Less Bang For The Buck

June 17, 20918

by Steve Stofka

There are two types of inputs into production, human and non-human. Over a hundred years ago, Henry Ford realized that he had to invest in his human inputs as well as his equipment, land and factories. Once he started paying his employees a decent wage, they were able to buy the very cars they were producing on Ford’s assembly line.

The total return on our stock investments depends on two inputs: dividends and capital gains, which is the increase in the stock price. Both are dependent on profits. Dividends are a share of the profits that a company returns to its shareholders. Capital gains arise from the profits/savings of other investors who are willing to buy the shares we own (see end for explanation of mutual funds).

In the past three decades, a growing share of total return has come from capital gains. Because of that shift from dividend income to capital gains, market corrections are harsh and swift.

In the 1970s, stocks paid twice the dividend rate that they do today. It took an oil embargo and escalating oil prices, a continuing war in Vietnam, the impeachment of President Nixon, a long recession and growing inflation to sink the market by 50% beginning in early 1973 to the middle of 1974.

In 2000, the dividend rate or yield was a third of what it was in 1973. Total return was much more dependent on the willingness of other investors to buy stocks. In 2-1/2 years, the market lost 45% because of a lack of investor confidence in the new internet industry, a mild recession and 9-11. Dividends act as a safety net for falling stock prices and dividends were weak.

In 2008, the dividend yield was about the same as in 2000. In 1-1/2 years, the market again lost 45% of its value because of a lack of confidence brought on by a financial crisis and a long and deep recession.

Other bedrock shifts have occurred in the past three decades. Corporate debt is an input to production. In the post-WW2 period until 1980, corporate debt as a percent of GDP was a stable 10-15%. $1 of debt generated $7 to $10 of GDP. Following the back-to-back recessions of the early 1980s until the height of the dot-com boom in 2000, that percentage almost doubled to 27%. Each $1 of corporate debt generated less than $4 of GDP.


Today $1 of corporate debt generates just $3 of GDP. Debt is a liability pool. GDP is a flow. That pool of debt is generating less flow. It is less efficient. In 1973, $1 of corporate debt generated 46 cents in profit. Now it generates just 30 cents.

To hide that inefficiency and make their stocks appealing to investors, companies have used some of that debt to buy back their own stock. This reduces the P/E ratio many investors use to gauge value, and it increases the leverage of profit flows.

Here’s a simple example to show how a stock buyback influences the P/E ratio. If a company makes a $10 profit and has 10 shares of stock outstanding, the profit per share is $1. If the company’s stock is priced at $20, then Price-Earnings (P/E) ratio is $20/$1 or 20. If that company borrows money and buys back a share of stock, then a $10 profit is divided among 9 shares for a per-share profit of $1.11. The P/E ratio has declined to 18. When the company buys stock back from existing shareholders, that often drives up the price, and thus lowers the P/E ratio further.

The P/E ratio values a company based on the flow of annual profits. A company’s Price to Book (P/B) ratio values the company based on a pool of value, the equity or liquidation value of the firm. If we divide one by the other, we get an estimate of how much profit is generated by each $1 of a company’s equity, or Return On Equity (ROE).

1982 was the worst recession since the Great Depression. Stocks were out of favor with investors and were at a 13 year low. In 1983, $8.70 of equity generated $1 of profit for companies in the SP500. Seventeen years later, at the height of the dot-com boom in 2000, companies had become more efficient at generating profits. $6 of equity generated $1 in profit. In the last quarter of 2017, companies have become less efficient. $7.30 of equity generated $1 of profit.

Let’s look at another flow ratio, one based on the flow of dividends. It’s called the dividend yield, and the current yield is 1.80, about the same as a money market account. I can put my $100 in a money market account or savings account and earn $1.80. If I need that $100 a year from now, it will still be there. I could use that same $100 and buy a fraction of a share of SPY, an ETF that represents the SP500. I could earn the same $1.80. However, if I wanted my $100 back in a year, it might be worth $120 or $50. A stock’s value can be very volatile over a short time like a year, and the current dividend rate does not compensate me for that extra risk.

Why don’t investors demand more dividends? After the early 1980s, economists at the Minneapolis Federal Reserve noted (PDF) that, on a global scale, companies’ profits grew at a faster rate than the dividends they paid to shareholders. The dividend yield of the SP500 companies fell from 6% in the early 1980s to 1% in 2000 (chart).

Those extra profits are counted as corporate savings. The same paper showed that global corporate savings as a percent of global GDP increased from 10% in the early 1980s to 15% this decade. Each year companies were adding on debt at a faster pace than during the post-war decades, but undistributed profits were growing even faster. The net result was an increase of 5% in the rate of corporate saving. Companies around the world were able to shift dividends from the savings accounts of shareholders to the savings of the companies themselves.

From the early 1980s to the height of the dot-com boom, stock prices increased more than ten-fold. Investors that had depended on company dividends for income in previous decades now depended on other investors to keep buying stocks and driving up the price. The source of an investor’s income shifted slightly from the pocketbooks of corporations to the pocketbooks of other investors. Investors adopted a shorter time horizon and now look to other investors to read the mood of the market.

The bottom line? If investors rely on each other for a greater part of their total return, price corrections will be dramatic.


Notes: Price to Book (P/B) ratio was 1.5 in 1983 (article).
P/B graph since 2000
When mutual funds sell some of their holdings, they assign any capital gains earned to their fund holders. This amount appears on the mutual fund statements and the yearly 1099-DIV tax form.

A November 2017 article on share buybacks at the accounting firm DeLoitte


Price Dividend and CWI

August 11th, 2013

Last week I wrote about viewing trends in the market through the lens of hard cold cash; that is, the dividends paid by the companies in the SP500.  Today, I’ll revisit that subject in a bit more depth.  Beginning in the last quarter of 2008, reported earnings of companies in the SP500 dropped precipitously, plunging about 90% in the first two quarters of 2009.

The portion of those earnings paid as dividends fell 24% from peak to trough, far less than earnings.

Robert Shiller, a Yale economist and co-developer of the Case-Shiller housing index, uses a smoothing technique for calculating a Price Earnings ratio and graciously makes his data available.  He calculates the 10 year average of real, or inflation-adjusted, earnings and divides the inflation adjusted price of the SP500 by that average.  Because of the low inflation environment for most of the past decade, the difference between the two earnings figures, nominal and real, is slight.

The drop in corporate earnings was extreme, more so than any recession, including the Great Depression of the 1930s.   In the 2001 recession, earnings declined to about half of their prerecession peak.  In the recession of the early nineties, it was about 30%.  In the back to back recessions of the early 1980s, corporate earnings fell about 25%.

While Shiller’s method evens out earnings, it has one drawback, one that no one could have foreseen until 2008 simply because it had never occurred.  The severity of the decline in earnings skewed the ten year average of earnings down over the 2002 – 2012 period.  Since the earnings average is the divisor in the Shiller P/E ratio, it correspondingly makes the ratio of the price of stocks a bit higher than it might otherwise be.

For that reason, I’ll look at a less volatile ten year average of dividends; that is, the inflation adjusted price of the SP500 divided by the ten year average of inflation adjusted dividends.

Today’s market prices are at the twenty year average of the real price dividend ratio, which is about 61.  For a number of factors, market prices as measured by this dividend ratio are higher for the past twenty years than the thirty year average of 51.  The tech and real estate bubbles over-inflated prices but investors have been willing to pay more for stocks as bond yields have declined steadily from their nosebleed levels of thirty years ago.

Let’s crank up the time machine and go back a year.  Here are a few quotes from an October 13, 2012 Reuters article after the market had dropped about 2%:

“Central bank-fueled gains took markets within reach of five year highs in September, but now U.S. stock market participants are shifting their focus back to corporate outlooks, and the picture is not pretty.”

The article quoted the director of investment strategy at E-Trade Financial, Michael Loewengart: “The overall tone is so pessimistic that we may see some upside surprises, but we could still suffer considerable losses if the news is bad.”

“Profits of SP500 companies are seen dropping 3% this quarter from a year ago, the first decline in three years”

It was close to being almost the end of the world.  As you read various comments in the news, keep in mind that these remarks are coming from active traders who see a 5% drop as catastrophic if they have not anticipated it through options and other hedging strategies.  For longer term investors, a 5% drop after a 5% rise over several months is more yawn provoking than cataclysmic.

Through the middle of November 2012, the market would drop another 5%.  Slowing corporate profits and the looming – yes, looming – fiscal cliff spooked investors.  Then, on the hopes that the Fed would do something to offset these negatives, the market regained the 5% lost in the previous month.  In mid-December, the Fed announced that it would double its bond purchasing program and the market has been rising since, gaining 20%.  Has this been a new bubble, one we’ll call the “Fed Bubble?”  Some say yes, some say no.

As we read the daily news, let’s keep in mind that in ten years we will have forgotten most of it.  Some fears will seem silly, some may seem prescient.  Each day there are many predictions, some like this one from December 30, 2001: “By the year 2003, there will be 2 types of businesses, those doing business on the internet and those out of business.” (Sorry, I didn’t write down the attribution).  Some predictions will seem rather silly like the one in March 2009 that the SP500 would be below 500 in a month.

Farmers and businessmen in ancient Rome consulted soothsayers who threw chicken bones and read the pattern in the bones to tell their clients whether there would be rains in the spring and how hot the summer would be.  Sometimes they were right, sometimes they were wrong.

Each day the market goes up – or it goes down.  For the past twenty years it has gone up 54% of the time, down 46% of the time.  Going up seems like an odds on favorite but this is complicated by the fact that the market usually goes down faster than it goes up.  There is also a well documented behavioral phenomenon of risk aversion; people respond more emotionally to loss than we do to gains.

This past Monday came the release of the ISM monthly survey of Non-Manufacturing businesses.  Like the manufacturing survey released a few days earlier, this index also surged upward in July, a welcome relief after the declining numbers in June.  I’ve updated the composite CWI that I introduced a while back and compared it to the SP500 and the Business Activity Index of the Non-Manufacturing Survey.

This composite index is weighted 70% to non-manufacturing, 30% to manfacturing.  Because this CWI relies on past months’ activity as a predictor of future conditions, it responds with less volatility to a one month surge in survey data.  As we can see, the tepid growth that began appearing this past spring is still showing in this index, although it is a strong 55.5, indicating sure footed, if not surging, growth.  It has been above the neutral mark of 50 since August 2009.