加载中...
Why Do Companies Choose Labor Outsourcing: An Explanation Based on Economic Policy Uncertainty
劳务外包是近年来备受企业青睐的一种灵活用工方式。尽管实践中劳务外包的市场规模呈现迅速扩张趋势,却鲜有研究对其成因展开分析。本文通过手工整理上市公司披露的劳务外包信息,发现经济政策不确定性会导致企业更多地选择劳务外包,且上述现象在民营企业中更加显著。进一步地,本文利用自然语言处理方法,从宏观、行业和企业三个层面构建不确定性指标,发现与企业自身和行业的不确定性相比,宏观层面的不确定性更可能导致企业选择劳务外包;通过区分劳务外包的业务水平发现,经济政策不确定性更可能导致企业采用劳务外包来完成低端业务而非高端业务。异质性检验发现,当企业的调整成本较高时,经济政策不确定性更可能导致企业选择劳务外包;而企业在产品市场上更稳定的客户关系、较高的议价能力以及政府对企业的政策支持有助于降低经济政策不确定性对企业劳务外包的影响。本文还发现,当经济政策不确定性较高时,企业通过劳务外包能够提高生产效率、缓解劳动力成本刚性,但同时可能挤出对低学历员工的用工需求,造成劳务外包人员遭受不平等待遇等问题。
Abstract
Labor outsourcing is a special form of labor division that allows companies to outsource repetitive, non-core activities to third-party labor contractors, thereby focusing their core resources and efforts on core business activities. The fundamental characteristic of labor outsourcing is that it helps enterprises convert long-term employment decisions into short-term employment contracts, which is reflected in increased employment flexibility and reduced labor costs. These two features make labor outsourcing a more effective employment option when dealing with uncertain market conditions. In practice, the increase in labor outsourcing is closely related to economic policy uncertainty. Economic policy uncertainty leads to the inability of firms to form consistent and stable expectations about the future decision-making environment, hence a decline in firms' demand for long-term employment. Therefore, this paper attempts to explain the logic of the abnormal boom in the labor outsourcing market in recent years from the perspective of increased economic policy uncertainty. By manually collecting the labor outsourcing information disclosed in the annual reports of listed companies from 2012 to 2021, this paper finds that economic policy uncertainty makes firms be inclined to opt for labor outsourcing, which is more significant among private firms. Furthermore, this paper uses natural language processing to construct uncertainty indicators at the macro, industry, and enterprise levels, and finds that compared with the uncertainty of enterprises and the industry, the uncertainty at the macro level is more likely to lead enterprises to choose labor outsourcing. Meanwhile, by distinguishing the business level of labor outsourcing, this paper finds that enterprises tend to outsource low-end rather than high-end business when confronted with economic policy uncertainty. The heterogeneity test shows that economic policy uncertainty is more likely to lead to labor outsourcing in firms with higher adjustment costs, while stable customer relationships in the product market, higher bargaining power, and government's policy support help to reduce the impact of economic policy uncertainty on firms' labor outsourcing. Finally, the paper finds that when economic policy uncertainty is high, labor outsourcing can improve productivity and reduce labor cost rigidity. However, it may crowd out the demand for low-skilled labor and create unequal treatment among outsourced workers.
经济政策不确定性显著提高企业采用劳务外包的概率
以2012—2021年A股上市公司为样本,采用Probit模型检验,EPU_Baker系数为0.0453且在1%水平显著,表明经济政策不确定性越高,企业越可能采用劳务外包。该结论在控制行业和省份固定效应后依然成立。
经济政策不确定性显著增加企业劳务外包工作量
以劳务外包报酬支出占员工总薪酬之比度量劳务外包工作量,采用Tobit模型回归,EPU_Baker系数为0.0069且在1%水平显著,说明不确定性上升不仅提高外包概率,也扩大外包规模。
民营企业受经济政策不确定性影响更显著
加入EPU与所有制交互项后,EPU_Baker×SOE系数为-0.0506且在1%水平显著为负,表明相较于国有企业,民营企业风险承担能力较弱、融资约束较强,在不确定性上升时更倾向于采用劳务外包。
宏观层面不确定性比行业和企业层面更易引致劳务外包
利用Word2Vec从年报管理层讨论与分析文本构建宏观、行业和企业三重不确定性指标,回归显示仅宏观不确定性系数显著为正(0.3033),行业和企业层面不显著,表明劳务外包主要是企业应对宏观政策环境变化的策略。
经济政策不确定性主要促使企业外包低端业务
以劳务外包平均小时工资区分业务水平,EPU对低端业务外包(Low_end)系数为0.0535且在1%水平显著,对高端业务(High_end)不显著,说明不确定性主要推动企业将替代性高、技术含量低的业务外包。
高调整成本企业更易因不确定性选择劳务外包
以劳动力依赖程度和机器密集度度量调整成本,EPU×LaborInt和EPU×RobotsInt系数均显著为正,表明劳动力调整成本或资本调整成本较高的企业,在不确定性上升时更有动力通过劳务外包增加用工灵活性。
稳定客户关系、议价能力和政策支持削弱外包倾向
EPU×StableClient、EPU×BargainPower系数显著为负,EPU×Subsidy、EPU×GovPurchase系数显著为负,说明稳定的客户关系、较强的议价能力、稳岗补贴及政府采购支持有助于降低不确定性对企业劳务外包行为的推动作用。
劳务外包提高生产效率并缓解劳动力成本粘性
以劳动生产率与全要素生产率度量生产效率,EPU×D_LaborOuts系数为2.4049且在1%水平显著;以未来一期劳动力成本粘性为被解释变量,EPU×D_LaborOuts系数为-0.0606且在1%水平显著,表明劳务外包有助于降本增效。
劳务外包挤出低学历员工需求并加剧薪酬不平等
EPU×D_LaborOuts对低学历员工占比系数为-0.0035且在1%水平显著,对高学历占比显著为正;同时劳务外包人员小时工资长期在25—35元波动,而在册员工工资持续上升,显示外包可能加剧弱势群体不平等。
核心解释变量
经济政策不确定性(EPU_Baker),采用Baker et al.(2016)构建的月度中国经济政策不确定性指数年内算术平均后除以100,并滞后1期;稳健性中使用Davis et al.(2019)基于《人民日报》《光明日报》构建的EPU_Davis、几何平均指数rEPU_Baker、中位数虚拟变量DEPU_Baker,以及基于年报文本的宏观、行业、企业三层不确定性指标(EPU_macro、EPU_industry、EPU_firm)
被解释变量
企业是否采用劳务外包(D_LaborOuts),为哑变量;以及企业劳务外包工作量(LaborOuts),以劳务外包报酬支出占企业员工总薪酬之比度量;稳健性中区分高端业务外包(High_end,小时工资≥50元)与低端业务外包(Low_end,小时工资<50元)
样本与数据
2012—2021年中国A股上市公司,初始观测值33274个,剔除ST、金融类及数据缺失样本后得到28059个观测值。劳务外包数据由作者从上市公司年报手工整理,经济政策不确定性来自Baker et al.(2016),其他变量来自CSMAR、CNRDS和国家统计局,连续变量进行上下1%缩尾处理
识别方法 / 模型设定
基准采用Probit模型(式1)和Tobit模型(式2)估计,并采用公司—年度双维聚类标准误;控制企业特征变量和宏观经济变量,加入行业和省份固定效应;异质性分析通过加入EPU与调整成本、客户关系、政策支持等变量的交互项进行识别;进一步分析采用OLS并控制个体固定效应
内生性及稳健性检验
(1)基于年报管理层讨论与分析文本,利用Word2Vec从宏观、行业和企业三个层面构建不确定性指标替换核心解释变量;(2)区分高端与低端业务外包;(3)更换EPU度量(Davis等指数、几何平均、虚拟变量);(4)仅保留披露劳务外包的样本并用OLS重新估计;(5)加入GDP增速标准差(GDP_sd)控制经济波动的不确定性,并对解释变量和被解释变量取差值进行动态检验
更多相关数据正在补充