Hi:欢迎来到高赢论文网!     

所有论文科目分类


主页 > 经济论文 > 万科房地产公司的盈余管理研究

万科房地产公司的盈余管理研究

作者2021-06-08 14:40未知
 

摘  要
盈余管理是近年来我国上市公司普遍存在的问题。虽然企业通过盈余管理可以在短期内满足自身的目的。但是,从公司的长期发展的角度来看,这不是很理想。过度的收入管理不仅会导致会计信息变化,还会使投资者做出较差的投资决策,从而在经济市场上不合理地分配资源。在这种情况下,上市公司盈余管理已成为理论和实践研究的重要课题。对于房地产公司而言,适当的收入管理可以向外界发出积极的信号,降低资本成本,避免业务风险,并帮助公司实现和完善长期运营。
本文以万科集团为研究对象,在相关领域国内外研究结论和选题背景作为切入点,运用相关理论和方法万科房地产企业的行为模式、管理手段和管理结果收益管理的研究,以及其他各方面,并提出相关措施根据收益管理的现状和存在的问题,对规范我国房地产上市企业的盈余管理行为具有一定的借鉴意义。
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
关键词:上市公司;房地产;盈余管理


Abstract
Earnings management is a common problem of listed companies in recent years. Although enterprises can meet their own goals in the short term through earnings management. However, from the perspective of the long-term development of the company, this is not ideal. Excessive income management will not only lead to changes in accounting information, but also make investors make poor investment decisions, thus unreasonable allocation of resources in the economic market. In this case, earnings management of listed companies has become an important subject of theoretical and practical research. For real estate companies, proper revenue management can send a positive signal to the outside world, reduce the cost of capital, avoid business risks, and help the company to achieve and improve long-term operations.
In this paper, vanke group, as the research object, in the related field at home and abroad research conclusion and the selected topic background as the breakthrough point, using relevant theories and methods of vanke real estate enterprise behavior patterns, management means and management research results of revenue management, and other various aspects, and puts forward relevant measures according to the current situation and existing problems of earnings management, to regulate the behavior of earnings management of listed companies of real estate in our country has a certain reference significance.
 
 
 
 
 
 
 
 
 
 
Key words:Listed company; Real estate; Earnings management

目  录
第1章 绪论 1
1.1研究背景及研究意义 1
1.1.1研究背景 1
1.1.2研究意义 1
1.2国内外文献综述 1
1.2.1国内文献综述 1
1.2.2国外文献综述 2
1.3研究方法与研究思路 2
1.3.1研究思路 2
1.3.2研究方法 2
第2章 盈余管理基本理论概述 3
2.1盈余管理概念的界定 3
2.2盈余管理的计量模型 3
2.2.1总体应计利润法 3
2.2.2具体项目法 3
2.2.3盈余频率分布模型 3
2.3盈余管理产生的动机 4
2.3.1资本市场动机 4
2.3.2节税动机 4
2.3.3政治动机 4
2.4盈余管理的手段 4
2.4.1利用非经常性损益 4
2.4.2利用公允价值计量 5
2.4.3利用资产减值准备 5
第3章 万科集团盈余管理行为分析 6
3.1万科集团企业概况 6
3.2万科集团盈余管理行为存在性分析 6
3.2.1初步判断分析 6
3.2.2检验模型选择分析 7
3.2.3模型检验与结果分析 8
3.3万科集团盈余管理产生的动机 8
3.3.1资本主义市场动机 8
3.3.2节税动机 9
3.3.3政治动机 10
3.4万科集团盈余管理手段分析 10
3.4.1利用非经常性损益 10
3.4.2利用公允价值计量 10
3.4.3利用资产减值准备 11
第4章 万科集团盈余管理结果评价分析 12
4.1财务绩效指标分析 12
4.1.1财务绩效定量评价指标 12
4.1.2管理绩效定性评价指标 13
4.2市场绩效指标分析 13
4.2.1托宾Q值 13
4.2.2市盈率 14
第5章 加强企业盈余管理的措施 16
5.1加强财务风险意识 16
5.2保障偿债能力 16
5.3尝试海外融资 16
5.4锁定市场需求 16
第6章 结论 17
参考文献 18
致谢 19
附录一 中文译文 20
附录二 外文原文 22
 

第1章 绪论

1.1研究背景及研究意义

1.1.1研究背景

城市化已成为未来中国甚至国际上经济社会发展的必由之路,为了加速我国发展城市化,房地产产业已成为行业中不可或缺的重要力量。对于房地产行业,收入信息对于许多投资者而言变得尤为重要和敏感。在经济统计数据中,我国第三产业中房地产业的增加值从2010至2018年间平均水平为12%,而房地产行业上市公司的净利润总额已达到4485300万元,在深交所的A股流通股数平均市盈率达到35%。由于近年来房地产行业的持续发展,它已成为高利润行业。由于需要进行战略规划,上市公司所披露的会计信息的质量将下降,监管体系的成熟将无法承受股市的发展。企业的会计盈余信息关系着投资者的决策、股票市价的走势、企业内部契约的签订等方面,在得到企业内部管理者重视的同时也受到了企业外部信息使用者的高度关注。因为企业希望通过披露的会计信息向外界传递某种信号,所以会计信息的信息导向作用会被人为的调控,企业盈余管理在此看来尤为必要。

1.1.2研究意义

作为一个蒸蒸日上的朝阳行业,对于很多投资者而言,它的收益信息显得尤为重要和敏感。因此,对这样一个持续不断发展的高收益行业的特点进行分析,以及对其盈余管理行为和手段的研究,成为了我国当前经济发展的一个重要方面。通过查看及分析其公开披露的财务报告是投资者获取上市公司财务状况的重要手段,然而财务报告所公布的盈余质量及其背后又隐藏的盈余管理却值得我们进一步探索。

1.2国内外文献综述

1.2.1国内文献综述

鲍辰璐(2016)将之前有关盈余管理的资料进行翻阅整理,得出“发行价格的提高、配股资格的获得、避免面临退市风险”这三点盈余管理的动机。徐敏霞(2017) 也同样觉得企业进行盈余管理使资金转亏为盈,而且通过探索发现上市公司在公司经营业绩较差,盈利能力预计比往年要大幅度减弱的时候,存在盈余管理动机。刘晓玢(2017) 认为上市公司普遍利用关联交易进行盈余管理,且关联交易度越高,盈余管理的行为越明显。盈余管理现象是多种因素引起的综合性结果,且盈余管理并不能毫无破绽地粉饰报表,审计工作仍能有迹可循。桂良军等(2016)以国有持股比例分析企业盈余管理动机强弱,分别以应计和真实盈余管理考察盈余管理手段的差别。

1.2.2国外文献综述

美国会计师威廉·斯科特在《财务会计理论》一书中认为,作为会计政策选择的经济后果具体标志的收入管理,如果公司管理人能够随意选择不一样的会计方针,他们就可以选择一种能够最大化企业效用或最大化企业市值的会计政策。
美国著名会计学者凯瑟琳●雪珀( Kathehne.Schipper)提出的收入管理概念是有用的信息,它基于广泛的基于会计数字的“开放式管理”概念,获得私人利润并有意管理外部财务报告。因此,损益表中盈余数字的管理,以及资产负债表管理和其他支持信息属于广泛的收入管理的范畴,其中的其他信息主要包括财务报告中的注释及其他。
保罗(Paul M.Healy)和詹姆斯(JamesM.Wahlen)指出,盈余管理是企业在生产经营过程中,在会计准则允许的情况下对企业的财务报表进行粉饰,展现更好的财务状况,使投资者对企业的经营状况满意,从而影响投资者的决策,有时候也会通过有关会计规划的选择和会计预估的办法,管理企业的会计盈余。

1.3研究方法与研究思路

1.3.1研究思路

针对企业进行盈余管理日益增多的现象,提出选题的背景及意义,查阅相关盈余管理的国内外文献,选定研究方法与研究对象。阐述盈余管理的基本理论,对计量方法、手段、产生的动机进行归纳总结。通过分析万科集团盈余管理,重点分析2010年—2018年年报数据,从中找出盈利管理的所带来的利益点,证明盈余管理应在企业中存在的必要性。最后,对案例进行总结,并从实际情况出发对万科集团企业管理提出意见与建议。

1.3.2研究方法

关于盈余管理与企业治理的关系,国内外学者大多采用实证研究法,在已有理论依据的基础上对特选的研究对象进行研究分析。本文在前人研究的基础上,采用案例研究法,以在房地产企业中口碑较好的万科房地产企业为案例研究,并通过具体分析、采集万科集团2010年—2018年的年报数据,证明了万科公司盈余管理行为存在的必要性,验证了有效的盈余管理对企业经营管理发展的重要性,深入研究盈余管理利益点背后存在的相关问题,并提出完善企业治理结构的相关建议与措施。


第2章 盈余管理基本理论概述

2.1盈余管理概念的界定

盈余管理是管理机构通过在会计准则允许的范围内处理会计信息来最大化自身利润的行为。公司治理主体是收入管理的主体,财务报告是通过利用会计绩效的可选绩效会计标准本身中的漏洞实现的目标。尽管公司的利润根据会计期间而变化,但公司整个利润周期的实际利润水平不会改变,最终导致决策或判断中的错误。

2.2盈余管理的计量模型

2.2.1总体应计利润法

更改交易时间和累积项目数是公司使用累积收益进行利润管理的一种常用方法。改变交易时间是指提前或推迟确认收入的时间;更改累计项目的数量是指使用会计估计影响公司的会计盈余。例如,一家上市公司可能会在年末列出收入,并在第二年或适当时候代表另一方的回报或会计错误调整收入,以累积项目折旧。从上一年开始累积。

2.2.2具体项目法

特定项目方法具有高度特定性,尤其是对于易于操作的会计帐户的研究。可以建立一个特定的度量模型来检测利润管理和利润管理行为的存在。此方法的优点是更直观,可以针对某些特定问题进行专门分析。这种方法通常用于研究某个或某些特定的行业,具体项目法的特点是目的明确,检测对象针对性强,因此检测模型没有一样的衡量尺度,通常是根据行业背景及检测目标设定,因此该方法研究往往局限于小样本或具体的行业和部门,因此研究结果难以推广。

2.2.3盈余频率分布模型

Burgstahler and Dichev(1997).F.J. Patel, and R.Zeckhauser(1999)等学者提出测试报告的收入分配,以查看是否有新的方式来管理收入。此方法假定未管理的盈余大致呈正态分布,并且密度函数平滑。如果管理人员正在执行盈余管理,则会反映所选样本的利润分配函数图。此方法可以测试收入管理的存在而无需操纵所产生的利润,并且仅通过分配功能的中断来判断收入管理的行为。

2.3盈余管理产生的动机

2.3.1资本市场动机

股票发行动机,首次发行股票或债券的企业会通过盈余管理使招股说明书所披露的经营指标最大化,吸引投资者,从而使公司股票或债券达到理想的价位。在上市收购中,公司的价值是根据联合股价和盈余资本法确定的。因此,收购前经理有动机低估盈余,因此目标是降低收购价格。在公司合并和收购过程中,收购方公司的股票价值与购买目标公司所需的股票成反比。

2.3.2节税动机

国家向企业收取的税费是企业一项很重要的成本支出,一般情况下企业所得税为净利润的25%。想要减少企业所得税的缴纳就要降低企业净利润的数值,但是净利润的降低必然会向市场释放出企业经营状况滑落的信号,企业管理层需要深思熟虑的是如何在两者之间抉择取舍。合理避税可以减轻企业运营负担,也可以为企业准备出更多的发展资金,从而提升营运能力和盈利能力。我国税收体制不完善,制度规定存在一定的空白区域和时间上的滞后性。我国的会计体系还是不够完善的,财务方向和税务方向无论是学校授课还是实际操作都是区分开来的,两者独立运行缺少必要的交叉地带,沟通不畅会使得信息传递出现问题。这些都为企业在税负这个点进行盈余管理提供了可能的空间。

2.3.3政治动机

随着近年来中国房地产市场的持续升温,房价持续上涨,房地产市场投资持续增长。在这种情况下,国家在市场中起主导作用,而政府对市场经济具有宏观控制权。政府新政策的出台加强了对房地产业的监管,在一定程度上抑制了中国房地产市场价格和利润的过度增长。驱动市场的主要因素是房地产公司的利益。为了不通过政府发布的新政策为企业提供政治成本,房地产行业通过合理管理公司的会计盈余来实现降低公司利润和避免政治成本的目标。

2.4盈余管理的手段

2.4.1利用非经常性损益

非经常性损益是由于公司的性质,数量或发生频率而影响公司正常获利能力的收入和支出,尽管公司与业务运营没有直接关系,与业务运营也没有关系。非经常性损益通常反映在损益表中,具体显示为“以前的收入计算”、“营业外利润”、“其他营业利润”等。一般情况下,只有那些不经常发生的交易或事项才可以界定为非经常性损益。比如说出售固定资产、转让公司股权、进行债务重组等。

2.4.2利用公允价值计量

根据中国现行会计准则的有关规定,公司必须从市场上获得相似或相似的房地产价格及其他相关信息。公司需要从公开市场报价中获得公允价值,并且如果无法从公开市场中获得类似项目的市场价格,则可以使用公允技术来设置公允价值。尽管到目前为止该机制正在改善,但是还不完善,中国的市场经济已经发展,但是房地产行业的市场仍处于不成熟阶段,在这种情况下,公司无法轻易获得市场的公开估计。由于该行业存在广泛的猜测,因此很难获得公开报价。

2.4.3利用资产减值准备

为了按照新会计准则中的审慎原则使资产的价值和损益更加可靠和准确,公司必须单独发布八项减值准备:坏账准备,短期投资减少准备金存货资产减值准备,长期投资减值准备,固定资产减值准备,无形资产减值准备等。由于属于该内容,且会计准则中没有专门规定,因此公司在独立确定各项减值准备负债的比例和方法后,收入管理行为很可能满足各项减值准备的规定。中国证券监督管理委员会引入了一个相关制度,明确规定了损害赔偿准备的一些要求,例如,上市公司不能随意改变损害赔偿的计算方法和准备率,而在撤回巨额赔偿金后可以大笔收回不行尽管银监会的政策已在一定程度上减少了公司内部收益管理的损益准备金规定,但中国证监会并未对此通知进行处罚,从而准确地提供了公司的收益管理空间。

第3章 万科集团盈余管理行为分析

3.1万科集团企业概况

万科集团1991年在我国上市,年营业收入复合增长率约27.35%,其复合净利润增长率约29.78%,在中国A股上市公司行列中万科集团于连续盈利年限、连续现金分红年限上均排在首位。2019年,万科将专注于主渠道,深入了解经营方面基础,在稳步推进城乡建设与生活服务商战略的同时,巩固其核心业务,致力于改善关键能力,加强管理,奠定创造卓越业绩、实现高质量发展的坚实基础。
2019年,万科继续入选恒生可持续发展基准指数,获评 MSCI—ESG 评级BB级,首次参评全球房地产可持续发展指数并获得亚太住宅家庭用房开发商第一名。为了进一步提升可持续发展管理水平,2019年我们对公司的社会责任管理架构进行了全面梳理和完善,成立了由董事会牵头组成的三级管理架构,持续提升公司各项可持续发展工作。

3.2万科集团盈余管理行为存在性分析

3.2.1初步判断分析

通过现金流量和会计收益的比较分析,很容易发现公司的收益管理行为。收入具有虚拟性,为了进行更详细的分析,可以将经营活动,投资活动和融资活动产生的净现金流量分别与主要净利润,商业收益和投资利润进行比较,以确定公司的盈余的质量。
表3-1 万科集团利润获利能力和现金获利水平对比
权责发生制 收付实现制
报告期 2016 2017 2018 报告期 2016 2017 2018
项目 比率值 项目 比率值
营业利润/营业收入 30.60% 13.75% 13.47% 经营现金净流量/营业收入 -1.17% -9.36% -45.37%
营业利润/总资产 6.53% 2.16% 2.10% 经营现金净流量/总资产 -0.25% -1.47% -7.07
营业利润/总股份 48.45% 20.24% 18.51% 经营现金净流量/总股份 -1.85% -13.77% -62.35%
 
(数据来源:万科集团官网年度报告)
 
通过表3-1可以看出企业在权责发生制下计算出来的2016-2018年企业获利能力高于对应的现金获取能力,两者有一定的差距;但是2016年的数据显示,二者的差距极为明显。在权责发生制下计算出来的获利能力指标数值均为正值,对应的现金获取能力指标为负值,万科企业现金净流量与盈利的关联性小,盈余质量低,企业很可能存在盈余管理行为。

3.2.2检验模型选择分析

以上介绍了三种典型的测量模型,其中频率分布检测方法使用利润分布直方图中每个利润区间中包含的公司数等于平均值相邻两个利润区间中的观察值,而包含在该区间中的公司数和实际观察数,则两者之间的差除以该区间的方差值就成为统计检验的参数,因此,在检查阈值参数和其他参数是否具有不同的特征。此方法不适用于测试单个公司的利润管理行为,因此无法测试该公司的利润管理行为。本文不使用盈余频率分布检测方法分析财务状况。
具体项目法针对于存在较大范围的可以操纵的应计利润空间所存在的特定项目,采用该种分析方式要依靠分析人员能够对企业正在滚动实行的羡慕进展有详尽的了解。本文对于万科集团的研究数据大多来自万科集团的年报,无法获取详尽的内部项目信息,所以本文也未采用具体项目法。
关于检验万科集团是否有盈余管理行为,综合考虑下使用总体应计利润法从可行性和操作性上分析更强。虽然总体应计利润模型的使用前提和房地产行业市场环境有一定的偏差,但是这种偏差对于本文的研究是属于可容许的范围。在本文中,万科集团通过建立总利润模型和平均利润累计模型来进行自我测试。如果结果相同,则可以解释为本文中的假设是正确的,并且验证万科集团具有盈余管理行为。

3.2.3模型检验与结果分析

由万科2010-2018年的净利润和经营活动现金流量计算出每年总体应计利润,如下表3-2所示,由此可计算出2010年至2015年的应计利润均值为188453.16 万元。则由此可推算出NDA2016。则万科集团2012年年非操控性应计利润为:
NDA2016= A2015X TA1-6 + A2014= 233854.20 (万元)
则DA2016 = TA2016 - NDA2016=11171.21 (万元)
2016年非操控性应计利润高达11171.21万元,可以推测万科集团实行盈余管理活动的可能性是非常大的。在此需要说明的是,由于模型构建的假设基础是万科集团经营现金流量未被企业管理层操纵,所以模型验证过程中会存在一定的误差。对于本文的研究而言,这些误差属于可接受误差。通过上述模型检验,证明万科集团从2016年开始施行明显的盈余管理行为。
表3-2 万科集团2010年—2018年总体应计利润表
年份 净利润 经营活动现金流量 期末总资产 总应计利润
2010 23954.01 -88023.13 701957.63 111977.14
2011 43396.26 -43973.16 1029370.25 87369.42
2012 114231.00 -131870.19 1383307.96 246101.19
2013 122623.00 251903.46 2204769.94 -129280.46
2014 173423.13 -240548.19 2845310.60 413971.32
2015 145610.00 -254970.37 3530786.26 400580.37
2016 233447.99 -11577.42 4645959.98 245025.41
2017 97454.18 -86163.25 5857925.13 183617.43
2018 114600.44 -500843.25 7084923.32 615443.69
 
(数据来源:万科集团官网年度报告)

3.3万科集团盈余管理产生的动机

3.3.1资本主义市场动机

表3-3 万科集团2013年—2018年盈余公积与每股收益
年份 盈余公积(单位:亿元) 每股收益(单位:元)
2013 87.38 0.48
2014 105.88 0.66
2015 136.49 0.88
2016 144.37 1.02
2017 168.06 1.34
2018 173.30 1.45
 
(数据来源:万科集团官网年度报告)
2017年8月22日,万科集团发布了关于增发A股的A股指南,并开始公开发行增发的A股。本次发行采用A股优先认购方式,优先认购其余网上价格,网下发行价格为31.53元。如果上市公司在最近三个会计年度中一直保持盈利,则该财务报表尚未发布或无法发表意见,同时还有一份带有保留意见和负面评论的特许会计师审计报告,并且该公司的资产质量良好且现金充裕资金流是正常的,并且公司必须在过去三年中使用现金或股票,平均分配收入大于可分配收入的30%。此外,公司的财务状况也对股票市场的动荡产生了重大影响。下图显示了万科集团2013年至2018年的盈余储备和每股收益。在此图中,可以看到两者之间存在正相关。因此,为了提高企业的股票价格,公司也应进行盈余管理。

3.3.2节税动机

房地产行业必须在项目完成后进行结转并计算成本,而未完成的项目无法合理地结算土地增值税,因此可以合理地避免此环节上的税收。万科集团在其2017年年度报告中表示:2017年,公司的新建筑面积同比增长55.2%至777万平方米,竣工面积增加4,600万至4.53亿平方米,目标下降了25.8%。2017年,新建房屋占翻新房屋的53.4%。在宏观调控政策下,一些城市改善了完成项目申报的条件,推迟了项目的完成和验收,这些措施推迟了公司的项目完成。由于房地产开发过程中缺乏装饰策略,公司无法准确预测项目进度,因此一些公司的项目未能按时完成,他们按时出售装潢精美的房屋,并成立了一家收取合理税款的装饰公司。万科集团2017年在上海,深圳和广州先后成立了装饰公司,房地产销售支付五个百分点的营业税,而装修收入仅需支付三个百分点的营业税,相比低两个百分点。

3.3.3政治动机

在信贷方面,房地产行业是资本密集型行业,房地产公司对资金的需求很高,因此要增加信贷风险,采取相应的措施来提高房地产项目的资本比率,并使人们成为第二购买住房贷款的首付的人。信贷增加的百分比和贷款利率的增加,降低了房地产行业信贷的风险。在税收方面,从户主二手房转让收入中征收所得税,使得房地产行业营业税的征收和管理得到加强。同时,征收土地增值税,进一步加强土地流转管理,严格管理各类非农建设用地和土地流转审批流程,力图通过相关措施保护土地资源。
3.4万科集团盈余管理手段分析

3.4.1利用非经常性损益

万科集团对外披露的2016年年度会计报告中注明公司属于上市公司股东的净利润98313.75万元,与2015年相比减少60%,然而在扣除非经常性损益后该值仅为25616.50

万元,与2015年减少90%。经研究发现2016年企业利润中计入8479.50万的营业外收入,这一部分主要来自万科集团与浙江省湖州市经济技术开发区管委会达成协议,退还本公司前期支付的土地认购款,并另行向本公司支付补偿款6500.00万元。另外,2016 年万科集团取得了一项特殊项收入10883.02万元,在年报中的解释为会计期内公司提供的咨询服务所取得的收入。其实,万科集团巧妙的利用从政府手中获取企业落户的政策性贴补企业利润的行为并不是初次。2014年万科集团净利润达145609.95万元,将非经常性收益从其中刨除之后减至83369.43万元,即近一半利润源自非经常性损益62240.53 万元。当万科集团面临较大的资金运转压力的时候,虽然也有机会获取的政府补贴款项和企业的经营业务,但是也正是为此被大额的调增利润后被追查的可能性非常低。

3.4.2利用公允价值计量

表3-3  2014年—2018年采用公允值价值计量的项目对当期利润的影响
年份 采用公允价值的项目对当期利润的影响(元) 增幅
2014 26,808,814.06  
2015 169,112,774.55 530.81%
2016 169,112,774.55 -2.80%
2017 404,763,600.00 146.25%
2018 441 ,261 ,600.00 9.02%
 
(数据来源:万科集团官网年度报告)
如表3-3中,万科集团在其年度财务报表中指出了合并公司的公允价值指标,该指标由活跃市场的金融资产或负债的公允价值乘以活跃市场的发行价确定,未表达财务报表的处置方式以及扣除产生的交易费用。公司持有的金融资产和其持有的金融负债的公开价格为当前报价,当前要求的价格为公司持有的金融资产或金融负债的估计价格。针对在市场上动态不均的银行产品,公司确定公允价值通过估值模型来估计,并且所使用的估值模型是指在类似情况下基本相同的金融产品的市场公允价值或当前公允价值或现金流量。期权价格和其他估值模型。以公允价值项目的增加和明显的变化来衡量,年度报告中披露的总金额为数亿元人民币,公司使用公允价值进行收益管理的可能性大大增加。

3.4.3利用资产减值准备

从表3-4中可以看出,万科集团的存货跌价准备金根据2015年年报,与2014年相比,2015年净利润比2014年减少8.1亿元,基本与存货跌价准备金相同。根据2016年最新的市场和销售情况,公司已完成所有项目。公司进行资产减值测试,尤其是在2015年底对具有跌价准备的项目进行评估和分析。根据测试结果,跌价规定中有九个已经完全冲销,其余四个项目有跌价规定。后续行动仍存在不确定性,因此2016年的股价储备仍为6.6亿元。
表3-4 万科集团2014年—2018年资产减值准备情况(单位:元)
年份 坏账准备 存货跌价准备 可供出售金融资产减值准备 长期投资减值准备
2014 105,704,325.53 11,970,772.52 5,860,000.00 2,600,000.00
2015 142,462,053.38 1,241,095,109.96 5,860,000.00 2,600,000.00
2016 164,299,563.23 624,680,5209    
2017 197,730,067.99 7,449,684.72    
2018 254,463,406.33 7,449,684.72    
 
(数据来源:万科集团官网年度报告)


第4章 万科集团盈余管理结果评价分析

根据万科集团在上面所分析的资产减值、公允价值、非经常性损益等在盈余管理方面的分析,为了使该公司的财务报告更公允真实地反应交易的实质,本文从财务绩效指标和市场绩效指标两个方面进行分析。

4.1财务绩效指标分析

4.1.1财务绩效定量评价指标

从下表中可以看出万科集团的资产收益率在2017年和2018年出现了大幅度下降。特别是2017年的资产收益率为7.62%,下降幅度为58.52% ,小于上一年资产收益率的一半。这也从侧面反映出2016年的盈余管理对于万科集团的负面影响,导致总资产收益率的异常波动。而在经营活动现金流量净额这个指标上,万科集团一直处于负值状态,说明企业经营活动成果并不像净利润反应的那么好。在2016年经营活动流量净额比2015年增加了24.34 亿元,2017年减至8.6亿元。特别是2018年,这一值达到了-50.08亿元是2017年的4倍,2015年的1.96倍。2016年万科集团利确认收储收入44.3亿,该笔收入将在未来12个月内实现经营性现金流入使得当年的经营活动现金流量净额数值回升,但是这种增长并不具有持续性和可复制性。万科集团的经营活动并没有明显好转,万科集团仍然沿用原有的土地开发和房地产项目的经营模式,2017年和2018年经营活动的净现金流量急剧下降。过度的盈余管理行为并没有有效地解决万科集团经营状况不佳的问题,反而导致了万科集团财务绩效指标的异常波动。市场对这种波动非常敏感,财务业绩的损失加剧了市场业绩的下滑。
表4-1 万科集团财务绩效指标表
年份 资产收益率(%) 净利润 扣除非经常性损益后的净利润 经营活动现金流量净额
2013 18.24 122622.99 110923.19 251903.46
2014 18.32 173423.13 163035.91 -240548.19
2015 14.39 145609.95 83369.43 -254970.37
2016 18.37 233447.99 191703.92 -11577.42
2017 7.62 97454.18 24756.93 -86163.25
2018 5.63 114600.44 70257.64 -500843.25
 
 
 
表4-2 财务绩效指标增长率一览表
 
年份 扣除非经常性损益后的净利润/净利润 资产收益率同比增长 归属净利润滚动同比增长 扣非净利润
滚动同比增
经营活动现金流量净额同比增长
2010 0.91        
2011 0.99 -6.99% 81.16% 98.03% -50.04%
2012 0.80 69.50% 163.23% 110.69% 199.89%
2013 0.90 -2.62% 7.35% 22.08% -291.02%
2014 0.94 0.44% 41.43% 46.98% -195.49%
2015 0.57 -21.45% -16.04% -48.86% 6.00%
2016 0.82 27.66% 60.32% 129.95% -95.46%
2017 0.25 -58.52% -58.25% -87.09% 644.24%
2018 0.61 -26. 12% 17.59% 183.79% 481.27%
 
(数据来源:万科集团官网年度报告)

4.1.2管理绩效定性评价指标

从万科集团的公司结构来看,从整体上看,万科集团已经根据公司治理框架建立了合适的公司治理结构。从公司的相关项目来看,有理由怀疑公司的组织结构没有发挥足够的作用。根据公告第八届董事会第64次会议的万科房地产(集团)有限公司有限公司监事会、董事会、审计委员会、独立董事同意的计算相关项目公司的会计,也没有环境损害公司和中小股东的合法权益。

4.2市场绩效指标分析

4.2.1托宾Q值

托宾Q值(企业市场价值/企业的重置成本),体现产业资产与金融资产二者之间的均衡关系。托宾Q值(以下简称Q )以1为分界点将企业分为财富创造型( Q>1入财富低效型( Q=1 )和财富缩水性(Q<1 )股价越高,托宾Q值也会随之增加,表明投资者看好公司未来成长前景,愿意支付高于企业资产价值的价格投资公司股票。

4.2.2市盈率

市盈率(每股市场价格/每股税后利润)是衡量股价高低和企业盈利能力的一个重要指标。一般来说,市盈率水平在0-13标明企业价值被低估,14-20是市盈率的正常水平区间,21-28标明企业价值被高估,大于28标明该企业股价存在投机性泡沫。在一定程度上,市盈率高标明投资人对于企业盈余增长信心大。表7万科房地产公司的一系列市场绩效指标及增长率:
表4-3 市场绩效指标及增长率一览表
年份 市盈率 托宾Q值 市盈率环比增长率 托宾Q值环比增长率
2010 49.66 1. 54    
2011 64.90 2.73 30.69% 77.27%
2012 13. 18 0. 82 -79.69% -69.96%
2013 27. 41 1. 53 107.97% 86.59%
2014 17. 76 1. 08 -35.21% -29.41%
2015 14. 59 0.60 -17. 85% -44.44%
2016 11. 55 0.58 - 20.84% -3.33%
2017 10. 55 0.34 -8.66% -41.38%
2018 11.31 0.43 7.20% 34.67%
 
(数据来源:万科集团官网年度报告)
由表4-3可知,万科集团从2015年起,托宾Q值一直处于1以下,在2017年达到最低值0.34。说明2015-2018年公司创造的价值小于投资资产的成本,企业资产处于缩水状态。在2012年万科集团盈余管理使得当年的托宾Q值降速放缓减至3.33%,但是201 7年的托宾Q值再次出现大幅度降低,此现象一直延续到2018年。2015-2018年间,市场对万科集团的盈利能力持有保留态度。虽然2016年市场对万科集团的信任度有所提升,但是万科集团的业绩提高大部分来自于盈余调节而不是企业经营状况的真实水平进步,所以当这种人为修饰的“业绩增加”无法持续的时候,市场对于企业的信任度还是会呈现下降状态,甚至更低。这也是2016年万科集团托宾Q值再次大幅度下降的原因。万科集团市盈率在2016年开始处于正常值以下,企业能力一直被低估。2016年减至11.56,降幅最大为20.84%,2017年有所缓解为8.66%。万科集团的市场表现并不如预期的好,特别是在盈余管理之后,托宾Q值和市盈率都有所下降。
 
 

 

第5章 加强企业盈余管理的措施

5.1加强财务风险意识

企业经营决策影响着企业财务风险的高低。在融资环节、投资环节、运营环节、利润分配环节都需要考虑公司的整体财务风险。国家宏观调控政策对房地产业的发展过于从紧。在“去产能、去产能、去杠杆”的政策背景下,万科房地产面临更加严峻的考验,万科集团需要转型。过去的商业策略只会让万科更加担心自己的未来。因此,寻求合理、科学、长期的经营战略是非常重要的。由于资金链紧,万科地产需要适当降低财务风险,谨慎选择投资项目,从而加快资金的流动,使资金得到有效利用。

5.2保障偿债能力

2016年以来,万科集团的负债率不断上升,资金链越来越紧。“拆东墙补西墙”的做法越来越明显。资产负债率在合理范围内,公司需要利用杠杆提高公司的盈利能力。但是过多的债务会给公司带来很大的偿债压力。万科需要调整长期负债和短期负债的金额和比例,优化资产负债结构,找到适合公司经营模式的资产负债率。2016年,万科开始出现大量坏账,应收账款周转率迅速下降。因此,万科资本的利用率和回收率亟待提高。应收账款周转率的增加一方面可以缓解资金短缺,另一方面也体现了企业经营能力的提高。

5.3尝试海外融资

万科集团的融资渠道一直是国内融资,这种单一的融资渠道在当前行业发展的背景下是不可行的。近年来,在经济下行压力和房地产业趋紧的情况下,单纯依靠国内融资已不能满足企业发展的需要。通过引入海外资金来降低融资成本的方法在一些重点房地产企业中取得了成功。因此,万科集团也需要跟上经济发展的步伐,灵活选择扩大自身发展的融资渠道,优化企业的融资成本通过多个融资渠道的有机结合,并加强企业的能力来应对资金链压力。

5.4锁定市场需求

“供过于求”、“刚性需求”无疑已成为当前房地产业的标签。房地产市场面临着产能和库存下降的任务。住宅住房的需求仍在增长。在一二线城市,房价仍处于高位,小户型面积的房价在大多数购房者的可接受范围内。万科地产需要准确定位未来开发项目的定位和定价门槛。这是万科寻找市场需求,开发符合市场需求的产品的又一次考验。


第6章 结论

经过上述研究分析,自2016年万科集团极大可能存在过度的盈余管理行为。万科集团通过对收入、费用、资产减值的调节,实现了主观控制利润。万科集团深知企业的生存发展对于地区经济的影响,政府在欢迎企业贡献经济增长的同时也会加大对于企业的监管力度,甚至在资金紧缺的时候利用政府政策上的保护补贴来增加企业的盈利水平。同时,借助金融资产和负债的公允价值计量属性,可以更加灵活地调节企业的利润,还有降低政治成本的政治动机和减轻税收负担的节税动机。根据指数分析,过度的盈余管理已经损害了万科集团的利润、现金流和股价。尽管一系列举措帮助万科实现了短期利润目标,但公司的长期利润将受到影响。偏离长期最优决策的后果表现为财务业绩和市场业绩的下降。通过对万科集团的案例研究可以发现,公司管理层能否准确把握盈余管理的规模是非常重要的。较低程度的盈余管理可以符合企业的发展战略,减轻企业在政治成本和税收成本上的负担,这有利于企业投入更多的资金和精力拓展业务市场;盈余管理程度越高,负面影响越严重,房地产企业盈余管理过度难以达到预期效果。

参考文献

[1] Paul M. Healy, James M. Wahlen, A review of the earnings management literature and its implications for standards setting, Accounting Horizons,Dec.1999
[2] William R. Scott. Financial Accounting Theory. Prentice-Hall, Inc, 1997
[3] Katherine Schipper, Commentary on earning management, Accounting Horizons, 1989
[4] Paul M. Healy, James M. Wahlen, A review of the eanings management literature and its implications for standards setting, Accounting Horizons,Dec.1999
[5] 刘芹.银行债务契约与债务人盈余管理关系的实证研究[J].经济纵横,2012,(06):114-117.
[6] 孙姝毅.基于公司治理结构下我国上市公司盈余管理研究[D].硕士论文.镇江:江苏大学,2006
[7] 陆宇建.上市公司基于配股权的盈余管理行为实证分析[J].经济学研究,2002.3
[8]  叶志锋,胡玉明,纳超洪.基于银行借款融资动机的盈余管理研究——来自中国证券市场的经验证据[J].太原:山西财经大学学报,2008,(01):118-124.
[9]  马永强,赖黎,曾建光.盈余管理方式与信贷资源配置[J].会计研究,2014,(12):39-45+95.
[10] 赵路明.ST公司有效利用盈余管理的实证研究[D].硕士论文.重庆:重庆大学,2004
[11] 郝煜.我国上市公司治理结构与盈余管理研究[D].硕士论文.北京:对外经济贸易大学硕士学位论文,2004
[12] 宁虎.我国上市公司治理结构对盈余管理的影响及博弈分析[J].审计月刊,2008年8期
[13] 陆建桥.中国亏损上市公司盈余管理实证研究[M].北京中国财政经济出版社,2002:46—88
[14] 赵佳雯.盈余管理动机对资产减值准备处理影响的实证研究[D].上海:东华大学,2015
[15] 张伟.盈余管理的计量方法文献综述[J] .时代金融, 2015 , 09 : 263-264
[16] 朱根云.非经常性损益持续性及价值相关性研究[D].上海:华东交通大学,2014
[17] 姚莹.真实盈余管理动机及其经济后果研究[D] .重庆:重庆大学, 2013
[18] 沈村.公允价值计量在投资决策中的风险防范—基于某公司投资决策过程中公允价值计量风险防范分析[D]. 上海:复旦大学, 2013
[19] 桂良军,于洋.混合所有制企业盈余管理对信贷融资的影响研究[J].财会月刊,2016,(30):40-45.
 


致  谢

在学习期间,承蒙各位老师的指导与教诲,使我圆满完成了学业。四年求学经历,学校浓厚的学习、学术氛围,令我终身难忘。在此,向给予我关系和帮助的各位老师表示感谢。在此篇论文的研究阐述过程中,我的导师杨晓清从论文的选题、思路、研究方法以及内容、格式等各方面都给予了悉心的指导,先后提出了很多建设性的意见与建议。我谨向张晓清导师致以崇高的敬意和衷心的感谢!
感谢全体同学的帮助与勉励,同窗四年,情同手足,我将终生难忘。同时,我还要感谢我的家人,感谢你们默默的关怀与鼓励,使我有足够的毅力完成全部学业。
最后,向审阅本文付出辛劳的各位专家、学者表示深深地谢意。


附录一 中文译文

中国中央国有企业盈余管理一国家层面激励的效果
摘要
本研究建立在利益趋同效应的基础上,假设中国中央国有企业(CSOEs)与中央政府之间的利益紧密结合,为国有企业的收益管理提供了国家级激励(如降低GDP波动)。与我们的命题一致,我们发现中国国有企业进行盈余管理是为了降低GDP的波动性。此外,我们发现中国国有企业只使用真实的盈余管理方法来降低企业收益波动,以减轻GDP波动。
1.介绍
本研究从广义趋同效应的角度对我国中央企业的盈余管理进行了研究。趋同效应的论证表明 , 当企业的控股股东足够大时,控股股东的利益与企业的利益是密切相关的。控股股 东的利益与公司的利益紧密结合可能会影响公司的盈余管理行为(Shleifer和Vishny 1986)。 在法律和制度环境较弱的国家, 如中国,联盟.效应更强(Ding, Zhang, and Zhang 2007)。 本文进 一步分析了国有企业利益与国有企业所有者利益之间的利益一致性,认为国有企业利益与国有企业所有者利益之间的利益一致性可以激励国有企业进行收益管理。
传统上,国有企业一直是中国的主导商业形式。虽然传统国有企业在中国的数量和重要性都大幅下降,但中国国有企业作为一种新型的国有企业,在中国经济中的地位却越来越重要。中国国有企业是规模庞大、资金充足的国有冠军企业集团,它们在国内垄断战略行业,在国际投资和收购领域展开激烈竞争。中国中央政府不仅通过控制股权,还通过其监督权力和决策权力对国有企业保持密切控制(Lin and Milhaupt 2013)。 我们认为 ,中国中央政府对国有企业的控制导致了国家层面的盈余管理激励。中国政府最伟大的成就之一,就是在保持稳定的同时,开创了前所未有的经济增长。因此,我们假设,国家层面的激励激励中国国有企业从事收入管理,以减少国内生产总值的波动,使中国人民感到更稳定和安全。
2.数据
我们的测试样本包括100个来自中国证券指数(CSI)中央国企100指数的中国国企, 其中包括最终由中央政府控制的最大的100个公共国企。1使用一 个控制样本100名当地国有企业包括沪深当地国企指数和200年non-SOEs包含在CSI non-SOE指数,从2003年到2015年,我们发现中国方案选择实际收益管理方法,使他们能够增加(减少)contri- butions低(高)GDP增长的国家GDP年 此外,我们发现收益管理方法选择的方案也会增加(减少)公布的企业收益低(高)GDP增长。
3.结论
本研究对国有企业盈余管理文献的贡献是双重的。首先,就我们所知,本研究是首次探索中国国有企业的盈余管理的国家层面的激励。虽然之 前对中国国有企业盈余管理的研究有限,但没有一项研究关注强大的中国国有企业。此外,虽然国家层面的激励已经被发现会影响国有企业的税收行为(如Chen et al. 2015;2011年;国家层面的激励对国有企业盈余管理的影响从未被研究过。中国国有 企业与中央政府的利益紧密结合,这使得国有企业不同于地方国有企业, 地方国有企业的利益与地方政府的利益是一致的。 这项研究的结果提供了对世界最大经济体之- -的主导力量的行为的理解。其次,本研究通过考察各种权责发生制和实际盈余管理方法(如对GDP波动率)的附加影响,进一步深化 了盈余管理的文献。研究表明,股权集中度不仅影响企业盈余管理目标,而且影响盈余管理方法的选择。虽然在美国 ,多元化的公司所有权结构占主导地位,但与国有企业类似的更集中的所有权结构的公司在中国和世界上许多其他国家和司法管辖区更常见(La所有权结构占主导地位,但与国有企业类似的更集中的所有权结构的公司在中国和世界上许多其他国家和司法管辖区更常见(LaPorta et al, 2000)。在这个全球化的时代,研究人员、政策制定者、企业高管和投资者都应该关注国有企业的盈余管理行为。本研究为进一步研究国有企业盈余管理方案提供了一个新的视角。本研究的发现也有实际意义。中国国有企业在中国经济增长和发展中发挥着独特而重要的作用。中国政府最近发布了一项新的政策指令,寻求扩大国有企业的混合所有制.并放松对私人投资者作用的限制,以提高国有企业的业绩。中国政府的新改革举措可能为国际投资者在这个全球第二大、增长最快的经济体创造巨大的投资机会。在中国等 新兴市场经营的国际企业和投资者所面临的挑战。与成然市场的信总不对称更为严重(Chan, Menkveld,Yang 2008)。 因此,本研究的结果也应该是潜在的国际企业和投资者在评估中国国有企业会计的相关性和可靠性的兴趣信息和在制定商业和投资决策。
本文组织如下。第一部分总结了文革后中国国有企业在经济改革过程中的演变,并介绍了中国国有企业;下一节回顾盈余管理的文献,特别是有关国有企业盈余管理的文献,并提出假设;接下来的章节介绍了研究的设计,结果的讨论和最后的评论。
 

附录二 外文原文

Earnings management of Chinese central state-owned enterprises – the effects of state level incentives*
Hong Fan and Xiaofei Song
Department of Accounting, sobey’s school of Business, saint mary university, Halifax, canada
 
 
ARTICLE HISTORY
received 11 september 2016
Accepted 19 september 2017
KEYWORDS
Earnings management; central state-owned enterprises; alignment effect; china
 
 
 
Introduction
This study investigates the earnings management of Chinese central state-owned enterprises (CSOEs) from the perspective of an extended alignment effect. The argument of alignment effect suggests that when the controlling shareholder of a firm is sufficiently large, the controlling shareholder’s inter- ests will be closely aligned with the firm’s interests. The close alignment between the interests of the controlling shareholder and those of the firm could affect the firms’ earnings management behavior (Shleifer and Vishny 1986). Alignment effects are stronger in countries with weaker legal and institu- tional environments such as China (Ding, Zhang, and Zhang 2007). We extend the alignment effect to argue that the close alignment between the interests of Chinese CSOEs and those of their state owner could induce state-level incentives for CSOEs to manage earnings.
State-owned enterprises (SOEs) have traditionally been dominant forms of business in China. Although traditional SOEs have declined significantly in both number and importance in China, Chinese CSOEs, a new type of Chinese SOE, have become more important in the Chinese economy. Chinese CSOEs are over-sized and well-funded national-champion conglomerates that monopolize strategic industries domestically and that compete aggressively in investment and acquisitions arenas internationally. The Chinese central government maintains close control of the Chinese CSOEs, not only by controlling shareholdings but also through its supervisory authority and policy-making power (Lin and Milhaupt 2013). We argue that the Chinese central government’s control over Chinese CSOEs gives rise to state-level earnings management incentives. One of the Chinese government’s greatest achievements has been its initiation of unprecedented economic growth while maintaining stability. Therefore, we posit that state-level incentives motivate Chinese CSOEs to engage in earnings man- agement to mitigate GDP volatility, allowing the Chinese population to feel more stable and secure.
CONTACT   Xiaofei song     xiaofei.song@smu.ca
*Accepted by suresh radhakrishnan upon recommendation by Yuan Zhang
© 2017 city university of Hong Kong and national taiwan university
 
Our test sample includes 100 Chinese CSOEs drawn from the China Securities Index (CSI) Central SOEs 100 Index, which includes the largest 100 public CSOEs that are ultimately controlled by the central government.1 Using a control sample of 100 local SOEs included in the CSI local SOE index and 200 non-SOEs included in CSI non-SOE index, from 2003 to 2015, we find that Chinese CSOEs chose real earnings management approaches that enabled them to increase (decrease) their contri- butions to the national GDP in low (high) GDP growth years Furthermore, we find that the earnings management approaches chosen by CSOEs also increase (decrease) reported enterprise earnings in low (high) GDP growth years.
This study’s contributions to the SOE earnings management literature are twofold. First, to the best of our knowledge, this study is the first to explore earnings management by Chinese CSOEs for state- level incentives. While limited prior studies have examined earnings management by Chinese SOEs, none have focused on powerful Chinese CSOEs. Also, although state-level incentives have been found to affect SOE tax behaviors (e.g. Chen et al. 2015; Deng and Luo 2011; Li, Feng, and Cao 2016), the effects of state-level incentives on SOEs’ earnings management have never been explored. The close alignment of the interests of Chinese CSOEs and the central government differentiates CSOEs from local SOEs, whoes interests are aligned with local governments. The findings of this study provide understandings of the behavior of the dominant force in one of worlds largest economies. Second, this study furthers the literature on earnings management by examining the additional effects of various accrual and real earnings management approaches (e.g. on GDP volatility). We argue and show evi- dence that ownership concentration not only influences firms’ objective of earning management but also choices of earnings management approaches. While diverse corporate ownership structures are dominant in the United States, firms with more concentrated ownership structures similar to those of SOEs are more commonly found in China and many other countries and jurisdictions around the world (La Porta et al. 2000). Information on the earnings management behaviors of SOEs should be of inter- est to researchers, policy makers, business executives and investors in this time of globalization. This study presents a new perspective to further the investigation of SOE earnings management schemes. The findings of this study also have practical implications. Chinese CSOEs play unique and signif- icant roles in China’s economic growth and development. The Chinese government recently issued a new policy directive seeking the expansion of the mixed-ownership of CSOEs and the relaxation of restrictions on roles played by private investors in an effort to improve the performance of SOEs. The Chinese government’s new reform initiative could potentially create vast investment opportunities for international investors in the world’s second largest and fastest growing economy.2 A challenge faced by international businesses and investors operating in emerging markets such as China pertains to the more profound information asymmetry compared to those of established markets (Chan, Menkveld, and Yang 2008). Thus, the findings of this study should also be of interest to potential international businesses and investors in their assessments of the relevance and reliability of Chinese SOE accounting
information and in making business and investment decisions.
This paper is organized as follows. The first section summarizes the evolution of Chinese SOEs throughout post-Cultural Revolution economic reforms and introduces the CSOEs; the next section reviews the earnings-management literature, particularly the literature regarding SOE earnings man- agement, and develops hypothesis; the sections following present the research design, a discussion of the results and final remarks.
The evolution of Chinese SOEs
Chinese SOEs have evolved significantly since the start of post-Cultural Revolution economic reforms. Prior to these economic reforms, all Chinese enterprises were state-owned. Beginning in the late 1970s, along with the rise of both commercial enterprises and a market economy, many small- and medi- um-sized SOEs that suffered from low levels of productivity and persistent losses were privatized. By the late 1990s, less than 40% of Chinese business enterprises were SOEs, and they controlled roughly 70% of all assets (Xu, G. 2010). Since then, Chinese SOEs have continued to decline in number. By the end of 2008, less than 5% of all Chinese enterprises were SOEs. However, the remaining small number of Chinese SOEs controlled nearly half of all total assets and generated nearly half of all industrial and business profits (Cary 2013). The shrinking number of SOEs was a result of concerted efforts to consol- idate disparate SOEs, which led to the emergence of CSOEs in strategic industries (Anonymous 2012). Of the 144,777 Chinese SOEs currently in operation, less than 0.1% are CSOEs controlled by the central government through the State-Owned Assets Supervision and Administration Commission (SASAC) (Cary 2013). These centrally controlled SOEs are the largest and most important Chinese enterprises. Through their multiple subsidiaries, Chinese CSOEs include almost all the most familiar Chinese companies (Scissors 2011). Three Chinese companies were among the top 10 of the 2016 Fortune Global 500 list, and they were all CSOEs: State Grid, China National Petroleum and Sinopec Group.3 These Chinese CSOEs also serve as key instruments of China’s international expansion; of China’s $60 billion in investments and acquisitions overseas in 2011, 80% came from CSOEs (Cary 2013). For instance, in February 2017 ChemChina acquired Swiss agribusiness giant, Syngenta. The
two entities will merge to create what will be the world’s largest chemical group.4
The SASAC is a government agency that reports to the State Council (China’s cabinet). Through the various wholly state-owned holding companies, the SASAC maintains controlling ownership (>50%) over all Chinese CSOEs. More importantly, the SASAC exercises additional control as a supervisory authority and policy-maker, subjecting Chinese CSOEs to tight governmental controls not typically applied to businesses (Lin and Milhaupt 2013). For instance, the SASAC has the responsibility to ‘appoint or remove the responsible persons for the invested enterprises and evaluate their performance in accordance with the statutory procedures and granting rewards impose punishments based on the evaluation results’ (State Council of the People’s Republic of China’s 2003). While SOEs are for-profit enterprises according to the Chinese government, countless anecdotes show that when state interests do not fully coincide with business interests, state interests trump business decisions (Canadian Trade Commissioner Service in China). As quasi-state apparatus, Chinese CSOEs also enjoy tremendous support and privileges, which include access to more markets (Luo 2003), lower borrowing rates and preferential policy treatment (Chen et al. 2016; Fan and Hope 2013).
As China is rapidly becoming the world’s largest economy and the second-largest trading part- ner to both the US and Canada,5 Chinese CSOEs are becoming key players in the global economy. Understating the behaviors of Chinese CSOEs could be helpful to companies that compete directly or that are given opportunities to form partnerships with Chinese CSOEs. In an effort to promote contin- uous economic growth and to reduce inefficiency levels, the third Plenum of the Chinese Communist Party’s 18th Congress (October 2013) and the National Peoples’ Congress have formally approved a plan to expand CSOE mixed ownership to attract private and foreign capital (March 2014). Fifty-two percent of the CSOEs and their subsidiaries already host some form of private investment according to the SASAC, but the roles of private investors have remained rather restricted thus far. Such expansion could create significant investment opportunities for international investors. Knowledge that could help reduce information asymmetry often encountered in emerging market investment should be of great interest to international investors.
Information on the accounting decisions of Chinese CSOEs could also further our understanding of the financial reporting of SOEs in general. SOEs play significant roles in many emerging and tran- sitional economies (Government of Canada 2014). Even in established market economies such as the US and Canada, SOEs are critical in certain economic sectors. The recent global financial crisis that resulted in increased governmental intervention in certain industries also brought to light certain hidden problems in these enterprises. A better understanding of SOE behaviors could facilitate more effective regulation and supervision in these jurisdictions.
Earnings management research
The traditional earnings management literature is based on agency theory (Jensen and Meckling 1976), which focuses on conflicts of interests between shareholders (principals) and management (agents).
SOEs and corporations with more concentrated ownership face a different set of institutional factors and corporate governance issues from those faced by typical North American corporations. According to La Porta, Lopez-De-Silanes, and Shleifer (1999) agency problems faced by firms like SOEs involve the expropriation of minority shareholders by controlling shareholders rather than the expropriation of shareholders by management.
One of the main research questions of SOE earnings management studies is whether SOEs engage in more earnings management than non-SOEs. Corporate governance studies often associate state ownership with poor corporate governance, low levels of corporate efficiency, corporate corruption and fraud, more frequent earnings management and low earning quality (Boardman and Vining 1989; Megginson, Nash, and Van Randenborgh 1994; Shleifer 1998). On the other hand, Wang and Yung (2011) argue that political costs and state protection could ‘mitigate the pressure on managers to manipulate firm-specific information.’ The empirical evidence on this matter is inconclusive. Liu, Saidi, and Bazaz (2014) note that Chinese SOEs manage their earnings more frequently and achieve lower levels of earnings quality than non-SOEs. Contrary to this, Chen et al. (2011) and Wang and Yung (2011) show that compared to their private counterparts, Chinese SOEs exhibit significantly lower levels of earnings management.
Ding, Zhang, and Zhang’s (2007) non-linear model may reconcile the seeming conflicting empir- ical findings of SOE earnings management studies. Ding, Zhang, and Zhang (2007) used an inverted U-shaped configuration to capture the relationship between SOE state ownership and SOE earnings management levels and referred to it as the alignment effect. Specifically, they argued and found evi- dence that, compared to non-SOEs, SOEs exhibited higher levels of earnings management because the state as a controlling owner had a greater capacity to expropriate minority owners. However, highly concentrated ownership aligned the interests of controlling owners and firms to such an extent that earnings management activities that could harm firm value were also detrimental to controlling owners. They concluded that extremely concentrated majority ownership was associated with reduced earnings management.
We argue that the unique institutional characteristics of Chinese CSOEs subject them to incentives to manage earnings that can be different from those of other companies. First, all Chinese CSOEs are tightly controlled by the same central Chinese government. Through the SASAC, the Chinese central government controls all aspects of CSOEs, including management performance evaluations and rewards. Thus, it is highly important for the managers of Chinese CSOEs to gain government approval by supporting the government’s political, social, and financial goals. Second, there are fun- damental differences between the ultimate goals of a government and those of a corporation. While a government is responsible for the political, social, and financial benefits of its citizens or constituents, corporations are primarily profit seeking. The Chinese government requires SOEs to assist the gov- ernment in achieving its goals (Deng and Luo 2011). For instance, in its 13th five-year plan,6 the state proposes to maintain a GDP growth rate of 6.5% over the next five years (2016–2020)7 and requires SOEs to increase efficiency levels and to transform to a mixed-ownership economy.8 In China, it is a widely held view that the success of SOEs as national champion enterprises are often measured by their contributions to national interests, such as successful prestige projects, not necessarily by firm level profits (Cary 2013).
One of China’s chief accomplishments lies in its unprecedented economic growth, which has created the foundation for China’s political and social stability. Although the ruling Chinese Communist Party has had no serious political challengers over more than two decades of democratic election history, the government is keen to prevent murmuring discontent concerning issues of inequality, corruption, pollution, human rights, etc. from escalating into open protests. Continuous and sustained economic growth requires careful pacing. Growing too fast could cause the economy to overheat, which inevitably would lead to a slowdown. For instance, like most countries, China experienced astonishingly rapid growth in 2006 and 2007 prior to the recent global financial crisis. Economic policies imposed by the Chinese government throughout this period were mainly designed to prevent the economy from becoming overheated. During the major slowdown of the global financial crisis of 2008 and 2009, the
government applied various stimulus policies to revive growth. Chinese CSOEs should be motivated to collaborate with the government in its efforts to maintain economic stability. Stated alternatively, our hypothesis regarding Chinese CSOE state-level earnings management incentives is as follows:
Hypothesis: Chinese CSOEs are motivated to engage in GDP-increasing (GDP-decreasing) earnings management in low (high) GDP growth years.
Although Chinese CSOEs are influenced by institutional factors that are different from those faced by other types of business enterprises, Chinese CSOEs are business enterprises nevertheless. As is the case for all for-profit businesses, profitability is a relevant performance measure and has implications for valuation. Prior studies have established that a steady reported earnings stream could enhance firm value and could benefit managers in many ways (Lambert 1984; Trueman and Titman 1988). We argue that in addition to state-level incentives, Chinese CSOEs are subject to similar institution-level earnings management incentives, e.g. to mitigate earnings volatility. Given that the economic factors that give rise to GDP volatility are also likely to affect Chinese CSOE earnings volatility, Chinese CSOE earning smoothing efforts could be associated with GDP smoothing efforts.
Three methods of GDP calculation are available: the Expenditure Approach, the Production Approach and the Income Approach.9 Chinese GDP calculations are mainly based on the Production Approach, while the results of the other two methods are used as a reference (Xu, X. 2010). The Production Approach is also referred to as the Net Product or Value Added method. When using this method, the GDP is the difference between the output, which is calculated as the quantity of products multiplied by the market value of products, and the intermediate consumption (e.g. materials, labor costs, etc.). Different earnings management approaches affect Chinese GDP differently. As a result, we expect that Chinese CSOEs prefer the earnings management approaches that can mitigate both GDP volatility and firm earnings volatility.
Research design
The model
To test the hypothesis of this study, we use the regression model expressed as Equation (1):
EMit = /30 + /31 CSOEi + /32 HighGDPGrowtht−1 + /33 LowGDPGrowtht−1
+ /34 CSOEi × HighGDPGrowtht−1 + /35 CSOEi × LowGDPGrowtht−1
+ /36 SIZEi,t−1 + /37 ROAi,t−1 + /38 LEVi,t−1 + /39 MTBi,t−1
+ /310 Marketizationi,t−1 + /311 AUDITORi,t−1 + E
In Equation (1), EM denotes the various measures of both accrual-based and real earnings manage- ment; CSOE, LowGDPGrowth and HighGDPGrowth are dummy variables that identify Chinese CSOEs and extremely low and extremely high GDP growth years, respectively; ROA, LEV, SIZE, AUDITOR, Marketization and MTB are control variables (returns on assets, financial leverage, firm size, big-8 auditors, the Marketization index, and the market-to-book ratio, respectively) that are identified by the earnings management literature as having an impact on a firm’s earnings management activities (e.g. Cheng, Wang, and Wei 2015). ε is the error term. Subscripts i and t identify the firm and year, respectively. The GDP growth year dummy variables and all control variables in the model are lagged by one year.
Dependent variables – The earnings management variables
The literature has identified different approaches to earnings management. We follow Cohen and Zarowin (2010) in measuring accrual earnings management and Roychowdhury (2006) in computing three metrics of real earnings management: overproduction, abnormal cash flows from operations, and reductions of discretional expenses.
The earnings-management measure estimates and their impacts on GDP are described below.
(1) Accrual earnings management (Accrual EM)
We use the following model to estimate normal accruals.
TAi,t Assets i,t−1
= k 1
1 Assets
i,t−1+ k2
ΔSalesi,t
Assets i,t−1+ k3
PPEi,t +
Assets i,t−1
In Equation (2), TAi,t denotes Total Accruals calculated as earnings before extraordinary items and discontinued operations minus operating cash flows; Assetsi,t−1 denotes total assets from the preceding year; ∆Salesi,t denotes changes in revenue calculated as the current year’s revenue minus last year’s revenue; and PPEi,t is the gross value of property, plants and equipment. The normal accruals (NA) of each firm and year are estimated using Equation (2). Discretionary accruals is the difference between
actual accruals and normal accruals, TAi,t/Asset i,t−1 − NAi,t.
Accrual-based earnings management achieves desirable reported earnings by manipulating the timing of revenue and expense recognition. To illustrate the impacts of an accrual-based earnings management approach on GDP, we use the example of a manufacturer who ‘manages’ inventories and costs of goods sold in a low GDP growth year. According to the cost of goods sold model, the sum of Beginning Inventory and Production equals the sum of Ending Inventory and Cost of Goods Sold. In a low GDP growth year, to increase reported earnings using this approach, a CSOE would need to assign a higher value to Ending Inventory and a lower value to Cost of Goods Sold. This would not have any impact on the output of the CSOE’s total production level and would thus have no effect on production based GDP. As a result, accrual EM is not an effective approach for state-level incentives since there is no significant impact on GDP.
(2) Real earnings management
Roychowdhury (2006) presents a detailed discussion of the three measures of real earnings man- agement: abnormal production costs, abnormal cash flows from operations (CFO), and abnormal discretionary expenses.
(1) Abnormal production costs (APROD)
Normal production costs are estimated using Equation (3):
PRODi,tAssets i,t−1 = k 1
1 Assets  i,t−1 + k2
Salesi,t Assets i,t−1+ k3
ΔSalesi,t Assets i,t−1 + k4
ΔSalesi,t−1 + E Assets i,t−1
In Equation (3), PRODi,t is the sum of the cost of goods sold and the change in inventory for this year, while all other variables are defined as those in Equation (2). The APROD is the difference between
the actual and normal production cost, which is estimated using Equation (3). Holding unit sales constant, a higher APROD indicates that firms have lowered the cost of goods sold, and thus reported higher earnings, through overproduction (Roychowdhury 2006). Overproduction would contribute positively to Chinese production-based GDP. Thus, the APROD is an effective earnings management approach to smooth GDP as well as enterprise earnings simultaneously.
(2) Abnormal cash flow from operations (ACFO) A normal CFO is estimated using Equation (4):
CFOi,t Assets i,t−1= k 1
1 Assets i,t−1+ k2
Salesi,t Assets i,t−1+ k3
ΔSalesi,t +Assets i,t−1
In Equation (4), CFOi,t denotes cash flow from operations in period t, and all other variables are as defined as above. Abnormal CFO is the difference between the actual and normal CFO calculated
 
using the estimated coefficients from Equation (4), and the ACFO is the negative value of Abnormal CFO. We take the negative value here to capture the upward effect on GDP and earnings. A higher ACFO (i.e. lower abnormal CFO) indicates that a firm might have provided additional discounts and/or more lenient credit terms to temporarily increase sales and thus the earnings of the period. Such forms of sales management are expected to lead to lower current period CFO levels, as the cash collection of such sales often are slower than normal (Roychowdhury 2006). Additional sales would be counted as additional spending by customers, and spending would be expected to lead to more production, thus resulting in more indirect contributions to GDP from the customers. Consequently, ACFO could serve as an effective earnings management approach for both GDP smoothing as well as enterprise earnings smoothing.
Both real earnings management approaches based on overproduction and on abnormal CFO have the same direction effects on GDP and CSOE earnings, therefore, could reduce the volatilities of earn- ings and GDP at the same time, if the volatilities in GDP and CSOE earnings happen concurrently, which we expect. We argue that overproduction-based real earnings management is more efficient than abnormal CFO-based real earnings management for GDP smoothing. First, the overproduction-based approach manipulates the timing of production and is not likely to cause real economic losses, whereas the abnormal CFO-based approach manipulates the timing of sales at a real economic cost, i.e. the additional discount offered for early sales. Second, overproduction may require additional purchases during the period, whereas the abnormal CFO-based approach may require additional sales from a willing partner. As companies tend to have more control over buying than selling, the overproduc- tion-based approach is easier to apply than the abnormal CFO-based approach. We predict that the overproduction-based approach is preferred by Chinese CSOEs to the abnormal CFO-based approach for the purposes of earnings management.
(3) Abnormal discretionary expenses (ADISX)
We model the normal level of discretional expenses using Equation (5):
 
 
DISXi,t
 
Assets i,t−1
 
= k 1
1 Assets
 
i,t−1
 
+ k2
 
Salesi,t−1 Assets i,t−1
 
+ Eit
 
(5)
 
In Equation (5), DISXi,t denotes discretionary expenses defined as the sum of advertising expenses, research and development expenses and sales and general and administrative expenses. The ADISX
is the difference between actual and normal discretionary expenses predicted through Equation (5). A higher ADISX denotes more firm spending. Other things being equal, a higher ADISX should lead to higher GDP, but lower firm earnings.
All measures of real earnings management are signed measures. Further, the signs are set so that all measures of real earnings management have an upwards effect on GDP. Accruals based earnings management has no impact on GDP.
We used a cross-sectional model to measure accrual based earnings management and the three metrics of real earnings management, where we estimated the models for each 2-digit industry for each year. Following Cohen and Zarowin (2010), we required at least eight observations for each industry and year group and sufficient data to calculate all four earnings management variables. This approach reduced our sample size and might have biased our sample by including more successful firms. However, we believe that this approach also reduces variations in earnings management met- rics, ‘resulting in a more conservative test of our research questions’ (Cohen and Zarowin 2010, 6).
 
The independent variables
Chinese CSOE is a dummy variable that is equal to 1 if a firm is included in the CSI Central State- Owned Enterprises 100 Composite Index and that is equal to 0 otherwise. The GDP growth rate is calculated as the current year’s GDP minus last year’s GDP divided by last year’s GDP. LowGDPGrowth
 
is a dummy variable that takes a value of 1 if GDP growth falls within the lowest 25% for 2003–2015 (i.e. equal to or lower than −6%) and that takes a value of 0 otherwise. Low GDP growth years include 2008, 2011, and 2012. HighGDPGrowth is a dummy variable taking a value of 1 when GDP growth is in the top 25% for 2003–2015 (i.e. equal to or higher than 11%) and taking a value of 0 otherwise. High GDP growth years include 2005, 2006, 2007 and 2010. In our regressions, the lagged variables are used. The use of high and low GDP growth years to proxy for high economic growth pressure periods is supported by the GDP growth targets set periodically by the Chinese central government in its Report on the Work of the Government. The Chinese government set fairly constant GDP growth targets dur- ing our sample period: from 7% to 8%. The years identified as the periods in which China experienced economic growth pressure, i.e. the years with low or high lagged GDP growth years, are supported by patterns of Chinese economic and financial cycles: China experienced rapid growth before 2007; it was hit by the global financial crisis in 2008, 2011 and 2012.10 The Chinese central government uses fiscal and monetary policies to boost GDP growth when it slows down and to cool down GPD growth when it is high. For instance, from 2006 to 2007, which are the high GDP years in our sample, the Chinese central bank increased the required reserve ratio seven times for commercial banks to cool the invest- ment and the economy.11 In late 2008, which is the low GDP year in our definition, the Chinese central government announced a four trillion yuan ($586 billion) investment plan to stimulate the economy and to avoid the economic downturn caused by the financial crisis.12 In addition, we calculated the GDP difference between the current year’s GDP growth rate and the previous year’s GPD growth rate, and we define the top 25% as the high GDP growth years and the bottom 25% as the low GDP growth
years. Using this definition yielded the same low and high economic growth pressure years.
Our hypothesis posits that Chinese CSOEs manage earnings to smooth GDP. A negative β4 (CSOEi × HighGDPGrowtht−1) and a positive β5 (CSOEi × LowGDPGrowtht−1) in Equation (1) from regressions on all real earnings management dependent variables will support our hypothesis of GDP
smoothing. To the extent that the same economic factors that cause GDP volatility also cause CSOEs’ earnings volatility, a negative β4 and a positive β5 from regressions on ARPOD and ACFO mean the real earnings management to smooth GDP with APROX and ACFO also results in lower earnings volatility. However, a negative β4 and a positive β5 from regressions on ADISX mean the real earn- ings management to smooth GDP with ADISX results in higher earnings volatility. Accruals based
earnings management has no impact on GDP. Consequently, we expect that Chinese CSOEs prefer APROD- and ACFO-based real earnings management approaches over accrual- and ADISX-based earnings management approaches to smooth GDP.
Control variable SIZE is the natural logarithm of total assets. LEV is calculated as long-term debt at the current year-end divided by the book value of equity at the current year-end. ROA is calculated as the net income of the current year divided by the beginning balance of total assets of this year. MTB is the market value of a firm divided by the book value of a firm for this year. AUDITOR is a dummy variable that is equal to 1 if the auditor is a big-8 auditor in China (Cheng, Wang, and Wei
2015), and it is equal to 0 otherwise. Big-8 auditors include the big 4 international auditing firms (i.e. PwC, E&Y, Deloitte, and KPMG) and the four largest Chinese auditing firms identified by Cheng, Wang, and Wei (2015), which are RSM China, Shu Lun Pan, Zhejiang Pan-China, and Shine Wing. Marketization denotes the extent of marketization development in each province, and the index is obtained from Fan (2011).13
Most of the data we used (e.g. ROA, LEV, SIZE, AUDITOR, MTB and data for calculating earnings
management levels) were retrieved from Capital IQ. Chinese GDP data were collected from Statistics of China.14 Firm types (i.e. central SOEs, local SOEs and non-SOEs) are categorized based on indexes provided through the China Securities Index Co. Ltd.15
Sample and summary statistics
The sample includes 400 public firms from the Shanghai and Shenzhen Stock Exchanges: 100 firms from the CSI Central State-Owned Enterprises 100 Index as a test sample, 100 firms from the CSI
 
Local State-Owned Enterprises 100 Index and 200 firms from the CSI Private-Owned Enterprises 200 Index as a control sample. The index defines CSOEs as public companies that are ultimately controlled by the central government through various holding companies wholly-owned by the SASAC.16 For instance, Datong Telecom Technology Co. Ltd, a public CSOE listed on the Shanghai Stock Exchange, is controlled by the holding company Datang Telecom Technology & Industry Group, which is fully owned and supervised by the SASAC.17 All firm the in indexes are the largest, most liquid and most representative of their respective constituents.
The sample period runs from 2003 to 2015 and is inclusive. The initial sample includes 5200 observations (firm-years): 1300 CSOE observations, 1300 local SOE observations and 2600 non-SOE observations. Our sample section is based on indexes for 2015, and our sample was reduced because some firms went public after 2003. After eliminating observations with missing values for earnings management measures and control variables, the final sample consisted of 2475 observations (647 CSOE observations, 739 local SOE observations and 1089 non-SOE observations). Table 1 describes the sample selection process.
Table 2 presents statistical descriptions of the variables for CSOE, local SOE and non-SOE sub- samples. The average accrual EM level is 0.064 for CSOEs and is 0.091 for local SOEs and non-SOEs. CSOEs appear to engage in less upward accrual-based earnings management activities, and the dif- ference is significant. Differences between the means of all real earnings management measures for the two subsamples are insignificant.
The CSOEs are significantly larger on average compared to local SOEs and non-SOEs. However, the CSOEs are more highly leveraged and are less profitable with a leverage of 109% and an ROA of 5.2% compared to the mean leverage of 59.1% and mean ROA of 9.20% for the control group. These differences are statistically significant. The CSOE market-to-book ratio is lower on average (3.6) than that of the control companies (3.9), but the difference is not significant. More than half of the CSOEs use big-8 auditors, while only 35% of local SOEs and non-SOEs use big-8 auditing firms.
This difference is also statistically significant. The two subgroups do not have significant differences in marketization scores.
The correlation coefficients of the key regression variables are presented in Table 3. Overall, the variable correlations are low, except the correlation between MTB and LEV. We removed one of them from the regressions, and the results do not change qualitatively. Therefore, we keep both variables in the regressions.
 
Results
Hypothesis tests
Regressions of the coefficient test results are presented in Table 4. We performed a Hausman test to compare the random effects model and ordinary least squares model with the fixed effects model. The
 
 
Table 1. sample selections.
Variables Central SOEs Local SOEs & Non-SOEs
 
initial sample (firm-year) 1300 3900 Deletions due to missing data for:
ACFOi,t −442 −1393
ROAi,t−1 −6 −20
Market to book ratioi,t−1 −194 −641
Marketizationi,t−1 −11 −18
final sample 647 1828
total final sample (firm-year) 2475
notes: the initial sample consists of 100 central soEs, 100 local soEs and 200 non-soEs for the period of 2003–2015 (400 firms × 13 years).
the variables are defined in table 2.
 
Table 2. statistic description.
Panel A: Continuous variables
Central SOEs (n = 647) Local SOEs & Non-SOEs (n = 1828)
 
 
APRODi,t −0.002 0.000 0.093 −0.007 −0.001 0.170 0.508
ACFOi,t 0.008 0.001 0.081 0.004 0.005 0.112 0.517
ADISXi,t −0.002 0.003 0.051 −0.007 0.002 −0.107 0.256
SIZEi,t−1 10.560 10.339 2.334 9.042 8.881 1.556 0.000
ROAi,t−1 0.052 0.034 0.067 0.092 0.070 0.105 0.000
LEVi,t−1 1.093 0.617 4.519 0.591 0.394 0.783 0.000
MTBi,t−1 3.598 2.118 13.797 3.915 2.950 3.225 0.361
AUDITORi,t−1 0.586 1.000 0.493 0.358 0.000 0.479 0.000
Marketizationi,t−1 8.934 9.630 2.320 8.952 9.360 2.074 0.839
Panel B: Discrete variables
Value ob. Percent Value ob. Percent
CSOEi 1 647 26.14 0 1828 73.86
HighGDPGrowtht−1 1 782 31.60 0 1693 68.40
LowGDPGrowtht−1 1 768 31.03 0 1707 68.97
AUDITORi,t−1 1 1033 41.74 0 1442 58.26
 
Variable definition: Accrual EM: is abnormal accrual estimated based on modified Jones’model (1991); APROD: abnormal production following roychowdhury (2006); ACFO: abnormal cash flow from operations following roychowdhury (2006) multiplied by −1; ADISX: abnormal discretionary expenses following roychowdhury (2006). SIZE: the natural logarithm of total assets; ROA: calculated as the net income in the current year divided by the beginning balance of total assets in this year; LEV: calculated as long-term debt at current year-end divided by book value of equity at current year-end; MTB: calculated as the market value of the firm divided by the book value of the firm in this year; AUDITOR: a dummy variable that is equal to 1 if the auditor is the big-8 auditor in china as identified by cheng, Wang, and Wei (2015), and 0 otherwise; Marketization: the marketization development index of the province the firm is located (fan 2011); CSOE: a dummy variable that takes value of 1 if the company is included in the china security index (csi) state-owned Enterprises 100 index (central soE 100), and 0 otherwise; HighGDPGrowth: a dummy variable that takes value of 1 for years 2005, 2006, 2007 and 2010, and 0 otherwise; LowGDPGrowth: a dummy variable that takes value of 1 for years 2008,
2011, 2012 and 2015, and 0 otherwise.
 
test results favor the fixed effects model (untabulated). Therefore, we chose to use fixed effects model and controlled industry fixed effects to account for all industry level factors.
In Table 4, dependent variables for the four regressions are the four earnings management measures: Accrual EM, APROD, ACFO and ADISX. The adjusted R-squares based on real earnings management measures are comparable to those of prior studies on Chinese SOE earnings management (Chen et al. 2011; Ding, Zhang, and Zhang 2007; Liu and Lu 2007; Wang and Yung 2011).
As discussed in detail above, positive values of Accrual EM increase reported enterprise earnings. However, Accrual EM has no effects on GDP. Positive values of real earnings management variables APROD and ACFO increase GDP as well as enterprise reported earnings. Positive values of ADISX also increase GDP but decrease enterprise earnings.
For  the  regression  on  Accrual  EM,  the  regression  coefficients  for  interaction   terms CSOE × HighGDPGrowth and CSOE × LowGDPGrowth are both negative but not significant. This result indicates that CSOEs do not engage in additional accrual-based earnings management activities in extremely high and low GDP growth years relative to the control companies.
For the regression on abnormal production-based real earnings management (APROD), regression coefficients for the interaction terms CSOE × HighGDPGrowth are negative at −0.043 and significant at 1%. This means that during high GDP growth years, Chinese CSOEs decrease production levels, resulting in lower contributions to national GDP and lower reported firm earnings. Compared to the summary statistics of APROD in Panel A of Table 2, this result suggests that CSOEs on average reduced abnormal production by 0.46 of standard deviation in extremely high GDP growth years to reduce earnings as well as contributions to GDP. The regression coefficient for the interaction term CSOE × LowGDPGrowth is positive at 0.035 and significant at 5%, meaning that during low GDP
 
Table 3. regression variable correlations.
CSOEi Accrual EMi,t     APRODi,t ACFOi,t ADISXi,t        SIZEi,t−1        ROAi,t−1        LEVi,t−1         MTBi,t−1        AUDITORi,t−1        Marketizationi,t−1        HighGDP  Growtht−1 Accrual EMi,t −0.135 1.000
(0.000)
APRODi,t 0.013 0.100 1.000
(0.508) (0.000)
ACFOi,t 0.013 −0.088 −0.544 1.000
(0.517) (0.000) (0.000)
ADISXi,t −0.023 0.102 −0.485 0.090 1.000
(0.256) (0.000) (0.000) (0.000)
SIZEi,t−1 0.349 −0.411 −0.010 0.023 0.046 1.000
(0.000) (0.000) (0.628) (0.253) (0.022)
ROAi,t−1 −0.181 0.297 −0.189 0.194 −0.033 −0.255 1.000
(0.000) (0.000) (0.000) (0.000) (0.105) (0.000)
LEVi,t−1 0.091 0.056 0.045 −0.025 0.029 0.106 −0.151 1.000
(0.000) (0.005) (0.026) (0.216) (0.153) (0.000) (0.000)
MTBi,t−1 −0.018 0.277 −0.041 0.042 −0.035 −0.181 0.116 0.831 1.000
(0.361) (0.000) (0.044) (0.035) (0.084) (0.000) (0.000) (0.000)
AUDITORi,t−1 0.203 −0.144 −0.008 −0.010 −0.047 0.405 −0.163 0.037 −0.078 1.000
(0.000) (0.000) (0.699) (0.620) (0.019) (0.000) (0.000) (0.069) (0.000)
Marketizationi,t−1 −0.004 −0.048 0.034 −0.035 −0.004 0.121 0.002 −0.002 −0.022 0.122 1.000
(0.839) (0.016) (0.092) (0.084) (0.856) (0.000) (0.931) (0.927) (0.266) (0.000)
HighGDPGrowtht−1 0.033 0.037 −0.018 0.027 −0.022 −0.174 0.039 0.023 0.105 −0.045 −0.132 1.000
(0.103) (0.069) (0.363) (0.178) (0.272) (0.000) (0.053) (0.253) (0.000) (0.026) (0.000)
LowGDP Growtht−1 −0.039 −0.017 0.022 −0.040 0.007 0.042 0.013 −0.015 −0.078 0.061 0.076 −0.456
(0.051) (0.406) (0.280) (0.046) (0.735) (0.039) (0.521) (0.467) (0.000) (0.002) (0.000) (0.000)
notes: the variables are defined in table 2.
the p-values of correlations are presented in parentheses.
Table 4. test of hypothesis.
Accrual EM APROD ACFO ADISX
 
CSOEi −0.009* 0.005 −0.010 −0.003
(1.66) (0.38) (1.26) (0.39)
HighGDPGrowtht−1 0.009** 0.009 −0.002 −0.001
(2.30) (1.03) (0.38) (0.17)
LowGDPGrowtht−1 0.003 −0.005 0.001 0.003
(0.82) (0.65) (0.18) (0.52)
CSOEi × HighGDPGrowtht−1 −0.002 −0.043*** −0.000 0.011
(0.26) (2.59) (0.01) (1.08)
CSOEi × LowGDPGrowth t−1 −0.005 0.035** 0.019* −0.007
(0.65) (2.11) (1.68) (0.71)
SIZEi,t−1 0.014*** −0.011*** −0.008*** −0.000
(10.93) (3.82) (4.27) (0.19)
ROA i,t−1 0.410*** −0.241*** −0.196*** −0.027
(24.72) (6.83) (8.11) (1.19)
LEV i,t−1 −0.010*** 0.015*** 0.007*** −0.008*** (7.16) (5.26) (3.51) (4.32)
MTB i,t−1 0.003*** −0.005*** −0.003*** 0.003*** (7.30) (5.33) (3.90) (4.28)
Marketization i,t−1 −0.002** 0.004** 0.002* 0.000
(2.40) (2.22) (1.82) (0.11)
AUDITOR i,t−1 −0.002 −0.007 0.002 0.016***
(0.51) (1.07) (0.42) (3.63)
Intercept −0.144*** 0.091*** 0.074*** −0.001
(10.15) (3.02) (3.61) (0.04)
Industry fixed effect included included included included
R2 0.27 0.06 0.05 0.02
N 2475 2475 2475 2475
notes: the variables are defined in table 2.
*p < 0.1; **p < 0.05; ***p < 0.01.
 
growth years, Chinese CSOEs increase production levels, leading to increases in their contributions to GDP as well as increases in firm earnings. This result suggests that CSOEs on average increased abnormal production by 0.38 of standard deviation in extremely low GDP growth years to increase earnings as well as contributions to GDP. Taken together, these results support the predictions of our hypothesis, i.e. indicate that CSOEs engage in production-based real earnings management to smooth GDP. Additionally, the CSOEs’ production-based earnings management activities in extreme high and low GDP growth years are likely to have resulted in smoother CSOE earnings as well.
For the regression on abnormal CFO-based real earnings management (ACFO), the regression coefficient of the interaction term CSOE × HighGDPGrowth is negative but not significant. However, the regression coefficients of the interaction term CSOE × LowGDPGrowth are positive at 0.019 and marginally significant at 10%. Although results are not significant to support the prediction that Chinese CSOEs engage in GDP decreasing CFO-based earnings management in high GDP growth years, they are significant to indicate that CSOEs do engage in GDP increasing CFO-based earnings management in low GDP growth years. One possible explanation for the asymmetric results with CFO-based earnings management is, that for the purpose of maintaining stability, it is more impor- tant to avoid perceived economic decline than economic boom. In addition, as discussed previously, CFO-based real earnings management could cause real economic loss due to the additional discount offered. CSOEs may be cautious and more likely to use it to boost the economy than to slow down the economy. The results partially support the predictions of our hypothesis.
For the regression on the discretionary expense-based earnings management measure, ADISX, the regression coefficients of interaction terms CSOE × HighGDPGrowth and CSOE × LowGDPGrowth are both insignificant. This result indicates that CSOEs do not engage in more discretionary expense-based real earnings management activities during extremely high and low GDP growth years compared to the control companies.
 
In combining the results of the regressions on all four earnings management measures, we make the following observations. First, Chinese CSOEs do engage in additional earnings management activities compared to the control companies in extreme high and low GDP growth years. Second, the docu- mented earnings management activities of Chinese CSOEs are consistent with the predictions of our hypothesis, i.e. Chinese CSOEs are motivated by state-level incentives to engage in earnings manage- ment to mitigate GDP volatility. Further, Chinese CSOEs select earnings management approaches with consideration of their impact on enterprise earnings as well. Chinese CSOEs prefer earnings management approaches that smooth both GDP and firm earnings during extremely high or low GDP growth years (e.g. APROD and ACFO based earnings management). Finally, Chinese CSOEs apply earnings management approaches to manage earnings effectively. Of the two earnings management approaches that satisfy both state- and institution-level incentives (APROD and ACFO), the results for the more effective earnings management approach is much stronger (i.e. the results of APROD). Consistent with the prior literature, the control variables have significant impact on the earnings management metrics, but the signs differ for different earnings management metrics (Wongsunwai
2013).
 
Robustness tests
Our study focuses on the earnings management of Chinese CSOEs due to their critical roles in the Chinese economy and close relationships with the central government. We believe that the interests of Chinese local SOEs, which are controlled by local governments that are primarily motivated by regional interests, are less aligned with those of the central Chinese government. Anecdotal evidence suggests that local SOEs pursue diverse and sometimes conflicting interests (Walder 1995). We thus included local SOEs and non-SOEs in our control sample to highlight the homogeneity of Chinese CSOEs in terms of the unique earnings management incentives that they share. However, it may be argued that state-level incentives can trickle down to local SOEs through the central government’s effects on local governments. To test this possibility and possible misspecification of the model associated with it, we introduce an additional dummy variable, LSOE, to distinguish local SOEs from non-SOEs in all four regressions. LSOE is a dummy variable that takes a value of 1 if the firm is included in the CSI Local SOEs 100 Index and takes a value of 0 otherwise. The regression results are presented in Table 5. In Table 5, the regression coefficients of test variables CSOE  ×  HighGDPGrowth  and CSOE × LowGDPGrowth for all four regressions remain qualitatively the same, though a marginally significant regression coefficient for CSOE × LowGDPGrowth drawn from the regression on ACFO is now no longer significant. The regression coefficients of variables LSOE × HighGDPGrowth and LSOE × LowGDPGrowth for all four regressions are not significant, indicating the LSOEs do not engage in more earnings management to smooth GDP in extremely high or low GDP growth years
relative to non-SOEs.
Interestingly, the coefficient for the Local SOE dummy variable, LSOE of the regression on accrual EM is positive and significant, meaning that local SOEs generally engage in earning increase manage- ment that do not affect GDP. This may be related to competition between the numerous local SOEs for resources and recognition. Why and how local SOEs engage in earnings management is beyond the scope of this paper. We encourage further research on this relevant issue.
Additional tests of robustness are conducted by first removing CSOEs from all regressions and by then focusing on only comparisons of local SOEs and non-SOEs. The results, untabulated, show that the earnings management activities of local SOEs and non-SOEs are not significantly different in extremely high or low GDP growth years. We also run all the regressions with local SOEs removed and focus only on comparisons of CSOEs and non-SOEs. The corresponding results, untabulated, confirm our main finding that CSOEs manage production to smooth both GDP and earnings during extremely high or low GDP growth years. Overall, our finding that Chinese CSOEs are motivated by both state- and institution-level incentives to manage earnings is strong and robust.
 
Table 5. Additional test – CSOEs, Local SOEs, and Non-SOEs.
Accrual EM APROD ACFO ADISX
 
CSOEi 0.004 0.007 0.015* −0.004
(0.61) (0.54) (1.65) (0.51)
HighGDPGrowtht−1 0.012** 0.008 0.002 0.000
(2.32) (0.73) (0.23) (0.05)
LowGDPGrowtht−1 0.007 −0.004 −0.002 −0.005
(1.33) (0.33) (0.29) (0.67)
CSOEi × HighGDPGrowtht−1 −0.006 −0.042** 0.000 0.010
(0.72) (2.35) (0.00) (0.86)
CSOEi × LowGDP Growtht−1 −0.008 0.033* −0.018 −0.001
(1.03) (1.87) (1.48) (0.06)
LSOEi 0.025*** 0.005 0.008 −0.004
(4.24) (0.40) (0.97) (0.51)
LSOE × HighGDPGrowtht−1 −0.011 0.001 −0.001 −0.003
(1.39) (0.05) (0.08) (0.32)
LSOE*LowGDPGrowth t−1 −0.009 −0.005 0.003 0.020
(1.13) (0.31) (0.23) (1.80)
SIZEi,t−1 0.012*** −0.011*** 0.007*** −0.001
(9.01) (3.74) (3.56) (0.27)
ROA i,t−1 0.413*** −0.241*** 0.197*** −0.026
(24.98) (6.81) (8.15) (1.18)
LEV i,t−1 −0.010*** 0.016*** −0.007*** −0.008*** (6.95) (5.26) (3.40) (4.31)
MTB i,t−1 0.003*** −0.005*** 0.002*** 0.003*** (7.11) (5.33) (3.79) (4.26)
Marketization i,t−1 −0.001 0.004** −0.002 0.000
(1.41) (2.26) (1.48) (0.14)
AUDITOR i,t−1 −0.003 −0.008 −0.003 0.016***
(0.99) (1.10) (0.56) (3.62)
Intercept −0.142*** 0.091*** −0.072*** 0.002
(9.89) (2.97) (3.40) (0.10)
Industry fixed effect included included included included
R2 0.30 0.06 0.06 0.02
N 2475 2475 2475 2475
notes: the variables are defined in table 2.
*p < 0.1; **p < 0.05; ***p < 0.01.
 
Conclusion
This study empirically investigates the earnings management of Chinese CSOEs. We found that CSOEs engage in certain forms of real earnings management activities to help reduce GDP volatility in low (high) GDP growth years, which is consistent with the principles of the extended Alignment Effect that high level state ownership induces strong earnings management incentives at state level. Such earnings management activities help lessen the appearance of economic volatility, which can be con- ducive to state political and social stability. In addition, the earnings management methods chosen by Chinese CSOEs for state-level incentives also result in reduced enterprise earnings volatility. This study contributes to the SOE earnings management literature by extending the existing Alignment Effect argument and by contributing to our understanding of SOE earnings management incentives and opportunities. In doing so, we present a new perspective that other scholars can use to examine other non-traditional earnings management incentives that SOEs may be subjected to.

最新更新

热门推荐

[经济理论]宝钢集团有限公司成本管
摘 要 在这个经济技术快速发展的时代,钢铁企业面临的市场竞争日益激烈,艰难的经营环境使国内钢铁企业之间的竞争环境更是不容乐观。钢铁企业为了适应多变的竞争环境,必须不断对传统成本管理进行改革,同时也必须持续不断的创新摸索,寻求更加符合适合企业本身的成本管理办法,来应对日益激烈的竞争环境。新的管理环境下,宝钢集团的成本管理方法需要根据企业现...[全文]
[会计审计]佰酚佰广告有限公司会计
摘 要 随着信息技术以势不可挡之势迅速地发展,其在会计中的应用也已经成为了一种必然的趋势。大多数企业也已经逐步实现了财务和业务上的一体化,因此,内外部会计环境发生了重大变化,信息共享对作为管理信息系统重要组成部分的会计信息系统的内部控制提出了挑战。挑战来了自然要迎战,所以我们要好好建立会计信息系统内部控制的一个大框架,是非常有必要的。本...[全文]
[农业经济]傲源有限责任公司筹资风
摘 要 目前,我国房地产企业筹资风险存在的原因主要有:资本成本较大、受宏观政府调控影响范围广、项目完工周期长、资金总额流动率低。但是房地产行业照比其他行业也存在一些优点,现在房地产行业整体处在蓬勃发展的阶段,行业内部蕴藏很多待发现的潜力,以及拥有许多待挖掘的发展空间。所以对房地产公司筹资风险问题研究还是必要的。 本文围绕着傲源有限责任公司...[全文]
[经济理论]鞍山大厦有限责任公司并
摘 要 对于企业来说,如果想要扩大经营规模,一个有效途径就是企业并购。并购是十九世纪出现的,起缘于西方资本主义国家。截止到今天,西方世界先后经历了四次并购浪潮,当前正处于第五次浪潮当中,近年来,我国社会主义市场经济发展速度越来越快,在世界上的影响力也与日俱增,很多国内企业为了提高发展速度也开始尝试使用并购这种资本运用手段。目前,并购是企...[全文]
[会计审计]非营利组织财务管理存在
摘 要 随着当今世界格局的不断变化与发展,社会组织日渐呈现多样化趋势,营利组织与非营利组织并存,并且非营利组织的数量不断增加,占据了社会组织的绝大部分,但是随着不断发展的过程中就会暴露出很多的问题。相比较来说我国非营利组织的覆盖面积相当广泛,在社会生活中也帮助政府承担一部分社会的责任,我们都知道非营利组织它并不会存在利润,它会无偿性的进...[全文]
[经济理论]中海油田服务股份有限公
摘 要 在新中国建立之后,制定了一五计划,国家的重工业发展日渐强大,国民经济水平越来越高,其中我国的石油化工企业也越来越多,发展至今在国内以及国际上都有着一定的地位,为国民经济的增长做出了巨大贡献。但是石油企业在生产经营过程中无法避免的对环境造成一定的伤害,同时也对人类的生活以及身体健康造成伤害。近几年来,中国石油企业漏油等相关的环境污...[全文]
关闭窗口 论文咨询