The URL can be used to link to this page

Home My WebLink AboutResolution 09-54 12/07/2009 RESOLUTION NO. 09-54 A RESOLUTION OF THE COMMON COUNCIL OF THE CITY OF SPRINGFIELD ADOPTING THE 2009 PRELIMINARY SPRINGFIELD RESIDENTIAL LAND AND HOUSING NEEDS ANALYSIS, FULFILLING ITS STATUTORY OBLIGATION TO "COMPLETE" THE PRELIMINARY INVENTORY, ANALYSIS AND DETERMINATION BEFORE JANUARY l, 2010. WHEREAS, in 2007 the Oregon Legislature passed and the Governor signed into law 2007 Or Laws Chapter 650, codified as ORS 197.304 and commonly known as "House Bill 3337; and WHEREAS, HB 3337, as codified, provides as follows: 197.304 Lane County accommodation of needed housing. (1) Notwithstanding an intergovernmental agreement pursuant to ORS 190.003 to 190.130 or acknowledged comprehensive plan provisions to the contrary, a city within Lane County that has a population of 50,000 or more within its boundaries shall meet its obligation under ORS 197.295 to 197.3l4 separately from any other city within Lane County. The city shall, separately from any other city: (a) Establish an urban growth boundary, consistent with the jurisdictional area of responsibility specified in the acknowledged comprehensive plan; and (b) Demonstrate, as required by ORS 197.296, that its comprehensive plan provides sufficient buildable lands within an urban growth boundary established pursuant to statewide planning goals to accommodate estimated housing needs for 20 years. (2) Except as provided in subsection (l) of this section, this section does not alter or affect an intergovernmental agreement pursuant to ORS 190.003 to 190.130 or acknowledged comprehensive plan provisions adopted by Lane County or local governments in Lane County. [2007 c.650 S2] Note: Section 3, chapter 650, Oregon Laws 2007, provides: Sec. 3. A local government that is subject to section 2 ofthis 2007 Act [l97.304] shall complete the inventory, analysis and determination required under ORS 197.296 (3) to begin compliance with section 2 of this 2007 Act within two years after the effective date of this 2007 Act [January l, 2008],[2007 c.650 S3]; and WHEREAS, the "inventory, analysis, and determination required under ORS 197.296(3)" is a preliminary determination of capacity; and WHEREAS, the inventory, analysis, and determination required by ORS 197.296(3) is based upon recent trends; and WHEREAS, the remaining steps required by HB 3337 and ORS 196.296 and state land use goals require consideration of a variety of legal, policy, and factual issues before adoption of a final inventory, analysis, and determination of capacity; and WHEREAS, formal adoption ofthe preliminary inventory, analysis, and determination" by a resolution recognizing the nonfinal nature of this preliminary step meets the express requirement ofHB 3337 that the city "complete" this step by January l, 20l0;and WHEREAS, the Springfield City Council had previously directed staff to begin a 20-year residential lands study pursuant to ORS 197.296 in December, 2005; and WHEREAS, to complete the preliminary inventory, analysis, and determination required by HB 3337, the City of Springfield commissioned ECONorthwest to prepare a Residential Land and Housing Needs Analysis outlining Springfield's housing needs for the next 20 years; and WHEREAS, Springfield has conducted the Residential Lands Study planning process to date in a manner consistent with Statewide Planning Goals 1 and 2, and evidence of the citizen involvement and intergovernmental coordination processes thus far is fully documented in the public record: application file number LRP2007-00030; and WHEREAS, timely and sufficient notice of the public hearing, pursuant to Springfield Development Code Section 5.2-115, has been provided; and WHEREAS, on October 20,2009, a public hearing on the Springfield Residential Land and Housing Needs Analysis was held before the City of Springfield Planning Commission and the Development Services Department staff report, the oral testimony, letters received, written submittals of the persons testifying at the hearing, and the public record for file # LRP2007-00030 have been considered and hereby are incorporated into the record for this proceeding. WHEREAS, on October 20,2009, the City of Springfield Planning Commission made a recommendation to the City Council to approve the determination set forth in the Springfield Residential Land and Housing Needs Analysis; and WHEREAS, the Springfield Residential Land and Housing Needs Analysis is consistent with ORS 197.296(3) as described in the attached staff report; and WHEREAS, adoption of the initial stage housing needs determination does not include adoption or amendment of an urban growth boundary or the adoption or amendment of any comprehensive plan policies or designations; and WHEREAS, adoption ofthe final 2030 Springfield Residential Land and Housing Needs Analysis will occur when the Springfield City Council and the Lane County Board of Commissioners adopt the Springfield 2030 Refinement Plan, a refinement plan of the Eugene-Springfield Metro Plan. RESOLUTION NO. 09-54 NOW THEREFORE BE IT RESOLVED, that the Common Council of the City of Springfield hereby declares its intention as follows: Section 1: The Common Council of the City of Springfield provisionally adopts, subject to further public input, refinement, correction, and revision, pending completion ofthe HB 3337 process, the determinations set forth in the Springfield Residential Land and Housing Needs Analysis, as submitted and revised in the course of these proceedings presented herein at (Case No. LRP 2007-00030), and attached hereto as Exhibit "A." (Revision date December l, 2009). ADOPTED by the Common Council and approved by the Mayor of the City of Springfield, Oregon, this ~ th day of Decembe ,2Q09. by te of 6 for and 0 against. ATTEST~)ntnL- REVIEWED & APPROVED Affi~ DATE: I OFFICE OF CITY ATTORNEY RESOLUTION NO. 09-54 Exhibit A-1 . Springfield Residential Land and Housing Needs Analysis Prepared for . City of Springfield by ECONorthwest 99 W. Tenth, Suite 400 Eugene, OR 97401 (541) 687-0051 Draft Report December 2009 . Attachment 1-1 Exhibit A-2 . . Written by: Robert Parker, Project Director Beth Goodman, Project Manager Whit Perkins, Research Assistant . Date submitted: December 2009 ECO Project Number 7139 ECONorthwest 99 W Tenth, Suite 400 Eugene, OR 97401 (541) 687-0051 . Attachment 1-2 Exhibit A-3 . Table of Contents . Page EXECUTIVE SU M MARy.......................................................................................................................... I CHAPTER 1: INTRODUCTION ................................................................................................................ 1 BACKG ROU N D ................................................................................................................................... 1 PURPOSE ........................................................................................................................................... 2 ORGAN IZA TI ON ................................................................................................................................. 3 CHAPTER 2: FRAMEWORK FOR A HOUSING NEEDS ANALySiS.............................................................. 4 OREGON HOUSI N G POll CY ................................................................................................................ 4 CHAPTER 3: RESIDENTIAL LAND INVENTORY9 METHODS, DEFINITIONS, AND ASSUMPTIONS ................................................................................... 9 RESUL TS........................................................................................................................................... 12 La nd base..................................... ............. .... ...................................................................... 12 Vaca nt bu ilda ble la nd .................................................. ......................................... ............... l5 Redevelopment potentia I.......................... ................ ................. .............................. ............ 18 Residential capacity................... ........................................ ............................ ...................... 18 CHAPTER 4: HISTORICAL DEVELOPMENT TRENDS............................................................................... 20 RESI DENTIALDEVELOPM ENT TREN OS .............................................................................................. 20 TRENDS IN HOUSING MIX AND TENURE ........................................................................................... 23 DENSITY. ......................... .'......... ........................... .............................. ......................:................. ...... 2S CHAPTER S: HOUSING DEMAND AND NEED ....................................................................................... 27 STEP 1: PROJECT NUMBER OF NEW HOUSING UNITS NEEDED IN THE NEXT 20 YEARS .................. 27 Popu lation .................. ..................... ................................ ................................................... 27 Persons in grou p quarters ...................................... .................................... .......................... 28 Household size a nd com position ................. ......... ................ .................... ........ .................... 29 Vacancy rate................ .................................................. .................... ...................... ........... 29 STEP 2: IDENTIFY RELEVANT NATIONAL, STATE, AND LOCAL DEMOGRAPHIC AND ECONOMIC TRENDS AND FACTORS THAT MAY AFFECT THE 20-YEAR PROJECTION OF STRUCTURE TYPE MIX.. 31 . Nationa I Housing Trends .............................. .................. ...................................................... 31 STEP 3: DESCRIBE THE DEMOGRAPHIC CHARACTERISTICS OF THE POPULATION AND, IF POSSIBLE, HOUSING TRENDS THAT RELATE TO DEMAND FOR DIFFERENT TYPES OF HOUSING ..................... 34 Attachment 1-3 Exhibit A-4 . Summary of key demographic and housing trends................................................................. 45 Implications of demographic and housing trends for housing need......................................... 47 Step 4: Determine the types of housing that are likely to be affordable to the projected popu lation based on household income ............................................... ..................... ............ 49 Step 5: Estimate the number of additional needed units by structure type and tenure ............ 58 Step 6: Determine the needed density range for each plan designation and the average needed net density for a II designations.......................... ................. .................. ........................... ..... 59 CHAPTER 6: COMPARISON OF SUPPLY AND DEMAND ........................................................................ 63 TOTAL RESIDENTIAL LAN 0 NEED, 2010-2030 .................................................................................... 63 land needed for new residential dwelling units ..................................................................... 63 land needed for other uses............................................ ................................ ...................... 63 BUILDABLE LAND INVENTORY AND CAPACITY ...............................................:.................................. 65 COM P ARISON AN 0 CONCLUSiONS................................................................................................... 66 APPENDIX A: CONTEXT FOR ASSESSING HOUSING NEEDS.................................................................. 68 WHAT IS AFFORDABLE HOUSI NG? .................................................................................................... 68 WHAT OBJECTIVES DO HOUSING POLICIES TYPICALLY TRY TO ACHIEVE? .......................................... 69 DEMAN 0 VERSUS NEED ................................................................................................................... 70 APPEN DIX B: NATIONAL H OUSI NG TREN OS ........................................................................................ 74 . . Attachment 1-4 . . . Exhibit A-5 Executive Summary The 2007 Oregon Legislature passed HB 3337 which requires Springfield to establish a separate urban growth boundary (UGB). In response to HB 3337, the City is conducting this study to evaluate the sufficiency of land available for residential uses in its UGB. To make this determination, the draft Residential Lands Study (RLS) presents a housing needs analysis consistent with requirements ofHB 3337, Goall4, ORS 197.296, and OAR 660-008. The Springfield Residential Lands Study is intended to provide technical analysis required to determine the 20-year need for residential land for Springfield's jurisdictional share of the area subject to the Eugene-Springfield Metropolitan Area, i.e., the area east ofInterstate 5, and whether the city has enough capacity within the area east ofI-5 inside the current regional UGB to meet that need. The Executive Summary provides key findings from the Springfield Residential Lands Study. The purpose of the Residential Study is to (1) present growth forecasts, (2) inventory how much buildable residential land the City has, (3) identify housing needs, (4) identify land needed for housing and other uses, and (5) determine how much land the City will need to accommodate growth between 2010 to 2030. How MUCH GROWTH IS SPRINGFIELD PLANNING FOR? Population forecasts provide the foundation for assessing land needs. Springfield must have a population forecast to project expected population change, over the 20-year planning period (in this instance, 2010-2030). Lane County adopted coordinated population forecasts for the County and its incorporated cities in June 2009. The forecasts include figures for Springfield for 2030 and 2035. ' Table S-1 shows the coordinated population forecast for the area within the current Springfield city limits, the current unincorporated urban area (the area between the city limit and UGB), and within Springfield's jurisdictional share f the current Metro Plan UGB for 2010 to 2030. The Springfield UGB forecast for 2030 is 81,608 persons-an increase of 14,577 persons during the 20-year planning period. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page i Attachment 1-5 Exhibit A-6 Table S-1. Springfield coordinated population . forecast, Springfield UGB, 2010 to 2030 Urban Year City Limit Area UGB 2010 58,891 8,140 67,031 2030 74,814 6,794 81,608 Change 2010-2030 Number 15,923 (1,346) 14,577 Percent 27% -17% 22% AAGR 1,2% -0,9% 1.0% Source: Lane County Rural Comprehensive Plan, 1984 (Amended in 2009), Table 1-1, pg 5 How MUCH BUILDABLE RESIDENTIAL LAND DOES SPRINGFIELD CURRENTLY HAVE? Springfield has 2,485 acres in tax lots that are designated for residential uses. Of these, about 1,447 acres within the Urban Growth Boundary (UGB) are considered vacant and buildable. Table S-2 shows vacant land by plan designation. Table S-2. Vacant residential land by plan designation, Springfield UGB, 2008 . Total Acres Developed Constrained Buildable Plan Designation Tax Lots in Tax Lots Acres Acres Acres Low Density Residential 981 2,137 71 765 1,301 Medium Density Residentia 126 329 142 58 128 High Density Residential 8 19 1 0 18 Total 1,115 2,485 214 824 1,447 Source: City of Springfield GIS data; analysis by ECONorthwest The purpose of the residential buildable lands inventory is to estimate the capacity of buildable land in dwelling units. The capacity of residential land is measured in dwelling units and is dependent on densities allowed in specific zones as well as redevelopment potential. In short, land capacity is a function of buildable land and density. The buildable lands inventory indicates that Springfield has about 1,447 acres of vacant and partially-vacant residential land and an additional 21 acres in the Glenwood mixed-use refinement plan area (these acres were included in the commercial and industrial lands inventory and are included here only for the . Page ii ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-6 Exhibit A-7 . purpose of estimating residential capacity).l This yields a total of 1,468 buildable acres. Table S-3 provides an estimate of how much housing could be accommodated by those lands based on needed densities after making deductions for development constraints. It includes capacity for areas with approved master plans that were not included in the acreage estimates. This includes Marcola Meadows (518 dwellings in the MDR designation) and RiverBend (730 dwellings in the MDR designation). Additionally, the housing needs analysis assumes that 5% of new housing (299 dwelling units) will be a result of redevelopment. Table S-3 shows that Springfield has capacity for 9,021 dwelling units within the existing UGB. Table S-3. Estimated residential development capacity, Springfield UGB, 2009 Plan Designation Low Density Residential Medium Density Residential High Density Residential Mixed-Use (Glenwood) Rede\.elopment Total Buildable Acres 824 95 16 21 Residential Capacity (OU) 5,379 2,718 355 270 299 9,021 Percent of Capacity 60% 30% 4% 3% 3% 100% na 956 . Source: City of Springfield residential BLI; analysis by ECONorthwest Note: Estimated residential development capacity includes sites with approved master plans (RiverBend - 730 DU and Marcola Meadows - 518 DU, All of this capacity is in the Medium Density Residential plan designation), How MUCH HOUSING WILL THE CITY NEED? Springfield will need to provide about 5,980 new dwelling units to accommodate growth between 2010 and 2030. About 3,588 dwelling units (60%) will be single-family types, which includes single-family detached, manufactured dwellings, and single-family attached housing. About 2,392 units (40%) will be multi-family housing. How MUCH LAND WILL BE REQUIRED FOR HOUSING? Table S-4 shows the capacity for residential development by plan designation both before and after subtracting acreage needed for other uses, such as parks, schools, churches, etc.). ECO estimates Springfield will need 463 acres for other uses during the 2010-2030 period. The results lead to the following findings: 1 Capacity in the Glenwood mixed-use area was calculated as follows: 21 buildable acres (45% of the 47-acre site; the policy requires 30% to 60% of the site be used for housing) multiplied by 15 dwelling units per gross acre equals 317 dwelling units, minus 47 dwelling units that would be displaced'from the River Bank Mobile Home Park equals 270 dwelling units. . DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page iii Attachment 1-7 Exhibit A-8 . . Springfield has an overall surplus of residential land. The Springfield UGB has enough land for 9,021 new dwelling units including redevelopment capacity without taking into account the need for 463 acres of this land for other uses. The housing needs forecast projects a need for 5,980 dwelling units and 145 group quarter dwellings. . The Low Density Residential designation has a surplus of approximately 72 gross acres. · The Medium Density Residential designation has a surplus of approximately 18 gross acres. · The High Density Residential designation has a deficit of approximately 34 gross acres. · The total residential land surplus is 59 gross acres. Table 6-4. Residential capacity for needed dwelling units by plan designation, Springfield UGB, 2010-2030 1 2 3 4 5 6 7 8 ' 9 Housing Housing Total land Surplusl Surplusl Surplusl Needed Need Deficit Other Deficit Need Capacity Deficit Density (Gross (Gross Residential (Gross Plan Designation (DU) (DU) (DU) (DUlGRA) Acres) Ac) land Need Ac) . Low Density Residential 3,468 5,379 1,911 5 -422 422 347 75 Medium Density Residential 1,794 3,137 1,343 12 0 111 93 18 High Density Residential 718 505 -213 20 11 -11 23 -34 Total 5,980 9,021 3,041 0 411 522 463 59 Source: ECONorthwest Column Notes: 1, Plan designations 2, Needed dwellings by plan designation (table 5-30) 3, Capacity by plan designation (table 6-2); Note: MDR capacity includes capacity in master planned areas (Glenwood, Marcola Meadows, Riverbend); redevelopment capacity is included in MDR (150 DU) and HDR (150 DU) 4, Capacity (column 3) minus Need (column 2); Note: a positive number denotes enough capacity within the existing UGB 5, Needed Gross Density (from bottom of page 5) 6, Total additional land needed (if a deficit exists), Equals -column 4 divided by column 5 7, Surplus/deficit gross acres, Equals Column 4 divided by Column 5 8, Other residential land need (land needed for parks, etc) 9, Total surplus/deficit. Equals column 7 minus column 8, Note: Total Surplus/Deficit (column 9) adds to 344 acres due to rounding errors, . Page iv ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-8 . . . Exhibit A-9 Chapter 1 Introduction This report presents a housing needs analysis for the City of Springfield. The primary purpose of this report is to address the requirement ofRB. 3337 that Springfield "demonstrate, as required by ORS 197.296, that its comprehensive plan provides sufficient 'buildable lands within an urban growth boundary established pursuant to statewide planning goals to accommodate estimated housing needs for 20 years." The study is intended to comply with statewide planning policies that govern housing, including Goal 10 (Housing), ORS 197.296, and OAR 660 Division 8. The primary goals of this study are to (1) project the amount ofland needed to accommodate the city's future housing needs of all types, and (2) evaluate the existing residential land supply within the Springfield Urban Growth Boundary to determine if it is adequate to meet that need. The methods used for this study generally follow the Planning for Residential Growth guidebook, published by the Oregon Transportation and Growth Management Program (1996). BACKGROUND The City of Springfield has not conducted a housing needs analysis since the Eugene-Springfield Residential Lands and Housing Study was completed in 1999. In the six years since the study was completed, Springfield's population has increased by nearly 3,000 residents, an increase of more than 5% over the six-year period. In 2007, the Oregon State Legislature passed House Bill 3337 which requires Springfield to: (a) Establish an urban growth boundary, consistent with the jurisdictional area of responsibility specified in the acknowledged comprehensive plan; and (b) Demonstrate, as required by ORS 197.296, that its comprehensive plan provides sufficient buildable lands within an urban growth boundary established pursuant to statewide planning goals to accommodate estimated housing needs for 20 years. The analysis and determination of land sufficiency required under section (b) must be completed by December 31,2009. This study is intended to meet the requirements of section (b) by determining whether the City has sufficient land within the Springfield Urban Growth Boundary (UGB) to accommodate expected future housing needs. To make this determination, this report presents a housing needs analysis consistent with requirements of Goall4, ORS 197.296, and OAR 660-008. As required by HB 3337, the City intends to "complete the inventory, analysis and determination required under ORS 197.296(3)" before the end of 2009, and to complete the remainder of its obligations under HB 3337 and ORS DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 1 Attachment 1-9 . . . Exhibit A-10 197.296 early in 2010. Consistent with the requirements ofORS 197.296(2) the planning period for this study is 2010-2030. PURPOSE The purpose of this study is to provide an assessment of residential development capacity and demand for residential land. The study will serve two purposes: (1) to inform policy makers about planning options and (2) to fulfill state planning requirements for a twenty-year supply of residential land. Consistent with the requirements of ORS 197.296, communities engaged in a buildable lands analysis and housing need assessment must complete, in part, the following: . Inventory the supply of buildable lands within the current urban growth boundary; . Determine the actual density and the actual mix of housing types of residential development that have occurred within the urban growth boundary since the last periodic review or five years, whichever is greater. Development activity used for this review was between 1999 and June 2008.2 . Conduct an analysis of housing need by type and density range, in accordance with ORS 197.303 and statewide planning goals and rules related to housing, to determine the amount of land needed for each needed housing type for the next 20 years (2010-2030). This report presents an analysis consistent with the above outlined requirements, and draws upon previous work that ECON orthwest for a number of Oregon cities and regions. The report is intended to serve as the basis for subsequent discussions and policy choices regarding the management of growth in Springfield and to enable the city to complete the residential lands inventory, analysis and determination required by ORS 197.296(3) and Section 3 of 2007 Or Laws Chapter 650 (HB 3337). It does not address land use efficiency measures as required by ORS 197.296 and OAR 660-024. Land use efficiency measures will be addressed through a separate process. In general, a housing needs analysis contains a supply analysis (existing housing, planned housing, and buildable land) and a demand analysis (population and employment growth leading to demand for more built space: housing by type and density). The geographic scope of the housing needs analysis is all land inside the current acknowledged Eugene-Springfield Metropolitan Urban Growth Boundary east of Interstate 5. 2 The City uses the 1999-2006 period for analysis due to limited availability of penn it data that can be cross-referenced to tax lot data to develop density estimates. Moreover, the 1990 and 2000 Census provides an accurate source for analysis of housing mix trends during the 1990s. Page 2 December 2009 DRAFT: Springfield Housing Needs Analysis ECONorthwest Attachment 1-10 . . . Exhibit A-11 ORGANIZATION The rest of this report is organized as follows: . Chapter 2, Framework For A Housing Needs Analysis, describes the theoretical and policy underpinnings of conducting a Goal lO housing needs analysis for Oregon cities. . Chapter 3, Residential Land Inventory, describes the supply of residential land available to meet the 20-year need for housing. · Chapter 4, Historical Development Trends, summarizes building permit and subdivision data to evaluate residential development by density and mix for the period beginning September I, 1988, through June 30, 2000. . Chapter 5, Housing Needs Analysis, presents a housing needs analysis consistent with HB 2709 requirements and the HB 2709 Workbook. · Chapter 6, Comparison of Supply and Need, compares buildable land supply with estimated housing need. The report also includes two appendices: · Appendix A, Context for Assessing Housing Needs provides an overview of planning for housing and typical local policy objectives related to affordable housing. . Appendix B, National and Regional Housing Trends presents research ECO has performed over the course of several years describing key factors affecting housing at the national and regional level. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 3 Attachment 1-11 Exhibit A-12 Chapter 2 Framework for a Housing Needs Analysis3 . Economists view housing as a bundle of services for which people are willing to pay: shelter certainly, but also proximity to other attractions Gob, shopping, recreation), amenity (type and quality of fixtures and appliances, landscaping, views), prestige, and access to public services (quality of schools). Because it is impossible to maximize all these services and simultaneously minimize costs, households must, and do, make tradeoffs. What they can get for their money is influenced by both economic forces and government policy. Moreover, different households will value what they can get differently. They will have different preferences, which in turn are a function of many factors like income, age of household head, number of people and children in the household, number of workers and job locations, number of automobiles, and so on. Thus, housing choices of individual households are influenced in complex ways by dozens of factors; and the housing market in Lane County and ' Springfield are the result of the individual decisions of thousands of households. These points help to underscore the complexity of projecting what types of housing will be built between 2010 and 2030. . The complexity of a housing market is a reality, but it does not obviate the need for some type of forecast of future housing demand and need, and its implications for land demand and consumption. Such forecasts are inherently uncertain. Their usefulness for public policy often derives more from the explanation of their underlying assumptions about the dynamics of markets and policies than from the specific estimates of future demand and need. Thus, we start our housing analysis with a framework for thinking about housing and residential markets, and how public policy affects those markets. OREGON HOUSING POLICY The passage of the Oregon Land Use Planning Act of 1974 (ORS Chapter 197), established the Land Conservation and Development Commission (LCDC), and the Department of Land Conservation and Development (DLCD). The Act required the Commission to develop and adopt a set of statewide planning goals. Goal 10 addresses housing in Oregon and provides guidelines for local governments to follow in developing their local comprehensive land use plans and implementing policies. At a minimum, local housing policies must meet the requirements of Goal 10 (ORS 197.295 to 197.3l4, ORS 197.475 to 197.490, and OAR 600-008). Goal 10 requires incorporated cities to complete an inventory of buildable residential lands . 3 This chapter is based on studies ECONorthwest has completed for other Oregon cities and regions. Page 4 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-12 . . . Exhibit A-13 and to encourage the availability of adequate numbers of housing units in price and rent ranges commensurate with the financial capabilities of its households. Goal 10 defines needed housing types as "housing types determined to meet the need shown for housing within an urban growth boundary at particular price ranges and rent levels." ORS 197.303 defines needed housing types: (a) Housing that includes, but is not limited to, attached and detached single- family housing and multiple family housing for both owner and renter occupancy; (b) Government assisted housing;' (c) Mobile home or manufactured dwelling parks as provided in ORS 197.475 to 197.490; and (d) Manufactured homes on individual lots planned and zoned for single- family residential use that are in addition to lots within designated manufactured dwelling subdivisions. ORS 197.296 defines factors to establish sufficiency of buildable lands within urban growth boundary and requires analysis and determination of residential housing patterns. It applies to cities with populations of25,000 or more and requires cities to: . Demonstrate that its comprehensive plan or regional plan provides sufficient buildable lands within the urban growth boundary established pursuant to statewide planning goals to accommodate estimated housing needs for 20 years (ORS 197.296(2)); · Inventory the supply of buildable lands within the urban growth boundary and determine the housing capacity of the buildable lands (ORS 197.296(3)(a)); and · Conduct an analysis of housing need by type and density range to determine the number of units and amount of land needed for each needed housing type for the next 20 years (197 .296(3)(b )). ORS 197.296 also defines a process for cities to following when considering UGB expansions to meet identified residential needs. ORS 197.296(6) requires cities to take one or more of the following actions if the housing need is greater than the housing capacity to accommodate the additional housing need: a. Amend its urban growth boundary to include sufficient buildable lands to accommodate housing needs for the next 20 years. As part of this process, 'Government assisted housing can be any housing type listed in ORS 197.303 (a), (c), or (d). DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 5 Attachment 1-13 . . . Exhibit A-14 the local government must consider the effects of "land use efficiency measures." The amendment must include sufficient land reasonably necessary to accommodate the siting of new public school facilities; b. Amend its comprehensive plan, regional plan, functional plan or land use regulations to include new measures that demonstrably increase the likelihood that residential development will occur at densities sufficient to accommodate housing needs for the next 20 years without expansion of the urban growth boundary; or c. Adopt a combination of the actions described in paragraphs (a) and (b) of this subsection. ORS 197.296 is also explicit about what must be considered in a housing needs analysis and the buildable lands inventory. For the purpose of the inventory, "buildable lands" includes: (A) Vacant lands planned or zoned for residential use; (B) Partially vacant lands planned or zoned for residential use; (C) Lands that may be used for a mix of residential and employment uses under the existing planning or zoning; and (D) Lands that may be used for residential infill or redevelopment. To visually display the buildable lands inventory, the inventory includes a map that identifies lands that are vacant, partially vacant, or designated for mixed- use development. The needs analysis includes an analysis of historical housing density and mix. This analysis, which must include data in the last periodic review or five years, whichever is greater.5 (A) The number, density and average mix of housing types of urban residential development that have actually occurred; (B) Trends in density and average mix of housing types of urban residential development; (C) Demographic and population trends; (D) Economic trends and cycles; and 5 A local government can make a determination to use a shorter time period than the time period described if the local government finds that the shorter time period will provide more accurate and reliable data related to housing capacity and need, The shorter time period may not be less than three years, Page 6 DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Attachment 1-14 . . . Exhibit A-15 (E) The number, density and average mix of housing types that have occurred on the buildable lands. Figure 2-1 provides a graphic representation of the housing needs analysis process as defined in ORS 197.296. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 7 Attachment 1-15 . . Exhibit A. Figure 2-1. Process for determining the sufficiency of residential lands I Land Supply I I Land Demand I I Land SupplylDemand Analysis II Actions Related to UGB No UGB expansion required, .. I Yes I I Priority 1 Amend plan/regulations to ... Does UGB contain include new measures that enough buildable land increase likelihood that Conduct housing needs analysis, needed at actual residential densities will occur at Inventory supply of ORS 197,296(3)(c) and ORS 197,296(7) residential densities? densities sufficient to ORS 197,296(4) accommodate housing needs buildable' residential lands .. for the next 20 years without within the UGB:2 Use population forecast from . , expansion of the UGB, _Determine 20-year supply coordinating body. ORS of buildable lands for 195,036 housing, Priority 2 ORS 197,296(2) and I I Adopt a combination of Priorities 197,296(3)(a) Is needed density the same as or less than No 1 and 3, II actual density? Is needed mix the same as actual mix? ORS 197.296(5) Determine actual I Priority 3 density/mix of housing Yes to I No I Amend the UGB to include ORS 197,296(3)(b) both y sufficient buildable lands to .. Identify and evaluate ~ Yes I accommodate housing needs,. measures to increase Do the measures for Take No measures likelihood that needed needed housing forego ~ one of for housing residential development the need to expand the several needed, will occur, UGB? l No actions: ORS 197.296(6) and (7) Footnotes: 1 Buildable lands means vacant and redevelop-able lands in urban and urbanizable areas that are suitable, available and necessary for residential uses, ORS 197.295(2) 2 Goal 14 requires UGB amendments to be adopted by City and County County, OAR 660-015-0000(14) Page 8 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-16 . . . Exhibit A-17 Chapter 3 Residential Land Inventory The residential lands inventory is intended to identify lands that are available for development within the UGE. The inventory is sometimes characterized as supply of land to accommodate growth. Population and employment growth drive demand for land. The amount of land needed depends on the density of development. This chapter presents the residential buildable lands inventory for the City of Springfield. 6 The results are based on analysis of Geographic Information System data provided by City of Springfield GIS and Lane County Assessment data. The analysis also used aerial orthophotographs for verification. METHODS, DEFINITIONS, AND ASSUMPTIONS The first step of the residential buildable lands inventory was to identify the "land base." The land base includes all lands in the Springfield portion of the Metro UGB that are either fully or partially within a residential plan designation. The following plan designations were included in the residential land base: . High Density Residential . Medium Density Residential . Low Density Residential The foundational assumptions for the residential lands inventory were reviewed and discussed by the ResidentialLands Stakeholder Committee. The committee recommended a package of definitions and assumptions for use in the residential land inventory. These were reviewed with the Planning Commission and Council and approved for use in the study. The draft acreages presented in this chapter utilize the definitions and assumptions and also incorporate more detailed information from the Lane County Assessor's Office to determine the character of the parcels. Property Class and Stat Class codes from the Lane County Assessor's Office were used to help determine if a property is vacant and what type of structure (if any) is present on the land. Property Class is a three digit code to define the current use of the land (residential, commercial, industrial, multi-family, etc) and whether is vacant or developed. Stat Class is also a three digit code used by the Assessor's Office to describe the type of structure on a parcel (single-family home, multi-family structure, agricultural outbuilding, etc.). Aerial Photos were 6 The residential buildable lands inventory was a collaborative effort between City of Springfield staff and ECON orthwest. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 9 Attachment 1-17 Exhibit A-18 . also used in some cases to help determine presence and extent of development on a site if other information was not clear. A key step in the buildable lands analysis was to classify each tax lot into a set of mutually exclusive categories. All tax lots in the UGB are classified into one of the following categories: . Vacant and Partially Vacant Land. This category includes parcels with no structures or with structures with a value of less than $l 0,000; parcels have not been precluded from development by a CUP or other commitment. . Unbuildable, Not Serviceable Land. This category includes land that is undevelopable. It includes tax lots or areas within tax lots with one or more ofthe following attributes: (1) slopes greater than 25%; (2) within the floodway; (3) in areas with severe landslide potential (DOGAMI map); (4) within wetlands and riparian corridors and setbacks; (5) with an easement a 230KV transmission line; (6) small irregularly shaped lots; and (7) publicly owned land. · Developed land. Land that is developed at densities consistent with zoning and improvements that make it unlikely to redevelop during the analysis period. Lands not classified as vacant, partially-vacant, or undevelopable are considered qeveloped. . · Potentially redevelopable land. Land on which development has already occurred but on which, due to present or expected market forces, there exists the potential that existing development will be converted to more intensive uses during the planning period. Redevelopable land is a subset of developed land and includes lands in MDR and HDR plan designations that have single-family dwellings. The initial classifications, while not perfect, provided a starting point. The initial classification was used to help City staff to define a list of parcels that meet the assumptions and criteria in the definitions listed below. The next step in the process was verification. City staff and ECONorthwest spent considerable effort to review and verify land classifications. Verification steps included review of classifications on top of 2008 aerial photographs, cross referencing data with LCOG land use data, and in selected instances, field verification. The land classifications result in identification of lands that are vacant or partially vacant. The inventory includes all lands within the Springfield UGH. Public and semi-public lands are generally considered unavailable for development. Map 3-l shows residential lands by plan designation within the Springfield UGB. . Page 10 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-18 . ~I . ~ 3: w ~ CJ 30TH ~ . \~ -Resioenfial-Lan a-fiy Plan Designation City of Springfield Oregon Legend D City Limits D Urban Growth Boundary Plan Designation _ High Density Residential I I Low Density Residential Medium Density Res Mixed Medium Density Residential N . . . . Exhibit A-20 RESULTS LAND BASE The first step in the residential land inventory was to determine the land base. This step was necessary because the inventory only covers a subset of land in the . Springfield UGB. The land base is the subset of tax lots that fall within the plan designations included in the residential portion of the inventory. Table 3-1 shows acres within the Springfield UGB and city limits in 2008. According to the City GIS data, Springfield has about] 4,603 acres within its UGB. Of the 14,603 acres, 12,139 acres (about 83%) are in tax lots. Land not in tax lots is primarily in streets and waterways. Springfield has about 9,958 acres within its City Limits;/of th~se 8,060 acres (about 8l % of total acres in the City Limit) are in tax!~ Additfonally, the City has about 4,645 acres between the City Limits and Urban Growth Boundary (the UGA); of this about 4,079 acres are in tax lots. Table 3-1. Acres in Springfield UGB and City Limit, 2008 Area Percent Total Acres in in Tax Tax Lots Acres Tax Lots Lots City Limits 19,477 9,958 8,060 81% Urban Growth Area 3,150 4,645 4,079 88% Total 22,627 14,603 12,139 83% Source: City of Springfield GIS data; analysis by ECONorthwest Note: Urban Growth Area is the unincorporated area between the City Limits and Urban Growth Boundary Table 3-1 summarizes all land in the Springfield UGB. The next step is to identify residential land base (e.g., lands with plan designations that allow housing or "residential lands"). The land base includes traditional residential designations, as well as mixed-use designations Note that not all of the land in mixed-use designations will be used for employment. Table 3-2 shows that about 7,482 acres within the Springfield UGB is included in the residential land base. Thus, about 62% of land within the Springfield UGB is included in the residential land base. The database includes all land in tax lots that have any portion that is in a residential plan designation. Page 12 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-20 . . . Exhibit A-21 Table 3-2. Lands designated for residential uses, Springfield UGB, 2008 Area Value Springfield UGB Number of Tax Lots Acres in Tax Lots Springfield CIBL Tax Lots in Residential Designations Acres in Land Base in Residential Designations Source: analysis by ECONorthwest 22,627 12,139 20,159 7,482 Table 3-3 shows residential acres by classification and constraint status for the Springfield UGB in 2009. Analysis by constraint status (the table columns) shows that about 4,832 acres are classified as built or committed (e.g., unavailable for development), l,203 acres were classified as constrained, and 1,447 were classified as vacant buildable. Table 3-3. Residential acres by classification, Springfield UGB, 2009 Land not avialable for Land available housing for housing Developed Constrained Capacity Classification Tax Lots Total Ac Ac Ac Buildable Ac (DU) Land with no development capacity De-.eloped 18,745 4,408 4,124 284 0 0 Park/School 96 335 314 21 0 0 Public 58 79 35 44 0 0 Right of Way 145 175 145 30 0 0 Subtotal 19,044 4,997 4,618 379 0 0 Land with development capacity Master Planned 18 151 138 13 See notes 1,248 Partially Vacant 234 841 77 170 595 3,206 Vacant 863 1 ,493 0 641 852 4,039 Subtotal 1,115 2,485 214 824 1,447 8,493 Total 20,159 7,482 4,832 1,202 1,447 8,493 Source: City of Springfield data; analysis by ECONorthwest Note: No buildable acres are shown for master planned areas because the master plan identifies the number of dwelling units, This capacity is reflected in Table 3-7, DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 13 Attachment 1-21 . \\ Cl~ a:: ::J lD o t) -g o . 30TH !) . Residential Land tiy Classification City of Springfield Oregon Legend D City Limit D Urban Growth Boundary 'I Class ifi cati ons MASTER PLAN PARTIALLY VACANT VACANT _ DEVELOPED N ! . . . Exhibit A-23 VACANT BUILDABLE LAND The next step in the buildable land inventory is to net out portions of vacant tax lots that are unavailable for development. Areas unavailable for development fall into two categories: (1) developed areas of partially vacant tax lots, and (2) areas with physical constraints (in this instance areas with steep slopes, waterway buffers, or wetlands). Table 3-4 shows land with development capacity by constraint status. The data show that about 214 acres within tax lots with development capacity are developed. An additional 824 acres have development constraints that are unbuildable, leaving about 1 ,447 vacant buildable residential acres within the UGB. Table 3-4. Residential land with development capacity by constraint status, Springfield UGB, 2009 Acres unavailable for housing Acres in Developed Unbuildable Classification Tax Lots Tax Lots Acres Acres Master Planned 18 151 138 Partially Vacant 234 841 77 Vacant 863 1,493 0 Total 1,115 2,485 214 Source: City of Springfield GIS data; analysis by ECONorthwest Note: No buildable acres are shown for master planned areas because the master plan identifies the number of dwelling units, This capacity is reflected in Table 3-7, Buildable Acres 13 170 641 824 See notes 595 852 1,447 Table 3-5 shows vacant land by plan designation. Map 3-3 shows the location of vacant land by plan designation. Map 3-4 shows vacant land with constraints that are unbuildable. Table 3-5. Residential land with development capacity by plan designation, Springfield UGB, 2008 Total Acres Developed Constrained Buildable Plan Designation Tax Lots in Tax Lots Acres Acres Acres Low Density Residential 981 2,137 71 765 1,301 Medium Density Residentia 126 329 142 58 128 High Density Res.idential 8 19 1 0 18 Total 1,115 2,485 214 824 1,447 Source: City of Springfield GIS data; analysis by ECONorthwest DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 15 Attachment 1-23 . f- U) N E J: E b; N J: f- "'" o \l, 30TH "tj !;: Z o '" -. ~<esldential Land by Classification City of Springfield Oregon Legend D City Limits D Urban Growth Boundary Classifications MASTER PLAN PARTIALLY VACANT VACANT N c::::;. ~~ l~ ~ . 30TH j) C,;}t . KeSlaentlal Lana by Classification and Constraint Status t; City of Springfield Oregon Legend D City Limit D Urban Growth Boundary Classifications MASTER PLAN PARTIALLY VACANT VACANT Constraints ~ Slope >25% ITit):r~i:~W;l Riparian Resource Areas Floodway c:J 100-yr Floodplain Wetlands ~ BPA Easement N . . . , , Exhibit A-26 REDEVELOPMENT POTENTIAL Redevelopment potential addresses land that is classified as developed that may redevelop during the planning period. While many methods exist to identify redevelopment potential, a common indicator is improvement to land value ratio. Different studies use different improvement to land value ratio thresholds. This study does not use improvement-to-land value ratios as a redevelopment threshold. The City of Springfield understands that low-value housing is an integral part of the City's affordable housing stock and that encouraging redevelopment of such housing will likely result in an overall loss of affordable housing in Springfield. Springfield uses a capacity-based method to identify redevelopment potential. Redevelopment capacity is estimated based on historical redevelopment rates. Historical rates of redevelopment are analyzed in Chapter 4. RESIDENTIAL CAPACITY The final step in a residential buildable lands inventory is to estimate the capacity of buildable land in dwelling units. The capacity of residential land is measured in dwelling units and is dependent on densities allowed in specific zones as well as redevelopment potential. In short, land capacity is a function of buildable land and density. The buildable lands inventory indicates that Springfield has about l,44 7 acres of vacant and partially-vacant residential land and an additional 21 acres in the Glenwood mixed-use refinement plan area (these acres were included in the commercial and industrial lands inventory and are included here only for the purpose of estimating residential capacity).7 This yields a total of 1,468 buildable acres. Table 3-7 provides an estimate of how much housing could be accommodated by those lands based on the needed densities identified in Table 5-25 after making deductions for development constraints. It includes capacity for areas with approved master plans that were not included in the acreage estimates. This includes Marcola Meadows (518 dwellings in the MDR designation) and RiverBend (730 dwellings in the MDR designation). Table 3-7 shows that Springfield has capacity for 8,722 dwelling units within the existing UGB. Note that this figure does not include capacity for redevelopment." 7 Capacity in the Glenwood mixed-use area was calculated as follows: 21 buildable acres (45% of the 47-acre site; the policy requires 30% to 60% of the site be used for housing) multiplied by 15 dwelling units per gross acre equals 317 dwelling units, minus 47 dwelling units that would be displaced from the River Bank Mobile Home Park equals 270 dwelling units, "Note that the revised capacity estimate of8,722 dwelling units does not make any adjustments for the City's Hillside Development Ordinance which requires minimum lot sizes of 10,000 or greater on slopes 15% or higher, or areas above 670' in elevation. Future drafts Page 18 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-26 . . . Exhibit A-27 Table 3-7. Estimated residential development capacity, Springfield UGB, 2009 Residential Buildable Capacity Plan Designation Acres (DU) Low Density Residential 824 5,379 Medium Density Residential 95 2,718 High Density Residential 16 355 Mixed-Use (Glenwood) 21 270 Total 956 8,722 Source: City of Springfield residential BLI; analysis by ECONorthwest Note: Estimated residential development capacity includes sites with approved master plans (RiverBend - 730 DU and Marcola Meadows - 518 DU, All of this capacity is in the Medium Density Residential plan designation), Percent of Capacity 60% 30% 4% 3% 97% of the residential could use such deductions, including deductions for land in floodplains, ifthe Springfield Planning Commission and City Council determine they yield more accurate results. DRAFT: Springfield Housing Needs Analysis ECONorthwest Page 19 Attachment 1-27 December 2009 . . . Exhibit A-28 Chapter 4 Historical Development Trends Analysis of historical development trends in Springfield provides insights into how the local housing market functions. The housing type mix and density are also key variables in forecasting future land need. Moreover, such an analysis is required by ORS 197.296. The specific steps are described in Task 2 of the DLCD HB 2709 Workbook: 1. Determin~ the time period for which the data must be gathered 2. Identify types of housing to address (all needed housing types) 3. Evaluate permit/subdivision data to calculate the actual mix, average actual gross density, and average actual net density of all housing types ORS 197.296 requires the analysis of housing mix and density to include the past five years or since the most recent periodic review, whichever time period is greater.9 The City of Springfield used the 1999- July 2008 period for this analysis. The rationale for using this period is that permit data prior to 1999 could not be associated with tax lots to develop density estimates. Moreover, the most recent housing needs analysis and inventory for the Eugene-Springfield Metropolitan Area was conducted in 1999. With respect to housing mix, the 1990 and 2000 Census provide more accurate counts. RESIDENTIA~ DEVELOPMENT TRENDS Figure 4-1 shows dwelling units approved in the Springfield city limits between 1980 and July 2008. Springfield approved 5,836 dwellings during this 26-year period. The number of dwellings approved annually ranges from a low of 14 in 1985 toa high of 616 in 1994. Springfield averaged about 217 dwelling unit approvals per year during this period. The rate of development, however, shows considerable variation from year to year. That variation can be largely tied to economic conditions in the region. 9 Specifically,ORS 197.296(5) (b) states: "A local government shall make the determination described in paragraph (a) of this subsection using a shorter time period than the time period described in paragraph (a) of this subsection if the local government finds that the shorter time period will provide more accurate and reliable data related to housing capacity and need. The shorter time period may not be less than three years," Page 20 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-28 I- - - Exhibit A-29 Figure 4-1. Dwelling units approved through building permits issued for new residential construction, Springfield, 1980 - July 2008 700 600 500 ". CIl e 400 0. 0. <C is 300 200 100 0 0 ;;; N C') <t 10 CD I"- <Xl (]) 0 a; N C') <t 10 CD I"- co (]) 0 0 N C') <t 10 CD I"- <Xl co co <Xl co co <Xl <Xl <Xl <Xl (]) (]) (]) (]) (]) (]) (]) (]) (]) 0 0 0 0 0 0 0 0 ~ ~ ~ (]) (]) (]) (]) (]) (]) (]) (]) ~ ~ ~ (]) ~ ~ ~ ~ ~ 0 0 0 0 0 0 0 0 0 ~ ~ ~ ~ ~ ~ ~ ~ ~ N N N N N N N N N Year Source: City of Springfield Planning Department, 2008 Note: 2008 includes January through July, Between July 1999 and July 2008, Springfield issued a total of 1,97l building permits for new residential construction that allowed 2,860 dwelling units. Figure 4-l shows that the number of dwelling units approved varies from year to year and peaked at 515 in 2002. The number of dwellings approved was slower in 1999 and 2001. Between 2003 and 2005, the number of dwellings approved remained relatively steady at around 360 annually. By 2006, residential permits reflected the downturn in the national housing market, but still remained relatively strong averaging around 200 permits per year. DRAFT: Springfield Housing Needs Analysis ECONorthwest December2009 Page 21 Attachment 1-29 . . . Exhibit A-3D Figure 4-1. Dwelling units approved through building permits issued for new residential construction, Springfield, July 1999 - July 2008 600 500 III b.Q 400 @ CII 3: c .... 300 0 .. CII ,g E 200 :::l Z 100 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 thru July 2008 Year Source: City of Springfield Planning Department, 2006 Table 4-1 shows dwelling units approved through building permits issued for new residential construction by type within Springfield. The data indicate that about 54% of residential dwellings approved were for single-family detached dwellings, manufactured homes accounted for about 10% of all permits issued, and multifamily housing of all types accounted for 36% of permits issued. Page 22 DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Attachment 1-30 Exhibit A-31 Table 4-1. Dwelling units approved through building permits issued for . new residential construction by type, Springfield, July 1999 - July 2008 Year Single Manufact- Duplex Tri-Plex Four- Apart- Total Family ured Plex ment Units Home 1999 30 9 22 0 O. 0 61 2000 209 38 30 3 4 40 324 2001 121 46 16 6 0 6 195 2002 252 45 14 0 4 200 515 2003 230 31 18 6 84 0 369 2004 155 26 38 6 12 122 359 2005 144 31 38 6 140 0 359 2006 116 27 17 3 56 0 219 2007 180 30 0 4 61 275 thru July 2008 92 27 10 0 0 55 184 Total Units 1529 280 233 30 304 484 2860 % of Units 53.5% 9.8% 8.1% 1.0% 10.6% 16.9% 100.0% Source: City of Springfield Planning Department, 2006 TRENDS IN HOUSING MIX AND TENURE . The housing mix by type (i.e., percentage of single family, multi-family, and mobile/manufactured home units) is an important variable in any housing needs assessment. Distribution of housing types is influenced by a variety of factors, including the cost of new home construction, area economic and employment trends, demographic characteristics, and amount of land zoned to allow different housing types and densities. Table 4-2 shows changes in Springfield's housing mix from 1990-2000. Between 1990 and 2000, Springfield increased its housing stock by 19%, adding 3,451 dwelling units. The mix of housing did not change substantially. In 1990 and 2000, 54% of dwelling units were single-family detached units. Over the ten- year period, Springfield added more than 2,000 single- family detached dwellings. Thirty-one percent of the new dwellings added between 1990 to 2000 were multifamily or manufactured. However, the share of these more affordable housing types did not increase in Springfield over the ten-year period. In 1990, these housing types accounted for 37% of the housing stock and in 2000 they accounted for 37% ofthe housing stock. . With respect to tenure, Springfield experienced a 4% increase in the ownership rate between 1990 and 2000. About 49% of housing in the Springfield city limits was owner-occupied in 1990 and 54% was owner-occupied in 2000. Homeownership rates in Springfield are lower than County and State averages. In 1990, about 61% of homes were owner-occupied in Lane County, a figure that increased to 63% by 2000. State homeownership rates were 63% in 1990 and 64% in 2000. DRAFT: Springfield Housing Needs Analysis ECONorthwest December2009 Page 23 Attachment 1-31 . . . Exhibit A-32 Table 4-2. Dwelling units by type and tenure, Springfield city limits, 1990 and 2000 1990 Census 2000 Census New DU 90-00 Housing Units Number Percent Number Percent Number Percent % Increase Single-family detached 9,687 53,5% 11,721 54,3% 2,034 58,9% 21% Single-family attached 1,755 9,7% 1,794 8,3% 39 1,1% 2% Multifamily 4,777 26,3% 6,118 28.4% 1,341 38,9% 28% Mobile/Manufactured 1,902 10,5% '1,939 9.0% 37 1,1% 2% Total housing units 18,121 100.0% 21,572 100.0% 3,451 100.0% 19% Occupied Housing Units 17,447 100.0% 20,514 100.0% 3,067 100.0% 18% Owner-occupied 8,599 49,3% 10,987 53,6% 2,388 77,9% 28% Renter-occupied 8,848 50,7% 9,527 46.4% 679 22,1% 8% Source: U,S, Census of Population and Housing; SF-3 1990 and 2000, Table 4-3 shows type of dwelling by tenure (owner/renter-occupied) in 2000. The results show that single-family and manufactured housing types have a much higher ownership rate than other housing types-about 95% of owner-occupied units were in these housing types. Multifamily housing types, including duplexes were predominately renter occupied. It is also notable that 88% of the single- family attached dwellings were renter occupied. By contrast, 20% of single- family detached and 13% of mobile homes were renter occupied in 2000. Table 4-3. Housing units by type and tenure, Springfield city limits, 2000 Owne r-Occupied Renter-Occupied Total %by %by %by %by %by Housing Type Number Tenure Type Number Tenure Type Number Type Single-family detached 8,989 80% 82% 2,219 20% 23% 11,208 55% Single-family attached 204 12% 2% 1 ,494 88% 16% 1,698 8% Multifamily-duplex 118 10% 1% 1,113 90% 12% 1,231 6% Multifamily-3+ units 89 2% 1% 4,447 98% 47% 4,536 22% Mobile home 1,581 87% 14% 244 13% 2% 1,825 9% Total 10,981 54% 100% 9,517 46% 100% 20,498 100% Source: US Census 2000, Summary File 3; Percentages calculated by ECONorthwest. Note: Total number of units is slightly different than reported in Table 4-2 due to different data sources (this table uses Summary File 3 sample data; Table 9,30.2 uses Summary File 1,100% count data, Table 4-4 shows changes in Springfield's housing mix from 2000-July 2008 based on 2000 Census and residential building permit data provided by the City of Springfield. Between 2000 and July 2008, Springfield increased its housing stock about 13%, adding 2,799 dwelling units. The mix of housing changed slightly, with multifamily dwellings accounting for about 0.9% greater share in July 2008 than 2000. Page 24 DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Attachment 1-32 . . . Exhibit A-33 Table 4-4. Estimated dwelling units by type, Springfield city limits, 2000 and July 2008 2000 Census 2006 Est. New DU 00-06 Housing Units Number Percent Number Percent Number Percent % Increase Single-family detached 11,721 54,3% 13,220 54,2% 1 ,499 53,6% 13% Single-family attached 1 ,794 8,3% 1,794 7.4% na na 0% Multifamily 6,118 28.4% 7,147 29,3% 1,029 36,8% 17% MobilelManufactured 1,939 9,0% 2,210 9,1% 271 9,7% 14% Total housinQ units 21,572 100.0% 24,371 100.0% 2,799 100.0% 13% Source: U,S, Census of Population and Housing; SF-3 1990 and 2000; City of Springfield Building Permit Data, 2006, Note: the City building permit data does not distinguish between single-family attached and detached dwellings, Thus, the 2008 estimate probably overestimates single-family detached dwellings and underestimates single-family attached dwellings, DENSITY Table 4-5 summarizes approved net residential densities by housing type from July 1999 through July 2008. During this period, 2,860 dwelling units were approved by residential building permits. The dwellings are associated with individual tax lots to calculate the net residential density (expressed indwelling units per acre).1O This development consumed 436.3 net vacant acres. New housing in Springfield developed at an average net density of 6.6 dwelling units per net buildable acre between 1999 and July 2008. The data indicate that single-family detached housing types averaged a density of 5.4 dwelling units per net acre, while manufactured homes achieved a lower density of 4.6 dwelling units per net acre. Multifamily housing types show more variation-from 25 units per net acre for triplexes, to 8.5 dwelling units per net acre for fourplexes, and 24.4 dwellings per net acre for apartment buildings with five or more units. 10 OAR 660-024-0040(9) defines a net buildable acre as follows: For purposes of this rule, a "Net Buildable Acre" consists of 43,560 square feet of residentially designated buildable land, after excluding present and future rights-of-way, restricted hazard areas, public open spaces and restricted resource protection areas. DRAFT: Springfield Housing Nee?s Analysis ECONorthwest December2009 Page 25 Attachment 1-33 Exhibit A-34 Table 4-5. Actual residential density by housing type, in net acres, . Springfield, July 1999 - July 2008 Dwelling Percent Net DU/Net Housing Type Units ofDU Acres Acre Single-Family Detached 1,529 '53% 280.7 5.4 Manufactured Home 280 10% 61.2 4.6 Duplex 233 8% 37.5 6,2 Triplex 30 1% 1,2 25,0 Fourplex 304 11% 35,9 8,5 Apartments 5+ Units 484 17% 19.8 24.4 Total 2,860 100'16 436.3 6.6 Source: City of Springfield building permit data REDEVELOPMENT TRENDS Analysis of historical redevelopment of residential lands provides context for determining how much redevelopment will occur over the 20-year planning period. Specifically, the analysis addressed redevelopment by analyzing new dwellings on developed lots. This includes lots that had addresses coded before 1999 and received additional addresses after 1999. In other words, it focuses on lands that were identified as "developed" in the buildable lands inventory, but had additional residential development in the 1999-2008 period. . The analysis found 102 new dwellings were added on developed lots between 1999 and 2008. This is about 4% of2,860 dwellings added in Springfield during this period. . Page 26 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-34 . . . Exhibit A-35 Chapter 5 Housing Demand and Need Chapter 2 described the framework for conducting a housing "needs" analysis. ORS 197.296 (HB 2709) requires cities over 25,000 or fast growing cities to conduct a housing needs analysis. A recommended approach is described in Task 3 of the HB 2709 Workbook. The specific steps in the housing needs analysis are: 1. Project number of new housing units needed in the next 20 years. 2. Identify relevant national, state, and local demographic and economic trends and factors that may affect the 20-year projection of structure type mIX. 3. Describe the demographic characteristics of the population and, if possible, housing trends that relate to demand for different types of housing. 4. Determine the types of housing that are likely to be affordable to the projected households based on household income. 5. Estimate the number of additional needed units by structure type. 6. Determine the needed density ranges for each plan designation' and the average needed net density for all structure types. STEP 1: PROJECT NUMBER OF NEW HOUSING UNITS NEEDED IN THE NEXT 20 YEARS Step 1 in the housing needs analysis is to project the number of new housing units needed during the planning period. This section describes the key assumptions and estimates of new housing units needed in Springfield between 2000 and 2020. POPULATION Springfield must have a population forecast to project expected population change over the 20-year planning period (in this instance, 2010-2030). Lane County adopted coordinated population forecasts for the County and its incorporated cities in June 2009. The forecasts include figures for Springfield for 2010 and 2030. Table 5-1 shows the coordinated population forecast for the Springfield city limit, urban area (the area between the city limit and UGB), and the UGB for 2010 to 2030. The UGB forecast for 2030 is 8l,608 persons-an increase of 14,577 persons during the 20-year planning period. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 27 Attachment 1-35 Exhibit A-36 . Table 5-1. Springfield coordinated population forecast, Springfield UGB, 2010 to 2030 Urban Area Year City Limit 58,891 74,814 8,140 6,794 UGB 67,031 81,608 2010 2030 Cha nge 2010-2030 Number 15,923 (1,346) 14,577 Percent 27% -17% 22% AAGR 1.2% -0,9% 1,0% Source: Lane County Rural Comprehensive Plan, 1984 (Amended in 2009), Table 1-1, pg 5 PERSONS IN GROUP QUARTERS Persons in group quarters do not consume standard housing units: thus, any forecast of new people in group quarters is typically backed out of the population forecast for the purpose of estimating housing need. Group quarters can have a big influence on housing in cities with colleges (dorms); prisons, or a large elderly population (nursing homes). In general, one assumes that any new requirements for these lodging types will be met by institutions (colleges, state agencies, health- care corporations) operating outside what is typically defined as the housing market. Group quarters, however, require land and are typically built at densities that are comparable to multiple-family dwellings. . Table 5-2 shows persons in group quarters in the City of Springfield as reported by the 1980, 1990, and 2000 Census. Table 5-2. Persons in group quarters, City of Springfield, 1980, 1990, and 2000 VARIABLE 1980 Total Population 41,621 Persons in Group Quarters 184 Percent in Group Quarters 0.44% Source: U,S, Census of Population and Housing, Summary File 1 1990 44,683 298 0.67% 2000 52,864 635 1.20% For the purpose of estimating housing needs for Springfield, ECO assumed that 1% of new persons (148 persons) will reside in group quarters. The majority of these new persons will live in assisted living quarters. A final note on persons in group quarters: persons in group quarters require land. While the HB 2709 workbook backs this component of the population out of total population that needs housing, it does not otherwise make accommodations for land demand for new group quarters. For the purpose of this analysis, we assume that persons in group quarters require land at approximately the same density as multiple family housing. Land needed for group quarters is estimated at the end ofthis chapter. . Page 28 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-36 Exhibit A-37 HOUSEHOLD SIZE AND COMPOSITION . Twenty years ago, traditional families (married couple, with one or more children at home) accounted for 29% of all households in Oregon. In 1990 that percentage had dropped to 25%. It wi11likely continue to fall, but probably not as dramatically. The average household size in Oregon was 2.60 in 1980 and 2.52 in 1990. One and two person households made up the majority of Oregon households in 1990. The direct impact of decreasing household size on housing demand is that smaller households means more households, which means a need for more housing units even if population were not growing. Table 5-3 shows average household size for Springfield as reported by the 1980, 1990, and 2000 Census. OAR 660-024-0040(7)(a) established a "safe harbor" assumption for average household size-which is the figure from the most recent Census (2.54 persons). The estimate of future housing needs uses an average household size of 2.54 persons, as allowed by the safe harbor. Table 5-3. Average household size, Springfield, 1980, 1990 and 2000 Average household size Year 1980 1990 2000 2,57 2,54 2.54 . Source: U,S, Census of Population and Housing, Summary File 1 VACANCY RATE Vacant units are the final variable in the basic housing need model. Vacancy rates are cyclical and represent the lag between demand and the market's response to demand in additional dwelling units. Vacancy rates for rental and multiple family units are typically higher than those for owner-occupied and single-family dwelling units. Table 5-4 shows that the average vacancy rate for Springfield varies by time period. The most recent Census showed an overall vacancy rate of 5%. The RCS housing needs model, however, requires separate vacancy rate figures for single- family and multifamily units. The vacancy rate in 2000 was 4.7% for single- family units and 5.7% for multifamily units. Table 5-4. Average vacancy rate, Springfield, 1980, 1990 and 2000 . Variable 1980 1990 Housing Units 17,469 18,121 Occupied Housing Units 16,173 17,447 Vacant Housing Units 1,296 674 Vacancy Rate 7.42% 3,72% Source: U,S, Census of Population and Housing, Summary File 1 2000 21,500 20,426 1,074 5.00% DRAFT: ,Springfield Housing Needs Analysis ECONorthwest December 2009 Page 29 Attachment 1-37 . . . Exhibit A-38 Thus study assumes an average vacancy rate of 5%--the same figure as reported in the 2000 Census. The countywide vacancy rate was 6.1 % in 2000. FORECAST OF NEW HOUSING UNITS, 2010-2030 The preceding analysis leads to a forecast of new housing units likely to be built in Springfield during the 2010 to 2030 period. Based on the assumptions shown in Table 5-5, Springfield will need 5,980 new dwelling units to accommodate forecast population growth between 2010 and 2030. These figures do not include new group quarters. The forecast assumes 60% will be single- family housing types (single-family detached and manufactured) and 40% will be multifamily. The rationale for the household mix is described in the housing needs analysis section of this chapter. The results indicate that Springfield will need to issue permits for about 299 new dwelling units annually during the planning period. This figure is consistent with the 300 dwelling units approved annually during the 1999 to July 2008 period, but is still significantly below the 515 dwellings approved in 2002. The forecast of new units does not include dwellings that will be demolished and replaced. This analysis does not factor those units in; it assumes they will be replaced at the same site and will not create additional demand for residential land. Table 5-5. Demand for new housing units, Springfield UGB, 2010-2030 Variable Assumptions I Results 14,577 145 14,432 2,54 5,682 5% 5,980 Change in persons minus Change in persons in group quarters equals Persons in households Average household size New occupied DU Average vacancy rate Total new DU Single-family dwelling units Percent single-family DU New occupied single-fam ily DU Multiple family dwelling units Percent multiple familyDU New occupied m ultiple-fam ily DU Totals equals Total new occupied dwelling units 5,980 Dwelling units needed annually 299 Source: Calculations by ECONorthwest based on safe harbor population forecast and assumptions described above, 60% 3,588 40% 2,392 Page 30 DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Attachment 1-38 Exhibit A-39 . STEP 2: IDENTIFY RELEVANT NATIONAL, STATE, AND LOCAL DEMOGRAPHIC AND ECONOMIC TRENDS AND FACTORS THAT MAY AFFECT THE 20-YEAR PROJECTION OF STRUCTURE TYPE MIX NATIONAL HOUSING TRENDS The overview of national, state, and local housing trends builds from previous work by ECO and conclusions from The State of the Nation's Housing, 2008 report from the Joint Center for Housing Studies of Harvard University. The Harvard report summarizes the national housing outlook for the next decade as follows: "Housing markets contracted for a second straight year in 2007. The national median single-family home price fell in nominal terms for the first time in 40 years of recordkeeping, leaving several million homeowners with properties worth less than their mortgages. With the economy softening and many home loans resetting to higher rates, an increasing number of owners had difficulty keeping current on their payments. Mortgage performance---especially on subprime loans with adjustable rates---eroded badly. Lenders responded by tightening underwriting standards and demanding a higher risk premium, accelerating the ongoing slide in sales and starts. . "It is still uncertain how far, and for how long, the housing crisis will drive down household growth. Regardless, given the solid underpinnings of long-term demand-including the recent strength of immigration and the aging of the echo-boom generation into young adulthood-household growth will pick up again once the economy recovers. But if the nation suffers a prolonged economic downturn that results in lower immigration and more doubling up, household growth in 2010-2020 may fall short of the 14.4 million level currently projected. This evaluation presents a bleak outlook for housing markets and for homeownership in the short-term brought on by the subprime mortgage crisis. However, the image painted of the future looks brighter, as the increase in housing demand is naturally induced by the growth of the population in the necessary age groups. Following is a summary of key national housing trends: · By 2006, higher prices and rising interest rates had a negative impact on market demand. Investor demand, home sales and single-family starts dropped sharply. Growth in national sales prices also slowed. By 2007 and early 2008, housing market problems had reached the rest of the economy, resulting in a nationwide economic slowdown and fear of recession. · Homeownership rates are decreasing. After l2 successive years of increases, the national homeownership rate slipped in 2005, again in 2006 to 68.8%, and again in 2007 to 68.l %. The Joint Center for Housing Studies predicts that once the corrections made to work off the housing oversupply and prices start to recover, a return to traditional mortgage . DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 31 Attachment 1-39 . . . Exhibit A-40 products and the strength of natural demand will invigorate the homeownership rate. . The long-term market outlook shows that homeownership is still the preferred tenure. Over the next decade, 88% of net household growth is expected to come from gains in the number of homeowners. While further homeownership gains are likely during this decade, they are not assured. . Population increases will drive future demand. The Joint Center for Housing Studies indicates that demand for new homes could total as many as l4.4 million units nationally between 2010 and 2020. Nationally, the vast majority of these homes will be built in lower-density areas where cheaper land is in greater supply. . People and jobs have been moving away from central business districts (CBDs) for more than a century: the number of the country's largest metropolitan areas with more than half of their households living at least 10 miles from the CBD has more than tripled from 13 in 1970 to 46 in 2000; in six metropolitan areas more than a fifth of households live at least 30 miles out. While people older than 45 years are generally continuing to move away from CBDs, younger people have begun to move nearer to CBDs. . Demand for higher density housing types exists among certain demographics. They conclude that because of persistent income disparities, as well as the movement of the echo boomers into young adulthood, housing demand may shift away from single-family detached homes toward more affordable multifamily apartments, town homes, and manufactured homes. Supply-side considerations, however, outweigh these demographic forces. · Immigration will playa key role in accelerating household growth over the next 10 years. Between 2000 and 2006, immigrants contributed to over 60% of household growth. Minorities will account for 68% of the l4.6 million projected growth in households for the 2005 to 2015 period. Immigrants now comprise a growing share of young adults and children in the United States. Twenty percent of Americans ages 25-34 are foreign born, and an additional 9% are second generation Americans. · An aging population, and of baby boomers in particular, will drive changes in the age distribution of households in all age groups over 55 years. A recent survey of baby boomers showed that more than a quarter plan to relocate into larger homes and 5% plan to move to smaller homes. Second home demand among upper-income homebuyers of all ages also continues to grow. Households aged 50 to 69 are expected to account for the purchase of nearly half a million second homes between 2005 and 2015. Page 32 ECONorthwest DRAFT: Springfield Housing Needs Analysis December 2009 Attachment 1-40 Exhibit A-41 . . The Joint Center for Housing studies expects rental housing demand to grow by 1.8 million households over the next decade. Minorities will be responsible for nearly all of this increased demand. The minority share of renter households grew from 37% in 1995 to 43% in 2005. The minority share is forecast to exceed 50% of renter households in 2015. Demographics will also playa role. · Ratios of rent to income are forecast to continue to increase. In 2006, one in three American households spent more than 30% of income on housing, and more than one in seven spent upwards of 50%. The national trend towards increased rent to income ratios is mirrored regionally in that a salary of two to three times the 2007 Federal minimum wage of $5.85 is needed to afford rents in Lane County. The U.S Bureau of Census Characteristics of New Housing Report presents data that show trends in the characteristics of new housing for the nation, state, and local areas. Several trends in the characteristics of housing are evident from the New Housing Report: . · Larger single-family units on smaller lots. Between 1997 and 2007 the median size of new single-family dwellings increased l5%, from l,975 sq. ft. to 2,277 sq. ft. nationally and 18% in the western region from 1,930 sq. ft. to 2,286 sq. ft. Moreover, the percentage of units under l,200 sq. ft. nationally decreased from 8% in 1997 to 4% in 2007. The percentage of units greater than 3,000 sq. ft. increased from 15% in 1997 to 26% of new one-family homes completed in 2007. In addition to larger homes, a move towards smaller lot sizes is seen nationally. Between 1994 and 2007 the percentage of lots under 7,000 sq. ft. increased by 13% from 29% oflots to 33% of lots. A corresponding 4% decrease in lots over ll,OOO sq. ft. is seen. · Larger multifamily units. Between 1999 and 2007, the median size of new multiple family dwelling units increased by 15%. The percentage of multifamily units with more than 1,200 sq. ft. increased from 26% to 47% in the western region and from 28% to 50% nationally. The percentage of units with less than 600 sq. ft. stayed at 1 % both regionally and nationally. · More household amenities. Between 1994 and 2007 the percentage of single-family units built with amenities such as central air conditioning, fireplaces, 2 or more car garages, or 2 or more baths all increased. The same trend in increased amenities is seen in multiple family units. . A clear linkage exists between demographic characteristics and housing choice. This is more typically referred to as the linkage between life-cycle and housing choice and is documented in detail in several publications. Analysis of data from the Public Use Microsample (P~MS) in the 2000 Census to describe DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 33 Attachment 1-41 . . . Exhibit A-42 the relationship between selected demographic characteristics and housing choice. Key relationships identified through this data include: , . Homeownership rates increase as income increases; . Homeownership rates increase as age increases; . Choice of single-family detached housing types mcreases as mcome mcreases; . Renters are much more likely to choose multiple family housing types than single-family; and . Income is a stronger determinate of tenure and housing type choice for all age categories. STEP 3: DESCRIBE THE DEMOGRAPHIC CHARACTERISTICS OF THE POPULATION AND, IF POSSIBLE, HOUSING TRENDS THAT RELATE TO DEMAND FOR DIFFERENT TYPES OF HOUSING State and regional demographic and housing trends are important to a thorough understanding of the dynamics of the Springfield housing market. Springfield exists in a regional economy; trends in the region impact the local housing market. This section documents state and regional demographic and housing trends relevant to Springfield. DEMOGRAPHIC TRENDS This section reviews historical demographic trends in the Lane County and Springfield. Demographic trends provide a broader context for growth in a region; factors such as age, income, migration and other trends show how communities have grown and shape future growth. To provide context, we compare the Springfield with Lane County and Oregon where appropriate. Characteristics such as age and ethnicity are indicators of how population has grown in the past and provide insight into factors that may affect future growth. State Demographic Trends Oregon's 2006-2010 Consolidated Plan includes a detailed housing needs analysis as well as strategies for addressing housing needs statewide." The plan concludes that "Oregon's changing population demographics are having a significant impact on its housing market." It identified the following population and demographic trends that influence housing need statewide: . II th fastest growing in the United States . Facing dramatic housing cost increases 11 http://www.ohcs.oregon.gov/OHCS/HRS_Consolidated_Plan _ 5yearplan.shtml Page 34 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-42 Exhibit A-43 . . Facing median and adjusted incomes less than those of 1999 . Growing faster than national rates: 4.0% v. 3.3% and expecting a non- entitlement growth during this consolidated plan of about 6%, 82% of which will come from in-migration. . Increasingly older · Increasingly diverse . Increasingly less affluent12 Richard Bjelland, State Housing Analyst at the Housing and Community Services Department of the State of Oregon, analyzed recent demographic changes taking place in Oregon and discussed their implications in a 2006 presentation "Changing Demographics: Impacts to Oregon and the US." Some of Bjelland's most significant findings are summarized below: · Oregon's minority population is growing quickly. Minorities made up 9.2% of the population in 1990 and 16.5% of the population in 2000, a 52% increase. . · Hispanics and Latinos make up a large share of that population and their growth rate is higher than non-Hispanics/ Latinos. The growth rate of Oregon's non-Hispanic/ Latino population between 1990 and 2000 was l5.3% compared to 144.3% for Hispanics and Latinos. · The birth rates of Hispanic/ Latino residents are higher than non- Hispanic/ Latino residents. In 1998, for the US, white non-Hispanic/ Latino residents had a birth rate of 12.3 per l,OOO, lower than Asians and Pacific Islanders (16.4 per 1,000), black non-Hispanics (18.2 per 1,000) and Hispanic/ Latino (24.3 per 1,000). . The share of resident births and deaths in Oregon shows the implications of that birthrate: Hispanic/ Latino residents accounted for 17.4% of births but only 1.4% of deaths in Oregon for 2001. In addition, Hispanic/ Latino Oregonians are younger than non-Hispanic/ Latino residents: in 2000, 75.9% of Hispanic/ Latino residents of Oregon are under age 35, compared to 45.7% of non-Hispanic/ Latino residents. · In Oregon, Hispanic/ Latino per capita income in 2005 was only 44% of white per capita income. · Hispanic/ Latino residents of Oregon become homeowners at younger ages than non-Hispanic/ Latino residents. Table 5-6 shows that Hispanic/ Latino Oregonians under 45 have higher homeownership rates than non- Hispanic/ Latino residents. . 12 State of Oregon Consolidated Plan, 2006-2010, pg. 23. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 35 Attachment 1-43 . . . Exhibit A-44 Table 5-6. Oregon homeownership rates by age of householder, 2000 Age of Non-Hispanicl Hispanicl householder Latino Latino 25-34 10.2% 25.7% 35-44 20,6% 31.0% 45 and older 68.1 % 39.4% Source: Richard Bjelland, State Housing Analyst at the Housing and Community Services Department of the State of Oregon, "Changing Demographics: Impacts to Oregon and the US" 2006, He obtained his data from US Census 2000, Note: Percentages represent percent of households in each age group that own homes; columns do not sum to 100%, Regional Demographic Trends Regional demographic trends largely follow the statewide trends discussed above, but provide additional insight into how demographic trends might affect housing in Springfield. Figure 5-l shows the populations of Oregon, Lane County, and Springfield by age for 2000. Springfield has a greater proportion of its population less than 40 years old than Oregon and Lane County, especially residents aged 20-29 and under 9 years. Springfield has comparatively fewer residents over 40 than the state. Figure 5-1. Population distribution by age, Oregon, Lane County, and Springfield, 2000 70 and older 60-69 50-59 40-49 Gl Cl <C 30-39 20-29 10-19 -. - -- -. ..__._.M_____ _ --.-. - ---~---- - -- --- - ---- ---- --- -~------ - ----- ~--- - ---- ------.--.-- Under 9 - -I 0% 2% 4% 6% 8% 10% 12% Percent of populatic 14% 16% 18% I . OR . Lane County 0 Springfield' Source: U.S, Census, 2000 Some outlying communities in the region have populations similar in age distribution to Springfield. Outlying communities with the largest percent of Page 36 DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Attachment 1-44 Exhibit A-45 . households with children from the 2000 census were: Creswell (41%), Veneta (40%), Junction City (40%), and Coburg (38%). The communities with the smallest percent of households with children were Eugene (27%), Oakridge (28%), and Cottage Grove (35%). In the communities with larger shares of children, attendance rates of children in elementary school are not declining, unlike districts such as Oakridge, McKenzie, and Pleasant Hill. School districts that have experienced increases in the Kindergarten-2nd grade populations are Fern Ridge District 28J (increased since 2003), Lowell 7l (since 2004), Creswell 40 (since 1999 with a dip in 2004), and Junction City 69 (from 2002 to 2005). However, this data is based on small districts with small class sizes, so it is not entirely conclusive. Outlying communities with the largest percent of persons 65 and over from the 2000 Census were: Oakridge (21 %) and Cottage Grove (15%). The community with the smallest percent of persons 65 and older was Veneta (9%). These data indicate that some outlying communities' trend toward older populations, others trend towards younger populations with families with younger children. . Table 5-7 shows population by age for Lane County for 2000 and 2006. The data show that Lane County grew by 13,479 people between 2000 and 2006, which is a 4% increase. The age breakdown shows that the County experienced an increase in population for every age group over age 25. The fastest growing age groups were aged 45 to 64 years and 65 and over. The group that experienced the fastest negative growth was ages l8-24. Table 5-7. Population by age, Lane County, 2000 and 2006 2000 2006 Chanae Aae Group Number Percent Number Percent Number Percent Share Under 5 18,584 6% 18,056 5% -528 -3% 0% 5-17 55,230 17% 52,730 16% -2,500 -5% -1% 18-24 38,662 12% 34,666 10% -3,996 -10% -2% 25-44 88,849 28% 95,171 28% 6,322 7% 1% 45-64 78,680 24% 88,926 26% 10,246 13% 2% 65 and over 42,954 13% 46,889 14% 3,935 9% 1% Total 322,959 100% 336,438 100% 13,479 4% 0% Source: U,S, Census, 2000 and Claritas, 2006 Table 5-8 shows Claritas Inc. population forecast by age for Lane County from 2006 to 2011. The data show that, with the exception of the 5-17 and l8-24 year old groups, each age group will experience growth and that groups aged 65 years and older and 45 to 64 years will grow at the fastest rates. The forecast shows that the 5 to 17 and 18 to 24 year age groups will decline. . DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 37 Attachment 1-45 Exhibit A-46 . Table 5-8. Claritas Inc. population projection by age, Lane County, 2006 and 2011 2006 2011 Chanae Age Group Number Percent Number Percent Number Percent Share Under 5 18,056 5% 18,615 5% 559 3% 0% 5-17 52,730 16% 51,098 15% -1,632 -3% -1% 18-24 34,666 10% 31,827 9% -2,839 -8% -1% 25-44 95,171 28% 99,401 29% 4,230 4% 0% 45-64 88,926 26% 94,999 27% 6,073 7% 1% 65 and over 46,889 14% 52,765 15% 5,876 13% 1% Total 336,438 100% 348,705 100% 12,267 4% 0% Source: Claritas, 2006 The data in Tables 5-7 and 5-8 suggest that Lane County is attracting older people and experiencing comparatively slow growth (or negative growth) in people under 44 years old. The age distribution in Figure 3 suggests a higher percentage of young adults (20-29) and children live in Springfield, indicating that Springfield's population and age trends are somewhat different from the projections for the county as a whole. . Between 1990 and 1999, almost 70% of Oregon's total population growth was from net migration (in-migration minus out-migration), with the remaining 30% from natural increase (births minus deaths).l3 Migrants to Oregon tend to have many characteristics in common with existing residents, with some differences- recent in-migrants to Oregon are, on average, younger and more educated, and are more likely to hold professional or managerial jobs, compared to Oregon's existing population. The race andethnicity of in-migrants generally mirrors Oregon's established pattern, with one exception: Hispanics make up more than 7% ofin-migrants but only 3% of the state's population. The number-one reason cited by in-migrants for coming to Oregon was family or friends, followed by quality of life and employment.14 Migration is a significant component of population growth in Lane County. Seventy-three percent of population growth in Lane County between 1990 and 2000 was from in-migration. This figure remained at 73% for the 2000-2005 period.15 The U.S. Census collects information about migration patterns. Specifically, it asks households where their residence was in 1995 (5 years prior to the Census count). Table 5-9 shows place of residence in 1995 for Oregon, Lane County, and Springfield. The data show that Springfield residents are more mobile than Lane County and Oregon residents. Less than half of residents in Oregon, Lane County or Springfield lived in the same residence in 1995 as in 2000. Twenty-four l3 Portland State University, Population Research Center, 2000.1990-2000 Components of Population Change 14 State of Oregon, Employment Department. 1999.1999 Oregon In-migration Study. . 15 Portland State University, Population Research Center, 2005. 2005 Oregon Population Report and contents Page 38 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-46 Exhibit A-47 . percent of Oregonians, 20% of residents of Lane County and 19% of residents of Springfield lived in a different county in 1995. Eleven percent of residents of Springfield and 13% of residents of Lane County lived in a different state in 1995, compared with 12% of Oregonians. Table 5-9. Place of residence in 1995, Oregon, Lane County, and Springfield, persons 5 years and over Oreoon Lane County Sorinofield Persons Percent Persons Percent Persons Percent Population 5 years and older 3,199,323 100% 304,463 100% 48,403 100% Same house in 1995 1,496,938 47% 142,447 47% 20,023 41% Different house in 1995 1,702,385 53% 162,016 53% 28,380 59% Same county 863,070 27% 94,788 31% 18,610 38% Different county 755,954 24% 61,639 20% 9,085 19% Same state 356,626 11% 23,526 8% 3,599 7% Different state 399,328 12% 38,113 13% 5,486 11% Source: U,S, Census, 2000 Table 5-10 shows the number of persons of Hispanic or Latino origin for Oregon, Lane County, and Springfield for 1990 and 2000. Springfield has a lower proportion of Hispanic/Latino residents as Oregon and a higher proportion than Lane County. In 2000, Springfield's population was 6.6 % Hispanic/Latino, compared with 4.5% of residents in Lane County. . The Hispanic/Latino population grew faster in Springfield than in Lane County from 1990 to 2000. Springfield's Hispanic/Latino population grew by 168% between 1990 and 2000. During the same period, Lane County's Hispanic/Latino population grew by III % and Oregon' Hispanic/Latino population grew by 143%. Table 5-10. Persons of Hispanic or Latino origin, Oregon, Lane County, and Springfield, 1990 and 2000 Lane Oregon County Springfield 1990 Total population 2,842,321 282,912 44,683 Hispanic or Latino 112,707 6,852 1,299 Percent Hispanic or Latino 4,0% 2.4% 2.9% 2000 Total population 3,421,399 322,959 52,729 Hispanic or Latino 273,938 14,488 3,475 Percent Hispanic or Latino 8.0% 4.5% 6,6% Change 1990-2000 Hispanic or Latino 161,231 7,636 2,176 Percent Hispanic or Latino 143% 111% 168% Source: U,S, Census, 2000 Table 5-l1 shows the number of Hispanic and Latino residents and the percent of Hispanic/ Latino residents as a percent of the total population between 1990 and 2000. The number of Hispanic and Latino residents is growing in all outlying . DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 39 Attachment 1-47 . . . Exhibit A-48 areas, especially in Cottage Grove and Junction City, according to the US Census 1990 and 2000. Table 5-11. Persons of Hispanic or Latino origin, outlying communities, 1990 and 2000 1990 2000 Change Percent Percent Number of total Number of total Number Percent Coburg 18 2% 29 3% 11 61% Cottage Grove 162 2% 417 5% 255 157% Creswell 109 4% 251 7% 142 130% Eugene 3,051 3% 6,843 5% 3,792 124% Junction City 73 2% 391 8% 318 436% Oakridge 141 5% 158 5% 17 12% Springfield 1,299 3% 3,651 7% 2,352 181% Veneta 50 2% 115 4% 65 130% Source: US Census 1990 and 2000 Table 5-12 shows household size by ethnicity for Oregon, Lane County, and Springfield. The number of people per household is similar for Oregon, Lane County, and Springfield for non-Hispanic households and Hispanic households. In each area, non-Hispanic households have a little less than 2.5 people per household. Households for Hispanic residents are larger, with between 3.2 and 3.9 people per household. The data show that Hispanic residents have between 0.7 and 1.4 additional people per household than non-Hispanic residents. Table 5-12. Household size by ethnicity for Oregon, Lane County, and Springfield, 2000 Oregon Lane County Springfield 2.42 2,39 2.49 3,87 3,19 3.50 Non-Hispanic/ Latino Hispanicl Latino Source: U,S, Census, 2000 In conclusion: (1) Springfield residents are younger than residents of Lane County, even as county-wide age levels are trending older; (2) Springfield has a growing population of Hispanic/ Latino residents, whose higher average household size is larger than non-Hispanic/ Latino residents. Household type and relationship also has implications for housing needs. For example, one-person households need smaller dwellings than family households with children. Table 5-13 shows household type and relationship in Springfield for 1990, 2000, and the 2005-07 period. The data show an increase in all household types during this period. With respect to share of household types, one- person households increased from 25% to 30% of Springfield households. A cOl-responding decrease in share occurred in two or more person households, with most ofthe decrease in'share coming from married couple family households. Page 40 DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Attachment 1-48 Exhibit A-49 . Table 5-13. Household type and relationship, Springfield, 1990, 2000 and 2005-07 1990 2000 2005-07 ACS Chanae 1990-2005/07 Household Type Number Percent Number Percent Number Percent Number Percent Share 1-person household 4,346 25% 5,206 25% 6,646 30% 2,300 53% 5% 2 or more person household 13,101 75% 15,308 75% 15,707 70% 2,606 20% -5% Family households: 11 ,593 66% 13.479 66% 13,915 62% 2,322 20% -4% Married-couple family 8,572 49% 9,373 46% 9,832 44% 1,260 15% -5% Other family: 3,021 17% 4,106 20% 4,083 18% 1,062 35% 1% Male householder, no wife present 658 4% 1,164 6% 1,017 5% 359 55% 1% Female householder, no husband present 2,363 14% 2,942 14% 3,066 14% 703 30% 0% Nonfamilv households: 1508 9% 1,829 9% 1,792 8% 284 19% -1% Total 17,447 100% 20,514 100"10 22,353 100% 4,906 28% Source: U,S, Census, 1990,2000, American Community Survey (2005-07) Note: 2005-07 American Community Survey is based on pooled data from household surveys conducted in 2005, 2006 and 2007, HOUSING TRENDS Table 5-14 shows the total number of permitted dwellings (single-family and multi-family) by year for selected Lane County cities between 2000 and 2007. Table 5-14 shows that Eugene had the highest number of permitted units during the period, with Springfield and Creswell having the second- and third-highest. Junction City and Oakridge had the lowest number of permitted units. Most cities showed the highest numbers of permitted units over the time period either in 2004 or in 2005, although Springfield's highest total was in 2003. . Table 5-14. Total permitted dwellings (all types) by year, selected Lane County cities, 2000-2007 City 2000 2001 2002 2003 2004 2005 2006 2007 Total Eugene 744 760 828 611 876 1,327 731 ~55 6432 Springfield 274 272 290 324 164 231 211 265 2031 Creswell 26 67 82 93 153 62 56 84 623 Cottage Grove 29 17 28 68 44 86 53 32 357 Junction City 15 12 12 13 10 13 8 78 161 Veneta 11 24 43 96112 117 128 62 593 Oakridge 1 4 1 0 8 4 9 13 40 Total 1,100 1,156 1,284 1,205 1,367 1,840 1,196 1,089 10,237 Source: U,S, Census, Building permits data site, http://censtats,census,gov/bldg/bldgprmt.shtml Note: These numbers a different than those provided by the City of Springfield that were used for the historical density analysis, We believe the data provided by the City are more accurate, Table 5-15 shows the permits issued for new single-family dwellings in selected Lane County cities between 1996 and 2007. Table 5-l5 shows that Springfield's number of permits issued for single-family dwellings remained consistently between 220 and 245 between 1998 and 2003, and has recently fluctuated at lower levels. . DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 41 Attachment 1-49 . . . Exhibit A-50 Table 5-15. Permits issued for new single-family dwellings, selected Lane County cities, 1996-2007 City 1996 1997 1998 1999 2000 2001 2002 2003 2004 Eugene 845 721 665 656 619 633 673 559 583 Springfield N/A 192 221 239 222 225 243 232 128 Coburg 12 9 11 10 3 1 7 6 2 Creswell 30 43 45 32 26 67 80 91 133 Cottage Grove 37 19 54 45 29 17 15 19 34 Junction City 53 19 13 28 15 12 34 13 10 Veneta 13 10 11 19 11 24 43 96 112 Oakridge 5 2 1 12 1 2 1 0 8 TOTAL 995 1,015 1,021 1,041 926 981 1,096 1,016 1,010 2005 2006 756 528 98 134 6 4 60 56 70 39 13 8 117 128 4 9 1,124 906 2007 297 170 1 84 22 78 62 11 725 Source: www.citv-data.com. Table 5-16 shows the total permitted single-family and multifamily dwellings (aggregated) by year between 2000 and 2007 for selected Lane County cities. Table 5-16 shows that Eugene consistently issues permits for the most multi- family units among the cities shown, whereas Oakridge, Veneta, Junction City and Creswell only issue permits for the occasional multifamily unit. Springfield typically issues permits for around 50 multifamily units each year, although it issued permits for 133 units in 2005. Page 42 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-50 . . . Exhibit A-51 Table 5-16. Total permitted single-family and multifamily dwellings (aggregated) by year, selected Lane County cities, 2000-2007 City 2000 2001 2002 2003 2004 2005 2006 2007 Eugene Single family 619 633 673 559 583 756 528 297 Multifamily 125 127 155 52 293 571 203 258 Springfield Single family 222 225 243 232 128 98 134 170 Multifamily 52 47 47 92 36 133 77 95 Coburg Single family N/A N/A N/A N/A N/A N/A N/A N/A Multifamily N/A N/A N/A N/A N/A N/A N/A N/A Creswell Single family 26 67 80 91 133 60 56 84 Multifamily 0 0 2 2 20 2 0 0 Cottage Grove Single family 29 17 15 19 34 70 39 22 Multifamily 0 0 13 49 10 16 14 10 Junction City Single family 15 12 12 13 10 13 8 78 Multifamily 0 0 0 0 0 0 0 0 Veneta Single family 11 24 43 96 112 117 128 62 Multifamily 0 0 0 0 0 0 0 0 Oakridge Single family 1 2 1 0 8 4 9 11 Multifamily 0 2 0 0 0 0 0 2 Source: U.S. Census, Building permits data site, http://censtats.census.gov/bldg/bldgprmt.shtml Figure 5-2 and Table 5-17 show where residents of Springfield worked in 2006. Figure 5-2 and Table 5-17 show that more than 80% of residents of Springfield worked in Lane County, with 26% of Springfield residents working in Eugene and 28% working in Springfield. About 27% of Springfield residents worked in unincorporated Lane County. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 43 Attachment 1-51 . . . Exhibit A-52 ) 001 \ .0 "i... ",-..}__I- -1.;/~ ~. \J ., ...12:.1, ,. , \yJ--- ,(one \ -;Y /.1 ir- ~~ Jj J"l, 0 f r:: \., -j":; ... ..j \... 5 \ -~_., /' ./ Table 5-17. Places where residents of Springfield were employed, 2003 Page 44 ECONorthwest Location Lane County Springfield Eugene Other Lane County Linn County Washington County Multnomah County Marion County Douglas County All Other Locations Total Number Percent 18,706 81% 6,512 28% 6,034 26% 6,160 27% 641 3% 619 3% 488 2% 468 2% 463 2% 1,837 8% 23,222 100% Source: US Census Bureau, LED Origin-Destination Data Base (2nd Quarter 2003) Note: Percent column adds to 101% due to rounding errors The implication of the data presented in this section is that majority of Springfield's workforce lives in Lane County, but many do not reside in the City of Springfield. Residents of Springfield are more likely to work in Eugene than in December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-52 . . . Exhibit A-53 Springfield. This analysis shows that businesses in Springfield have access to the labor force in parts of Lane County. SUMMARY OF KEY DEMOGRAPHIC AND HOUSING TRENDS Springfield has a larger share of young people than Lane County as a whole . Springfield has a higher percentage of people under age 30 than Lane County. · Between 2000 and 2006, Lane County experienced changes in the age structure of its residents. Age groups under age 25 experienced negative growth; the fastest growing age groups were people aged 45 to 64 and 65 and over. This indicates that retirees or people nearing retirement are moving to Lane County; Springfield's share of young people shows that its age structure is experiencing different age trends. Migration is an important component of recent growth in Lane County and will continue to be a key factor in future population growth. · In-migration accounted for 73% of population growth in Lane County between 1990 and 2000 and between 2000 and 2005. · Springfield's population was more mobile than the County's as a whole. Only 41 % of the residents of Springfield lived in the same house in 2000 as they did in 1995 compared to 47% for all of Lane County. A greater share ofthe population in Springfield moved within Lane County during that time period (38%) than for Lane County as a whole (31 %). Single-person households are increasing faster than other household types. . Between 1990 and 2005/07 one-person households increased from 25% to 30% of Springfield households. A corresponding decrease in share occurred in two or more person households, with most of the decrease in share coming from married couple family households Springfield is becoming more ethnically diverse. · Springfield's HispaniclLatino population grew by 168% (2,352 persons) between 1990 and 2000, compared with 111 % growth in Lane County's HispaniclLatino population during the same period. · Other smaller communities near Springfield experienced significant growth in Hispanic/ Latino populations. The communities experiencing the largest increase in the Hispanic/ Latino populations were Eugene (3,792), Junction City (318), Cottage Grove (255), and Creswell (142). DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 45 Attachment 1-53 . . . Exhibit A-54 Hispanic/Latino residents have larger, younger households. . The birth rates for Hispanic/ Latino residents (1998 data) are 24.3 per 1,000 compared to 12.3 per 1,000 for non-Hispanic/ Latino residents. . Hispanic/ Latino residents accounted for 17.4% of births and only 1.4% of deaths in Oregon in 2001. - . In 2000, 75.9% of Hispanic/ Latino Oregonians are under 35 compared to 45.7% of non-Hispanic/ Latino residents. . The average size of a HispaniclLatino household in 2000 in Lane County was 3.2 people, compared with 2.4 people in non-Hispanic households. Household sizes in Springfield were larger: 2.5 for non-Hispanic households and 3.5 for Hispanic/ Latino households. Hispanic/Latino'residents typically have lower incomes but become homeowners at younger ages than non-Hispanic/ Latino residents. . Per capita income in Oregon in 2005 for Hispanic and Latino residents was only 44% of white per capita income/ · 56.7% of Hispanic/ Latino residents of Oregon under age 45 are homeowners, compared to 30.8% of non-Hispanic/ Latino residents Springfield is part of a complex, interconnected regional housing market. · Among selected Lane County cities, Springfield has the third-highest permit average permit valuation for 2005 (behind Coburg and Eugene) and average construction costs for 2005 were highest in Springfield. . · However, median sales prices for Springfield were lower between 1999 and 2007 than median prices in Lane County, and Springfield had the lowest median sales prices in 2007 among all of the selected cities. · Commuting is typical throughout the region: Springfield's workforce lives in Lane County, but many do not reside in the City of Springfield. Since 2000, housing starts in the selected cities within Lane County have been dominated by single-family types. · The data show that new housing development in the 2000-2007 period was predominately single-family housing types. In fact, only 32% of all units for which building permits were issued in the 2000-2007 were for multifamily housing types. · Springfield's number of permits issued for single-family dwellings remained consistently above 220 between 1998 and 2003, and dropped to below 135 per year between 2004 and 2007. Page 46 DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Attachment 1-54 . . . Exhibit A-55 Housing types are trending towards larger units on smaller lots. · Between 1997 and 2007 the median size of new single-family dwellings increased 15%, from 1,975 sq. ft. to 2,277 sq. ft. nationally and 18% in the western region from 1,930 sq. ft. to 2,286 sq. ft. Moreover, the percentage of units under 1,200 sq. ft. nationally decreased from 8% in 1997 to 4% in 2007. The percentage of units greater than 3,000 sq. ft. increased from 15% in 1997 to 26% of new one-family homes completed in 2007. · In addition to larger homes, a move towards smaller lot sizes is seen nationally. Between 1994 and 2007 the percentage oflots under 7,000 sq. ft. increased by 13% from 29% of lots to 33% oflots. A corresponding 4% decrease in lots over 11,000 sq. ft. is seen. · Even when controlling for income and savings, level of education, age, marital status, family size, the housing market in which the unit was located [and other factors], compared to whites both black families and Hispanic families had significantly lower likelihood of homeowners hip, lower house values (for owners) and lower rents (for renters).I6 . Minority households have substantially lower rents than white households. 17 · Hispanic households, particularly low-income families, have higher levels of mortgage debt than do white households, although their house values are lower than whites. This suggests a substantial difference in borrowing or loan terms for Hispanics. 18 IMPLICATIONS OF DEMOGRAPHIC AND HOUSING TRENDS FOR HOUSING NEED The purpose of the analysis thus far has been to give some background on the kinds of factors that influence housing choice, and in doing, to convey why the number and interrelationships among those factors ensure that generalizations about housing choice are difficult and prone to inaccuracies. There is no question that age affects housing type and tenure. Mobility is substantially higher for people aged 20 to 34. People in that age group will also have, on average, less income than people who are older. They are less likely to have children. All of these factors mean that younger households are much more likely to be renters; renters are more likely to be in multi-family housing. 16 Boehm, Thomas P. and Alan M. Schlottmann, "Housing Tenure, Expenditure, and Satisfaction Across Hispanic, African American, and. White Households: Evidence from the American Housing Survey." US Department of Housing and Urban Development, February 2006. 17 Boehm, Thomas P. and Alan M. Schlottmann, "Housing Tenure, Expenditure, and Satisfaction Across Hispanic, African American, and White Households: Evidence from the American Housing Survey." US Department of Housing and Urban Development, February 2006. 18 Boehm, Thomas P. and Alan M. Schlottmann, "Housing Tenure, Expenditure, and Satisfaction Across Hispanic, African American, and White Households: Evidence from the American Housing Survey." US Department of Housing and Urban Development, February 2006. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 47 Attachment 1-55 . . . Exhibit A-56 The data illustrate what more detailed research has shown and what most people understand intuitively: life cycle and housing choice interact in ways that are predictable in the aggregate; age of the household head is correlated with household size and income; household size and age of household head affect housing preferences; income affects the ability of a household to afford a preferred housing type. The connection between socioeconomic and demographic factors, on the one hand, and housing choice, on the other, is often described informally by giving names to households with certain combinations of characteristics: the "traditional family," the "never marrieds," the "dinks" (dual- income, no kids), the "empty nesters."19 Thus, simply looking at the long wave of demographic trends can provide good information for estimating future housing demand. Thus, one is ultimately left with the need to make a qualitative assessment of the future housing market. Following is a discussion of how demographic and housing trends are likely to affect housing in Springfield for the next 20-years: . On average, future housing will look a lot like past housing. That is the assumption that underlies any trend forecast, and one that allows some quantification of the composition of demand for new housing. As a first approximation, the next five years, and maybe the first 10 years, of residential growth will look a lot like the last five years. · If the future differs from the past, it is likely to move in the direction (on average) of smaller units and more diverse housing types. Most of the evidence suggests that the bulk of the change will be in the direction of smaller average house and lot sizes for single-family housing. In summary, smaller households, an aging population, increasing housing costs, and other variables are factors that support the conclusion of smaller and less expensive units and a broader array of housing choices. · No amount of analysis is likely to make the long-run future any more certain: the purpose of the housingforecasting in this study is to get an approximate idea about the long run so policy choices can be made today. It is axiomatic among economic forecasters that any economic forecast more than three (or at most five) years out is highly speculative. At one year one is protected from being disastrously wrong by the shear inertia of the economic machine. But a variety of factors or events could cause growth forecasts to be substantially different. 19 See Planningfor Residential Growth: A Workbookfor Oregon's Urban Areas (June 1997). Page 48 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-56 Exhibit A-57 . STEP 4: DETERMINE THE TYPES OF HOUSING THAT ARE LIKELY TO BE AFFORDABLE TO THE PROJECTED POPULATION BASED ON HOUSEHOLD INCOME Step four of the housing needs assessment results in an estimate of need for housing by income and housing type. This requires some estimate of the income distribution of future households in the community. ECO developed these estimates based on estimated incomes of households that live in Springfield. INCOME AND AFFORDABILlTY OF HOUSING This section summarizes regional and local income trends and housing cost trends. Income is one of the key determinants in housing choice and households' ability to afford housing. A review of historical income and housing price trends provides insights into the local and regional housing markets. Table 5-18 shows a set of inflation adjusted income indicators for Eugene, Springfield and Lane County. The results paint a mixed picture, but generally suggest that income (by most measures) decreased during the I 980s, and increased during the 1990s. Overall, median household and median family incomes remained relatively flat during the 20-year period between 1979 and 1999. The data show that the percentage of persons below the poverty level increased in Springfield and Lane County, and decreased slightly in Eugene between 1979 and 1999. . Table 5-18. Inflation adjusted income indicators (in 1999 dollars), Eugene, Springfield and Lane County, 1979, 1989, and 1999 City Eugene Median HH income Median Family income Per Capita Income % Persons Below Poverty Level Springfield Median HH income Median Family income Per Capita Income % Persons Below Poverty Level Lane County Median HH income Median Family income Per Capita Income % Persons Below Poverty Level Source: U.S. Census. Notes: All dollar amounts in 1999 dollars. 1979 income converted to 1999 dollars using 3.06 inflation factor. 1989 income converted to 1999 dollars using 1.35 inflation factor. . 1979 $34,493 $46,960 $18,029 14.7% $34,248 $38,981 $14.,676 15.2% $37,521 $44,920 $16,837 12.8% Year 1989 $34,248 $46,107 $18,746 17.0% $29,608 $34,332 $13,800 16.5% $34,112 $41,530 $16,970 14.5% 1999 $35,850 $48,527 $21,315 14.4% $33,031 $38,399 $15,616 17.1% $36,942 $45,111 $19,681 17.9% DRAFT: Springfield Housing Needs Analysis Page 49 ECONorthwest Attachment 1-57 December 2009 Exhibit A-58 . A typical standard used to determine housing affordability is that a household should pay no more than 30% of its total monthly household income for housing, including utilities. According to the U.S. Census, nearly 19,000 households in the region-about one-third-paid more than 30% of their income for housing in 2000. One way of exploring the issue of financial need is to review wage rates and housing affordability. Table 5-19 shows an analysis of affordable housing wage and rent gap for households in Springfield at different percentages of median family income (MFI). The data are for a typical family of four. The results indicate that a household must earn about $14.00 an hour to afford a two-bedroom unit according to RUD's market rate rent estimate. Table 5-19. Analysis of affordable housing wage and rent gap by HUD income categories, Eugene-Springfield, 2007 . $15,000 to $24,999 $25,000 to $34,999 $35,000 to $49,999 $50,000 to $74,999 Lane County MFI: $52,200 $75,000 to $99,999 1,066 6% $100,000 to $149,999 573 3% $150,000 or more 188 1% Total 18,865 100% Source: HUD, Oregon office; analysis by ECONorthwest MFI: Median family income 3,254 2,870 3,625 3,476 17% 15% 19% 18% $375 to $625 $625 to $875 $875 to $1,250 $1,250 to $1,875 $1,305 $1,875 to $2,450 $2,450 to $3,750 More than $3,750 $37,500 to $62,500 $62,500 to $87,500 $87,500 to $125,000 $125,000 to $187,500 $130,500 $187,500 to $245,000 $245,000 to $375,000 More than $375,000 Est. Est Number Number o! Owner of Renter Surplus Units Units (Deficit) Notes 33 706 (1,501) 14 825 (735) 2007 HUD FMR studio: $478; 172 6,523 3,441 1 bdnn: $581; 2 bdnn: $654 1,019 959 (892) HUD FMR 2 bdnn: $735 4,791 152 1,318 HUD FMR 3 bdnn: $1028 2,938 42 (496) Income Level Less than $10,000 $10,000 to $14,999 Crude Estimate of Number Affordable Monthly Affordable Purchase of HH Percent Housing Cost Owner-Occupied Unit 2,240 12% $0 to $250 $0 to $25,000 1,574 8% $250 to $375 $25,000 to $37,000 495 9 (563) 133 0 (440) 56 0 (132) 9,650 9,215 0 The total amount a household spends on housing is referred to as cost burden. Total housing expenses are generally defined to include payments and interest or rent, utilities, and insurance. RUD guidelines indicate that households paying more than 30% of their income on housing experience "cost burden" and households paying more than 50% of their income on housing experience "severe cost burden." Using cost burden as an indicator is consistent with the Goal 10 requirement of providing housing that is affordable to all households in a community. Table 5-20 shows housing costs as a percent of income by tenure for Springfield households in 2000. The data show that about 26% of Springfield households experienced cost burden in 2000. The rate was much higher for homeowners (31 %) than for renters (18%). This finding is unusual for Oregon cities-it is much more common for renters to experience higher rates of cost burden. . Page 50 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-58 . . . Exhibit A-59 Table 5-20. Housing cost as a percentage of household income, Springfield, 2000 Owners Renters Total Percent of Income Number Percent Number Percent Number Percent Lest than 20% 4,125 12% 11,965 64% 16,090 30% 20% - 24% 8,852 26% 1,238 7% 10,090 19% 25% - 29% 6,376 19% 1,018 5% 7,394 14% 30% - 34% 4,437 13% 989 5% 5,426 10% 35% - 49% 5,551 16% 1,338 7% 6,889 13% 50% or more 4,988 15% 2,036 11% 7,024 13% Total 34,329 100% 18,584 100% 52,913 100% Cost Burden 10,539 31% 3,374 18% 13,913 26% Severe Cost Burden 4,988 15% 2,036 11% 7,024 13% Source: 2000 Census Table 5-21 shows a rough estimate of affordable housing cost and units by income levels for Springfield in 2000. Several points should be kept in mind when interpreting this data: . Because all of the affordability guidelines are based on median family income, they provide a rough estimate of financial need and may mask other barriers to affordable housing such as move-in costs, competition for housing from higher income households, and availability of suitable units. They also ignore other important factors such as accumulated assets, purchasing housing as an investment, and the effect of down payments and interest rates on housing affordability . · Households compete for housing in the marketplace. In other words, affordable housing units are not necessarily available to low income households. For example, if an area has a total of 50 dwelling units that are affordable to households earning 30% of median family income, 50% of those units may already be occupied by households that earn more than 30% of median family income. The data in Table 5-21 indicate that in 2000: · About 20% of Springfield households could not afford a studio apartment according to RUD's estimate of $478 as fair 'market rent; · Approximately 45% of Springfield households could not afford a two- bedroom apartment at HUD's fair market rent level of$735; · A household earning median family income ($52,200) could afford a home valued up to about $130,500. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 51 Attachment 1-59 Exhibit A-60 . Table 5-21. Rough estimate of housing affordability, Springfield, 2000 Est. Est. Affordable Crude Estimate of Number of Number of Monthly Housing Affordable Purchase Owner Renter Percent Cost Owner-Occupied Unit Units Units 11.9% $0 to $250 $0 to $25,000 33 706 8.3% $250 to $375 $25,000 to $37,000 14 825 Income Level Number ofHH 2,240 1,574 SurpluS (Deficit) -1,501 -735 Notes Less than $10,000 $10,000 to $14,999 $15,000 to $24,999 3,254 17.3% $375 to $625 $37,500 to $62,500 $25,000 to $34,999 2,870 15.2% $625 to $875 $62,500 to $87,500 $35,000 to $49,999 3,625 19.2% $875 to $1,250 $87,500 to $125,000 $50,000 to $74,999 3,476 18.4% $1,250 to $1,875 $125,000 to $187,500 Lane County MFI: $52,200 $1,305 $130,500 $75,000 to $99,999 1,066 5.7% $1,875 to $2,450 $187,500 to $245,000 495 9 -563 $100,000 to $149,999 573 3.0% $2,450 to $3,750 $245,000 to $375,000 133 0 -440 $150,000 or more 188 1.0% More than $3,750 More than $375,000 56 0 -132 Total 18,866 100.0% 9,651 9,215 0 Sources: 2000 Census, HUD Section 8 Income Limits, HUD Fair Market Rent. Based on Oregon Housing & Community Services. Housing Strategies Workbook: Your Guide to Local Affordable Housing Initiatives, 1993. Notes: FMR-Fair market rent 172 1,019 4,791 2,939 6,523 959 152 42 3,441 -893 1,318 -495 2007 HUD FMR studio: $478; 1 bdrm: $581; 2 bdrm: $654 HUD FMR 2 bdrm: $735 HUD FMR 3 bdrm: $1028 The conclusion based on the data presented in Table 5-21 is that in 2000 Springfield had a significant deficit of more than 2,200 affordable housing units for households that earn less than $15,000 annually. Housing prices have increased significantly in the past five years; the affordability gap for lower income households has probably increased considerably. The next section examines changes in housing cost since 2000. . Changes in housing cost According to the Office of Federal Housing Enterprise Oversight, the average sales price of a single-family home in the Eugene-Springfield MSA increased 229% between 2000 and 2006. A key concern expressed by the City was that the housing needs analysis and runs of the HCS housing needs model reflect recent trends in the regional housing market. To quantify these trends, ECO analyzed data from two sources: (1) sales data from the Lane County Assessor; and (2) rental data from Duncan & Brown, an Eugene-based real estate analysis firm that conducts rent surveys for the Metropolitan Region. The sales database provided to ECO by the City of Springfield included 34,680 property sales.2o For purposes of comparison, the database included Creswell, Cottage Grove, Eugene, Junction City, Springfield, and Veneta. Table 5-22 shows sales prices for single-family dwellings for Lane County and Springfield between 1999 and 2006. Table 5-22 shows that Springfield median sales prices have been lower than median sales prices in Lane County over the entire time period. Median sales prices also increased at a slower rate in Springfield; percent change in median sales prices between 1999 and 2006 for Lane County was 73%; in Springfield it was 64%. Sales prices for single-family dwellings peaked in 2007 and had declined to about $175,000 by the first quarter of 2009. . 20 The sales data was obtained through queries of the Regional Land Information Database (www.rlid.org). Page 52 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-60 Exhibit A-61 . Table 5-22. Sales price for single-family dwellings, Lane County and Springfield, 1999-2006 Lane County Springfield Average Median Average Median Sales Sales Sales Sales Year # of Sales Price Price # of Sales Price Price 1999 3,940 140,564 127,900 843 118,520 112,745 2000 3,171 144,142 129,900 687 119,152 112,750 2001 3,808 149,252 133,000 881 122,700 118,450 2002 4,291 156,603 138,165 886 129,432 121,900 2003 4,761 168,780 149,000 1,042 135,719 128,000 2004 5,092 183,497 162,500 1,112 149,082 137,900 2005 5,326 222,835 194,000 1,157 177,260 165,000 2006 4,291 249,438 221,000 973 201,000 185,000 Change 1999-2006 Number 351 108,874 93,100 130 82,480 72,255 Percent 9% 77% 73% 15% 70% 64% Source: RLlD, Analysis by ECONorthwest . Table 5-23 shows the average and median sales prices for single-family dwellings in selected Lane County cities between 1999 and 2006. Table 5-23 shows that median sales prices increased throughout the county during this period. In 2006, the highest median sales prices were in Eugene, the rest of the county, and Creswell. Lowest median sales prices in 2006 were in Springfield and Junction City. Prices increased the most in Creswell (87%) and Eugene (80%). Prices increased the least in Springfield (64%) and Junction City (67%). Table 5-23. Average and median sales price, single-family dwellings, Lane County cities, 1999-2006 Year Increase (1999-2006) City 1999 2000 2001 2002 2003 2004 2005 2006 Dollars Percent Median Sales Price Cottage GrOl.e 112,000 103,500 109,750 110,000 120,000 128,000 157,000 195,000 83,000 74% Creswell 112,500 118,000 109,000 121,750 125,000 142,500 180,750 210,500 98,000 87% Eugene 136,900 140,000 143,500 149,900 163,000 179,900 215,000 247,000 110,100 80% Junction City 113,250 112,500 115,150 119,638 120,750 138,000 162,000 189,000 75,750 67% Springfield 112,745 112,750 118,450 121,900 128,000 137,900 165,000 185,000 72,255 64% Veneta 115,250 110,000 112,000 119,950 126,500 139,500 173,635 200,000 84,750 74% Rest of County 111,000 108,750 110,000 121,250 127,750 160,000 212,500 216,000 105,000 95% Average Sales Price Cottage Grol.e 118,112 106,767 113,150 116,152 122,298 134,854 168,828 193,157 75,045 64% Creswell 115,662 121,697 114,497 130,475 129,891 162,095 200,008 223,307 107,645 93% Eugene 152,872 159,920 165,366 173,351 188,484 202,750 246,272 275,674 122,802 80% Junction City 120,218 116,282 120,164 131,761 130,170 149,294 169,287 191,574 71,356 59% Springfield 118,520 119,152 122,700 129,432 135,719 149,082 177,260 201,000 82,480 70% Veneta 121,039 111,754 111,961 118,976 134,297 148,313 178,916 213,220 92,181 76% Rest of County 124,741 120,724 136,013 134,572 152,744 181,894 234,178 246,311 121,570 97% Source: RLlD, Analysis by ECONorthwest . DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 53 Attachment 1-61 . . . Exhibit A-62 Table 5-24 shows the median contract rent for Lane County cities. The highest median contract rents from the 2000 Census were in Eugene and Springfield. The lowest median contract rents were in Oakridge and Creswell. Table 5-24. Median contract rent, Lane County cities, 1999 Location Rent Eugene $ 566 Springfield $ 518 Veneta $ 502 Coburg $ 498 Junction City $ 491 Cottage Grove $ 456 Creswell $ 417 Oakridge $ 384 Source: US Census 2000 Vacancy rates have generally decreased in Eugene-Springfield rental market since 2000. Vacancy rates for studio, 1- and 2-bedroom apartments all decreased from between 4.1-4.7% to between 1.1-2.1 % between fall 2000 and 2006. Apartment rents have remained relatively stable, increasing between 4% and 10% between 2000 and 2005.21 Table 5-25 shows average monthly cost of rental units in Springfield for the 2000 to 2005 period. Rental units were separated into two categories: (1) units built prior to 1988 and (2) units built since 1988. The majority of Springfield's units were built prior to 1988. Rents increased based on the number of bedrooms. Rents ranged from $392 for a studio unit in 2000 to $646 for a three-bedroom unit in 2004. Rents for units with a similar number of bedrooms were higher for newer units. For instance, the average rental cost of a two-bedroom unit built prior to 1988 was $529 compared to $620 for a two-bedroom unit built since 1988, a difference of $91 per month. Over the six-year period, rents increased by between $19 and $56 per month. Monthly rental costs of two-bedroom units had the largest increases, $34 per month for older units and $56 per month for newer units. Rent for studio, one- bedroom, and three-bedroom units increased all increased by about $20 per month. 21 Duncan & Brown Apartment Report. Fall 2000-Fall 2006. Daniel J. Puffinburger, Corey S. Dingman, Duncan & Brown Real Estate Analysts Page 54 DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Attachment 1-62 Exhibit A-63 . Table 5-25. Average rental monthly costs by unit type, Springfield, 2000 to 2005 Units Built Prior to 1988 Units Built Since 1988 One Two Three One Two Three Year Studio Bedroom Bedrooms Bedrooms Studio Bedroom Bedrooms Bedrooms 2000 $392 $428 $514 $594 -- - $588 - 2001 $394 $423 $523 $601 -- - $583 -- 2002 $389 $431 $526 $619 -- $575 $615 -- 2003 $386 $438 $531 $600 $550 $550 $642 -- 2004 $388 $437 $533 $633 -- $575 $646 -- 2005 $414 $447 $548 $615 -- $575 $644 -- Change 2000 to 2005 Amount $22 Percent 5.6% AAGR 1.10% $19 4.4% 0.87% $34 6.6% 1.29% $21 3.5% 0.70% $56 9.5% 1.84% Source: Duncan & Brown Apartment Rent Report, 2000 to 2005; Calculations by ECONorthwest Note: Blank values indicate that there were too few units in the survey to include in the summary. Table 5-26 shows a comparison of change in rental costs during the 2000 to 2005 period for Springfield and Eugene. Rental costs were higher in Eugene than in Springfield. The difference in rental costs for all units, regardless when they were built, ranged from $39 per month for a studio unit to $211 per month for a three-bedroom unit, increasing with the number of bedrooms. . The difference in average rental costs was greater for newer and larger units. Newer one-bedroom units cost an average of$74 per month more to rent in Eugene than Springfield. Newer two-bedroom units cost an average of $166 more to rent in Eugene than Springfield. Table 5-26. Comparison of average rental monthly costs by unit type, Springfield and Eugene, 2000 to 2005 One Two Three Studio Bedroom Bedrooms Bedrooms Springfield Built prior to 1988 $394 $434 $529 $610 Built since 1988 $569 $620 All rentals $416 $488 $574 $610 Eugene Built prior to 1988 $400 $483 $611 $719 Built since 1988 $623 $645 $786 $924 All rentals $456 $564 $699 $822 Difference (Eugene minus Springfield) Built prior to 1988 $6 $49 $82 $109 Built since 1988 $76 $166 All rentals $40 $74 $124 $211 Source: Duncan & Brown Apartment Rent Report, 2000 to 2005; Calculations by ECONorthwest Note: Blank values indicate that there were too few units in the survey to include in the summary. . Figure 5-4 shows a comparison of change in average rental costs and average sales price in Springfield between 2000 and 2005. Over the five-year period average sales price increased by 46%, compared to a 7% change in average rental DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 55 Attachment 1-63 1---- Exhibit A-64 . costs. The greatest increases in average sales price occurred since 2003, while average rental costs remained relatively flat since 2003. Since 2005, average sales prices have continued increasing at a faster rate than average rental costs. The increase in average sales price in Springfield between 2005 and 2006 was about 13%. According to the Fall 2006 Duncan & Brown Apartment Report, changes in average rental costs in Springfield were comparable to increases in recent years.22 Figure 5-4. Comparison of annual change in average rental costs and average sales price, Springfield, 2000 to 2005 20.0% 18.0% 16.0% Q) 14.0% l:l'l c ca .c 12.0% 0 ... c 10.0% Q) (J .. Q) 8.0% ll.. iii 6.0% :l c c 4.0% . <( 2.0% 0.0% 2001 2002 2003 Year 2004 2005 -Average Rental Costs ~ Average Sales Price Source: Duncan & Brown Apartment Rent Report, 2000 to 2005; RUD; Calculations by ECONorthwest The analysis of housing starts, sales prices, and rents presented in this section leads us to several conclusions: · The housing market peaked in 2007 and sales prices declined in 2008 and the first quarter of2009. Springfield single-family housing starts have declined since 2003. The overall number of permits for new single-family residences issued regionwide has remained remarkably stable; . 22 The Fall 2006 Duncan & Brown Apartment Report did not present average rent by unit type like they did in previous reports. As a result, we were not able to include 2006 average rents in this analysis. Page 56 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-64 . Exhibit A-65 . . New construction costs are higher than regional averages. Springfield's permit valuations and construction costs have generally been on or near the middle or towards the high end compared with selected Lane County cities; . Price increases are lower than in other cities. Springfield's median sales prices for single-family dwellings have increased the smallest amount compared with selected Lane County cities; . Single-family development has dominated new construction. Multi-family dwelling units do not make up a high percentage of units constructed in Springfield and other selected Lane County cities; . Sales prices increased much faster than rental rates. Over the five-year period between 2000 and 2005 average sales price increased by 46%, compared to a 7% change in average rental costs. . The implications of the data shown above are that ownership costs increased much faster than rents and incomes, but declined as the housing bubble burst in 2008. Table 5-27 underscores this trend for the Eugene-Springfield MSA.2J Between 1990 and 2000, incomes increased about 46% while median owner value increased 115%. Rents increased 44%--about the same as incomes. Since 2000, the data show housing costs have increased faster than incomes. The owner values include all units in the MSA; the sales data presented earlier in this section suggest that owner costs have increased much faster than the Census data suggest. Finally, the results show that the median owner value was 2.6 times median household income-a figure that increased to 4.7 by 2005. Table 5-27. Comparison of income, housing value, and gross rent, . Eugene-Springfield MSA, 1990, 2000, and 2005 Indicator Median HH Income Median Family Income Median Owner Value Median Gross Rent Percent of Units Owned Housing Value/Income Median HH Income 2.6 3.8 4.7 Median Family Income 2.1 3.1 3.5 Source: U.S. Census of Population and Housing, 1990 and 2000; American Community Survey, 2005 1990 $25,268 $30,763 $65,600 $418 61% 2000 $36,942 $45,111 $141,000 $604 62% Change 1990-2000 2000-2005 2005 $37,290 $49,555 $173,600 $683 63% 46% 47% 115% 44% 1% 10% 23% 13% In summary, the data indicate that homeownership is increasingly expensive in Springfield and that the cost of homeownership is prohibitive for low- and . 23 2005 data from the American Community Survey is not available for Springfield. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 57 Attachment 1-65 Exhibit A-66 . moderate-income households. The data indicate that homeownership rates in the Metropolitan area and Springfield have increased, despite the rapid increase in sales prices. This is probably due in large part to a much broader array of financing options available to households than existed previously. STEP 5: ESTIMATE THE NUMBER OF ADDITIONAL NEEDED UNITS BY STRUCTURE TYPE AND TENURE24 Step five of the housing needs assessment results in an estimate of need for housing by income and housing type. This requires some estimate of the income distribution of future households in the community. ECO developed these estimates based on (1) secondary data from the Census, and (2) analysis by ECONorthwest. The next step in the analysis is to relate income levels to tenure and structure type. Table 4-3 showed tenure by structure type from the 2000 Census. Table 5- 28 shows an estimate of needed housing by st~cture type and tenure for the 2010- 2030 planning period. The housing needs analysis suggests that a higher percentage of multifamily units will be needed, thus, the housing mix changes from approximately 63% single-family/37% multifamily during the 1999-July 2008 period to 60% single-family/40% multifamily.25 The housing needs analysis also suggests the City will see a higher rate of homeownership in the future. Thus, the tenure split is increased from 54% owner-occupied/46% renter occupied to 57% owner-occupied/43% renter occupied. . Table 5-28. Estimate of needed dwelling units by type and tenure, Springfield, 2010-2030 Owner-Occupied Renter-Occupied Total Housing Type New DU Percent New DU Percent New DU Percent Needed Units, 2010-2030 Single-family types Single-family detached 2,756 81% 353 14% 3,109 52% Manufactured in Parks 54 2% 6 0% 60 1% Single-family attached 343 10% 75 3% 419 7% Subtotal 3,153 92% 435 17% 3,587 60% Multi-family Multifamily 256 8% 2,136 83% 2,392 40% Subtotal 256 8% 2,136 83% 2,392 40% Total 3,409 100% 2,571 100% 5,980 100% . 24 Note: Manufactured dwellings are a permitted use in all residential zones that allow 10 or fewer dwellings per net buildable acre. As a result, Springfield is not required to estimate the need for manufactured dwellings on individual lots per OAR 660-024-0040 (7) (c). 25 Single-family attached dwellings typically achieve densities closer to multifamily housing types. Ifthese higher density housing types are included with multifamily, the housing mix is 53% lower density, and 47% higher density types. Page 58 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-66 . . . Exhibit A-67 The analysis (Table 5-28) indicated that Springfield needs 5,980 new dwelling units for the 2010-2030 period. The next step in estimating units by structure type is to evaluate income as it relates to housing affordability. Table 5-29 shows an estimate of needed dwelling units by income level for the 2010-2030 period. The analysis uses market segments consistent with HUD income level categories. The analysis shows that about 49% of households in Springfield could be considered high or upper-middle income in 2007 and that about 49% of the housing need in the 2010-2030 period will derive from households in these categories. Table 5-29. Estimate of needed dwelling units by income level, Springfield, 2010-2030 Market Segment by Income High (120% or more of MFI) Financially Attainable Products Number of Percent of Owner- Renter- Households Households occupied occupied 1,822 30% All housing All housing types; higher types; higher prices prices 1,141 19% All housing All housing types; lower types; lower values values 1,296 22% Manufactured on Single-family lots; single- attached; family attached; detached; duplexes manufactured on lots; apartments Primarily New Housing Primarily Used Housing Income ra nge $68,640 or more 1 Upper Middle (80%- $45,760 to 120% of MFI) $68,640 Lower Middle (50%- $28,600 to 80% of MFI $45,760 Low (30%-50% or less of MFI) $17,160 to $28,600 Apartments; manufactured in parks; duplexes Apartments; new and used gOl.ernment assisted housing 13% 756 Manufactured in parks Very Low (Less than 30% of MFI) Less than $17,160 965 16% None Source: ECONorthwest STEP 6: DETERMINE THE NEEDED DENSITY RANGE FOR EACH PLAN DESIGNATION AND THE AVERAGE NEEDED NET DENSITY FOR ALL DESIGNATIONS This section summarizes the forecast of needed housing units in Springfield for the period 2010-2030. Table 5-30 shows the forecast of needed housing units in Springfield for the period 2010-2030. Springfield makes the following findings in support of the density assumptions used in Table 5-30: · Springfield had an average residential density of 6.6 dwelling units per net acre or about 6,600 square feet of land per dwelling unit between 1999 and 2008 (Table 4-5). Average single-family detached density was 5.4 units per net acre. Manufactured homes averaged 4.6 dwelling units per net DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 59 Attachment 1-67 . . . Exhibit A-68 acre, while all multifamily housing types averaged 11.1 dwelling units per net acre. National homeownership rates increased to nearly 70% in 2006 before declining as the housing bubble burst. The homeownership rate in Springfield in 2000 was considerably lower at 54%. It is the policy of the City to provide homeownership opportunities to Springfield residents. . National trends are towards larger units (both single-family and multifamily) on smaller lots. . . More than 28% of dwelling units in Springfield in 2000 were multifamily types. . The "needed" density for single-family dwellings in the housing needs analysis is 5.5 dwelling units per net acre. This assumption is a slight increase overthe historical density of 5.4 dwellings per net acre for single- family detached units. Increasing the average density of single-family detached dwellings should result in the provision of more affordable single-family detached units as a result of decreased lot sizes. . Topography, lot configurations, and other factors typically reduce land use efficiency. The achieved density may be lower for single-family detached dwellings in areas with slopes. · The City assumes an average multifamily density of 18.0 dwellings per net acre or a land area of about 2,420 square feet per dwelling unit. This assumption is an increase of about 62% over historical density of II. I dwellings per net acre for all multifamily types. · The City assumes an average density for all housing types of7.9 dwelling units per net acre. This is an increase of about 20% over the historical density of 6.6 dwelling units per net acre. In summary, the City assumes that average densities will increase significantly (by about 20% over average historical densities) during the planning period, that ownership rates will increase, and that an increasing percentage of households will choose single-family attached housing types. These assumptions are consistent with the housing needs analysis presented in this chapter. These findings support the City's overall density assumption of7.9 dwelling unit per net acre. The forecast indicates that Springfield will need about 752 net residential acres, or about 927 gross residential acres to accommodate new housing between 2010 and 2030. The forecast results in an average residential density of7.9 dwelling units per net residential acre and of 6.3 dwelling units per gross residential acre. This represents a 20% increase in density over the historical average of 6.6 dwelling units per net acre. Page 60 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-68 Exhibit A-69 Table 5-30. Forecast of new dwelling units and land needed by type, . Springfield 2010-2030 Density Net to Gross Density (DUlnet Net Res. Gross Res. (DUlgross Housing Type New DU Pe rce nt res ac) Acres Factor Acres res ac) Needed Units, 2010-2030 Single-family types Single-family detached 3,109 52% 5.5 565 20% 707 4.4 Manufactured in parks 60 1% 8.0 7 18% 9 6.6 Single-family attached 419 7% 9.0 47 15% 55 7.7 Subtotal 3,588 60% 5.8 619 770 4.7 Multi-family Multifamily 2,392 40% 18.0 133 15% 156 15.3 Subtotal 2,392 40% 18.0 133 156 15.3 Total 5,980 100% 7.9 752 927 6.5 Source: ECONorthwest . Table 5-31 provides an allocation of housing units by Springfield's three residential plan designations. Dwelling units were allocated to plan designations based, in part, on historic development trends within each plan designation and on the type of development allowed in each plan destination. Table 5-31 also provides an estimate of the gross acres required in each designation to accommodate needed housing units for the 2010-2030 period. The acreages are based on the gross density assumptions shown in Table 5-30. The residential land needs presented in Table 5-31 may change based on policy decisions related to land use efficiency measures, which may result in increased or decreased land need. Based on the housing needs analysis, dwellings have been allocated by plan designation and type: . The overall needed housing mix is 60% single-family (including manufactured and single-family attached units) and 40% multifamily. · The density assumptions increase by plan designations as shown in Table 5-30. . · Fifty-eight percent of needed dwelling units will locate in the Low Density residential designation, which allows single-family detached and manufactured homes. This designation also allows duplex, single-family attached, and some multifamily dwellings in conjunction with discretionary review. · Thirty percent of needed dwellings will locate in the Medium Density residential designation, which allows single-family detached, single- family attached, manufactured home parks, townhomes, duplexes, and multifamily dwellings. . Twelve percent of needed dwelling units will locate in High Density or Mixed-Use residential designations, which allow single-family detached, DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 61 Attachment 1-69 . . . Exhibit A-70 townhomes, manufactured (single detached and manufactured home parks), duplexes, and multifamily. . Manufactured units in parks will locate in the Low-Density plan designation. Table 5-31. Allocation of needed housing units by plan designation, Springfield 2010-2030 Plan Designation High Densityl Low Density Medium Density Mixed-Use Total Housing Type DU Gross Ac DU Gross Ac DU Gross Ac DU Gross Ac Single-family Single-family detached 3,229 734 0 - 0 - 3,229 734 Manufactured in parks 60 9 0 - 0 - 60 9 Single-family attached 179 23 299 39 0 - 478 63 Subtotal 3,468 766 299 39 0 - 3.767 806 Multi-family Multi-family 0 - 1 ,495 109 718 36 2,213 145 Subtotal 0 - 1 ,495 109 718 36 2,213 145 Total 3,468 766 1,794 148 718 36 5,980 950 Percent of Acres and Units Single-family Single-family detached 54%" 77% 0%" 0% 0%" 0% 54% 77% Manufactured in parks 1%" 1% 0%" 0% 0%" 0% 1% 1% Single-family attached 3%" 2% 5%" 4% 0%" 0% 8% 7% Subtotal 58%" 81% 5%" 4% 0%" 0% 63% 85% Multi-family Multi-family 0%" 0% 25%" 11% 12%" 4% 37% 15% Subtotal 0%" 0% 25%" 11% 12% " 4% 37% 15% Total 58%" 81% 30%" 16% 12%" 4% 100% 100% Source: ECONorthwest In addition to the housing types shown in Table 5-31, Springfield needs to plan for additional group quarters. The analysis assumes the City will add 145 persons in group quarters between 2010 and 2012.The City will need to add a similar number of group quarter units during this period. Assuming that group quarters achieve densities comparable to multifamily units, the City will need approximately nine gross residential acres for these units (145 divided by 15.3 units per gross acre). The majority of these units will probably be residential care facilities which are permitted as a discretionary use in the Low Density residential designation and a special use in the Medium- and High-Density designations. Page 62 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-70 . . . Exhibit A-71 Chapter 6 Comparison of Sulm!Y and Demand This chapter summarizes from data and analysis presented in Chapters 2 through 5 to compare "demonstrated need" for vacant buildable land with the supply of such land currently within the Springfield UGB and city limits. Chapter 2 described the policy framework, Chapter 3 described land supply, Chapter 4 described historical development patterns, and Chapter 5 described residential land needs. The following section estimates land needed for other uses; the chapter concludes with a comparison of land supply and land demand for the 20 10-2030 time period. TOTAL RESIDENTIAL LAND NEED, 2010-2030 This section estimates total residential land need for the period between 2010 and 2030. In additional to land needed for new residential units, it .estimates land needed for parks, public facilities, and other semi-public uses to arrive at an estimate of total need for land designated for residential purposes. LAND NEEDED FOR NEW RESIDENTIAL DWELLING UNITS Chapter 5 presented estimates of land needed for new residential dwellings (see Tables 5-30 and 5-31). Table 6-1 summarizes land needed for new housing by plan designation for the 2010-2030 period. Note that group quarters is a separate category that can locate in any plan designation. , Table 6-1. Land needed for new housing by plan designation, Springfield UGB, 2010-2030 . Plan Designation DU Gross Ac Low-Density Residential 3,468 766 Medium-Density Residential 1,794 148 High-Density Residential/Mixed-Use 718 36 Group Quarters 145 9 Total 6,125 959 Source: Table 5-31 LAND NEEDED FOR OTHER USES Cities need to provide land for uses other than housing and employment. Public and semi-public facilities such as schools, hospitals, governments, utilities, churches, parks, and other non-profit organizations will expand as population increases. Many communities have specific standards for parks. School districts typically develop population projections to forecast attendance and need for DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 63 Attachment 1-71 . . . Exhibit A-72 additional facilities. All of these uses will potentially require additional land as a city grows. This section considers other uses that consume land and must be included in land demand estimates. Demand for these lands largely occurs independent of market forces. Many can be directly correlated to population growth. For the purpose of estimating land needed for other uses, these lands are classified into three categories: . Lands needed for public operations and facilities. This includes lands for city offices and maintenance facilities, schools, state facilities, substations, and other related public facilities. Land needs are estimated using acres per 1,000 persons for all lands of these types. · Lands needed for parks and open space. The estimates use a parkland standard of 14 acres per 1,000 persons based on the level of service standard established in the Willamalane Park and Recreation Comprehensive Plan, which projected need for parkland in Sprin~field between 2002 and 2022. · Lands needed for semi-public uses. This includes hospitals, churches, non- profit organizations, and related semi-public uses. The analysis includes land need assumptions using acres per 1,000 persons for all lands of these types. Table 6-2 shows land in public and semi-public uses by type. The data show a total of 1,636 acres in public and semi public uses in the Springfield UGB in 2009. This equates to 24.8 acres per 1,000 persons. Table 6-2. Summary of public and semi-public land need by type, Springfield UGB, 2010-2030 Type of Use Govemment Utilities Parks Schools Church/Charities/Other Total Acres 581 134 563 277 81 1,636 Acres / 1000 Persons 8.8 2.0 8.5 4.2 1.2 24.7 Assumed Need (Ac/1000 Pe rsons) 3.0 2.0 14.0 0.9 1.2 21.1 Estimated Acres 2010. 2030 44 30 357 14 18 463 Source: City of Springfield GIS data; analysis by ECONorthwest Table 6-2 shows that there will be an additional need of about 463 acres of land for all new public and semi-public uses or 21.1 acres per 1,000 people between 2010 and 2030. The information in Table 6-1 is based on the following assumptions: Page 64 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-72 Exhibit A-73 . · Government land in 2007 includes a 271-acre site that is owned by the Bureau of Land Management (BLM) and the liS-acre Booth-Kelly mixed-use site. Not including these sites, Springfield has 195 acres of government land or 3.0 acres per 1,000 people. The assumed land need for 2010 to 2030 is 3.0 acres per 1,000 people, assuming that the City's land need will not include more sites like the BLM or Booth- Kelly site. . Park land needs are based on the level-of-service established in Willamalane's parks plan of 14 acres per 1,000 persons, which will require 207 new acres of parkland. In addition, park land includes need for 150 acres of parkland for need identified in the Park and Recreation Comprehensive Plan and to serve residents that moved to Springfield between 2002 and 2008.26 · School land needs are based on the fact that the Springfield School District will need to add one 14 acre site in the Jasper-Natron area over the planning period. 27 The land need of 0.9 acres per 1,000 persons was based on population growth and the District's need for one 14 acre site. . Land needs for utilities, recreation, and churches/charities/other are based on maintaining the same ratio of acre to population as currently exists for these land uses. BUILDABLE LAND INVENTORY AND CAPACITY . The capacity of residential land is measured in dwelling units and is dependent on densities allowed in specific zones as well as redevelopment potential. In short, land capacity is a function of buildable land and density. The buildable lands inventory indicates that Springfield has about 1,447 acres of vacant and partially-vacant residential land and an additional 21 acres in the Glenwood mixed-use refinement plan area (these acres were included in the commercial and industrial lands inventory and are included here only for the purpose of estimating residential capacity). 28 This yields a total of 1,468 buildable acres. 26 According to Greg Hyde, the Planning and Development Manager with the Willamalane Park & Recreation District, Springfield has acquired 37 acres of park land between 2002 and 2008. The Park and Recreation Comprehensive Plan identified a deficit of 130 acres to serve population in 2002 (at the 14 acres per 1,000 person level of service). That deficit was reduced to 93 acres with the addition of the 37 acres of parkland. In addition, Springfield's population grew by 4,095 people between 2002 and 2008, resulting in an additional need for 57 acres of parkland. Together, Springfield has a need for 150 acres of parkland to serve the City's population in 2008 at the 14 acres per 1,000 person level of service. 27 According to Jeff DeFranco, the Springfield Public Schools Director of Communications and Facilities, the school district has one 14- acre site that will be sold (the Rainbow (Chase) Property). The City owns a 65-acre site in East Springfield has no services. The District owns a IS-acre site in the Clear Water area that is outside of the UGB, which will be developed when there is more residential development in the area. . 28 Capacity in the Glenwood mixed-use area was calculated as follows: 21 buildable acres (45% of the 47-acre site; the policy requires 30% to 60% of the site be used for housing) multiplied by 15 dwelling units per gross acre equals 317 dwelling units, minus 47 dwelling units that would be displaced from the River Bank Mobile Home Park equals 270 dwelling units. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 65 Attachment 1-73 Exhibit A-74 . Table 6-3 provides an estimate of how much housing could be accommodated by those lands based on the needed densities identified in Table 5-30 after making deductions for development constraints. It includes capacity for areas with approved master plans that were not included in the acreage estimates. This includes Marcola Meadows (518 dwellings in the MDR designation) and RiverBend (730 dwellings in the MDR designation). Total residential capacity includes capacity for redevelopment, which is assumed as 5% of needed new dwellings, or 299 dwellings. The basis for this assumption is presented in Chapter 4. Table 6-3 shows that Springfield has capacity for 9,021 dwelling units within the existing U GB. Table 6-3. Estimated residential development capacity, Springfield UGB, 2009 . Residential Buildable Capacity Plan Designation Acres (DU) Low Density Residential 824 5,379 Medium Density Residential 95 2,718 High Density Residential 16 355 Mixed-Use (Glenwood) 21 270 Rede\elopment na 299 Total 956 9,021 Source: City of Springfield residential BLI; analysis by ECONorthwest Note: Estimated residential development capacity includes sites with approved master plans (RiverBend - 730 DU and Marcola Meadows - 518 DU. All of this capacity is in the Medium Density Residential plan designation). Percent of Capacity 60% 30% 4% 3% 3% 100% COMPARISON AND CONCLUSIONS Table 6-4 shows the capacity for residential development by plan designation. It also shows an estimate oflands needed for other uses (e.g., parks, schools, churches, etc.). ECO estimates Springfield will need 463 acres for other uses during the 2010-2030 period. The results lead to the following findings: . Springfield has an overall surplus of residential land. The Springfield UGB has enough land for 9,021 new dwelling units. The housing needs forecast projects a need for 5,980 dwelling units and 145 group quarter dwellings. · The Low Density Residential designation has a surplus of approximately 72 gross acres. · The Medium Density Residential designation has a surplus of approximately 18 gross acres. . · The High Density Residential designation has a deficit of approximately 34 gross acres. Page 66 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-74 . . . Exhibit A-75 . The total residential land surplus is 59 gross acres. Table 6-4. Residential capacity for needed dwelling units by plan designation, Springfield UGB, 2010-2030 1 2 3 4 5 6 7 8 9 Housing Housing Total Land Surplusl Surplusl Surplusl Needed Need Deficit Other Deficit Need Capacity Deficit Density (Gross (Gross Residential (Gross Plan Designation (DU) (DU) (DU) (DUlGRA) Acres) Ac) Land Need Ac) Low Density Residential 3,468 5,379 1,911 5 -422 422 347 75 Medium Density Residential 1,794 3,137 1,343 12 0 111 93 18 High Density Residential 718 505 -213 20 11 -11 23 -34 Total 5,980 9,021 3,041 0 -411 522 463 59 Source: ECONorthwest Column Notes: 1. Plan designations 2. Needed dwellings by plan designation (table 5-30) 3. Capacity by plan designation (table 6-2); Note: MDR capacity includes capacity in master planned areas (Glenwood, Marcola Meadows, Riverbend); MDR and HDR includes capacity for redevelopment. 4. Capacity (column 3) minus Need (column 2); Note: a positive number denotes enough capacity within the existing UGB 5. Needed Gross Density (from bottom of page 5) 6. Total additional land needed (if a deficit exists). Equals -column 4 divided by column 5 7, Surplus/deficit gross acres (negatives mean a UGB expansion). Equals Column 4 divided by Column 5 8. Other residential land need (land needed for parks, etc) 9. Total surplus/deficit. EqualS column 7 minus column 8. Note: Total Surplus/Deficit (column 9) adds to 344 acres due to rounding errors. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 67 Attachment 1-75 . . . Exhibit A-76 Appendix A Context for Assessing Housing Needs WHAT IS AFFORDABLE HOUSING? The terms "affordable" and "low-income" housing are often used interchangeably. These terms, however, have different meanings: . Affordable housing refers to households' ability to find housing within their financial means. Households that spend more than 30% of their income on housing and certain utilities are considered to experience cost burden. 29 As such, any household that pays more than 30% experiences cost burden and does not have affordable housing. Thus, affordable housing applies to all households in the community. · Low-income housing refers to housing for "low-income" households. HUD considers a household low-income if it earns 80% or less of median family income. In short, low-income housing is targeted at households that earn 80% or less of median family income. These definitions mean that any household can experience cost burden and that affordable housing applies to all households in an area. Low-income housing targets low-income households. In other words, a community can have a housing affordability problem that does not include only low-income households. It is important to underscore the point that many households that experience cost burden have jobs and are otherwise productive members of society. A household earning 80% of median family income in Springfield earns about $39,000 annually-{)r about $18.50 per hour for a full-time employee. The maximum affordable purchase price for a household earning $39,000 annually is about $120,000. Depending on household size, many of these households are eligible for government housing assistance programs. In summary, any household can face housing affordability problems. Because they have more limited financial means, the incidence of cost burden is higher among low-income households. Statewide planning Goal 10 requires cities to adopt policies that encourage housing at price ranges commensurate with incomes. In short, state land use policy does not distinguish between households of different income levels and requires cities to adopt policies that encourage housing for all households. 29 Cost burden is a concept used by HUD. Utilities included with housing cost include electricity, gas, and water, but do not include telephone expenses. Page 68 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-76 Exhibit A-77 . WHAT OBJECTIVES DO HOUSING POLICIES TYPICALLY TRY TO ACHIEVE? The Practice of State and Local PlanningO classifies goals that most government housing programs address into four categories: · Community life. From a community perspective, housing policy is intended to provide and maintain safe, sanitary, and satisfactory housing with efficiently and economically organized community facilities to service it. In other words, housing should be coordinated with other community and public, services. Although local policies do not always articulate this, they are implicit in most local government operations. Comprehensive plans, zoning, subdivision ordinances, building codes, and capital improvement programs are techniques most cities use to manage housing and its development. Local public facilities such as schools, fire and police stations, parks, and roads are usually designed and coordinated to meet demands created by housing development. · Social and equity concerns. The key objective of social goals is to reduce or eliminate housing inadequacies affecting the poor, those unable to find suitable housing, and those discriminated against. In other words, communities have an obligation to provide safe, satisfactory housing opportunities to all households, at costs they can afford, without regard to income, race, religion, national origin, family structure, or disability. . · Design and environmental quality. The location and design of housing affect the natural environment, residents' quality of life, and the nature of community life. The objectives of policies that address design and environmental quality include neighborhood and housing designs that meet: household needs, maintain quality of life, provide efficient use of land and resources, reduce environmental impacts, and allow for the establishment of social and civic life and institutions. Most communities address these issues through local building codes, comprehensive land use plans, and development codes. · Stability of production. Housing is a factor in every community's economy. The cyclical nature of housing markets, however, creates uncertainties for investment, labor, and builders. The International City Manager's Association suggests that local government policies should address this issue-most do not. Moreover, external factors (e.g. interest rates, cost of building materials, etc.) that bear upon local housing markets tend to undermine the effectiveness of such policies. Despite the various federal and state policies regulating housing, most housing in the U.S. is produced by private industry and is privately owned. While the land . 30 The Practice of Local Government Planning, r Edition, International City Managers Association, 1988. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 69 Attachment 1-77 Exhibit A-78 . use powers of local government have been an important factor in the production of housing, the role of local government has largely focused on regulation for public health and safety and provision of infrastructure. More recently, awareness has grown regarding the impact policies and regulations have had on the other aspects of community life such as costs of transportation and other infrastructure, access of residents to services and employment, and social interactions. DEMAND VERSUS NEED The language of Goal lO and ORS 197.296 refers to housing need: it requires communities to provide needed housing types for households at all income levels. Goal 10's broad definition of need covers all households-from those with no home to those with second homes. State policy, however, does not make a clear distinction between need and demand. Following is our definition, which we believe to be consistent with definitions in state policy: . . Housing need can be defined broadly or narrowly. The broad definition is based on the mandate of Goal 1 0 that requires communities' plan for housing that meets the needs of households at all income levels. Thus, Goal 10 implies that everyone has a housing need because everyone needs housing. However, definition used by public agencies that provide housing assistance (primarily the Department of Housing and Urban Development - HUD, and the Oregon Housing and Community Services Department - HCS) is more narrow. It does not include most of the households that can purchase or rent housing consistent with the requirements of their household size for a price that is affordable. Households that cannot find and afford such housing have need: they are either unhoused, in housing of substandard condition, overcrowded, or paying more than their income and federal standards say they can afford. · Housing market demand is what households demonstrate they are willing to purchase in the market place. Growth in population leads to a growth in households and implies an increase in demand for housing units that is usually met primarily by the construction of new housing units by the private sector based on developers' best judgments about the types of housing that will be absorbed by the market. ORS 197.296 includes a market demand component: buildable land needs analyses must consider the density and mix of housing developed over the previous five years or since their most recent periodic review, whichever is greater. In short, a housing needs analysis should make a distinction between housing that people might need (housing needs) and what the market will produce (housing market demand). . Figure A-I shows a schematic that distinguishes between housing needs that are unmet and those that are met via market transactions. All housing need is the total number of housing units required to shelter the population. In that sense, it is approximately the number of households: every household needs a dwelling place. But some of that need is met through market transactions without much Page 70 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-78 Exhibit A-79 . government intervention because households have the income to demand (purchase) housing services (as owners or renters). That demand is shown in the box on the right. Other households, however, have needs unmet, usually because they lack the resources to purchase housing services (financial need), but because of special needs as well (though, even here, the issue is still one of financial resources). Figure A-1. Relationship between housing need and housing demand ~ All Housing ,..- , .I, ~ Housing Need I-- Demand for New Housing I (housing market) Financial Need Special Need . Most housing market analyses and housing elements of comprehensive plans in Oregon make forecasts of new demand (what housing units will get built in response to market forces). Work by housing authorities is more likely address housing need for special classes, especially low-income. It is the role of cities under Goal 10 to adopt and implement land use policies that will encourage provision of housing units that meet the needs of all residents. It is unlikely that housing markets in any metropolitan area in the US provide housing to meet the needs of every household. Even many upper-income households probably believe they "need" (want) more housing than their wealth and income allows them to afford. Goal 1 0 does not require communities address the housing "want" of residents. More important, however, are more basic housing needs. At the extreme there is homelessness: some people do not have any shelter at all. Close behind follows . substandard housing (with health and safety problems), space problems (the . structure is adequate but overcrowded), and economic and social problems (the structure is adequate in quality and size, but a household has to devote so much of its income to housing payments that other aspects of its quality of life suffer). Location can also be a burden-households that live further from work and shopping opportunities will have to spend more money on transportation. Moreover, while some new housing is government-assisted housing, public agencies do not have the financial resources to meet but a small fraction of that need. New housing does not, and is not likely to, fully address all these needs because housing developers, like any other business, typically try to maximize their profits. . DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 71 Attachment 1-79 . . . Exhibit A-80 In fact, many of those needs are much more likely to be satisfied by existing housing: the older, used stock of structures that is usually less expensive per square foot than new housing. Thus, forecasting the type of new units that might be built in a region (by type, size, and price) is unlikely to bear any relationship to the type of housing to which most people with acute housing needs will turn to solve their housing problems. One key reason for this is the dynamics associated with housing construction. The cost of building new housing is largely prohibitive for building dwelling units affordable to low-income households. This "trickle- down" effect is well known among housing specialists. In most communities a quick comparison of new home prices with income distributions will underscore the fact that developers tend to focus on the move-up market and not on entry- level housing. Viewed in the light of those definitions (e.g., housing demand and housing need), the requirements of Goal 10 need clarification. Goal 10 mandates that communities plan for housing that meets the needs of households at all income levels. Thus, Goal 10 implies that everyone has a housing need. As we have noted, however, it is hard to justify spending public resources on the needs of high-income households: they have the income to purchase (demand) adequate housing services in the housing market. The housing they can afford may not be everything they want, but most policymakers would agree that the difference does not classify as the same kind of need that burdens very-low-income households. This study is not the place to resolve debates about definitions of housing need and the purposes of Goal 10. Here are our assumptions about the distinction between demand and need in the rest of this study: . Our analysis of need addresses the Goal 1 0 requirements regarding financial need (ability to obtain housing) as they relate to future households and to those households whose circumstances suggest that they will have special problems in finding adequate and affordable housing services. That analysis occurs after, and largely independent of, the forecast of new housing that is likely to be built to supply effective demand. · Our forecast includes a comparison of demand for new housing: what kind of housing of what type is likely to get built in the region over the next 20 years. The baseline forecast is the housing "demand" forecast, the alternative forecast is the housing "need" forecast. In summary, Goal 10 intends that cities identify housing need and develop a land use policy framework that meets identified needs. One of the key issues that gets addressed in a 'housing needs analysis is to determine how much land is needed for different housing types, and therefore must be designated for different housing types. Providing sufficient land in the proper designations is one of the most fundamental land use tools local governments have to meet housing need. Page 72 DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Attachment 1-80 . . . Exhibit A-81 DRAFT: Springfield Housing Needs Analysis Page 73 ECONorthwest December 2009 Attachment 1-81 . Appendix B . . Exhibit A-82 National Housing Trends The overview of national, state, and local housing trends builds from previous work by ECO and conclusions from The State of the Nation's Housing, 2008 report from the Joint Center for Housing Studies of Harvard University. The Harvard report summarizes the national housing outlook for the next decade as follows: "Housing markets contracted for a second straight year in 2007. The national median single-family home price fell in nominal terms for the first time in 40 years of recordkeeping, leaving several million homeowners with properties worth less than their mortgages. With the economy softening and many home loans resetting to higher rates, an increasing number of owners had difficulty keeping current on their payments. Mortgage performance-especially on subprime loans with adjustable rates---eroded badly. Lenders responded by tightening underwriting standards and demanding a higher risk premium, accelerating the ongoing slide in sales and starts. "It is still uncertain how far, and for how long, the housing crisis will drive down household growth. Regardless, given the solid underpinnings of long-term demand-including the recent strength of immigration and the aging of the echo-boom generation into young adulthood-household growth will pick up again once the economy recovers. But if the nation suffers a prolonged economic downturn that results in lower immigration and more doubling up, household growth in 2010-2020 may fall short of the 14.4 million level currently projected. This evaluation presents a bleak outlook for housing markets and for homeownership in the short-term brought on by the subprime mortgage crisis. However, the image painted of the future looks brighter, as the increase in housing demand is naturally induced by the growth of the population in the necessary age groups. Long run trends in home ownership and demand Last year (2007) was a continuation of the significant departure from the recent housing boom that had lasted for 13 consecutive years (1992-2005). While strength in early 2005 pushed most national housing indicators into record territory, the market began to soften and sales slowed in many areas in the latter half of2005. By 2006, higher prices and rising interest rates had a negative impact on market demand. Investor demand, home sales and single-family starts dropped sharply. Growth in national sales prices also slowed. By 2007 and early 2008, housing market problems had reached the rest of the economy, resulting in a nationwide economic slowdown and fear of recession. After 12 successive years of increases, the national homeownership rate slipped in 2005, again in 2006 to 68.8%, and again in 2007 to 68.1 %. Page 74 DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Attachment 1-82 Exhibit A-83 . The Joint Center for Housing Studies concludes that the cooling housing market in 2006 had an immediate impact on homeownership. Increasing interest rates and decreasing housing affordability contributed to the recent market correction. Homebuilders could not react quickly enough to changing market conditions, resulting in an oversupply of housing and a rising inventory of unsold homes. The Joint Center for Housing Studies predicts that once the corrections made to work off the housing oversupply and prices start to recover, a return to traditional mortgage products and the strength of natural demand will invigorate the homeownership rate. The long-term market outlook shows that homeownership is still the preferred tenure. Over the next decade, 88% of net household growth is expected to come from gains in the number of homeowners. While further homeownership gains are likely during this decade, they are not assured. Additional increases depend, in part, on finding ways to ease the difficulties faced by low and moderate income households in purchasing a home. It also rests on whether the conditions that have led to homeownership growth can be sustained. . From 2000 to 2005 housing starts and manufactured home placements appeared to have been roughly in line with household demand. In 2005, with demand for homes falling but construction coming off record levels, the surplus of both new and existing homes was much higher than in recent years. In late 2007 and early 2008, the excess supply of new single-family homes retreated by about 12%, though the simultaneous drop in sales left the supply at 11 months, a figure not seen since the 1970s. This resulted in a strong buyer's market, leaving many homes lingering on the market and forcing many sellers to accept prices lower than what they were expecting. The Joint Center for Housing Studies predicts the oversupply will eventually balance as housing starts continue to fall, lower prices motivate unforeseen buyers, and the rest of the economy begins to recover. The Joint Center for Housing Studies indicates that demand for new homes could total as many as 14.4 million units nationally between 2010 and 2020. Nationally, the vast majority of these homes will be built in lower-density areas where cheaper land is in greater supply. People and jobs have been moving away from central business districts (CBDs) for more than a century: the number of the country's largest metropolitan areas with more than half of their households living at least 10 miles from the CBD has more than tripled from 13 in 1970 to 46 in 2000; in six metropolitan areas more than a fifth of households live at least 30 miles out. While people older than 45 years are generally continuing to move away from CBDs, younger people have begun to move nearer to CBDs. . The Joint Center for Housing Studies also indicates that demand for higher density housing types exists among certain demographics. They conclude that because of persistent income disparities, as well as the movement of the echo boomers into young adulthood, housing demand may shift away from single- family detached homes toward more affordable multifamily apartments, town homes, and manufactured homes. Supply-side considerations, however, outweigh these demographic forces. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 75 Attachment 1-83 . . . Exhibit A-84 Recent trends in home ownership and demand Conditions that had previously bolstered the housing market and promoted homeownership weakened in 2005 and eroded further in 2006 and 2007. Increasing interest rates and weakening housing prices combined to slow the housing market. In 2007, new home sales were down 40% from the record 2005 level, and existing home sales were down 20%. Regionally, using housing permits issued as a proxy for new home ownership, Lane County's issued housing permits fell between 25% and 50% between 2005 and 2007. Figure B-1. Change in housing permits issued by county, U.S., 2005-2007 Change in Permits 2005-2007: . More lhan 50% Decline . 25% to 50% Decline Less than 25% Decline . Increase Note: The largest decline in permits from 2005 to 2007 was in Palm Coast, R. (-8G%I. while !.he largest increase was in Hauiescurg, MS 1t-369)1.1 Source: Census Bureau, Construction Statistics, Building Permits by County. As cited in The State of The Nation's Housing, 2008, The Joint Center for Housing Studies of Harvard University, p. 8 Page 76 DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Attachment 1-84 1- . . . Exhibit A-85 Figure B-2. Change in housing permits issued by county, Oregon, 2005-2007 .r? r- -- "'{ I . .L,~-l" , :".-.----. .. . '- " ........... __ -1 .... Source: Census Bureau, Construction Statistics, Building Permits by County. As cited in The State of The Nation's Housing, 2008, The Joint Center for Housing Studies of Harvard University, p. 8 Demographic trends in home ownership According to the Joint Center for Housing Studies, immigration will playa key role in accelerating household growth over the next 10 years. Between 2000 and 2006, immigrants contributed to over 60% of household growth. Minorities will account for 68% of the 14.6 million projected growth in households for the 2005 to 2015 period. Immigrants now comprise a growing share of young adults and children in the United States. Twenty percent of Americans ages 25-34 are foreign born, and an additional 9% are second generation Americans. Members of this generation will probably earn more than their parents becoming an even greater source of housing demand in the coming decades. The Joint Center for Housing Studies suggests that an aging population, and of baby boomers in particular, will drive changes in the age distribution of households in all age groups over 55 years. A recent survey of baby boomers showed that more than a quarter plan to relocate into larger homes and 5% plan to move to smaller homes. Second home demand among upper-income homebuyers of all ages also continues to grow. Households aged 50 to 69 are expected to account for the purchase of nearly half a million second homes between 2005 and 2015. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 77 Attachment 1-85 . . .' Exhibit A-86 People prefer to remain in their community as they age.3! The challenges that seniors face as they age in continuing to live in their community include: changes in healthcare needs, loss of mobility, the difficulty of home maintenance, financial concerns, and increases in property taxes.32 Not all of these issues can be addressed through housing or land-use policies. Communities can address some of these issues through adopting policies that: . Diversify housing stock to allow development of smaller, comparatively easily maintained houses in single-family zones, such as single story townhouses, condominiums, and apartments. · Allow commercial uses in residential zones, such as neighborhood markets. . Allow a mixture of housing densities and structure types in single- family zones, such as single-family detached, single-family attached, condominiums, and apartments. · Promote the development of group housing for seniors that are unable or choose not to continue living in a private house. These facilities could include retirement communities for active seniors, assisted living facilities, or nursing homes. · Design public facilities so that they can be used by seniors with limited mobility. For example, design and maintain sidewalks so that they can be used by people in wheel chairs or using walkers. Home rental trends Nationally, the rental market continues to experience growth, adding 2 million rental households from 2004 to 2007. Demand strengthened in every region except the Northeast. Vacancy rates in the West continue to decline, leading to. strong increases in rental rates. Over the longer term, the Joint Center for Housing studies expects rental housing demand to grow by 1.8 million households over the next decade. Minorities will be responsible for nearly all of this increased demand. The minority share ofrenter households grew from 37% in 1995 to 43% in 2005. The minority share is forecast to exceed 50% of renter households in 2015. Demographics will also playa role. Growth in young adult households will increase demand for moderately priced rentals, in part because echo boomers will reach their mid-20s after 2010. Meanwhile growth among those between the ages of 45 and 64 will lift demand for higher-end rentals. Given current trends in home prices and interest rates, conditions will become increasingly favorable for rental markets in the coming years. 3! A survey conducted by the AARP indicates that 90% of people 50 years and older want to stay in their current home and community as they age. See htto:/Iwww.aaro.org/research. 32 "Aging in Place: A toolkit for Local Governments" by M. Scott Ball. Page 78 ECONorthwest December 2009 DRAFT: Springfield Housing Needs Analysis Attachment 1-86 . . . Exhibit A-87 Despite only modest increases in rents in recent years, growing shares of low- and moderate-wage workers, as well as seniors with fixed incomes, can no longer afford to rent even a modest two-bedroom apartment anywhere in the country. In 2006, one in three American households spent more than 30% of income on housing, and more than one in seven spent upwards of 50%. The national trend towards increased rent to income ratios is mirrored regionally in that a salary of two to three times the 2007 Federal minimum wage of$5.85 is needed to afford rents in Lane County (see Figure B-3). According to the Joint Center for Housing Studies, these statistics understate the true magnitude of the affordability problem because they do not capture the tradeoffs people make to hold down their housing costs. For example, these figures exclude the 2.5 million households that live in crowded or structurally inadequate housing units. They also exclude the growing number of households that move to locations distant from work where they can afford to pay for housing, but must spend more for transportation to work. Among households in the lowest expenditure quartile, those living in affordable housing spend an average of$100 more on transportation per month than those who are severely housing cost-burdened. With total average monthly outlays of only $1,000, these extra travel costs amount to 10 percent of the entire household budget. Figure B-3. Hourly wages needed to afford rent by county, U.S., 2008 "'- Housing Wage: .. :.. $9.48 to S11.69 . $11.70 to $11.54 . $11.55 to $29.24 .- -> . . $29.25 or Higher Source: HUD's Fair Market Rents for 2008, based on methodology developed by the National Low Income Housing Coalition. As cited in The State of The Nation's Housing, 2008, The Joint Center for Housing Studies of Harvard University, p. 30 Note: Every county in Oregon had a housing wage between $11.70 and $17.54 in 2008, DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 79 Attachment 1-87 . . . Exhibit A-88 Trends in housing affordability Despite widespread falling house prices, affordability problems have not improved significantly. A median-priced single-family home under conventional' terms in 2007 (10% downpayment and 30-year fixed rate loan) only costs $76 per month and $1,000 downpayment less than a house bought in 2006, the year in which the sales prices of single-family homes were at their highest real price in history. Only 17 of the 138 National Association of Realtors-covered metropolitan areas have lower costs in 2007 than they did in 2003 when interest rates were bottomed out. With low-wage jobs increasing and wages for those jobs stagnating, affordability problems will persist even as strong fundamentals lift the trajectory of residential investment. The number of severely cost-burdened households (spending more than 50% of income on housing) increased by almost 4 million households from 2001 to 2006, to a total of nearly 18 million households in 2005. Nearly 40% of low-income households with one or more full-time workers are severely cost burdened, and nearly 60% of low-income households with one part- time worker are severely cost burdened. The Joint Center for Housing Studies points to widening income disparities and decreasing federal assistance as two factors exacerbating the lack of affordable housing. While the Harvard report presents a relatively optimistic long-run outlook for housing markets and for homeownership, it points to the significant difficulties low- and moderate-income households face in finding affordable housing, and preserving the affordable units that do exist. Trends in Housing Characteristics The V.S Bureau of Census Characteristics of New Housing Report presents data that show trends in the characteristics of new housing for the nation, state, and local areas. Several trends in the characteristics of housing are evident from the New Housing Report: · Larger single-family units on smaller lots. Between 1997 and 2007 the median size of new single-family dwellings increased 15%, from 1,975 sq. ft. to 2,277 sq. ft. nationally and 18% in the western region from 1,930 sq. ft. to 2,286 sq. ft. Moreover, the percentage of units under 1,200 sq. ft. nationally decreased from 8% in 1997 to 4% in 2007. The percentage of units greater than 3,000 sq. ft. increased from 15% in 1997 to 26% of new one-family homes completed in 2007. In addition to larger homes, a move towards smaller lot sizes is seen nationally. Between 1994 and 2007 the percentage of lots under 7,000 sq. ft. increased by 13% from 29% oflots to 33% oflots. A corresponding 4% decrease in lots over 11,000 sq. ft. is seen. · Larger multifamily units. Between 1999 and 2007, the median size of new multiple family dwelling units increased by 15%. The percentage of multifamily units with more than 1,200 sq. ft. increased from 26% to 47% in the western region and from 28% to 50% nationally. The Page 80 DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Attachment 1-88 . . . Exhibit A-89 percentage of units with less than 600 sq.. ft. stayed at 1 % both regionally and nationally. . More household amenities. Between 1994 and 2007 the percentage of single-family units built with amenities such as central air conditioning, fireplaces, 2 or more car garages, or 2 or more baths all increased. The same trend in increased amenities is seen in multiple family units. A clear linkage exists between demographic characteristics and housing choice. This is more typically referredto as the linkage between life-cycle and housing choice and is do'cumented in detail in several publications. Analysis of data from the Public Use Microsample (PUMS) in the 2000 Census to describe the relationship between selected demographic characteristics and housing choice. Key relationships identified through this data include: . Homeownership rates increase as income increases; . Homeownership rates increase as age increases; · Choice of single-family detached housing types increJases as income Increases; . Renters are much more likely to choose multiple family housing types than single-family; and . Income is a stronger determinate of tenure and housing type choice for all age categories. DRAFT: Springfield Housing Needs Analysis ECONorthwest December 2009 Page 81 Attachment 1-89