Indian telecom powerhouse Reliance Infocomm has signed a deal with China Telecom and will establish a direct cable connection between those two countries. Now the calls will fly directly over Reliance’s FLAG Telecom cables that will connect to China Telecom network in Hong Kong. When compared to meaty issues (such as Feedburner’s two-way RSS,) it doesn’t seem like a big deal, but it could have longer term implications on the bandwidth business.
Why? For starters the calls between the two countries that till recently were routed through Europe/US, will no longer need to travel half way across the planet. This means reduced role for major US carriers like Level 3, MCI and AT&T (currently the world’s largest ISP.) It is a trend that is gathering momentum, and slowly and slowly the network traffic that almost always used to flow through US is becoming more and more regional. Through most of the 1990s it was cheaper to connect the European Union countries through US, but then small regional networks started to siphon off traffic.
Asia network traffic is following the same trajectory. At some point in the future this is bound to have an impact on the US carriers. US is currently the top Internet hub country with 1.4 terabits/second of bandwidth. World’s fattest pipes (metaphorically speaking) are between London and New York, about 320 GB/s of bandwidth.
Will the regionalization of the traffic mean that the price wars will rear their ugly head? Will the prices plunge on the London-New York routes as once again capacity outstrips demand? Some folks who work in the bandwidth business say that the transit fees, that once used to make up nice profit center for global carriers are beginning to head sound.
But going beyond that – is this a general global economic trend? I can even pretend to know anything about global trade, but my broadband prism forces me to think, that this could be forbearer of a global trade shift. I believe that what sea routes, air routes and highways were to the 20th century, broadband pipes are to the 21st century. And from that perspective, things could be shifting away from the US being the hub of global trade. What do you think? I am seeking your intelligence and input to get a better understanding, on what might be just dumb thinking on my part?
PS: David Isenberg, Dave Burstein, Paul Kedrosky, Joe ‘The Stalwart’ Weisenthal, Mike Masnick, Andy Kessler, Carl Haacke, Nicholas Carr, and Pip Coburn…. I will be looking for your insights in particular. Jeff Pulver, you are not off the hook either!
Requesting all of us to respond? Hmm…sounds like a shamelss ploy to get atop Memeorandom.
Nevertheless, a response will be forthcoming.
Om,
I see a trend, but I don’t think it will be the direct cause of a large loss of revenues to N.Am. carriers. This is because I don’t think that a significant fraction of US (or North American) carriers’ revenues come from IP transit or bulk international bandwidth sales to other carriers.
This move that you mention from transit to direct routes is an engineering economics decision that occurs in the background continuously. The difference as I see it is that new players are becoming more important on the world stage. Hong Kong (HKTI and its corporate descendents and competitors) have had direct routes to Europe for a long time – so it is not new to see direct routes between Europe and HK. What I see is an intelligent reaction to network growth between the carriers in India and China. This cable capacity is not new, I think this capacity has been in place while. This looks to me like carriers trying to optimize their networks based on the available resources.
You do mention “the largest ISP” and that is interesting. How do you measure “largest IP carrier” and how important is this measure? Are the biggest transit carriers the most important or are those with the biggest base of end customers and associated end customer traffic? One thing that many carriers have found is that revenues and size are no guarantee to success.
KM, the largest internet services provider tag came from telegeography folks, who say that ATT has 52 landing points and are a network which runs through many different countries. I don’t think i meant it as ISP in the traditional sense … you know like Earthlink.
secondly, the network optimization is something which is what caught my attention. okay perhaps maybe i am making too much out of it, but i think will the regionalization is bound to have an impact.
by the way, great comment
Actually, Joe, I was hoping for your response here. in the comments section. Little or no interest in memeorandum’s top slot. however, it is something which I thought should be discussed.
Om,
It is a significant development, but not a leading indicator. It is another step in an already established trend – the decline of the empire. Our dependence on foreign everything (oil, goods, labor) and ROW growing independence from us (less reliance on US$, technology, innovation, etc). Not that we are going to collapse overnight – it should be a long, gradual slope 🙂
This question reminds me of a debate that took place in the 1980s … Was there causality in the correlation between telephone penetration and per capita GDP? There were a few papers published on this. For example, Hardy published a transnational study in Telecommunications Policy in 1980 in which he showed that residential penetration (not commercial) was predictive of economic activity. This study was criticized on methodological grounds, and followup studies were less clear about the correlation.
IMHO, the primary limitation of extending some of these questions has to do with the transnational nature of the questions rather than the technological ones. In the end, I believe that infrastructure is a necessary but not sufficient enabler of economic success. Many other factors are critical as well.
Om there are a variety of drivers changing the distribution of traffic around the Internet. Fiver years ago nearly all trffic went in and out of the US, that’s where all the significant web sites were physically located.
That centralisation has, and continues, to decline.
There is a regionalisation of content; driven by local internet brands, the demand for native language content (some of which is stimulated by governemt policy; Italy, Spain, China, Korea, Japan), the global/US brands (Google, Microsoft, Yahoo) have distributed their servers (as volume served has increased) in order to improve survivability and reduce the latencies to end users.
That demand is also stimulated by global population migration. Either economicly driven migration or refugee migration. The UN have a great deal of detail on what populations are moving and where – they also mirror the changes of traffic on the Internet.
Overlaying that are the differentials in economic growth (as you mention). Eastern Europe, Latin America and some Asian countries have some of the fastest growing economies (and fastest growing broadband deployments) and so traffic to and from them is growing faster than on some more established routes like in and out of the US. And yes, I think that economic strength and economic ties are proxies for Internet traffic.
In summary Intra Asia and Intra European traffic is growing faster than traffic into and out of the US. There is a dispersion of traffic that within the next five years will see the US as the smallest of those regions in terms of total traffic – although not by much.
Having said all that there will continue to be substantial growth in all regions and all routes. This is an equalisation of traffic, not a gain in one region and a decline in another – no one will end up with empty pipes. The far bigger problem still, for all carriers, will be how to carry all the traffic.
Just teasing, and it’s definitely something worth being discussed.
I suspect, as the above commenter states, that revenues from this aren’t a huge portion of their top-line, though I’ll have to check. Still, it’s symbolic the the major transit routes still flow through the U.S., even though it appears to be circuitous and uneconomical.
As I stated, on my blog, I think there are interesting parallels between this, and international finance networks. Money made overseas, still flows through NYC as foreign producers like to park their money in relatively stable dollars. But, as we’ve seen this year especially, the NYC banks take a huge cut of that cash, and, like the broadband map, it doesn’t seem particularly economical to go this way.
Ultimately, this doesn’t look zero-sum to me. There may be some pain for the incumbent telco’s, but a more connected world should benefit everyone. And while shareholders in AT&T, sorry at&t, might not like a price war, I think most businesses would be quite happy to see another one.
“Can Broadband Predict Economic Shifts?”
If anything, I think broadband would be a lagging indicator. Asian intra-regional trade already accounts for 50.3% of the region’s total merchandise exports (Source: wto.org).
“And from that perspective, things could be shifting away from the US being the hub of global trade.”
US exports as a share of total world exports, dropped from 12.7% in 1993 to 9.2% in 2004 – that’s over a 25% fall in a little over a decade. From this perspective, the US was declining as a “hub of global trade” as its dominance of broadband thouroughfares was being erected.
Amazingly, during the same time period, US imports as a share of total world imports rose(!) from 16.0% to 16.5%.
I think most mainstream economists would agree that this disparity – the result of massive borrowing, both consumer and government – is not sustainable. This would lead to a decline in the US share of world imports, and hasten its already declining share of world exports.
Without making an essay out of this comment, broadband statistacs as an indicator of world trade and economic activity would seem to be more reflective of the 1950’s. And now the indicator is simply catching up with trends that have been well underway and apparent for quite a while now in other fields, and that have been masked somewhat by US borrowing.
In any event, it will be interesting to see the results of the inaugural East Asian Summit taking place in Kuala Lumpur this week, where Indian PM Manmohan Singh has called for the creation of a pan-Asian free trade zone that would serve as an analogue to the EU in some respects.
[Offtopic, for Internet]
If I were to head any Telecom company in India, I would invest little more and drop my packets in Japan. Japan & South Korea have the best internet conectivity and cordial relations between Japan and India would never risk my investments. While my competitor, Bharti, has chosen to be smart and is already droping his traffic on SingTel networks, again one of the best bets in Asia.
Om,
Da Mayor (ex San Francisco mayor) Willie Brown told this story at newspaper legend Herb Caen’s funeral:
Herb was covering the opening of a new train station in some small town way to the south of San Francisco and San Jose. Morro Bay perhaps. Herb interviewed one of the local residents and asked him, “Are you excited about all the people and business that will be coming into town?”
The local resident turned to Herb Caen and said, “Excited? Sure. I’m going to use this new train station to get the hell out of here.”
Now, don’t read too much into this, except sometimes, the benefits of new technology are exactly the opposite of what you might think.
Andy Kessler
Andy Kesler got the general drift (though he was too subtle 😉
Asian ISPs originally sent all their traffic to the US because of two major points:
* Most of the Content was there originally
* The US was the most “neutral” and advanced country at the time
There was a long period where there was no peering within even the same country. So for instance if you were on one Singporian ISP as a customer and wanted to send email to another Singaporian on the other Singaporian ISP, the mail had to go all the way to the US and back.
But as many of these Asian countries have dramatically outpaced the US in both backbone and local access bandwidth and developed their own local content, and developed some peering. Traffic has stayed more within each country.
Now the trend that Om points out, they are bypassing the US as the way to route traffic between countries.
And here back in the US, we have become what the Asian Countries use to be: Bandwidth completely controlled by monopolies, cronyism between the monopolies and the people running the regulatory environment. All of which has stopped innovation in its tracks and the US is now 16 – 27th in the world in terms of broadband.
The US is becoming the Morro Bay of the world….
The less the world depends on the U.S., the less the world will be f***ed in the a** by the Bush administration. I think it’s a positive thing.
My view is “Not really”. While international communications is important to growing economic activity, what is even more important is in-country communication. While a direct fiber link between, say Hong Kong and New Delhi is useful, the economies of both countries actually need more bandwidth from their major cities to more modest concentrations of people than better international links. with only about 300 million fixed-line telephones and about the same number of cell phones in China, telecom penetration is still only about 25% of the population. India has a similar profile. This wiring (both fixed line and celluar) of these countries’ populations will be a much larger effect than what route their international bits take.
The way I see it, about 2.3 billion people potentially disintermediated the USA and Europe in their communications.
It also shows that there is significant communication taking place (duh!), which would suggest a shift in trade patterns merely due to the size of the respective populations. Transactions begin and end with communication.
Om
Good observation, especially with MCI’s acquisition by Verizon, one of the key USPs was revenues from trunks and bandwidth, which definitely appears to be diminishing as you point out. Another thing is web servers, there was a time when most of the web server hostings were based out of the US, even Indian sites as a matter of fact, but that is diminishing as well. I think the thought seems reasonable especially with Asian economies being the booming broadband markets of the near future.