3 ms·

The stock market has always been a bit of a mystery to me. There is no requirement that company revenue be tied to a stock's value. Rather, the stock for a give

The stock market has always been a bit of a mystery to me. There is no requirement that company revenue be tied to a stock's value. Rather, the stock for a given company is entirely based on human sentiment. Person A buys stock S at time t0 because he thinks it will be more valuable to person B at time t1. Person B buys stock S at time t1 because he thinks it will be more valuable to person C at time t2. ...and so on. Buying a small-volume of stock in company X provides no tangible value. You aren't getting a seat on the board with a few thousand dollars. Obviously, things like dividends and retirement tax-breaks have value, but I'm talking about pure speculative investing. My question to the investors out there: Why should be there be a person C or D who buys the stock from you in the future? In the extreme case, you could buy stock X, and then never find someone to buy it from you.

peripitea 5y ago
A public company's value is based on a collective estimate of what people think its cashflows will look like in the future. For a mature company that you don't expect to change much, looking at present financials is a solid way to estimate future financials. For an earlier-stage, or more speculative company like Rivian with a wide range of possibilities in its future, current financials are not typically helpful, with a few exceptions (e.g. growth indicators, signs of poor management). For example: Five years ago, I thought Tesla had a 30% chance of failing, 20% chance of being bought by a major auto manufacturer (for let's say ~30B), 30% chance of turning into a major auto manufacturer (est value ~100B), 10% chance of turning into the top auto manufacturer (est value ~300B), and 10% chance of turning into something that transcended auto manufacturing (est value ~1T). If you multiply this out, you get an expected value of 0+6+30+30+100 = 166B. At the time the market cap was around 40B, so I thought it was underpriced and bought some. Obviously I'm glossing over some things, but it's sensible enough and should illustrate how you can get there. In terms of the notion of "what if you never find someone to buy it from you?" The short answer is that that's extremely unlikely. The longer answer is that you can also just price that into your probability distribution, but because of the low likelihood it's unlikely to affect your price much. FWIW I don't know much about Rivian, but my sense is also that they are overpriced. You could go through the same little exercise above for Rivian and see if you agree. To summarize what my distribution would look like, I think they have a much higher chance of failure than Tesla in my example above (because they're much less far along), and not enough potential upside to justify the valuations being discussed. But I could easily be wrong.

peripitea 5y ago

A public company's value is based on a collective estimate of what people think its cashflows will look like in the future. For a mature company that you don't expect to change much, looking at present financials is a solid way to estimate future financials. For an earlier-stage, or more speculative company like Rivian with a wide range of possibilities in its future, current financials are not typically helpful, with a few exceptions (e.g. growth indicators, signs of poor management). For example: Five years ago, I thought Tesla had a 30% chance of failing, 20% chance of being bought by a major auto manufacturer (for let's say ~30B), 30% chance of turning into a major auto manufacturer (est value ~100B), 10% chance of turning into the top auto manufacturer (est value ~300B), and 10% chance of turning into something that transcended auto manufacturing (est value ~1T). If you multiply this out, you get an expected value of 0+6+30+30+100 = 166B. At the time the market cap was around 40B, so I thought it was underpriced and bought some. Obviously I'm glossing over some things, but it's sensible enough and should illustrate how you can get there. In terms of the notion of "what if you never find someone to buy it from you?" The short answer is that that's extremely unlikely. The longer answer is that you can also just price that into your probability distribution, but because of the low likelihood it's unlikely to affect your price much. FWIW I don't know much about Rivian, but my sense is also that they are overpriced. You could go through the same little exercise above for Rivian and see if you agree. To summarize what my distribution would look like, I think they have a much higher chance of failure than Tesla in my example above (because they're much less far along), and not enough potential upside to justify the valuations being discussed. But I could easily be wrong.