In December last year I published an article on submarine cable landscape set to unfold in India in 2019-21. An afterthought stayed on thereafter - why is growth of new submarine cables abysmally slow in India, despite skyrocketing growth in data consumption. The answer may not be straightforward, given the number of intertwined variables involved. One key reason is, submarine cables and their landing in the last 15 years have been under monopolistic control of telco carriers, whose dominance manifested primarily as duopoly of the two leading providers and that possibly throttled the growth of submarine cables.
This continued unchallenged due to negligible participation of the incumbent provider, that remained marginalized and confined to domestic market, very unlikely of many other countries where the incumbent providers have formidable global footprint. While the regulator has been actively pursuing policies to bring open access to submarine cables and break down commercial barriers, their efforts are stemmed by long drawn litigations. Hence there is need for new approach to develop new submarine cables. This article aims to explore the role of Independent Infrastructure Provider or IIP and uncover how an IIP in India can orchestrate design, development and operation of open access submarine cable system.
In markets globally where construction of new submarine cables is progressing steadily, nearly 15 IIPs over the last 4 years are leading more than 30 submarine cables investing more than $8 billion with 5 cable systems already operational. This encouraged me to publish an article to profile the top 10 IIPs in November 2018. Evidently their fast paced growth will reach a scale of readiness by 2020 to disrupt telco dominance of submarine cables. These developments are shaping up in US, Singapore, Hong Kong, Australia, UK, Brazil, Nigeria and even Mauritius. In comparison, India does not have an IIP till date that is focused on developing open access telco-neutral submarine cable(s) to bring subsea connectivity options unshackled from telco dominance.
Primarily by ownership control of the cable landing station (CLS), dominance of the Indian telco is assured as the pricing mechanisms discriminate against competition even within the same cable consortium. Further, backhaul and wider terrestrial fiber ecosystem which one is then restrained to go with creates attractive long term profit generator that unless challenged will continue to restrictive practices. For tomorrow’s opportunities rapidly unfolding in India this is unacceptable to turbocharge the growth engines necessary for a globally competitive digital industrialization in India.
More than ever before the need for Indian IIP(s) is felt now to develop, own and operate new submarine cables, with never-before design specifications and unparalleled economics that would seriously question the efficacy of submarine fiber pair lease with submarine cable built and above all lower price of IPL and IP transit by no less than 60-70% by 2022. This is also pertinent with respect to analysis presented by Telegeography in PTC 2019 that shows plateauing of price declines (in Mumbai) in the last 3-years.
India has always had an unique position in the Europe-India-Asia submarine connectivity corridor. All the 5 submarine cables from South East Asia to Middle East and Europe (SMW3, SMW4, AAE-1, BBG, FEA) land in India. Only exception is SMW5. 4 cable systems from Europe (EIG, IMEWE, TGN-EA, FALCON), 2 cable systems to Africa (SAFE, SEACOM) terminate in India. Only 2 privately owned cable systems, from Singapore (TIC, i2i), terminate in Chennai. The irony is, while the 2 cable systems are privately owned, the private owners are incumbent telco carriers. 3 cable systems are under construction (CACS, IOX, SAEx1), but its impact if they finally make it would be marginal at best. Evidently that’s too less for the fastest growing wireless data market in the world and the pent up demand for new submarine cables needs to be addressed.
It’s anyone’s guess that telco carriers will join new submarine cable consortiums and be the lead driver. Airtel is expected to rope in SingTel along with OTTs (Google and/or Facebook!) and likely to go for Chennai-Singapore submarine cable, to start with, extending to Europe. Reliance Jio is expected to focus on Mumbai-Marseilles route and also rope in OTTs (Google and/or Facebook!). If this happens Microsoft and Amazon are likely to pull up their sleeves to participate in new submarine cables. Plan of action for Tata Communications, the czar of submarine cables in India with arsenal of 7 cable landings in 5 CLS, and Vodafone Idea is not clear yet, though they will be force to recon with, once they reveal their plans. However none of these new telco carrier led cable systems will be any different from other cables that have landed with these telco carriers till date. Hence the market needs a new entrant, an Indian IIP, that will orchestrate new scheme of collaboration for development of one and possibly more open access submarine cable systems with telco carrier agnostic distributed capacity breakout. Such never before development will drive the telco dominated market to think afresh on next-gen subsea connectivity.
The IIP as new entrant in submarine cables market in India will have an uphill journey to establish credibility and present compelling value proposition. Possibility of leading IIPs like RTI, Aqua Comms and Seaborn Networks that have proven track record in design, deployment and operation of submarine cables doesn’t seem practical, yet. While capital is one of the most important success factor, it should be considered with a word of caution. Coming to India with truckload of money will provide the complete answer to develop submarine cable system only if the overall approach is harmonized with evolving dynamics of the fiber infrastructure and datacenter market. Hence the need is for integrating expertise and local knowledge to navigate the regulatory process and orchestrate collaboration of external international industry expertise brining in the latest architectures and network topologies with local on the ground expertise in navigating the complexity of regulatory and tax compliance.
Thus the IIP is likely to start the journey ground up, from scratch. That is also the reason the IIP would very unlikely go solo in India and take on the entire investment, for the risks involved would outweigh the rewards. Hence it has to engage OTT players and telco carriers, gain their confidence and onboard one or more as consortium members that commit to pre-agreed capital infusion. However between the two, IIP is highly likely to favour OTT players as consortium members and signup telco carriers as anchor customers. The IIP by itself also needs to demonstrate commitment and conviction as lead consortium member. The best way to uphold skin in the game is to bring capital from PE funds or any other credible source.
It also needs to put forth a differentiated approach. That’s would in all likelihood be construction of new open-access CLS. Here comes a critical question – how many CLS should be constructed. In case it’s planning one sided cable i.e. Chennai-Singapore or Mumbai-Marseilles, one CLS in Mumbai or Chennai would suffice. However if it wants to go for larger Europe-India-Asia play, one in Chennai and one in Mumbai would be necessery. Notably sharing of telco owned CLS will restrict flexibility and independence and defeat the objective of open access CLS. To move forward, it would be necessary for the IIP to be a licensed operator in India. Hence it has to go for NLD and ILD licence. This requires heavy lift of preparatory activities in order to comply with license conditions. Once the license is granted, IIP will be able to proceed with other permits for construction of CLS.
Such a submarine cable system will be first of its kind in India where the consortium members other than the IIP and OTTs, could be competing telco carriers, with capacity holding in the same cable system, along the same route and without ownership and control of the CLS and hence no CLS based monopolistic differentiation (think SMW4 and BBG). While the OTTs will also evaluate options of joining more than one submarine cable consortiums, the IIP needs to have a pitch to influence the OTTs to go with the submarine cable initiative that will see light of the day first and reap the benefits of first mover advantage.
One of the most important and tricky question to addressed by the IIP is - what should be the size of the consortium membership. Empirically, if not experientially,IIP+3 would seem good fit. Arguably the telco consortium members may feel the IIP led submarine cable will have limited usefulness, as they would not be able to sell to OTT consortium members. However that will happen otherwise also as the hyperscale OTT players will join consortiums of other submarine cables along with telco carriers as they would be interested in 4 and even 6 diverse routes in the next 3-5 years. This could also be the reason the IIP may not prefer to have Indian telco carriers in the consortium, so that it could have the agile and nimble footed approach of a startup venture.
The consortium members need to agree on including provisions to onboard additional providers during or even after the deployment for landing in countries enroute, without impacting the construction of the trunk cable system. This is likely to happen with providers, with or without collaboration with OTTs, that are agreeable to fund the branch submarine cable to their home country. Allowing a provider to fully fund the branch to a given country, deprives existing consortium members from having any right to land capacity in that country. Some submarine cables in the past have suffered because of this reason. Hence the IIP led consortium for the trunk submarine cable should have part ownership of the branch submarine cable. This could in the form of reciprocal commercial arrangement as the new consortium member(s) also need to have capacity on the main submarine cable. In addition the IIP as the cable administrator needs to have operational ownership of the dry plant at the end of the branch submarine cable.
A fundamental attribute that needs to be considered to characterize this IIP led submarine cable system before proceeding further. What should its span or spread be. Chennai-Singapore or Mumbai-Marseilles as standalone cable systems could raise apprehension of incompleteness or lack of marketable value proposition in either Asia or Europe. That’s because connectivity requirements to/from India alone many not be attractive enough to make investment remunerative for 4 or 5 consortium members. Hence for a submarine cable system from India to be truly next-gen, global and attractive investment proposal, it needs to connect Europe and Asia. An Europe-India-Asia cable system, that does not land in too many countries, and with access to transatlantic and transpacific routes would make differentiated value proposition. The route would be Marseilles-Egypt-Djibouti-Mumbai-Chennai-Singapore. Going further through Gulf of Gibraltar to UK has remained questionable for SMW3 and EIG, and avoided by most other cable systems, given multiple options of cost effective Dark Fiber connectivity from Marseilles to transatlantic, Africa-facing and other European cable landing stations. At the other end, going from Singapore to Hong Kong though attractive as destination and to get access to more transatlantic cables, would involve additional span of 2600 Kms and increase cost.
A cable system spanning Marseilles-Egypt-Djibouti-Mumbai-Chennai-Singapore will be as long as SMW4, that’s 20,000 kms, with 16 consortium members investing $500 million. If the branching to other countries are excluded, it’s 16,000 kms and nearly $400 million initiative. Is this ruinously expensive! Possibly is not for a 4 or 5 member consortium, given the fact that has gained prominence in the last few month about AAE-1 and BBG facing insurmountable challenge with overland connectivity with Malaysia to Singapore due to restrictive regulatory provision of traffic reciprocity in Singapore, leading to idling capacity inventory. While aspiring to make it big, a pragmatic approach with respect to time and money needs to adopted to make a start. Hence it could be divided in two segments -Marseilles-Egypt-Djibouti of 7,000 kms and Djibouti-Mumbai-Chennai-Singapore of 9,000 kms. Between the two the second segment - Djibouti-Mumbai-Chennai-Singapore of 9,000 kms, is expected to witness sharper year-on-year growth in demand and could be deployed first and Djibouti-Egypt-Marseilles of 7,000 kms taken up in the second phase based on success of the first phase. Further Djibouti is a well-known interconnect location to reach Africa and Middle East. The 9,000 kms system with two CLS in Mumbai and Chennai would be close to but less than $300 million. With 4 or 5 consortium members sharing it equally, its $75 million or $60 million. Alternately if it’s a 6 fiber pair system, it’s $50 million per fiber pair. If the IIP is able to retain 2 fiber pair with $100 million investment, the 4 consortium members pool in $50 million each for 1 fiber pair. Considering 120x200G or 60x400G technology per fiber pair is deployed, $50 million investment brings capacity ownership of 24 Tbps. This however seems to be usual business model and nothing really differentiating about it. However if the IIP could have a higher count of fiber pairs with itself, it could play the game very differently. It should be noted that the market is poised to see the first 12-fiber-pair and 16-fiber-pair cables being announced in the course of 2019. The technology is available with top-4 submarine cable system vendors. Active evaluation is underway. It’s not difficult to foresee that economics in the highly competitive landscape will not be roadblock for someone wanting to be the first-mover to go ahead with 12-fiber-pair or 16-fiber-pair system. In fact the next leap of 24-fiber-pair system could also emerge, much sooner than expected.
While the cable span and architecture is firmed up and consortium membership finalized, the IIP also needs to firm up its business model. The likely business model will be carrier’s carrier and OTT’s carrier. This would mean services the IIP offers would be 100G, 200G, 400G, spectrum and virtual fiber pair. Given the expected RFS of the submarine cable system would be Q1 2022, Tier1 telco carriers and Tier1 OTTs are likely to go for spectrum and virtual fiber pairs, while Tier2 telco carriers, Tier 2 OTT players and datacentre providers would prefer to go with capacity wavelengths.
The IIP, while leading the overall program management to develop the submarine cable system and retain command and control as the cable administrator, needs to develop an interconnect approach for India, Singapore, Djibouti and Marseilles. For India, the IIP led submarine cable system targeting CIF in Q4 2019 and eventually commissioned in Q1 2022 needs to integrate with the connectivity ecosystem that continues to growth steadily with infrastructure deployment by multiple providers in Mumbai and Chennai. Hence the IIP needs to partner with telco carriers, datacenter providers, and metro fiber providers and harmonize its plan of action. Alliance with providers beyond the consortium members needs to be forged based on outright purchase or collaborative partnership based on sharing of risks and rewards.
While the high level game plan for India was discussed at length, the cable landing in Singapore and Djibouti also needs ratification. In Singapore there are two options – either the IIP creates Singaporean subsidiary, acquires FBO license and based on it contracts with SingTel to land the submarine cable in any of the 4 CLS owned and operated by them. This however would be a conventional and restrictive approach followed by nearly every cable system till date. The IIP could prefer not to have dependency on SingTel and develop new CLS ground up in order to build truly open access cable system. Developing new CLS would expectedly preferred by other IIPs too that are planning or started construction of cable systems with Singapore landing, but hesitant of the rigor involved. Hence if likeminded IIPs decide to join hands and the lead IIP acquires FBO license to develop new CLS in Singapore, it would be a remarkable milestone, hailed by the entire submarine cable fraternity. It will set a trend that will be replicated progressively in other countries.
The Singapore episode does not end with CLS. While the CLS houses the SLTE with facility for third party SLTEs to be hosted, a Dark Fiber network needs to be built from the CLS to create capacity breakout facilities in major datacenters and interconnect hubs. For onward connectivity within APAC and to Australia and US, consortium members and other customers can independently least Dark Fiber for interconnect with CLS where their preferred submarine cables land.
For Djibouti, the only option is to execute commercial cooperation contract with the incumbent Djibouti Telecom as the landing party to land the cable in one of the 3 CLS owned and operated by them. There is also dependency to interconnect with submarine cables that land or scheduled to land in 2 other CLS over resilient DWDM network operated by Djibouti Telecom. It’s highly unlikely that Djibouti Telecom would offer Dark Fiber connectivity.
How will the IIP starting ground up with humble beginning design, deploy and operate a submarine cable system with terrestrial extensions for multi-point capacity breakout and offer highest level of committed service levels. Needless to say it requires infrastructure, teamwork and effective leadership from well-regarded figures from the submarine cable industry. This would require sizable capital infusion even before the initiative reaches the milestone of CIF, after which the real action begins. What about team size. The IIP is unlikely to have a large workforce occupying multiple floors of a business high-rise with usual corporate flamboyance and elaborate business operations in multiple countries. It would likely be a lean team of domain specialists, with multi-functional responsibilities, fiercely committed to make the initiative a great success, increasing headcount progressively based on milestones achieved. A judicious balance of inhouse and outsourced activities need to be put in place. Several partnerships and alliances need to be forged to develop the connectivity architecture to achieve the goal of open access cable system. Not everything will be first time right and some midway course corrections would be needed. Being a new initiative in India without precedence there will be enough and more challenges to deal with in every stage of program management. Every member of such a team will create history the day the submarine cable becomes operational.