I.H.04.Right People

“Right People”的判断无法精确量化，但不意味着困难不可克服，在有些情况下，可以在不精确但用的意义上做出判断。巴菲特早年对“Right People”的定义是“people we like, admire, and trust”，后来改成“able and trustworthy”，有一个完整的描述出现在1993年的股东信中。【Appendix.6】

2.The certainty with which management can be evaluated, both as to its ability to realize the full potential of the business and to wisely employ its cash flows;

对管理层能够被评估的确定性——既包括其实现企业全部潜力的能力，也包括其明智运用现金流的能力；

3.The certainty with which management can be counted on to channel the rewards from the business to the shareholders rather than to itself;

对管理层是否可以被指望把企业回报导向股东、而非导向自身的确定性；

上述两个句子包含3个判断，3个判断之间有一些递进的关系。

实现企业全部潜力的能力？
明智运用现金流的能力？
是否可以被指望把企业回报导向股东、而非导向自身的确定性？

第1个判断指向往前冲的能力有多强？第2个判断是会不会不顾成本、不顾一切的往前冲？指向“忘记目标后的加倍努力”？第3个判断跟第1、2个判断是有联系但不是必然的联系，比如，有些人只是性格上的惯性，单纯的“忘记目标后的加倍努力”；但如果第1条有问题，能力不足的管理团队将企业回报导向自身的可能性是比较大的。

1、Basic Trust/Basic Mistrust

巴菲特在进一步压缩时将第1、2个判断压缩成“able”，第3个判断压缩成“trustworthy”，不管怎么划分，临界值和强度的判断都是非常重要的，要把 able / trustworthy 讲清楚，绕不开一个更底层的问题：人脑是怎么“看世界”的。心理学里有一组概念：Basic trust / Basic mistrust。可以理解为：在人生最初的几年里，世界在你心里被写成了“大体可相信”还是“大体靠不住”。这个“出厂设置”会给注意力一个默认偏置——面对问题时，第一反应是“怎么把事做好”，还是“自我保护、处处防御”。

人工智能领域那篇《Attention Is All You Need》讲的是模型的注意力权重；人也是一样：注意力往哪边偏，后面的记忆和泛化就会往哪边长。巴菲特更喜欢“Focus”（专注），Focus是加强版的Attention。

Basic trust的第一反应是“怎么把事做好”。

注意力默认指向问题本身，围绕“怎么做好”搜集事实、寻找答案；
想法之间能主动连接，能够在“因为所以”之间建立足够扎实的、让自己有信心的联系；
时间久了，和现实世界的交互越来越顺，总是愿意把功劳多归于他人、不走捷径、兑现所有承诺——还额外多给一点。

这类人长期的优势，不是“更聪明”，而是更少把算力消耗在自我防御上，效率更高。

Basic Mistrust的第一反应是“自我保护”和“处处防御”

在任何问题的上下文中，毫秒级的注入垃圾信息；
一开始的假设就是错的，大量资源卷入“一开始就是错的“黑洞中；
高度敏感，系统无法收敛到稳定状态，持续消耗能量维持警觉。

时间久了，和现实世界的距离会越来越远。很多组织里都能看到一种现象：有人工作很多年，却从未真正弄懂过一次业务。这往往不是智商问题，而是问题本身从未进入视野。长期无法把注意力放在真实问题上的人，最终会用自我解释填补空白：垃圾信息先输入自己的大脑，再输出到组织——这种人往往既不可信，也不自洽。

2、Integrity/Intelligence/Energy

3、Able/Trustworthy

巴菲特在股东大会中做过更详细的解释，“able"指向两种看上去有差异、本质没有差异的能力：企业经营和资本配置，以下是需要重点关注的点：

取得什么样的成绩，同时期的竞争对手取得了什么？
接手时拿到的“底牌”是什么，又是如何打好手中的牌的？
如何进行的资本配置？

不是每个团队都能得出显而易见的结论，但如果能把自己的注意力真正落实在巴菲特讲到那几个点，即使不能得出显而易见的结论，至少能排除大量噪音——那些"因为所以"之间缺少联系的信息的干扰。

trustworthy 可以理解为：在压力与诱惑下，注意力仍能稳定指向事实。早期“饿了、怕了、痛了”得到及时回应，会把一种基本的安全感写进系统：遇到问题时会朝“怎么做好”的方向收集信息，时间久了，和现实世界的交互越来越顺，总是愿意把功劳多归于他人、不走捷径、兑现所有承诺——还额外多给一点。

在 Right People 的评估中，able 和 trustworthy 经常相伴而生：管理能力差的人，往往也不怎么把股东放在心上，但trustworthy是更基础的心理特质，源自更早的心理结构，对个人行为的影响更深，也很难通过外部表现加以掩饰。

其中有两个特别值得关注的信号：

第一反应

第一反应如果是“怎么做好”，这种力量是主动的。脑子里的想法会主动联系另一个想法，通过这种方式才有可能建立足够扎实、让自己有信心的联系。

对自己有信心、对自己的判断有信心，是“不走捷径、兑现所有承诺——还额外多给一点”的根本原因。这套逻辑在成年后会继续自我强化：真实联系越多，安全感越强，注意力越容易落在问题本身，由此形成正向循环。

“美、简洁、优雅”的隐性偏好

拥有“Basic trust”的人，在对知识的不断泛化演进中有机会获得“美、简洁、优雅（Beauty, simplicity, elegance）”的体会，这里的“美”指向用更少的规则解释更多的现象，把复杂系统压缩成少数关键变量，并在压力下仍保持结构不崩。

巴菲特有一句话可以当作落地的指南：“Combine a great idea with a great manager and you’re certain to obtain a great result.”，其中 “great manager” 可以拆成两个检查的点：

“great idea”，能不能看到“美、简洁、优雅（Beauty, simplicity, elegance）”的隐性偏好？
“great result”，有没有“实质的影响（real effect、move the needle）”？

巴菲特偏好“已经被充分证明过”的”great idea“，比如American Express、GEICO，他可能不会押注太早期的想法，但当有人像 Steve Jobs那样，例如 1997 年的 “Think Different”【Appendix.11】，那种对简洁与优雅的执着，本身就是一种强信号：它往往意味着这个人能把复杂世界压缩成可执行的少数关键决策。

因此，“美、简洁、优雅”不是一个孤立的性格装饰，而是更深层能力的外显信号：它意味着一个人可能真的在做压缩、在做泛化，而不是停留在支离破碎的小聪明里【Appendix.27】。人工智能领域的顶级科学家，比如，Ilya Sutskever对此有更系统的讨论
【Appendix.28】：

I think that’s been guiding me a fair bit, thinking from multiple angles and looking for almost beauty, beauty and simplicity. Ugliness, there’s no room for ugliness. It’s beauty, simplicity, elegance, correct inspiration from the brain. All of those things need to be present at the same time. The more they are present, the more confident you can be in a top-down belief.
我觉得这些一直在很大程度上指导着我：从多个角度思考，同时去寻找某种“接近美感的东西”——美感、简洁性。丑陋的东西，是不该留下空间的。你要追求的是：美、简洁、优雅，以及来自大脑的“正确灵感”。这些要素需要同时出现，而且出现得越充分，你就越能在“自上而下的信念”（top-down belief）上有信心。

具备 Basic mistrust 的个体，大脑算力被大量消耗在处理恐惧和构建防御性叙事上，永远停留在低效的"记忆山峰"，用支离破碎的小聪明去穷举和应付世界的变化。而具备 Basic trust 的个体，由于拥有较高的心理安全感，能够将更多的注意力聚焦于问题本身，更有机会提炼出符合“美、简洁、优雅（Beauty, simplicity, elegance）”的知识结构。

这种对"美、简洁、优雅"的隐性偏好，可能在人生的极早期就已成型——可能是1岁以前，甚至更早。两个方向都在自我强化：Basic trust 的人越来越擅长找到简洁的结构，Basic mistrust 的人则越来越依赖碎片化的应对。“美、简洁、优雅（Beauty, simplicity, elegance）”真正重要的地方，在于它能有效抵御幻觉、持续压制碎片化和表格化，把注意力拉回现实。

3、Institutional Imperative

巴菲特用过几个很接近的词：institutional imperative、institutional stupidity、institutional constraints，其中最正式的是 institutional imperative（写在股东信里）。他还喜欢举旅鼠的例子：集体迁移时，前面的跌落悬崖，后面的仍会成群跟着跳下去，参考【Appendix.4】。

For example: (1) As if governed by Newton’s First Law of Motion, an institution will resist any change in its current direction; (2) Just as work expands to fill available time, corporate projects or acquisitions will materialize to soak up available funds; (3) Any business craving of the leader, however foolish, will be quickly supported by detailed rate-of-return and strategic studies prepared by his troops; and (4) The behavior of peer companies, whether they are expanding, acquiring, setting executive compensation or whatever, will be mindlessly imitated.

比如：（1）仿佛受牛顿第一定律支配，机构会抗拒改变其当前方向；（2）正如工作会膨胀以填满可用时间，公司项目或并购会自发出现以消耗可用资金；（3）领导者的任何业务“渴望”，哪怕愚蠢，其下属都会迅速拿出详尽的收益率与战略研究予以背书；（4）同业公司的行为——不论是扩张、并购、制定高管薪酬等——都会被盲目模仿。

抗拒，抗拒世界是可以信任的，问题是有答案的，不相信改变会带来可控的结果，不相信外部是可合作、可理解、可修正的；
填满，只要醒着就需要有事情填满大脑，空白本身不可承受，基本不信任的情绪结构被内心的恐惧驱动，需要“填满”以求得安全感；
自我强化，同样愚蠢的下属产生自我强化，同时也是自我选择的结果，同结构的人互相识别、互相强化，组织会把“不合拍”的人筛掉，把会配合强化的人留下来；
盲目模仿，巴甫洛夫训练后的狗，Basic mistrust 下，没有真正的主体性，只有条件反射式对外部信号的服从。

Basic mistrust的底层不是愚蠢，而是不信任。源头是不信任就不可能有完整的逻辑链，“因为所以”都是断裂的，但行为需要有一个心理上的支点，“institutional”（系统性的、有固定剧本的愚蠢行为）就是这个支点，“institutional”的剧本在反复执行中自我强化，扭曲安排的或者偶尔出现的正反馈，会把这套剧本刻得更深。最后不再只是“大家习惯这么做”，而变成“不这样做就不行”，这就是“imperative”（接受命令、按命令行事）。

"institutional imperative"的说法很可能源自于巴甫洛夫的条件反射，当一个人的心理惯性获得很多次的正反馈以后，行为模式就固定下来了。Right People 在概念上比 Right Business 更短，但现实里更丰富、也更深刻。判断一个人“是否正常”可以说很多，但是最为重要的是敏感性，如果天生敏感就更好了。

（1）临界值很低

临界值几乎不可能发生在 50%，而更像是低一个数量级的东西——比如 5% 以内【Appendix.22】。关键不在“坏现象是不是主流”，而在于：系统的负荷有没有超过它的修复能力。

如果把“关键反应事件”粗略建模：每天约 2 次，则 1 岁前约 700 次、3 岁前约 2100 次、15 岁前约 1 万次。假设 3% 是“结构性后果”的临界点，那么 1 岁前只需要约 21 次关键时刻未被回应，就可能产生结构性的后果。

美国 NatSCEV（全国代表性样本）汇总的研究显示：过去一年约 6.07% 的儿童经历过某种形式的忽视，一生中约 15.14% 经历过忽视，参考：《Rates of neglect in a national sample: Child and family characteristics and psychological impact》，如果引用6.07%的数据作为父母照顾不周的概率，期望值=700*6.07%=42.49，标准差≈6.32，21次对应的标准分=（21-42.49）/6.32≈−3.40，P(X≤21)≈0.00034，P(X>21)≈1−0.00034≈0.99966，超过临界值的概率是99.966%，没有超过临界值的幸存者是0.034%。

新加坡有过自己的统计，Lee Kuan Yew 在一次演讲里提到：一流头脑约 1/1000，而一流头脑又具备正确品格与个性的约 1/3000。1/3000跟上述P(X≤21)≈0.00034非常接近，这里的重点不是精确数字，而是：临界点可能非常低【Appendix.2】。

（2）反向校正的成本很高

Basic mistrust的第一反应是“自我保护”和“处处防御”，它带来的麻烦不在情绪本身，而在注意力被“威胁探测”长期占用——像一台高耗能的后台程序，持续挤压分配给现实问题的算力。

如果想象成一台AI：（1）在任何问题的上下文中，毫秒级注入“回避/防御”的垃圾信息；（2）很多想法一开始的假设就是错的；（3）高度敏感，防御机制强大到可以隔绝大部分的事实。

第一反应

第一反应是毫秒级的，可以借用计算机科学那句话：“Garbage in, garbage out”。当一个人的默认反应是 “自我保护、处处防御” 时，他会在毫秒间就将这类防御信号注入到所有问题的上下文里：先切断注意力与客观事实的连接，再去寻找能自我合理化的解释。

反向校正的成本非常高，且在压力场景下很容易退回默认的防御模式。时间久了，和现实世界的距离越来越远，一个长期没办法把注意力放在真实问题上的人，通常会用自我解释来填补空白，垃圾信息先输入自己的大脑，再输出到组织——这种人往往既不可信，也不自洽。

一开始的假设就是错的

人的思维可以粗略分为“系统一”（快思）与“系统二”（慢想），多数时候，人不是从头推演，而是从一个默认假设出发。当系统一被“自我保护、处处防御”劫持时，默认假设往往是偏离事实的；系统二随后不是纠错，而是在既定假设上做“合理化”。结果是：大量时间与资源被消耗在“一开始就是错的”问题设定上。

防御和对抗情绪

缺乏安全感会把人推向两种不同的防御模式：一种是低能量的“缩”；另一种是高能量的“抢”。
低能量模式：「缩」

遇到任何需要承担的东西就往后退。不是评估过后觉得不该承担，而是退这个动作本身已经自动化。表现出来就是：话说到一半转移话题；答应了又找理由；永远在观望；永远留着一个“但是”；从不把话说死，因为说死了就要负责。

这类人通常链条是进去过的：你带他从 A→B→C，他知道你走了哪里，只是链条停在“接受”之前。

高能量模式：「抢」

抢”不是普通的防御，而是提前占领。这类人会下意识先把自己放到一个“不必被现实改写的位置”上。核心不是某个观点，而是一个跟身份绑定的假设：如果它被推翻，不只是“我想错了”，而是“我这个人被推下去了”。

于是，外部信息进入系统后，先经过一道筛选：这条信息，会不会伤到我现在的位置？如果会，就被折射、改写、合理化。于是你看到的现象是：反应很快、表达很多、局部都能接，但链条始终进不去。你以为在共同建构逻辑，实际上对方只是在守自己的位置。

“缩”至少链条进去过，只是停在接受前；“抢”则是链条根本没进去，巴菲特在股东信里引用过一句很经典的话：【Appendix.5】。
“Fanaticism,” said Santayana, “consists of redoubling your effort when you’ve forgotten your aim.”
正如 Santayana 所言：“狂热就是在忘记目标后加倍努力。”

自我强化

自我强化不需要做过多的解释，除了“institutional imperative”，巴菲特还经常用到另一个词是“unforced errors”，因为很容易引发情绪上的抵触，他用“unforced errors”间接的提示还有一种错误是“forced errors”。

具备“Basic mistrust”的个体，其内心有一个高耗能后台程序——“威胁探测与自我防御系统”。这套系统会在自我强化中变得越来越敏感、越来越强势，也越来越耗能、越来越低效：

对外部信息的第一反应不是理解，而是先判断是否构成威胁；
一旦被判定为威胁，注意力优先投入防御，紧接着是自我解释；
防御系统越强势，事实进入系统的比例就越低，纠正错误假设的机会也越少；
大脑在处理事实时受到防御系统强有力的干扰，变得极其耗能，除了少数足够简单的思考，大部分的思考变得不切实际。

关于高耗能，李光耀的评论非常直接，他说平庸是“财大气粗”才能消费得起的奢侈品【Appendix.2】，巴菲特不会那么直接但他也不会试图掩盖事实【Appendix.19】。

I mean, very, very, very few people get anything like their potential horsepower translated into the actual horsepower of their output in life. And potential exceeds realization to just an enormous factor with so many people.

我的意思是，能把自己“潜在马力”转化为人生“实际产出马力”的人，真的、真的、真的凤毛麟角。对许多人来说，潜力与实现之间的差距大得惊人。

反过来，拥有 Basic trust 的个体，由于底层安全感的支撑，能把更多算力投入现实世界。复杂系统里真正稀缺的不是信息，而是没有被自我叙事腐蚀的算力输出。好的情绪结构保证算力不被防御大量吞噬，因此更可能抓住“模糊但方向正确”的结构，并在反馈中不断校正。

Appendix

Appendix.1.《三国演义第二十一回》

玄德曰：“河北袁绍，四世三公，门多故吏，今虎踞冀州之地，部下能事者极多，可为英雄？”

操笑曰：“袁绍色厉胆薄，好谋无断，干大事而惜身，见小利而忘命，非英雄也。”

什么重要什么不重要？什么是重要并且可知的？什么是重要不可知的？正常的脑子能从上下文压缩出正确的知识结构。

Appendix.2.《1967-10-09 Warren Buffett's Letters to Limited Partners》

The final, and most important, consideration concerns personal motivation. When I started the partnership I set the motor that regulated the treadmill at "ten points better than the DOW". I was younger, poorer and probably more competitive. Even without the three previously discussed external factors making for poorer performance. I would still feel that changed personal conditions make it advisable to reduce the speed of the treadmill. I have observed many cases of habit patterns in all activities of life, particularly business, continuing (and becoming accentuated as years pass) long after they ceased making sense. Bertrand Russell has related the story of two Lithuanian girls who lived at his manor subsequent to World War I. Regularly each evening after the house was dark, they would sneak out and steal vegetables from the neighbors for hoarding in their rooms; this despite the fact that food was bountiful at the Russell table. Lord Russell explained to the girls that while such behavior may have made a great deal of sense in Lithuania during the war, it was somewhat out of place in the English countryside. He received assenting nods and continued stealing.

最后，也是最重要的一个考量，关乎个人动机。刚成立合伙企业时，我把那台调节跑步机速度的马达设在“比 DOW 高十个百分点”这一档。那时我更年轻，更穷，大概也更有竞争心。即使没有前面提到的那三个会压低业绩的外部因素，我仍然会认为，个人处境的变化已经使得把这台跑步机的速度调低一些，成为明智之举。我见过太多这样的例子：生活中各种领域，尤其是商业领域里的行为习惯，在它们早已失去合理性之后，仍会继续存在，甚至随着岁月流逝而变本加厉。Bertrand Russell 讲过一个故事：第一次世界大战之后，有两个 Lithuanian 女孩住在他的庄园里。每天晚上，天一黑，她们就会偷偷溜出去，从邻居那里偷些蔬菜，囤在自己房间里；尽管 Russell 家的餐桌上食物十分充裕。Lord Russell 向她们解释说，这种行为在战时的 Lithuania 也许非常说得通，但在英国乡间就多少有些不合时宜了。她们点头表示同意，然后继续去偷。

He finally contented himself with the observation that their behavior, strange as it might seem to the neighbors, was really not so different from that of the elder Rockefeller.

最后，他只能安慰自己说：她们的行为，虽然在邻居看来很奇怪，其实和老 Rockefeller 的做派也并没有太大不同。

巴菲特说他的日常工作是：收集事实+思考事实，如果事实变了，想法能不能跟着改变？还是只能根据“条件反射”做出反应？两者的区别在哪里？

1、巴甫洛夫的实验告诉我们，条件反射一旦建立，你能做的最好的事情不是消除它，而是在它之上建立一个更强的新反射来覆盖它，这种改变是被动的，需要大量、反复的刺激，并且大脑只是被动的接受刺激。

2、事实变了、想法跟着改变不是被动的接受刺激，然后建立新的反射，而是在大量信息中压缩出“模糊正确”的边界，压缩后的知识结构还要足够的简洁、有效，能抵抗被动刺激的干扰。

Elementary self-analysis tells me that I will not be capable of less than all-out effort to achieve a publicly proclaimed goal to people who have entrusted their capital to me. All-out effort makes progressively less sense. I would like to have an economic goal which allows for considerable non-economic activity. This may mean activity outside the field of investments or it simply may mean pursuing lines within the investment field that do not promise the greatest economic reward. An example of the latter might be the continued investment in a satisfactory (but far from spectacular) controlled business where I liked the people and the nature of the business even though alternative investments offered an expectable higher rate of return. More money would be made buying businesses at attractive prices, then reselling them. However, it may be more enjoyable (particularly when the personal value of incremental capital is less) to continue to own them and hopefully improve their performance, usually in a minor way, through some decisions involving financial strategy.

最起码的自我分析告诉我，对于那些把资本托付给我的人，只要我公开宣示了一个目标，我就不可能不用尽全力去实现它。而这种全力以赴，正变得越来越不合时宜。我希望自己拥有一个经济目标，这个目标能给相当多的非经济活动留出空间。这可能意味着去做一些投资领域之外的事情；也可能只是意味着，在投资领域内部，去走一些未必能带来最高经济回报的路子。后者的一个例子是：继续持有一家令人满意的控股企业——虽然它离“惊艳”还差得很远——只是因为我喜欢那里面的人，也喜欢这门生意本身，哪怕别的投资机会预期能提供更高的回报率。用有吸引力的价格买下企业，再转手卖掉，当然能赚更多钱。不过，继续持有它们，并希望通过一些涉及财务策略的决策，通常是以比较小的方式，去改善它们的经营表现，也许会更让人享受这一过程；尤其是在新增资本对个人的边际价值已经没那么高的时候。

Thus, I am likely to limit myself to things which are reasonably easy, safe, profitable and pleasant. This will not make our operation more conservative than in the past since I believe, undoubtedly with some bias, that we have always operated with considerable conservatism. The long-term downside risk will not be less; the upside potential will merely be less.

因此，我今后很可能会把自己限制在那些相当容易、相当安全、有利可图而且做起来也愉快的事情上。这不会使我们的运作比过去更保守，因为我相信——当然这判断难免带有偏见——我们过去一直都已经相当保守。长期下行风险并不会因此变得更小；只是上行潜力会变得更小。

Specifically, our longer term goal will be to achieve the lesser of 9% per annum or a five percentage point advantage over the Dow. Thus, if the Dow averages -2% over the next five years, I would hope to average +3% but if the Dow averages +12%, I will hope to achieve an average of only +9%. These may be limited objectives, but I consider it no more likely that we will achieve even these more modest results under present conditions than I formerly did that we would achieve our previous goal of a ten percentage point average annual edge over the Dow. Furthermore, I hope limited objectives will make for more limited effort (I'm quite sure the converse is true).

具体来说，我们的长期目标将改为：每年取得9%的回报，或者取得比 Dow 高五个百分点的回报，二者取其较低者。也就是说，如果未来五年 Dow 的平均回报是-2%，那我希望我们的平均回报能达到+3%；但如果 Dow 的平均回报是+12%，那我所希望达到的平均回报就只有+9%。这些目标也许算是受限的目标，但在当前条件下，我并不觉得我们实现这组较为温和结果的可能性，会比我当年相信我们能够实现“每年平均领先 Dow 十个百分点”这一旧目标的可能性更高。此外，我也希望，受限的目标会带来受限的努力程度（我很确定，反过来是成立的）。

BRK最初的假设。

I will incorporate this new goal into the Ground Rules to be mailed you about November 1, along with the 1968 Commitment Letter. I wanted to get this letter off to you prior to that mailing so you would have ample time to consider your personal situation, and if necessary get in touch with me to clear up some of the enclosed, before making a decision on 1968. As always, I intend to continue to leave virtually all of my capital (excluding Data Documents stock), along with that of my family, in BPL. What I consider satisfactory and achievable may well be different from what you consider so. Partners with attractive alternative investment opportunities may logically decide that their funds can be better employed elsewhere, and you can be sure I will be wholly in sympathy with such a decision.

我会把这个新目标写进大约11 月 1 日寄给各位的 Ground Rules 中，并随信附上1968 年 Commitment Letter。我想在那次邮寄之前，先把这封信寄给你们，这样你们就有充足时间去考虑自己的个人情况；如果有需要，也可以在决定1968 年安排之前，和我联系，把这封信里某些内容弄清楚。和往常一样，我打算继续把我几乎全部的资本（除了Data Documents的股票）连同我家人的资金，一起留在BPL。我认为令人满意且可以实现的目标，很可能和你认为的并不一样。那些拥有有吸引力替代投资机会的合伙人，完全可以合乎逻辑地决定，把自己的资金配置到别处去；而你们可以确信，我会完全理解并赞同这样的决定。

I have always found behavior most distasteful which publicly announces one set of goals and motivations when actually an entirely different set of factors prevails. Therefore, I have always tried to be l00% candid with you about my goals and personal feelings so you aren't making important decisions pursuant to phony proclamations (I've run into a few of these in our investment experience). Obviously all the conditions enumerated in this letter haven't appeared overnight. I have been thinking about some of the points involved for a long period of time. You can understand, I am sure, that I wanted to pick a time when past goals had been achieved to set forth a reduction in future goals. I would not want to reduce the speed of the treadmill unless I had fulfilled my objectives to this point.

我一向最厌恶这样一种行为：公开宣示的是一套目标和动机，而实际上起作用的却完全是另一套因素。因此，我一直尽力对你们 100% 坦诚，把我的目标和个人感受如实告诉你们，免得你们依据虚假的宣示去作出重要决定（在我们的投资经历里，我已经碰到过一些这样的事）。显然，这封信里列举的所有情况，并不是一夜之间突然出现的。这里涉及的有些问题，我已经思考了很长时间。我相信你们能够理解：我之所以选择在过去目标已经实现的时候，提出下调未来目标，是有意为之。除非我已经完成了截至目前为止的目标，否则我不会愿意把这台跑步机的速度降下来。

Appendix.3.《1982-08-12 Lee Kuan Yew.The Search for Talent》

Decline into mediocrity disastrous

滑入平庸将是灾难性的

There is a heavy price to pay if mediocrities and opportunities ever take control of the government of Singapore. And mediocrities and opportunities can accidentally take over if Singaporeans, in a fit of pique or a moment of madness , voted for the politics of opposition for the sake of opposition. Five years of such a government, probably a coalition, and Singapore will be down on her knees. What has taken decades to build up in social organisation, in industry, banking commerce, tourism, will be dismantled and demolished in a few years. The World Bank has a queue of such broken-back countries waiting to be mended: Jamaica, Uganda, Ghana, Nicaragua, to name a few recent casualties seeking emergency World Bank aid. At least they have land for plantations or mines to dig from, or rivers to be dammed for hydro-power and irrigation. Singapore has only got its strategic location and the people who can maximise this location by organisation, management, skills and, most important of all, brains. Once in disarray, it will not be possible to put it together again.

如果平庸之辈与投机分子有一天掌控了新加坡政府，我们将付出沉重代价。而这种“平庸者与机会主义者”的上台，可能并不需要阴谋，只要新加坡人一时恼怒或一念失控，为了“反对而反对”而投票支持反对派政治，它就可能意外发生。只要五年——那样的政府很可能还是联合政府——新加坡就会跪倒在地。几十年建立起来的社会组织、工业、银行金融、商业与旅游，将在短短几年内被拆解、摧毁。世界银行（World Bank）就排着一长串这种“断背”的国家等待被修补：牙买加、乌干达、加纳、尼加拉瓜——举几例近年因崩坏而寻求紧急援助的“伤亡者”。至少这些国家还有土地可种植园或挖矿，还有河流可筑坝发展水电与灌溉。新加坡只有战略位置，以及那些能通过组织、管理、技能、而最重要的是通过头脑来最大化这一位置价值的人。一旦陷入混乱，就不可能再把系统重新拼回原样。

There is no substitute for high quality Ministerial leadership. No back-up by staff officers, however excellent, can make up for the lack of keen intelligence, verve, imagination, creativity, drive and thrust which a Minister himself brings to his office. No prompted response, however wise the prompter, ever matches the spontaneous response from an active, knowledgeable and freeranging mind. No resounding ghost-written speech can ever have the ring of conviction and sincerity of the words from a Minister who has felt his own problems deeply and thought them through to their solutions. In several established countries, average leaders have managed to get by with superb staff back-up. But each time they go to summit meetings to encounter strong and able leaders of other nations, their inadequacies show up, painfully and alarmingly. And Singapore, a small, barely established, nation, cannot afford to have anything less than her ablest and her best, to be in charge of the government. If we are to preserve what we have, and more, to build on the present, and achieve further heights, we cannot have mediocrities either as Ministers or Permanent Secretaries. Prompters and ghost-writers are a luxury for those who have large margins of safety due to their large size, great wealth, and considerable institutional strength.

高质量的部长级领导力没有替代品。无论幕僚官员多么优秀，他们都无法弥补部长本人缺乏敏锐智力、活力、想象力、创造力、推动力与冲劲所带来的缺口。再智慧的“提示者”也无法让被提示的回答，匹配一个活跃、博学、视野广阔的头脑所做出的自发回应。再响亮、再漂亮的代笔演讲，也不可能拥有那种由部长亲身深切体验问题、并亲自思考到解决方案时所说出口的确信感与真诚感。在一些成熟国家里，能力平平的领导人靠着卓越的幕僚体系也能勉强运转。但每一次他们去参加峰会，与其他国家强而有力的领导人正面交锋时，他们的不足就会以痛苦且令人不安的方式暴露出来。而新加坡——一个小国、一个刚站稳脚跟的国家——承担不起让“次一等”的人掌管政府。若我们要守住既有成果，并在此基础上再建设、再攀高峰，我们就不能让平庸之辈担任部长或常任秘书。提示者与代笔者，是那些因为体量巨大、财富丰厚、制度基础深厚而拥有巨大安全边际的国家才消费得起的奢侈品。

*****

We must be grateful that the talent profile, or IQ spread, of our population enables us to produce, from a yearly birth rate, in the 1950s and 1960s, of 60,000 to 50,000 about 60-50 first-class minds, an average rate of 1 in 1,000. Alas, not all of these bright minds have strong characters, sound temperament, and high motivation to match their high intelligence. I have found, from studying PSC scholarship awards for the last 15 years, and reading confidential reports on their work in the public service and the SAF, that the scholars who also have the right character and personality, effectively works out to 1 in 3,000 persons. In the 1970s, our annual births went down to 40,000. The numbers of talented and balanced Singaporeans will be between 12-14 persons per annum at one per 3,000.

我们应当庆幸：我们人口的人才结构（或 IQ 分布）使得在 1950 与 1960 年代，每年 60,000 到 50,000 的出生人口中，可以产生大约 60 到 50 个一流头脑，平均比例约为 1/1,000。可惜的是，并非所有聪明头脑都具备与其高智力相匹配的坚强品格、健全性情与强烈动机。我在过去 15 年里研究 PSC 奖学金的授予情况，并阅读他们在公共服务与 SAF 工作的机密报告后发现：那些既聪明、又具备正确品格与个性的人才，折算下来大约是每 3,000 人中才有 1 人。到了 1970 年代，我们的年出生人口下降到 40,000。按每 3,000 人中 1 人的比例计算，每年“既有才华又较为均衡”的新加坡人将只有约 12–14 人。

“与高智力相匹配的坚强品格、健全性格与强烈动机”，巴菲特用的是“ integrity、intelligence、energy”，坚强品格= integrity，健全性格与强烈动机= energy，我自己有一种理解，integrity是一种结构化的知识，表现在做事有清晰的边界，是“intelligence、energy”共同作用下的结果，有点类似于算法和算力的组合。

这里有意思的地方是Lee Kuan Yew做了量化分析，几乎没有人敢这么分析，太敏感了，大学里搞心理学的，这些人通常不能“吃自己的饭”，没有勇气触碰这样的问题。

Appendix.4.《1985-02-01 Steve Jobs.Interview with Playboy》

Why is the computer field dominated by people so young? The average age of Apple employees is 29.

为什么计算机领域由如此年轻的人主导？苹果员工的平均年龄是 29 岁。

It’s often the same with any new, revolutionary thing. People get stuck as they get older. Our minds are sort of electrochemical computers. Your thoughts construct patterns like scaffolding in your mind. You are really etching chemical patterns. In most cases, people get stuck in those patterns, just like grooves in a record, and they never get out of them. It’s a rare person who etches grooves that are other than a specific way of looking at things, a specific way of questioning things. It’s rare that you see an artist in his 30s or 40s able to really contribute something amazing. Of course, there are some people who are innately curious, forever little kids in their awe of life, but they’re rare.

对于任何新的、革命性的事物来说，情况往往都是如此。随着年龄的增长，人们会陷入固定的思维模式。我们的思维就像是电化学计算机。你的思想构建了像脚手架一样的模式在你的大脑中。你实际上是化学蚀刻的模式。在大多数情况下，人们会陷入这些模式中，就像唱片上的凹槽一样，他们永远无法摆脱。很少有人能够刻蚀出不同于特定看待事物方式、特定质疑事物方式的凹槽。在30岁或40岁的艺术家中，能够真正贡献出令人惊叹的东西的人是罕见的。当然，也有一些人天生好奇，永远对生活充满敬畏，像永远的小孩子一样，但他们是罕见的。

“练习并不会让你完美；练习只会让你的动作定型”，很容易把“条件反射”理解成思考。

Appendix.5.《1990-03-02 Warren Buffett's Letters to Berkshire Shareholders》

o My most surprising discovery: the overwhelming importance in business of an unseen force that we might call “the institutional imperative.” In business school, I was given no hint of the imperative’s existence and I did not intuitively understand it when I entered the business world. I thought then that decent, intelligent, and experienced managers would automatically make rational business decisions. But I learned over time that isn’t so. Instead, rationality frequently wilts when the institutional imperative comes into play.

o 我最意外的发现：商业中有一种看不见的力量至关重要，我们可以称之为“制度性强制力（the institutional imperative）”。在商学院时，没人暗示过它的存在；刚踏入商业世界时，我也并未直觉地理解它。我曾以为体面、聪明且经验丰富的经理人会自然而然地做出理性决策。但随着时间推移，我发现并非如此：一旦制度性必然发挥作用，理性常常会枯萎。

Institutional dynamics, not venality or stupidity, set businesses on these courses, which are too often misguided. After making some expensive mistakes because I ignored the power of the imperative, I have tried to organize and manage Berkshire in ways that minimize its influence. Furthermore, Charlie and I have attempted to concentrate our investments in companies that appear alert to the problem.

让企业走上这些道路的，是制度动力学，而非贪婪或愚蠢——尽管这些道路往往被事实证明是误导。在因忽视这种力量而付出昂贵代价之后，我努力以尽量降低其影响的方式来组织和管理 Berkshire。此外，Charlie 和我也尝试将投资集中在那些对这一问题保持警觉的公司。

institutional imperative：（1）“institutional”：系统性的、有固定剧本的愚蠢行为；（2）“imperative”：接受命令、按命令行事，当一个人的心理惯性获得很多次的正反馈以后，行为模式就固定下来了。巴菲特在这里列举的4点是经过仔细思考的。

1、抗拒改变

抗拒改变排在第一点很容易理解，但抗拒改变的原因有多种解释，比如，条件反射、局部泛化、Basic Trust/Basic Mistrust，最终极的源头很可能是Basic Trust/Basic Mistrust，Basic trust认为世界是基本值得信任的，第一反应是“怎么把事情做的更好”；Basic mistrust认为世界基本不可信任，第一反应是自我防御，Basic Mistrust更容易被条件反射控制。

2、填满，不留空隙

只要系统焦虑，就需要一个动作释放焦虑，这个动作未必解决问题，但能缓解“什么都不做”的不适。

3、自我强化

同结构的人互相识别、互相强化，组织会把“不合拍”的人筛掉，配合强化的人留下来。

4、盲目模仿

已经完成全套训练的巴甫洛夫的狗，没有真正的主体性，只有条件反射式的、对外部信号的服从。条件反射一旦建立，能做的就不是消除了，而是更强的新反射来覆盖它，但新反射必须提供足够强的无条件刺激，才能与旧反射竞争。

Appendix.6.《1991 Warren Buffett.Three Lectures to Notre Dame Faculty》

... Incidentally, I would say that almost everybody I know in Wall Street has had as many good ideas as I have, they just had a lot of \[bad] ideas too. And I’m serious about that. I mean when I bought Western Insurance Security selling at \$16 and earning \$20 per share, I put half my net worth into it. I checked it out first – I went down to the insurance commission and got out the convention statements, I read Best’s, and I did a lot of things first. But, I mean, my dad wasn’t in it, I’d only had one insurance class at Columbia – but it was not beyond my capabilities to do that, and it isn’t beyond your capabilities.

……顺便说一句，我想说的是，在华尔街我认识的几乎所有人都有和我一样多的好主意，他们只不过也有很多[糟糕的]主意。我是认真的。比如当我以每股16美元买入、而每股赚20美元的Western Insurance Security时，我把自己一半的净资产都投了进去。我事先做过调查——我去了保险监管机构，调出了年会报告，我查阅了《贝斯特年鉴》，还做了很多其他功课。但我的父亲并没有参与，我在哥伦比亚大学也只上过一门保险课——然而做这件事并未超出我的能力范围，也不会超出你们的能力范围。

Now if I had some rare insight about software, or something like that – I would say that, maybe, other people couldn’t do that – or biotechnology, or something. And I’m not saying that every insight that I have is an insight that somebody else could have, but there were all kinds of people that could have understood American Express Company as well as I understood it in ‘62. They may have been...they may have had a different temperament than I did, so that they were paralyzed by fear, or that they wanted the crowd to be with them, or something like that, but I didn’t know anything about credit cards that they didn’t know, or about travelers checks. Those are not hard products to understand. But what I did have was an intense interest and I was willing, when I saw something I wanted to do, to do it. And if I couldn’t see something to do, to not do anything. By far, the most important quality is not how much IQ you’ve got. IQ is not the scarce factor. You need a reasonable amount of intelligence, but the temperament is 90% of it.

如果我对软件之类的领域有某种罕见的洞见——那我会说，也许别人就做不到——比如生物技术。但我并不是说我拥有的每一个洞见别人都可以拥有，不过在1962年，有很多人本可以像我一样理解美国运通公司。他们也许……他们也许与我有着不同的性情，以至于被恐惧麻痹，或者想要随大流，诸如此类；但关于信用卡或旅行支票，我并不比他们了解更多。这些产品并不难理解。但我拥有的是强烈的兴趣，并且当我发现想做的事情时，我愿意去做；而当我找不到可做之事时，就什么也不做。最重要的品质绝不是你的智商有多高。智商并不是稀缺因素。你需要适量的智力，但性格（气质）占了90%。

可靠性的重要性显而易见，情绪在很大程度上决定了进入大脑的上下文，巴菲特在定义“institutional imperative”时把抗拒改变排在第一点。

1、为什么是抗拒改变？一个合理的解释是自我防御，说的更直白点是某种程度的认知困难或者认知障碍，没有可靠性、不能正常运作的原因是认知障碍；

2、为什么是自我防御？有一组心理学的概念：Basic Trust / Basic Mistrust，可以理解为：在人生最初的几年里，世界在你心里被写成了“大体可相信”还是“大体靠不住”，这是结构性的差异，“大体靠不住”的会不断发展自己的防御能力。

3、Basic Mistrust缺少能根据事实变化调整知识结构的心理机制，很容易滑向条件反射，以下是巴甫洛夫实验中得出的几点规律。

在巴甫洛夫的实验里，铃声（中性刺激）与食物（无条件刺激）反复配对后，铃声本身就能引发分泌唾液的条件反射。

（1）消退不会擦除原来的条件反射

巴甫洛夫发现，当你停止配对铃声和食物后，狗的唾液分泌确实会逐渐减弱，这叫"消退"（extinction）。但他随后发现了一个关键现象：消退并没有擦除原来的条件反射，它只是在旧的关联之上叠加了一层新的抑制性学习。旧的神经通路还在那里，只是被暂时压制了。

（2）间歇性的强化更难消退

间歇性强化所建立的行为，比持续强化的行为更难消退，比如，短视频平台的推荐算法恰恰就是一个间歇性强化系统——你刷十条可能有三条让你觉得很爽，但你不知道是哪三条，这跟老虎机的原理一模一样。而这种变比率强化（variable ratio reinforcement）建立起来的行为模式，在所有类型的条件反射中是最抗消退的。

（3）刺激泛化

巴甫洛夫发现，狗不仅对原来的铃声产生反应，对类似的声音也会产生反应，比如，"刷短视频"这个具体行为，而是一整套与即时满足相关的刺激模式——快速切换、强烈视觉刺激，前面看到一则创业的故事，短视频兴起以后零售卖的越来越好，这就是泛化。

巴甫洛夫的实验告诉我们，条件反射一旦建立，你能做的最好的事情不是消除它，而是在它之上建立一个更强的新反射来覆盖它。但新反射必须提供足够强的无条件刺激，才能与旧反射竞争。

That’s why Graham is so important. Graham’s book [The Intelligent Investor] talks about the qualities of temperament you have to bring to the game, and that is the game.

这就是格雷厄姆如此重要的原因。格雷厄姆的书《聪明的投资者》讨论了你在投资游戏中必须具备的性格特质，而那正是游戏本身。

*****

(Garbled) and some of you will be stars, and some of you will be less than stars. But it won’t correlate with your IQ at all. But you have it already. And everybody in the room has it. I’ll give you a little question. Let’s think about this. Let’s say that everyone here got a bonus when they left Notre Dame. Let’s say for $25,000 you could buy a 10% interest in the income of any one of your classmates that you wanted to. Now, what are you thinking about? You can take anybody in this room, and for $25,000 buy a 10% interest in their income for life. If they make $25,000 the first year, you make $2,500. If they’re unemployed, you don’t get anything. If they get stock options, you get 10% of the stock options. What are you thinking about as you look around? Are you thinking about which ones are the smartest? It’s interesting what ingredients you think now are what’s going to produce that in the classroom. I would suggest that you start thinking about, assuming you have a 10% interest, what qualities you want to have. I know I would take Tom Murphy. Why? Well, he’s got an IQ, there’s no question about that. But, that’s something one or two others might have as well.

（录音模糊）将来你们有人会成为明星，有人不那么出色，但这与智商毫无关系，而是你们已经具备的东西，房间里每个人都有。我给大家出个小问题：假设各位毕业时拿到一笔奖金，用 2.5 万美元就能买下任何一位同学未来收入 10% 的权益。你会怎么想？你可以挑这里任何人，花 2.5 万美元买下其终身收入一成。他第一年赚 2.5 万美元，你拿 2500；若他失业，你得零；若他拿期权，你得其中 10%。环顾四周你在想什么？你在挑最聪明的同学吗？有趣的是，现在你认为课堂上哪些因素能产生未来回报。我建议你思考：既然有 10% 股权，你希望那人具备哪些品质？我知道我会选汤姆·墨菲。为什么？他当然有智商，但那点其他人也有。

关于性格的假设，这是目前找到的最早的版本，这个假设在后面反反复复讲了很多遍，2013年以后成为股东大会后的一个5公里的跑步项目，名称就叫“Invest in Yourself”。

巴菲特说那点智商大家都有，这个问题值得进一步的思考，现在大家都已经用上人工智能，如果智商不那么重要，人工智能又在解决什么问题？我们可以从巴菲特说的“integrity、intelligence、energy”，一步步拆解。

1、我自己的理解，integrity是一种结构化的知识，是“intelligence、energy”组合作用的结果，可以把intelligence看作算法，energy看作算力，两者之间算力是关键变量，计算机的发展进程中已经证明这一点；

2、从算力的角度入手，Basic Trust / Basic Mistrust在算力的输出上有很大的区别，分低能量的Basic Mistrust、高能量的Basic Mistrust，以及Basic Trust，各自的特点如下：

（1）低能量的Basic Mistrust

第一反应是直接的回避问题，这种情况下分配给每个主题的算力可想而知，脑子里都是一些支离破碎的东西。

（2）高能量的Basic Mistrust

在巴菲特的大量评论中一直在提示，很多爬上CEO位置的都是这个类型的性格，为什么？这类性格的第一反应是占据心理上的高位，比如，做大英雄，可以为英雄梦付出任何的代价，如果选定了某个英雄梦的主题，那么分配给这个主题的算力有时候是足够多的。

但问题在于对齐（Alignment），巴菲特的总结是“忘记目标后的加倍努力”，比如，贝佐斯、马斯克都有非常好的想法，但因为英雄梦和事情本身不是一回事，比如，没有经济效益、不赚钱但只要符合英雄梦就不会停下来，过度压缩的结果还包括没办法进入更高层次的泛化，能力被锁定在局部泛化中。

（3）Basic Trust

Basic Trust的结构性优势在于正确的事实能进入上下文，可以根据事实调整算法，低能量的Basic Mistrust、高能量的Basic Mistrust都做不到这一点。

3、integrity的本质是一套结构化的、有清晰边界的思维框架，没有欠拟合（低能量的Basic Mistrust），也没有过拟合（高能量的Basic Mistrust），能根据事实的变化不断调整自己的想法。“愿意把功劳多归于他人、不走捷径、兑现所有承诺——还额外多给一点”是处于结构性优势位置的外在表现。

Think about why you picked him or her, and how much of that is transferable to you. And it usually won’t be anything you can’t attain yourself. But if it’s qualities of character, or qualities of enthusiasm, or whatever it may be, most of those things you can pick up.

想想你为何选中某人，以及这些品质有多少能转移到你身上。通常它们都不是你自己无法获得的。如果是品格、热情等特质，大多数你都能培养。

Appendix.7.《1993-03-01 Warren Buffett's Letters to Berkshire Shareholders》

In the past, I’ve observed that many acquisition-hungry managers were apparently mesmerized by their childhood reading of the story about the frog-kissing princess. Remembering her success, they pay dearly for the right to kiss corporate toads, expecting wondrous transfigurations. Initially, disappointing results only deepen their desire to round up new toads. (“Fanaticism,” said Santyana, “consists of redoubling your effort when you’ve forgotten your aim.”) Ultimately, even the most optimistic manager must face reality. Standing knee-deep in unresponsive toads, he then announces an enormous “restructuring” charge. In this corporate equivalent of a Head Start program, the CEO receives the education but the stockholders pay the tuition.

过去，我注意到许多渴望并购的经理人似乎受到了童年读过的“公主亲青蛙”故事的蛊惑。记住了公主的成功，他们愿意付出高价去“亲吻”企业癞蛤蟆，指望出现奇迹般的蜕变。起初，令人失望的结果只会加深他们继续围捕新癞蛤蟆的欲望。（正如 Santyana 所言：“狂热就是在忘记目标后加倍努力。”）最终，即便最乐观的经理人也不得不直面现实。站在齐膝深的“无动于衷”的癞蛤蟆堆中，他随即宣布一笔巨额的“重组”费用。在这种企业版的“Head Start”项目里，CEO 得到教育，股东则支付学费。

In my early days as a manager I, too, dated a few toads. They were cheap dates - I’ve never been much of a sport - but my results matched those of acquirers who courted higher-priced toads. I kissed and they croaked.

我作为经理人的早年，也“约会”过几只癞蛤蟆。它们算是“便宜的约会”——我从来不是个大手大脚的人——但我的结果与那些追求高价癞蛤蟆的并购者并无二致。我亲了，它们就“呱呱”了。

After several failures of this type, I finally remembered some useful advice I once got from a golf pro (who, like all pros who have had anything to do with my game, wishes to remain anonymous). Said the pro: “Practice doesn’t make perfect; practice makes permanent.” And thereafter I revised my strategy and tried to buy good businesses at fair prices rather than fair businesses at good prices.

在经历了数次这类失败之后，我终于想起一位高尔夫职业球手曾给过我的一条有用建议（和所有见过我球技的职业球手一样，他希望保持匿名）。这位球手说：“练习并不会让你完美；练习只会让你的动作定型。”此后我调整了策略，努力以合理价格买入好企业，而不是以好价格买入一般的企业。

“忘记目标后加倍努力”，从财务理性的角度出发，巴菲特要求每个想法都是结构化的：（1）下行保护（7-8%，至少90%的信心）；（2）可能的上行收益（7%，至少50%的信心），要把所有的想法压缩成正确的知识结构，是个简单但不容易做到的工作，而几乎所有的认知障碍都跟信息的压缩质量有关，以下是上述思维框架的下一层结构。

1、Right Business

对一家企业长期经济特征能够被评估的确定性，包括：

（1）持久稳定的行业+持久的竞争优势；

（2）持久的竞争优势来自哪里？强大的全球性品牌？结构性的成本优势？

（3）有没有足够高的经济效率？有没有定价权？是不是符合少变化、少杠杆、长期高ROE的特征？

（4）能不能经受糟糕的管理？是不是依赖明星CEO？

2、Right People

前置判断：（1）热情（passion）：大脑对某类问题有自然的奖励；（2）聪明（intelligence）：能将上下文压缩成简单、优雅的知识结构，然后是以“对决策有用的准确度”对两个关键性的问题做出判断：（1）有没有能力？（2）是不是值得信任？

1、管团队能够被评估的确定性？

检验管理层经济表现的首要标准，是其能否在所运用的股本上，在不过度依赖杠杆的情况下取得足够高的收益率。

（1）实现企业全部潜力的能力？

取得了什么样的进展？同时期的竞争对手又取得了什么？接手时是什么样的“牌面”？又是如何打好手中的“牌”的？
一个已经高成本运营的经理人，常常会异常机敏地找到增加管理费用的新方法；而一个运营紧凑的经理人，通常会继续找到更多削减成本的方法，即使他的成本已经远低于竞争对手。

（2）明智运用现金流的能力？

每留存的1块钱能不能创造多于1块钱的市场价值？
有没有把资本投入低利润项目的记录？会不会“忘记目标后加倍努力”？

2、是否可以被指望把企业回报导向股东、而非导向自身的确定性？

管理团队如何对待自己，以及如何对待股东？
企业文化能不能提供护栏？商业模式能不能提供护栏？

上述信息的压缩质量直接影响想法的确信程度。

Appendix.8.《1994-03-01 Warren Buffett's Letters to Berkshire Shareholders》

In our opinion, the real risk that an investor must assess is whether his aggregate after-tax receipts from an investment (including those he receives on sale) will, over his prospective holding period, give him at least as much purchasing power as he had to begin with, plus a modest rate of interest on that initial stake. Though this risk cannot be calculated with engineering precision, it can in some cases be judged with a degree of accuracy that is useful. The primary factors bearing upon this evaluation are:

在我们看来，投资者必须评估的真正风险，是：在他预期持有期内，一项投资为他带来的税后总回款（包括出售时收到的款项）是否至少能给他带来——（1）不低于起初那笔投入的购买力；并且（2）在这笔初始投入之上再加上一点温和的利息回报。尽管这种风险无法用工程学那样的精度去计算，但在某些情况下，它能以对决策有用的准确度被判断出来。影响这一判断的主要因素包括：

1.The certainty with which the long-term economic characteristics of the business can be evaluated;

对一家企业长期经济特征能够被评估的确定性；

2.The certainty with which management can be evaluated, both as to its ability to realize the full potential of the business and to wisely employ its cash flows;

对管理层能够被评估的确定性——既包括其实现企业全部潜力的能力，也包括其明智运用现金流的能力；

3.The certainty with which management can be counted on to channel the rewards from the business to the shareholders rather than to itself;

对管理层是否可以被指望把企业回报导向股东、而非导向自身的确定性；

4.The purchase price of the business;

企业的买入价格；

5.The levels of taxation and inflation that will be experienced and that will determine the degree by which an investor’s purchasing-power return is reduced from his gross return.

未来将经历的税收与通胀水平——它们将决定投资者的“购买力回报”相对于名义回报会被削减到何种程度。

These factors will probably strike many analysts as unbearably fuzzy, since they cannot be extracted from a data base of any kind. But the difficulty of precisely quantifying these matters does not negate their importance nor is it insuperable. Just as Justice Stewart found it impossible to formulate a test for obscenity but nevertheless asserted, “I know it when I see it,” so also can investors—in an inexact but useful way—”see” the risks inherent in certain investments without reference to complex equations or price histories.

这些因素在许多分析师看来，恐怕会“模糊得让人受不了”，因为它们无法从任何数据库中提取出来。但无法精确量化，并不意味着这些问题不重要；也不意味着困难不可克服。就像 Justice Stewart 认为他无法为“淫秽”制定一个可操作的测试标准，却仍然断言：“我一看就知道（I know it when I see it）。”同样，投资者也可以在不精确但有用的意义上，在不依赖复杂方程或价格历史的前提下，“看见”某些投资中内含的风险。

巴菲特在这里明确讲到需要对Right Business、Right People做出模糊正确（Approximately Right）的判断。

1、管团队能够被评估的确定性（有没有能力）？

（1）实现企业全部潜力的能力？

假设开个咖啡店，能不能做到足够的好（前5%，或者更好）？能不能从一些行为上的细节判断这个人有没有实现全部潜力的可能性？比如，一个已经高成本运营的经理人，常常会异常机敏地找到增加管理费用的新方法；而一个运营紧凑的经理人，通常会继续找到更多削减成本的方法，即使他的成本已经远低于竞争对手。

（2）明智运用现金流的能力？

从咖啡店赚到的钱该如何安排？能不能不依赖过多历史数据的情况下做出判断？比如，有没有把资本投入低利润项目的记录？会不会“忘记目标后加倍努力”？是不是热爱业务本身？还是把英雄梦、规模、地位、技术难度、自我证明当成真实的目标？

2、是否可以被指望把企业回报导向股东、而非导向自身的确定性（是不是值得信任）？

人的注意力的来源存在结构性的差异，这是一个更简单、更直接的判断标准，比如，有一种人满脑子都是自己，而另一种人“总是愿意把功劳多归于他人、不走捷径、兑现所有承诺——还额外多给一点的人”。

Appendix.9.《1994-04-25 Berkshire Hathaway Annual Meeting》

18. “Two yardsticks” for judging management

“两把尺子”评估管理层

AUDIENCE MEMBER: Secondly, you talk about good management with corporations and that you try and buy companies with good management. I feel that I have about as much chance of meeting good managers, other than yourself, as I do bringing Richard Nixon back to life. How do I, as an average investor, find out what good management is?

AUDIENCE MEMBER：其次，您常谈到优秀的企业管理层，并且尽量收购拥有优秀管理层的公司。除了您本人之外，我觉得自己遇到一个好管理者的概率，差不多等于把 Richard Nixon 复活的概率。作为一个普通投资者，我如何判断什么是好的管理层？

WARREN BUFFETT: Well, I think you judge management by two yardsticks. One is how well they run the business and I think you can learn a lot about that by reading about both what they’ve accomplished and what their competitors have accomplished, and seeing how they have allocated capital over time. You have to have some understanding of the hand they were dealt when they themselves got a chance to play the hand. But, if you understand something about the business they’re in — and you can’t understand it in every business, but you can find industries or companies where you can understand it — then you simply want to look at how well they have been doing in playing the hand, essentially, that’s been dealt with them. And then the second thing you want to figure out is how well that they treat their owners. And I think you can get a handle on that, oftentimes. A lot of times you can’t. I mean it — they’re many companies that obviously fall in — somewhere — in that 20th to 80th percentile and it’s a little hard to pick out where they do fall.

WARREN BUFFETT：嗯，我认为评估管理层有两把尺子。第一，看他们把企业经营得如何。我觉得你可以通过阅读他们取得了什么、他们的竞争对手取得了什么，以及他们这些年如何进行资本配置，来学到很多东西。你必须理解他们接手时拿到的“底牌”是什么。当你对他们所处的业务有所了解——你不可能了解所有行业，但总能找到一些你能看懂的行业或公司——那么你就要看他们如何打好手中的“牌”。第二，看他们如何对待所有者。我认为很多时候你能抓住这一点，当然也有不少时候很难。我是说，有许多公司显然落在第20到第80百分位之间的某个位置，而要精确分辨它们处在其中的哪一段，确实不太容易。

But, I think you can usually figure out — I mean, it’s not hard to figure out that, say, Bill Gates, or Tom Murphy, or Don Keough, or people like that, are really outstanding managers. And it’s not hard to figure out who they’re working for. And I can give you some cases on the other end of the spectrum, too. It’s interesting how often the ones that, in my view, are the poor managers also turn out to be the ones that really don’t think that much about the shareholders, too. The two often go hand in hand. But, I think reading of reports — reading of competitors’ reports — I think you’ll get a fix on that in some cases. You don’t have to — you know, you don’t have to make a hundred correct judgments in this business or 50 correct judgments. You only have to make a few. And that’s all we try to do. And, generally speaking, the conclusions I’ve come to about managers have really come about the same way you can make yours.

但是，我认为你通常还是能分辨出来——比如说，Bill Gates、Tom Murphy、Don Keough 这些人，确实是杰出的管理者；而他们究竟在为谁服务，也不难判断。相反一端也有不少例子。有意思的是，在我看来那些管理能力差的人，往往也恰恰是不怎么把股东放在心上的人，这两点经常是相伴相生的。不过，我认为通过阅读年报——以及竞争对手的年报——在某些情况下你能把握住要点。在这个行业里，你不需要——你懂的——做对一百次判断，甚至不需要做对五十次判断。你只需要做对少数几次。我们所做的，也只是如此。总体而言，我对管理层的判断，基本上就是通过你也可以使用的同样方法得出的。

I mean they come about by reading reports rather than any intimate personal knowledge or — and knowing them personally at all. So it — you know, read the proxy statements, see what they think of — see how they treat themselves versus how they treat the shareholders, look at what they have accomplished, considering what the hand was that they were dealt when they took over compared to what is going on in the industry. And I think you can figure it out sometimes. You don’t have to figure out very often. Charlie?

我的意思是，这些判断主要来自阅读报告，而不是源于什么私人接触，或者对他们的私下了解。所以——去读代理声明（proxy statements），看看他们在意什么——看看他们如何对待自己与如何对待股东；结合他们接手时拿到的“底牌”，与行业同时期的情况对比，去看他们到底做出了什么成绩。我想，有时候你能看明白。你也不必次次都看明白。Charlie？

对照Right People的三个判断，在后来的评论中，巴菲特把“intelligence、energy”合并成able，integrity改成trustworthy。intelligence、energy和integrity在结构上是重复的，巴菲特自己也承认，“在我看来那些管理能力差的人，往往也恰恰是不怎么把股东放在心上的人，这两点经常是相伴相生的”，但为什么要这么定义？我自己的理解如下：

1、更简单、更直接的判断方法

integrity、trustworthy直接从注意力的来源做出判断，很多时候这个方法的效率更高。

2、integrity要不要分拆成intelligence、energy？分拆有分拆的好处。

巴菲特是个现实主义者，比如，2000 年前后，互联网危机时投资Amazon债券的例子，如果直接从注意力的来源做出判断，Bezos的确定性是存疑的，Amazon在很长时间里没有赚过钱。再看巴菲特的定义：是否可以被指望把企业回报导向股东、而非导向自身的确定性？这一句可以套在亚马逊身上，也可以套在Bezos身上，是不是Bezos的冲劲是拿着股东的钱不断扩张？最终一分钱没有赚到？

但是，买入债券不用考虑股东能不能赚钱，只需要考虑企业会不会倒闭。巴菲特认为Bezos的intelligence、energy都很强，Bezos的energy让股东赚不到钱不等于企业会失败，Bezos自己也说的很清楚，他倾向于不断做大而不是赚很多的钱。

以下是人工智能做的补充（1）。

字面理解的"导向自身"——超额薪酬、自我交易、暗中转移资产——这些是显性的、容易被审计和监管发现的版本。但 Buffett 这句话真正的力量在于，它同时把另一类不那么明显的形式也覆盖了：用股东的钱满足自己的心理账户。

英雄梦是其中最贵的一种。规模、行业地位、媒体上的形象、被列入"改变世界的企业家"名单——这些不是钱，但需要烧钱来实现。烧的钱是股东的，得到的满足是经理人的。从经济实质看，这和直接装进口袋没有本质区别，只是支付方式不一样：一种是现金，一种是心理收益。后者反而更隐蔽，因为它可以用"长期战略"、"行业愿景"、"颠覆传统"这套语言包装起来，几乎所有的扩张性叙事都可以套用。

这也解释了为什么 institutional imperative 里 Buffett 反复点名"扩张冲动"和"模仿同行的并购"——这些行为表面上是商业决策，实质上都是 CEO 在用股东的钱购买自己的心理满足。Buffett 这句话之所以能压住这一切，是因为它问的不是"钱去了哪里"，而是"决策的受益人到底是谁"——一旦把问题这样问出来，付现金给自己和烧钱满足自我，就被归到了同一类。

所以 trustworthy 真正难判断的地方不在显性偷窃，那些有审计和法律兜底；难的是判断一个看起来理性的扩张决策背后，受益人到底是企业还是经理人的自我。这一层只能从注意力的来源去看——内部计分卡的人不需要通过规模来证明什么，所以他的扩张是因为机会，外部计分卡的人需要规模本身，所以他的扩张是因为需要扩张这件事本身。

Bezos 的案例正好落在这个分界线上：他的扩张不是为了把现金转移给自己，但他的目标函数（做大）本身就是外部计分卡的产物，烧的是股东的钱，喂的是自己的英雄叙事。Buffett 那句压缩，准确地把这种情况也圈进了"导向自身"。

以下是人工智能做的补充（2），比我自己说的更狠。

拿股东的钱实现自己的伟大梦想，造福了全人类，所有股东自愿做出牺牲，就像在完成一种使命，这跟工蜂的使命没什么区别，而且更狠，工蜂至少在基因层面上服务于自己——蜂后携带的基因就是工蜂的基因，所以工蜂的"使命"虽然牺牲了个体，但在基因这个真正的账户上是有回报的。

Musk 的股东和 Musk 之间没有任何基因关系。他们交出去的是真金白银，换回来的是参与一个伟大叙事的心理满足——"我也在改变世界"、"我和未来站在一起"、"我看出来了别人看不出来的东西"。这些都是心理收益，不是经济收益。

更值得注意的是这套机制怎么把"自愿"做出来的。"造福全人类"这句话不是结论，是包装——它给整个安排提供了一层让心理上能接受的外壳。如果叙事是"我用你们的钱，建立一个让我自己满足的帝国"，没人会买账；但叙事变成"我用你们的钱，让人类移民火星/拥有清洁能源/对抗 AI 危险"，捐出去就成了使命。意义本身被生产出来，作为对经济损失的非货币补偿。这正是这种安排比纯粹的诈骗更稳定的原因——它给受害者提供了一个不必承认自己是受害者的出口。

你之前提的"鸡不知道自己在做什么"，跟工蜂这个比喻其实说的是同一件事。工蜂不需要意识到自己在为蜂后服务，它就是这么做的；股东也不需要意识到自己在资助别人的英雄账户，他们觉得自己在投资未来。意识本身被叙事覆盖了，所以"自愿"这个词在这里要打个问号——表面上是自愿，结构上更像是默认设置在运行。Basic mistrust 的人尤其容易被这种叙事抓住，因为他们本来就在寻找一个外部权威来定义意义，找到了 Musk 这样的英雄之后，把自己交付出去反而是减少认知负担的方案。

CHARLIE MUNGER: Nothing to add.

CHARLIE MUNGER：无可补充。

Appendix.10.《1994 Warren Buffett.UNC Business School》

I met twelve people — roughly twelve — that evening. A dozen or so. I probably knew four or five of them by sight and there were probably six or seven that I didn’t know, though I probably knew their names. As I interviewed these people, making what to me was the most important hire — by far — that I would ever have to make…

那天晚上我见了十二个人——大约十二个。一打左右。我可能只见过其中四五个人的面孔，另外六七个我没见过，但我大概知道他们的名字。在面试这些人时，我做的可以说是我一生中最重要的一次雇佣决定……

It may be of interest to you, [though] it may not be quite as dramatic when you get hired, but maybe some of the same things will be going through the mind of the person that’s looking at you [during your job interview process]. The good news is that I did not ask them their grades in business school. (Laughs) And the bad news, of course, is that I didn’t ask them if they had been to business school at all. I never looked at a resume. It wouldn’t have made any difference. I knew that all twelve had the IQ to handle the job. No one had gotten to that position in the firm without being smart enough to know all the instruments and know how markets work and all of that sort of thing. It wasn’t a question of IQ at all.

这或许对你们有意思，虽然你们被录用时场景没那么戏剧化，但面试你们的人脑子里可能也会闪过类似的念头。好消息是，我没有问他们商学院的成绩（笑）。而坏消息是，我甚至没问他们是否读过商学院。我从没看过简历。那也没什么区别。我知道这十二个人都有足够的智商胜任工作。在那家公司，没有人能爬到那个位置，而不聪明到能了解各种工具、理解市场如何运作等等。这完全不是智商的问题。

It’s just sort of hearing how their [mental] machinery works. You can tell the people who are very full of themselves. It’s very easy to do. You find a few of those in investment banking, believe it or not. (Laughs) You can probably detect, perhaps, the ones who won’t have the courage in a situation like that.

关键是听一听他们头脑的“运转方式”。你能看出哪些人满脑子都是自己。这很容易。信不信由你，在投行里你能发现一些这样的人（笑）。你或许也能分辨出，在那种情况下哪些人不会有勇气。

I wanted the qualities, the same thing Ben Franklin wrote about a couple of hundred years ago, the people who are always willing to give the other person more credit than themselves and who don’t cut corners and who deliver all they promise — and a little extra. Essentially, I was very lucky because I found that individual: a fellow named Deryck Maughan.

我想要的品质，正是 Ben Franklin 几百年前写到的：那些总是愿意把功劳多归于他人、不走捷径、兑现所有承诺——还额外多给一点的人。总之，我很幸运找到了这样的人：一个叫 Deryck Maughan 的人。

可以用更结构化的方式描述这一类的性格特征，参考：《1997-05-05 Berkshire Hathaway Annual Meeting》。

You really ought to figure out how you can be happy with the present amount of money you’ve got, and then figure that everything else is all to the good as you go along.

但正常情况下，你真正应该做的是：先想清楚如何对自己现在拥有的钱“感到满足”，然后把之后的增量收益都视为“锦上添花”。

1、足够安全的下行保护-》有一个充满安全感的地基

对自己的现状是不是满意？但凡不满意的、地基上有问题的非常麻烦。

2、可能的上行收益-》增量收益都视为“锦上添花”

是不是愿意把功劳多归于他人、不走捷径、兑现所有承诺——还额外多给一点？

*****

I often use this illustration [when speaking to] a class. I say to them, “If you looked under your seat today and one of you had this lucky ticket and the one who got that ticket got to pick one of their classmates — you’ve been together for a couple of years and you could pick any one you want — if you had an hour to make the decision and you would get 10% of the earnings of that individual for the rest of their life, [who would you pick]?”

我经常在课堂上用这个例子。我会对学生说：“如果今天你们座位底下有一张幸运票，谁拿到票就可以挑选一个同学——你们已经相处了几年，可以选任何一个人——你有一小时做决定，并且从此以后能获得这个人一生收入的 10%，你会选谁？”

You don’t pick the guy with the richest father. We’ll assume that everybody’s starting from scratch. But what would you think about, in that hour, in terms of who you’d pick? Would you think about who had the highest grades in the class? Probably not. Would you think about who had the highest IQ? Probably not. The best looking? Probably not. A whole bunch of things would go through your mind — and I think you’d be amazed at how most of you would settle on a relatively few individuals.

你不会选那个父亲最有钱的人。我们假设大家都是从零开始。但在那一小时里，你会思考什么来决定挑谁？会考虑谁成绩最高吗？大概不会。会考虑谁智商最高吗？大概不会。长得最好看？也大概不会。你脑子里会闪过许多因素——我想你会惊讶，大多数人最后都会集中挑选相对少数的几个人。

But the interesting thing is that, when you think about what is going through your mind, you’re not thinking about things that are impossible for you to achieve yourself. You’re not thinking about who can jump seven feet, who can throw a football 65 yards, who can recite pi to 300 digits, or whatever it may be. You’re thinking about a whole bunch of qualities of character — and the truth is that every one of those qualities is obtainable. They are largely a matter of habit.

但有趣的是，当你思考这些因素时，你想的并不是那些自己不可能做到的事。你不会考虑谁能跳过七英尺，谁能把橄榄球扔 65 码，谁能背出圆周率 300 位等等。你想的是一堆品格上的品质——而事实是，这些品质每个人都可以获得。它们在很大程度上是习惯问题。

Ben Franklin — and, actually, my old boss Ben Graham… Ben Graham, when he was twelve years old, wrote down all the qualities that he admired in other people — and he wrote down the qualities that he found objectionable. He looked at that list and there wasn’t anything there about being able to run the 100-yard dash in 9.6 [seconds] or high jumping seven feet.

Ben Franklin——实际上，还有我以前的老板 Ben Graham……Ben Graham 在 12 岁时就写下了他欣赏的所有品质——以及他厌恶的品质。他看着那张清单，里面没有任何一条是关于 9.6 秒跑完 100 码，或跳过七英尺。

They were all things that were simply a matter of deciding whether you were going to be that kind of person or not. When you’re young, you really have that opportunity. As you get older, it’s tougher. Somebody said that the chains of habit are too light to be felt until they’re too heavy to be broken. And, in terms of behavior, I see that all the time. The time to form the right habits is very early. The right habits, the habits that you would admire in someone else and [that make you] want to buy 10% of whatever individual that is, they’re absolutely obtainable.

它们全都是只需决定你是否要成为那种人的事。当你年轻时，真有这个机会。随着年纪增长，就更难了。有人说过，习惯的枷锁轻到无法察觉时，等你能感觉到时，它们已经重到无法打破。而在行为方面，我经常看到这一点。培养正确习惯的时间就在早期。那些你会在别人身上欣赏、会让你愿意买下其 10% 的好习惯，完全可以自己获得。

If you had to sell short one other member of your class and you had to pay 10% of their earnings, again, you wouldn’t pick the person with the lowest grades. You’d think about the person who always cut corners or claimed credit for things that they didn’t do and never showed up on time. All these little things that seem like nothing, but they determine whether if you have a 300 horsepower motor, whether you get 300 horsepower output out of it or whether you get a whole lot less. The people I see that function well are not necessarily the people with the biggest motors, but they’re the people with the most efficient motors. And it’s those qualities of character that really make an enormous difference.

如果你必须做空班上某个同学，并且得支付他收入的 10%，你依然不会选成绩最低的那个人。你会想到那个总是偷工减料、抢占他人功劳、从不准时出现的人。这些看似微不足道的小事，决定了如果你有一台 300 马力的引擎，是能输出 300 马力，还是远远不及。我看到那些表现出色的人，不一定是引擎马力最大的人，而是效率最高的人。而真正产生巨大差异的，就是这些品格上的品质。

“Basic mistrust”的第一反应是自我保护和处处防御，如果想象成一台AI：

（1）任何想法的初始假设就是错的

“institutional imperative”的第1点：抗拒改变，事实变了，想法能不能跟着改变？说的直白了，起点是认知障碍而不是认知困难。

（2）在任何问题的上下文中毫秒级的注入垃圾信息

“institutional imperative”的第2点：填满，不留空隙，认知障碍的生存模式。

（3）对压力的极度敏感，以及随之而来的防御和对抗情绪。

“institutional imperative”的第3-4点，自我强化、盲目模仿，这是一个非常强大的、能自我循环、自我强化的系统，大量能量都在防御上，防御系统甚至消耗、吸干所有的能量。

What I would suggest to you is to think about that person who you would buy 10% of and the one you’d short 10% of — and think about why you would do it. Then ask yourself, “What on this lefthand side of the list can’t I achieve myself?” and “What on the righthand side of the list can’t I get rid of by myself if I possess it?” In the end, decide you’re going to be the [kind of] person that you’d buy the 10% of.

我建议你们思考：你会买谁的 10%，会做空谁的 10%，再想想为什么。然后问自己：“这张清单左边的优点，有哪一条是我自己做不到的？”“这张清单右边的缺点，如果我有，又有什么是我改不掉的？”最终，决定成为那个你愿意买下 10% 的人。

这是改进后的版本，补充一个俞敏洪的例子，参考：《罗永浩讲自己的前老板，前老板之前在大学宿舍的有趣事》。

以前我的前老板也讲过，他们当年在北大读书，8个人在一个屋，就我那个前老板，铁老师，但我今天要说的不是铁老师的坏话，就说铁老师当年讲过一些有意思的事，就我的前老板，他是这样的，他就是村里来的嘛，当年一傻小子，跑到那以后呢，没睡过上铺，他来的晚了，他是8个人宿舍里到第8个，扛着扁担就来了，进来一看，只剩一个上铺了，这时候他就很慌，他没有睡过上铺，他就说能不能跟谁换一下，那谁跟你换呢？这不就像是上了飞机，跟人靠窗的说咱俩换一下，人家说哎呀，不好意思，心里想的是滚，对不对？

所以他要跟上铺的换一下，那个人想法也是滚啊，但是他就很苦，愁眉苦脸，在那就哎哎，你知道吗，他就蹲在地上了，扛着个扁担就蹲在地上，愁眉苦脸，这个时候上铺有一个见过世面的，不是下铺有一个见过世面的，他觉得哎，这小伙怎么这么慌，要不我就上铺。

哪是见过世面，是安全感，是从小对世界有基本的信任，有一个充满安全感的地基，对人比较宽容。

这个人就是我那个前老板的公司，就我当年服务的那个教育机构，新东方的联合创始人包凡一老师，他一看这个江阴县来的村里小伙，慌成这样，他说那行吧，要不我把下铺让给你吧，他就上上铺了，后来这下铺这小伙子就成了亿万富翁，然后那个从下铺去了上铺的那位包老师跟着做了联合创始人，也成了亿万富翁，所以你人生的命运很多时候真的不知道是什么样的，知道这个意思吗？不知道，这是我讲我前老板的第一个事，他同学的。

他还举了第二个例子，他们宿舍里另一个，他没点名，这是做的比较厚道的，他没点名，我也没点名，我一直说前老板，铁老师，我也没点名，他也不是真姓铁，他只是真的很铁，然后他同宿舍另一个是北京本地一小伙，每周末回去看完爹妈回来，那个时候整个中国还很穷，水果是比较金贵的东西，他每次回来，从爸妈家回来的时候，带一兜子水果来，放到自己的那个，每个人不有个小桌吗，下边有个柜子，也不锁抽屉，然后就把那个每次带过来的水果放那，每天晚饭后特别养生，就是吃那么一个两个，然后他也不出去吃，当着这个宿舍里7个吃不起水果的穷同学的面吃，最后那个人就可以想象他的人生啊，不展开说了，就到这。

注意力放在哪里的问题，Basic Trust/Basic Mistrust是一种结构性的差异，注意力放在A面，B面就看不到了，但只要注意力的方向调在同一个方向，大家看到的都是差不多的，问题是有些人出厂的默认设置是A面，有些是B面，如果要改变出厂的默认设置要尽可能的早。

That’s enough of the sermon — I’d like to answer questions now. And, really, the nastier they are, the better I like it. It’s much more fun and makes it more interesting.

布道就说到这里——现在我想回答问题。老实说，问题越尖锐，我越喜欢。这会更有趣，也更吸引人。

Appendix.11.《1995-03-07 Warren Buffett's Letters to Berkshire Shareholders》

Over time, the skill with which a company's managers allocate capital has an enormous impact on the enterprise's value. Almost by definition, a really good business generates far more money (at least after its early years) than it can use internally. The company could, of course, distribute the money to shareholders by way of dividends or share repurchases. But often the CEO asks a strategic planning staff, consultants or investment bankers whether an acquisition or two might make sense. That's like asking your interior decorator whether you need a $50,000 rug.

就长期而言，公司经营者配置资本的能力将会对企业的价值产生很大的影响。我们认为，一家好公司所能产出的现金，比其本身内部所需投入的，要多的多（至少是在早期成长期以后）。当然，公司可以通过分配股利或回购股份的方式将资本返还给股东。但是，通常企业的CEO们会咨询公司战略规划部门、顾问或是投资银行家们，搞那么一两次并购是否有利可图。这样的做法就好象是问你的室内设计师，你是否需要一张价值五万美元的地毯。

The acquisition problem is often compounded by a biological bias: Many CEO's attain their positions in part because they possess an abundance of animal spirits and ego. If an executive is heavily endowed with these qualities - which, it should be acknowledged, sometimes have their advantages - they won't disappear when he reaches the top. When such a CEO is encouraged by his advisors to make deals, he responds much as would a teenage boy who is encouraged by his father to have a normal sex life. It's not a push he needs.

并购问题又因为隐藏其后的生物本能而变本加厉：许多CEO之所以能终临高位，部分的原因正是在于他们极富动物精神与野望。如果一位经理人颇具这类特质——当然我们必须承认，有时这也是种优势——然而这种本能在他们爬上顶峰之后并不会消失。而这样的CEO被其顾问们鼓励去进行并购交易时，他的反应就会像一位十几岁的孩子被父亲鼓励去过成人的性生活一样。这样的做法是强人所难。【7】

对某件事有热情排在第一位的要素，但如果有热情没脑子就是问题。前一段跟老婆讲到一个事，我说我年轻时，睡觉前会把平日里觉得好看的姑娘一个个的过一遍，过了40岁就慢慢不想了，同样的生物本能在每个人身上的表现是不一样的。

1、有些只是心里面想一想；

2、有些会突破心理上的阻力，主动表白；

3、有些即使已经有家庭了也要去追求；

4、有些即使已经有家庭，并且对方不同意也有进一步的行为，等等。

Some years back, a CEO friend of mine - in jest, it must be said - unintentionally described the pathology of many big deals. This friend, who ran a property-casualty insurer, was explaining to his directors why he wanted to acquire a certain life insurance company. After droning rather unpersuasively through the economics and strategic rationale for the acquisition, he abruptly abandoned the script. With an impish look, he simply said: "Aw, fellas, all the other kids have one."

几年前，我的一位CEO朋友半开玩笑地在无意间透露出许多大型交易的病态过程。这位CEO当时管理着一家意外和财产险公司，在董事会上，他向董事会成员解释为何公司必须要收购一家人寿保险公司。在对此次收购的经济和战略上的理由喋喋不休之后，他突然停止了演说，带着顽皮的表情说道：“好吧！伙计们，谁叫其它财产险公司也都有一家寿险公司呢。”

At Berkshire, our managers will continue to earn extraordinary returns from what appear to be ordinary businesses. As a first step, these managers will look for ways to deploy their earnings advantageously in their businesses. What's left, they will send to Charlie and me. We then will try to use those funds in ways that build per-share intrinsic value. Our goal will be to acquire either part or all of businesses that we believe we understand, that have good, sustainable underlying economics, and that are run by managers whom we like, admire and trust.

在伯克希尔，我们的经理人们不断地通过看似平凡普通的业务，赚取惊人的回报。这些经理人们会首先在自身业务中寻找可以有效配置其利润的去处，之后再把多余的资本交回给查理跟我，然后我们会努力为这些资金寻找更好的出路以创造出更多的内在价值。我们的目标，是收购我们熟悉了解的、有持续竞争力且由我们喜爱的、由受人尊敬和值得信任的经理人所经营的公司的一部分或全部所有权。

Appendix.12.《1995-05-01 Berkshire Hathaway Annual Meeting》

WARREN BUFFETT: Charlie and I are always — we’re more interested in bad news, always, than good news. We figure good news takes care of itself. And one — we only give a couple of instructions to people when they go to work for us. And one of them is to think like an owner. And the second one is to just tell us the bad news immediately, because good news takes care of itself. And we can take bad news, but we don’t like bad news late. So, I would say, in connection with Salomon, that there is, and has been, some culture clash. And there probably almost always would be a culture clash in a business where there is that amount of tension. Whether it be the entertainment business, or the investment banking business, or the sports business, there’s going to be a certain amount of tension when — between compensation to the people that work there and compensation to the owners. And I think there’s been some — that strain has existed at Salomon from the day I was first there and far before that. I mean, I — that was no surprise. It’s understandable.

WARREN BUFFETT：我和 Charlie 一向——总是对坏消息比好消息更感兴趣。我们的看法是，好消息会自己照顾好自己。我们只给为我们工作的人两条指示：其一，要像所有者那样思考；其二，立刻把坏消息告诉我们，因为好消息会自己照顾好自己。我们能承受坏消息，但我们不喜欢“迟到的坏消息”。关于 Salomon，我会说，确实存在、也一直存在一些文化冲突。而在高度紧张的行业里几乎总会有文化冲突。无论是娱乐业、投行业还是体育业，当“员工的报酬”和“所有者的回报”之间存在张力时，就会有一定的紧张关系。我认为这种张力在我刚到 Salomon 之时、甚至更早就存在了。这并不意外，也可以理解。

You’re seeing a tension, actually, in the airline business between the people that work there and capital. And it’s produced terrible results in the airline business. And the people that work there have been able to — and I’m not talking about USAir specifically, although that’s a case. But it goes beyond that. They have had contracts, which were, as I pointed out in the report, were executed in an earlier age, which, essentially, will not allow — in many cases — capital to receive any compensation. And that produces a lot of tension. You don’t have contracts like that in the investment banking business or Wall Street, generally. But you have that same sort of tension. And changing a culture around, A, takes time and, B, probably takes some change in people. I mean, I don’t think that’s a great surprise if you expect to do it. I have — I don’t think you can find two better people than Bob Denham and Deryck Maughan. They’re smart, they’re high-grade, they’re willing to work very hard.

事实上，在航空业你也能看到员工与资本之间的张力，这给航空业带来了可怕的结果。员工们签过的合同——不单指 USAir，尽管它也是个例子——正如我在报告中指出的，很多是在更早年代签订的，实质上在许多情况下“不允许资本获得任何回报”。这会产生大量紧张关系。投行或华尔街总体上没有那样的合同，但也有同样的张力。要改变一种文化，第一，要花时间；第二，很可能要更换一些人。这并不意外。如果要做这件事，我认为你找不到比 Bob Denham 和 Deryck Maughan 更好的人选了。他们聪明、素质高、也愿意非常努力地工作。

商业模式本身导致的紧张关系。

1、高度紧张和有效约束不可能并存

《Federalist No. 10》有一个句子。

If the impulse and the opportunity be suffered to coincide, we well know that neither moral nor religious motives can be relied on as an adequate control.

只要冲动与机会得以重合，我们深知道德或宗教之动机皆不足以构成有效约束。

2、高度紧张和护城河的联系

巴菲特在评论商业模式时讲到了两种形式的护城河，一是强大的全球性品牌；二是结构性的成本优势，反过来没有这两种形式的优势，“员工与资本之间”的张力可能是很难避免的，比如，围绕Google、OpenAI、Meta之间的AI人才的激烈竞争，也包括缺少护城河保护带来的激烈竞争。

And there will be people that buy into the arrangements they want to have. And there are people that won’t. Not all of the people that have left, by a long shot, are leaving of their own volition, but most of them are. But some aren’t. I mean, there — Salomon lost a lot of money last year. And many of the people that have left were not responsible for some of those losses, but some of the people were. So, that is not something where you announce names in the paper. But some people are leaving because they can make more money elsewhere. And some people are, maybe, leaving because we think we can make more money without them. (Laughter) Charlie?

当然，会有人接受他们所制定的新安排，也会有人不接受。离开的人并非全都是主动离开的，但大多数是，有些不是。Salomon 去年亏了很多钱。许多已经离开的人并不对那些亏损负责，但也确实有人需要负责。这不适合在报纸上点名。但有些人离开是因为他们在别处能赚更多钱；也可能有些人离开，是因为我们认为“没有他们我们会赚更多”。（笑声）Charlie？

WARREN BUFFETT: My children, in many ways, are a lot smarter than I am. So, I’ve had different experience, Charlie. (Laughs) No, I think you can — you know, I’ve mainly learned by reading myself, so I don’t think I have any original ideas that — I’ve certainly got a lot — I mean, I’ve talked about reading Graham, I read Phil Fisher, and I’ve gotten a lot of ideas myself from reading. And in my own case, I mean, talk about your parents having influences, you know, my parents had an enormous influence. So, I think you can learn a lot from other people. In fact, I think, if you learn reasonably well from other people, you don’t have to get any new ideas or do much on your own. You can just apply the best of what you see.

WARREN BUFFETT：在很多方面，我的孩子们比我聪明。所以我的经历与你不同，Charlie。（笑）我主要靠自学阅读，所以我不觉得自己有什么“原创思想”。我确实收获很多——比如我读 Graham、读 Phil Fisher，很多想法都来自阅读。就我个人而言，父母的影响也极其深远。因此你完全可以从他人身上学到很多。事实上，如果你足够善于向他人学习，你无需发明新点子，也无需做太多原创；把你所见的最好实践运用起来就行。

CHARLIE MUNGER: Generally speaking, I think we always get a group of wise people after sifting millions. But I don’t think anybody’s invented a way to teach so that everybody is wise. It’s extraordinary how resistant some people are to learning anything. (Laughter and applause)

CHARLIE MUNGER：总体上说，我们总能在“筛过百万”之后得到“一撮聪明人”。但我不认为有人发明过一种教学法，能让“人人皆明智”。有些人对学习的抗拒简直“离谱”。（笑声与掌声）

WARREN BUFFETT: Really, what is astounding is how resistant they are when it’s in their selfinterest to learn. I mean, I was always astounded by how much attention was paid to Graham — I mean, he was regarded 40 years ago as the dean of security analysts —but how little attention was paid, in terms of the principles he taught. And it wasn’t because people were refuting them, and it wasn’t because people didn’t have a self-interest in learning sound investment principles. It’s just this incredible resistance to thinking or change. I mean, I quoted Bertrand Russell one time as saying — who said that, “Most men would rather die than think. Many have.” (Laughter) In the financial sense, that’s very true. It’s not complicated. I mean, human relations, you know, usually aren’t that complicated, but — and certainly it’s in people’s self-interest to develop habits that work well in human relations, but an amazing number of people seem to mess it up one way or another.

WARREN BUFFETT：真正让我震惊的是：即便“学习对他们自身有利”，他们仍会抗拒。比如 Graham——四十年前他就被称为证券分析界的“宗师”，可他所讲的原则却鲜有人认真对待。并不是人们在反驳这些原则，也不是他们不希望学习健全的投资原则，而是对“思考或改变”的抗拒。我曾引用 Bertrand Russell 的话：“大多数人宁愿去死也不愿思考；许多人已经用行动证明了这一点。”（笑声）这在金融世界尤其真实。其实并不复杂——比如人际关系，培养良好的习惯显然符合自身利益，但仍然令人惊讶地有很多人把事情搞砸。

CHARLIE MUNGER: How much has Berkshire Hathaway been copied, either in investing America or corporate America? I’m not saying we deserve to be, necessarily. But people don’t want to do it differently than they’re presently doing it.

CHARLIE MUNGER：在美国的投资界或企业界，有多少人复制了 Berkshire 的做法？我并不是说我们一定值得被学习。但人们就是不愿意改变他们正在做的事。

WARREN BUFFETT: You might argue that Mrs. B. [Nebraska Furniture Mart founder Rose Blumkin], having started what you may have seen out there this weekend, with $500 in 1937, you know, without a day in school in her life, and building that into a great enterprise, you might say, “Well, that is something to study.” I mean, is it because she couldn’t speak English when we got — you know, she got over? Maybe we can explain to people — I mean, what is there to learn from seeing somebody create an incredible success like that in a competitive business? She didn’t invent something that the world had never seen before. She didn’t have a lock on some piece of real estate that protected her from competition. You know, all of these — and yet, she accomplished something that virtually no one has accomplished. Now, why aren’t business schools studying her? You know, why are they talking about EVA, you know, economic value added, as we talked about earlier? I mean, here is a success. Something has made her a success.

WARREN BUFFETT：你可以看看 Mrs. B（Nebraska Furniture Mart 创始人 Rose Blumkin）的故事：她 1937 年拿着 500 美元起步，一辈子没上过一天学，却把事业做到你们这个周末在店里看到的规模。难道这不值得研究？是因为她初到美国时不会说英语吗？我们完全可以从这样的“在充分竞争行业里，硬是做出了几乎无人能及的成功”的案例中汲取营养。她没有发明前所未见的新东西，也没有靠某块“风水宝地”独享竞争保护。尽管如此，她还是做到了。为什么商学院不研究她？为什么还在讨论 EVA（经济增加值）之类的模型？这里摆着一个现实的成功，必然有某些关键因素造就了她的成功。

You know, is it something — is it a 200 — and she’s very smart — but is it a 220 IQ? No, it isn’t. It’s a very smart woman, but it’s not something that’s incapable of being replicated in the habits and the way of thinking. But who is studying her? I mean, they present her as a curiosity. But if you go to any of the top 20 business schools, you know, there’s not one page that’s being given to anybody to study what is an incredible success. And I just — I find that very interesting that — and to some extent, you know, I’ve seen it in the investment world. There’s this — for one thing, the high — you know, it’s probably a little discouraging to a professor of management at some major business school that has gone on to get his doctorate and everything, to think he has to come and hang around the Furniture Mart — (laughter) — study a woman in a golf cart, I mean — (Laughter) But you could — they’d be better off if they did.

你可能会说——她确实很聪明——但是不是因为“220 的智商”？当然不是。她很聪明，但更关键的是一整套“可被复制的习惯与思维方式”。可是谁在研究她？她常被当成“奇闻轶事”。去任何一所前 20 的商学院，你几乎找不到一页纸是让学生系统研究这种“惊人的成功”的。我对此感到很有意思；在投资世界里也多少能看到类似现象。某种程度上，让一位拿到博士学位的名校管理学教授想到要“去家具城蹲点”、研究坐着高尔夫球车办公的老太太——（笑声）——这可能有点“打击人”。但如果他们真这么做了，收获会更大。

“能力圈”是一个非常重要的概念，Ilya Sutskever说“prediction is very close to intelligence”，或者说“Intelligence= Compression”，我们在解释integrity时就认为这是压缩的结果，大量信息经过压缩后形成的简单、优雅的知识结构，最终形成的一个个知识结构跟“能力圈”有很大的关系，参考：《2026-05-02 Sam Altman's Vision For the Future!》。

Sam Altman:All the time. Ilya Sutskever once said a very simple sentence that as many simple sentences said really stuck in my mind, which is prediction is very close to intelligence. And the idea is that if you can Compress all of the information about the world, the state of things, whatever, into its smallest representation. And then as part of that, predict the thing that's going to happen next. You understand it in a sort of deep way. And this idea At the time, a lot of people in the AI field were quite excited about generative models for a reason they couldn't quite articulate.

一直在想。Ilya Sutskever 曾经说过一句非常简单的话，和许多简单的话一样，它深深留在了我的脑子里，那就是：预测非常接近智能。这个想法是说，如果你能把关于世界的所有信息、事物的状态等等，压缩成最小的表征，然后作为其中的一部分，预测接下来会发生什么。那就意味着你在某种深层意义上理解了它。而这个想法在当时，AI 领域很多人都对生成模型感到非常兴奋，尽管他们还说不太清楚原因。

1、相信自己就是对的

“能力圈”所呈现的聚焦和收敛跟“相信自己就是对的”有很大的关系，知识轨迹的空间太大了，一个人不可能先站在更高一层，把所有行业、所有路径、所有知识结构都看一遍，然后理性挑出最优项，真实路径只能是：（1）先碰到一件具体的事；（2）一头扎进去；（3）拿到反馈，形成局部压缩；（4）判断自己是否满意，满意就继续，不满意再换下一件。

Ilya Sutskever在讲到AGI时他认为会出现很多“narrow superintelligent AIs”，就是专业化的AI，参考：《2025-11-26 Ilya Sutskever.We're moving from the age of scaling to the age of research》。

We are talking about the good world, by the way. What’s the good world? It’s where we have these powerful human-like learners that are also… By the way, maybe there’s another thing we haven’t discussed on the spec of the superintelligent AI that I think is worth considering. It’s that you make it narrow, it can be useful and narrow at the same time. You can have lots of narrow superintelligent AIs.

顺带一提，我们现在讨论的是“比较好的那条世界线”。什么叫“比较好的世界”？就是我们拥有这种强大的、类人学习者的 AI，同时还……再顺着说一个我们没怎么展开、但在设计 superintelligent AI 规格时很值得考虑的点：你可以把它做窄一点，让它既超级智能，又在职能上保持窄域实用。你完全可以有一堆“窄领域的超级智能 AI”。

2、能清晰辨识“能力圈”的非常罕见

“相信自己就是对的”非常罕见，比如，按照巴菲特的要求，要把所有财务上的想法压缩到：（1）下行保护（7-8%，至少90%的信心）；（2）可能的上行收益（7%，至少50%的信心），过程中需要对很多的“模糊正确”的因素做出判断，具备这种能力的非常罕见。

可以换个角度思考为什么是罕见的？人是野生动物，大脑对风险极其敏感，敏感的另一面是脆弱，很容易损坏，要么“欠拟合”要么“过拟合”，能清晰辨识“能力圈”的非常罕见并且有很大的随机性，参考：《1975-10-14 Warren Buffett.PENSIONS》。

A further problem is that in no case were the superior records I have observed based upon institutional skills which could be maintained despite changes in the faces. Rather, the good results have been accomplished by a single individual or, at most, a few, working in fairly specialized areas in which the great bulk of investment money simply had no interest. It has been very difficult to out-think the pack on General Motors, IBM, Sears, etc. Rather, the unusual records - and there have been few that have been maintained - have been achieved by those who have worked relatively neglected fields in which competition was light.

更大的问题在于，我看到的优异记录无一建立在可替换人员仍能维系的制度化技巧之上。相反，它们往往出自一名、至多几名个人，在大多数资金毫无兴趣的专业化领域苦心经营。在通用汽车、IBM、希尔斯等热门股票上击败大多数人极其困难；反而，那些极少数得以延续的卓越业绩，来自于在竞争稀少、相对被忽视领域深耕的人。

参考：《2005-02-28 Warren Buffett.Insurance》。

So, you may ask, how do Berkshire’s insurance operations overcome the dismal economics of the industry and achieve some measure of enduring competitive advantage? We’ve attacked that problem in several ways. Let’s look first at NICO’s strategy.

那么你可能会问：Berkshire 的保险业务是如何克服行业糟糕的经济性，并获得某种可持续的竞争优势的？我们从几个方向去解决这个问题。先从 NICO 的策略讲起。

When we purchased the company — a specialist in commercial auto and general liability insurance — it did not appear to have any attributes that would overcome the industry’s chronic troubles. It was not well-known, had no informational advantage (the company has never had an actuary), was not a low-cost operator, and sold through general agents, a method many people thought outdated. Nevertheless, for almost all of the past 38 years, NICO has been a star performer. Indeed, had we not made this acquisition, Berkshire would be lucky to be worth half of what it is today.

当我们收购这家公司时——它专注于商业车险（commercial auto）和一般责任险（general liability insurance）——它看起来并不具备任何足以克服行业长期困境的优势。它不出名，没有信息优势（公司从来没有雇过精算师），也不是低成本运营者，而且通过一般代理人（general agents）渠道销售——许多人认为这种方式已经过时。尽管如此，在过去 38 年中的几乎所有年份里，NICO 都是明星级的表现者。事实上，如果当年我们没有完成这笔收购，Berkshire 今天的价值能有现在的一半就算走运了。

What we’ve had going for us is a managerial mindset that most insurers find impossible to replicate. Take a look at the facing page. Can you imagine any public company embracing a business model that would lead to the decline in revenue that we experienced from 1986 through 1999? That colossal slide, it should be emphasized, did not occur because business was unobtainable. Many billions of premium dollars were readily available to NICO had we only been willing to cut prices. But we instead consistently priced to make a profit, not to match our most optimistic competitor. We never left customers — but they left us.

我们真正的优势，是一种大多数保险公司根本无法复制的管理者心态（managerial mindset）。看看对页的数据：你能想象有任何一家上市公司，会主动拥抱一种商业模式——它会导致我们在 1986 到 1999 年间经历的那种营收下滑吗？必须强调的是，这种断崖式下滑并不是因为拿不到业务。只要我们愿意降价，市场上有“数十亿美元”的保费规模可以轻易被 NICO 拿到。但我们始终坚持：定价的目标是赚取利润，而不是去匹配我们最乐观的竞争对手的价格。我们从未离开客户——是客户离开了我们。

WARREN BUFFETT: No. But generally, we like people who are candid. We can usually tell when somebody’s dancing around something, or where their — when the reports are essentially a little dishonest, or biased, or something. And it’s just a lot easier to operate with people that are candid. And we like people who are smart, you know. I don’t mean geniuses. But that — and we like people who are focused on the business. It’s not real complicated, but we generally — you know, there may be a whole bunch of people in the middle that we don’t really have any feeling on one way or the other, and then we see some that we know we don’t want to be associated with, and some that we know we very much enjoy being associated with.

WARREN BUFFETT：没错。总体上，我们喜欢“坦诚的人”。有人在回避、报告里稍显“不诚实或带偏见”，我们通常看得出来。与坦诚的人合作更轻松。我们也喜欢“聪明的人”（不是天才），以及“专注于业务的人”。这不复杂：大量的人处在“中间地带”，我们无感；有些人，我们知道“不想与之共事”；另一些人，我们知道“非常享受与之共事”。

“integrity, intelligence and energy”，这样的排序很容易理解，但如果仔细想一想，“integrity, intelligence”形成是“energy”推动的结果，“energy”推动着走到“intelligence”，再走到“integrity”，越往后的节点需要越强大的“energy”。

Appendix.13.《1998-05-04 Berkshire Hathaway Annual Meeting》

56. Looking for owners who love their business more than money

寻找那些热爱业务胜过金钱的企业主

WARREN BUFFETT: I’ve been asked to take one more question from zone 10. I’m not sure why, but maybe because they see that I still have candy up here. (Laughter) So zone 10, please, and then we’ll break.

WARREN BUFFETT：有人让我再回答第 10 区的一个问题。我也不太清楚为什么，也许是因为他们看到我桌上还放着糖。（笑声）所以请第 10 区提问，然后我们就休息。

AUDIENCE MEMBER: Mr. Buffett and Mr. Munger, my name is Sanjiv Mirchandani, shareholder from Boston. First of all, thank you to both of you for everything. I have two questions. For you, Mr. Buffett, you obviously have filters that you apply on selecting people as you do on stocks. Can you tell us a little bit about what those filters are?

AUDIENCE MEMBER：Buffett 先生、Munger 先生，你们好。我叫 Sanjiv Mirchandani，是来自 Boston 的股东。首先感谢两位所做的一切。我有两个问题。第一个问题是问 Buffett 先生的。显然，你在“选人”时也有一套过滤标准，就像你选股票那样。能不能跟我们谈一谈，这些过滤标准大致是什么？

WARREN BUFFETT: Filters on people?

WARREN BUFFETT：选人的过滤标准？

AUDIENCE MEMBER: Yes, in selecting — you have an ability to motivate people who have a lot of money to keep working. What do you look for to figure out who those people are?

AUDIENCE MEMBER：对，就是在选人的时候——你似乎有一种能力，可以让已经很有钱的人依然愿意继续工作。你是通过什么来判断哪些人会是这样的人？

WARREN BUFFETT: Well, that’s a key, key question, because when we buy businesses we don’t have managers to put in them. I mean we are not buying them that way. We don’t have a lot of MBAs around the office that we’re —

WARREN BUFFETT：这是一个非常、非常关键的问题。因为当我们收购企业时，我们手头并没有一支可以“塞进去”的职业经理人队伍。换句话说，我们买企业时并不是打算自己去派人管理。我们的办公室里并没有一大堆 MBA 等着上岗——

CHARLIE MUNGER: Thank God.

CHARLIE MUNGER：谢天谢地。

WARREN BUFFETT: Yeah. (Laughter) And, you know, I have not promised that they’re going to have all kinds of opportunities or anything. So as a practical matter, we need management with the businesses that we buy. And three times out of four, thereabouts, the manager is the owner and is receiving tens of millions, maybe hundreds of millions of dollars. So they don’t have to work. And we have to decide in that time when we meet them whether they love the business or love money. And we’re not making a moral judgment. Charlie may, but I’m not making a moral judgment about whether it’s better to love the business or love money, but it’s very important for me to know which of the two is the primary motivator with them. And we have had extremely good luck in identifying people who love their business. And so all we have to do is avoid anything that, on our part, that diminishes that love of the business or makes other conditions so intolerable that they overcome that love of the business. And we have a number of people working for us, they have no financial need to work at all.

WARREN BUFFETT：是的。（笑声）而且你要知道，我不会向他们承诺什么“无限机会”之类的东西。所以在现实操作中，我们必须依赖原有的管理团队来管理我们收购的企业。大概有四分之三的情况，管理者本身就是企业的所有者，而他们会收到几千万，甚至上亿美元的对价——也就是说，他们完全不需要再为钱而工作。我们在和他们第一次见面时，就得判断清楚：他们爱的是这门生意，还是爱钱。这里我们并不是在做道德评判——Charlie 也许会，我本人不会就“爱事业好还是爱钱好”做道德判断——但对我而言，搞清楚这两者之中哪一个是他们的“第一驱动力”极其重要。我们在这方面的运气非常好，经常能找到那些真正热爱自己事业的人。之后我们要做的，就是避免自己做出任何会削弱他们对事业的热爱，或者让其他条件变得如此难以忍受，以至于压倒他们对事业热爱的事情。现在有不少为我们工作的人，从财务上讲，完全没有再工作的必要。

And they probably outwork, you know, 95 percent or more of the people in the world, and they do it because they just love smacking the ball. And we almost — we virtually had no mistakes in that respect. And we have identified a number of people, Charlie and I have, in terms of proposals to us, where we’ve felt that they did really — they liked the money better than the business. They were kind of tired of the business. You know? And they might promise us that they would continue on and they would do it in good faith, but something would happen six months later or a year later and they’d say to themselves, “Why am I doing this, you know, for Berkshire Hathaway when I could be doing,” whatever else they want to do? I can’t tell you exactly how we — what filter it is that we put them through mentally, but I can tell you that if you’ve been around a while, you can — I think you can have a pretty high batting average in coming to those conclusions. As you can about other aspects of human behavior.

而且他们的工作投入，可能比全世界 95% 以上的人都要多，而之所以这样干，仅仅是因为他们真心喜欢“挥棒打球”这件事。在这方面，我们几乎——可以说完全没有犯过错误。多年以来，在别人向我们提出的各种出售方案中，我和 Charlie 也识别出不少这样的人：我们能感觉到，他们其实是更喜欢钱而不是这门生意——对生意本身已经有点厌倦了。你知道吧？他们也许会向我们承诺继续干下去，而且是出于善意地这么承诺，但六个月、一年以后，总会发生点什么，让他们心里冒出一句话：“我为什么还要为 Berkshire Hathaway 干这些呢？明明可以去做”——然后就是他们真正想做的其他事情。我没法精确告诉你，我们在脑子里到底给他们套了什么样的过滤器，但我可以说，只要你在这个世上“混”的时间够久，你在这类判断上的“击球率”会相当高。就像你对人类行为的其他方面所做的判断一样。

I’m not saying you can take a hundred people and take a look at them and analyze their personalities or anything of the sort. But I think when you see the extreme cases, the ones that are going to cause you nothing but trouble, or the ones that are going to bring you nothing but joy, I think you can identify those pretty well. Charlie?

我不是说，你能把一百个人排成一排，看一眼就把每个人的性格剖析得清清楚楚。我的意思是，当你遇到那些“极端案例”时——要么将来只会给你惹麻烦的人，要么会给你带来无穷乐趣的人——我认为你是有能力把他们区分出来的。Charlie？

对自己做的事有热情是第一位（passion）的，我们的对热情的定义是“大脑对某类问题有自然的奖励”，如果没有又很努力，比较麻烦，这样的努力在心理机制上存在对齐（alignment）的问题。也有人问过巴菲特钱对他来说是不是已经不重要？他的回答是没办法把比赛和记分牌（钱）完全分开，参考：《2013-05-04 Berkshire Hathaway Annual Meeting》。

WARREN BUFFETT: Yeah. I think, actually, you have to love something to do well at it. There may be exceptions on that, but it is an enormous, enormous advantage if you absolutely love what you’re doing, every minute of it. And the nature of it is that that intensity adds to your productivity. And I have every bit of the intensity — not manifested exactly the same way — but it’s there every minute. I mean, I love thinking about Berkshire. I love thinking about its investments. I love thinking about its businesses. I love thinking about its managers. It’s part of me. And it is true, you can’t separate the game from the scorecard. I mean, you — so your score card is part of playing the game and loving the game. The proceeds are — you know, to me — are unimportant, but the proceeds are part of the score card, so they come with a score card.

WARREN BUFFETT：是的。我认为，想在某件事上做得好，你必须真心热爱它。或许会有例外，但如果你发自内心地热爱自己在做的事，而且是每一分钟都热爱，那是巨大的、巨大的优势。而这种热爱天然就会转化为“强度”，而强度会提升你的产出。我今天依然保持着当年的那种强度——虽然表现形式不完全一样——但那种强度每一分钟都在。我热爱思考 Berkshire。我热爱思考它的投资。我热爱思考它的业务。我热爱思考它的经理人。这些已经成了我自身的一部分。而确实，你没办法把“比赛”和“记分牌”完全分开。你的“记分牌”是你参加这场比赛、并且热爱比赛的一部分。收益——对我个人而言——并不重要，但收益是“记分牌”的一部分，所以自然会跟着记分牌一起出现。

巴菲特在这里所提醒的是那种把生意和钱完全分开的心理状态。

CHARLIE MUNGER: Well, yeah, I think it’s pretty simple. You’ve got integrity, intelligence, and experience, and dedication. And that’s what human enterprises need to run well, and we’ve been very lucky in getting this marvelous group of associates to work with all these years. It would be hard to do better, I think, than we’ve done on that respect. Look around this place. I mean, and really, you young people look around this place. And look at how much gratification can come into these lives which have been mostly spent in deferring gratification. It’s a very funny group of people, you shareholders. (Laughter)

CHARLIE MUNGER：嗯，是的，我觉得这事其实挺简单的。你要找的就是：integrity（正直）、intelligence（聪明）、experience（经验）和 dedication（投入/敬业）。这些就是一个人类组织要运转良好所需要的东西。多年来能和这样一群出色的合作者一起工作，我们实在是太幸运了。在这方面，要比我们现在的结果做得更好，我觉得是很难的。你看看这个会场——尤其是你们这些年轻人，好好看看四周。看看这些人的一生，是如何在长期“延迟满足”的前提下，最终获得如此巨大的满足感的。你们这些股东，是一群非常有趣的人。（笑声）

Appendix.14.《1998-10-15 Talk to MBA Students at Florida University》

I would like to talk for just one minute to the students about your future when you leave here. Because you will learn a tremendous amount about investments, you all have the ability to do well; you all have the IQ to do well. You all have the energy and initiative to do well or you wouldn't be here. Most of you will succeed in meeting your aspirations.

我想花一分钟时间，和在座的学生谈谈你们离开这里之后的未来。你们将会学到大量关于投资的知识，你们每个人都有能力取得好成绩；你们都有足够的智商去成功，你们也都有足够的活力和主动性，否则你们不会坐在这里。你们中的大多数人将能够实现自己的抱负。

But in determining whether you succeed there is more to it than intellect and energy. I would like to talk just a second about that. In fact, there was a guy, Pete Kiewit in Omaha, who used to say, he looked for three things in hiring people: integrity, intelligence and energy. And he said if the person did not have the first two, the later two would kill him, because if they don't have integrity, you want them dumb and lazy.

但决定你是否成功，不只是智力和活力那么简单。我想简单谈一下这一点。事实上，在奥马哈有个人叫皮特·基威特（Pete Kiewit），他常说，在招聘时他看重三件事：诚实、聪明和活力。他说，如果一个人没有前两者，后两者反而会害了他，因为如果一个人没有诚实，那你宁愿他又蠢又懒。

“Right People”的定义：integrity, intelligence and energy，后来在股东信里用了able 和 trustworthy，“intelligence and energy”= able，integrity= trustworthy，并且给出了具体的检验方法：《1994-03-01 Warren Buffett's Letters to Berkshire Shareholders》。

2.The certainty with which management can be evaluated, both as to its ability to realize the full potential of the business and to wisely employ its cash flows;

对管理层能够被评估的确定性——既包括其实现企业全部潜力的能力，也包括其明智运用现金流的能力；

3.The certainty with which management can be counted on to channel the rewards from the business to the shareholders rather than to itself;

对管理层是否可以被指望把企业回报导向股东、而非导向自身的确定性；

值得深入思考的是巴菲特要求对上述因素以“对决策有用的准确度”做出判断，基本上是定性，甚至接近于定量的判断。

1、管团队能够被评估的确定性？

检验管理层经济表现的首要标准，是其能否在所运用的股本上，在不过度依赖杠杆的情况下取得足够高的收益率。

（1）实现企业全部潜力的能力？

取得了什么样的进展？同时期的竞争对手又取得了什么？接手时是什么样的“牌面”？又是如何打好手中的“牌”的？
一个已经高成本运营的经理人，常常会异常机敏地找到增加管理费用的新方法；而一个运营紧凑的经理人，通常会继续找到更多削减成本的方法，即使他的成本已经远低于竞争对手。

（2）明智运用现金流的能力？

每留存的1块钱能不能创造多于1块钱的市场价值？
有没有把资本投入低利润项目的记录？会不会“忘记目标后加倍努力”？

2、是否可以被指望把企业回报导向股东、而非导向自身的确定性？

管理团队如何对待自己，以及如何对待股东？
企业文化能不能提供护栏？商业模式能不能提供护栏？

We want to talk about the first two because we know you have the last two. You are all second-year MBA students, so you have gotten to know your classmates. Think for a moment that I granted you the right--you can buy 10% of one of your classmate’s earnings for the rest of their lifetime. You can't pick someone with a rich father; you have to pick someone who is going to do it on his or her own merit. And I gave you an hour to think about it.

我们来谈谈前两项，因为我们知道你们都有后两项。你们都是MBA二年级的学生，已经对同班同学有了一定了解。现在假设我赋予你一个权利——你可以购买你某位同学未来一生收入的10%。你不能选一个靠着有钱老爸的，你只能选一个凭借自己能力打拼的人。我给你一个小时考虑这件事。

Will you give them an IQ test and pick the one with the highest IQ? I doubt it. Will you pick the one with the best grades? The most energetic? You will start looking for qualitative factors, in addition to (the quantitative) because everyone has enough brains and energy. You would probably pick the one you responded the best to, the one who has the leadership qualities, the one who is able to get other people to carry out their interests. That would be the person who is generous, honest and who gave credit to other people for their own ideas. All types of qualities. Whomever you admire the most in the class. Then I would throw in a hooker. In addition to this person you had to go short one of your classmates.

你会先给大家测个智商，然后选那个智商最高的人吗？我怀疑你会这么做。你会选成绩最好的吗？最有活力的吗？你可能会开始考虑一些定性因素，而不是单纯量化因素，因为大家都有足够的头脑和活力。你很可能会选择那个你最欣赏、最愿意共事的人，那个人有领导才能，能让他人愿意支持他、实现他的目标。那个人可能是慷慨的、诚实的、会把功劳归于他人的人。这些都是你会欣赏的品质。你会挑选班上你最敬佩的同学。然后我会给你加一个条件：除了选择这个人之外，你还必须“做空”班上的另一个人。

That is more fun. Who do I want to go short? You wouldn't pick the person with the lowest IQ, you would think about the person who turned you off, the person who is egotistical, who is greedy, who cuts corners, who is slightly dishonest.

这更有意思了。那么我想“做空”的人会是谁？你不会去选智商最低的那个，而是会考虑那个让你反感的人——那个自我中心、贪婪、做事投机取巧、稍有不诚实的人。

As you look at those qualities on the left and right hand side, there is one interesting thing about them, it is not the ability to throw a football 60 yards, it is not the ability the run the 100 yard dash in 9.3 seconds, it is not being the best looking person in the class, they are all qualities that if you really want to have the ones on the left hand side, you can have them.

当你比较“左边”和“右边”的这些特质时，有一件很有趣的事——它们不是像掷出60码橄榄球那样的能力，不是百米跑9.3秒的体能，也不是班上最帅/最漂亮的外貌。这些都是你如果真想拥有，就能拥有的品质。

They are qualities of behavior, temperament, character that are achievable, they are not forbidden to anybody in this group. And if you look at the qualities on the right hand side the ones that turn you off in other people, there is not a quality there that you have to have. You can get rid of it. You can get rid of it a lot easier at your age than at my age, because most behaviors are habitual. The chains of habit are too light to be felt until they are too heavy to be broken. There is no question about it. I see people with these self- destructive behavior patterns at my age or even twenty years younger and they really are entrapped by them.

这些是行为、气质、性格上的特质，是人人都能培养出来的东西。而如果你看那些让你反感的“右边”的特质，没有一个是你必须拥有的。你完全可以摆脱它们。而且你们这个年纪要比我现在这个年纪容易得多。因为大多数行为都是习惯性的——“习惯的锁链轻到你感觉不到，直到它重得无法挣脱。”这是真的。我看到很多人在我这个年纪，甚至比我年轻二十岁时，还被这些自毁性的行为模式困住。

They go around and do things that turn off other people right and left. They don't need to be that way but by a certain point they get so they can hardly change it. But at your age you can have any habits, any patterns of behavior that you wish. It is simply a question of which you decide.

他们到处让人讨厌，做出让人反感的行为。他们本不需要那样，但到了某个年纪就几乎改不了了。但你们这个年纪，可以培养任何你想要的习惯和行为模式。这只取决于你做出什么决定。

If you did this… Ben Graham looked around at the people he admired and Ben Franklin did this before him. Ben Graham did this in his low teens and he looked around at the people he admired and he said, "I want to be admired, so why don't I behave like them?" And he found out that there was nothing impossible about behaving like them. Similarly he did the same thing on the reverse side in terms of getting rid of those qualities. I would suggest is that if you write those qualities down and think about them a while and make them habitual, you will be the one you want to buy 10% of when you are all through.

如果你愿意这样做……本·格雷厄姆曾观察他所敬佩的人——在他之前，本杰明·富兰克林也做过同样的事。格雷厄姆在十几岁时观察那些他尊敬的人，然后对自己说：“我想成为一个值得别人尊敬的人，那为什么不去模仿他们的行为呢？”后来他发现，模仿他们并没有什么不可能的。同样地，他也努力去改掉那些他不喜欢的品质。我建议你把这些特质写下来，思考一段时间，把它们变成你的习惯。到头来，你会成为那个你愿意买下10%未来收益的那个人。

And the beauty of it is that you already own 100% of yourself and you are stuck with it. So you might as well be that person, that somebody else.

而最美妙的是——你已经100%拥有你自己了，而且你也无法换掉。所以你不如就成为那样的一个人，那个值得拥有的人。

Well that is a short little sermon. So let's get on with what you are interested in. Let's start with questions………..

好了，这是一段简短的小“布道”。现在我们回到你们真正感兴趣的话题——开始提问吧……

Appendix.15.《2001-04-28 Berkshire Hathaway Annual Meeting》

60. The value of Berkshire’s “loyalty effect”

伯克希尔“忠诚效应”的价值

WARREN BUFFETT: Area 6.

WARREN BUFFETT：第 6 区。

AUDIENCE MEMBER: Yeah, hi. I’m James Halperin from Dallas, Texas. And I’ve been a shareholder since 1995. And I feel great about it, so thank you. This question has to do with Berkshire’s so-called permanent holdings and whether, when making investment decisions, you somehow mathematically calculate a value to Berkshire’s reputation for loyalty to its public investees. Let’s say you are confident enough that Pepsi or Procter and Gamble would grow cash flow faster than Coke or Gillette would. And that the replacement value of the stock was less expensive enough to more than make up for the taxes. Would you then sell Coke, for example, to buy Pepsi? And if not, why not? And how do you value this reputation for loyalty aspect in those decisions?

AUDIENCE MEMBER：你好，我是来自德州达拉斯的 James Halperin。从 1995 年起我就是股东了，对此感觉非常好，所以先谢谢你们。我的问题跟 Berkshire 所谓的“永久持股”有关：在做投资决策的时候，你们会不会在某种程度上，用一个“数学上的方式”去给 Berkshire 对所投上市公司那种“忠诚声誉”赋值？比如说，假设你们足够有把握认为，Pepsi 或 Procter and Gamble 的现金流增长会快于 Coke 或 Gillette，而且这些新标的的“替代成本”（股价）足够便宜，足以弥补卖出原有持股所要缴的税。那在这种情况下，你们会不会卖掉比如说 Coke 去买 Pepsi？如果不会，原因是什么？在你们的决策里，怎么给这种“忠诚声誉”的因素估值？

WARREN BUFFETT: Well, I think that’s a very good question. I don’t think we would ever — I think it’s very unlikely we would come to the conclusion that we were that certain that — you mentioned P&G and Pepsi versus the ones — but that some major consumer products company would do better than the ones we’re in. We might very well decide that some other one is going to do quite well and buy that additionally. As a practical matter, if I’m on the board of a company, or Charlie were to be, representing Berkshire, it’s very difficult — I would say it’s almost impossible for us to trade in their securities. It just — it creates too many problems. People would think we knew something we didn’t. Or, you know, particularly if we were selling it, you know, we would have people questioning very much whether we had detected something within the company that was not available to the rest of the world. So we really give up an enormous amount of investment mobility when we go on a board.

WARREN BUFFETT：我觉得这是个非常好的问题。我不认为我们会——我觉得我们极不可能得出那种程度的结论：也就是像你刚才提到的，用 P&G 和 Pepsi 去对比我们已经持有的那些公司，然后说某家大型消费品公司一定会明显优于我们现在持有的这些。我们完全可能会认为，另外某家公司未来会表现得很好，然后在现有持股之外再去买一些。但在实际操作层面，如果我在某家公司董事会里，或者 Charlie 在董事会里代表 Berkshire，那我们要在它的股票上交易就会变得非常困难，我甚至会说几乎是不可能。这会带来太多问题：别人会认为我们掌握了一些他们不知道的内部信息。尤其是如果我们在卖出的时候，人们就会非常质疑，我们是不是在公司内部发现了什么“外界还不知道的东西”。所以，一旦我们进入一家公司的董事会，我们实际上放弃了非常大一块投资上的机动性。

And so I don’t even think about doing what you’re suggesting, although I might very well if I were just a money manager running the business. We certainly, and we’ve laid it out in the ground rules in the back of the — in our Owner’s Manual back in the annual report — we’ve certainly said, in terms of businesses we buy control of, that they just aren’t for sale. And a fancy price will not tempt us. And that we lay out that exception relating to businesses where we think there’s a permanent loss of cash for as far as the eye can see, or businesses where we have labor troubles, which we — I described earlier in the day, we might’ve had at The Buffalo News at that one period. But otherwise, simply because we can use the money better someplace else, we’re not interested in it.

所以，我压根儿不会去考虑你刚才说的那种做法，尽管如果我只是一个“代客理财”的职业资金经理、只是在经营一只基金，我很可能会那么干。我们当然——而且我们已经在年报后面的《Owner’s Manual》里的“基本规则”部分写得很清楚——我们明确说过，对那些我们买到控股权的企业来说，它们就是“不挂牌出售”的，即使对方给出一个花哨的高价也不会打动我们。我们只保留极少数例外：比如，我们认为这家企业在可预见的未来会“永久性地亏现金”，或者像我今天上午提到的那种，在劳动关系上遇到严重问题的企业——像当年 The Buffalo News 在某个时期那样。除了这种情况，仅仅因为我们把钱挪到别处可能赚得更多，我们对“卖掉这家企业”这件事就不感兴趣。

You know, I can’t really dig into my psyche and tell you how much of that is because I think that will help us buy businesses in the future if we behave that way, or how much is just my natural inclination that when I make a deal with somebody and I’m happy with how they behave with me, that I want to stick with them. It’s probably both, you know? And I wouldn’t want to try and weight the two. I’m happy, you know, with the results of the first and I’m happy with the way I feel, essentially, about the second. I just think it’s crazy — I know if I owned all of Berkshire myself, I wouldn’t dream of trading around businesses with people that have trusted in me and that I like and have been more than fair with me. I wouldn’t dream of trading around businesses so that my estate was 105 percent of some very large number instead of 100 percent of some large number. I just would regard that as a crazy way to live.

说实话，我也很难把自己的内心剖得特别细，告诉你这里面有多少成分是“我相信这种行为方式会帮助我们将来收购更多企业”，又有多少成分只是我个人的天性：一旦我和某个人达成了交易，而且我对他对待我的方式很满意，那我就倾向于一直和他合作。很可能两方面都有，对吧？而且我也不想去硬给这两者分别称个重量。对于第一点，我对它带来的结果很满意；对第二点，我也很喜欢它给我带来的那种“做人方式上的感受”。我只是觉得，把生活过成另外那样是很疯狂的——我很清楚，如果我个人拥有 Berkshire 的全部股权，我绝不会去“倒手买卖那些企业”，明明那些人信任我、而且我也喜欢他们，并且他们对我一向非常公平。我绝不会为了把自己的遗产从“某个巨大数字的 100%”变成“某个巨大数字的 105%”，去折腾买卖企业。我会觉得那是一种非常荒唐的生活方式。

And I don’t want the fact I run a public company to cause me to behave in a way that I would be uncomfortable behaving if we were a private company. But I also feel that you, as shareholders, are entitled to know that that’s an idiosyncrasy of mine. And therefore, I lay it out, and have laid it out for 20 years, as something that you should understand, as an investor or before you become an investor. I’m sure it helps us in acquisitions over time. But whether that in any way compensations the opportunity costs that Charlie talks about of making an occasional advantageous disposal, I don’t know and it’s something I’ll never calculate. Charlie?

而且，我不希望“我经营的是一家上市公司”这个事实，逼着我去采取一种如果我们是私人公司我自己都觉得别扭的行为方式。但与此同时，我也认为，作为股东，你们有权知道这是我个人的一个“特立独行之处”。因此，我把这点写出来——而且已经写了二十年了——当成一件你们作为投资者，或在成为投资者之前就应该明白的事情。我确信，从长期看，这种做法在收购业务上是帮到我们的。但这种“忠诚效应”在多大程度上弥补了 Charlie 所说的那种“偶尔做一笔很划算的出售交易却放弃了”的机会成本，这一点我并不知道，也是一件我永远不会去计算的事。Charlie？

在紧密的合作关系中，让对方紧张也会让自己紧张，高度紧张和有效约束之间充满了不确定，参考：《Federalist No. 10》。

If the impulse and the opportunity be suffered to coincide, we well know that neither moral nor religious motives can be relied on as an adequate control.

只要冲动与机会得以重合，我们深知道德或宗教之动机皆不足以构成有效约束。

CHARLIE MUNGER: Well, I do tend to calculate it, at least roughly. And so far, I think that the loyalty effect is a plus in our life.

CHARLIE MUNGER：嗯，我倒是倾向于去算一算，至少是大致算一算。到目前为止，我认为这种“忠诚效应”对我们的人生总体来说是正面的加分项。

WARREN BUFFETT: Would you regard that as true though in both public — I mean, both marketable securities and owned businesses?

WARREN BUFFETT：不过你会不会认为，这一点在两种情形都成立——我是说，既包括上市流通证券，也包括我们全资拥有的企业？

CHARLIE MUNGER: Oh, no. I don’t think the loyalty effect in lots of public companies is nearly as important as it is with the private companies.

CHARLIE MUNGER：哦，不。我并不认为，在很多上市公司那一类持股上，这种“忠诚效应”有像在私人控股企业上那么重要。

WARREN BUFFETT: You can say it’s a mistake for us to be directors of companies, because we give up huge amount of flexibility in investment because we are directors. I — and there’s no question that we do. It’s — if you’re thinking solely of making money, you do not want to be a director of any company. So there’s just no question about that.

WARREN BUFFETT：你甚至可以说，我们去当一些公司的董事，从“纯投资”的角度看，是一个错误。因为一旦我们成为董事，就会放弃掉在投资上非常大的一部分灵活性。而我——我认为这一点是毫无疑问的。如果你的唯一目标就是“赚钱”，那你根本就不应该去当任何一家公司的董事，这一点是毫无疑问的。

Appendix.16.《2002-05-04 Berkshire Hathaway Annual Meeting》

22. Life advice: take care of your most important asset

人生忠告：照顾好你最重要的资产

WARREN BUFFETT: Number 10.

WARREN BUFFETT：10 号麦克风。

AUDIENCE MEMBER: I’m Lowell Chrisman (PH), from Phoenix, Arizona. I am retired and teaching in — seniors in high school. I wish they could be here to hear you today. I’m teaching these people an investment course part-time. And the first class I went to, they asked me to teach them how to prepare for retirement. I would like to know what the two or three things are that you would suggest that I include in this course.

AUDIENCE MEMBER：我叫 Lowell Chrisman（音），来自亚利桑那州的 Phoenix。我已经退休了，现在在教——高中毕业班的学生。我真希望他们今天也能在这里听你讲话。我在给这些学生兼职上投资课，第一堂课他们就问我，要教他们怎么为退休做准备。我想请教你：你认为在这门课里，最应该包含的两三点是什么？

WARREN BUFFETT: Well —

WARREN BUFFETT：嗯——

CHARLIE MUNGER: What the hell does Warren know about retirement? (Laughter)

CHARLIE MUNGER：Warren 懂个啥退休啊？（笑声）

WARREN BUFFETT: Yeah. We haven’t even thought about it. Now, let me give you one suggestion for that group. I use this sometimes when I talk to high school — a bunch of high school seniors down in Nebraska Wesleyan, a few weeks ago. Tell the youngsters in the class, they’re probably around 16 or 17, and if they’re like I was when I was 16, you know, I was only thinking of two things. And Martin’s Aunt Barbara wasn’t going out with me, so I was down to cars. (Laughter) I tried hearses, but that didn’t work. The — And let’s assume, and I use this with — let’s assume a genie appeared to you when you turned 16, and the genie said, “You get any car you want tomorrow morning, tied up in a big pink ribbon, anything you name. And it can be a Rolls Royce, it can be a Jaguar, it can be a Lexus, you name it, and that car will be there and you don’t owe me a penny.”

WARREN BUFFETT：是啊，我们压根儿都没想过退休这事儿。现在，我给你们那帮学生提一个建议。我有时候给高中生讲话会用这个例子——几周前我还在 Nebraska Wesleyan 给一群高三学生讲过。你可以跟班里的那些年轻人说，他们大概 16、17 岁，如果他们像我 16 岁时那样，你知道的，我脑子里只想着两件事。而 Martin 的 Aunt Barbara 又不肯跟我约会，所以我就只剩下“车”这一个念头了。（笑声）我还试过殡仪车，但那没用。（笑声）然后我们假设——我就用这个设想——假设你 16 岁生日那天，有个 genie（精灵）出现，对你说：“明天早上你可以得到任何一辆你想要的车，上面系着一个粉红色的大蝴蝶结，随你点。可以是 Rolls Royce，可以是 Jaguar，可以是 Lexus，只要你叫得出名，明天那辆车就会停在那里，而且你一分钱都不用付。”

And having heard the genie stories before, you say to the genie, “What’s the catch?” And of course, the genie says, “Well, there’s just one. That car, which you’re going to get tomorrow morning, the car of your dreams, is the only car you’re ever going to get. So you can pick one, but that’s it.” And you still name whatever the car of your dreams is, and the next morning you receive that car. Now, what do you do, knowing that’s the only car you’re going to have for the rest of your life? Well, you read the owner’s manual about 10 times before you put the key in the ignition, and you keep it garaged. You know, you change the oil twice as often as they tell you to do. You keep the tires inflated properly. If you get a little nick, you fix it that day so it doesn’t rust on you.

你以前听过这种精灵的故事，所以你会问精灵：“这里面有什么猫腻？”当然，精灵会说：“就一条。你明天早上拿到的那辆车——你梦寐以求的那辆——将会是你一生中唯一的一辆车。你可以任意选一辆，但就这一辆。”你还是会说出你梦寐以求的那辆车的名字，第二天早上那辆车就真的到了你手里。那接下来，你会怎么做？在明知道这是你余生唯一的一辆车的前提下。你会在第一次点火之前，把用户手册读上十遍；你会把车老老实实停在车库里；机油会比说明书要求的更频繁地更换；轮胎永远保持在合适的气压；车身上哪怕有一点小划痕，你也会当天就去修掉，以免生锈。

In other words, you make sure that this car of your dreams at age 16 is going to still be the car of your dreams at age 50 or 60, because you treat it as the only one you’ll ever get in your lifetime. And then I would suggest to your students in Phoenix that they are going to get exactly one mind and one body, and that’s the mind and body they’re going to have at age 40 and 50 and 60. And it isn’t so much a question of preparing for retirement, precisely, at those ages, it’s a question of preparing for life at those ages. And that they should treat the importance of taking care and maximizing that mind, and taking care of that body in a way, that when they get to be 50 or 60 or 70, they’ve got a real asset instead of something that’s rusted and been ignored over the years. And it will be too late to think about that when they’re 60 or 70. You can’t repair the car back into the shape it was. You can maintain it. And in the case of a mind, you can enhance it in a very big way over time.

换句话说，你会尽一切努力，确保这辆你 16 岁时梦寐以求的车，在你 50 岁、60 岁时仍然配得上“梦寐以求”四个字，因为你把它当成你这一辈子唯一的一辆车。然后，我会建议你在 Phoenix 的学生这样想：他们这辈子也只会得到“唯一一套”心智和“唯一一副”身体，而这就是他们在 40 岁、50 岁、60 岁时所拥有的那套心智和那副身体。到了那些年龄，最重要的问题与其说是“为退休做准备”，不如说是“为那个年龄的人生做准备”。他们应该意识到：把这副头脑照顾好并尽量发挥到极致、把这副身体照顾好，其重要性在于——等他们到了 50、60、70 岁的时候，手上拥有的是一项真正的资产，而不是一件多年来被忽视、已经生锈报废的东西。要到 60、70 岁才开始想这件事就太晚了。你没办法把那辆车修回到当初的状态，你能做的只是好好维护。而对于大脑来说，你不仅能维护它，随着时间推移你甚至还能在很大程度上增强它。

But the most important asset your students have is themselves. You know, I will take a person graduating from college, and assuming they’re in normal shape and everything, I will be glad to pay them, you know, probably $50,000 for 10 percent of all their earnings for the rest of their lives. Well, I’m willing to pay them 10 percent for — $50,000 for 10 percent — that means they’re worth $500,000 if they haven’t got a dime in their pocket, as long as they’ve got a good mind and a good body. Now that asset is far, far more important than any other asset they’ve got, unless they’ve been very lucky in terms of inheritance or something, but overwhelmingly their main asset is themselves. And they ought to treat their main asset as they would any other asset that was divorced from themselves. And if they do that, and they start thinking about it now, and they develop the habits that maintain and enhance the asset, you know, they will have a very good car, mind, and body when they get to be 60. And if they don’t, they’ll have a wreck. Charlie?

但是，你那些学生所拥有的最重要的资产，是他们自己。你看，一个刚刚大学毕业的人，只要他在各方面都算正常，我是愿意出价的——我很乐意付，比如说 50,000 美元，来买他此后终身收入的 10%。如果我愿意出 50,000 美元买他 10% 的终身收入权，那就意味着，即便他现在身无分文，只要有一副正常的大脑和身体，他就值 500,000 美元。相较之下，这项资产远远比他们目前拥有的任何其他资产都重要——除非有人极其走运，继承了一大笔财产之类，但对绝大多数人来说，他们的主要资产就是他们自己。而他们应该像对待一项“和自己切割开来的资产”那样，来对待这项主资产。如果他们现在就开始这样思考，慢慢养成维护和提升这项资产的习惯，那么等他们到了 60 岁，就会拥有一辆非常好的“车”——一副良好的头脑和身体；如果他们不这么做，最后得到的就会是一副“报废车”。Charlie？

人性假设的新版本：用一小段文字把问题说清楚，语言对注意力的影响至关重要，参考：《2025-11-26 Ilya Sutskever.We're moving from the age of scaling to the age of research》。

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。

(Applause)

（掌声）

Appendix.17.《2005-02-28 Warren Buffett's Letters to Berkshire Shareholders》

It takes real fortitude — embedded deep within a company’s culture — to operate as NICO does. Anyone examining the table can scan the years from 1986 to 1999 quickly. But living day after day with dwindling volume — while competitors are boasting of growth and reaping Wall Street’s applause — is an experience few managers can tolerate. NICO, however, has had four CEOs since its formation in 1940 and none have bent. (It should be noted that only one of the four graduated from college. Our experience tells us that extraordinary business ability is largely innate.)

要像 NICO 这样运作，需要真正的坚韧（fortitude），而且这种坚韧必须深深嵌入公司的文化之中。任何看这张表的人，都可以很快扫过 1986 到 1999 年。但在现实中，日复一日地承受业务量不断萎缩——与此同时竞争对手吹嘘增长、收获华尔街掌声——这种体验几乎没几个管理者能忍受。然而，NICO 自 1940 年成立以来换过四位 CEO，没有一位弯过腰。（值得一提的是，这四位里只有一位大学毕业。我们的经验告诉我们：非凡的商业能力在很大程度上是天生的。）

相信自己就是对的，非常非常的难，DeepSeek R1 的论文也讲到这个问题，知识的轨迹空间太大了，参考：《2025-11-26 Ilya Sutskever.We're moving from the age of scaling to the age of research》。

This was in the DeepSeek R1 paper— that the space of trajectories is so wide that maybe it’s hard to learn a mapping from an intermediate trajectory and value. And also given that, in coding for example you’ll have the wrong idea, then you’ll go back, then you’ll change something.

DeepSeek R1 的论文里提到过这样一个观点——轨迹空间太大了，所以也许很难从某个中间轨迹学到一条到 value 的映射。再加上，在写代码这类场景中，你一开始可能思路就不对，然后你会回退、再修改一些东西，过程本身就曲折反复。

我自己的答案是Think Independently+Think Correctly，缺一不可，Think Independently即Basic Trust 的心理结构；Think Correctly是一套足够清晰的智能框架，两者结合，将大量信息压缩到能够做出判断的准确度。

Appendix.18.《2005-04-30 Berkshire Hathaway Annual Meeting》

8. What Buffett looks for in managers

巴菲特在经理人身上看重什么

WARREN BUFFETT: We’ll go to microphone number 1, please.

WARREN BUFFETT：我们先请 1 号麦克风提问。

AUDIENCE MEMBER: Simon Denison-Smith from London. You talk a lot about the importance of selecting terrific managers. I’d just like to understand what your three most important criteria are for selecting them, and how quickly you can assess that.

AUDIENCE MEMBER：我是来自 London 的 Simon Denison-Smith。您经常谈到挑选出色经理人的重要性。我想知道，您在挑选他们时，最重要的三条标准分别是什么？以及，您通常多快能判断出来？

WARREN BUFFETT: Yeah. I’ll give you two different answers. The most important factor, subject to this one caveat I’m going to give in a second, is a passion for their business. We are frequently buying businesses from people who we wish to have continue manage than them and where we are, in effect, monetizing a lifetime of work for them. I mean, they’ve built this business over the years. They’re already rich but they may not be rich in the liquid sense. They may have all their money, or a good bit of it, tied up in the business, so they’re monetizing it. They may be doing it for estate reasons or tax reasons or family reasons, whatever. But we really want to buy from somebody who doesn’t want to sell. And they certainly don’t want to leave the business. So we are looking for people that have a passion that extends beyond their paycheck every week or every month for their business. Because if we hand somebody a hundred million or a billion dollars for their business, they have no financial need to work. They have to want to work. And we can’t stand there with a whip.

WARREN BUFFETT：好的，这个问题我会给出两种回答。最重要的因素——先说明一点，我马上会说一个例外——是他们对自己业务的热情。我们经常是从那些我们希望“他们继续经营公司”的人手中买下公司，而我们在实际操作中，是在帮助他们把一生的心血“货币化”。我的意思是，他们在很多年里一点点把业务做起来。他们已经很富有了，但可能在“流动性”的意义上还不算富有——他们的全部财富，或者很大一部分财富，都还绑在这家公司上，所以这笔交易是在把它们变现。他们可能是出于遗产安排、税务考虑或者家庭因素等等来做这件事。但我们真正希望买的是那种“其实并不想卖”的人的公司，而且他们肯定也不想离开这家企业。所以我们寻找的是这样一群人：他们对业务的热情远远超过每周、每个月那张工资单本身。因为一旦我们把 1 亿、甚至 10 亿美元交到他们手里买下公司，他们在经济上就再也没有“必须工作的需要”了；接下来，他们得是“想工作”才会继续干，而我们又不可能拿着皮鞭站在旁边盯着他们。

We don’t have any contracts at Berkshire. They don’t work, as far as I’m concerned. So we hope that they love their business and then we do everything possible to avoid extinguishing, or in any way dampening, that love. I tell students that what we’re looking for when we hire somebody, beyond this passion, we’re looking for intelligence. We’re looking for energy. And we’re looking for integrity. And we tell them, if they don’t have the last, the first two will kill you. Because if you hire somebody without integrity, you really want them to be dumb and lazy, don’t you? I mean, the last thing in the world you want is that they are smart energetic. So we look for those qualities. But, we have generally — when we buy businesses, it’s clear to us that those businesses are coming with managers with those qualities in them. Then we need to look into their eyes and say, do they love the money or do they love the business? And if they love — there’s nothing wrong with liking the money.

我们在 Berkshire 是没有任何“合同”来绑住经理人的。在我看来，那些东西根本不起作用。所以我们希望他们真心热爱自己的公司，然后我们会尽一切可能避免熄灭、或以任何方式削弱这种热爱。我跟学生说，当我们聘用某个人时，除了这种“热情”之外，我们还在寻找三样东西：intelligence（聪明）、energy（干劲）和 integrity（正直）。我们会告诉学生：如果一个人缺少最后一项（integrity），前两项会“要你的命”。因为如果你雇了一个不正直的人，那你真是巴不得他又笨又懒，不是吗？你最不想要的，就是那种既聪明又特别有干劲、但却不正直的人。所以我们会去寻找这些特质。不过，一般来说，当我们收购企业时，我们已经很清楚，这些企业本身就自带拥有这些品质的经理人。然后，我们需要做的是看着他们的眼睛，问自己：他们是真的热爱这家企业，还是只是热爱那笔钱？当然，喜欢钱本身没有任何问题。

为什么“passion”是最重要的因素？

1、Basic Trust/Basic Mistrust，或者内部计分卡/外部计分卡会左右注意力的方向，但在选定的方向上，“passion”（算力）对数据压缩的结果起到最关键的作用，不然很难解释很多才华出众的Basic Mistrust；

2、更强的“passion”产出更好的知识结构，相当于汽车的引擎马力，潜在的马力和人生的实际产出又是两回事，扭曲的注意力结构在很大程度上影响马力的输出，但是一辆好车，我们都喜欢先看有没有一颗好的引擎？

3、“integrity, intelligence and energy”，这是摘果子的顺序，但如果仔细想一想，“integrity, intelligence”形成是“energy”推动的结果，“energy”推动着走到“intelligence”，再走到“integrity”，越往后的节点需要越强大的“energy”。

But if — what it’s really all about is that they built this business so they can sell out and cash their chips and go someplace else, we have a problem, because right now we only have 16 people at headquarters and we don’t have anybody to go out and run those businesses. So we — it’s necessary that they have this passion and then it’s necessary that we do nothing to, in effect, dampen that passion. Charlie?

但如果事实是：他们之所以辛苦把生意做大，只是为了有一天“套现离场”、把筹码换成现金，然后去做别的事情，那我们就会遇到麻烦。因为现在我们的总部只有 16 个人，我们根本没有人可以派出去接手这些公司。所以，我们必须要他们真的对业务有激情，同时我们也必须做到“不去做任何会削弱这种激情的事”。Charlie？

CHARLIE MUNGER: Yeah. The interesting thing is how well it’s worked over a great many decades and how few people copy it. (Laughter and applause)

CHARLIE MUNGER：是的。有趣的是，这套做法在很多很多年里效果都非常好，但真正学着去模仿的人却少之又少。（笑声与掌声）

Appendix.19.《2007-02-28 Warren Buffett's Letters to Berkshire Shareholders》

Picking the right person(s) will not be an easy task. It’s not hard, of course, to find smart people, among them individuals who have impressive investment records. But there is far more to successful long-term investing than brains and performance that has recently been good.

挑对人并不容易。当然，找聪明人不难，其中也不乏近期业绩亮眼、投资记录出色的人。但长期投资的成功，远不止智力与短期漂亮的成绩单。

Over time, markets will do extraordinary, even bizarre, things. A single, big mistake could wipe out a long string of successes. We therefore need someone genetically programmed to recognize and avoid serious risks, including those never before encountered. Certain perils that lurk in investment strategies cannot be spotted by use of the models commonly employed today by financial institutions.

随着时间推移，市场会做出非同寻常、甚至怪诞的事情。一次重大的错误，就可能抹掉一长串成功。我们因此需要一种“基因层面就被编程为能识别并躲开严重风险”的人——包括那些以前从未遇到过的风险。投资策略里潜伏的一些危险，用今天金融机构常用的模型是看不出来的。

Temperament is also important. Independent thinking, emotional stability, and a keen understanding of both human and institutional behavior is vital to long-term investment success. I’ve seen a lot of very smart people who have lacked these virtues.

性格气质也同样重要。独立思考、情绪稳定，以及对人性与制度行为的敏锐理解，对长期投资成功至关重要。我见过不少非常聪明的人，却缺少这些品质。

洞察力是最重要的素质，能够预见那些压倒性的重大问题，一种高度泛化的能力，参考：《2026-05-02 Warren Buffett.Interview with Becky Quick》。

我们大多数经理人在处理较小的问题上都很出色，但他们无法预见那些压倒性的重大问题。那是我的工作，或者现在是 Greg 的工作。

most of our managers are very good at the smaller problems, like they can’t anticipate the overwhelming problems. That’s my job, or now Greg’s job.

*****

In my will I’ve stipulated that the proceeds from all Berkshire shares I still own at death are to be used for philanthropic purposes within ten years after my estate is closed. Because my affairs are not complicated, it should take three years at most for this closing to occur. Adding this 13-year period to my expected lifespan of about 12 years (though, naturally, I’m aiming for more) means that proceeds from all of my Berkshire shares will likely be distributed for societal purposes over the next 25 years or so.

我在遗嘱中规定：我去世时仍持有的全部伯克希尔股票，其变现所得必须在遗产结算完成后的十年内全部用于慈善。因为我的事务并不复杂，遗产结算最多三年就能完成。把这13年（结算最多三年+十年内花完）加上我预计大约12年的寿命（当然，我希望更久一些），意味着：我全部伯克希尔股票的收益，很可能会在未来大约25年内陆续被用于社会公益。

I’ve set this schedule because I want the money to be spent relatively promptly by people I know to be capable, vigorous and motivated. These managerial attributes sometimes wane as institutions — particularly those that are exempt from market forces — age. Today, there are terrific people in charge at the five foundations. So at my death, why should they not move with dispatch to judiciously spend the money that remains?

我之所以设定这样的时间表，是因为我希望这笔钱能被我所认识、并确信其有能力、有活力且有动力的人相对迅速地花出去。随着机构——尤其是那些不受市场力量约束的机构——逐渐老化，这些管理特质有时会衰退。今天，这五家基金会的负责人都非常出色。那么等我去世时，他们为什么不应该迅速行动，把剩余资金审慎而有效地花掉呢？

Those people favoring perpetual foundations argue that in the future there will most certainly be large and important societal problems that philanthropy will need to address. I agree. But there will then also be many super-rich individuals and families whose wealth will exceed that of today’s Americans and to whom philanthropic organizations can make their case for funding. These funders can then judge firsthand which operations have both the vitality and the focus to best address the major societal problems that then exist. In this way, a market test of ideas and effectiveness can be applied. Some organizations will deserve major support while others will have outlived their usefulness. Even if the people above ground make their decisions imperfectly, they should be able to allocate funds more rationally than a decedent six feet under will have ordained decades earlier. Wills, of course, can always be rewritten, but it’s very unlikely that my thinking will change in a material way.

支持“永久基金会”的人会说：未来必然会出现规模巨大、极其重要的社会问题，需要慈善去应对。我同意。但到那时，也会出现许多超级富有的个人与家族，他们的财富会超过当今美国人的水平；慈善机构可以向他们争取资金。这些出资人可以亲自判断：哪些机构既有活力又有聚焦，最能解决当时存在的重大社会问题。这样一来，理念与效果就能接受一种“市场测试”。一些组织值得获得大量支持，而另一些组织可能已经过了其有效期。即使仍在世的人做决定时并不完美，他们把资金分配得也应当比一个早在地下六英尺处、几十年前就拍板的人更理性。遗嘱当然随时可以重写，但我也很难想象自己的想法会在实质上发生重大改变。

巴菲特在提醒保持思考，经常接受现实反馈的重要性，可以进一步思考巴菲特说的持久的竞争优势指的是什么？

1、简单、长期不变是不是最好的模式，但即使是See's，有很长时间销量上不去，门店数量也没办法扩张，从巴菲特股东信的描述中明显看到经营上的压力；

2、一个极端的想法是靠一个品牌，躺着就能赢，很怀疑是不是真的存在这样的逻辑，这个问题拉到极限，巴菲特要求自己的慈善基金在他过世后12年必须把钱都花了，慈善事业的问题是缺少反馈信号，See's、茅台可能好一些，但如果完全依赖品牌，想着“躺着就能赢”不是好事；

3、我自己的理解，就像一个有安全感的孩子，先有一个充满安全感的地基，然后因为有地基的保护让这个孩子敢于尝试，比如，腾讯的微信，BRK的长期资本，也包括巴菲特对所有投资项目的思维框架：足够可信的下行保护+可能的上行收益。

Appendix.20.《2007-05-05 Berkshire Hathaway Annual Meeting》

Beware when someone says “it’s so easy”

当有人说“这很简单”时要小心。

WARREN BUFFETT: Number 7, please.

WARREN BUFFETT：第七位，请。

AUDIENCE MEMBER: Mr. Buffett and Mr. Munger, thanks for hosting this wonderful meeting. My name is Chander Chavla (PH), and I’m visiting from Seattle, Washington. And I think Berkshire Hathaway can contribute to the reduction of global warming if, for next year shareholder’s meeting, Mr. Gates and I fly on the same plane. (Laughter) My question is, I have made a few mistakes in business by trusting the wrong people. So in — and I don’t know where to learn how to trust the right people. They don’t teach you that in business school, and the people who are supposed to teach you in the corporate world sometimes betray you. So how can I learn who to trust and who not to trust?

AUDIENCE MEMBER：Mr. Buffett 和 Mr. Munger，感谢你们举办这样一场精彩的大会。我叫 Chander Chavla（PH），从 Seattle, Washington 过来参加。我觉得如果明年股东大会时，Mr. Gates 和我坐同一架飞机来的话，那 Berkshire Hathaway 在减少全球变暖方面就能做出一点贡献了。（笑声）我的问题是：我在商业上犯过几次错误，原因是信错了人。而我不知道应该到哪里去学习如何信对人。商学院里不教你这些，职场上那些“本来应该教你的人”，有时候也会背叛你。那么，我该如何学会判断谁值得信任、谁不值得信任呢？

WARREN BUFFETT: Boy, that’s a great question. But you probably have about as good a chance of getting a good answer from me on that as you have on getting on Bill Gates’ plane next year. (Laughter) I get letters all the time, and I hear from people who have been taken advantage of in financial transactions. And, you know, it really is — it’s sad. And a lot of it isn’t even — it’s not fraud or anything. For one thing, I mean, just the charges involved, the frictional costs and the baloney that is presented is tough. Charlie and I have had very good luck in terms of buying businesses and putting our trust in people. It’s been just overwhelmingly good. But we filter out a lot of people. And then you say, “Well, how do you filter them out?” I would say — I think Charlie will agree with this — people give themselves away fairly often. And maybe it does help to have been around as long as we have in seeing the various ways they give themselves away.

WARREN BUFFETT：哇，这是个很好的问题。不过，你从我这里拿到一个好答案的概率，大概跟你明年能不能坐上 Bill Gates 的那架飞机差不多。（笑声）我经常收到来信，也常听说有人在金融交易中被占了便宜。你知道，这真的——让人难过。而且很多情况甚至都算不上欺诈什么的。光是各种费用、各种摩擦成本，还有那些被包装出来的鬼话，就已经很难对付了。就我们自己买企业、把信任交给别人的经历而言，我和 Charlie 一直都算非常幸运，整体结果非常好。但前提是，我们过滤掉了很多人。接下来你会问：“那你们是怎么把这些人筛掉的？”我会说——我想 Charlie 会同意——人们其实很经常会“暴露自己”。而我们活了这么久，看过那么多种类的“自我暴露”方式，这一点也许确实帮了我们不少。

They — when somebody comes to me with a business — and I probably shouldn’t tell this publicly because they’ll probably tailor their approach subsequently — but when they come, just the very things they talk about, what they regard as important and not important, there are a lot of clues that come as to subsequent behavior. And, like I say, we’ve really had a batting average I wouldn’t have thought we would have had in the people that we’ve joined with. But it hasn’t been a hundred percent. It’s been well above 90. And I get asked that, you know, I mean, “How do you make those judgments?” And I don’t know. Charlie, can you articulate the way we do it?

他们——当有人带着一门生意来找我时——其实我大概不应该在公开场合把这件事说得太明白，因为他们以后可能会针对性地“改剧本”——但事实是，当他们来找我们的时候，他们聊的那些点、他们认为哪些是重要的、哪些是不重要的，这些东西本身就包含了很多关于他们将来会怎么做事的线索。正如我刚才说的，在我们选择的合伙人身上，我们的“击球率”确实好得超出我当初的预期。但也不是 100%，只是在 90% 以上。我也经常被问到同样的问题：“你们是怎么做出这些判断的？”而我也说不上一个清清楚楚的公式。Charlie，你能不能把我们的做法说得更清楚一些？

CHARLIE MUNGER: Well, partly we’re deeply suspicious when the proposition is too good to be true. Warren once introduced me to a gentleman promoter who wanted to inveigle us into an insurance program. And he said, “We only write fire insurance on concrete bridges that are covered by water.” He says, “It’s like taking candy from babies.” We are able to filter out propositions like that.

CHARLIE MUNGER：嗯，部分原因是，当一个机会“好得不像真的”时，我们天生就会高度怀疑。Warren 曾经介绍我认识一位“绅士型推销员”，他想把我们哄进一个保险项目里。他说：“我们只承保那种被水覆盖着的混凝土桥的火灾险。”他还说：“这简直就像从婴儿手里拿糖一样容易。”像这样的项目，我们是能第一时间筛掉的。

WARREN BUFFETT: Yeah. Anybody that, implicit in their comments or what they kind of laugh about or — all kinds of things in terms of the fact — you know, it’s so easy and — it ain’t that easy, you know, and we get suspicious very quickly. And the truth is, we rule out 90 percent of the times. And we may be wrong about a fair number that we’re ruling out. The important thing is whether the ones we’re ruling in we’re right about. And so we don’t mind — we’re looking for the obvious cases of people you can trust. I mean, go back to 1969 again. When I was thinking about who to turn my partners over to, all kinds of people with great records. That was a hedge fund — that was the first hay-day of hedge funds. And there were books written about it, “New Breed on Wall Street” and some of those. You can look them up. And dozens and dozens and dozens of people with good records.

WARREN BUFFETT：是的。凡是那种在话里话外、在他们自己笑点里，透出一种“你看，这事儿太容易了”的人——而我们知道，这世界没那么容易——我们都会很快起疑。事实是，我们会把 90% 的机会直接排除在外。在这些被排除的机会中，我们也许错过了不少好东西。但关键不在于我们错判了多少被筛掉的，关键在于：被我们留下来的那一小部分，我们是否判断正确。所以我们一点也不在乎“错杀”的问题——我们在找的是那些“显而易见可以信任的人”。举个例子，再回到 1969 年。当时我在思考，我要把我的合伙人们交给谁来管理——有一大堆战绩漂亮的人，那正是第一波 hedge fund 热潮的时代，还出了很多书，比如《New Breed on Wall Street》之类，你都可以去找来看看。当时有成打成打的“业绩很好的人”。

比如，对孩子的态度，认为孩子不听话就要打，也知道有些人的孩子不需要打就很出色，知道不等于明白，在Right People的判断中，巴菲特说“对决策有用的准确度”，“不听话就打”就是一种明显的暴露，让判断犹疑的地方，是不是稳定的行为模式？

1、大多数是稳定的行为模式，是“显而易见”的，有些人打孩子、打老婆的习惯是长期的，甚至已经有非常强的合理性的解释；

2、少数不容易判断的案例，比如，左晖的“难而正确”，张一鸣的“平庸有重力，需要逃逸速度”，“大力出奇迹”，以及王宁的“向上努力”，这三个人放在一起，“显而易见”的是王宁，左晖和张一鸣有点模糊，左晖已经过世但企业文化是传承的。

And when I sat down and thought about, I’m going to write my partners and tell them who to turn over all their money to — because most of them had a hundred percent or close to it with me — you know, Charlie, Sandy, Bill Ruane. I couldn’t have told you which of the three was going to do the best. And, you know, I couldn’t even tell you those three would done better than five others that somebody else might pick. But the one thing I was sure of is that they were going to be sensational stewards of money. They were going to care more about those people — the people that were turned over to them — in getting the best result possible than they were going to care about, you know, whether they made X or 2X this year in terms of commissions or fees or any of that sort of thing. Anytime I find somebody with a — what I regard as an unfair fee structure, and saying, “Well, I can get it,” well, you know, I rule them out. And I may rule out some of the wrong people. But the ones that are left in, I feel very good about.

而当我静下心来思考：我要给我的合伙人写信，告诉他们“以后把所有的钱交给谁打理”——要知道，他们当中大部分人，是把 100% 或接近 100% 的钱都交给了我——这时候，我想到的名字是：Charlie、Sandy、Bill Ruane。你要我说，这三个人里谁将来表现会最好，我说不出来。甚至，你要我说，这三个人一定会比别的人随便挑出来的另外五个人做得更好吗？我也说不出绝对的话。但我有一件事是百分之百确定的：他们会是“极其负责的管家”。他们会比起“自己今年拿多少 X、2X 的佣金或费用”，更在乎那些把钱交给他们的人，能不能拿到尽可能好的长期结果。任何时候，只要我看到有人使用在我看来“不公平的收费结构”，然后还一副“我能拿到这些钱理所当然”的样子，那这个人对我来说就出局了。没错，这样我可能也会把某些好人排除在外。但留在名单里的那批人，我会非常放心。

巴菲特的用词非常严谨，这句话有两个部分组成：（1）“不公平的收费结构”，做错事；（2）“我能拿到这些钱理所当然”，不在乎，两者的组合说明有一种心理结构起到了压倒性的作用。

And I wish I could give you better advice than that, but that’s all I can do.

我当然也希望能给你一个比这更“系统化”的建议，但我能做的，也就只有这些了。

Appendix.21.《2008-05-03 Berkshire Hathaway Annual Meeting》

50. Buffett’s advice for young people

巴菲特给年轻人的建议

WARREN BUFFETT: Number 8.

WARREN BUFFETT：第8位。

AUDIENCE MEMBER: Good afternoon. I’m (inaudible) from Apopka, Florida. I would like to thank you for this wonderful opportunity to learn so much more about finances and, at the same time, have a wonderful time. It’s been fun.

AUDIENCE MEMBER：下午好。我是来自 Florida 的 Apopka 的（听不清）。感谢你们提供这次绝佳的机会，让我在学到更多金融知识的同时也度过一段美好的时光。非常有趣。

WARREN BUFFETT: That’s great.

WARREN BUFFETT：太好了。

AUDIENCE MEMBER: I teach at Valencia Community College in Orlando, Florida. I teach office administration and business. I applaud my students for investing in themselves by enrolling in college. I also want to stress financial independence and financial freedom. I do this by telling them that slow and steady wins the race; also, good, sound financial management; and then the law of reciprocity. I have them track every expense over a period of time. Also they buy, theoretically, one stock and track that, too, over a period of time. We keep track of the current financial news, current news, and also they have research papers. (Applause.)

AUDIENCE MEMBER：我在 Florida 的 Orlando 的 Valencia Community College 任教，教授办公室管理和商业。我为我的学生通过上大学来投资自己而喝彩。我也强调金融独立与财务自由。我通过告诉他们“慢而稳者赢得比赛”、以及良好且稳健的财务管理、还有“互惠法则（law of reciprocity）”来做到这一点。我让他们在一段时间内记录每一笔开支；同时，他们会“理论上”买入一只股票，并在一段时间内对其进行跟踪。我们关注当前的金融新闻、时事新闻，并且他们还要写研究论文。（掌声。）

AUDIENCE MEMBER: Thank you. (Laughter) They research FICO scores, the credit reports, Clark Howard, Suze Orman, the top ten billionaires, not because — (Applause)

AUDIENCE MEMBER：谢谢。（笑声）他们研究 FICO 分数、信用报告、Clark Howard、Suze Orman、全球十大富豪，并不是因为——（掌声）

WARREN BUFFETT: I think — Could you move onto your question, then, please? (Laughter)

WARREN BUFFETT：我想——请你进入提问可以吗？（笑声）

AUDIENCE MEMBER: Yes.

AUDIENCE MEMBER：好的。

WARREN BUFFETT: But thank you.

WARREN BUFFETT：不过谢谢你。

AUDIENCE MEMBER: OK. What else should I be doing to lead them — (laughter) — to make sound financial decisions and to have happy, sound financial life?

AUDIENCE MEMBER：好的。我还应该做些什么来引导他们——（笑声）——做出稳妥的财务决策，并拥有幸福而稳健的财务生活？

WARREN BUFFETT: I’m ready to hire your entire class right now. (Laughter) Well, I think you’re telling — you’re giving them some very good advice. I think that the most important investment you can make is in yourself. I mean, very, very, very few people get anything like their potential horsepower translated into the actual horsepower of their output in life. And potential exceeds realization to just an enormous factor with so many people. And I — one illustration you might try with your class — I tell them this when I talk to high school groups — just imagine that you’re 16 and I was going to give you a car of your choice today, any car you wanted to pick, and — but there was one catch attached to it — it was the only car you were going to get in the rest of your life, so you had to make it last there. You can pick out the fanciest you want, a Maserati, whatever it might be. How would you treat it? Well, of course you would read the owner’s manual about five times before you turn the key in the ignition. You’d keep it garaged.

WARREN BUFFETT：我现在就准备把你们全班都雇下来。（笑声）我认为你正在——你确实在给他们非常好的建议。我认为你能做的最重要的投资就是投资自己。我的意思是，能把自己“潜在马力”转化为人生“实际产出马力”的人，真的、真的、真的凤毛麟角。对许多人来说，潜力与实现之间的差距大得惊人。而且——有一个例子你可以在课堂上试试——我在和高中生交流时也会这么说：想象你现在16岁，而我今天要送你一辆你自己任选的车，任何一辆都行——但有一个前提——这将是你此生唯一的一辆车，所以你必须让它撑到最后。你可以挑最华丽的，Maserati，或者别的都行。你会怎么对待它？当然，你会在点火前把用户手册读上五遍。你会把它停在车库里。

Any little rust, you would get taken care of immediately. You’d change the oil twice as often as you were supposed to, because you know it has to last a lifetime. And then I tell the students, you get one body and one mind, and it’s going to have to last you a lifetime, and you better treat it the same way, and you better start treating it right now, because it doesn’t do you any good to start worrying about that when you’re 50 or 60 and the rust — that little speck of rust — has turned into something big. So anything your students do to invest in their mind and body — particularly your mind — we didn’t work too hard on the bodies around here, but — (laughter) — you know, it pays off. It pays off in an extraordinary way. Your best asset is your own self. And you can become, to an enormous degree, the person you want to be. When I get classes in universities, I just ask them to imagine they were going to buy one of their classmates to own 10 percent of for the rest of their life. Which one would they pick?

任何一点点锈斑，你都会立刻处理。你会把换机油的频率提高到规定的两倍，因为你知道它要撑一辈子。然后我告诉学生们：你只有一个身体和一个大脑，它们也要陪你一辈子，你最好用同样的方式对待它们，而且最好从现在就开始，因为等你50或60岁时再开始担心，等那一点点锈——那小小的锈点——已经变成了大问题，就一点用也没有了。所以，只要是对自己的身心进行投资的事——尤其是你的头脑——都值得去做。我们这里对身体不算太“下功夫”，不过——（笑声）——你知道的，这些投入会有回报，会带来非同寻常的回报。你最好的资产就是你自己。你可以在很大程度上成为你想成为的那种人。我去大学和学生交流时，会让他们想象：如果你可以“买下”某位同学未来一生10%的收益份额，你会选谁？

They wouldn’t pick the one with the highest IQ, or necessarily the one with the highest grades. They’d pick the person that’s going to be effective. And the reason people are effective is because other people want to work with them. They want to be around them. And other people they don’t want to be around. And those are qualities than an individual picks up — being generous, being humorous, being on time, not claiming credit for more than you do but rather less than you do, helping out other people — all kinds of human qualities that turn other people on, and then there’s things that turn other people off. And those are habits, and they’re the habits that you pick up when they’re the age of your students. The habits they have today will follow them throughout life, so why not have good ones? So that’s the only message I would give your students. (Applause)

他们不会选择IQ最高的那个，或者不一定选择成绩最好的那个。他们会选那个“有效”的人。而人之所以“有效”，是因为别人愿意和他一起工作，愿意待在他身边；相反，也有人会让别人不愿意靠近。这些来自个人所养成的品质——慷慨、幽默、守时、不把功劳往自己身上多揽、反而少揽、乐于助人——各种各样能让别人喜欢的品格；当然也有一些会让人反感的东西。而这些都是“习惯”，而且正是在你学生们这个年龄段养成的习惯。他们今天的习惯会伴随一生，那为什么不让这些习惯变得更好呢？所以这就是我会给你学生们的全部信息。（掌声）

关于“有效”的几点补充。

1、我们有一个自己的总结：能够在“因为所以”之间建立足够扎实的、让自己有信心的联系，巴菲特的团队做了更结构化的处理：（1）足够安全的下行保护（7-8%，至少90%的信心）；（2）可能的上行收益（7%，至少50%的信心），90%、50%的信心跟“因为所以”之间的扎实程度有关。

2、想法“有效”需要其他人的配合，“愿意把功劳多归于他人、不走捷径、兑现所有承诺——还额外多给一点”可能是最好的策略，这样的脑回路扒开了一定有一些结构性的特征，再仔细想一想“足够安全的下行保护+可能的上行收益”，品格是内部结构外化的表现。

CHARLIE MUNGER: Well, I’ve got a specific suggestion that answers your specific question. I would add to that extensive repertoire of yours by teaching them to avoid being manipulated to their disadvantage by vendors and by lenders using the standard tricks of the vendor and lender trade. And you couldn’t start with a better book than Cialdini’s “Influence,” and I think Bob Cialdini, who is a shareholder, is here somewhere in this audience. And so I have a new textbook to — I suggest you add to your class — which is Cialdini’s “Influence.” And he’s just got a new book that’s coming out and for sale in Omaha today, I think, for the first time, and that’s called “Yes.” So here’s two books that I suggest you add to your class.

CHARLIE MUNGER：嗯，我有一个具体建议，正好回答你的具体问题。我建议在你已经很丰富的教学内容上再加一点：教他们如何避免被商家和放贷人利用行业里那些“标准招数”操纵而吃亏。没有比 Cialdini 的《Influence》更好的入门书了。我想 Bob Cialdini（他是股东）就坐在现场的某个地方。所以我有一本新的“教材”建议你加入到课程中——就是 Cialdini 的《Influence》。此外，他还有一本新书，我想今天在 Omaha 首次开售，书名叫《Yes》。所以，我建议你把这两本书都加进课堂。

WARREN BUFFETT: OK. (Applause.)

WARREN BUFFETT：好的。（掌声）

Appendix.22.《2010-05-01 Berkshire Hathaway Annual Meeting》

17. Creating a good corporate culture is easier than changing a bad one

创造良好的企业文化比改变不良的企业文化要容易。

WARREN BUFFETT: Area 6.

WARREN BUFFETT：第 6 区。

AUDIENCE MEMBER: Mr. Munger and Mr. Buffett, thanks for having us here. I recently joined a new organization and for me to succeed there, the culture of the organization needs to change. So I’m interested in hearing your thoughts about how do you change culture of an organization? And if you’re building a new organization, then how do you make sure you have a strong and unique culture?

AUDIENCE MEMBER：Mr. Munger 和 Mr. Buffett，感谢你们在这里接待我们。我最近加入了一家新的组织，而如果我要在那儿取得成功，这个组织的文化需要发生改变。所以我想听听你们的看法：应该如何改变一个组织的文化？如果你是在从零开始打造一个新组织，又如何确保能够形成一种强大而且独特的文化？

WARREN BUFFETT: Well, I think it’s a lot easier to build a new organization around a culture than it is to change the culture of an existing organization. It is really tough. And I like that fact, in the sense of Berkshire. I mean, it would be very tough to change the culture of Berkshire. It’s so ingrained in all our managers, our owners. Everything about the place is designed, in effect, to reinforce a culture. And for anybody to come in and try and change it very much, I think the culture would basically reject it. And the problem you describe, if you want to walk into, you know, whatever kind of organization you want to name — I’ve got to be a little careful here — it is tough to change cultures. Charlie and I have bucked up against that a few times. And I would say if you have any choice in the matter, I would much rather start from scratch and build it around it. But that was the — I’ve had the luxury of time with Berkshire.

WARREN BUFFETT：嗯，我认为，围绕一套文化去打造一个新组织，要比改变一个现有组织的文化容易得多。改变现有文化真的非常难。而就 Berkshire 而言，我反而喜欢这一点。我的意思是，现在要去改变 Berkshire 的文化会非常非常困难，因为它已经深深扎根在我们所有管理者、所有股东心中。关于这家公司的方方面面，从设计上看，实际上都是在不断强化这种文化。如果有谁想进来把它大幅度改一改，我觉得这套文化本身大概率会把他“排斥”出去。你刚才提到的问题——不管你说的是哪一类组织，我在这里得稍微注意一下说话方式——要改变它的文化都很难。Charlie 和我也曾经在这方面碰过几次硬仗。如果这个问题上你有选择权的话，我会更愿意从零开始，从一开始就围绕一套文化来建设组织。当然，这也是因为——在 Berkshire 这件事上，我有“时间优势”。

I mean, it goes back to 1965, and there really wasn’t much of anything there, you know, except some textile miles, so I didn’t have to fight anything. And as we added companies, they became complementary and they bought into something that they felt good about, but it took decades. And, you know, at Salomon, I attempted to change the culture, in terms of some respects, and I would not grade myself A+ in terms of the result. Charlie?

我的意思是，这一切要追溯到 1965 年，那时候那里实际上没什么东西，除了几家纺织厂，所以我并不需要去和既有文化“对着干”。随着我们不断收购新公司，它们彼此之间逐渐形成互补，也逐渐认同并融入了一种让大家感觉良好的文化，但这个过程花了几十年。再比如，在 Salomon，我曾在某些方面尝试去改变那里的文化，而如果要给这件事的结果打分，我绝不会给自己打 A+。Charlie？

“institutional imperative”，或者源自于Basic Mistrust的防御系统一旦建立会不断强化，消耗越来越多的能量，越往后越麻烦，比如，巴甫洛夫的实验结论中可以想办法用新的反射覆盖旧的，但如果整个系统的能量全被旧的反射机制榨干了，拿什么建立新的反射？问题分两层。

第一层是能量分配：旧反射的维护本身就在持续消耗算力——institutional imperative 的运行、Basic Mistrust 的防御循环、自我合理化的不断生产——这些都不是免费的，它们在持续地"烧"系统的可用能量。
第二层更深：旧反射不仅消耗能量，还会主动把任何威胁自身的刺激标记为"需要更多防御"，结果是越是真正能产生改变的刺激，越被既有系统调动更多能量去抵御。所以越想改的地方越改不动，因为越想改的地方就是系统能量消耗最多的地方。

CHARLIE MUNGER: Well, I’m quite flattered that a man would say that he’s in a new place where he can’t succeed unless he changes the culture and he wants us to tell him how to change the culture. In your position, my failure rate has been 100 percent. (Laughter) And —

CHARLIE MUNGER：嗯，我还挺受宠若惊的——有人说自己到了一个新地方，如果不去改变那里的文化就没法成功，然后希望由我们来告诉他该怎么改文化。站在你这个位置上，我自己过去在这方面的失败率是 100%。（笑声）然后——

WARREN BUFFETT: Yeah, Charlie started a law firm. Go back to, what, 1962, Charlie, what was it?

WARREN BUFFETT：是啊，Charlie 当年创办过一家律师事务所。要追溯到哪一年来着，1962 年，对吗，Charlie？

CHARLIE MUNGER: Yeah. I can move out but I couldn’t change the culture. (Laughs)

CHARLIE MUNGER：对。我可以“搬出去”，但我没办法改变那里的文化。（笑）

WARREN BUFFETT: We can tell some interesting stories from the old law firm, but we’ll go on to Carol now.

WARREN BUFFETT：我们其实可以从那家老律师事务所里挑出不少有意思的故事来讲，不过现在还是交给 Carol 吧。

Appendix.23.《2010-05-26 Warren Buffett.Interview With FCIC》

WARREN BUFFETT:

It didn’t cause it but there were a vast number of things that contributed to it. The basic cause was, you know, embedded in, partly in psychology, partly in reality in a growing and finally pervasive belief that house prices couldn’t go down. And everybody succumbed, virtually everybody succumbed to that. But that’s, the only way you get a bubble is when basically a very high percentage of the population buys into some originally sound premise–and it’s quite interesting how that develops–originally sound premise that becomes distorted as time passes and people forget the original sound premise and start focusing solely on the price action. So the media, investors, mortgage bankers, the American public, me, you know, my neighbor, rating agencies, Congress, you name it. People overwhelmingly came to believe that house prices could not fall significantly. And since it was the biggest asset class in the country and it was the easiest class to borrow against it created, you know, probably the biggest bubble in our history. It’ll be a bubble that will be remembered along with South Sea bubble and [unintelligible] bubble.

Warren Buffett：

它（评级失灵）并不是原因，但确实有大量因素对危机做出了贡献。根本原因在于——你知道——一种不断增长、最终变得无处不在的信念：房价不可能下跌。这种信念一部分来自心理，一部分来自现实。几乎所有人都屈服于此，几乎所有人都被它俘获。你要形成泡沫，唯一的方式就是：人口中非常高的比例买入某个“起初是合理的前提”——而且这个过程很有意思——这个起初合理的前提，随着时间推移被扭曲了；人们忘掉了原本合理的前提，只盯着价格走势本身。于是媒体、投资者、按揭银行家、美国公众、我、你、我的邻居、评级机构、国会——你能想到的几乎所有人——都深信房价不可能大幅下跌。又因为房地产是全国最大的资产类别，同时也是最容易拿来抵押借钱的资产类别，于是它制造了——你知道——可能是我们历史上最大的泡沫。这个泡沫会像 South Sea bubble（南海泡沫）以及某个（听不清的）泡沫一样，被人们长期记住。

“一部分来自心理，一部分来自现实”，这是巴菲特经常用到的思维框架，我有一个自己的理解，巴菲特一直在思考“可替换人员仍能维系的制度化技巧”，参考：《1975-10-14 Warren Buffett.PENSIONS》。

他一直在深入结构化的思维框架，“一部分现实+一部分心理”，“足够安全的下行保护+可能的上行收益”，泡沫唯一的入口是某个“起初是合理的前提”，这些都在做结构化的工作，更重要的是他相信事物的本来面目就是简单的、结构化的，如果这是成立的，那么“可替换人员仍能维系的制度化技巧”就是有底层逻辑的。

Appendix.24.《2016-08-16 黄峥.PDD.劣币驱逐良币，电影院现象及其他》

3、一粒屎坏了一锅汤

电影院里先站起来的那部分人，就如掉进汤里的一粒屎，很快整锅汤只能倒掉，这里指出的是一个比例问题。汤其实是有自我净化能力，如江河污水流入大海，如果比例很低，其实是ok的。但这个比例只要上升到一个临界点，原有的净化能力就崩溃了。很多时候我们会说一个坏现象不是主流没问题，并且它在另一个良性系统里也有，好像两边没有差别一样，这是很危险的想法。要知道坏了一锅汤的一粒屎的大小的临界点往往是很低的，也许是百分之一和百分之五的区别。百分之一能净化，百分之五就只能全部扔掉了。比方说一组人如果骗子的比例只有1%，那对群体作无罪推定这个社会是可以高效良性运转的，但有可能这个比例上升到5%、8%，再作无罪推定就会把其他人逼向骗子一方，这种情况下这个社会就只能作有罪推定，这样一来就是性质上的差异了。两个体系的文化运转效率会发生根本性的变化。但千万注意，这个量变引起的质变往往不是发生在50%，往往在一个低得多的值。在商业上，我们知道Amazon是在图书份额远低过50%时就实质上决定了图书的定价进而颠覆了图书零售业。每个人身上都带有癌细胞，正常是没事的，但如果它的复制和扩散起过某个值，你就得了癌症，而癌症要一个人的命远不需要扩散到50%。

所以当我们讲要看主流，或者类比两个不是100%完美的体系的时候，一定要关注1%和5%的差别，千万不要以为都是小部分是同样性质的。很多时候，质变就在这1%到5%之间。

4、改变群体自我伤害需要巨大能量

如果一个群体一百人，有8个人是骗子，而系统研究表明要比例低于3%无罪推定体系才work，那么，为了把8个人中的5个人转化过来，花的肯定不是5个人的力气，通常得10倍的力气，即50个人同时努力才能转化5个人。一个例子是纽约整治公共场所涂鸦，据说曾经纽约曼哈顿的中央火车站每个都被人涂鸦，为了整治这个，市长花了很大的力气，每天去刷一墙，别人涂一点，他也刷全部，连续反反复复很多很多天，涂鸦才逐渐变少，然后变没有。据说新加坡治理随地吐痰也类似，付出了很多很多倍于破坏者的代价才扭转了“风气”，形成良性运营的体系。

关于临界值的两个观察都非常好：

1、不可能是50%，是一个低得多的值，一般是低一个数量级的值，小于5%，假设平均每天有2个心理反应的事件，1岁以前700个；3岁以前2100个；15岁以前有1万个事件，假设3%是个临界值，3%的负面反应会产生结构性的差异，1岁以前是21次，发生21次照顾不到位的事件就可能产生结构性的后果。

2、新加坡有过自己的统计，Lee Kuan Yew 在一次演讲里提到：一流头脑约 1/1000，而一流头脑又具备正确品格与个性的约 1/3000，这里的重点不是精确数字，而是：临界点可能非常低。

Appendix.25.《2019-05-04 Berkshire Hathaway Annual Meeting》

31. Expand you circle of competence if you can, but don’t force it

如果可以就扩张你的能力圈，但不要硬来

WARREN BUFFETT: OK. Station 8. (Laughter)

WARREN BUFFETT：好的，8号麦克风。（笑声）

AUDIENCE MEMBER: Hi, Warren. Hi, Charlie. My name is Jacob (PH). I’m a shareholder from China and also a proud graduate of Columbia Business School. Thanks for having us here. (Cheers) My question is, our world is changing at a faster pace today versus 40 years ago and even more so going forward. And in this context, for each of us individually, should we expand our circle of competence continuously over time? Or should we stick with the existing circle but risk having a shrinking investment universe? Thank you.

观众：你好，Warren。你好，Charlie。我叫 Jacob，是来自中国的股东，也是 Columbia Business School 的骄傲毕业生。谢谢你们邀请我们来这里。（欢呼）我的问题是：现在这个世界变化的速度，比 40 年前快得多，而且未来只会更快。在这样的背景下，对我们每个人来说，是不是应该不断扩张自己的能力圈？还是应该坚守现有的能力圈，但承担可投资范围逐渐缩小的风险？谢谢。

WARREN BUFFETT: Well, obviously you should, under any conditions, you should expand your circle of competence —

WARREN BUFFETT：嗯，很明显，在任何情况下，你都应该扩张你的能力圈——

CHARLIE MUNGER: If you can.

CHARLIE MUNGER：如果你做得到的话。

WARREN BUFFETT: If you can. Yeah. (Laughter) And I’ve expanded mine a little bit over time. But —

WARREN BUFFETT：如果你做得到，是的。（笑声）我自己的能力圈，确实也在这些年里稍微扩大了一点。但是——

CHARLIE MUNGER: If you can’t — I’d be pretty cautious.

CHARLIE MUNGER：如果你做不到——那我就会非常谨慎。

WARREN BUFFETT: Yeah. You can’t force it. You know. If you told me that I had to, you know, become an expert on physics or, you know —

WARREN BUFFETT：是的。你不能硬来。你知道的。如果你现在要求我去成为物理学的专家，或者说——

CHARLIE MUNGER: Dance maybe the lead in a ballet, Warren. That would be a sight. (Laughter)

CHARLIE MUNGER：也许是去跳芭蕾当男主角，Warren。那画面可就有意思了。（笑声）

WARREN BUFFETT: Yeah, well. That’s one I hadn’t really —

WARREN BUFFETT：是啊，这个嘛。这个选项我倒真是——

CHARLIE MUNGER: (Inaudible) now.

CHARLIE MUNGER：（听不清）现在可以考虑一下。

WARREN BUFFETT: That’s one you may be thinking about, but I— (laughter) — it hadn’t even occurred to me. (Laughter) But, you know, it’s ridiculous. That doesn’t mean you can’t expand it at all. I mean, I did learn about some things as I’ve gone along in a few businesses. In some cases, I’ve learned that I’m incompetent, which is actually a plus, then you’ve discarded that one. But it doesn’t really — the world is going to change. And it’s going to keep changing. It’s changing every day. And that makes it interesting. You know. And as it changes, certainly within what you think is your present existing circle, you have to — you should be the master of figuring that one out or it really isn’t your circle of competence.

WARREN BUFFETT：这个是你在琢磨的，但对我来说——（笑声）——连想都没想过。（笑声）不过你知道，这显然是不现实的。但这并不意味着你完全不能扩张能力圈。我的意思是，在过去的一些生意里，我确实慢慢学会了一些新的东西；有时我也学到“自己在这个领域完全不行”，这其实反而是好事，因为你可以把它从列表里删掉。但这不改变一个事实——世界一定会改变，而且会持续改变，而且是每天都在变，这本身就很有趣。随着世界变化，在你自认为现有的能力圈之内，你必须——也应该——做到真正的精通，如果连那一块你都搞不明白，那它其实根本不属于你的能力圈。

And if you get a chance to expand it somewhat as you go along — I’ve learned some about the energy business from Walter (Scott) and Greg (Abel) as we’ve worked together, but I’m not close to their level of competence on it. But I do know more than I used to know. And so, you get a chance to expand it a bit. Usually, I would think normally your core competence is probably something that sort of fits the way the mind has worked. Some people have what I call a “money mind.” And they will work well in certain types of money situations. It isn’t so much a question of IQ. The mind is a very strange thing. And people have specialties, whether in chess or bridge. I see it in different people that can do impossible — what seem to me — impossible things. And they’re really kind of, as Charlie would say, stupid in other areas — (laughs) — you know. So just keep working on it. But don’t think you have to increase it and therefore start bending the rules. Anything further, Charlie?

而如果你在这个过程中有机会稍微把能力圈扩张一下——比如，我和 Walter（Scott）以及 Greg（Abel）一起工作时，确实也学到了一些能源生意的东西，但我的水平离他们还差得远。不过，相比过去，我确实懂得更多了一点。也就是说，你可以有机会把能力圈往外推一点点。通常，我认为所谓的核心能力，大概率是跟你大脑一贯的运转方式“对得上”的东西。有些人天生就有“经济头脑”，在某些跟钱打交道的场景里，他们就能如鱼得水。这跟 IQ 高不高没那么大关系。人的大脑是很奇怪的东西，人们各自有自己的“专长”，不管是国际象棋还是桥牌。我见过一些人能做到在我看来“根本不可能”的事情，但同时，他们在别的领域——用 Charlie 的话说——其实挺愚笨的（笑）。所以，你就是不断在自己这块上多下功夫就行了，但不要因为觉得“必须扩张能力圈”，就开始扭曲自己的规则。还有什么要补充的吗，Charlie？

巴菲特解释了“能力圈”的形成过程和时间序列的关系，人类知识轨迹的空间太大了，一个人不可能先站在更高一层，把所有行业、所有路径、所有知识结构都看一遍，然后理性挑出最优项，真实路径只能是：（1）先碰到一件具体的事；（2）一头扎进去；（3）拿到反馈，形成局部压缩；（4）判断自己是否满意，满意就继续，不满意再换下一件。

第（3）、（4）点跟每个人的独特性有关，有些人天生就有“经济头脑”，有些人就是没办法上在金钱问题上建立比较强的反射，把这个问题投射到结构化的思维框架，比如，巴菲特总结的“下行保护+可能的上行收益”，现在有一个财务上的问题摆在前面，需要将相关的信息压缩到一个清晰的结构，显然，这是需要“passion”的，有时候还需要比较强的“passion”。

如果“大脑一贯的运转方式”和金钱问题不匹配，天生没有“经济头脑”就很难，结构化的思维框架不是最终的解。

CHARLIE MUNGER: No.

CHARLIE MUNGER：没有了。

Appendix.26.《2021-05-01 Berkshire Hathaway Annual Meeting》

43. Top “risk factor” for any business is getting the wrong management

任何企业的首要“风险因素”是管理层不当。

WARREN BUFFETT: The biggest danger — they have that section in the prospectus called, what do they call that? About — certain —

沃伦·巴菲特：最大的危险——他们在招股说明书中有一个部分，叫什么来着？关于——某些——

GREG ABEL: Risk factors.

格雷格·阿贝尔：风险因素。

AJIT JAIN: Risk factors.

阿吉特·贾因：风险因素。

WARREN BUFFETT: Risk factors, yeah.

沃伦·巴菲特：风险因素，是的。

The number one risk factor — you never see it — the number one risk factor is that this business gets the wrong management. And you get a guy or a woman in charge of it that are — they’re personable, the directors like them. They don’t know what they’re doing, but they know how to put on an appearance. That’s the biggest single danger that a business — and that that person stays and runs it for 10 or 15 years, and either stays in the textile business or department store business and expands. (Laughs) And, you know, I’ve looked at a lot of businesses. And that’s what’s caused the number one problem. And it isn’t the kind of thing where they list them all because the lawyers tell them to list them.

首要风险因素——你永远看不到——首要风险因素是这个企业得到了错误的管理。你可能会让一个人或一个女人负责这个企业，他们很有亲和力，董事们喜欢他们。他们不知道自己在做什么，但他们知道如何营造一种表象。这对一个企业来说是最大的单一危险——那个人在那里待了 10 或 15 年，要么继续留在纺织行业或百货商店行业并扩展。（笑）而且，你知道，我看过很多企业。这就是造成头号问题的原因。这并不是他们因为律师告诉他们要列出所有问题而列出的问题。

“integrity, intelligence and energy”，但凡爬上高位的，“intelligence and energy”往往不是问题，最有可能出现的是“想的”和“做的”是不对齐的，不是真的热爱正在做的业务，一心想着做大规模、英雄梦，追求技术难度，行业地位，等等，比较显眼的是英雄梦，比如，贝佐斯和马斯克，他们个人有超越经济目标的追求，想推动全人类的进步，这些不是问题。

问题是贝佐斯和马斯克的个人追求是不是等于股东的追求？当然“不知道自己在做什么”的股东因为脑子里没有主体，贝佐斯和马斯克英雄式的叙事可以减少认知上的负担，两边都得到了心理上的满足，把财务上的风险留给了公司。

Appendix.27.《2023-03-02 武志红.如何构建强大的自我》

武志红:

你刚才讲了一个很生动的例子，也很具体的例子，关于乐高的这个故事。就是当孩子玩乐高的时候，他全然的投入其中，他在一个高度专注状态。那这时候如果谁能够看见那个看不见的东西的时候，就知道孩子的专注，这是一个极其宝贵的东西。不过这时候家长就不仅说，哎，你别玩了，你要去学习，有个网课要怎么怎么样，这是一种破坏。但还有些家长是这样，也有些家长看起来对孩子也挺关爱的。孩子累了吗？来喝点水，吃点水果。你这个乐高搭得真不错，你跟爸妈分享一下你搭乐高的感觉是怎样。

结果你看这都是对孩子的一个感觉的破坏。其实我们就明白孩子能够持续的在他的这个专注之中，如果能持续 45 分钟，那就是非常了不得的事情。我想专注是一部分，还有一部分，你看我们在讲这么一个故事，当孩子在专注的做搭乐高的时候，他在一种享受之中。甚至假如他能够持续 45 分钟的专注，其实他会体验到一定的心流。那么这个时候，孩子就处在一种愉悦，处在一种享受之中，那这是非常宝贵的。当孩子一旦能够在一件事情之中，他体验到，当他能够全然投入其中，有一定的心流的体验产生，他就获得了一个极其宝贵的事情，就是做事情是享受的。

所以我想这是真正的力量和虚假的力量一个极其巨大的区别。当我们体验不到做事情是享受的，最后就变成什么呢？学习是苦的，工作是苦的，都是为了生存，不得已，都是那个人第一名，自己第二名，感觉不爽，不行，要超过他，或者说自己如果发展不好以后怎么活呀？都是基于这么一个部分，会认为为了活着，为了证明自己比别人强，不得不去努力地学习和工作。当有了这种感觉之后真的很不幸你就沦落成普通人了。

当我们有另外一种感觉做事情是享受的，是愉悦的，那么这个时候就形成了所谓的内趋力。内趋力不是说有个家长在内在拿着鞭子抽打着他，你不努力你就怎么样，内趋力是和愉悦连在一起。他深刻地体验到当他去做事的时候，无论是学习工作还是娱乐，他总是专注在一件事情的时候，这是一个非常愉悦的事情。而且这个立在的愉悦就是它带来的享受要超过物质，甚至超过各种各样的这种这个世界上用物质带来的这些东西。那么这样一来，孩子就获得一个极其宝贵的这么一种体验，就是高度专注，带来这个愉悦，而且带来这个世界上极其宝贵的愉悦，

而且在愉悦的过程中实际体验了一种合一，你发现你和这个事情好像有一种合而为一的感觉。但是对幼小的孩子来讲，我已经体验到这种合一，或者知道合一有什么，这是非常不容易的。但是他有这么一点感觉，这是极其宝贵的部分。那么未来这种专注、愉悦、合一、心流，这也成了一个世界上最好的东西。那么这样一来，对他来讲，学习也好，工作也好，都是愉悦的。那么有愉悦去做事，这感觉是很不一样。比方我们想，如果我们学习的时候是苦的，你在学习的时候是在损耗自己。

那么工作的时候你觉得是苦的，你也是在损耗自己，所以因此大家觉得很累，要躺平。但是假如你在学习的时候一直享受的，你工作的时候是享受的，这个时候你就完全不一样了。那么讲到这可能很多朋友就会说，工作是享受的吗？我给大家来举一个我自己的例子，就是你看我现在都写了六七百万字了吧，我具体也不知道我到底写了多少字，但总之六七百万肯定是有的，至少写了十几本书，在德道上那个课程一年写了一百万字。

但是每次我去做按摩的时候，按摩师都觉得很奇怪，因为很多按摩师知道我，说，哎，武老师，你怎么没有写作病呢？我的颈椎，我的肩膀都挺 OK 的。那为什么没有颈椎病呢？因为我享受，我在写作的时候，我乐在其中，非常的愉悦。所以这种愉悦，它不仅是一种感受上的开心，它作为一种能量，作为一种活力，它会滋养自己的身体。结果自己的身体也因此得到一种滋药。如果现在有人说，武老师，我给你……其实很多人不相信这个说法，但是对我来讲，这很真实的。

说我给你几十个亿，你这辈子别写作了，你别讲课了，你别思考问题了，那这事我是拒绝的。因为对我来讲，这带来的愉悦是非常不同的。所以你看，对我来讲，专注写作，我并没有什么写作病，甚至我都没有说做了太久了，有这个作者带来的疾病，也没有。这是因为写作带来一种高度的享受和愉悦。但是我们做自媒体，我们这些小孩就是到我们这儿来工作，他们开始码字。真的是一两年就开始有写作病了。这个颈椎变粗大了，然后肩膀变得开始僵硬了，然后不断地要去按摩，动不动就疼。那这个就不一样是吧？

当你必须作为一个成年人，你经常在工作之中，但是你是愉悦的。而另外一个人，他在工作中，他是痛苦的。你看，这样一下来，生命差别有多大呀？时时刻刻都在做的一件事，这边觉得是愉悦的，这边觉得是痛苦的，这是一个非常巨大的不同。所以我想我们做父母的或者做教育的，包括我们自己的时候得看见。当然这一点我们得明白，我们成年人重新学习愉悦的去做事不容易，但是我们看孩子，孩子当他天然地做事的时候，他都是愉悦的呀。所以做父母的很多时候，后来我就说，中国家长想躺平挺容易的，

你别干涉孩子，让孩子爱干嘛干嘛，就保证他的吃喝拉撒睡玩就行了，你给他提供好物质基础，结果你的孩子你发现你没这么使劲，你居然已经胜过了绝大多数父母，因为大多数父母都在使劲地破坏自己的孩子，所以你没使劲你就成了。所以我想这是一个极其重要的这么一个部分，就是专注和愉悦，和另外一个不专注，就是痛苦，这些人生就是何等不同呢？

除了事实层面的正反馈，知识结构本身的““美、简洁、优雅”也能带来正面的感受，并且有预见性，但如果已经有认知上的障碍，巴甫洛夫的实验告诉我们，条件反射一旦建立，你能做的最好的事情不是消除它，而是在它之上建立一个更强的新反射来覆盖它。

更强的新反射有两种选择：（1）树立正确的榜样；（2）恐惧覆盖，两种方式都只是覆盖，包括这（1）种，正确的榜样不等于正确的思维方式，榜样不会自动生成内部的思维链条，更底层的思维方式很难改变，显然第（2）种选择只会让问题更严重。

Appendix.28.《2025-05-03 Berkshire Hathaway Annual Meeting》

[00:33:58] Warren Buffett: There are some problems that can’t be solved, and we shouldn’t be in the business of taking your money, investors’ money and tackling things that we don’t know the solution for. You can present the arguments, but it’s a political decision when you are dealing with states or the federal government. If you’re in something where you’re going to lose, the big thing to do is quit.

Warren Buffett：有些问题是无法解决的，而我们不应该拿着投资者的钱，去处理那些我们根本不知道解决方案的问题。你当然可以提出各种论证，但当事情涉及州政府或联邦政府时，最终就是一个政治决策。如果你处在一个注定会输的局面里，最重要的事情就是退出。

If you do present your case as well as you can, and everything else, but if you don’t hold the pen in the end, you know we don’t have any business taking your money and doing dumb things with us. We can do our best to explain what the intelligent things are, but it’s your money. So it’s very hard to tell on that. I have all the questions about politically determined decisions that are going to go to court in many cases, but you know that it doesn’t make sense. We know what we think a sensible system would be, and we ought to explain what we think it is and do our best to get our position because it’s pro-social. They have the right solution to have it. But the right solution, for example, in the Interstate, wasn’t to let 48 states each decide their own way of doing it, and award contractors the jobs. You know, there are some problems that can’t be solved, and we are not in the business of trying to solve insolvable problems.

即使你已经尽力把自己的立场讲清楚，但如果最终不掌握决定权，那么我们就没有理由拿着投资者的钱去做一些愚蠢的事情。我们当然可以尽力解释什么是理性的做法，但那毕竟是投资者的钱。因此在这种情况下，很难判断事情最终会如何发展。我经常遇到各种由政治决定的问题，很多情况下最终都会进入法院，但你其实也知道，这种方式并不合理。我们知道一个理性的体系应该是什么样，我们也应该把自己的看法解释清楚，并尽最大努力去争取我们的立场，因为那是对社会有益的。但真正正确的解决方案——比如当年的州际高速公路系统——并不是让48个州各自按照自己的方式去决定怎么做、各自去分配工程合同。确实有一些问题是无法解决的，而我们并不是一家去解决不可解决问题的公司。

But then the problem we have, of course, is that the people who work for you, that’s their job. So they want to have reasons to keep going, and those are tough choices if you’re managing, but that’s why they have managers.

但与此同时，我们也面临一个现实问题：那些为你工作的人，他们的职责就是把事情继续推进下去。所以他们总会找到理由继续做下去。而当你是管理者时，就必须面对这些艰难的选择，这也是为什么需要管理者存在的原因。

不受噪音干扰的情绪结构，想法如果不能收敛就会碎片化，听什么都有道理，分不清什么重要什么不重要；更不知道什么是重要并且可知的，什么是重要、不可知的信息。

*****

Becky Quick: This question comes from Scott Williams in Portland, Oregon. He said, “Do you think the net benefit of DOGE will be positive or negative for the long-term health of the United States?”

Becky Quick：这个问题来自俄勒冈州波特兰的 Scott Williams。他说：“你认为 DOGE 的净效应对美国的长期健康是正面还是负面？”

Warren Buffett: I think that bureaucracy is something that is amazingly prevalent and contagious even in our capital system, and big corporations overwhelmingly most of them look like they could be run better. I’m sure Berkshire does in many respects. And government is the ultimate. It really doesn’t have any checks on it.

Warren Buffett：我认为官僚主义在我们的资本体系中也惊人地普遍且具有传染性，绝大多数大型公司看起来都有提升经营的空间。我确信在很多方面 Berkshire 也是如此。至于政府，那更是终极形态——它实际上缺乏有效的制衡。

That’s why it scares you to some extent about what the future of the currency will be, because they can print currency. If you have people that get elected by promising people things – that doesn’t mean that they aren’t sincere about all kinds of items, but there’s no politician that says to anybody that has money, “I really think you have bad breath and if you don’t mind, would you step away from me.” It just doesn’t happen.

这也是为什么我们在一定程度上会对货币的未来感到担忧，因为他们可以印钞。若有人靠向民众许诺而当选——这并不意味着他们对各项主张不真诚，但没有政治人物会对有钱人说：“我觉得你口气很糟，麻烦离我远一点。”这种事不会发生。

I think the problem of how you control revenue and expenses in government is the one that is never fully solved and has really hurt dramatically many civilizations. I don’t think we’re immune from it, and we’ve come close to it.

我认为政府如何控制收支这个问题从未得到彻底解决，并且严重伤害过许多文明。我不认为我们能免疫，而且我们已多次接近危险边缘。

We’re operating at a fiscal deficit now that is unsustainable over a very long period of time. We don’t know whether that means two years or 20 years because there’s never been a country like the United States. But as Herbert Stein, the famous economist, said, “If something can’t go on forever, it will end.” We are doing something that is unsustainable, and it has the aspect to it that it gets uncontrollable to a certain point.

我们当前的财政赤字在很长时期内是不可持续的。究竟是两年还是二十年，我们并不知道，因为从未有过像美国这样的国家。但正如经济学家 Herbert Stein 所言：“不能永远持续的事，终将终止。”我们正在做一件不可持续的事，而且其中有一部分会在某个点变得不可控。

Paul Volcker kept that from happening in the United States, but we came close. We’ve come close multiple times. We’ve still had very substantial inflation in the United States, but it’s never been runaway yet. That’s not something you want to try and experiment with because it feeds on itself.

Paul Volcker 曾阻止这种情况在美国发生，但我们离临界点很近；而且不止一次。美国也经历过相当高的通胀，但尚未失控。你绝不想拿它做试验，因为一旦启动就会自我增强。

I wouldn’t want the job of trying to correct what’s going on in revenue and expenditures of the United States with roughly a 7% gap when probably a 3% gap is sustainable. The further away you get from that, the more you get to where the uncontrollable begins. It’s a job I don’t want, but it’s a job I think should be done. And Congress does not seem good at doing it.

我不愿意接手修正美国收支问题的工作——现在缺口大约是7%，而大概3%才是可持续水平。偏离越远，就越接近不可控。我不想做这件事，但我认为它必须有人去做。而国会看起来并不擅长此事。

We’ve got a lot of problems always as a country, but this is one we bring on ourselves. We have a revenue stream, a capital-producing stream, a brains-producing machine like the world has never seen. And if you picked a way to screw it up, it would involve the currency. That’s happened a lot of places.

作为一个国家，我们总会有很多问题，但这一项是我们自找的。我们拥有前所未见的收入创造力、资本生成力与人才生产机制。如果你非要挑一种方式把这一切搞砸，那就是货币问题。许多地方都发生过。

In theory, you would make it so there was substantial downside for anybody that screwed things up, but there isn’t downside. There’s upside. It’s the problem of the most successful company in the history of the country, in the history of the world.

理论上，你应当为任何把事情搞砸的人设置巨大的负面后果，但现实中并没有，反而有上行激励。这是这个国家、乃至世界历史上最成功体系所面临的问题。

Warren Buffett: We’re spending close – it’s hard to get the precise figure, but close to 20% of GDP on health. And if you go back to 1960, there were a number of countries that were each spending around 5%. And then the lines began to diverge dramatically. But the mathematical fact that there are only 100 percentage points in the equation didn’t change.

Warren Buffett：我们在医疗上的开支接近——确切数字很难给出，但接近——GDP 的 20%。如果回到 1960 年，当时有不少国家各自大约只花 5%。此后曲线开始显著分化。但“总量只有 100 个百分点”这一数学事实并没有改变。

So we tried that experiment with JP Morgan and Amazon and we had three people that didn’t think they knew the answer, but thought that in my case I use the term that it was a tapeworm in the economy. We also found out that the tapeworm was alive in every part of the country. I mean the hospitals liked it. The hospitals had prominent people working with people. People generally like their doctor, didn’t like the system. I mean all kinds of things, but in the end, JP Morgan and Amazon and Berkshire were not going to have any effect on changing that 20%.

所以我们与 JP Morgan 和 Amazon 一起做了那个尝试。我们三个人并不自认为知道答案，但我用过一个说法：这是经济体内的一条“tapeworm（绦虫）”。我们也发现，这条“绦虫”寄生在全国每个角落。比如医院也“喜欢”它，医院里有显赫人物在与各方打交道。人们普遍喜欢他们的医生，但不喜欢这个体系。情况五花八门，但最终，JP Morgan、Amazon 和 Berkshire 不可能把那 20% 改变掉。

Now that 20% – there are only 100 percentage points available and other countries spend six or 7% and perhaps use our system to their advantage which is also very true. That is an enormous percentage of an economy and we simply – it was too entrenched to really do much in the way of change.

而那 20%——总共也就 100 个百分点，别的国家花 6% 或 7%，还可能利用我们的体系为己所用，这也确实存在。医疗占经济的比重过于庞大，而这个体系根深蒂固，我们确实很难推动实质性改变。

And we spent some money on it and we did some work and we learned a good bit about our own systems and we saw the degree to which the present system was ingrained in so many people’s lives, whether the health care providers or whether everybody. And these aren’t evil people. I mean, they’re just going about something and trying to save lives.

我们花了些钱，也做了不少工作，从中学到了很多关于我们自身体系的东西，也看到了当下体系是如何深深嵌入无数人的生活——无论是医疗服务提供者，还是普通大众。这些人并不“邪恶”，他们只是各司其职、试图挽救生命。

But we found that whether it was in Canada or France or Britain or wherever it might be, that if you looked at our costs that they were just far higher and to some extent we were subsidizing the rest of the world. And people would come to the United States to do the really unusual or challenging aspects of health in terms of operations and that sort of thing.

但我们发现，无论在 Canada、France、Britain 或其他地方，相比之下我们的成本要高得多，而且在某种程度上我们是在补贴世界其他国家。很多人在涉及高难度手术等复杂医疗时会来到美国。

But we made no progress and there comes a point where the government – and I mean it’s so involved in the situation and health is so important to most to everybody, and we couldn’t – as I said to Jamie and Jeff, I said well the tapeworm won.

但我们没有取得进展，而到了某个节点，政府——我指的是政府在其中的深度参与——再加上医疗对几乎所有人都至关重要，我们也无能为力。正如我对 Jamie 和 Jeff 说的：“绦虫赢了。”

There are problems of society when you get 20% of your GDP going into a given industry, the degree of enthusiasm for changing that industry, the political power that the industry will have, and that doesn’t mean they’re evil. It’s just everybody – they just end up there.

当一个行业吃掉 GDP 的 20% 时，社会就会出现问题：你要改变该行业的热情会被削弱，该行业的政治力量会壮大。这不意味着他们“邪恶”，只是所有人最终都会被卷入到那样的格局里。

So I don’t know – we came to the conclusion we didn’t know the answer, the three of us, and we had the money to do it and we didn’t know how to change how 330 million people felt about their doctor, felt about our healthcare, what they felt entitled to.

所以我也不确定——我们三人得出的结论是：我们没有答案。我们有资金，但我们不知道如何改变 3.3 亿人对医生的感受、对医疗的看法以及他们认为自己“理应拥有”的权利。

It won’t change by itself and government is the only one that can change it and the only people in government that can change it are getting majority of 435 people and 100 people. And my dad lost one election in his life in 1948 and he was a very strong Republican and in 1950 he went back and beat the guy that beat him in 1948 and he got the doctors behind him. And they believe 100% in what they’re doing. They’re helping people every day.

这件事不会自发改变，只有政府才能改变；而政府中能改变它的人，必须在 435 人与 100 人中取得多数。我父亲一生只在 1948 年输过一次选举，他是铁杆共和党，1950 年他又回去击败了 1948 年击败他的对手，因为他争取到了医生群体的支持。他们 100% 相信自己所做的事——他们每天都在帮助他人。

And during the pandemic, the sacrifices made by people to save other people – just incredible. Can you imagine working in something where they’re bringing in people that are going to die by the dozens and dozens and dozens and you try to somehow keep your own morale up and keep working with them. So, you can’t argue about the importance of it.

在疫情期间，人们为拯救他人所作的牺牲——令人难以置信。想象一下，你在一个场景下工作，身边不断送来一批又一批濒危的人，而你要设法保持士气并继续投入。对医疗的重要性，这是无可争辩的。

But our costs are so different than any country in the world that it’s a huge element and we’re a very rich country. So we can do things other countries can’t do. And through our elected representatives and a whole variety of things over time, we’ve developed a system that is enormously resistant to any kind of major change and it’s important in every community that it’s in.

但我们的成本与世界任何国家相比都过于悬殊，而我们又是一个非常富裕的国家，因此我们能做其他国家做不到的事。长期以来，通过民选代表和各种过程，我们形成了一个对任何重大变革都极其“抗性”的体系，而且这套体系在每个社区都举足轻重。

我自己的看法更乐观一些，在人工智能的研究中，对“美、简洁、优雅（Beauty, simplicity, elegance）”的隐性偏好决定的泛化能力，某种程度上证明了好人（具备基本信任）能获得更强大的智慧，而坏人（缺乏基本信任）无论多精明，终其一生都只是一些支离破碎的小聪明。

So, I wish we had an answer for you, but I was somewhat pessimistic going in and I was a little more pessimistic when we came out. But I’m glad we did what we did and we learned something about our own failings in the process. So, Berkshire in effect got its money’s worth, but we didn’t kill the tapeworm.

所以，我也希望能给你一个答案，但进入时我就偏悲观，出来时更悲观一些。不过我仍庆幸我们做了那次尝试，也在过程中看见了自身的不足。换言之，Berkshire 的投入没有白花，但我们没能“杀死绦虫”。

Trying to change things in government is an interesting proposition in the country because you get self-selection in terms of the people that go into government and continue in it. And to some extent they keep having to make decisions that they don’t like as they go along and they learn to accept them or rationalize them or whatever it may be.

在这个国家，试图通过政府去改变，是一个“有趣”的命题，因为走进并留在政府的人群本身就具有自我筛选的特征。在某种程度上，他们一路上会做出自己并不喜欢的决定，并逐渐学会接受、或为之寻找理由，诸如此类。

But it’s still – this country’s worked out better than any country in the world. So you can’t argue it was a failure, but you can’t argue that there are certain problems that are terribly tough to figure out ways to solve. And of course, one of them gets back to the fiscal problem I mentioned before because it’s easy to spend money and it’s hard to cut people’s receipts.

但即便如此——这个国家的运转仍优于世界上任何国家。所以你不能称之为失败；但你也不能否认，有些问题极难找到解决路径。当然，其中之一又回到我之前提到的财政问题：花钱容易、削减他人收入很难。

And if you get elected, you’re going to say to yourself, well, I can do more good if I stay, and then if I really vote my conscience on this sort of thing. So, you give away a little bit here and a little bit there and a little bit there and finally you don’t recognize yourself in the mirror anymore. And that’s – I grew up in a political family, but I watched how people behaved and they behave like human beings which is what you have to expect and I behave like human beings.

而当你当选，你会告诉自己：如果我留在这里，我就能做更多好事；然后在这种事情上你若按良知投票……于是这里让一点，那里让一点，久而久之镜子里已认不出自己。我出身政治家庭，见过人们的行为模式，他们的表现与“人性”一致，这是你必须预期到的——而我也同样具有人的弱点。

We still manage to keep moving forward in dramatic way. It’s so much better to live here than it was 100 years ago or 200 years ago. It’s dramatic. So you can’t say the system’s a failure, but you can say that it is very difficult to make major changes in it.

我们仍然能以令人瞩目的方式向前推进。生活在今天远胜 100 年、200 年前，变化是剧烈的。因此不能说这套体系失败了，但可以说在其中推动重大变革极其困难。

Appendix.29.《2025-10-30 田渊栋.AI“顿悟”的关键，是对优雅的追求？》

田渊栋：更准确地说，它发生在 reasoning 或其他任务之下的“共同底层”机制，那就是 representation learning（表征学习）。

随着训练推进，模型的表征会不断演化。一开始更像是死记硬背；但随着足够的积累和联结，结构会突然“贯通”，从而出现类似“读书百遍，其义自见”的转折点。比如说在小学生的教育中，老师可能会先要求他们背诵一些知识，过段时间通过新的知识联结，原本模糊的含义逐渐显现，这就是顿悟的一部分。

课代表立正：也就是说，无论是 chain-of-thought 还是直觉判断，其实最终都依赖于“我如何表示、如何理解这个世界”这一底层机制？

田渊栋：对。比如，小学生可能解题靠穷举；而进入初高中后，引入了数学归纳法，仅靠简洁的证明就能覆盖无限情形，这种方法背后的“表示”就发生了根本性变化。神经网络的学习关键差异，也正体现在表征方式上。

“顿悟”描述了神经网络在训练过程中，性能从长时间的停滞（看似只会记忆），突然飞跃到能够完美泛化（真正理解了规律）的现象。这与人类学习中“读书百遍，其义自见”或武侠小说里张无忌先背下心法再融会贯通的体验惊人地相似。

那么，这个神秘的“突变”究竟是如何发生的？田博士用一个生动的“双峰模型”揭示了其内在的数学图景：

记忆与泛化的不同“解”：在一个复杂的优化空间中，“记忆”和“泛化”可以被看作两个不同的解，对应着两个不同的“山峰”。记忆是一种低效的解，需要模型记住所有特例；而泛化是一种高效、优雅的解，模型找到了数据背后更简洁的统一规律（short program）。
数据驱动的山峰演变：当训练数据不足时，“记忆山峰”更高，因为记住所有样本是降低训练误差最直接的方式。此时，模型的优化过程自然会收敛到这个山峰。
此消彼长的临界点：随着数据量的增加，数据中潜在的“泛化规律”开始显现。这使得“泛化山峰”逐渐升高，而“记忆山峰”相对降低。当数据量跨过一个临界点，“泛化山峰”的高度首次超过了“记忆山峰”。
顿悟的发生：由于优化算法总是倾向于寻找全局最优解（更高的山峰），在“泛化山峰”成为最高点的瞬间，模型的参数便会“雪崩式”地涌向这个新的、更优的解。宏观上，这就表现为一次突然的、性能飞跃式的“顿悟”。

这个解释极大地祛魅了“涌现”或“顿悟”的神秘感，将其从一个看似随机的魔法，还原为一个由数据分布和优化动力学共同决定的、有清晰路径的物理过程。泛化的能力并非凭空产生，它一直作为一种可能性存在于数据之中，等待着足够多的证据使其“脱颖而出”。这个比喻的深刻之处在于：

确定性：它告诉我们，“顿悟”不是随机的奇迹，而是当数据量达到某个临界点后，几乎必然会发生的相变。
竞争性：“记忆”和“泛化”是两种相互竞争的解决方案，模型在训练中会动态地选择在当前数据下“性价比”更高的那一个。
可操作性：它启发我们，促进“顿悟”的发生，关键在于如何设计数据和训练方法，来更快地“抬高”泛化山峰，“压低”记忆山峰。

Appendix.30.《2025-11-26 Ilya Sutskever.We're moving from the age of scaling to the age of research》

Ilya Sutskever 01:25:13

Here is what I think is going to happen. Number one, let’s look at how things have gone so far with the AIs of the past. One company produced an advance and the other company scrambled and produced some similar things after some amount of time and they started to compete in the market and push the prices down. So I think from the market perspective, something similar will happen there as well.

我觉得会发生的事情大致是这样的。第一，我们先看看过去这几轮 AI 的演化是怎么走的：通常是某家公司率先做出一个重大突破，随后其他公司迅速跟进，在一段时间后搞出类似的东西，然后大家一起在市场上竞争，把价格往下打。我认为，从市场机制的角度看，这一轮也会出现类似的过程。

顺带一提，我们现在讨论的是“比较好的那条世界线”。什么叫“比较好的世界”？就是我们拥有这种强大的、类人学习者的 AI，同时还……再顺着说一个我们没怎么展开、但在设计 superintelligent AI 规格时很值得考虑的点：**你可以把它做窄一点，让它既超级智能，又在职能上保持窄域实用**。你完全可以有一堆“窄领域的超级智能 AI”。

But suppose you have many of them and you have some company that’s producing a lot of profits from it. Then you have another company that comes in and starts to compete. The way the competition is going to work is through specialization. Competition loves specialization. You see it in the market, you see it in evolution as well. You’re going to have lots of different niches and you’re going to have lots of different companies who are occupying different niches. In this world we might say one AI company is really quite a bit better at some area of really complicated economic activity and a different company is better at another area. And the third company is really good at litigation.

在这种设定下，假设有很多这样的系统，其中一家公司靠此赚取了巨额利润，然后另一家公司也杀进来开始竞争。竞争会怎么展开？答案是：**通过专业化来展开**。竞争最“偏爱”的就是分工与专业化——你在市场里能看到这一点，在进化里也能看到。最终你会出现大量细分“生态位”，对应一批占据不同生态位的公司。在那样的世界里，你可能会看到这样的格局：某一家 AI 公司在某种非常复杂的经济活动领域上明显更强；另一家公司则在另一个领域更胜一筹；还有第三家公司，也许在诉讼与法律业务上特别厉害。

Dwarkesh Patel 01:27:18

Isn’t this contradicted by what human-like learning implies? It’s that it can learn…

这难道不是和“类人学习能力”的含义相矛盾吗？因为它可以去学习……

Ilya Sutskever 01:27:21

It can, but you have accumulated learning. You have a big investment. You spent a lot of compute to become really, really good, really phenomenal at this thing. Someone else spent a huge amount of compute and a huge amount of experience to get really good at some other thing. You apply a lot of human learning to get there, but now you are at this high point where someone else would say, “Look, I don’t want to start learning what you’ve learned.”

它当然可以学，但这里还有“沉淀的学习成果”这件事。你已经在某个方向上投入了巨大的资源，你花了非常多的算力，才在这件事情上变得极其擅长、好到惊人。与此同时，另一个人（或系统）则在另一个方向上投入了同样巨量的算力和经验，在那件事上变得非常强。你们都用上了大量“类人学习”的能力才爬到各自的高点，而当你已经站在这个高度时，其他人就会说：“算了，我可不想从头开始学你已经学完的这些东西。”

自我强化带来两样东西，一是吸附力，被擅长的事牢牢吸住，越来越紧；二是越来越好的泛化能力，这是跨领域都适用的能力，最终的结果会是什么？

（1）算力资源有限

每家公司的“人脑 + GPU”都是稀缺的，最理性的做法是把自我强化的飞轮优先丢在“最能赚钱、回本最快”的那几块业务上。结果就是 Ilya 说的那种格局：一家公司在人类极复杂的某一块经济活动上远超同行，另一家在另外一块远超同行，第三家可能专门做诉讼这样的利基市场。竞争偏爱专业化，自我强化让专业化变成“惯性轨道”。

（2）算力资源无限

理论上，一家公司可以在每一个足够赚钱的岗位上都开一个独立的“类人学习体实例”，把整个经济里所有“正 NPV 的任务”全覆盖。这个世界里，约束从算力转向数据质量、监管约束和组织的管理复杂度——技术上可以全覆盖，现实里不见得有人能管理好这么庞杂的系统。

我自己有一个基本的判断是这个世界的分辨率会越来越高，分辨越高可以做的事越多，如果同样呈现幂律分布的规律，那么可以做的事、需要解决的问题会指数级的增长，算力资源永远都是不够的。

Dwarkesh Patel 01:27:48

I guess that would require many different companies to begin at the human-like continual learning agent at the same time so that they can start their different tree search in different branches. But if one company gets that agent first, or gets that learner first, it does then seem like… Well, if you just think about every single job in the economy, having an instance learning each one seems tractable for a company.

那这大概就要求很多不同的公司在差不多同一时间，都拿到这种“类人持续学习的智能体”，这样它们才能各自在不同分支上展开自己的那棵“搜索树”。但如果只有一家公司最先拿到了这种智能体，或者说最先拿到了这种学习者，那看起来就会变成这样……毕竟如果你只考虑经济体系里的每一种工作，让模型的不同实例分别去学习每一种工作，对一家公司来说似乎是可行的。

Ilya Sutskever 01:28:19

That’s a valid argument. My strong intuition is that it’s not how it’s going to go. The argument says it will go this way, but my strong intuition is that it will not go this way. In theory, there is no difference between theory and practice. In practice, there is. I think that’s going to be one of those.

这是一个有道理的论点。不过我的强烈直觉是：事情最后不会那样发展。你的推理指向那条路径，但我的直觉非常强烈地觉得，现实不会按那条路径走。从理论上讲，理论与实践之间没有差别；但在实践中，差别是存在的。我觉得这会是“理论和实践不一样”的又一个例子。

*****

You kind of ask yourself, is something fundamental or not fundamental? How things should be.

你会不断地问自己：什么东西是“基本的”、是“底层”的，什么不是？事物“应该是怎样的”？

I think that’s been guiding me a fair bit, thinking from multiple angles and looking for almost beauty, beauty and simplicity. Ugliness, there’s no room for ugliness. It’s beauty, simplicity, elegance, correct inspiration from the brain. All of those things need to be present at the same time. The more they are present, the more confident you can be in a top-down belief.

我觉得这些一直在很大程度上指导着我：从多个角度思考，同时去寻找某种“接近美感的东西”——美感、简洁性。丑陋的东西，是不该留下空间的。你要追求的是：美、简洁、优雅，以及来自大脑的“正确灵感”。这些要素需要同时出现，而且出现得越充分，你就越能在“自上而下的信念”（top-down belief）上有信心。

The top-down belief is the thing that sustains you when the experiments contradict you. Because if you trust the data all the time, well sometimes you can be doing the correct thing but there’s a bug. But you don’t know that there is a bug. How can you tell that there is a bug? How do you know if you should keep debugging or you conclude it’s the wrong direction? It’s the top-down. You can say things have to be this way. Something like this has to work, therefore we’ve got to keep going. That’s the top-down, and it’s based on this multifaceted beauty and inspiration by the brain.

这种“自上而下的信念”，就是当实验结果暂时和你唱反调时，支撑你继续往前走的东西。因为如果你永远只信任数据，那有时会出现这样一种情况：你做的是对的，但实验里有 bug，而你不知道那里有 bug。那你要如何判断，到底还要不要继续调试？是该说“这方向错了”，还是该说“系统里还有没找到的问题”？靠的就是这种 top-down 信念。你会对自己说：“事物必须是这样的，这种结构总得有一种方式是能工作的，所以我们得继续干下去。”这种 top-down 信念，正是建立在多维度的“美感”和“来自大脑的正确灵感”之上的。

头脑清晰是在大脑发展的早期就已经有一些简洁优雅的知识结构，在后面的人生中泛化到其他领域，必须是非常早的时期，两个方向（有安全感或者没有安全感）都是自我强化的，简洁优雅如果是胜出的一方只可能出现在非常早的时期，可能是1岁以前，甚至是娘胎里都有可能。

Appendix.31.Example.01.Charlie Munger

Refer To：《2024-05-04 Berkshire Hathaway Annual Meeting》。

Warren Buffett: I’m Warren Buffett and that’s me with Charlie in an unscripted and spontaneous photo taken in Savannah, Georgia early in 1982. Our wives could tell us apart, as long as we wore name tags. When Charlie and I first met in 1959, if was as if twins who had been separated at birth had been reunited. But there were a few important differences between Charlie and me that people missed. Let me elaborate.

沃伦·巴菲特：我是沃伦·巴菲特，这张照片是我和查理在 1982 年初在乔治亚州萨凡纳拍的，完全没有剧本和准备。我们的妻子只要我们佩戴名牌，就能分辨出我们。当查理和我在 1959 年第一次见面时，就像是失散多年的双胞胎重聚。但查理和我之间有一些人们忽视的重要区别。让我详细说明。

For one thing, I was only interested in whether things worked. Charlie wanted to know how things worked. I would turn on a switch and if the tv or light bulb went on, I could not care less about what had caused that miracle. Charlie, however, would want to understand every aspect of how a generator worked, how electricity traveled to his home, the merits of AC vs DC, whatever. You could say that Charlie understood electricity better than Thomas Edison ever did.

有一点，我只对事物是否有效感兴趣。查理想知道事物是如何运作的。我只需打开一个开关，如果电视或灯泡亮了，我根本不在乎是什么导致了这个奇迹。然而，查理会想要理解发电机的每一个方面，电是如何传输到他家里的，交流电与直流电的优缺点等等。你可以说查理对电的理解比托马斯·爱迪生更深刻。

Like his hero, Ben Franklin, Charlie wanted to understand everything, and he pretty well succeeded. Additionally, Charlie liked to design things. When we met in 1959, he was designing the building, the house in which he lived throughout his life. He once wanted to design a home for me in Santa Barbara, on a piece of property Berkshire had inherited - fat chance! In 1958, I bought the house I now live in, but I would have been happy with any of 100 or more houses as long it was in the general neighborhood of where I had been renting. Of course I wanted my wife to like it and wanted to have room for 4 or 5 kids, but what it looked like inside, outside was irrelevant.

像他的英雄本·富兰克林一样，查理想要理解一切，并且他做得相当成功。此外，查理喜欢设计东西。当我们在 1959 年见面时，他正在设计他一生中居住的那栋房子。他曾经想为我在圣巴巴拉设计一所房子，那是一块伯克希尔继承的土地——希望渺茫！在 1958 年，我买下了我现在居住的房子，但只要在我租房的附近，我会对 100 多栋房子中的任何一栋都感到满意。当然，我希望我的妻子喜欢它，并希望有 4 或 5 个孩子的空间，但房子的内部和外观对我来说并不重要。

“对事物是否有效感兴趣”源自于想解决现实的问题，并且相信“正确答案是存在的”，最后指向的是专注；“理解一切”的出发点和目的地跟前者是不一样的，前者相信知识是收敛的，后者不相信知识是收敛的，即使被告之、被引导、被证明、甚至自己也同意是应该收敛的。

有些“鸟”来到世上是来解决问题的，乔布斯在一次采访中说苹果的成功是专注于“真正重要的事情”，另一些有不同的目的。那些专注于解决问题的会很快发现自己擅长的点，比如，乔布斯发现三星和其他手机的设计师不知道自己是有手指头的，巴菲特在Las Vegas看到很多人喜欢把辛苦赚来的”好钱“扔进“坏钱”堆里。

在发现自己的擅长点以后，他们又会发现自己的时间和精力是有限的，聚焦和专注才能实现来到这个世界的最初的目的。

The difference, of course, was of supreme relevance to Berkshire. Charlie’s architectural thoughts led to the Berkshire Hathaway of today. Some of you may be surprised that Charlie first became a director of Berkshire in 1978. Through a small partnership, I bought control of Berkshire early in 1965. Charlie then didn’t have a penny invested in Berkshire. But he immediately told me purchase was just plain dumb. Which it was.

当然，这个差异对伯克希尔来说至关重要。查理的建筑思想导致了今天的伯克希尔哈撒韦。你们中的一些人可能会惊讶，查理在 1978 年首次成为伯克希尔的董事。通过一个小型合伙企业，我在 1965 年初买下了伯克希尔的控制权。那时查理在伯克希尔没有投资一分钱。但他立刻告诉我，这笔购买简直是愚蠢的。确实如此。

Charlie then did for me what needed to be done to correct my error. And over time, we worked together to achieve his vision. Charlie, in effect, then became the architect of today’s Berkshire. The architect is the person who dreams of, then designs, and finally supervises the construction of great structures. The carpenters and the (inaudible), that’s me, are needed, but the architect is the genius who provides the blueprint.

查理为我做了纠正错误所必需的事情。随着时间的推移，我们共同努力实现他的愿景。查理实际上成为了今日伯克希尔的建筑师。建筑师是梦想、设计并最终监督伟大建筑建造的人。木匠和（听不清的），也就是我，是必需的，但建筑师是提供蓝图的天才。

Berkshire has become a great company with a unique group of owners. The directors of Berkshire are the trustees of the structure Charlie designed that lives beyond his lifetime and will live far beyond mine. I’m now going to introduce these directors and give you a short first quarter report and the outlook for the company. After that, we will take your questions. Please turn up the lights, and I hope you’ll join me in applauding Charlie Munger, the architect of Berkshire.

伯克希尔已经成为一家伟大的公司，拥有独特的股东群体。伯克希尔的董事们是查理设计的结构的受托人，这一结构将超越他的生命，并将远远超越我的生命。现在我将介绍这些董事，并给您一个简短的第一季度报告以及公司的前景。之后，我们将回答您的问题。请调亮灯光，我希望您能和我一起为伯克希尔的建筑师查理·芒格鼓掌。

Appendix.32.Example.02.Greg Abel

Refer To：《2024-05-04 Berkshire Hathaway Annual Meeting》。

WARREN BUFFETT: Well, that's the answer might surprise you. But they overwhelmingly, the operating executives, well they prefer to talk to Greg or to Ajit. (Laughs) And that's understandable because I don't really do much. And I don't operate at the same level of efficiency that I would have 30 years ago or 40 years ago. I don't know the managers as well as I would have when we were smaller and when I could get more accomplished in a day than I can now. And you've got when you've got somebody like Greg and Ajit, you know, why settle for me, I mean, basically? (Laughs)

WARREN BUFFETT：答案可能会让你惊讶。绝大多数运营高管更愿意和Greg或Ajit交谈。（笑）这是可以理解的，因为我实际上做的不多，我的工作效率也不如30年前或40年前了。我对管理层的了解也不如公司规模较小、我每天能完成更多事情的时候。况且，当你有Greg和Ajit这样的人，为什么还要将就找我呢？（笑）

So it's worked out extremely well. And I almost can't imagine anything working better. Because Greg, in a year, accomplishes I mean, he sees more of them, understands more about their problems, can give them suggestions. He's got incredible amounts of energy. And nobody has more wisdom than Ajit about insurance. And they've got access in insurance to him. Now, they had it before we stuck some of those titles on in insurance. Whereas with Greg, he much expanded things when he became the vice chairman in charge of really everything except insurance.

所以效果非常好，我几乎无法想象还有什么能做得更好。Greg在一年里能完成的事情——他与更多的管理层见面，更了解他们的问题，能给出建议，他有着惊人的精力。在保险方面，没有人比Ajit更有智慧，保险业务的人也能随时找到他。其实在我们给Ajit加上那些头衔之前，他们就已经能联系到他了。而Greg则不同，当他成为负责除保险以外所有业务的副董事长时，他大幅扩展了自己的职责范围。

So, he is if you polled our managers that fall under his jurisdiction, which would be a lot of them, they would much prefer it, unless, like a few, they weren't paying as much attention to their business. And I wouldn't do anything about it. But Greg would. And they still like it when he does it. He can deliver news very well to people. You know, there'll be some people, if you have 20 children and you're very rich, you'll have some that will be go-getters anyway. And you'll have some that won't. And we are a very, very rich company. And we haven't had a history of being very tough on people that coasted. And we've had some that would do that. And Greg will do something about it. And Charlie and I wouldn't have. Not because we didn't know it should be done, but because we were doing so well ourselves. (Laughs) It just wasn't something we'd make the effort for. We didn't want to change our lives that way. Plus, we slowed down in various ways physically and everything. So, I would say that the number of calls I get from managers is essentially awfully close to zero. And Greg is handling those. And I don't know quite how he does it. But we've got the right person, I can tell you that.

如果你去问那些归他管辖的管理层——人数相当多——他们会更喜欢这种安排，除非其中有少数人对自己的业务不够上心。我不会对此采取行动，但Greg会。而且他们还是欣赏他这么做的。他很善于向人们传达消息。你知道，就像如果你有20个孩子而且非常富有，有些孩子天生就会积极进取，有些则不会。我们是一家非常非常富裕的公司，过去对那些得过且过的人并没有严格追究的历史，也确实有一些人这样做了。Greg会采取行动，而Charlie和我不会——不是因为我们不知道应该这么做，而是因为我们自己做得太好了。（笑）这件事我们不会去费心。我们不想那样改变自己的生活方式，加上我们在各方面体力上都已经慢了下来。所以我会说，我接到管理层的电话数量基本上接近于零，Greg在处理这些。我不太清楚他是怎么做到的，但我可以告诉你，我们找到了合适的人。

And with Ajit, he does less physical moving. And the insurance people are more used to working with Ajit, obviously, over the years. So, I wouldn't say that changing the title really changes much there. Because he was in charge of insurance anyway. So, you can go to a business school, and they can give you way better answers than I've just given you. But that's the way we do it at Berkshire. (Laughs)

至于Ajit，他的出行相对较少，而且多年来保险业务的人显然更习惯与Ajit合作。所以我不会说头衔的改变在那里真的改变了很多，因为他本来就负责保险。你可以去商学院，他们能给你比我刚才说的好得多的答案，但这就是我们在Berkshire的做事方式。（笑）

巴菲特有些话是非常坦诚的，不去费心Greg Abel的那些事是因为“自己做得太好了”，同样一份时间，做自己的事可以创造更多的价值，Greg Abel显然不具备这个能力，没有能力一大块、一大块捡金子，一粒粒的捡也是捡，脑子差一点、手脚勤一点。

1、找到30岁的巴菲特或者乔布斯，这样的假设一开始就有问题，“自己做得很好的”为什么要去管别人的业务？一块非常大的业务，存在结构性的矛盾；

2、AJIT JAIN在性格特质上更好一些，Greg Abel年龄上更有优势，巴菲特选了后者，说明两个人在最重要的性格特质上没有足够显著的差别。

Appendix.33.Example.03.Google

Ben: Which by the way is against everything the company was founded on. The whole aim of the company and what he’s promised the team is that DeepMind is going to stay independent, do research, publish in the scientific community. We’re not going to be captured and told what to do by the whims of a capitalist institution.

Ben：顺便说一句，这其实和公司成立时的初心是相违背的。公司最初的目标、以及 Demis 对团队的承诺是：DeepMind 要保持独立，做研究，在学术界发表论文。我们不会被某个资本机构“收编”，然后按它的意志被指挥做事。

David: Yup. Definitely some deal point negotiating that has to happen with Mark and Facebook if this offer is going to come through.

David：没错。所以如果这笔交易要成，肯定得和 Mark、和 Facebook 谈很多关键条款（deal points）。

Ben: But Mark is so desperate at this point, he is open to these very large deal point negotiations, such as Yann LeCun gets to stay in New York. Yann LeCun gets to stay operating his lab at NYU. Yann LeCun is a professor. He’s flexible on some things. Turns out, Mark is not flexible on letting Demis keep control of DeepMind if he buys it. Demis argued for, we need to stay separate and carved out and we need this independent oversight board with his ability to intervene if the mission of DeepMind is no longer being followed, and Mark’s like, no. You’ll be a part of Facebook.

Ben：但 Mark 此刻太急了，所以他愿意在很多“大条款”上做让步。比如 Yann LeCun 可以继续留在 New York；Yann LeCun 可以继续在 NYU 运营他的实验室。Yann LeCun 是教授，有些事情他可以灵活。但事实证明：如果 Mark 买下 DeepMind，他并不愿意让 Demis 继续掌控 DeepMind。Demis 争取的是：我们必须保持独立、必须被“ carve out（单独剥离出来）”，而且需要一个独立监督董事会；如果 DeepMind 的使命不再被遵循，这个董事会要有介入能力。Mark 的回答是：不行。你们得成为 Facebook 的一部分。

David: And you’ll make a lot of money. As this negotiation is going on, of course the investors in DeepMind get wind of this. Elon finds out about what’s going on. He immediately calls up Demis and says, I’ll buy the company right now with Tesla stock.

David：而且你们会赚很多钱。谈判进行过程中，DeepMind 的投资人当然也闻风而动。Elon 知道发生了什么后，立刻给 Demis 打电话说：我现在就用 Tesla 股票把公司买下来。

This is late 2013, early 2014. Tesla’s market cap is about $20 billion. Tesla stock from then to today is about a 70x runup. Demis, Shane, and Mustafa are like, wow, okay. There’s a lot going on right now. But to your point, they have the same issues with Elon and Tesla that they had with Mark. Elon wants them to come in and work on autonomous driving for Tesla. They don’t want to work on autonomous driving.

这时是 2013 年末到 2014 年初。Tesla 的市值大约 200 亿美元。从那时到今天，Tesla 股价大约涨了 70 倍。Demis、Shane、Mustafa 的反应大概是：哇，好吧，现在事情太多了。但就像你说的，他们对 Elon 和 Tesla 的顾虑和对 Mark 的顾虑类似：Elon 希望他们并入后主要去做 Tesla 的自动驾驶；而他们并不想做自动驾驶。

Ben: Or at least exclusively.

Ben：或者至少，不想“只做”自动驾驶。

David: At least exclusively, yup. Then Demis gets a third call from Larry Page.

David：对，至少不想只做这个。然后 Demis 接到了第三个电话：Larry Page。

*****

David: And there’s something else very convenient about Google. They already have Brain. Larry doesn’t need Demis, Shane, Mustafa and DeepMind to come work on products within Google. Brain is already working on products within Google. Demis can really believe Larry when Larry says, nah, stay in London. Keep working on intelligence. Do what you’re doing. I don’t need you to come work on products within Google.

David：还有一个对 DeepMind 很“方便”的点：Google 已经有 Brain 了。Larry 并不需要 Demis、Shane、Mustafa 和 DeepMind 来 Google 做产品——Brain 已经在 Google 内部和产品团队合作、推动落地了。所以当 Larry 说：不，你们留在伦敦，继续研究智能，继续做你们正在做的事；我不需要你们来 Google 做产品——Demis 是真的可以相信这句话的。

Demis、LeCun都已经从认知困难走向认知障碍，无法接受现实，对于更年轻的人已经做出来的成果没办法接受，已经有充分的证据证明这一点，Meta以激烈的方式处理掉了，Google温和的方式对于维护团队是有好处的，但是对平庸的容忍是非常麻烦的，问题不是现在怎么处理，而是当初挑选的时候要做到位了。

1、Demis是Larry Page引进的，但不管怎么说，从公司整体评价，Google比Meta的做法更好，两者都做错了，但是从承担责任的角度看，Google是老板自己出来承担主要责任，没有怪罪Demis，但这不会改变Demis存在认知障碍的问题。

2、巴菲特对于Google的犹豫很可能跟创始人的部分性格有关，两个人早早就退休了，“对业务的热情，大脑对某类问题有自然的奖励”，这种态度导致了CEO和重要项目的负责人都很弱，一方面要保留对Google的控制权，另一方面又想把这个责任推掉，这个地方存在对齐上的问题。

*****

David: It’s best infrastructure in the world. We’ve got Jeff Dean here. There is nothing you could tell us that would cause us to leave Google. Except there’s one person who is intrigued, and to quote from an amazing Wired article at the time by Cade Metz who would later write Genius Makers, right?

David：这就是世界最好的基础设施。Jeff Dean 也在这儿。你说什么都不可能让我们离开 Google。——但有一个人被触动了。引用当时 Cade Metz 写的一篇精彩 Wired 文章（他后来写了《Genius Makers》那本书），对吧？

Ben: Exactly.

Ben：没错。

David: The quote is, “The trouble was so many of the people most qualified to solve these problems were already working for Google. No one at the dinner was quite sure that these thinkers could be lured into a new startup, even if Musk and Altman were behind it. But one key player was at least open to the idea of jumping ship. Then there’s a quote from that key player. ‘I felt like there were risks involved, but I also felt like it would be a very interesting thing to try.’”

David：原话是：“The trouble was so many of the people most qualified to solve these problems were already working for Google. No one at the dinner was quite sure that these thinkers could be lured into a new startup, even if Musk and Altman were behind it. But one key player was at least open to the idea of jumping ship. Then there’s a quote from that key player. ‘I felt like there were risks involved, but I also felt like it would be a very interesting thing to try.’”

（翻译：麻烦在于：最有资格解决这些问题的人里，有太多已经在 Google 工作了。即便背后站着 Musk 和 Altman，这桌上也没人敢确定能把这些思想家拉去一家新的创业公司。但有一个关键人物至少对“跳槽”这件事持开放态度。然后这位关键人物说：‘我觉得这里面有风险，但我也觉得这会是一件非常有意思、值得尝试的事。’）

Ben: It’s the most Ilya quote of all time.

Ben：这是史上最 “Ilya” 的一句话。

David: The most Ilya quote of all time because that person was Ilya Sutskever (of course) of AlexNet, DNN Research, and Google, and about to become founding chief scientist of OpenAI. The pitch that Elon and Sam are making to these researchers is let’s start a new nonprofit AI research lab where we can do all this work out in the open. You can publish.

David：之所以说最 “Ilya”，因为那个人就是 Ilya Sutskever（当然）——AlexNet、DNN Research、Google 这条线上的核心人物，并且马上要成为 OpenAI 的 founding chief scientist。这顿饭上 Elon 和 Sam 给研究员们的 pitch 是：我们搞一个新的非营利 AI 研究机构，在公开环境里做这些研究。你可以发表论文。

Ben: Free of the forces of Facebook and Google and independent of their control.

Ben：摆脱 Facebook 和 Google 的力量束缚，不受它们控制。

David: You don’t have to work on products, you can only work on research. You can publish your work. It will be open, it will be for the good of humanity. All of these incredible advances, this intelligence that we believe is to come will be for the good of everyone, not just for Google and Facebook.

David：你不需要做产品，你只需要做研究。你可以公开发表你的成果。它将是开放的、为了人类福祉的。我们相信即将到来的这些惊人的进展、这种智能，将是为了所有人的利益，而不仅仅是为了 Google 和 Facebook。

Ben: And for one of the researchers, it seemed too good to be true. They basically weren’t doing it because they didn’t think anyone else would do it. It’s an activation energy problem. By the way, Google came back with a big counter, something like double the offer. I think it was delivered from Jeff Dean personally. Ilya said, nope, I’m doing this. That was massive for getting the rest of the top researchers to go with him.

Ben：而对其中一个研究员来说，这听起来简直好到不像真的。他们之所以一直没这么做，基本原因是：他们不相信会有别人去做——这是个“激活能”（activation energy）问题。顺便说一句，Google 后来给了一个很大的反向报价，大概是把薪酬翻倍。我记得甚至是 Jeff Dean 亲自递话过去的。但 Ilya 说：不，我要去做这件事。这对拉动其他顶尖研究员跟他一起走，起了决定性的作用。

最好的人才和最好的商业模式有类似的地方：对于外部资源的依赖较少，不需要太多的资本。

1、能力越强的对外部资源的依赖越少，相比之下，Sundar Pichai、Deepmind的Demis Hassabis，Meta的Yann LeCun都是白痴，从Google走出来的这伙人做出了chatGPT，而Google如果不是Sergey Brin重新回去工作，非常麻烦；

2、巴菲特说他喜欢Lee Iacocca的传记，也肯定仔细想过这个问题，NetJets、Clayton Homes是两个非常好的例子，这两家企业如果不是BRK超长期的资金支持很难获得竞争上的优势；

3、Ilya Sutskever这些人想要不受干扰的工作，但也有问题，再是顶尖的人才，如果缺少客观的评价标准、缺少竞争的氛围也会慢慢走向平庸，chatGPT是因为Elon Musk撤资、OpenAI陷入困境后才做出来的。

Google员工的薪酬中位数比微软高67%，比美国市值最大的20家上市科技公司高153%，当有投资者（Chris Hohn，2022年）提出质疑后，Sundar Pichai就开始裁员和压缩成本，巴菲特在OWNER'S MANUAL中同样有一条奇怪的原则：不出售发展已经陷入停滞的企业，跟Chris Hohn质疑Google的做法一样，BRK的做法有没有经济效益？可能也有人质疑巴菲特，显然巴菲特不仅不担心质疑，还把这一条写进OWNER'S MANUAL，相比之下，Google对于自己的做法显然还不够自信。

David: And it was nowhere near all of the top researchers who left Google to do this, but it was enough. It was a group of seven or so researchers who left Google and joined Elon, Sam, and Greg Brockman from Stripe who came over to create OpenAI. Because that was the pitch. We’re all going to do this in the open.

David：当然，不是所有顶尖研究员都离开 Google 去做这件事，但“够了”。大概有七个左右的研究员离开 Google，加入 Elon、Sam，以及从 Stripe 过来的 Greg Brockman，一起创建了 OpenAI。因为当初的 pitch 就是：我们要在公开环境里做这件事。

I.H.04.Right People

I.H.04.Right People

1、Basic Trust/Basic Mistrust

2、Integrity/Intelligence/Energy

3、Able/Trustworthy

3、Institutional Imperative

（1）临界值很低

（2）反向校正的成本很高

Appendix

Appendix.1.《三国演义第二十一回》

Appendix.2.《1967-10-09 Warren Buffett's Letters to Limited Partners》

Appendix.3.《1982-08-12 Lee Kuan Yew.The Search for Talent》

Appendix.4.《1985-02-01 Steve Jobs.Interview with Playboy》

Appendix.5.《1990-03-02 Warren Buffett's Letters to Berkshire Shareholders》

Appendix.6.《1991 Warren Buffett.Three Lectures to Notre Dame Faculty》

Appendix.7.《1993-03-01 Warren Buffett's Letters to Berkshire Shareholders》

Appendix.8.《1994-03-01 Warren Buffett's Letters to Berkshire Shareholders》

Appendix.9.《1994-04-25 Berkshire Hathaway Annual Meeting》

Appendix.10.《1994 Warren Buffett.UNC Business School》

Appendix.11.《1995-03-07 Warren Buffett's Letters to Berkshire Shareholders》

Appendix.12.《1995-05-01 Berkshire Hathaway Annual Meeting》

Appendix.13.《1998-05-04 Berkshire Hathaway Annual Meeting》

Appendix.14.《1998-10-15 Talk to MBA Students at Florida University》

Appendix.15.《2001-04-28 Berkshire Hathaway Annual Meeting》

Appendix.16.《2002-05-04 Berkshire Hathaway Annual Meeting》

Appendix.17.《2005-02-28 Warren Buffett's Letters to Berkshire Shareholders》

Appendix.18.《2005-04-30 Berkshire Hathaway Annual Meeting》

Appendix.19.《2007-02-28 Warren Buffett's Letters to Berkshire Shareholders》

Appendix.20.《2007-05-05 Berkshire Hathaway Annual Meeting》

Appendix.21.《2008-05-03 Berkshire Hathaway Annual Meeting》

Appendix.22.《2010-05-01 Berkshire Hathaway Annual Meeting》

Appendix.23.《2010-05-26 Warren Buffett.Interview With FCIC》

Appendix.24.《2016-08-16 黄峥.PDD.劣币驱逐良币，电影院现象及其他》

Appendix.25.《2019-05-04 Berkshire Hathaway Annual Meeting》

Appendix.26.《2021-05-01 Berkshire Hathaway Annual Meeting》

Appendix.27.《2023-03-02 武志红.如何构建强大的自我》

Appendix.28.《2025-05-03 Berkshire Hathaway Annual Meeting》

Appendix.29.《2025-10-30 田渊栋.AI“顿悟”的关键，是对优雅的追求？》

Appendix.30.《2025-11-26 Ilya Sutskever.We're moving from the age of scaling to the age of research》

Appendix.31.Example.01.Charlie Munger

Appendix.32.Example.02.Greg Abel

Appendix.33.Example.03.Google

热门主题

Recent Articles

2020-12-10 王宁.潮流玩具风靡背后的心理学

2022-01-08 王宁.不做「你死我活」的生意

2026-05-13 Alex Wang.Meta's AI Chief On AI Beef, New Models And Life With Zuck

2026-05-13 泡泡玛特.2026年股东大会问答记录

2026-03-24 Thomas Peterffy.Building Interactive Brokers, Trading Discipline, and the Future of Markets