TIP776：斯蒂格·布罗德森的心智模型与投资组合更新

十年前，我无法代表你克莱回话，但我当时确实认为亚马逊已经是个非常成熟的玩家了。

A decade ago, I can't speak for you Clay back, I certainly thought of Amazon as a very established player.

然后再想想你今天使用它的频率。

And then think about how much you use it today.

所以每当谈到优步时，我认为我们还处于非常早期的阶段，未来优步在按需匹配人员和货物交付方面会做得更好，选择范围也会更广。

And so whenever it comes to Uber, I think we're just in the very early stages and Uber will in the future be even better at matching the delivery of people and goods at your convenience with a much wider selection.

那么为什么要在出行和配送领域同时进军新市场呢？

So why there's certain entering new markets both with mobility and delivery?

我认为使用频率将成为主要的增长杠杆。

It's just I think frequency is going to be the major growth lever.

克莱：是的。

Clay Yeah.

就像我提到优步时说的，我们每个人对这只股票都有自己的偏见。

As I mentioned with Uber, we all come to this stock with our own biases.

我目前在美国，而美国现在仍是他们业务的主要市场。

I'm based in The US and The US is a major part of their business today.

你住在丹麦，有趣的是你其实没有车。

You're based in Denmark and interestingly, you actually don't own a car.

我不确定你使用优步的频率，是每天、每周一次还是不用。

I'm not sure how often you use Uber, whether it's every day, once a week or not.

你的使用情况是怎样的？

What's the use look like for you?

我只是好奇。

I'm just curious.

斯蒂格 是啊。

Stig Yeah.

我想我大概每月用一两次，但我觉得自己不是最佳代表。

I think I probably use once or twice a month, but I don't think I'm the best representative.

我有点晕车。

I have a bit of motion sickness.

我不介意步行一小时。

I don't mind walking for an hour.

我知道对很多人来说这听起来很荒谬，但我真的很喜欢有声书和播客。

And I know to a lot of people it probably sounds absolutely ridiculous, but I love audiobooks, I love podcasts.

如果我要去某个地方，距离大概五公里左右，我就会步行，然后边走边听，比如听你或别人的节目。

If I'm going somewhere and it's five kilometers away or whatnot, I would just walk and then listen to, I don't know, you or someone else.

特别是在城市地区，这里没有车其实挺常见的，这有点像...

I don't know, but it is sort of like, especially if you live in a city area, it's not too uncommon not to have a car here.

某种程度上这就是这里的常态。

It's kind of just the landscape.

想想在美国的时候，我能理解为什么很多地方没有车会很不安全，因为那里甚至没有人行道。

Think for whenever I'm in The States, can see many places why it would be, I don't know, outright dangerous because there are no sidewalks or whatnot.

一切都显得有点不同。

Everything just looks a little bit different.

你来过克莱，这里的情况有点不一样。

You've visited Clay, it's a little bit different here.

我没有车，但用优步让我非常满意。

I don't have a car and I'm perfectly happy with Uber.

所以这就是我的出行方式吧。

So that's my mode of transportation, I guess.

确实挺有意思的。

Trey It is certainly interesting.

你住的城市步行便利性很高，可以轻松步行去买咖啡、采购日用品。

The city you live in is very walkable, so you could easily go walk and get coffee, go walk and get groceries.

这根本不算什么事。

It's not really a big deal.

在我居住的内布拉斯加这边，有车是必备条件。

Where I'm based in Nebraska here, having a car is a non negotiable.

要是没车的话，那简直不可理喻。

If you don't have a car, you're kind of crazy.

但看看优步，投资者很快发现大城市占据了其大部分订单量。

But when you look at Uber, investors quickly find that the biggest cities capture a lot of these Uber rides.

比如如果我住在伦敦，养车可能会很头疼，完全能理解为什么这些城市的居民会用优步通勤或点优食外卖午餐。

If I lived in London, example, owning a car would probably be a big headache and I could see why in many of these cities people would use Uber to commute to work or use Uber Eats to get lunch and whatnot.

像分析这类股票时总是存在许多细微差别。

There's a lot of nuance in looking at a stock like this as always.

在财报电话会议上，优步管理层也谈到他们持续通过再投资和创新来扩展用户服务，这就引出了优步会员计划。

And on earnings calls, Uber's management team has also talked about their continued reinvestment and innovation to expand their offering to users, which brings me to mention Uber One.

优步会员是他们的付费订阅服务，整合了打车和外卖的双重福利。

So Uber One's their paid subscription service that combines the benefits for both rides and deliveries.

标准会员费为每月9.99美元或每年96美元。

A standard membership is $9.99 per month or $96 per year.

许多大型科技公司都在转向订阅模式。

And many of these big tech companies are turning into the subscription model.

比如亚马逊、网飞、YouTube这些公司。

You look at Amazon, Netflix, YouTube, for example.

优步会员可享受打车折扣和返现、符合条件的外卖订单免配送费、优先服务以及更多专属特权。

With Uber One, members get discounts and credits on rides, dollars 0 delivery fees on eligible Uber Eats orders, priority service, and just more exclusive perks.

对于经常使用叫车服务的人来说，购买这个会员资格显然是明智之选。

And for those who utilize their ride hailing service often, it just seems like a no brainer to pay for this membership.

对优步来说，这确实能锁定那些可能会不时尝试其他应用（如Lyft）的客户，并激励他们使用已付费的服务，就像我们因为支付了月费或年费而更倾向于在亚马逊购物或去Costco消费一样。

For Uber, it really allows them to lock in customers who might otherwise explore different apps such as Lyft from time to time and also incentivize them to utilize the offerings that they're paying for, similar to how we're all incentivized to order off Amazon or visit Costco because of the monthly or annual fee we're paying.

在优步1.89亿月活跃用户中，估计约15%至20%是优步One会员。

Of their 189,000,000 monthly active users, it's estimated around 15% to 20% of customers are Uber One members.

优步One会员的消费金额大约是非会员的3.5倍。

An Uber One member spends around three and a half times more money than non Uber One members.

我认为他们在将许多用户转化为优步One会员方面做得相当不错。

I think they're doing a pretty good job of converting many of their users to Uber One.

我前几天刚预约了优步，在叫车时他们提供了6%返现和优步One免费试用，基本上就是在邀请我加入。

I actually booked an Uber just the other day and as I was booking the ride, they offered 6% cash back with a free trial of Uber One, asking me to join essentially.

你对优步这个优步One会员计划怎么看？

What do you think of this membership program of Uber One?

斯蒂格：不，我超爱这个计划，克莱。

Stig No, I love it, Clay.

另外我想正式声明一下，我现在是居家办公状态。

And I also just want to say for the record that I'm work from home now.

基于这个原因以及其他诸多因素，我工作时不会选择乘坐优步。

For that reason and so many other reasons, I don't take an Uber for work.

但实际上我以前上班大多时候都是打出租车的。

But I actually used to take a cab for work most days.

这可能听起来有些奢侈，但我觉得这也取决于你住在哪里、税收情况以及其他一系列因素。

And it probably sounds extravagant to some, but I think it also really depends on where you live and taxes and a bunch of other stuff.

在某些城市，买个停车位简直就像买套公寓一样。

And certainly in some cities, buying a parking spot, it's almost like buying a condo.

价格贵得离谱，简直难以置信。

It's just so ridiculously expensive that it's just unbelievable.

实际上有段时间——虽然严格来说不是优步，但效果和优步差不多——我基本上是用订阅服务接送上下班的。

So there was actually a time whenever It wasn't Uber for the record, but it might as well have been Uber, where I more or less used a subscription service to bring me back from work.

不过话说回来，我觉得优步One会员计划很棒。

But anyways, I think Uber One is wonderful.

总的来说，他们为建立经常性收入和提高使用频率所做的一切都非常合理。

And I would generally say that everything they do to have a somewhat recurring revenue and increased frequency of use, it just makes a lot of sense.

我知道这可能不是我所能提供的最详细的反馈。

I know that's probably not the most detailed feedback I can give.

我会尽量说得更具体些，但有时候当你审视一个企业并退一步思考时，'这真的合理吗？'这种想法本身就蕴含着强大的力量。

I'll try to be a bit more specific, but sometimes whenever you're looking at a business and you take a step back, there's something very thirty:fifty powerful about thinking, Does

这真的合理吗？

this really make sense?

作为用户和股东，你能说'这确实合理'吗？

Can you as a user and shareholder say, This makes sense?

在我们开始录音前，克莱问我最近怎么样，当然我的生活一如既往地平淡无奇。

So before we hit record, Clay was asking me what was going on, and of course nothing is going on in my life.

倒是有个TIP团队的成员这周末要来看我。

I do have someone from the TIP team that's visiting me here over the weekend.

所以我给她注册了绑定在我账户下的优步账号。

And so I set her up with an Uber account, so it's tied to my account.

她在城市里出行时，就可以直接叫优步了。

Whenever she's navigating here in the city, she would do an Uber.

我们团队的卡米尔在纽约和你们一起办活动时，也是类似安排——她只需预订优步，然后走公司账户报销。

Camille from my team, whenever she was doing the event together with you guys in New York City, she was on a similar arrangement where she would just book an Uber and then she would just go over the company account.

所以我知道企业订阅服务并非优步首创。

So I know that having a company subscription, Uber didn't invent that.

这不是我想表达的重点。

So that's not what I'm saying.

我想说的是优步确实做得非常出色，

What I'm saying is that Uber is doing a really thirty:fifty good

无论是通过Uber One会员计划还是其他方式，都能让你以最合适的方式轻松使用他们的服务。

job, whether it's in the Uber One membership or not, to make it very easy for you to use their service in the way that's most suitable for you.

而且我能清晰看到他们如何区分不同服务场景。

And it's very easy for me to see how they discriminate under different service.

这里的区分不含任何负面意味，而是非常直观——比如去机场时你很可能需要预约功能。

And I don't mean discriminate like any kind of negative way, but it's so intuitive in terms of, okay, you're going to the airport, you probably should have the reserve function.

现在推出的团队功能是否契合你当前的生活场景需求？

Does it make sense for you in whatever kind of life situation you have that you now have the team function?

对你来说，共享功能是否更有意义？

Does it more make sense for you to have the share?

所有这些都非常合理。

All of it just makes a lot of sense.

Uber One会员也非常合理。

And Uber One also makes a lot of sense.

它的年化收入超过10亿美元，同比增长超过60%。

It's more than a billion dollar run rate and more than a 60% year on year growth.

有趣的是你把它和Costco会员进行了比较。

And it's interesting that you compared it to Costco's membership.

我认为这是个很好的观点。

I think that's a good point.

会员业务的利润很高，但正如你之前所说，其中部分会来自核心产品的利润，比如那6%的返现。

The margins of the memberships are great, but then also to what you said before, some of that would come out of the margins of the core product, let's say the 6% cash back there.

但无论如何，他们在提升使用频率方面做得非常出色。

But anyways, they're really, really good at driving that frequency.

这对送餐和叫车服务来说都是重大胜利。

And it's just a major win both for delivery and for ride hailing.

我不知道是不是出于个人偏见，但我们经常使用送餐服务。

I don't know if I'm just talking out my own biases, but we use delivery all the time.

也不知道是不是只有我这样，我总想着要多在家做饭，结果却总是忍不住点外卖。

And I don't know if it's just me, but I have all these intentions of I'm going to do more home cooked meals and then I always find myself ordering in.

在提升使用频率方面效果非常显著。

It's just very powerful in terms of increasing frequency.

这有点像...不知道这个比喻是否恰当，就像我过去办的健身卡，付了钱却从不去用。

It's a bit like I don't know if I should even make this comparison, but the gym membership, I used to pay for it, but I never used.

总之，Uber One会员计划历来侧重于送餐服务，但现在我们看到它正更多地转向叫车出行领域。

But anyways, the membership program, Uber One, has historically been geared towards delivery, but we see it now lean more into ride hailing or the mobility segment.

我认为这是非常强大的组合。

I think that's a very powerful combination.

早在2018年，CEO就说过叫车服务对Uber而言就像图书之于亚马逊。

Back in 2018, the CEO said that ride hailing was like books for Amazon.

他将Uber的雄心描述为打造'万物运输商店'。

He described Uber's ambition as being the everything transportation store.

这是类比亚马逊作为'万物商店'的说法。

So this was the analogy to Amazon being the everything store.

我想说，即便作为股东，我也不完全认同他的热情，但大体方向我是一致的。

I'd say that even as a shareholder, I don't think I completely share his enthusiasm, but I think I'm directionally in the same camp.

我确实认为Uber可以拓展新业务领域，但将其比作亚马逊Prime可能有些牵强。

I do think that Uber can include new verticals, but comparing it to Amazon Prime is probably a bit of a stretch.

我猜他是在为自己造势，这无可厚非。

I guess he's talking his own book and I don't fault him for that.

Stig 另外我要说的是，虽然我之前对Uber One有些批评，但现在看来它确实有其合理性。

Stig And then the other thing I will say, because I think I was doing a bit of a knock there on Uber One previously, I think that it also makes sense.

他们建立了这些合作关系，虽然未公开具体成果，但目前确实在搭建大量合作。

You have these partnerships and they don't really disclose the results, but they're setting up a lot of partnerships right now.

我们提到达美航空，克鲁格也刚合作不久——这些案例都历历在目。

We talk about Delta, Kruger comes to mind, was not too long ago.

所以Stig，这3000万会员并非都是主动选择付费加入的。

And so Stig it's not 30,000,000 members who deliberately chose to pay for that membership.

其中很多会员是通过那些合作渠道被动获得的，无论他们是否想要。

A lot of that is baked into those partnerships where people are getting it, whether they want it or not.

他们的数据中并未明确说明是否包含这些合作渠道的会员。

It's not disclosed whether or not in their numbers that includes those memberships.

我强烈怀疑他们确实计入了这些数据，毕竟他们总是往对自己有利的方向说。

I strongly expect that they do because they're talking their book.

这就是那种需要我们现在开始深入研究的细节问题。

It's one of those things where, okay, we just now start to look more into details like that.

Clay，关于Uber最让我着迷且意外的是它竟然还有广告业务。

Clay What also really intrigued me about Uber and really caught me by surprise is that this is an advertising business too.

他们的广告业务部门创造了超过10亿美元的收入。

They have an advertising segment that generates north of $1,000,000,000 in revenue.

如果我们再次将Uber与亚马逊类比，亚马逊零售业务可能几乎不赚钱，但依靠Prime会员和广告等业务依然能获得巨额利润。

And if we correlate Uber to Amazon again, Amazon can make little to no money from their retail business, but still be hugely profitable thanks to things like their membership of Amazon Prime or advertising.

我不禁也看到了Uber的相似之处，这充分展示了像Uber这样的公司所具备的选择权优势——你拥有平台所有权，还能通过其他渠道实现平台变现。

And I just can't help but see the similarities here with Uber as well, which really illustrates the power of optionality with a company like Uber where you do own the platform and you have these other avenues of monetizing that platform.

凭借他们的Uber One会员计划，如果他们未来能持续开发新的业务板块，我一点都不会感到惊讶。

And with their Uber One membership, I wouldn't be surprised if they can continue to spin up new ways to create new business segments in the future.

斯蒂格 没错，这确实是投资逻辑的一部分。

Stig Yeah, that is certainly a part of the thesis.

你在Spotify身上也能看到这种情况。

You see that with Spotify too.

他们最初只做音乐，后来增加了播客，最近又加入了有声读物。

They just started with music and then they added podcasting and then audiobooks here more recently.

你可能会预期会有一些内部竞争，但实际上你只会看到每个用户的使用率和参与度更高。

So you might expect some cannibalization, but really you just see more usage and engagement per user.

因此整体上其实是增长的。

So that has actually grown in total.

Stig，我认为Uber在这方面会做得比Spotify更好。

Stig I think Uber would do it even better than Spotify.

网约车和外卖业务之间的蚕食风险确实较小，我猜。

Ride hailing delivery, the cannibalization risk is just less, I guess.

而且它们彼此互补性很强，能很好地强化Uber的核心服务。

And also they complement each other really well and really lean into Uber's core service.

单是解决最后一公里配送问题，就比表面看起来复杂得多。

And just matching on that last mile is just much more complicated than what it looks.

但正如你所说，无论是亚马逊、Spotify还是Uber，企业真正需要的是让用户持续使用平台。

But really to your point, what you want, whether you're Amazon, you're Spotify, Uber, is that you want people to engage on your platform.

一旦用户使用平台，你就能收集数据并实现变现。

And then once you use the platform, you can collect data and you can monetize them.

举个例子，这是个全新的业务单元。

So one example that comes to mind is this is a brand new business unit.

目前盈利很少，我们静观其变吧。

They make very little money, so let's see what happens.

但Uber正将自己定位为零工经济工作者的平台。

But Uber are seeing themselves as a platform for gig workers.

所以当你是一名司机时，假设你此刻没有在做叫车或送货服务，你可以在等待时用手机做些其他事情，比如AI标注之类的。

So whenever you're a driver and let's say you're not doing ride hailing or delivery that second, you can do something else like AI labeling or whatever on your phone while you're waiting.

你总能赚到钱。

You can always make money.

在消费者端，Uber掌握着大量关于你的数据，可以用来投放定向广告。

Stig On the consumer side, Uber just has so much data in you that they can use to deliver targeted advertising.

目前这是一项十亿美元营收的业务，但正以30%到50%的速度快速增长。

And right now it's a billion dollar revenue business, but it's growing thirty:fifty fast.

虽然看起来这只是整体业务中的一小部分，但我们仍处于非常早期的阶段，比网约车和配送业务的成型期要早得多——当然，我承认自己有偏见，我认为这些领域也仍处于初期。

And so it might seem like it's a small part of the overall business, but again, we're very much in the early innings and way earlier than formability and delivery in general, which again, I'm biased, I also think we're in the early innings of that.

克莱，给你一些具体数字：配送业务占1.6%，而整体来看仅占优步总预订量的0.9%。

And so just to throw some numbers at you, Clay, for delivery, they make 1.6% and then overall for Uber's business, it's 0.9% of gross booking.

他们的目标是整体占比达到2%。

And they want to get to 2% overall.

让我们拭目以待吧。

Let's see if that happens.

需要记住的是，广告业务通常能提供70%到80%的利润率。

Keep in mind that advertising typically offers margins of 70% to 80%.

所以一美元实际上不止一美元。

So a dollar is not really a dollar.

为了给你一个对比参考，配送业务的抽成比例是订单总额的30%到20%。

And just to give you something to compare it to, from ability, there's a take rate of gross booking of 3020% on delivery.

克莱，我们来谈谈你所说的优步房间里的大象——自动驾驶车辆。

Clay Let's talk about what you refer to as the elephant in the room for Uber, its autonomous vehicles.

我想和你进行一个思维实验，这个实验不是为了证明特斯拉或自动驾驶的看涨观点，也不是为了看空优步，更多是做一个目的地分析练习。

I wanted to present a thought experiment to you that isn't meant to serve as the bull case for Tesla or AVs or the bear case for Uber, but more so an exercise of destination analysis here.

如你所知，埃隆·马斯克做过无数预测，有些比其他预测准确得多。

Elon Musk, as you know, has made countless predictions, some much more accurate than others.

不管他做过什么预测，在我看来他在车辆生产方面取得的成就相当惊人。

And regardless of the predictions he's made, what he's done in terms of producing vehicles to me is pretty amazing.

从2013年到2023年这十年间，他们的车辆总交付量从2万辆增长到了180万辆。

So over a ten year period from 2013 to 2023, their total vehicle deliveries went from 20,000 to 1,800,000.

所以我认为说他们未来有可能推出大量自动驾驶出租车并非夸大其词，不一定是在2026年，也许是在2030年或更晚。

So I don't think it's a stretch to say that there's at least a possibility of them rolling out a substantial number of robo taxis in the future, not necessarily in 2026, perhaps in 2030 or later.

马斯克的万亿美元薪酬方案中，部分激励结构就是要生产超过百万辆自动驾驶出租车。

Musk's trillion dollar pay package, part of the incentive structure is producing a million plus robo taxis.

如果特斯拉或任何公司真的推出大量这种自动驾驶出租车，并以比Uber更低的成本提供相同的出行服务，或者有迹象表明某家公司即将做到这一点，你会如何看待对Uber的投资及其业务影响？

And if Tesla or any company for that matter were rolling out many of these robo taxis and providing the mobility service that Uber is providing and doing so at a much lower cost, or it became clear that a player was on their way to doing that, how would you think about your investment in Uber and how it's going to impact their business?

斯蒂格 是的。

Stig Yeah.

这确实是个风险因素，希望我没给人留下我不重视这个问题的印象。

It's certainly a risk factor and I hope it didn't come across like I don't think it is.

可能在我看来这个风险比很多投资者认为的要小些，但谁知道呢？

Perhaps in my book, it's a little bit smaller than a lot of investors look, but who knows?

我想首先要做的是对这个前提提出一些质疑。

I think the first thing I want to do is to challenge the premise a bit.

每当我展望2030年甚至更远的未来时，我认为对Uber来说配送业务将比网约车更重要。

Whenever I look at 2030 and certainly further into the future, I think delivery will be more important than ride hailing for Uber.

这只是其中一点。

That's just one thing.

配送服务比叫车服务更难自动化，因为人就是人。

Delivery is so much harder to automate than ride hailing because people are, well, people.

他们有眼睛，可以四处走动，走进建筑物又走出来。

They have eyes and they can move around and walk into a building and walk out again.

其中有些事情对机器人来说确实很困难。

Some of those things are just really hard for robots.

除了像某些大学校园这样的小范围区域外，各种货物的配送自动化难度极高。

Aside from some minor pockets such as select college campuses, it's just very hard to automate delivery of various goods.

Stig 所以如果你看看从亚洲到北美的集装箱船运输，按每集装箱每英里计算非常经济。

Stig And so if you look at, let's say, shipment on a container ship from Asia to North America, It's very economical to do that per container, per mile.

从餐厅取餐、开门、跨过门槛、引起餐厅员工注意、确保拿到正确的食物、再送到两英里外的住宅——这些对人类来说很基础的30:50操作，要完全自动化却出奇复杂且难以经济高效地实现。

Really picking up food from a restaurant and then opening the door, stepping over the doorstep, getting the attention of the staff of the restaurant, ensuring you have the right food, dropping off at a house two miles away, pretty basic thirty:fifty stuff for a It is surprisingly complicated to fully automate and it's very difficult to do it economically.

所以世界上最笨的披萨公司已经证明其中部分环节可以实现自动化。

So you have the dumbest pizzas of the world who have shown that some of that can be automated.

这在更大范围内就是无利可图的。

It's just not profitable on a wider scale.

实现这一点非常、非常复杂。

It's very, very complicated to do that.

配送是个极其丰富的话题，你越深入探究，就越能看出Uber的强大之处。

Delivery is such a rich topic, and the deeper you go down the rabbit hole, the more you see the power of Uber.

例如，他们在全球各个城市打造了这个绝妙的双端餐厅市场。

For example, they created this wonderful two sided restaurant marketplace in cities all around the world.

作为餐厅，你可以在平台上打广告，还能为用户提供各种消费激励。

And as a restaurant, you can advertise and you can offer users different incentives to buy from you.

市场上竞争越激烈，用户获得的好处就越多。

And the more competition you have in a marketplace, the more the users benefit.

因此餐厅——其实任何能被Uber配送的商品或服务都可以代入——通过拍卖系统竞价，不仅竞标点击曝光量，还会为用户提供'买一赠一'之类的优惠。

And so the restaurant and you can essentially replace restaurant with any goods or service that can be delivered by Uber, they bid through an auction system, not only through impressions of clicks, but also gifts for users like buy this meal and get another meal for free.

斯蒂格 所以本质上，你是在将边际成本与边际收益进行对比。

Stig And so implicitly, you compare the marginal cost to the marginal revenue.

因此，如果你看到这种模式在规模上的高效性，并将其自动化，在全球众多市场推行，这正是谷歌在其广告业务中所采用的策略。

And so if you see how effective that is on scale and you automate that and you have a lot of markets all around the world, It is exactly what Google is doing with their advertising business.

这个过程本质上是一样的。

The process is really the same.

每当出现拍卖机制时，作为股东，你应当高度关注，因为你能从中获取大量的消费者剩余。

Every time that there is something called auctions, you should, as a shareholder, very much pay attention because you capture so much of that consumer surplus.

因此在餐厅等低利润行业中竞争非常困难，这些近乎于大宗商品。

And so it is very hard to compete in low margin businesses such as restaurants, which is your borderline commodities.

你很难比那些最不自律的竞争对手赚更多钱。

It's really difficult to make more money than your least disciplined competitors.

如果你像优步或谷歌那样提供促进竞争的工具，无论如何你都是赢家。

If you provide the tools to facilitate the competition like Uber or like Google, you win either way.

但克莱，回到你的观点，如果有人能以远低于优步的成本提供叫车服务呢？

But really to your point, Clay, what if someone can do ride hailing at a much lower cost than Uber?

我不抱任何幻想。

I make no illusions.

这无疑会打破我理论的核心部分，并彻底颠覆我对Uber第二大业务部门的预期。

That would definitely break a core part of my thesis and certainly disrupt what I would expect to be the second most important business unit for Uber.

因为在选择假想的特斯拉网络而非Uber时，几乎不存在任何转换成本。

Because there are hardly any switching costs in choosing the imaginary Tesla network over Uber.

你基本上只需下载另一个应用，就能让特斯拉或其他自动驾驶出租车上门服务。

You can basically just download a different app and then have your Tesla or whatnot robotaxi come to your door.

如果比Uber更便宜，你有什么理由不这么做呢？

And why would you not do that if it's cheaper than Uber?

斯蒂格，每当我查看数据时，制造自动驾驶汽车的成本仍然非常高昂。

Stig Whenever I look at the numbers, it's still very expensive to build AVs.

随着人们继续在自动驾驶领域竞争，薪资水平会被压低以保持竞争力——显然不会归零，但全球范围内仍有很大下调空间。

As people continue to compete with AVs, the salaries would be pushed down to stay competitive, clearly not to zero, but globally there's a lot of room.

我的意思是，资本主义就是如此残酷。

I mean, capsulism is just that brutal.

特斯拉还面临着其他阻力。

Tesla also faces other headwinds.

我必须首先承认，美国比欧洲要友好得多。

I'll be the first to say that US is friendlier than let's say Europe.

在欧洲这事根本行不通。

Don't even get this out in Europe.

美国可能仅次于中国，是最友好的地方，但要实现监管30:50框架还有很长的路要走。

And US is probably aside from China, where they're most friendly, but there's still a long way before the regulatory thirty:fifty framework is in place.

所以我必须说，通常押注反对埃隆·马斯克不是明智之举，这次也不例外。

And so I'll be the first to say it's usually not good business to bet against Elon Musk, and I don't think this is any different.

话虽如此，我不确定特斯拉是否真能成为这里的最低成本供应商。

That said, I'm not sure if Tesla is really going to be the lowest cost provider here.

我不仅指制造电动车的成本，还包括需要最佳指标技术来保持尽可能高的利用率，这才是真正降低成本的关键。

I'm not really only talking about the cost of making EVs, but also the need for the best metric technology to keep utilization rate as high as possible because that is really what's driving down cost.

确实，你还需要廉价电池和廉价材料，但利用率真的非常关键。

And yes, you also need cheap batteries and you need cheap materials in general, but utilization rate is really, really key.

Stig 即便是与优步在某些地区合作、在其他城市竞争的Waymo，他们也是通过优步的匹配技术获得最佳利用率。

Stig And so even Waymo who are partnering some places with Uber, but also competing with them in other cities, they get the best utilization rate with Uber's matching technology.

所以如果你看看Uber，我觉得我可能过于乐观且带有偏见，请理性看待——他们在全球各地无法与本土企业竞争时，都持有竞争对手的知名股份。

And so if you look at Uber, and I feel like I'm probably way too bullish and I'm way too biased, please take it for what it is, but they have famous stakes in competitors around the world where they could not compete with the locals.

东南亚的Grab、中东的Karim、中国的滴滴就是例子。

Grab in Southeast Asia comes to mind, Karim in The Middle East, Didi in China.

或许较少被宣传的是，他们实际上正在循环利用这些股权，并将其投入与Uber合作的各种自动驾驶汽车供应商中。

Stig And perhaps what gets a little less publicity is that they're really recycling these equity stakes and then putting them into various AV providers that work together with Uber.

总会有人——我不知道是谁——但必定会有人在自动驾驶领域胜出。

And someone, I don't know who, but definitely someone will emerge as the winner in the AV space.

我不确定那会不会是特斯拉。

I don't know if that will be Tesla.

但即便是特斯拉最终胜出，我认为有充分的理由证明与Uber合作将是双赢。

But even if it's going to be Tesla who's going to win, I think I see a pretty strong thirty:fifty case why teaming up with Uber is going to be a win win.

我们稍事休息，听听今天赞助商的内容。

Let's take a quick break and hear from today's sponsors.

想象一下用真正理解你客户的技术来扩展业务。

Stig Imagine scaling your business with technology that understands your customers, literally.

这就是Alexa和AWS AI背后的故事。

That's the story behind Alexa and AWS AI.

每天，Alexa处理超过10亿次交互，覆盖17种语言，同时将客户摩擦减少40%。

Every day, Alexa processes over 1,000,000,000 interactions across 17 languages, all while reducing customer friction by 40%.

这不仅关乎简化生活，更关乎重塑客户参与度并开辟新的收入来源。

It's not just about making life easier, it's also about transforming customer engagement and generating new revenue streams.

幕后由AWS AI驱动的70多个专业模型协同工作，打造自然对话，证明企业如何能安全可靠地规模化部署AI。

Behind the scenes, AWS AI powers more than 70 specialized models working together to create natural conversations, proving how enterprises can deploy AI at scale with confidence and security.

Alexa的AI能力在亚马逊庞大业务中经过实战检验，实现了规模化可衡量的真实影响。

Alexa's AI capabilities were battle tested across Amazon's massive operations, delivering real measurable impact at scale.

这些创新成果如今为其他企业提供了成熟框架，用以提升效率、开辟新收入来源并获得持久市场优势。

These same innovations now give other businesses a proven framework to boost efficiency, unlock new revenue streams and gain a lasting market edge.

访问aws.comai/rstory了解Alexa的故事。

Discover the Alexa story at aws.comai/rstory.

网址是aws.com/ai/rstory。

That's aws.com/ai/rstory.

新年伊始，正是实现你创业梦想的最佳时机。

It's the new year, which means that it's the best time to finally start the business you've been dreaming about.

就在几年前，我创办了自己的电商业务，Shopify正是我起步所需的完美工具。

Just a couple years ago, I launched my own e commerce business and Shopify was exactly the tool I needed to get started.

当许多人不断将梦想推迟到明年时，我要告诉你：现在就是把握眼前机遇的时刻。

While many people continually push off their dreams until the next year, I am here to tell you that now is the time to capitalize on the opportunities right in front of you.

Shopify为你提供线上线下销售所需的一切。

Shopify gives you everything you need to sell online and in person.

包括我在内的数百万创业者，都已实现从普通家庭到初次创业者的跨越。

Millions of entrepreneurs, including myself, have already made this leap from household names to first time business owners just getting started.

从数百款精美模板中任选定制，使用内置AI工具撰写产品描述或编辑商品照片。

Choose from hundreds of beautiful templates that you can customize and use their built in AI tools to write product descriptions or edit product photos.

随着你的成长，Shopify将全程伴你同行。

And as you grow, Shopify grows with you every step of the way.

2026年，别再等待，立即通过Shopify开启销售之旅。

In 2026, stop waiting and start selling with Shopify.

立即注册享受每月1美元试用优惠，今天就开始在shopify.com/wsb上销售吧。

Sign up for your $1 per month trial and start selling today at shopify.com/wsb.

请访问shopify.com/wsb。

Go to shopify.com/wsb.

网址是shopify.com/wsb。

That's shopify.com/wsb.

让Shopify陪伴您迎接新年的第一声钟响。

Hear your first this new year with Shopify by your side.

初创企业行动迅速。

Startups move fast.

借助人工智能，它们产品交付更快，也能更早吸引企业买家。

And with AI, they're shipping even faster and attracting enterprise buyers sooner.

但大订单会带来更大的安全与合规要求。

But big deals bring even bigger security and compliance requirements.

仅通过SOC2认证往往还不够。

A SOC two isn't always enough.

合适的安全措施能促成交易，也能毁掉交易。

The right kind of security can make a deal or break it.

但哪位创始人或工程师能抽得出时间，放下公司建设去处理这些呢？

But what founder or engineer can afford to take time away from building their company?

Vanta的人工智能和自动化技术让大额交易准备在几天内轻松完成。

Vanta's AI and automation make it easy to get big deals ready in days.

而且Vanta持续监控您的合规状态，确保未来交易永不受阻。

And Vanta continuously monitors your compliance so future deals are never blocked.

更重要的是，Vanta随您共同成长，全程提供及时的支持服务。

Plus Vanta scales with you, backed by support that's there when you need it every step of the way.

面对AI不断改变的法规和买家预期，Vanta清楚何时需要什么，并为您打造了最快、最简单的达标路径。

With AI changing regulations and buyers' expectations, Vanta knows what's needed and when, and they've built the fastest, easiest path to help you get there.

这就是为什么严肃的初创企业都选择早期通过Vanta实现安全合规。

That's why serious startups get secure early with Vanta.

我们的听众可享专属优惠：访问vanta.com/billionaires立减1000美元。

Our listeners get $1,000 off at vanta.com/billionaires.

访问vanta.com/billionaires可享1000美元优惠。

That's vanta.com/billionaires for $1,000 off.

好的，回到节目中来。

All right, back to the show.

在非正式讨论中，我们曾聊过如何称呼那些仅次于科技巨头的企业层级。

In the off air, we've talked a bit about what we can call the layer below the biggest tech companies.

当然，所有人都盯着'神奇七巨头'，但下面还有这个层级——我们可以称之为全球的Uber、Spotify和Shopify们。

Of course, everyone has their eyes on the magnificent seven, but there's also this layer below that we can call it, Think the Ubers, the Spotifys, the Shopifys of the world.

我在上周的节目中也和肖恩·奥马利简单讨论过这个话题。

I also discussed this topic a little bit with Sean O'Malley on last week's episode.

不如你来和我们聊聊你如何看待这个框架？

How about you talk to us about how you view this framework?

斯蒂格 价值投资圈有个著名案例：当1848年加利福尼亚南部发现金矿时——

Stig So you have this famous example in value investing circles about whenever gold was found in Southern Mills, California back in 1848,

第二年就出现了矿工的大规模涌入。

and then you have the massive influx of miners the following year.

于是就有了‘49人队’这个名字。

So you have the names of the 49ers.

但不管怎样，斯蒂格，真正靠淘金发财的只有少数人，大多数人都破产了。

But anyways, Stig really only a few gold diggers made a fortune, most went broke.

所以更稳妥的选择其实是成为提供镐和铲子的人。

So the safer bet was really to be the one providing the picks and shovels.

许多投资者自然会问自己：在当今的AI热潮中如何复制这种模式？

Many investors have naturally asked themselves, how can that be replicated in today's AI boom?

许多人得出的合理结论是必须投资英伟达、台积电、ASML这类公司，或是亚马逊、Alphabet和微软这样的大型云服务提供商。

Many have come to the logical conclusion that they must invest in Nvidia's and TSMC's and ASML's of the world or perhaps the biggest cloud providers such as Amazon, Alphabet, and Microsoft.

斯蒂格：基于这点，我一直在思考是否有其他方式可以布局？

Stig With that in mind, I've been thinking about, are there other ways we can place our bets?

回到你之前的观点，我认为在优步、Spotify这类公司之下还存在一个层级。

And so to your point before, I think there is a layer below companies such as Uber, Spotify.

另一个例子可能是Netflix。

Another example could be Netflix.

他们拥有大量宝贵的数据。

They have a lot of valuable data.

虽然他们的技术栈不及行业巨头，但可以购买这些资源的使用权，然后在自己的细分领域占据主导地位，因为他们掌握着最优质的数据。

Now, don't have the tech stack as the biggest players, but they can buy access to those resources and then in their own niche, they can dominate it because they have the best data available.

看看Spotify如何收集你的数据，又如何反过来利用谷歌和亚马逊的云服务将这些用户数据变现。

And so look at how Spotify collects data on you and how they in turn uses Google and Amazon's cloud to monetize that data on you as a user.

无论是过去还是现在，他们都拥有海量的用户行为数据。

And historically, and I would say also today, they have this vast amount of data of how you use their service.

比如当你暂停一首歌时，他们过去可能需要猜测原因，但现在完全不需要了。

So for example, if you stop a song, they might be figuring out why is it that you did that, but now they don't even have to guess anymore.

现在Spotify甚至有个功能，你可以直接告诉他们：'我很喜欢这首歌，但我刚才在做早餐'之类的原因。

Now there's actually a feature in Spotify where you can speak directly with them and tell you, Hey, I love this song, but I was just baking breakfast, whatever.

当你看到AI的力量时，本质上比拼的是谁拥有最好的数据，以及谁能最优地利用这些数据。

So whenever you see here the power of AI, it's really about who has the best data and who can use that data the best possible way.

这样他们就能为你定制产品，从而获得更强的定价权。

Then they can customize the product for you and it gives them more pricing power.

因此，借助底层的新技术，你实际上是在以更强大的方式实现这一点。

And so you're really doing that on steroids with that layer below with the new technology that becomes available.

30:30 所以像Alphabet和亚马逊这样的公司，仅通过提供云计算就能获胜，但它们仍难以与拥有更具体需求的Spotify、优步、奈飞等公司竞争。

Thirty:thirty So you certainly have companies such as Alphabet and Amazon that wins simply by providing the cloud computing, but it's still hard for them to compete with Spotify, Uber, Netflix who have more need specific data.

在数据丰富和人工智能的新世界里，传统企业更难与大型科技公司以及我称之为它们之下的层级竞争。

It's even harder for traditional businesses in the new world of abundant data and AI to compete with big tech and then what I would call the layer below them.

斯蒂格 所以我花了很长时间才跟上这一点。

Stig So it took me a long time to catch up to that.

长期以来，可能是因为我作为价值投资者的背景，但我一直非常关注那些倍数。

Had for the longest time, it's probably because of my background as a value investor, but I've been focusing heavily on those multiples.

这并不是说你不应该关注倍数。

This is not my way of saying you shouldn't be looking at multiples.

这可能是你能给出的最糟糕的建议。

Think that's probably the worst piece of advice you can give.

但我认为我们都对一些企业能持续增长多久感到惊讶，它们如何继续为自己再投资，以及你需要30:50来规范化那些利润率。

But I think we've all been surprised by how long some of those businesses can grow, how much they continue to reinvest in themselves, and how much you need to thirty:fifty normalize those margins.

举个例子，我在2020年买入了Spotify，2021年又卖出，然后在2022年底再次买入。

And so, for example, I bought Spotify back in 2020, I sold out again in 2021, and then I bought it back in late twenty twenty two.

如果我不提它后来涨了八倍就太失职了，不过我得第一个承认我投得太少，所以别太佩服我。

And I would be remiss if I didn't say it was an eight bagger since, but then I'll be the first to say that I invested so little, so please don't be impressed.

你们或许该对我失望——虽然我讲了为什么要回马枪投资，但自己却没做到。

Should probably be disappointed in me that now that I talked about why you need to swing back, I certainly didn't.

总之问题总是接踵而至，不过现在我对于在科技巨头之下的层级挑选赢家这个框架更有信心了。

Stig But anyways, one problem really just follows the next year because now I have much more conviction in the framework of picking winners in the layer beneath the biggest tech companies.

但同时，我也开始怀疑这种信心是否源于所谓的‘结果导向’——

Then at the same time, I'm also starting to doubt whether my conviction is due to the what is called resulting.

是不是因为股市证明了我的正确，我才找借口继续投资这个火热的股市？

Is that because the stock market is telling me that I'm right and I'm now coming up with excuses for continuing to investing in a really hot stock market?

时间会给出答案。

So time will tell.

但我觉得重要的是不要非黑即白地看待投资——不是对就是错。

But I think it's important to not think too binary about investing, as in either you're right or you're wrong.

如今，我更倾向于将其视为一场扑克游戏。

Today, I see this a bit more as a game of poker.

在扑克中，你玩的是概率和人性心理。

In poker, you play probabilities and you play human psychology.

所以如果你玩无限注德州扑克，你会渴望在翻牌前用AA对KK全押对抗对手。

So if you play no limit hold'em and you would love to go all in against your opponent with aces versus kings before the flop.

如果你完全听不懂我在说什么也没关系，但这意味着你有82%的胜率。

And if you have no idea what I'm talking about, that's completely fine, but that means that you're going to win 82% of the time.

但你仍可能做出正确决策全押筹码后输掉，因为毕竟有18%的失败概率。

But you can still make the right decision, get all the chips in, and then still lose because you're going to lose 82% of the time.

因此你需要区分信号与噪音，毕竟你无法控制荷官会发出什么公共牌。

And so you need to separate the signal from the noise and you just can't control which cards the dealer is going to put on the board.

投资也是如此，所有事情都处于0到100%的区间内，非黑即白的情况极少。

And so investing, everything is on the spectrum from zero to 100% and there are very few things that are either or.

真相往往介于两者之间。

It's always somewhere in between.

所以你根据概率做出投资决策，然后相应调整投资规模，因为你可能会判断错误，也可能运气不佳。

So you make your investment decisions based on probabilities and then you size them accordingly because you can be wrong, you can be unlucky.

而且很多时候很难区分你是判断错误还是运气不好

And very often it's hard to tell whether you've been wrong or one:thirty

无论是扑克还是投资。

unlucky, whether it's poker or investing.

在扑克中，我必须指出，我觉得很讽刺的是，当年我和朋友们如果遭遇亏损，我们总是互相安慰说'唉，我们只是运气不好'。而每当我们盈利时，却总归功于自己是出色的扑克玩家。

In poker, I have to call this out, I found it ironic how my friends and I back in the day, if we had a losing session, thirty:thirty we always told each other, Ugh, we've just been And then whenever we had a winning session, it was always because we're fantastic poker players.

我发现这种思维框架在投资中相当有用。

I found that framework to be quite useful to have in investing.

当然不是完全照搬，我应该说。

Not the exact one, I should say.

我试图判断自己在某项投资中是否正确。

I tried to figure out if I've been right on an investment.

我是真的判断正确，还是这只是一个80%概率的事件，由于结果导向才显得是个好决策？

Have I really been right or is this an 80% probability that just due to resulting was actually a good decision?

所以这是那种你必须保持谦逊的情况，如果

And so it's one of those where you have to stay humble one:thirty if

你是对的，但如果你错了也不要自责。

you're right, but also don't beat yourself up if you're wrong.

归根结底这都是概率问题，这就是我们玩的游戏。

It's all about probabilities at the end of the day, and that's the game we're playing.

科技确实造就了这些'壮丽七巨头'中的超级赢家。

Technology has certainly enabled these mega winners in the magnificent seven to emerge.

如果这个新AI时代能让第二梯队的大型科技公司成为丰沃的猎场，我一点都不会感到惊讶。

I wouldn't be surprised if this new AI era enables the second layer of big tech to be a fruitful hunting ground.

鉴于这次讨论，这让我想起了最近几个月思考的一个问题：投资小公司相较于这些美国消费者熟知的大型公司的机会成本。

And in light of this discussion, it reminds me of something I've thought about in recent months, and that's the opportunity cost of investing in smaller companies versus these larger companies that are in The US consumer facing and everyone knows about.

因此投资小盘股或小型公司的理由可能是：它们可能存在更多定价错误，被发现程度较低，且由于规模较小可能有更大成长空间。

So the case for small caps or small companies can be that there might be more mispricings, they might be less discovered, and there might be more room to grow since they're smaller.

你经常能获得相当有吸引力的估值。

You can oftentimes get pretty attractive valuations.

但多亏了我在智囊团社区遇到的许多人，我现在开始更关注这些规模更大、知名度更高的公司，比如全球的Alphabet和Uber这类企业。

But thanks to many of the people I've met in our Mastermind community, I've come around to putting a bit more focus on these larger, more well known companies like the Alphabets and Ubers of the world.

至少我还没聪明到能把这两家公司中的任何一家纳入我的投资组合。

Haven't been smart enough at least to add either of those to my portfolio.

虽然你过去投资过各种类型的公司，但我感觉你可能也经历过类似的转变。

Although you've owned many types of companies in the past, I sense that you've gone through maybe a bit similar evolution.

我知道我们在2010年代初、2010年代中期没有一对一讨论过你持有的股票，但我偶尔会听到你持有过一些冷门公司的故事。

I know we didn't talk one on one back in the early 2010s, mid-2010s about some of the stocks you owned, but I hear stories one:thirty about some obscure companies that you've owned.

不如你来谈谈，在投资市场时，

How about you talk about how you figure out what pond you want to fish in when it

你是如何决定要在哪个领域投资的？

comes to investing in the market?

是的，这是个好问题。

Yeah, it's a good question.

如果我们从公司规模说起，我对这点考虑得并不多。

So if we start with the size of the company, I don't think too much about that.

我曾经是。

I used to.

我曾经非常关注这一点，如果是最大的公司，他们是否具有规模优势？

I used to very much be thinking about, if this is the biggest company, do they have scale advantages?

如果是小公司，这是否意味着他们有成长空间？

If it's a small company, does that mean they have room to grow?

我仍然认为这是重要的考量因素。

I still think that's important consideration.

如果你看我的投资组合，我持有的最小公司市值不到20亿。

If you look at my portfolio, the smallest company I have, it has a market cap of less than 2,000,000,000.

同时我也持有Alphabet。

And then I also have Alphabet.

它的市值超过3万亿。

It's more than 3,000,000,000,000.

所以归根结底取决于我认为自己理解哪些公司，以及是否符合30:50投资理论。

So it really comes down to which companies I think I understand and then to the investment thirty:fifty thesis.

我越来越倾向于考虑的另一件事，在其他条件相同的情况下，这或许让我更青睐大公司，就是我倾向于投资那些

Another thing I've increasingly been considering, and that perhaps tells me towards larger companies, everything else equal, is that I tend to invest in businesses thirty:fifty where

我觉得有趣且使用它们产品的企业。

I find them interesting and it's products that I use.

Alphabet和Uber就很好地符合这一标准。

Alphabet and Uber fits well into that bucket.

所以当我开始价值投资生涯时，我会投资那些名不见经传的海上油气设备公司。

And so whenever I started my value investing career, I would invest in this obscure offshore oil and gas equipment company.

我曾经投资过United Therapeutics，因为我觉得自己对肺动脉高压药物有独到见解。

I once invested in United Therapeutics because I felt I had this solid thesis about pulmonary arterial hypertension medication.

事实证明我并没有。

I don't, it turned out.

这不仅完全超出了我的能力范围，而且也不是我真正感兴趣的研究课题。

But not only was it just far top of shortcomatous, but it wasn't a topic I really cared to study.

如果不是因为投资，我可能永远不会研究它，这也意味着我不会真正去深入了解它。

If it wasn't for my investment, I probably never would study it, which also means that I don't really read up on it.

所以那种被迫式的研究对我来说效果并不好。

And so that type of forced reading just hasn't worked out well for me.

斯蒂格，我发现自己开始研究自动驾驶汽车和云计算的未来。

Stig I found myself studying the future of AVs and cloud computing.

对我来说，这本身就很有趣。

To me, that's just interesting.

我在投资之前就阅读相关领域，因为我觉得它很有意思。

I read about it before I invested in it because I found it to be interesting.

因此我认为这种求知欲确实能带来优势，因为在某个时间点你可能被迫卖出或想要卖出。

And so I do think that that curiosity gives you an advantage because at some point in time you may be forced to sell or you want to sell.

你需要对这个行业的各个动态部分有深入了解。

You need to have a lot of knowledge about the moving parts in the industry.

根据我个人痛苦的经验来看，如果你进入一个你本身并不感兴趣的领域，你可能只会研究那家特定公司，而不会太关注竞争对手。

And if you or at least I probably just speak from my own painful experience, if you're entering into an area where you're not naturally curious, you would mainly be reading up about that specific company and not so much about the competitors.

你无法总是获得同样全面的行业概览。

You don't always get the same thirty:fifty overview of what's going on.

但回到你关于公司规模的论点，我认为这也取决于我们如何定义规模。

But again, back to your point about the size of companies, I also think it depends on what do we mean by that.

比如在上次Mastermind讨论中，我推荐了Remedly。

So for example, in the last Mastermind discussion, I was pitching Remedly.

这是一家市值26亿美元的公司。

It's a 2,600,000,000 market cap company.

我认为他们有可能在汇款领域成为重要参与者。

So I think that they could become a large player in their remittance space.

假设他们市值增长到100亿美元，这算小公司还是大公司？

Let's say you would take them to a $10,000,000,000 market cap, is that a small or big company?

如果在汇款领域，这实际上就是大公司了。

Well, if it's in the remittance space, it's actually a large company.

Stig然后他们表示想拓展传统商业转账业务。

Stig And then they said, Oh, we want to expand this traditional business transfers.

这种情况下可能会出现市值300亿美元的公司，但在更大的市场中他们只是小玩家。

And in that case, there might be a 30,000,000,000 market cap company, but then they're a small player in a much larger market.

你们有些人可能会说，三百亿还是五百亿真的有那么重要吗？

And some of you might be saying, Well, does it really thirty:fifty matter?

300亿难道不比100亿更好吗？

Isn't 30,000,000,000 better than 10,000,000,000?

我们难道不应该感到高兴吗？

Shouldn't we just be happy?

然后我们可以更进一步说，如果他们必须大幅稀释股东权益才能达到那300亿市值呢？

And then we could take it one step further and say, Well, what if they had to significantly dilute shareholders to get to that thirty:fifty 30,000,000,000

市值？

market cap?

如果他们坚守汇款领域，通过有机增长达到100亿市值，同时支付股息并回购股票，你可能会获得更好的回报

You might get a much better return if they stayed in the remittance space and they organically grew to a $10,000,000,000 market cap while they're paying out dividends and buying back

这让我不禁想到你们投资组合中的Constellation Software公司，他们正在利基市场中收购近乎垄断的企业

And so I can't help but think of one of the companies in your portfolio, your Constellation Software, who are buying near monopolies in niche markets.

而这些市场的价值可能只有数百万或数千万美元

And those markets might only be worth millions or tens of millions of dollars.

但如果你拥有足够多的这类企业并建立起护城河，突然间你就能获得价值数百亿美元的市场估值。

But then if you have enough of them and you have a moat around them, all of a sudden you have a market cap worth tens of billions of dollars.

克莱，我们转到你持有的另一家公司，至少今年它表现优异。

Clay Transitioning to another one of your holdings, it's been a winner this year at least.

这就是字母表公司（Alphabet），截至录制时年内涨幅已达54%。

It's Alphabet, it's up 54% year to date as of the time of recording.

沃伦·巴菲特的伯克希尔哈撒韦也加入了这场盛宴，最近一个季度购入了价值超40亿美元的股票。

Warren Buffett's Berkshire Hathaway joined the party by purchasing over $4,000,000,000 worth of stock in the most recent quarter.

正如你过去多次强调的，Alphabet是你投资组合中的核心持仓。

And as you've outlined many times in the past, Alphabet's a core position in your portfolio.

众所周知，迈克尔·伯里近期频频直言AI泡沫问题，指出科技巨头通过低估折旧来虚报盈利。

Recently, as many know, Michael Burry hasn't been shy to call out the AI bubble and how Big Tech's overstating their earnings by understating depreciation.

你对当前AI竞赛怎么看？Alphabet在这场竞赛中扮演什么角色？

What's your take on just the AI race, Alphabet's role in the AI race?

你如何看待这些科技巨头在资本支出上永无止境的军备竞赛？

How do you think about the never ending spend on CapEx that a few of these big tech companies are in a race.

他们的资本支出正膨胀到前所未有的水平。

Their CapEx is just ballooning to unprecedented levels.

你如何看待这个问题？

How do you think about this?

斯蒂格 确实，Alphabet与微软和亚马逊一起，看起来已为成为AI革命的中坚力量做好了准备。

Stig Definitely, Alphabet looks well positioned together with Microsoft and Amazon to be the backbone of the AI revolution.

当然还有其他玩家正在加入这场游戏。

There are certainly other players joining the game.

目前我看不到任何能与这三大巨头三足鼎立的竞争者。

I don't see anyone competing thirty:fifty with them, the big three, right now.

谁知道呢？

Who knows?

我们都明白盛极必衰的道理。

We know that the mighty fall.

如果你看看1990年世界银行市值最大的公司，其中有八家是日本企业。

If you look at the biggest companies in the World Bank Market cap in 1990, eight of them were Japanese.

如今，日本最大的三十比五十公司

Today, the biggest Japanese thirty:fifty company

是丰田。

is Toyota.

它排名第54位。

It's placed 54.

资本主义是残酷的。

Capitalism is brutal.

快进到现在，如果我们看2007年全球最大的公司，会发现前10名中有四家都是大型银行。

And so fast forward, if we look at the biggest companies in the world in 2007, you find four of the all majors in the top 10.

而今天一家都没有了。

You will find zero today.

斯蒂格 可能有人会说，这是因为石油会枯竭而数据会复利增长，并认为这次不一样，AI巨头将长期保持霸主地位。

Stig Then some might say, Well, that's because all depletes and then data compounds, and then argue that this time it's different and the AI titans will remain titans for a very long time.

也许确实如此。

Perhaps that is the case.

我认为我们很容易受到近因偏差的影响。

I think we're very susceptible to recency bias.

在使用这些产品之前，我很难想象ChatGPT或iPhone的存在，而现在我已无法想象没有它们的生活。

I had a really hard time imagining ChatGPT or the iPhone before I started using the products and now I can't imagine life without them.

我毫不怀疑这三大巨头终将被拉下神坛。

And I don't make any illusions that the big three won't eventually be dethroned.

我的意思是，我之前说过，资本主义就是如此残酷。

I mean, I've said it before, capitalism is just that brutal.

所以问题在于，这需要多久才会发生？

And so the question is, how long would it take before it happens?

在此期间你能赚多少钱？

How much money will you make in the meantime?

如果巨头倒下，你能在市场意识到之前及时退出吗？

And if the mighty fall, can you exit your position before the market realizes it?

这绝对是说起来容易做起来难。

And it's definitely easier said than done.

抛开近因偏差不谈，我深知要与那些以极低成本甚至免费为你提供定制化服务的人竞争是非常困难的。

Stig But recency bias aside, it's not lost in me that it's very hard to compete with someone who provides you with a service that's customized specifically for you at a very low or no cost to you.

Alphabet在这方面是专家。

Alphabet is an expert in that.

我认为随着AI的兴起，这种情况会更加普遍。

I would argue with the rise of AI, it would be even more prevalent.

我从2018年开始买入Alphabet股票，从初始价格算大约翻了五倍，从均价算翻了四倍。

I've been buying Alphabet since 2018, roughly a five bagger from a starting price and a four bagger from average price.

把这个与关于大盘股和小盘股的问题联系起来，你会发现即使投资这些巨头公司，依然能获得很高的回报。

Tying this to a question about the big to small cap stock question, you still see very high returns on your investment, even with these massive companies.

这确实让我感到惊讶。

I've certainly been surprised by that.

再次强调，这不仅关乎市值增长（尽管Alphabet在这方面的表现已经足够惊艳），还包括股息和股票回购带来的收益。

Again, it's not just about market cap growth, even though that has been impressive enough with Alphabet, but it's really also in dividends and share buybacks.

关于资本支出问题，2015年他们给出的指引是910亿到930亿美元，而明年他们表示会有显著增加，但尚未给出具体金额。

Stig To your question about CapEx spend, in 2015, the guidance they've given has been 91,000,000,000 to $93,000,000,000 and then for next year they said significant increase, but they haven't given the exact amount.

一种方法是询问其中有多少是维护性资本支出，有多少是增长性支出？

So one way is to ask how much of this is maintenance CapEx, how much of that is growth?

我最好的猜测是目前增长性支出多于维护性支出。

My best guess is that there's more growth than maintenance right now.

首先，你可以查看过去业绩并与资本支出进行比较。

You can, for starters, look at the trailing results and compare it to CapEx.

与此同时，如果谷歌想保护搜索和广告业务，就必须保持领先地位。

At the same time, if Google wants to protect search and ads, it has to stay at the forefront.

与微软不同，我应该说微软是租用英伟达的设备。

Unlike Microsoft, I should also say that rents Nvidia.

而谷歌正在自主构建整个技术栈——从GPU、数据中心到建模工具等等。

Have Google that's building their entire stack themselves, GPUs, data centers, modelers, and so on and so forth.

我倾向于认为这是高回报但也可能是高风险的投资。

I'm inclined to say high rewards and perhaps high risks.

时间会证明一切。

Time will tell.

我有个以收购公司为生的朋友，他跟我开玩笑说，当他寻找新公司收购时，从经纪人那里听到最糟糕的一句话是‘看看这些高资本支出’。

I have a friend who buys companies for a living and he jokes with me and he says the worst line that he hears from brokers whenever he's looking for new companies to acquire is, Look at all this high CapEx.

它会为你筑起护城河。

It's going to give you a moat.

他总是这样，唉，这简直痛苦死了

And he's always like, Ugh, this is just painful thirty:fifty

因为高资本支出基本上意味着未来你将面临高额的替换成本，这不是我们想看到的

because high CapEx basically just means that you're going to have high replacement costs later and not something we like.

但我也认为这其中存在微妙之处。

But I also think there is a nuance here.

我不会假装大量资本支出是件很棒的事。

I'm not going to pretend that a lot of CapEx is awesome.

这完全不是故意的。

That's not by intention at all.

但在Alphabet、微软和亚马逊这样的运营规模下，支出如此巨大，实际上限制了能够参与竞争的对手。

But at the scale that Alphabet and Microsoft and Amazon operates, the spending is so massive that it actually limits who can even show up to compete.

大多数公司根本负担不起每年烧掉数百亿美元的成本。

Most companies simply can't afford to burn tens of billions a year.

因此关键不在于资本支出本身，而在于企业是否有足够实力年复一年地维持这种级别的投入。

And so the mode isn't the CapEx itself, the mode is having a business strong enough to sustain that type of spend year after year.

所以在人工智能领域，你根本无法半心半意地参与竞争。

And so whenever it comes to AI, you can't really half compete.

我不确定是否该用这个词来形容。

I don't know if I'm going to commit to that word.

这不像你投入50亿而非500亿就能成为缩小版谷歌——根本是两码事。

It's not like if you spend $5,000,000,000 instead of $50,000,000,000 or instead of 100, oh, now you just have a smaller Google.

你基本上只会被边缘化。

You basically just become irrelevant.

这就是场规模经济游戏：赢家获得更廉价的计算资源和更优的模型，小玩家则被彻底碾压。

So it is an economies of scale game where the winner gets cheaper computing and better models, and then the smaller players just get crushed.

但我要补充说明：这种模式成立的前提是投入必须能带来回报，而有人认为这点尚待验证。

But then I also just want to preface that in saying, it's of course only a mode if the spending is going to give you a return, and some would say that's still left to be seen.

克莱，你说得很有道理，如果Alphabet在资本支出上花费少于900亿美元，可能会是个代价高昂的错误。

Clay You do make a good point that if Alphabet spends somewhere less than $90,000,000,000 on CapEx, it could be quite an expensive mistake.

正如你所说，在人工智能领域他们确实没有半心半意竞争的选择。

They don't really have the choice of half competing, as you say, in AI.

信不信由你，苹果的资本支出在过去五年里一直保持相对平稳。

Believe it or not, when you look at Apple, their CapEx has remained relatively flat over the past five years.

他们刻意避开了这场人工智能军备竞赛。

They've deliberately stayed away from this AI arms race.

是的，继续观察事态发展会很有趣。

And yeah, it'll be interesting to watch it continue to play out.

我最近听了其他几家科技巨头的财报电话会议，尽管他们计划在2026年投入巨额资金，却仍担心投资不足，这种局面看起来非常有意思。

I was tuning into some of the other earnings calls for some of these big tech companies and they almost fear under investing despite how much they're looking to spend in 2026, which is just amazing to watch.

我听你提到过另一个思维模型，我们可以简单称之为'不公平优势'。

And I've heard you talk about another one of your mental models, which we can refer to as just simply having an unfair advantage.

你能详细谈谈这个思维模型吗？在构建投资组合时它意味着什么？

How about you talk more about this mental model and what it means when you're building out your portfolio?

Stig 是的，这是个有趣的概念。

Stig Yeah, it's an interesting concept.

前几天我和一个朋友聊过。

I was speaking with a friend the other day.

让我先把这个问题退一步讲。

Let me just take that one step back here.

他对自己当时的工作不太满意，所以问我他可以开创什么事业。

And he wasn't too happy with his day job, so he asked me which business he could start.

我在这里也运用了同样的不公平优势心智模型。

And I used the same mental model here of unfair advantages.

于是我们发现他妻子是退伍军人，所以他们可以零首付买房。

So together we found his wife is a veteran, so they could buy a house with no money down.

他岳父是承包商，所以他们能获得帮助并以成本价购买材料。

His father-in-law is a contractor so they could get help and buy materials at cost.

他们夫妻俩都很能干，所以很多工作可以自己动手完成。

Both him and his wife are quite handy so they can do a lot of the work themselves.

把这些因素加起来，再加上他们没有孩子，行动相当灵活，可以四处搬迁。

And so if you add that and then you also add they don't have kids, they're quite flexible, they can move around.

如果你住在美国的主要居所，装修后可以在两年后免税出售。

If you live in your primary residence and you're based in America, you can fix it up and then sell it tax free after two years.

所以他们拥有所有这些优势。

So they had all of those advantages.

这就是我所说的不公平优势。

So that is what I would call an unfair advantage.

我这么说的时候感觉可能会听起来很糟糕，因为我不想让人觉得爱国和为国家服务是不公平的。

I kind of feel like, as I'm saying it, this is going to sound terrible because I don't want to make it sound like it's unfair if you're patriotic and you've been serving your country.

这不是我的本意。

That's not my intention.

我所说的不公平优势，是指如果你自己没有这些优势，就很难与之竞争。

But what I mean whenever I say unfair advantage is that it's hard to compete with if you don't have those advantages yourself.

如果你和他们正面交锋，你只会输掉。

If you're going toe to toe to them, you're just going to lose.

斯蒂格，同样地，当你构建和管理投资组合时，应该充分利用你的不公平优势。

Stig Similarly, whenever you build and manage your portfolio, you should really lean into your unfair advantages.

所以我要说的一个可能不公平的地方是——我不确定——但我见过很多人一有钱就忍不住挥霍。

So one of the things that I would say that might be unfair, I don't know, but I've met a lot of people whenever they have money, that money is just burning a hole in their pocket.

而我出于某种原因，并没有同样的冲动。

And I, for whatever reason, don't feel the same way.

所以我只靠收入中相对较少的部分生活。

So I live on a relatively small part of my income.

这让某些事情变得稍微容易些，因为可以让我更专注于长期目标。

So it makes some things a little bit easier in the sense of it allows me to be more long term focused.

我不必依赖市场先生

I don't have to rely on Mr.

来决定何时出售股票。

Market whenever I sell my stock.

当然间接上可能还是需要，因为我希望获得好的估值，但我没有被强迫出售的压力。

Well, indirectly, probably have to because I want a good valuation, but I'm not forced to sell.

因此，如果你量入为出地生活，可以说这是一种不公平的优势，因为你有三到五成的概率拥有这种优势

And so if you live within your means, you can argue that that is an unfair advantage if you like, because thirty:fifty you have

这种优势让你可以自主决定何时出售，而不必被迫行动

that advantage that you decide whenever you sell because you're not forced to do

不公平优势的框架类似于彼得·林奇'投资你所知'理念的表亲，但显然并不相同

And so the framework of unfair advantage is like of a cousin of Peter Lynch's invest in what you know, but it's certainly not the same.

生在马厩里不会让你变成一匹马

Being born in a stable doesn't make you a horse.

假设你每天带着笔记本电脑去星巴克，因为你对咖啡很了解，能盲测出他们用的混合咖啡豆——没错，这很厉害

Just let's say you go every day to Starbucks with your laptop and because you know stuff about coffee, you can blind test whatever kind of blend they're using and it's, yes, awesome.

这很棒

That's great.

如果你看不懂星巴克的财务报表，也不知道它的价值，你或许就不该投资它

If you can't read the financial statements of Starbucks and you don't know what it's worth, you probably shouldn't invest in it.

因此你仍需要找到真正深入了解的领域，才能发挥这些不公平优势

Stig And so you still need to find something you really understand really well to lean into those unfair advantages.

有时候它就明摆在眼前

And then sometimes it's just hiding in plain sight.

有时候你得深挖一下才能发现

Sometimes you have to dig a bit for it.

有位大师班社区的嘉宾谈到他为何不投资奢侈品

Had a guest in the Mastermind community and he was talking about how he didn't invest in luxury goods.

我觉得他对手提包不感兴趣的观点挺有意思

I kind of felt that he had an interesting point about he was just not interested in handbags.

我完全理解他的意思

I completely understand what he means.

我对包包确实没兴趣

I'm not interested in handbags.

我真正感兴趣的是为什么有人愿意花数万美元买个包

What I am interested in is why is it that someone is willing to pay tens of thousands of dollars for a handbag?

顺便声明一下，我虽未投资奢侈品行业，但觉得这现象非常迷人

And I should just say for the record, I'm not invested in the luxury sector, but I find it deeply fascinating.

我认为如果你能看透背后的原因，弄清楚为什么有些人愿意支付高价，具体是什么产品其实并不太重要。

I think if you can look behind that and figure out why is it that some people are willing to pay that, it doesn't really matter too much.

产品是法拉利吗？

Is the product a Ferrari?

产品是手提包吗？

Is the product a handbag?

如果你理解消费者行为，更重要的是理解消费者行为的变化，你仍然能找到自己的不公平优势。

If you understand consumer behavior, and more importantly understand the changes in consumer behavior, you can still find your unfair advantage.

因此，识别你的不公平优势可能需要一些时间，但一旦找到，它应该会相当明显。

And so identifying your unfair advantage may take a little time, but once you find it, it should be quite evident.

如果能用数据验证这一点也会很有帮助。

It's also helpful if you can validate it with numbers.

比尔·米勒曾与我们分享过，优势分为三种类型。

Bill Miller shared with us that there are three types of advantages.

第一种是信息优势，即比别人更早获取重要信息。

There are the information advantage, which is basically getting important information before others.

这在当今很难做到，因为新闻传播得太快了。

That's hard today because news out there is contagiously.

当然，如果你有内幕消息，那是违法的，所以你最好别那么做。

Of course, you have insider information, then that's illegal, so you probably shouldn't be doing that.

其次是分析优势，即比市场更善于理解公开信息。

Stig Then there is analytical advantage, understanding public information better than the market.

这相当难做到。

That's quite difficult to do.

然后是性格优势，基本上就是比其他投资者拥有更好的心理素质。

And then there's a temperamental advantage, basically having a better psychological profile than other investors.

克莱，不知道你还记不记得，就在几年前——或者说过去——这只是在我的回音室里。

I don't know if you remember this, Clay, but just a few years ago or past it was just in my echo chamber.

这是许多价值投资者讨论的框架，几乎所有人，包括我自己，都曾有过这种想法：'哦，我的性格比别人好'。

This was a framework a lot of value investors were talking about and almost everyone, including yours truly, had this idea that, Oh, I have a better temperament than other people.

我倾向于不相信这种说法，包括对我自己，因为这是很难衡量的。

And I'm hesitant to believe that is the case, including with myself, I should say, because it's difficult to measure.

如果你说'我有信息优势'，对方会说'说来听听'。

If you say, I have an information advantage, the other person would be like, Tell me about it.

证明给我看。

Show it.

或者说'我有分析优势'。

Or if you say, Analytical advantage.

那好，告诉我你是怎么解读那份10-K报表的，为什么你比我理解得更透彻。

Okay, tell me how you read that 10 ks and why you understand it better than me.

大多数人其实并不想读10-K报表。

Most people don't really want to read a 10 ks.

他们只想听个好故事。

They want a good story.

不，真正愿意告诉你他们10-K报表内容的人少之又少。

No, very few people actually want to tell you what they've written in their 10 ks.

所以大多数人会说'我只是更有耐心'，或者'我不受确认偏误影响'之类的话。

And so most people would say, I'm just more patient, or I'm not susceptible to confirmation bias, or whatever it is.

这某种程度上是件便利的事，因为确实难以衡量。

And it's kind of a convenient thing because it's really difficult to measure.

所以我认为，当你审视自己的不公平优势时，关键是要能实际衡量它。

So I think it's important whenever you're looking at your unfair advantage that you can actually one:thirty measure it.

这并不意味着无法衡量的优势就不存在。

That doesn't mean that you cannot have an advantage if it can't be measured.

只是说你应该更加怀疑——

It just means that you should be a lot more suspicious whether or

它是否真的算不公平优势。

not it truly is an unfair advantage.

接下来我想谈谈我称之为比较优势的概念，它虽不相同但也不无关联。

Then I wanted to talk about something I call comparative advantage, which is not the same, but it's not too remote.

举例来说，根据我的居住地和注册身份，我在黄金上支付的相对税负低于股票。

For example, where I live and how I'm registered, I pay less comparative tax on gold than I do on stocks.

在读完巴菲特的所有著作后，我本应发誓绝不投资黄金。

Going through all of Buffett's writings, I would have sworn never to invest in gold, of course.