【正文】 據(jù)支持 在手之鳥(niǎo)理論 。 我們不期望大多數(shù)的受訪者同意剩余的股利模型，這意味著股利支付的是“剩余”收益。我們 期望 經(jīng)理 們 普遍 相信股利 政策 很重要， 因?yàn)樗麄児ぷ?的 世界中市場(chǎng) 不完全 可以使 股利 政策 與公司價(jià)值 有關(guān)。林特納的著名 的 股利政策調(diào)查 研究 強(qiáng)調(diào)，當(dāng)管理層認(rèn)為，公司收入一直增加時(shí)，公司只會(huì)增加股利 。因此， 通過(guò)降低未來(lái)現(xiàn)金流的風(fēng)險(xiǎn)，不會(huì)使得 股利越高企業(yè)價(jià)值越大。 瑟夫 （ 1982 年），朗和 利真伯格 （ 1989 年），阿格拉瓦爾和杰亞拉曼 （ 1994），詹森，索伯格和佐恩（ 1992 年）為這些 代理解釋提供實(shí)證支持。 另一種關(guān)于公司為什么支付股利的解釋是基于企業(yè)之間的代理關(guān)系。 其他研究探討除息日價(jià)格下降。這種稅收優(yōu)惠的解釋測(cè)試支付或不付股 利 采取兩種形式。 增加股 利 作為一個(gè) 可信的信號(hào)，當(dāng)其他公司不具備良好的內(nèi)部信息無(wú)法模仿股 利 增加，而不會(huì)過(guò)度增加后招致股息削減的機(jī)會(huì)。因此，股利政策 中哪種 解釋最正確仍然沒(méi)有得到解決。其他 的 如 利真伯格和拉馬斯瓦米 （ 1979， 1982），布魯姆（ 1980）， 安格 和彼得森（ 1985）采取相反的立場(chǎng)。在 包括 法瑪 和 班貝克（ 1968）， 貝克，法雷利，和愛(ài)德曼（ 1985）和貝克和鮑威爾（ 1999 年）的 幾個(gè)研究 支持林特納的行為模式。 決定公司的股利政策 林特納（ 1956） 在他的經(jīng)典研究中 說(shuō)， 公司要 有長(zhǎng)期目標(biāo)派息 率，還應(yīng)該把他們的注意力放在對(duì)股利 的 變化超過(guò)絕對(duì)的股 利 水平。我們也回顧 了有關(guān)林特納的一些調(diào)查結(jié)果的后續(xù)研究。第四部分討論方法和局限性。通過(guò)采取這種方法，我們可以評(píng)估相對(duì)重要的不同的原因，源于支付股利基于各個(gè)層次的各類報(bào)表分歧。米歇爾（ 1979）和貝克（ 1988）研究表明，股利政策各行業(yè)各不相同。在引進(jìn)和快速擴(kuò)張階段，企業(yè)通常不支付或支付很低的股息。 在這項(xiàng)研究中，我們不集中這個(gè)典型的納斯達(dá)克公司管理 人員股利政策的觀點(diǎn)，因?yàn)榇蠖鄶?shù)納斯達(dá)克企業(yè)要么不支付股息，要么不定期支付股息。 我們的研究探討經(jīng)理人如何檢視股息政策，但使用不同的數(shù)據(jù)開(kāi) 始展開(kāi)研究和提煉過(guò)去的調(diào)查結(jié)果。為公司的股息支出和重要的見(jiàn)解股利政策的各種問(wèn)題的看法。 特曼（ 1997）認(rèn)為忽視正常投資者的行為模式，而去解決股利之謎是不可能的?！蹦敲?，為什么企業(yè)支付股息，為什么投資者會(huì)關(guān)心？布萊克（ 1976）曾經(jīng)這樣描述這 個(gè)問(wèn)題，股息“謎”與“該理論似乎并不合適。Number 3s description of how firms set their dividend payments. We expect the NASDAQ firms studied, all of which have established patterns of paying dividends, to hold similar views. Second, do corporate managers of dividendpaying NASDAQ firms believe a firm39。s exdividend price. Thus, increasing the dividend today will not increase a firm’s value by reducing the riskiness of future cash flows. Although virtually no empirical support exists for the birdinthehand explanation for paying dividends, we want to determine if managers39。s (1970) version of the capital asset pricing model, dividendpaying stocks must offer higher pretax returns than nondividendpaying stocks, all else equal. Brennan39。s Dividend Policy In his classic study, Lintner (1956) reports that firm have longrun target dividend payout ratios and place their attention more on dividend changes than on absolute dividend levels. He also finds that dividend changes follow shifts in longrun sustainable earnings (managers’ smooth earnings). And managers are hesitant to make dividend changes that may later need to be reversed. Managers also try to stabilize dividends and avoid dividend cuts. Lintner developed a partialadjustment model to describe the dividend decision process that explained 85 percent of yeartoyear dividend changes. Several studies including Fama and Babiak (1968), Baker, Farrelly, and Edelman (1985), and Baker and Powell (1999) support Lintner39。s life cycle. During the introduction and rapid expansion stages, firms typically pay no or very low dividends. Such firms characterize a large portion of firms trading on NASDAQ. Our study differs from previous research on dividend policy in several ways. First, unlike prior fieldwork and surveys that focus only on NYSElisted firms from a few industries, we study managers from dividendpaying NASDAQ firms from numerous industries. Michel (1979) and Baker (1988) present evidence that dividend policies vary across industries. Our rationale for examining NASDAQ firms rests on the belief that the views of NASDAQ managers may differ from those of NYSElisted firms because of different firm characteristics such as size. Second, we investigate several areas not examined in previous surveys such as views about historical patterns of dividends, dividend life cycle, and residual dividend policy. Finally, unlike most research that focuses on a single explanation of why panies pay dividends, we examine multiple explanations. By taking this approach, we can assess the relative importance of different reasons for paying dividends based on the level of agreement or disagreement with various statements involving each explanation. The remainder of the paper is organized as follows. The next section provides a brief review of the relevant dividend literature. The third section presents our research questions and empirical predictions followed by a disc