Valuing recreational water clarity and quality: evidence from hedonic pricing models of lakeshore properties

ABSTRACT

This study extends the economics literature on the relationship between recreational water quality and residential property pricing by separately testing the relationships between both water clarity and quality and the price of lakeshore residential properties in Ontario, Canada. Our econometric approach includes objective measures of both attributes: Secchi disk readings for water clarity and phosphorus concentrations for water quality. Econometric results presented in this study suggest that a one-metre improvement in water clarity results in an 8 per cent residential property price premium, while a one standard deviation improvement in water quality boosts lakeshore residence prices by 32 per cent.

KEYWORDS:

• Water clarity
• water quality
• hedonic pricing models
• real estate markets
• waterfront properties

JEL CLASSIFICATION:

• Q25
• Q51
• Q53
• R30
• R31

Acknowledgements

The authors thank John Fincham of REMAX Parry Sound-Muskoka Realty Ltd. Brokerage in Magnetawan, Ontario, for providing much of the data used in the study. The usual caveat applies.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Guignet et al. (2017) discuss a scenario wherein submerged aquatic vegetation provides ecosystem functions that are viewed by bay-shore residents as beneficial. Specifically, they find that waterfront and near-waterfront homes around the Chesapeake Bay sell at a premium of 6.5 per cent when such vegetation is present. Relatedly, Walsh et al. (2017) find that EPA-mandated water quality improvements on the Chesapeake Bay also yield significant benefits to waterfront and near-waterfront homeowners.

² This is the measure used by tax assessment authorities in Canada, which, as pointed out below, is the source of our data.

³ The associated t-ratios are −2.61 and 2.54, respectively. This regression also produces an R² of 0.054, which is almost seven times that of 0.008 from a simple regression including only PHOS, which is insignificant.

⁴ The t-ratio in this case is 2.01. This simple regression also produces an R² of 0.054, which compares favourably to that of 0.057 from a regression including both SECCHI and SECCHI².

⁵ The parameter estimates for PHOS and PHOS2 are −274.7 and 11.09, respectively. These are significant at the .05 and .03 levels, respectively. That for SECCHI is 102.2, and it is significant at only the .32 level.