As part of my recently completed graduate degree in Urban Planning I undertook an independent project that looked at the effects of dam removal on property values in Traverse City, MI. Basically, the city has proposed the removal of four dams along a 17-mile stretch of the Boardman River. However, at least 27 owners of waterfront property along one of the impoundments are vehemently opposed to the plan.
Property owners claim, logically, that removal of the dams (and the subsequent loss of their waterfront property) will result in a massive decline in the value of their land. In response, Traverse City put out a study that claimed property values along the river would not fall due to dam removal, in fact, they could rise by as much as 1% a year over twenty years (that’s in addition to otherwise expected increases in property values).
Like the residents of Traverse City, I had trouble reconciling these two viewpoints and went on a mission to understand the city’s rationale. I read all the publicly available documents, e-mailed officials and residents, and read the reports that the reports were based on (and sometimes the reports those were based on as well!). What I found was that the city’s logic for predicting a net gain in property values was at the very least poorly explained and at the worst grossly misguided.
As an Urban Planner, it pains me when public planning processes are not seen as fair and transparent. I support the goals of restoring natural river flows and ecosystems, but if cities cannot be honest with their residents, they will increasingly face sterner opposition from property owners effected by the dam removal process.
The Boardman River Dams Project:
An economic analysis critique of the predicted effects of dam removal
on property values in Traverse City, MI
The Boardman River runs for 130 miles through northwest Michigan. Between its source in Kalkaska County and the river mouth in Traverse City, it drains 300 square miles of the Boardman Watershed before emptying into Grand Traverse Bay, which in turn empties into Lake Michigan. The Boardman River is one of the top-ten fisheries in Michigan with more than 36 miles designated as Blue Ribbon trout fishery. It is also a State of Michigan designated “Natural River” (Orientation, 2006). Traverse City is a major summer tourism destination for the region and the river receives its share of recreational visitors looking to fish, hike, camp, canoe, and kayak. However, both the river’s Blue Ribbon trout fishery and its recreational uses are limited by four dams.
The Boardman River Dams are located over a 17-mile stretch of the river (see Map 1). The first dam was constructed in 1867. Known as the Union Street dam, it is just 10 feet tall and approximately 1.5 miles from the river mouth. It was originally built to supply power for a now defunct flourmill. Its current purpose is to maintain the level of Boardman Lake, a naturally occurring water body on the river, and to keep invasive species like the sea lamprey from migrating upstream (Preliminary, 2006). The other three dams – Sabin Dam, Boardman Dam, and Brown Bridge Dam – were originally built between 1894 and 1921 but were all reconstructed in 1930 for use as hydroelectric plants. Traverse City Light and Power operated these plants until their delicensing in 2005, when it was no longer considered economically viable to continue their operation. The heights and distances from the river mouth are shown in Table 1.
The Boardman River Dams disrupt the natural ecosystem of the river. As water slows and is stored behind the impoundments it warms. As this water is released from the dams it increases the average temperature of the entire downstream river. This creates a habitat for coolwater fish species, such as northern pike and walleye, which would not otherwise be found in the river, and disrupts the original coldwater habitat preferred by brook trout and brown trout. The warming of the river also creates an ideal environment for the invasive zebra mussel (Preliminary, 2006).
In addition to harming the natural ecosystem, the dams also disrupt recreational uses of the river. For example, the annual number of paddling days along the Boardman is estimated to be between 5,500 and 16,900 (Bingham, 2008). However, 82% of these occur above the Boardman Dam where the river is free of impoundments. This is most likely because the dams are seen as dangerous by boaters and because it is considered a hassle to have to portage boats over multiple dams. The figures are roughly the same for anglers. Former Traverse City Mayor Mike Estes puts it like this, “the river system is blocked by so many dams that using the Boardman as a recreational tool really doesn’t exist. If we can open it up, people will be able to canoe, kayak, and raft from areas way outside the city, on the edges of the county, all the way into downtown Traverse City. I see that as a huge recreation opportunity” (Estes qtd. in Mackinac, 2009).
It was ultimately these two viewpoints – that the river should be returned to its ecologically natural state and provide a better recreational experience for users – that lead to the Boardman River Dams Committee and Implementation Team recommending the removal of three of the dams on January 22, 2009. Only the Union Street Dam in downtown Traverse City is planned to remain intact, though an improved fish passage will provide better access to the river for Great Lakes fish while continuing to limit access for the invasive sea lamprey.
While several legitimate reasons for keeping the Union Street Dam were given, others have to be deduced. For example, dam removal can be a politically sensitive issue, as property owners near and on the impounded waterfront come to perceive the area as a lake. Boardman Lake, behind the Union Street Dam, has 40 waterfront properties (Abrams, 2009) and 1,778 residential parcels within ½ mile (Bingham, 2008) – more than any other impoundment. Therefore, keeping the dam was likely to satisfy the largest segment of the population potentially impacted by dam removal. The remaining impoundments have only 27 waterfront properties (Abrams, 2009) and 111 residential parcels within ½ mile (Bingham, 2008). Nevertheless, the effect of dam removal on property values has become a central issue in the fight over the future of the dams.
|Table 1: Boardman River Dams|
|Dam||Miles from River Mouth||Dam Height||Age (years)||Fish Passage||Purpose||Pooled Acres||Condition|
|Union St.||1.5 miles||10 feet||143||Yes||Maintain level of Boardman Lake||350||Good|
|Sabin||5.3 miles||32 feet||80||No||Hydroelectric||40||Good|
|Boardman||6.1 miles||56 feet||80||No||Hydroelectric||104||Insufficient spillway|
|Brown Bridge||18.5 miles||46 feet||80||No||Hydroelectric||191||Insufficient spillway|
Two diametrically opposing viewpoints have emerged concerning the effects of dam removal on property values. Landowners along the Boardman River impoundments believe that having water frontage confers a premium on the value of their property. The research produced by the Boardman River Dams Committee (BRDC), however, states that “a residential property near a free-flowing river is worth more than an identical property located near a small impoundment,” and that, “the values of an identical property near an existing impoundment and near a recently removed impoundment are not statistically different” (Bingham, 2008). Two quotes from a 2009 documentary by the Mackinac Center for Public Policy help to sum up these contradictory views:
“It’s an ugly scene here. The property values have dropped dramatically. The day that they even mentioned draining [Keystone Pond] down I believe that I probably lost well over $100,000 between the two lots that I have here.”
– Bill Lane, Homeowner
“I can’t imagine how the property value could have dropped substantially. I don’t believe any of them have lost any waterfront. They had something that they never essentially paid for when the original property was purchased. If these residents feel that they were sold this property with the knowledge that they had actual ownership up to the current water’s edge then I think they need to go back after the real estate people who sold them that property under false conditions.”
– Mike Estes, Mayor, Traverse City
In order to understand how these contrasting beliefs were formed and endure today, as Traverse City begins planning the first dam removal in Spring 2011, a closer look at the perceptions and economic analysis is needed.
Property Owner Perceptions
There are an estimated 2,500 dams in Michigan and 79,000 in the United States. (Orientation, 2006). Over the past 50 years only around 600 have been removed across the entire country (Dam, 2010). Little research has been directed towards assessing the impact of dam removal on property values. In this vacuum of information, property owners are often left to believe their own worst fears. “One of the biggest and most consistent concerns expressed is about the appearance of the former impoundment after dam removal,” states Helen Sarakinos in her paper Social Perspectives on Dam Removal (Graf, 2003). “Two of the most commonly expressed concerns are that the river will dry up without the dam and that removal will leave an eyesore in the form of a permanent mudflat” (Graf, 2003). Robert Kaufman, an attorney representing property owners on an impoundment states, “one of my original clients was foreclosed on and went bankrupt because his former lakefront property became more or less unsalable because it looked like a lunar landscape or mudflat” (Kaufman qtd. in Mackinac, 2009).
These concerns highlight two common misconceptions. The first is that impoundments formed by small dams create lakefront property. The truth is that small impoundments more often merely alter the landscape from one type of wetland, a forest or scrub-shrub wetland to another type of wetland, either an emergent-vegetation or open-water wetland. It is important to note that these properties are near wetlands, not lakes, and housing prices should reflect this difference. The second misconception is that the exposed bottomlands will remain indefinitely as mudflats. The truth is that ecological succession begins as soon as the bottomlands are exposed but takes time to yield results. While mudflats may persist in the short-term, after a couple years they will be replaced with woody vegetation typical of a scrub-shrub wetland. This type of wetland supports an increased quantity of wildlife habitat and is more aesthetically pleasing than property owners perceive the initial mudflat phase to be. Ultimately, the question comes down to how different types of wetlands effect property values.
Boardman River Dams Committee Perceptions
In January 2009 BRDC released a detailed analysis of alternatives characterizing six different management options for the river. This report states:
“The value of an individual residential parcel in the vicinity of a removed dam could fall, on average, by as much as 6%, if all other influences on property values are held constant. About two years after the removal, the effected properties are predicted to begin to increase in value. 20 years after removal, the properties, on average, could increase in value by as much as 18%, or approximately 1% per year, relative to current conditions” (Environmental, 2009).
This assessment was applied to the 4,000 properties within ½ mile of the Boardman River and shows that, 20 years after dam removal, property values will increase by $1.9 million, holding all other values constant (see Table 2). Spread out over 4,000 properties, this results in an average increase of $475 per property over 20 years due solely to dam removal – a relatively small increase that at least seems to show that property values wont decline.
This conclusion was based on an economic and social analysis of the Boardman River Dams performed by Veritas Economic Consulting (Veritas). The study was outsourced by BRDC’s principle consultant, Environmental Consulting & Technology, Inc. (ECT), and was completed in July 2008. Rather than being an analysis, the report is actually a quantification of existing conditions that discusses in detail how Veritas will go about performing an economic and social analysis during a “future analysis phase.” However, it appears that rather than having Veritas conduct the “future analysis phase,” ECT used the Veritas recommendations to draw their own conclusions.
The Veritas Study
In regards to property values, Veritas recommended a “structurally calibrated transfer approach” to economic modeling. “This approach uses established theory to link the dam-management options to human welfare and available data to empirically specify the linkages” (Bingham, 2008). In other words, rather than conducting their own study, which would take time and money, Veritas suggested looking at other studies and applying their findings to the Boardman River Dams Project. Specifically, the Veritas report recommended using a study done by Provencher, Sarakinos, and Meyer entitled Does Small Dam Removal Affect Local Property Values? Veritas states, “their study uses hedonic analysis as the methodology, which statistically decomposes the value of a residential property into its component parts. In this way, the approach identifies the relative contribution to a porperty’s value that can be uniquely attributed to proximity to an existing impoundment, to a recently removed impoundment, or to a free-flowing river” (Bingham, 2008).
The Veritas study also highlights potential problems with using the Provencher report as a transfer study. They note that the study “contains sufficient information to estimate a potential change in property values within the first two years of dam removal and 20 years after removal.” They also note, “the underlying data provides only a handful of properties that contained frontage near recently removed dams.” Veritas believed that these data gaps could be addressed with statistical techniques during an expected future analysis phase. However, since this future analysis phase either never occurred or at least was never cited by ECT’s Alternatives Analysis Report, the reader is left to question how ECT arrived at their property value estimates. It seems likely that ECT simply took the Veritas recommendations and drew their own conclusions. In order to understand these conclusions, the Provencher study recommended by Veritas will need to be researched more thoroughly.
The Provencher Study
The Provencher, Sarakinos, and Meyer study looked at 773 properties within ¼ mile of each site. Sites looked at included properties near a recently removed dam, near an intact dam, and near a free flowing stream. Only 116 observations were from parcels with water frontage and only 6 had water frontage on a removed dam site. The analysis includes market sales data over the period 1993-2002 for sites in south-central Wisconsin. The small impoundments in the study held between 8 and 194 surface acres of water. The average impoundment size was 57.9 acres. The study notes, “many of the reservoirs formed by impoundments at the study sites are quite small and shallow” (Provencher, 2006).
The analysis showed results that conflicted with intuition. Mainly that property value increases with distance from a small impoundment and that water frontage on a small impoundment is a nonsignificant contributor to property value (Provencher, 2006). Possible explanations for these effects include risk of flood damage, perennial damage issues such as water seepage into basements, mosquito infestations on impoundments, foul odors associated with algae blooms and decaying vegetation (Provencher, 2006). The analysis does not include findings used by ETC to estimate changes in property values. Namely, the Provencher Report does not find that property values will decrease for two years nor that they will rise by 1 percent a year for the subsequent 18 years. Instead, the report finds, “removing a dam does little harm to property values in the short run and serves to increase property values in the long run, as the stream and associated riparian zone matures to a ‘natural’ free-flowing state, or is managed as desirable open space” (Provencher, 2006).
The only reference Provencher et al. make to property values decreasing for two years and then rebounding is a mention of an unpublished study done by Carmen Wagner in 2001, which found that “riparian property values either remained unchanged, or dropped temporarily and rebounded within two years” (Wagner qtd. in Provencher, 2006). When contacted for this report Miss Wagner conveyed that her study was done as an assignment for a class she was taking in Regional and Urban Planning and that it’s findings should not be used for determining property values. “My advisor’s comments on the paper,” states Wagner, “were that it was a good start, but more analysis was needed to make any firm conclusions or recommendations on property values.” Clearly, an informal student paper is not a reliable source for predicting changes in property values due to small dam removal.
ETC’s Detailed Analysis of Alternatives Report includes a table reproduced here (see Table 2) that plainly predicts the changes in property values within ½ mile of the Boardman River due to dam removal. Unfortunately, following their sources for these findings does not reveal how these conclusions were deduced. At best it appears that ETC utilized a transfer study of Provencher et al. for their general findings and then utilized a second non-cited source to inform their final figures. At worst, they followed the general principles of Provencher et al. and then flushed it out with numbers that seemed correct. Either way, the ETC Report does not leave the reader confident of their predictions for how removing the Boardman River dams will effect property values.
Critique of Proposed Methods
The ETC report supposedly conducted a structurally calibrated transfer approach using the Provencher et al. study of changes in property values due to small dam removal. Even if this report is given the benefit of the doubt and it is assumed that the transfer approach was used as intended, there are still reasons to question the legitimacy of the report’s conclusions.
It is unclear why the ETC Report looks at property values within ½ mile of the river when Provencher et al. uses only properties within ¼ mile of each site. There is no justification for the doubling of this distance and it risks including properties that are too far away to be impacted by the dam removal. The ETC study includes 3,082 properties (out of a total of 3,921) that are located within ½ mile of Union St. Dam between Boardman Lake and the river mouth in Traverse City. Since Union St. Dam is not being removed these properties will not be impacted by dam removal. All of these properties are over 2 miles away from Sabin Dam, the nearest dam proposed for removal. Yet these properties make up 78.6% of all property values considered in the ETC Report. The ETC Report’s transfer study would be more accurate if it was limited to only properties within ¼ mile of the impoundments proposed for removal and instead of all properties located within ½ mile of the entire river.
The ETC report’s transfer study may be flawed due to a difference in the impoundment sizes studied in Provencher et al. and the sizes of the Boardman River impoundments. Provencher et al. looked at the effects of small dam removal on property values. Though the report does not explicitly define the dimensions of a small dam, one of the report’s authors, Helen Sarakinos wrote another paper entitled Social Perspectives on Dam Removal in which she defines small dams as being less than 25 feet high (Graf, 2003). Only the Union Street Dam is less than 25 feet high. The others are 32, 56, and 46 feet high and thus do not fit Sarakinos’ definition of a small dam. A second indicator is the pooled acres of the impoundments studied by Provencher et al. Despite including a range of impoundment sizes from 8 to 194 acres, the average impoundment size in Provencher et al. was 57.9 acres. The average impoundment size of the Boardman River Dams is 168.25 acres. Even if Union St. dam is not included (because it is the largest impoundment and not scheduled for removal) the average is 111 Acres, nearly twice the size of the average impoundment in Provencher et al.
The size of the impoundments is important because it can dramatically effect the recreational uses of the wetlands. The small impoundments studies by Provencher et al. are more likely to be seen as swamps with little opportunity for fishing or boating. The Provencher et al. report states, “the conclusion that free-flowing rivers confer a price premium on residential property compared to impounded waters is likely due to the small size of the impoundments at our study sites. The conclusion should not be extended to large impoundments where such activities as fishing, boating, and swimming are especially attractive” (Provencher, 2006). In fact, the Veritas report on existing conditions notes that recreational fishing and paddling are activities currently enjoyed on the Boardman impoundments lending further weight to the notion that these may not be considered “small dams” as defined by the Provencher et al. study.
Recommendations for Alternate Studies
Instead of looking at the effects of small dam removal on property values, it is recommended that ETC instead look at studies using a hedonic approach to valuing wetlands. Two such studies are Valuing Urban Wetlands: A Property Price Approach by Mahan, Polasky, and Adams (2000) and A Semiparametric Hedonic Model for Valuing Wetlands by Okmyung Bin (2005). These studies look at the relationships between the four different types of wetlands as well as the shape of the wetlands (linear vs. areal). This is relevant because the Boardman River Dams Project is basically proposing to turn areal emergent-vegetation wetlands into linear scrub-shrub wetlands. Understanding how each type effected property values could yield a more accurate estimate than the Provencher et al. model.
Mahan et al. conducted two analysis models. Model I found that, “while there is a preference to be closer to wetlands, the type of wetland does not seem to matter” (Mahan, 2000). This would seem to indicate that changing the wetland type along the Boardman River would not effect property values. Model II found “a preference by homeowners for scrub-shrub and open-water wetlands, over forested and emergent-vegetation types of wetlands (Mahan, 2000). This would seem to indicate that the proposed change in wetlands might actually increase property values along the Boardman River. However, it’s important to note that the wetland type varies around the shore of each impoundment. Some homes may be on open-water wetlands that will change over time to forested wetlands. Model II would suggest that this could result in a decrease in property values. Mahan et al. indicate a preference for the findings of Model I and suggest that Model II may be inferior.
Bin finds that the proximity to the nearest open water wetland has an overall positive effect on property values. “Property values decline as the distance to the nearest open water wetland increase.” Conversely, proximity to the nearest emergent-vegetation wetland shows a clear negative effect on property values (Bin, 2005). These results show that property owners place a premium on open water wetlands and that changing the wetland type to any other type of wetland will have a negative effect on values. This is the result that most closely reflects our intuition about how people value open water vs. swamps and marshes.
The Boardman River Dams Project is the most comprehensive dam removal project in Michigan history (Orientation, 2006). It could dramatically alter who benefits from the river. Currently, a few local residents enjoy the impoundments while the dams deter tourists and upstream users. If the dams are removed, residents will lose the access they once had to open water while tourists and upstream users will benefit from longer boating trips and an improved coldwater fishery. The city will also benefit through an increase in tourism dollars spent by those beginning and/or ending their river voyages on Boardman Lake in Traverse City. For these reasons it is important that all the rationale for dam removal be clearly stated and easy to understand. This economic analysis of the documents prepared for and by Environmental Consulting & Technology, Inc. on behalf of the Boardman River Dams Committee shows that their findings are inconsistent and poorly explained.
This report has made suggestions and offered advice on how a revised estimate could be approached. It may seem like it’s in the BRDC’s best interest to deny that any loss of property values will take place, however, this also prevents an honest dialogue between those who stand to lose value in their homes and the city/county government that manages the dams. An honest dialogue could produce compromises without either party having to resort to lawsuits. The exposed bottomlands, which have been legally determined to belong to the city/county, are one potential source of compromise. Landowners losing frontage on an impoundment could be offered some percentage of this land along with a tax reduction to offset their losses. While it is unlikely that all parties can be satisfied, the more parties that are satisfied, the less political backlash will be faced by local governments.
This report recognizes that the main goals of the BRDC are to restore the natural flow and ecosystem of the river as well as to increase opportunities for recreational users. Because dam removal is increasingly recognized as a cost-effective river restoration tool, state and federal grants and other funding are becoming more readily available for dam removal (Graf, 2003), in spite of this, the process must be seen as fair and transparent or else Michigan will increasingly face sterner opposition from property owners effected by the dam removal process.
Bin, Okmyung. 2005. “A Semiparametric Hedonic Model for Valuing Wetlands.” Applied Economic Letters 12: 597-601.
Bingham, Matthew F., Mathews, Kristy E., Morton, Dawn M., Abrams, Elliot S., and Li, Zhimin. July 2008. “Economic and Social Analysis of the Boardman River Dams: Quantification of Existing Conditions.” Available at: http://theboardman.org/d/Econonic_Social_Analysis_Veritas.pdf.
Dam Removal: Frequently Asked Questions. American Rivers. Website accessed December 2010. Available at: http://www.americanrivers.org/our-work/restoring-rivers/dams/background/faqs.html.
Environmental Consulting & Technology, Inc. October 2008. “Boardman River Feasibility Study: Detailed Analysis of Alternatives Report.” Available at: http://theboardman.org/d/Detailed_Analysis_of_Alternatives1.pdf.
Graf, William L., editor. Dam Removal Research: Status and Prospects. 2003. Washington, DC: The H. John Heinz III Center for Science, Economics and the Environment. xii + 152 pages, illustrated. ISBN 0-9717592-4-3.
Mackinac Center for Public Policy. 2009. “Boardman River Dams Grand Traverse County. Episode One: Drawdown & Devastation.” Documentary available at: http://vimeo.com/6987904.
Mahan, Brent L., Polasky, Stephen, and Adams, Richard M., February 2000. “Valuing Urban Wetlands: A Property Price Approach.” Land Economics 76(1): 100-113.
Orientation Packet. 2006. “The Boardman River Dams Project: Using Public Process to Determine the Fate of the Dams.” Available at: http://theboardman.org/d/orientation_packet.pdf.
Preliminary Restoration Plan. July 2006. “Preliminary Restoration Plan: Boardman River, Section 506 Great Lakes Fishery and Ecosystem Restoration.” Available at: http://theboardman.org/project/doc/prp/.
Provencher, Bill, Helen Sarakinos, and Meyer, Tanya. “Does Small Dam Removal Affect Local Property Values? An Empirical Analysis.” Staff Paper No. 501. Madison, WI: University of Wisconsin, Agricultural and Applied Economics.