Wednesday, October 18, 2006
Valuation stuff
This post is intended to answer a question in the comments section in my last post about Epolin (EPLN) here.
First off, at this point, the present value of future free cash flow is so well established as a method for determining the value of a security that I don't think I need to explain why. If you don't know what net present value is or how to compute it or if you disagree about that being the basis for valuation, then that's a totally different issue.
Below is a chart of three cash flows, with the x-axis being time.
If you were to rely on the immediate P/E ratio alone, A and C would have roughly the same value while B would have a higher value. If you were to rely very heavily on the rate of growth in free cash in the recent past and near future, then you'd say that A was the most valuable, followed by C and then B. In reality, we should be figuring out the area under the curve, discounting by increasing amounts the further we look forward in time (not that I'm actually going to sit down and do the integral).
Now I know that I very often just slap a P/E ratio on free cash flow streams (and often I just use earnings, when I believe that's a reasonable measure of free cash flow), but in every case [hopefully], I'm always considering what I know about the future of that free cash flow stream. Very often, I really don't know and I assume something a bit like B above and slap a P/E of 15 onto it and use that as the value.
Sometimes I believe that a business model is such that the current free cash flow is in danger. It could be due to future competition and a lack of any sort of moat. It could be due to big changes in the future or even just big uncertainties in the future. Technology companies usually deal with short lived cash flow streams like A and they constantly need to create new cash flow streams with new products. But in reality, nearly all cash flow streams look like A over a long enough time period. For a business to survive for any length of time, it needs to find new sources of free cash flow to replace ones which are declining.
One point I've tried to make on various message boards (mostly TMF Berkshire) is that your valuation of these three cash streams will differ depending on what discount rate you use. An unusually high discount rate will make near-term cash seem much more desirable than cash farther out in time. It will distort your perception of value. Likewise, an unusually low discount rate will make cash in the distant future look better than it really is. With a 15% discount rate, you have no patience. With a 2% discount rate, you'll ignore cash today in favor of cash too far into the future.
So anyway, my reason for considering the value of EPLN to be around a dollar is due to my sense of what the future free cash flow looks like. I see it more like C than B above. But I could be wrong. For me, EPLN is a fairly small investment and if I had more companies like ETLT, CXTI, and STHJF then I probably wouldn't be invested in EPLN.
First off, at this point, the present value of future free cash flow is so well established as a method for determining the value of a security that I don't think I need to explain why. If you don't know what net present value is or how to compute it or if you disagree about that being the basis for valuation, then that's a totally different issue.
Below is a chart of three cash flows, with the x-axis being time.
If you were to rely on the immediate P/E ratio alone, A and C would have roughly the same value while B would have a higher value. If you were to rely very heavily on the rate of growth in free cash in the recent past and near future, then you'd say that A was the most valuable, followed by C and then B. In reality, we should be figuring out the area under the curve, discounting by increasing amounts the further we look forward in time (not that I'm actually going to sit down and do the integral).
Now I know that I very often just slap a P/E ratio on free cash flow streams (and often I just use earnings, when I believe that's a reasonable measure of free cash flow), but in every case [hopefully], I'm always considering what I know about the future of that free cash flow stream. Very often, I really don't know and I assume something a bit like B above and slap a P/E of 15 onto it and use that as the value.
Sometimes I believe that a business model is such that the current free cash flow is in danger. It could be due to future competition and a lack of any sort of moat. It could be due to big changes in the future or even just big uncertainties in the future. Technology companies usually deal with short lived cash flow streams like A and they constantly need to create new cash flow streams with new products. But in reality, nearly all cash flow streams look like A over a long enough time period. For a business to survive for any length of time, it needs to find new sources of free cash flow to replace ones which are declining.
One point I've tried to make on various message boards (mostly TMF Berkshire) is that your valuation of these three cash streams will differ depending on what discount rate you use. An unusually high discount rate will make near-term cash seem much more desirable than cash farther out in time. It will distort your perception of value. Likewise, an unusually low discount rate will make cash in the distant future look better than it really is. With a 15% discount rate, you have no patience. With a 2% discount rate, you'll ignore cash today in favor of cash too far into the future.
So anyway, my reason for considering the value of EPLN to be around a dollar is due to my sense of what the future free cash flow looks like. I see it more like C than B above. But I could be wrong. For me, EPLN is a fairly small investment and if I had more companies like ETLT, CXTI, and STHJF then I probably wouldn't be invested in EPLN.
// posted by Bruce @ 5:49 PM 
Comments:
<< Home
Bruce, thanks for this post.
Sure, cash flows, NPVs and DCF models are the basis for all. Nothing new here.
It would be great, however, for you to simply write down the numbers you used to place the $1 tag on EPLN. IMO examples are the best way to educate.
p.s. not that EPLN is particularly interesting. I use it just as an example.
Sure, cash flows, NPVs and DCF models are the basis for all. Nothing new here.
It would be great, however, for you to simply write down the numbers you used to place the $1 tag on EPLN. IMO examples are the best way to educate.
p.s. not that EPLN is particularly interesting. I use it just as an example.
# posted by 
: 3:12 AM, October 19, 2006
: 3:12 AM, October 19, 2006
I don't have numbers because I can't see the business that well into the future. I've seen a lot of people jump into details like that to come up with numbers and I've done it myself lots of times, but I find that it tends to push away the bigger picture and you get lost in details. I think this is what Buffett was talking about in terms of being roughly right vs precisely wrong.
If you look at the market for infra-red dyes and these kind of specialty chemicals, it just seems that as China and India grow a gigantic middle class, that the market and also the "usefulness" of what EPLN does overall will grow as well. In my professional work over the last 24 years, I've seen the same sort of slow transition which concentrates importance in places which generate the knowledge: the technology providers emerge. This business is already there, they just haven't established that relation with the rest of the world.
Post a Comment
If you look at the market for infra-red dyes and these kind of specialty chemicals, it just seems that as China and India grow a gigantic middle class, that the market and also the "usefulness" of what EPLN does overall will grow as well. In my professional work over the last 24 years, I've seen the same sort of slow transition which concentrates importance in places which generate the knowledge: the technology providers emerge. This business is already there, they just haven't established that relation with the rest of the world.
<< Home
