Natural Hazards ( IF 3.7 ) Pub Date : 2021-01-04 , DOI: 10.1007/s11069-020-04484-w Steven J. Dundas , Roger H. von Haefen
Credible empirical estimation of the economic impacts of climate change is dependent on data structure (e.g., cross sectional, panel) and the functional relationship between weather data and behavioral outcomes. We show here how these modeling decisions lead to significantly different results when estimating the effects of weather and simulating the potential welfare impacts of climate change on outdoor recreation. Using participation data from 1.6 million households in the United States from 2004 to 2009, we estimate the impact of temperature and precipitation on participation decisions for marine shoreline recreational fishing. Results from linear models suggest temperature positively impacts participation and, by implication, climate change is likely to improve welfare associated with outdoor recreation in all regions of our study area. Conversely, nonlinear specifications suggest more days with extreme heat reduce participation and lead to significant declines in welfare under future climate scenarios. Differences in the treatment of how weather enters recreation participation decisions change both the sign and magnitude of welfare effects by nearly $1 billion annually. Differences in data structure, however, only affect the magnitude of welfare impacts but not the sign. Disaggregation of welfare estimates suggests warmer baseline climates are more susceptible to these choices. Our results demonstrate the critical nature of modeling decisions about data structure and the use of weather data to assess the future impacts of climate change, especially with nonmarket goods where value is related to environmental quality such as outdoor recreation.
中文翻译:
数据结构和非线性在评估气候对户外休闲的影响中的重要性
对气候变化的经济影响进行可靠的经验估计取决于数据结构(例如横截面,面板)以及天气数据与行为结果之间的功能关系。我们在这里展示了这些建模决策如何在估计天气影响并模拟气候变化对户外休闲的潜在福利影响时导致显着不同的结果。利用2004年至2009年美国160万个家庭的参与数据,我们估算了温度和降水对海洋海岸线休闲捕鱼参与决策的影响。线性模型的结果表明温度对参与者的参与有积极影响,并且暗示着,气候变化可能会改善我们研究区域所有地区与户外休闲相关的福利。相反,非线性规范表明,在极端高温下,更多的日子会减少人们的参与并导致未来气候情景下福利的显着下降。天气如何进入娱乐参与决策的差异每年会改变福利影响的迹象和幅度,将近10亿美元。但是,数据结构的差异只会影响福利影响的大小,而不会影响征兆。对福利估计数的分解表明,基线气候变暖更容易受到这些选择的影响。我们的结果证明了对数据结构进行建模决策的关键性质,以及使用天气数据评估气候变化的未来影响,特别是对于那些与环境质量相关的价值的非市场商品,例如户外休闲。非线性规范表明,在未来的气候情景下,高温持续多天会减少参与,并导致福利显着下降。天气如何进入娱乐参与决策的差异每年会改变福利影响的迹象和幅度,将近10亿美元。但是,数据结构的差异只会影响福利影响的程度,而不会影响征兆。对福利估计数的分类显示,基线气候变暖更容易受到这些选择的影响。我们的结果证明了对数据结构进行建模决策的关键性质,以及使用天气数据评估气候变化的未来影响,特别是对于那些与环境质量相关的价值的非市场商品,例如户外休闲。非线性规范表明,在未来的气候情景下,高温持续多天会减少参与,并导致福利显着下降。天气如何进入娱乐参与决策的差异每年会改变福利影响的迹象和幅度,将近10亿美元。但是,数据结构的差异只会影响福利影响的程度,而不会影响征兆。对福利估计数的分类显示,基线气候变暖更容易受到这些选择的影响。我们的结果证明了对数据结构进行建模决策的关键性质,以及使用天气数据评估气候变化的未来影响,特别是对于那些与环境质量相关的价值的非市场商品,例如户外休闲。