Buffett's 50% per year on small sums

[wabuffo]

(No, I'm not one of the 6 Buffetts in the poll).

 

he he - I'm not either (actually I didn't even answer the poll)  8)

 

I always go back to Buffett's quote about making 50% returns in the early 1950's (before even the BPL days). IIRC, he said something like "you should check my record ... I killed the Dow".  He reportedly started with $7800 at the beginning of 1950 and finished with $174,000 in 1956. (various sources, Snowball, Kilpatrick's books, etc).  That's a 55% CAGR if my math is right and assuming no capital additions by him during that period.

 

We can get a small insight into his portfolios of that period.  Its Buffett's year-end portfolios from 1950 and 1951 (he made a 75.8% return in '51).  The source is from one of Andy Kilpatrick's books.  His 1950 portfolio did not included margin but was only at $9800 but his 1951 portfolio includes margin of $5000 ("bank loans") on a long portfolio of $24,800.  Since we know that Buffett continued to use borrowed money in his Buffett Partnership days he probably used increasing amounts of leverage in this period (1950-1956) as his capital base grew.

 

I'm not saying that it's not theoretically possible to get 50%, 100% returns from a single stock that you hold throughout some ups and downs (particularly if the business retains its earnings and grows quickly) -- but I don't think anyone can generate 50% returns from even a concentrated portfolio of 2-5 stocks, long-only, held over 5 years.  Not without margin or options (and a lot of luck).

 

Even Buffett's own example (1950-56) was one that was helped by margin.  This does not diminish the accomplishment by Buffett or my respect for his record (we're still talking probably 30% unlevered returns).  I just worry that he unintentionally misleads when he talks about the levered record but omits mentioning the use of leverage in helping to achieve it.

 

wabuffo

[alwaysinvert]

The idea of a long-term 50% CAGR is ridiculous. Someone asked for a fund posting those numbers. What do you think would be happen to such a fund? Do you actually realize how fast that compounds thus hampering returns? Not to mention how fast hot money would chase down the returns by providing size.

 

But OK, I grant the possibility that a Buffett like genius could compound at 50% for 10 years plus if he had an extremely high withdrawal rate and didn't take OPM. Why that would ever be the case.

[thepupil]

Are there other examples of 50% gross for 10 years in public stocks?

 

https://www.finect.com/file/download/blog/0020e47b2da014b31069eb60e47b2da014b31069eb6

 

Joel Greenblatt is the managing principal

and co-chief investment officer of Gotham

Asset Management, the successor to Gotham

Capital, an investment firm he founded

in 1985. He has a long history of being a

successful value-driven investor. At Gotham

Capital Greenblatt built one of the best

10-year track records around: compounding his

capital at 50% per year from 1985-1994 (net return

for this period was 34.4%). At the end of that period

Greenblatt returned all of his partners’ capital.

 

 

[Gregmal]

I would also think parameters should be established. With leverage or if you are able to borrow it would actually be quite easy. I don't think its a stretch to say that generating a positive return is difficult at all. One should be able to be net positive over any 3 year period regardless of market conditions. Sure a curveball comes ones way like in 08, yea, you'll be down, but a running 3 year period gives enough time to adjust. So if you can generate a positive return, how difficult would it be to take 10k of your own capital along with a 100k home equity loan at a 4% rate. With margin buying power at 2%, you have essentially 220k to work with in order to pay 4-6k in interest and make an additional 5k for your 50% return. Using the 220k figure you'd need to net ~5%. Your returns would need to increase but lets say you do 15% the first year, then you're already at 34k off a 10k starting point.

[Jurgis]

Another person who wished to remain anonymous send me this via private message:

 

I'm one of the people who have earned 50%+ for 10 years or more. I'm right at the 10 year mark. I don't invest in stocks, but real estate. This probably doesn't count or isn't exactly what you're looking for, but I have made money from it.

[wabuffo]

I don't invest in stocks, but real estate.

 

I would presume a fair amount of borrowed money is employed in the generation of those returns?

 

I found this study interesting...

http://blog.alphaarchitect.com/2016/02/02/even-god-would-get-fired-as-an-active-investor/

 

Basically - it assumes with perfect foresight, one could pick the top decile stocks performers for the next five years starting in 1927 and continuing until end of 2009.  At the end of each five year period, portfolio would rebalance to the best performing stocks for the next five years and so on...  In other words, its as if you get the stock pages exactly five years from now and buy the top decile stocks based on price performance.  Dividends are included and an assumption is made for transaction costs.

 

Even in "god-mode", the returns for the top decile long-only portfolio are 28.9% vs S&P at 9.6%.  The other thing to note is you are still not insulated from some nerve-wracking declines (which might shake out weaker hands).

 

http://i66.tinypic.com/r9n2qa.jpg

 

The author also attempts a long-short portfolio with the same methodoloy (ie, short a basket of the worst-performing stocks) that attains close to 50% but also experiences some bone-rattling swoons.

 

I think the key points are:

1) Achieving 20%+ CAGRs with a long-only portfolio is very, very hard over a long period of time (maybe even impossible)

2) To attain these strong results, you also have to have the emotional capability to sit through declines of 50% without bailing (and cementing a permanent loss of capital).

 

Again - portfolio concentration + using margin could be used to amp up returns, but those declines can be a killer.

 

I think Buffett did a disservice talking about guaranteeing 50% returns because it could lead people into thinking they can do it.  They will be disappointed, ultimately.

 

wabuffo

[KCLarkin]

Wabuffo, that is a great study. But it is explicitly limited to the 500 largest US stocks. We know this is not where Buffett thinks he can get 50% returns. He is talking about nano-caps and emerging markets.

 

A decile portfolio is also 50 stocks. Way too diversified to get exceptional returns.

[wabuffo]

A decile portfolio is also 50 stocks. Way too diversified to get exceptional returns

Well yes -- but I would think the offset that makes it worthwhile is you're like Marty McFly in Back to the Future getting the BEST 50 stocks from a USA Today newspaper printed five years from now!

http://i64.tinypic.com/2nimbe8.jpg

 

At the risk of becoming that boring dinner guest who won't stop talking....  ;D

 

I'd like to also contrast Buffett's stock picking during the 80s and 90s in the Berkshire Hathaway stock portfolio (before BRK transformed into more of a holding company of wholly-owned subs in the double-naughts) with this upper bound from my previous post.

 

I once tried to separate Buffett's raw returns as a stock picker from the leveraged returns from 1982-2000 (using insurance float as a zero-cost margin account).

 

The results:

S&P 500 - 16.8%  (1982-2000 was a great bull market after all!)

BRK Return on investments - 19.1% (Pre-tax return/avg. investments)

BRK Return with leverage - 26.6% (Pre-tax return less leverage costs/avg. equity)

 

Again - I'm not trying to disparage Buffett, but his results were just a bit better than the S&P (+2.3%) and there were probably a few top-tier mutual funds that did as well or better in terms of stock picking performance.  His genius is how he uses different sources of margin to amplify returns.  That's always been what makes him so good -- he thinks hard about how he engineers great results because he is very good at avoiding permanent losses and smart about how he finds and uses leverage.

 

wabuffo

[KCLarkin]

The highest rates of return I’ve ever achieved were in the 1950s. I killed the Dow. You ought to see the numbers. But I was investing peanuts then. It’s a huge structural advantage not to have a lot of money. I think I could make you 50% a year on $1 million. No, I know I could. I guarantee that.

 

Buffett, guarantees this with $1M. Over $10M, it definitely doesn't work. Let's say he has 5 positions, so $200k - $2M per each position. Let's say the max he wants to own per company is 10%. So the sweet spot seems to be nano-caps with market caps of $2M to $20M.

 

I once tried to separate Buffett's raw returns as a stock picker from the leveraged returns from 1982-2000

 

In 1982, his portfolio value was almost $1B. Instead of investing in $2M nano-caps, he needed to invest in $1B+ companies to move the needle. Size was already a massive anchor. Today, he needs to invest in $100B+ companies.

 

 

[no_free_lunch]

This is some interesting analysis wabuffo.  Thanks for posting.  In particular the comments on the engineered use of leverage.  I don't really have anything to add but it is interesting that he just focuses on those areas where he can gain an advantage (insurance leverage, etc) without having to out think everyone (stock picking).

[netnet]

Frankly, I wasn't expecting this discussion to be as long and vigorous as it turned out. My thoughts are:

 

[*]Thanks for the better research on Buffett's quotes.  Clearly he was not talking about control situations, by and large.

[*]I had not realized that he had used some margin early on, 1950 on, but that really magnifies the necessity of being right and being able to withstand a reversal.  His and Munger's friend Rick Guerin, used too much margin and had to sell out during a slump.  His now comfortably rich but not as rich as he would have been.

[*]Great study Wabuffo, but as KC points out you ain't going to make that return in the S&P 500, at least buying 50 stocks.  Wells common on margin close to the bottom tick or Wells' preferreds on margin would have gotten 45% simple interest yearly though.

[*]I don't quite understand some people's thinking that Buffett meant he could make 50% marked to market yearly.  That would be ridiculous.

[*]I have definitely seen real estate guys (high margin) who have made 50% on their capital, but they don't want my money and they may not survive the next blowup.

[*]Buffett also said that he was not the only one.  He knew others who could do this.  Practically it means this is something to aspire to, but realistically by the time you are able to do this, you probably too much money to do it!

[*]Reading this board for ideas would not be the waste as someone said, but writing up posts explaining how you think that Warren....nevermind ;)

 

[tombgrt]

I don't invest in stocks, but real estate.

 

I would presume a fair amount of borrowed money is employed in the generation of those returns?

 

I found this study interesting...

http://blog.alphaarchitect.com/2016/02/02/even-god-would-get-fired-as-an-active-investor/

 

 

Interesting study, thanks.

 

But the stocks picked are the largest 500 and those are often assumed to be the most efficiently priced. It's unlikely any data is available for the first few decades but it would be fun to see how small caps would fair. I'm assuming they would easily beat the large caps.

 

Prices might also be more attractive both in the lows and highs somewhere during the 5-year period. As a normal investor you aren't bound to this holding period of 5 years and you are able to roll over investments more often. One of the strongest tools of mentally stable investors is their ability to use volatility to their advantage.

 

As someone else stated, 50 stocks are a lot. Randomly picked you would be labeled an index hugger. (On the other hand, I believe Peter Lynch had plenty of stocks in Magellan and still kicked ass.)

 

 

You could make a study with 10 small cap stocks over 6 months periods and you'd get a CAGR that is a multiple of 28.89%.Reversly, this wouldn't indicate that you could achieve absurd CAGR returns over very long time periods either. But these studies are interesting regardless.

[KCLarkin]

IIRC, Buffett used leverage for low risk arb ops not for Generals. So leverage isn't required.

 

If we assume he that he would allow leverage, then we know it is possible because Greenblatt and others have done 50% with small sums plus leverage.

[no_free_lunch]

I have been thinking about this and all I can think of is there are market anomalies that you could exploit with small sums.  They are so rare that I just disregard them but every now and then you see things where it is near risk-less profit.  For instance, I once saw the subsidiary of a company announce a major deal.  The subsidiary stock shot up immediately but for a an hour or 90 minutes as I recall the parent barely budget.  It was just sitting there because there just wasn't a major investor analyzing this who had the time to put in a buy order. There was just no reason for the stock not to have moved up and sure enough an hour or two later it did.

 

These things come up so rarely, I really only see them every year or so but if you had the time and work ethic you might be able to make some great returns looking for these anomalies. The thing is it would literally just be a full time job and you wouldn't be able to scale past a few hundred k.

[DooDiligence]

Given the discussion here; I'm surprised at the lack of comments on the following:

 

http://www.cornerofberkshireandfairfax.ca/forum/books/warren-buffett's-ground-rules-jeremy-miller/msg262646/#msg262646

 

I found it to be an excellent read!

[netnet]

Given the discussion here; I'm surprised at the lack of comments on the following:

 

http://www.cornerofberkshireandfairfax.ca/forum/books/warren-buffett's-ground-rules-jeremy-miller/msg262646/#msg262646

 

I found it to be an excellent read!

 

No comments because most have read his partnership letters already, (only slightly tongue in cheek  ;))

[DooDiligence]

Given the discussion here; I'm surprised at the lack of comments on the following:

 

http://www.cornerofberkshireandfairfax.ca/forum/books/warren-buffett's-ground-rules-jeremy-miller/msg262646/#msg262646

 

I found it to be an excellent read!

 

No comments because most have read his partnership letters already, (only slightly tongue in cheek  ;))

 

I figured that was the case but as someone who's never understood the appeal of all the little highly levered, cruddy looking ideas posted here (I've been reading for nearly 3 years and just recently joined officially) this book has caused the scales to fall from my eyes.

 

I'd love to hear comments but in truth they're all over this forum (just not in this thread...)

[antao]

I believe such returns are only attainable by either deep concentration (5 positions or less making at least 90% of the entire portfolio) and/or the use of moderate levels of leverage. For it to work security selection is of paramount importance and each new valid idea must come about only once a year or so on market conditions such as today's.

[DooDiligence]

I believe such returns are only attainable by either deep concentration (5 positions or less making at least 90% of the entire portfolio) and/or the use of moderate levels of leverage. For it to work security selection is of paramount importance and each new valid idea must come about only once a year or so on market conditions such as today's.

 

Seems like you'd also need to get eyes on the assets to evaluate (has valuable property been depreciated, etc.?)

 

Fishers scuttlebutt:

 

Calling/visiting customers to see if product is sitting on loading bays or to hear if they offer generous credit on products they themselves got generous credit on.

 

Are vendors getting paid or are they selling receivables to factors?

 

What's the parking lot look like (does everyone take off intermittently for long lunches & are they driving hoopties?)

 

Probably a lot of work; fun work though! (the sound of cackling from an evil supervillain who turns out to be misunderstood & is really a good guy who likes the idea of saving & creating jobs...)

[HowMuchValue]

I am finding it unbelievable that you guys on this value oriented board are doubting that Warren Buffett could rack up 50% returns on small sums of capital. You can get those returns too. No, you don't need leverage. No, I am not crazy  8).

 

How? Let's take some value investor mindset and combine it with math.

 

Assumptions:

 

• Use Owner Earnings for all calculations as defined by Warren Buffett (we all know reported "Net Income" is sometimes ridiculously off base)
  
• An average company earns a 10% return on invested capital (Owner Earnings/Invested Capital) and can re-invest capital at 10% rates of return
  
• An average company is worth 15x Owner Earnings (simplistic, but you can and should adjust for debt. You should not subtract 100% of debt necessarily - low yielding debt is actually value creating if the company is used to generate returns higher than its cost. Also increased borrowing base is something that increases as a company earns more money which can cause a virtuous circle)
  
• It takes 3 years, on average, for a company to reach its fair value
  
• Some companies will take less than 3 years and some will take more. The ones that become fairly valued sooner you sell and invest in companies that are more undervalued.
  

 

From the above assumptions, if you buy X company when it is at 15x Owner Earnings, you will earn about a 10% rate of return per year.

 

What happens if you buy the company at 10 x Owner Earnings?

 

The value of this stock, on average after 3 years, would be (15 x Owner Earnings/10 x Owner Earnings) x (1 + 10%)^3 = 1.99x the value today (i.e. doubling in value after 3 years).

 

If your stock doubles in value after 3 years, this is a compounded annual return:

 

(10^(LOG(1.99)/3)-1) = 25.9% annual rate of return.

 

So, if you buy average companies when they are undervalued by about 33%, and sell them when they reach fair value, you will earn about a 26% rate of return over time.

 

If we required a 50% rate of return, then, we would need a stock to go up by this much every 3 years:

 

(10^(LOG(X)/3)-1)=50%

 

Solved for X: X = 3.38

 

We would need each stock, on average, to increase by 238% (3.38 minus the original value of 1) over 3 years.

 

How undervalued would this average stock need to be when we bought it to earn a 238% return over 3 years?

 

(15 x Owner Earnings/Y x Owner Earnings) x (1 + 10%)^3 = 3.38x the value today

 

Y = 5.9

 

Therefore if you bought average companies by the definition above at 5.9 x Owner Earnings and they are revalued to their fair value in times averaging 3 years, you will earn about a 50% rate of return.

 

Companies at 6 x Owner Earnings definitely exist. Generally you have to dig a little deeper than the income statement to understand depreciation versus maintenance capital expenditures (talk about a walloping difference for some real estate companies), cases where amortization is recorded as an expense but definitely isn't one, investments and/or expenses (even on the income statement) that will generate future growth, etc.

 

On top of that, businesses that are above average may deserve a higher multiple and will grow at higher rates than 10%.

 

So yes, it is definitely possible Buffett could beat 50%+ returns.

[thepupil]

Are there other examples of 50% gross for 10 years in public stocks?

 

https://www.finect.com/file/download/blog/0020e47b2da014b31069eb60e47b2da014b31069eb6

 

Joel Greenblatt is the managing principal

and co-chief investment officer of Gotham

Asset Management, the successor to Gotham

Capital, an investment firm he founded

in 1985. He has a long history of being a

successful value-driven investor. At Gotham

Capital Greenblatt built one of the best

10-year track records around: compounding his

capital at 50% per year from 1985-1994 (net return

for this period was 34.4%). At the end of that period

Greenblatt returned all of his partners’ capital.

 

bump. any other examples of a moderately institutional, somewhat verifiable, 50% per year  for a decade? Sorry if i missed it.

[finetrader]

Buffett said that if you look at stocks 52 week highs and lows, more often than not, the spread between the two is around 100%, from low to high.

 

So I think he is banking on capturing this spread when he is saying he could do 50% per year on small sums.

[DooDiligence]

Buffett said that if you look at stocks 52 week highs and lows, more often than not, the spread between the two is around 100%, from low to high.

 

So I think he is banking on capturing this spread when he is saying he could do 50% per year on small sums.

 

He's a closet market timer...

[peter1234]

Here is a good chart from Frederik Vanhaverbeke's book Excess Returns: A comparative study of the methods of the world's greatest investors.

 

Spoiler alert: return scale "only" goes to 30% per year...(I believe it is returns over S&P 500, so maybe add another 10% to get absolute returns).

 

:)

 

 

 

[hillfronter83]

Are there other examples of 50% gross for 10 years in public stocks?

 

https://www.finect.com/file/download/blog/0020e47b2da014b31069eb60e47b2da014b31069eb6

 

Joel Greenblatt is the managing principal

and co-chief investment officer of Gotham

Asset Management, the successor to Gotham

Capital, an investment firm he founded

in 1985. He has a long history of being a

successful value-driven investor. At Gotham

Capital Greenblatt built one of the best

10-year track records around: compounding his

capital at 50% per year from 1985-1994 (net return

for this period was 34.4%). At the end of that period

Greenblatt returned all of his partners’ capital.

 

bump. any other examples of a moderately institutional, somewhat verifiable, 50% per year  for a decade? Sorry if i missed it.

 

"[Renaissance] won the [Labor Department]'s permission to put pieces of Medallion inside Roth IRAs. That means no taxes -- ever -- on the future earnings of a fund that averaged a 71.8 percent annual return, before fees, from 1994 through mid-2014."

https://en.wikipedia.org/wiki/Renaissance_Technologies