It was a disquieting year , as political and economic news stories shook the foundations of the post-war economic order, built around global trade and the US dollar. In fact, if you had been read just the news all through the year, and were shielded from financial markets, and been asked what stocks did during the year, you would have guessed, based on the news, that they had a bad year. You would have been wrong, though, as equity markets proved resilient (yet again) and delivered another solid year of returns for investors. In this post, I will focus on US equities, starting with the indices, and then deconstructing the data to see the differences in the cross section. As has been my practice for the last few years, I will also use this post to update the equity risk premium for the S&P 500, my composite indicator for whether the market is richly priced or not, and estimate a value for the index, with a "reasonable" equity risk premium.
Back from the Brink: US Equities in 2025
At the start of 2025, the consensus view was that stocks were primed to do well, helped by what investors perceived would be a business-friendly administration and a Federal Reserve, ready to cut rates. In keeping with Robert Burn's phrase that the best-laid plans of mice and men go awry, the year did not measure up to those expectations at least in terms of policy and rate changes, but stocks still managed to find a way through. Let's start with a look at the S&P 500 and the NASDAQ, day-to-day through the year:
The first few weeks of 2025 saw of continuation of the momentum built up after the 2024 elections and stock prices continued upwards, but February and March saw a drawdown in stock prices as talk of tariffs and trade wars heated up before culminating in a dramatic sell off in early April, after liberation day, when breadth and magnitude of the tariffs blindsided markets. The sell off was brutal and short, and stocks hit their low point for the year on April 11, 2025. Over the next few months, stocks mounted a comeback, before leveling off at the end of September and coasting for the rest of the year. Early in the year, the S&P 500 held its value better than the NASDAQ, generating talk of a long-awaited tech sell off, but as stocks recovered in the subsequent months, the NASDAQ ended up moving ahead the S&P 500.
Across the entire year, the S&P 500 rose from 5881.6 to 6845.5, delivering price appreciation of 16.39% for the year. The dividends on the companies in the index for the year, based upon dividends in the first three quarters of 2025 and estimates for dividends in the last quarter amounted added a yield of 1.34%.
The S&P 500's return in 2025 of 17.72% was a solid year, but to provide perspective on how it measures up to history, I looked at annual returns from US stocks from 1928 to 2025, and computed distributional statistics:
While 2025's annual returns put it in the right in the middle of the distribution, close to the median and ranked 45th of the 98 years of US equity returns from 1928-2026, it represented a third consecutive year when the annual stock return exceeded the median returns, the longest streak since the mid 1990s; US equities between 2023 and 2025, a period where many market timers were suggesting not just caution but staying out the market, returned 85.32% to investors.
Deconstructing US Stock Price Performance
While stocks had a good year overall, the spoils were dividend unequally, as if often the case, across industries and sectors. To take a closer look at where the best and worst performance was in 2025, I started by looking at a breakdown by sector, where I computed the returns based on the change in aggregate market capitalization in 2025:
I have tracked the performance of each sector, by quarter, and across the year a measured the returns. The best performing sector in percentage returns was communication services (which includes Alphabet and Meta), up 30.63% for the year, followed by technology, which continued it sustained run of success by delivering 23.65% as an annual return; on a dollar value basis, it was not close with technology companies posting an increase of $4.17 trillion in market cap during the year. The worst performing sectors were consumer staples and real estate where the returns were about 2% for the year.
The problem with sector categorizations is the they are overly broad and include very diverse industry groupings, and to overcome that problem, I looked at returns by industry, with a breakdown into 95 industry groups. While you can find the full list at the end of this post, I ranked the industry returns in 2025, from best to worst, and extract the ten best and worst performing industry groups:
The surge in gold and silver prices in 2025 carried precious metals companies to the top of the list, with a return of 169.2% for the year, and other energy and mining companies also made the best performer list, with a scattering of technology standouts. The worst performing businesses were primarily old economy, with chemicals, consumer product companies and food processing all struggling during the year.
One of the major changes that we have seen in cross sectional differences in the twenty first century has been the fading or even disappearance of two well documented phenomena from the twentieth century, the first being the small cap premium, where small market cap companies delivered much higher risk-adjusted returns that large market cap companies, and the value premium, where low price to book stocks beat high price to book stocks in the return game. I focused in how these categorizations behaved in 2025, and we did see small cap stocks and low price to book stocks return, at least in part, to favor:
If you are small cap or a value investor, though, I would not be celebrating the return on these premia, but I do think that we will start to see a return to balance, where the groupings will trade off winning in some years for losing in others.
As a final assessment, I did look at the seven stocks that have not only carried the market for the last few years, the Mag Seven, but have been the source of much hand wringing about how markets are becoming top-heavy and concentrated. I started by looking at the individual companies, and how they performed in 2025:
While the Mag Seven saw their collective market capitalization increase by 22.36%, Apple and Amazon lagged with single digit increases, and Nvidia (up 37.8%) and Alphabet (up 62.7%) for the year. Increasingly, the Mag Seven are diverging in their price paths, and that should be expected since they operate in very different businesses and have very different management running them. To examine how much the Mag Seven have carried the market, I tracked the market cap of the Mag Seven against the rest of US equity (close to 6000 companies) from 2014 through the four quarters of 2025.
The aggregate market cap of the Mag Seven has increased from 11% of the US equity market (composed of close to 6000 stocks) in 2014 to 30.89% of the market at the end of 2025, with the $3.9 billion in market cap added in 2025 accounting for 39.3% of the overall increase in market capitalization of all US equities during the year. While this Mag Seven party will undoubtedly end at some point, it did not happen in 2025.
US Equities: Too high, too low or just right?
This post, at least so far, has been a post mortem of the year that was, but investing is always about the future, and the question that we all face as investors, is where stocks will go this year. In my unscientific assessment of stock market opinion, from experts and market timers, there seems to a decided tilt towards bearishness at the start of 2026, for a variety of reasons. There are some who note that having had three good years in a run, stocks will take breather. Others point to history and note that stocks generally don't do well in the second years of presidential terms. The most common metric that bearish investors point to, though, is the PE ratio for stocks at the start of 2026 is pushing towards historic highs, as can be seen in the graph below, where I look at three variants on the PE ratio - a trailing PE, where I divide the index by earnings in the most recent 12 months, a normalized PE, where I divide the index by the average earnings over the last ten years and a Shiller PE, where I average inflation-adjusted earnings over the last ten years:
Using every PE ratio measure, it is undeniable that the PE ratio for the S&P 500, at the start of 2026, is much higher than it has been at any extended period in history, perhaps with the exception with the late 1990s. While this may sound like a slam dunk argument for US stocks being over priced, it is worth remembering that this indicator would have suggested staying out of US equities for much of the last decade. The problem with the PE pricing metric is that it is noisy and an unreliable indicator, and before you use it to build a case that equity investors in the US have become irrational, you may want to consider reasons why US stocks have benefited able to fight the gravitational forces of mean reversion.
1. Robust Earnings Growth & Earnings Resilience: In this century, US stocks have increased more than four-fold, with the S&P 500 rising from 1320.28 at the end of 2000 to 6845.5 at the end of 2025, but it is also worth noting that US companies have also had a solid run in earnings, with earnings increasing about 356% during that same time period.
It is also notable that not only did earnings register strong growth over this period, there were only three years in this century when earnings declined - 2001 (dot com bust), 2009 (2008 crisis) and 2020 (Covid). US companies have become more resilient in terms of delivering earnings through recessions and other crises, pointing to perhaps less risk in equities. I will return in a later post to examine why that may be, with some of the answers rooted in changes in US equity market composition and some in management behavior.
2. Healthy cash returns: In conjunction with delivering earnings growth, US companies have also been returning large amounts of cash to their shareholders, albeit more in buybacks than in conventional dividends. In 2025, the companies in the S&P 500 alone returned more than a trillion dollars in cash flows in buybacks, and in the graph below, I look at how the augmented cash yield (composed of dividends and buybacks) has largely sustained the market:
While the dividend payout ratio, computed using only dividends, has been on a downward trend all through this century, adding buyback to dividends and computing a cash yield ratios yields values that are comparable to what dividend yields used to be, before the buyback era.
In sum, you can see why both bulls and bears retreat to their favored arguments, and there is no obvious tie breaker. The level of stock prices (PE ratios) should be a concern, but you cannot dismiss the benefits of growing and resilient earnings, and substantial cash return. To break the tie, in a very self serving away, I will revert to my favored metric for the US equity market, the implied equity risk premium, which in addition to looking at stock price levels, the growth in earnings and the cash return, also brings in the level of rates. The implied equity risk premium, as I compute it, is the based upon the index level and the expected cashflows (from dividends and buybacks, augmented by earnings growth), and very simply, is an internal rate of return for stocks. Netting out the riskfree rate yields an equity risk premium. The table below contains the computation of the implied ERP at the start of 2026:
Given the index level on January 1, 2026, of 6845.5, and the expected cash flows that I computed on that date (using the dividends and buybacks in the trailing 12 months as my starting point, and growing them at the same rate as earnings), I obtain an expected return on stocks of 8.41%. Subtracting out the US T. Bond rate (dollar riskfree rate) of 4.18% (3.95%) on that day yields an equity risk premium of 4.23% (4.46%) for the US. I want to emphasize again that this estimate is entirely a market-driven number and is model-agnostic.
If you are wondering how estimating this numbers lets you make a judgment on whether US stocks are over priced, all you need to reframe the equity risk premium by asking whether the current ERP is, in your view, too high, too low or just right.
If you believe that the market is pricing in too low an ERP, given the risks that are on the horizon, you are contending the stocks are over priced.
If your view is that the current ERP is too high, that is equivalent to arguing that stocks today are under priced.
If you are not a market timer, you are in effect arguing that the current ERP is, in fact, the right ERP for the market.
To illustrate this point, I have estimated the value of the index at equity risk premiums ranging from 2% to 6%:
With a 2% equity risk premium, you get an astounding value of 14834 for the S&P 500, which would make the index undervalued by 53%. At the other end of the spectrum, with a 6% equity risk premium, the index should trade at 4790, translating into an overvaluation of 43%.So, is the ERP of 4.23% (I will revert to this number, since my historical numbers did use the US treasury bond rate as the riskfree rate) at the start of 2026 a high, low or just-right number? Rather than make that judgment for you, I have computed the implied ERP for the S&P 500 going back to 1960:
There is something in this graph that almost every investor group can take comfort in, If you are market neutral, you will take comfort from the fact that the current ERP is almost exactly equal to the average for the 1960-2025 period. If you are bearish you will point to the fact that the ERP now is lower than it has been in the post-2008 period, backing up your case that an adjustment is overdue. I am leery of the bubble word, especially used in the context of this market, since unlike the end of 1999, when the ERP got as low as 2.05%, the current ERP is more in the middle of the historic range.
The Bottom Line
US equities had a good year in 2025, and there are signs of excess in at some parts of the market, especially related to AI. That said, the capacity of US companies to continue to deliver earnings and return cash flows even in the face of a tsunami of bad news continues to sustain the market. I am, at my core, a non market-timer, but I have held back on putting idle cash back into US equities in the last year, preferring to keep that cash in treasury bills. It is entirely possible that the market will continue to prove the naysayers wrong and post another strong year, but much as it may pain equity investors, the healthiest development for the market would be for it to deliver a return roughly equal to its expected return (8-9%) and clean up on pricing overreach along the way. For the bears, this may also be the year when the bad news stories of last year, including tariffs and political whiplash, will finally start to hit the bottom line, reducing aggregate earnings and cash flows, but waiting on the sidelines for this to happen has not been a good strategy for the last decade.
In my musings on valuation, I have long described myself as more of a number cruncher than a storyteller, but it is because I love numbers for their own sake, rather than a fondness for abstract mathematics. It is that love for numbers that has led me at the beginning of each year since the 1990s to take publicly available data on individual companies, both from their financial statements and from the markets that they are listed and traded on, and try to make sense of that data for a variety of reasons - to gain perspective, to use in my corporate financial analysis and valuations and to separate information from disinformation . As my access to data has improved, what started as a handful of datasets in my first data update in 1994 has expanded to cover a much wider array of statistics than I had initially envisioned, and my 2026 data updates are now ready. If you are interested in what they contain, please read on.
The Push and Pull of Data
After a year during which we heard more talk about data and data centers than ever before in history, usually in the context of how AI will change our lives, it is worth considering the draw that data has aways had on not just businesses but on individuals, as well as the dangers with the proliferation of data and the trust we put on that data.
In a world where we feel adrift and uncertain, the appeal of data is clear. It gives us a sense of control, even if it is only in passing, and provides us with mechanisms for making decisions in the face of uncertainty.
Signal in the noise: Anyone who has to price/value a stock or assess a project at a firm has to make estimates in the face of contradictions, both in viewpoints and in numbers. The entire point of good data analysis is to find the signals in the noise, allowing for reasoned judgments, albeit with the recognition that you will make mistakes.
Coping mechanism for uncertainty: Investors and businesses, when faced with uncertainty, often respond in unhealthy ways, with denial and paralysis as common responses. Here again, data can help in two ways, first by helping you picture the range of possible outcomes and second by bringing in tools (simulations, data visualizations) for incorporating uncertainty into your decision-making.
Prescription against tunnel vision: It is easy to get bogged down in details, when faced with having to make investment decisions, and lose perspective. One of the advantages of looking at data differences over time and across firms is that it can help you elevate and regain perspective, separating the stuff that matters a lot from that which matters little.
Shield from disinformation: At the risk of getting backlash, I find that people make up stuff and present it as fact. While it is easy to blame social media, which has provided a megaphone for these fabulists, I read and hear statements in the media, ostensibly from experts, politicians and regulators, that cause me to do double takes since they are not just wrong, but easily provable as wrong, with the data.
While data clearly has benefits, as a data-user, I do know that it comes with costs and consequences, and it behooves us all to be aware of them.
False precision: It is undeniable that attaching a number to something that worries you, whether it be your health or your finances, can provide a sense of comfort, but there is the danger with treating estimates as facts. In one of my upcoming posts, for instance, I will look at the historical equity risk premium, measured by looking at what stocks have earned, on an annual basis, over treasury bonds for the last century. The estimate that I will provide is 7.03% (the average over the entire period), but that number comes with a standard error of 2.05%, resulting in a range from a little less than 4% (7.03% - 2 × 2.05%) to greater than 11%. This estimation error plays out over and over again in almost every number that we use in corporate finance and valuation, and while there is little that can be done about it, its presence should animate how we use the data.
The Role of Bias: I have long argued that we are all biased, albeit in varying degrees and in different directions, and that bias will find its way into the choices we make. With data, this can play out consciously, where we use data estimates that feed into our biases and avoid estimates that work in the opposite direction, but more dangerously, they can also play out subconsciously, in the choices we make. While it is true that practitioners are more exposed to bias, because their rewards and compensation are often tied to the output of their research, the notion that academics are somehow objective because their work is peer-reviewed is laughable, since their incentive systems create their own biases.
Lazy mean reversion: In a series of posts that I wrote about value investing, at least as practiced by many of its old-time practitioners, I argued that it was built around mean reversion, the assumption that the world (and markets) will revert back to historic norms. Thus, you buy low PBV stocks, assuming (and hoping) that those PBV ratios will revert to market averages, and argue that the market is overpriced because the PE ratio today is much higher than it has been historically. That strategy is attractive to those who use it, because mean reversion works much of the time, but it is breaks down when markets go through structural shifts that cause permanent departures from the past.
The data did it: As we put data on a pedestal, treating the numbers from emerge from it as the truth, there is also the danger that some analysts who use it view themselves as purely data engineers. While they make recommendations based upon the data, they also refuse to take ownership for their own prescriptions, arguing that it is the data that is responsible.
As the data that we collect and have access to gets richer and deeper, and the tools that we have to analyze that data become more powerful, there are some who see a utopian world where this data access and analysis leads to better decisions and policy as output. Having watched this data revolution play out in investing and markets, I am not so sure, at least in the investing space. Many analysts now complain that they have too much data, not too little, and struggle with data overload. At the same time, a version of Gresham's law seems to be kicking in, where bad data (or misinformation) often drives out good data, leading to worse decisions and policy choices. My advice, gingerly offered, is that as you access data, it is caveat emptor, and that you should do the following with any data (including my own):
(a) Consider the biases and priors of the data provider.
(b) Not use data that comes from black boxes, where providers refuse to detail how they arrived at numbers.
(c) Crosscheck with alternate data providers, for consistency.
Data Coverage
As I mentioned at the start of this post, I started my data estimation for purely selfish reasons, which is that I needed those estimates for my corporate financial analyses and valuations. While my sharing of the data may seem altruistic, the truth is that there is little that is proprietary or special about my data analysis, and almost anyone with the time and access to data can do the same.
Data Sources
At the risk of stating the obvious, you cannot do data analysis without having access to raw data. In 1993, when I did my first estimates, I subscribed to Value Line and bought their company-specific data, which about 2000 US companies and included a subset of items on financial statements, on a compact disc. I used Value Line's industry categorizations to compute industry averages on a few dozen items, and presented them in a few datasets, which I shared with my students. In 2025, my access to data has widened, especially because my NYU affiliation gives me access S&P Capital IQ and a Bloomberg terminal, which I supplement with subscriptions (mostly free) to online data. It is worth noting that these almost all the data from these providers is in the public domain, either in the form of company filings for disclosure or in government macroeconomic data, and the primary benefit (and it is a big one) is easy access.
As my data access has improved, I have added variables to my datasets, but the data items that I report reflect my corporate finance and valuation needs. The figure below provides a partial listing of some of these variables:
As you can see from browsing this list, much of the data that I report is at the micro level, and the only macro data that I report is on variables that I need in valuation, such as default spreads and equity risk premiums. In computing these variables, I have tried to stay consistent with my own thinking and teaching and transparent about my usage. As an illustration for consistency, I have argued for three decades that lease commitments should be treated as debt and that R&D expenditures are capital, not operating, expenses, and my calculations have always reflected those views, even if they were at odds with the accounting rules. In 2019, the accounting rules caught up with my views on lease debt, and while the numbers that I report on debt ratios and invested capital are now closer to the accounting numbers, I continue to do my own computations of lease debt and report on divergences with accounting estimates. With R&D, I remain at odds with accountants, and I report on the affected numbers (like margins and accounting return) with and without my adjustments. On the transparency front, you can find the details of how I computed each variable at this link, and it is entirely possible that you may not agree with my computation, it is in the open.
There are a few final computational details that are worth emphasizing, and especially so if you plan to use this data in your analyses:
With the micro data, I report on industry values rather than on individual companies, for two reasons. The first is that my raw data providers are understandably protective of their company-level data and have a dim view of my entry into that space. The second is that if you want company-level data for an individual company or even a subset, that data is, for the most part, already available in the financial filings of the company. Put simply, you don't need Capital IQ or Bloomberg to get to the annual reports of an individual company.
For global statistics, where companies in different countries are included within each industry, and report their financials in different currencies, I download the data converted into US dollars. Thus, numbers that are in absolute value (like total market capitalization) are in US dollars, but most of the statistics that I report are ratios or fractions, where currency is not an issue, at least for measurement. Thus, the PE ratio that I report would be the same for any company in my sample, whether I compute it in US dollar or Chilean pesos, and the same can be said about accounting ratios (margins, accounting returns).
While computing industry averages may seem like a trivial computational challenge, there are two problems you face in large datasets of diverse companies. The first is that there will be individual companies where the data is missing or not available, as is the case with PE ratios for companies with negative earnings. The second is that the companies within a group can vary in size with very small and large companies in the mix. Consequently, a simple average will be a flawed measure for an industry statistic, since it weighs the very small and the very large companies equally, and while a size-weighted average may seem like a fix, the companies with missing data will remain a problem. My solution, and you may not like it, it to compute aggregated values of variable, and use these aggregated values to compute the representative statistics. Thus, my estimate the PE ratio for an industry grouping is obtained by dividing the total market capitalization of all companies in the grouping by the total net income of all companies (including money losers) in the grouping.
Since my data is now global, I also report on these variables not only across all companies globally in each industry group, but for regional sub-groupings:
I will admit that this breakdown may look quirky, but it reflects the history of my data updates. The reason Japan gets its own grouping is because when I started my data grouping two decades ago, it was a much larger part of both the global economy and markets. The emerging markets grouping has become larger and more unwieldy over time, as some of the countries in this group had or have acquired developed market status and as China and India have grown as economies and markets, I have started reporting statistics for them separately, in addition to including them in the emerging markets grouping. Europe, as a region, has become more dispersed in its risk characteristics, with parts of Southern Europe showing the volatility more typical of emerging markets.
-
Data Universe
In the first part of this post, I noted how bias can skew data analysis, and one of the biggest sources of bias is sampling, where you pick a subset of companies and draw the wrong conclusions about companies. Thus, using only the companies in the S&P 500 or companies that market capitalizations that exceed a billion in your sample in computing industry averages will yield results that reflect what large companies are doing or are priced at, and not the entire market. To reduce this sampling bias, I include all publicly traded companies that have a market price that exceeds zero in my sample, yielding a total sample size of 48,156 companies in my data universe. Note that there will be some sampling bias still left insofar as unlisted and privately owned businesses are not included, but since disclosure requirements for these businesses are much spottier, it is unlikely that we will have datasets that include these ignored companies in the sample in the near future.
In terms of geography, the companies in my sample span the globe, and I will add to my earlier note on regional breakdowns, by looking at the number of firms listed and market capitalizations of companies in each sub-region:
As you can see, the United States, with 5994 firms and a total market capitalization of $69.8 trillion, continues to have a dominant share of the global market. While US stocks had a good year, up almost 16.8% in the aggregate, the US share of the global market dipped slightly from the 48.7% at the end of 2024 to 46.8% at the end of 2025. The best performing sub-region in 2025 was China, up almost 32.5% in US dollar terms, and the worst, again in US dollar terms, was India, up only 3.31%. Global equities added $26.3 trillion in market capitalization in 2025, up 21.46% for the year.
While I do report averages by industry group, for 95 industry groupings, these are part of broader sectors, and in the table below, you can see the breakdown of the overall sample by sector:
Across all global companies, technology is now the largest share of the market, commanding almost 22% of overall market capitalization, followed by financial services with 17.51% and industrials with 12.76%. There is wide divergence across sectors, in terms of market performance in 2025, with technology delivering the highest (20.73%) and real estate and utilities the lowest. There is clearly much more that can be on both the regional and sector analyses that can enrich this analysis, but that will have to wait until the next posts
Usage
My data is open access and freely available, and it is not my place to tell you how to use it. That said, it behooves me to talk about both the users that this data is directed at, as well as the uses that it is best suited for.
For practitioners, not academic researchers: The data that I report is for practitioners in corporate finance, investing and valuation, rather than academic researchers. Thus, all of the data is on the current data link is data as of the start of January 2026, and can be used in assessments and analysis today. If you are doctoral student or researcher, you will be better served going to the raw data or having access to a full data service, but if you lack that access, and want to download and use my industry averages over time, you can use the archived data that I have, with the caveat being that not all data items have long histories and my raw data sources have changed over time.
Starting point, not ending point: If you do decide to use any of my data, please do recognize that it is the starting point for your analysis, not a magic bullet. Thus, if you are pricing a steel company in Thailand, you can start with the EV/EBITDA multiple that I report for emerging market steel companies, but you should adjust that multiple for the characteristics of the company being analyzed.
Take ownership: If you do use my data, whether it be on equity risk premiums or pricing ratios, please try to understand how I compute these numbers (from my classes or writing) and take ownership of the resulting analysis.
If you use my data, and acknowledge me as a source, I thank you, but you do not need to explicitly ask me for permission. The data is in the public domain to be used, not for show, and I am glad that you were able to find a use for it.
The Damodaran Bot!
In 2024, I talked about the Damodaran Bot, an AI entity that had read or watched everything that I have put online (classes, books, writing, spreadsheets) and talked about what I could do to stay ahead of its reach. I argued that AI bots will not only match, but be better than I am, at mechanical and rule-based tasks, and that my best pathways to creating a differential advantage was in finding aspects of my work that required multi-disciplinary (numbers plus narrative) and generalist thinking, with intuition and imagination playing a key role. As I looked at the process that I went through to put my datasets together, I realized that there was no aspect of it that a bot cannot do better and faster than I can, and I plan to work on involving my bot more in my data update next year, with the end game of having it take over almost the entire process.
I do think that there is a message here for businesses that are built around collecting and processing data, and charging high prices for that service. Unless they can find other differentials, they are exposed to disruption, with AI doing much of what they do. More generally, to the extent that a great deal of quant investing has been built around smart numbers people working with large datasets to eke out excess returns, it will become more challenging, not less so, with AI in the mix.
Stock markets have always rewarded winners with large capitalizations, and with each new threshold, the questions begin anew of whether animal spirits or fundamentals are driving the numbers. A few weeks ago, Nvidia seemed unstoppable as its market capitalization crested $5 trillion, and while markets have turned skeptical since, the core questions have not gone away, and the answers come from two extremes. At one end are the "realists”, who view themselves as rational, above the fray and entirely data-driven, who argue that there is no business model that can support a value this high, and that Nvidia is overvalued. At the other end are the “AI true believers”, who believe that if the market the company is going after is big enough, and they see AI as such a market, the upper bounds on value are released, the sky is the limit. As someone who entered the Nvidia sweepstakes early (in 2018) and has held it through much of its magical run, while expressing reservations about its pricing running ahead of its value, especially in the last three years, I will try to thread the needle (unsuccessfully, I am sure) in this post. In fact, rather than try to convince you that the company is under or overvalued, which is really your judgment to make, I will offer a simple model to reverse engineer from any given market capitalization, the revenues and profitability thresholds you have to meet, and allow you to come to your own conclusions.
A History of Market Cap Thresholds
In 1901, US Steel was created when Andrew Carnegie and J.P. Morgan consolidated much of the US steel business, with an eye to monopolizing the steel business, and the company became the first global firm with a market capitalization of a billion dollars, a small number in today's terms, but a number that was three times larger than the Federal budget in that year. The twentieth century was a good one for the US economy and US stocks, and the thresholds for highest market cap rose along the way:
Note the long stretch between Microsoft hitting the half-a-trillion dollar market cap in 1999, as the dot com boom peaked, and Apple doubling that threshold in 2018. Note also the quickening of the pace, as Apple hit the $2 trillion and $3 trillion market capitalization thresholds in the next four years, and Nvidia continued the streak hitting $4 trillion in 2024 and $5 trillion in 2025.
The table does provide a starting point to discussing multiple themes about how the US economy and US equities have evolved over the last century. You can see the shift away from the smokestack economy to technology , in the companies hitting the thresholds, with US Steel and GM firmly in the old economy mode, Microsoft, Apple, and Nvidia representing the new economy, and GE, with its large financial service arm, operating as a bridge. Having been in markets for all of the thresholds breached since 1981, the debate about whether the company breaking through has risen too much in too short a time period has been a recurring one.
Substance: To get a measure of operating substance, I looked at the revenues and net income in the year leading into the year in which each company broke through the threshold. As you can see, US Steel had revenues of $0.56 billion and net income of $0.13 billion in 1901, the year in which its market cap exceeded $1 billion. GM, at the time its market cap breached $10 billion, had revenues of $9.83 billion, on which it generated net income of $0.81 billion; if PE ratios are your pricing metric of choice, that would have translated into a PE ratio of 12.35. Between 2018 and 2022, as Apple's market cap tripled from $1 trillion to $3 trillion, its annual revenues increased by 72%, and its net profits almost doubled. Finally, coming to Nvidia, the surge in market cap to $4 trillion in 2024 and $5 trillion in 2025 has come on revenues and net income that are about a quarter of the size of Apple's revenues and net income.
Life cycle: Every company that climbed to the top of the market cap tables and hit a market cap threshold historically has had single-digit revenue growth in the year leading up, with two exceptions: Microsoft in 1999, which was coming off a 28% revenue growth rate in 1998, and Nvidia in both 2024 and 2025 coming off even higher growth rates. Using this revenue growth rate in conjunction with the ages of the companies involved, I think it is fair to conclude that there has been a shift across time, with the mature companies (older, lower growth) that were at the top of the list for much of the twentieth century to much younger companies with growth potential rising to the top in this one.
Investment returns: Looking at the returns in the years after these companies hit their market cap thresholds, the results are mixed. While buying Apple in 2018, 2020, or 2022 would have yielded winning returns, at least over the next year or two, buying Microsoft in 1999 would not. In some of these cases, extending the time horizon would have made a difference, for the positive with Microsoft and for the negative with GE.
From a rational perspective, you could argue that these thresholds (billion, half a billion, trillion, etc.) are arbitrary and that there is nothing gained by focusing on them, but in a post that I wrote in September 2018 on Apple and Microsoft becoming trillion-dollar companies, I argued that crossing these arbitrary thresholds can draw attention to the numbers, with the effects cutting both ways, drawing in investors who regret missing out on the rising market cap in the periods before (a positive) and causing existing investors to take a closer look at what they are getting in return (perhaps a negative).
Market Caps: Pathways to Intrinsic Value Break Even
Debates about whether a company is worth what it is trading for, whether it be a billion, ten billion, a hundred billion, or a trillion, devolve into shouting matches of "he said, she said", with each side staking out divergent perspectives on value and name-calling the other. Having been on the receiving end of some of that abuse, I decided to take a different pathway to examining this question. Rather than wonder whether Nvidia is worth five trillion or Eli Lilly is worth a trillion, I framed the question in terms of how much Nvidia or Eli Lilly would have to generate in revenues to justify their market capitalizations. The reason for my focus on revenues is simple since it is relatively unaffected by accounting games and can be compared to the total market size to gain perspective.
The tool that I plan to use to arrive at this breakeven revenue is intrinsic valuation, and I chose not to use the acronym "DCF" deliberately. A discounted cash flow valuation (DCF) sounds like an abstraction, with models driving discount rates and financial modeling driving cash flows. To me, a DCF is just a tool that allows you to assess how much you would pay for a business or the equity in the business, given its capacity to generate cash flows for its owners. Since it is easy to get lost in the labyrinth of estimates over time, I will simplify my DCF by doing two things. First, since our discussion is about market capitalization, i.e., the market's estimate of the value of equity, I will stay with an equity version of the model, where I focus on the cash flows that equity investors can get from the business and discount these cash flows back at a rate of return that they would demand for investing in that equity. In its most general form, this is what an equity valuation yields:
To simplify the assessment further, I structured this model to value equity in a mature company, i.e., one growing at or below the nominal growth rate of the economy in the very long term and again for simplicity, assumed that it could do this forever. The value of equity in this mature, long-lasting firm can be written as follows:
To put this model into use, let's take the $5 trillion dollar market capitalization that Nvidia commanded a few weeks ago and assign the following general inputs:
Cost of equity: Every month, I estimate the implied cost of equity for the S&P 500, and that number is model-agnostic and driven by what investors are willing to pay for stocks, given their fears and hopes. At the start of November 2025, that number was about 8%, with higher required returns (9-12%) for riskier stocks and lower expected returns (6-7%) for safer stocks.
Inflation rate: While inflation has come down from its 2022 highs, it has stayed stubbornly above 2%, which the Fed claims as its target, and it seems more realistic to assume that it will stay at 2.5%, which is consistent with the riskfree rate being about 4%.
Stable growth rate (nominal growth rate in the economy): This is a number that is in flux, as economists worry about recessions and economic growth, but since this is a long-term number that incorporates expected inflation, it seems reasonable to assume an expected nominal growth of 4% for the economy (about 1.5% real growth).
The net profit margin for Nvidia in the most recent twelve months has been 53.01%, an exceptionally high number, and the return on equity it has earned, on average over the last five years, is about 64.44%. I know that these numbers will come under pressure over time, as competition for AI chips picks up, and Nvidia's biggest customers (and chip maker) push for their share of the spoils, but even if you assume that Nvidia can maintain these margins, the revenue that Nvidia would have to deliver to justify its value is $483.38 billion.
Since Nvidia is still growing and you may need to wait, as equity investors, to get your cash flows, this breakeven number will get larger, the longer you have to wait and the lower the cash yield that equity investors receive during the growth period. In fact, with Nvidia, if you assume that it will take five years for them to grow to steady state, and that equity investors will receive a cash yield (cash flow as a percent of market cap) of 2% a year, the estimated breakeven revenue increases to $677.97 billion. The table below maps out the effects of waiting on breakeven revenues for a range of cash yield:
If, as seems reasonable, you assume that net margins and return on equity will decrease over time, the revenues you would need to break even will expand:
In fact, if you are a low-margin company, with net margins of 5% (as is the case with even the very best-run discount retailers) and a more modest return on equity of 10%, you will need revenues of $8 trillion or more to be able to get to a market capitalization of $5 trillion.
This framework can be used to compute breakeven revenues at other firms, and in the table below, we do so for the twelve largest market cap companies in the world, at their market capitalizations on November 20, 2025:
Note that, for simplicity, I have used a 2% cash yield and 4% growth rate in perpetuity for all of these firms, and that the breakeven revenues reflect current net margins and returns on equity at each of these firms, but with that said, there is still value in looking at differences. To allow for this comparison, I forecast out breakeven revenues five years from now, and estimated the growth that each company would need over the five years to justify its current market cap. Not surprisingly, Aramco can get to its breakeven revenues in year 5 with almost no growth (0.59% growth rate) but Tesla needs to deliver revenue growth of 86.4% to break even. Broadcom, another company that has benefited from the market's zeal for AI, has the next highest cliff to climb in terms of revenue growth. In fact, for all of the Mag Seven stocks, growth has to 15% or higher to breakeven, a challenge given their scale and size. In dollar value terms, three companies will need to get to breakeven revenues that exceed one trillion by year 5 to breakeven, Apple, Amazon and Tesla, but the first two are already more than a third of the way to their breakeven targets, but Tesla has a long, long way to go.
From Breakeven Revenues to Investment Action
While some are more comfortable replacing conventional intrinsic valuation, where you estimate value and compare it to price, with a breakeven assessment, the truth is that the two approaches are born out of the same intent.
The Economics of Breakeven Revenues
The model that I used to compute breakeven revenues is a vastly simplified version of a full equity valuation model, but even in its simplified form, you can see the drivers of breakeven revenues.
Market Capitalization: Since we work back from market capitalization to estimate breakeven revenues, the larger the market capitalization, holding all else constant, the greater the breakeven revenues will be. Using just Nvidia as an example, the company has seen its market capitalization rise from less than $400 billion in 2021, to $1 trillion in 2023, $2 trillion and $3 trillion thresholds in 2024 and crossed the $4 trillion and $ 5 trillion market cap levels in 2025. As the market cap has risen, the breakeven revenues have increased from $200 billion at the $1 trillion mark to $600 billion at the current market cap.
Operating Profitability: There are two profitability metrics in the drivers, with net margins determining how much of the revenues a company can convert to profits and the return on equity driving the reinvestment needed to sustain growth. Higher profitability will allow a company to deliver a higher market capitalization, at any given level of revenues. One reason manufacturing firms like Tesla will need higher breakeven revenues than software firms is that the unit economics are not as favorable.
Interest rates and equity risk premiums: The level of interest rates and equity risk premiums determine the cost of equity for all company, with higher values for the latter pushing up the costs of equity for riskier companies higher, relative to safer companies.
Operating and leverage risk: The riskiness in a business will push its cost of equity higher, and a higher debt load (relative to market cap) will have the same effect. A higher cost of equity will raise the breakeven revenues needed to deliver the same market capitalization.
In sum, while the breakeven revenue that you need to justify a given market cap always increases as the market cap increases, its level and rate of rise will be governed by business economics.
The 3Ps: Possible, Plausible, and Probable
Replacing a conventional intrinsic valuation with a breakeven revenue analysis still leaves open the final investment question of whether that breakeven revenue is a number that you are comfortable with, as an investor. To address this question, I will draw on a structure that I use for intrinsic valuation, where I put my assessment through what I call the 3P test.
It is possible that once you compute the breakeven revenues for a firm and measure it up against reality that it is impossible, i.e., a fairy tale. The most obvious case is when the breakeven revenues that you compute for your firm exceeds the total market for the products or services that it provides. If there is a lesson that tech companies learned in the last decade, it was in making the total addressable market (TAM) for their market into almost an art form, adding zeros and converting billion dollar markets into trillion dollar TAMs. If you pass the "it is possible" test, you enter the plausibility zone, and nuance and business economics enter the picture more fully. Thus, assuming that a luxury retailer with sky-high margins and small revenues, by staying with a niche market, can increase its revenues ten-fold, while keeping margins intact, is implausible, as is a net margin of 40% in stable growth for a company with gross margins that are barely above that number. Finally, assuming that revenues can multiply over time, without reinvesting in acquisitions or projects to deliver those revenues are also pushing the boundaries of what is plausible. Once breakeven revenues pass the possible and plausible tests, you should be on more familiar ground as you look at the entire story line for the company, and assess whether the combination of growth, profitability and reinvestment that you are assuming with your story has a reasonable probability of being delivered.
To apply these tests, consider Nvidia and Tesla. Nvidia needs about $590 billion in revenues by 2030 to break even at its current market capitalization of $4.3 trillion, requiring a growth rate in revenues of about 26% for the next five years. While that is a reach, it is both possible and plausible, with continued growth in the AI chip market and a dominant market share for Nvidia providing the pathway. It is on the probable test that you run into headwinds, since competition is heating up, and that will put pressure on both growth and margins. The problem for Tesla is that if the net margin stays low (at 5.31%), the revenues needed to breakeven exceed $2.2 trillion, and even with robotics and automated driving thrown into the business mix, you are pushing the limits of possibility. A Tesla optimist, though, would argue that these new businesses, when they arrive, will bring much higher net margins, which, in turn, will push down breakeven revenues and bring it into plausible territory.
The Aggregated 3P Test - Big Market Delusion
We tend to ask the 3P question at the company level with the companies that we choose to invest in (and like), but as we construct what look like plausible and probable stories for these companies, and invest in them accordingly, there are other investors are asking the same questions about the companies that they invest in, many of which compete in the same business as yours. That may sound unexceptional to you, but when the market that these companies are competing in is very large and still in formation, you can end up with what I described almost a decade ago as the big market delusion. In a paper on the topic, I used the dot.com boom, the cannabis stock surge and online advertising as case studies to explain how this behavior is a feature of big markets
The AI storyline clearly fits the big market delusion. There is talk of a "huge" market for AI products and services, with little to show as tangible evidence of that market’s existence right now, and that potential has drawn massive investments in AI architecture from tech companies. Along the way investors have also fallen under the spell of the big market, and have pushed up the market capitalizations of almost every company in the space. Using the language of breakeven revenues, investors in each of these companies is attributing large breakeven revenues to their chosen companies, but the delusion comes from the reality that if you aggregated these breakeven revenues across companies, the market is not big enough to sustain all of them. In short, each company passed the possible and plausible test, but in the aggregate, you are chasing an impossible target.
While the big market delusion is at play in every aspect of AI, one segment where it is most visible right now is in the Large Language Models (LLM) space, where high profile players like ChatGPT, Gemini, Grok and Claude are vying for users, and their creators are being rewarded with nosebleed pricing. OpenAI, while still unlisted, has used the early lead that ChatGPT gave it in the LLM race to attract investments from a host of big tech companies (including Nvidia and Amazon) and venture capitalists, with the most recent investors pricing it at $500 billion, an astonishing number, given that the company reported revenues of only $13 billion in the most recent twelve months. Anthropic, the creator of Claude, has seen its pricing jump in the most recent funding round (from Microsoft and Nvidia in November 2025) to $350 billion, fifty times its revenues of $7 billion in the last twelve months. Elon Musk's owners stake in xAI, Grok's originator, was estimated to be worth $230 billion in November 2025, again an immense multiple of its revenues of $3.2 billion (if you include combined revenues with X). Expanding the list to the large tech companies, it is undeniable that some of Alphabet's massive rise in market capitalization in 2025 is because of its ownership of Gemini, and that Meta (with Llama) and Amazon (with Nova) have also seen bumps in market capitalization. Finally, while Deepseek is no longer making headlines, it is also in the space, competing for business. In the aggregate, LLM ownership is being priced at $1.5 trillion or more, and the collective revenues, even generously defined, are less than $100 billion. It is entirely plausible that a big market exists for LLMs, and that one or even two of the players in this space will be winners, but in the aggregate, the market is overreaching.
The Management Effect
The mechanics of the breakeven revenue process may make it seem like managers are bystanders in the process and that investing can be on autopilot, but they are not. In fact, when market capitalizations rise, and breakeven revenues run well ahead of current revenues, I would argue that management matters more than ever. Going back to the breakeven revenues that we computed for the twelve largest market cap companies in the world, I would make the case that management matters much less (if at all) in Aramco and Berkshire Hathway, where breakeven revenues are close to current revenues, and the investments needed to deliver those revenues have already been made, that at the companies that still have steep climbs ahead of them to get to breakeven revenues.
In this context, I will reemphasize a concern that I raised at the height of Meta's metaverse investing fiasco, which is that investors at many tech companies, including most on the large cap list, have given up their corporate governance rights, often voluntarily (through the acceptance of shares with different voting rights), to founders and top management in these companies. When traditional corporate governance mechanisms break down, and top managers have unchecked power, there is an increased risk of overreach. That concern is multiplied in the LLM space, where Sam Altman (at OpenAI) and Elon Musk (at xAI) are more emperors than CEOs.
The Investing Bottomline
I started this post with mentions of market cap thresholds being breached, as the market pricing pushes up into the trillions for some of the biggest stock market winners. But what are the implications for investors?
Highly priced ≠ Overpriced: If you are an investor who considers any highly priced company to be overvalued, I hope that this post leads you to reconsider. By reframing a pricing in terms of breakeven revenues, profitability and reinvestment, it allows you to consider whether a stock, even if priced at $4 trillion, may still be a good buy.
The 3P test: Once you compute the operating metrics you need to breakeven on an investment in a highly priced company, passing those metrics through the 3P test (Is it possible? Is it plausible? Is it probable?) allows you to examine each company on its merits and potential, rather than use a broad brush or a rule of thumb (based on PE ratios or revenue multiples).
Room to disagree: I have never understood why, even if you believe strongly that a stock is over or under priced, that you need to evangelize that belief or contest people with alternate views. I think that the pathway that you would need (in terms of revenue growth and profitability) to justify Nvidia's and OpenAI's current pricing is improbable, but that is just my view, and it is entirely possible that you have an alternate perspective, leading to the conclusion that they are undervalued.
Reality checks: No matter what your view, optimistic or pessimistic, you have to be open to changing your mind, as you are faced with data. Thus, if you have priced a company to deliver 20% growth in revenues over the next five years (to break even) and actual revenues growth comes in at 10%, you have to be willing to revisit your story, admit that you were wrong, and adapt.
If you came into this post, expecting a definitive answer on whether Nvidia is overpriced, you are probably disappointed, but I hope that you use the breakeven spreadsheet to good effect to make up your own mind.
