技术冲击和劳动供给对经济波动的影响分析--基于可分劳动RBC模型的实证检验

文章根据国内外经济波动的不同特征,构建了用于研究中国经济波动的可分劳动RBC模型,并对改革开放以来的中国经济进行了实证检验,从供给角度考察技术冲击对中国经济波动的影响,并在RBC模型框架下分析了技术进步对中国劳动市场的影响。研究发现,一是在固定劳动和可分劳动RBC模型中,技术冲击可以解释中国经济波动的主要部分;二是可分劳动RBC模型的实证结果表明,劳动供给变动对经济波动的影响较小,技术进步对改革后的产出、居民消费和就业都产生了正向冲击效应。技术进步对就业增长效应较小,致使我国的劳动需求增长率明显小于劳动供给增长率,劳动市场供需严重不平衡,这也是导致就业波动较为平缓、失业问题日趋严峻的一个重要原因。     

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

本文采用随机动态一般均衡方法,将政府支出作为外生随机冲击变量,构建中国三部门实际经济周期(RBC)模型,并对改革后的中国经济进行了实证检验,从而考察中国宏观经济波动的周期特征及财政政策的效应问题。研究发现,在包含政府部门的RBC模型中,技术冲击和政府支出冲击可以解释70%以上的中国经济波动特征,中国经济波动是技术因素、供给因素和需求因素综合影响的共同产物。此外,本文证实改革后政府支出对居民消费产生了一定的挤出效应。     

家庭消费行为变迁、经济波动与居民消费率——基于Bayes估计的DSGE模型

在动态随机一般均衡(DSGE)模型的理论框架下,结合转型期中国经济的特征事实,引入不确定性、信贷约束以及习惯形成等居民消费特征,采用贝叶斯(Bayes)估计方法分析1978~2012年间信贷约束与消费习惯形成特征对我国宏观经济波动和居民消费率的影响机制。研究结果表明,模型能够分别解释实际产出、消费、投资和就业波动的94.1%、84.5%、50.7%和74.5%;同时探讨家庭消费行为变迁对居民消费率的影响,研究发现,短期内收入冲击与偏好冲击对居民消费率均有刺激作用,但长期内两者对消费率的影响有较大差异,即收入冲击的正向作用有明显的持续性,而偏好冲击使居民消费率更加萎缩;同时,习惯形成特征有效平滑了居民消费,有效减弱了不确定性对居民消费率和储蓄率的影响。     

中国的实际经济周期与税收政策效应

本文构建了财政税收的实际经济周期(RBC)模型,实证研究了中国宏观经济波动的周期特征及税收政策的经济效应。研究发现:(1)引入财政收支冲击的RBC模型能够解释70%以上中国经济波动的特征事实,政府支出冲击加剧中国实体经济波动,而税收冲击对经济波动的影响不显著。(2)降低劳动收入和资本收入的税率都能促进经济增长和带动资本和劳动的供给增加,降低劳动收入税率有利于促进居民消费增长,降低资本收入税率则起到抑制作用,调整劳动收入税率的政策效果更强。(3)资本收入税率与社会福利呈正相关和非对称性,劳动收入税率与社会福利呈负相关和非对称性,技术冲击和财政冲击的共同作用使得结构性税收调整政策的福利效应具有非对称性,两种税率的同向变动对社会福利具有放大作用,两者的反向变动对社会福利具有削弱作用。     

资本市场的市场化与中国的实际经济周期

本文在RBC模型框架下引入垄断竞争,来解释宏观经济的周期波动和考察市场化进程冲击对经济波动的影响,并利用中国改革开放以来的经济数据,对该模型进行了实证检验。研究发现该模型能够更好地解释资本市场的波动特征,更好地解释各变量与劳动生产率之间的协动关系;能够合理地解释中国居民消费波动大于产出波动的特征事实,说明该模型对中国经济的解释力比传统的RBC模型更好。另外,还发现市场化进程冲击是中国经济波动的重要冲击来源,可以解释约23%的产出波动,53%的消费波动,33%的就业与投资波动,40%的资本存量与劳动生产率波动,说明其是中国宏观经济波动不可忽略的重要影响因素。     

趋势性冲击、政府消费与中国经济周期波动——基于开放经济RBC模型的比较分析

趋势冲击是指对于技术增长率的冲击,即技术增长率服从一个随机过程。趋势冲击在新兴市场国家经济周期波动中起着非常重要的作用,由于中国的经济波动特征更为接近新兴市场,因此趋势冲击可能是驱动中国周期波动的重要因素。本文首先构建了一个包含趋势性冲击和暂时性冲击的开放经济RBC模型,模型较好地拟合了中国的大部分经济波动特征;包含政府消费和趋势性冲击的开放经济模型则出现了过度拟合,拟合结果较差。这可能是因为两种冲击存在"叠加效应"。     

政府支出的外部性和中国政府支出的宏观效应:动态随机一般均衡视角

基于1978-2007年的中国宏观数据和VAR模型的研究发现一些经验事实:总产量、社会投资以及居民消费对政府支出正向冲击反应为正。由于中国仍是一个发展中国家,政府支出的正外部性可能是解释这些经验事实的重要因素,于是,本文主要在动态随机一般均衡分析框架下对此问题加以深入探讨,尝试将消费性和生产性的政府支出分别引入效用函数和生产函数,构建模型、经济模拟,以解释中国总产量、居民消费、社会投资对政府支出冲击响应的经验事实。分析发现所构建的模型能较好地拟合中国宏观经济运行的一些特征。     

中国技术进步技能偏向性水平:1979—2010

当前技术进步朝技能劳动方向发展已形成共识,但技术进步技能偏向强度的定量研究还相对匮乏。本文设定双层嵌套型CES生产函数,构建技术进步技能偏向性指数,采用标准化系统的贝叶斯方法和FGNLS方法测算我国1979-2010年技术进步技能偏向性水平,考察技术进步以多大强度偏向于提高技能劳动生产率,并通过回归分析考察技术进步偏向性成因。结果发现:(1)1979-2010年技术进步明显偏向于技能劳动,但其变化强度逐年减弱。(2)技术进步技能偏向性影响因素的实证检验结果显示,无论是中性技术进步还是资本体现式技术进步都引致技术进步偏向于技能劳动,人力资本投资对技术进步技能偏向性起促进作用,但市场化制度对技能和非技能劳动的激励作用相同即呈现技能无偏特征。     

随机货币冲击下的中国经济波动研究——基于RBC模型

中国居民消费存在明显的现金先行约束.在真实经济周期(RBC)模型中通过现金先行约束引入货币冲击符合中国经济运行特征.笔者在真实经济周期理论框架内分析了外生货币冲击对中国经济波动产生的影响.研究表明,货币发行冲击主要影响消费、投资和通胀率的波动,而其他主要经济变量(产出、就业等)的波幅基本不变.     

消费攀比、劳动调整成本与中国经济波动

引入消费攀比与劳动调整成本两种内生机制于动态随机一般均衡模型,显著改善了模型对于中国经济动态特征的匹配性,它能够增强消费的持久性动态特征,而且准确地复制出了实际经济对中性技术冲击的“驼峰”反应动态以及劳动力市场的反周期动态.进一步,我们定量分析了中国经济波动的来源问题,发现中性技术冲击和投资冲击对中长期波动的显著解释力.     

Aggregate labor market fluctuations under news shocks

The standard RBC model fails to replicate the relationship between aggregate hours worked and average productivity. We propose a DSGE model that incorporates habit formation preferences, capital adjustment costs, and news shocks to solve the puzzle implied in the standard RBC model with only technological shocks. The aggregate labor supply curve is shifted due to the wealth effect caused by the variation of consumption under a news shock. Moreover, capital adjustment costs help amplify the variation of consumption, and thus the movement of the aggregate labor supply curve under the news shock. Also, the aggregate demand curve will be shifted, as it operates in the standard RBC model after the realization of the news shock. As a result of the joint movement of the aggregate labor supply curve and aggregate labor demand curve under the news shock, the model achieves a relationship quite close to the empirically observed relationship.     

Learning-by-doing or habit formation?

In a recent paper, Chang et al. [2002. Learning-by-doing as a propagation mechanism. American Economic Review 92 (5) 1498–1520] extend the standard real business cycle (RBC) model to allow for a learning-by-doing (LBD) mechanism whereby current labor supply affects future productivity. They show that this feature magnifies the propagation of shocks and improves the matching performance of the standard RBC model. In this paper, we show that the LBD model is nearly observationally equivalent to an RBC model with habit formation in labor (or, equivalently, in leisure). Under the same calibration of the parameters, the two models share the same equilibrium paths of output, consumption, and investment, but have different implications for hours worked. Using Bayesian techniques, we investigate which of the LBD and Habit models fits the US data better. Our results suggest that the Habit specification is more strongly supported by the data.     

Can Habit Formation Be Reconciled with Business Cycle Facts?

Many asset pricing puzzles can be explained when habit formation is added to standard preferences. We show that utility functions with a habit then gives rise to a puzzle of consumption volatility in place of the asset pricing puzzles when agents can choose consumption and labor optimally in response to more fundamental shocks. We show that the consumption reaction to technology shocks is too small by an order of magnitude when a utility includes a consumption habit. Moreover, once a habit in leisure is included, labor input is counterfactually smooth over the cycle. In the case of habits in both consumption and leisure, labor input is even countercyclical. Consumption continues to be too smooth. Journal of Economic Literature Classification Numbers: E13, E21, E32.     

Why have business cycle fluctuations become less volatile?

This paper shows that a standard Real Business Cycle model driven by productivity shocks can successfully account for the 50% decline in cyclical volatility of output, its components, and labor input that has occurred since 1983. The model is successful because the volatility of productivity shocks has also declined significantly over the same time period. We then investigate whether the decline in the volatility of the Solow Residual is due to changes in the volatility of some other shock operating through a channel that is absent in the standard model. We therefore develop a model with variable capacity and labor utilization. We investigate whether government spending shocks, shocks that affect the household’s first order condition for labor, and shocks that affect the household’s first order condition for saving can plausibly account for the change in TFP volatility and in the volatility of output, its components, and labor. We find that none of these shocks are able to do this. This suggests that successfully accounting for the post-1983 decline in business cycle volatility requires a change in the volatility of a productivity-like shock operating within a standard growth model. We thank Stephen Parente, Ed Prescott, John Taylor, and two anonymous referees for helpful comments and suggestions.     

Labor's share fluctuations, biased technical change, and the business cycle

We extend the basic RBC model to allow for biased technical changes. One broad definition of biased technical changes is changes that directly affect factor elasticities. Given the link between changes in factor elasticities and factor shares, observed fluctuations in US labor's share are motivation for this study. We find that when the technology shock process is calibrated according to US labor's share dynamics, 93 percent of US GDP volatility is accounted for. The observed countercyclical nature of labor's share is accounted for, although the model correlation is too high. As well, the model exhibits business cycles that are qualitatively similar to those of the standard model with neutral technology shocks. These findings, while robust to the short-run properties of various measures of labor's share, are sensitive to the average labor's share used in calibration, e.g., departing from a baseline calibration value of 63 percent, for steady-state labor's shares of 50 percent and 70 percent the model accounts for 107 percent and 84 percent of US GDP volatility respectively.     

Demand shocks and economic fluctuations

This paper studies conditions under which demand-side shocks can generate realistic business cycles in RBC models. Although highly persistent demand shocks are necessary for generating procyclical investment, variable capacity utilization and habit formation can reduce the required degree of persistence.     

BUSINESS CYCLES IN THE EQUILIBRIUM MODEL OF LABOR MARKET SEARCH AND SELF‐INSURANCE*

I introduce risk‐aversion, labor‐leisure choice, capital, individual productivity shocks, and market incompleteness to the standard model of labor search and matching and investigate the model’s cyclical properties. I find that the model can generate the observed large volatility of unemployment and vacancies with a reasonable replacement rate of unemployment insurance benefits of 64%. Labor‐leisure choice plays a crucial role through additional utility from leisure when unemployed and further amplification from adjustments of hours worked. On the other hand, the borrowing constraint or individual productivity shocks do not significantly affect the cyclical properties of unemployment and vacancies.     

Labor adjustment costs and complex eigenvalues

Aggregate variables display both persistence and damped oscillations in response to temporary external shocks. The standard real business cycles (RBC) model cannot explain these patterns, because its stable eigenvalues are positive and real. We demonstrate that this model with labor adjustment costs can yield complex eigenvalues. However, numerical experiments suggest that the model cannot display distinguishable damped oscillations of aggregate variables.     

News shocks and labour market dynamics in matching models

We enrich a baseline real business cycle (RBC) model with search and matching frictions on the labour market and real frictions that are helpful in accounting for the response of macroeconomic aggregates to shocks. The analysis allows shocks to have an unanticipated and a news (i.e., anticipated) component. The Bayesian estimation of the model reveals that the model that includes news shocks on macroeconomic aggregates produces a remarkable fit of the data. News shocks in stationary and non‐stationary TFP, investment‐specific productivity and preference shocks significantly affect labour market variables and explain a sizeable fraction of macroeconomic fluctuations at medium‐ and long‐run horizons. Historically, news shocks have played a relevant role for output, but they have had a limited influence on unemployment.