The advantage
of writing a blog for 10+ years is that I don’t have to go anywhere to find
examples of mistakes. I can always find one I have committed and written about
it.
To see the
example of a PE driven investment gone wrong, read the analysis of Facor alloys
here.
In a year’s time, I realized that I had made a mistake and exited this position
with a 12% loss – you can read my analysis here. If I had
held on to the position, I would have lost close to 85% of my investment, even
as the stock continued to sell at a very low valuation (current PE being 3)
Reasoning from
first principles
Reasoning from first principles leads one to understand the fundamentals factors driving the issue in question. So how do we apply this concept to investing?
I wrote the
following on it
Lets break the above point down into its key components
- Free cash flow
- Lifetime
- Discounted
You can find
multiple definitions of cash flow, but the one which I like to use is the cash
you can receive from the asset, without impairing its long term earning
capacity. Lets apply this to a simpler example than a company – A house or a
flat where it is easier to analyze the cash flows.
Free cash flow –
Can be estimated as follows : Gross rent – taxes – maintenance expense – other
overheads
Maintenance
expense usually involves repairing the house, cleaning it after the tenant
vacates it and any other expense incurred to keep it in a rentable conditions.
Other overheads can be society & broker charges to let out a house etc. So
after paying out all these costs, the cash left behind would be the recurring free
cash flow to an owner.
Lifetime - This is the period an asset can be expected to
generate a cash flow. In case of a flat or house one can take as it as 30
years, before one has to permanently replace it with a new construction. In an
extreme condition you can stretch it to 50 years, however try letting out a
very old house and you may realize that the rentals are much below the market
rates.
Discount - The definition
of discounting can be found here. Usually this depends on the riskiness of an
asset.
So how would
you value the house or flat now? The gross rental yields these days are usually around 2-3%. At these yields , one is in effect paying 50 times pre-tax free cash flow. This of course assumes 100% occupancy and no taxes.
Over the long
term these rentals usually follow the inflation rate. So over the life of a
flat or a house, you will earn back around 60% of the cost in the form of
rental. The value of land underlying a house or a flat has been known to
appreciate at the nominal GDP rate (GDP growth + inflation rate).
If you put all
these cash flows together and discount it at around 10%, the final DCF value
comes to about 1.5X purchase price. In other words, the asset is generating an
IRR of 12%.
Is this cheap
or expensive? It depends on what you believe the price of land will be 30 years
from now and if 12% is good enough for the risk and effort of managing a rental
property.
The problem with
PE ratiosAs you can see from the above example, the PE ratio is dependent on several variables which we had to estimate upfront. In the case of some assets such as a rental property, it may be possible to estimate it with a certain level of confidence.
This is however
not always the case
Let say, for
the sake of example, that the house turns out to be on an old burial ground
where there are ghosts and so one want to rent or buy that land J . What happens then? Well the entire
investment goes to zero.
On the other
hand, lets assume that the government announces a large IT park close to the
property and the rental go up by 5X. Irrespective of the actual increase in the
property price, the cash flow based valuation definitely goes up as the rentals
have increased drastically. This is what has occurred in several cities across
the country in the last 10 years.
So the initial
PE turns out to be cheap or expensive depending on the subsequent cash flows
and terminal value of the asset
PE ratio in
equities is even more misleading
In the case of
companies, the problem we face is that the cash flows are quite difficult to
estimate, there is no fixed duration and the terminal value in the real long
run for any business is usually 0.
In the example
of Facor alloys, the PE appeared to be low based on the recent cash flow (as of
2010) which had been in excess of 30 Crs. As a result, if one assumed that
these cash flows would persist, the company appeared cheap at 3-4 times cash flow.
The above
assumption turned out to be wrong. The cash flows were at a peak due to a
cyclical high in demand from the steel industry. In addition to a crash in the
demand, the management diverted the cash flows to another sister firm which
demonstrated poor corporate governance.
In effect, the
expected cash flow and duration turned out to be wrong. In such a scenario, the
PE ratio was simply misleading.
As a counter
example, consider the case of CRISIL
(a past holding) which has always appeared expensive based on the usual
measures of valuation. However the company has delivered above average returns
as it has generated the expected cash flows without much variability in a
fairly predictable fashion. The competitive position continues to improve and
the company is likely to keep growing with a high return on invested capital
for the foreseeable future.
Understand the
business
The only way to
evaluate if a company is over or underpriced is to be able to predict its cash
flow. The higher the valuation, the longer the prediction period.
So if a company
is selling for 2 times earning and you are fairly confident that the current
cash flow will persist for 5 years, then you have a bargain. On the other hand
if you are looking at a commodity company whose cash flows depend on the price
of a volatile commodity, then making any prediction is usually a waste of time.
You may be able to look at some long price charts of the underlying commodity and
get lucky from time to time, but good luck with trying to make it keystone of
your investing strategy.
On the flip
side, if you are looking at a company selling for 100 times earnings, one needs
to have a high degree of confidence on the expected cash flow for 20+ years and
beyond. Anyone claiming such clairvoyance is worth of worship !!
The sweet spot
is usually when the valuations appear reasonable (in 15-25 range) and one can
make a reasonable estimate of the cash flow based on an in-depth understanding
of the company, its industry and the competitive situation.
In summary, the
best way to approach an investment candidate is to filter out the extreme cases
and then dig into the business as much as possible. This should help one make a
reasonable estimate of the cash flows and its duration. Once you have a
reasonable fix on these key inputs, doing a valuation and comparing it with the
market price is the easy part.
Homework: Is
coal india Ltd a value stock?
It is selling
for 10 times earnings net of cash for sure. Personally I think the PE ratio
here is meaningless. One is making a bet that Coal will continue to be a
dominant fuel for us for the next 10-20 years in face of dropping cost of solar
and other energy sources such as Natural gas. In addition there is also the
headwind of climate change regulations and drop in prices globally. In short I don’t
know enough to predict the cash flow and hence the idea is a pass for me. If you plan to buy or hold it, you need to answer the above questions with a high degree of confidence.Stocks discussed in this post are for educational purpose only and not recommendations to buy or sell. Please contact a certified investment adviser for your investment decisions. Please read disclaimer towards the end of blog.
I am amused about what you mentioned on discount rate - Usually this depends on the riskiness of an asset.
ReplyDeleteI have heard value investors harping about how discount rate has nothing to do with risk and you cannot reduce risk by increasing discount rate. At the same time, they are fine with increasing margin of safety.
Both are tools to deal with uncertain cash flows. You can use either, but not together to cause double counting. Both discount rate and margin of safety can be arbitrary and can do the same job quite satisfactorily.
What do you have to say about this? This is not a criticism, just like to hear your thoughts. You can choose to ignore this message if you feel that way.
Hi anon
ReplyDeletediscount rate has been discussed indefinitely by academics and I was just sharing the theory.
personally I just use a standard hurdle rate of 10-11% and then account for riskiness via cash flow estimation. in the example show I used a discount rate of around 10% ..
MOS and discount rates are ofcourse arbitrary to that extent and it comes to the main point of the post - academics try to run DCF model without trying to dig deep into a biz. they want some math which works in all cases. I think that's asinine ..you cannot value biz without understanding it well and no MOS or discount rate can help
as an example, applying any discount rate or MOS to MTNL would not have helped in early 2000 without understanding what was happening in the telecom space.
Thank you for your response on discount rate, Rohit
ReplyDeleteSo , In analysing a company first comes the business and management then numbers can come afterwards
ReplyDeleteThanks for a very simplified way of explaining over dependence on PE. I am a relatively beginner but from my experiences believe that the real screener test is Quality of business and Quality of Management. If the management is adept and agile tuned to current global conditions it can manage business troughs and steer the ship accordingly. At the risk of sounding naive, would like to make a contrarian statement. Current business climates are evolving rapidly as compared to businesses a decades ago. In this conditions forecasting cash flow is like crystal gazing. Hence if we just use the Quality of Management and Quality of business filter almost surely you can go wrong. Valuation principles will help squeeze out a little extra points. In my quest to learn appreciate your comments.
ReplyDelete"If you put all these cash flows together and discount it at around 10%, the final DCF value comes to about 1.5X purchase price. In other words, the asset is generating an IRR of 12%."
ReplyDeleteI am sorry, didn't understand the calculation.
Thanks Rohit for another great post.
ReplyDeleteYes, I can see your changing attitude towards PE factor over the years..:-)
Facor Alloys..what an example!! I still hold it despite being reduced to ashes..:-) It reminds me to "act fast" when a mistake has been made or vice versa.
On another note, it is always exciting to have anon participate in these discussions. Great to see fellow longtime readers of your blog.
Vikas
Dear Rohit,
ReplyDeleteA fantastic write up on P/E. What could otherwise go as a discussion for hours, you have covered in a 10 minutes write up. It is quite unfortunate that the investing community has not looked beyond P/E for a valuation tool.
The wrongly held opinion... Low P/E, Cheap company to buy as it is under valued, High P/E, Avoid as it is over valued. P/E is a concept that has been sold too far in these years, while others like DCF is known to hand full of investors (I don't mean traders in guise of investors!)
Buying at a low P/E could even lead to complete wash of invested capital. I too had taken a hit in Facor Alloys. In the initial years, i sincerely made this mistake to look at only low P/E stocks. It was long before i learned that something was wrong and I was not making money. But before the recognition came it was too late.
Keep your good work with more such though provoking articles.
Venkatesh.
Rohit
ReplyDeleteKudos to you sir. I am a big fan of your writings and learned a lot from you. Infact I have started writing my own blog. I hope to have you visit that sometime. I am intentionally not mentioning my blog name here as I would like the blog to become successful one day and then you yourself will give a visit and that will be a proud moment for me to have you who is my inspiration.
Thanks
Sachin