As we peeled back the layers on Axia, its undervaluation stunned us. Axia owns and/or operates fiber broadband networks in several geographies. Its first network is in Alberta, Canada, where this virtual duopoly business produces $22m in annual EBITDA. With Axia’s Enterprise Value at $59m, just this high quality asset alone prices Axia at an absurdly low 2.6x EV/EBITDA.

In addition to Canada, Axia also owns 50% of a fiber broadband network in France. The other 50% was sold a few months ago to a European infrastructure fund for $92m ($42m upfront cash payment + $50m 3-year earn out). That implies Axia’s stake in the French broadband network alone is worth more than the EV of Axia. 

Further still, Axia has 3 more broadband networks–in Massachusetts, Spain, and Singapore. These fiber networks–which are just now beginning to “light-up”–stand to produce another $8m-$16m in pre-tax income.

As absurd as this sounds, Axia’s fiber broadband networks altogether stand to generate$28m-$57m in pre-tax income. Yet, the company’s EV today is $59m.

This tremendous earnings power is not yet visible to the market. It is masked because Axia has spent money to buildout its fiber networks over the past few years, i.e., “to lay pipe in the ground.” However, most of the build outs are now complete, and as these networks “light-up,” income from them should rapidly fall to the bottom line. That is the catalyst that should cause a repricing of the stock. 

Being virtually the only provider of fiber broadband in a geography is a tollbooth-like business. It provides a product people need to use everyday; it has stickiness, recurring revenues, no obsolescence risk, and huge tailwinds with the ever-growing use of data. We find that when we calculate what Axia is worth based on what its networks will earn, we get a valuation 3x-6x higher than the current share price.

About Axia - Selling Broadband Services Wholesale

Usually, we all purchase our broadband from our local telco. In your area, for example, Verizon may be the telco that built out and owns the fiber infrastructure. Owning the fiber, Verizon then proceeds to sell you services directly (TV, internet, phone, etc.)–no one else can sell services over Verizon’s infrastructure. Governments have grown to hate this model because they think it puts “monopoly” power in the hands of telcos.

Axia offered governments the chance to replace that model. Axia proposed to own and/or operate the fiber infrastructure, but to never sell services directly to the end user. Instead, Axia would sell wholesale broadband to RSPs (Retail Service Providers), who would then sell services to the end-user (TV, internet, phone, etc.). Any person can an open a business as an RSP, purchase wholesale broadband from Axia on the same terms offered to everyone else, and then try to make money by selling services to the end-user.

This creates a dynamic where Axia is the only company that owns the fiber infrastructure in a city, but yet there are dozens of RSPs competing to sell broadband services to businesses and residents.

Governments love the model that Axia proposes. They usually put up the capital for Axia to buildout fiber to areas where it is unprofitable for the private sector to do so.  

Axia’s first network was created seven years ago in Alberta, Canada, and it has turned out to be a huge success. 

Alberta, Canada (SuperNet): $10m-$21m pre-tax income

It all started with Alberta, Canada in 2005. The Government of Alberta put up several hundred million dollars for Axia to build out and operate a fiber network in rural areas. The network consists of over 13,000 kms of fiber. This network began operations in 2005 and has since produced a consistent annual stream of ~$15m net income.

The Alberta network is a great asset, with one caveat. Axia’s contract with the government of Alberta comes up for renewal in 2015. Speaking with Art Price, the CEO, one will get the impression he is confident the contract will be renewed on terms that will allow Axia to continue to earn at least as much as it did in the past.

The reason is because Axia’s contract with the Government of Alberta already has price adjustment language in the agreement–a clause where the government can demand a lower price if Axia is not offering a competitive price. So at any point, if the government thought Axia was overcharging, they could have already requested a price adjustment. So price is not that big of a motivating factor, because the government could argue if Axia’s prices aren’t reasonably competitive, that they should be adjusted. 

Furthermore, Axia currently earns ~40% margins on its Alberta network. This is in line with the 30%-40% margins that Canadian telcos earn (Tellus, Roger Communications, Bell Canada, etc.). Nevertheless, to be conservative, post-2015 we assume it is possible Axia might have to cut its margins down to way-below market rates of 20% in order to have its contract with the government renewed. Though Axia thinks Alberta will not get a profit cut (they think profits will actually be higher because data broadband use grows everyday), to be conservative, we provide a range of $10m-$21m pre-tax for Alberta’s network post-2015. 

As Alberta turned out to be a huge success for the government, other governments began to take notice. With such a success under its belt, Axia was able to propose many similar models to governments worldwide and secure very profitable assets for itself. Its biggest asset is the one growing in France.

France (Covage): $10m-$20m pre-tax income

Covage is a new 3,700 kms fiber broadband network in France. Axia owns 50% of Covage and is the operator of the network. Cube Infrastructure Fund just purchased the other 50% for $92m ($42m upfront + $50m three year earn out). (Cube is primarily a passive financial investor in European infrastructure assets.) All the calculations below are based on Axia’s remaining 50% interest in Covage.

The Covage network has been recently activated and just generated a positive EBITDA. In FY 2011, Covage generated EBITDA of ~$4m. Covage’s current market is the 34,000 businesses that are currently within 100m of Covage’s fiber network. Penetration has consistently grown 0.6% each quarter (see Axia’s September 2011 presentation, Slide 29, for an illustration), reaching 6% penetration as of the latest quarter. Axia’s management has repeatedly stated they will earn ~$14m pre-tax at 15% penetration. (In Axia’s MD&A filings, it states “At a market penetration of 15%, we would expect to earn approximately 20% return before tax on our capital investment [in Covage].”)

Covage should reach 15% penetration by 2015. The only reason penetration is not increasing faster is because there is more demand than Covage can supply. It takes time for Covage to build out fiber from its major backhaul to the premise of each business (the current average wait time is ~50 days).

It is important to note that Covage has reached an inflection point. In its 2011 MD&A, Axia states “We are currently selling Broadband Services to about 5.4% of Covage’s addressable market although the cost of products and services sold for this NGN include a full complement of sales, marketing and operational staff and related activities which are capable of servicing a much larger customer base.”

Management has also stated several times on conference calls that at 5% penetration, operating expenses become relatively flat and they expect incremental revenue to largely drop to the bottom line. Management also stated that margins should be in the 30%-35% range (which is line with the margins that other telcos produce, such as France Telecom, SFR, Tellus, Roger Communications, Shaw Communications). They reiterate that the nature of a fiber network is that its cost base is fairly fixed (it virtually costs nothing to light up an extra strand of fiber), and so the margin is based on the growth of revenues. “[A]ny incremental dollar now is effectively an incremental dollar that essentially drops to the bottom line.”

Covage is the only fiber broadband provider in its market. Currently, the majority of the remaining 94% of the market not served by Covage receive their broadband through copper (DSL). Copper is inferior to fiber–it cannot provide the speeds, the asynchronous upload/download, the constant connectivity, etc. that fiber provides. (A business that needs speeds in excess of 100 Mbps or constant connectivity to the cloud cannot use a copper connection.) Interestingly, our research revealed that the price for a fiber connection is now equivalent to a copper one, creating an inflection point.

The RSPs (Retail Service Providers that buy bandwidth wholesale from Covage and then resell it to end-users) we spoke with are aggressively pitching Covage’s fiber broadband. For them, it is better if a customer is on fiber rather than copper because it generates additional revenues. One RSP we spoke with has his business with Covage growing in excess of 20% per year. This RSPs business currently consists of customers on 10,000 lines of copper and on 1,200 lines of fiber–an 8:1 ratio. In 2012, he stated he is confident he will sell 3 lines of fiber for each 7 lines of copper. He is certain 30% of his customer base will soon be on fiber, and eventually, 100%.

Furthermore, all of Covage’s estimates do not count the residential market–that is a free option. (Residential users use much less broadband than businesses, so they don’t have the same dire need for fiber broadband. But as iPads, Apple TV, 3D HDTV, etc. grow, they will eventually have to switch to fiber, as well.) 

Also, there is nothing stopping Axia from achieving higher than 15% penetration of the business market. 85% of the market will not stay on copper DSL forever.

Given all the above, Covage should produce at least $10m-$20m in pre-tax income by 2015.

(Also, Axia just added the city of Nantes to its network. Based on our due diligence, that increased Axia’s addressable market to 39,000 businesses–a significant add.)

Singapore (OpenNet): $5m-$11m pre-tax income

The Singapore government put up SGD$750m to outlay the entire country with fiber to every premise–yes, fiber to every single premise. The Singapore government is hellbent on switching over the country to a fiber network. The cost of a fiber broadband account will be virtually the same price for DSL. (OpenNet has proposed wholesale prices of SGD$15 (US$10) per month per residential fiber connection and S$50 (US$35) per month per non-residential connection.)

Axia owns 30% of OpenNet and management has consistently stated its stake will produce SGD$6m-$14m per year, or $5m-$11m. (The only reason it is not more than SGD$14m is because the Singapore government capped it at that amount.)

Spain (Xarxa Oberta): $3m pre-tax income

Axia signed a contract with the province of Catalonia about a year ago, where it will so far have 1,900 kms of fiber. Axia will turn on their first services in April 2012. Spain will be an extension of France. The current market commitments (contracts) that Axia has in Spain is about 1/5 the size of France contracts.

Massachusetts, U.S. (MassBroadband 123): $1m-$3m pre-tax income

This network will be activated in 2013 and will serve the western part of Massachusetts. It will have 2,153 kms of fiber serving 124 communities, with 38,000 households and 44,300 commercial premises. 

Axia has revenue guarantees from 1,400 government institutions. The revenue from these committed sites will be sufficient to cover the operating and maintenance costs of the network. Any incremental sale of services to the private sector will be accretive. 

In our due diligence and speaking with players in the Massachuessets market, Axia does not stand to make a lot of money serving the communities that will be getting fiber. The estimates we received are about $1m-$3m. So that Massachusetts network should be viewed as more of a strategic foothold for Axia to establish itself in the U.S. market, rather than a substantial profit generator.

Management

We visited Axia’s headquarters and found Art Price and the management team to be very candid and transparent. Art sees Axia’s stock as being undervalued and has recently bought shares himself and initiated a share buyback program for the company. Alberta turned out to be a huge return on capital. The current projects looks like they’ll have great returns as well, while maintaing a solid balance sheet. 

VALUATION

All the telcos trade at around 10x EV/EBIT–and that’s with the fact that a large part of that EBIT is in old, legacy fixed-line business. To be conservative, we assume Axia should command at least a 7x EV/EBIT multiple. 

We also will not count the cash that Axia will generate over the next few years, as we will assume that will go towards CAPEX to finish building out its remaining networks. In several years, based on all the above, Axia should produce in the range of:

Bottom Line

We find Axia to be a highly asymmetrical investment. They sell fiber broadband in a world that is increasingly using more data everyday. Axia is in markets where it is usually the only provider of fiber broadband and it is uneconomical for another competitor to enter its market. The earnings for all its networks have not yet fallen to the bottom line, which creates the dynamic of a high quality earnings stream that is trading for an absurdly cheap price. 

We believe Axia is cheap enough where it is hard to imagine a scenario where an investor would lose money, yet easy to see how they would make a 3x-6x return.