1、 1 外文翻译 原文 The balance of power in closely held corporations Material Source: Journal of Financial Economics 58 (2000) Author: Morten Bennedsen , Daniel Wolfenzon The vast majority of firms in developing and transitional economies do not have their shares traded in an exchange.Even in the U.S.,out o
2、f almost 4 million corporations that fled taxes in 1993(Statistics of Income),a mere7842 corporations were listed in the NYSE,Nasdaq and Amex combined(1994 Nasdaq Fact Book). Closely held corporations typically have an ownership structure comprised of several signifcant shareholders. However, the co
3、rporate finance literature has focused on firms with either a dispersed ownership structure (Berle and Means, 1932;Grossman and Hart,1980)or a single controlling shareholder(Jensen and Meckling, 1976; Shleifer and Vishny, 1986; Burkart et al., 1997). In the former case, shareholders are too small an
4、d disorganized to impose their will. As a result, control resides in the hands of the manager.In the latter,the dominant shareholder dictates corporate policy either by managing the firm directly or by closely monitoring the managing team. Remaining shareholders lack either the power or the incentiv
5、es to oppose the controlling shareholders decisions. In this paper,we focus on firms in which shareholders are large enough not to surrender control to the manager even though no individual shareholder is large enough to control the firm alone.In these firms,corporate policy is the result of interac
6、tion among shareholders. In the model we develop, an initial owner chooses an ownership structure with multiple large shareholders to prevent a single shareholder from taking unilateral actions that might hurt other shareholders. For instance, in the presence of multiple large shareholders,a decisio
7、n to divert funds from the firm requires the consent of a coalition of shareholders. Since a coalition of shareholders diverts fewer funds than would any of its individual members,the initial owner can commit to low levels of fund diversion by seeking the participation of other large shareholders.In
8、 other 2 words,by diluting her power,the initial owner commits not to undertake unilateral actions. In our model,a firms initial owner,in need of external capital,sells votes and cash flows to wealth constrained outside investors. Once the ownership structure is determined,coalitions compete to seiz
9、e control of the firm. Shareholders value control because it allows them to enjoy private benefits. We assume that these benefits come at the expense of noncontrolling shareholders and, moreover, that such benefits are inefficient to extract. The outcome of the control contest is infuenced by the ow
10、nership structure,i.e.,by the number of votes and the size of the cash flows of each shareholder. Therefore, when deciding the ownership structure, the initial owner takes into account both the need to raise. Our results about ownership are driven by two opposing effects,which we call the alignment
11、and coalition formation effects. The alignment effect is the positive relation that exists between the cash flow stake of the controlling coalition and total firm value.The greater the cash flow possessed by the controlling coalition,the more this coalition internalizes the costs of its actions. Hen
12、ce, the fewer costly private benefits it extracts.This effect is similar to that described in Jensen and Meckling (1976). The second and novel effect arises from shareholder equilibrium behavior at the time of coalition formation.Once votes and cash flows are distributed,many different coalitions ha
13、ve suffcient voting power to control the firm. However, out of these coalitions,the one with the smallest cash flow stake wins the control contest because it has the largest group of shareholders (in term of cash flows) from whom to expropriate. This implies that, conditional on having suffcient vot
14、ing power to control the firm, the coalition formation effect minimizes the cash flow stake of the winning coalition. Consider the initial owners choice between retaining absolute control and diluting control of the firm. Note that, through the price of the securities sold,the initial owner bears an
15、y inefficiency caused by diversion. If the initial owner has enough wealth to fnance the firm and retain a large portion of the cash flows,say 80%,then it is optimal for her to retain absolute control.By doing so,the winning coalition, which in this case will be the initial owner herself, will have
16、a considerable fraction of the cash flows. However, if the initial owners wealth allows her to retain a smaller fraction,say only 55%, then retaining control is not optimal. Consider an alternative ownership 3 structure in which the initial owner sells 331/3% of the cash flows and votes to each of t
17、wo investors. With this ownership structure, two shareholders will form a coalition to control the firm. With 662/3% of the cash flows, the controlling body will take more effcient actions than one with only 55%.Clearly,the initial owner would like the controlling coalition to include all three shar
18、eholders (the alignment effect). However, once votes and cash flows are distributed, shareholders have incentives to form a controlling coalition that expropriates the largest set of shareholders (the coalition formation effect), and consequently, one shareholder is left out. Not all ownership struc
19、tures in which control is diluted generate the same value for the firm.Indeed,by studying how control can be diluted optimally,our analysis yields a number of results about ownership structure. First, we define a control structure as the collection of coalitions with suffcient power to control the f
20、irm. We find that for any possible control structure, there is a one-share-one-vote ownership structure that maximizes effciency. The intuition behind this finding is that deviations from a one-share-one-vote structure create shareholders with a high ratio of votes to cash flows. Thus, the coalition
21、 formation effect tends to include these share-holders in the controlling coalition. As a result, this coalition ends up with a small cash flow stake, which is a bad outcome in terms of effciency. Following the initial contributions from Grossman and Hart (1988) and Harris and Raviv (1988), an exten
22、sive literature on the optimality of one-share-one-vote has developed.Our finding differs from this literature in that it is not derived in the context of takeovers and applies to closely held corporations. Second, we establish the relation between effciency and the number of shareholders. In partic
23、ular, we show that the larger the equilibrium number of shareholders, the smaller the value of the firm. Third, we show that the best ownership structure is one with either a single large shareholder or shareholders of roughly the same size. The intuition for these two results is similar: as cash fl
24、ows are distributed among more shareholders,or as they are distributed more unevenly, it will be easier to find a coalition with suffcient voting power and a relatively small cash flow stake.A finding similar to our third result is obtained by Zwiebel(1995) in a model in which the ownership structur
25、e is determined by investors who allocate their wealth across firms to receive a larger share of the private benefits. This paper is related to a growing literature on the expropriation of minority shareholders. Shleifer and Vishny (1997) argue that, in most countries, the relevant 4 agency problem
26、that corporate governance should address is the expropriation of minority shareholders by controlling shareholders rather than the expropriation of all shareholders by the manager. A number of studies suggest that large shareholders function as a mechanism to mitigate such expropriation. La Porta et
27、 al. (1999) and Pagano and Roel (1998) argue that other large shareholders reduce diversion by monitoring the controlling shareholder. Gomes and Novaes (1999) focus on how expost bargaining problems among large shareholders protect minority shareholders by preventing large shareholders from undertak
28、ing actions that would reduce minority shareholders payoffs. Gomes and Novaes analysis and ours are complementary since we focus on the formation process of the controlling coalition without considering expost bargaining problems, while they concentrate on bargaining problems and do not consider str
29、ategic issues relating to coalition formation.A relatively small cash flow stake. A finding similar to our third result is obtained by Zwiebel(1995) in a model in which the ownership structure is determined by investors who allocate their wealth across firms to receive a larger share of the private
30、benefits.However, the two analyses are similar in that they endogenize the process by which large shareholders mitigate diversion rather than starting with an exogenous monitoring technology. Our paper contributes to this literature by introducing control dilution as a mechanism to reduce diversion.
31、 Interestingly, our result that control dilution is an effective mechanism to reduce diversion counters the suggestions in LaPorta et al.(1999)and Bebchuk (1999) that argue that,in countries with poor investor protection,control should be concentrated to prevent someone seizing it without fully payi
32、ng for it.However, this scenario cannot occur in our model since, once the initial owner sells the securities, there is no market for them. In addition, the initial owner is not worried about losing control since she receives full payment for the benefits of control that shareholders expect to recei
33、ve. On a more general level, ONeal (1987), in his discussion of shareholder disputes in closely held corporations, observes The most frequently occurring conflict of interest is between active shareholders, i.e. shareholder-officers or employees, and shareholders who are not active in the business (
34、p. 122).Furthermore, he explicitly acknowledges the role of a power contest and the difficulties in enforcing contracts among shareholders. Holders of a majority of the voting shares in a corporation, through their ability to elect and control a majority of the directors and to determine the outcome
35、 of 5 shareholdersvotes on other matters, have tremendous power to benefit themselves at the expense of minority shareholders. Traditionally, American courts have been reluctant to interfere in the internal affairs of corporations, even when minority shareholders claim they are being squeezed out or
36、 otherwise oppressed. Furthermore, many courts apparently feel that there is a legitimate sphere in which controlling shareholders can act in their own interest even if the minority suffers. In this paper we analyze a model of a closely held corporation with nontransferable shares and potentially mo
37、re than one significant shareholder. In our model the ownership structure determines which group of owners seizes control over the firm. The model shows that it may be in the initial owners interest to dilute her own power by distributing votes among several large shareholders. This dilution of powe
38、r commits the initial owner to form a coalition to obtain control, and thus creates a controlling body that has more cash flow, and that diverts less. In other words, we propose dilution of power as a mechanism to commit to low levels of diversion.In addition,when we consider the optimal way to dilu
39、te control, we obtain several implications for the ownership structure of the firm. The model suggest interesting topics about closely held corporations for future research.As discussed in the last section,among these topics is the design of the optimal corporate governance mechanism in the presence
40、 of many active shareholders. The analysis of this paper can also be extended to study the decision by firms to go public. It can be shown that, in this setting, letting the shareholders freely trade shares will render an ownership structure with many significant owners unstable. Therefore, firms th
41、at go public forgo a mechanism to mitigate diversion. This is a cost that firms must weigh against the potential benefits of going public. 译文 紧密持有型公司的权力制衡 资料来源 :金融经济学期刊 58（ 2000 年） 作者： Morten Bennedsen , Daniel Wolfenzon 绝大多数的企业在 发展中国家和转型经济体下并不具备其股票在交易所买6 卖的条件 ，即使在美国，几乎有 400 万家公司于 1993 年逃离（收入统计）纳税，仅
42、 7842 公司已在纽约证券交易所，纳斯达克和联合（ 1994 年纳斯达克股市资料）美国证券交易所上市了 。 紧密持有型公司的股权结构通常有几个显著性股东组成。但是企业的财政都集中在分散的所有权结构（伯利和手段， 1932 年，格罗斯曼和哈特， 1980）或单一的控股股东（ Jensen 和 Meckling， 1976; Shleifer 和 Vishny， 1986; Burkart等， 1997）。在前一种情 况下，股东太小，杂乱无章地强加他们的意愿。因此，控制权掌握在经理人的手里。掌握在后者手中，占主导地位的股东通过直接密切监测管理公司的管理队伍来决定支配企业的政策。其余的股东缺乏力量
43、或激励来反对控股股东的决定。 在本文中，我们着重研究股东足够大的公司，即使没有足够大的个人股东独自来控制公司也不放弃控制经理人。在这些公司，企业的决策是股东之间相互作用的结果。 在该模型中，我们发现，公司的创立者选择具有多个大股东的所有权结构，以防止采取单方面行动，可能会伤害其他单一股东的股权结构。例如，有多个大股东存在时，决定从 公司转移资金，需要一个股东联盟的同意。由于联盟的股东挪用资金会比其个人的任何成员难，公司的创立者可以承诺较低的资金转移水平，寻求其他大股东的参与。换句话说，分散他的集中度，公司的创立者承诺不采取单方面行动。 在我们的模型中，一个公司的最初所有者，用外部资金的需求，销
44、售量，现金流量来限制外来投资者的财富。所有制结构一旦确定，股权制衡竞争，制衡该公司的控制权。股东价值的控制，因为它允许他们享受私人利益。我们认为这些优势来自于非控股股东的费用，此外，这种利益是没有效率的提取物。该制衡比赛的结果是受所有制结构的影响 ，即由通过表决的数量和尺寸来决定股东现金流量。因此，在决定股权结构时，公司的最初所有者应该考虑需要加薪。有关所有权我们的结果是由两个对立的效果影响的 ,这段时间我们称之为协调与联合形成的影响。在股份现金流的制衡中和总公司的价值是有积极效果的。拥有越大的现金流量制衡联盟， 这个联盟的内部有更多费用化的行为。因此，需要减少昂贵的私人利益分配费用。这种效果
45、类似的 Jensen 和 Meckling（ 1976）的描述。 第二个戏剧化的影响来自于股东联盟的形成到制衡行为形成的时间。一旦股票和现金流量分布，许多不同的联盟有足够 的投票权来控制该公司。然而，在这联盟中，用最小的现金流量控制的股份之一赢得比赛，因为它有剥夺而来的最大的股东集团（以现金流术语）。这意味着，在有足够的投票权的情况下，以控制该公司为条件，联盟形成的影响最小的是获胜联盟的现金流的股份。考7 虑了初始所有者之间保持绝对控股和削弱该公司选择与控制行为。注意的是，通过出售证券的价格，由最初的所有者承担分流造成的任何风险。如果最初的业主有足够的财富，可以与公司的资产相提并论并保持大部份
46、的现金流量，比如说占有 80，那么它有绝对控制权。这样，获胜的联盟，在这种情况下最初的所 有者将会有相当一部分的现金流量流入。 但是，如果最初的所有者的财富只能使他保留一个较小的部分，只有 55，则保留控制权不是最佳的。考虑替代的所有权结构，其中初始所有者出售 331 / 3的现金流和股票至每个投资者。有了这种所有权结构，两公司的股东将结成联盟，以控制该公司。随着 662 / 3的现金流量，控制机构将采取只有比 55更有效的行动。显然，最初的拥有者希望控制联盟，包括所有三个股东（对齐的效果）。然而，一旦股票和现金流量分布，股东有动机形成一个联盟，控制并征用最大的股东集（联盟形成中的作用），因此
47、， 其中一个股东是被排除在外的。 并非所有的所有权，控制权分散是对公司产生同样价值的结构。事实上，通过研究如何控制分散程度可优化股权，我们的分析得出了关于所有制结构的成果数。 首先，我们定义一个有足够的权力来控制公司的联盟集散控制结构。我们发现，对于任何可能的控制结构，有一股一票所有制结构，最大限度地追求效率。这一发现背后的直觉是，从一股一票结构偏差创建具有高现金流量的票数比例的股东。因此，联盟形成中的作用往往要包括联合控制这些股份持有人。因此，这个联盟的目的与少量现金流的股份，这是一个在不好的方面在追求效率的 结果。 继 Grossman 和 Hart（ 1988）和 Harris 和 Ra
48、viv（ 1988 年）之后，一股一票是最优最广泛的学术发展初期的贡献。我们发现不同之处在于它不是在收购范围内，并适只用于少数人持股公司。 第二，我们之间建立了绩效和股东人数的关系。特别是，我们证明了较大的股东数量平衡导致较小的公司价值。第三，我们表明，最好的所有制结构是不符合任何一个单一的大股东或大致相同的大小股东之一。对于这两种结果的直觉是相似的：现金流分布在股东以上，或者因为它们分布更不均匀，这将是更容易找到足够的投票权和现金流量相对较小 的股权联盟。一个类似于我们的第三个结果发现是得到茨威贝尔（ 1995）在所有制结构中，由投资者决定谁的财富分配在公司收到的私人收益较大份额模型。 本文
49、收集了有关对少数股东征用越来越多的文献。 Shleifer 和 Vishny（ 1997）认为，在大多数国家，有关机构的公司治理问题应该解决的是少数股东通过控制征用，而不是由全体股东经理征用。多项研究表明， La Porta 等人（ 1999 年）和帕加诺和罗尔（ 1998）认为大股东的机制，以减轻这种征用功能，8 降低其他大股东，控股股东通过监测分流。 Gomes 和 Novaes（ 1999）认为大股东之间讨价还价的问题，如何暴露重点保护，防止其从事的行动，将减少小股东的回报大股东小股东。 Gomes 和 Novaes 的分析和我们是互补的，因为我们对联盟形成过程的控制，而不考虑暴露谈判的焦点问题，同时他们集中精力谈判的问题并没有考虑到联盟的形成有关的战略问题，一个相对较小的现金流量的股份。一个类似于我们的第三个结果发现是得到茨威贝尔（ 1995）在所有制结构中，投资者谁是他们的财富分配在公司收到的私人收益更大份额确定模型。然而，两种分析方法是相似的，他们向内发展的过程，减轻大股东的持股比例，而 不是一开始转移外源监控技术。我们的文献有助于实现这一假设作为一种机制引入控制，以减少
