Final Report. City of Arvada, Colorado. Vauxmont/Cimarron Park ODP

Transcription

1 Final Report City of Arvada, Colorado Vauxmont/Cimarron Park ODP by Brian Lewandowski, MBA Leeds School of Business University of Colorado-Boulder Thomas G. Thibodeau, PhD Global Real Estate Capital Markets Professor Leeds School of Business University of Colorado-Boulder Boulder, Colorado Phone: Richard Wobbekind, PhD Associate Dean for External Affairs Leeds School of Business University of Colorado-Boulder Boulder, Colorado Phone: June 25, 2007

2 Table of Contents Contents Page Table of Contents i List of Tables ii Executive Summary iii Introduction 1 Methodology 1 The Distribution of Economic Activity in Arvada 3 Population and Households 3 Persons 3 Households 4 Residential Real Estate 5 Housing Stock 5 Permits 5 Housing Stock Projections 6 Demand for Residential Land 6 Employment 8 Employment Projections 8 Office Market Projections 9 Industrial Market Projections 9 Demand for Office and Industrial Land 9 Retail Expenditures 10 For Denver, Colorado 10 For Arvada, Colorado 11 Demand for Retail Land 13 Demand for Land Summary 13 Fiscal Implications of Real Estate Development Budget 14 Property, Sales and Use Tax Revenues 15 Future Real Estate Development 16 The Questions 17 Responses to Property Market Questions 17 Responses to Fiscal Impact Questions 19 Concluding Remarks 20 References 23 i

3 List of Tables Table Title Page 1 Table Comparing Existing Approved Comprehensive Plan/ODP to the 25 Proposed Plan 2 Denver Metro Population: Jefferson County Population: Race/Hispanic Origin of Arvada Population: Persons per Household, by Tenure, Race and Hispanic Origin 28 6 Arvada Population and Household Forecast, by Race, Hispanic Origin and Tenure 29 6A DRCOG Population and Household Forecasts 30 7 Arvada Housing Stock, by Tenure and Structure Type 31 8 Residential Permits: Projecting Arvada s Housing Stock: The Distribution of Real Property in the City of Arvada Forecasting Demand for Residential Land with the Current Distribution of Housing 35 11A Forecasting Demand for Residential Land with the Comprehensive Plan 36 Distribution of Housing 12 City of Arvada 2005 Employment and Wages by 2-Digit NAICS City of Arvada Employment: Assumptions for Office and Industrial Land Demand City of Arvada Office Employment: Demand for Office Space in Arvada (Square Feet): City of Arvada Industrial Employment: Demand for Industrial Space in Arvada (Square Feet): City of Arvada Demand for Office and Industrial Land: Denver Metro Consumer Expenditures Arvada Households Expenditures City of Arvada s Share of Arvada Households Expenditures Additional Retail Space Required With Higher Retail Sales Capture Rates Forecast for Retail Space: Forecast for Retail Land Demand: Land Demand Summary City of Arvada, Colorado 2005 Statement of Net Assets City of Arvada 2005 Statement of Revenue City of Arvada 2005 Statement of Activities City of Arvada Allocating Cost of Government to Sectors City of Arvada 2005 Sales and Use Taxes City of Arvada 2006 Property Taxes City of Arvada Property Tax Rates City of Arvada Property Tax Revenues, by Property Type City of Arvada Revenue and Expenditure Implications of Real Estate 60 Development 36 Sale Rates for Single-Family New Construction in the Arvada Market 61 ii

4 Executive Summary The Cimarron Park/Vauxmont site is a 1,500 acre track of land located in northwest Jefferson County that was annexed into the City of Arvada, Colorado in At the time, property owners and City officials envisioned developing a site that would facilitate a concentration of primary jobs and provide mixed-used development for the area. Nineteen years have passed and the site remains undeveloped. Excluding land allocated to roadways, the current Cimarron Park/Vauxmont Outline Development Plan (ODP) allocates about 37.4% of the undeveloped site to residential properties 29.5% to single-family and 7.9% to multifamily, 19.8% to office and retail land uses and 42.9% to office/industrial uses. Property owners have requested that City officials alter the existing ODP to increase the amount of land allocated to residential uses and decrease the share allocated to industrial uses. The owners proposal allocates 55.5% of the site (net of roadways) to residential uses (48.5% to single-family and 7% to multifamily), 35.7% to office/retail land uses, 2.2% to office/industrial uses and 6.6% to other real property (a hotel, a medical facility, a fire station, and an educational facility). The owners have argued that these changes are driven by the relatively stronger market for single-family detached homes in this area and by the need to use the residential property tax revenues for financing the infrastructure required at the site. This report examines the implications that expected economic growth in Arvada will likely have on the demand for land in Arvada. Population and employment forecasts are used to generate the expected demand for residential, office, industrial and retail real estate consumers in the City of Arvada through The residential analysis converts population projections to households, by race, Hispanic origin and tenure (e.g. renters and homeowners), and employs residential land use intensities obtained from the City of Arvada 2005 Comprehensive Plan to convert the expected demand for dwellings to demand for residential land. Similarly, the analysis derives the demand for office and industrial land from expected growth in office and industrial employment. Projections for retail land use are based on households expected expenditures for goods and services. The report also examines the fiscal implications of alternative real estate development strategies. The fundamental conclusions of this report are: Based on population and employment forecasts, the predominant demand for City of Arvada land for the foreseeable future will come from the residential sector, Expected economic growth in the City of Arvada is expected to generate demand for 3,600 acres of land with about 79.5% of the land demand coming from residential users, 7% from office properties, 10.6% from industrial properties and 2.8% from retail properties, The City of Arvada can reduce sales tax revenue leakage by encouraging the development of Health and Personal Care retail and Clothing and Accessories retail, The limitations on financing public services with property taxes imposed by the State of Colorado s Gallagher Amendment force local governments in Colorado in general, and the City of Arvada in particular, to finance public services with sales tax revenues and user fees. iii

5 Introduction The current approved Vauxmont/Cimarron Park Outline Development Plan (ODP) permits acres of residential ( acres for single-family and acres for multi-family) and 856 acres of non-residential (251 acres for commercial/office/ retail, 19 acres for a Town Center, and 586 acres of office-industrial). The property owners have petitioned the City of Arvada to amend the approved Comprehensive Plan/ODP to permit acres of residential (a 48.6% increase) and acres of non-residential (a 29.1% decrease). The property owners have also proposed to change the mix of residential land use from 78.9% single-family to 87.4% single family. Table 1 compares the approved Comprehensive Plan/ODP to the proposed plan. Property owners have requested the change because they feel the primary market for land use in this area is residential and because tax revenues from the single-family homes are required to finance the infrastructure needed to develop the entire site. This research has two objectives. The first is to examine the demographic and employment trends in the area to identify alternative mixes of land uses that the real estate market will support, now and for the foreseeable future. The second objective is to examine the fiscal implications of alternative land use proposals. Methodology The first objective is accomplished by examining the population and employment trends in Jefferson County in general and in the City of Arvada in particular. This research examines how population and employment growth generate demand for alternative land uses. Population projections through 2030 from the State of Colorado Demography Office are used to establish a baseline scenario for future residential and retail land demand. Population projections are allocated to households based on household size and rates of homeownership by race, Hispanic origin, and structure type (single-family vs. multifamily). Future demand for residential properties is converted into residential land demand using densities obtained from the City of Arvada 2005 Comprehensive Plan. In addition to population projections, this analysis employed data from the 1990 and 2000 Census, the 2004 Denver American Housing Survey, the 2004 Bureau of Labor Statistics Consumer Expenditure Survey, the 2004 Economic & Planning 1

6 Systems Arvada Economic Analysis Final Report, the Census 2005 American Community Survey and data from the 2006 Jefferson County Assessor s office. We follow a similar strategy for commercial real estate land uses. We begin with current employment by two-digit North American Industry Classification System (NAICS) codes from the Colorado Department of Local Affairs and from the Denver Regional Council of Governments (DRCOG). Employment projections for Jefferson County from Woods and Poole, Inc. are used to develop a base case expected demand for office and industrial space. Expected employment growth is allocated to office and industrial jobs using data from the Urban Land Institute (ULI) and from the 2004 Economic and Planning Systems, Inc. Final Report. Assumptions about the intensity of land use (floor-area ratios, or FARs) were obtained from the 2005 City of Arvada Comprehensive Plan. We then use expected household expenditures for goods and services to estimate the demand for retail space in the City of Arvada. We estimate demand for retail space twice: once using current retail expenditure capture rates and again assuming the City of Arvada can add sufficient retail space to cut the retail sales tax leakage in half. We convert the demand for retail space to demand for retail land using FARs provided in the City s 2005 Comprehensive Plan. Finally, we examine the fiscal implications of alternative real estate development strategies. We compare the expected costs associated with the population and employment growth patterns to the expected revenues from property, sales and use taxes. It should come as no surprise that the City of Arvada should do whatever it can to capture as much sales tax revenues as possible. We conclude the report by answering the questions posed in the statement of work: A. Property Market Questions 1. Is there a market demand for the City s vision for a business/industrial park, town center, retail development, single family and high density residential housing for the Vauxmont/Cimarron properties? What is the timeframe for market demand? Is it realistic to expect build-out of the market demand within the urban renewal district timeframe? 2. Based on market trends and conditions, what public actions or circumstances will this help, hinder or prohibit the City from achieving its vision within the planning time frame? What role does the Jefferson Parkway play in market demand and its timing? 2

7 3. Should the City consider revising its vision to incorporate other land uses that the market study has identified? 4. Is there a projected market for a larger town center as reflected in the Property Owners plan? 5. Can the City s vision of generating primary jobs and expanding the tax base be achieved through an alternative plan? B. Fiscal Implications 1. How does the fiscal impact of the two proposals, or an alternative program based on likely market demand, compare in terms of property and sales tax revenues. 2. Are the property tax/assessed valuation numbers provided by DA Davidson legitimate? 3. Are the owners projected absorption levels realistic? If not, what is the realistic absorption levels on residential units, commercial space and industrial space? 4. In terms of sales tax revenue, address the issue of leakage. What amount of total sales tax revenue is leaked to other communities? 5. How much residential must be built to stimulate retail development? The Distribution of Economic Activity in Arvada Population and Households Persons The population in the Denver metro area grew at a compound annual growth rate of 2.11% between 2000 and Douglas County experienced the greatest rate of population growth in the region, increasing 7.4% per year over this period. Adams County and Arapahoe County increased 2.1% and 1.8% annually, respectively, while Jefferson County experienced the lowest rate of growth over the 2000 through 2005 period (only 0.21% annually). Broomfield County was established as the 64 th county in Colorado in November 2001, effectively capturing some population from the surrounding counties and skewing growth rates. In 2005, Broomfield County population was estimated at 45,755 (Table 2). The State of Colorado s Demography office reports that the 2005 population of Arvada was 103,704. Of these, 100,918 individuals lived in the Jefferson County portion of the City of Arvada and 2,786 lived in the Adams County portion of Arvada. In 2005, 18.95% of Jefferson County s population lived in Arvada, the second largest municipality in the county. The City of Arvada population remained fairly constant between 2000 and 2005, with Arvada s population increasing only 0.32% annually over this period (Table 3). 3

8 According to the State of Colorado s Demography office, the Denver metro population is expected to increase at an annual rate of about 1.65% through This rate of population growth is expected to slow to annual rates of 1.52% from 2015 to 2020; to 1.35% from 2020 to 2025 and to 1.21% from 2025 to Jefferson County is expected to be the slowest growing county in the Denver Metro region, with annual population growth rates just under 0.9% annually for the next ten years; increasing to 0.93% per year between 2015 and 2020; declining to 0.8% annually between 2020 and 2025 and to 0.67% annually between 2025 and According to the US Census Bureau, the population of Arvada increased from 89,235 in 1990 to 102,153 in 2000, for an annual growth rate of 1.36% over the decade. Of the 12,918 person increase in total Arvada population, 8,870 were white (5,482 white, non-hispanic and 3,388 Hispanic white). Table 4 provides the 1990 and 2000 distributions for Arvada s population by race and by Hispanic origin. Individuals of Hispanic origin may report themselves as white, black or some other race. Whites made up 91.0% of the population in 2000, while black and American Indian residents each made up less than 1%. Asians made up 2.2% in Persons of Hispanic origin grew at 4.21% annually from 1990 to 2000, and made up nearly 10% of the population in Households The next step in the analysis is to allocate persons to households (then households to dwellings, and finally dwellings to residential land use). It is important to separately identify persons of Hispanic origin for two reasons: (1) Hispanic households are larger (have more persons per household) than white, non-hispanic households; and (2) rates of homeownership vary significantly by race and by Hispanic origin. Data from the 2004 Denver American Housing Survey indicates that 81% of white, non-hispanic persons reside in owner-occupied dwellings while only 57.6% of Hispanic persons reside in owner-occupied dwellings. Furthermore, there are 2.54 persons per household for white, non-hispanic owner-occupied housing and 3.24 persons per household for Hispanic owner-occupied housing (Table 5). 4

9 According to the 2005 American Community Survey, Arvada s median household income was $61,353. This is slightly above Jefferson County s median household income of $60,944, and more than 21.1% above Colorado s median household income of $50,652. The number of households in Arvada is expected to increase 0.76% annually over the next 30 years. The highest overall growth is expected in renter-occupied households, increasing 1.08% annually between 2005 and 2035, while the slowest growth is expected in homeowners (0.68% annually). While the Hispanic-occupied households will see the fastest growth rate in the future, white, non-hispanic households will add the most residents and households to the City of Arvada (Table 6). According to the State of Colorado Demography Office, the City of Arvada is expected to have a population of 127,613 in 2030, for an annual growth rate of 0.83% over the next 25 years. The Denver Regional Council of Governments (DRCOG) has forecasted much slower growth (Table 6A). DRCOG has projected a 2030 Arvada population of 115,081. This amounts to an annual rate of growth of 0.41% and is less than half of the State Demography Office forecast. We use population forecasts from the State Demography Office. Residential Real Estate Housing Stock The stock of housing in Arvada is predominately single-family. According to the 2000 Census, 72.0% of the Arvada housing stock is single-family detached and 7.2% is single-family attached (Table 7). The remaining stock is evenly distributed across multifamily dwellings. In 2000, 74.2% of the housing stock was owner-occupied. It is important to note that in 2000, approximately 8.9% of single-family dwellings were occupied by renters and about 15.1% of multifamily dwellings were owner-occupied (e.g. residential condominiums). Permits Arvada s building permits have had a high level of variability compared to Jefferson County, with an average of 398 single-family permits and 166 multi family permits issued annually since The peak number of issued permits was 1,058 permits in The following year permits declined by 28.5% (Table 8). 5

10 Housing Stock Projections The growth in Arvada s housing stock was projected to 2035 by assigning the expected growth in households, by race, Hispanic origin and tenure (e.g. owner-occupied vs renter-occupied), to dwellings. The assumption going forward is that 95.8% of owner-occupied households will occupy single-family homes while the other 4.2% homeowners will own condominiums 1. Similarly, 70.7% of renters will reside in multifamily dwellings while 29.3% will occupy singlefamily detached dwellings. Finally, vacancy rates in the 2000 Census were abnormally low. For the housing stock projections, it is assumed that the vacancy rates for both single-family and multifamily housing will be 8%. A 92% occupancy rate is considered to be a normal, long-term occupancy rate. These assumptions produce the housing stock forecasts revealed in Table 9. The City of Arvada will need to add about 280 single-family units per year and 110 multifamily units per year for the next decade. Housing demand is expected to grow at a faster rate between 2015 and 2020 yielding a total need for 1,619 single-family dwellings and 624 multifamily dwellings over this five-year period. Demand for Residential Land Converting the expected growth in the number of residential dwellings to residential land use requires an assumption about how intensively residential real estate will use land in the future. The Jefferson County Assessors data provides some guidance on the current intensity of land use for the Arvada real estate market. The 2006 Jefferson County Assessors file has approximately 45,000 parcels located in the City of Arvada. Table 10 indicates that nearly 90.3% of these parcels are residential (single-family, multifamily, condominiums, town homes and vacant land zoned residential). This table also indicates that about 38.4% of the single-family properties are on lots with less than 7,500 square feet and another 34.1% are on lots with between 7,500 and 10,000 square feet. Only 5.3% of the single-family properties in Arvada are on lots exceeding ½ acre. These are approximately evenly divided between properties on lots exceeding one acre and properties on lots between ½ and one acre. The City of Arvada s 2005 Comprehensive Plan also provides guidelines for the future intensity of land use. The Comprehensive Plan allows three low density single-family districts and one 1 We will examine an alternative assumption when we calculate residential land demand. 6

11 suburban district (with up to 5 dwellings per acre). The Comprehensive plan also provides for four attached housing/multifamily housing districts: Mixed Use-Residential (25% - 8 dwellings/acre); Mixed Use-Residential (60% at 8 dwellings/acre); Medium Density Residential (8 dwellings/acre) and High Density Residential (15 dwellings/acre). According to the Comprehensive Plan, about 68% of the single-family properties that will be built in the City of Arvada will be in Suburban Residential districts (up to 5 units per acre). The remaining 32% will be built in Low Density Residential (with Cluster Option) districts. 2 The expected build-out for attached/multifamily housing is 12.2% in Mixed Use Residential (25%-8 du/ac); 63.0% in Mixed Use Residential (60% at 8 du/ac); 19.1% Medium Density Residential and 5.6% High Density Residential (Table 11). Assigning population growth to households and households to dwellings yields a demand for 9,959 residential dwelling units between 2005 and Converting housing demand to demand for residential land in the City of Arvada is done twice: once using the current distribution of the housing stock in Arvada (which is about 72% single-family and 28% multifamily) and again using the Comprehensive Plan distribution of structure types (40% single-family and 60% multifamily). Using the current 72/28 split in structure types, the expected demand for residential dwellings through 2030 is 9,959 units-7,187 single-family properties and 2,772 multifamily units. Using the expected residential land use intensities provided by the Comprehensive Plan, this demand for dwellings results in a demand for 3,859 residential acres about 3,409 acres for single-family homes and about 450 acres for multifamily units. Changing the structure type mix to correspond to the Comprehensive Plan yields a demand for 3,984 single-family units and 5,975 multifamily units. This converts to a demand for 2,860 residential acres-1890 acres for single-family properties and 970 acres for multifamily properties (Table 11A). 2 Table B-1: Arvada Land Use Plan-Buildout 7

12 Employment Employment Projections According to the Colorado Department of Local Affairs, Arvada s 2005 employment was 29,816. Arvada employment is concentrated in construction, manufacturing, retail trade, and accommodation and food services. These sectors represent nearly 51% of Arvada s 2005 employment. Table 12 reports City of Arvada employment by two-digit North American Industry Classification System (NAICS) Code. The NAICS is a demand oriented product classification system collectively developed by the governments of the United States, Canada and Mexico to collect and report business activity in North America. Employment forecasts were obtained from Woods & Poole Economics, Inc. Woods & Poole Economics, Inc. is an independent firm that specializes in long-term county economic and demographic projections. Woods & Poole's database for every county in the U.S. contains projections through 2030 for more than 900 variables. Each year Woods & Poole updates the projections with new historical data. The projections are sold in printed books and in spreadsheet files on CD-ROMS. Woods & Poole has been making county projections since The Woods & Poole forecasts for Jefferson County, Colorado were allocated to the City of Arvada using the same proportions that existed in Using these forecasts, employment in the City of Arvada is expected to increase to 43,342 by 2030, an increase of about 1.51% per year (Table 13). Converting employment growth to commercial demand for land is a three step process. In the first step, some of the growth in employment is allocated to office employment and some new employment is assigned to industrial employment. The office and industrial jobs are then converted to square feet of office and industrial space using average employee space requirements. These conversion factors were obtained from several sources including the Urban Land Institute and from the 2004 EPS report. The office and industrial space requirements are then converted to land demand using the floor-area ratios (FARs) reported in the City of Arvada s 2005 Comprehensive Plan. Table 14 provides a summary of the assumptions used to 3 Source: 8

13 convert employment growth to demand for office and industrial land. For example, the assumption is that 100% of employment growth in Finance and Insurance will require office space and that these office employees will use an average of 300 square feet of space per employee. In manufacturing, on the other hand, 85% of the employment growth will require industrial space and the typical industrial employee consumes 500 square feet of space. Office Market Projections Expected growth in office employment through 2030, by NAICS code, is reported in Table 15. Office employment is expected to grow at an annual rate of 1.54% over the period with the largest growth occurring in the Accommodation and Food Services industry. If office employees consume an average of 300 square feet of space per employee, the demand for office space in the City of Arvada is expected to increase from its current level of 4.3 million square feet of space to 6.3 million, an increase of 2 million square feet of space over the next 25 years (Table 16). Unfortunately, the Jefferson County Assessors data does not separately identify office space so it is not possible to confirm the 4.3 million square feet of current consumption of office space. Industrial Market Projections Tables 17 and 18 provide a similar analysis for industrial space. Table 17 assigns employment forecasts to industrial employment using the assumptions provided in Table 14. Table 18 converts the industrial employment to industrial space. Industrial employment in the City of Arvada is expected to increase by 3,256 employees over the next 25 years. This increase in industrial employment is expected to increase the aggregate demand for industrial space by 2.2 million square feet of space--from the current 4.6 million square feet of space to 6.8 million square feet of space in Demand for Office and Industrial Land Converting the demand for office and industrial space to the demand for office and industrial land requires some assumptions about how intensively land will be used. Assumptions about the future intensity of land use must be made with care. Small differences in assumed FARs can result in large differences in projected land demand. To illustrate this point, we have computed 9

14 the expected demand for office and industrial land demand using for alternative assumptions about future FARs. All these assumptions are legitimate possibilities: one is from the 2004 City of Arvada EPS Final Report; a second is from the City of Arvada s 2005 Comprehensive Plan; a third comes from the City of Westminster s 2005 Comprehensive Plan and a fourth possibility is from the City of Arvada s current Land Development Code (Table 19). The employment projections generate a 2 million square foot increase in the aggregate demand for occupied office space over the next 25 years. With an assumed 10% vacancy rate, this means that the City of Arvada will require an additional 2.25 million square feet of office space over the next 25 years. The effective office space FAR used in the 2004 EPS report is With a FAR of 0.13, the City of Arvada will need 386 acres of land to accommodate a 2.25 million square foot increase in the aggregate demand for office space. Using the City of Arvada s Comprehensive Plan FAR of 0.2, the City will require 257 acres of land to accommodate the expected increase in office employment. Using the City of Westminster s Comprehensive Plan FAR of 0.5, the City of Arvada would need 147 acres of land to accommodate the expected increase in office employment. Finally, the City of Arvada s current Land Development Code permits three story office improvements that utilize 35% of the lot. This effectively translates to an office FAR of With an FAR of 1.05, the City of Arvada would only need 49 acres of land to accommodate the expected increase in the demand for office space. The difference in acreage required to meet the (same) expected demand for office employment between the EPS report and what is permitted under the current Arvada building code is 337 acres. Assuming a long-run vacancy rate for industrial space of 10% and the 0.15 FAR found in the City of Arvada s 2005 Comprehensive Plan yields a demand for industrial land of 382 acres over the next 25 years. Retail Expenditures For Denver, Colorado Next we examine the demand for City of Arvada land that is driven by retail users. We begin by computing average expenditures on retail goods and services for the Denver metro area. Table 10

15 20 provides the average dollar and percent expenditures on consumer goods and services for 2004 obtained from the Bureau of Labor Statistics Consumer Expenditure Survey. The median household income for the Denver metro area in 2004 was $65,224. Denver area households spent about $50,000 on consumer goods and services. The remainder went to taxes and (very little) to savings. About 9.6% of household incomes were spent on food (5.8% on food at home and 3.8% on food away from home); 24.2% on housing; 2.7% on apparel; 13.3% on transportation; 4.2% on healthcare and 5.0% on entertainment. For Arvada, Colorado According to the 2005 Census American Community Survey, the median household income for the City of Arvada in 2005 was $61,353. With 41,370 households, Arvada residents spent $1.9B on consumer goods and services in Expenditures for each consumption category for Arvada households were estimated by applying the average Denver metro percentage expenditures obtained from the 2004 Consumer Expenditure Survey (CES) to Arvada households. So Arvada households spent approximately $147.4 million on food consumed at home in 2005 and an additional $95.8 million on food away from home. Table 21 provides estimates of 2005 aggregate consumer expenditures for each CES expenditure category for City of Arvada residents. Some of this money was spent in the City of Arvada and some spent elsewhere. To estimate how much of Arvada households expenditures were captured by City of Arvada retailers, we examined the City s sales tax revenues for 2005 and We first converted 2006 expenditures to 2005 dollars using the Consumer Price Index. We then averaged total retail sales for 2005 and 2006 and combined appropriate revenue categories. This procedure produced the total City of Arvada expenditures provided in Table 22. According to the 2004 Consumer Expenditure Survey, Arvada households spent $159.4 million on food consumed at home in According to the City of Arvada s tax records, grocery stores, convenience stores, dairies, bakeries and produce markets generated $248.8 million in sales. Apparently, City food stores are capturing grocery store revenues from outside City boundaries. This is the only retail category where this occurs. For all other retail categories, Arvada residents spent some of their income outside City boundaries. For example, Arvada residents spend only 37.4% of their health and personal care 11

16 expenditures at retailers in the City of Arvada and only 38.1% of their clothing and accessories expenditures in Arvada stores. The next steps are to convert retail expenditures to retail space and then convert retail space to demand for retail land. The Urban Land Institute and the International Council of Shopping Centers annually publish a volume in the series Dollars and Cents of Shopping Centers. This publication summaries annual operating revenues and expenses for a variety of retail centers ranging from Super Regional Shopping Centers to Neighborhood and Convenience Centers. We converted annual retail expenditures to gross leaseable area using revenues for Super Community and Community Shopping Centers. This category includes traditional community shopping centers, power centers, town centers, lifestyle centers and out/off-price centers. 4 This publication reports that, on average, grocery stores have annual revenues of $ per square foot. Consequently to accommodate $248.8 million in food consumed at home expenditures, the City of Arvada needs 634,810 square feet of grocery store (and convenience store, dairy, bakery and produce market) space. This is square feet of food store space per City of Arvada household. Health and Personal Care stores sell $ per square foot while General Merchandise outlets sell $ per square foot. Converting expenditures in these categories to square feet of retail space suggests the City of Arvada has 155,671 square feet of Health and Personal Care retail space and 2.3 million square feet of General Merchandise retail space. Table 23 provides the Arvada Retail Expenditures, the Required Dollars per square foot (from the ULI) and the implied square feet of retail space require to accommodate the observed expenditures for convenience goods, shoppers goods, eating and drinking establishments and building materials. In the aggregate, the City of Arvada has 3.9 million square feet of retail space. This is about 94.4 square feet of retail space for every household in the City of Arvada. Table 23 examines how much additional retail space the City of Arvada would need if it were able to capture more retail sales. The assumed Arvada share of consumer expenditures in Table 23 splits the difference between the current Arvada retail expenditure capture rate and a 100% capture rate. For example, the current capture rate for Health and Personal Care items in the City 4 Dollars and Cents of Shopping Centers/The SCORE 2006, Urban Land Institute, page 7. 12

17 of Arvada is 37.4%. Splitting the difference between the current capture rate and a 100% capture rate for Health and Personal Care items yields a capture rate of 68.7%. If the City of Arvada were able to increase retail expenditure capture rates to the rates assumed in Table 23, then the City of Arvada could justify adding 529,076 square feet of retail space today (assuming a 0.2 FAR). Demand for Retail Land The future demand for retail space is determined using the projected increases in the number of households determined earlier. The number of households in Arvada is expected to steadily increase at 0.76% annually over the next 30 years. If real household incomes are held constant, then future demand for additional retail space can be estimated by applying the per household requirements for retail space. The additional demand for retail space through 2030 is provided in Table 24. These forecasts were computed using current retail expenditure capture rates. On average, the City of Arvada is going to need about 34,000 square feet of retail space annually to accommodate the expected growth in the number of households. Future demand for retail land is estimated using an assumed floor-area ratio. We examine four forecasts that employ FARs from the 2004 EPS report, from the City of Arvada s 2005 Comprehensive Plan, from the City of Westminster s 2005 Comprehensive Plan and using the maximum FAR permitted under Arvada s current land use ordinance. Using the FAR from the City s 2005 Comprehensive Plan, the City of Arvada will need 99 acres of land to accommodate the expected increase in demand for retail land over the next 25 years (Table 25). Demand for Land Summary We forecast the future demand for the City of Arvada land in two ways first assuming the current capture rates for retail sales tax revenues and second assuming the City of Arvada will capture more sales tax revenues. Both projections employ the 40%/60% split between singlefamily and multifamily contained in the 2005 Comprehensive Plan. With current sales tax revenue capture rates, Arvada will need 3,598 acres of land to accommodate population and economic growth through Residential uses (both single-family and multifamily) will require 79.5% of the land needed to accommodate the expected increase in economic activity; 13

18 office 7.1%; industrial 10.6% and retail 2.8%. If the City of Arvada adds enough retail to split the difference between current retail expenditure capture rates and 100% capture rates, then projected land demand increases to 3,659 acres with residential uses comprising 78.2% of total land demand, office 7.0%, industrial 10.4% and retail 4.4% (Table 26). Fiscal Implications of Real Estate Development Budget We now turn our attention to the fiscal implications of alternative development plans. Table 27 provides the 2004 and 2005 revenues and expenditures for the City of Arvada as reported in the Comprehensive Annual Financial Report. On the revenue side, sales taxes and user fees provided approximately 39.2% of general revenues and 88.7% of tax revenues (Table 28). Property taxes provided nearly 5.0% of Government Activities total revenue and about 11.3% of tax revenues. In 2005, General Government expenses accounted for 28.6% of the $118.7 million in City expenditures; public safety, 16.3%; public works 11.8%; and water, 13.1%. Some of these expenditures are reimbursed through user fees. Table 29 illustrates that water services provided $18.8 million in revenues--$13.5 million in user fees and $5.3 million through capital grants and contributions. To examine the fiscal implications of alternative development strategies we must associate government costs with residential and business users of public services. According to the City of Arvada, approximately 79% of water services are consumed by residential users and 21% are consumed by business users. Most of the waste water services, 86.9%, are consumed by residents and the remaining 13.1% by businesses. To allocate City of Arvada government costs for most categories of net expenses we averaged these allocations 83% to residents and 17% to business users. The net expenses associated with the Arvada Urban Renewal Authority (AURA) are allocated 75% to the business sector and 25% to the residential sector. The net expenses associated with the Arvada Economic Development Association (AEDA) are allocated entirely to the commercial sector. The details of this allocation are provided in Table 30: Allocating 14

19 Cost of Government Services to Sectors. In 2005, net government expenses totaled $69 million. This amounts to $ per resident and $ per Arvada job. Property, Sales and Use Tax Revenues How much of these expenses are financed with sales, use and property taxes? Table 31 provides City of Arvada Sales and Use Taxes for 2005 by revenue category. In 2005, the City of Arvada received about $ per capita or $ per household, resulting in about $35.7 million in sales tax revenues. The City also received about $12.75 per person and $31.55 per household in associated use taxes, totaling approximately $1.3 million in Table 32 provides 2006 Property Tax revenues by property class. These were obtained from the Jefferson County Assessors office. In 2006, Arvada real property generated about $97.7 million in property tax revenue. Five million of this went to the City of Arvada. Table 33 provides the current property tax mill rates for real property located in the City of Arvada. Now we can estimate the sales, user and property tax revenues associated with alternative real estate developments. For residential properties, our property tax revenue estimates are based on median transaction prices for single-family and multifamily properties sold in Arvada during 2005 and Market values for multifamily properties are reported separately for condominiums (e.g. owner-occupied) and for apartments (e.g. renter-occupied). Commercial property values are estimated using the income approach, a standard technique for estimating the market value of income producing properties. To estimate the market value for a commercial property using the income approach, you capitalize the income produced by the property net of all expenses at the prevailing market capitalization rate. The capitalization rate (or cap rate) is simply the rate used to convert the flow of net benefits generated by a property to an asset value, or market price. Information on market rents, vacancy rates, operating expenses and capitalization rates for office, industrial and retail properties were obtained from a variety of sources. The details are provided in Table 34. For example, the current market rent for office space in northwest Jefferson County (an area that includes Arvada) is $15.50 per square foot. The current vacancy rate for this office market is 15%. Operating expenses average about 20% and, according to Real Capital Analytics, 15

20 the current market cap rate for office space in Denver is 7.2%. We estimate value for a 10,000 square foot office property. The expected net operating income for this property (rent less a vacancy allowance and less operating expenses) is $105,400. A market cap rate of 7.2% yields an estimated market value of about $1.46 million. The Gallagher Amendment requires that commercial properties be assessed at 29% of their market value. The assessed value for a $1.46 million office property is $424.5 thousand. Using the current City of Arvada mill rate, this yields property tax revenues of $2,000. With a 15% vacancy rate and 300 square feet of office space per employee, a 10,000 square foot office property is expected to provide 28.3 jobs and generate about $70.67 in City property tax revenue per employee. This office job costs the City $ A similar valuation methodology yields property tax revenues of $1,077 for 10,000 square feet of industrial space and $1,093 for 10,000 square feet of retail space. The median transaction price for a single-family home built and sold during 2005 and 2006 in the City of Arvada was $396,900. The median price for a condo was $153,200 and the median market value for an apartment unit was about $75,000. Gallagher requires that residential properties be assessed at 7.96% of market value. Using the current City of Arvada mill rate, a recently constructed single-family home in Arvada is expected to generate $149 per year in property tax revenue for the City, a condominium $57, and an apartment unit $28. Future Real Estate Development Finally, we have examined the fiscal implications for two alternative real estate development plans-one that examines the consequences of developing a future housing stock that is similar to the current stock (72% single-family and 28% multifamily) and a second that develops a housing stock that is 40% single-family and 60% multifamily (according to the 2005 Comprehensive Plan). We have examined the fiscal implications for development over the period. In both cases, population is expected to increase by 4,724 people and the City of Arvada is expected to add 2,718 jobs. On the revenue side, the additional population and jobs are going to generate additional real estate which, in turn, generates additional property tax. The additional households will also spend some of their income in Arvada, increasing Arvada sales tax revenues. 16

21 On the expense side, each person costs $ in City services and each job costs $ in City services. Using the existing mix of residential property, incremental revenues total about $1.64 million while additional expenses total $3.39 million for a deficit of $1.74 million. Using the 2005 Comprehensive Plan mix of single-family/multifamily, the expected deficit increases to $1.80 million. These results are listed in Table 35. The Questions Responses to Property Market Questions 1. Is there a market demand for the City s vision for a business/industrial park, town center, retail development, single family and high density residential housing for the Vauxmont- Cimarron properties? Yes, there are markets for these property types. Some markets are stronger than others. The primary demand for land in Arvada, Colorado is residential land. The City of Arvada should allocate at least 75% of undeveloped land to residential uses going forward. Using population projections from the State of Colorado Demography Office, the City of Arvada should be adding about 400 residential dwellings per year for the next 25 years. Using density assumptions provided in the 2005 Comprehensive Plan, the City will need about 3,600 acres to accommodate the expected population and employment growth. Residential use is expected to consume about 2,800 acres. The City should also aggressively pursue retail properties. Most of the market demand for retail space is coming from existing residents. If the City were able to split the difference between current retail expenditure capture rates and 100% capture rates, the City could add one-half million square feet of retail space today. At a sales tax rate of 3.46%, this space could potentially generate sales tax revenues of nearly $13 million annually. The markets for office and industrial space in Arvada are considerably weaker. Expected increases in employment are expected to require 250 acres of land devoted to office use and 360 acres devoted to industrial use for the entire City of Arvada over the next 25 years. What is the timeframe for market demand? Is it realistic to expect build-out of the market demand within the urban renewal district timeframe? Based on current population and employment projections, the market for residential properties in the City of Arvada will be able to absorb 400 dwellings per year for the 17

22 foreseeable future; the office market about 90,000 square feet of space per year; the industrial market about 100,000 square feet of space and the retail market about 35,000 square feet of space per year. For residential property, both the approved Vauxmont/Cimarron Park ODP and the proposed plan are easily achievable within the urban renewal district timeframe. 2. Based on market trends and conditions, what public actions or circumstances will help, hinder or prohibit the City from achieving its vision within the planning time frame? What role does the Jefferson Parkway play in market demand and its timing? The Jefferson Parkway will speed up residential and retail absorption rates. The Parkway is essential if any of the office and industrial space is to locate in this area. Fundamentally, the City of Arvada is competing with FasTracks for employment. The City is at a significant disadvantage because there is no employment currently at the site. Central city employment locations have an advantage for two reasons. First, many employers seek to locate near other firms like themselves when making location decisions. They do this to take advantage of scale economies in the market for inputs (e.g. labor markets). Second, providing tax incentives to create jobs does not necessarily mean the City will be able to attract investors. Capital markets prefer central city locations to suburban locations for office property. When office demand increases in the central business district (CBD), rents increase because CBD locations are typically supply constrained. Higher rents mean higher investment returns (in both income and capital appreciation). When office demand increases in suburban locations, firms just build more office space. Rents do not increase as fast as they do in the CBD. So if the City wants to attract employment, the City should focus on young industries that can locate anywhere (because the scale economies are not established) and focus on firms wanting to locate their corporate headquarters as opposed to speculative office construction. This is a challenge young firms with enough of their own capital to invest in real estate. 3. Should the City consider revising its vision to incorporate other land uses that the market study has identified? Yes, the City s vision for 586 acres of office/industrial is too optimistic. The primary demand for land in the City of Arvada going forward is coming from the residential sector. 4. Is there a projected market for a larger town center as reflected in the Property Owners plan? A larger town center? No. If you add the current demand for one-half million square feet of retail space to the demand coming from population growth going forward, you end up with a demand for 160 acres. The proposed plan calls for 312 acres for the Town Center. 18

23 5. Can the City s vision of generating primary jobs and expanding the tax base be achieved through an alternative plan within the same time frame of the existing urban renewal district? For reasons described above, the City s vision of generating jobs at this location is challenging for several reasons. First, the City is not in a position to compete with FasTracks for jobs. Second, jobs cost money more money than residents. If the City wants to expand the tax base it needs to compete more aggressively for retail. Gallagher is what it is and it has municipalities across Colorado competing for retail sales tax revenues. Responses to Fiscal Impact Questions 1. How does the fiscal impact of the JCMD proposal, and all alternative programs based on likely market demand, compare in terms of property and sales tax revenues. Assuming that approximately 83% of net government expenses can be appropriately allocated to residents and that 100% of AURA and AEDA costs can be allocated to businesses, the City of Arvada government costs $ per resident and $ per job. In addition to user fees, the primary sources of revenue to finance public services are property and sales taxes. With Gallagher, the property tax burden is imposed on commercial real estate. But even with a disproportionate share of the property tax burden placed on commercial real estate, the revenues from commercial real estate taxes don t cover costs. For a typical office property, an office employee generates about $71 in property tax revenue well short of the $466 per employee cost of government. The scenario is similar for industrial and retail property taxes. The property tax scenario is worse for residential properties. A $397,000 home in Arvada generates about $150 in property tax revenues. With 2.5 persons per household, $150 per household is $60 per person, well short of the $ per capita cost. In 2005, the City of Arvada received about $37 million in sales and use tax, about $356 per person or $880 per household. Revenues from retail sales taxes and user fees finance a significant portion of public services in the City of Arvada (as well as in every other municipality in Colorado.) The City should do whatever it can to increase the retail options for Arvada households. In addition, the City should encourage residential development, particularly when those new households can be encouraged to make their purchases within City limits. 2. Are the property tax/assessed valuation numbers provided by DA Davidson legitimate? The property tax/assessed valuation numbers provided by DA Davidson have several component parts that should be evaluated separately. First, the DA Davidson projections assume 173 residential units will be sold in 2007: 0 Estate Properties, 24 large (80 lots), 36 Standard (75 lots), 77 Medium (65 lots) and 36 patio homes (50 to 55 lots). 77 Medium sized lot sales annually translates to 6.4 units per month. This is high, but within the range of single-family absorption rates for this market (see response to #3 below). It is also fair to say that absorption is behind schedule for Total 19