← 返回任务列表

2003年股东大会 · 巴菲特对股票投资的最低回报要求是多少?

71 段 · 3 位说话人 · 原片 9:02
M1
M10:00

下午好。

Good afternoon.

我是来自 Virginia 州 Arlington 的 Patrick Wolff。

I'm Patrick Wolff from Arlington, Virginia.

Charlie,我实在忍不住要告诉你,我其实就是那个蒙着眼下棋的人。

Charlie, I can't resist telling you that I'm actually the fellow who plays the chess games blindfolded.
M2
M20:08

对。

Yeah.
M1
M10:08

所以我很期待明天在那里“看不见”你。

So I look forward to not seeing you there tomorrow.
M2
M20:12

没错。

Right.
M1
M10:12

我其实有个两部分的问题。

I actually have a 2-part question.

我想请你再展开讲一讲,你是怎么思考机会成本的。然后我——我想我接下来会在很大程度上顺着刚才最后那个问题继续问。

I'd like to ask you to elaborate a bit how you think about opportunity cost, and I'm— I think I'm going to be elaborating very much on the very last question that was asked.

首先,在年报里,你明确说过,你在看普通股投资的时候,会寻找税前百分之十的净资产回报率。

First of all, in the annual report, you say explicitly that you look for 10% pre-tax return on equity in looking at common stocks.

我记得你之前还谈到过,你会在这个基础上推导出税后百分之五到百分之六的回报,然后再把通胀和税收一层层加上去。

And I think you talked earlier about how you build up from that for 5% to 6% after-tax return, and then you layer on inflation and then layer on taxes.

我的第一个问题是,你会怎么根据不同时期来调整这个要求的回报率?

My first question would be, how do you adjust that required rate of return across periods of time?

比如说,当利率更高的时候,你会不会在不同时间阶段要求不一样的 equity premium return?

So, for example, when interest rates are higher, do you look for a different equity premium return over different periods of time?

我的第二个问题是,你刚才说过,你实际上会对不同股票使用同样的 discount rate。我相信你也知道,现代金融理论其实认为你不该这么做,它认为你应该考虑现金流发生的时间,尤其是它和整体市场之间的 covariance。

My second question is, you just said that you actually would apply the same discount rate across the stocks, and I'm sure you know that modern finance actually suggests that you should not do that, that you should be thinking about the timing of cash flows and in particular the covariance with the general market.

你一直很强调一点,就是当你思考风险的时候,你主要是从这样一个角度来看:随着时间推移,你最终能不能拿到你原先预测会拿到的那些现金流?

Now, you've made a point of emphasizing that when you think of risk, you think of risk primarily in terms of will you get the cash flows that you predict you will get over time?

如果用 discounted cash flow 的框架来说,这有点像是分子端的风险。我想在场所有人都得承认——你的业绩已经说明一切了——这大概一直都是一种非常有效的风险思考方式。

Sort of numerator risk, if you think in terms of discounted cash flow, which I think everyone here will have to acknowledge— your results speak for themselves— has probably been a very effective way of thinking about risk.

但是,从真正的经济成本来看,现金流发生时间本身也同样值得考虑。这个成本也许小得多,但它依然是真实存在的成本。

But there is a true economic cost to think about the timing of cash flows as well, and it may be a much smaller cost, but it is still a real cost.

比如说,我可以请你设想一个人在两份工作之间做选择。

I might, for example, suggest you think about somebody deciding between 2 jobs.

这两份工作完全一样,而且这个人预计从每份工作里赚到的钱也一样多,但有一个区别。

The jobs are completely identical, and the person expects to make the same amount of money from each job, but there's one difference.

区别在于,一份工作会在经济很差的时候给他付更多钱,而另一份工作会在经济特别火热的时候给他付更多钱。

And the difference is one job will pay him more when the economy's in the tank, and the other job will pay him more when the economy is going gangbusters.

那么,如果他问你哪份工作实际上更值钱,我猜你的回答会是:那份在经济低迷时付他更多钱的工作更值钱。

Now, if he asked you which job was actually worth more, my guess is you would tell him that the one that would pay him more when the economy's in the tank.

原因是,如果他想靠兼职或者做点别的事情来多赚些钱,那么在经济好的时候,这会容易得多。

And the reason is, if he wanted to make more money by moonlighting or doing something else, it'd be much easier when the economy is doing better.

这其实就是你要去看现金流进入的时点和整体市场之间 covariance 的核心逻辑。

That's the essential logic behind the idea that you look at the covariance of when cash flows come in with the overall market.

这是一种真实的成本,虽然它很难衡量;而且即便它比所谓分子端风险——也就是真正拿到那些现金流的风险——要小一些,它毕竟还是实实在在的成本,所以我想你一定也会考虑它。

It's a real cost, even though it is difficult to measure, and even if it is a smaller risk than numerator risk, the risk of getting the actual cash flows, since it's a real cost, I imagine you must think about it.

所以,我的第二个问题就是,你是怎么考虑这个问题的?如果你决定不考虑,原因又是什么?

And so my second question to you would be, how do you think about it, and if you decide not to, why?
M3
M32:45

那么,首先我想说,Patrick,我们很感谢你专程过来,因为 Patrick——具体多少年了我现在说不准,但已经有好些年了——每到星期天都会自愿来参加对弈。

Now, first of all, I would like to say, Patrick, we appreciate you coming out because Patrick— now, I don't know how many years it's been, but it's been a number— has volunteered on Sunday to play.

我想他现在大概是在同时蒙眼跟六个人左右下棋,听完刚才那个问题,你也就能明白他为什么能做到这一点了。

Now, I think he's playing 6 people or so blindfolded simultaneously, and after hearing that question, you can understand how he does it.

Patrick,还有 Bob Hamman,以及今年的 Peter Morris 和 Bill Robertie,都会专程过来。

Patrick, as well as Bob Hamman, and then this year Peter Morris and Bill Robertie, all come out.

到了星期天,他们——我们这边就有这些非凡的人才在现场。

And on Sunday, they— we've got these extraordinary talents out there.

对于那些喜欢他们所玩的各种游戏的人来说,他们会专门拿出一个下午的时间来陪大家,而且完全不求任何回报。

And for people who like the various games they play, they devote an afternoon for it and ask nothing in return.

所以,Patrick,我们真的非常感谢你。

So we really appreciate it, Patrick.

我很期待看看你明天会不会从眼罩后面偷看。

And I'll look forward to seeing if you're peeking out of your blindfold tomorrow.

关于机会成本,还有我们提到的百分之十,你知道,基本上那就是我们划的线。

The question on opportunity costs and the 10% we mentioned, you know, basically that's the figure we quit on.

也就是说,在买入这件事上——如果我们实际预期回报低于百分之十,我们就不想买股票。

And we quit on buying— we don't want to buy equities where our real expectancy is below 10%.

这一点不管短期利率是百分之六,还是短期利率是百分之一,都一样。

Now, that's true whether short rates are 6% or whether short rates are 1%.

我们只是觉得,

We just feel that it would get very sloppy.

一旦开始往这个标准下面滑,就会变得很松。

To start dipping below that.

另外,我们也觉得,很明显,我们将来还是会遇到至少达到这个水平、甚至可能明显高于这个水平的机会。

And we would add, we feel also obviously that we will get opportunities that are at least at that level and perhaps substantially above.

所以到了某个点,我们就退出这个游戏了。

So there's just a point at which we drop out of the game.

这个点是人为定的。

And it's arbitrary.

没有什么——我们没有做过什么科学研究之类的。

There's no— we have no scientific studies or anything.

但我敢跟你打赌,很多年以后,我们或者你,还是能找到你能看懂、或者我们能看懂,而且有百分之十或更高回报概率的股票。

But I will bet you that a lot of years in the future, we or you will be able to find equities that you understand or we understand and that have a probability of returns at 10% or greater.

一旦你找到一批处在这个范围内的股票,先不考虑我们这种资金规模特别大的问题,那我们就只买最有吸引力的。

Now, once you find a group of equities in that range, and leaving aside the problem of huge sums of money which we have, then we just buy the most attractive.

通常这就意味着,买那些我们把握最大的。

That usually means the ones we feel the surest about.

我是说,现实里确实有些企业,它们具备某些经济特征,会让它们未来前景比别的企业更容易预测得多。

I mean, as a practical matter, there's just some businesses that possess economic characteristics that make their future prospects far more predictable than others.

也有各种各样的企业,你根本没法预测它们能赚多少钱,那你就只能把它们忘掉。

There's all kinds of businesses that you just can't remotely predict what they'll earn, and you just have to forget about them.

所以,随着时间过去,我们越来越偏爱那些我们觉得可预测性很高的企业。

But when we get— so we have over time gotten very partial to the businesses where we think the predictability is high.

但即便这样,我们还是希望门槛回报率是百分之十,反正税后来看,这也不算多高。

But we still want a threshold return of 10%, which is not that great after tax anyway.

Charlie,你要不要先就这个——这个问题的这一部分说几句?

Charlie, do you want to comment on that portion of the— that question first?
M2
M25:36

对,我觉得说到底,我们做的每件事最终都要回到机会成本。

Yeah, the— I think in the last analysis, everything we do comes back to opportunity cost.

但在某种程度上,其实是相当大的程度上,我们是在猜自己未来的机会成本。

But to some extent, in fact, to some considerable extent, we are guessing at our future opportunity cost.

Warren 基本上是在说,他判断今后总会有机会把钱投出去,拿到相当不错的回报率,所以他现在不打算把太多火力浪费在那些较低回报的项目上。

Warren is basically saying that he's guessing that it's in due course to put out money at pretty attractive rates of return, and therefore he's not going to waste a lot of firepower now at lower returns.

但这本质上就是一种机会成本的

But that's an opportunity cost
M3
M36:13

计算。

calculation.
M2
M26:13

如果利率差不多永久地稳定在百分之一,或者类似那样的水平,而 Warren 重新评估了他对未来机会成本的看法,他就会调整这些数字。

And if interest rates were to more or less permanently settle at 1% or something like that, and Warren were to reappraise his notions of future opportunity cost, he would change the numbers.

就像 Keynes 说的,当你对事实的看法变了,你会怎么做?

It's like Keynes said, what do you do when you change your view of the facts?

那你就改变你的行为。

Well, you change your conduct.

但至少到目前为止,我们心里是有门槛的,而这些门槛本质上——嗯,里面其实隐含了对未来机会成本的判断。

But so far, at least, we have hurdles in our mind which are basically— well, they involve implicitly future opportunity cost.
M3
M36:45

现在我们这 一百六十亿美元,税前收益率是百分之一点二五,这样一年就是两亿美元。

Right now, with our $16 billion that's getting 1.25% pretax, that's $200 million a year.

我们完全可以很轻松地去买二十年后到期的 government bonds,拿到大概百分之五的收益。

We could very easily buy governments due in 20 years and get roughly 5%.

这样一来,我们每年两亿美元的收入,就可以变成八亿美元。

So we could change that $200 million a year to $800 million a year of income.

正像 Charlie 说的,我们是在做一个决定:我们宁可暂时先拿着两亿美元,基于这样一种想法——将来我们会找到一个能给我们百分之十或者更高回报的东西——也不愿意现在就把每年八亿美元投进去,结果过了一年左右,更好的机会来了,却发现我们之前投进去的东西本金已经出现了很大的亏损。

And we're making a decision, as Charlie says, that it's better to take $200 million for a while on the theory that we'll find something that gives us 10% or better than to commit to the $800 million a year, and then find that in a year or thereabouts when the better opportunities came along, that what we had committed to had a big principal loss in it.

不过这个嘛,你知道,这并不是——并不是特别科学,但我能告诉你的只是,从实际操作来看,这似乎相当管用。

But that's, you know, that's not— it's not terribly scientific, but it— all I can tell you is in practice it seems to work pretty well.
M2
M27:35

很多年前,Warren 管合伙企业的时候,还有某种程度上我管的那个合伙企业,运作方式也差不多,我们其实就在暗中做你说的这种事,也就是把一部分合伙资金放进所谓的 event arbitrage 投资里。

And years ago, when Warren ran a partnership, and to some extent the partnership that I ran was the— operated in the same way, we implicitly did what you're suggesting, in that part of the partnership funds were in so-called event arbitrage investments.

而这类投资有时候会在整个市场普遍很糟糕的时候,反而带来回报。

And those tended to generate returns occasionally when the market generally was in the tank.

而其他类型的投资,表现就会更像整个大盘。

And alternative investments would more mimic the general market.

所以我们其实在四十年前就在做这种学院派理论所主张的事了,呃,只不过我们那时候没用现在这些时髦术语。

So we were doing what this academic theory prescribes, you know, 40 years ago, and, uh, but we didn't use the modern lingo.
M3
M38:21

对,我们确实会偏好让很多钱持续不断地流进来,不过呢,我们也会做那种波动性非常大的保险交易,这就意味着可能会有一大笔钱在某一个时点,或者很短时间内流出去。

Yeah, we've got some preferences for having a lot of money coming in all the time, but, well, we do go into insurance transactions with huge volatility, which could mean that a big chunk of money could go out at one time or in a very short period of time.

但我们不会为了平稳性,就放弃很多本来可以预期得到的回报。

And we won't give up a lot in expectable return for smoothness.

但是如果让我们选,一种是每周都有钱进来,另一种是同样现值的钱以非常不均匀、而且我们事先也不知道什么时候会来的方式进来,那我们会选更平稳的那种。

But if you give us a choice of having money come in every week and the same present value of money coming in in very lumpy ways that we wouldn't know about, we would choose the smooth.

但如果你给我们的选择是,作为这种不均匀性的补偿,现值更高一些,那我们就会选这种不均匀的。

But if you give us a choice of a higher present value for the lumpiness, we will take the lumpiness.

而这通常就是那种——我是说,这通常就是——别人摆在我们面前的选择。

And that's usually the choice that's— I mean, that's usually— we get offered that choice.