Tuesday, January 6, 2015

Valuation thoughts on Sembcorp Industries - Part 1 (DCF Analysis)

Oil related companies make the headlines again as the trio of Keppel, SCI and SMM saw heavy decline with just less than a week of trading in the new year. Since this is now a hot topic and I am vested in SCI, I probably had more interest in it than the other two.

I've blogged about SCI related material a couple of times in the past and you can view them here:

Sembcorp Industries - Updates from Investor Relations
Sembcorp Industries - Segmentation Weakness in Utilities
Sembcorp Industries - Q1 FY14 Results

In this post, I will present my valuation on SCI which I have played with the template I have at hand during my lunch hour today. I wanted to split the segments using different valuation method but found that it's probably too time consuming at the moment. Furthermore, I am not too sure on how I can best value SMM mostly based on project orders and revenue recognition timeliness projection can be very tricky. Thus, I have reverted back to using the DCF methodology for all the segments.

I will divide them into two parts of the DCF methodology. The first is the EBITDA multiple and the second by using the Gordon Growth Model to value its "intrinsic value". As you probably know, there are many assumptions used in this methodology, so I will explain which number I have used my calculation.

Assumptions Overview

The goal is to come up with a final unlevered free cash flow figure for a 5 year period.

To start, I begin with the EBIT figure for 2013 which I am able to obtain via the financial statement. I assumed a forward growth rate of -1% for the first year and 5% for the next subsequent years. Q314 YTD figure is out and pro-rating them to an annualized basis would give me a rough indication of negative growth rate for 2014. Hopefully, we'll see better results in the coming years.

I come up with an estimated provision taxes of 17% corporate rate constantly throughout for the years and this figure is conservative since the company usually has deferred tax provision to offset.

I added back the depreciation which is in line with what they usually has for the normal years. This is also the biggest item that is impacting the operating cash flow other than the changes to the working capital, which can go either way.

I tried to put in more CAPEX figures which is way more than what analysts out there are predicting. The assumption is the company is still trying to grow and the growth in CAPEX is in tandem with the growth figure mentioned earlier.

With that, we have our unlevered free cash flow for the next 5 years.

1.) DCF Analysis (EBITDA Multiple Method)

The Weighted Average Cost of Capital (WACC), which is the discount factor used here is 10% and a sensitivity analysis between -1 and +1 is added for further reference.

The EBITDA multiple used is 11x and again a sensitivity analysis between -2 and +2 is added for further reference.

Based on the above assumptions, we compute the Enterprise Value using the discounted cash flow method. Next, we deduct the $4,513 debt borrowings which the company has to arrive at the Equity Value. We divide them by the outstanding shares of 1,784 million as of 17 Dec 2014 and we get an intrinsic value of $3.30.

You might wonder how could my calculated intrinsic value be so low as compared to those analysts out there and I will explain this later in my conclusion. For now, just remember that it's all assumptions.

DCF Analysis (2013 - 2018) - EBITDA Multiple Method

2.) DCF Analysis (Perpetuity Gordon Growth Model)

The same assumption is used with regard to the WACC (discount factor).

The only difference here is finding the terminal growth rate which can substantiate the value of this model.

The problem with this model is that the denominator is noded as k-g, which means the growth cannot be larger than your discount factor because if it does, then this will become the best investment in the world that everyone will invest. This is akin to asking why the trees cannot be taller than the cloud.

Anyway, I've decided on my terminal growth rate at 7% with a sensitivity variation of -1 and +1.

It appears that this method isn't the best of valuing SCI because of the much lower equity value because of the increasing debt they have on their balance sheet.

DCF Analysis (2013 - 2018) - Perpetuity Gordon Growth Method


If there is one thing I note something new about SCI, it is that their debt borrowings are increasing, almost tripling that of last year. I don't know why I did not notice this in the past. I am not too sure why they would want to start leveraging too much but my feeling is because they are ramping up their CAPEX aggressively over the coming years and they need the money to do that.

My intrinsic value also came up much shorter than most analysts' projection. This is because I have put in aggressive CAPEX over the next 5 years which brings down the FCF figure. Other analysts put in almost half the amount and as a result this boosted the intrinsic value of the company.

We already know that other risks involving higher competition for its utilities segment and the falling margin for its marine segment means that the growth we are projecting may be over optimistic. The key is probably how they fast the overseas segment can step in to cover this drop for the local market.
Companies such as SMM whose revenue depends on project order execution are hardest to value because of the irregularity of the timeliness of revenue recognition and lumpy order execution. On bad days, there could also be cancellation of such orders or over-accruing for such costs. I think this is harder than projecting property companies because for properties at least we know their schedule up to the TOP. Nevertheless, I'll see if there are better ways in valuing project order companies like SMM.

vested with 9 lots of SCI as of writing. Just playing around the spreadsheet while awaiting for a good time to average down SCI. Anyone has taken the bite for SCI in the new year already??


  1. splendid analysis. this is better than your previous work. we should all go study masters if the improvement is so dramatic.

    one thing i felt is ur comment on a 5% growth rate as well as a 1%growth fall. a 1% looks seriously small! a 5% looks an above average net profit.

    i felt its better you come up with a conservative free cash flow estimate, either a realistic cut, and then an optimistic free cash flow estimate.

    peraps u can also say abit more about how is the capex numbers derive even though you increase it. is it a % or something.

    the perpetual growth rate at 6,7,8% seems very high

    looks iike the intrinsic value is quite low.

    1. Hi Kyith

      Thanks for your input and comments. Appreciate it.

      The negative 1% growth for this year might be rather optimistic, they are currently on a negative 4% for the past 3 quarters and we have only one last quarter to make it up. I'm actually including the TPCIL consideration from 2015 onwards because I think this will be one of the main growth over the next few years. The management seems optimistic about this JV so I think it will contribute positively to the bottomline.

      We discussed privately about the CAPEX yesterday and agree that this figure could be overweight. Still, I think the Brazilian yards operations for SMM will require plenty of investment CAPEX for coming years to come.

      The perpetual thing doesnt really work, i don't want to put the terminal growth rate too high nor i want to discount them too low, as a result, I get weird figures for their intrinsic value. Would be interesting to see how others do for this.


  2. Hi B,

    Very chim to me, but it is amazing how much work you put in to generate this blog post. Clap Clap.

    Intrinsic value is a subjective word. This Warren Buffett causing a lot of problem with this word. Haha.

    Your low intrinsic value is worrying lots of SCI investors, who invested because the whole blogspace is advocating the same on SCI.

    New readers are beginning to follow the crowd and be happy. Next moment, another piece of news, they become worry.

    I am invested in SCI using gut feel! Walk one step…see one step..approach for the price….


    1. Hi Rolf

      Thanks for your kind comments.

      My estimated intrinsic value is based on a very high consistent CAPEX numbers which kind of pulled the cash flow figures down. This is in contrast to analysts reports whose target price is $5 or $6 as they use a much lower CAPEX figure with lower discounted rate at 7%.

      Still a good company and not too big of a concern, we just want to know where a good margin of safety is probably at :)

  3. May be someone set up a poll to count how many of us in SCI? Ha Ha!

    1. Hi Uncle CW

      I think there are many retail investors who owns them because they have been dropping massively.

      When 19 Jan comes, more retail investors will come in for SCI and Keppel. Akan datang :D

  4. My iv is also its nav....$3.02?
    Actually, the value is what you pay at certain price and it's dependance on CO price now.

    1. Hi Anonymous

      Care to share how you came up to $3.02 and what kind of assumptions do you use?

  5. Hi B,

    Your analysis is very impressive even though it is kind of complicated for the novice trader in me. Haha. I'm also looking for a good time to enter the market for SCI!

    1. Welcome to SCI Small Retail Investors Club soon!

      All are welcome. Next year AGM we can have free gathering there and personally witness how Mr. B questions the CEO.


    2. Hi Dividends101

      Thanks for your kind words.

      I remember you are quite heavy in your portfolio on O&G sector in SMM. They are somewhat correlated, I think it's a good idea to wait a while to see how things are going. Otherwise, the portfolio would be heavy and very much correlated to the oil industry.

    3. Hi Uncle CW

      Cannot la.

      This year AGM sure a lot of angry retail investors ask the management what is their plan and so on. I just sit back and take note :)

  6. Hi B,

    I'm not an expert in DCF and GGM valuation models, though one lesson I've learnt when building such models is garbage in, garbage out. As long as the assumptions are realistic and so are the inputs, I think the low IV should be correct.

    Fantastic analysis by the way!

    1. Hi SRSI

      Thanks for your kind words.

      Analysts like to put in some optimistic growth rate, which is why their target price is always very high and unattainable. They do this constantly with growth stocks like Osim and Super as well and many investors buy their story as a result.

      It's good to be on the conservative side of the fence in case anything gets wrong. The other way is to put in higher discounted value.

  7. HI B,

    I spend a bit of time looking at the formula of WACC and the components, has difficulty comprehending.

    Makes me want to learn more =) when I am free...

    1. Hi Sillyinvestor

      They are certainly interesting and forces us to think beyond. I think it's a good thing to know them even though we all know they are not the most accurate thing in the world.

      Looking forward to sparing with you on exchanging knowledge next time :)

  8. interesting read

    For SCI's Utilities business I use simple PE ratio to compare it against the industry, it makes around 20 cents per share so I value this unit as $3.00 per share alone (15 times earnings, comparable to other utilities such as CIT and Ausnet Services)

    The marine business is the difficult one.. but at $1.00++ I think its definitely cheap even though the business is cyclical

    If I want to be hyper conservative, SCI is worth at least 1 times book value (which is around $3.00) but given its high ROE of around 15%, logically the market should price it 1.5 times higher at around $4.50 as the base value (compared to REITS, which ROE are usually just 5-10% and priced near book value)

    1. Hi Felix

      Thanks for your input.

      If we just value SCI based on the earnings perspective, they are indeed looks undervalued compared to their peers.

      What I have used is a more conservative approach by drilling it down from the cash flow perspective, taking into account the huge CAPEX they would have to incur every year.

      The RoE is also certainly an interesting perspective because it masks the hidden leverage that the company is undertaking. The best way is to disect them to see if the increase is due to better margins or high leverage, though I highly suspect its due to the latter.

      I am not trying to pick down on SCI as I am rather heavily vested myself with 9 lots but I think it's good that we know a little bit more than what they actually are :)

  9. Oh ya also to add, i think managment had mentioned before that they are looking to spin off the utlities busines (maybe 3-5 years down) when it becomes matured (now still considered growth stage), so i think maybe after seperating the marine and utilities business will investors have a better time valuing SCI


    1. Hi Felix

      I hope they disposed off the marine business slowly. They used to be the cash flow generator but no longer now. They can be a drag if O&G doesn't improve.

  10. Hi B,

    A very detailed analysis. Appreciate your sharing.

    1. Thanks David for your kind comments as always :)

  11. Anyway see all of u guys in the coming agm haha, i currently hold 15 lots and not selling unless way above $5


  12. Make sure this AGM got good food!!! LOL. U lost paper value, but at least still eat back good food!

  13. Hi B,

    Interesting exercise here, and I was very interested in all the numbers being presented. It's good practice to run a DCF analysis and also to use an EV/EBITDA rule of thumb analysis. A few questions though:-

    1) If you start off with EBIT, shouldn't you deduct the finance costs as well before you arrive at profit before tax, and then apply the tax rate? Since SCI has so much net debt, the EBT should be quite affected by the interest costs. Won't this then further lower the value of SCI?

    2) How did you derive the higher capex over the years? Reading your post on maintenance versus expansionary capex was helpful, but I am curious to find out the basis for the assumption of higher capex. Granted, the company is taking on more debt but it does not necessarily mean there's going to be higher capex. The debt could be a result of having to cope with higher WC requirements.

    3) Why is the EBITDA multiple 11x? Is this in comparison to other competitors which are trading at similar EV/EBITDA valuations?

    4) SCI is itself made up of SCI (Utilities) and its stake in SCM. Would it be possible to value just the utilities portion by taking the segment EBIT, less finance costs and taxes, and to find unlevered FCF for the utilities division? You can then add on the SCM stake at market price to get the total equity value of the Company.

    5) The DCF uses g of 7% which I view as being rather aggressive. A more common g would be anywhere between 2% to 5%, with 5% being the higher limit.

    6) Why has SCI's debt burden increased so much over the years? Could this be a concern for a shareholder? What is their cost of debt?

    7) By using the method which produces the "lower" valuation, I do not necessarily see it as "wrong". The problem here is that many investors are anchored to the analysts' target prices and are thus unwilling to accept anything lower as a possible true value for SCI. The evidence shows that profitability is going to suffer for FY 2014 and net debt is also rising, so shouldn't the Company be valued more conservatively?

    8) It would also be good to have a certain % margin of safety to calculated valuation to ensure that one has a buffer in case things go wrong. You could use a smaller discount of 10-15% since SCI is a blue-chip and is run by competent management.

    9) I am very interested in the Greenwald method and the EPV as this would imply zero growth, and is a lot easier than estimating growth and using a "g". The EPV would represent a kind of "baseline" for one to benchmark the company to and you can also see what kind of growth rate the market is implying by observing the premium (of the market price) over EPV.


    1. Hi MW

      Thanks for your length questions about DCF.

      It definitely allows everyone, including myself to learn about the assumptions we have input into the model. I will try my best to attempt your questions so here we go.

      1.) Good Point there. On hindsight, I could have just used the net profit instead of the EBIT as it appears that I am discounting the interest portion which could be substantial if the company has quite a high borrowings.

      2.) The capex I have keyed into the model contained both maintenance and growth capex, which I have adjusted year on year in line with the profitability growth for each year, which is -1% for the first year and 5% for the next 4 years.

      The rationale here is to ensure that they rise in tandem with the profitability since higher profit comes from higher capex costs which is required to maintain and grow them.

      3.) The EV/EBITDA for the stock is currently trading at around 9x while competitors outside is trading at 15x. 11x is currently my basis for the intrinsic value though the sensitivity analysis there provides an indication of how much should the value be should this be changed.

      4.) It's possible actually, though I don't quite see the point at the moment since there are no better ways to value the SMM portion (at least I can't do it).

      5.) Yes agree. I was just trying to play around with the g in this case since the GGM model doesnt work in this instance. In my other post, I have reduced this to around 3-4% on average.

      Since g is a very subjective matter, I would think the reverse DCF would be an easier method to value the stock.

      6.) The risk has increased with more debt on the balance sheet but they will use it to grow the business and jv. Not necessarily a bad deal if they can meet the required rate of return on the investment so still finger cross for now.

      7.) For a company like SCI, I think it has proven over the years to be a great company under a very good hands of the management. Valuing them conservatively gives us more margin of safety but I think it skewed the true intrinsic value that the company can offer. As always, my strategy is to ensure that the growth is conservative and not exaggerated. The entry point for individual investors is another point for another story.

      8.) My way of computing the WACC is usually the risk free rate + risk premium, which is usually at around 10% for a stable blue chip companies proven over the years. Again, my reasoning of a much lower discount factor would be that it does not give justice to what the company is truly valued in.

      How much investor is willing to pay for the stock based on the margin of safety is another point altogether.

      9.) I've not tried the EPV method, but another way we can do this quickly is by doing the reverse DCF, using the same discount rate I am applying right now. We'll then see if the g is worth what they are.

      Thanks for all your comments despite being a seasoned investor. Always looking to learn more and willing to accept any feedback.

    2. Hi B,

      Nice post there and I believed many readers enjoyed it as well :-)

      With regards to point 1 raised above about EBIT, I think what is presented in the post itself is correct. Essentially, B has removed the effects of debt via deducting it from the overall valuation. By including the finance cost in the calculation and again removing the net debt would mean there is 'double-counting' for the company's debt situation. I guess to present a fair valuation, either you can include finance costs or deduct the net debt but not both.

      Regarding point 2, I also find capex a problem when introducing it to our own model. Some of the past capex may be used for future profit growth and it may not mean that more or the same amount of capex will be used again in future. I applaud B for bravely doing and showing them in the post :)

      Regarding point 6, the debt increased due to consolidation of TPCIL and Marine's new yard in Brazil. This is the part I don't like about Sembcorp too. Recurrent earnings from the Utilities business may justify higher than average debt but not sure about the Marine's side. Let's hope that the debt situation will not worsen anymore.

      Thanks B for your nice, comprehensive post about Sembcorp's valuation and also contributing great insights in my post on the analysis of the same company!