Impact of Green Infrastructure on Property Values within the Milwaukee Metropolitan Sewerage District Planning Area: Four Case

Studies
Spring 2013 Presented by Catherine Madison, AICP Center for Economic Development, UWM For further information or comments please contact cmadison@uwm.edu (414) 229-6155
1

Case Studies: Background

Study began as an EPA request for information from MMSD (and other sewerage districts throughout US) Measuring the $ Impact of Green Infrastructure (GI) projects on property values In comparison to other forms of public infrastructure (roads, water and sewer lines, utilities)
Estimates on the financial benefits of roads, water/sewer lines, utilities, etc on property values are well established GI is relatively new, impacts are not established at this point

EPA and MMSD: Provide case studies to show the impact of GI projects on property values
Using Regression Modeling to measure impact
2

Case Studies: Background

Green Infrastructure examples
Greenways Rain Gardens Wetlands Stormwater Trees Green Roofs Bio-Swales Porous Pavement Native Landscaping Rain Barrels/Cisterns Alleys/Streets/Parking Lots

Multi-purpose and multi-functional, all methods for reducing stormwater runoff

3

Case Studies: Methodology

Hedonic Regression Analysis
Standard method for real estate valuation modeling The value of a property is based upon its components: Physical characteristics: square footage, lot size, number of bathrooms and bedrooms, number of units Neighborhood quality: proximity to parks/rivers/lakes or other amenities, access to transit and transportation Socio-Economic data: educational attainment, age, household size, household income Others: Presence of Business Improvement District or Tax Increment Finance District Proximity/adjacency to GI Modeling determines the portion of the value contributed by GI
4

Case Studies: Methodology

Fiscal Impact Analysis and Return On Investment
Estimating the amount of new property tax revenue created by the increase in property values Payoff of GI investment Comparing results of the models to the real world conditions

Four MMSD Study Areas Identified:

The Menomonee Valley The Brewery (former Pabst site) Lincoln Creek neighborhood Shorewood Downspout Disconnection Program and Raingardens

Case Studies: Results

The Menomonee Valley Redevelopment
Industrial Area
History (from Milwaukee Road Shops, to Menomonee Valley Partners) Significantly blighted

Tax Increment Financed Redevelopment

Four TIFs as mechanism to spur private redevelopment = $31M Brownfield cleanup and environmental/landuse planning Major infrastructure projects - increasing connectivity and access, reducing stormwater runoff

GI integrated into significant redevelopment project (stormwater runoff)

12 projects including bioretention facilities, green roofs, green alleys/streets, porous pavement - the largest being a 3 acre stormwater park Design contest for stormwater park (Wenk Associates and HNTB) greater aesthetic appeal and recreational facilities GI Costs = $1.8M, with MMSD portion being $835,000
6

Case Studies: Results

Menomonee Valley Model

Study Group = 117 industrial properties near GI Control Group = 510 comparables 6,319 observations over a 9 year cycle A Robust Sample

Case Studies: Results

The Menomonee Valley Model
Variables Green Infrastructure (Key Independent Var) Building Area (in square feet) Dependent Var = Total Assessed Value Parcel Area (acres) Distance to Lake Michigan Distance to Closest River or Stream 2 R = .273 Distance to Nearest Freeway Ramp Distance to Nearest Bus Stop Model performed Tax Increment Finance District (TID) Business Improvement District (BID) well (coefficients) Number of Brownfields within 1,000 feet Number of Vacant Properties within 1,000 feet Number of Parks within 1,000 feet Percent of Population with High School Diploma Percent of Population In Poverty Percent of Population Black Median Household Income (log) Population Density Parking Lot (dummy variable) Constant Coefficient (Std. Error) Sig. (P <|z|) 0.058 (0.022) 0.010** 4.70E-06 (0.000) 0.000** 1.74E-06 (0.000) 0.000** 3.92E-06 (0.000) 0.327 -1.63E-04 (0.000) 0.000** -1.02E-05 (0.000) 0.453 6.74E-05 (0.000) 0.318 -0.088 (0.026) 0.001** 0.118 (0.018) 0.000** 0.017 (0.004) 0.000** -0.001 (0.000) 0.121 0.02 (0.009) 0.033** 2.232 (0.122) 0.000** 0.946 (0.108) 0.000** -0.112 (0.095) 0.242 0.415 (0.051) 0.000** 0.007 (0.005) 0.199 -0.094 (0.035) 0.007** 5.826 (0.557) 0.000** 8

Case Studies: Results

The Menomonee Valley Model Holding all other variables constant, in any give year, the assessed property values were 5.8 percent higher than they otherwise would have been without the GI/Redevelopment This translates into a $13,317 increase in assessed value for the average industrial property in Milwaukee (which is about $223,555 and is not in a TID or BID, nor is it a parking lot), which adds up to roughly $1.56M in added assessed value for 117 properties.. BUT..still some unanswered questions

Case Studies: ROI

The Menomonee Valley Return On Investment
Figure1: Property Value Growth Rates in the Menomonee Valley and Control Group Areas, 2002-2011

Source: City of Milwaukee MPROP database and UWM-Center for Economic Development
10

Case Studies: ROI

Comparison of Assessed Values for the Study and Control Groups within the Menomonee Valley Model

Year

Study Group Total Assessed Value* 75,233,145 80,886,664 89,207,703 91,241,883 93,330,453 96,813,589 110,674,367 143,790,214 157,543,696 167,517,374 170,256,000 Average Assessed Value* 723,396 763,082 803,673 814,660 833,308 820,454 892,535 1,141,192 1,211,875 1,213,894 1,224,863 Growth Rate -7.5 10.3 2.3 2.3 3.7 14.3 29.9 9.6 6.3 1.6

Control Group Total Assessed Value* 115,807,484 125,930,508 129,198,672 137,486,387 139,712,174 152,952,429 168,107,658 187,855,065 185,947,938 188,298,577 186,818,348 Average Assessed Value* Growth Rate -265,613 8.7 286,206 2.6 6.4 1.6 9.5 9.9 11.7 -1.0 1.3 -0.8

Properties 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 104 106 111 112 112 118 124 126 130 137 140

Properties 436 440 447 454 453 466 473 481 488 498 509

289,035 302,833 308,415 328,224 355,407 390,551 381,041 378,110 367,030

Note: Assessed values are adjusted for inflation to reflect 2011 dollars. Source: City of Milwaukee MPROP database and UWM-Center for Economic Development

11

Case Studies: ROI

The Menomonee Valley Return On Investment

Note: Assessed values are adjusted for inflation to reflect 2011 dollars. Source: City of Milwaukee MPROP database and UWM-Center for Economic Development

12

Case Studies: ROI

Table 3: Non-Industrial Land Uses and Assessments in the Menomonee Valley, 2011

2011 Number of Parcels

35 163 24

Land Use Type

Commercial Residential Under Development1

Total Assessed Value

52,025,300 40,485,600 8,061,100

SUBTOTAL
Industrial Properties

222
140

100,572,000
170,256,000

TOTAL
1

362

270,828,000

A total of 24 properties within the Valley study area were identified as under development in 2011. These properties will likely add additional assessed value upon completion. Source: City of Milwaukee MPROP database and UWMCenter for Economic Development

13

Case Studies: ROI

The Menomonee Valley Return On Investment Based on a Total Assessed Value of $270.8M in 2011 (was $168M in 2001, adjusted to 2011 dollars):
MMSD should anticipate receiving about $400,825 in annual tax revenues generated from all taxable properties within the value Based on MMSDs $834, 711 initial investment, payoff would be in 3 years, assuming that all of the revenues generated are directed to payoff the investment Estimate does not include any additional user charges nor any costs that were likely to accrue had the GI not been developed

14

Case Studies: Results

The Brewery Redevelopment
Commercial/Mixed Use
History former Pabst Brewery location Significantly blighted, abandoned

Tax Increment Financing

28 Pabst Buildings added to National Register of Historic Place historic preservation tax credits 1 TID $29M (over half went towards demo/abatement and hist pres easements) Significant amount of private investment in GI as well

GI integrated directly into infrastructure (ex roadside bioswales, Zilber Park)

LEED Platinum Certification: GI is integral to this designation GI Costs = $3.2M (both public and private sources), with MMSD portion being $1.04M Anticipated future GI investment = $6.5M from public and private sources
15

Case Studies: Results

The Brewery Redevelopment

16

Case Studies: Results

The Brewery Redevelopment
Variables Dependent Var = Total Assessed Value R2 = .503 Model performed well (coefficients) Green Infrastructure (Key Independent Var) Building Area (square feet) Parcel Area (square feet) Distance to Lake Michigan Distance to Closest River or Stream Distance to Nearest Freeway Ramp Distance to Nearest Bus Stop Tax Increment Finance District (TID) Business Improvement District (BID) Number of Brownfields within 1,000 feet Number of Vacant Properties within 1,000 feet Number of Parks within 1,000 feet Percent of Population: High School Graduates Percent of Population: In Poverty Percent of Population: Black Median Household Income (log) Population Density Parking Lot (dummy) Constant Coefficient 0.114 3.29E-06 0.000 0.000 -0.001 9.17E-06 0.000 -0.002 0.100 0.044 0.007 0.033 0.647 1.054 0.383 0.203 0.012 -0.497 9.694

(Std. Error) (0.057) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.064) (0.093) (0.011) (0.003) (0.022) (0.174) (0.185) (0.229) (0.093) (0.012) (0.082) (0.992)

Sig. (P <|z|) 0.047** 0.000** 0.000** 0.000** 0.000** 0.873 0.174 0.973 0.283 0.000** 0.054* 0.151 0.000** 0.000** 0.095* 0.031** 0.337 0.000** 17 0.000**

Case Studies: Results

The Brewery Model Holding all other variables constant, in any give year, the assessed property values were 11.4 percent higher than they otherwise would have been without the GI/Redevelopment This translates into a $22,062 increase in assessed value for the average commercial property in Milwaukee (which is $161,504 and is not in a TID or BID, nor is it a parking lot), which adds up to roughly $485,700 in added assessed value for 22 Brewery properties.. Again.. unanswered questions

18

Case Studies: ROI

The Brewery Return On Investment
Figure 3: Property Value Growth Rates in The Brewery and Control Group Areas, 2004-2011

Source: City of Milwaukee MPROP database and UWM-Center for Economic Development
19

Case Studies: ROI

Comparison of Assessed Values for the Study and Control Groups within The Brewery Model

Year

Study Group Total Assessed Value 6,589,099 6,448,788 6,425,208 7,549,300 9,258,389 12,699,894 23,327,749 39,892,359 43,082,200 Average Assessed Value 439,273 429,919 401,576 444,076 544,611 577,268 777,592 1,246,636 1,267,124 Growth Rate --2.1 -0.4 17.5 22.6 37.2 83.7 71.0 8.0

Control Group Total Assessed Value 138,014,476 143,688,207 157,172,395 161,954,462 175,430,314 222,720,767 226,746,028 199,808,920 200,038,500 Average Assessed Value 624,500 644,342 698,544 701,102 724,919 909,064 875,467 771,463 757,722 Grow th Rate -4.1 9.4 3.0 8.3 27.0 1.8 -11.9 0.1

Properties 2003 2004 2005 2006 2007 2008 2009 2010 2011 15 15 16 17 17 22 30 32 34

Properties 221 223 225 231 242 245 259 259 264

Note: Assessed values are adjusted for inflation to reflect 2011 dollars. Source: City of Milwaukee MPROP database and UWM-Center for Economic Development

20

Case Studies: ROI

The Brewery Return On Investment

Note: Assessed values are adjusted for inflation to reflect 2011 dollars. Source: City of Milwaukee MPROP database and UWM-Center for Economic Development

21

Case Studies: ROI

The Brewery Return On Investment Based on a Total Assessed Value of $43.1M* in 2011 (was $6.6M in 2003, adjusted to 2011 dollars):
MMSD should anticipate receiving about $62,038 in annual tax revenues generated from all taxable properties within the value Based on MMSDs $1.04M initial investment, payoff would be in 17 years, assuming that all of the revenues generated are directed to payoff the investment Estimate does not include any additional user charges nor any costs that were likely to accrue had the GI not been developed Development within The Brewery is ongoing these estimates are based on 2011 conditions
*Includes addition of 19 properties (total of 34)
22

Case Studies: Results

Lincoln Creek
Residential
History concrete channels Stormwater and flooding Infrastructure was significantly blighted

Redevelopment was entirely GI

Limited this model to 2 very similar reaches (otherwise number of cases within model would have been unwieldy) GI Costs = $10.95M
Widen/deepen creek Concrete removal Grading and erosion control Adding low flow habitat Bridge revetments, culvert

23

Case Studies: Results

Lincoln Creek

24

Case Studies: Results

Variables Green Infrastructure Building Area (square feet) Parcel Area (square feet) Dependent Var = Age Total Assessed Value Age2 Bedrooms Bedrooms2 R2 = .754 Baths Distance to Lake Michigan Model performed Distance to Closest River or Stream very well (coefficients) Distance to Nearest Freeway Ramp Distance to Nearest Bus Stop Number of Brownfields within 1,000 feet Number of Vacant Properties within 1,000 feet Number of Parks within 500 feet Number of Parks within 1,000 feet Percent of Population In Poverty Percent of Population: Black Percent of Population: Hispanic Percent of Population: High School Graduates Median Household Income (log) Population Density Constant

Lincoln Creek

Coefficient 0.204 4.102E-04 1.690E-05 -4.387E-03 2.630E-05 0.267 -0.039 0.099 4.890E-06 7.300E-06 9.800E-06 1.930E-05 0.015 -0.0070 -0.0671 0.0500 0.7305 -0.2601 1.1105 2.4755 0.1624 -0.010 6.691

(Std. Error) (0.003) (0.000) (0.000) (0.000) (0.000) (0.015) (0.002) (0.007) (0.000) (0.000) (0.000) (0.000) (0.002) (0.000) (0.005) (0.004) (0.026) (0.013) (0.026) (0.028) (0.013) (0.000) (0.154)

Sig. (P <|z|) 0.000** 0.000** 0.000** 0.000** 0.000** 0.000** 0.000** 0.000** 0.000** 0.005** 0.000** 0.027** 0.000** 0.000** 0.000** 0.000** 0.000** 0.000** 0.000** 0.000** 0.000** 0.000** 25 0.000**

Case Studies: Results

Lincoln Creek Model Holding all other variables constant, in any give year, the assessed property values were 20.4 percent higher than they otherwise would have been without the GI/Redevelopment This translates into a $19,174 increase in assessed value for the average residential property in Milwaukee (which is $90,253 single family home with 3 bedrooms) This adds up to roughly $18.33M in added assessed value for study area properties Again.. unanswered questions

26

Case Studies: ROI

Lincoln Creek Return On Investment
Property Value Growth Rates in Lincoln Creek and Control Group Areas, 1999-2011

Source: City of Milwaukee MPROP database and UWM-Center for Economic Development
27

Case Studies: ROI

Comparison of Assessed Values for the Study and Control Groups within the Lincoln Creek Model

Year Properties 959 958 958 957 958 958 960 960 960 959 959 959 960 961

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Study Group Total Average Assessed Assessed Value Value 58,248,826 60,739 57,323,088 59,836 64,951,010 67,799 63,230,130 66,071 71,184,495 74,305 74,249,642 77,505 80,598,302 83,957 88,553,413 92,243 97,480,808 101,543 94,862,660 98,918 93,929,775 97,946 80,507,676 83,950 77,835,773 81,079 75,500,500 78,565

Growth Rate --1.5 13.3 -2.5 12.5 4.3 8.3 9.9 10.1 -2.6 -1.0 -14.3 -3.4 -3.1

Properties 1,193 1,195 1,195 1,196 1,199 1,197 1,202 1,201 1,205 1,205 1,205 1,205 1,206 1,207

Control Group Total Average Grow Assessed Assessed th Value Value Rate 111,163,714 93,180 -109,255,756 91,427 -1.9 105,703,368 88,455 -3.3 113,901,821 95,236 7.7 129,096,419 107,670 13.1 134,126,591 112,052 4.1 143,472,795 119,362 6.5 153,853,368 128,104 7.3 165,910,313 137,685 7.5 161,841,232 134,308 -2.5 157,657,314 130,836 -2.6 148,303,668 123,074 -5.9 140,358,096 116,383 -5.4 136,347,600 112,964 -2.9

Note: Assessed values are adjusted for inflation to reflect 2011 dollars. Source: City of Milwaukee MPROP database and UWM-Center for Economic Development

28

Case Studies: ROI

Lincoln Creek Return On Investment
Table 8: Land Uses and Assessments within the 600 Foot Buffer Boundary in Lincoln Creek Reaches 2 and 6, 2011 2011 Land Use Type Commercial Industrial Under Development1 Residential Properties Not in Study Group2 SUBTOTAL Residential Properties in Study Group TOTAL
1

Number of Parcels 30 11 33 343 417 961 1,378

Total Assessed Value 6,622,300 2,070,300 102,600 34,914,100 43,709,300 75,500,500 119,209,800

A total of 33 properties within the Lincoln Creek study area were identified as under development in 2011. These properties will likely add additional assessed value upon completion. 2 Non-single family home residential properties.

29

Case Studies: ROI

Lincoln Creek Return On Investment Based on a Total Assessed Value of $119.2M in 2011:
MMSD should anticipate receiving about $176,430 in annual tax revenues generated from all taxable properties within the value Based on MMSDs $10.95M initial investment, payoff would be in 62 years, assuming that all of the revenues generated are directed to payoff the investment Estimate does not include any additional user charges nor any costs that were likely to accrue had the GI not been developed, nor the negative impact on property values given the history of flooding and basement backups within the study area

30

Case Studies: Results

Shorewood
Residential
History Stormwater: flooding and basement backups Downspout Disconnection is an invisible infrastructure

GI: Downspout Disconnection Program and Rain Gardens

GI Costs = Minimal Permit required for downspout disconnection

Regression model was different than others.

Cross-sectional regression Drawback: Not as robust No annual assessment data therefore needed to rely on sales data could include only properties that were sold between 2001 and 2010 for both sample and control groups
31

Case Studies: Results

Shorewood

32

Case Studies: Results

Shorewood
2 Independent GI Variables Dependent Var = Total Assessed Value R2 = .744 Model performed very well (coefficients) Results are spurious
Variables Green Infrastructure (Downspout Disconnect) Key Var Green Infrastructure (Rain Garden) Key Var Building Area (sq. ft.) Parcel Area (sq. ft.) Age Age2 Bedrooms Bedrooms2 Full Bathrooms Half Bathrooms Fireplace Openings Enclosed Porch (sq. ft.) Rec Room (sq. ft.) Distance to Lake Michigan Distance to Lake Michigan2 Distance to Closest River/Stream Distance to Nearest Freeway Ramp Distance to Nearest Bus Stop # Brownfields within 500 ft. Distance to Nearest Park % Pop. In Poverty % Pop. Black % HS Grad. Rate Median Household Income (log) Population Density Coefficient -0.054 -0.077 2.19E-04 -0.026 0.023 -1.54E-04 0.196 -0.022 0.057 0.072 0.069 2.24E-04 1.18E-04 -2.64E-04 4.27E-08 2.96E-05 1.66E-05 -3.27E-06 -0.026 1.78E-05 -0.266 2.816 3.383 -0.005 0.003 (Std. Error) (0.020) (0.051) (0.000) (0.055) (0.003) (0.000) (0.035) (0.004) (0.011) (0.012) (0.010) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.012) (0.000) (0.402) (2.548) (1.330) (0.014) (0.004) Sig. (P <|z|) 0.009** 0.133 0.000** 0.633 0.000** 0.000** 0.000** 0.000** 0.000** 0.000** 0.000** 0.039** 0.000** 0.000** 0.000** 0.292 0.252 0.858 0.040** 0.245 0.509 0.269 0.011** 0.733 33 0.522

Case Studies: Results

Shorewood Model
Holding all other variables constant, in any give year, the property sales prices were 5.4 percent lower than they otherwise would have been without the downspout disconnection; rain gardens had neither a positive nor negative impact on sales prices This translates into a $13,192 decline in sales price for the average single family home in Shorewood ($337,753 w/ 3 bedrooms) PROBLEMS:
Cross-sectional data is less comprehensive Invisible infrastructure Selection Bias (those with flooding problems may be choosing to disconnect their downspouts with greater frequency) Contamination (properties that disconnect without registering)

Shorewood results are inconclusive

34

Thanks for your input!

Questions and Comments:
Draft report WILL BE available online at CEDs website soon!
https://www4.uwm.edu/ced/

Please contact Kate Madison at cmadison@uwm.edu or by phone at (414) 229-6155

35