In my spare time I have taken up financial statement / valuation analysis of public companies again.
I have always felt that there was something missing from my skill set from a financial statement analysis perspective. This problem only really arises when I look at a public company, as its a bit more open-ended and vague to me.
I can do all of the technical stuff, and build everything out, but when it comes to actually sitting there and looking at the historical numbers, I really struggle to contextualize it all when looking at individual public companies.
To use an exaggerated example, if an investor says a company has "great FCF generation." - I can see what the FCF number is, but it means absolutely fucking nothing to me.
As of right now I take a fairly individual approach to companies. I do look at comps, but don't really dive deep, as I don't not have that much time.
Its as if there piece of the puzzle that's missing, and keeping it from truly clicking. I have tried to approach things a few different ways to improve it, but I don't think I am tackling it as effectively as I could be.
Has anyone else experienced this?
What helped you best develop a "feel for it"? What would be the most effective use of my time to sharpen this skill?
Is it as simple as diving deeper into the comps to get a better feel for a companies relative position? Or is it a bit more than that?
You're on the right track. Just rinse and repeat my brother
A lot of this job, especially in the earlier days, is building up pattern recognition and contextualization skills. The short answer is: comps. The long answer is comps against your accumulated knowledge base.
Over time, you build up an understanding of business models and cost structures and typical growth algos and typical cost re-orgs, etc. Software biz vs. tech distributor or consumer biz; very different margin profiles. I think DCFs, at least starting out, can be beneficial, as it helps you to understand the implicit assumptions that get hidden when people slap multiples on things, and it gives you a more explicit sense of the cash and earnings generation through cycles over long time periods that people are willing to underwrite (and how those assumptions reflect things about business quality + moats + ROIC etc.), and why stocks can be cheap at 30x and expensive at 5x NTM EPS.
You are on the right track, but it takes time to contextualize these figures. Just have to keep at it. So as you get further along, you will be able to contextualize the difference behind an E&P trading at a +10% FCF yield and an SPGI/INTU trading at a 5% FCF yield.
How's it going these days man if you dont mind me asking? Have you made the move that you wanted to, or do you have a change in plans :)
Hey! Some ups and downs; from my vantage point, the opportunities vanished in 2H23, but are slowly showing up again it seems. Still working on transition, but keeping my options a bit broader in terms of opportunities I am pursuing
May I ask what transition you have been trying to make? I got long-covid last year and accordingly got caught up in the banking layoffs. Took a corp dev role recently to see if I like it and to get some money coming in. So I can relate in terms of ups and downs
Thanks its super helpful when you lay it out that way. Much appreciated.
Experience helps for sure, and makes a world of a difference.
But also how much effort you put into things, a lot of people just don't dig (even when they're applying to buyside).
You might have 0 insight on whether a 30x multiple is cheap or what its actually implies, but if you chart out P/E quarterly for the last 10 years and comb through 10-k/whatever throughout that time you can get a sense on where the multiple is today and what is implying on growth.
I think that's part of my problem as well. Like sure I can look at historical financial statements, but I'm realizing it means fuck all when you aren't tracking how it matches up to the price across time. Looking at how historical decisions made with respect to capital investment, and strategy etc. have affected a company's value.
A little bit. What I was trying to get at above is that over time you will build more contextualization skills when it comes to valuing different types of businesses (and eventually you'll understand why certain people will talk about "excellent FCF generation", and the relevancy and/or implications of such comments, which varies widely. So in that sense, DCFs are an early tool that can be useful as you build that up.
The buyside likes to dunk on people who build DCFs, as at the end of the day, it arguably is a bit useless for the game many are playing (in the sense that arguing that a company has this intrinsic north star valuation that it deserves to trade at isn't applicable in a lot of settings + since the multitude of assumptions required drives so much variation in outcomes, and your range of outputs aren't necessarily that helpful depending on the timeline you are playing for).
What these people take for granted when they say that, is that they probably built a ton of DCFs over time already (whether through school, or IB I assume, or as an early value investor enthusiast), and that through these process they have learned the context behind valuation multiples, and how topline growth + margins + ROIC feed into valuation ranges. The most important part of the DCF is not the outputs, but the ability to flex your model through different scenarios and get a better grasp in understanding what you need to underwrite to justify the stock price across a range of valuations (and whether the quality of the biz or the biz model is capable of doing that).
I'd also recommend checking out Mckinsey's Valuation book
When you are working on a stock thesis, depending on what type of strategy applies, the entire thing may come down to: stock de-rated on a big two quarters ago and has been in dog house, but intra-q checks + peers + leading indicators in sector indicate accelerating KPIs. That one KPI may be the most important thing in driving this business model over the long term in your DCF/3yr quarterly model, and so if the stock reports an acceleration here, the market is likely to extrapolate out the result and you are playing for the re-rating upwards to historical average multiple, etc. So can you properly diligence year 10's estimates in your DCF? Does the DCF range matter for your thesis? Not significantly - but you know that company's that LOOK like they can grow +8% over the long term with stability trade at a much different multiple than companies that trade at 3% LT with lots of cyclicality. If you can nail when the market changes their attitude on the stock, you are likely to make a lot of money. This context for stocks in your coverage, and cyclicality, and peak on peak on trough, etc., accumulates over time. The sell side may "this stock is a buy because FCF generation is solid - but you need to eventually contextualize both the importance and relevancy of that statement, especially depending on the type of stock thesis you are crafting and your investment philosophy.
Sorry I kind of ranted here - needed a break from earnings.
This was great thank you
The commenters above are all correct. It's really repetition and accumulated knowledge/experience in your coverage sector. One thing I like to do is really understand connection between numbers on a page and business model/operations. I ask myself: how these numbers got there and what is the story being told?
For example: Company A runs a 80% opex margin drops to 70% by Year 10 and Company B runs a 75% opex margin but held steady by Year 10. (All things being equal). I think about what are the factors that causes the discrepancy. Maybe the number got there b/c CompA over hired or have a bad supplier contract and was able to change the terms. Whereas CompB margin is the best they can do. The story told is that CompA has some awkward cost items that will get "resolved" in the long run.
I try to conceptualize all the numbers and line items changes and think about how that translates into actual business practice. It's one thing to say revenue grows by 15%, it's another to put it into practice. (IB forecast is the worst offender, we all seen those hockey stick rocket ship go to the moon charts). Once you seen enough financial statements and read the MDNA you'll get an understanding the driving factors that gets the company to be "great FCF generation".
This was a mistake I made when I was an analyst. I would just see numbers go up and think great! Or see a stack of P/E ratio comps then just pick the average or median. Not really thinking why is this PE ratio applicable or why certain comps are excluded. Take a step back from just seeing numbers on a page and really seek to understand why the revenues go up (macro and micro reasons).
This was a bit me rambling but hope it helps and gets you thinking about the actual business operations & underlying business/economic driving factors.
In short, time and reps... read transcripts... i'm a visual learner, so visualize HQ/plants/units/salesforces... learn how the sausage is made and picture yourself as running the enterprise on your own - what info would you rely on to make $$$ and grow? ... speak to people in the industry and see what frameworks they assume for whether a particular point-in-time or player or event is "good" or "bad" ... spread comps using your own dimensions on whatever factors/variables you care about/you think matter to make $$$... a mindshift into owning your own thought process/logic is very valuable for an investor (versus relying on some rules/codes that exist in the third party greater world - that you fear you don't fully understand). Always learn. Master these rules - that's table stakes. But investing requires a level on conviction that shifts from learning into making decisions based on info/knowledge (that's never perfect) & wisdom.
It's an "art and science" combination. As you go to pitches and listen to what your boss / your client has to say, you'll get more context on the comps - i.e.client will say something like "yeah Company X spent capex on that crap project last year" or "Company Y's new CEO...". These will form the "qualitative" cues and context for you to start digging into - google "company X fired CEO" and you'll get more of the story, and oftentimes the article will also tell you how their peers are doing.
On the quantitative side of things, you need to sense check - for example if you notice that an LTM multiple is unnaturally high (because of unnaturally depressed EBITDA or NI), they probably had a bad earnings year so you should look into research reports / company presentations / google the company to see what happened. This will also make you look very smart when your seniors ask you why does the multiple look off. Rule of thumb: multiples should decline over time if earnings are increasing, meaning 25E multiples 24E multiples 23. So anything that doesnt fit the shape should be looked into - e.g. "A" shape where 24E multiples are higher than 23 and 25E multiples - means that brokers are forecasting a big earnings dip - you should probably dig into it and figure out why because a senior will definitely ask you about it.
I agree with everything above. Repetition and experience are key.
One thing that really helped me was reviewing equites that performed exceptionally well (or poorly) over a certain time period and reverse engineering the driver of their success. I'm a LO generalist and we typically hold for 3-5 years so we focus on quality and FCF per share.
So, get a handle on:
1. sales (break this down as granularly as possible: geographic, segment, unit level)
2. margins (incremental margins too)
3. capex (growth and maintenance capex, RR)
4. div and buyback policies
5. ultimately FCF/sh
Try and triangulate what's driving performance and then look at 10-ks or SS reports to get up to date on the story around those KPIs. This exercise should improve your ability to tie numbers and narratives together and more efficiently conduct deep dives into the things that matter for the stock. Your due diligence will aid you in finding a probabilistic inflection points in one of the above drivers (differentiated view). Your job then is to flex a DCF with those assumptions and come up with a range of values.
expectations tie into this as well. read expectation investing by Michael Mauboussin.
Going to very specifically respond on the "phenomenal FCF generation" piece, as it's something I used to think about to. You are correct that without benchmarking it to something, it means absolutely nothing at all, and needs to be contextualized.
Usually, I think of it either in terms of benchmarking relative to EV / Market Cap (i.e. FCF Yield), or in terms of FCF conversion from EBITDA to FCF. At the end of the day, even though buying off of EBITDA multiples is the norm in the industry, what you really care about is FCF. As such, you could have a 6x NTM EBITDA biz that seems really cheap, but it ends up spending all of it on capex - on the other hand, you can have a 15x business which has little to no capex and maintains growth fully organically, which might end up "cheaper" as such on buying off an EV / FCF multiple. Either way, both of these end up pointing to FCF yield regardless.
Natus voluptatum nostrum nulla. Dignissimos rerum impedit dicta. Velit est omnis quasi sed perspiciatis aperiam.
Similique animi harum eos possimus expedita. Quas soluta suscipit atque necessitatibus quaerat fugiat vero sit. Magni eligendi velit aperiam quibusdam laboriosam recusandae.
Dolor neque earum ullam quasi sed consequatur. Est quas perferendis facere voluptatem in. Dolores sit in cumque beatae.
Animi expedita culpa et voluptatem. Magnam et eum quis voluptas exercitationem et. Sit nobis animi cum laudantium. Unde similique totam itaque iste. Ratione impedit atque exercitationem aliquam quis.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...