Garfield County Land Values and Solutions Study BBC

Garfield County Land Values and Solutions Study BBC Research & Consulting 3773 Cherry Creek North Drive, # 850 Denver, Colorado 80209 800 -748 -3222 www. bbcresearch. com June 1, 2006 Report Presentation

Introduction n Two Economic Studies Are Underway Ø The Economic Impact Model Ø The Land Values and Solutions Study 1

Land Values Study: Project Objectives n Demonstrate what factors drive residential land values in Garfield County. n Document how rural industrialization (gas, gravel, power lines, etc. ) affects the value of residential property. n Offer mitigation strategies for situations where value losses occur. 2

Land Values Study: Three Phases Phase I. Data Collection and Analysis Phase II. Statistical Analysis and Conclusions Phase III. Solutions and Mitigation Evaluation 3

Phase I: Process n Assembled a data base of 7, 600 sales transactions Ø Unincorporated, residential sales; 1987 -04 Ø Cleaned and added data: gravel roads, geographic features, water & sewer, etc. Ø Ultimately used 20+ variables per property n Assembled gas drilling and industrial data Ø Location of power lines, gravel pits, highway, railroads Ø Location & dates of gas wells n Integrated GIS with Community-Viz mapping software n Analyzed data in light of interviews and anecdotal observations. 4

Phase II: Statistical Analysis n Meet with committees; revised conceptual approach n Completed Statistical analysis Ø Tested 20+ property variables • Land characteristics (e. g. size, presence of water) • Location (e. g. ; RFRV vs. CRV, distance to town) • Structural characteristics (e. g. ; size, age, number of bedrooms) Ø Determined factors that explain value Ø Provided a basis for understanding impact on property value and strategies for mitigation 5

What is Hedonic Regression Analysis n Hedonic regression analysis is a method of explaining demand or prices for a particular good (e. g. a housing unit) by attaching estimates of value to its component characteristics (e. g. size of structure, age, quality of construction) n Why Use? Produces results with statistical authority 6

Variables Tested for the Property Value Models — Included in Model Land Characteristics: • Size (acreage) • Presence of water features • Presence of “good” vegetation (CRV only) Structural Characteristics: • Size of home • New home (less than 10 years old) • Presence of garage (CRV only) • Presence of outbuildings • Heated space in outbuildings • Central wastewater system (RFV only) Locational characteristics/ industrial proximity • Distance from Glen. Sprgs. (CRV only) • Distance from Pitkin County (RFV only) • North of Colorado River (CRV only) • View of Mt. Sopris (RFV only) • Distance to nearest paved road • Distance to nearest gravel pit (CRV only) • Gas well completed within 90 days after sale (CRV only) • Gas well completed less than 2 years prior to sale (CRV only) • Gas well completed more than 2 years prior to sale (CRV only) Value appreciation over time: • Increase in value per acre by year • Increase in value per square foot by year 7

Variables Tested for the Property Value Models — Tested and Rejected Land Characteristics: • South facing percentage • All flat terrain Structural Characteristics: • Number of bedrooms • Number of bathrooms • Construction type (e. g. , modular, condominium, etc. ) • Additional house age groupings (e. g. , 10 to 20 years old) • Water system other than a private well Locational characteristics/ industrial proximity • • • Distance to nearest town Adjoins Federal land Distance to I-70 Distance to railroad Proximity of high voltage lines Proximity of land fill 8

Challenges n Wide variation in property characteristics and locational influences n Value effects across three key dimensions — property characteristics, size and time of sale n Sample sizes diminish with multiple variables n Difficult to measure some key factors n All data sets have some inaccuracies 9

Results n We can explain influences on property values with a reasonable level of accuracy: Ø 76% of value variation in Roaring Fork Valley (2, 726 observations) (95% confidence level) Ø 81% of value variation in Colorado River Valley (4, 727 observations) (95% confidence level) n Provides a reliable basis for overlaying impacts of gas drilling and other industrial effects. 10

Industrial Impacts n We tested effects of highways, railroads, gravel pits, power landfills lines and gas drilling n Also tested positive site attributes: vegetation, views, proximity to USFS lands, rivers n Proximity to highways, power lines, landfills and railroads were not proven to have an impact on values n Proximity to gravel pits and gas drilling has an apparent (but not statistically significant) impact on property values 11

Gas Drilling Data Issues n Gas well permits 5, 010 n Operational wells 2, 674 n Parcels with operational wells 354 n Valid single parcel sales of parcels with operational wells 140 n Final sample “Well impacts” (less than 160 acres) 32 12

Revised Colorado River Valley Property Value Model Total Value = $303, 079 Baseline Property: (Average Property With Gas Well) Well activity at time of Sale $254, 736 Well completed at time of sale $271, 623 Well completed less than 90 days after sale (16%) Well completed long before sale $280, 070 Well older than two years (11%) (8%) 13

Generalized Gas Drilling Impact on Property Value in Colorado River Valley Value Loss/Gain Exploration Phase +$100, 000 Drilling Phase Completion Phase +$50, 000 +$53, 000 Typical Residential Change in Value +$25, 000 $303, 0791 ($32, 000) ($50, 000) Drill Site Properties Change in Value ($49, 000) ($100, 000) 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Months Typical property with a well — 40 acres, small home, 24 miles from Glenwood. 14

Generalized Gas Drilling Impact on Property Value in Colorado River Valley Value Loss/Gain Exploration Phase +$100, 000 Drilling Phase Completion Phase +$50, 000 +$53, 000 Typical Residential Change in Value +$25, 000 $303, 0791 ($32, 000) ($50, 000) Drill Site Properties Change in Value ($49, 000) Perception of Risk Institutional Uncertainty Quality of Life Impacts ($100, 000) 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Months Typical property with a well — 40 acres, small home, 24 miles from Glenwood. 15

Generalized Gas Drilling Impact on Property Value in Colorado River Valley Value Loss/Gain Exploration Phase +$100, 000 Drilling Phase Completion Phase +$50, 000 +$53, 000 Typical Residential Change in Value +$25, 000 $303, 0791 Drill Site Properties Change in Value ($50, 000) Impact of Gas Employment Demand ($100, 000) 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Months Typical property with a well — 40 acres, small home, 24 miles from Glenwood. 16

Impacts of Gas Drilling: Conclusions n Properties that experience drilling see a reduction in market value, but seemingly temporary n On average, net residential loss of value of about 16% during drilling and about 8% three years after drilling ceases n Anecdotal data suggest: Ø There is no average well site Ø Some drilling instances have more severe impacts Ø Problem compounds with contiguous site operations or multiple drilling Ø Micro site issues are hard to capture Ø Recent wells tend to be closer to residential uses 17

Impacts of Gas Drilling: Conclusions (cont. ) n Gas activity also has countervailing positive impacts: Ø Gas employment drives housing demand Ø Property lease payments Ø Site improvements Ø Tax revenues n Mineral owners have legitimate property rights, which can’t be ignored n Drilling is not locally regulated so operational restrictions are limited 18

Mitigation Possibilities Institutional • • Quality of Life • Define and enforce best practices • IGA with COGCC Perception of Risk • • • Education material/seminar Ombudsman Recommend cooperative lenders/brokers Fund property purchase or buy down Education Remedial funds Insurance Certification of completeness Environmental monitoring reporting 19

Land Values and Solutions Study Institutional Changes n Ombudsman/Advocate Ø Document county land value changes over time Ø Represent Owners Ø Clearing House of Information for Appraisers, Realtors and Buyers Ø Environment Response Agent n Intergovernmental Agreement with COGCC n Lending or Property Purchase 20

Quality of Life Mitigation Measures n Phase I Exploration Ø Landowner notification Ø Negotiated surface damage provisions Ø Ground water testing n Phase II Drilling and Field Organization Ø Reasoned environmental protections Ø Reasoned well-siting practices Ø Noise and nuisance abatement 21

Quality of Life Mitigation Measures n Phase III Production and Stimulation Ø Responsible Ø Proper Ø Air stimulation techniques waste disposal and water quality monitoring n Phase IV Abandonment and Reclamation Ø Certification of proper abandonment Ø Reclamation with native topsoil and vegetation 22

Garfield County Land Values and Solutions Study