The Myth of the Rising Stocks over Time
#10
Investment math is hard.

Some people in this thread have already mentioned some of this, but I think it's both harder to calculate and a more positive picture than has been mentioned. But first, some stock fundamentals from the perspective of a value investor:

In theory, the total valuation of a company (which is called "market cap" and is calculated as price per share * number of shares) reflects the intrinsic value of a company. Buying stock in a company is buying partial ownership in that company, so you are entitled to the profits of a company in proportion to your ownership interest. The valuation of a company is determined by investors and typically reflects (in the long run) the price at which an investor is likely to make approximately 10-11% return per year, adjusting for risk.

When you have companies like Tesla being traded at large multiples of their earnings, this means that the current profits do not support such a high long-term payout, but enough investors believe that the company will grow large enough quickly enough that they will pay a premium now to buy it at current prices, so that in the future they will be making more than that.

The price per share of a stock is entirely arbitrary and determined in large part by the company issuing the stock. I say it is determined by that company because they get to decide how many shares they issue. As I described previously, the market cap is determined by the expected future profitability of the company (adjusted for risk), and the formula for market cap is market cap = price per share * number of shares. The company gets to set the number of shares, therefore, they also get to set the price per share (market cap / number of shares = price per share).

This leads me to my first agreement with previous commentators, which is that the Dow is kind of ridiculous. It picks 30 large companies that it believes represent the US economy and the price of the index is determined by the price per share (an arbitrary number) of each component. This means that if, say, Apple were to have done a 2:1 stock split (which does not change any fundamentals of the company) prior to being added to the Dow, it would have had (approximately) half of the influence on the overall performance of the Dow. Once companies are in, however, they get magic adjustment factors to try to keep things like stock splits from changing the price of the index. You can read a similar take from someone who knows far more about investing than I do here: https://www.joshuakennon.com/what-do-you...nt-system/

The decision to graph stock price over time is also strange, but sadly, common. As mentioned, the compounding effect of dividends are huge. There are two main ways for a company to pay its shareholders, and they make sense only in the context of thinking of shares as actual partial ownership in a business, and share price being a mostly meaningless number (with total company valuation, or market cap, being the interesting number).

The first way for companies to pay their owners (shareholders), which people are probably most familiar with, is to pay out dividends. A dividend is when a company has some cash on hand and decides that rather than reinvest all of it into the company to create future growth, they want to pay some of it out to all of the shareholders in proportion to how much they own. Some companies have a policy of always paying out some amount on a fixed schedule, others pay out irregularly. When a company decides to pay out dividends of $100,000 (so if they have 1 million shares, a dividend of $0.10 / share), it means that their market cap immediately drops by $100,000. This makes sense, because the company now has $100,000 less in assets. This means that all of the stock that everyone owned is less valuable, but they have cash exactly equal to that drop in value.

The second way for companies to pay their owners (shareholders), which is more common in recent times, is through a stock buyback. With dividends, the company has money and pays all of its owners, and the owners all maintain the same ownership interest they had before the dividend. With a stock buyback, the company instead spends its money to buy shares back from some of the shareholders. The remaining owners (those who did not sell their shares) do not have any more money than they had before, but instead their shares are now worth a larger portion of the company. If there were previously 1 million shares outstanding and the company bought 100,000, then if I had 100,000 shares I would now own 1/9 of the company instead of 1/10. If I want to get my money out, I can sell some number of my shares. If I sell 10,000 shares I would have 90,000 / 900,000 shares, meaning I would go back to owning 10% of the company. My shares were worth $1 / share before the buyback, but the shares are now more valuable (because there are fewer of them), so I can sell them for $1.11 / share. This means I get back $10,000 in cash for the sale. This is exactly the amount of money I would have gotten in the dividend case.

Whether a company chooses to use dividends or stock buybacks to pay its owners is mostly driven by tax consequences, but otherwise they have the same effect. Given this, it would be crazy to show them differently on a graph, but that's exactly what almost all financial graphs do. If you are not looking at a total return graph, you are saying you want to ignore dividends but include stock buyback (and this is probably not what you actually want to do). The total return graph (which is not the graph in the first post) includes dividends and assumes you reinvest them back into the stock (which is what happens by default in a stock buyback scenario).

Combining all of this together, and using the dqydj calculator you linked (which seems to give the correct numbers), we can compare to the graph in your first post going back to 1947. When I do so and do not adjust for inflation, we get about 11.3% total annual return on the S&P500. Adjusting for the CPI (which has its flaws, I would rather have the 2018 basket of goods than the 1970 basket of goods, even if they are considered the "same" price in inflation adjusted terms), this gives us an annualized return of about 7.5%. This means that $1000 invested in 1947 gives you an inflation adjusted $170,000 today. There were some periods in there where you did not make as much, but there were others where you made a lot more.

In short, I believe that if I put $1000 into a broad-based index of stocks today, that 10 years from now I will have about $2000 adjusting for inflation (or the same in about 7 years not adjusting), because that is a long enough time horizon for me to expect any short-term trends to revert to the mean. For most people who don't really want to think about their stock portfolio, they should not feel at all bad about putting all of the money they have for stock investments into the US Total Stock Market Index (https://investor.vanguard.com/mutual-fun...file/VTSMX) if you can, otherwise the S&P 500 Index is just fine, too. The important thing is to put your money into a broad base of companies and then leave it there until you need it.
Reply


Messages In This Thread
RE: The Myth of the Rising Stocks over Time - by Obi2Kenobi - 07-22-2018, 05:50 PM

Forum Jump:


Users browsing this thread: 4 Guest(s)