgtp31 wgtp32 wgtp33 wgtp34 wgtp35 wgtp36 wgtp37 wgtp38 wgtp39 wgtp40 wgtp41 wgtp42 wgtp43 wgtp44 wgtp45 wgtp46 wgtp47 wgtp48 wgtp49 wgtp50 wgtp51 wgtp52 wgtp53 wgtp54 wgtp55 wgtp56 wgtp57 wgtp58 wgtp59 wgtp60 wgtp61 wgtp62 wgtp63 wgtp64 wgtp65 wgtp66 wgtp67 wgtp68 wgtp69 wgtp70 wgtp71 wgtp72 wgtp73 wgtp74 wgtp75 wgtp76 wgtp77 wgtp78 wgtp79 wgtp80 Single unit: missing Strata 1: <one> SU 1: <observations> FPC 1: <zero>

191 . 192 . /* Monthly rent by length of tenure in housing unit */193 . epctile monthlyrent if sp==1, p(50 75 85 90 95 ) svy over(twobedroomrental years2)



.................................................. 50

..............................


Over: twobedroomrental years2 _subpop_1: 0 0 _subpop_2: 0 1 _subpop_3: 1 0 _subpop_4: 1 1


__000006 _subpop_1 -.003482 .0144301 -.0317645 .0248006 _subpop_2 -.0102701 .0283688 -.065872 .0453319 _subpop_3 -.0884656 .0296175 -.1465148 -.0304163 _subpop_4 -.01718 .0605797 -.1359141 .1015541

__000007 _subpop_1 -.0009154 .0142519 -.0288486 .0270178 _subpop_2 -.06019 .0294456 -.1179022 -.0024777 _subpop_3 -.0234284 .0227869 -.0680898 .021233 _subpop_4 -.0129823 .0503848 -.1117348 .0857701


__000009 _subpop_1 -.002265 .0088128 -.0195377 .0150077 _subpop_2 -.0164607 .0206084 -.0568524 .023931 _subpop_3 -.0219265 .0194005 -.0599507 .0160978


_subpop_4 -.0110864 .0396652 -.0888288 .0666561

__00000A _subpop_1 -.005184 .0069954 -.0188947 .0085268 _subpop_2 -.0115059 .0168001 -.0444334 .0214217 _subpop_3 -.0004523 .0144886 -.0288495 .0279448 _subpop_4 -.0157178 .0346601 -.0836503 .0522148



p50 0_0 930 12.5 74.40 0.000 905.5005 954.4995 0_1 1100 60 18.33 0.000 982.4022 1217.598 1_0 1100 25 44.00 0.000 1051.001 1148.999 1_1 1300 50 26.00 0.000 1202.002 1397.998

p75 0_0 1300 25 52.00 0.000 1251.001 1348.999 0_1 1500 50 30.00 0.000 1402.002 1597.998 1_0 1300 25 52.00 0.000 1251.001 1348.999 1_1 1600 125 12.80 0.000 1355.005 1844.995

p85 0_0 1500 50 30.00 0.000 1402.002 1597.998 0_1 1900 75 25.33 0.000 1753.003 2046.997 1_0 1400 50 28.00 0.000 1302.002 1497.998 1_1 2000 150 13.33 0.000 1706.005 2293.995

p90 0_0 1700 50 34.00 0.000 1602.002 1797.998 0_1 2000 125 16.00 0.000 1755.005 2244.995 1_0 1500 50 30.00 0.000 1402.002 1597.998 1_1 2200 200 11.00 0.000 1808.007 2591.993

p95 0_0 2000 25 80.00 0.000 1951.001 2048.999 0_1 2400 175 13.71 0.000 2057.006 2742.994 1_0 1700 100 17.00 0.000 1504.004 1895.996 1_1 2400 150 16.00 0.000 2106.005 2693.995

194 . 195 . log close

name: <unnamed> log: D:\Dropbox\Expert\Winstar\log\WinstarAnalysis.smcl log type: smcl closed on: 26 Jun 2017, 11:12:40


Appendix C

Map Used to Identify PUMAs


Appendix D

Tables and Figures


Table 1. Distribution of Monthly Rent for a 2 Bedroom Apartment, by Race/National Origin

Rent Percentile for Each Group

Median 75th 90th 95th

Hispanic Immigrant

$1000 $1200 $1300 $1400

Hispanic Non‐Immigrant

$1100 $1300 $1500 $1600

Non‐Hisp White $1400 $1900 $2300 $2600

Black $1100 $1400 $1500 $1700

Asian $1200 $1400 $1700 $2300

Other $980 $1200 $1300 $1700

Mixed Race $1400 $2100 $2600 $2600



Table 2. Simulated Change in Affordability for a 2 Bedroom Apartment, by Race/National Origin

Affordable Rent Simulated Change $1,250 $2,000

Hispanic Immigrant 32% 20% ‐37%

Hispanic Non‐Immigrant 26% 26% 0%

Non‐Hisp White 14% 21% +50%

Black 2% 2% 0%

Asian 24% 27% 13%

Other 1% 1% 0%

Mixed Race 2% 3% 50%



adela hernandez, et al. v winstar properties, et al. - …... et al. v winstar properties, et al....

Documents