Sunday, December 18, 2016

Planning for green infrastructure: The spatial effects of parks, forests, and fields on Helsinki's apartment prices

Highlights
• Spatial effects of forests, parks, and fields on apartment prices are estimated.
• Forests generate indirect benefits in the urban core and direct in the urban fringe.
• Parks generate direct and indirect benefits in the urban core.
• Fields generate direct benefits in the urban fringe and no indirect spillovers.
• Successful green interventions are location-, benefit-, and goal-sensitive.

Abstract
As the importance of urban green spaces is increasingly recognised, so does the need for their systematic placement in a broader array of socioeconomic objectives. From an urban planning and economics perspective, this represents a spatial task: if more land is allocated to various types of green, how do the economic effects propagate throughout urban space? This paper focuses on the spatial marginal effects of forests, parks, and fields and estimates spatial hedonic models on a sample of apartment transactions in Helsinki, Finland. The results indicate that the capitalization of urban green in apartment prices depends on the type of green, but also interacts with distance to the city centre. Additionally, the effects contain variable pure and spatial spillover impacts, also conditional on type and location, the separation of which highlights aspects not commonly accounted for. The planning of green infrastructure will therefore benefit from parameterizing interventions according to location, green type, and character of spatial impacts.
...
The full-sample estimation explained 78% of price variation and returned the expected signs for all hedonic coefficients, except for that of distance to a forest. An increase in the debt and maintenance costs and a decrease in the condition of the property decreases price/m2. Additional rooms have a negative effect, reflecting the diminishing marginal utility of additional units of space. Increase in the property's age decreases price until historical status becomes relevant and price increases again. The yearly dummy variables are significant, indicating a drop in the average level of selling price/m2 from 2000 to 2001, followed by an increase from 2002 onwards. Increased distance to the city centre and coastline decrease price, whereas lot size is not significantly different from zero. The coefficients of the proxies for noise and air pollution disamenities are significant; a 100-meter increase in distance to rails increases average m2 price by 0.15%, while the corresponding increase for over-ground metro line is 0.19% and for major road is 0.36%.
Kuva: Roy Koto/ Viherosasto

The estimation supported the assumption of a CBD gradient in the marginal effects of parks and fields. Increased distance to a park decreases prices in the city centre, or, conversely decreasing the distance of a downtown property to a park increases its price, with the effect gradually declining as distance to the CBD increases. The maximum effect is estimated to a decrease of 1.5% in the m2 price when distance to a park increases 100 m, which is in the same range to the effect of recreational forests in the study of Tyrväinen (1997) that reports a corresponding increase of 0.5% (after currency conversion and average price normalization).... Increased distance to fields decreases price in the urban fringe, or conversely, decreasing the distance of a suburban property to fields increases its price. The maximum effect along this gradient is a decrease of 1.1% in m2 price when distance to a field increases by 100 m.

The regression is problematic in understanding the effect of forests. It indicates that increased distance to a forest increases price throughout the city with no statistically significant CBD gradient. Interestingly, a similar result is reported by Tyrväinen (1997) for the effect of distance to forest parks, who attributed it to non-fulfilment of the conditions for capitalization (Starret, 1981) and to dweller preferences on the specific tree type in forest parks. Additionally, Tyrväinen (1997) and Tyrväinen and Miettinen (2000) note that samples that are aggregated from years with varying macroeconomic conditions may pose estimation problems. Table 3 indicates that the present sample does have such variations as indicated by the somewhat sharp fluctuations in regional unemployment rates.

The regression is problematic in understanding the effect of forests. It indicates that increased distance to a forest increases price throughout the city with no statistically significant CBD gradient. Interestingly, a similar result is reported by Tyrväinen (1997) for the effect of distance to forest parks, who attributed it to non-fulfilment of the conditions for capitalization (Starret, 1981) and to dweller preferences on the specific tree type in forest parks. Additionally, Tyrväinen (1997) and Tyrväinen and Miettinen (2000) note that samples that are aggregated from years with varying macroeconomic conditions may pose estimation problems. Table 3 indicates that the present sample does have such variations as indicated by the somewhat sharp fluctuations in regional unemployment rates.
...
The maximum effect of distance to a forest or park is at the urban core, while that of distance to a field is in the urban fringe. On a multiyear average, the effect of a 100 m increase of distance to a forest is a decrease of 3.7% in price/m2 at 0 km from the CBD, which gradually drops to zero at 6 km from the CBD. The maximum effect is close to that reported by Tyrväinen and Miettinen (2000), which corresponds to a 5.3% decrease in price/m2 for a 100 m increase in distance to a forested area for the average floorspace of 90 m2 of their sample. The difference in estimates may be attributed to the fact that the valuation of Tyrväinen and Miettinen (2000) was conducted on a sample of terraced apartments as opposed to block apartments in this study. Terraced houses in Finnish housing markets have higher m2 price than block apartments and are typically associated with wealthier households; it is assumed that the difference between the two studies relates to the higher WTP of wealthier households for green amenities. The maximum effect of distance to a park is estimated to 1.8% at the CBD, gradually dropping to zero at approximately 8 km from the CBD. As in the full-sample regression, the slope of the gradient of distance to a field is reversed; the maximum effect is 0.8% in the urban fringe (indicatively at 15 km from the CBD) and gradually drops to zero at approximately 8 km from the CBD. The difference between these estimates and the estimates of the full-sample regression is small (0.3% for parks and fields), except of the notable difference in the forest effects.

The estimated gradients show that at approximately 6 to 8 km from the CBD, the marginal value of forests and parks diminishes to zero, while that of fields rises from zero. The estimations return negative effects in areas further than 6–8 km from the CBD for forests and parks, and in areas closer than 8 km to the CBD for fields. This is due to the assumed unbounded linear form of the gradient; it is therefore interpreted not as an actual discount, but as zero benefit. The maps in Fig. 3 (reproduced in color in the article's electronic version and in greyscale in its paper version) display the multiyear mean gradients (black lines in Fig. 2) as surfaces over Helsinki's urban morphology and also indicate the spatial distribution of Helsinki's GI and densities of residential building stock.
...
The maximum direct impact of a 100 m increase of distance to a forest is a decrease in m2 price by 1% at the urban fringe, gradually dropping to zero at approximately 9 km from the CBD; the maximum indirect impact is reverse with approximately 3.4% at the CBD, gradually dropping to zero at 4 km from the CBD; and the maximum total impact is 2% at the CBD, gradually dropping to zero at 3 km from the CBD. Concerning the effects of a 100 m increase of distance to a park, the maximum direct impact is 0.1% at the CBD, gradually dropping to zero at 3 km from the CBD; the maximum indirect impact is 2% at the CBD, dropping to zero at 10 km from the CBD; and the maximum total impact is 2.2% at the CBD, dropping to zero at 9 km from the CBD. The maximum direct impact of a 100 m increase in distance to a field is 2.5% at the urban fringe, gradually dropping to zero at 3 km from the CBD; the maximum total impact is 0.7% at the CBD, declining to zero at 8 km from the CBD; indirect impacts are negative and assumed as zero-benefit.
...
The above figures indicate a few important differences in the spatial character of the marginal price effects of distance to forests, parks, and fields. Given the separation of pure and spillover effects, it is reasonable to suggest that decreased distances to all three green types capitalize positively in Helsinki's apartment prices, but only at the correct locations within the urban area and with a specific spatial impact character in mind. In particular, fields capitalize exclusively in the urban fringe and the effects concern exclusively changes at a certain property; spatial spillover of the price effects to/from neighbouring properties is zero and it takes a city-wide change (total impacts) to observe more widespread price changes. In contrast, parks capitalize exclusively at the city centre; the price effects are small at the concerned property and mostly spill over to (and from) neighbours. The capitalization of forests is double-natured as also found in Tyrväinen (1997); they capitalize at the concerned property only in the urban fringe, while the price effects in the urban core are spillovers to and from the neighbourhood.
...
by Athanasios Votsis, Finnish Meteorological Institute, Research Group for the Socioeconomic Impacts of Climate and Weather, Erik Palménin aukio 1, P.O. Box 503, FI 00101 Helsinki, Finland University of Helsinki, Faculty of Science, Department of Geosciences and Geography, Helsinki, Finland
Ecological Economics via Elsevier Science Direct www.ScienceDirect.com
Volume 132; February, 2017, Pages 279–289; Available online 18 November 2016
Under a Creative Commons license, Open Access
Keywords: Green infrastructure; Land use options; Spatial effects; Housing prices

No comments:

Post a Comment