The stock market sure looks like a bubble now but it’s actually very difficult to distinguish bubbles from rational behavior when dividend growth rates can change and must be forecast. Ironman has an excellent post on this at Political Calculations, titled Acceleration, Amplification and Shifting Time. See also the simple model discussed in my paper with Gary Santoni which Ironman kindly cites.
I don’t deny that bubbles exist, by the way, my point is that the fact that they are difficult to distinguish from rational behavior is one reason that they exist.
Just suppose that real GDP is increasing at 3% per year, while real stock market yields summed over the whole market are increasing at 5% a year. And the common wisdom is that this can continue indefinitely.
Is there a way to see that as something other than a bubble?
I considered writing some personal anecdotes that made it look like a bubble to me, but there’s no point, and there isn’t anything that special about them. Basicly, GDP went up 3% and my profits were 30%. Was somebody (pension funds?) losing the money I made? Or was there enough new money coming in that prices rose without anybody losing — yet?
Or perhaps was the economy about to start expanding at 30% too? I was pretty sure I could discount that one.
Jason, I think it’s better to compare inflation-adjusted results in both cases, or if you don’t trust the inflation numbers then use nominal results in both cases.
It sounds to me as if you’re saying the stock market is more volatile than the economy, bigger swings in both directions. Kind of … bubbly?
Ironman, I may very well be making a false assumption. But here’s where I have trouble — why would you expect dividends to keep growing at an increasing rate, when GDP does not grow nearly as fast? How can that be sustainable?
Of course it makes sense that stock prices might fit expected future dividends, and that future dividend increases might correlate more with even-more-future GDP growth. But when GDP doesn’t grow, to keep increasing dividends don’t you at some point have to extract more and more of the money from the economy so you can pass them out again as dividends? And won’t that be subject to logistic limits?
When dividends are only distributed quarterly, you can’t very well collect them back as profits so you can distribute them again more than four times a year….
J Thomas wrote:
Again, growth in GDP doesn’t drive dividend growth or, by extension, stock prices.
With the stock market, you’re looking at a very small subset of the entire economy. More specifically, you’re looking at those companies that have already proven to be successful enough to even get to issue stock to shareholders in a public offering.
Looking at the S&P 500 index, in which companies are represented in proportion to their market capitalization, you’re looking at an even smaller subset of the economy – the companies that have been successful enough to grow to be among the 500 largest companies in the stock market as measured by their market cap.
When you weight stocks by market capitalization, you find that an even smaller subset of companies can even affect the overall direction of the market based on changes in their future business prospects. For example, if #1 S&P 500 company Exxon Mobil announces it will have lower earnings next quarter, that news will more than offset a hypothetical blowout quarter announced by #500 S&P 500 company Jones Apparel on the same day. (Here are the current component stock weightings.)
And that should help explain why dividends generally grow faster than GDP. Over time, dividend growth in the stock market is driven by the most successful of a very small portion of the whole economy.
Indeed. Hence the focus on dividends as a measure of a company’s “sustainable” level of earnings. CEOs have huge incentives to deliver growth, but even huger incentives (take that, 7th grade English teacher!) to avoid having to reduce their dividends. This makes changes in stock market dividends highly useful as a tool by which to anticipate its future direction for investors – the company is communicating what level of payouts its leadership believes it can sustain into the indefinite future based on its best available revenue and cost projections.
Where most companies’ stocks run into trouble is when they try to deliver dividends their businesses can’t sustain. Some turn to fraud or highly questionable business practices to artificially inflate their bottom lines (Enron, Fannie Mae), others just get their stock prices whacked because of the disconnect between what they pay out in dividends compared to what their expected future business prospects indicate that they can afford to pay out (GM, New York Times).
There are a number of reason GDP growth and dividend growth might be decoupled.
1. Stocks could account for more/less GDP growth than forms of equity ownership.
2. Taxation changes
3. As was said previously companies could be producing goods in other countries like China
4. Labour’s share of income could go down and capital’s share go up
5. Consumption could be funded by increasing debt or consuming savings.
Maybe instead of looking at GDP growth it would be better to look at consumption growth to predict corporate earnings growth. But this still doesn’t take into account earnings due to consumption in foreign countries.
J Thomas,
It is possible for a stock market index to increase faster than the rate of GDP growth because the component stocks of the index are continually changing over time. In other words, index investing is not passive buy-and-hold investing (see also here).
The S&P 500 itself outpeforms the market due to survivorship bias; the component stocks at any given date in the past, fixed and frozen going forward, in general do not.
Ironman, depending on survivorship bias is a con.
It’s like, “Look, on average the people who bet at this casino won more money than they bet. Here’s a list of people who did just that.” And then you look at it, and it turns out that 30% of the people who bet at the casino actually lost all their money and committed suicide, and the casino didn’t include them in the list because they were dead.
It’s kind of true if you look at it wrong, but it’s more an exceptionally cruel joke.
There are a couple things i think you are all missing.
1. stock market returns are intinsically based on changes in earnings attributable to shareholders.
2. GDP is the value of all final goods and services produced by an economy, so GDP growth is really revenue growth, not earnings growth.
3. most companies have a “fixed” component to their cost structure, so they benefit from operating leverage, ie. a 3% increase in GDP results in a >3% increase to earnings.
4. in addition, companies also typically employ financial leverage by borrowing money to fund projects/acquire fixed assets, etc. the use of financial leverage increases earnings growth.
Some of your factors may also be true, but even in a steady state, “mature” economy, returns to eequity holders should outpace GDP growth.
Comments on this entry are closed.