Present Value of Future Cash Flows

[Ves]

Hi Craft / McLovin,

I have actually started to take the plunge and delve into some NPV calculations for Free Cash Flow lately.

I am happy to play around the the ROE / ROIC, the retention rate, working capital requirements etc. I usually try to use a very conservative estimate of these, to come up with a basic growth rate that may vary slightly over a 5-10 year period. I then discount these basic over a range of discount rates. I take it when you say "sensitivity analysis" you mean that various scenarios (say they make an unexpected loss in a year) should be run using fluctuations of these figures? Is there a systematic way of doing this?

It is the "perpuity" part of the valuation that has me the most uncomfortable. It is obviously the biggest component of the NPV. Generally I set this growth rate to match inflation. The problem I have, even with this, is that no company lasts forever. How do you account for this? Are there any alterations or adjustments that can be made? I guess the whole learning how to forecast for myself, and realising that you will never be accurate to the nth degree, and sometimes not at all, is quite daunting for a relative beginner.

 

[odds-on]

Good article by Geoff on gurufocus.

http://www.gurufocus.com/news/175393/free-cash-flow-isnt-everything

 

[Ves]

Good article by Geoff on gurufocus.

http://www.gurufocus.com/news/175393/free-cash-flow-isnt-everything

Cheers Oddson - I read the first part and seems interesting. I will read the rest of it over the weekend when my head feels a bit clearer.

 

[McLovin]

It is the "perpuity" part of the valuation that has me the most uncomfortable. It is obviously the biggest component of the NPV. Generally I set this growth rate to match inflation. The problem I have, even with this, is that no company lasts forever. How do you account for this? Are there any alterations or adjustments that can be made? I guess the whole learning how to forecast for myself, and realising that you will never be accurate to the nth degree, and sometimes not at all, is quite daunting for a relative beginner.

Sorry, I missed this. I usually use 0 as my growth rate for the perpetuity. Sometimes, I'll include some growth 1-2%, never more than 3% though and usually it's when assessing a company like COH or CSL which has a global market for its products. I wouldn't include any growth in something like TGA, for example.

I take it when you say "sensitivity analysis" you mean that various scenarios (say they make an unexpected loss in a year) should be run using fluctuations of these figures? Is there a systematic way of doing this?

Yeah, you can learn a lot about a company by running a DCF under various scenarios I often compare across an industry, especially with things like margins and working capital which tend to mean revert (one of my old finance lecturers used to drum in that the benefit of ratio analysis was in comparing across time and sector, the actual calculation of the ratio was of secondary importance as long as you were consistent in applying it). And then see how that reversion would effect DCF. I guess the important thing to remember is that you're not trying to come up with an answer, rather a guide. Like any way of attempting to estimate IV really. I don't make mine overly complex or systematic but I do often look for discrepencies in ratios as a good starting point for sensitivity analysis.

Hope that helps.

 

[craft]

Hi Craft / McLovin,

I have actually started to take the plunge and delve into some NPV calculations for Free Cash Flow lately.

I am happy to play around the the ROE / ROIC, the retention rate, working capital requirements etc. I usually try to use a very conservative estimate of these, to come up with a basic growth rate that may vary slightly over a 5-10 year period. I then discount these basic over a range of discount rates. I take it when you say "sensitivity analysis" you mean that various scenarios (say they make an unexpected loss in a year) should be run using fluctuations of these figures? Is there a systematic way of doing this?

It is the "perpuity" part of the valuation that has me the most uncomfortable. It is obviously the biggest component of the NPV. Generally I set this growth rate to match inflation. The problem I have, even with this, is that no company lasts forever. How do you account for this? Are there any alterations or adjustments that can be made? I guess the whole learning how to forecast for myself, and realising that you will never be accurate to the nth degree, and sometimes not at all, is quite daunting for a relative beginner.

Does post 21 or 40 in this thread help at all with your thinking?

You have to be incredibly careful with terminal values. You only have to change the inputs by small margins to produce any number you like. Consider that if future perpetual growth is 5% and future cost of capital is 9% then the terminal value multiple is 25. Changing those two variables by just one 1% can produce multiples from 16 to 50 times.

I prefer to discount only the cash flow horizon I can have some certainty about and then calculate a terminal value based on the replacement cost of tangible assets and possible some multiple of that if the business has a true sustainable competitive advantage.

 

[Ves]

Does post 21 or 40 in this thread help at all with your thinking?

You have to be incredibly careful with terminal values. You only have to change the inputs by small margins to produce any number you like. Consider that if future perpetual growth is 5% and future cost of capital is 9% then the terminal value multiple is 25. Changing those two variables by just one 1% can produce multiples from 16 to 50 times.

I prefer to discount only the cash flow horizon I can have some certainty about and then calculate a terminal value based on the replacement cost of tangible assets and possible some multiple of that if the business has a true sustainable competitive advantage.

Thanks to you both - it does give me some ideas.

I think the trouble I am having is relates to the practice, rather than the theory. I really enjoy reading both of your posts and a few others on this subject and related topics. Theoretically most of it makes sense in my mind. I have been trying to focus on the first two things on your list: earnings risk and default risk. Playing around with a few ratios and looking at business models / financial structures hasn't been so bad. You can usually discard the dogs pretty easy from just doing some work (or even a glance) at the balance sheet. Most companies do not fit in with what I am trying to achieve in this respect. I wouldn't bother valuing them. It seems far easier to find a company I wouldn't invest in. Probably a good thing for "risk management."

However I get confused when I have to start coming up with my own forecasts (or the numbers going forward). Perhaps I need to re-read Greenwald. I sometimes watch 5-10 of Sky Business when eating my dinner. Roger Montgomery was on the other night saying how his Skaffold system relies on broker forecasts and "conservative past assumptions" and you should buy when price is lower than both of these. I feel frustrated when I hear things like this, because the point is to do your own forecasting (not rely on brokers who always get it wrong by large margins). Relying on the past without any thought for how it will change in the future is also lunacy. This guy is supposed to be a professional, yet he seems to lack any clue as the theory behind forecasting, even when compared to a beginner.

If anything, the opportunity is there, becoming good (not perfect) at valuation puts you far ahead of the pack if coupled with in-depth analysis and an intimate understanding of the company.

I will try to explain what I try to include in my models:

Once I have attempted to make a judgment on the earnings risk or the default risk of the company and I am satisfied I started looking more closely at the underlying cash flow. I like trying to compare it to NPAT to see if I can discern the differences.

I then try to calculate a normalised version of Free Cash Flow to Firm for as many years as I can find using the published financials.

To do this I look for how much it costs to fund working capital and capex as a % of the cash flow or NPAT depending on what I feel is most relevant (ie. what Damodaran calls the reinvestment rate.

The left over after this reinvestment rate is the FCFF. The firm can either pay this out as a dividend or use it to fund future expansion (retention rate).

I then forecast for the next, say five years or as far as I am comfortable with, the growth rate of the FCFF using a more conservative than historical calculation using the retention and reinvestment rates compared to ROE or ROIC.

The terminal value is added to this. I will play around with some of the suggestions that you both mentioned, to see what disparities are thrown up!

I discount all of these cash flows and terminal back over a wide range of discount factors to compare. I usually feel most comfortable with the results between 12-16%.

My way of doing it seems to be pretty big picture.

I more than understand that it is mostly an 'airy fairy' scenario and in the end it is your gut feel stemming from your analysis and intuition vs the market's perception. I often wonder if it is just better to chase businesses that I find have the least earnings and default risk and substitute an earnings multiple (or yield) or something similar. Doing industry wide comparisons etc.

I don't really know what else you can say to me, or guide me with! I feel like I could be doing much, much better.

 

[Ves]

Probably another thing that I also need to focus on and did not mention:

Incremental capital in the respect of it is fine to have a high ROE / ROIC, but if you can no longer re-invest capital then growth must slow down or become non-existent going forward (other than perhaps inflation or GDP growth). Finding companies that can utilise high amounts of incremental capital going forward is important to me.

edit:

Example.

Company A - Cashflow $1,000 - 50% ROIC - But can only re-invest 10% of this profit. Growth is only 5% or $50.

Company B - Cash flow $1,000 - 10% ROIC - But can re-invest 50% of this profit. Growth is also 5% or $50.

 

[skc]

Probably another thing that I also need to focus on and did not mention:

Incremental capital in the respect of it is fine to have a high ROE / ROIC, but if you can no longer re-invest capital then growth must slow down or become non-existent going forward (other than perhaps inflation or GDP growth). Finding companies that can utilise high amounts of incremental capital going forward is important to me.

edit:

Example.

Company A - Cashflow $1,000 - 50% ROIC - But can only re-invest 10% of this profit. Growth is only 5% or $50.

Company B - Cash flow $1,000 - 10% ROIC - But can re-invest 50% of this profit. Growth is also 5% or $50.

Ves - have you tried say pick a company and look at their reports from 2001 to 2005, then try to come up with some forecasts / DCF / valuation for 2006 to 2012 (or 2008 if you want to avoid the GFC)? Preferrably companies that you understand (so you can assess the industry) but don't really know much about (so you are not biased). You can then check your answers against the real figures.

Sounds like a fun exercise and I might just try it myself.

 

[Ves]

Ves - have you tried say pick a company and look at their reports from 2001 to 2005, then try to come up with some forecasts / DCF / valuation for 2006 to 2012 (or 2008 if you want to avoid the GFC)? Preferrably companies that you understand (so you can assess the industry) but don't really know much about (so you are not biased). You can then check your answers against the real figures.

Sounds like a fun exercise and I might just try it myself.

Actually did think of that, but have never tried. I like the disclaimer about not knowing much about the company, but I wonder how you would over-come this if you were to practice a full-analysis on it? I guess you could use the search feature for announcements on ASX.com and limit yourself to 2005 and before, reading some of their presentations etc. Not quite the same, as you couldn't use other sources, but what do you think?

 

[craft]

Probably another thing that I also need to focus on and did not mention:

Incremental capital in the respect of it is fine to have a high ROE / ROIC, but if you can no longer re-invest capital then growth must slow down or become non-existent going forward (other than perhaps inflation or GDP growth). Finding companies that can utilise high amounts of incremental capital going forward is important to me.

edit:

Example.

Company A - Cashflow $1,000 - 50% ROIC - But can only re-invest 10% of this profit. Growth is only 5% or $50.

Company B - Cash flow $1,000 - 10% ROIC - But can re-invest 50% of this profit. Growth is also 5% or $50.

Company A can fund a dividend of $900 and fund 5% growth

Company B can fund a dividend of $500 and fund 5% growth

Dividend discount model with a 15% RR

Company A value = 9000- Investment capital required 2,000 (value/Invested capital = 4.5)
Company B value = 5000- Investment capital required 10,000 (value/Invested capital = .5)

 

[Ves]

How did you calculate the "investment capital required" part down the bottom? Agree with the dividend, I must admit I overlooked this.

Also will note that if I had have used ROE instead of ROIC, the dividend would depend on debt repayments (if any). That's why I like to use ROIC.

 

[Ves]

Never mind - Profit / ROIC

 

[craft]

How did you calculate the "investment capital required" part down the bottom? Agree with the dividend, I must admit I overlooked this.

Also will note that if I had have used ROE instead of ROIC, the dividend would depend on debt repayments (if any). That's why I like to use ROIC.

50% ROIC / 1000 = 2000
10% ROIC / 1000 = 10000

 

[Ves]

Apologies for not being very sharp tonight!

 

[craft]

I don't really know what else you can say to me, or guide me with! I feel like I could be doing much, much better.

I think you are teaching yourself more than 99.9% of people do and that will give you an advantage.

The journey goes from complexity to simplicity with lots of light bulb moments along the way that give you the knowledge of when and where to utilise simple.

 

[Ves]

I will practice a few older ones like SKC suggested. It couldn't hurt. I think the main thing is if I keep enjoying it and maintain patience with myself (and importantly the market) I will get somewhere eventually.


edit: Maybe I will also play around with some "blue chips" like the Banks. Run some simulations using some American style figures or European even that have been reported since the GFC. Since our banks have gotten out of it much better and may see some pain to come. See what the possibilities throw out in terms of valuation.

 

[craft]

There are 4 cards E, 4, K, 7. Each card has a letter on one side and a number on the other.

If I tell you that an E card has a 4 on the other side, which cards would you like to turn over to verify I was telling the truth?

 

[Ves]

There are 4 cards E, 4, K, 7. Each card has a letter on one side and a number on the other.

If I tell you that card E has a 4 on the other side, which cards would you like to turn over to verify I was telling the truth?

I don't quite understand, but if you turned E over and it had a 4 on it, that does not prove or disprove the theory of letter on one side, number on the other. You need to turn them all over. I take it that this is the moral of the question.

 

[craft]

I don't quite understand, but if you turned E over and it had a 4 on it, that does not prove or disprove the theory of letter on one side, number on the other. You need to turn them all over. I take it that this is the moral of the question.

No I'm just saying that E cards should hvae 4 on the other side. Which cards? You don't need to turn them all over.

 

[Ves]

I guess another way of looking at it is to ignore your comment about the E and the 4. And turn over the other two cards, K and 7.