Attached is a DCF template that i used to value the hyperscalers.I am just sharing it here in case anyone wants to experiment with it.
[https://docs.google.com/spreadsheets/d/1oexuOkRTpnQCPWyQDTiE-Z2OEihBfci6FLysPJIysR4/edit?usp=sharing](https://docs.google.com/spreadsheets/d/1oexuOkRTpnQCPWyQDTiE-Z2OEihBfci6FLysPJIysR4/edit?usp=sharing)
Please read the notes in the Google Sheet tab. I have listed down the assumptions:
**Notes on the valuation**
**Purpose**: i try to have a standard template across MSFT, META and GOOGL
And i try to keep the number of moving parts small.
And as new information come online, i will modify the assumptions
The thing about valuation, almost nobody will agree with me on the assumptions. So do change it yourself as you see fit.
Areas of contention: WACC, Terminal Growth rate vs Terminal multiple, composition of debt etc
**High level assumptions**
1. The duration of abnormal growth is 10 years. All three companies exhibit evidence of a moat, each have strong competitive advantage for many years. Normally for tech companies, i would use a 5 year duration, however for MSFT-META-GOOGL i think they are in a a strong position to do really well and i use a 10 year duration.
2. The discount rate is 9% and terminal growth is 3%
3. Formula of Free Cash Flow **FCFF = NOPAT + D&A – CAPEX – Δ Net WC**
4. We assume the following to be stable unless specified differently:
EBIT margin
Capex / Revenue Unless FADE
Tax Rate
1. Long live assets vs Short lived assets
For MSFT, it is suggested that ⅔ are short lived assets depreciated over 6 years
For META and GOOGLE, 75% is the ratio for short lived assets
For META, they specified 5.5 years but i round it to 6 as with Google
For long lived assets they depreciate over 15 to 25 years.
**Specific Tabs**
For each company there are 4 valuation tabs drawing the assumptions off the assumption tabs
FLAT means that 1 dollar of capex generate a fix amount of revenue. This means that capex/revenue is quite fixed. Management does not seek to optimise efficiency. This is the opposite of the Tab labelled “FADE” where there are efficiency implied. For example Capex/Sales could start at 45.5% and overtime, due to efficiency, capex fades to 25% of sales in 10 years.
There are two additional tabs, one is FLAT Escalating, and FADE escalating. In theory, over the long term at the terminal stage, Capex = D&A you a bit like maintenance capex. However, it is suggested that some companies with continue to outspend capex over D&A over the long term even in the terminal stage.
I thought i would include all 4 tabs for each company, because depending on the company, management discipline, and CFO’s ability to manage finance, you can use the appropriate tabs.
For example, MSFT’s CFO is famous for being very disciplined and executing on her plans. It is likely that MSFT Capex behavior will fall under the “FADE” tab, where capex/sales will fade from 45.5 to 25%.
As for Google, my wager is that the company will “fade” as well since they have their own TPU and will continue to improve the efficiencies but I think they will be very hypercompetitive and try and outspend in order to win, hence i believe “Fade Escalation” is more appropriate.
As for Meta, they have demonstrated that they like to overspend like a drunken sailor, just looked at how they spent 100bn in Reality Labs since 2020. So even though they do not have a cloud business for business customers, i think MZ will continue to spend on capex. I think the most appropriate tab is “FLAT escalation”
In terms of the formula between no escalation and escalation, the calculation is different.
In terms of Revenue growth rates, i purposely did not choose a wildly high growth rate, i looked at the past years growth, the next 5 years expected growth rates, the log liner analysis etc. And i just used my judgement. ShaunSheep went through my assumptions and felt that my year 10 target revenue growth rates are too pessimistic. I am sticking to my guns, that if i guide conservatively and the valuation numbers are still higher than the current share price then perhaps the company is undervalued.
The biggest issue i faced was calculating debt. Where possible i took a conservative calculator of bonds, finance leases and operating leases. I used AI to check for me on the nos.
