Geoff Gannon May 29, 2017

Over the Last 17 Years: Have My Sell Decisions Really Added Anything?

I get a lot of emails from people saying that my strategy has changed – I’ve become more of a growth investor and less of a value investor – over time.


It’s true that the investments I’ve made in recent years have definitely changed.


But, my philosophy has changed less than it would appear from my stock picks. I concentrate heavily and go where I see opportunities I consider “nearly certain” rather than being the highest return opportunities based on pure probabilities.


There is, however, one area where my philosophy really has changed:


I’m convinced that I should simply hold stocks indefinitely.




Let’s start with two spin-offs I bought. One spin-off happened 2 years ago. The other spin-off happened a little over 10 years ago.


First, the 2-year-old spin-off. I have 25% of my portfolio in BWX Technologies (BWXT). I bought that as part of a spin-off from Babcock & Wilcox. The stock has returned more than 30% a year in the two years since the spin-off. It now trades at a P/E of 26. Normally, this is when a value investor would sell the stock. However, I think the company can grow earnings per share by 10% a year for the next several years. I also think a company of this quality should always trade at a P/E no less than 25. So, with no new ideas that seem more likely to deliver returns of 10% a year or better – I have no intention of selling BWXT. With the catalyst from the spin-off gone and the P/E above 25 – no value investor would keep holding this stock. But I intend to. Does that mean I’m not a value investor?


It might mean that. But, it also may just mean I learned from the last spin-off I liked a lot.


About ten years ago, I picked a spin-off called Hanesbrands (HBI). Here’s a quote from a roundtable discussion I did back in 2006 (share prices are not adjusted for subsequent splits):


However, there are many situations (and here is usually where you find some bargains) where the EV/EBIT measure is not the most useful. When I can predict a high free cash flow margin with confidence, I use a very long-term discounted cash flows calculation. For instance, this is what I would do with Hanes Brands which was recently spun-off from Sara Lee. On an EV/EBIT basis, it may not look cheap. But, looking truly long-term, I’m convinced the intrinsic value of each share is much closer to the $45 – $65 range than the roughly $23 a share at which it now trades. But, that’s a special case – Hanes is a special business.”


I gave that quote back in October of 2006. Hanesbrands stock has compounded at 12% a year in the 10 years since I made that comment (it’s compounded at 15% a year since the actual spin-off date).


I probably should have just held on to that stock. When we include the crisis years of 2007-2008 in my results, it’s hard to see how selling out of Hanesbrands and buying other things – while paying taxes, etc. – has materially benefited the long-term compounding power of my investment account. That’s typical of my good stock picks, by the way – selling them to buy something else has never really been necessary to get an adequate long-term return in my overall account.


Investing is all about taking decisive action. Making a decision and then acting on that decision. So, the question is: Do my sell decisions add value?


Any decision I make that doesn’t add value is an area I either need to improve or stop worrying about altogether. In other words, if selling hasn’t helped my returns – I either need to learn to become a better stock seller or I need to ignore selling all together.


Have my sell decisions added to my returns?


Let’s start our investigation with three stocks I liked best around the time I started investing as a teenager. They were:


  1. Village Supermarket (VLGEA)
  2. J&J Snack Foods (JJSF)
  3. Activision (ATVI)


I started investing at age 14. I read the Intelligent Investor at age 16. I originally bought these stocks after I started investing but before I started trying to implement more of a Ben Graham type approach. These were stocks from my self-directed, philosophy-less days – before I called myself a value investor.


How have those stocks done?


Let’s use January 1st, 2000 till today as a convenient measurement day. I don’t own any of these stocks anymore. So, the idea here is – since we know I did buy them – let’s ask: what if I never sold them?


Measuring from January 1st, 2000, the holding period would be 16.5 years. Over the last 16.5 years, those 3 favorite stocks of teenage me have compounded annually at:


  1. Village Supermarket: 13%
  2. J&J Snack Foods: 17%
  3. Activision: 26%


Those long-term results are good. Because of the way compounding works, a three stock portfolio consisting only of those stocks would have actually outperformed what I did myself by buying and selling dozens of different stocks over the last 16.5 years. I could have saved myself a lot of trouble and gotten a better result just by holding on to everything I owned when I was 15 years old.


Activision skews those results. That could be pure luck. So, we can either take the median result (J&J Snack Foods’s 17% result) or even an average of the other two stocks (15% is the average of Village and J&J over the last 16.5 years) or we could even just take the bottom result (13%). Whichever way we slice it, I’ve added anywhere from negative value, to fairly neutral value, to an almost immaterial amount of value to my compound result by buying and selling dozens of stocks over the years instead of holding what I bought around 15 years ago.


It certainly seems that I was capable – as a teenager – of picking a stock that would return 13% a year over the next 15 years. When you consider that a 13% annual result that comes mostly in the form of unrealized gains isn’t taxed till you sell it – it becomes very unclear if all my buying and selling over the years has added any value beyond the result I could have gotten by simply holding the worst of those 3 stocks for the last 15 years.


Some of you may know this number better, but I’d guess the overall stock market has returned more than 3% but less than 5% a year over the last 16.5 years. Let’s be generous and call it 5% (which from January 1st, 2000 as your start date – I’m sure it isn’t any higher than). In all 3 of these stocks, we are talking about compounding at comfortably more than 10% a year when the market was doing worse than it normally does. The market was doing 5% a year or worse. These stocks were all doing 10% a year or better. If you can compound at better than 10% a year forever, you can achieve all your long-term financial goals.


Here’s an illustration.


An initial investment of $10,000 that compounds at 5% a year for 15 years becomes $20,789.


An initial investment of $10,000 that compounds at 10% a year for 15 years becomes $41,772.


And an initial investment of $10,000 that compounds at 15% a year for 15 years becomes $81,370.


The key is obviously avoiding the first result. You never want a 15-year period where you compound at only 5% a year. Roughly speaking, in 15-year chunks: a 5% annual result gives you a double, a 10% annual result gives you a quadruple, and a 15% annual result gives you an octuple.


So, a good goal is to find quadruples and octuples. Stocks that can be held for 15 years while returning 10% to 15% a year.


That, at least, is what my own experience has taught me.


Obviously, I’ve owned stocks that have had much higher annualized returns than that. I bought Bancinsurance when there was an offer to take the company private at $6 a share. Here’s what I wrote when I sold that position:


“My cost was $5.82 a share. Bancinsurance’s board agreed to an $8.50 a share buyout. I sold my shares between $8.00 and $8.20…That’s a better than 38% return in less than 7 months. If I’d held Bancinsurance through the buyout I would’ve done better with a 46% return in less than 10 months.”


Here, it’s worth noting that I sold my Bancinsurance shares which would have go on to return another 5% raw which is more than 20% annualized by the time the deal closed. I used some of the proceeds to buy Barnes & Noble (BKS), which was completely flat for me as a stock. No gain. So, even here, I sold too soon. I should’ve just held Bancinsurance till the deal closed.


So, can I sometimes find opportunities that return 40% or more in a single year? Sure. It’s happened. In 2009, my entire account did better than 40% a year. But, that was 2009. Stocks started that year in a crisis. In the eight years since, there’s never been a market wide opportunity like that.


There are better investors than me – people like Warren Buffett and Charlie Munger when they were running their partnerships – who were able to routinely find opportunities to fill an entire portfolio with things that could compound at 20% to 30% a year if you bought them, sold them, and then rotated the proceeds into something else over and over again. They worked hard and found a steady flow of Bancinsurance type opportunities.


I have not found enough such opportunities. What I have been able to find – a lot of actually – is stocks that can do between 10% and 15% a year over the next 10 to 15 years. That doesn’t sound very impressive. But, when you chart those stocks against the S&P 500, they come out way ahead.


Let’s go back to the Hanesbrands example. From the moment Hanesbrands was spun off till today – so 10.5 years – it has returned 15% a year. That meets our goal of a stock that can return 10% to 15% a year over 10 to 15 years. Meanwhile, the market has compounded at closer to 6% a year.


Now, you can make mistakes and get a worse cumulative result than this even over a very long period of time. For example, in the same year I picked Hanesbrands, I also picked Posco (PKX). Posco was actually much more of the value stock. It was even – in some ways – perhaps more of a “moat” stock in an academic analysis. But, clearly not the better, safer company. If I had to put all my life savings into just Posco or just Hanesbrands in 2006 – it’d take half a second to decide on Hanes. Steel is a bad business. Korea is right next to China (which overbuilds things like steel plants). And I don’t usually want a Korean management team making capital allocation decisions for me. So, Posco was never the kind of company you’d want to own forever. Hanes could be. Over the last 10.5 years, Posco has returned exactly nothing. The market has done 6% or so.


So, if you put equal amounts of money into both Hanes and Posco on the same day and held them till today – your result would be closer to the S&P 500 than if you just bought Hanes alone. Because the result in each stock is compounded (and therefore Hanes has an outsized effect on the portfolio in later years), I think you’d have gotten a 9% to 10% annual result over the 10.5 years. Let’s call it 9%. So, picking one good compounder like Hanes and one completely flat stock – like Posco – and holding them both for a full decade could get you down to 9% versus 6% a year for the S&P 500.


Again, we see a basic problem that I have noticed with my investment results. Simply holding a stock I picked a long, long time ago has been surprisingly competitive with all my hard work spent buying and selling dozens of stocks in the ensuing years. I’ve done better than 9% a year over the last 10.5 years. But, I haven’t done better than 15% a year over the last 10.5 years. So, I would have had to combine a good pick with a bad pick in equal proportions for a buy and hold strategy to underperform all the buying and selling I’ve done over the last decade. Also, even combining one of my bad ideas with one of my good ideas from 2006 and holding them through today would have still outperformed the market.


So, why am I working so hard? Why am I worrying about selling at all?


Maybe the real duds are the problem. Maybe knowing when to sell a loser is key. What about the duds? I concentrate. So, I can lose a lot in one stock. I put 25% of my account into Weight Watchers (WTW) at one point. That stock was a disaster. Here’s what I said when I reported my sale:


My Weight Watchers position was eliminated at an average sale price of $19.40 a share…My Weight Watchers position had an average cost of $37.68 a share. So, I realized a loss of 49% on Weight Watchers.”

So, I sold Weight Watchers at a price of $19.40 a share. What has happened with Weight Watchers stock since then?

WTW now trades at $26.90. So, I missed out on 39% upside by selling.

I realized a loss of 49% on my Weight Watchers investment. If I had held on to that stock through today, my unrealized loss would now be only 29%. No one likes a 30% paper loss. But, a 50% paper loss is worse. As usual, I did worse by selling.

More importantly, I know the experience of some blog readers and newsletter readers who followed me into WTW was even worse than my own. Weight Watchers fell from over $38 a share when I bought it to a low of around $4 a share. Someone emailed me recently asking me at what price I thought various people who talked to me about WTW finally threw in the towel and sold their shares. I looked back through my emails, and best guess – this is a pure guess based on those who said they were selling in their emails – is that they exited the position around $8 to maybe $12 (at the very most). The anecdotal evidence (it’s a small sample size) says $8. But, let’s say it was $12 a share. Today, the stock is more than twice that level.

If look back at the stock’s chart, and I pick a time when WTW was selling at around $12 a share as the hypothetical sale date for many readers who followed me into the stock – I get a comparison between WTW and the market which tells me WTW is up 130% since then versus the market being up 15% since then. Odds are, if you followed me into WTW and then you sold that stock – you ended up selling an asset that would go on to return 130% while you put the proceeds into something that did just 15%. You see the problem with selling out of even a “broken” investment idea. It was a mistake to buy Weight Watchers. But, it was also a mistake for me to sell at $19 a share and an even bigger mistake for most people to sell at $12, $10, $8, or even $4 a share – as some people really did.

Of course, Weight Watchers survived its crisis. Not all stocks do. Some go to zero. And Weight Watchers very well could have. For example, if Weight Watchers’s decline in membership had coincided with the financial crisis – I’m convinced the company would have entered bankruptcy. The company needed access to credit. Because of high debt levels, the corporation would have failed – wiping out all shareholders – even if the brand survived. The reason the corporation survived was that credit was loose when the company experienced its crisis.

But, this is a post about my own investment philosophy. What has it been? What should it be? I’m looking for process improvements I can implement. I’m not looking for process improvements to suggest to everyone else out there.

One thing that has hurt the compounding power of my account has been holding on to a “dead money” stock. This is a stock with little growth and no catalyst. I owned a stock like that for the past 7 years. I bought George Risk in 2010. I sold it earlier this year. Let’s pretend I never sold it.

Looking at the stock from the time I bought it – 7 years ago – to today, we can see the annual return has been 10.3% a year. That’s the return you would have gotten if you bought George Risk shares when I did and continued to hold them through today. The S&P 500 has done well since the summer of 2010. I think it’s done about 12.3% a year since the day I bought George Risk. So, buying and holding George Risk for too long would have cost you about 2 percentage points a year over the last 7 years versus owning an index fund.

What could I have done differently? George Risk once traded at a higher price than it does now. It hit an all-time high of roughly $9 a share several times while I owned the stock. Let’s pick the earliest of those times. It was August of 2013. I bought the stock almost exactly 3 years before that day. So, it would’ve been possible – if I timed my sale perfectly – to make 25% a year in George Risk over 3 years. Instead, I made 10% a year over 7 years. My rate of compounding was 15% lower because I didn’t sell then.

In my defense, you might ask: how could you time the sale perfectly?

Well, Oddball Stocks timed this post “Is It Time to Dump George Risk?” pretty much perfectly. That post was written in August of 2013:

I’m also going to continue thinking about selling the position.  I love this at $6, but at $9 not as much.”

That turned out to be right. Oddball Stocks also wrote this:

I feel that selling stocks is a weakness for most investors, if not all investors.  It’s easy to spot a bargain and buy it, but it’s hard to tell when the music is over and it’s time to sell.  In general I try to keep a few things in mind when I look to sell a stock.  The first is whether the company has a margin of safety at the current level.  For most investments this means an asset backed margin of safety.  I do own one moat stock, Mastercard, and the margin of safety for that stock is the brand.  If there is still a margin of safety at the current price, and nothing else has changed with the business I will continue to hold.”

So, maybe knowing when to sell was a weakness of mine with George Risk.

If so, how do I correct that weakness in the future? Do I sell stocks after a certain period of time? Do I sell them when they reach my estimate of fair value? Do I continue to hold them indefinitely?

What is a practical rule I can adopt?

The simplest pattern I can see is that it’s been very hard for me to do better than just buying and holding the stocks I liked most. I was most “certain” of George Risk when I picked it in 2010. I was most certain of Hanesbrands when I picked it in 2006. If I held George Risk for a full 7 years, I’d be sitting on a 10% compound annual return without paying much of anything in taxes and without creating more work for myself (by having to find another stock to replace it with). If I held Hanesbrands for a full 10 years, I’d be sitting on a compound 12% annual return without paying much in taxes or having to do any additional work. So, again, it seems I’m capable of finding stocks that can – if held indefinitely – return 10% to 15% a year. George Risk and Hanes would both – if held through today – have given me what I consider an “adequate” return of 10% a year. I’d prefer 15% a year. And it was possible to make 25% a year in George Risk if you adopted a Ben Graham type mindset and bought around $4.50 when I did in 2010 and sold around $9 in 2013 (when Oddball Stocks raised the possibility it was time to sell George Risk). But, you are going to do better than most investors if you keep finding stocks that return 10% a year regardless of the market environment and just hold on to them.

It may seem like I’ve cherry picked specific investment ideas of mine that have continued to have especially high compound returns after I sold them. I assure you I haven’t. In fact, I’ve actually excluded some outliers. I bought Fair Isaac (FICO) in 2010. Here’s what that stock has done since then. To downplay the timing of my purchase – the market was still cheap in 2010 – I’m going to assume I bought at the highest price FICO ever traded at in 2010. The high for that year was a little under $26 a share. Today, the stock trades for $133 a share. That’s a compound annual return of 26% over 7 years. Obviously, I should have just kept holding FICO. The stock now trades at a very un-value-investor like 33 times P/E. But, almost all the good stock picks I’ve made over the year have followed a pattern of being priced at below average to average P/Es versus the market when I buy them and then being re-evaluated by the investment community such that they are then consistently priced at a premium to the market in later years.

I think the best case for why I shouldn’t make sell decisions at all is Omnicom (OMC). I bought that stock in 2009 at around $28 a share. The stock has returned 12% a year since then. The S&P 500 has also returned 12% a year since then. This would seem to suggest there was no harm in me selling Omnicom when I did – at a time when it had outperformed the market while I held it. And that might be true, but I actually like Omnicom at today’s price. So, I bought a stock at $28 a share in 2009 that I am now considering – in 2017 – buying again at $83 a share. I wrote a whole newsletter issue about it. It’s on my personal watchlist with just 3 other stocks right now. So, I obviously still really, really like the stock. And yet, it’s 8 years later and the price is 3 times higher. If you pick a stock today that you’re still going to like at triple the price 8 years from now – then why ever sell that kind of stock?

I think there are strategies that can benefit from including good sell decisions. And I think there are investors who pick the kinds of stocks that should be held for a few years and sold. Those are just the stocks they’re most comfortable owning. Over the years, I’ve learned that I’m not that kind of investor. The stocks I pick don’t benefit much from well-timed sales. There’s usually little harm in holding on to them much, much longer than I do.

So, I’ve decided to hold the stocks I own indefinitely. When I find a really, really good stock idea – which might happen once a year – I will need to sell pieces of the stocks I already own to raise cash for that purpose. I’ll do that. So, if I’m fully invested and want to put 20% of my money into a new stock – I’ll have to sell 20% of each stock I now own. But, I’m not going to eliminate my entire position in a stock anymore. Those decisions to completely exit a specific stock haven’t added value for me. So, I’m not going to try to make them anymore.

From now on, I’m going to be a collector of stocks.

The hard part will be implementing this – no full sale, ever – policy once I’m fully invested. Right now, I have about 35% of my portfolio in cash. So, I have no need to sell anything when I next find a stock to buy.


Geoff’s Verdict

From now on: Geoff will never voluntarily exit a position entirely. Once he owns a stock, he’ll keep owning at least some of that stock forever unless that company is taken over or goes bankrupt. He will simplify things down to a true “buy and hold” approach. No thought will be given to selling a stock ever again.


Your Verdict

Is Geoff’s resolution to never sell a position entirely too extreme? Have your own selling decisions boosted your account’s long-term compounding power? What can you do to become a better seller of stocks?

Comment below.

(Geoff and Andrew read and respond to these comments.)