TacticScaleMarkets · Valuation
Valuation Is Not a Property of the Company
From the A/H share gap to what a market price really is — and why the better question isn’t what a stock is worth, but what its price already believes.
We are used to asking what a company is “really” worth. Spend enough time in markets, though, and you start to suspect the question itself is slightly wrong.
Investment analysis carries a stubborn intuition: a company has an intrinsic value, the analyst’s job is to calculate it, and the market price merely wobbles around that true number. It’s an appealing picture, because it offers certainty. Feed revenue, margins, growth, ROIC, cost of capital, free cash flow and a terminal growth rate into a model, and out comes an answer: the company is worth 80, or 120, or 137.42.
But the market keeps reminding us of something the model leaves out. Valuation isn’t produced inside the company. It is produced where the company meets the capital market.
The company brings profits, cash flow, assets, growth and competitive advantage. The market brings money, liquidity, risk appetite, valuation anchors, positioning, rules, narratives and trading constraints. A share price is what forms, moment to moment, when those two systems collide.
One company, two markets, two prices
Companies listed in both Shanghai or Shenzhen (A shares) and Hong Kong (H shares) are one of the cleanest natural experiments there is. Same company, trading in two places at once. Management hasn’t split in two. Neither have the factories, the brand, the revenue or the profit — none of it belongs to one market or the other. In theory, both markets are looking at the very same business.
Yet the two listings can trade at quite different multiples, and for years at a time. Eastroc Beverage, the Chinese energy-drink maker, is a recent example: its A shares and H shares carry visibly different prices and price-to-earnings ratios. The exact gap on any given day matters less than the fact that it exists at all. The same profit can be priced differently in different markets.
The company’s earning power decides the first term. The second — whether the market pays 12 times, 18 times or 30 times — is plainly not something the company decides for itself. So a sentence like “this company is worth 15 times earnings” deserves a second look. A more accurate version is:
A P/E is not a property of the company. It belongs to the company and the market together.
A listing is a coordinate system
A market is never a neutral price display. Each has its own mix of money, its own participants, its own rules.
Investors in mainland China, Hong Kong and the US are not the same people, and they don’t ask the same things of a stock — on risk, on growth, on how scarce a sector is, on governance, on the cost of capital. A Chinese consumer company listed at home may be compared with the leading liquor and food-and-beverage names. Listed in New York, it gets compared with global consumer brands. In Hong Kong, it is priced through global dollar liquidity, the risk premium attached to Chinese assets, and Hong Kong’s generally lower valuation levels.
Listing the same company in a different market means placing it in a different valuation coordinate system. What changes isn’t just where the shares trade. It is who sets the price, what they compare it with, what risk they’ll carry, and how many times they’ll pay for the same stream of future cash.
This is also why a good company is never automatically a good stock. A business can be excellent and still deliver ordinary returns if the market has already paid a very high multiple for it. And a merely average business can be a fine investment if the price already assumes the worst, and reality turns out less bad.
Why the Buffett Indicator and the P/E tell different stories
The same logic holds for the market as a whole. By the Buffett Indicator — total US stock market value against GDP — US equities sit near the top of their historical range, and the Shiller CAPE is not far from its own extremes. Switch to the S&P 500’s forward P/E, or the Nasdaq-100’s current earnings multiple, and the picture looks rich but far less extreme than the 2000 dot-com peak.
That looks like a contradiction. It isn’t.
A high market-cap-to-GDP ratio doesn’t have to come entirely from a high P/E. It can also come from corporate profits taking a larger share of GDP over decades, from US-listed companies earning more of their revenue abroad, from structurally higher margins at the largest technology companies. Valuation ratios are not truths that float free of the economy. Each is one projection of the market — one angle of view.
Fundamental analysis isn’t useless — it’s misused
Fundamental analysis earns its keep. It answers the questions that matter: does the company actually make money? Are the profits durable? Is the free cash flow real? Are returns on capital high enough? Can the moat be breached? Can earnings grow over the next ten years?
Where it goes wrong is when all of that gets squeezed into a single, precise-looking “fair price.” The discounted-cash-flow model is the classic case. It looks mathematically rigorous, but nudge the long-term growth rate, the discount rate, the margin or the terminal assumption, and the answer can swing enormously.
The formula is right. The trouble is that future FCF and r aren’t known constants. A model that outputs 137.42 hasn’t discovered an objective number hidden in nature. What it is really saying is: if these assumptions about cash flow, growth, cost of capital and terminal value hold, I’m willing to value the company in this range. That is an entirely different question from what the market will actually pay.
A DCF doesn’t find the value. It states the conditions under which you’d accept a value.
A price carries far more than fundamentals
A real market price looks less like the output of a valuation model and more like this:
Fundamentals are one input among many. A stock can rise today because earnings expectations went up — or because index funds were buying, shorts were covering, institutions were underweight, the sector story changed, rates fell, liquidity improved, risk appetite rose, or simply because the sellers stepped away for a while.
More importantly, none of these inputs sit still. A rising price changes how participants behave.
This is what makes “sentiment” interesting. It isn’t simply optimism versus pessimism; it’s a feedback system. A rising price reinforces a narrative. The narrative draws in money. Inflows dampen volatility and lift the multiple. The higher price then pulls in trend-followers, passive flows and performance-chasers. The price is no longer just a result — it becomes an input to the next round of price formation. Markets, in this sense, are reflexive. The loop runs in reverse on the way down, too.
So is the market price the “true” value?
If “true value” means some permanent economic constant, then plainly not. A company can fall 20% in a day on a liquidity crunch, a forced deleveraging or a large fund’s redemptions. Its factories didn’t vanish. Its customers didn’t leave. Its future cash flows did not shrink by a fifth in a few hours.
But ask a different question — what is this share worth, right now, this second? — and the market price holds a position no other method can claim. It is the only number that has been confirmed by real money changing hands.
A DCF is a model. A price target is an opinion. P/E is a ratio, P/B is a reference point, EV/EBITDA is a vocabulary. Only the market price is a price at which people actually exchanged cash. A stricter way to put it:
The market price may not equal objective intrinsic value. But at any moment, it is the only clearing value that is observable, tradable at once, and verified by real money.
There are at least three kinds of value
| Concept | What it means |
|---|---|
| Economic value | The economic benefit the business will create for its owners in the future. It exists, but it can’t be observed directly — only estimated. |
| Perceived value | The market participants’ collective judgement of those future benefits. |
| Market price | Those judgements after they pass through money, positioning, liquidity and trading mechanics — the instant clearing price. |
Separate the three and many investing arguments simply dissolve. Fundamental investors are usually talking about economic value. Analysts argue over a reasonable range for perceived value. Traders face the market price every day. These aren’t rival views; they describe different layers of the same system.
Studying the company isn’t studying the stock
Experienced traders often ask questions a valuation model rarely does:
- Who, exactly, is buying right now — and why do they have to?
- How much new money is still to come?
- Whose positioning is crowded, and who might be forced to sell?
- What multiple is the market willing to pay for this story?
- Is the valuation regime expanding, or contracting?
That isn’t because fundamentals don’t matter. It’s because company analysis and stock analysis are not the same thing. Company analysis studies the business. Stock analysis has to study the participants too.
Where a stock’s return actually comes from
Compress the whole discussion into one framework and it looks something like this:
The first term is the business genuinely getting better or worse. The second is the market changing its mind about the future. The third is a change in the multiple the market is willing to pay at all. The most violent moves are rarely the work of the first term alone. A company grows earnings 20% and the stock doubles: the difference usually came from expectations and a richer multiple. Run it backwards — earnings still growing, but the market re-rates the stock from 40 times earnings to 20 — and the share price can fall hard all the same.
Being right about the company is not the same as being right about the stock.
The better question: what does the price already believe?
Traditional valuation loves one question: what is a fair price for this company? A more useful one may be: what has the market already come to believe, at today’s price?
They sound alike and are nothing alike. The first tries to derive a price from inside the company. The second works backwards from the price to the crowd’s collective expectations. Accept that valuation is generated between a company and its market, and the focus of research shifts. It is no longer only the company, but also:
- the business’s real economic capacity;
- how much growth the current price already assumes;
- which valuation anchors investors are using;
- where the money is flowing;
- which valuation regime the market is in;
- and how the price itself is changing the behaviour of the people trading it.
None of this makes fundamentals meaningless. What loses its meaning is treating a fundamental model as a machine that prints the one correct price. Businesses create cash. Investors form expectations. Money sets the marginal trade. Market structure constrains how trading happens. And the price keeps forming, out of all of it.
Valuation is not a nameplate bolted to the company. It is a process — and the market is that process.
Reading the price backwards on TacticScale
TacticScale’s fair-value view is built around that second question. It still runs one uniform owner-earnings model for every company, but it treats the result as a set of stated assumptions, not the answer.
What the price implies. A market-implied marker reads the model in reverse: at today’s price, what annual return — and so what growth — is the market already pricing in?
Cash versus belief. The no-growth floor splits the price into the part the current cash already supports and the part that is a bet on growth not yet seen — the layer most sensitive to a change in expectations or valuation regime.
The crowd’s own range. For liquid names, an options-implied curve shows the price distribution traders are actually paying for — perceived value, read straight off the option chain.
See what any price already believes → A model estimate, not advice — the decision stays yours.
This essay is a discussion of market structure and valuation method, for education only. It is not investment advice, and nothing here is a recommendation to buy or sell any security. Companies and indices are mentioned only to illustrate ideas. Do your own research; the decisions are yours.
市场 · 估值
估值并非公司本身的属性
从 A/H 股价差，到市场价格究竟是什么——以及为何更好的问题不是一只股票值多少钱，而是它的价格里已经相信了什么。
我们习惯于追问一家公司「真正」值多少钱。然而，在市场里待得够久，您会开始怀疑：这个问题本身，或许就略有偏差。
投资分析里有一种根深蒂固的直觉：公司存在一个内在价值，分析师的工作就是把它算出来，而市场价格只是围绕那个真实数字上下波动。这幅图景很有吸引力，因为它给人确定感。把营收、利润率、增长、资本回报率（ROIC）、资本成本、自由现金流和永续增长率输入模型，就会得到一个答案：这家公司值 80、120，或者 137.42。
但市场一再提醒我们，模型漏掉了一样东西。估值并不是在公司内部生成的，而是在公司与资本市场相遇之处生成的。
公司带来利润、现金流、资产、增长与竞争优势；市场带来资金、流动性、风险偏好、估值锚、持仓结构、规则、叙事与交易约束。股价，就是这两套系统每时每刻碰撞时所形成的东西。
同一家公司，两个市场，两种价格
同时在上海或深圳（A 股）与香港（H 股）上市的公司，是现成的、最干净的自然实验之一：同一家公司，同时在两地交易。管理层没有一分为二，工厂、品牌、营收和利润也没有——这些都不专属于某一个市场。理论上，两个市场看的是同一门生意。
然而这两个上市地却可能以相当不同的倍数交易，而且一连数年如此。中国功能饮料厂商东鹏饮料就是近期的例子：它的 A 股与 H 股有着肉眼可见的不同价格和市盈率。某一天的具体差距有多大并不要紧，要紧的是这个差距竟然存在。同样的利润，可以在不同市场被定出不同的价。
公司的盈利能力决定第一项。至于第二项——市场愿意给 12 倍、18 倍还是 30 倍——显然不是公司自己说了算。所以，像「这家公司值 15 倍盈利」这样的说法，值得再想一想。更准确的写法是：
市盈率并非公司本身的属性。它同时属于公司与市场。
一个上市地，就是一套坐标系
市场从来不是一块中立的价格显示屏。每个市场都有自己独特的资金构成、参与者和规则。
中国内地、香港与美国的投资者并非同一批人，他们对一只股票的要求也不一样——在风险、增长、行业稀缺性、公司治理、资本成本上各有偏好。一家中国消费公司在本土上市，可能被拿来与龙头白酒和食品饮料对标；在纽约上市，则被与全球消费品牌相比；在香港上市，则要透过全球美元流动性、中国资产的风险溢价，以及香港整体偏低的估值水平来定价。
把同一家公司放到不同市场上市，等于把它放进了不同的估值坐标系。改变的不只是股票在哪里交易，还有：谁来定价、他们拿什么对标、他们愿意承担多少风险，以及他们愿意为同一串未来现金支付多少倍。
这也是为什么，好公司从来不会自动等于好股票。一门生意可以非常出色，却只带来平庸的回报——如果市场早已为它支付了极高的倍数。而一门仅仅中等的生意，也可能是一笔不错的投资——如果价格里早已假定了最坏的情形，而现实没那么糟。
为什么「巴菲特指标」与市盈率讲着不同的故事
同样的逻辑也适用于整个市场。按「巴菲特指标」——美股总市值与 GDP 之比——美股正处于其历史区间的高位附近，席勒周期调整市盈率（CAPE）也离自身的极值不远。可一旦换成标普 500 的前瞻市盈率，或纳斯达克 100 的当期盈利倍数，画面看起来虽贵，却远不及 2000 年互联网泡沫顶峰那般极端。
这看似矛盾，其实不然。
高企的市值与 GDP 之比，不必全然来自高市盈率。它也可能来自：数十年间企业利润在 GDP 中占比的上升、美国上市公司越来越多的营收来自海外、最大型科技公司结构性更高的利润率。估值比率并不是脱离经济而独立存在的真理。每一个比率，都只是对市场的一种投影——一个观察的角度。
基本面分析并非无用——而是被误用
基本面分析自有其价值。它回答那些真正要紧的问题：这家公司是否真的在赚钱？利润是否持久？自由现金流是否真实？资本回报率是否足够高？护城河会不会被攻破？未来十年盈利能否增长？
它出问题，是在把这一切都硬塞进一个看似精确的「合理价格」之时。现金流折现（DCF）模型就是典型。它看上去数学严谨，但只要稍稍拨动长期增长率、折现率、利润率或永续假设，答案就可能剧烈摆动。
公式本身没错。麻烦在于，未来的自由现金流（FCF）与 r 并不是已知的常数。一个输出 137.42 的模型，并没有发现藏在自然里的客观数字。它真正说的是：如果关于现金流、增长、资本成本与终值的这些假设成立，我愿意把这家公司估在这个区间。这与市场实际会支付多少，是完全不同的两个问题。
现金流折现（DCF）并不能找出价值。它只是陈述了：在哪些条件下，您愿意接受某一价值。
价格承载的，远不止基本面
真实的市场价格，与其说像估值模型的输出，不如说更像这样：
基本面只是众多输入中的一个。一只股票今天上涨，可能是因为盈利预期上修——也可能是因为指数基金在买入、空头在回补、机构仓位偏低、行业叙事变了、利率下行、流动性改善、风险偏好上升，或者仅仅是卖方暂时离场。
更重要的是，这些输入没有一个是静止的。上涨的价格，会改变参与者的行为。
这正是「情绪」有意思的地方。它不仅仅是乐观对悲观，而是一套反馈系统：上涨的价格强化了某种叙事，叙事吸引资金入场，资金流入压低波动、抬升倍数，更高的价格又引来趋势追随者、被动资金和追逐业绩的人。价格不再只是结果——它成了下一轮价格形成的输入。从这个意义上说，市场是反身的。下跌时，这个循环会反向运转。
那么，市场价格就是「真实」价值吗？
如果「真实价值」指的是某种永恒的经济常数，那显然不是。一家公司可以在一天内因流动性紧缩、被迫去杠杆或某只大基金的赎回而暴跌 20%。它的工厂没有消失，客户没有离开，未来现金流也没有在几个小时里缩水五分之一。
但换一个问题——此时此刻，这一股究竟值多少？——市场价格就占据了其他任何方法都无法宣称的位置。它是唯一一个被真金白银的成交所确认过的数字。
DCF 是模型，目标价是观点，市盈率是比率，市净率是参照，EV/EBITDA 是一套话术。只有市场价格，是人们真正用现金完成交换的价格。更严格地说：
市场价格或许不等于客观的内在价值。但在任一时刻，它都是唯一可观测、可即时交易、并由真金白银验证过的出清价值。
价值至少有三种
| 概念 | 含义 |
|---|---|
| 经济价值 | 企业未来将为其所有者创造的经济利益。它确实存在，却无法直接观测——只能估计。 |
| 认知价值 | 市场参与者对那些未来利益的集体判断。 |
| 市场价格 | 这些判断在经过资金、持仓、流动性与交易机制之后的产物——即时的出清价格。 |
把这三者分开，许多投资争论便自行消解。基本面投资者谈的通常是经济价值，分析师争的是认知价值的合理区间，交易员每天面对的则是市场价格。它们并非彼此对立的观点，而是在描述同一系统的不同层次。
研究公司，不等于研究股票
经验丰富的交易员，常常会问估值模型很少问的问题：
- 此刻究竟是谁在买入——他们为何不得不买？
- 还有多少增量资金尚未入场？
- 谁的持仓过于拥挤，谁又可能被迫卖出？
- 市场愿意为这个故事支付多高的倍数？
- 估值体系正在扩张，还是收缩？
这并不是因为基本面不重要，而是因为「公司分析」和「股票分析」本就不是一回事。公司分析研究的是生意；股票分析，还必须研究参与者。
一只股票的回报，究竟来自哪里
把整段讨论压缩成一个框架，大致是这样：
第一项，是生意真正变好或变坏；第二项，是市场改变了对未来的看法；第三项，是市场愿意支付的倍数本身发生了变化。最猛烈的行情，很少只是第一项的功劳。一家公司盈利增长 20%，股价却翻倍：差额通常来自预期的抬升与倍数的扩张。反过来——盈利仍在增长，但市场把估值从 40 倍重估到 20 倍——股价照样可以大跌。
看对了公司，并不等于看对了股票。
更好的问题：价格里已经相信了什么？
传统估值钟爱一个问题：这家公司的合理价格是多少？更有用的，或许是：在今天的价格下，市场已经形成了怎样的共识？
两者听起来相似，实则天差地别。前者试图从公司内部推导出一个价格；后者则从价格反推出群体的集体预期。一旦接受「估值是在公司与市场之间生成的」，研究的重心就会转移——不再只是公司，还包括：
- 企业真实的经济能力；
- 当前价格已经假定了多少增长；
- 投资者正在使用哪些估值锚；
- 资金正流向何处；
- 市场正处于哪一种估值体系；
- 以及价格本身，正如何改变交易者的行为。
这些都不会让基本面失去意义。失去意义的，是把基本面模型当成一台能打印出「唯一正确价格」的机器。企业创造现金，投资者形成预期，资金决定边际成交，市场结构约束着交易如何发生。而价格，始终在这一切之中不断形成。
估值不是钉在公司上的一块铭牌。它是一个过程——而市场，正是这个过程。
在 TacticScale 上，反向读懂价格
TacticScale 的公允价值视图，正是围绕第二个问题构建的。它依然对每家公司运行同一套「所有者收益」模型，但把结果视为一组明示的假设，而非答案本身。
价格隐含了什么。一个市场隐含标记会反向解读模型：在今天的价格下，市场已经在定价多高的年化回报——因而是多高的增长？
现金，还是信念。无增长底价把价格拆成两部分：当前现金已经支撑的那一部分，以及押注于尚未兑现之增长的那一部分——后者对预期或估值体系的变化最为敏感。
群体自己的区间。对于流动性充足的标的，期权隐含曲线会显示交易者实际在为之付费的价格分布——即从期权链上直接读出的认知价值。
看看任一价格里已经相信了什么 → 这是模型测算，而非投资建议——决定权始终在您。
本文探讨市场结构与估值方法，仅供学习参考。它不构成投资建议，文中任何内容都不是买卖任何证券的推荐。提及的公司与指数仅用于说明观点。请自行研究；决定权在您。