Read on Twitter
1/ Quick thread on intrinsic value...
How should we define intrinsic value?
The academic view is the following: It is the discounted value of the cash that can be taken out of a business during its remaining life.
There might be a better way to think about this though
How should we define intrinsic value?
The academic view is the following: It is the discounted value of the cash that can be taken out of a business during its remaining life.
There might be a better way to think about this though
2/ The DCF definition is great in a theoretical sense, but it is mired with practical difficulties.
- It is difficult to ascertain WHEN a company will reach its "terminal" or boring state. This usually doesn't happen in cookie cutter 5/10 year periods.
- It is difficult to ascertain WHEN a company will reach its "terminal" or boring state. This usually doesn't happen in cookie cutter 5/10 year periods.
3/
- Figuring out cash that can be taken out in a given year is nearly impossible. Investments also don't follow annual time stamps - this is particularly important with companies that have LT investments in infrastructure - think AWS
- Figuring out cash that can be taken out in a given year is nearly impossible. Investments also don't follow annual time stamps - this is particularly important with companies that have LT investments in infrastructure - think AWS
4/
- The output of a DCF is a single number. Once you get anchored to that number, few will "modify" their DCF to account for changing expectations and changes in discount rates. The temporal changes in a company's value are lost.
- The output of a DCF is a single number. Once you get anchored to that number, few will "modify" their DCF to account for changing expectations and changes in discount rates. The temporal changes in a company's value are lost.
5/
Here's the working definition I use.
Fair value = The price at which the expected forward returns of a company equal the returns of the overall index.
Here's the working definition I use.
Fair value = The price at which the expected forward returns of a company equal the returns of the overall index.
6/
For example, if the average company grows at 5% and trades at a P/E of 15x, the "fair" multiple of a company growing at 10% is double that, or 30x, assuming multiples revert to the mean.
Why do I like this definition?
For example, if the average company grows at 5% and trades at a P/E of 15x, the "fair" multiple of a company growing at 10% is double that, or 30x, assuming multiples revert to the mean.
Why do I like this definition?
7/
- It forces you to have a view on expected earnings growth (i.e. everything that goes into a DCF) while ALSO having a view on how multiples - or sentiment - will change over time. People are scared to project multiples but truth is they have a huge effect on your return...
- It forces you to have a view on expected earnings growth (i.e. everything that goes into a DCF) while ALSO having a view on how multiples - or sentiment - will change over time. People are scared to project multiples but truth is they have a huge effect on your return...
8/
- Continuously evaluating forward returns relative to the market allows you to update your expectations accordingly. If an event occurs that increases your certainty that above-average growth will materialize, you should be willing to pay more.
- Continuously evaluating forward returns relative to the market allows you to update your expectations accordingly. If an event occurs that increases your certainty that above-average growth will materialize, you should be willing to pay more.
9/
- You can easily deconstruct your "thesis" into your views on multiples and earnings growth. You can adjust your internal confidence interval of expected returns based on the certainty of each of these points. Also allows you to understand where your return actually came from
- You can easily deconstruct your "thesis" into your views on multiples and earnings growth. You can adjust your internal confidence interval of expected returns based on the certainty of each of these points. Also allows you to understand where your return actually came from
10/
- Everyone will have different views on expected index returns and expected growth of certain companies. So, this fair value is different for people with different information/views. In a sense, everyone has their personal "internal benchmark".
- Everyone will have different views on expected index returns and expected growth of certain companies. So, this fair value is different for people with different information/views. In a sense, everyone has their personal "internal benchmark".
11/
Personally, I think the market will return ~5-6%/year over the next long period of time using assumptions on growth in corporate earnings. So, I look for companies that will generate enough earnings growth to top this AND any multiple contraction that may occur.
Personally, I think the market will return ~5-6%/year over the next long period of time using assumptions on growth in corporate earnings. So, I look for companies that will generate enough earnings growth to top this AND any multiple contraction that may occur.
12/
A P/E ratio of 50x nor a ratio of 5x seem cheap or expensive to me. It is all relative to expected growth and where these multiples can go over time.
I like it when the lower bound on my confidence interval is ~5.5%. I don't pay too much attention to the upper bound.
A P/E ratio of 50x nor a ratio of 5x seem cheap or expensive to me. It is all relative to expected growth and where these multiples can go over time.
I like it when the lower bound on my confidence interval is ~5.5%. I don't pay too much attention to the upper bound.
13/
There are some problems with this definition though. First, it throws the word "intrinsic" out the door since forward returns are company AND market-determined. Second, most assumptions of forward returns will be wrong - that's where the confidence interval comes in.
There are some problems with this definition though. First, it throws the word "intrinsic" out the door since forward returns are company AND market-determined. Second, most assumptions of forward returns will be wrong - that's where the confidence interval comes in.
14/
Despite potential theoretical failings, I like this definition because it forces you to focus on what's important instead of relying on some elusive, unchanging output of a static spreadsheet.
Despite potential theoretical failings, I like this definition because it forces you to focus on what's important instead of relying on some elusive, unchanging output of a static spreadsheet.
15/
I've been "valuing" companies using expected returns for 2 years now, and others are transitioning from DCFs to this approach as well. For example, I know @borrowed_ideas does this too [check out his writing!].
If any academics follow me, I'd love to get your views!
I've been "valuing" companies using expected returns for 2 years now, and others are transitioning from DCFs to this approach as well. For example, I know @borrowed_ideas does this too [check out his writing!].
If any academics follow me, I'd love to get your views!
