Someone who reads my articles asked me this question:
From what I can infer from your posts, your own style of investing has parallels of both Graham and Buffett. How come you opt to be less diversified than Graham when having a portfolio concentrated in net-nets? What is your reasoning for straying from his approach? You say that Buffett bought 20 companies when he was investing in less familiar foreign small companies, so he was more diversified than his usual focused approach. How come you don't when investing in net-nets as well?
Thanks,
Tom
I understand why you say my own investing style has parallels to both Buffett and Graham – everybody says that – but it’s really not true. The first explicit rules of value investing I ever learned were from reading Ben Graham. But I never invested like Ben Graham. I always invested like Warren Buffett. I wrote an article entitled “How Long Does it Take to Develop an Investing Style?”
Read that article. It’s the best description I’ve given of how Buffett actually invested when he was starting out. Warren Buffett was never like Ben Graham in terms of his overall approach to investing. He never ran his portfolio the way Ben Graham would. Yes, they both bought some of the same stocks in the 1950s. But this similarity in terms of stock selection did not carry over into portfolio construction.
I concentrate on my top ideas. Buffett always did that. He put 75% of his net worth into GEICO in the 1950s. And he put 40% of the partnership’s money into American Express (AXP). Graham never concentrated on his top ideas – or, almost never. He actually did concentrate his investments in companies like Northern Pipeline. These usually had some sort of activist slant. I’m not sure activist is the right word. It means something a little different today. But Graham did buy big (for him) positions when he wanted a company to pay out cash, or he could have control of the board, or something like that. Still, these were rarely large positions as I define them. They were just very large relative to Graham’s median position size.
I only invest in simple businesses. I’m extremely restricted in terms of the companies I’ll buy into. I wouldn’t buy Intel (INTC) at any price – except maybe below its net current assets, and certainly below its net cash. Intel is not in my circle of competence.
I mentioned that three of the first investment I ever made were: J&J Snack Foods (JJSF), Village Supermarket (VLGEA), and Activision (ATVI). A fourth – the stock I paid the highest price for in all my time investing – was Bio-Reference Labs (BRLI). This was about 10 years ago. They were not involved in some of the more esoteric tests they do today. It was a simpler business back then.
So, we’re talking about things like soft pretzels, groceries, video games, blood and urine tests. Simple stuff. Stuff with a “consumer” angle of some sort. Not capital goods. And not something where a corporate buyer has a pretty free hand in determining demand. Yes, the actual buyer in some of these situations – the “customer” – is a corporation. But that’s like saying that Hanes (HBI) serves Wal-Mart. To an extent it does. But the customer that matters is the mom who buys the socks and underwear and T-shirts for her family. And the people in the family who influence her shopping decisions. If Hanes’s products click with the folks who actually buy and wear the stuff, Wal-Mart will order it. If not, they won’t.
I’m more interested in looking at a situation like that than at some sort of system you put in a factory. I have a hard time knowing what the demand is for that product. And I have a very hard time knowing how much you can charge for the product without alienating customers.
Take some of the stocks I’ve mentioned recently. Three of the most notable recent mentions were DreamWorks (DWA, Financial), Carnival (CCL, Financial) and Royal Caribbean (RCL, Financial). I’m a high school drop-out. I’m not going to understand something technical. But I can understand the value of an experience somebody is selling. I can understand a business like DreamWorks, Carnival or Royal Caribbean.
I rarely spend time looking at companies where the product is bought by a corporation. There are a few exceptions to this:
· The company has a lot of untapped pricing power (I will even buy a capital good if you could raise prices 20% tomorrow).
· The corporate buyer has limited freedom of action because they must respond to the desires of their own customers.
Graham didn’t care about these things. Buffett does.
Now, sometimes people say I must be a Ben Graham type investor because I buy things like net-nets. But as I explained in my articled entitled “How Warren Buffett Made His First $100,000” – Warren Buffett bought net-nets too.
A cheap price is always better than an expensive price. And a good business is always better than a bad business. Where possible, you try to buy a good business at a cheap price. Ben Graham sometimes paid lip service to this approach. But I don’t think he was really interested in the quality of a business when it came to his own investing.
Buffett was interested in business quality from the very beginning. I was interested in business quality from the very beginning. I know this is hard to believe.
Because we know Warren Buffett bought some bad businesses. Yes, he bought bad businesses, when they were offered at really, really good prices. Like I said, Buffett has always been a return on investment investor. He’s always looked at how fast his capital will compound.
If you are offered a business for a 50% discount to its cash and securities net of total liabilities – and that business is usually profitable – you’re going to tend to compound your money faster in that business than in Coca-Cola (KO). It doesn’t matter that the business earns low returns. There’s a decent chance you could get a 150% gain in that stock in one year, or two years, or three years, or five years. As we move out in time – as long as the business is eking out some profitable return – the chance gets higher of you realizing that return. Even if it takes five years to see a 150% rise in the valuation of that business and cash combo you bought – you will still earn an annual return of 20% a year. Actually, you’ll probably earn a few percentage points more a year since book value will keep rising for a profitable business.
You have to sort through thousands of opportunities, but you will find some situations – especially in micro cap stocks – where the business really would be sold for 150% more than the current stock price if it was offered up to a control buyer today.
Usually, these businesses are controlled by families. Yes, this is similar to Ben Graham’s approach. But Graham diversified widely. I don’t.
Why not?
Because I think Ben Graham was wrong to diversify. Anyone who looks through Graham’s portfolio and his track record and the track record of his students – would have to at least entertain that possibility.
Graham continued to do some unnecessarily complex stuff. For example, he kept buying some bonds, preferred stocks, etc. There was no reason for this. From the very early days, it was obvious that Graham-Newman could make more money – and make it consistently – in common stocks.
Graham-Newman had a handful of successful techniques:
· Control positions
· Net-nets
· Arbitrage
· Liquidations
· Related hedges
In theory, all of these opportunities are open to today’s investor. Some of them still work. Even related hedges still exist. They’re rare. But if you pay very careful attention to convertibles, you’ll find sometimes when it’s clear either the common stock or the convertible is mispriced. Convertibles aren’t something individual investors spend time on. And institutional investors need ideas where they can put a lot of money to work.
As a result, even in 2012, you can work through a list of every known convertible and find a few smaller ones where shorting the common stock and owning the convertible looks like a free lunch.
I’m not sure it is. In fact, your annual returns may not be all that impressive relative to what you could do simply by picking good businesses at decent prices. But it is a very high probability bet. And you won’t live and die by which way the market is moving every week, month or quarter.
I wouldn’t suggest hedging as a technique individual investors should use, but within the last couple years there have been occasions where you could guarantee yourself at worst a small loss and at best (if the market tanked) a very decent gain doing exactly what Ben Graham did. Literally, you wouldn’t have to touch options or anything – just go long the convertible and short the common stock the same way Graham did almost 90 years ago.
I don’t do related hedges.
And I don’t have enough money to do control positions.
I’ve invested in net-nets, arbitrage, and liquidations. They’ve all worked well. They haven’t all been fun. And some of these techniques are inappropriate for people with certain personalities. I’ll lump “workouts” together. When I say “workouts” I mean arbitrage, liquidation, and litigation – mostly. And when I say liquidation I mean a situation where it is obvious some of the assets of the business will be returned to shareholders. In some cases, the exact mechanism used is not a total liquidation of the business with all capital returned as cash.
While some people associate this stuff with Ben Graham exclusively – Warren Buffett did all of these things. Within the last 10 years, he owned Comdisco (a liquidation) and Dow Jones (a merger). In the 1980s, he owned Arcata. Read the 1988 letter to shareholders for details.
Arcata is a good example of the kind of situation that will always pop up from time to time in this area. Basically, it’s a situation that involves uncertainty in a rather low-risk way. And people may confuse high uncertainty with high risk. You can make a lot of money over time selling people protection against uncertainty.
Most investors have this totally bizarre need to not just know that they are going to make money – they need to know exactly how and when they’ll become filthy rich.
Anyway, if you go back and read the newspapers from around the time of the Arcata deal, it was more of a slam dunk than Buffett portrays. He says only, “Finally, we had to ask ourselves what the redwood claim might be worth. Your Chairman, who can’t tell an elm from an oak, had no trouble with that one; he coolly evaluated the claim at somewhere between zero and a whole lot.”
The evidence at the time was strongly towards “a whole lot.” And Buffett knew this. This was not the first time a redwood claim had been litigated. And there had been substantial inflation since certain earlier headline making redwood deals had occurred. While it’s true there was a lot of uncertainty about the value of the claim – and how interest would be determined – there was not as much risk in Buffett’s Arcata position as you’d expect in an investment where you could make as much money as Berkshire did.
The reason why people shy away from a good special situation like this one is that it involves uncertainty in both time and value. But it’s not hard to sketch out a couple rows of possible annual(ized) returns under different payout dates and values.
Graham taught Buffett how to think about workouts. But I’m not sure Buffett needed the education. These are very simple calculations. What you need to be able to make money in these kinds of situations is a belief in Graham’s Mr. Market metaphor. If you really believe the market can be wrong when it comes to something as simple as a liquidation, then you can buy into these situations. I should point out – as Buffett did in 1988 – that these are not typical returns. Ben Graham and Warren Buffett both made money in arbitrage by passing on the vast majority of possible deals. They had much higher standards in terms of a hurdle rate than funds that focus on arbitrage. Because it was just one arrow in their quiver, Buffett and Graham only resorted to arbitrage when it offered very attractive annual returns. As a result, the arbitrage they did tended to improve returns rather than just smooth out returns.
Graham was willing to buy into deals like that. Buffett is willing to buy into deals like that. And I’m willing to buy into deals like that.
If I do the math on some tiny liquidation and decide it is almost certain to return 15% a year and most likely will return 30% a year – I will buy into that liquidation. As much as I like owning a good business – as much as it is much more fun to research an actual company – you don’t pass up a very high probability of 15% or better returns when you see it. Especially if you know the new opportunity is not connected to things you are already invested in.
But a willingness to buy a stock that is liquidating – and I don’t do this often (I’ve averaged a little under one liquidation a year) – isn’t really something that makes you more Buffett like or more Graham like. It’s something both Warren Buffett and Ben Graham did throughout their careers.
Where I differ with Ben Graham is in how much attention I pay to the business, its products, etc.
And how much I concentrate on a few positions.
And how much I focus on understanding a business.
Ben Graham was willing to buy any business anywhere as long as it was statistically cheap. I’m not willing to do that. You know I owned Barnes & Noble (BKS, Financial), which was a situation I thought I understood (I was wrong). It’s one of the very rare examples of a retailer appearing in my portfolio. I am not a good judge of retailers. And I know this.
I would never buy a bank or retailer or insurer because it was statistically cheap. That doesn’t mean I would never buy a bank or retailer or insurer at all – at different times a retailer and insurer have been the biggest position I owned. But I knew a lot more about them than just their price-to-earnings, price-to-book, debt-to-equity, etc.
Really, the overwhelming difference between my approach and Ben Graham’s approach is focus. I like to focus on my best ideas. I like to focus on trying to understand the businesses I invest it. Ben Graham liked to spread out his portfolio. He took a broad and shallow approach. I take a narrow and deep approach.
Which is better?
I think you know my answer. Ben Graham did not achieve the kinds of annual returns he was capable of. He did not focus exclusively on common stocks. He passed up some opportunities to take very big positions in select stocks.
Which is odd. Because he had some success in stocks like Northern Pipeline that you would think would teach him to focus on his best common stock ideas – to focus on places where he could add value (like activism and reallocation of capital).
He didn’t. Graham wanted a system. In fact, he really wanted a system that the average individual investor could apply at home. Activism was not part of that system.
During his actual investing career, the approach that seemed best for individual investors was simply buying net-nets. But by the time Graham died, net-nets were much less common than they had once been. So at the end of his life, Graham tried to develop other systems. He wanted simple, formulaic ways to buy value stocks – beyond just buying the increasingly rare net-nets.
Is a formulaic approach – investing by pure statistics in dozens of stocks at the same time – a viable value strategy?
Yes. In fact, it’s very simple to boil down the essence of Graham’s approach into a strategy any investor could profitably follow for years:
1. Dividend Yield > Median Dividend Yield of all stocks
2. P/E< Median P/E of all stocks
3. P/B (Tangible)< Median P/B (Tangible) of all stocks
4. Tangible Equity/Total Liabilities > Median Tangible Equity/Total Liabilities of all stocks
Rank by tangible equity/total liabilities. Buy the top 20 stocks. Swap out stocks to the extent required – but never more frequently than once per year. By the way, you would swap out fewer than 15 stocks on average, because sorting by tangible equity divided by total liabilities tends to keep causing the same stocks to appear in much the same order.
The approach works. It generates a list of Ben Graham type companies – not net-nets, but companies he’d suggest for the “defensive investor” – and it outperforms the market most of the time.
It even outperforms when applied to huge stocks.
In most years, there would be at least 20 such stocks in the S&P 500, and just buying these 20 stocks instead of spreading your money across all 500 stocks would beat the market in most years and especially decline less in bad years.
That idea would hold a lot of appeal for Ben Graham. You could design a fund around it. In fact, you could apply it to just the biggest companies if you wanted to. And the sorting mechanism I mentioned is not important (tangible equity to liabilities above the average for all stocks is all you need to prove a stock is safe enough to be part of a value group – beyond that it’s just superfluous safety).
Now, you might say this proves the power of investing in 20 different stocks, inside of selectively picking a few.
I think this approach is fine. But even here the diversification is unnecessary. If you take the same group which is again stocks with:
1. Dividend Yield > Median Dividend Yield of all stocks
2. P/E< Median P/E of all stocks
3. P/B (Tangible)< Median P/B (Tangible) of all stocks
4. Tangible Equity/Total Liabilities > Median Tangible Equity/Total Liabilities of all stocks
And you say – as I would – there’s no point in picking 20 of these things. Anything that passes this screen is cheap. At most, the issue is about long-term earnings stability. So, why not cut the list of stocks from 20 to 5 and just sort by total dividends paid. This will give you the biggest companies that pay out the most in dividends and retain the least earnings. My expectation is that when those companies are cheaper than the median stocks in terms of yield, price-to-tangible-book and P/E and they have more tangible equity, Mr. Market is wrong. And he’ll be consistently wrong whenever he prices stocks like that at such low levels.
It turns out I’m right. The top 5 dividend payers (in terms of total dollars paid) that pass a median yield, median P/E, median tangible equity total liabilities test very rarely go down in price. I’m not exaggerating. They tend – not just as a group, but each stock individually – to have a very high chance of being at least flat for the next year. Their longer term record isn’t too bad either.
By the way, this is what the list would look like today:
1. ConocoPhillips (COP, Financial)
2. Duke Energy (DUK, Financial)
3. NextEra Energy (NEE, Financial)
4. American Electric Power (AEP, Financial)
5. Public Service Enterprise Group (PEG, Financial)
No one reading this article is going to go home and buy that group. They’re all energy companies. One is an oil company. The other four are utilities.
You wouldn’t be diversified. What if utilities underperform?
It would be crazy to buy this group, right?
I don’t think so. And that is where I differ with Ben Graham. And with all modern conventional wisdom on diversification.
I look at that group and I say: “What’s the chance these stocks will be worth less in the future?”
Even though there are only five of them, the risk of any one stock being worth much less in the future is quite low.
Their yields range from 3.4% to 4.9%. So, we’re talking about a dividend yield equal to or greater than the 30-Year U.S. Treasury bond. Okay, what about price-to-tangible book?
The range is 1.3 to 1.7. Let’s think of that in terms of a long-term rate of return. If I pay 1.3 to 1.7 times tangible book for a collection of U.S. utilities will I do at least as well as the S&P 500?
I might not this year. But, even if these stocks earn just 10% on tangible equity, I will be buying them at normalized earnings yield of 5.9% to 7.7%. And they will pay most of that yield out to me in cash.
Now, compare that to reasonable return expectations in the S&P 500. Pick your favorite predictor of future stock market success – the Shiller P/E, Buffett’s market value to GDP ratio, etc.
I’m much less convinced than most people that you are actually taking more risk by buying these five stocks:
1. ConocoPhillips (COP)
2. Duke Energy (DUK)
3. NextEra Energy (NEE)
4. American Electric Power (AEP)
5. Public Service Enterprise Group (PEG)
I think it’s entirely possible that if you repeated this kind of pattern of buying over time you would not take more risk than buying the S&P 500. You might take less.
Yes, it’s buying just five stocks. But we know they were selected on the basis of:
· Adequate price (Low P/E, Low P/B, High Yield)
· Adequate safety (Low liabilities to tangible equity)
· Adequate stability (Big size, high dividends)
But, of course, it’s a completely undiversified group.
I don’t care. I don’t care that it’s basically just stocks from one sector. If they were selected with safety and cheapness in mind they may be more reliable investments than 500 stocks that were selected without any concern for their safety or cheapness.
Now, this is skirting the issue. I’m a value investor. Of course I prefer a list of value investments – however short – to the entire S&P 500.
Why do I prefer a portfolio of 4 to 12 value investments to a portfolio of 20 to 50 value investments?
Because I’m not sure how much safety is added by increasing the number of stocks you buy.
It’s a complicated issue. Where the greatest risk comes from the risk of catastrophic failure – for example, if you are buying the cheapest banks without regard to their Texas ratio, etc. – I would be all for buying a very big group of stocks because there’s a good chance one of the four or five or eight stocks you pick could prove a total loss even if the group does well.
But I think wide diversification is usually like using the coarse adjustment knob on that microscope you used in high school science. People think diversification reduces risk. Instead, they should be thinking about what risk they need to reduce.
If your answer to every risk was diversification, you would design the Ben Graham screen I spoke of before with just three criteria:
· Low P/E
· Low P/B (Tangible)
· High Yield
Or even just one criteria (low P/E). And then you’d try to fix the problem of all the lousy stocks slipping into the portfolio by diversification.
But, if we take out the tangible equity to total liabilities part of the screen – the results are okay, but they no longer have much to do with Ben Graham. Basically, low price can outperform regardless of safety – but this is only because big winners make up for big losers.
It takes two seconds to stop and think about the problem: “Low P/E stocks might be on the verge of bankruptcy” and then solve it with one of the world’s simplest stock ratios “tangible equity/total liabilities” to create a kind of loss absorption measure for your portfolio.
This is why I question the usefulness of diversification. It makes you think you can’t separate the winners from the losers. And you don’t really need to anyway.
Actually, sometimes it’s quite easy to notice that one stock is in super shaky financial condition and another one is in rock solid shape. I mentioned this same ratio – tangible equity to total liabilities – when talking about net-nets. I also mentioned the ratio of NCAV to total liabilities. This is a very common sense ratio that people who diversify widely in net-nets sometimes forget to apply.
The ideal net-net is safe and cheap. Just cheap may work as a group. But I’d rather try investing in a moderate number of safe and cheap net-nets rather than an obscene number of just cheap net-nets.
Mostly this has to do with risk. It doesn’t matter if you back test a purely cheap net-net portfolio for half a century and prove it works all the time, every time. It’s not a strategy I can get comfortable with.
It is very easy for me to get comfortable holding five stocks I picked myself – and which I believe to be both safe and cheap.
Even in Japan?
Your question about why I only own 6 Japanese net-nets rather than the 20 or so stocks Warren Buffett bought in Korea is a good one.
Basically, I was not comfortable paying for Japanese receivables and inventory (especially inventory). And I was not comfortable investing in loss-making Japanese companies. I decided I wanted consistently profitable companies bought at a discount to their net cash.
I couldn’t find 20 of those. I could find six of them.
Yes, it would’ve been possible to buy 20 Japanese net-nets. But I felt safer buying six profitable Japanese net cash stocks.
It was easier for me to wrap my head around the idea of intrinsic value for those six stocks. I knew the companies were usually profitable. I knew they were selling for less than their net cash. Therefore, I knew the market was valuing these profitable operating businesses at some negative number. I figure a profitable business is worth more than zero. Therefore, I felt comfortable buying them.
This would be more complicated if I was looking at the value of receivables, inventory, etc. since these are “invested” assets. If you add in the idea of net-nets with recent losses – we are now talking about trying to figure out the normal earning power of the actual operations of some Japanese company.
That’s a lot tougher for me to do.
Some people – and they’re not wrong to do this – prefer buying 20 net-nets with mixed profitability to buying six net cash stocks with consistent profitability. I don’t believe the six net cash stocks I bought will necessarily outperform a basket of Japanese net-nets. In fact, I would lean towards the belief that as a group all Japanese net-nets will outperform the net cash stocks I bought.
There is more upside potential in the net-nets. If their businesses recover, they will one day trade at much higher stock prices.
The problem with that approach is the person implementing it.
I know I can hold six profitable Japanese companies selling for less than their net cash – no matter what happens in Japan. Remember, I was buying these shortly net-nets after a terrible natural disaster and subsequent nuclear plant failure. As it turned out, my ability to stomach adverse developments in Japan hasn’t been tested. Things have generally gotten better since I bought my first Japanese net-net.
But I didn’t know that when I bought these net-nets. I didn’t know that the headlines out of Japan weren’t going to keep getting worse. I didn’t know how serious – and lasting – the disaster’s impact on the economy would be. I didn’t know how bad sentiment would get on Japan.
I’m half a world away from Japan. Whatever gets reported in the worldwide financial press is basically all I’m going to hear about Japan. So, if people got very, very fearful about Japanese stocks – that might rub off on me.
With a handful of profitable net cash companies, I could withstand that wave of fear quite easily. Nothing would shake me from holding those companies at those prices.
With a basket of 20 net-nets – I’d like to think I would have the willpower to hold those stocks no matter how bad the headlines out of Japan got. But I couldn’t be sure.
Usually, I pick stocks I know I am most likely to hold. The two top issues for me in stock selection are whether the stock will give me an adequate return if I buy it today and whether I will stick with the stock long enough to let that happen.
I’ve diversified very little at times – and a whole lot more at other times. I’ve never owned more than 20 stocks or anything like that. But I have owned 3 stocks and I have owned 15 stocks.
It has never been my experience that I felt more convinced of my ability to stick with 15 stocks than my ability to stick with 3 stocks. In fact, the more stocks I have in my portfolio, the less sure I am that I will be able to stay calm and rational when nobody else is.
So that’s my reason for avoiding diversification. I feel more comfortable the less I diversify. And I invest best when I’m most comfortable. Plenty of other people feel less comfortable when they are less diversified. If you’re one of them – you probably should own a lot of stocks.
Under no circumstances should you ever adopt an investing plan that is going to make you uncomfortable enough to sell when you shouldn’t.
It’s hard enough being right about stocks. Too many investors make it harder by selling things they were right about before Mr. Market has conceded the point.
So my advice is not to do what I do. My advice is to do what makes you comfortable. For me, that’s holding very few stocks. For you, it might be holding a lot of stocks.
Your top priority should be your comfort. Your ability to hold good stocks through bad times.
Read Geoff’s Other Articles
Ask Geoff a Question
Check out the Buffett/Munger: Bargain Newsletter
Check out the Ben Graham: Net-Net Newsletter
From what I can infer from your posts, your own style of investing has parallels of both Graham and Buffett. How come you opt to be less diversified than Graham when having a portfolio concentrated in net-nets? What is your reasoning for straying from his approach? You say that Buffett bought 20 companies when he was investing in less familiar foreign small companies, so he was more diversified than his usual focused approach. How come you don't when investing in net-nets as well?
Thanks,
Tom
I understand why you say my own investing style has parallels to both Buffett and Graham – everybody says that – but it’s really not true. The first explicit rules of value investing I ever learned were from reading Ben Graham. But I never invested like Ben Graham. I always invested like Warren Buffett. I wrote an article entitled “How Long Does it Take to Develop an Investing Style?”
Read that article. It’s the best description I’ve given of how Buffett actually invested when he was starting out. Warren Buffett was never like Ben Graham in terms of his overall approach to investing. He never ran his portfolio the way Ben Graham would. Yes, they both bought some of the same stocks in the 1950s. But this similarity in terms of stock selection did not carry over into portfolio construction.
I concentrate on my top ideas. Buffett always did that. He put 75% of his net worth into GEICO in the 1950s. And he put 40% of the partnership’s money into American Express (AXP). Graham never concentrated on his top ideas – or, almost never. He actually did concentrate his investments in companies like Northern Pipeline. These usually had some sort of activist slant. I’m not sure activist is the right word. It means something a little different today. But Graham did buy big (for him) positions when he wanted a company to pay out cash, or he could have control of the board, or something like that. Still, these were rarely large positions as I define them. They were just very large relative to Graham’s median position size.
I only invest in simple businesses. I’m extremely restricted in terms of the companies I’ll buy into. I wouldn’t buy Intel (INTC) at any price – except maybe below its net current assets, and certainly below its net cash. Intel is not in my circle of competence.
I mentioned that three of the first investment I ever made were: J&J Snack Foods (JJSF), Village Supermarket (VLGEA), and Activision (ATVI). A fourth – the stock I paid the highest price for in all my time investing – was Bio-Reference Labs (BRLI). This was about 10 years ago. They were not involved in some of the more esoteric tests they do today. It was a simpler business back then.
So, we’re talking about things like soft pretzels, groceries, video games, blood and urine tests. Simple stuff. Stuff with a “consumer” angle of some sort. Not capital goods. And not something where a corporate buyer has a pretty free hand in determining demand. Yes, the actual buyer in some of these situations – the “customer” – is a corporation. But that’s like saying that Hanes (HBI) serves Wal-Mart. To an extent it does. But the customer that matters is the mom who buys the socks and underwear and T-shirts for her family. And the people in the family who influence her shopping decisions. If Hanes’s products click with the folks who actually buy and wear the stuff, Wal-Mart will order it. If not, they won’t.
I’m more interested in looking at a situation like that than at some sort of system you put in a factory. I have a hard time knowing what the demand is for that product. And I have a very hard time knowing how much you can charge for the product without alienating customers.
Take some of the stocks I’ve mentioned recently. Three of the most notable recent mentions were DreamWorks (DWA, Financial), Carnival (CCL, Financial) and Royal Caribbean (RCL, Financial). I’m a high school drop-out. I’m not going to understand something technical. But I can understand the value of an experience somebody is selling. I can understand a business like DreamWorks, Carnival or Royal Caribbean.
I rarely spend time looking at companies where the product is bought by a corporation. There are a few exceptions to this:
· The company has a lot of untapped pricing power (I will even buy a capital good if you could raise prices 20% tomorrow).
· The corporate buyer has limited freedom of action because they must respond to the desires of their own customers.
Graham didn’t care about these things. Buffett does.
Now, sometimes people say I must be a Ben Graham type investor because I buy things like net-nets. But as I explained in my articled entitled “How Warren Buffett Made His First $100,000” – Warren Buffett bought net-nets too.
A cheap price is always better than an expensive price. And a good business is always better than a bad business. Where possible, you try to buy a good business at a cheap price. Ben Graham sometimes paid lip service to this approach. But I don’t think he was really interested in the quality of a business when it came to his own investing.
Buffett was interested in business quality from the very beginning. I was interested in business quality from the very beginning. I know this is hard to believe.
Because we know Warren Buffett bought some bad businesses. Yes, he bought bad businesses, when they were offered at really, really good prices. Like I said, Buffett has always been a return on investment investor. He’s always looked at how fast his capital will compound.
If you are offered a business for a 50% discount to its cash and securities net of total liabilities – and that business is usually profitable – you’re going to tend to compound your money faster in that business than in Coca-Cola (KO). It doesn’t matter that the business earns low returns. There’s a decent chance you could get a 150% gain in that stock in one year, or two years, or three years, or five years. As we move out in time – as long as the business is eking out some profitable return – the chance gets higher of you realizing that return. Even if it takes five years to see a 150% rise in the valuation of that business and cash combo you bought – you will still earn an annual return of 20% a year. Actually, you’ll probably earn a few percentage points more a year since book value will keep rising for a profitable business.
You have to sort through thousands of opportunities, but you will find some situations – especially in micro cap stocks – where the business really would be sold for 150% more than the current stock price if it was offered up to a control buyer today.
Usually, these businesses are controlled by families. Yes, this is similar to Ben Graham’s approach. But Graham diversified widely. I don’t.
Why not?
Because I think Ben Graham was wrong to diversify. Anyone who looks through Graham’s portfolio and his track record and the track record of his students – would have to at least entertain that possibility.
Graham continued to do some unnecessarily complex stuff. For example, he kept buying some bonds, preferred stocks, etc. There was no reason for this. From the very early days, it was obvious that Graham-Newman could make more money – and make it consistently – in common stocks.
Graham-Newman had a handful of successful techniques:
· Control positions
· Net-nets
· Arbitrage
· Liquidations
· Related hedges
In theory, all of these opportunities are open to today’s investor. Some of them still work. Even related hedges still exist. They’re rare. But if you pay very careful attention to convertibles, you’ll find sometimes when it’s clear either the common stock or the convertible is mispriced. Convertibles aren’t something individual investors spend time on. And institutional investors need ideas where they can put a lot of money to work.
As a result, even in 2012, you can work through a list of every known convertible and find a few smaller ones where shorting the common stock and owning the convertible looks like a free lunch.
I’m not sure it is. In fact, your annual returns may not be all that impressive relative to what you could do simply by picking good businesses at decent prices. But it is a very high probability bet. And you won’t live and die by which way the market is moving every week, month or quarter.
I wouldn’t suggest hedging as a technique individual investors should use, but within the last couple years there have been occasions where you could guarantee yourself at worst a small loss and at best (if the market tanked) a very decent gain doing exactly what Ben Graham did. Literally, you wouldn’t have to touch options or anything – just go long the convertible and short the common stock the same way Graham did almost 90 years ago.
I don’t do related hedges.
And I don’t have enough money to do control positions.
I’ve invested in net-nets, arbitrage, and liquidations. They’ve all worked well. They haven’t all been fun. And some of these techniques are inappropriate for people with certain personalities. I’ll lump “workouts” together. When I say “workouts” I mean arbitrage, liquidation, and litigation – mostly. And when I say liquidation I mean a situation where it is obvious some of the assets of the business will be returned to shareholders. In some cases, the exact mechanism used is not a total liquidation of the business with all capital returned as cash.
While some people associate this stuff with Ben Graham exclusively – Warren Buffett did all of these things. Within the last 10 years, he owned Comdisco (a liquidation) and Dow Jones (a merger). In the 1980s, he owned Arcata. Read the 1988 letter to shareholders for details.
Arcata is a good example of the kind of situation that will always pop up from time to time in this area. Basically, it’s a situation that involves uncertainty in a rather low-risk way. And people may confuse high uncertainty with high risk. You can make a lot of money over time selling people protection against uncertainty.
Most investors have this totally bizarre need to not just know that they are going to make money – they need to know exactly how and when they’ll become filthy rich.
Anyway, if you go back and read the newspapers from around the time of the Arcata deal, it was more of a slam dunk than Buffett portrays. He says only, “Finally, we had to ask ourselves what the redwood claim might be worth. Your Chairman, who can’t tell an elm from an oak, had no trouble with that one; he coolly evaluated the claim at somewhere between zero and a whole lot.”
The evidence at the time was strongly towards “a whole lot.” And Buffett knew this. This was not the first time a redwood claim had been litigated. And there had been substantial inflation since certain earlier headline making redwood deals had occurred. While it’s true there was a lot of uncertainty about the value of the claim – and how interest would be determined – there was not as much risk in Buffett’s Arcata position as you’d expect in an investment where you could make as much money as Berkshire did.
The reason why people shy away from a good special situation like this one is that it involves uncertainty in both time and value. But it’s not hard to sketch out a couple rows of possible annual(ized) returns under different payout dates and values.
Graham taught Buffett how to think about workouts. But I’m not sure Buffett needed the education. These are very simple calculations. What you need to be able to make money in these kinds of situations is a belief in Graham’s Mr. Market metaphor. If you really believe the market can be wrong when it comes to something as simple as a liquidation, then you can buy into these situations. I should point out – as Buffett did in 1988 – that these are not typical returns. Ben Graham and Warren Buffett both made money in arbitrage by passing on the vast majority of possible deals. They had much higher standards in terms of a hurdle rate than funds that focus on arbitrage. Because it was just one arrow in their quiver, Buffett and Graham only resorted to arbitrage when it offered very attractive annual returns. As a result, the arbitrage they did tended to improve returns rather than just smooth out returns.
Graham was willing to buy into deals like that. Buffett is willing to buy into deals like that. And I’m willing to buy into deals like that.
If I do the math on some tiny liquidation and decide it is almost certain to return 15% a year and most likely will return 30% a year – I will buy into that liquidation. As much as I like owning a good business – as much as it is much more fun to research an actual company – you don’t pass up a very high probability of 15% or better returns when you see it. Especially if you know the new opportunity is not connected to things you are already invested in.
But a willingness to buy a stock that is liquidating – and I don’t do this often (I’ve averaged a little under one liquidation a year) – isn’t really something that makes you more Buffett like or more Graham like. It’s something both Warren Buffett and Ben Graham did throughout their careers.
Where I differ with Ben Graham is in how much attention I pay to the business, its products, etc.
And how much I concentrate on a few positions.
And how much I focus on understanding a business.
Ben Graham was willing to buy any business anywhere as long as it was statistically cheap. I’m not willing to do that. You know I owned Barnes & Noble (BKS, Financial), which was a situation I thought I understood (I was wrong). It’s one of the very rare examples of a retailer appearing in my portfolio. I am not a good judge of retailers. And I know this.
I would never buy a bank or retailer or insurer because it was statistically cheap. That doesn’t mean I would never buy a bank or retailer or insurer at all – at different times a retailer and insurer have been the biggest position I owned. But I knew a lot more about them than just their price-to-earnings, price-to-book, debt-to-equity, etc.
Really, the overwhelming difference between my approach and Ben Graham’s approach is focus. I like to focus on my best ideas. I like to focus on trying to understand the businesses I invest it. Ben Graham liked to spread out his portfolio. He took a broad and shallow approach. I take a narrow and deep approach.
Which is better?
I think you know my answer. Ben Graham did not achieve the kinds of annual returns he was capable of. He did not focus exclusively on common stocks. He passed up some opportunities to take very big positions in select stocks.
Which is odd. Because he had some success in stocks like Northern Pipeline that you would think would teach him to focus on his best common stock ideas – to focus on places where he could add value (like activism and reallocation of capital).
He didn’t. Graham wanted a system. In fact, he really wanted a system that the average individual investor could apply at home. Activism was not part of that system.
During his actual investing career, the approach that seemed best for individual investors was simply buying net-nets. But by the time Graham died, net-nets were much less common than they had once been. So at the end of his life, Graham tried to develop other systems. He wanted simple, formulaic ways to buy value stocks – beyond just buying the increasingly rare net-nets.
Is a formulaic approach – investing by pure statistics in dozens of stocks at the same time – a viable value strategy?
Yes. In fact, it’s very simple to boil down the essence of Graham’s approach into a strategy any investor could profitably follow for years:
1. Dividend Yield > Median Dividend Yield of all stocks
2. P/E< Median P/E of all stocks
3. P/B (Tangible)< Median P/B (Tangible) of all stocks
4. Tangible Equity/Total Liabilities > Median Tangible Equity/Total Liabilities of all stocks
Rank by tangible equity/total liabilities. Buy the top 20 stocks. Swap out stocks to the extent required – but never more frequently than once per year. By the way, you would swap out fewer than 15 stocks on average, because sorting by tangible equity divided by total liabilities tends to keep causing the same stocks to appear in much the same order.
The approach works. It generates a list of Ben Graham type companies – not net-nets, but companies he’d suggest for the “defensive investor” – and it outperforms the market most of the time.
It even outperforms when applied to huge stocks.
In most years, there would be at least 20 such stocks in the S&P 500, and just buying these 20 stocks instead of spreading your money across all 500 stocks would beat the market in most years and especially decline less in bad years.
That idea would hold a lot of appeal for Ben Graham. You could design a fund around it. In fact, you could apply it to just the biggest companies if you wanted to. And the sorting mechanism I mentioned is not important (tangible equity to liabilities above the average for all stocks is all you need to prove a stock is safe enough to be part of a value group – beyond that it’s just superfluous safety).
Now, you might say this proves the power of investing in 20 different stocks, inside of selectively picking a few.
I think this approach is fine. But even here the diversification is unnecessary. If you take the same group which is again stocks with:
1. Dividend Yield > Median Dividend Yield of all stocks
2. P/E< Median P/E of all stocks
3. P/B (Tangible)< Median P/B (Tangible) of all stocks
4. Tangible Equity/Total Liabilities > Median Tangible Equity/Total Liabilities of all stocks
And you say – as I would – there’s no point in picking 20 of these things. Anything that passes this screen is cheap. At most, the issue is about long-term earnings stability. So, why not cut the list of stocks from 20 to 5 and just sort by total dividends paid. This will give you the biggest companies that pay out the most in dividends and retain the least earnings. My expectation is that when those companies are cheaper than the median stocks in terms of yield, price-to-tangible-book and P/E and they have more tangible equity, Mr. Market is wrong. And he’ll be consistently wrong whenever he prices stocks like that at such low levels.
It turns out I’m right. The top 5 dividend payers (in terms of total dollars paid) that pass a median yield, median P/E, median tangible equity total liabilities test very rarely go down in price. I’m not exaggerating. They tend – not just as a group, but each stock individually – to have a very high chance of being at least flat for the next year. Their longer term record isn’t too bad either.
By the way, this is what the list would look like today:
1. ConocoPhillips (COP, Financial)
2. Duke Energy (DUK, Financial)
3. NextEra Energy (NEE, Financial)
4. American Electric Power (AEP, Financial)
5. Public Service Enterprise Group (PEG, Financial)
No one reading this article is going to go home and buy that group. They’re all energy companies. One is an oil company. The other four are utilities.
You wouldn’t be diversified. What if utilities underperform?
It would be crazy to buy this group, right?
I don’t think so. And that is where I differ with Ben Graham. And with all modern conventional wisdom on diversification.
I look at that group and I say: “What’s the chance these stocks will be worth less in the future?”
Even though there are only five of them, the risk of any one stock being worth much less in the future is quite low.
Their yields range from 3.4% to 4.9%. So, we’re talking about a dividend yield equal to or greater than the 30-Year U.S. Treasury bond. Okay, what about price-to-tangible book?
The range is 1.3 to 1.7. Let’s think of that in terms of a long-term rate of return. If I pay 1.3 to 1.7 times tangible book for a collection of U.S. utilities will I do at least as well as the S&P 500?
I might not this year. But, even if these stocks earn just 10% on tangible equity, I will be buying them at normalized earnings yield of 5.9% to 7.7%. And they will pay most of that yield out to me in cash.
Now, compare that to reasonable return expectations in the S&P 500. Pick your favorite predictor of future stock market success – the Shiller P/E, Buffett’s market value to GDP ratio, etc.
I’m much less convinced than most people that you are actually taking more risk by buying these five stocks:
1. ConocoPhillips (COP)
2. Duke Energy (DUK)
3. NextEra Energy (NEE)
4. American Electric Power (AEP)
5. Public Service Enterprise Group (PEG)
I think it’s entirely possible that if you repeated this kind of pattern of buying over time you would not take more risk than buying the S&P 500. You might take less.
Yes, it’s buying just five stocks. But we know they were selected on the basis of:
· Adequate price (Low P/E, Low P/B, High Yield)
· Adequate safety (Low liabilities to tangible equity)
· Adequate stability (Big size, high dividends)
But, of course, it’s a completely undiversified group.
I don’t care. I don’t care that it’s basically just stocks from one sector. If they were selected with safety and cheapness in mind they may be more reliable investments than 500 stocks that were selected without any concern for their safety or cheapness.
Now, this is skirting the issue. I’m a value investor. Of course I prefer a list of value investments – however short – to the entire S&P 500.
Why do I prefer a portfolio of 4 to 12 value investments to a portfolio of 20 to 50 value investments?
Because I’m not sure how much safety is added by increasing the number of stocks you buy.
It’s a complicated issue. Where the greatest risk comes from the risk of catastrophic failure – for example, if you are buying the cheapest banks without regard to their Texas ratio, etc. – I would be all for buying a very big group of stocks because there’s a good chance one of the four or five or eight stocks you pick could prove a total loss even if the group does well.
But I think wide diversification is usually like using the coarse adjustment knob on that microscope you used in high school science. People think diversification reduces risk. Instead, they should be thinking about what risk they need to reduce.
If your answer to every risk was diversification, you would design the Ben Graham screen I spoke of before with just three criteria:
· Low P/E
· Low P/B (Tangible)
· High Yield
Or even just one criteria (low P/E). And then you’d try to fix the problem of all the lousy stocks slipping into the portfolio by diversification.
But, if we take out the tangible equity to total liabilities part of the screen – the results are okay, but they no longer have much to do with Ben Graham. Basically, low price can outperform regardless of safety – but this is only because big winners make up for big losers.
It takes two seconds to stop and think about the problem: “Low P/E stocks might be on the verge of bankruptcy” and then solve it with one of the world’s simplest stock ratios “tangible equity/total liabilities” to create a kind of loss absorption measure for your portfolio.
This is why I question the usefulness of diversification. It makes you think you can’t separate the winners from the losers. And you don’t really need to anyway.
Actually, sometimes it’s quite easy to notice that one stock is in super shaky financial condition and another one is in rock solid shape. I mentioned this same ratio – tangible equity to total liabilities – when talking about net-nets. I also mentioned the ratio of NCAV to total liabilities. This is a very common sense ratio that people who diversify widely in net-nets sometimes forget to apply.
The ideal net-net is safe and cheap. Just cheap may work as a group. But I’d rather try investing in a moderate number of safe and cheap net-nets rather than an obscene number of just cheap net-nets.
Mostly this has to do with risk. It doesn’t matter if you back test a purely cheap net-net portfolio for half a century and prove it works all the time, every time. It’s not a strategy I can get comfortable with.
It is very easy for me to get comfortable holding five stocks I picked myself – and which I believe to be both safe and cheap.
Even in Japan?
Your question about why I only own 6 Japanese net-nets rather than the 20 or so stocks Warren Buffett bought in Korea is a good one.
Basically, I was not comfortable paying for Japanese receivables and inventory (especially inventory). And I was not comfortable investing in loss-making Japanese companies. I decided I wanted consistently profitable companies bought at a discount to their net cash.
I couldn’t find 20 of those. I could find six of them.
Yes, it would’ve been possible to buy 20 Japanese net-nets. But I felt safer buying six profitable Japanese net cash stocks.
It was easier for me to wrap my head around the idea of intrinsic value for those six stocks. I knew the companies were usually profitable. I knew they were selling for less than their net cash. Therefore, I knew the market was valuing these profitable operating businesses at some negative number. I figure a profitable business is worth more than zero. Therefore, I felt comfortable buying them.
This would be more complicated if I was looking at the value of receivables, inventory, etc. since these are “invested” assets. If you add in the idea of net-nets with recent losses – we are now talking about trying to figure out the normal earning power of the actual operations of some Japanese company.
That’s a lot tougher for me to do.
Some people – and they’re not wrong to do this – prefer buying 20 net-nets with mixed profitability to buying six net cash stocks with consistent profitability. I don’t believe the six net cash stocks I bought will necessarily outperform a basket of Japanese net-nets. In fact, I would lean towards the belief that as a group all Japanese net-nets will outperform the net cash stocks I bought.
There is more upside potential in the net-nets. If their businesses recover, they will one day trade at much higher stock prices.
The problem with that approach is the person implementing it.
I know I can hold six profitable Japanese companies selling for less than their net cash – no matter what happens in Japan. Remember, I was buying these shortly net-nets after a terrible natural disaster and subsequent nuclear plant failure. As it turned out, my ability to stomach adverse developments in Japan hasn’t been tested. Things have generally gotten better since I bought my first Japanese net-net.
But I didn’t know that when I bought these net-nets. I didn’t know that the headlines out of Japan weren’t going to keep getting worse. I didn’t know how serious – and lasting – the disaster’s impact on the economy would be. I didn’t know how bad sentiment would get on Japan.
I’m half a world away from Japan. Whatever gets reported in the worldwide financial press is basically all I’m going to hear about Japan. So, if people got very, very fearful about Japanese stocks – that might rub off on me.
With a handful of profitable net cash companies, I could withstand that wave of fear quite easily. Nothing would shake me from holding those companies at those prices.
With a basket of 20 net-nets – I’d like to think I would have the willpower to hold those stocks no matter how bad the headlines out of Japan got. But I couldn’t be sure.
Usually, I pick stocks I know I am most likely to hold. The two top issues for me in stock selection are whether the stock will give me an adequate return if I buy it today and whether I will stick with the stock long enough to let that happen.
I’ve diversified very little at times – and a whole lot more at other times. I’ve never owned more than 20 stocks or anything like that. But I have owned 3 stocks and I have owned 15 stocks.
It has never been my experience that I felt more convinced of my ability to stick with 15 stocks than my ability to stick with 3 stocks. In fact, the more stocks I have in my portfolio, the less sure I am that I will be able to stay calm and rational when nobody else is.
So that’s my reason for avoiding diversification. I feel more comfortable the less I diversify. And I invest best when I’m most comfortable. Plenty of other people feel less comfortable when they are less diversified. If you’re one of them – you probably should own a lot of stocks.
Under no circumstances should you ever adopt an investing plan that is going to make you uncomfortable enough to sell when you shouldn’t.
It’s hard enough being right about stocks. Too many investors make it harder by selling things they were right about before Mr. Market has conceded the point.
So my advice is not to do what I do. My advice is to do what makes you comfortable. For me, that’s holding very few stocks. For you, it might be holding a lot of stocks.
Your top priority should be your comfort. Your ability to hold good stocks through bad times.
Read Geoff’s Other Articles
Ask Geoff a Question
Check out the Buffett/Munger: Bargain Newsletter
Check out the Ben Graham: Net-Net Newsletter
