An overview of methodologies used to value “green space”

Posted on April 1, 2014 · Posted in Blog

From time to time, we will public a blog post on a specific environmental economics topic. Today’s is brought to you by Economist at Large, William Li, who put together this excellent summary of methodologies used to value ‘green space’. We did some work with William on this in 2013 and thought this might be a handy resource for anybody working in the area.

It is not difficult to see that green spaces are valuable in highly urbanised settings. There are numerous benefits in being able to access green spaces including recreational opportunities, aesthetic enjoyment, environmental and agricultural functions, as well as value from preserving open spaces for future generations (Brander and Koetse, 2011).

However, as green spaces typically have public good characteristics, they tend to be underprovided in the absence of policy intervention (Kotchen and Powers, 2006; Smith et al, 2002), due to an inability to calculate exactly how much we value these types of assets.

Economists have developed a number of techniques to evaluate the value of environmental assets, but the values produced by each technique can vary noticeably. The most commonly used techniques employed by economists in the literature are hedonic pricing models (HPM), contingent valuation studies (CVM) and travel cost analyses (TCM).


Hedonic pricing model

What is it?

A hedonic pricing model(HPM) estimates the impact of economic value that parks and environmental variables to market prices. This is most commonly seen in housing prices, where it is assumed the price of houses reflect the characteristics of that house, including its access to particular environmental assets.

Unlike other valuation methods such as the travel cost method and the contingent valuation method, hedonic pricing does not rely on survey data, and instead uses property data, which is typically more robust.


Although there have been many studies using HPMs, the studies display wide variation in their characteristics with respect to model specification, sample size, study area and time period. Brander and Koetse (2011) collected more than 52 hedonic pricing studies on open space, and performed a meta-analysis on the results of 12 of those studies. They found that house prices demonstrated an average increase in house price of 0.1% when they are 10 m closer to open space. However, they found this relationship is non-linear; increases in house prices are stronger the closer the house is to the green space. These findings reveal that the further a house is located from open space, the smaller the price effect of moving closer to the open space. This implies that HPMs largely capture personal consumption values, and that aspects such as preservation for future generations are not as significant a factor.

Other papers have looked at more specific questions using hedonic pricing models, such as the value of tree cover (McPherson et al, 2011), the value of bodies of water (Kerstens et al, 2004) and the value of farmland (Johnston et al, 2001), all of which have found positive relationships between sale price and greater accessibility to the environmental asset.


As with all regressions, the first step is to collect data. In the case of a hedonic pricing model, this would be the selling price and location of residential properties in the area, as well as the details of the qualities of each house that would affect the selling price, including a number of property and neighbourhood characteristics. Included in this would be the accessibility and proximity to an environmental asset.

This data is then regressed against the house price, and the relationship between housing price and the key environmental attributes is defined.

Potential issues

  • Hedonic pricing will only capture benefits from environmental assets insofar as they affect housing prices.
  • The results depend heavily on the model specification.
  • Hedonic pricing is generally more suited to general questions as opposed to a more specific valuation. The value of parks on housing prices in general would be easier for the model to identify as opposed to the value of a particular park, as there would be less pertinent data.

Travel cost method

What is it?

The travel cost method (TCM) aims to value assets such as ecosystems or parks by inferring the demand and economic surplus for these assets through visitor travel costs. As travel costs and time increase with distance from the location, each zone of travel is treated as a different “price” at which visitors are willing to pay to visit the site.
Once calculated, this demand is compared to the cost of operating and maintaining the asset, and when combined, these two values provide the total economic surplus that the asset provides.
The travel cost method is most appropriate when the environmental asset is in itself a destination that attracts visitors, and so is best suited for valuation of assets such as national parks and/or wildlife reserves.


In most cases, the travel cost method has been used in the economic literature to value national parks or particular tourists attractions (Loomis et al, 2000; Twerefou and Adjei-Ababio, 2012). Ultimately, comparison across these different studies is fairly arbitrary, as each site will have its own particulars and context that will differ from each other site, and so it is difficult to say whether these values are high or low. However, a meta-analysis by Shrestha and Loomis (2003) of outdoor recreation over the past 30 years in the USA predicted an average consumer surplus of $47.10 USD per day per person. Other meta-analyses have also been undertaken for studies in other locations (Zanderson and Tol, 2009; Johnston et al, 2005).


The TCM uses survey data taken from a sample of visitors to the site. Firstly, zones are defined by their distance from the location. The exact number and definition of the zones is largely arbitrary, and more zones will result in greater differentiation, but will typically be more work and may require more data.

Visitors are then grouped into “zones”, determined by the distance they travelled to get to the location, and the total visits per zone is calculated. The visitation rate is then determined by dividing the total visits per zone by the zone’s population in thousands.

The average round trip travel costs is then calculated for each zone, assuming that people in the closest zone (Zone 0) have a zero cost of travel to get to the location. This will involve the average cost of travel (petrol/plane tickets) as well as the average cost of time (usually the average hourly wage). Any additional admission fees are also included in determining the average WTP of visitors.

These figures are then used (with a number of other variables) to produce a regression model, which can then be used to derive the demand curve for the asset. The area under this curve is thus the total estimate of the economic benefits of the asset.


Potential issues

  • The TCM assumes that the visitation or usage of the site is the primary reason for the individual’s trip. If there are other reasons that an individual has travelled to the location, that visitor’s travel cost will be overestimated.
  • The TCM does not capture non-use values of the asset, as it would only survey individuals who have come to visit the site. This could also (and would likely) introduce sampling bias to the results and is likely to underestimate the value of the asset.
  • There could be individuals who live near the site and thus have low travel costs to the location but nonetheless value the asset highly.


Contingent valuation method

What is it?

The contingent valuation method (CVM) involves directly asking individuals how much they would be willing to pay to use a particular asset or the amount of compensation they would require to give up the asset. The name of the method derives from the fact that individuals are asked for their willingness to pay contingent on a specific hypothetical scenario.
CVM is quite flexible in being able to value more or less anything, and also captures individual’s valuations for both use and non-use purposes.


While there are many questions regarding how accurately stated preferences translate into actual behaviour, CVM remains a widely used technique in the literature. Brander and Koetse (2011) performed a meta-analysis of 20 CVM studies, finding the value of open space in an area with ‘average characteristics’ (average hectare size, GDP per capita and population density) has a value of approximately $1,550 USD/ha/year.

They also found recreation was much more highly valued (322% higher value per hectare) than environmental/agricultural benefits, which contrasted with previous results found by Kline and Wilchelns (1998) and Kotchen and Powers (2006), who found evidence of stronger preference for agricultural land over other types of open space. Brander and Koetse (2011) also observed a significant and positive relationship between the value of open space and population density.

There have also been a number of papers that have found no significant income effect for the value of open space (Romero and Lieserio, 2002; Kline and Wilchens, 1994) although Brander and Koetse (2011) suggest this may be the case people prefer to consume private open space (e.g, private gardens) rather than public open space as their income increases.


More so than any other technique, the design of the survey for the CVM is the most important part of the method, as it deals with a high level of subjectivity and conjecture that arises from using non-observational data. A number of different aspects of the survey must be considered when designing the questions:

  • Who is the asset valuable/relevant to?
  • How large will the sample space be?
  • What method of survey distribution and collection will be used?

Further to this, the questions and survey must be tested and refined to ensure that all the relevant information is provided and there are no ambiguities in the phrasing of questions. This ensures the data is as accurate and representational as possible, although the reality is it is highly unlikely to remove all biases from the survey.

Potential issues

  • Of the methods used to evaluate non-use goods, CVM is one of the most controversial, as the valuation data is based on what respondents say they would do rather than observing their actions. For any number of reasons, individuals may overstate or understate their valuation of an asset and this may bias the result. This is a major criticism of CVM, and while proper survey design and effective complementing analysis can help minimize this issue, it is unlikely to remove its impact entirely.
  • CVM relies on responses to a hypothetical situation, and it can be reasonably argued that what people say they will do in a hypothetical situation will differ to when they are required to pay the amount.
  • The design of the survey and questions can heavily influence the results.

Other methods of valuation

While not as analytically rigorous as the above three methods, the values below should not be ignored, as they may provide complementary or supporting valuation of environmental assets.


Land value (simple)

What is it?

Land value looks at the prices of the land of the asset in question. This usually reflects the total commercial value of the land.

Potential issues

  • Land value primarily takes into account market-based and commercial benefits of the land, and will usually fail to incorporate all of its non-market and non-use benefits.

Investment or development value (in the case of a park)

What is it?

The development value of a park represents the potential commercial gains due to the increased tourism/jobs that could be realised if amenities and facilities such as houses, hotel and restaurants were built in the area.


The development value will typically be a function of the number of individuals that visit a location multiplied by the average amount of money a visitor would spend at the location.

Potential issues

  • Like land value, development value primarily deals with commercial benefits, and would ignore the potential damage that development could have on the non-market benefits of the asset, such as congestion, additional waste etc…

Urban heat island effects

What is it?

An urban heat island is a metropolitan area that is significantly warmer than its surrounding area due to human activities. This leads to increased rainfall in these areas, as well as a decrease in local air and water quality. Green space can be valuable for its contribution in mitigating the urban heat island effect.


The effect of green spaces in mitigating urban heat islands can be seen in measuring the air temperature around the green spaces, typically several years after their setup.

Potential issues

  • While the mitigation of urban heat islands is a positive aspect of green space, it is difficult to value the benefits of the impact of the green space, given how little consumers are likely to feel the impacts. Ultimately, it may be implicitly captured in other models, but it would be difficult to capture this value in its entirety.




Brander, L.M. and Koetse, M.J. (2011). The value of urban open space: meta-analyses of contingent valuation and hedonic pricing results, Journal of Environmental Management, 92: 2763-2773. 

Kotchen, M.J. and Powers, S.M. (2006), Explaining the appearance and success of voter referenda for open-space conservation, Journal of Environmental Economics and Management, 52(1): 373-390.

Smith, V.K., Poulos, C., Kim, H. (2002), Treating open space as an urban amenity, Resource and Energy Economics, 24: 107-129.

McPherson, E.G., Simpson, J.R., Xiao, Q., Wu, C. (2011) Million trees Los Angeles canopy cover and benefit assessment, Landscape and Urban Planning, 99(1): 40-50.

Kerstens, Y., Theriault, M., Des Rosiers, F. (2004) The impact of surrounding land use and vegetation on single-family house prices, Environment and Planning B, 31: 539-567.

Johnston, R.J., Opaluch, J.J., Grigalunas, T.A., Mazzotta, M.J., (2001), Estimating amenity benefits of coastal farmland, Growth and Change, 32: 302-325.

Johnston, R.J., Besedin, E.Y., Iovanna, R., Miller, C.J., Wardwell, R.F., Ranson, M.H. (2005) Systematic variation in willingness to pay for aquatic resource improvement and implications for benefit transfer: a meta-analysis, Canadian Journal of Agricultural Economics, 53: 221-248.

Carson, R.T., Flores, N.E., Meade, N.F. (2001), Contingent valuation: Controversies and evidence, Environmental and Resource Economics, 19(2): 173-210.

Kline, J., Wilchelns, D. (1998), Measuring heterogeneous preferences for preserving farmland and open space, Ecological Economics, 26(2): 211-224.

Kline, J., Wilchelns, D. (1994), Using referendum data to characterize public support for purchasing development rights to farmland, Land Economics, 70: 223-233.

Romero, F., Liserio, A. (2002), Saving open spaces: determinants of 1998 and 1999 antisprawl ballot measures, Social Science Quarterly, 83: 341-352.

Twerefou, D.K., Adjei-Ababio, D. (2012), An economic valuation of the Kakum National Park: An individual travel cost approach, African Journal of Environmental Science and Technology, 6(4): 199-207.

Loomis J., Yorizane, S., Larson, D. (2000), Testing significance of multi-destination and multi-purpose trip effects in a travel cost method demand model for whale watching trips, Agricultural and Resource Economics Review, 19(3):  183-191

Zandersen M., Tol, R.S (2009), A meta-analysis of forest recreation values in Europe, Journal of Forest Economics, 15: 109-130.

Shrestha, R.K., Loomis, J.B. (2003), Meta-analytic benefit transfer of outdoor recreation economic values: testing out-of-sample convergent validity, Environmental and Resource Economics, 25: 79-100.