1. How to use GuruFocus - Tutorials
  2. What Is in the GuruFocus Premium Membership?
  3. A DIY Guide on How to Invest Using Guru Strategies
Tannor Pilatzke
Tannor Pilatzke
Articles (76)  | Author's Website |

Law of Diminishing Returns - Warren Buffett's Advised Experiment

Warren Buffett gave the following example as a law of diminishing returns and future revenue growth, and I decided to take him up on it:

“A growth rate of that magnitude can only be maintained by a very small percentage of large businesses. Here’s a test: Examine the record of, say, the 200 highest earning companies from 1970 or 1980 and tabulate how many have increased per-share earnings by 15% annually since those dates. You will find that only a handful have. I would wager you a very significant sum that fewer than 10 of the 200 most profitable companies in 2000 will attain 15% annual growth in earnings-per-share over the next 20 years.”

I used the year 1990's Fortune 500 and the year 2013's Fortune 500 for my criteria. (Yes, I used 23 years instead of 20, but don't think much difference is made.)

Based on a capitalization rate of 15% and 23 years as a time interval, I arrived at a future value factor of 24.891 (1.15^23) and multiplied that by the 200th ranking company’s revenue in 1990, using it as a base for 2013 qualification.

2.172 x 24.891 = 54.063 Billion 2013 Annual Revenue as base for criteria.

I did not have much of a list so I decided to add in companies that had 23-year sales growth over 13%
($36.113 billion 2013 sales as the base).

Still only 86 companies in total made the cut. I manually sorted through the list for companies that were both in the top 200 in 1990 as well as the top 86 in 2013. I found 28 that had over 13% revenue growth for the last 23 years or 14% of the 1990 Fortune 200.

Company 2013 Rank 1990 Rank
1 Exxon Mobile 2 3
2 Chevron 3 11
3 Phillips 66 (SpinOff) 4 30
4 Berkshire Hathaway 5 179
5 Apple 6 96
6 General Motors 7 1
7 General Electric 8 5
8 Ford 10 2
9 Hewlett-Packard 15 33
10 IBM 20 4
11 Archer Daniels Midland 27 57
12 Procter & Gamble 28 14
13 Caterpillar 42 38
14 Pepsi 43 23
15 ConocoPhillips 45 30
16 Johnston & Johnston 41 47
17 Pfizer 48 80
18 United Technologies 50 17
19 Dow Chemical 52 20
20 Intel 54 137
21 Coca-Cola 57 121
22 Merck 58 70
23 Lockheed Martin 59 45
24 Johnson Controls 67 126
25 Abbott Labortories 70 90
26 Dupont 72 9
27 Honeywell 78 65
28 Deere 85 66

How many of the 28 companies had over 15% annual EPS growth for the last 20 years? Buffett’s wager was that fewer than 10 had done so. I used net income as a proxy for EPS.

Lockheed Martin (NYSE:LMT) had the largest CAGR of the bunch, making the earnings cut with $2 million 1990 earnings compared to $2.745 billion in 2013 or 36.29% CAGR over the last 23 years.

Coca-Cola (NYSE:KO) also makes the cut from 1990 earnings of $71.7 million to $9.019 billion or a 23.39% CAGR for the last 23 years.

Intel (NASDAQ:INTC) was another company with over 15% growth for the last 23 years, growing from $391 million net income in 1990 to $11 billion in 2013. Intel managed a 15.614% CAGR for the last 23 years.

ConocoPhillips (COP) went from 1990 net income of $219 million to $8.428 billion in 2013 or good for a 17.2% CAGR.

Apple (NASDAQ:AAPL) was one of four in the 20%-plus club, growing from $454 million 1990 net income to a phenomenal $41.733 billion in 2013 or a 21.72% CAGR.

Berkshire Hathaway (NYSE:BRK.A) (BRK.B) is really no surprise here considering the CEO and team who are running the place. Berkshire had 1990 net income of $447.5 million and 2013 net income of $14.824 billion or a 23-year CAGR of 16.44%.

Chevron (NYSE:CVX) also had phenomenal growth numbers over the 23-year span, growing from $251 million to $26.179 billion or a whopping 22.39% CAGR.
Looks like Buffett’s bet would have paid off with only seven companies from the 1990 Fortune 500 growing both revenue at 13% or higher and net income at 15% or higher. It is crazy to think that if one took the Fortune 500 in 1990 (or possiblly now), with the goal of at least a 15% CAGR from investment, the chances of doing so would only be about 3.5%.

Score one for the small caps?

About the author:

Tannor Pilatzke
I am a self taught investor through Warren Buffett, Charlie Munger, Ben Graham, Peter Lynch, Joel Greenblatt, David Einhorn, Seth Klarman, Howard Marks, Phillip Fisher and Thornton O'Glove. My focus is a bottoms up Value-GARP strategy with a mix of top down contrarianism.

"When you find yourself on the side of the majority, it is time to pause and reflect." - Mark Twain

Visit Tannor Pilatzke's Website

Rating: 3.8/5 (24 votes)



Seanickson - 6 years ago    Report SPAM
I appreciate the article, but I think net income is a fairly poor proxy for EPS, the two can differ fairly significantly for a lot of companies.
BEL-AIR - 6 years ago    Report SPAM
Thats why most of your gains should come from the appreciation from under valuation to full valuation, and not from the growth of the company, buy for 50 cents to 25 cents and sell for 95 cents, might take you a few years to happen, but you get the valuation growth plus the business growth with dividends if any on top, then you look for another 50 cent bargain... You only buy and hold for 35 years if you are to big, other wise returns will be alot smaller.

Take buffet's returns as an example, his returns are significantly less today compared to the days he bought and sold every few years, so instead of buying a dollar bill for 50 cents or less and selling once it goes to full value, then wash, rinse, and repeat. Now he is so big he is forced to buy and hold forever, and as your example above shows, odds of getting high returns are slim indeed even for the oracle.
Tannor - 6 years ago    Report SPAM
I agree Seanickson, depending on shares outstanding but it was the best proxy I had for the time constraints I was dealing with and don't think the outcome would be changed much.

BEL-AIR - I see where you are coming from, I guess it comes down to timing and the holding period. $1 available for 50 cents is a great bargain provided it does not take 20+ years to materialize. The problem with equally weighting the valuation components and the growth components is that valuation usually has a hypothetical limit and does not grow exponentially where as growth can/will until the law of large numbers sets in. I guess the point (that I drew from the experiment) is that unless you "have to" invest in large and mega caps (dictated by policy) you should view lower capitalization niche companies first.

I should have gone further and split the companies that under performed the DJIA or SPY from the ones that did not over the same time interval. I would be interested to see how many still outperform the market as a whole, even after missing the 15% hurdle rate.

BEL-AIR - 6 years ago    Report SPAM
One thing you have to remember and I just pointing out, the obvious and true 50 cent bargains do not come along often or stay that low for long.... Yes some companies always appear to be cheaper but that is were a person with experience valuing companies and knowing historical valuations comes in. So either not buying 50 cents on the dollar or buying companies or even industries that always trade at lower valuations historically or not true bargains at all.

Now of course the kicker in this all is in buying a company you think is 50 cents but then the fundamentals deteriorate after you buy it and you end up actual paying full value or more.... Still this is were your gains come from, not in the growth in most cases, since like your research shows not many companies can grow much consistently over a long time period.
Tannor - 6 years ago    Report SPAM
I respectfully disagree and that the most of returns come from growth as shown with the companies above. The CAGR from multiple expansion from 50cents to $1 over the 23 years is 3.06% and even if the entire basket of 7 companies were bought at 0.25 cents on the dollar at the time that would add 6.2% to the returns.

Look at the returns for the companies, the growth is the much larger component. 50 cents on the dollar can be found in combination with growth and those are the hardest to find but best investments. This is the premise of "Its far better to buy a wonderful company at a fair price than a fair company at a wonderful price" - Warren Buffett

I agree with the difficulty of finding 50 cent dollars but that is why one must constantly read and seek out ideas. Valuation mistakes can happen and it segues perfectly into a quote from Peter Lynch.

In this business, if youre good, youre right six times out of ten. Youre never going to be right nine times out of ten.

BEL-AIR - 6 years ago    Report SPAM
"Its far better to buy a wonderful company at a fair price than a fair company at a wonderful price" - Warren Buffett

He never said that in his first 10 years of investing when he was making 30% yearly returns... You really think he held companies for 23 years when he was making those kind of returns? Or Graham, or Schloss, or Templeton? They bought for 50 cents... Even guys like Tepper and Prabrai are not owning companies for 23 years... This is were you are wrong.

You like many others are confusing the huge returns he made in his early years to what he is saying now in the last 20 years, big big difference.

In fact hear from the oracles own mouth how he would invest if he was starting out with millions and not hundreds of billions he has today...


"I would buy classic Graham type stocks, low pe's or below working capital"

You don't hold these for 23 years or have fundamentals increasing by 15% for 23 years...

You should emulate buffett in his early years, not now when his returns are a fraction of what they used to be as they are totally different styles....

I just trying to help you see the difference...

Sapporosteve premium member - 6 years ago
I am with Bel-air on this one. It amazes me how many people think that investing like Buffett does today will make them wealthy. As Bel-air points out, when Buffett had less money (the same amount as most of us do today), his turnover was a lot higher. But so was his returns. It is the same with Ben Graham also. After 40 years Graham is on the record as saying basically just buy a lot of cheap stocks on 2 criteria and they will return about 15% on average. Yep some losers but the winners more than make up. So in the end Graham was actually a quant. But many people seem to spend an inordinate amount of time checking companies management and all the ratios and their future prospects and there history etc. And I am not sure whether it actually works out to be worth it. The magic formula (which I dont use) but I do use something similar has great returns, turnover once a year and basically ignores a lot of fundamental research. But basically you are probably better off with the Graham approach since my holding period is not "forever" as Buffett says. You can do that if you own 5 or 10% of a mega-cap and cant really sell down that quickly. regards Steve
Haoafu - 6 years ago    Report SPAM
While I think there's an argument to be made on both sides, I agree with the author that growth(or company quality) is a more important factor mainly because it's way more difficult to find pure vaue stocks in last 30-50 years compared to the Graham days.

Buffet and Charlie once said if you take out their best 10 or so inverstments in the past, the overall return would be just average. Those 10 best investments include Geico, Washington Post, American express, Cocacola... Dig deeper and you'll find they enjoy the best of both worlds: those investments are bought at deep discounts and they are also great companies. Also you have Greenblatt's magic formula and his own fund producing 30+% for over 20 years not based on pure Graham style stocks.

Overall I think buying good companies at relatively cheap(or below) price is just a natrual evolution of investment philosophy. I know people who bought great companies at bargain basement prices and get rich quickly, but those are lifetime opportunities well capitalized.

Tannor - 6 years ago    Report SPAM
Thanks everyone and I can see both sides of the argument as Buffett has said, "Growth and Value are joined at the hip." Ironically, Graham's best investment was GEICO.. giving a higher returns than most other investments made, combined.

"By 1972, at the Companys then peak operating level, a single original Graham-Newman share of $27 had grown in worth to $16,349 not quite the classic single cigar-puff, textbook Graham and Doddsville stock. Graham-Newmans original 1948 investment of $712,000 was worth over $400,000,000 25 years later. Graham had scored a Peter Lynchian 500-bagger. The gain in GEICO would come to represent a much larger percentage of the firm's profits than its other investments combined. "
Jk815 - 6 years ago    Report SPAM
don yacktman answer 16. For small and amateur investors, which approach do you think is better? Cigar Butts or Wide Moat? A: That really depends on price. Whats better? Whats cheaper? It is easy to find the moat and stick with it and the problem with cigar butts is that very often there is no value to begin with because management is burning the money. Margin of safety is just a function of the price you pay. At the right price, either approach will work. My advice for investors is to be open to all kinds of opportunities and focus on what the return on the investment.
BEL-AIR - 6 years ago    Report SPAM
Geico was Grahams largest gain by total monetary gain, but not annual percent wise, this is due to the fact that he held it the longest, not that it had the best yearly returns.

If you do the calculations he made 17.5% a year on Geico over 25 years by buying and holding, yet he made 21% a year over 30 years buying net nets... So it is rather easy math to do. But he bought over 4,000 stocks in his life and only ONE turned out a real compounder... The others were not, so like your research shows it is hard to find, plus you have to buy it at a discount as well to even get the returns he got from Geico... He bought it at a liquidation type price.

Again this is from Grahams own mouth...

Graham: "I have a considerable amount of doubt on the question of how successful analysts can be overall when applying these selectivity approaches. The thing that I have been emphasizing in my own work for the last few years has been the group approach. To try to buy groups of stocks that meet some simple criterion for being undervalued - regardless of the industry and with very little attention to the individual company.

My recent article on three simple methods applied to common stocks was published in one of your Seminar Proceedings. I am just finishing a 50-year study - the application of these simple methods to groups of stocks, actually, to all the stocks in the Moody's Industrial Stock Group. I found the results were very good for 50 years. They certainly did twice as well as the Dow Jones. And so my enthusiasm has been transferred from the selective to the group approach. What I want is an earnings ratio twice as good as the bond interest ratio typically for most years.

One can also apply a dividend criterion or an asset value criterion and get good results. My research indicates the best results come from simple earnings criteria, and the thing that I have been talking about so much this afternoon is applying a simple criterion of the value of a security. But what everybody else is trying to do pretty much is pick out the Xerox companies, the 3Ms, because of their long-term futures or to decide that next year the semiconductor industry would be a good industry. These dont seem to be dependable ways to do it. There are certainly a lot of ways to keep busy.

These are Graham words not mine....

Thanks for the lively discussion though, take care all.
LwC - 6 years ago    Report SPAM
@BEL-AIR, I'm wondering how you calculated the Geico return on investment over 25 years as 17.5%.

Using an equation for calculating the compound annualized return, I find the 25 year return to be about 29% per year. Here's the equation that I used:

Compound annualized return = (400M 712,000) ^ 1/25 - 1= .29= 29%

When I assume your 17.5%, I get to $40M:

Final value = 712,000 x (1+.175) ^ 25 = 40M

Doing the same calculation for 29%:

Final value = 712,000 x (1 + .29) ^ 25 = 400M

[edit] It occurred to me that maybe you left a zero off of 400M when you did your calculation?
Vgm - 6 years ago    Report SPAM
In support of comments above that Buffett would do things differently if he was running much smaller sums of money, here's his response to a question from Univ of Maryland MBA students November 15 2013:

"WB: I developed my investment strategy under Graham. I went to Columbia and learned from Graham. With Grahams approach, you cannot lose money over time. Its very quantitative in nature, and you have to do reasonably well. On the other hand, it has less and less application as you get into bigger and bigger companies with larger sums of money. Its better to buy wonderful businesses at fair prices than so-so businesses at low prices.

With the cigar approach, you can find a nasty cigar on the ground, with one puff left, can pick it up, light it and you get a free puff. You can keep doing this and get many free puffs. Thats one approach, thats what I did. I looked for very cheap stocks quantitatively. After exposure to Fisher and Charlie, I started looking for better companies. Previously I was doing both. Now we are looking for good companies, not just cheap companies. Railroads are huge, and they will be good in 10 years, and 100 years from now. Burlington Northern is now earning $6 billion pre-tax, as compared to $3 billion a few years ago before we bought it. Moving much towards Fisher now and less Ben Graham because we are working with larger sums. With smaller sums, we would be looking at better margins/cheaper stocks."

The notes taken are worth a read.


Please leave your comment:

Performances of the stocks mentioned by Tannor Pilatzke

User Generated Screeners

pascal.van.garsseHigh FCF-M2
kosalmmuseBest one1
DBrizanall 2019Feb26
kosalmmuseBest one
DBrizanall 2019Feb25
MsDale*52-Week Low
Get WordPress Plugins for easy affiliate links on Stock Tickers and Guru Names | Earn affiliate commissions by embedding GuruFocus Charts
GuruFocus Affiliate Program: Earn up to $400 per referral. ( Learn More)