TGS has been written up once on VIC by finn520 in 2010; this writeup will focus on the core situation of the business today and why I think it is still a good long in this environment. 

Description:
TGS-Nopec is the leading independent multi-client geo-seismic surveying company in the world. For those of us who are not familiar with geo-seismic surveying, TGS’ core business is shooting and processing 2D and 3D ocean-floor maps, which are collected in a data library and then sold on average to between 3 and 20 oil companies who use these maps to make CAPEX decisions on where to drill for off-shore oil. The company based in Oslo, Norway but also has an HQ in Texas, USA. It is listed on the Oslo Stock Exchange under ticker TGS. It also trades under an ADR in NY under TGSGY.

Brief History
TGS-Nopec was formed in 1998 through the merger between TGS, a US business who had a strong focus on multi-client seismic data in the Gulf of Mexico and Nopec, a Norwegian listed company with a strong focus on the same sort of data but in the North Sea. In the years between 1998 and 2013, TGS-Nopec continued growing mostly organically but made a few technology acquisitions to improve their processing and shooting capabilities. To give an idea of the expanded scope in TGS-Nopec, total investments into shooting seismic data in 1998 were USD 26 million; in 2012, the figure was USD 496 million i.e. 19x times more.  Similarly, the un-amortized book value of the collected multi-client data library TGS-Nopec has built had a value of USD 27 million at the end of year 1998; at the end of Q1 2013, the figure was USD 707 million i.e. 26x times. In total, TGS-Nopec today has a global market share by revenues of approximately 22% of the entire multi-client surveying market in a market that has grown very considerably in the last 10 years (source: Pareto Research).

Industry Overview
To look at a company in solitary confinement is almost always a bad idea and TGS-Nopec is no exception. The seismic exploration sector is dominated by publicly listed companies. Lucky for us, quite a few of these are from Europe – Scandanavia in particular, so we had a chance to meet more than one or two of them. The two big distinctions in this sector are:
(1) between operators who do “contract surveying” i.e. they have a customer like Exxon who pays them to shoot a set of maps exclusively for them, and operators who do “multi-client” surveying i.e. they get some pre-funding often from customers like Exxon, but almost always retain the rights to the maps they shoot, which they will then sell to numerous customers.
(2) between operators who own their own ships, and those that charter their ships.
Out of the competitor set, the largest other players are Petroleum Geo-Services (PGS), CGG Veritas, Western Geico. Several smaller players also exist in Polarcus, Electromagnetic Geoservices, and Specturm.
Among the large players only TGS-Nopec is focused exclusively on multi-client surveying. PGS, CGG and Western Geico have historically all done contract business. Moreover, TGS Nopec is the only large competitor who does not own ships. The only player who is like TGS-Nopec without owning ships and focused on multi-client work is Spectrum, which has a market cap 1/10th of TGS. The importance of both of these aspects is explained in the next section.
One other aspect of the competitive situation that is important to understand is that because of the extremely high cost of shooting a seismic survey, it is very unusual that the different competitors will choose to shoot the same exact area of exploration. This is because operators are looking for a good return on their investment and if the potential revenues would be half – as another comparable product would be in the market if two competitors shoot the same area– then the returns instantly become less attractive. Hence, more frequently, operators either focus on their niches – for TGS this has included the North Sea and Gulf of Mexico, recently also off of the coast of Africa. Alternatively, players have entered into JVs of splitting the cost of shooting and then dividing revenues. A recent example of this is the work that TGS has done in a JV with Fugro and PGS in the Gulf of Mexico last year.
As a result, customers – the oil companies – when they want the seismic data of a specific area usually have only one or two operators they could choose who will also fit the technical specifications they require for a shoot. This means that their ability to escalate competition is somewhat limited. Put in a different way, quoting a management team who shall remain nameless, the industry works more often based on “quality partnerships” rather than cut throat competition.

Why is TGS an attractive business?
Besides the relatively benign competitive environment that clearly helps, the reason that TGS-Nopec delivers the consistently good results we will get to later has to do with its business model.
Foremost, there is a significant difference between multi-client surveying and contract surveying. To me it is self-explanatory that when you incur the same cost shooting maps of the ocean floor and then sell it to many companies, it is inherently more efficient than selling it to just one company. As evidence of this, I point to PGS, the aforementioned competitor, who does both multi-client and contract work and makes ca. 25%-30% EBIT margin in their multi-client business and much lower margins in their contract division. An argument could also be made that if your focus is on your own shooting technique and processing ability then you have more of a chance to build a special capability better than your peers over time who are just taking orders. I am less convinced of this second point.
A real second reason however seems to be the ship ownership structure. The argument for owning your own ships has always been that in times when the market tightens, rates for chartering ships will be higher and it is even possible for a player like TGS to not have access to enough ships to do the work that they want. In practice, over the last 15 years, a market which certainly included a very healthy cycle for exploration work, this has essentially never happened. Whenever the market started “getting hotter”, there were always new shipping players putting new surveying capacity into the market at reasonable rates. As a result, the hypothesized margin crunch has never happened so far. To be very exact, chartering rates have increased in some of those times but not so much as to destroy margins. If you look at TGS-Nopec’s EBIT margin between 1998-2012, it has always traded at a very healthy 30%-50% with an average of 42%. The lowest year’s 31% in 2003 and 39% in 2011 were not necessarily in years with especially booming exploration activity.

On the reverse side, however, it is clearly the case (verified from our discussion with PGS, CGG Veritas) that those operators who have owned their ships have had to keep running their ships for utilization on less profitable work and/or suffer lower utilization in tougher exploration markets. This negative effect does not even consider the general higher capital intensity of having ships on your balance sheet.

One more fact about owning ships vs not owning ships. When you own ships, you have to invest into the R&D of having the best surveying technologies available. For an operator like PGS for example, this means that ca. 5% of sales is spent annually on R&D. This compares to TGS-Nopec who spends < 1% of their sales on R&D. Nevertheless, as we know in practice, TGS has had access to all the latest technologies because they charter the latest ships from companies like Polarcus and in fact PGS themselves. As TGS with its scale has become one of the largest customers of PGS itself, it is unlikely that PGS does not let TGS charter its ships. And remember, we are operating in an environment of “quality partnerships”. So, TGS-Nopec also has to spend less on R&D in its setup.

The obvious question then becomes why not all seismic exploration companies ditch their ownership of ships. I don’t know the answer, but it could have to do with operators believing in continuing to have their own supposedly superior technologies (PGS touts its Geostreamer, CGG has its Broadseis, but from customer interviews, it seems that all the top technologies are relatively comparable). It could also be that inherently contract work has been more stable as it is locked in before it begins and some operators prefer this certainty even at the expense of poorer returns on capital.
What I do know is the historically very consistent financial performance that TGS-Nope has shown. I have calculated the numbers myself based directly on the annual reports, but results should be roughly comparable based on the data of the popular databases.  Between 1998 and 2012, TGS-Nopec has grown revenues, EBIT, and EPS by 20%, 19%, and 19% p.a. respectively over a 15 year period. Its EBIT margin has never dropped below 30%, which with a total capital employed base (I consider PPE , intangibles relating to the multi-client data library, and NWC) of 110% of revenues, meant a ROTCE of at least 27% before tax and ca. 20% after tax. On average over the 15 year time period, the EBIT margin has been 42%, and the after tax ROTCE 30% as I calculate. In 2012, the EBIT margin was 43% and the ROTCE slightly above 30% so we are close to a historical average.

All in all, I think I do not stretch when I say that this business has been an incredible compounder. The business model is simple, use very healthy cash-flows from operations to invest in more surveying activity, which is then sold to multiple customers for even more cash.

Current Situation
It seems that the stock markets has always been and continues to be of the belief that the demand for seismic data is closely dependent on the oil price and that this relationship holds at the month-by-month time-frame. In the last month as oil prices have come down from $110 to $95 due to fears of a China slow down and general economic malaise. As a result, the price of companies in seismic exploration, TGS-Nopec included, has come down ca. 15%.
I find this quite unjustified for two reasons. First, the reality in seismic exploration activity is that long term CAPEX decisions of oil majors do not change so quickly – at least not month by month. Yes, it would be true that in the case we have oil prices significantly below $80/barrel for the next 5 years there will probably be an effect on seismic data demand; but this is not the case now. I think the indications for a five year outlook is as likely to be up as down.
Second, the most direct relationship to TGS’ revenues for those that understand the business has always been the amount of investments that it has put into shooting 2D and 3D seismic data. This is because the very nature of the business is that TGS invests in shooting seismic maps followed by a rather short payback period where it monetizes the maps into revenues. On average, ca. 50% of revenues come within 1 year of the investment and ca. 80% coming within 3 years. Historically, TGS has gotten about USD 2 for every USD 1 it invests. In the last year, TGS-Nopec has invested more than it has ever before and it guides on being able to continue doing so for the foreseeable future. The management has had a history of under-promising and over-delivering, so that if they say that they are able to continue investing more in multi-client data for the foreseeable future, I do not see why we should believe that revenues will actually drop rather than increase over the next years.
To be very specific, in FY 2012, TGS-Nopec invested USD 393 million into shooting seismic data compared to USD 276 million in FY2011. For FY 2013, TGS has guided for investing an even more – USD 570 million, out of which they have already invested USD 126 million in Q1. I can go into the detail on which projects are already scheduled to be conducted for the remainder of the year, but the point should be that these are confirmed projects. The overall picture you should get is that from everything concrete we have seen and been told by the management, the company has had a healthy demand for continuing projects both in the past year and in the next year, which should be monetized over the next few years.

I have built a simple model which I have used to estimate on-going revenues. This is based on the simple assumptions based on the historical monetization patterns: the model assumes simply that investments for any one year will be monetized at 50% in year 1, 20% in year 2, 10% in year 3 and so on. The payback multiple for the entirety of revenues is 2x the invested amount, which is roughly consistent with historical patterns. As any model, there have been years where the model is a bit high and some years where the model has been a bit low, but overall, the correlation coefficient of the model has been 96%, which gives me quite a bit of confidence. According to the model, in FY 2012, TGS was to have revenues of ca. USD 839m. The actual figure was USD 932m. For FY 2013, due to the increased investment, the model predicts a revenue figure of USD 1115m.

If you want to think about it in a simpler way, without any models, you can look also at the book value of the data library; if more is invested (than is amortized), more exists to be monetized. TGS looks at this metric and it is discussed by them. I have used my model because if I model using the book value, the correlation coefficient (see above) is a very accurate 94%, but lower than 96%. Still, the main point is that this is a business where you usually know with some accuracy what you will get in revenues if you know exactly what you invest. For TGS, they clearly have invested more, not less, in FY 2012 and will do so in FY2013 than before, hence we should expect them to make more revenues and more profits, not less.

Historical Growth
As noted, the historical top line growth has been 19% per annum over the last 15 years. One thing I want to expand on is that this growth has not been just based on doing a larger scope of exploration, but also due to the technologies involved. The evolution in increased revenues is as much a function of gaining market share for TGS as it has been that the whole industry had moved to increasingly more sophisticated and more expensive mapping technologies. This evolved from 2D to 3D, to wide azimuth, and I believe will continue to evolve into more sophisticated methodologies for accurately mapping the ocean floor. This means that the risk of running out of work and an immediate glass ceiling on revenue growth is not quite accurate. I believe that TGS-Nopec does have a continued ability to increase revenues.

Valuation
See above. I will let you do the valuation yourself based on cash based or other non-traditional metrics, but I would make the clear argument that at a PER multiple of 11 times and an EV/EBIT of 6 times for such a good business with a proven track record of compounding well intrinsically (high ROTCE and growth reinvestment), this is very cheap. PER metrics look slightly more expensive due to TGS-Nopec having net cash, but still is also cheap.

What is the upside/fair value?
As a value investor, I do not like to pay for growth so the first thing I look at is the current earnings power of the business. Using the FY2012 EV/EBIT 6.4x as an indication, I would say based on the earnings that are there already, I would be buying the business for a fair price not assuming any growth. (For a high ROTCE business with ability to return cash but NO GROWTH, I think 7.0x-8.0x EV/EBIT is a fair multiple) But in this case of TGS-Nopec, there clearly is growth and there clearly is a value of being able to compound at an ROTCE after tax of 30%.

How to value this can be done in many ways. You can run your own valuation models to value this growing business with high ROTCE, but I am going to just use a very simple model as an illustration: I am going to attribute a fair multiple, in this case 7x EV/EBIT, on earnings 5 years from now and discount the figure back to today at a discount rate of 8% p.a. This is just a way to ballpark the value of growth – as a value investor I do not pay for this, but I do think the assumptions I make are conservative. If I assume that TGS-Nopec would continue to growth its EBIT by 19% p.a. then in 5 years, its EBIT would be USD 959 million. Multiplied by 7x, its fair EV should be USD 6715m. Netting by net cash (I assume unchanged at USD 412 million), a fair market cap would be USD 7172m. Discounted by 8% p.a. for 5 years, the today’s fair value would be USD  4850m or USD 47.45 per share (assuming share count has remained unchanged at 102.2 million shares outstanding). This is > 50% more than today’s price.

Please note that this is a ballpark figure only to give an indication of the rough value of the growth. As you can see, I did not assume any dividend distributions. On the other hand I did not assume any additional external capital required to finance the growth (I think a business with an after tax ROTCE>30% should generate enough cash to finance a 20% growth).
Another way to look at the valuation would simply be that a business with these economic characteristics is worth more than 6.0x this year’s EBIT or 10.0x this year’s PER. If I had to guess, it should be northward of 10x EV/EBIT and 15x PER. But again, you should decide how much a business with a consistent historical 15 year track record of EBIT growth of 19% per annum and a ROTCE after tax of 30% and a decent management should be worth.  I think it is at least more than 50% above what it is being valued at in the market.

Potential Risks
Lower Oil Prices:
One risk we already mentioned is that if oil prices were to be lower than ca. USD 80/ton indefinitely AND expectations of oil majors would be that this would remain the case, then oil majors could change their CAPEX decisions to less off-shore as few such sources would be profitable in that case. This would of course dampen the demand for seismic data. This is not the case currently, but let us look at what happened in 2008-2009, which is probably the closest case to this that we saw in the last few decades.
If we look at the low in 2009, the amount of revenue for TGS-Nopec indeed dropped 15% for that year. But this immediately recovered in 2010 to 2008 peak levels. EBIT margin in 2009 remained equal at 44% compared to 2008. One thing that happened in 2009 was that helped TGS-Nopec was that the chartering of vessels also got cheaper and most of the work still remained, which became sold when markets recovered again and more work was done by the oil majors in exploration.
I would argue that although there would definitely be an effect if we had a severe recession, even in the 2008-2009 case, the drawdown was rather muted. The company kept generating cash and demand was postponed for the most part.

Competition:
Competition is certainly also a legitimate concern. There has been some movement from the competitors i.e. PGS, CGG Veritas, etc. to also do more multi-client work. This has not affected TGS’ numbers so far, but is something we should be mindful of. My belief is that a company that has been entrenched in the multi-client, no ships owned business model for a long time will have an advantage over any player trying to adopt some of its characteristics while maintaining their bread and butter contract work stream. In any case, the effect of this in the overall picture of rather friendly competition is also likely to be more muted. If this dynamic changed as well, then this would add my concerns. So far, as said we have not seen any negative parts in the numbers of TGS.

Conclusion
To summarize, I think TGS-Nopec is a great quality business, with a proven management team, who has increased EPS by 19% p.a. over the last 15 years. The reason they were able to do so is because TGS is inherently a business with high cash generation and the ability to reinvest at a high ROTCE. In the last 15 years the ROTCE calculated at an after-tax basis has been ca. 30%.
Currently the market seems to mark down the price of TGS-Nopec due to month-to-month movements in oil prices. I think this is not justified because oil majors do not change their CAPEX decisions so short term and in fact all concrete data points towards TGS-Nopec having been able to/will be able to invest much more in FY2013 and in FY2014 into shooting seismic data. If historical patterns are any indication, they are very likely to monetize all this new data in much higher revenues in the next 2 years.

At today’s valuation of 6.1x 2013A EV/EBIT, I think TGS-Nopec is very undervalued. If it could continue at its historical 20% revenue and 19% EPS growth rate and continue to do this from internally generated cash (which is possible because of its > 30% after tax ROTCE), the business’ fair value should be at least 50% more than what it is priced at today.

(1) Significant undervaluation of great quality compounding business with fundamental growth drivers
(2) The subsiding of short-term negative investor sentiments which I find mostly unjustified.