How to Frame an Investment Problem

Author's Avatar
Jul 29, 2013
Article's Main Image
We can break the investment process down into a few steps. One, you get an idea. Two, you analyze the idea. Three, you choose whether or not to buy the stock. This makes it sound like the last step is the only one that requires choosing. But you are actually choosing at each step along the way. It’s rare to have only one potential stock idea – one candidate for research – on your radar.

Choice starts from the very beginning of the investment process. And it’s hard. What makes choosing hard?

As long as you keep your options open – as long as there doesn’t seem to be any risk of making a mistake – choosing seems pretty easy. If I asked you to sit down and list 10 stocks you might be interested in researching, you wouldn’t find that very taxing.

If I asked you to cross off one of those names, you’d find it a bit harder. If I asked you to cross off 9 of those names – you’d find it very hard to do.

In a sense, though, that’s what you’re doing even at the very start of your investment process. It takes me a good chunk of time to research a stock fully. Sometimes I’m focused on a single stock for a few weeks. That means ignoring other stocks – crossing them off my list – during that time. In that time, they might increase in price. Other ideas might appear. There are a lot of ways that a stock – once I decide to put off researching it – might end up never getting my attention.

So how should you choose which ideas to focus on?

Circle of competence is a good place to start. How familiar are you with this kind of company? The more familiar you are, the better a place this might be to start. Conspicuous cheapness is another. If a stock doesn’t look obviously cheap, you might be wasting your time. You’ll find it’s wonderful in many respects – almost all respects – other than price. Often that means you’ll learn a lot about the company, but you still won’t buy the stock. This is a dead end to avoid if you have stocks that are obviously cheap you’re also interested in.

The next question is a process question. It’s a matter of specialization. Do you know how to analyze net-nets? Have you done it before? Do you have an explicit checklist? An unwritten routine?

If you do, net-nets are a good place for you to start.

Process itself is a matter of choices. It starts with framing the problem. This is one of the biggest parts of the investment process. A lot of investors probably don’t give it much thought. But it determines how long your research will take and how clear a conclusion you’ll be able to come to.

Framing the problem means trying to define – often in simple, logical terms – the basis on which you will or won’t make an investment. Comparing a new stock to a stock you already own is a good way to frame the problem.

I’ve mentioned a stock I own – George Risk (RSKIA, Financial) – as an example of this framing before. When I first bought the stock, it was trading a bit below net cash. And I thought it was a fairly high quality company. So once the stock was in my portfolio the test that any new stock had to pass was simple:

Does this stock trade below net cash?

OR is this stock higher quality than George Risk?

If the answer to both questions seemed – at first glance – to be no, then that was probably a stock it wasn’t worth following up on. After all, I could just buy more of the stock I owned and knew better.

If the stock passed one of the two tests – for example, it traded below net cash but was lower quality than George Risk – then it might warrant a follow up. In that situation, the difference is easier to quantify. Something that’s easy to quantify is often easier to analyze immediately. So, if the stock was trading at a 10% discount to net cash while George Risk was trading at only a 5% discount, any deficiency in the quality of the company would probably eliminate it. A small quality difference can easily make up for such a tiny discrepancy in price.

Qualitative differences are harder to immediately analyze. So they’re usually not the first way you want to frame a problem. And when you do frame a problem using a qualitative distinction, it’s often best to use a relative distinction.

One of my favorite tests of whether a stock is worth researching is a combination of two questions – both of which must be answered in the affirmative:

Is it an above average business?

Is it selling for a below average price?

Here we have two questions that can be answered relative to something else. Is it an above average business? That’s often easy enough to answer in cases of clear quality. Is Copart (CPRT) an above average business?

I think so.

Is Carnival (CCL, Financial)?

That’s hard to say. Lately, the financial results suggest not. The competitive position – and performance before oil prices rose in the early 2000s – would say definitely yes. And so that’s how you’d have to elaborate on that analysis. It would quickly become a question of framing where you are looking at whether the future will be like the recent past, the distant past, or none of the above.

Personally, I find that frame a tough one to provide a clear conclusion. It involves issues like oil prices. Parts of the analysis are easy enough to answer. For example, the capacity of the industry – the supply of beds for passengers – is very easy to answer. There are only a few major competitors, and they all provide details about what ships they have contracted for several years out.

In most industries, you wouldn’t be able to predict supply so well. In the cruise industry, it’s very easy.

But, in most industries, you’d be able to predict input costs – here, we’re talking bunker fuel – a lot better than at Carnival. Carnival doesn’t control that cost. And – if you could predict oil prices – why not make money in oil futures, oil stocks, airlines, etc. rather than betting on Carnival’s stock price.

This doesn’t mean you can’t invest in Carnival. But, for me, it means that framing the problem so that I know Carnival is an above average business is too hard. As a result, when the stock trades much above book value – as it almost always does – it is probably too hard an analytical problem to solve.

You don’t need to do a full analysis over several weeks to realize that problem. It is clear when you choose the frame to see the problem through. If you know the price-to-book is well above 1, then you know you need to believe this is clearly a high quality company. Otherwise, there’s no margin of safety. This framing choice allows you to decide quickly whether the stock is worthy of a deep analysis. If it’s a company you can evaluate qualitatively or it’s trading well below book value, an analysis shouldn’t be hard.

We can see a similar price issue with DreamWorks (DWA, Financial). Right now, I think the company is way too hard to analyze as a stock. This isn’t because the business’s competitive position, etc. is especially difficult to evaluate. It’s because intrinsic value is very squishy here. You have to choose a number – a line in the sand – that you can hypothetically prove the stock is more valuable than.

It would be very hard to prove DreamWorks is clearly worth more than $24 a share. It would be easy to prove it is worth more than $12 a share. I don’t mean you need to do the analysis to prove this. You actually don’t – at first – need to look very hard at DreamWorks to figure out the actual intrinsic value. That’s not necessary. What is fairly easy to do is to look at the existing intangible assets – not what’s on the books, but what economic assets you think the company actually has in terms of franchise rights, etc. – and compare this to the stock price.

The stock might be worth more than it currently trades for. But, if it is, it’s because of the ability to make money in TV (or with Netflix) and in foreign countries that are growing fast.

It’s not hard to frame an analysis in a way where you can dig into exactly what DreamWorks is worth – and possibly prove it is worth well over $20 a share. But it will have to be a speculative – forward looking – analysis. It will have to depend on questions like:

· How much do I think Kung Fu Panda 3 (etc.) can make in China?

· Can DreamWorks exploit the Classic Media content it bought?

· Will the Chinese joint venture create a lot of value?

These are speculative questions. They are growth oriented. They aren’t value oriented. That doesn’t mean they are wrong. It just means that unless you are prepared to ask and answer speculative questions – you probably shouldn’t follow up on this stock at this price.

You can see that immediately if you think about how to frame your analysis. The way you frame an analysis is by asking:

· What questions will I have to get answers to – and what will those answers have to be – to prove I should buy this stock?

· And what questions could I ask – and what answers could I get to those questions – that would prove I should not buy the stock?

In other words, what would make this stock rocket to the top of your favorite investments list? And what would kill the idea?

This is something you can see in how you initially frame the problem.

That probably sounds wrong to you. It probably sounds like you should gather all the evidence – without any questions, theories, etc. – and then analyze the data.

That sounds scientific.

But it’s not. The best approach is first to design an experiment. And only then gather the data. The idea of evidence independent of a theory to hang it on is appealing – but it’s not practical.

For one thing, there’s simply too much data to gather. Simply by picking a dozen areas of interest about how some company works you’ll be closing off much of what you could be analyzing.

I don’t consider a dozen different factors when making an investment. I always narrow it down further than that. My point is that even if I did go as absurdly wide as trying to answer 12 questions at once – just deciding on those 12 questions is already applying a frame. It’s already deciding how you will see the problem. What you will investigate.

You might think you don’t design a frame for each investment problem you face. In that case, it’s very likely you are just using the same frame over and over again. You are applying one frame to all problems. You approach every investment like a net-net, like a wide moat company, like a growth company, etc.

That might be fine for you. In fact, I think that is fine – if you know you’re specializing. Some investors have a few frames they use by rote. A common combo is Ben Graham style value stocks and special situations. The two frames are different. But they tend to appeal to the same sorts of investors.

Most of the great investors you’ve read about:

· Warren Buffett

· Charlie Munger

· Ben Graham

· Phil Fisher

· Joel Greenblatt

· Peter Lynch

Had only a few frames. Some had basically one (Phil Fisher). Some had a couple (Ben Graham had net-nets and special situations).

For a good discussion of different frames, read Peter Lynch’s books. He breaks down his investments into a sort of playbook of different types of situations. This is probably something he had to do because his favorite investments – small, growing companies – were not a good way to make money with a big fund.

Warren Buffett evolved into using different frames over time. Much of that evolution – again – was probably spurred by the necessity of managing large amounts of money.

Individual investors don’t face this problem. If you’re really great at solving problems framed as special situations (thinking like an arbitrageur) then go for it.

There are a few keys to framing investment problems:

1. Monitor your actual behavior – make your favorite categories explicit

2. Admit when you are using a favorite frame – accept man with a hammer syndrome applies to you too

3. Don’t force yourself to use a certain frame out of a desire to impose a certain self-identity

This last one sounds odd. I think it’s the most important. Some folks want to see themselves as long-term investors focused on moats – but their actual actions tell a different story. This is why monitoring your behavior is key.

A record is a mirror. It forces you to see yourself as you are rather than as you’d like to see yourself. You can always change your actual self to better mesh with what you want to be.

But that’s not the same as trying to force yourself into framing problems a certain way – because you want to be like Charlie Munger.

If your transaction log shows you invest like Ben Graham and your self-identity says you’re a Charlie Munger style investor – you’re in trouble. You probably aren’t admitting the frames you actual use. You are using a lot of unspoken process rules rather than making your process explicit.

Once you monitor and break down your actual analytical process you’ll be able to see the frames you really use. And you’ll be able to focus your efforts on honing the tools you actually use – not the ones you claim to use.

See Ben Graham’s “The Intelligent Investor” for a good example of this. Graham started out doing all sorts of things. He even did long/short pairs. He realized only a few categories – net-nets, special situations, and control investments – drove almost all of his returns. He decided to focus on those frames.

You can do the same.

Talk to Geoff about How to Frame an Investment Problem