Ask HN: What is the actual cost basis of the stock market?
Does anyone know what is the actual cost basis for the entire US stock market.
I mean at what price did the current stock holders got/bought their stocks.
Does anyone know what is the actual cost basis for the entire US stock market.
I mean at what price did the current stock holders got/bought their stocks.
That is an interesting question. I guess one way to approximate a very rough, back-of-the-envelope type of answer would be to look at the average daily dollar volume traded and the total market capitalization. So for instance with the Nasdaq we have approx. $30 trillion in market cap and approx. $400 billion of average daily volume, a factor of 75. So in the best-case situation where each share trades exactly once, it would take approx. 75 trading days for the market to turn over once. However, let's take a more pessimistic assumption (though probably still not pessimistic enough) that shares are selected randomly to trade. We can use the classic ball-and-bins type of probability to estimate how many days it would take until let's say 95% of dollar value had turned over at least once, which produces an estimate of roughly 225 trading days, so let's round it up to 1 trading year (these are very rough estimates). So then assuming that roughly 95% of the Nasdaq market cap has turned over in the past year, then the average market cap over the past year is not a bad ball-park estimate of the cost basis. (I invite any real mathematicians out there to tell me where I went wrong.)
My fear in such calculations is that we leave the big whales that do not move, outside. For example the CEO of Nvidia has 4% Nvidia that in today's value is 160B, but the basis is from the IPO, so more like $40M. So now our back-of-the envelope calculation is 160B off.
In my humble opinion,It would be 0 as there is no transaction taking place. The original question is about stock buyers not stock holders which is often related but not always.
How is there no transaction taking place? The CEO is exchanging their labor for shares
I don't think that's a very relevant calculation. Or at least you haven't articulated why.
It's just a measure of liquidity. A stock can trade 1$ or 0$ per day and still have a market cap of 300B ( see private eq.)
(Maybe you are onto something, if stock iliquid, they got stocks at 0. If stock very liquid they got stocks at actual price.)
I dont think your pessimistic scenario is nearly slow enough. There is a large buy and hold component of the market, with high frequency trading on the margin.
The random sale model doesn't account for pensions, mutual funds, and individuals that hold. Where would firms like Berkshire Hathaway fit into this model?
I imagine the distribution of stock hold duration is nonlinear/logarithmic
What about dark pools?
What about them?
It'd be hard to figure out. A lot of hedge funds run on mark-to-market accounting, which basically taxes the unrealized gains as gains every year, as opposed to keeping track of the cost basis when bought and taxed upon selling. There are quarterly reports made to the SEC though for firms of a certain size, or holding a certain percentage of one stock, typically 10% of a company involves additional paperwork and regulation.
Also, there's many times where the cost basis is stepped up, for example when someone dies and passes on assets to their beneficiaries. In this case the cost basis changes from what the dead person paid for it to what the beneficiary gets it for upon inheriting it.
Overall, I don't think this is a useful measure of anything though. What you might be interested in is the volume weighted average price, which gives an idea on how much people have paid on a stock in a recent window, based on the volume of trades and the price of those trades, reflecting what the average buyer's cost basis could be around.
> Overall, I don't think this is a useful measure of anything though
This would be useful to know how much outstanding capital gains could be paid, if we removed the step up.
Even that isn't really true, since you don't pay taxes on Roth IRAs, and a normal IRA is taxed based on the total amount distributed (since you didn't pay taxes on the amount put in), rather than cost basis.
It's pretty complicated, look up the Secure Act 2.0 for some of the recent changes to try to get more taxes out of retirement accounts.
It's all a tower of cards based on what people think the cards are worth. Unless stocks pay dividend, valuation is afloat on the expectations of shareholders.
They either pay dividends or investors hope they will pay dividends one day (even META and GOOGL do now!) Or the company can get bought out.
Those are all very concrete reasons for valuations
yep
What does it tell you though?
If someone has 50% of say Google from the start, them selling it all over say 5 years would put the cost basis way up, but without changing much in the economy or in those companies.
They would probably in turn buy other stocks and increase their cost basis.
Interesting Q though!
There is no meaningful answer to this question. There are hundreds of millions of answers that change with varying frequency with intervals going from decades to milliseconds.
Since the cost can correctly only be evaluated when the transaction (cycle) has taken place, take a look at that for cost calculation of previous trades, and use some algo to predict the same for future scenario which you are supposed to be dealing of.
My advice on approach would be that you should consider it as a big auction house with many items, if you remove the items you're still left with the house, what's the value of the house ? There can be different approaches, yet there can't be a repetitive answer.
You have regulatory reports on quarters that help determine pricing based on sales and book value.
Price stocks were bought at is hard to calculate, but someone suggested liquidity as a proxy.
Price stocks bought at on primary market is interesting but probably irrelevant.
I dont think anyone has that data. Individual companies dont have a record of who owns their stock and what they bought it at. Even individual owners sometimes dont know their own cost basis, let alone their broker or some centralized authority. This is further complicated by the fact that >50% of trades now occur outside of exchanges [1]
You might be able to estimate a similar attribute by looking at IPO history and additional primary offerings, but that also has major challenges. It doesn't include pre-IPO owners that retain stake, buybacks, ect. What is Zuckerberg's cost basis in Meta?
Interesting question though....
https://www.bloomberg.com/news/articles/2025-01-24/wall-stre...
I don’t even directionally have an idea how much is it. Is it like 80% of the current total market value or 8%? My back of the envelope math plus very coarse assumptions lead me closer to 80%.
To be clear, you are talking about the aggregate for the current holders, understanding that actual cost basis will be different for every person with stock? Some people will have long invest duration with tiny cost basis, some will have cost basis over 100%.
My ballpark guess is less than 80%. Most of the stock market is owned by households, either directly through equities, or indirectly through mutual funds and pensions [2, p14]
Most individuals and pensions are long term positions, with buying and selling to increase or decrease a position, not buying or liquidating entire portfolios.
If I guess the average hold time is ~5 years, that would be a about a 60% basis.
US total stock market cap in feb 25' was 67 Trillion, Total market cap in feb 20' was 40 trillion. [1]
What conclusion are you trying to draw from it? From an individual perspective, I care more about my cost basis in my market account, than the cost basis for an individual stock. I might have a 95% cost basis my current holding, but a 5% cost basis for my entire account(what I put in the market)
https://focus.world-exchanges.org/issue/february-2020/market... https://www.sifma.org/wp-content/uploads/2019/10/SIFMA-Insig...
I am trying to understand how much the stock market can drop before we destroy realized value (cash). Or how much money is not used in the economy, but rather it is tied to the stock market.
Leaving aside any drops, the entire market cant be sold/liquidated at once at it's current market cap / value. The current "value" / market cap of any given stock / the market is based on whatever tiny fraction of shares traded today.
Outside the ~ 5 Trillion USD that went to primary market (IPOs, secondary offerings and share issuances) since 1970 and flowed to the actual companies, there is a significant amount of cash that is tied in the secondary market speculating the value of the stocks instead of e.g. providing liquidity to the economy. I am trying to understand how much is that.
Hmm, not really. If you buy some shares from me for $1000, we settle and I give you the shares and you give me the $1000, and I can spend the $1000 on plane tickets or wine or put it in the bank or whatever, there is no $1000 that is "tied" to any shares.
Perhaps you value those shares at $1000 on your balance sheet as their "book value" or "cost basis" or whatever, but the $1000 doesn't exist (you gave it to me), all you have is shares.
I think the answer to that question is close zero. The US economy doesn't need more liquidity, and is essentially maxed out.
When you add large amounts of liquidity, you get proportional inflation/devaluation destroying the money you have added.
Covid Stimulus and related inflation showed how they economy cant even handle hundreds of billions more liquidity in the hands of consumers, let alone tens of trillions. The worker productive capacity doesn't exist to make and sell more stuff, so stocks and real-estate soak up as much liquidity as they can, and inflation takes the rest.
If the money in circulation doubled tomorrow, I would expect rent and eggs to double as well, because there isnt actually more stuff. In this way, excess liquidity and associated inflation heavily favors holders of debt, equity, capital, and real-estate.
As a well off American, I expect I would be the winner. My house dollar value of my house and 401k would double, while my mortgage would stay the same.
I would be cautious about using this assumption or analysis to answer those questions, and your phrasing.
Cash is different than realized value, as anyone who has watched their cash inflate away will tell you as they use it for kindling.
Similarly, money tied up in the stock market doesn't mean it could be used in the economy. In many ways, the economy is not money limited and adding more does nothing.
I’d hazard that average hold time is much shorter. What makes you say 5 years? My gut tells me that the majority of stocks are held by whales moving them in and out on a constant basis nowadays.
I stated most of my rationale in the post, but I don't think whales buy and sell most of their portfolio. I think they move a tiny portion over and over again.
Companies ofc have a record of how much they were issued at. Not that it necessarily would reflect its market value.
It is driven by supply and demand!
All the brokers and individuals logged into trading platforms enter the prices at which they wish to execute a buy or sell.
The system matches buyers to sellers, and that creates a pressure which moves the price this way and that way.
Firstly, consider the situation when there is a disparity between the prices: most asking prices are too high relative to bids, the volume of trading will be low. In this situation, sellers who want to get rid of the stock are motivated to drop their prices. Or, alternatively, if the market believes the stock has good near term prospects, the buyers have to raise their bids. Raised bids cause upward pressure on the price, lowered asks downward pressure. (No brainer!)
(You can see a big spread in bid/ask prices after hours when an exchange is closed. After hours trading is possible. If you want to buy and have that execute, you have to specify at least the asking price. Otherwise your bid will just be added to the others that are sitting there, waiting. After-hours trading volumes are significantly lower, which is associated with the wide bid/ask spread.)
When people execute market orders, that also moves the price. Selling something at market order means you urgently want to sell at whatever price is current. That means that your asking price will be lowered in order to meet a bidder.
Short-selling activity can drive a price down. And of course, short selling is motivated by the belief that a stock is about to tank or tanking. Under short selling, someone borrows a stock and sells it right away, hoping to buy it back at a lower price. They sell it ASAP because they believe they are buying a tanking stock, so they faster they sell it, the more money they will fetch, and the bigger will be their take when they buy it back. Short selling, I think, typically uses market orders, and so thereby exerts downward pressure. When bad news about a company invites short sellers, that can accelerate the drop in the stock price.
Anyway the end of the day, the price is the result of psychology, plus a few fundamentals of the stock like company earnings and whatnot. People have beliefs about which way something will go, and that effects their buying and selling behavior. The prices come from all the individual trades.
Ask yourself, what caused the Beanie Baby craze? What was the cost basis for someone paying $200 for a stuffed toy? Simply the belief that someone will later pay them $300, together with a casual dismissal of the risk that they will not even get back their $200 --- why, because they are in the middle of a craze that is obviously going to go on longer than the time scale of their intended action!
What is your ultimate goal with this analysis?
Will probably be very hard to calculate and super skewed since market makers are constantly buying/selling...
One way to look at this is to look at the valuation of each company at the time of IPO, and see where it is now. Stock changed hands but you could think of growth relative to when the public could trade that stock. It’s not perfect - just an idea.
The concept of an "actual cost basis" for the entire U.S. stock market is not straightforward or directly observable, as it would require tracking every individual purchase price of every stock held by all investors—a task that is both logistically impossible and economically meaningless. Here’s why:
Key Challenges in Defining "Cost Basis" for the Market Diverse Ownership Structures:
Stocks are owned by individuals, institutions (pensions, mutual funds, hedge funds), governments, and foreign entities. Each group buys/sells stocks at different times and prices. Even within individual stocks, ownership is fragmented across millions of buyers with varying entry points. Dynamic Pricing:
Stock prices fluctuate constantly. A stock’s current price ($X today) does not reflect the average price paid by all current holders. For example: Some holders bought at 10,othersat50, and still others at $100. New buyers enter at today’s price, while existing holders may have held for decades. No Centralized Tracking:
There is no system to aggregate all historical purchase prices for every stock. Exchanges track trades, but not individual investor purchase histories. Alternative Metrics That Approximate "Cost Basis" While no single number exists, here are related concepts that might shed light:
1. Market Capitalization (Market Cap): Definition: Current stock price × total outstanding shares. Use: Reflects the market’s valuation of a company or the entire market (e.g., S&P 500 market cap). Limitation: Not a cost basis—it measures today’s value, not what investors paid. 2. Average Purchase Price for Index Funds: For passive investors (e.g., S&P 500 funds), cost basis depends on when they bought. Example: A fund using dollar-cost averaging buys more shares during downturns and fewer during rallies, lowering its average cost basis over time. Data for major index funds (e.g., Vanguard S&P 500 ETF) is public but only applies to specific funds, not the entire market. 3. Book Value: Definition: A company’s historical cost of equity minus depreciation/amortization (for accounting purposes). Use: Reflects the book value of shareholders’ equity, not the average price paid by investors. Example: If a company issued stock at 10pershareandneverrepurchasedshares,itsbookvaluemightbe10/shares, but current market value could be $50/shares. 4. Average Price Paid by Retail Investors: Surveys (e.g., by brokerage firms) occasionally estimate this for specific stocks or sectors. Example: A study might find that retail investors bought Tesla stock at an average price of $800/share over the past year. Limitation: Data is sparse, skewed toward recent activity, and not comprehensive. Why This Matters (or Doesn’t) For Individual Investors: Your personal cost basis matters for tax calculations (e.g., capital gains/losses). For the Market: The collective "cost basis" of all holders has no direct economic significance. What matters is: Current Valuation: How much the market values companies today. Market Sentiment: Investor psychology driving prices up or down. Fundamentals: Earnings, growth, debt, etc. Final Takeaway There is no single "actual cost basis" for the entire U.S. stock market. If you’re asking this question out of curiosity about market valuation, focus on metrics like market cap, P/E ratios, or index performance (e.g., S&P 500 history). For personal financial planning, track your own cost basis for individual stocks using brokerage records.