失业率波动对就业者家庭消费和配偶劳动供给的冲击
|
摘要
基于中国城镇住户调查2010—2012年月度数据,本文实证分析了失业率波动对就业者家庭消费和成员劳动供给的影响。研究发现:(1)群组失业率每提高1%将导致就业者家庭消费增速降低0.48%,失业预期引起的预防性储蓄动机明显。(2)所有制异质性方面,群组失业率每提高1%,国有、集体和股份制就业者所受影响在统计意义上并不显著,个体或私营企业、没有固定性职业的就业者家庭消费增速分别降低0.50%和1.22%。(3)年龄异质性方面,临近退休的户主对群组失业率并不敏感。位于26—35岁和36—50岁的户主,群组失业率每提高1%,家庭消费增速分别降低0.49%和0.35%。(4)群组失业率每提高1%,将引起户主配偶劳动供给概率上升8.13%,与此同时,单身家庭消费增速显著下降0.69%,说明婚姻制度发挥着自我保险功能。以上结论对于评判中国失业保险制度,理解居民部门消费制约因素,以及劳动力市场结构调整具有现实意义。
Based on China's monthly urban household survey(UHS) from 2010 to 2012, this paper empirically tests the effect of unemployment rate fluctuation on employees' consumption and spousal labor supply.The results show that:(1) 1% increase in the grouped unemployment rate would lead to 0.48% reduction in households' consumption growth. That is, unemployment expectation will cause significant pre-cautionary saving.(2)In terms of the heterogeneity of ownership structure, 1% increase in the grouped unemployment rate would not statistically significantly affect the consumption growth for those who work in state-owned, collectively-owned or joint-stock firms, whereas there will be 0.5% and 1.22% reduction in consumption growth for those who have no fixed jobs or those who work in private enterprises.(3) In terms of the heterogeneity in ages, the consumption growth for those who are near to retirement is not sensitive to the grouped unemployment rate. For those aged between 26 and 35(or aged between 36 and 50), 1% increase in the grouped unemployment rate would respectively lead to the 0.49%(0.35%) reduction in consumption growth.(4) 1% increase in the grouped unemployment rate would lead to the 8.13% probability increase in terms of spousal labor supply. Meanwhile, the single-families' consumptions growth rate will decrease statistically significantly by 0.69% indicating that the getting marriage could provide some sorts of self-insurance. This paper has far reaching implications to the assessment of China's unemployment insurance, interpretation of the constraining factors for China's low consumption rate as well as the adjustment of China's labor market structure during business cycles.
Based on China's monthly urban household survey(UHS) from 2010 to 2012, this paper empirically tests the effect of unemployment rate fluctuation on employees' consumption and spousal labor supply.The results show that:(1) 1% increase in the grouped unemployment rate would lead to 0.48% reduction in households' consumption growth. That is, unemployment expectation will cause significant pre-cautionary saving.(2)In terms of the heterogeneity of ownership structure, 1% increase in the grouped unemployment rate would not statistically significantly affect the consumption growth for those who work in state-owned, collectively-owned or joint-stock firms, whereas there will be 0.5% and 1.22% reduction in consumption growth for those who have no fixed jobs or those who work in private enterprises.(3) In terms of the heterogeneity in ages, the consumption growth for those who are near to retirement is not sensitive to the grouped unemployment rate. For those aged between 26 and 35(or aged between 36 and 50), 1% increase in the grouped unemployment rate would respectively lead to the 0.49%(0.35%) reduction in consumption growth.(4) 1% increase in the grouped unemployment rate would lead to the 8.13% probability increase in terms of spousal labor supply. Meanwhile, the single-families' consumptions growth rate will decrease statistically significantly by 0.69% indicating that the getting marriage could provide some sorts of self-insurance. This paper has far reaching implications to the assessment of China's unemployment insurance, interpretation of the constraining factors for China's low consumption rate as well as the adjustment of China's labor market structure during business cycles.
引文
[1]白重恩,李宏斌,吴斌珍.医疗保险与消费:来自新型农村合作医疗的证据[J].经济研究,2012a,(2):41-53.
[2]白重恩,吴斌珍,金烨.中国养老保险缴费对消费和储蓄的影响[J].中国社会科学,2012b,(8):48-71.
[3]蔡昉.农民工就业面临各种风险[J].农村工作通讯,2013,(9):48.
[4]陈斌开,陆铭,钟宁桦.户籍制约下的居民消费[J].经济研究,2010,(S1):62-71.
[5]陈斌开,杨汝岱.土地供给、住房价格与中国城镇居民储蓄[J].经济研究,2013,(1):110-122.
[6]程令国,张晔.早年的饥荒经历影响了人们的储蓄行为吗?---对我国居民高储蓄率的一个新解释[J].经济研究,2010,(8):119-132.
[7]樊潇彦,袁志刚,万广华.收入风险对居民耐用品消费的影响[J].经济研究,2007,(4):124-136.
[8]黄少安,孙涛.非正规制度、消费模式和代际交叠模型---东方文化信念中居民消费特征的理论分析[J].经济研究,2005,(4):57-65.
[9]李宏彬,施新政,吴斌珍.中国居民退休前后的消费行为研究[J].经济学(季刊),2015,(1):117-134.
[10]卢锋,刘晓光,姜志霄,张杰平.劳动力市场与中国宏观经济周期:兼谈奥肯定律在中国[J].中国社会科学,2015,(12):69-89.
[11]吕丹,曲展.典型国家失业保险制度[J].中国劳动,2014,(10):29-33.
[12]温兴祥.失业、失业风险与农民工家庭消费[J].南开经济研究,2015,(6):110-128.
[13]杨继生,徐娟.环境收益分配的不公平性及其转移机制[J].经济研究,2016,(1):155-167.
[14]杨继生,阳建辉.行政垄断、政治庇佑与国有企业的超额成本[J].经济研究,2015,(4):50-61.
[15]张军,赵达,周龙飞.最低工资标准提高对就业正规化的影响[J].中国工业经济,2017,(1):81-97.
[16]赵达,谭之博,张军.中国城镇地区消费不平等演变趋势---新视角与新证据[J].财贸经济,2017,(6):115-129.
[17]赵达,周龙飞.非线性定价有效吗[J].统计研究,2018,(8):58-68.
[18]邹红,喻开志.退休与城镇家庭消费:基于断点回归设计的经验证据[J].经济研究,2015,(1):124-139.
[19]Aguiar,M.,and E.Hurst.Consumption versus Expenditure[J].Journal of Political Economy,2005,113(5):919-948.
[20]Aguiar,M.,and E.Hurst.Life-cycle Prices and Production[J].American Economic Review,2007,97(5):1533-1559.
[21]Aguiar,M.,E.Hurst,and L.Karabarbounis.Time Use during the Great Recession[J].American Economic Review,2013,103(5):1664-1696.
[22]Aladangady,A.Housing Wealth and Consumption:Evidence from Geographically-linked Microdata[J].American Economic Review,2017,107(11):3415-3446.
[23]Apergis,N.,and Y.Georgellis.Regional Unemployment and Employee Loyalty:Evidence from 12 UK Regions[J].Regional Studies,2018,52(9):1283-1293.
[24]Bai,J.Panel Data Models with Interactive Fixed Effects[J].Econometrica,2009,77(4):1229-1279.
[25]Bartik,T.J.Who Benefits from State and Local Economic Development Policies[M].Kalamazoo,MI:Upjohn Institute,1991.
[26]Battistin,E.,A.Brugiavini,E.Rettore,and G.Weber.The Retirement Consumption Puzzle:Evidence from a Regression Discontinuity Approach[J].American Economic Review,2009,99(5):2209-2226.
[27]Blundell,R.,L.Pistaferri,and I.Preston.Consumption Inequality and Partial Insurance[J].American Economic Review,2008,98(5):1887-1921.
[28]Blundell,R.,L.Pistaferri,and I.Saporta-Eksten.Consumption Inequality and Family Labor Supply[J].American Economic Review,2016,106(2):387-435.
[29]Browning,M.,and T.F.Crossley.Shocks,Stocks,and Socks:Smoothing Consumption over a Temporary Income Loss[J].Journal of the European Economic Association,2009,7(6):1169-1192.
[30]Cameron,A.C.,J.B.Gelbach,and D.L.Miller.Bootstrap-based Improvements for Inference with Clustered Errors[J].Review of Economics and Statistics,2008,90(3):414-427.
[31]Campos,R.G.,and I.Reggio.Consumption in the Shadow of Unemployment[J].European Economic Review,2015,(78):39-54.
[32]Chodorow-Reich,G.,J.Coglianese,and L.Karabarbounis.The Macro Effects of Unemployment Benefit Extensions:A Measurement Error Approach[J].Quarterly Journal of Economics,2019,134(1):227-229.
[33]Diamond R.The Determinants and Welfare Implications of U.S.Workers’Diverging Location Choices by Skill:1980-2000[J].American Economic Review,2016,106(3):479-524.
[34]Feng,S.,Y.Hu,and R.Moffitt.Long Run Trends in Unemployment and Labor Force Participation in Urban China[J].Journal of Comparative Economics,2017,45(2):304-324.
[35]Gan,L.,M.D.Hurd,and D.L.McFadden.Analyses in the Economics of Aging[M].Chicago:University of Chicago Press,2005.
[36]Giles,J.,P.Albert,and J.Zhang.What is China’s True Unemployment Rate[J].China Economic Review,2005,16(2):149-170.
[37]He,H.,F.Huang,Z.Liu,and D.Zhu.Breaking the Iron Rice Bowl:Evidence of Precautionary Savings from Chinese State-owned Enterprises Reform[J].Journal of Monetary Economics,2018,(94):94-113.
[38]Helliwell,J.F.,and H.Huang.New Measures of the Costs of Unemployment:Evidence from the Subjective Well-being of 3.3 Million Americans[J].Economic Inquiry,2014,52(4):1485-1502.
[39]Hendren,N.Knowledge of Future Job Loss and Implications for Unemployment Insurance[J].American Economic Review,2017,107(7):1778-1823.
[40]Johnston,A.C.,and A.Mas.Potential Unemployment Insurance Duration and Labor Supply:The Individual and Market-level Response to a Benefit Cut[J].Journal of Political Economy,2018,126(6):2480-2522.
[41]Knabe,A.,and S.R覿tzel.Scarring or Scaring?The Psychological Impact of Past Unemployment and Future Unemployment Risk[J].Economica,2011,78(31):283-293.
[42]Kolsrud,J.,C.Landais,P.Nilsson,and J.Spinnewijn.The Optimal Timing of Unemployment Benefits:Theory and Evidence from Sweden[J].American Economic Review,2018,108(4-5):985-1033.
[43]Kroft,K.,and M.J.Notowidigdo.Should Unemployment Insurance Vary with the Unemployment Rate?Theory and Evidence[J].Review of Economic Studies,2016,83(3):1092-1124.
[44]Krueger,A.B.,and A.I.Mueller.Time Use,Emotional Well-being,and Unemployment:Evidence from Longitudinal Data[J].American Economic Review,2012,102(3):594-99.
[45]Meng,X.Unemployment,Consumption Smoothing,and Precautionary Saving in Urban China[J].Journal of Comparative Economics,2003,(31):465-485.
[46]Moon,H.R.,and M.Weidner.Linear Regression for Panel with Unknown Number of Factors as Interactive Fixed Effects[J].Econometrica,2015,83(4):1543-1579.
[47]Nalewaik,J.Current Consumption and Future Income Growth:Synthetic Panel Evidence[J].Journal of Monetary Economics,2006,53(8):2239-2266.
[48]Ravn,M.O.,and V.Sterk.Job Uncertainty and Deep Recessions[J].Journal of Monetary Economics,2017,(90):125-141.
[49]Stephens,M.Job Loss Expectations,Realizations,and Household Consumption Behavior[J].Review of Economics and Statistics,2004,86(1):253-269.
[50]Wei,S.J.,and X.Zhang.The Competitive Saving Motive:Evidence from Rising Sex Ratios and Savings Rates in China[J].Journal of Political Economy,2011,119(3):511-564.
[51]Zhang,J.,L.Xu,and H.Zhang.Uncovering the Truth about Chinese Urban Unemployment Rates:2005-2012[J].China and World Economy,2016,24(6):1-18.
[52]Zhao,D.,T.Wu,and Q.He.Consumption Inequality and Its Evolution in Urban China[J].China Economic Review,2017,(46):208-228.
(1)在1995—1999年的国有企业改革中,约有1500万员工下岗(He et al.,2018)。
(2)农民工平均受教育年限为9.6年,第三产业所要求的受教育年限为13.3年,两者并不完全匹配(蔡昉,2013)。
(1)He et al.(2018)从理论层面指出,失业率上升给就业者带来的不确定性可细分为两点:未来失业风险增加,以及即使不失业,未来降薪的概率也会增加。
(1)(Et-Et-1)Δyt+j也可以表示为:E(Δyt+j|Ωt)-E(Δyt+j|Ωt-1)。
(2)从图2看,这一假设较为合理。
(1)具体包括:户口在本地的常住非农业户;户口在本地的常住农业户;户口在外地,居住在本地半年以上的非农业户;户口在外地,居住在本地半年以上的农业户。
(2)即仅包括本市(县)非农业户口和本市(县)农业户口。Giles et al.(2005)认为,非户籍失业人口既可以回归农村,也可以流入其他城镇。因为迁徙具有一定自选择属性,所以利用当地户籍人口(相比于全体样本)计算失业率,能够更为准确地把握中国劳动市场变化规律。
(3)主要变量描述性统计详见附录,可在《中国工业经济》网站(http://www.ciejournal.org)下载。
(1)就业状态为“其他”的个体占比仅为2.24%,年龄均值为69.15岁,无论是将其归为失业还是非劳动力,均不会对本文结果造成较大影响。
(1)邹红和喻开志(2015)以及李宏彬等(2015)发现,退休使得与工作相关的消费支出降低25.1%和33%。
(2)失业率波动可能与其他未被控制的宏观变量相关,造成内生性。因此,这里参考Kroft and Notowidigdo(2016),将ΔUg,t与时间固定效应交叉项、ΔUg,t与地区固定效应交叉项分别控制,从而捕捉特定时间、特定地区劳动需求冲击对失业率的影响,发现未对基准结果造成显著影响。
(3)户主年龄一般在30岁以上,受教育水平基本保持不变,因此,属于前定变量。
(4)群组失业率可能与家庭成员就业状态相关,因此需要控制;其他住房数量用以控制家庭财富变化;借贷支出用以刻画是否面临借贷约束;年龄用以控制消费的生命周期效应。
(5)为了反映不同群组人口占比,这里采用加权最小二乘法进行相应调整。与此同时,由于群组失业率变动会导致群组内部产生序列相关性,这里在群组层面对标准误进行聚类(Cluster)。此外,根据Cameron et al.(2008),当核心变量群组类别较少(本文为32组)时,需要基于t分布而非正态分布计算P值。
(1)根据Meng(2003),在其他条件不变的情况下,预测失业率每增加1%,对于保持就业的家庭来说,其年度消费下降额为66元左右,约为本文测算值的26.7%。这一差异既可能来自识别技术,也可能来自样本时间范围(该文所用数据为1995年和1999年),甚至可能受到数据频率影响。
(1)子女年龄、是否和父母居住等信息,与家庭成员劳动参与决策密切相关(Blundell et al.,2016),这里一并予以控制。
(2)相比而言,张军等(2017)将在国有、城镇集体、联营经济、股份制经济、外商和港澳台经济、其他经济单位中工作,并由其支付工资的人员归为正规部门就业,同时将城镇个体或私营企业主、城镇个体或私营企业被雇者、其他就业者归为非正规部门就业者。
(1)Moon and Weidner(2015)证明,如果共同因子个数n超过最优值,将等价于控制若干无关变量,这不会对一致性造成影响。因此,拇指法则(Rule of Thumb)是连续增加公共因子个数,直至估计值保持稳定。
(1)具体指所有家庭消费支出占比的平均值。
(1)与消费方程不同,此处在控制变量中剔除了收入变量。
[2]白重恩,吴斌珍,金烨.中国养老保险缴费对消费和储蓄的影响[J].中国社会科学,2012b,(8):48-71.
[3]蔡昉.农民工就业面临各种风险[J].农村工作通讯,2013,(9):48.
[4]陈斌开,陆铭,钟宁桦.户籍制约下的居民消费[J].经济研究,2010,(S1):62-71.
[5]陈斌开,杨汝岱.土地供给、住房价格与中国城镇居民储蓄[J].经济研究,2013,(1):110-122.
[6]程令国,张晔.早年的饥荒经历影响了人们的储蓄行为吗?---对我国居民高储蓄率的一个新解释[J].经济研究,2010,(8):119-132.
[7]樊潇彦,袁志刚,万广华.收入风险对居民耐用品消费的影响[J].经济研究,2007,(4):124-136.
[8]黄少安,孙涛.非正规制度、消费模式和代际交叠模型---东方文化信念中居民消费特征的理论分析[J].经济研究,2005,(4):57-65.
[9]李宏彬,施新政,吴斌珍.中国居民退休前后的消费行为研究[J].经济学(季刊),2015,(1):117-134.
[10]卢锋,刘晓光,姜志霄,张杰平.劳动力市场与中国宏观经济周期:兼谈奥肯定律在中国[J].中国社会科学,2015,(12):69-89.
[11]吕丹,曲展.典型国家失业保险制度[J].中国劳动,2014,(10):29-33.
[12]温兴祥.失业、失业风险与农民工家庭消费[J].南开经济研究,2015,(6):110-128.
[13]杨继生,徐娟.环境收益分配的不公平性及其转移机制[J].经济研究,2016,(1):155-167.
[14]杨继生,阳建辉.行政垄断、政治庇佑与国有企业的超额成本[J].经济研究,2015,(4):50-61.
[15]张军,赵达,周龙飞.最低工资标准提高对就业正规化的影响[J].中国工业经济,2017,(1):81-97.
[16]赵达,谭之博,张军.中国城镇地区消费不平等演变趋势---新视角与新证据[J].财贸经济,2017,(6):115-129.
[17]赵达,周龙飞.非线性定价有效吗[J].统计研究,2018,(8):58-68.
[18]邹红,喻开志.退休与城镇家庭消费:基于断点回归设计的经验证据[J].经济研究,2015,(1):124-139.
[19]Aguiar,M.,and E.Hurst.Consumption versus Expenditure[J].Journal of Political Economy,2005,113(5):919-948.
[20]Aguiar,M.,and E.Hurst.Life-cycle Prices and Production[J].American Economic Review,2007,97(5):1533-1559.
[21]Aguiar,M.,E.Hurst,and L.Karabarbounis.Time Use during the Great Recession[J].American Economic Review,2013,103(5):1664-1696.
[22]Aladangady,A.Housing Wealth and Consumption:Evidence from Geographically-linked Microdata[J].American Economic Review,2017,107(11):3415-3446.
[23]Apergis,N.,and Y.Georgellis.Regional Unemployment and Employee Loyalty:Evidence from 12 UK Regions[J].Regional Studies,2018,52(9):1283-1293.
[24]Bai,J.Panel Data Models with Interactive Fixed Effects[J].Econometrica,2009,77(4):1229-1279.
[25]Bartik,T.J.Who Benefits from State and Local Economic Development Policies[M].Kalamazoo,MI:Upjohn Institute,1991.
[26]Battistin,E.,A.Brugiavini,E.Rettore,and G.Weber.The Retirement Consumption Puzzle:Evidence from a Regression Discontinuity Approach[J].American Economic Review,2009,99(5):2209-2226.
[27]Blundell,R.,L.Pistaferri,and I.Preston.Consumption Inequality and Partial Insurance[J].American Economic Review,2008,98(5):1887-1921.
[28]Blundell,R.,L.Pistaferri,and I.Saporta-Eksten.Consumption Inequality and Family Labor Supply[J].American Economic Review,2016,106(2):387-435.
[29]Browning,M.,and T.F.Crossley.Shocks,Stocks,and Socks:Smoothing Consumption over a Temporary Income Loss[J].Journal of the European Economic Association,2009,7(6):1169-1192.
[30]Cameron,A.C.,J.B.Gelbach,and D.L.Miller.Bootstrap-based Improvements for Inference with Clustered Errors[J].Review of Economics and Statistics,2008,90(3):414-427.
[31]Campos,R.G.,and I.Reggio.Consumption in the Shadow of Unemployment[J].European Economic Review,2015,(78):39-54.
[32]Chodorow-Reich,G.,J.Coglianese,and L.Karabarbounis.The Macro Effects of Unemployment Benefit Extensions:A Measurement Error Approach[J].Quarterly Journal of Economics,2019,134(1):227-229.
[33]Diamond R.The Determinants and Welfare Implications of U.S.Workers’Diverging Location Choices by Skill:1980-2000[J].American Economic Review,2016,106(3):479-524.
[34]Feng,S.,Y.Hu,and R.Moffitt.Long Run Trends in Unemployment and Labor Force Participation in Urban China[J].Journal of Comparative Economics,2017,45(2):304-324.
[35]Gan,L.,M.D.Hurd,and D.L.McFadden.Analyses in the Economics of Aging[M].Chicago:University of Chicago Press,2005.
[36]Giles,J.,P.Albert,and J.Zhang.What is China’s True Unemployment Rate[J].China Economic Review,2005,16(2):149-170.
[37]He,H.,F.Huang,Z.Liu,and D.Zhu.Breaking the Iron Rice Bowl:Evidence of Precautionary Savings from Chinese State-owned Enterprises Reform[J].Journal of Monetary Economics,2018,(94):94-113.
[38]Helliwell,J.F.,and H.Huang.New Measures of the Costs of Unemployment:Evidence from the Subjective Well-being of 3.3 Million Americans[J].Economic Inquiry,2014,52(4):1485-1502.
[39]Hendren,N.Knowledge of Future Job Loss and Implications for Unemployment Insurance[J].American Economic Review,2017,107(7):1778-1823.
[40]Johnston,A.C.,and A.Mas.Potential Unemployment Insurance Duration and Labor Supply:The Individual and Market-level Response to a Benefit Cut[J].Journal of Political Economy,2018,126(6):2480-2522.
[41]Knabe,A.,and S.R覿tzel.Scarring or Scaring?The Psychological Impact of Past Unemployment and Future Unemployment Risk[J].Economica,2011,78(31):283-293.
[42]Kolsrud,J.,C.Landais,P.Nilsson,and J.Spinnewijn.The Optimal Timing of Unemployment Benefits:Theory and Evidence from Sweden[J].American Economic Review,2018,108(4-5):985-1033.
[43]Kroft,K.,and M.J.Notowidigdo.Should Unemployment Insurance Vary with the Unemployment Rate?Theory and Evidence[J].Review of Economic Studies,2016,83(3):1092-1124.
[44]Krueger,A.B.,and A.I.Mueller.Time Use,Emotional Well-being,and Unemployment:Evidence from Longitudinal Data[J].American Economic Review,2012,102(3):594-99.
[45]Meng,X.Unemployment,Consumption Smoothing,and Precautionary Saving in Urban China[J].Journal of Comparative Economics,2003,(31):465-485.
[46]Moon,H.R.,and M.Weidner.Linear Regression for Panel with Unknown Number of Factors as Interactive Fixed Effects[J].Econometrica,2015,83(4):1543-1579.
[47]Nalewaik,J.Current Consumption and Future Income Growth:Synthetic Panel Evidence[J].Journal of Monetary Economics,2006,53(8):2239-2266.
[48]Ravn,M.O.,and V.Sterk.Job Uncertainty and Deep Recessions[J].Journal of Monetary Economics,2017,(90):125-141.
[49]Stephens,M.Job Loss Expectations,Realizations,and Household Consumption Behavior[J].Review of Economics and Statistics,2004,86(1):253-269.
[50]Wei,S.J.,and X.Zhang.The Competitive Saving Motive:Evidence from Rising Sex Ratios and Savings Rates in China[J].Journal of Political Economy,2011,119(3):511-564.
[51]Zhang,J.,L.Xu,and H.Zhang.Uncovering the Truth about Chinese Urban Unemployment Rates:2005-2012[J].China and World Economy,2016,24(6):1-18.
[52]Zhao,D.,T.Wu,and Q.He.Consumption Inequality and Its Evolution in Urban China[J].China Economic Review,2017,(46):208-228.
(1)在1995—1999年的国有企业改革中,约有1500万员工下岗(He et al.,2018)。
(2)农民工平均受教育年限为9.6年,第三产业所要求的受教育年限为13.3年,两者并不完全匹配(蔡昉,2013)。
(1)He et al.(2018)从理论层面指出,失业率上升给就业者带来的不确定性可细分为两点:未来失业风险增加,以及即使不失业,未来降薪的概率也会增加。
(1)(Et-Et-1)Δyt+j也可以表示为:E(Δyt+j|Ωt)-E(Δyt+j|Ωt-1)。
(2)从图2看,这一假设较为合理。
(1)具体包括:户口在本地的常住非农业户;户口在本地的常住农业户;户口在外地,居住在本地半年以上的非农业户;户口在外地,居住在本地半年以上的农业户。
(2)即仅包括本市(县)非农业户口和本市(县)农业户口。Giles et al.(2005)认为,非户籍失业人口既可以回归农村,也可以流入其他城镇。因为迁徙具有一定自选择属性,所以利用当地户籍人口(相比于全体样本)计算失业率,能够更为准确地把握中国劳动市场变化规律。
(3)主要变量描述性统计详见附录,可在《中国工业经济》网站(http://www.ciejournal.org)下载。
(1)就业状态为“其他”的个体占比仅为2.24%,年龄均值为69.15岁,无论是将其归为失业还是非劳动力,均不会对本文结果造成较大影响。
(1)邹红和喻开志(2015)以及李宏彬等(2015)发现,退休使得与工作相关的消费支出降低25.1%和33%。
(2)失业率波动可能与其他未被控制的宏观变量相关,造成内生性。因此,这里参考Kroft and Notowidigdo(2016),将ΔUg,t与时间固定效应交叉项、ΔUg,t与地区固定效应交叉项分别控制,从而捕捉特定时间、特定地区劳动需求冲击对失业率的影响,发现未对基准结果造成显著影响。
(3)户主年龄一般在30岁以上,受教育水平基本保持不变,因此,属于前定变量。
(4)群组失业率可能与家庭成员就业状态相关,因此需要控制;其他住房数量用以控制家庭财富变化;借贷支出用以刻画是否面临借贷约束;年龄用以控制消费的生命周期效应。
(5)为了反映不同群组人口占比,这里采用加权最小二乘法进行相应调整。与此同时,由于群组失业率变动会导致群组内部产生序列相关性,这里在群组层面对标准误进行聚类(Cluster)。此外,根据Cameron et al.(2008),当核心变量群组类别较少(本文为32组)时,需要基于t分布而非正态分布计算P值。
(1)根据Meng(2003),在其他条件不变的情况下,预测失业率每增加1%,对于保持就业的家庭来说,其年度消费下降额为66元左右,约为本文测算值的26.7%。这一差异既可能来自识别技术,也可能来自样本时间范围(该文所用数据为1995年和1999年),甚至可能受到数据频率影响。
(1)子女年龄、是否和父母居住等信息,与家庭成员劳动参与决策密切相关(Blundell et al.,2016),这里一并予以控制。
(2)相比而言,张军等(2017)将在国有、城镇集体、联营经济、股份制经济、外商和港澳台经济、其他经济单位中工作,并由其支付工资的人员归为正规部门就业,同时将城镇个体或私营企业主、城镇个体或私营企业被雇者、其他就业者归为非正规部门就业者。
(1)Moon and Weidner(2015)证明,如果共同因子个数n超过最优值,将等价于控制若干无关变量,这不会对一致性造成影响。因此,拇指法则(Rule of Thumb)是连续增加公共因子个数,直至估计值保持稳定。
(1)具体指所有家庭消费支出占比的平均值。
(1)与消费方程不同,此处在控制变量中剔除了收入变量。