无形资产对基础比率的影响
反观点全球洞察
Counterpoint Global Insights
无形资产的影响
The Impact of Intangibles
在基准比率方面
on Base Rates
CONSILIENT OBSERVER | 2021 年 6 月 23 日
CONSILIENT OBSERVER | June 23, 2021
引言 2017 年 3 月 25 日，《经济学人》报纸的封面展示了一幅城市景观，下方是一艘带有亚马逊标志的威胁性飞船。标题是“亚马逊帝国”。该期的一篇文章标题问道：“投资者是否对亚马逊过于乐观了？”此后四年间，该股票的年化涨幅达到 37.8%，而同期标普 500 指数的股东总回报率为 15.8%。这一增长转化为市值增加超过 1 万亿美元。
Introduction The cover of the Economist newspaper on March 25, 2017 showed a city landscape beneath a menacing spacecraft marked by the Amazon logo. The headline was, “Amazon’s empire.” An article in the issue had a title that asked, “Are investors too optimistic about Amazon?” 1 The stock would go on to appreciate 37.8 percent annually in the next four years versus a total shareholder return of 15.8 percent for the S&P 500 Index. That growth translated into an increase in market capitalization of more than $1 trillion.
文章正文引述了一位分析师对亚马逊的预测，称其销售额在 2025 年前将以 16% 的年复合增长率增长。值得注意的是，亚马逊 2016 年的总销售额为 1360 亿美元，这意味着该公司到预测期末的销售额将达到 5170 亿美元。
The body of the article cites an analyst who forecasted that Amazon would grow its sales at a 16 percent compound annual rate through 2025. Of note, Amazon’s total sales were $136 billion in 2016, suggesting the company’s sales at the end of the period would be $517 billion.
同一段落还提到我们在 2016 年所做的一项研究，该研究显示，在基年销售额达到 1000 亿美元或以上的公司中，从未有哪家公司能够以中等偏上十位数（约 15% 左右）的增长率持续那么长时间。2 我们的数据覆盖 1950 年至 2015 年，反映的是经通胀调整、但未针对收购和剥离进行调整的销售额数字。该分析并非针对特定企业，但明确的含义是：一家规模如此之大的公司，不可能以那样快的速度增长。如果市场普遍预测准确，到 2022 年第二季度，亚马逊的销售额年化运行率将超过 5150 亿美元，到 2022 年结束时的 6 年销售额复合增长率将达到 27.6%。这一增长率比该分析师看似“过于乐观”的观点还要高出 11 个百分点以上。如果亚马逊实现了这一业绩，其结果将
The same paragraph mentions work that we did in 2016 revealing that no company with $100 billion or more in base year sales had ever grown at that mid-teens rate for that long. 2 Our data were from 1950-2015 and reflected sales figures unadjusted for acquisitions and divestitures but adjusted for inflation. The analysis was not specific to any particular business, but the clear implication was that it was improbable that a company that big could grow that fast. Amazon will be at a $515 billion-plus sales run rate by the second quarter of 2022 and will have a 6-year sales growth rate ended 2022 of 27.6 percent, if the consensus estimates are accurate. That growth rate is more than 11 percentage points above the analyst’s seemingly “too optimistic” view. If achieved, Amazon’s results will
迈克尔·莫布森（Michael J. Mauboussin）[email protected] 丹·卡拉汉（Dan Callahan, CFA）[email protected]
Michael J. Mauboussin [email protected] Dan Callahan, CFA [email protected]
重新表述基础率数据。
recast the base rate data.
两种预测方式
Two Ways to Make a Forecast
预测未来——说到底是对未来做判断——大致有两种方法。第一种方法是因果思维，称为内视角。你大量收集与目标相关的信息，结合自己的投入和经验，然后投射到未来。分析师的模型就是这种方法的典型例子：分析师研究一家公司的业务，综合宏观经济因素、行业趋势和公司的竞争地位，来预测销售额和经营利润率。
There are broadly two ways to make a forecast, which is really a judgment about the future. 3 The first method is to think causally, which is called taking the inside view. You gather lots of information about what is of interest, combine it with your own input and experience, and project into the future. Analyst models are a good example of this approach. The analyst studies a company’s businesses and projects sales and operating profit margins based on a combination of macroeconomic factors, industry trends, and the company’s competitive position.
因果思维是一种自然而然产生的叙事形式。它是预测未来的有力方式，也是解释过去的令人信服的方法。我们的大脑极擅于编织顺口的故事，来解释周遭世界发生的一切。
Causal thinking is a form of storytelling that comes naturally. It is a compelling way to anticipate the future and a convincing way to explain the past. Our minds are great at creating facile narratives to explain what happens in the world around us.
第二种方法是统计思维，通常称为外部视角。这种方法不是基于因果联系编织故事，而是考察某个合适的参照系中过去发生过什么。这个参照系的结果被称为基础概率。于是，分析师构建模型时不是寻求因果联系，而是问自己：“那些与我正在研究的公司处境相似的企业，表现如何？”你不再依赖自己的经验，而是借用了别人的经验。
The second method is to think statistically, commonly referred to as the outside view. Rather than weaving a story based on causal links, the statistical approach examines what happened to an appropriate reference class of cases in the past. The results of the reference class are called base rates. Now the analyst builds her model not by seeking causal links but rather by asking, “how did other companies perform that were in a similar position to the one I am studying?” Instead of relying on your own experience, you tap the experience of others.
这种思维方式之所以不自然，是因为它依赖统计数据而非叙事。此外，基础概率可能并不现成。但研究表明，将内部视角与外部视角深思熟虑地结合，能够得出更准确的预测。
This type of thinking is unnatural because it features statistics rather than stories. Further, base rates may not be readily available. But research shows that a thoughtful combination of the inside and outside views leads to more accurate forecasts.4
有效运用基础概率的关键之一是找到恰当的参照系。在许多情况下，结果的分布是直观的——它们不会偏离平均值太远，极端值也很罕见。衡量公司业绩的指标，比如销售增长率，通常就属于这一类。对于遵循或近似钟形分布的结果，基础概率在评估时非常有效。
One of the keys to using base rates effectively is finding an appropriate reference class. In many instances, the distribution of outcomes is straightforward. In these cases, the outcomes don’t fall too far from the average, and outliers are rare. Measures of corporate results, such as sales growth rates, generally fit into this camp. Base rates are very effective for assessing outcomes that follow, or resemble, a bell-shaped distribution.
在其他情况下，结果的分布方差很大，平均数的概念没有意义，极端值会严重扭曲结果。例如，大多数书籍、歌曲和电影的销量都很一般，只有少数成为爆款。基础概率更难应用，但了解分布本身非常有价值。根据我们的经验，基础概率使用不足比过度使用更成问题。
In other instances, the distribution of outcomes has a variance that is large, the concept of an average is meaningless, and outliers skew the results.5 For example, most books, songs, and movies have very modest sales while only a handful are blockbusters. Base rates are more difficult to apply, but knowledge of the distribution itself is very useful. In our experience, underutilization of base rates is a bigger problem than overutilization of them.
需要牢记的一点是，恰当参照系的结果会随时间变化。亚马逊的销售增速即将超越过去 70 年间我们所见过的任何案例，这本身就证明了这一点。这意味着基础比率可以极具参考价值，但并非最终定论。事实上，有理由相信，由于商业本质已经发生改变，某些企业绩效指标正在发生偏移。
One important point to bear in mind is that outcomes of a proper reference class can change over time. 6 That Amazon’s sales growth is on pace to be greater than anything we have seen in the past 70 years proves the point. That means that base rates can be very instructive but are not the final word. Indeed, there is reason to believe that some measures of corporate performance are shifting because the nature of business has changed.
在一个无形资产的世界里，增长率
Growth Rates in a World of Intangible Assets
公司增长的基本路径，是通过投资赚取回报。回报用利润来衡量，而利润是销售额与利润率的乘积。投资可以是实物资产或无形资产。实物资产是你能触摸和感受到的东西，比如工厂、卡车和机器。无形资产没有物理形态，包括软件、可口可乐的秘方，以及某种救命药的配方。
The basic way that companies grow is by earning a return on investments. Return is measured by profits, which are the product of sales and margins. Investments can be tangible or intangible assets. Tangible assets are things you can touch and feel, such as factories, trucks, and machines. Intangible assets lack a physical existence and include software, the secret recipe for Coca-Cola, and the formulation of a life-saving drug.
企业有形资产与无形资产之间的一个重要区别在于可及性。同一时间，只能有一家公司使用有形资产，而无形资产则可以同时被多家公司使用。
One important distinction between corporate tangible and intangible assets is access. Only one company can use a tangible asset at a time, whereas many can use an intangible asset at the same time.7 In reality, the
这一区分在资产具有连续性的情况下并不那么分明。但核心要点是，共享一项无形资产的边际成本可能非常低。
distinction is less stark as assets fall on a continuum. But the main point is that the marginal cost of sharing an intangible asset can be very low.
无形资产有两个特点，对评估企业增长率很重要。第一，它们可以享受强大的规模经济效应，因为通常复制和共享成本很低。
Intangible assets have two characteristics that are important for considering corporate growth rates. 8 The first is that they can enjoy strong economies of scale because they are commonly cheap to reproduce and share.
规模经济是以单位成本衡量，且该成本随产出规模变化。以软件为例，最初的代码开发成本可能极其高昂，但随着销售量增加，单位成本会迅速下降，因为复制分享的成本很低。这是好消息。
Economies of scale are a measure of cost per unit as a function of output. Think of software as an example. The original code may be very expensive to produce but the cost per unit sold drops rapidly because it is inexpensive to share. That is the good news.
第二个是过时性及其相关的沉没概念。当一种更新、更好的版本出现，使旧版本变得过时时，无形资产的价值可能骤然下跌。而由于旧版本的价值极其有限，投资成本就成了沉没成本。我们继续用软件的例子来说明。
The second is obsolescence and the related concept of sunkenness. The value of intangible assets can drop precipitously when a new and better version comes along and makes the old version obsolete. And because the old version has very limited value, the investment cost is sunk. Let’s continue with our example of software.
一旦一家公司为电脑或手机推出新的操作系统，旧系统就几乎毫无意义或价值。这是坏消息。
Once a company introduces a new operating system for a computer or mobile phone, the old one is of little relevance or value. That is the bad news.
这些特点凸显了与增长率相关的有形资产与无形资产之间的差异。无形资产比有形资产更具可扩展性。这意味着依赖无形资产的成功公司，其增长速度可以快于基于有形资产的公司。随着整体投资结构从有形资产转向无形资产，我们应当预期，赢家的增长速度会比基础数据所显示的更快。
These characteristics highlight the contrast between tangible and intangible assets that are relevant for growth rates. Intangible assets are more scalable than tangible assets. That means successful companies that rely on intangible assets can grow faster than companies built on tangible assets. As the overall mix of investments shifts from tangible to intangible, we should expect to see faster growth rates for the winners than we have seen in the base rate data.
另一方面，过时意味着依赖无形资产的企业比那些以有形资产为基础的企业衰落得更快。无形资产很少是标准化的，不像有形资产那样，这意味着它们的残值有限。一家拥有过时软件的公司无法从中回收多少价值，而一家失败的商店则可以通过出售库存和家具来收回部分价值。这意味着，我们应该预期，失败者的增长率会更慢，或者衰退速度会大于基础数据所反映的水平。
On the other hand, obsolescence means that companies that rely on intangible assets can decline more rapidly than those built on tangible assets. Intangible assets are rarely standard, unlike tangible assets, which means they have limited salvage value. A company with obsolete software cannot get much for it while a company with a failed store can recoup some value by selling inventory and furnishings. This means that we should also expect to see slower growth rates, or a greater rate of decline, for the losers than the base rate data reflect.
这些关于增长率的观察很重要，因为近几十年来，整体投资支出已从以有形资产为主转向无形资产为主。我们估计，涵盖美国股票市场绝大多数可投资资产的罗素 3000 指数成分公司，在 2020 年的无形资产投资约为 1.8 万亿美元。这一金额是这些公司 8000 亿美元资本支出总额的两倍多。
These observations about growth rates are important because overall investment spending has shifted in recent decades from being predominately tangible to intangible. We estimate that intangible investments for companies in the Russell 3000, which captures the vast majority of the investable U.S. equity market, was around $1.8 trillion in 2020. This is more than double the $800 billion those companies spent on capital expenditures.
上述讨论提出了两个可供检验的假说。第一，以无形资产为核心的业务增速可能高于基准数据所显示的水平。本质上，增长率的分布曲线正在向均值右侧延伸——亚马逊的业绩为此提供了佐证案例。
This discussion suggests two hypotheses that we can test. The first is that intangible-based businesses can grow faster than what the base rate data show. In essence, the right tail of the distribution of growth rates is extending outward from the average. Amazon’s results provide anecdotal evidence for this.
第二个观察点是：以无形资产为核心的业务，其增长率分布的离散程度应该更大。这意味着增长率分布的左尾也在进一步偏离均值。黑莓公司（BlackBerry）就是一个典型的例子——在截至 2021 年 2 月的过去十年里，其年营收平均下降 26.7%。
The second is that we should observe greater variance in the distribution of growth rates for intangible-based businesses. That means that the left tail of the distribution of growth rates is also spreading further from the average. BlackBerry’s 26.7 percent average annual revenue decline in the past decade through February 2021 is a case in point.
这给投资者同时带来了好消息和坏消息。好消息是，会有一些企业的增长速度超出历史规律所暗示的水平，从而创造出机会。坏消息是，一些企业将失去其显赫地位，并以比前辈更快的速度走向衰落。基础比率依然极具参考价值，但我们必须在思维上保持灵活，承认企业群体随着时间推移已经发生了怎样的变化。
This provides investors with good and bad news. The good news is there will be some businesses that grow in excess of what history would suggest, creating opportunity. The bad news is some businesses will lose their positions of prominence and decline more rapidly than their predecessors did. Base rates remain extremely informative, but we must have the mental flexibility to acknowledge how the population of companies has changed over time.
基于无形资产密集度的基础概率
Base Rates Based on Intangible Asset Intensity
为检验这些观点，我们首先需要根据企业无形资产的密集程度进行分组。在近期的一篇论文中，三位金融学教授构建了一个模型，通过分析市场价格与并购交易数据来推算无形资产的价值。³ 相对账面价值而言的高市场价格，意味着部分无形资产未被确认。当一家公司收购另一家公司时，收购方的会计人员必须将收购价与有形资产之间的差额记为商誉或无形资产。他们以知识与组织资本的常用衡量指标作为基准，发现并购数据在估算无形资产价值方面的效果优于市场价格法。
To test these ideas, we first need to sort the companies based on their intangible asset intensity. In a recent paper, three finance professors built a model to infer the value of intangible assets by examining market prices and merger and acquisition (M&A) deals.9 High market prices relative to stated book values suggest a failure to recognize some intangible assets. When one company acquires another, the acquirer’s accountants have to record the difference between the purchase price and tangible assets as goodwill or intangible assets. Using popular measures of knowledge and organizational capital as a benchmark, they found that the M&A data did a better job of estimating the value of intangible assets than did the market price technique.
教授们将并购会计法应用于 1978 年至 2017 年间的大量公司样本，以找出无形资产密集度最高的领域。他们发现，按行业从高到低排序依次是：医疗健康、科技、消费品和制造业。他们还把约三分之一的公司归入“其他”类别，因为这些公司无法被整齐地划入上述任何一个行业。
The professors applied the M&A method to a large sample of companies from 1978-2017 to figure out where intangible asset intensity was highest. They found that the order of ranking from highest to lowest by industry was healthcare, technology, consumer, and manufacturing. They also placed about one-third of the companies into an “other” category because they didn’t fit neatly into one of the industries.
我们使用 1984 年至 2020 年罗素 3000 指数的成分股，计算了每个类别中公司的销售增长率中位数。我们还考察了分布的标准差（一种衡量离散程度的指标）。图表 1 显示了全样本的结果。
We calculated the median sales growth rate for companies in each of those categories using the constituents of the Russell 3000 from 1984-2020. We also examined the standard deviation, a measure of the dispersion, of the distributions. Exhibit 1 shows the results for the full sample.
表 1：各行业销售增长的基础概率，1984–2020 年
Exhibit 1: Base Rates for Sales Growth by Industry, 1984-2020
原件此处是表格，PDF 抽取时列结构已丢失，下面只剩按列读出的数字，行列对应关系无法还原。核对数据请打开来源正文。
| 行业 | 中位数复合年增长率 | 平均复合年增长率 | 标准差 | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 | |
| 医疗健康 | 11.5% | 10.8% | 10.4% | 9.3% | 52.6% | 16.8% | 12.6% | 9.3% | 406.3% | 45.9% | 30.6% | 22.5% |
| 科技 | 9.7% | 8.4% | 7.9% | 7.2% | 15.4% | 10.6% | 9.0% | 7.3% | 49.0% | 21.9% | 16.5% | 13.0% |
| 全部 | 7.4% | 6.9% | 6.5% | 6.2% | 16.6% | 9.5% | 8.0% | 6.7% | 177.3% | 23.2% | 16.4% | 12.0% |
| 消费品 | 6.9% | 6.4% | 6.0% | 5.9% | 13.5% | 8.9% | 7.7% | 6.6% | 164.7% | 18.8% | 13.9% | 9.5% |
| 制造业 | 5.4% | 5.1% | 5.0% | 5.5% | 9.3% | 6.8% | 6.1% | 6.0% | 50.4% | 17.5% | 13.1% | 9.4% |
| 其他 | 7.6% | 7.3% | 6.9% | 6.3% | 16.2% | 9.6% | 8.1% | 6.6% | 194.5% | 22.5% | 15.8% | 12.2% |
Median CAGR Mean CAGR Standard Deviation Industry 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr Healthcare 11.5% 10.8% 10.4% 9.3% 52.6% 16.8% 12.6% 9.3% 406.3% 45.9% 30.6% 22.5% Technology 9.7% 8.4% 7.9% 7.2% 15.4% 10.6% 9.0% 7.3% 49.0% 21.9% 16.5% 13.0% All 7.4% 6.9% 6.5% 6.2% 16.6% 9.5% 8.0% 6.7% 177.3% 23.2% 16.4% 12.0% Consumer 6.9% 6.4% 6.0% 5.9% 13.5% 8.9% 7.7% 6.6% 164.7% 18.8% 13.9% 9.5% Manufacturing 5.4% 5.1% 5.0% 5.5% 9.3% 6.8% 6.1% 6.0% 50.4% 17.5% 13.1% 9.4% Other 7.6% 7.3% 6.9% 6.3% 16.2% 9.6% 8.1% 6.6% 194.5% 22.5% 15.8% 12.2%
Source: FactSet.
Source: FactSet.
注：数据为截至年底的罗素 3000 指数成分股；增长率基于名义销售额；CAGR = 年复合增长率。
Note: Constituents of the Russell 3000 Index as of year-end; growth rates are based on nominal sales; CAGR=compound annual growth rate.
无论是从中位数还是平均值来衡量，销售增长率始终呈现出这样的规律：无形资产密集度最高的公司增速最高，密集度最低的公司增速最低。虽然短期数据存在波动，但这一关系在 1 年、3 年、5 年和 10 年区间内均成立。以五年期销售收入复合年增长率的中位数为观察指标：医疗保健行业增长 10.4%，科技行业 7.9%，消费品行业 6.0%，制造业 5.0%。全市场公司的中位数为 6.5%。
The sales growth rates, measured either as the median or average, consistently go from highest for companies that are most intangible-asset intentive to lowest for those that are least intensive. While the short-term numbers are noisy, the relationship holds true over 1-, 3-, 5-, and 10-year periods. Consider the median compound annual sales growth rates over five-year periods. Growth was 10.4 percent for healthcare, 7.9 percent for technology, 6.0 percent for consumer, and 5.0 percent for manufacturing. The median across all companies was 6.5 percent.
这一结论支持了第一个假设。
This supports the first hypothesis.
增长率的 标准差 也遵循同样的模式。在无形资产密集度高的行业，标准差同样很高。假设这些分布服从正态分布——这一假设虽不完美但具有说明性——那么大约三分之二的医疗保健公司，其 5 年销售额增长率介于 -18.0% 到 43.2% 之间。制造业公司的对应数值则是 -7.0% 到 19.2%。这与第二条论点一致。
The standard deviation of the growth rates follows the same pattern. Where intangible asset intensity is high, the standard deviation is also high. Assuming these distributions are normally distributed, an imperfect but illustrative assumption, about two-thirds of healthcare companies had 5-year sales growth rates between -18.0 and 43.2 percent. The comparable figures for manufacturing companies were -7.0 and 19.2 percent. This is consistent with the second thesis.
为了捕捉规模带来的潜在影响，我们根据起始年份的销售额将全市场样本划分为七个层级。从中浮现出几个规律。其一是，随着公司规模变大，平均和中等销售额增长率及其标准差往往都会下降。这一结论复现了已被实证确立的发现。
To capture the potential impact of size, we broke the universe into seven bins based on starting year sales. A couple of patterns emerge. The first is that the average and median sales growth rates and standard deviations tend to decline as companies get bigger. This replicates a finding that has been established empirically. 10 The
第二个发现是，高无形资产密度与高增长率之间的基本关系，在各类规模的公司群体中都普遍成立。（详见附录。）
second is that the basic relationship between high intangible asset intensity and high growth rates tends to hold across all size bins. (See the appendix for more detail.)
2020 年的全球疫情给世界卫生和经济增长带来了巨大挑战。一个积极面是，那些主要建立在无形资产之上的数字企业能够在混乱中蓬勃发展。我们考察了罗素 1000 指数（美国最大的 1000 家公司）中企业的销售增长率，以此判断哪些公司表现良好。医疗保健和科技，这两个无形资产密集度最高的行业，在增速排名前 20、前 50 和前 100 的公司中占比均超过 60%，尽管它们在总体公司中只占 29%（见图表 2）。
The global pandemic in 2020 was a substantial challenge for global health and economic growth. One silver lining was the ability of digital companies, built largely on intangible assets, to thrive in the chaos. We examined the sales growth rates of the companies in the Russell 1000, the largest one thousand companies in the U.S., to see which companies fared well. Healthcare and technology, the industries with the highest intangible asset intensity, represented over 60 percent of the top 20, 50, and 100 growers despite being only 29 percent of the universe (see exhibit 2).
表 2：2020 年增长最快的行业中，无形资产密集型行业名列前茅
Exhibit 2: Intangible Asset-Intense Industries Among the Fastest Growers in 2020
| 行业 | 前 20 名 | 前 50 名 | 前 100 名 | 全指数 |
|---|---|---|---|---|
| 医疗保健 | 8 | 12 | 22 | 81 |
| 科技 | 5 | 19 | 42 | 203 |
| 消费 | 4 | 8 | 15 | 189 |
| 制造 | 1 | 2 | 3 | 192 |
| 其他 | 2 | 9 | 18 | 327 |
| 总计 | 20 | 50 | 100 | 992 |
| 医疗保健 + 科技，数量 | 13 | 31 | 64 | 284 |
| 医疗保健 + 科技，占比 | 65% | 62% | 64% | 29% |
Industry Top 20 Top 50 Top 100 Full Index Healthcare 8 12 22 81 Technology 5 19 42 203 Consumer 4 8 15 189 Manufacturing 1 2 3 192 Other 2 9 18 327 Total 20 50 100 992 Healthcare + Technology, Number 13 31 64 284 Healthcare + Technology, Percent of Total 65% 62% 64% 29%
Source: FactSet.
Source: FactSet.
注：包含日历年 2019 年和 2020 年均有销售数据的公司。
Note: Includes companies with sales data for calendar years 2019 and 2020.
Conclusion
Conclusion
精确的预测需要将因果思维与统计思维进行恰当结合。统计思维依赖于识别出合适的过往结果参照系。如果参照系的统计特性随时间发生变化，过度依赖基础概率也可能导致错误的预测。尽管如此，我们认为预测者对基础概率的使用频率仍远未达到应有的水平。
Accurate forecasts combine causal and statistical thinking in proper measure. Statistical thinking relies on identifying an appropriate reference class of past outcomes. An overreliance on base rates can lead to faulty forecasts if the statistical properties of a reference class change over time. That said, we believe that forecasters don’t use base rates as frequently as they should.
企业通过投资产生回报来实现增长。近几十年来，投资的性质发生了显著变化，从以有形资产为主转变为以无形资产为主。无形资产有一些区别于有形资产的特点，包括更大的规模经济潜力和更高的过时风险。好消息是，无形资产密集型企业可以比有形资产密集型企业发展得更快。坏消息是，它们也可能迅速变得无关紧要并萎缩。
Companies grow by generating a return on investment. The nature of investment has changed markedly in recent decades, from one dominated by tangible assets to one mostly in the form of intangible assets. Intangible assets have some characteristics that distinguish them from tangible assets, including greater potential economies of scale and higher risk of obsolescence. The good news is that intangible-intensive companies can grow faster than their tangible counterparts. The bad news is they can also become irrelevant and shrink fast.
因此，我们应该会在数据中看到两种效应：更高的增长，以及结果之间更大的离散度。
As a consequence, we should see two effects in the data: higher growth and more dispersion in the outcomes.
我们对 1984 年至 2020 年罗素 3000 指数成份股公司业绩的分析揭示了这两个趋势。销售增长率的基本概率在正负两个方向上均较均值有所扩大。
Our analysis of the results from companies in the Russell 3000 from 1984-2020 reveals both of these. The base rate of sales growth is getting stretched from the average in both the positive and negative direction.
投资者可以从中得到两个主要教训。第一，必须意识到无形资产的崛起可能导致基准利率发生潜在变化。第二，精明的投资者或许有能力识别出那些增速将超出预期的公司，从而为获得可观回报提供可能。
There are two main lessons for investors. First, it is important to be mindful of the potential shift in the base rate as the result of the rise of intangibles. Second, skillful investors may be able to identify the companies that will grow faster than expected, hence providing the potential for attractive returns.
Appendix A
Appendix A
在将公司分入不同类别时，我们先把起始年份的销售额换算成 2020 年的美元价值。计算增长率时，所有数字均保留名义值。我们剔除基准年销售额低于 100 万美元（按 2020 年美元计算）的公司。
When sorting the companies into bins, we translate the starting year sales into 2020 dollars. When calculating the growth rates, we keep all figures in nominal terms. We exclude companies with base year sales of less than $1 million in 2020 dollars.
表 3：按规模与行业划分的销售增长率基础概率，1984–2020 年
Exhibit 3: Base Rates for Sales Growth by Size and Industry, 1984-2020
原件此处是表格，PDF 抽取时列结构已丢失，下面只剩按列读出的数字，行列对应关系无法还原。核对数据请打开来源正文。
| 全样本 | 年复合增长率中位数 | 年复合增长率均值 | 标准差 | 行业 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 医疗保健 | 11.5% | 10.8% | 10.4% | 9.3% | 52.6% | 16.8% | 12.6% | 9.3% | 406.3% | 45.9% | 30.6% | 22.5% | |||||
| 科技 | 9.7% | 8.4% | 7.9% | 7.2% | 15.4% | 10.6% | 9.0% | 7.3% | 49.0% | 21.9% | 16.5% | 13.0% | |||||
| 全行业 | 7.4% | 6.9% | 6.5% | 6.2% | 16.6% | 9.5% | 8.0% | 6.7% | 177.3% | 23.2% | 16.4% | 12.0% | |||||
| 消费品 | 6.9% | 6.4% | 6.0% | 5.9% | 13.5% | 8.9% | 7.7% | 6.6% | 164.7% | 18.8% | 13.9% | 9.5% | |||||
| 制造业 | 5.4% | 5.1% | 5.0% | 5.5% | 9.3% | 6.8% | 6.1% | 6.0% | 50.4% | 17.5% | 13.1% | 9.4% | |||||
| 其他 | 7.6% | 7.3% | 6.9% | 6.3% | 16.2% | 9.6% | 8.1% | 6.6% | 194.5% | 22.5% | 15.8% | 12.2% | |||||
| 样本数量 | 行业 | 1 年 | 3 年 | 5 年 | 10 年 | ||||||||||||
| 医疗保健 | 6,493 | 5,278 | 4,282 | 2,511 | |||||||||||||
| 科技 | 14,491 | 12,009 | 9,874 | 6,012 | |||||||||||||
| 全行业 | 90,725 | 76,369 | 64,252 | 41,145 | |||||||||||||
| 消费品 | 19,568 | 16,798 | 14,318 | 9,329 | |||||||||||||
| 制造业 | 19,659 | 17,333 | 15,139 | 10,544 | |||||||||||||
| 其他 | 30,514 | 24,951 | 20,639 | 12,749 | |||||||||||||
| 销售额：0–10 亿美元 | 年复合增长率中位数 | 年复合增长率均值 | 标准差 | 行业 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 | |
| 医疗保健 | 14.7% | 13.4% | 12.6% | 11.9% | 66.5% | 19.6% | 14.2% | 10.1% | 465.4% | 52.4% | 35.1% | 26.5% | |||||
| 科技 | 11.6% | 10.2% | 9.4% | 8.4% | 18.4% | 12.3% | 10.5% | 8.5% | 56.5% | 24.0% | 17.5% | 13.6% | |||||
| 全行业 | 9.7% | 9.0% | 8.4% | 7.8% | 24.3% | 12.5% | 10.3% | 8.3% | 237.7% | 28.6% | 19.8% | 14.4% | |||||
| 消费品 | 9.5% | 8.4% | 7.8% | 7.4% | 22.2% | 12.5% | 10.2% | 8.2% | 256.3% | 25.2% | 18.1% | 12.1% | |||||
| 制造业 | 7.3% | 7.2% | 6.9% | 7.2% | 13.9% | 9.9% | 8.5% | 8.0% | 66.5% | 21.5% | 15.8% | 11.1% | |||||
| 其他 | 9.2% | 8.9% | 8.4% | 7.6% | 21.8% | 12.0% | 10.0% | 8.0% | 248.9% | 26.2% | 17.8% | 13.4% | |||||
| 样本数量 | 行业 | 1 年 | 3 年 | 5 年 | 10 年 | ||||||||||||
| 医疗保健 | 4,930 | 3,945 | 3,151 | 1,748 | |||||||||||||
| 科技 | 10,157 | 8,346 | 6,795 | 4,062 | |||||||||||||
| 全行业 | 49,819 | 41,290 | 34,214 | 21,125 | |||||||||||||
| 消费品 | 7,992 | 6,811 | 5,744 | 3,603 | |||||||||||||
| 制造业 | 8,274 | 7,262 | 6,268 | 4,259 | |||||||||||||
| 其他 | 18,466 | 14,926 | 12,256 | 7,453 |
Full Universe Median CAGR Mean CAGR Standard Deviation Industry 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr Healthcare 11.5% 10.8% 10.4% 9.3% 52.6% 16.8% 12.6% 9.3% 406.3% 45.9% 30.6% 22.5% Technology 9.7% 8.4% 7.9% 7.2% 15.4% 10.6% 9.0% 7.3% 49.0% 21.9% 16.5% 13.0% All 7.4% 6.9% 6.5% 6.2% 16.6% 9.5% 8.0% 6.7% 177.3% 23.2% 16.4% 12.0% Consumer 6.9% 6.4% 6.0% 5.9% 13.5% 8.9% 7.7% 6.6% 164.7% 18.8% 13.9% 9.5% Manufacturing 5.4% 5.1% 5.0% 5.5% 9.3% 6.8% 6.1% 6.0% 50.4% 17.5% 13.1% 9.4% Other 7.6% 7.3% 6.9% 6.3% 16.2% 9.6% 8.1% 6.6% 194.5% 22.5% 15.8% 12.2% Count Industry 1-yr 3-yr 5-yr 10-yr Healthcare 6,493 5,278 4,282 2,511 Technology 14,491 12,009 9,874 6,012 All 90,725 76,369 64,252 41,145 Consumer 19,568 16,798 14,318 9,329 Manufacturing 19,659 17,333 15,139 10,544 Other 30,514 24,951 20,639 12,749 Sales: $0-1 Billion Median CAGR Mean CAGR Standard Deviation Industry 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr Healthcare 14.7% 13.4% 12.6% 11.9% 66.5% 19.6% 14.2% 10.1% 465.4% 52.4% 35.1% 26.5% Technology 11.6% 10.2% 9.4% 8.4% 18.4% 12.3% 10.5% 8.5% 56.5% 24.0% 17.5% 13.6% All 9.7% 9.0% 8.4% 7.8% 24.3% 12.5% 10.3% 8.3% 237.7% 28.6% 19.8% 14.4% Consumer 9.5% 8.4% 7.8% 7.4% 22.2% 12.5% 10.2% 8.2% 256.3% 25.2% 18.1% 12.1% Manufacturing 7.3% 7.2% 6.9% 7.2% 13.9% 9.9% 8.5% 8.0% 66.5% 21.5% 15.8% 11.1% Other 9.2% 8.9% 8.4% 7.6% 21.8% 12.0% 10.0% 8.0% 248.9% 26.2% 17.8% 13.4% Count Industry 1-yr 3-yr 5-yr 10-yr Healthcare 4,930 3,945 3,151 1,748 Technology 10,157 8,346 6,795 4,062 All 49,819 41,290 34,214 21,125 Consumer 7,992 6,811 5,744 3,603 Manufacturing 8,274 7,262 6,268 4,259 Other 18,466 14,926 12,256 7,453
原件此处是表格，PDF 抽取时列结构已丢失，下面只剩按列读出的数字，行列对应关系无法还原。核对数据请打开来源正文。
| 销售额：10 亿—50 亿美元 | 中位数复合年增长率 | 均值复合年增长率 | 标准差 | |||||||||
| 行业 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 |
| 医疗保健 | 8.3% | 8.3% | 8.1% | 7.7% | 9.7% | 9.2% | 9.3% | 8.6% | 18.7% | 12.2% | 9.8% | 7.0% |
| 科技 | 7.2% | 6.2% | 6.0% | 5.7% | 8.8% | 6.8% | 6.0% | 5.3% | 22.9% | 15.9% | 14.1% | 12.0% |
| 全部 | 6.1% | 5.7% | 5.6% | 5.6% | 8.2% | 6.7% | 6.2% | 5.8% | 30.1% | 14.2% | 11.4% | 9.1% |
| 消费品 | 6.3% | 6.1% | 5.9% | 5.9% | 8.4% | 7.2% | 6.8% | 6.3% | 21.7% | 13.1% | 10.4% | 7.8% |
| 制造业 | 4.7% | 4.5% | 4.6% | 5.1% | 6.9% | 5.4% | 5.2% | 5.5% | 39.0% | 13.8% | 10.7% | 8.1% |
| 其他 | 6.5% | 6.4% | 6.1% | 5.6% | 8.6% | 7.0% | 6.3% | 5.4% | 30.8% | 14.8% | 11.9% | 10.1% |
| 数量 | ||||||||||||
| 行业 | 1 年 | 3 年 | 5 年 | 10 年 | ||||||||
| 医疗保健 | 915 | 765 | 637 | 409 | ||||||||
| 科技 | 3,103 | 2,606 | 2,177 | 1,348 | ||||||||
| 全部 | 26,234 | 22,253 | 18,888 | 12,403 | ||||||||
| 消费品 | 6,940 | 5,938 | 5,053 | 3,343 | ||||||||
| 制造业 | 7,166 | 6,289 | 5,506 | 3,885 | ||||||||
| 其他 | 8,110 | 6,655 | 5,515 | 3,418 | ||||||||
| 销售额：50 亿—100 亿美元 | 中位数复合年增长率 | 均值复合年增长率 | 标准差 | |||||||||
| 行业 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 |
| 医疗保健 | 7.9% | 8.6% | 8.2% | 8.3% | 8.6% | 9.0% | 8.7% | 8.2% | 12.1% | 9.5% | 7.9% | 5.0% |
| 科技 | 5.3% | 4.9% | 3.9% | 2.5% | 7.4% | 6.5% | 5.4% | 4.4% | 22.1% | 15.4% | 13.1% | 10.4% |
| 全部 | 4.9% | 4.5% | 4.5% | 4.5% | 6.0% | 5.1% | 4.7% | 4.4% | 20.9% | 13.3% | 10.8% | 8.5% |
| 消费品 | 5.5% | 5.4% | 5.1% | 5.2% | 6.8% | 5.8% | 5.4% | 5.2% | 19.5% | 10.9% | 8.6% | 6.1% |
| 制造业 | 3.7% | 3.6% | 3.9% | 4.5% | 5.1% | 4.2% | 4.2% | 4.5% | 22.1% | 13.2% | 9.7% | 6.2% |
| 其他 | 4.9% | 4.5% | 4.4% | 3.7% | 5.6% | 4.4% | 3.9% | 3.1% | 21.2% | 15.2% | 13.4% | 12.0% |
| 数量 | ||||||||||||
| 行业 | 1 年 | 3 年 | 5 年 | 10 年 | ||||||||
| 医疗保健 | 194 | 166 | 138 | 79 | ||||||||
| 科技 | 519 | 441 | 367 | 232 | ||||||||
| 全部 | 6,453 | 5,573 | 4,764 | 3,155 | ||||||||
| 消费品 | 1,936 | 1,673 | 1,432 | 950 | ||||||||
| 制造业 | 1,933 | 1,710 | 1,494 | 1,045 | ||||||||
| 其他 | 1,871 | 1,583 | 1,333 | 849 | ||||||||
| 销售额：100 亿—250 亿美元 | 中位数复合年增长率 | 均值复合年增长率 | 标准差 | |||||||||
| 行业 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 |
| 医疗保健 | 7.3% | 7.2% | 6.9% | 6.3% | 7.9% | 7.5% | 7.1% | 6.5% | 14.7% | 9.0% | 7.0% | 6.1% |
| 科技 | 5.7% | 5.2% | 4.9% | 4.4% | 6.4% | 5.4% | 5.0% | 4.7% | 17.3% | 11.4% | 9.6% | 7.2% |
| 全部 | 4.8% | 4.0% | 3.8% | 3.6% | 5.4% | 4.3% | 3.9% | 4.0% | 20.0% | 11.5% | 9.1% | 6.5% |
| 消费品 | 5.1% | 4.6% | 4.4% | 4.5% | 5.9% | 5.2% | 4.9% | 5.0% | 16.9% | 10.6% | 8.7% | 6.6% |
| 制造业 | 3.2% | 2.5% | 2.4% | 2.7% | 4.0% | 2.8% | 2.4% | 2.5% | 21.0% | 11.5% | 8.8% | 6.1% |
| 其他 | 4.9% | 4.0% | 3.8% | 3.2% | 5.7% | 4.2% | 3.6% | 3.8% | 23.5% | 12.6% | 9.7% | 6.5% |
| 数量 | ||||||||||||
| 行业 | 1 年 | 3 年 | 5 年 | 10 年 | ||||||||
| 医疗保健 | 284 | 252 | 223 | 175 | ||||||||
| 科技 | 431 | 370 | 320 | 228 | ||||||||
| 全部 | 5,226 | 4,597 | 4,031 | 2,834 | ||||||||
| 消费品 | 1,582 | 1,383 | 1,208 | 819 | ||||||||
| 制造业 | 1,577 | 1,434 | 1,296 | 960 | ||||||||
| 其他 | 1,352 | 1,158 | 984 | 652 |
Sales: $1-5 Billion Median CAGR Mean CAGR Standard Deviation Industry 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr Healthcare 8.3% 8.3% 8.1% 7.7% 9.7% 9.2% 9.3% 8.6% 18.7% 12.2% 9.8% 7.0% Technology 7.2% 6.2% 6.0% 5.7% 8.8% 6.8% 6.0% 5.3% 22.9% 15.9% 14.1% 12.0% All 6.1% 5.7% 5.6% 5.6% 8.2% 6.7% 6.2% 5.8% 30.1% 14.2% 11.4% 9.1% Consumer 6.3% 6.1% 5.9% 5.9% 8.4% 7.2% 6.8% 6.3% 21.7% 13.1% 10.4% 7.8% Manufacturing 4.7% 4.5% 4.6% 5.1% 6.9% 5.4% 5.2% 5.5% 39.0% 13.8% 10.7% 8.1% Other 6.5% 6.4% 6.1% 5.6% 8.6% 7.0% 6.3% 5.4% 30.8% 14.8% 11.9% 10.1% Count Industry 1-yr 3-yr 5-yr 10-yr Healthcare 915 765 637 409 Technology 3,103 2,606 2,177 1,348 All 26,234 22,253 18,888 12,403 Consumer 6,940 5,938 5,053 3,343 Manufacturing 7,166 6,289 5,506 3,885 Other 8,110 6,655 5,515 3,418 Sales: $5-10 Billion Median CAGR Mean CAGR Standard Deviation Industry 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr Healthcare 7.9% 8.6% 8.2% 8.3% 8.6% 9.0% 8.7% 8.2% 12.1% 9.5% 7.9% 5.0% Technology 5.3% 4.9% 3.9% 2.5% 7.4% 6.5% 5.4% 4.4% 22.1% 15.4% 13.1% 10.4% All 4.9% 4.5% 4.5% 4.5% 6.0% 5.1% 4.7% 4.4% 20.9% 13.3% 10.8% 8.5% Consumer 5.5% 5.4% 5.1% 5.2% 6.8% 5.8% 5.4% 5.2% 19.5% 10.9% 8.6% 6.1% Manufacturing 3.7% 3.6% 3.9% 4.5% 5.1% 4.2% 4.2% 4.5% 22.1% 13.2% 9.7% 6.2% Other 4.9% 4.5% 4.4% 3.7% 5.6% 4.4% 3.9% 3.1% 21.2% 15.2% 13.4% 12.0% Count Industry 1-yr 3-yr 5-yr 10-yr Healthcare 194 166 138 79 Technology 519 441 367 232 All 6,453 5,573 4,764 3,155 Consumer 1,936 1,673 1,432 950 Manufacturing 1,933 1,710 1,494 1,045 Other 1,871 1,583 1,333 849 Sales: $10-25 Billion Median CAGR Mean CAGR Standard Deviation Industry 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr Healthcare 7.3% 7.2% 6.9% 6.3% 7.9% 7.5% 7.1% 6.5% 14.7% 9.0% 7.0% 6.1% Technology 5.7% 5.2% 4.9% 4.4% 6.4% 5.4% 5.0% 4.7% 17.3% 11.4% 9.6% 7.2% All 4.8% 4.0% 3.8% 3.6% 5.4% 4.3% 3.9% 4.0% 20.0% 11.5% 9.1% 6.5% Consumer 5.1% 4.6% 4.4% 4.5% 5.9% 5.2% 4.9% 5.0% 16.9% 10.6% 8.7% 6.6% Manufacturing 3.2% 2.5% 2.4% 2.7% 4.0% 2.8% 2.4% 2.5% 21.0% 11.5% 8.8% 6.1% Other 4.9% 4.0% 3.8% 3.2% 5.7% 4.2% 3.6% 3.8% 23.5% 12.6% 9.7% 6.5% Count Industry 1-yr 3-yr 5-yr 10-yr Healthcare 284 252 223 175 Technology 431 370 320 228 All 5,226 4,597 4,031 2,834 Consumer 1,582 1,383 1,208 819 Manufacturing 1,577 1,434 1,296 960 Other 1,352 1,158 984 652
原件此处是表格，PDF 抽取时列结构已丢失，下面只剩按列读出的数字，行列对应关系无法还原。核对数据请打开来源正文。
| 销售额：250 亿 - 500 亿美元 | 中位数复合年增长率 | 均值复合年增长率 | 标准差 | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 行业 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 |
| 医疗保健 | 5.5% | 5.2% | 4.4% | 5.3% | 7.6% | 6.2% | 5.4% | 4.7% | 15.8% | 9.7% | 7.9% | 4.6% |
| 科技 | 5.2% | 4.0% | 3.2% | 4.9% | 6.4% | 5.1% | 4.3% | 4.1% | 14.7% | 12.7% | 11.4% | 8.5% |
| 全部 | 4.6% | 3.7% | 3.4% | 3.7% | 5.2% | 4.2% | 3.6% | 3.7% | 17.6% | 12.2% | 9.6% | 6.6% |
| 消费 | 4.7% | 4.0% | 3.7% | 4.3% | 5.2% | 4.4% | 4.1% | 4.1% | 15.7% | 10.1% | 8.2% | 6.3% |
| 制造业 | 3.7% | 2.4% | 2.2% | 2.2% | 4.0% | 1.7% | 1.0% | 1.9% | 21.9% | 14.7% | 11.2% | 7.1% |
| 其他 | 4.6% | 4.1% | 3.5% | 3.5% | 5.4% | 5.4% | 4.6% | 4.5% | 17.0% | 12.2% | 8.9% | 6.0% |
| 样本数量 | ||||||||||||
| 行业 | 1 年 | 3 年 | 5 年 | 10 年 | ||||||||
| 医疗保健 | 93 | 80 | 71 | 60 | ||||||||
| 科技 | 145 | 130 | 118 | 86 | ||||||||
| 全部 | 1,714 | 1,524 | 1,359 | 976 | ||||||||
| 消费 | 614 | 544 | 486 | 360 | ||||||||
| 制造业 | 414 | 369 | 332 | 224 | ||||||||
| 其他 | 448 | 401 | 352 | 246 | ||||||||
| 销售额：500 亿 - 1000 亿美元 | 中位数复合年增长率 | 均值复合年增长率 | 标准差 | |||||||||
| 行业 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 |
| 医疗保健 | 3.5% | 1.7% | 2.0% | 1.6% | 2.6% | 1.0% | 1.3% | 2.2% | 17.1% | 10.4% | 7.4% | 4.4% |
| 科技 | 5.5% | 5.5% | 6.1% | 8.8% | 6.6% | 5.9% | 5.9% | 6.9% | 16.3% | 12.3% | 10.4% | 7.3% |
| 全部 | 4.3% | 3.5% | 3.3% | 3.9% | 5.1% | 3.1% | 2.9% | 2.9% | 21.7% | 12.3% | 9.7% | 7.0% |
| 消费 | 4.7% | 4.2% | 4.4% | 4.0% | 5.2% | 4.0% | 3.7% | 3.0% | 14.8% | 9.9% | 8.3% | 6.0% |
| 制造业 | 3.3% | 2.2% | 1.5% | 2.8% | 1.7% | -0.1% | 0.0% | 0.8% | 28.9% | 15.1% | 12.0% | 9.6% |
| 其他 | 5.0% | 3.3% | 3.7% | 5.2% | 8.4% | 4.6% | 4.1% | 4.3% | 25.8% | 12.8% | 9.1% | 5.2% |
| 样本数量 | ||||||||||||
| 行业 | 1 年 | 3 年 | 5 年 | 10 年 | ||||||||
| 医疗保健 | 61 | 56 | 50 | 33 | ||||||||
| 科技 | 77 | 63 | 48 | 22 | ||||||||
| 全部 | 797 | 712 | 629 | 416 | ||||||||
| 消费 | 298 | 274 | 250 | 169 | ||||||||
| 制造业 | 177 | 163 | 145 | 99 | ||||||||
| 其他 | 184 | 156 | 136 | 93 | ||||||||
| 销售额：超过 1000 亿美元 | 中位数复合年增长率 | 均值复合年增长率 | 标准差 | |||||||||
| 行业 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 | 1 年 | 3 年 | 5 年 | 10 年 |
| 医疗保健 | 5.5% | 5.7% | 6.1% | 7.9% | 7.4% | 7.3% | 7.4% | 7.8% | 7.5% | 4.1% | 2.9% | 0.4% |
| 科技 | 5.1% | 3.0% | 1.8% | 2.0% | 3.7% | 2.1% | 0.9% | 0.3% | 11.1% | 7.9% | 6.3% | 3.6% |
| 全部 | 4.1% | 3.4% | 2.9% | 2.2% | 3.2% | 2.6% | 2.1% | 1.5% | 13.8% | 9.0% | 7.4% | 5.9% |
| 消费 | 4.5% | 4.0% | 3.6% | 3.4% | 4.9% | 4.4% | 4.0% | 3.5% | 9.1% | 6.6% | 5.4% | 4.8% |
| 制造业 | 0.4% | 0.6% | 0.8% | 1.0% | -0.2% | -0.1% | 0.1% | 0.4% | 20.7% | 12.2% | 9.9% | 7.6% |
| 其他 | 4.4% | 0.4% | 0.7% | -1.3% | 2.4% | 1.4% | 0.9% | -1.2% | 13.1% | 8.7% | 7.0% | 4.4% |
| 样本数量 | ||||||||||||
| 行业 | 1 年 | 3 年 | 5 年 | 10 年 | ||||||||
| 医疗保健 | 16 | 14 | 12 | 7 | ||||||||
| 科技 | 59 | 53 | 49 | 34 | ||||||||
| 全部 | 482 | 420 | 367 | 236 | ||||||||
| 消费 | 206 | 175 | 145 | 85 | ||||||||
| 制造业 | 118 | 106 | 98 | 72 | ||||||||
| 其他 | 83 | 72 | 63 | 38 |
Sales: $25-50 Billion Median CAGR Mean CAGR Standard Deviation Industry 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr Healthcare 5.5% 5.2% 4.4% 5.3% 7.6% 6.2% 5.4% 4.7% 15.8% 9.7% 7.9% 4.6% Technology 5.2% 4.0% 3.2% 4.9% 6.4% 5.1% 4.3% 4.1% 14.7% 12.7% 11.4% 8.5% All 4.6% 3.7% 3.4% 3.7% 5.2% 4.2% 3.6% 3.7% 17.6% 12.2% 9.6% 6.6% Consumer 4.7% 4.0% 3.7% 4.3% 5.2% 4.4% 4.1% 4.1% 15.7% 10.1% 8.2% 6.3% Manufacturing 3.7% 2.4% 2.2% 2.2% 4.0% 1.7% 1.0% 1.9% 21.9% 14.7% 11.2% 7.1% Other 4.6% 4.1% 3.5% 3.5% 5.4% 5.4% 4.6% 4.5% 17.0% 12.2% 8.9% 6.0% Count Industry 1-yr 3-yr 5-yr 10-yr Healthcare 93 80 71 60 Technology 145 130 118 86 All 1,714 1,524 1,359 976 Consumer 614 544 486 360 Manufacturing 414 369 332 224 Other 448 401 352 246 Sales: $50-100 Billion Median CAGR Mean CAGR Standard Deviation Industry 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr Healthcare 3.5% 1.7% 2.0% 1.6% 2.6% 1.0% 1.3% 2.2% 17.1% 10.4% 7.4% 4.4% Technology 5.5% 5.5% 6.1% 8.8% 6.6% 5.9% 5.9% 6.9% 16.3% 12.3% 10.4% 7.3% All 4.3% 3.5% 3.3% 3.9% 5.1% 3.1% 2.9% 2.9% 21.7% 12.3% 9.7% 7.0% Consumer 4.7% 4.2% 4.4% 4.0% 5.2% 4.0% 3.7% 3.0% 14.8% 9.9% 8.3% 6.0% Manufacturing 3.3% 2.2% 1.5% 2.8% 1.7% -0.1% 0.0% 0.8% 28.9% 15.1% 12.0% 9.6% Other 5.0% 3.3% 3.7% 5.2% 8.4% 4.6% 4.1% 4.3% 25.8% 12.8% 9.1% 5.2% Count Industry 1-yr 3-yr 5-yr 10-yr Healthcare 61 56 50 33 Technology 77 63 48 22 All 797 712 629 416 Consumer 298 274 250 169 Manufacturing 177 163 145 99 Other 184 156 136 93 Sales: >$100 Billion Median CAGR Mean CAGR Standard Deviation Industry 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr 1-yr 3-yr 5-yr 10-yr Healthcare 5.5% 5.7% 6.1% 7.9% 7.4% 7.3% 7.4% 7.8% 7.5% 4.1% 2.9% 0.4% Technology 5.1% 3.0% 1.8% 2.0% 3.7% 2.1% 0.9% 0.3% 11.1% 7.9% 6.3% 3.6% All 4.1% 3.4% 2.9% 2.2% 3.2% 2.6% 2.1% 1.5% 13.8% 9.0% 7.4% 5.9% Consumer 4.5% 4.0% 3.6% 3.4% 4.9% 4.4% 4.0% 3.5% 9.1% 6.6% 5.4% 4.8% Manufacturing 0.4% 0.6% 0.8% 1.0% -0.2% -0.1% 0.1% 0.4% 20.7% 12.2% 9.9% 7.6% Other 4.4% 0.4% 0.7% -1.3% 2.4% 1.4% 0.9% -1.2% 13.1% 8.7% 7.0% 4.4% Count Industry 1-yr 3-yr 5-yr 10-yr Healthcare 16 14 12 7 Technology 59 53 49 34 All 482 420 367 236 Consumer 206 175 145 85 Manufacturing 118 106 98 72 Other 83 72 63 38
Source: FactSet.
Source: FactSet.
注：数据截至年末，成分股为罗素 3000 指数；增长率基于名义销售额。
Note: Constituents of the Russell 3000 Index as of year-end; growth rates are based on nominal sales.
Appendix B
Appendix B
我们采用尤金·法马和肯尼斯·弗伦奇五行业分类的微调版本，将公司划分为五个组别。法马和弗伦奇依据公司的标准行业分类（SIC）代码进行划分。我们则遵循迈克尔·尤恩斯所做的细微调整——他将医院从医疗保健行业重新划归消费者行业，同时将广播和电视提供商从科技行业调整至消费者行业。11 下表列出了五大行业所包含的主要子行业。
We sort companies into five groups using a slightly modified version of the five-industry classification of Eugene Fama and Kenneth French. Fama and French assign companies based on their Standard Industrial Classification (SIC) codes. We follow the minor modifications of Michael Ewens, who reassigns hospitals from healthcare to consumer, as well as radio and TV providers from technology to consumer.11 The table below shows the main sub-industries included in the five primary industries.
行业名称 子行业 消费 耐用消费品、非耐用消费品、批发、零售、医院、部分服务业 制造 制造业、能源、公用事业 科技 电子、计算机硬件与软件、电信 医疗 医疗健康、医疗设备、制药 其他 其他所有行业，包括矿业、建筑、建材、运输、酒店、商业服务、娱乐、金融
Industry Name Sub-Industries Consumer Consumer durables, nondurables, wholesale, retail, hospitals, some services Manufacturing Manufacturing, energy, utilities Technology Electronics, computer hardware and software, telecommunications Healthcare Healthcare, medical equipment, pharmaceuticals Other Everything else, including mines, construction, building materials, transportation, hotels, business services, entertainment, finance
尾注 1 “Are investors too optimistic about Amazon?”《经济学人》，2017 年 3 月 25 日。
Endnotes 1 “Are investors too optimistic about Amazon?” Economist, March 25, 2017.
2 Michael J. Mauboussin、Dan Callahan 和 Darius Majd 合著的《基础率手册：整合过去以更好地……》（The Base Rate Book: Integrating the Past to Better……）
2 Michael J. Mauboussin, Dan Callahan, and Darius Majd, “The Base Rate Book: Integrating the Past to Better
“预期未来，”瑞信全球金融策略，2016 年 9 月 26 日。
Anticipate the Future,” Credit Suisse Global Financial Strategies, September 26, 2016.
3 丹尼尔·卡尼曼、奥利维耶·西博尼和卡斯·R·桑斯坦，《噪声：人类判断的缺陷》（纽约：利特尔，
3 Daniel Kahneman, Olivier Sibony, and Cass R. Sunstein, Noise: A Flaw in Human Judgment (New York: Little,
Brown Spark, 2021), 156-158.
Brown Spark, 2021), 156-158.
4 Mauboussin, Callahan, 和 Majd，《基础利率手册》，7-9 页。
4 Mauboussin, Callahan, and Majd, “The Base Rate Book,” 7-9.
这段话也可以理解为温和、缓慢且狂野的随机性。参见伯努瓦·B·曼德尔布罗特的《分形与……
5 This can also be understood as mild, slow, and wild randomness. See Benoit B. Mandelbrot, Fractals and
《金融中的规模效应：不连续性、集中度与风险》（纽约：斯普林格出版社，1997 年），第 117-125 页。
Scaling in Finance: Discontinuity, Concentration, and Risk (New York: Springer, 1997), 117-125.
这意味着生成该分布的整个过程并非平稳的。在平稳序列中，统计
6 This means that the process that generates the distribution is not stationary. In a stationary series, the statistical
均值与标准差这类属性不会随时间改变。我们讨论的分布虽然并非一成不变，但稳定到足以产生价值。
properties such as the mean and standard deviation do not change over time. The distributions we are discussing are not stationary but stable enough to be beneficial.
这类商品的术语叫做“竞争性商品”。
7 The technical term for this is a rivalrous good.
如需更深入地了解无形资产的特征，请参阅乔纳森·哈斯克尔（Jonathan Haskel）与斯蒂安
8 For a more thorough review of the characteristics of intangible assets, see Jonathan Haskel and Stian
Westlake，《没有资本的资本主义：无形经济的崛起》（普林斯顿，新泽西：普林斯顿大学出版社，2017 年），第 56–88 页。
Westlake, Capitalism Without Capital: The Rise of the Intangible Economy (Princeton, NJ: Princeton University Press, 2017), 56-88.
迈克尔·尤恩斯、瑞安·H·彼得斯和肖恩·王合著的《用市场价格衡量无形资本》，工作论文。
9 Michael Ewens, Ryan H. Peters, and Sean Wang, “Measuring Intangible Capital with Market Prices,” Working
Paper, October 2020.
Paper, October 2020.
10 Michael H. R. Stanley, Luís A. N. Amaral, Sergey V. Buldyrev, Shlomo Havlin, Heiko Leschhorn, Philipp
10 Michael H. R. Stanley, Luís A. N. Amaral, Sergey V. Buldyrev, Shlomo Havlin, Heiko Leschhorn, Philipp
Maass, Michael A. Salinger, 和 H. Eugene Stanley,《公司增长中的标度行为》（Scaling Behaviour in the Growth of Companies）,《自然》（Nature）,第 379 卷，1996 年 2 月 29 日，第 804–806 页。另见 Rich Perline、Robert Axtell 和 Daniel Teitelbaum,《随时间跨度增加小企业增长率的波动性与不对称性》（Volatility and Asymmetry of Small Firm Growth Rates Over Increasing Time Frames）,《小企业研究摘要》（Small Business Research Summary）,第 285 号，2006 年 12 月。
Maass, Michael A. Salinger, and H. Eugene Stanley, “Scaling Behaviour in the Growth of Companies,” Nature, Vol. 379, February 29, 1996, 804-806. Also, Rich Perline, Robert Axtell, and Daniel Teitelbaum, “Volatility and Asymmetry of Small Firm Growth Rates Over Increasing Time Frames,” Small Business Research Summary, No. 285, December 2006.
11 Michael Ewens。详见 https://github.com/michaelewens/Intangible-capital-stocks。
11 Michael Ewens. See https://github.com/michaelewens/Intangible-capital-stocks.
术语定义
DEFINITIONS OF TERMS
投资回报率是一项用于评估某笔投资效率或比较多种不同投资效率的绩效衡量指标。
Return on investment is a performance measure used to evaluate the efficiency of an investment or to compare the efficiency of a number of different investments.
罗素 3000® 指数衡量美国最大 3000 家公司的表现，约占美国可投资股票市场的 98%。罗素 3000 指数的构建旨在提供一个全面、无偏见且稳定的广泛市场晴雨表，并且每年完全重新调整，以确保新上市和成长中的股票都能被反映出来。
The Russell 3000® Index measures the performance of the largest 3,000 U.S. companies, representing approximately 98% of the investable U.S. equity market. The Russell 3000 Index is constructed to provide a comprehensive, unbiased, and stable barometer of the broad market and is completely reconstituted annually to ensure new and growing equities are reflected.
罗素 1000® 指数衡量罗素 3000® 指数中最大 1000 家公司的表现。
The Russell 1000® Index measures the performance of the 1,000 largest companies in the Russell 3000® Index.
标普 500 指数衡量美国股票市场中大盘股板块的表现，覆盖美国股票市场约 80% 的市值。该指数包含美国经济领先行业中的 500 家龙头企业。
The S&P 500® Index measures the performance of the large cap segment of the U.S. equities market, covering approximately 80% of the U.S. equities market. The Index includes 500 leading companies in leading industries of the U.S. economy.
股东总回报反映了股价变化加上股息再投资。
Total shareholder return reflects the change in stock price plus reinvested dividends.