10 Comments
User's avatar
DIY Investor's avatar

Another way to look at this for the Terminal Value calculation: I will grow this cash by small amount (in line with GDP growth, 3% for instance) after the explicit growth years of 10 years, then assume it will go on for 100 years (reversing 1/(discount rate - perpetuity growth rate)% yields 100 divided by a number). Holding growth rate constant (assuming 3% growth in line with GDP is being conservative, but that is the point) and assuming higher discount rate such as 13% aka desired return means you are taking a smaller portion of those 100 years as terminal value. Less of those 100 years you add to your intrinsic value, the better because 100 year is too long with many unknowns. Hence, assuming higher TV multiple is riskier.

Numbers explain better:

Cash flow x (1+3%) x 1/(13-3)% = Cash flow x (1+3%) x 100/10. Since 100/10=10, we are taking only 10 years out of 100 for the TV calculation.

Maybe, I'm complicating it by trying to break it down.))

Expand full comment
Speedwell Research's avatar

I'm not sure i'm following where the 100 in this part comes from "Cash flow x (1+3%) x 100/10"

If you are using the perpetual growth formula, I don't believe you can assume a set number of years (even if 100).

I could be misunderstanding you though

Expand full comment
DIY Investor's avatar

Maybe I'm babbling here. I will try again. 13%-3%=10%=10/100. Since it is 1/(10/100) in the formula it becomes 100/10. So we are basically multiplying the cash flows by 100 years then dividing by ten.

Expand full comment
Speedwell Research's avatar

Oh I see. Yeah I don't believe you can multiple it by "100 years" if you are using the perpetual formula.

If you are going to make an explicit forecast period (the 100 years) then you need to estimate cash flow for each year and discount it.

I'm not exactly sure how, but you are getting the same answer as if the "100 years" didn't exist though.

The perpeutual formula for the figures you have [ 1 / (13%-3%) ] -> 1 / 10% is 10x.

Expand full comment
DIY Investor's avatar

Yes. We are not multiplying it with 100 but 10 which is a lower multiple. The lower the multiple the less of the "100 years" we are using, less riks we incorporate to our DCF. My last attempt to make sense. Thanks for replying!

Expand full comment
DIY Investor's avatar

This is how you internalize the valuation process like Buffet. Must read for everyone regardless of experience. I have recently written about this. I will edit and add this to my post for my subscribers. Thank you.

Expand full comment
Chris Stevo's avatar

Outstanding! I’ve used the reverse DCF for years. It’s always annoyed me when people whine about DCF’s being “garbage in, garbage out” or “too complicated and not useful”…. Any DCF value or model is almost INHERENTLY wrong, but that’s entirely irrelevant. It’s an abstraction of reality that allows you to simplify your thinking around the key drivers and the sensitivity of value to those different drivers. It helps you focus your time and effort where it will give you the most value. Reverse DCF takes that a step further and tells you what the market is telling you about the stock today. It might be right and it might be wrong, but you’d be a fool to ignore it. Figuring out where you have a differentiated view from the market and whether that view can be refuted is the name of the game. As Munger said, “Invert - always invert!”

Expand full comment
Speedwell Research's avatar

Thrilled you liked it! And agreed!

Expand full comment
George Norris's avatar

Awesome. Thank you

Expand full comment
Speedwell Research's avatar

Glad you liked it!

Expand full comment