I.H.184.Warren Buffett.Wrong Incentives
错误的激励就是错误的注意力方向，方向错了，大脑在提取信息、解释信息，以及其他所有环节都会朝着错误的方向自我强化，在注意力管理的细节中有一个需要特别注意的地方是语言。
语言、公式、图表，都是帮我们压缩世界的“工具”，一旦一个群体采纳了新的词汇或类目（比如 scaling、pre-training），整个群体的思维空间真的会变形，语言 ≈ 注意力分配的“操作界面”，谁给你按钮、菜单和词，你就容易按谁的玩法去理解世界。
广告的本质，以及巴菲特说的消费品的心智份额就是市场份额都是进一步延伸的产物。
1、《2005-04-30 Berkshire Hathaway Annual Meeting》
23. The danger of earnings expectations
预测收益的风险
WARREN BUFFETT: Number 14. Do we have a second microphone in the overflow room? Oh, there’s only one in there. OK, then we’ll go back to number one.
WARREN BUFFETT：14 号提问。溢出会场那边我们有第二个话筒吗？哦，那边只有一个。好，那我们就回到 1 号提问。
AUDIENCE MEMBER: Andrew Noble (PH) from England. Apart from the catastrophic insurance events or Armageddon scenarios that you’ve been talking about earlier, in relation to the operations of Berkshire Hathaway, what are your greatest fears?
AUDIENCE MEMBER：我叫 Andrew Noble（PH），来自英国。撇开你之前谈到的那些灾难性的保险事件或者“世界末日”式的情景不谈，就 Berkshire Hathaway 的运营而言，你最大的担心是什么？
WARREN BUFFETT: Well, the greatest, I would say — we don’t worry about the economics of the businesses we have. We’ve got a very diverse group. By and large, they’re very good businesses. By and large, they’re run by some of the best managers in the country. But we worry about something going wrong. I mean, you know, you heard about the Salomon [Brothers] thing in the movie. And, I mean, we’ve got 180,000 people out there, and I’ll guarantee you something is going wrong someplace, as we talk. That’s just the nature of it. But, what you hope is that it’s relatively unimportant or that we catch it. And — but that’s, you know, that is something that we know will happen. We try to have a culture that minimizes that. And I think we do have a culture that minimizes that. We — it’s very important, in our view, to have the right incentives. And many places, we think, have incentives that aren’t so good.
WARREN BUFFETT：嗯，我想说，最大的——是这样的：我们并不担心自己所拥有业务的经济属性。我们有非常多元化的一组业务，大体来说，它们都是非常优秀的生意，而且总体上都由美国最优秀的一批经理人来经营。但我们担心会发生“某件坏事”。你们在影片里也看到过 Salomon [Brothers] 那件事。我们旗下有 18 万名员工，我可以向你保证，就在我们讲话的这一刻，一定有哪怕一件事在某个地方出错。这是必然的。但我们希望，出错的事情是相对不重要的，或者我们能及时发现并纠正。我们很清楚，这种情况一定会发生。我们努力营造一种可以把这种情况降到最低的文化氛围。而我认为，我们确实做到了这一点。我们非常重视的一点，就是设定正确的激励机制。而在我们看来，许多地方在激励上做得并不好。
I mean, when I get on an airplane, and we own the company, like NetJets, the last thing in the world I want to tell the pilot is that I’m running late and I hope we can get to New York a little faster. I mean, that is dumb, to incent a pilot, who may be worried that, you know, somehow you affect his job or something, to get in a hurry about the takeoff or the checklist or whatever it may be. And companies do that time after time in their compensation plans or things that they incent people, in our view at least, some of the wrong things. That doesn’t mean nothing wrong will happen under any circumstances. But you should not have a system that causes people to, for example, worry about quarterly earnings. None of our — our managers do not submit budgets into Omaha. I have no idea what we’re going to earn next quarter. And I have no implicit body language out there, or anything of the sort, to our managers that I’m hoping to earn X dollars per share in the quarter.
举个例子：当我坐上飞机，而那家公司是我们旗下的，比如说 NetJets，我最不想做的一件事，就是跟机长说“我有点晚了，希望我们能快点飞到纽约”。这样说太蠢了。你等于在给机长一种激励，让他为了讨好你、为了担心你可能影响他的饭碗，而在起飞、检查单之类的问题上匆忙行事。而很多公司一次又一次在薪酬计划或各种激励上，做的正是这种在我们看来“错误的激励”。当然，这并不意味着在任何制度之下都不会出事。但你至少不应该设计一种制度，去鼓励大家去担心“下个季度的盈利”。我们——我们的经理人不用向 Omaha 提交预算。我完全不知道我们下个季度会赚多少钱。我在对他们的言语或“身体语言”当中，也不会传递出任何类似的信息——比如我希望这个季度每股能赚多少之类的暗示。
错误的激励函数决定了错误的方向。
Because in the insurance business particularly, you can report any numbers, basically, that you want to, if you write long-tail business for a short period of time. And, you know, we’ve got 45 billion of loss reserves. Well, who knows whether the right figure is 45 billion or 46 or 44? And if the desire is to report some given number in a given quarter, instead of saying 45 billion, you say 44-and-three-quarters billion, or something of the sort. So we have no incentives, in terms of how people are paid, or in terms of the fact they just don’t want to let me down. Let’s assume at the start of the year I asked everybody to submit budgets and then I went on Wall Street and preached a bunch of numbers. Even if their compensation didn’t depend on it, the managers would feel, you know, we don’t want to let Warren down on this.
尤其是在保险业务里，如果你承保的是“长尾”业务、但观测期很短，那基本上你想报什么数字都可以。比如说，我们现在有 450 亿美元的损失准备金。谁能说正确的数字一定是 450 亿，而不是 460 亿或者 440 亿呢？如果有人一心想在某个季度报出一个特定的盈利数字，那他就可以不说 450 亿，改成 447.5 亿之类的。所以在激励机制上——不论是薪酬设计，还是“大家不想让我失望”那种心理层面——我们都尽量不让这类因素起作用。假设一开年我就要大家上交预算，然后我跑到 Wall Street 去对外“布道”一堆数字。即使他们的薪酬不直接跟这些数字挂钩，这些经理人心里也会想：我们可不能在这件事情上让 Warren 失望。
So, you know, we’ll take an optimistic view of reserves, and that’s easy to do, at the end of the quarter and we won’t let him down and then he won’t look like a jerk in front of Wall Street. So we try to avoid that sort of thing. But even then, that is what we worry about. We don’t worry about this place making money. I mean, we’ll make money. And if we don’t, it’s my fault. That’s not that’s not the problem. The problem right now, in addition to the one we just talked about, the problem is deploying capital, and that’s my job, too. And, you know, I haven’t done a very good job of that recently, but with a little luck, you know, we will — and a different kind of market situation — we will get a chance to do that. Charlie?
于是你可以想象，到了季度末，我们就会对准备金采取更乐观的估计——而这么做非常容易——这样我们就不会让他失望，他在 Wall Street 面前也不会显得像个傻瓜。所以我们极力避免这种情况。但即便如此，这类事仍然是我们真正担心的。我们一点都不担心这家公司能不能赚钱——我的意思是，我们肯定会赚钱；要是哪一年没赚到钱，那是我的责任，但那还不是根本问题。眼下的问题，除了刚才提到的那些，更大的问题是如何配置资本，而这同样是我的工作。老实讲，最近在这方面我做得并不好，不过要是运气好一点，再碰上不一样的市场环境，我们总会有机会把资本好好投出去的。Charlie？
CHARLIE MUNGER: Yeah. Well, if you stop to think about it, the history of much of which we don’t like in modern corporate capitalism comes from an unreasonable expectation, communicated from headquarters, that earnings have to go up with no volatility and great regularity — corporate earnings, I mean. That kind of an expectation from headquarters is not just the kissing cousin of evil. It’s the blood brother of evil. And we just don’t need that blood brother in our headquarters.
CHARLIE MUNGER：是的。如果你仔细想一想，我们之所以不喜欢当代公司资本主义中的许多现象，有很大一部分历史根源，都是源自总部传递下来的那种不合理预期：盈利一定得不断上升、不能有波动，而且要非常平滑——我是指公司盈利。这种从总部发出的预期，不仅仅是“邪恶的远房表亲（kissing cousin of evil）”，简直就是“邪恶的亲兄弟（blood brother of evil）”。而我们在总部，根本就不需要这么一个“邪恶的亲兄弟”。
WARREN BUFFETT: Businesses do not meet expectations quarter after quarter and year after year. It just isn’t in the nature of running businesses. And, in our view, people that predict precisely what the future will be are either kidding investors, or they’re kidding themselves, or they’re kidding both. Charlie and I have been around the culture, sometimes on the board, where the ego of the CEO became very involved in meeting predictions which were impossible, really, over time. And everybody in the organization knew, because they were very public about it, what these predictions were and they knew that their CEO was going to look bad if they weren’t met. And that can lead to a lot of bad things. You get enough bad things, anyway, I mean. But setting up a system that either exerts financial or psychological pressure on the people around you to do things that they probably really don’t even want to do, in order to avoid disappointing you, I mean, I just think that that’s — it’s a terrible mistake. And, you know, we’ll try to avoid it.
WARREN BUFFETT：现实中，企业的盈利表现不可能季度季度、年复一年地总是符合预期。这根本不是企业运行的本性。而在我们看来，那些对未来做出精确预测的人，要么是在糊弄投资者，要么是在糊弄他自己，要么就是两边一起糊弄。Charlie 和我也见过那种企业文化，有时候我们还在董事会里亲身经历过：CEO 的自尊心深度卷入到“必须完成那些实际上长期来看根本不可能完成的预测目标”这件事上。全公司的人都知道这些预测是什么，因为这些目标被公开宣传得很厉害；大家也都知道，如果达不到这些目标，自己的 CEO 在外面脸上会挂不住。这样的文化会引发很多坏事。即便没有这些，你在商业世界里也已经会遇到足够多的麻烦了。可如果你再弄出一套机制，让周围的人在财务上、心理上都承受压力，去做一些他们其实并不想做的事情，只是为了不让你失望——在我看来，这是一个巨大的错误。而我们会尽量避免这种错误。
2、《2008-05-03 Berkshire Hathaway Annual Meeting》
36. No “extreme frugality” but don’t spend more than you make
不主张“过度节俭”，但绝不要超支
WARREN BUFFETT: OK. We’ll do one more, and then we’ll break for lunch. Number 7.
WARREN BUFFETT：好的。我们再回答一个问题，然后去吃午饭。第7位。
AUDIENCE MEMBER: Good morning. My name is Tim Fam (PH). I’m from Austin, Texas. For my children, I would like to hear from both of you as far as the temptation to keeping up with the Joneses. And can you give them advice that they can live by with respect to frugality, debt, and work ethic?
AUDIENCE MEMBER：早上好。我叫 Tim Fam（音），来自德州 Austin。为了我的孩子们，我想听听你们双方对“keep up with the Joneses”（与人攀比）这种诱惑的看法。关于节俭、负债和职业道德，你们能给一些他们可以奉为日常准则的建议吗？
WARREN BUFFETT: Yeah. Just tell them to keep up with the Buffetts. (Laughter) Well, Charlie and I have always been big fans of living within your income, and if you do that, you’ll have a whole lot more income later on. And, you know, it — I think they will, to a considerable extent, not a perfect extent, they will follow the example of their parents. I mean, if their parents are coveting, you know, every possession of their neighbor, you know, or trying to figure out ways to increase their cost of living without necessarily their standard of living, the kids are likely to pick up on it. But now you can get the reverse effect. If you get too tough with them, they go crazy later on. (Laughs) But the — it’s — people make that election. Incidentally, there are people — there are plenty of people that I don’t advise to save.
WARREN BUFFETT：可以。就告诉他们“跟上 Buffetts 的步伐”就行了。（笑声）嗯，Charlie 和我一直非常推崇量入为出。如果你能做到这一点，将来你的收入会多得多。而且，我认为，孩子们在相当程度上——当然不是完美程度——会以父母为榜样。也就是说，如果父母觊觎邻居拥有的每一样东西，或者总想办法提高自己的生活成本而不一定提高生活质量，孩子们很可能会学到这一套。但反过来也可能出现：如果你对他们太苛刻，他们以后可能会“报复性”放飞（笑）。总之，人们要自己做选择。顺便说一句，也有一些人——其实很多人——我并不建议他们去积极储蓄。
I mean, the real — if you’re struggling along and making a reasonable income and you have a job with a 401(k) being put aside for you, and you have Social Security, who’s to say whether it’s better to defer a dollar of expenditure on your family on a trip to Disneyland or something that they’ll get enormous enjoyment out of so that when you’re 75 you can have a, you know, 30-foot boat instead of a 20-foot boat? I mean, there are choices and there are advantages to spending money in various forms for your family when it’s young, and giving them various forms of enjoyment or education or whatever it may be. So I don’t — I don’t advocate — I may practice — but I don’t advocate extreme frugality. The — and I don’t say that it’s always better to be saving 10 percent of your income instead of 5 percent of your income. I think it’s crazy to be spending 105 percent of your income, and I think that leads to all kinds of problems, and I get letters from people every day that have experienced those problems.
我的意思是，真正的情况是——如果你过得紧巴巴但有一份还算体面的收入，有一份工作替你缴 401(k)，也有 Social Security，那么谁能武断地说，把一美元原本可以花在一家人去 Disneyland 或其他能给他们带来极大快乐的事情上，推迟下来，只为了等你75岁时能买一艘30英尺而不是20英尺的船，这样就更好呢？人生有很多选择。家里还年轻的时候，在不同形式上为家人花钱、给他们不同形式的快乐或教育，或其他种种，是有其好处的。所以我并不——我也许在个人生活中会这么做——但我并不主张“极端节俭”。另外，我也不认为把收入的10%存起来就一定优于存5%。我倒是认为花到收入的105%是疯狂之举，这会带来各种各样的问题。我几乎每天都能收到经历过这些问题的人来信。
苛刻可能是一种错误激励。
But, you know, in the end you want to have an internal score card. I mean, you are not a better person or a worse person because you live a different kind of life than your neighbor. Live a life that, you know, is true to yourself. Charlie?
不过，归根到底，你需要的是一张“内在记分卡”。我的意思是，因为你过的生活与邻居不同，你并不会因此成为更好或更差的人。过一种忠于自我的生活。Charlie？
CHARLIE MUNGER: Yeah. It’s obviously the best method to train your children to provide the proper example. (A person in the audience is shouting)
CHARLIE MUNGER：是的。显而易见，教育孩子的最好方式就是以身作则。（现场有观众在喊）
WARREN BUFFETT: I think we’re hearing a child that didn’t get that advice. (Laughter)
WARREN BUFFETT：我猜我们刚听到的是一个没得到这种建议的孩子。（笑声）
CHARLIE MUNGER: But of course, even if you do provide the proper example, it’s likely not to work —
CHARLIE MUNGER：但当然，即使你树立了正确榜样，它也有可能——
WARREN BUFFETT: It’s noon.
WARREN BUFFETT：到中午了。
CHARLIE MUNGER: — some of the time anyway.
CHARLIE MUNGER：——有时候也未必奏效。
WARREN BUFFETT: It’s noon now. We’ll take about a 30 to 40-minute break. We’ll come back. Some of those who are in the other room — we always have some openings here after lunch so you might be able to move into the main room. We’ll reconvene, we’ll say, at 12:45. We’ll go to 3 o’clock.
WARREN BUFFETT：现在是中午。我们休息大约30到40分钟，然后回来。另一个厅里的朋友——午饭后主会场通常会空出一些座位，你们也许可以移步过来。我们就定在12:45 重新开始，一直进行到3点。
3、《2015-05-02 Berkshire Hathaway Annual Meeting》
WARREN BUFFETT: We had a case at National — it’s interesting. You really have to understand — should understand — human behavior, if you’re going to run a business, because when National Indemnity — we’re going back to the late 1960s — Jack Ringwalt was a marvelous man, and he ran it, and he had another marvelous man who worked for him, his tennis partner, and that fellow was in charge of claims. And when the claims man would come in to Jack and say, “I just received a claim for $25,000” or something, for some long-haul truck or something, Jack would say — Jack — it was just his personality. He would start berating the fellow and say, “How could you do this to me?” and “These claims are killing me,” and all of that, and he was joking. But the fellow he was joking with couldn’t take it, really, and he started hiding claims. And he just didn’t — he stuck them in a drawer.
沃伦·巴菲特：我们在National公司有一个案例——这很有趣。你真的必须理解——应该理解——人类行为，如果你打算经营一家公司，因为当国家赔偿公司——我们回到 20 世纪 60 年代末——杰克·林格沃特是一个了不起的人，他负责管理一切，还有另一个了不起的人为他工作，他的网球搭档，那位先生负责索赔。当理赔人员来找杰克说，“我刚收到一个25000美元的索赔”或其他类似的事情，杰克会说——杰克——这只是他的个性。他会开始责备那个家伙，说，“你怎么能这样对我？”以及“这些索赔正在毁了我，”所有这些，他是在开玩笑。但那个他开玩笑的家伙真的不能接受，他开始隐藏索赔。他只是——把它们放在一个抽屉里。
这里隐含了一个事实，能把例子说的这么生动是因为脑子有这回事，讲不出来的很可能就没有这回事。
And that caused us to not only misreport fairly minor figures, but it also caused us to misinform our reinsurers, because they had an interest in the size of claims. And the fellow that was hiding the claims had no financial interest in doing it at all, but he just didn’t like to walk into the office and have Jack kid him about the fact that he was failing him. And you really have to be very careful in the messages you send as a CEO. And if you tell your — if you tell your managers you never want to disappoint Wall Street, and you want to report X per share, you may find that they start fudging figures to protect your predictions. And we try to avoid all that kind of behavior at Berkshire. We’ve just seen too much trouble with it. (Applause)
这导致我们不仅错误报告了一些相对较小的数字，还误导了我们的再保险公司，因为他们对索赔的规模有兴趣。那个隐藏索赔的人根本没有任何经济利益，但他就是不想走进办公室，让杰克嘲笑他失败的事实。作为首席执行官，您在发送的信息中必须非常小心。如果您告诉您的经理，您永远不想让华尔街失望，并且希望报告每股 X 的收益，您可能会发现他们开始篡改数据以保护您的预测。我们尽量避免在伯克希尔出现这种行为。我们已经见识到太多麻烦了。（掌声）
4、《2017-05-08 张一鸣-如何应对公司变大之后的管理挑战》
第三，弱化层级跟title。
我们鼓励年轻人多提想法。我第一次担任CEO是26岁，我相信我们公司26岁的人有很好的实践经验，受过很好的教育，只要给他们好的Context，他们也能做出好的决策。为了避免形式感给基层节点带来压制，我们弱化层级，首先是不允许这种称呼——“老大”、“某某总”、“老师”，这种称呼一旦出来之后，很多想法就不能涌现出来了。他们可能会倾向于先听听“老师”有什么意见，自己不能先说出来。我们没有title带来的日常可见的待遇区别，比如什么样的人配备什么样的电脑，什么样的人配备什么样的办公桌，这样也会带来层级感，也会影响不同的同事发表意见。
5、《2019-05-02 Warren Buffett.Influencers with Andy Serwer》
WARREN BUFFETT: Yeah, Walmart. But it's been accentuated, I think. We have a new retailing environment now. It isn't like it goes from night to day, but it moves somewhat. And brands that people spent billions of dollars developing and sponsoring TV shows or sponsoring radio shows in the old days-- Campbell's soup was always on there with Jack Benny or something when I was a kid, and it was big.
WARREN BUFFETT：对，Walmart。但我觉得这种趋势被进一步放大了。我们现在处在一个新的零售环境中。它不是一夜之间天翻地覆，但确实在变化。过去人们花数十亿美元打造品牌、赞助电视节目或广播节目——我小的时候，Campbell's soup 经常出现在 Jack Benny 的节目里之类，那影响可大了。
And adult brands, and people obviously like the product, too. But people are more willing to change, and it's a somewhat different world than what-- it is night and day. You're very unlikely to keep changing brands everyday. But it really surprised me that Gillette lost position. Men don't like to experiment much. Women are better at experimenting.
还有面向成年人的品牌，而且人们显然也喜欢这些产品。但如今大家更愿意改变偏好，世界与以往相比在某种程度上不同——并不是天翻地覆。你不太可能每天都换品牌。但让我吃惊的是 Gillette 的地位下滑。男性并不喜欢尝试，女性在尝试方面更积极。
When you were a kid, Gillette cavalcade of sports was your pal, and brought you the Rose Bowl and the World Series, and all that sort of thing. You just shaved with Gillette the rest of your life. And you still do to a great degree. But it's not exactly the same as it was, even five years ago or so, when we bought Kraft.
你小时候，Gillette 的体育转播就是你的伙伴——给你带来 Rose Bowl、World Series 之类。然后你这辈子就用 Gillette 刮胡子。到今天很大程度上依然如此。但情况已不完全相同了——甚至和大约五年前我们买 Kraft 时相比都不同。
学界大概从三个方向系统地研究了这种现象：
1. “消费者社会化”：小孩怎么被训练成带着品牌偏好的大人
正式名字叫 consumer socialization of children，研究小孩如何从父母、媒体、广告、同伴那里学会“做消费者”、形成品牌偏好和忠诚。
- John 1999 在 Journal of Consumer Research 做过一个 25 年综述，系统梳理了儿童在成长过程中如何发展消费知识、价值观和品牌态度
- 研究发现：
- 父母、电视广告、后来是网络与同伴，是儿童品牌偏好的核心“社会化代理”
- 早期形成的偏好可以持续影响青少年乃至成年期的选择，尤其是高频、低风险的消费品（零食、饮料、日用品）
跟巴菲特讲的小男孩一边看 Rose Bowl、一边看 Gillette 广告，本质一模一样：父母+媒体+仪式化场景（体育、重要比赛）一起，把“剃须 = Gillette”写进了孩子的消费脚本。
2. 早期品牌接触 + 怀旧：偏好可以真的跟一辈子
这条线更接近你问的“能不能终身”的问题：
- 有研究专门看 childhood nostalgia 与成年后品牌忠诚 的关系：
- 儿时的品牌体验，会在大脑里和“安全感、家庭、快乐记忆”绑定；
- 成年后看到同一品牌，会触发怀旧和情绪加成，显著提高品牌忠诚度和复购意愿
- 有营销综述指出，大约 25% 的品牌偏好可以从童年延续到成年，而且很大程度由童年时期的情感联结和怀旧驱动
这跟巴菲特那句话几乎可以一一对照：“When you were a kid, Gillette Cavalcade of Sports was your pal… You just shaved with Gillette the rest of your life.”
学界的版本就是：
- 高频曝光 + 强情绪场景 + 家庭和同伴的正向强化
→ 在童年形成一个“带情绪标签的品牌节点”
→ 这个节点在成年后会通过怀旧反复被激活，推动长期忠诚。
当然，研究也强调：
- 环境变化（新渠道、电商、对价格更敏感、同伴文化等），会削弱这种粘性，所以现在的代际忠诚比巴菲特那一代弱很多
3. 更基础的一层：心理学里的 “mere exposure effect”（单纯曝光效应）
在品牌之前，心理学早就研究过“反复曝光 → 喜好提升”的机制：
- Zajonc 1968 的经典工作提出：
仅仅反复看到某个刺激（词、符号、面孔），就足以让人更喜欢它，即使没有任何理性理由。
- 后来的 meta 分析显示：
- 单纯曝光效应非常稳健，在各种刺激类型上都能复现；
- 多次曝光后，偏好会提升，而且延时测量时效果往往更强——说明它能留在记忆和态度系统里一段时间
广告界直接把这套理论拿来用：不一定要讲特别有道理的 story，只要在高情感场景（体育、节日、家庭时刻）反复出现，就能在消费者心中种一个“熟悉 + 好感”的种子。
Buffett 小时候的 Campbell’s、Gillette，就是被“单纯曝光效应 + 情绪场景（体育转播、家庭氛围）”长期叠加的经典例子。
6、《2022-05-02 Berkshire Hathaway Annual Meeting》
32. How a “culture of lying” can develop at a company
公司如何形成“撒谎文化”
WARREN BUFFETT: Yeah. Station six. OK.
沃伦·巴菲特：好的，第六麦克风。
MARTIN WEGAND: My name is Martin Wegand. I live in Nashville, Tennessee. Mr. Buffett and Mr. Munger, thank you for your lifetime of teachings and for hosting us back in Omaha this year. (Applause)
MARTIN WEGAND：我叫 Martin Wegand，住在田纳西州的 Nashville。巴菲特先生、芒格先生，感谢你们一生的教诲，也感谢你们今年再次在奥马哈接待我们。（掌声）
WARREN BUFFETT: Well, thank you.
沃伦·巴菲特：谢谢你。
MARTIN WEGAND: You have mentioned that companies get the shareholders they deserve. And in this year’s letter, you mentioned a great satisfaction of yours is working for the individual long-term shareholders. With the growing influence of institutional index funds, how can management teams foster a shareholder culture like the one we have at Berkshire? Thank you.
MARTIN WEGAND：你曾提到过，公司会得到它应得的股东。在今年的股东信里，你也说，让你非常满足的一件事，就是为这些长期持有股票的个人股东工作。随着指数基金这些机构投资人的影响力越来越大，管理层要怎样才能培育出像伯克希尔这样的一种股东文化？谢谢。
WARREN BUFFETT: Well, fortunately we have it, and we know more about how to keep it than to institute one. And it’s very interesting. We have a 1,470,000 class A shares outstanding today. Fewer than we had a year ago. And those seats are filled. I mean you are the shareholders in place. We like the group we have. So why in the world, when we got a fixed number of seats, should we go out and recruit other people to replace you? You know? I mean the ideal shareholder group we can have is the group we have today. And, you know, if we had a church, we’d want the people to keep coming back week after week after week. If we had a limited number of seats, and we had some wonderful parishioners, we would not go out and recruit another 50 or 100 of them and have to throw out 50 or 100 of the ones we already had. We’ve got. And every company I know, virtually, you know, is wooing new people to come in. And whether they’re improving the group they get or not, I mean it strikes us as basically crazy.
沃伦·巴菲特：嗯，幸运的是我们已经有了这种文化，而且我们比起“怎么建立”更清楚“怎么维持”。这点非常有意思。现在我们大概有 1,470,000 股 A 类股在外流通，比一年前还少。而这些座位都坐满了。也就是说，你们就是已经在位的股东。我们非常喜欢现在这批股东。那么既然“座位”是固定的，为什么这个世界上我们还要到处出去招募其他人来把你们换掉呢？你想想看。在我们看来，我们所能拥有的最理想股东群体，就是我们今天已经拥有的这群人。打个比方，如果我们办一间教堂，我们当然希望这些教友一周又一周地持续回来。如果座位数量有限，而我们已经有了一群非常好的教友，那我们绝不会再出去多招募 50 个、100 个新的教友，然后把原来的 50 个、100 个从座位上挤出去。可是我看到，几乎所有我认识的公司，都在“求新欢”，想办法去讨好、去吸引新的一批人进来。我根本搞不清楚，他们得到的新股东群体到底有没有比原来那批更好——在我们看来，这种做法基本上是疯了。
We don’t want anybody different (Laughter) than we have now. And, you know, we’re not going to get rid of the index fund, so we have to get rid of people like you, and we don’t want to get rid of people like you. (Laughs) And I just don’t understand why if you had a neighborhood, and the size of the terrain or whatever it was would be such that you could have ten neighbors, and they were all great neighbors, why in the world would you go out and say to a whole bunch of people going up and down the street, you know, “Why don’t you buy the house of the guy next to me?” You know, (Laughs) it is weird, but there’s an awful lot of people that make their living by doing that, and they never really question. I would sort of ask any company that’s making analyst presentations every month or something, “Which of the present ones are you trying to get rid of?” You know, basically, because I hope you’re not going to have more shares outstanding at the end of the year than you have now.
我们一点也不想要跟现在不同的股东。（笑声）而且你知道，我们也不可能把指数基金赶走，除非把你们这类人赶走——而我们可一点都不想把你们这种人赶走。（笑）我实在不明白，如果你住在一个社区，这个地方的面积大小，决定了你最多只能有十个邻居，而且这十个邻居全都特别好，那你为什么还要跑到大街上，对一堆路过的人说：“嘿，你要不要买下我隔壁那家的房子？”（笑）这太奇怪了。不过外面确实有一大群人是靠干这个谋生的，他们从来不会真正质疑这个逻辑。我倒是很想问问，那些每个月都在做分析师路演的公司一个问题：“你们现在的股东里，你们到底是想摆脱哪一些人？”因为我真心希望，你们今年年末的流通股股数，不要比现在还多。
And am I supposed to, you know, get out of the way so (Laughs) some other fund that is thinking about what your stock is going to do next week replaces me? It is a very, very, very weird situation. And of course, the really crazy process that has developed is people talking to, we’ll say, analyst group, you know, sort of the high priests of finance, you know, some companies are doing it more than once a month. Well, just imagine if you work for that company, you go to work for that company, and every month people are repeating these things about their company that, “It’s important that we have more services per customer at 6.2 and we got to get to seven,” or something like that. And they’d say that month after month after month, so it becomes a catechism. And CEO says it or his or her representative says it, and how do you go on the next month and say, “By the way, we were really wrong, and this is what we should be working on.” You don’t say that.
那我是不是应该，你懂的，让到一边（笑），好让某个只关心你们下周股价走势的基金来把我换掉？这真是一个非常、非常、非常怪异的局面。更疯狂的是，已经发展出那一整套流程：大家去跟所谓的分析师团体讲话，把他们当成金融界的“大神父”。有些公司一个月不止开一次这种会。你想象一下，如果你在这样的公司上班，你每天去公司，而公司里每个月都要对外重复一套说辞：“对我们来说很重要的一点是，每个客户的服务项目要从 6.2 提高到 7”，类似这一类的话。他们一个月又一个月地讲，慢慢就变成了“教义问答”。CEO 说，或者他/她的代表说，你下个月怎么好意思站出来说：“顺便说一句，我们之前完全搞错了，我们真正应该做的是这个”？你不会那么说的。
And it’s a terrible problem the new CEO has coming in after a previous CEO has said the important thing to do is to hit your earnings targets. Well, you know, he’s been meeting them, in all probability, by cheating from some time to time. And this guy hands you the baton, and are you going to come out and say, “Well, we’ve really been cheating a little and it’s really counterproductive to the development of the, you know, companies, not to make earnings projections and just to give you the results as they come, rather than making up a few things. And the accounting department, you know, they can’t do it. It’s not human nature, and besides, you wouldn’t get appointed the successor. But you just don’t go in and say, “We’ve been perpetuating these myths that we can always deliver 8% growth or we can do this or do that, or the most important thing is this.”
而这对新上任的 CEO 来说，简直是个可怕的问题——前任 CEO 已经反复对外宣称，“最重要的事情就是完成盈利目标”。那么你也知道，这个家伙十有八九是靠时不时“动点手脚”来完成这些目标的。现在他把接力棒交给你，你会不会站出来说：“其实我们这些年一直在小打小闹地造点假，而这种做法对公司长期发展其实是适得其反的。我们根本不该做盈利预测，只应该如实把结果报出来，而不是凭空编点东西。财务部的人其实也做不到长期这么干，因为这不符合人性，更别说，你要真这么说，你根本不可能被选成接班人。” 但你就是不会冲进去说：“我们这些年一直在延续一个神话：好像我们总能保证 8% 的增长，我们能做到这个、能做到那个，或是说某件事是最重要的。”
You can’t go in and change that if every month you’ve been preaching to people that this is what we stand for, and just ask another question and carry this message out to the masses, to the analysts and all that. And it’s a totally destructive policy. I mean, you know, I can, within gap accounting, I can play a lot of games with numbers. We’ve done a lot-ta dumb things at Berkshire. We have never told anybody that the number had to be this or that or to change anything. I mean once you start it, it’s all over. You can’t quit. It’s like taking $5 out of the cash register. You know, the first time you take the five bucks out, you’ll say, “Well, I’m going to put it back.” And then do it a few times, and you’ll never stop. In fact, do it once and you probably never stop. But if something is going to be destructive, the thing to do is not start it. And forecasting earnings, I can’t imagine anything more destructive.
如果你每个月都在对外布道：“这就是我们所坚持的东西”，然后再让别人不断发问，把这些信息层层传导出去，传给普通投资者、传给分析师，那你之后就完全没法回头说“其实我们错了”。这种政策是彻底具有破坏性的。我的意思是，你知道，在 GAAP 会计框架之内，我也可以在数字上玩很多游戏。我们在 Berkshire 干过不少蠢事，但我们从来没有对任何人说过：“这个数字必须是这样或那样”，或者要求他们“把数字改成什么样”。一旦你开启了这条路，游戏就结束了，你再也停不下来。这就像从收银机里顺走一张 5 美元钞票。你第一次拿走这 5 块钱时，会说：“我之后会放回去的。”可是一旦你这么干了几次，你就永远停不下来了。事实上，只要你干过一次，你大概率就再也不会停手了。如果一件事从根本上是具有破坏性的，最好的办法就是一开始就别去做。而“预测每股盈利”这件事，我想不出有什么比它更具破坏性的。
注意力盯着结果而不是原因：Attention Is All You Need。
I’ve got 360,000 people out there, and they know whether I’m lying or not. Many of them. And they know what they send in figures, and they get changed? You know, what message are you telling? We’ve got one dramatic illustration of that within Berkshire. And it’s just, you know, if you start lying you’ve got a big problem. It’s that simple. And if you start saying to your team that somehow you’ve got a job — you’ve got shareholder relations, your job is to go out and tell everybody that our stock is the best thing among thousands of choices to buy, every day. Well, that’s crazy. So what do you tell them? Well, they try to, you know, see which way the wind is blowing and figure out what they have to tell people. And that they go out and tell them, and then if you’re human, and you’ve said, “We’re going to earn $3.59 a share,” you can get to $3.59. And get there quite a while.
我手下有 36 万名员工，他们很清楚我是不是在撒谎——至少有很多人看得出来。他们知道自己往上报了哪些数字，也看得到这些数字有没有被改过。那你到底是在向他们传递什么信息？我们在 Berkshire 内部有一个非常典型、非常戏剧化的例子可以说明这一点。事情很简单：一旦你开始说谎，你就摊上大麻烦了。如果你对团队说，他们的工作之一——比如所谓“投资者关系”——是每天出去对所有人说，我们的股票是在上千种选择里最值得买的一只，那就太疯狂了。那你要他们怎么说呢？他们只好去感受哪边风向对自己有利，琢磨自己“该”对外说些什么，然后就跑出去照着说。而如果你是个正常人，只要你事先对外说过一句“我们今年每股要赚 3.59 美元”，你就总能想办法把账做到 3.59——而且还能这样撑很长一段时间。
And, you know, you can have all these processes, but if you have a culture of lying, the processes really don’t — they just disappear. And Charlie and I have seen it, well, probably every time we’ve gone on a board. Charlie, tell them about it? (Laughs)
而且你知道，你可以设计出各种流程，但只要你有一种“说谎文化”，这些流程最后都形同虚设、烟消云散。Charlie 和我基本上每次进董事会，都见过这种情况。Charlie，跟大家讲讲？（笑）
CHARLIE MUNGER: Well, I think Berkshire’s culture’s going to last a long time after we’re gone. And I think it should, and I think it’ll prosper pretty well. The rest of corporate America is quite different, and it gets more different, I think, with each passing decade. And it’s getting very peculiar. Pretty soon they’re going to hold all the shareholders’ meetings online, and the shareholders won’t even come. And it’s just it’s getting very peculiar. And the index fund’s getting more and more important than the voting. And it’s like everything else in life. It changes, and not always in ways you like.
CHARLIE MUNGER：嗯，我认为 Berkshire 的文化在我们离开之后还能延续很长时间。我觉得它理应如此，我也认为它会延续得相当不错。美国公司界的其他企业则很不一样，而且在我看来，每过十年，这种差异都在加大，整个局面变得越来越奇怪。很快，他们可能会把所有股东大会都改成在线开会，股东连人都不来了。事情正朝着一种非常怪异的方向发展。而指数基金在投票中的影响力是越来越大。跟人生中其他事情一样，一切都在变化，而且变化的方式并不总是你所喜欢的那种。
WARREN BUFFETT: And it ends up for selecting different CEOs and all kinds of things. I mean, you’re not going to appoint a successor CEO that’s going to come in and say everything that’s been done before, you know, has been kind of fraudulent. You know? I mean if we needed to book an extra sale after the end of the quarter, if we needed to adjust the reserves, once you start lying, it’s all over. And I just don’t know any way around that, except to try every way you can to not — if you set the wrong example at the top, you’ve got a real problem. You know? And we’ve never told anybody to change a figure. And we never will. And if they had been changing figures, you know, we’d be in all kinds of trouble, because they know it and I’d know it and the next person would know it. And it just deteriorates. And we’ve really seen it time after time. The way boards operate, you know, it has to be process-oriented.
WARREN BUFFETT：而这一切最终会反映在 CEO 的选拔以及各种各样的具体事情上。我的意思是，你不可能任命一个接班 CEO，让他上台就说：“你知道吗，以前干的那些事其实都有点造假。” 比如说，到了季末如果我们需要多记一笔销售收入，或者需要去调整准备金……一旦你开始说谎，一切就完了。除了用尽一切办法从一开始就不要走上这条路之外，我真想不出有什么其他办法。如果在最高层树立了错误的榜样，你就会有一个大麻烦。你知道的。我们从来没有对任何人说过“把数字改一改”，以后也不会这么做。要是他们真的改过数字，你可以想象我们会遇到多少麻烦，因为他们知道，我会知道，接任的人也会知道，整个事情就会一路恶化下去。而这样的事情，我们一次又一次在别的公司身上见到。董事会运作的方式，在现实中就不得不变得“极度流程化”。
I mean I understand the problems that Delaware has in writing a statute that judges face when they look at things, but it’s just extraordinary what an emphasis on process can do to an organization, because they think they can do anything if it’s allowed. And, you know, eventually the foundation crumbles.
我当然理解 Delaware 在起草公司法时面临的难题，也理解法官在审理案件时面对的约束。但令人震惊的是，一旦过度强调“流程”，会对一个组织产生多大的副作用，因为大家会以为：只要程序上允许的事，我们就都可以做。久而久之，整个组织的基础就会开始崩塌。
7、《2025-11-26 Ilya Sutskever.We're moving from the age of scaling to the age of research》
Ilya Sutskever 00:19:00
Here’s a perspective that I think might be true. The way ML used to work is that people would just tinker with stuff and try to get interesting results. That’s what’s been going on in the past.
我有一个也许是对的看法。过去的 ML 基本是这样做的：大家东改改、西试试，看看能不能折腾出一些有趣的结果。以前大概就是这么个状态。
Then the scaling insight arrived. Scaling laws, GPT-3, and suddenly everyone realized we should scale. This is an example of how language affects thought. “Scaling” is just one word, but it’s such a powerful word because it informs people what to do. They say, “Let’s try to scale things.” So you say, what are we scaling? Pre-training was the thing to scale. It was a particular scaling recipe.
后来，“scaling” 这个洞见出现了——Scaling laws、GPT-3 出来之后，大家一下子意识到：我们应该去 scale。这其实是“语言如何影响思维”的一个例子。“scaling” 只是一个词，但它非常有力量，因为它在暗示大家应该做什么——“我们来把一切都 scale 一下吧。”那你会问：我们究竟在 scale 什么？答案是：pre-training 是那个被拿来 scale 的对象，它是一种特定的 scaling recipe。
热门主题
Recent Articles
2007-02-28 Warren Buffett's Letters to Berkshire Shareholders
Refer To：《2007-02-28 Warren Buffett's Letters to Berkshire Shareholders》。 To the Shareholders of Berkshire Hathaway Inc.: Our gain in net worth during 2006 was $16.9 billion, which increased the per-share book value of both our Class A and Class B ...2009-02-27 Warren Buffett's Letters to Berkshire Shareholders
Refer To：《2009-02-27 Warren Buffett's Letters to Berkshire Shareholders》。 To the Shareholders of Berkshire Hathaway Inc.: Our decrease in net worth during 2008 was $11.5 billion, which reduced the per-share book value of both our Class A and Class B ...2025-10-14 Tracy Britt Cool.What I Learned Working With Buffett
Refer To：《2025-10-14 Tracy Britt Cool.What I Learned Working With Buffett》。 INTRODUCTION 引言 SHANE PARRISH: Tracy, welcome to the show. SHANE PARRISH：Tracy，欢迎来到节目。 TRACY BRITT COOL: Thank you. It’s an honor to be here. I appreciate it. TRACY BRITT ...2010-05-26 Warren Buffett.Interview With FCIC
INTERVIEWER: Thank you. Mr. Buffett we’re with the staff of the Financial Crisis Inquiry Commission. We were formed by Congress in 2009 to investigate the causes of the financial crisis both globally and domestically. And to do a report, due at the ...1988-02-29 Warren Buffett's Letters to Berkshire Shareholders
Refer To：《1988-02-29 Warren Buffett's Letters to Berkshire Shareholders》。 To the Shareholders of Berkshire Hathaway Inc.: Our gain in net worth during 1987 was $464 million, or 19.5%. Over the last 23 years (that is, since present management took ...