Project Summary Housing Affordability Data for CLF Equity Atlas 2.0

Similar documents
Technical Description of the Freddie Mac House Price Index

Average Change Sale Price 07 v. 08 Clackamas $375,800-6% Columbia $230,700-9% Multnomah $331,400-1% Washington $317,100-5% Yamhill $270,700-5%

Research Report #6-07 LEGISLATIVE REVENUE OFFICE.

Preliminary Analysis

Closed Sales. Pending Sales

AVM Validation. Evaluating AVM performance

Briefing Book. State of the Housing Market Update San Francisco Mayor s Office of Housing and Community Development

7224 Nall Ave Prairie Village, KS 66208

State of the Nation s Housing 2008: A Preview

5. PROPERTY VALUES. In this section, we focus on the economic impact that AMDimpaired

Integrating SAS and Geographic Information Systems for Regional Land Use Planning

2.2 Future Demand Projection Methodology

Metro Boston Perfect Fit Parking Initiative

FY 2013 Fair Market Rent Documentation System

Wind Energy Device Valuation

Rent Stabilization, Vacancy Decontrol and Reinvestment in Rental Property in Berkeley, California

Housing and Mortgage Market Update

Town of Gilford, New Hampshire

The recent report which was released by real estate website Zillow show that Los

A Model to Calculate the Supply of Affordable Housing in Polk County

2011 ASSESSMENT RATIO REPORT

Comparables Sales Price (Old Version)

Residential Real Estate Market Overview: September 2017 Data

American Community Survey 5-Year Estimates

American Community Survey 5-Year Estimates

Year to Date Summary. Average and Median Sale Prices

An Alternate Approach to Address Creation and Maintenance Hernando County Property Appraiser Alvin R. Mazourek, CFA

Town of Prescott Valley 2013 Land Use Assumptions

RESIDENTIAL MARKET ANALYSIS

SELECTED HOUSING CHARACTERISTICS American Community Survey 5-Year Estimates

Shaping Our Future. Return-on-Investment Study. June 2017

Section 5: Fair Housing Index

MULTIFAMILY MARKET ANALYSIS

The Honorable Larry Hogan And The General Assembly of Maryland

June 6, Proposed FY Annual Automatic Adjustment for the Affordable Housing Unit Base Fee

The Impact of Using. Market-Value to Replacement-Cost. Ratios on Housing Insurance in Toledo Neighborhoods

April 12, The Honorable Martin O Malley And The General Assembly of Maryland

Average Sale Price. Closed Sales

Housing Price Forecasts. Illinois and Chicago PMSA, December 2015

Average and Median Sale Prices

Miami Beach, FL 33141

Volume Title: Well Worth Saving: How the New Deal Safeguarded Home Ownership

Closed Sales. Pending Sales

RESIDENTIAL MARKET ANALYSIS

Characteristics of Recent Home Buyers

Demonstration Properties for the TAUREAN Residential Valuation System

MARKET ACTION. Year-to-Date Trends

PROPERTY TAX IS A PRINCIPAL REVENUE SOURCE

RENTAL PROPERTY $ 1, % 4.1 % SINGLE-FAMILY 1,308 2, mi. SMALLER THAN 86 % This report provides an in-depth comparison of

American Community Survey 5-Year Estimates

RENTAL PROPERTY $ 1, % 14.2 % SINGLE-FAMILY 1,748 11, mi. LARGER THAN 22 %

HOUSINGSPOTLIGHT. The Shrinking Supply of Affordable Housing

Median Income and Median Home Price

Cook County Assessor s Office: 2019 North Triad Assessment. Evanston Residential Assessment Narrative Updated: April 8 th, 2019

Housing Affordability in Lexington, Kentucky

DRAFT. Development Impact Fee Model Ordinance. Mount Pleasant, SC. Draft Document. City Explained, Inc. J. R. Wilburn and Associates, Inc.

Chapter 12 Changes Since This is just a brief and cursory comparison. More analysis will be done at a later date.

RENTAL PROPERTY $ % 3.9 % SINGLE-FAMILY 1,088 NOT FOUND mi. SMALLER THAN 37 % This report provides an in-depth comparison of

City of San José. Produced by City of San José Department of Housing. Housing Market Update. First Quarter 2018

The rapidly rising price of single-family homes in. Change and Challenges East Austin's Affordable Housing Problem

RENTAL PROPERTY $ % 7.5 % SINGLE-FAMILY NOT FOUND 2,348 NOT FOUND mi. LARGER THAN 93 % This report provides an in-depth comparison of

MARKET AREA UPDATE Year: 2018 Report as of: 1Q 2Q 3Q 4Q

Estimate of the Percentage of Rent that Constitutes Property Taxes in Minnesota. Based on Rent and Property Taxes Paid in 2016

Census Tract Data Analysis

Assessment Year 2016 Assessment Valuations / Mass Appraisal Summary Report

RENTAL PROPERTY $ % 6.9 % SINGLE-FAMILY NOT FOUND mi. SMALLER THAN 14 %

Moorestown, NJ 08057

Methodological Appendix: The Growing Shortage of Affordable Housing for the Extremely Low Income in Massachusetts

Washington Department of Revenue Property Tax Division. Valid Sales Study Kitsap County 2015 Sales for 2016 Ratio Year.

BUILD-OUT ANALYSIS GRANTHAM, NEW HAMPSHIRE

Burlington Unincorporated Community Plan

IHS Regional Housing Market Segmentation Analysis

Town of Windham. Planning Department 8 School Road Windham, ME Voice ext. 2 Fax

FEBRUARY Published March 25, 2016

HOUSING MARKET ANALYSIS

FINAL REPORT AN ANALYSIS OF SECONDARY ROAD MAINTENANCE PAYMENTS TO HENRICO AND ARLINGTON COUNTIES WITH THE DECEMBER 2001 UPDATE

2013 Update: The Spillover Effects of Foreclosures

REDSTONE. Regression Fundamentals.

Monthly Indicators. September Monthly Snapshot + 2.0% + 7.4% %

RISK ASSESSMENT APPENDIX E HAZARD ANALYSIS RESULTS

Report on Nevada s Housing Market

Monthly Indicators % % - 1.1%

Rental Housing Finance Survey* *and a few observations about gubment data. Rich Levy US Census Bureau Washington, DC April 2014

Neighborhood Market Study/Housing Needs Assessment

RENTAL PROPERTY $ 1, % 8.2 % MULTI-FAMILY 1,408 NOT FOUND mi. LARGER THAN 26 % This report provides an in-depth comparison of

Estimating User Accessibility Benefits with a Housing Sales Hedonic Model

CHAPTER 2 VACANT AND REDEVELOPABLE LAND INVENTORY

RENTAL PROPERTY $ 2, % 6.8 % SINGLE-FAMILY 2,122 8, mi. LARGER THAN 74 % This report provides an in-depth comparison of

RENTAL PROPERTY $ 1, % 6.5 % SINGLE-FAMILY 1,555 NOT FOUND mi. LARGER THAN 71 % This report provides an in-depth comparison of

Data Sheets Introduction

Description of IHS Hedonic Data Set and Model Developed for PUMA Area Price Index

Joint Center for Housing Studies. Harvard University

School Quality and Property Values. In Greenville, South Carolina

RENTAL PROPERTY $ 2, % 3.7 % SINGLE-FAMILY 1,260 8, mi. SMALLER THAN 32 % This report provides an in-depth

RENTAL PROPERTY $ % 5.4 % SINGLE-FAMILY 3, mi. SMALLER THAN 80 % This report provides an in-depth comparison of

Residential December 2009

STEVEN J. DREW Assessor OFFICE OF THE ASSESSOR Service, Integrity, Fairness, Internationally Recognized for Excellence

RENTAL PROPERTY $ % 3.9 % SINGLE-FAMILY 1,200 NOT FOUND mi. SMALLER THAN 53 % This report provides an in-depth

Assessment-To-Sales Ratio Study for Division III Equalization Funding: 1999 Project Summary. State of Delaware Office of the Budget

RESIDENTIAL MARKET ANALYSIS

Transcription:

Project Summary Housing Affordability Data for CLF Equity Atlas 2.0 1. Scope The scope of this project is limited to the portion of the Equity Atlas that measures the housing affordability in the tri-county area, Clackamas, Multnomah and Washington County. Specifically the collected data will be used to generate maps similar to the following Equity Atlas maps. a. Single-Family Housing Affordability, http://equityatlas.org/maps/map3-1a.pdf b. Single-Family Housing Affordability: 2004 Incomes, 1995 Prices, http://equityatlas.org/maps/map3-1b.pdf c. Percent Change: 1990-2000: Median Value Owner-occupied Single-family Housing, http://equityatlas.org/maps/map3-2.pdf d. Percent Change:1995-2004: Median Sale Price Single Family Homes, http://equityatlas.org/maps/map3-3.pdf The updated maps will use the 2000 and 2010 tax lot and census information. 2. Stakeholders A number of groups are collaborating on the CLF Equity Atlas. Below is a list of key organizations and contacts. a. Coalition for a Living Future (CLF), contact: Kris Smock b. Portland State University (PSU), contact: Meg Merrick c. Metro, contact: Clint Chiavarini 3. Methodology The methodology for acquiring and mapping the data should be similar to the methodology used for Equity Atlas 1.0 1,2. The methodology must be consistent with the methods used by PSU, Metro and CLF. 4. Data a. Source Tax lot information has been provided by Metro for the years 2000 and 2010. The tax lot data will be used to determine median sales price per aggregate geography. The housing sales data will be used in all 4 maps within the scope of this project. The median sales price will be used to determine the percent change in house prices. b. Indices Following is a summary of standard indices that are calculated using Census Data and other resources. The indices are used for calculating regional values for the maps outlined in section 1, Scope. i. Median Income 2010 3 Median Income is used to calculate housing affordability. 1

County Median Household Income 2006-2010 3 Washington $62,574 Clackamas $62,007 Multnomah $49,618 Clark, WA $58,262 4 County Average $58,115 ii. Housing Affordability Index 4,5 The housing affordability index is $199,800. The affordability index was calculated based on the 2011/2012 lending environment. Assuming an annual income of $58,115, a down payment of $10,000, an interest rate of 4% and a household debt of less than 8% of the annual income, a buyer would qualify for a $189,800 30 year fixed rate loan. This assumes a 28/36 qualification ratio. Note: Since the down payment is less than 20% of the mortgage would be required as part of the mortgage payment. iii. Median Home Value 2000 8,9 The median value of a home in 2000 is used to calculate the percent change in housing values between 2000 and 2010. Multnomah $157,900 Washington $184,800 Clackamas $199,000 Clark,WA $147,000 Average Home Value $172,175 iv. Median Home Value 2010 3 The median value of a home in 2010 is used to calculate the percent change in housing values between 2000 and 2010. Multnomah $281,600 Washington $303,700 Clackamas $331,100 Clark, WA $260,800 Average Home Value $294,275 v. Median Housing Value Increase 2000-2010 The median house value increase from 2000 to 2010 is the regional value. The value is calculated by taking the difference between the average home values in 2010 and 2000 then dividing the difference by the average 2000 home value. (294,275 172,175)/172,175 =.71 2

vi. Inflation Between 2000 to 2010 7 The inflation rate between 2000 and 2010 is needed to calculate the affordability of houses if the housing prices kept pace with inflation. Inflation between 2000 and 2010 equals (218.1-172.2)/172.2 = 0.27 Consumer Price Index Table 1995 152.4 2004 188.9 2000 172.2 2010 218.1 (218.056) 5. Unknowns a. Determining Owner-Occupied Homes via Tax lot Data The Census Bureau changed how and what data would be collected in 2010. The long form census data was replaced by the American Community Survey (ACS). The ACS began in 2005. The housing information is now conducted as part of the ACS. The ACS collects data on a continuous basis over a 5 year period. The previous equity atlas used 2000 block data for map 3-2, Percent Change 1990-2000: Median Value Owner-occupied Single-family Housing. Block or tract data is unavailable in a single year ACS summary table. Only the data in 5 year ACS summary tables are optionally organized using census blocks and tracts. The data set for 2006-2010 is available. The prior map reflects a percent change in median home values between the values in 1990 and the values in 2000. If the 5 year ACS data were to be used, the map could not communicate the same information. If the ACS data proves to be inadequate, the tax lot data will be used to determine owner-occupied single family residents. The owner address can be compared with the site address to determine if the residence is owner-occupied. It has already been determined that condos are marked as multi-family residents. Whether or not this methodology will work has not been determined. b. Geographic Extent Map 3-2 uses tax lot data outside the tri-county, Portland-Metro area. The map includes data for Columbia, Yamhill and Clark counties. The subtitle specifically mentions the Portland-Vancouver metropolitan areas. The project description provided by CLF indicates that the tax lot data from Metro includes the tri-county area as well as Clark County. The data may not include Yamhill and Columbia County. 3

References c. Geographic Unit Three of the four maps included in the scope of this project use neighborhood as the geographic unit. One of the maps uses census block groups. Census tracts are used for this analysis. 1. The Regional Equity Atlas, Appendix B: Map Methods and Notes, pg 133-134, Chapter3: Housing, http://equityatlas.org/chapters/methodology.pdf 2. The Regional Equity Atlas, Appendix A: Neighborhood and City Summary Table, http://equityatlas.org/chapters/neighborhoodtable.pdf 3. U.S. Census Bureau, State & County Quick Facts: Oregon, Last Revised,17-Jan-2012 http://quickfacts.census.gov/qfd/states/41000.html, http://quickfacts.census.gov/qfd/states/53/53011.html 4. Freddie Mac, 30-Year Fixed Rate Mortgages Since 1971, Copyright 2012. http://www.freddiemac.com/pmms/pmms30.htm 5. Homefair, Mortgage Affordability Calculator, Copyright Move, Inc. 2006. http://www.homefair.com/tools/mortgage-affordability-calculator/index.asp 6. Oregon Property Tax Statistics, Fiscal Year 2011-2012, Oregon Department of Revenue, Pg 7, http://www.oregon.gov/dor/stats/docs/303-405-12/property-tax-stats_303-405_2011-12.pdf?ga=t 7. Historical CPI-U data from 1913 to the present, InflationData.com, Copyright 1996-2012, Capital Professional Services, http://inflationdata.com/inflation/consumer_price_index/historicalcpi.aspx 8. Oregon:2000 Summary Social, Economic, and Housing Characteristics, 2000 Census of Population and Housing Characteristics, US Census Bureau, Issued March 2003, page 154, Table 26 Home Value 2000, http://www.census.gov/prod/cen2000/phc-2-39.pdf 9. Clark County Washington, Census Quick Facts, Copyright 2012, Clark County, Washington, http://gis.clark.wa.gov/gishome/census/2000/?pid=quickfacts 4

Calculating Housing Indices Using Tax Lot Data Since we decided not to use the American Community Survey data to gather information on median home values, I m using the tax lot data to calculate the median home values as well as the sales prices. Below is series of tables with my calculations. Range 1 Median Sales Price Area Value Count % Total Total Clark County 222721 21250 0.233365 51975.33 Metro 254456 69809 0.766635 195074.8 Average 247050.2 Range 2 Median Sales Price Area Value Count % Total Total Clark County 256095 10942 0.224585 57515.07 Metro 310224 37779 0.775415 240552.4 Average 298067.4 Range 1 Median Value Area Value Count % Total Total Clark County 156050 75556 0.386077 60247.28 Metro 211686 120146 0.613923 129958.9 Average 190206.2 Range 2 Median Value Area Value Count % Total Total Clark County 246890 97246 0.235052 58032.02 Metro 305287 316475 0.764948 233528.6 Average 291560.7 Affordable Price = 199800 Affordability Index = Affordable Price/Median Sales Price = 0.685278 Change in Median Value (range2 - range1) /range1 = 0.532866 Change in Median Home Price (range2 - range1)/range1 = 0.206506 The Regional Values (RV) can be calculated using the values above. The affordable house price for the average Metro and Clark County income was calculated previously in the Project Summary. 5

Methodology for determining the values used for sales price for the two temporal ranges, 2001-2005 and 2008-2010, of home sales. 1. After selecting records for the Clark County and Metro sales within the temporal ranges, I combined the Clark County and Metro tables based on their temporal range. At the end of this step I had a table of 2001-2005 home sales and a table for 2008-2010 home sales. 2. For each record I calculated the percentage of difference between the sales price of a home and its total value. The formula is (100 * ((Sales Price Total Value)/Total Value). This formula will give values between -100 and 100. 3. I created histograms for the 2 temporal ranges. I had 3 expectations. One, a large percentage of the sales prices would be way below the home value in both datasets. Two, the early range of values would be skewed to the negative end of the graph because sales price was compared against a 2010 valuation. Three, most of the sales would be close to 0%, i.e. a house is more likely to listed and sold if the listing price is around the sales price. 4. I took the total number of records in each temporal range and subtracted the number of records with 0 sales values. The 2001-2005 range had 98,565 non-zero sales price records. The 2008-2010 had 56259. 5. I selected ranges of percentages that netted between 95 and 97% of the records for each temporal range. For 2001 to 2005 that range is -67% to 58% (96%). For 2008 to 2010 that range is -45% to 51% (97%). Below are the histograms for the datasets. 6

Note that the 2001 to 2005 histogram above is skewed towards that negative end of percentages. Also notice that small spike between -84 and -68. The sales indicate that the homes were significantly undervalued. The more current sales histogram shows a more normal bell curve. Both histograms include the zero sales, so they both have a spike around -100. Both graphs contain outliers at the positive end as well as the negative end. 7

TABLE ATTRIBUTES Housing indicators are based on owner occupied single family homes on less than 10 acres. The values are calculated per neighborhood. The data was acquired from tax lot information from Portland Metro and Clark County GIS, therefore any errors in the tax lot data has been inherited by this data. CURVAL - The current (2011 for Metro, 2010 for Clark County) median home value fcurval The number of samples included in the current median home value PREVAL The previous (2001 for Metro, 2000 for Clark County) median home value fpreval The number of samples included in the previous median home value CURPRC The median sales price of homes in the years 2008 to 2010 fcurprc The number of samples included in the 2008-2010 median sales price PREPRC The median sales price of homes in the years 2001 to 2005 fpreprc The number of samples included in the 2001-2005 median sales price VALCHG The change in value between PREVAL and CURVAL. The average change or regional value is 20.7% PRCCHG The change in value between PREPRC and CURPRC. The average change or regional value is 53.3%. AFBLTY The affordability of a home compared with the median sales price. The value is based on a 2006-2010 average salary of $58,115/year using the 2011/2012 lending environment. The affordable home should be at most $199,800. The average value or regional value is 68.5% INFPRC The median sales price of a home, if the home prices kept pace with inflation from 2000 to 2010. INAFF The affordability of a home based on current income and a 2000 median home value adjusted by the inflation rate between 2000-2010. 8

DATA PROCESSING Clark County Tax Lot Information Attributes Used ZIP1 Owner s zip code ZP1 Site Zip Code SHAPE_AREA Parcel size in square feet PT1 Land Use Code (10-19 Residential, 511-518 Residential on Commercial Land) SYEAR The year the property sold SAMOUNT The selling price of the property TOTPROP Total property value Metro Tax Lot Information Attributes Used OWNERADDR Address of the property Owner SITEADDR Address of the property AREA - Parcel size in square feet LANDUSE Land use code (SFR, MFR (Condos)) SALEDATE - The date the property sold SALEPRICE The selling price of the property TOTALVAL Total property value Census Tract Data Used Tract the tract number Neighborhood Data Used The neighborhood shapefile was taken from RLIS. The data was dissolved on the attribute NAME. One name had to be edited prior to the dissolve. The Milwaukie #1 neighborhood was made up of 2 polygons in the RLIS data, but the spellings didn t match, Milwaukie #1 and Milwaukie # 1. Need Per Geography (Tract or Neighborhood), Shapefile Attributes Average Selling Price (2000-2005, 2008-2010) Average total Value (2000, 2010) Affordability Index Difference in Median Price Between 2000 and 2010 Difference in Median Value Between 2000 and 2010 2010 Home Values if they only increased at the inflation rate 9

We have 4 data sets to process, 2000 and 2010 Clark County tax lot shapefiles, and 2001 and 2011 Metro tax lot shapefiles. Please note that the years do not precisely match. 1. Selecting Owner Occupied Single Family Residence on less than 10 acres 1.1. Select the properties that have land use codes for single family homes 1.1.1. Clark County, Select By Attribute on PT1 between 10 and 19 OR between 511 and 518. 1.1.2. Metro, Select By Attribute using PROP_CODE ( 18 codes represent owner occupied SFR) 10 Residential: 101, 102, 121, 122, 131, 132, 14, 15, 151, 191 Tract: 451 Farm: 501, 541, 551 Forest: 601, 641, 651, 681 1.2. Select the properties that are owner occupied 1.2.1. Clark County, Select By Attribute comparing zip codes (ZIP1 = ZP1) and building value is greater than 0 1.2.2. Metro, Select By Attribute comparing addresses (OWNERADDR = SITEADDR) and owner address is not blank 1.3. Select the properties on less than 10 acres. I used the shape area, because the taxed area was not always set. The shape area is in square feet. 1.3.1. Clark County, Select By Attribute, Shape_Area < 43602 1.3.2. Metro, Select By Attribute, AREA < 43602 2. Aggregate on Geometric Unit, results go in an.mdb database 2.1. Find centroids for each Property, Feature To Point 2.2. Join the properties to the geometric unit file, Spatial Join, Intersect, join one to one 3. Summarize Data By Tract, results go in an.mdb database 3.1. Median House Value 3.1.1. Clark County, Summary Statistics, TOTPROP Mean per TRACT 3.1.2. Metro, Summary Statistics, TOTALVAL Mean per Tract Metro sales dates are text values, so a new field needs to be created to convert the sales dates to range values 1. Convert Dates from TEXT to SHORT INT 1.1. Add a field isdate that represents the year ranges, 2001-05, 2008-10 1.2. Calculate Field values are 0, 1, 2 for no range, range 1, range 2 For the current data sets, 2010 for Clark County and 2011 for Metro 2. Summarize the Sales Prices, 2000-2005 and 2008-2010, per geography, results go in an.mdb database 2.1. Make two data sets, one for Sales between 2000 and 2005, the other between 2007 and 2010 2.1.1. Clark County 2.1.1.1. Select By Attribute, SYEAR between 2001 and 2005 AND SAMOUNT > 0 2.1.1.2. Select By Attribute SYEAR between 2008 and 2010 AND SAMOUNT > 0

2.1.2. Metro 2.1.2.1. Select By Attribute isdate = 1 AND SALEPRICE > 0 2.1.2.2. Select By Attribute isdate = 2 AND SALEPRICE > 0 2.2. Eliminate Outliers 2.2.1. For each dataset calculate the percent of home sales price to value (100 * ((Price Value)/Value)) 2.2.2. For the datasets of sales between 2001 and 2005, select records with a percentage between -67% and 58% (96% of records) 2.2.3. For the datasets of sales between 2008 and 2010, select records with a percentage between -45% and 51% (97% of records) 2.3. Summarize Data per Tract 2.3.1. Clark County, Summary Statistics, SAMOUNT Mean per Tract 2.3.2. Metro, Summary Statistics, SALEPRICE Mean per Tract Merge the tables with the Geometric Units Shapefile 1. If the geometric units are census tracts, remove the demographic data from the shapefile 2. Combine the Clark County summary tables with the Metro tables 2.1. Merge Clark County 2010 median value with Metro 2011 median value 2.2. Merge Clark County 2000 median value with Metro 2001 median value 2.3. Merge Clark County 2010 range 1 average price with Metro 2011 range 1 average price 2.4. Merge Clark County 2010 range 2 average price with Metro 2011 range 2 average price 3. Join the statistic field with its frequency into the shapefile For each summary statistic the frequency is also added. The frequency is needed to determine if the numbers of samples are sufficient to determine a trend. The previous Equity Atlas required at least 10 samples per geometric unit. 4. Calculate Affordability, Percent Change Median Value, Sales Prices differences per tract. 11