公共基础设施投资与全要素生产率:基于异质企业家模型的理论分析
|
摘要
本文构建一个异质性企业家模型,以中国现实数据为基础,利用数值模拟深入考察公共基础设施投资对全要素生产率(TFP)的影响机理,同时考虑金融摩擦、劳动摩擦和其他财税政策对公共基础设施投资TFP效应的抑制作用。研究表明,公共基础设施投资对TFP具有明显的"倒U型"影响,即存在TFP最大化的公共基础设施投资规模。这源于公共基础设施投资影响企业家的财富积累,即自融资机制,对企业间要素配置的重要影响。信贷约束对公共基础设施投资的TFP效应影响显著(企业家面临的信贷约束越严重,公共基础设施投资对TFP的影响越大)。存贷利差的影响则较弱。减少劳动摩擦可增强公共基础设施投资的TFP效应,较大的有效基础设施服务、企业间差异性和公共消费性支出也具有类似影响,利用消费税融资可更好地发挥公共基础设施投资对TFP的促进作用。
How public infrastructure investments affect economic growth is an important topic for both academicians and practitioners. A large body of theoretical research has explored the mechanism through which public infrastructure investments affect firm production and thus economic growth. However,studies have typically used the representative agent optimization model and assumed that total factor productivity( TFP) is exogenous and constant,therefore neglecting an important theoretical issue—the effect of public infrastructure investments on TFP,which has widely been considered a key driving force of long-run economic growth. This issue is particularly important for many developing countries,as public infrastructure investments have become a prevalent policy instrument for facilitating economic growth in those countries.This paper closes the current research gap by developing a general equilibrium model with heterogeneous entrepreneurs,financial frictions( collateral constraints and deposit-loan interest spreads),labor friction( overhead labor)and public infrastructure services as inputs to firm production. This public service entails congestion and is financed by a proportional income tax. The paper then quantitatively analyzes the effects of public infrastructure investments on TFP by calibrating the model to the Chinese economy from 1992 to 2014. The analysis shows that public infrastructure investments have an inverted-U effect on TFP,and that the size of public infrastructure investment that maximizes TFP is equal to the elasticity of public infrastructure investment with respect to aggregate output. Public infrastructure investments may affect the accumulation of entrepreneurs' wealth, including their self-financing practices, which can help entrepreneurs to overcome collateral constraints and thus reduce productivity losses by improving the allocation of production factors across firms.Furthermore,this paper extensively analyzes the importance of several factors shaping the effects of public infrastructure investments. Specifically,collateral constraints are shown to have a significant role in shaping the effects of public infrastructure investments on TFP—public infrastructure investments tend to have a larger effect on TFP in an economy with tighter collateral constraints,while deposit-loan interest spreads have a relatively weaker effect. This paper also extends the model to incorporate public consumption expenditure that affects entrepreneurs' utility,and finds that a larger size of public consumption expenditure increases the effects( particularly the negative effects) of public infrastructure investments on TFP. An extended analysis shows that,compared with corporate income taxation,using consumption taxes to finance an increase in public infrastructure investment increases the positive effects of that investment on TFP. The results of these analyses are both theoretically relevant and practically meaningful; they substantially improve understanding of the effects of public infrastructure investments and provide important and useful guidance for designing better public infrastructure investment policies.
How public infrastructure investments affect economic growth is an important topic for both academicians and practitioners. A large body of theoretical research has explored the mechanism through which public infrastructure investments affect firm production and thus economic growth. However,studies have typically used the representative agent optimization model and assumed that total factor productivity( TFP) is exogenous and constant,therefore neglecting an important theoretical issue—the effect of public infrastructure investments on TFP,which has widely been considered a key driving force of long-run economic growth. This issue is particularly important for many developing countries,as public infrastructure investments have become a prevalent policy instrument for facilitating economic growth in those countries.This paper closes the current research gap by developing a general equilibrium model with heterogeneous entrepreneurs,financial frictions( collateral constraints and deposit-loan interest spreads),labor friction( overhead labor)and public infrastructure services as inputs to firm production. This public service entails congestion and is financed by a proportional income tax. The paper then quantitatively analyzes the effects of public infrastructure investments on TFP by calibrating the model to the Chinese economy from 1992 to 2014. The analysis shows that public infrastructure investments have an inverted-U effect on TFP,and that the size of public infrastructure investment that maximizes TFP is equal to the elasticity of public infrastructure investment with respect to aggregate output. Public infrastructure investments may affect the accumulation of entrepreneurs' wealth, including their self-financing practices, which can help entrepreneurs to overcome collateral constraints and thus reduce productivity losses by improving the allocation of production factors across firms.Furthermore,this paper extensively analyzes the importance of several factors shaping the effects of public infrastructure investments. Specifically,collateral constraints are shown to have a significant role in shaping the effects of public infrastructure investments on TFP—public infrastructure investments tend to have a larger effect on TFP in an economy with tighter collateral constraints,while deposit-loan interest spreads have a relatively weaker effect. This paper also extends the model to incorporate public consumption expenditure that affects entrepreneurs' utility,and finds that a larger size of public consumption expenditure increases the effects( particularly the negative effects) of public infrastructure investments on TFP. An extended analysis shows that,compared with corporate income taxation,using consumption taxes to finance an increase in public infrastructure investment increases the positive effects of that investment on TFP. The results of these analyses are both theoretically relevant and practically meaningful; they substantially improve understanding of the effects of public infrastructure investments and provide important and useful guidance for designing better public infrastructure investment policies.
引文
盖庆恩、朱喜、程名望、史清华,2015:《要素市场扭曲、垄断势力与全要素生产率》,《经济研究》第5期。
龚关、胡关亮,2013:《中国制造业资源配置效率与全要素生产率》,《经济研究》第4期。
黄赜琳,2005:《中国经济周期特征与财政政策效应:一个基于三部门RBC模型的实证分析》,《经济研究》第6期。
贾俊雪,2014:《税收激励、企业有效平均税率与企业进入》,《经济研究》第7期。
金戈,2016:《中国基础设施与非基础设施资本存量及其产出弹性估算》,《经济研究》第5期。
经济增长前沿课题组,2004:《财政政策的供给效应与经济发展》,《经济研究》第9期。
娄洪,2004:《长期经济增长中的公共投资政策》,《经济研究》第3期。
张光南、李小瑛、陈广汉,2010:《中国基础设施的就业、产出和投资效应》,《管理世界》第4期。
张军、高远、傅勇、张弘,2007:《中国为什么拥有良好的基础设施?》,《经济研究》第3期。
张学良,2012:《中国交通基础设施促进区域经济增长了吗?》,《中国社会科学》第3期。
Achdou,Y.,Jiequn Han,J-M.Lasry,P-L.Lions,and B.Moll,2015,“Heterogeneous Agent Models in Continuous Time”,Working paper,Princeton University.
Aschauer,D.A.,1989,“Is Public Expenditure Productive?”,Journal of Monetary Economics,Vol.23,177-200.
Bai,Chong-En,Chang-Tai Hsieh,and Yingyi Qian,2006,“The Return to Capital in China”,Brookings Papers on Economic Activity,Vol.2,61-88.
Barro,R.J.,1990,“Government Spending in a Simple Model of Endogenous Growth”,Journal of Political Economy,Vol.98,103-125.
Bartelsman,E.,J.Haltiwanger,and S.Scarpetta,2013,“Cross-country Differences in Productivity:The Role of Allocation and Selection”,American Economic Review,Vol.103,305-334.
Baxter,M.,and R.G.King,1993,“Fiscal Policy in General Equilibrium”,American Economic Review,Vol.83,315-334.
Beck,T.,A.Demirguc-Kunt,and R.Levine,2000,“A New Database on the Structure and Development of the Financial Sector”,World Bank Economic Review,Vol.14,597-605.
Bom,P.R.D.,and J.E.Ligthart,2014,“Public Infrastructure Investment,Output Dynamics,and Balanced Budget Fiscal Rules”,Journal of Economic Dynamics and Control,Vol.40,334-354.
Brandt,L.,J.V.Biesebroeck,and Yifan Zhang,2012,“Creative Accounting or Creative Destruction?Firm-level Productivity Growth in Chinese Manufacturing”,Journal of Development Economics,Vol.97,339-351.
Buera,F.J.,J.P.Kaboski,and Y.Shin,2011,“Finance and Development:A Tale of Two Sectors”,American Economic Review,Vol.101,1964-2002.
Buera,F.J.,and Y.Shin,2013,“Financial Frictions and the Persistence of History:A Quantitative Exploration”,Journal of Political Economy,Vol.121,221-272.
Dekle,R.,and G.Vandenbroucke,2006,“A Quantitative Analysis of China's Structural Transformation”,Working Paper N0651,Institute of Economic Policy Research.
Feenstra,R.,Zhiyuan Li,and Miaojie Yu,2014,“Exports and Credit Constraints under Incomplete Information:Theory and Evidence from China”,Review of Economic and Statistics,Vol.96,729-744.
Fernald,J.G.,1999,“Roads to Prosperity?Assessing the Link between Public Capital and Productivity”,American Economic Review,Vol.89,619-638.
Fisher,W.H.,and S.J.Turnovsky,1998,“Public Investment,Congestion,and Private Capital Accumulation”,Economic Journal,Vol.108,399-413.
Hsieh,Chang-Tai,and P.J.Klenow,2009,“Misallocation and Manufacturing TFP in China and India”,Quarterly Journal of Economics,Vol.124,1403-1448.
Jia,Junxue,Jing Guo,and Zijie Wang,2015,“The Fiscal-monetary Policy Mix and Exchange Rate Stability:A Dynamic Stochastic General Equilibrium Model with Chinese Characteristics”,Emerging Markets Finance and Trade,Vol.51,729-746.
Levinsohn,J.,and A.Petrin,2003,“Estimating Production Functions Using Inputs to Control for Unobservable”,Review of Economic Studies,Vol.70,317-341.
Moll,B.,2014,“Productivity Losses from Financial Frictions:Can Self-financing Undo Capital Misallocation?”,American Economic Review,Vol.104,3186-3221.
Restuccia,D.,and R.Rogerson,2008,“Policy Distortions and Aggregate Productivity with Heterogeneous Plants”,Review of Economic Dynamics,Vol.11,707-720.
Song,Zheng,K.Storesletten,and F.Zilibotti,2011,“Growing Like China”,American Economic Review,Vol.101,202-241.
(1)大量文献实证考察了公共基础设施投资(或存量)的经济影响,如Aschauer(1989)、Fernald(1999)、张光南等(2010)和张学良(2012)。
(1)图1纵轴为1998—2006年间地级市TFP的均值(除以样本最大值),横轴为公共基础设施投资规模gI(基本建设支出与GDP比值)的均值,剔除小于1%和大于99%分位数的样本以避免异常值的影响。拟合曲线回归结果为:TFP=0.100(0.009)+0.816(0.341)gI-5.056(2.002)g2I,R2=0.02,样本数为326,小括号中数字为标准差。基本建设支出数据来自1998—2006年间《全国地市县财政统计资料》,TFP数据由企业TFP(用Levinsohn&Petrin(2003)方法估算)在地级市层面加权平均得到(中间投入比重为权重),企业数据来自1997—2007年间国家统计局规模以上工业企业调查数据。本文利用企业相关信息(如企业代码和所在地等)对样本企业进行匹配(Brandt et al.,2012),剔除了:总资产等关键变量缺失、从业人员少于10人(Feenstra et al.,2014)、总资产小于流动资产或固定资产、企业代码缺失或无法唯一识别和建立时间无效的企业,最终包含56.9万家企业。
(1)与Moll(2014)一样,本文中信贷约束为系统性的(企业家面临的杠杆率约束相同),但高生产率企业家的最优投资规模进而外部资金需求更大,故信贷约束对其影响更大。与本文不同,Song et al.(2011)先验区分了两类企业(国有和民营企业),且假定它们面临的信贷约束不同,Buera et al.(2011)则先验区分了两个部门(制造业和服务业),且假定不同部门的企业进入成本进而资金需求不同。
(2)1997-2007年间,我国规模以上工业企业的劳动生产率(劳均产出)与劳动规模的相关系数为0.55。Song et al.(2011)、Buera et al.(2011)、Buera&Shin(2013)和Moll(2014)忽略了这一典型事实,Bartelman et al.(2013)则没有考虑金融摩擦的影响。
(3)借鉴Barro(1990),本文考虑公共基础设施投资而非存量,这可简化模型,以给出整个经济TFP的解析解(见(17)式)。
(1)现实中,规模较大的企业通常具有较大能力要求政府提供较好的基础设施服务。例如,为吸引大中型企业投资办厂,我国地方政府热衷设立各种经济开发区,为这些企业提供良好的道路交通和电力通讯等服务,小企业则很难享受此类特殊待遇。
(1)由(16)式可得整个经济的有效劳均生产函数:y_t=Zt(kαt)υg1-υt,yt=Yt/(Lt-)、kt=Kt/(Lt-)和gt=Gt/(Lt-)。这与Barro(1990)的生产函数较一致,故本文由受信贷约束的异质性企业家模型加总得到类似Barro(1990)的模型。但Barro(1990)将TFP设为完全外生的,本文中企业家生产率为外生,但整个经济的TFP很大程度上为内生的,取决于企业间的要素配置。
(1)(22)式等价于连续的AR(1)过程,稳态分布服从均值为零、方差为σ2/(2θ)的正态分布,自相关系数为e-θ。此时,由Ito公式可知:(1)式中的漂移项μ(zit)=[-θlogzit+σ2/2]zit,扩散项为(σzit)2。
(2)数据来自国家统计局网站(www.stats.gov.cn)。也有研究利用全社会固定资产投资中来自国家预算的资金度量公共基础设施投资。全社会固定资产投资中来自国家预算资金与GDP的比值在1992-2014年间也为0.022,故使用这两个指标度量公共基础设施投资没有差异。
(1)总储蓄S=Y-C-G,Y为支出法核算的GDP,C为居民消费,G为政府消费(Song et al.,2011)。1992-2014年间,我国总储蓄率为44%。鉴于模型刻画的是企业家行为,而企业家储蓄率较高,故以一个较大值57%作为赋值依据。1992-2005年间,我国的资本回报率为21.8%但呈下降趋势(Bai et al.,2006),故以一个较小值17.4%作为赋值依据。我们也尝试(61.6%,15.3%)和(53%,18.5%)两组储蓄率和资本回报率为赋值依据,基准结论很稳健。
(2)为使数值有意义,对TFP和总产出进行标准化处理,即除以参照经济的TFP和总产出。参照经济为不存在存贷利差和信贷约束(ξ=0;为避免λ=∞带来的计算困难,以λ=10为不存在信贷约束的近似赋值),以及公共基础设施投资为最优规模的经济。公共基础设施投资的Samuelson效率条件为:Yt/G=1,故最优g*I=(1-υ)/[Z1-t YtZGZt/G=0,故t]。TFP最大化要求:TFP最大化的gI=1-υ(见图2(a)),但无法从理论上给出g*I。模拟结果表明:ξ=0和λ=10时,gI为0.25的产出最大,而TFP与gI=0.17时的几乎一样,故以0.25作为参照经济gI的赋值。=0时,企业劳动生产率与劳动规模不存在相关性,与经验事实不符,故未将=0的经济作为参照经济。
(1)我们也尝试改变其他参数(如ρ、1-υ、θ和σ等),在较合理的取值范围内,基准结论具有良好的稳健性。
(2)Bartelsman et al.(2013)发现Clpl随间接劳动增加而先增后降,给出相似解释。Hsieh&Klenow(2009)忽略了企业退出以避免此问题。
(1)这一结论有助于更好地认识2008年世界金融危机期间我国“四万亿元投资计划”的影响。此间,非常宽松的货币信贷政策极大放松了企业的信贷约束(λ明显增大),因此以公共基础设施投资为主要措施的“四万亿元投资计划”促进了经济的快速恢复。但就图4(a)和(b)的模拟结果来看:这可能主要表现为总量扩张(产出增加)而非效率改善(TFP提升)。
(1)本文其他部分,固定,Clpl的增加源于要素更多流向高生产率企业,即配置效率的提升。图6(d)考察的是变化的影响,此时Clpl增加并不代表配置效率提升。也引入了内生企业退出机制:模拟结果表明,企业退出率Ext在=0时为零且随增加而增加。Bartelsman et al.(2013)没有考虑信贷约束,故他们的模型中对Ext的影响很大,致使Clpl随小幅增加而先增后降。本文中信贷约束的存在使得对Ext的影响较小:不存在最优投资规模限制延缓了低生产率非信贷受限企业的退出,这更符合经济现实。模拟结果表明,λ越大,对Ext的影响越大,Clpl越倾向于随增加而先增后降,即Bartelsman et al.(2013)的结论是本文的一个特例。km
(2)由(αυ8)式可知:当zit<-z时,有it zitz km=,即非信贷受限企业的资本投资关于生产率的弹性随γ减小而增大。it(1-υ)γ
(1)一些研究表明,企业面临的特异性政策差异会导致要素错配和TFP损失(Resuccia&Rogerson,2008;Bartelsman et al.,2013)。但这些研究主要通过类似差异性税收的方式引入特异性政策差异,忽略了一些政策差异(如本文考虑的基础设施服务差异)对企业家的正向激励,得到这样的结论较为直观。本文可视为这些研究的有益补充,有利于更全面地认识特异性政策差异对企业间要素配置的影响。
(2)其他参数仍采用基准值,只将调整为0.301以使模型关于企业劳动生产率与劳动规模相关系数的预测值与现实数据相匹配。
(3)本文也尝试将η由0.5增加为2。此时,2mit/cgc<0,即公共消费支出的引入将降低私人消费的边际效用、减少企业家消费,但总体仍会抑制企业家财富积累(因为税率增加对企业家财富积累的抑制更突出),故结论不变。
(1)本文中的消费税类似现实经济中的商品服务课税,而非对特定消费(如烟酒)征收的“消费税”。这一理论拓展也是基于对中国经济的现实考量:商品服务课税和企业所得税在我国财政收入体系中占据重要地位:1992-2014年间,企业所得税占税收和财政总收入的比重分别为16.4%和14.9%,国内增值税、国内消费税和营业税三者占税收和财政总收入的比重分别为55.9%和51.2%。为确保τc很大时效用函数的凹性,采用对数效用函数(η=1)。2000-2008年间规模以上工业企业所得税有效平均税率为0.2(贾俊雪,2014),故将τf基准值设为0.2(也尝试0.33或0.25的法定税率,结论一致),可得τc基准值为-0.101。模型关于储蓄率、企业劳动生产率与劳动规模相关系数的预测值与现实数据相匹配,但资本租金率较低。
龚关、胡关亮,2013:《中国制造业资源配置效率与全要素生产率》,《经济研究》第4期。
黄赜琳,2005:《中国经济周期特征与财政政策效应:一个基于三部门RBC模型的实证分析》,《经济研究》第6期。
贾俊雪,2014:《税收激励、企业有效平均税率与企业进入》,《经济研究》第7期。
金戈,2016:《中国基础设施与非基础设施资本存量及其产出弹性估算》,《经济研究》第5期。
经济增长前沿课题组,2004:《财政政策的供给效应与经济发展》,《经济研究》第9期。
娄洪,2004:《长期经济增长中的公共投资政策》,《经济研究》第3期。
张光南、李小瑛、陈广汉,2010:《中国基础设施的就业、产出和投资效应》,《管理世界》第4期。
张军、高远、傅勇、张弘,2007:《中国为什么拥有良好的基础设施?》,《经济研究》第3期。
张学良,2012:《中国交通基础设施促进区域经济增长了吗?》,《中国社会科学》第3期。
Achdou,Y.,Jiequn Han,J-M.Lasry,P-L.Lions,and B.Moll,2015,“Heterogeneous Agent Models in Continuous Time”,Working paper,Princeton University.
Aschauer,D.A.,1989,“Is Public Expenditure Productive?”,Journal of Monetary Economics,Vol.23,177-200.
Bai,Chong-En,Chang-Tai Hsieh,and Yingyi Qian,2006,“The Return to Capital in China”,Brookings Papers on Economic Activity,Vol.2,61-88.
Barro,R.J.,1990,“Government Spending in a Simple Model of Endogenous Growth”,Journal of Political Economy,Vol.98,103-125.
Bartelsman,E.,J.Haltiwanger,and S.Scarpetta,2013,“Cross-country Differences in Productivity:The Role of Allocation and Selection”,American Economic Review,Vol.103,305-334.
Baxter,M.,and R.G.King,1993,“Fiscal Policy in General Equilibrium”,American Economic Review,Vol.83,315-334.
Beck,T.,A.Demirguc-Kunt,and R.Levine,2000,“A New Database on the Structure and Development of the Financial Sector”,World Bank Economic Review,Vol.14,597-605.
Bom,P.R.D.,and J.E.Ligthart,2014,“Public Infrastructure Investment,Output Dynamics,and Balanced Budget Fiscal Rules”,Journal of Economic Dynamics and Control,Vol.40,334-354.
Brandt,L.,J.V.Biesebroeck,and Yifan Zhang,2012,“Creative Accounting or Creative Destruction?Firm-level Productivity Growth in Chinese Manufacturing”,Journal of Development Economics,Vol.97,339-351.
Buera,F.J.,J.P.Kaboski,and Y.Shin,2011,“Finance and Development:A Tale of Two Sectors”,American Economic Review,Vol.101,1964-2002.
Buera,F.J.,and Y.Shin,2013,“Financial Frictions and the Persistence of History:A Quantitative Exploration”,Journal of Political Economy,Vol.121,221-272.
Dekle,R.,and G.Vandenbroucke,2006,“A Quantitative Analysis of China's Structural Transformation”,Working Paper N0651,Institute of Economic Policy Research.
Feenstra,R.,Zhiyuan Li,and Miaojie Yu,2014,“Exports and Credit Constraints under Incomplete Information:Theory and Evidence from China”,Review of Economic and Statistics,Vol.96,729-744.
Fernald,J.G.,1999,“Roads to Prosperity?Assessing the Link between Public Capital and Productivity”,American Economic Review,Vol.89,619-638.
Fisher,W.H.,and S.J.Turnovsky,1998,“Public Investment,Congestion,and Private Capital Accumulation”,Economic Journal,Vol.108,399-413.
Hsieh,Chang-Tai,and P.J.Klenow,2009,“Misallocation and Manufacturing TFP in China and India”,Quarterly Journal of Economics,Vol.124,1403-1448.
Jia,Junxue,Jing Guo,and Zijie Wang,2015,“The Fiscal-monetary Policy Mix and Exchange Rate Stability:A Dynamic Stochastic General Equilibrium Model with Chinese Characteristics”,Emerging Markets Finance and Trade,Vol.51,729-746.
Levinsohn,J.,and A.Petrin,2003,“Estimating Production Functions Using Inputs to Control for Unobservable”,Review of Economic Studies,Vol.70,317-341.
Moll,B.,2014,“Productivity Losses from Financial Frictions:Can Self-financing Undo Capital Misallocation?”,American Economic Review,Vol.104,3186-3221.
Restuccia,D.,and R.Rogerson,2008,“Policy Distortions and Aggregate Productivity with Heterogeneous Plants”,Review of Economic Dynamics,Vol.11,707-720.
Song,Zheng,K.Storesletten,and F.Zilibotti,2011,“Growing Like China”,American Economic Review,Vol.101,202-241.
(1)大量文献实证考察了公共基础设施投资(或存量)的经济影响,如Aschauer(1989)、Fernald(1999)、张光南等(2010)和张学良(2012)。
(1)图1纵轴为1998—2006年间地级市TFP的均值(除以样本最大值),横轴为公共基础设施投资规模gI(基本建设支出与GDP比值)的均值,剔除小于1%和大于99%分位数的样本以避免异常值的影响。拟合曲线回归结果为:TFP=0.100(0.009)+0.816(0.341)gI-5.056(2.002)g2I,R2=0.02,样本数为326,小括号中数字为标准差。基本建设支出数据来自1998—2006年间《全国地市县财政统计资料》,TFP数据由企业TFP(用Levinsohn&Petrin(2003)方法估算)在地级市层面加权平均得到(中间投入比重为权重),企业数据来自1997—2007年间国家统计局规模以上工业企业调查数据。本文利用企业相关信息(如企业代码和所在地等)对样本企业进行匹配(Brandt et al.,2012),剔除了:总资产等关键变量缺失、从业人员少于10人(Feenstra et al.,2014)、总资产小于流动资产或固定资产、企业代码缺失或无法唯一识别和建立时间无效的企业,最终包含56.9万家企业。
(1)与Moll(2014)一样,本文中信贷约束为系统性的(企业家面临的杠杆率约束相同),但高生产率企业家的最优投资规模进而外部资金需求更大,故信贷约束对其影响更大。与本文不同,Song et al.(2011)先验区分了两类企业(国有和民营企业),且假定它们面临的信贷约束不同,Buera et al.(2011)则先验区分了两个部门(制造业和服务业),且假定不同部门的企业进入成本进而资金需求不同。
(2)1997-2007年间,我国规模以上工业企业的劳动生产率(劳均产出)与劳动规模的相关系数为0.55。Song et al.(2011)、Buera et al.(2011)、Buera&Shin(2013)和Moll(2014)忽略了这一典型事实,Bartelman et al.(2013)则没有考虑金融摩擦的影响。
(3)借鉴Barro(1990),本文考虑公共基础设施投资而非存量,这可简化模型,以给出整个经济TFP的解析解(见(17)式)。
(1)现实中,规模较大的企业通常具有较大能力要求政府提供较好的基础设施服务。例如,为吸引大中型企业投资办厂,我国地方政府热衷设立各种经济开发区,为这些企业提供良好的道路交通和电力通讯等服务,小企业则很难享受此类特殊待遇。
(1)由(16)式可得整个经济的有效劳均生产函数:y_t=Zt(kαt)υg1-υt,yt=Yt/(Lt-)、kt=Kt/(Lt-)和gt=Gt/(Lt-)。这与Barro(1990)的生产函数较一致,故本文由受信贷约束的异质性企业家模型加总得到类似Barro(1990)的模型。但Barro(1990)将TFP设为完全外生的,本文中企业家生产率为外生,但整个经济的TFP很大程度上为内生的,取决于企业间的要素配置。
(1)(22)式等价于连续的AR(1)过程,稳态分布服从均值为零、方差为σ2/(2θ)的正态分布,自相关系数为e-θ。此时,由Ito公式可知:(1)式中的漂移项μ(zit)=[-θlogzit+σ2/2]zit,扩散项为(σzit)2。
(2)数据来自国家统计局网站(www.stats.gov.cn)。也有研究利用全社会固定资产投资中来自国家预算的资金度量公共基础设施投资。全社会固定资产投资中来自国家预算资金与GDP的比值在1992-2014年间也为0.022,故使用这两个指标度量公共基础设施投资没有差异。
(1)总储蓄S=Y-C-G,Y为支出法核算的GDP,C为居民消费,G为政府消费(Song et al.,2011)。1992-2014年间,我国总储蓄率为44%。鉴于模型刻画的是企业家行为,而企业家储蓄率较高,故以一个较大值57%作为赋值依据。1992-2005年间,我国的资本回报率为21.8%但呈下降趋势(Bai et al.,2006),故以一个较小值17.4%作为赋值依据。我们也尝试(61.6%,15.3%)和(53%,18.5%)两组储蓄率和资本回报率为赋值依据,基准结论很稳健。
(2)为使数值有意义,对TFP和总产出进行标准化处理,即除以参照经济的TFP和总产出。参照经济为不存在存贷利差和信贷约束(ξ=0;为避免λ=∞带来的计算困难,以λ=10为不存在信贷约束的近似赋值),以及公共基础设施投资为最优规模的经济。公共基础设施投资的Samuelson效率条件为:Yt/G=1,故最优g*I=(1-υ)/[Z1-t YtZGZt/G=0,故t]。TFP最大化要求:TFP最大化的gI=1-υ(见图2(a)),但无法从理论上给出g*I。模拟结果表明:ξ=0和λ=10时,gI为0.25的产出最大,而TFP与gI=0.17时的几乎一样,故以0.25作为参照经济gI的赋值。=0时,企业劳动生产率与劳动规模不存在相关性,与经验事实不符,故未将=0的经济作为参照经济。
(1)我们也尝试改变其他参数(如ρ、1-υ、θ和σ等),在较合理的取值范围内,基准结论具有良好的稳健性。
(2)Bartelsman et al.(2013)发现Clpl随间接劳动增加而先增后降,给出相似解释。Hsieh&Klenow(2009)忽略了企业退出以避免此问题。
(1)这一结论有助于更好地认识2008年世界金融危机期间我国“四万亿元投资计划”的影响。此间,非常宽松的货币信贷政策极大放松了企业的信贷约束(λ明显增大),因此以公共基础设施投资为主要措施的“四万亿元投资计划”促进了经济的快速恢复。但就图4(a)和(b)的模拟结果来看:这可能主要表现为总量扩张(产出增加)而非效率改善(TFP提升)。
(1)本文其他部分,固定,Clpl的增加源于要素更多流向高生产率企业,即配置效率的提升。图6(d)考察的是变化的影响,此时Clpl增加并不代表配置效率提升。也引入了内生企业退出机制:模拟结果表明,企业退出率Ext在=0时为零且随增加而增加。Bartelsman et al.(2013)没有考虑信贷约束,故他们的模型中对Ext的影响很大,致使Clpl随小幅增加而先增后降。本文中信贷约束的存在使得对Ext的影响较小:不存在最优投资规模限制延缓了低生产率非信贷受限企业的退出,这更符合经济现实。模拟结果表明,λ越大,对Ext的影响越大,Clpl越倾向于随增加而先增后降,即Bartelsman et al.(2013)的结论是本文的一个特例。km
(2)由(αυ8)式可知:当zit<-z时,有it zitz km=,即非信贷受限企业的资本投资关于生产率的弹性随γ减小而增大。it(1-υ)γ
(1)一些研究表明,企业面临的特异性政策差异会导致要素错配和TFP损失(Resuccia&Rogerson,2008;Bartelsman et al.,2013)。但这些研究主要通过类似差异性税收的方式引入特异性政策差异,忽略了一些政策差异(如本文考虑的基础设施服务差异)对企业家的正向激励,得到这样的结论较为直观。本文可视为这些研究的有益补充,有利于更全面地认识特异性政策差异对企业间要素配置的影响。
(2)其他参数仍采用基准值,只将调整为0.301以使模型关于企业劳动生产率与劳动规模相关系数的预测值与现实数据相匹配。
(3)本文也尝试将η由0.5增加为2。此时,2mit/cgc<0,即公共消费支出的引入将降低私人消费的边际效用、减少企业家消费,但总体仍会抑制企业家财富积累(因为税率增加对企业家财富积累的抑制更突出),故结论不变。
(1)本文中的消费税类似现实经济中的商品服务课税,而非对特定消费(如烟酒)征收的“消费税”。这一理论拓展也是基于对中国经济的现实考量:商品服务课税和企业所得税在我国财政收入体系中占据重要地位:1992-2014年间,企业所得税占税收和财政总收入的比重分别为16.4%和14.9%,国内增值税、国内消费税和营业税三者占税收和财政总收入的比重分别为55.9%和51.2%。为确保τc很大时效用函数的凹性,采用对数效用函数(η=1)。2000-2008年间规模以上工业企业所得税有效平均税率为0.2(贾俊雪,2014),故将τf基准值设为0.2(也尝试0.33或0.25的法定税率,结论一致),可得τc基准值为-0.101。模型关于储蓄率、企业劳动生产率与劳动规模相关系数的预测值与现实数据相匹配,但资本租金率较低。