That’s a Teddy Roosevelt quote. One thing that can drive even the best absolutely crazy is the concept of perpetually escalating expectations. It was so bad that Warren Buffett even contemplated early retirement because the track record of his early days had become more of a curse than a blessing. Never having a down year, beating the Dow Jones by ten percentage points annually, and the growing hype in Omaha started to choke him.
In his last partnership letter dated May 29, 1969, he mentioned that he considered stock market values generally frothy, but also spent time discussing the notion that he didn’t want to be chasing the “investment rabbit” for the rest of his life. He was circumspect in saying that his own past successes had forged their own chains; his early successes were so extreme that the Shakespearean fall awaited and he would spent the last fifty years of his life getting the Bill Miller treatment, “He used to be great.”
Eventually, Buffett was able to get around this by getting into the insurance business. That gave him a structural advantage over everyone else. As a stock picker, most of Buffett’s picks have compounded between 13% and 14% annually. There’s no indication that he is necessarily a better stock picker than Messrs. Munger, Yacktman, Lynch, Schloss, Whitman, and Neff. But he was able to create a legend for himself through the sustained use of leverage from insurance operations that give him low-cost (essentially free) access to upfront capital that he could perpetually deploy into intelligently chosen stocks.
The relationship between your investment returns and the S&P 500 is a fascinating topic. If you generated 5% annual returns with individual stocks from 1990 through 2015, and the S&P 500 returned 9% annually over that time frame, a re-evaluation is probably warranted. The difficult part is figuring out: (1) how to measure whether your stock picking ability is improving as you age so that the poor stock selections in years like 1995 don’t represent your improved selection ability by 2005, as a man can improve tremendously in ten years if he sets his mind on learning something; (2) at what does underperformance against the S&P 500 become more than market noise and start to become an actual indication of poor investment selection ability; and (3) whether a point exists at which your performance against the S&P 500 doesn’t matter anymore.
I’ll tackle the third one because that is the easiest to answer. Yes, if you are generating more in dividends than the 2% offered by the S&P 500, and you have a diversified portfolio of stocks that could support your entire desired lifestyle, there is no reason why you should be beating yourself up over anything. If you’re generating $150,000 per year in common stock dividends, you’ve made the top 1% of the 1%.
Who cares if you get 7% returns while the S&P 500 goes up by 9.5%? It has absolutely no practical application on your life. If you own 50 stocks paying quarterly dividends, you are getting 200 checks deposited into your account per year at an average value of $750. You are beyond set. Once you’ve reached your desired lifestyle, everything else about growth becomes largely academic. Your shift should focus to protecting what you’ve got and ensuring that the income streams continue to at least match inflation, and concerns about your relationship to the S&P 500 should dissipate.
The next question is trickier: At what point should an investor be concerned about a stock’s underperformance against the S&P 500? The answer given at Tweedy Browne (those are the guys who executed the administrative functions of Graham’s stock-picking) is that you should wait five years. They argued that is the point at which you should be able to tell whether or not a stock reaches fair value.
I have mixed feelings on that. Maybe that should be a general presumption, but I think there are many, many examples in which it takes more than five years to figure out whether a stock is a wise selection. Heck, I think we are living through one of those periods right now in that the stock market may advance for a seventh straight year. I’m not sure Tweedy Browne anticipated five straight years (or more) of straight gains when giving their advice. This is certainly one of the longest bull markets without a meaningful correction in modern history.
I think it takes ten years to figure things out in relation to the business cycle because it will almost assuredly capture the ebb and flow of a full economic cycle (and often the least riskiest stock selections prove their worth in poor economies.) If you bought something in 2005 that is not holding up well against the S&P 500 during the 2005 through 2015 comparison period, I think it is fair to question the usefulness of the holding (unless it’s a huge, dividend-paying cash cow that gives you a 5% to 6% yield and the purpose is to deliver income rather than beat the S&P 500.)
My addition to the Tweedy Browne commentary is that I believe investors should focus on the earnings per share growth rate of the firm during your holding period to figure out whether it makes sense. When people talk about those 10% percentage gains from the S&P 500, it is because the S&P 500 historically grows earnings at 6.7% and pays out a dividend at 3.3%.
I would pay attention to those measures in relation to the growth of the companies that I own rather than focusing on the market values. In the 2000s, Abbott Labs was paying out a 2.5% dividend and growing earnings at 8.5%. But the P/E ratio was declining moderately. The art part is figuring out whether you overpaid or whether the stock had gotten cheap over the course of your holding period.
A guidepost: If it is a large, non-cyclical company and you paid over 20x earnings for the stock, a P/E compression to the 16-20x earnings range is not the result of the stock getting cheap. It is a probably a reflection that you overpaid. If a business is growing while the P/E ratio declines from 20x earnings to 15x earnings, I would be more inclined to think that the stock had gotten cheap and it was not a reflection of my investment ability.
People like real-time examples, and so I’ll give one. Look at Aflac right now. The business has grown at 8.5% annually and the dividend is 2.5%. It makes most of its money in Japan, and the stock is cheap because the dollar has performed so well against the yen (seriously, almost 80% of Aflac’s profits come from Japan). If you underperform with Aflac for a while, that’s life. Eventually, that earnings growth will be rewarded. Usually, it happens right after you decide to sell it ;).
Looking at data back to 1984, Aflac has raised its dividend every year. It has compounded at 18% annually. $10,000 grew into $1.8 million. But there were long periods where earnings grow, and the stock price did not. With reinvested dividends, the compounding effects amplified because systematically putting cash every 90 days into a business that is selling for less than its truly worth tends to give you a nice reward at some point. If Aflac kept growing at 8.5% and raising its dividend every year, I can think of no point at which I’d say: “Okay, I’m selling it now for underperforming the S&P 500.” It’s not my style to bail on things that grow at over 7% over the long run.
To determine whether you are improving, I would compare the earnings growth rates of the companies that you select at different stages of life, and adjust for the dividend as necessary. Ideally, I’d want at least ten years to review from the start of an investment, but five years can be tolerable. If the companies you chose in 1995 grew earnings per share at 5% and paid dividends in the 2-3% range, and your 2005 companies grew earnings at 7.5%, you are probably improving your stock selection over time.
I would get worried and move straight to indexing and never look back if I had spent 15+ years being like the people from the Dalbar study that generated 3% annual returns while the S&P 500 returned over 9%. You are leaving so much money on the table that you need to do something else. But if you’re close, and it is clear that your investment selections have improved over time, I would continue to hone the craft by figuring out why I am coming up short.
Almost always, the failure has to deal with selling too quickly. Fidelity pointed out that the best investment accounts are the deceased ones. People don’t get into trouble usually by purchasing inferior assets and holding too long; the trouble comes from abandoning stocks at points of unfashionability and missing the return upward. Marty Whitman said he learns the most by studying the subsequent performance of companies that he has sold.
The best way to improve investment skill quickly probably involves developing strong presumptions against selling investments, and that is why constant comparison to things like the S&P 500 or the latest trend can be hazardous. If your Chevron stock hasn’t been beating the S&P 500, the answer isn’t to sell. It is to recognize that all businesses, especially commodities, run in cycles and holding through the downpoints is the way to make sure you receive all of the investment gains on the upturn. You don’t want to breed petty jealousies by perpetually analyzing what other people are doing because it will likely lead to further underperformance accompanied almost certainly by psychological dissatisfaction.