英国小企业CEO薪酬与公司业绩关系研讨.doc

淮 阴 工 学 院
毕业设计(论文)外文资料翻译

系 （院）：
经济管理学院
专 业：
财务管理
姓 名：
高施佳
学 号：
3062106438
外文出处：
xspringerlink.lib.tsinghua.edux/content/?k=does+the+market+for+top+executi
(用外文写)
ves+work%3f
附 件：
1.外文资料翻译译文；2.外文原文。

指导教师评语：
外文资料与课题相关性较强，翻译符合原文意思，语言表达顺畅，格式规范，符合外文翻译要求。

年

月

日
签名：

注：请将该封面与附件装订成册。附件1：外文资料翻译译文

市场是否对高管有作用？
——英国小企业CEO薪酬和公司业绩关系研究
Martin J. Conyon
Daphne Nicolitsas
摘要：本文介绍了我们对中小型制造业公司的劳动力市场管理运作情况的调查结果。在研究过程中我们使用大约40家的公司来作为研究薪酬与公司业绩敏感性的样本。在这类型的企业管理中首席执行官的6位数薪酬远远低于大型上市公司，但即使如此也比我们预期的要高很多。研究分析过程中，我们发现了一些数据，表明在小企业中任职期间的CEO薪酬与其自身业绩相关，CEO薪酬与公司业绩敏感性较强。

1 引言
在英国，高级管理人员的工资上涨已经引起社会公众的舆论，同时也引起了其他国家的关注。而在对这个问题上，真正的关注点不在于薪金水平的高低，而是他们未能反映薪酬与业绩之间的关系。（经济学家报纸，1995年6月3日）
在合理的市场环境下，薪酬的上升趋势，可以通过调整劳动供给曲线来稳定，也就是可通过左移或向上移动劳动供给曲线来实现。因为由于管理培训和商学院的课程扩散，供应给社会的合格管理人员人数就可能得以增加，劳动供给曲线就有可能转移到了右侧，因此，在一定范围内至少有一个管理人员愿意增加工资的数量来做这项工作；而对于上移需求曲线时，管理人员的边际产品必须增加，也就是说，管理者必须要更加富有成效。管理者为提高他们的生产力可以通过控制更多的资源或通过控制一个常数实现。
为了给“市场是否对高管期作用？”的问题提供一个答案，我们在这里要探讨以下方面：也就是我们机构测试的双重假设模式：（一）行政赔偿和经济绩效呈正相关；（二）业绩越差的公司，其CEO更迭的可能性越大。英国和美国同时对前一个命题进行了测试，总的来说 ，使用的数据都来自大型上市公司。
本文强调了在过去研究中受到较少关注的问题，英国小企业中就执行确定工资为重点的薪酬战略经验（一个例外：斯托里，沃森，威尼司克，1995年）。其次，是创新CEO薪酬的焦点。受詹森和墨菲（1990）论文的影响，以美国大公司的数据为研究基础，说明了提供统一的就业解雇规范化行为对管理人员和所有者造成了威胁。他们发现：公司高管薪酬与业绩关系比代理理论预测的更大。然而，已发表的证据证明：在英国小企业中缺乏薪酬——业绩关联的执行机制，此外，研究发现现实中存在大量比小公司更有效的相关数据。目前的研究阶段是希望通过增加对劳动力市场管理纪律的了解，评估小公司的经理在公司业绩不佳的情况下，薪酬上涨的可能性。
本文的其余部分组织如下：第２节进行分析；第３节简要介绍使用的数据，并提出管理人员的薪酬和公司规模关系的一些事实问题；第４节介绍使用的框架和性能；第５节介绍公司业绩与薪酬之间的关系。
2 分析框架
上世纪六十年代开始关注对高管薪酬的理论研究，但是随着信息经济的发展，它只持续了一段时间。从罗森（1990）的研究分析中可以找到所涉及问题的最佳解释和发展的模型。
本文以大量的工作理论为基础，够建了实证分析的框架。
2.1 企业规模与高管薪酬之间的关系
从CEO是公司的最高管理员的角度讲，罗森（1990）从高管经营业绩水平较低的实证研究中表明了高管的薪酬与组织规模大小的紧密关联性。我们对人才内部结构做出假设：CEO薪酬是与控制在每个层次（s）和对层（不适用若干层跨度增加）参与，这样就更能提高个人层次对这种结构的影响。因此，所期望的是有更多的管理信息是反映高管薪酬与企业规模大小是呈正相关的。（见罗森，1990）。
这种假设已经在对数形式的线性回归模型研究中得到了验证：
ln Wit =a + b 1n xit + gt + dj + ui + hit (1)
其中：
i表示该公司，t表示时间；
W表示CEO实际薪酬;x表示衡量的大小，一般是由其总就业人数，实际资产总额或实际销售组成;
t表示假设的受时间控制的总变化量;
j表示假设某一行业的生产仅限于受企业规模大小控制；
Ui表示企业受不确定性因素的影响;
it表示是由企业规模和时间带来的无法预测的影响。
2.2 薪酬和能力之间的关系
在研究模型中，我们一开始假设高管的个人能力对于高管薪酬的影响是不可预测的，可见，个人业绩的表现过程对薪酬结构是有很重要的影响的。所以设W，一些业绩表现；Y，特别是努力（e）和自然状态（e）作用于CEO的薪酬（尤其见：哈特，1995函数）。其中，“自然状态”表示大环境相近，即有类似的企业规模与生产力。
W = f(Y); f ¢ > 0 (2)
Y = f(e, e); f¢1 > 0, f¢2 ? 0 (3)
思考过程是这样的：由于是自然状态下的企业，所以受管理者个人业绩以外因素作用的差异程度就显得相对较小，这样也就更利于体现个人能力对企业的影响力程度。但是这个问题比较复杂，因为即使某一个行业可能受到宏观经济环境影响程度类似，但接触不利冲击的程度的大小却是一个由不同管理决定而导致。所以，这个设计就应该要达到薪酬结构风险的权衡和激励的最佳状态两方面的要求。
通过对上面（1）公式的实证研究分析 ，可估算确定薪酬和绩效之间的关系如下：
ln Wit = ai + b1 1n vi, t – 1 b2 1n xi, t – 1+ gt ´ t + dj ´ t + uit (4)
其中：
I 表示特定企业中固定不变的影响（常量）；
vi, t – 1表示一些业绩指标（如每名雇员带来的利润，会计收益率，股东回报率，销售，企业的利润相对于行业的利润水平）；
xi, t – 1 表示实施范围，uit表示是有关企业和个人时期都无法观察到的影响。其余的变量在方程（1）段中已经说明。
自从我们这一年来的研究发现，业绩与薪酬的关系越来越紧密，这很有可能他们之间存在着内源性。也就是说，薪酬反作用于企业绩效。此外，也可能是CEO的薪酬决策是在本财政年度结束之前决定，这就表明了利用右移变量可以使得内部更合理化。
2.3 营业额与业绩之间的关系
股东可以通过利用薪酬与绩效之间的关系，进行激励管理。就像在引言中指出的一样，在对英国的样本研究中表明，英国一直关注股东利益的最大化条件下，薪酬和绩效之间薄弱的联系可能导致公司的激励管理机制失效。然而，可以通过其他手段来规范管理行为。进一步来说，是所有者可以通过评价CEO在过去所作出的业绩来进行管理。
这个假设的结果是通过CEO轮换制来实现的：如果在任期间业绩表现良好，那么应该得到相应回报，反之亦然，不管其他原因，公司业绩差将直接导致首席执行官的更换。换言之，CEO将面临被解雇的威胁（见詹森，墨菲 1990; 威斯巴赫 1988）。
一个基本的模式是如下的形式：
Pr(DCEOit) = F(xb) =exp(xb)/(1 + exp xb) (5)
其中：
CEOit表示一位首席执行官以新的或其他方式存在；
在x向量中包括了在过去的P时期公司的平均表现；
加权是表示首席执行官的特点，qkt,一些不随时间变化的（如出生日期， 国籍）和有些是随时间变化（如选择任用，教育）的量。
关于预期业绩计量的符号为负。然后，现任首席执行官的业绩好，就获得嘉奖；业绩差，就必须受到惩罚，也就意味着将重新选举新的首席执行官。
在第4部分和第5部分将使用数据进一步验证以证明上面的假设。
2.4 最近的薪酬和营业额的调查数据
下面我们将提供一个最新研究数据，主要是关于英国薪酬和营业额调查的研究，这有助于证明我们的研究结果。
卡尼欧，格雷格和梅钦（1995年）审查了关于董事支付与公司治理的一些最新数据。这项研究中的大样本大多关系到上市公司。研究得出的主要结论是：高收入的董事和其采取的公司治理措施，薪酬与个人业绩联系很少。例如，格雷格，梅钦和斯曼斯基（1993年）使用的样本是1983年至1988年期间288个英国上市公司找到的一个股东回报期间的弹性薪酬为0.03。而对于1989-91年期间相同的取样结果显示企业的这个弹性为零。结果表明符合美国研究的结论。詹森和墨菲（1990）有影响的论文中也报告说，薪酬与绩效之间的敏感度低，股东财富每增加1000美元CEO的财富增加 3.25美元。
在上述研究中存在着的一个不足之处就是：在表明业绩向下的关系时，可能是因为董事担任CEO时股票期权是不算在薪酬尺度之内引起的，这就说明存在着薪酬的偏置区间。布鲁斯和巴克（1996）使用1990年60家大公司的FT-SE100指数进行研究，他们发现，当以包括期权的价值作为薪酬时，管理人员的薪酬与业绩关系表现还是比较敏感的。

附件2:外文原文(复印件)

Does the Market for Top Executives Work ?
CEO Pay and Turnover in Small U.K. Companies

Martin J. Conyon
Daphne Nicolitsas

ABSTRACT. This paper presents the results of our investigation into the operation of the managerial labour market in small and medium sized manufacturing companies. Using a sample of some 40 or so companies we study the sensitivity of managerial pay and tenure to company performance. Managerial pay in this type of firms is much lower than the six digit figures quoted for large public companies and CEO turnover is much higher than we expected. We find some evidence to suggest that pay in small companies is sensitive to sales growth and that CEO tenure in public companies responds to the growth in sales.

1. Introduction
The soaring pay of top executives has caused a public outcry in Britain, and rumblings of discontent in other countries The real problem is not the level of salaries, but their failure to reflect performance. The Economist Newspaper, 3 June 1995 (Our emphasis added)

Within the context of a market environment an upward trend in pay could be justified either by an upward shift of the labour demand curve with a constant labour supply curve, or by a shift to the left of the labour supply curve. The supply of suitably qualified managers has probably increased given the proliferation of management training and Business School courses. So the labour supply curve is likely to have shifted tothe right. Thus at least within a certain wage range there is an increase in the number of managers willing to do the job. For the demand curve to have shifted upwards the marginal product of managers must have increased; that is, managers must have become more productive. Managers can increase their productivity either by controlling more resources or by controlling a constant amount of resources more effectively.
We here address the following question: Does the market for top executive work? In order to provide an answer to this we test the twin predictions of the agency model that (i) executive compensation and economic performance are positively correlated and (ii) that poor company performance results in a higher probability of CEO turnover. The former proposition has been tested for both the U.K. and the U.S. using, in general, data for large quoted companies. A recent review of this literature is presented in Conyon, Gregg and Machin (1995).
This paper augments the U.K. empirical literature on the determination of top pay by focusing on executive wage setting in small companies; an area that has received comparatively little attention (an exception is Storey, Watson and Wynarczyk, 1995). The second innovation herein is the focus on CEO turnover. The influential paper by Jensen and Murphy (1990), based on data for large U.S. firms, illustrates that the discipline provided by the threat of job dismissal acts to align the objectives of managers and owners. They find that management turnover is greater in companies with poor economic performance as predicted by the agency theory. There is, however, a paucity of published evidence on the executive turnover mechanism in U.K. companies. Furthermore, the studies that do exist relate to large rather than small firms (see Cosh and Hughes, 1995).The current paper hopes to add to the understanding of managerial labour markets by assessing whether managers in small firms are disciplined for poor company performance by increased probability of turnover.
The rest of this paper is organised as follows; Section 2 sets out the framework used in the analysis, Section 3 briefly describes the data used and presents some facts on the issues of interest, Section 4 looks at the relationship between management pay and company size and performance. Section 5 presents information on the relationship between turnover and company performance. Section 6 concludes.
2. A framework for analysis
Theoretical interest in the market for top executives started in the 1960s but only took off recently in the last decade or so following developments in information economics. The best exposition of the issues involved and the development of the appropriate models can be found in Rosen (1990). This section draws heavily on this work to outline the theoretical framework which forms the basis for the empirical analysis.

2.1. The relationship between executive pay and firm size

Starting from the premise that firms are organized in a hierarchical structure with the CEO at the top, Rosen (1990) has shown that executive pay and organisation size are positively correlated since the marginal product of the CEO is reproduced at each lower level. Assuming that the distribution of talent is such that more able individuals are higher up the hierarchy the implications of this structure are that CEO pay is increasing with the span of control at each layer of the hierarchy (s) and the number of layers (n) involved. So, the expectation is that executive compensation and company size are positively correlated reflecting the greater managerial talent of those at the apex of the company (see Rosen, 1990).
In empirical work this hypothesis has been investigated using log-linear regressions of the form.
Ln Wit =a + b 1n xit + gt + dj + ui + hit (1)
where:
i represents the firm, and t the year;
W is CEO real pay;
x represents size measured, in general, by total employment, total real assets or real sales;
gt are time dummies to control for aggregate variations;
dj are industry dummies to control for factors specific to the industry in which the firm belongs;
ui are unobservable firm-specific effects; and
hit represents unobservable effects pertaining to both individual firms and time periods.

2.2. The relationship between pay and performance

In modelling the pay of top executives we start with the presumption that managers’ effort is unobservable. Thus pay has to be linked with a performance measure which in turn is a function of effort. Thus W, the CEO’s pay is a function of some performance measure Y which, in turn, is a function of, inter alia, effort (e) and the state of nature (e) (see, inter alia, Hart, 1995). Where th‘state of nature’ represents both firm-level and industry-wide shocks.
W = f(Y); f ¢ > 0 (2)
Y = f(e, e); f¢1 > 0, f¢2 ? 0 (3)
The idea is that the state of nature affects all firms and thus differences in relative performance reveal information about the manager’s effort. The issue is more complicated, however, since although it is not possible for a single firm to influence the macroeconomic environment, the level of exposure to different adverse shocks is a management decision. The design of the payment scheme should thus achieve the optimal trade-off between risk and motivation.
In the empirical analysis the interest is in identifying the relationship between pay and performance by estimating equations of the form.1
ln Wit = ai + b1 1n vi, t – 1 + b2 1n xi, t – 1+ gt ´ t + dj ´ t + uit (4)
where:
ai represents firm-specific unobservable fixed effects;
vi, t – 1 represents some measure of performance (e.g. profits per employee, accounting rates of return, shareholder rate of return, sales, firm profits relative to industry profits);
xi, t – 1 measures size, uit represents unobservable effects pertaining to both individual firms and time periods. The rest of the variables are as in equation (1) above.

The performance and size variables are lagged by a year since we are interested in the sensitivity of remuneration to past achievements whereas the current value of size and performance is likely to be endogenous. That is, corporate performance could also be affected by pay. Additionally, since it is likely that the pay of the CEO is decided before the end of the current financial year this provides an additional rationale for using lagged right hand side variables.

2.3. The relationship between turnover and performance

The relationship between pay and performance is only one way by which the shareholders can motivate management. Since, as noted in the introduction, the evidence for the U.K. suggests that the link between compensation and performance has been at best weak, this has led to concerns that the incentive structure faced by managers at leading U.K. companies in pursuing shareholders interests may not be effective. It is, however, possible to regulate management behaviour by other means. More specifically the owners could monitor management by making tenure a function of past performance.
The proposition under scrutiny then is whether good performance is rewarded by re-electing the CEO and, vice versa, whether bad company performance leads to replacement of the CEO. In other words the issue is whether the threat of dismissal could act as a discipline device (see Jensen and Murphy, 1990; Weisbach, 1988).
A basic model is thus of the form:
Pr(DCEOit) = F(xb) =exp(xb)/(1 + exp xb) (5)
where:
DCEOit indicates the presence or otherwise of a new CEO;
The x vector includes a measure of average firm performance during the last p periods;
, and CEO characteristics, qkt, some of which are time invariant (e.g. date of birth, nationality) and some of which are time varying (e.g. alternative appointments, education).

The expected sign on the performance measure is negative. That is good performance is rewarded by re-electing the current CEO and bad performance is ‘punished’ by bringing in a new CEO.
Sections 4 and 5 use the data described next to test the above hypotheses.

2.4. Some recent evidence on pay and turnover

Below we present a succinct review of some of the recent evidence, predominantly for the U.K., on executive compensation and management turnover to set the context within which to place our results.
Conyon, Gregg and Machin (1995) review some of the recent evidence on directors’ pay in relation to corporate governance. Most of the studies reviewed relate to samples of large quoted companies. The main conclusion is that the estimated link between compensation of the highest paid director and market measures of company performance is small. For instance, Gregg, Machin and Szymanski (1993) using a sample of 288 U.K. quoted companies find an elasticity of pay to shareholder returns of 0.03 for the period 1983– 88. Whereas for the same sample of firms over the period 1989–91 this elasticity is zero. The results appear consistent with the U.S. evidence. The influential paper by Jensen and Murphy (1990) also reports a low sensitivity of pay to company performance; CEO wealth increases by $3.25 for every $1000 increase in shareholder wealth.
One of the disadvantages of the