General Category > Fairfax Financial

Fairfax Valuation

(1/3) > >>

jfan:
Hey everyone,

Fairfax is a pretty complex beast to value with all its moving parts but thought I would give it an amateurish attempt.

Just wondering if all you nice people on this board won't be opposed to provide me with some criticism and feedback to my simple dcf model.

Thanks for your time.

Jerome

Spekulatius:

--- Quote from: jfan on April 16, 2019, 07:11:12 PM ---Hey everyone,

Fairfax is a pretty complex beast to value with all its moving parts but thought I would give it an amateurish attempt.

Just wondering if all you nice people on this board won't be opposed to provide me with some criticism and feedback to my simple dcf model.

Thanks for your time.

Jerome

--- End quote ---

It seems that you tax investment earnings twice - one time by putting a tax on investment earnings (1/3) and the second time, but assuming a company tax rate of 27%.

jfan:
The 2/3 discount on the investment income was just a margin of safety that was imposed in order to be conservative. I recognize that it will be quite difficult to forecast their future investment returns as it will be dependent on the skill of the team, market sentiment changes to revalue their stock picks, the weighted average interest rate changes, global macroeconomic Dynamics, ever changing capital shifts in their portfolio, etc. I just reasoned that given the uncertainty and the complexity of all the variables, trying to predict a precise estimate would be fool's game and that the central tendency would be the most least error prone value.

jfan:
I've attached an updated FFH valuation model spreadsheet for everyone.

I've included some historical data, estimated yield on their bond portfolio, estimated current investment portfolio yield, as well as their geometric total return portfolio (historically). I assumed a 2% GDP related growth of the float.

I realize everything here is a rough approximation and most likely everyone here will have a more precise model.

For what I gather, there are a few interesting points (at least to me):
1) The current investment portfolio likely will hover around the range of 4.5% given its cash, bond, stock composition. To get their investment returns to 7% will require a common stock return of >20% to achieve this level or much higher interest rates.
2) It seems they have been quite dependent on growing their share of profits/losses from their equity accounted investments over time (especially their non-insurance side, Fairfax AFrica and India fall here as well) to make up for their portfolio returns. This portion has been growing 33% yearly over the past 10 years.
3) Also getting to a 95% Combined ratio is likely a stretch as well. Their 11 year average CR is about 98.6%.



omagh:
Insurance companies are usually modelled using a sum of the parts (SOTP) and book value (BV) multiple.  So, something like WRB is at the top range of valuation (approaching 2x BV) while Fairfax (sub 1x BV; use US$ as financials are US$) is at the bottom range of valuation.  Your view on BV growth and a range of BV multiples is probably more meaningful than DCFs which are a trickier model to get 'right'. 

Fairfax is saddled with a 1xBV valuation because their portfolio management has been atrocious for years.  They play the Templeton-style deep value game, in the small percentage of their portfolio dedicated to equities, which hasn't worked for ages.  The bulk of their portfolio is in cash and bonds which was great during the bond bull market but tremendously limits their upside going forward.  At best, Fairfax will grow their book value by xx% and possibly get a 1.2 or 1.3xBV bump when one of their deep value plays finally pays off.  There are easier ways to make money.

DCFs are more appropriate for predictable cash-gushing businesses than insurance companies which have catastrophe years where cash is needed for payouts.  Even still, favorable assumptions in DCFs lead to a large tail value which is quite probably wrong.  Caveat emptor.


--- Quote from: jfan on April 16, 2019, 07:11:12 PM ---Hey everyone,

Fairfax is a pretty complex beast to value with all its moving parts but thought I would give it an amateurish attempt.

Just wondering if all you nice people on this board won't be opposed to provide me with some criticism and feedback to my simple dcf model.

Thanks for your time.

Jerome

--- End quote ---

Navigation

[0] Message Index

[#] Next page

Go to full version