Deviation Analysis on Willingness and Behavior of Residents' Earthquake Insurance Purchasing−Based on Logit Model
-
摘要: 为真实了解和深入研究居民地震保险购买意愿与行为的背离现象及其形成机制,以全国不同省份居民作为研究对象,开展专项问卷调查,并采用Logit模型进行回归分析,提出提升居民购买地震保险意愿并付诸行动的相关建议。研究结果表明,居民对地震断裂带的判断属于地震风险感知变量,对投保意愿和购买行为均产生显著的正向影响;是否购买其他保险和是否通过网络捐款属于普通风险感知变量,对投保意愿产生显著的正向影响,对购买行为产生正向影响但不显著;地震风险感知变量既可增强居民的投保意愿,又可促使居民付诸实际购买行动;普通风险感知变量虽能增强居民的投保意愿,但对购买行为无效力,使投保意愿和购买行为表现出一定程度的背离。Abstract: In order to understand and study the deviation of residents' willingness and behavior to purchase earthquake insurance and its formation mechanism, a special questionnaire survey was conducted with residents of different provinces across China as the research target, and a regression analysis was conducted using a Logit model to propose recommendations to improve residents' willingness to purchase earthquake insurance and put them into action. The results of the study show that residents' judgment of earthquake rupture zones is an earthquake risk perception variable, which has a significant positive impact on both willingness to insure and purchase behavior; whether or not to purchase other insurance and whether or not to donate through the Internet are general risk perception variables, which have a significant positive impact on willingness to insure and a positive but insignificant impact on purchase behavior; earthquake risk perception variables can both enhance residents' willingness to insure and prompt them to take actual purchase action; general risk perception variables can enhance residents' willingness to insure but have no effect on purchase behavior, causing a certain degree of divergence between willingness to insure and purchase behavior.
-
Key words:
- Earthquake insurance /
- Willingness expression /
- Purchase behavior /
- Regression analysis /
- Logit model
-
表 1 我国地震危险性区划与样本个数分布
Table 1. Seismic hazard zoning in China and sample number distribution
地震危险性区划 省份(直辖市、自治区) 调查人数/人 所占比例/% Ⅰ级区 北京、天津、云南、四川、西藏 775 45.35 Ⅱ级区 河北、山西、甘肃、青海、宁夏、新疆 465 27.21 Ⅲ级区 内蒙古、辽宁、吉林、上海、江苏、安徽、福建、
山东、河南、广东、海南、重庆、陕西330 19.31 Ⅳ级区 黑龙江、浙江、江西、湖北、湖南、广西、贵州 139 8.13 表 2 变量定义及赋值
Table 2. Variable definition and assignment
变量 变量定义及描述 因变量 投保意愿(Yint) 愿意投保=2,考虑中=1,不愿意投保=0 购买行为(Ybuy) 已购买地震保险=1,未购买地震保险=0 自变量 个体特征 性别(Sex) 男=1,女=0 年龄(Age) <18岁=1,18~25岁=2,26~30岁=3,31~40岁=4,41~50岁=5,51~65岁=6,≥65岁=7 教育程度(Edu) 小学及以下=1,初中=2,高中=3,大学专科=4,大学本科=5,硕士研究生=6,博士研究生=7 家庭状况 家庭年收入(Inc) <3万元=1,3~5万元=2,5~10万元=3,10~15万元=4,
15~20万元=5,20~50万元=6,≥50万元=7城乡聚落(Loc) 城市=1,农村=0 当前居住房屋价值(Pri) <5万元=1,5~10万元=2,10~20万元=3,20~50万元=4,50~100万元=5,
100~200万元=6,200~500万元=7,≥500万元=8风险感知 是否处在地震断裂带(Fauz) 是=1,否=0 是否购买其他保险(Ins) 购买=1,未购买=0 是否通过网络捐款(Don) 没有=0,有过1、2次=1,有过很多次=2 地震环境 地震危险性区划(Edd) Ⅰ级区=4,Ⅱ级区=3,Ⅲ级区=2,Ⅳ级区=1 注:年龄、家庭年收入与当前居住房屋价值3个变量的取值区间含下限,不含上限。 表 3 描述性统计
Table 3. Descriptive statistics
变量名称 均值 标准差 最小值 中位数 最大值 Yint 0.82 0.74 0 1 2 Ybuy 0.05 0.23 0 0 1 Sex 0.38 0.49 0 0 1 Age 3.23 1.35 1 3 7 Edu 4.05 1.19 1 4 7 Inc 3.37 1.54 1 3 7 Loc 0.61 0.49 0 1 1 Pri 3.95 1.74 1 4 8 Edd 3.10 0.98 1 3 4 Fauz 0.38 0.49 0 0 1 Ins 0.54 0.50 0 1 1 Don 1.18 0.71 0 1 2 表 4 整体Logit估计结果
Table 4. Overall Logit estimation results
变量名称 Yint Ybuy 模型1 模型2 模型3 模型4 模型5 模型6 Sex 0.207**
(2.183)0.182* 0.016* 0.525** 0.537** 0.021** (1.868) (1.824) (2.386) (2.416) (2.379) Age 0.079** 0.065* 0.006* −0.162* −0.197** −0.008** (2.327) (1.828) (1.737) (−1.854) (−2.133) (−2.198) Edu 0.015 −0.024 −0.002 0.058 0.030 0.001 (0.379) (−0.589) (−0.559) (0.610) (0.307) (0.282) Inc 0.215*** 0.122*** 0.011*** 0.404*** 0.346*** 0.014*** (5.583) (3.035) (2.893) (4.533) (3.761) (3.452) Loc 0.237** 0.234** 0.020** 0.062 0.032 0.001 (2.114) (2.034) (1.947) (0.227) (0.115) (0.107) Pri −0.064* −0.082** −0.007** −0.183** −0.181** −0.007** (−1.698) (−2.078) (−1.946) (−2.077) (−1.971) (−2.067) Edd — 0.396*** 0.035*** — 0.068 0.003 — (7.414) (5.587) — (0.556) (0.488) Fauz — 0.715*** 0.062*** — 0.740*** 0.029*** — (6.851) (5.902) — (3.107) (2.834) Ins — 0.572*** 0.050*** — 0.288 0.011 — (5.727) (4.849) — (1.179) (1.159) Don — 0.166** 0.014** — 0.221 0.009 — (2.448) (2.368) — (1.339) (1.528) 常数项 — — — −3.632*** −4.214*** — — — — (−6.832) (−6.735) — N(样本量) 1 709 1 709 1 709 1 709 1 709 1 709 决定系数R2 0.019 0.077 0.077 0.048 0.074 0.074 注:* 、**、***分别代表10%、5%、1%的置信水平,括号内数值为Z统计量。 表 5 稳健性检验
Table 5. Robustness test
变量名称 Yint Ybuy 最小二乘回归模型 Probit模型 Logit模型
最小二乘回归模型
Probit模型 Logit模型 Sex 0.065* 0.107* 0.153 0.027** 0.258** 0.539** (1.862) (1.840) (1.532) (2.211) (2.404) (2.351) Age 0.023* 0.036* 0.087** −0.008** −0.087** −0.190** (1.757) (1.694) (2.366) (−2.254) (−2.052) (−2.068) Edu −0.007 −0.012 −0.010 0.002 0.015 0.037 (−0.480) (−0.476) (−0.245) (0.308) (0.325) (0.345) Inc 0.048*** 0.076*** 0.116*** 0.018*** 0.169*** 0.347*** (3.348) (3.213) (2.843) (3.279) (3.639) (3.434) loc 0.075* 0.131* 0.254** 0.001 0.023 0.032 (1.836) (1.895) (2.151) (0.098) (0.169) (0.107) Pri −0.031** −0.051** −0.083** −0.009** −0.094** −0.180** (−2.198) (−2.120) (−1.989) (−2.139) (−2.283) (−2.100) Edd 0.140*** 0.233*** — 0.002 0.030 — (7.339) (7.049) — (0.363) (0.474) — Fauz 0.274*** 0.433*** 0.708*** 0.038*** 0.369*** 0.784*** (7.015) (6.887) (6.467) (2.787) (3.166) (2.928) Ins 0.198*** 0.331*** 0.549*** 0.012 0.141 0.288 (5.552) (5.551) (5.436) (1.071) (1.240) (1.159) Don 0.060** 0.100** 0.189*** 0.011* 0.115* 0.223 (2.484) (2.478) (2.761) (1.646) (1.653) (1.545) Int — — 0.552*** — — 0.000 — — (6.609) — — (0.001) 常数项 −0.052 — — −0.003 −2.260*** −4.074*** (−0.576) — — (−0.102) (−7.406) (−5.485) N(样本量) 1 709 1 709 1 709 1 709 1 709 1 709 注:* 、**、***分别代表10%、5%、1%的置信水平,括号内数值为Z统计量。 -
丁元昊, 2012. 巨灾保险需求研究−基于需求行为的实证. 成都: 西南财经大学.Ding Y. H., 2012. The demand for catastrophe insurance−behavioral evidence. Chengdu: Southwestern University of Finance and Economics. (in Chinese) 李姜, 张合, 刘志辉, 2021. 张家口地区精细化地震灾害风险评估. 震灾防御技术, 16(1): 134—145 doi: 10.11899/zzfy20210114Li J. , Zhang H. , Liu Z. H. , 2021. Refined risk assessment of earthquake disaster in Zhangjiakou Area. Technology for Earthquake Disaster Prevention, 16(1): 134—145. (in Chinese) doi: 10.11899/zzfy20210114 李俏, 肖遥, 底欣欣等, 2020. 辽宁省地震易发区地震危险性研究. 震灾防御技术, 15(2): 392—401Li Q. , Xiao Y. , Di X. X. , et al. , 2020. Study on seismic hazard in earthquake-prone areas of Liaoning Province. Technology for Earthquake Disaster Prevention, 15(2): 392—401. (in Chinese) 李文娟, 2009. 巨灾保险需求及其影响因素研究. 武汉: 武汉大学.Li W. J., 2009. The demand for catastrophe insurance and its influential factors. Wuhan: Wuhan University. (in Chinese) 刘沐泽, 2015. 我国地震指数保险定价研究及模式选择. 武汉: 武汉大学.Liu M. Z., 2015. Earthquake index insurance pricing and mode selection. Wuhan: Wuhan University. (in Chinese) 尚燕, 熊涛, 2020. 所为非所想?农户风险管理意愿与行为的悖离分析. 华中农业大学学报(社会科学版), (5): 19—28, 169Shang Y. , Xiong T. , 2020. The Paradox between farmers’ willingness and adoption of risk management. Journal of Huazhong Agricultural University (Social Sciences Edition), (5): 19—28, 169. (in Chinese) 田玲, 姚鹏, 王含冰, 2015. 政府行为、风险感知与巨灾保险需求的关联性研究. 中国软科学, (9): 70—81 doi: 10.3969/j.issn.1002-9753.2015.09.007Tian L. , Yao P. , Wang H. B. , 2015. The relationship among government behavior, risk perception and catastrophe insurance demand. China Soft Science, (9): 70—81. (in Chinese) doi: 10.3969/j.issn.1002-9753.2015.09.007 谢佳秋, 谢晓非, 甘怡群, 2011. 汶川地震中的心理台风眼效应. 北京大学学报(自然科学版), 47(5): 944—952Xie J. Q. , Xie X. F. , Gan Y. Q. , 2011. Psychological typhoon eye effect in the Wenchuan earthquake. Acta Scientiarum Naturalium Universitatis Pekinensis, 47(5): 944—952. (in Chinese) 于泳, 李晨阳, 2021-10-13(007). 发挥好巨灾保险“缓冲垫”作用. 经济日报. 张宗军, 2008. 基于公共性基础上的巨灾保险制度研究. 保险研究, (7): 15—17 doi: 10.13497/j.cnki.is.2008.07.006Zhang Z. J. , 2008. A research on the catastrophe insurance system from the aspect of its public nature. Insurance Studies, (7): 15—17. (in Chinese) doi: 10.13497/j.cnki.is.2008.07.006 郑祎, 伍国春, 2020. 影响公众地震应急准备因素分析. 震灾防御技术, 15(3): 591—600 doi: 10.11899/zzfy20200313Zheng Y. , Wu G. C. , 2020. Analysis of factors affecting the Chinese public earthquake emergency preparedness. Technology for Earthquake Disaster Prevention, 15(3): 591—600. (in Chinese) doi: 10.11899/zzfy20200313 中华人民共和国国家质量监督检验检疫总局, 中国国家标准化管理委员会, 2016. GB 18306—2015 中国地震动参数区划图. 北京: 中国标准出版社.General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China, Standardization Administration of the People's Republic of China, 2016. GB 18306—2015 Seismic ground motion parameters zonation map of China. Beijing: Standards Press of China. (in Chinese) 祝伟, 陈秉正, 2015. 我国居民巨灾保险需求影响因素分析——以地震风险为例. 保险研究, (2): 14—23 doi: 10.13497/j.cnki.is.2015.02.002Zhu W. , Chen B. Z. , 2015. The identification of factors affecting residents’ catastrophe insurance demands − the evidence from earthquake risks in China. Insurance Studies, (2): 14—23. (in Chinese) doi: 10.13497/j.cnki.is.2015.02.002 卓志, 周志刚, 2013. 巨灾冲击、风险感知与保险需求——基于汶川地震的研究. 保险研究, (12): 74—86 doi: 10.13497/j.cnki.is.2013.12.011Zhuo Z. , Zhou Z. G. , 2013. Catastrophic shock, risk perception and demand of insurance. Insurance Studies, (12): 74—86. (in Chinese) doi: 10.13497/j.cnki.is.2013.12.011 Ajzen I. , 1991. The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2): 179—211. doi: 10.1016/0749-5978(91)90020-T Athavale M. , Avila S. M. , 2011. An analysis of the demand for earthquake insurance. Risk Management and Insurance Review, 14(2): 233—246. doi: 10.1111/j.1540-6296.2011.01205.x Browne M. J. , Hoyt R. E. , 2000. The demand for flood insurance: empirical evidence. Journal of Risk and Uncertainty, 20(3): 291—306. doi: 10.1023/A:1007823631497 Burrell O. K. , 1951. Possibility of an experimental approach to investment studies. The Journal of Finance, 6(2): 211—219. Grace M. F., Klein R. W., Kleindorfer P. R., et al., 2003. Catastrophe insurance: consumer demand, markets and regulation. New York: Springer. Johnson E. J. , Hershey J. , Meszaros J. , et al. , 1993. Framing, probability distortions, and insurance decisions. Journal of Risk and Uncertainty, 7(1): 35—51. doi: 10.1007/BF01065313 Lai L. H. , Hsieh H. Y. , 2007. Assessing the demand factors for residential earthquake insurance in Taiwan: empirical evidence on spatial econometrics. Contemporary Management Research, 3(4): 347—358. Palm R. , Hodgson M. E. , 1992. Earthquake insurance: mandated disclosure and homeowner response in California. Annals of the Association of American Geographers, 82(2): 207—222. doi: 10.1111/j.1467-8306.1992.tb01905.x Slovic P., Fischhoff B., Lichtenstein S., 1977. Cognitive processes and societal risk taking. In: Jungermann H., Zeeuw G., eds., Decision Making and Change in Human Affairs. Dordrecht: Springer, 7—36.