Tuesday, 7 August 2012

Roman candles and breached moats

Getting paid for failure

People don't like to dwell on failure.

However for an investor (particularly one with the ability to sell stocks short) being able to identify companies which fail can be just as rewarding as picking the ones which succeed. And the bigger they are, the harder they fall.

A particularly rewarding category are companies which Warren Buffet calls "Roman Candles", which burn brightly for a while but then explode in a shower of sparks. As he wrote in wrote in 2007:

Business history is filled with “Roman Candles,” companies whose moats proved illusory and were soon crossed."

Because they are perceived as having strong moats such companies often earn very rich valuations for a time (and are overwhelmingly acclaim from market analysts), until their weaknesses become apparent. This obviously creates ample opportunity to sell the stock short, and even after an initial fall there is often much more downside to come as investors' anchoring means they are slow to cut their losses.

As the saying goes - what is the definition of a stock that's fallen 90%? A stock that's fallen 80% and then halved again...

How can the light that burned so brightly... 

Here are some of the Roman Candle's I've come across in recent years. One thing I have noticed is that belief in a false moat is often compounded by management arrogance. Often management are caught flat-footed by a new competitor, precisely because they believe their moat is unbreachable.

Nokia: Nokia's moats were design chops and scale. Certainly for low-end feature phones their usability and interface were far superior to their competition. However their business model also had real scale advantages - because of their scale and manufacturing reach they could work at a 20%+ operating margin while most of the competition was at half that (for example spreading fixed costs like R&D across a larger number of devices means a lower cost per device). They also had enough scale to offer direct distribution in emerging markets such as India or China while competitors had to go through local intermediaries.

However for all the strengths of these moats (and they were strong - after five years of turmoil Nokia is still the second largest vendor of mobile phones by volume) they were breached by Apple and android handset vendors. Quite simply Nokia at first dismissed Apple's chances of success. Then they retreated behind their supposed moats and failed comprehensively to bring out competitive products not only for one but two, and then three generations of device (warning, these reviews are painful to read!). When they finally got it brilliantly right at the fourth time of asking it was far too late (and they immediately junked that platform in favour of Windows Phone).

MySpace: Being the world's biggest and baddest social network (and having a user base replete with trendy Generation Ys) should have given MySpace a big network effect and created a massive moat for its business. That's what Rupert Murdoch thought when he splashed out $580m for it in 2005. However in reality that moat was illusionary - users proved fickle and by pursuing short-term page views and ad revenue rather than playing the long game as Facebook did to build an API and a platform, that moat simply melted away.

TomTom: For a brief moment in the late-noughties "TomTom" was as synonymous for SatNav device as "Hoover" was for vacuum cleaners or "Zamboni" was for, well, Zamboni's. By being first to the market, producing great, easy to use devices and creating a strong brand they seemed to have created a cosy duopoly between them and Garmin, fighting off low-end Far Eastern competition such as Mio/Mitac who only competed on price.

However once Google dropped the bombshell of free navigation on android smartphones that supposed brand loyalty was cruelly exposed. Compared to Google's brand, and the magical price point of "free" (plus an unwise debt-fuelled acquisition at the height of the credit bubble) TomTom went into a tailspin from which it still hasn't recovered.

RIM: RIM is a business which had two supposed moats. The first was its ubiquity within enterprise IT systems, and the way its high security Network Operating Centres (NOCs) could sit within a customer's firewall and offer notoriously fickle business customers enterprise-class security. The second was its Blackberry Messaging (BBM) instant messaging service which captivated a generation of consumers.

Actually despite some high profile outages the competitive advantage offered by the NOC model still remains in place (and would offer an enticing strategic asset to a potential acquiror, IMHO). However while BBM retains its loyalists, instant messaging services such as iMessage and Google Talk are two-a-penny nowdays and this has proved no moat at all against competition from low-cost android smartphones.

Where are the next Roman Candles?

"In retrospect" are the easiest words for an analyst to write. In retrospect everything looks inevitable. In retrospect everything is obvious. But "in retrospect" is never going to put bread on the table (unless you are a famous TV historian, perhaps). So where are some future Roman Candles.

I do not know, and if I did I would of course be on my way to becoming a millionaire. But here are a few which cross my mind.

Sage: Sage is a leading provider of accounting software for small businesses. Sound unglamorous? That's what's so great about it. Over decades it has assiduously built up an enormous customer base of small businesses who pay it annual support revenues (which now make up the majority of its business) which are high margin and largely recurring. This is a beautiful moat and while the company has displayed unspectacular growth set against racier tech companies, it has been a reliable defensive cash generator for many many years. The sort of company which Warren Buffet might like.

However the company has been excruciatingly slow to move towards a cloud computing model, partly because its fabulous customer reach is built on being as local as possible. Each country has its range of products (largely incompatible with those from the next country) which lets cosset its customers far more closely than its competitors. But as I have said this product spaghetti means it has been very slow to develop on-demand software which simply works through a browser. (Imagine if Google had to develop a completely different version of GMail for every country it operated in)

So the risk is, what if a new generation of customers grows up unused to the idea of buying and installing software locally and just uses on-demand software? And what if, growing up with the global homogenisation of Google and Facebook they don't give a fig about having your countries software being developed round the corner? Suddenly that moat won't be much of a moat after all.

Telecity: Sticking with the UK theme for a while, Telecity is one of Europe's largest providers of data-centre services. They operate a deliberately dumb model - they give you a room in a data-centre and then charge you by how much power you draw from the socket in the wall. You could fill the room with pinball machines if you like they don't care.

As I said, dumb, but brutally successful in a rising market. The shares have been a stock market darling and now trade on 30x forward earnings which would appear to price in a significant competitive advantage. CEO Mike Tobin is, along with Bill McDermott at SAP, the most impressive salesman I have ever met. The guy could sell oil to Arabs, and then get them to drive round again for a refill.

The competitive advantage Telecity enjoys is simply that in the short-term (say the next five years) it will be extremely difficult to add new capacity to the market due to the complexities of planning restrictions and the need to secure upfront investment. Telecity, which has access to more capital than it can use at the moment, is in a sweet spot. It earns ROIC in the mid-20% on a cost of capital which is more like 10%.

However in the long-term (and the shares now price in longer-term success) the prognosis is less bright. As I said the beauty of the model is that it is dumb. But that is also the weakness - there is no patent protection, no recurring revenues (customers renegotiate contracts on an annual basis), and although there are near-term constraints the outsized returns Telecity earns will one day attract competition. At that point they become eroded away and the bottom falls out of the model. One day.

Facebook? This is the biggie. As I wrote in the last post, Facebook has a big moat at the moment - over 900m active users and all their data and connections. It was also very early into the platform game, building precisely the API and app ecosystem which competitors like MySpace lacked.

However a little like Nokia I am continually bewildered by how much the platform seems to have stood still. Put simply in theory the app ecosystem should provide a huge moat, but since Farmville I have seen little or no innovation on it. And developers are now starting to make ominous rumblings about Facebook selling them short as it chases near term ad revenues.

True the installed base is another huge moat, but Google Plus has ramped from a standing start to 150m active users pretty damn fast. And remember every Android phone user has to have a Google account, making them a de facto part of Google's social network.

One thing's for sure - as MySpace showed when a social network loses its mojo it loses it big.


  1. Totally agree. Nice post.

    Short may the way of the future for investors. Seems safer nowadays. Sure, your loss is "unlimited" but I've never seen a stock goes to infinity! And science shows that corporations always die (google Geoffrey B. West - interesting presentations).

    1. True at the moment, although bear in mind in the long term markets tend to go up so that might not be a strategy for forever.

      Yes corporations always die (and long-term DCFs never seem to price in terminal decline. Funny that), but the problem is "the market can solvent irrational longer than you can remain solvent"!!

      As an aside what can really kills you in a short positiont (especially if you're trying to bottom pick distressed names) is when an M&A bid comes in - what if Google paid $8bn tmr for Nokia just to get its patent portfolio. (although a bid is not going to be an issue for something like Facebook for a long time, given founder's controlling stake...)

    2. Japan is the closet model of the future? Demography is destiny.

      Yes, M&A is a risk. On balance though, there is (or used to be) an optimism bias built in to prices. Plus dumb money is mostly on the long side? So all else being equal, the odds may be better on the short side (this is just a guess. But surely can't be worse than long)? And who knows, one might always luck on to a black swan!

  2. Bluehost is the best web-hosting company with plans for all of your hosting needs.

  3. Thanks for a nice share you have given to us with such an large collection of information. Great work you have done by sharing them to all. simply superb.

    amazon web services in india