为什么说股票回购正在妨碍创新

卡莉•费欧莉娜1999年7月入主惠普后立即采取的措施之一是回购股票。2005年费欧莉娜下台时，惠普已累计回购股票140亿美元，超过了这六年间创造的120亿美元利润。

费欧莉娜的继任者马克•赫德当了五年CEO，在此期间他为回购股票甚至投入了更多资金，共计430亿美元，而在他任期内，惠普的总利润为360亿美元。在赫德之后，任期仅11个月的李艾科在2011年卸任前也动用了100亿美元来回购股票。

这三位CEO在十几年时间里采取了同样的策略，而这项策略在过去20年中已经成为许多美国大公司的常态，那就是用大量现金来回购股票，同时借收购来提高收入水平。

这些回购行动让大批资金流入股东手中。长期而言，这样做能为惠普带来哪些好处并不明朗。多年来，惠普一直未能推出重量级产品。在更有利可图的软件与服务领域，惠普迟迟未能有所建树。该公司核心业务的收入和利润率也不断下降。

惠普的困境体现了全球市场的快速变化，大多数大型企业都因此面临压力。同时，在本轮经济扩张进入第六年之际，越来越多的批评人士都指出，耗资数百亿美元来回购股票制约了惠普等公司应对市场变化的能力。他们认为，把资金用在这方面妨碍了创新，延缓了增长，加剧了收入失衡局面，而且削弱了美国的竞争力。

马萨诸塞大学洛威尔分校经济学教授、工业竞争力中心主任威廉•拉左尼克说：“惠普是创新型企业的典范，这样的企业会把利润留下来，再将其用于提升员工的创造能力。但1999年以来，该公司不断压缩员工队伍，并将利润交给股东，这无异于自我毁灭。”

惠普方面拒绝就本文发表评论。

在现任CEO梅格•惠特曼指挥下，惠普刚刚完成了历史上规模最大的公司分拆，建立了从事个人电脑和打印机业务的惠普公司（HP Inc.）和从事企业硬件与服务的惠普企业（HP Enterprise）。为了扭亏和重组，惠普最终将裁员8万人。

路透社分析师指出，许多公司都在沿着同样的道路大踏步前进，它们的股票回购增速超过了为长期增长投入的资金，比如研发以及其他形式的资本支出。

路透社查看了3297家美国非金融上市公司，2010年以来一直在回购股票的公司几乎占60%。2014财年，这些公司在股票回购和分红方面的支出超过了它们的净利润总和，这是非经济衰退期首次出现这种情况。已经公布2015财年业绩的公司有613家，它们在这方面的开支也继续上升。

在2014财年，这3297家公司股票回购规模达到了创纪录的5200亿美元。再加上3650亿美元的分红，返还给股东的资金总额为8850亿美元，超过了这些公司8470亿美元的利润。

路透社分析师指出，和经营性投资相比，回购和分红费用呈迅猛增长态势。2010年以来一直在回购股票的1900家公司，它们用于回购和分红的资金是资本支出的113%。2000年这个数字为60%，1990年为38%。

回购股票并披露了研发支出的公司约有1000家。2009年以来，这些公司的创新支出/净利润比例平均不到50%；2014年这个数字才因净利润下降而增至56%。在20世纪90年代，该比率一直高于60%。

在经济学家的论述中，美国公司越发金融化，回购股票则是其中的一部分，而且这种金融工具投资正在不断挤占其他投资的空间。

这种现象由几个因素共同造成，包括来自维权股东的压力；高管薪酬和每股收益及股价挂钩，而股票回购可以提升这两项指标；全球竞争越发激烈；以及担心对产品和服务进行长期投资可能得不到回报。

如今，在一些最著名的创新型企业，高管们满脑子想的都是回购和分红。

2005年以来，IBM已斥资1250亿美元来回购股票，并发放了320亿美元红利，超过了同期1110亿美元的资本支出和研发投入。制药公司辉瑞在过去10年中为回购和分红投入了1390亿美元，而研发开支和资本支出分别为820亿美元和180亿美元。发明了便利贴和透明胶带的3M用了480亿美元来回购股票并分红，研发资金和资本支出则分别为160亿美元和140亿美元。

年报显示，作为路透社的母公司，汤森路透集团2014年的资本支出为9.68亿美元，其中逾一半用于研发。股票回购和分红总额达20.5亿美元，是资本支出的两倍多。2014年该集团全职员工人数为5.3万人，低于2011年的6.05万人。2015年初以来，汤森路透的资本支出为7.43亿美元，回购和分红规模则达到20.2亿美元。

汤森路透负责公司事务的高级副总裁大卫•克伦威尔说：“从资金分配的角度来说，我们总是把用于增长的再投资排在股票回购之前。”

“恐怖的情景”

理论上，除了分红，股票回购是另一条和股东分享利润的途径。由于回购能提高公司股票需求并减少供应，它往往会提高股价，从而放大其积极效应，至少短期内是这样。同时，总股本变少会推动每股收益上升，甚至是在净利润总额持平的情况下。

公司方面称，如果产品和服务需求不足以体现研发支出的合理性，或者他们认为股票遭低估时，就应当回购股票；而且正因为这样，股票回购是一种优于新项目的投资。

然而，要是这些回购是以牺牲创新为代价呢？要是股东财富的短期增长可能削弱长期竞争力呢？哈佛商学院教授、《创造繁荣：美国为什么需要制造业复兴》（Producing Prosperity: Why America Needs a Manufacturing Renaissance）一书的作者加里•皮萨诺说：“从平板电视到半导体，再到光伏电池，美国在制造领域处于全面落后状态。”

他认为，如果美国公司继续把现金派发给投资者，经济性投资“就会流向能得到充分利用的地方。如果德国、印度或巴西的公司有需要这些投资的地方，这些资金就会流向那里，而且也应该如此，从而为当地带来增长和活力，而不是为美国。这是一种恐怖的情景。”

随着国防预算不断萎缩，军工企业越来越难以为研究开支找到正当理由，这甚至可能威胁到美国的国土安全。

美国航空航天工业协会CEO大卫•梅尔彻指出，由于缺乏新的武器项目，再加上来自华尔街的压力，公司纷纷开始转向股票回购。

梅尔彻说：“这些公司的投资团体和关注它们的分析师都在说‘我们想要更高的回报率，我们希望每股收益上升’。除非这些公司都准备退市，否则这就不是一种可持续的长期策略……就连华尔街分析师有时也会异口同声地说，‘你们什么时候能提高业绩？’”

在几家最大的军工企业中，诺斯洛普格拉曼公司2010年以来共投入逾120亿美元来回购股票，即便过去五年收入水平连续滑坡。2010年至今，洛克希德-马丁公司的收入一直持平，而该公司的股票回购规模几乎也达到了120亿美元。

最近几个月，随着2016年大选逐渐升温，疯狂回购股票的长期效应所引发的担忧开始闯入公众的视野，也引起了政坛人物的注意。

民主党参议员伊丽莎白•沃伦和塔米•鲍德温已经要求美国证监会将股票回购作为操纵市场的潜在手段进行调查。

民主党总统候选人希拉里•克林顿的竞选纲领已经提出，要把公司的关注焦点从短期转向长期。她在2015年7月建议提高短期投资税收，并就披露股票回购和高管薪酬信息做出更严格的要求。她表示，这些措施将促进长期投资，带动创新并提高普通员工收入。

借助自己的企业高管背景，前惠普CEO费欧莉娜现已成为共和党总统候选人。记者曾多次要求她就此发表评论，但均遭到拒绝。

另一位前惠普CEO赫德现任甲骨文公司联合首席执行官。他在接受路透社采访时说，回购股票是使用资金的恰当方式。他认为：“惠普现金充足，能按自己想要的规模来回购股票。”赫德担任CEO期间，惠普创造的经营性现金流为620亿美元，比他为回购投入的资金多三分之一。赫德表示：“这是对资金的妥善利用。”

赫德在任时，惠普的收入和股价双双上升。他说股票回购规模和研发投资没有关系。在赫德的任期内，惠普共为研发投入了170亿美元。

赫德的继任者李艾科的发言人拒绝就此发表评论。

制约管理者

1982年以前，美国基本上不允许公司回购股票。在那一年，作为罗纳德•里根总统全面放松金融市场管制的举措之一，美国证监会放宽了限制，允许公司在公开市场上购买自己的股票。

当时的自由市场改革者指出，二战以后，经过几十年的增长，美国公司已变得臃肿而浪费，对于经营者如何使用资金，或者不使用资金的情况缺乏监督。

特拉华大学金融学教授、约翰•L•温伯格公司治理中心主任查尔斯•埃尔森说：“当时，公司董事都是经营者本人和他们的朋友。权力基本上在管理层手中，而且不受制约。”

不过，当时人们已经逐渐形成了一种观点，那就是公司的首要目标是尽量为股东创造价值，就算这意味着要通过股票回购和分红向股东派发现金，而这些资金本来可能用于长期经营性投资。

俄亥俄州立大学费舍尔商学院金融学教授伊萨克•本-大卫对回购股票持赞成态度。他在回答路透社问题的电子邮件中写道：“为消费者服务，推出创新型新产品，雇佣员工，保护环境……这些都不是公司的目标。它们是一个过程的组成部分，这个过程的目标则是尽量让股东获得更多价值。”

这个目标在最近的一些知名案例中备受关注。维权投资者要求公司高管分享财富，否则就可能赶他们下台。

2015年3月，通用汽车同意回购50亿美元的股票，以便让投资者哈里•威尔森感到满意。后者曾威胁说，通用汽车摆脱破产保护状态后，在短短几年时间里就积攒了250亿美元现金，如果不拿出部分资金跟股东分享，他就去争夺该公司的董事席位。

杜邦公司2014年宣布将斥资50亿美元回购股票，2015年初又披露了一项价值40亿美元的股票回购计划，目的是阻击维权投资者纳尔逊•佩尔茨的基金公司Trian Fund Management，后者打算拿下四个董事席位，以便控制杜邦。尽管如此，该公司CEO柯爱伦还是因销售额增速放慢以及股价下跌在2015年10月辞职。

2015年3月，迫于对冲基金公司Jana Partners的压力，高通公司同意在随后12个月里斥资100亿美元回购股票；此前高通已制定了78亿美元的股票回购方案，并承诺将把四分之三的自由现金流返还给股东。然而，在过去10年中，该公司股票在大多数时间里的表现都不如标普500指数。

但Jana Partners并不买账。7月份，高通宣布将裁员近5000人，而且还会采取其他措施来削减成本。该公司表示，研发支出仍将保持每年约40亿美元的水平。

无视股东要求的管理者需要自担风险，特别是在公司股价面临压力的时候。拉斯•丹尼尔斯是一位技术和管理高管，曾在苹果公司和惠普分别任职15年和13年。他在惠普的职位是企业服务首席技术官，于2012年离职。丹尼尔斯说：“没有哪条要求有选择的余地。如果不予理会，你就会下台。这些要求其实都跟资源配置有关……目前的情况是许多投资者都相信，在如何使用资源的问题上，自己的决定要比企业管理层的好。”

作为软件、硬件和服务供应商，IBM曾是美国科技公司中的佼佼者。2014年，该公司为研发投入了54.3亿美元。用于回购股票的资金则一直远高于这个数字。

几十年来，IBM的高管薪酬一直和每股收益挂钩，而通过股票回购就可以操控这项指标。2007年至今，IBM的每股收益飙升了66%，但净利润总额只增长了15%（该公司在提交给监管部门的文件中称，它在考核高管绩效时就股票回购对每股收益的影响做了相应调整）。

在借助财务手段来提高每股收益方面，IBM一直是做得最露骨的公司之一。2007年，在和股东进行沟通时，IBM拿出了第一份提升每股收益的“路线图”，当时提出的目标是到2010年让每股收益达到10美元。按照这个方案，该公司将对提高利润率、收购、收入增长率和股票回购给予同样的重视。IBM轻而易举就实现了这个目标。

2010年，时任IBM首席执行官的彭明盛更进一步，承诺到2015年让每股收益增长75%以上，达到20美元。这次，IBM预计逾三分之一的增长将通过回购股票来实现。彭明盛于2011年卸任，他任职最后三年的薪酬超过了8700万美元。

在一段时间里，这项计划起了作用。IBM的股价在2013年3月达到215美元的历史最高点。但该公司的经营业绩并未跟上股价上涨的脚步。

近三年IBM的收入已经减少。利润也连续两年下滑。和2013年的高点相比，该公司股价已回落三分之一，标普500指数则上涨了34%。为控制成本，IBM进行了裁员。目前该公司员工人数比2012年少了5.5万人。

汤姆•米奇利在IBM设在纽约州波基普西市的工厂工作了30年。他说：“看到公司裁员后，员工的士气不是太高。”米奇利指出，近年来自己的工资涨幅已经没有原来那么高，IBM为员工缴纳的401(k)养老金（美国一种特殊的退休储蓄计划）也在减少。

IBM发言人伊恩•科利称，汇率波动和业务剥离给该公司业绩带来了不利影响。IBM仍在增长，而且股票回购不会影响该公司的研发和创新工作。他说：“IBM的首要任务是对业务进行投资”，比如最近在云计算等领域实施的收购。

财富效应

在股票回购的帮助下，美国股市已从金融危机时的低谷攀升到了历史高点。因此，股东觉得自己比以前富有多了，薪酬和股价挂钩的公司高管也有同感。

一些经济学家指出，劳动者为这些财富买了单，因为它削减了支持长期增长的资本支出，进而影响了就业。此外，由于大多数美国公司的股票由最富裕的美国人持有，股价上升并不能给普通劳动者带来同等收益。

拉左尼克说，因此，尽量为股东创造价值的做法“让收入集中在社会上层，并且造成中产阶层失去工作机会。和40年前相比，美国经济的富裕程度提高了一倍，但大多数人都觉得自己更穷了。”

保罗•布卢姆在IBM管理层任职16年，担任过通信研究首席技术官等职位，2013年卸任。他是乐观派之一，认为风投资本以及研发投资的其他替代渠道将起到一些弥补作用，从而为创新和经济增长提供支持。

如今在风投公司担任顾问的布卢姆预计，大企业的注意力将从直接投资于研发转向收购初创公司和剥离试验性项目，这些公司和项目不怎么受制度和华尔街要求的约束。他说：“大家会发现，和以前相比投资于初创型企业的公司将越来越多。”

在过去的100年里，许多改变了世界的突破性进展，比如灯泡、激光、计算机、航空和航天技术都以实验室创新为基础，而且这样的创新都来自IBM、苹果、施乐和惠普等资金实力雄厚的公司。

有些人认为，在技术层面惠普和IBM等公司正在远离需要大量厂房和设备投资的传统制造业，并且转向以数据为基础的产品，这也改变了创新所需投资的计算方法。

马克•迪安在IBM研发部门工作了34年，曾参与研制了1981年诞生的第一台个人计算机。他说：“这些公司花钱的方式与众不同，此类投资也很难计算。大家也许认为它们的支出在原地踏步，但我觉得资金得到了更好的利用。创新正在发生变化。”

惠普之道

在很长一段时间里，惠普一直坚持着“惠普之道”，这种平等主义企业哲学得到了外界的普遍赞赏。经营性部门在发展业务方面高度自主。惠普建立了培育型环境并鼓励员工进行创造性思维，研发支出在收入中占的比重也总是超过10%。

1999年上任的费欧莉娜改变了这一切，她实施了全方位裁员，把工作机会转移到海外，并将控制权集中了起来。

比尔•穆特尔曾是惠普高级副总裁，2001年惠普斥资250亿美元收购康柏电脑后他成了后者的一员。在费欧莉娜竞选团队建议下，穆特尔接受了路透社的采访。他说，费欧莉娜做出了惠普需要的改变，原因是该公司当时在创新方面已经变得毫无建树，惠普“总是在瞄准，瞄准，再瞄准，却从未实施和执行。”

费欧莉娜上任前不久，惠普剥离了一项业务，后者成为现在的安捷伦科技公司，并且带走了很多高科技人才。

在研发方面，费欧莉娜的关注重点是把获取专利作为提高开支效率的手段。惠普提交给监管部门的文件显示，2002年该公司注册了1.7万项专利，2005年费欧莉娜卸任时这个数字已升至3万项。

尽管如此，所有这些新专利都未能带来任何持久而成功的创新。研发工作变得支离破碎，有些项目出现了重叠。

1999年上任时，费欧莉娜的薪酬取决于每股收益。从2003年开始，它又和股东总回报率挂了钩，这个业绩衡量标准把股价涨幅和分红加在一起，然后和标普500指数的回报率进行比较。

2000年网络泡沫破裂后，费欧莉娜的股票回购措施未能阻止惠普的股价下跌。在其任内，不均衡的利润以及对收购康柏的质疑不断打击着惠普的股价，这也是促成她2005年下台的原因之一。

收购康柏后惠普变得臃肿起来，赫德则理顺了惠普的结构。他把研究项目从6800个削减到40个左右，在个人电脑和打印机部门全面压缩成本，并将惠普的重心转移到打造利润率更高的软件与服务业务上。

各项业务的市场份额都得到了提升。但研究者指出，在个人电脑和打印机部门，限制支出的新规打乱了项目时间表。据一些曾在惠普工作的工程师介绍，为达到赫德提出的目标，每到季末，一些难以做到这一点的经理就会冻结支出，暂停物资采购。

其中一位工程师说：“不能像水龙头一样开开关关。不能为了季度业绩好看就节流，下一季度再重新投入也不会立即带来很棒的产品。”

此前在美国造出产品原型的惠普工程师如今得依靠设在亚洲的制造基地来进行生产。受开支压力影响，到亚洲出差的计划有时会被推迟。惠普实验室的工作人员也离开了更具试验性的项目，转而为现有产品线提供服务。

赫德在接受采访时说，自己当时不知道存在冻结支出和项目受到干扰的情况。

他做出的调整取得了辉煌成果。从2005年到2010年，惠普的净利润增长了265%，收入增幅则低得多，只有45%。赫德任职期间，惠普的股价上涨了一倍多，从20美元升至50美元。

由于大量回购股票，每股收益的表现更为优异，增长了350%。惠普用于股票回购的资金从2005年的35.1亿美元增至2006年的77.8亿美元，在随后五年里，有四个年头的股票回购规模都超过了90亿美元（其中约20-30%抵消了员工持股计划带来的摊薄效应）。

赫德说，在惠普任职期间，他最关注的一直是提高收入和扩大市场份额。

他表示：“股价是公司业绩的体现。提高每股收益进而推动股价上升的短期手段通常都不会奏效……对外宣称将减少分红，进行一次性股票回购，这些都类似于在家里玩的把戏，不会持久。”他还说，自己曾多次拒绝股东的要求，包括增加分红以及效仿IBM的“路线图”提高每股收益。

由于赫德几乎总是能实现每股收益等业绩目标，他的酬劳基本上也呈上升态势。举例来说，2008年赫德拿到了4200万美元薪酬，远高于上年的2500万美元（由于绩效不达标，2009年这个数字降至3000万美元）。

惠普业绩的好转打动了投资者。在费欧莉娜治下，该公司的营业利润率一度跌至5%以下，赫德上任后，营业利润率升至9%，股价也猛涨了200%。

2010年8月，赫德因为和承包商的关系丑闻曝光而黯然下课。

他的继任者李艾科担任CEO的时间不足一年，在此期间的一项重大决策是在2011年10月斥资110亿美元收购软件公司Autonomy。一年后，也就是李艾科下台后，惠普称调查发现Autonomy在此项收购前做了假账，自身利润将因此下调近90亿美元。

惠普的打算是从扭亏过渡到增长，现任CEO惠特曼一直想在这方面求得平衡。2014年，该公司研发开支略有增长，升至34.5亿美元，这是2008年以来的最高水平，并未受到收入下降的影响。同时，股票回购规模从2013年的15亿美元增至27亿美元。

惠普一分为二后，惠特曼随即就会面临一项挑战，那就是把惠普企业打造成高利润率公司。分拆后的两家公司都将继续大量回购股票。2015年9月份，惠普企业（HP Enterprise）预计，2016年它将通过回购和分红把至少50%的自由现金流返还给股东。惠普公司（HP Inc.）承诺的现金返还率更是达到75%。（财富中文网）

译者：Charlie

校对：詹妮

When Carly Fiorina started at Hewlett-Packard in July 1999, one of her first acts as CEO was to start buying back the company’s shares. By the time she was ousted in 2005, HP had snapped up $14 billion of its stock, more than its $12 billion in profits during that time.

Her successor, Mark Hurd, spent even more on buybacks during his five years in charge—$43 billion, compared to profits of $36 billion. Following him, Leo Apotheker bought back $10 billion in shares before his 11-month tenure ended in 2011.

The three CEOs, over the span of a dozen years, followed a strategy that has become the norm for many big companies during the past two decades: large stock buybacks to make use of cash, coupled with acquisitions to lift revenue.

All those buybacks put lots of money in the hands of shareholders. How well they served HP in the long term isn’t clear. HP hasn’t had a blockbuster product in years. It has been slow to make a mark in more profitable software and services businesses. In its core businesses, revenue, and margins have been contracting.

HP’s troubles reflect rapid shifts in the global marketplace that pressure most large companies. But six years into the current expansion, a growing chorus of critics argues that the ability of HP and companies like it to respond to those shifts is being hindered by billions of dollars in buybacks. These financial maneuvers, they argue, cannibalize innovation, slow growth, worsen income inequality, and harm U.S. competitiveness.

“HP was the poster child of an innovative enterprise that retained profits and reinvested in the productive capabilities of employees. Since 1999, however, it has been destroying itself by downsizing its labor force and distributing its profits to shareholders,” said William Lazonick, a professor of economics and director of the Center for Industrial Competitiveness at the University of Massachusetts-Lowell.

HP declined to comment for this article.

CEO Meg Whitman has just overseen one of the largest corporate breakups ever attempted, creating one company for the PC and printer business, called HP Inc. HPQ -0.52% , and one for the corporate hardware and services business, called HP Enterprise HPE -1.20% . Ultimately, HP’s turnaround efforts and restructuring will cost 80,000 jobs.

A Reuters analysis shows that many companies are barreling down the same road, spending on share repurchases at a far faster pace than they are investing in long-term growth through research and development and other forms of capital spending.

Almost 60% of the 3,297 publicly traded non-financial U.S. companies Reuters examined have bought back their shares since 2010. In fiscal 2014, spending on buybacks and dividends surpassed the companies’ combined net income for the first time outside of a recessionary period, and continued to climb for the 613 companies that have already reported for fiscal 2015.

In the most recent reporting year, share purchases reached a record $520 billion. Throw in the most recent year’s $365 billion in dividends, and the total amount returned to shareholders reaches $885 billion, more than the companies’ combined net income of $847 billion.

The analysis shows that spending on buybacks and dividends has surged relative to investment in the business. Among the 1,900 companies that have repurchased their shares since 2010, buybacks and dividends amounted to 113% of their capital spending, compared with 60% in 2000 and 38% in 1990.

And among the approximately 1,000 firms that buy back shares and report R&D spending, the proportion of net income spent on innovation has averaged less than 50% since 2009, increasing to 56% only in the most recent year as net income fell. It had been over 60% during the 1990s.

Share repurchases are part of what economists describe as the increasing “financialization” of the U.S. corporate sector, whereby investment in financial instruments increasingly crowds out other types of investment.

The phenomenon is the result of several converging forces: pressure from activist shareholders; executive compensation programs that tie pay to per-share earnings and share prices that buybacks can boost; increased global competition; and fear of making long-term bets on products and services that may not pay off.

It now pervades the thinking in the executive suites of some of the most legendary U.S. innovators.

IBM IBM 0.08% has spent $125 billion on buybacks since 2005, and $32 billion on dividends, more than its $111 billion in capital spending and R&D during the same period. Pharmaceuticals maker Pfizer PFE -0.87% spent $139 billion on buybacks and dividends in the past decade, compared to $82 billion on R&D and $18 billion in capital spending. 3M MMM -0.75% , creator of the Post-it Note and Scotch Tape, spent $48 billion on buybacks and dividends, compared to $16 billion on R&D and $14 billion in capital spending.

At Thomson Reuters Corp, owner of Reuters News, capital spending last year totaled $968 million, more than half of which went toward R&D, according to the company’s annual report. Buybacks and dividends for the year were more than double that figure, at a combined $2.05 billion. The company had 53,000 full-time employees last year, down from 60,500 in 2011. So far this year, capital spending is at $743 million, while buybacks and dividends total $2.02 billion.

“From a capital allocation perspective, we will always prioritize re-investments in our growth priorities over share buybacks,” said David Crundwell, senior vice president, corporate affairs, at Thomson Reuters.

“A Scary Scenario”

In theory, buybacks add another way, on top of dividends, of sharing profits with shareholders. Because buybacks increase demand and reduce supply for a company’s shares, they tend to increase the share price, at least in the short-term, amplifying the positive effect. By decreasing the number of shares outstanding, they also increase earnings per share, even when total net income is flat.

Companies say buybacks are warranted when demand for their products and services isn’t enough to justify spending on R&D, or when they deem their shares to be undervalued, and therefore a better investment than new projects.

But if those buybacks come at the expense of innovation, short-term gains in shareholder wealth could harm long-term competitiveness. “The U.S. is behind on production of everything from flat-panel TVs to semiconductors and solar photovoltaic cells,” said Gary Pisano, a professor at Harvard Business School and author of Producing Prosperity: Why America Needs a Manufacturing Renaissance.

If U.S. companies continue to dole out their cash to investors, he said, economic investment “will go where it can be used well. If a company in Germany, India, or Brazil has something to do with the money, it will flow there, as it should, and create growth and activity there, not in the United States. It’s a scary scenario.”

Even national security could be threatened as a shrinking defense budget has made it more difficult for contractors to justify research spending.

David Melcher, CEO of the Aerospace Industries Association, said companies have turned to buybacks because of a dearth of new weapons programs and under pressure from Wall Street.

“Their investment community and the analysts that cover them are all saying, ‘We want a better return and we want EPS to grow,’ ” Melcher said. “That’s not a sustainable long-term strategy unless all these companies are going to go private. … Even the Wall Street analyst crowd at some point will say, ‘When are you going to grow?’ ”

Among the largest U.S. defense contractors, Northrop Grummanhas spent more than $12 billion on share repurchases since 2010, even as revenue has declined in each of the past five years. Lockheed Martin’s revenue has been flat since 2010; it has spent almost $12 billion on buybacks in that time.

In recent months, as the 2016 election campaigns have gathered momentum, concern about the long-term effects of the buyback craze has crept into public discourse and caught the attention of politicians.

Democrat Senators Elizabeth Warren and Tammy Baldwin have called on the Securities and Exchange Commission to investigate buybacks as a potential form of market manipulation.

Democratic presidential candidate Hillary Clinton has made shifting companies’ short-term focus to the long term a plank of her campaign. In July, she proposed increasing taxes on short-term investments and more rigorous disclosure of share repurchases and executive compensation. These moves, she said, will foster longer-term investment, innovation and higher pay for workers.

Fiorina, now a Republican presidential contender running on her record as a corporate executive, declined multiple requests for comment.

Hurd, now a co-CEO at Oracle, told Reuters that repurchases were an appropriate use of capital. “HP had plenty of cash to buy back as much stock as it wanted to,” he said in an interview. Operating cash flow during his tenure was $62 billion, a third more than he spent on buybacks. “It’s a good use of capital,” he said.

HP’s revenue and share price rose while Hurd was in charge. He said decisions about the size of stock buybacks and investment in R&D, which totaled $17 billion during his tenure, were not related.

A spokesman for Apotheker, Hurd’s successor, declined to comment.

To Rein in Managers

Until 1982, companies were largely prohibited from buying their own shares. That year, as part of President Ronald Reagan’s broad moves to deregulate financial markets, the SEC eased its rules to allow companies to buy their own shares on the open market.

At the time, free-market reformers argued that corporate America had become fat and wasteful after decades of postwar growth, with no checks on how managers spent cash—or didn’t.

“The boards you had were managers themselves and their friends,” said Charles Elson, finance professor and director of the John L. Weinberg Center for Corporate Governance at the University of Delaware. “It was basically managerial power, unchecked.”

Over the years, however, a belief has taken hold that companies’ primary objective is to maximize shareholder value, even if that means paying out now through buybacks and dividends money that could be put toward long-term productive investments.

“Serving customers, creating innovative new products, employing workers, taking care of the environment … are NOT the objectives of firms,” Itzhak Ben-David, professor of finance at Ohio State University’s Fisher College of Business and a buyback proponent, wrote in an email response to questions from Reuters. “These are components in the process that have the goal of maximizing shareholders’ value.”

That goal has come to the fore in some high-profile cases of late as activist investors have demanded that executives share the wealth—or risk being unseated.

In March, General Motors GM 0.11% acceded to a $5 billion share buyback to satisfy investor Harry Wilson. He had threatened a proxy fight if the auto maker didn’t distribute some of the $25 billion cash hoard it had built up after emerging from bankruptcy just a few years earlier.

DuPont DD -0.25% early this year announced a $4 billion buyback program—on top of a $5 billion program announced a year earlier—to beat back activist investor Nelson Peltz’s Trian Fund Management, which was seeking four board seats to get its way. Even so, CEO Ellen Kullman stepped down in October after sales slowed and the stock slid.

In March, Qualcomm, under pressure from hedge fund Jana Partners, agreed to boost its program to purchase $10 billion of its shares over the next 12 months; the company already had an existing $7.8 billion buyback program and a commitment to return three quarters of its free cash flow to shareholders. Still, the stock had been underperforming the S&P 500 for most of the past 10 years.

Jana wasn’t satisfied, and in July, Qualcomm announced it would shed nearly 5,000 workers, among other moves to cut costs. R&D spending, it said, would stay at around $4 billion a year.

Managers ignore shareholder demands at their own risk, especially when the share price is under pressure. “None of it is optional. If you ignore them, you go away,” said Russ Daniels, a technology and management executive who spent 15 years at Apple Inc and then 13 years at HP, where he was chief technology officer for enterprise services when he left in 2012. “It’s all just resource allocation … The situation right now is there are a lot of investors who believe that they can make a better decision about how to apply that resource than the management of the business can.”

IBM, once the grande dame of U.S. tech companies, spent $5.43 billion on R&D in the most recent year. It has been spending a lot more on buybacks.

For decades, the computer hardware, software and services company has linked executive pay in part to earnings per share, a metric that can be manipulated by share repurchases. Since 2007, IBM’s per-share earnings have surged 66%, though total net income has risen only 15%. (The company says in regulatory filings that it adjusts for the impact of buybacks on EPS when determining pay targets.)

IBM has been among the most explicit in its pursuit of higher per-share earnings through financial engineering. In 2007, in communications with shareholders, it laid out the first of its “road maps” for boosting EPS, this time to $10 a share by 2010. It would do so, under the plan, through equal emphasis on improved margins, acquisitions, revenue growth, and share repurchases. It easily met its expectations.

In 2010, then-CEO Sam Palmisano doubled down, pledging to boost earnings by more than 75% to $20 a share by 2015. This time, more than a third of that increase was expected to come from buybacks. Palmisano left in 2011, having received more than $87 million in compensation in his last three years at the company.

For a while, the plan worked. Shares surged to an all-time high of $215 in March 2013. But the company’s operating results have lagged.

Revenue has declined for the past three years. Earnings have fallen for the past two. The stock is down a third from its 2013 peak, while the S&P 500 has risen 34%. To rein in costs, IBM has cut jobs. It now employs 55,000 fewer workers than it did in 2012.

“Morale is not too good when you see these cuts,” said Tom Midgley, a 30-year veteran of IBM’s Poughkeepsie, N.Y. plant. In recent years, he said, his wage increases have shrunk, as has the company’s contribution to 401(k) retirement savings.

IBM spokesman Ian Colley said that the company’s results have been hurt by currency shifts and business divestitures. He said that the company continues to grow, and that its buybacks have not affected research, development and innovation efforts. “IBM prioritizes investment in the business,” he said, citing recent acquisitions in cloud and other areas.

Wealth Benefit

Share repurchases have helped the stock market climb to records from the depths of the financial crisis. As a result, shareholders and corporate executives whose pay is linked to share prices are feeling a lot wealthier.

That wealth, some economists argue, has come at the expense of workers by cutting into the capital spending that supports long-term growth—and jobs. Further, because most U.S. stock is held by the wealthiest Americans, workers haven’t benefited equally from rising share prices.

Thus, said Lazonick, the economics professor, maximizing shareholder value has “concentrated income at the top and has led to the disappearance of middle-class jobs. The U.S. economy is now twice as rich in real terms as it was 40 years ago, but most people feel poorer.”

Paul Bloom, who was an executive at IBM for 16 years, including chief technology officer for telecom research before leaving in 2013, is among the optimists who argue that venture capital and other alternative channels of R&D investment will take up some of the slack, supporting innovation and economic growth.

Now a consultant to venture capital firms, Bloom expects large companies to shift away from investing directly in R&D, focusing instead on acquiring startups and spinning off experimental projects that will be less constrained by bureaucracy and Wall Street demands. “You are going to see more and more corporate investing in the startups than you have in the past,” he said.

Many of the transformative breakthroughs of the past century—light bulbs, lasers, computers, aviation, and aerospace technologies—were based on innovations coming out of the labs of companies that could afford rich funding, like IBM, Apple, Xerox, and HP.

Some say a technological shift at companies like HP and IBM away from traditional manufacturing, which requires large investments in buildings and equipment, and toward data-based products is also changing the calculation of how much investment is needed in innovation.

“The way these companies spend dollars is different, the type of investment is hard to count. While you might think their spending is flat, I think it’s better utilized,” said Mark Dean, who worked in R&D for 34 years at IBM and was a member of the team that created the first personal computer in 1981. “Innovation is changing.”

The HP Way

For years, HP adhered to “the HP way,” a widely admired egalitarian corporate philosophy. Operating divisions were given broad autonomy to develop their businesses. Employees were encouraged to think creatively in a nurturing environment. R&D spending regularly topped 10% of revenue.

When Fiorina arrived in 1999, she upended that, implementing companywide layoffs, shifting jobs overseas and centralizing control.

Bill Mutell, a former HP senior vice president who joined from Compaq after HP paid $25 billion for it in 2001, spoke to Reuters at the suggestion of Fiorina’s presidential campaign. He said that changes she implemented were needed because the company had become sluggish at innovation. HP would “aim, aim, and aim, and there was never any implementation and execution,” he said.

Fiorina joined soon after the company had spun off what is now Agilent Technologies A 2.81% , the arm that housed much of the company’s high-tech expertise.

In R&D, she focused on winning patents as a measure of the effectiveness of spending. The number of HP-registered patents rose from 17,000 in 2002 to 30,000 when she left in 2005, according to regulatory filings.

Even so, all of those new patents failed to yield any enduringly successful innovations. R&D efforts were scattered, and some projects overlapped.

Fiorina’s compensation was linked in part to earnings per share when she joined in 1999. And from 2003, it was also linked to something called total shareholder return, a measure of performance, including stock-price appreciation plus dividends, that was then compared to returns for the S&P 500 Index.

Fiorina’s buybacks failed to stop HP’s share price slide after the dot-com bubble burst in 2000. Uneven earnings and concern about the Compaq acquisition whipsawed the share price during her tenure, helping lead to her ouster in 2005.

Hurd streamlined the company’s structure, which had ballooned after the Compaq acquisition. He slashed the number of research projects, from 6,800 to about 40, and cut costs across the company’s PC and printer divisions, focusing instead on building higher-margin software and services businesses.

Market share in each division grew. But in the PC and printer divisions, researchers said, new limits on spending disrupted project timelines. Some managers struggling to meet Hurd’s targets implemented spending freezes as the end of a quarter neared, halting procurement of supplies, according to former HP engineers.

“You can’t turn it on and off like a faucet, turn it off one quarter to make the quarterly results look good, then turn it back on next quarter and have great products coming out the other end,” said a former HP engineer.

Engineers at HP who had previously created prototypes at U.S. facilities were also now relying on Asian manufacturing sites to build them. Travel to these regions was on occasion delayed due to spending pressures. Workers at the company’s labs were also moved off the more experimental projects and realigned to work on existing product lines.

In the interview, Hurd said he wasn’t aware of any spending freezes or project disruptions.

The changes he implemented led to sparkling results: From 2005 to 2010, net income rose 265% on a much smaller 45% increase in revenue. HP’s stock price more than doubled, from $20 to $50, during his tenure.

Thanks to hefty stock buybacks, earnings per share did even better, increasing 350%. HP increased share repurchases from $3.51 billion in 2005 to $7.78 billion in 2006, and again to more than $9 billion a year in four of the next five years. (Roughly 20 to 30% of annual repurchases offset dilution from employee stock-purchase plans.)

Hurd said improving revenue and market share during his term was always his first concern.

“The share price is the result that occurs if the company is performing well,” he said. “Short-term tricks to try to improve EPS, and eventually share prices, usually don’t work … Going out and saying I’m going to cut a dividend, make a one-time buyback, these are sort of like parlor tricks, they aren’t sustainable.” He said he declined shareholder requests that ranged from increasing dividends to adopting a specific EPS plan like IBM’s “road map.”

Because he nearly always met per-share earnings and other targets, his pay mostly rose, too. In 2008, for example, it jumped to $42 million from $25 million the year before. (It fell in 2009 to $30 million when he failed to meet targets.)

Investors were impressed by the turnaround. Operating margins, which had dropped below 5% under Fiorina, rose as high as 9% under Hurd, and the share price soared 200%.

Hurd resigned in August 2010 amid a scandal involving his relationship to an HP contractor.

His successor, Leo Apotheker, spent just shy of a year at the helm, marked by his decision to buy software firm Autonomy for $11 billion in October 2011. A year later—after Apotheker left—HP said an investigation had uncovered accounting fraud at Autonomy before the purchase. It took a charge against earnings of nearly $9 billion.

CEO Whitman has attempted to strike a balance with HP’s plans to move into a growth mode from a turnaround effort. R&D spending rose slightly to $3.45 billion in 2014, the highest since 2008, even as revenue declined. At the same time, share repurchases rose to $2.7 billion, from $1.5 billion in 2013.

Post breakup, her immediate challenge is to build the higher-margin HP Enterprise. Both companies will continue with generous buyback programs. HP Enterprise said in September that it expects to give shareholders at least 50% of free cash flow next year through buybacks and dividends. HP Inc. said it will give back 75%.