I would just like to say thank you for the work you put in! This was an extremely well-written and accessible analysis that left me with a lot of food for thought!
Hey Mathias, thank you so much for the feedback. Reading messages like this motivates me to keep producing well thought out analysis that is accessible to all.
Wolf I think your FCF is overstated because it includes the add back of SBC. I think you have implied that you are accounting for that with a dilution assumption. Is that correct? If so, can you share how you've done do?
I am accounting for dilution through an increase in the share count so the FCF is not overstated. I’ve included dilution of 1% per year for the 5 years.
How do you think it compares to Booking Holdings? It seems to be trading much cheaper with a similar growth outlook, superior long-term strategy, and larger market share.
We had the same question! Booking has also stepped up in private bookings too which could increase competition with Airbnb customers. Still, totally agree Airbnb can increase market share due to still being in its infancy currently. The advertising point is very interesting, I thought they already did this!
I can't say that I have studied Booking enough to given an intrinsic valuation but on a relative P/E valuation it appears cheaper. On growth, the consensus is that Airbnb will grow by double digits for the next number of years compared to 8/9% for Booking. If this transpires then it means that Airbnb is increasing its market share. The real value creation for Airbnb will be its own advertising platform which I feel is being ignored at the moment. Two great businesses in my opinion, I use both regularly.
Do you think the terminal growth in FCF is understated in the analysis?
If you assume a terminal growth of 4% (2% from volume and 2% from pricing due to inflation) then shares trade at a 43% discount.
You could assume that unit growth in property listings/stays grows at the rate of real growth in GDP (probably 2% over the long term). This is because we will probably build a net 2% more properties each year over the long term and as people become more productive they will have more discretionary time and income to travel.
You could also assume that the revenue per listing grows at the long term inflation rate (also probably 2% in the long term). This is because hosts charge daily rates in nominal terms and since revenues are a function of the product of take rate and GBV then the price per listing should increase by 2% over the long term as well.
Thank you for reading. I can see where you are coming from but I would never use a 4% terminal growth rate for any business. If you do that you are saying that the company will be bigger than the world’s GDP which is obviously impossible. Hope that makes sense.
Agree with this response. I think the assumption of 2% pricing growth due to inflation is entirely reasonable but another 2% (for a total of 4%) forever is extremely aggressive and as Wolf says, over the period of forever would eclipse the entire economy.
I can definitely get behind the reluctance of "4% terminal growth". You could instead reframe the question to talk about terminal multiples. For example, I don't think it would be that crazy to project that AirBNB could trade at a P/FCF of 17.3 instead of 12.75.
A terminal growth rate of 2% implies a 12.75 P/FCF multiple and a terminal growth rate of 4% implies a 17.3 P/FCF multiple. It doesn't seem to be *as* crazy when framed differently (even though "grows bigger than the world's GDP" is technically what 4% is saying).
What do you guys think? Is that still bad practice for a DCF?
If I recall correctly, the best practice if I wanted to model 4% growth, would be to add a second growth phase (at 4%) prior to the 2% terminal growth phase.
Yes that's right. If you want to project higher growth you would be better served extending your explicit forecast period and applying a growth rate, then using a GDP-like or inflation-like terminal growth rate from the time you think the business may enter "steady state" maturity.
Agree with Wolf's earlier reply to this as well re mixing multiple "valuation" with DCF valuation. If you include a multiple in your DCF (usually used to calculate the terminal value) then it isn't really a DCF.
I moment I realised a multiple is an "output" of a valuation model instead of an "input", my understanding completely shifted.
I guess the issue with using an exit multiple is now you are mixing an intrinsic valuation technique with a relative valuation technique. A true DCF represents the future cash flows generated by the company. If you use an exit multiple the calculation is no longer intrinsic.
FWIW I initially modelled using exit multiples but I moved to a terminal growth rate for the above reason. Nothing wrong with using valuation multiples in general but I’d avoid mixing with a DCF.
I would just like to say thank you for the work you put in! This was an extremely well-written and accessible analysis that left me with a lot of food for thought!
Hey Mathias, thank you so much for the feedback. Reading messages like this motivates me to keep producing well thought out analysis that is accessible to all.
Wolf I think your FCF is overstated because it includes the add back of SBC. I think you have implied that you are accounting for that with a dilution assumption. Is that correct? If so, can you share how you've done do?
I am accounting for dilution through an increase in the share count so the FCF is not overstated. I’ve included dilution of 1% per year for the 5 years.
How do you think it compares to Booking Holdings? It seems to be trading much cheaper with a similar growth outlook, superior long-term strategy, and larger market share.
We had the same question! Booking has also stepped up in private bookings too which could increase competition with Airbnb customers. Still, totally agree Airbnb can increase market share due to still being in its infancy currently. The advertising point is very interesting, I thought they already did this!
Thanks for the feedback 👍
I can't say that I have studied Booking enough to given an intrinsic valuation but on a relative P/E valuation it appears cheaper. On growth, the consensus is that Airbnb will grow by double digits for the next number of years compared to 8/9% for Booking. If this transpires then it means that Airbnb is increasing its market share. The real value creation for Airbnb will be its own advertising platform which I feel is being ignored at the moment. Two great businesses in my opinion, I use both regularly.
This was a fun read!
Do you think the terminal growth in FCF is understated in the analysis?
If you assume a terminal growth of 4% (2% from volume and 2% from pricing due to inflation) then shares trade at a 43% discount.
You could assume that unit growth in property listings/stays grows at the rate of real growth in GDP (probably 2% over the long term). This is because we will probably build a net 2% more properties each year over the long term and as people become more productive they will have more discretionary time and income to travel.
You could also assume that the revenue per listing grows at the long term inflation rate (also probably 2% in the long term). This is because hosts charge daily rates in nominal terms and since revenues are a function of the product of take rate and GBV then the price per listing should increase by 2% over the long term as well.
Thank you for reading. I can see where you are coming from but I would never use a 4% terminal growth rate for any business. If you do that you are saying that the company will be bigger than the world’s GDP which is obviously impossible. Hope that makes sense.
Agree with this response. I think the assumption of 2% pricing growth due to inflation is entirely reasonable but another 2% (for a total of 4%) forever is extremely aggressive and as Wolf says, over the period of forever would eclipse the entire economy.
100% agree 👍
I can definitely get behind the reluctance of "4% terminal growth". You could instead reframe the question to talk about terminal multiples. For example, I don't think it would be that crazy to project that AirBNB could trade at a P/FCF of 17.3 instead of 12.75.
A terminal growth rate of 2% implies a 12.75 P/FCF multiple and a terminal growth rate of 4% implies a 17.3 P/FCF multiple. It doesn't seem to be *as* crazy when framed differently (even though "grows bigger than the world's GDP" is technically what 4% is saying).
What do you guys think? Is that still bad practice for a DCF?
If I recall correctly, the best practice if I wanted to model 4% growth, would be to add a second growth phase (at 4%) prior to the 2% terminal growth phase.
Yes that's right. If you want to project higher growth you would be better served extending your explicit forecast period and applying a growth rate, then using a GDP-like or inflation-like terminal growth rate from the time you think the business may enter "steady state" maturity.
Agree with Wolf's earlier reply to this as well re mixing multiple "valuation" with DCF valuation. If you include a multiple in your DCF (usually used to calculate the terminal value) then it isn't really a DCF.
I moment I realised a multiple is an "output" of a valuation model instead of an "input", my understanding completely shifted.
I guess the issue with using an exit multiple is now you are mixing an intrinsic valuation technique with a relative valuation technique. A true DCF represents the future cash flows generated by the company. If you use an exit multiple the calculation is no longer intrinsic.
FWIW I initially modelled using exit multiples but I moved to a terminal growth rate for the above reason. Nothing wrong with using valuation multiples in general but I’d avoid mixing with a DCF.