M20:49
嗯，我当然也能回答，至少不比你差。
Well, I can certainly handle it as well as you can.
他每次把球横传给我，你知道，他都把这叫作临场改战术。
Every time he laterals them off to me, you know, he calls those audibles.
这个嘛，你要把所有变量都算进去，而且要算得相当保守，但也别在每一个层面上都额外留太多余量。
The, uh, you calculate—I think you take all of the variables and calculate them reasonably conservatively, but you don't try and put too much windage in at every level.
然后等你全部算完之后，再加上安全边际。
And then when you get all through, you apply the margin of safety.
所以我的意思是，不要太执着于在每一个变量上都单独做过多保守处理，比如折现率、增长率之类的。
So I, I would say don't focus too much on taking it on each variable in terms of the discount rate and the growth rate and so on.
而是尽量让这些数字贴近现实，不过如果有误差的话，要让误差偏向保守的一边。
But try to be as realistic as you can on those numbers, uh, but with any errors being on the conservative side.
然后等你全部弄完之后，再加上安全边际。
And then when you get all through, you apply the margin of safety.
Ben Graham 有一个非常简单的公式，他在那种最明显的情况下会用，就是看营运资本，准确说是净营运资本，然后尽量用比营运资本低三分之一的价格去买。
Ben Graham had a very simple formula he used for just the most obvious situations, which was to take working capital, net working capital, and try and buy it at a third off working capital.
总体来说，这个方法是有效的。
And overall, that worked.
对他来说是这样，不过后来当那种低于净营运资本交易的股票消失之后，这个办法也就慢慢失灵了。
For him, but that method sort of ran out of steam when the sub-working-capital stocks disappeared.
不过，这和我们在保险业务里做的是一回事。
But it's the same thing we do in insurance.
我是说，如果我们想判断某项保险该收多少钱，比如说，California 发生六点零级地震的概率。
I mean, if we're trying to figure out what we should charge for, we'll just say, the chances of a 6.0 earthquake in California.
那我们知道，在上个世纪里，我记得 California 大概发生过二十六次六点零级或以上的地震。
Well, we know that in the last century, I think there have been 26 or so 6.0 or greater quakes in California.
先别去管这些地震是不是发生在人烟稀少的地方。
And let's forget about whether they occur in remote areas.
就假设我们卖的是这样一种保单：只要 California 发生六点零级或以上的地震就赔，不管它是不是发生在沙漠里、有没有造成任何损失。
Let's just say we were writing a policy that paid off on a 6.0 or greater quake in California, regardless of whether it occurred in a desert and did no damage or anything.
那我们就会看历史数据，然后说，嗯，上个世纪有二十六次，那下个世纪我们大概会假设这个数字稍微高一点。
Well, we would look at the history, and we'd say, well, there have been 26 in the last century, and we would probably assume a little higher number in the next century.
这就是我们的天性。
That'd just be our nature.
但我们不会假设是五十次。
But we wouldn't assume 50.
要是那样假设，我们就什么生意都没法做了。
If we did, we wouldn't run any business.
所以我们不会那么做——我们也许会假设稍微高一点。
So we wouldn't—we might assume a little higher.
如果是我自己来定价，我大概会说，好吧，我就假设会有三十次，或者三十二次，差不多这样。
I would, if I was pricing it myself, I'd probably say, well, I'll assume they're going to be 30 or maybe 32 or something like that.
然后等我全部算完之后，我会希望在这个定价里再加上一笔保费溢价，把安全边际也放进去。
Then when I get all through, I'll want to price the—I'll want to put a premium on it that now puts in a margin of safety.
换句话说，如果我算出来三十二这个合适的比率，那它值一百万美元。
In other words, if I figured the proper rate for 32, it's $1 million.
我大概会想要收得比一百万美元再高一点，把那个安全边际留出来；但我不想在——我是希望在每一个层面上都尽量保守，然后最后还要有一个相当可观的安全边际。
I would probably want to charge something more than $1 million to build in that margin of safety, but I don't want to get it at— I want to be conservative at all the levels, and then I want to have that significant margin of safety at the end.
我想，如果我对这个问题的理解没错的话，这就是我的回答。
And I guess that, as I understand the question, that'd be my answer.
Charlie，你要不要再补充一点？
And Charlie, you want to add to that?