Saturday, August 16, 2008

IPTV - potential $100m revenue growth in 2009

Over the last year, UTStarcom has provided subscriber numbers to track the growth in their iptv system in China and worldwide. I have kept tract with the following chart.

During the last earnings call, we learned that the global subsriber base for UT has increased to 956k, increasing by about 100k in the previous 3 months, mostly from China. Out of the total, China has about 670k, Japan about 250k, and the rest of the world less than 50k. Aside from China and Japan, the company has contracts in India (MTNL, Bharti Airtel, BSNL, Goa, Sri Lanka), Taiwan (Markwell cable), and Brazil.

Global IPTV subscribers and growth projections - I had posted previously on the state of iptv globally from the end of 2006.

The report, entitled “IPTV: The Killer Broadband Application”, forecasts that there will be nearly 60mn subscribers to IPTV services around the world by 2010, with a growth rate of over 500% expected during the next three years.

2011. - Worldwide IPTV subscribers to reach 72.6 million by 2011, says MRG. the subscriber

Recently, here were iptv growth rates from Q1 2007 to Q2 2008:

Iptv subscribers in various regions:

Region Q1 2007 Q1 2008
Europe 3,875,266 8,425,370
Asia Pacific 1,129,355 2,619,035
North America 850,601 2,258,601
South and East Asia 1,353,000 2,086,000
Latin America 2,300 11,183
Middle East & Africa 10,000 10,000
Total 7,220,522 15,410,189

While Europe has led the way, most projections have the APAC region leading the way in the next 2 years. Here is a recent article (June 2008) that I posted previously but is a good read specifically on UT markets in Asia. Here are some excerpts:

"The research firm notes that IPTV currently is deployed in seven Asian markets: China, Hong Kong, Malaysia, Singapore, South Korea, Taiwan and Thailand. Service introductions are anticipated in India and the Philippines later this year." UT management has discussed the NGN in the Philippines but not iptv. We still have not heard anything official out of Hong Kong except for the local news interview with a UTStarcom executive.

"In-Stat Research predicted that APAC's IPTV market would grow 80% annually through 2010, when revenue would hit $4.2 billion. China, Japan, India and South Korea are expected to account for most of the region's growth." Japan has been stalled for some time because of regulatory issues regarding multi-cast/time shifting that negate iptv/UTs advantage but it seems like there is some growth potential as well (definitely massive if regulations change).

"China is the untapped market with the potential to drive the Asian IPTV numbers through the roof. While current deployments are limited, future rollouts are planned in conjunction with upcoming global events.
"The growth of IPTV will remain modest for now and will take off only after 2008," says Serene Fong, broadband analyst for ABI Research. "Adoption will be boosted by major events, such as the Beijing Olympics in 2008 and then the 2010 World Expo in Shanghai." She adds that the Chinese government is backing IPTV because the technology is aligned with its long-term plan to unify broadband, Internet and telephony services. ABI projects that the IPTV subscriber base in China will surpass 23 million by 2012.
In her report, Fong warns, however, that there are bottlenecks in China that may restrict IPTV growth. These include a shortage of content and too much government regulation."

"A report from Pyramid Research forecasts 10 million Chinese subscribers by 2011"

Alcatel-Lucent's Schutte says China "is a difficult market to enter and do business in" because it is complex and involves competing IPTV technologies and standards deployed in different regions. Its clear that foreign companies will be shut out in the iptv market in China.

End of 2008- The company will probably end the year with around 1.2m subscribers, maybe a little more due to the olympics. This is about double the 600k that the company started with and consistent with global trends of about 100% growth. UT will probably have about 900k in China. At 62% of the China market, that puts total China iptv subscribers at around 1.5m. Sometimes, the numbers for China are inflated due to counting other iptv services that are not comparable so I will use UT's subscriber and market share numbers.

China Growth for 2009 and beyond- Lets say China doubles every year. It will have 3m by end of 2009, 6m by end of 2010, 12m by 2011, and 24m by 2012. That will be around the projections reported above. If UT gets 50% market share in China, they will increase their own subscriber base by 600k in 2009 (1.5m-900k), 1.5m in 2010(3m-1.5m), 3m in 2011 (6m-3m), and 6m in 2012 (12m-6m). Assuming STB revenue of $125/box and $40 for infra/service/maintenance, revenues could be $100m, $250m, $500m, and about $1b in the next few years. UT started the year with about 400k in China and will end the year with close to 900k so about double. The growth in China will probably exceed the 100% in the earlier years and UT will have a higher than 50% of the market. If we use those trends to fine tune the numbers, UT could have 1.8m-2m or more subscribers in China by the end of 2009. That is a growth of about 1m in 2009.

India - UT has been winning a lot of contracts, topped off by the recent win with BSNL in 20 cities.

August 6, 2008 - Indian firm Aksh Optifibre is planning to invest US$100mn in its joint venture with Bharat Sanchar Nigam Ltd (BSNL) to deploy IPTV services in 20 cities in the regions of Rajasthan, Jammu & Kashmir, Punjab, Haryana and Uttar Pradesh (west), according to local reports.
K S Choudhary, Managing Director of Aksh Optifibre, commented that the company is targeting a subscriber base of 500,000 in the next three years: "[IPTV] technology allows the use of internet, telephone and television at the same time... It's difficult to ascertain the business dynamics of this highly volatile segment. We already have more than 7,000 subscribers in New Delhi and Mumbai where we are offering iControl on broadband network of MTNL."Under the terms of its agreement with BSNL, Aksh Optifibre needs to have 10,000 IPTV subscribers within the next nine months across 20 cities, however it is reportedly confident that it can reach this number. Aksh concluded a GDR/FCCB issue of US$40mn in January of this year to finance the IPTV and VoIP services in Delhi and Mumbai.
On one hand, the subscriber number target for 3 years and dollar amount seem to be good but then they are only targetting 10k users in 9 months. UTs broadband phase I contract was for 900 cities so the 20 city iptv deployment is only a fraction of BSNL's footprint. MTNL, Bharti Airtel have similar growth targets but the current numbers show slow movements in India. You also have Goa, Sri Lanka, Taiwan with high growth targets from 100k to 500k but slow movements as well.

Brazil - UT has a win with Brazil telecom but again it is only for 10k users. It also won a FMC contract and is deploying it currently but how quickly the market accelerates is unknown.

Russia - UT is said to be working on a couple of contracts by year end and iptv could be one of them. This is another huge market (heck all of them are huge but timing is the key).

Latin America, Middle East, Eastern Europe - UT has had some success in PAS, NGN, broadband so iptv will also be a part in the future. Blackmore mentioned an update in Q3 regarding these other regions. Obviously, they are even behind the other countries/regions if wins have not even been announced yet.

UT revenues - As I mentioned in the last post, if $210m of the $315m total revenues for the MMCBU comes from PAS infra, then only $105m is expected for iptv/ngn. IPTV users will increase by about 600k users for UT. That is $75m in STB. Infra/service/maintenance have longer revenue recognition cycles. If we consider half will be recognized, that could be $12m. NGN is about $20m, maybe more if PAS infra is less.

2009 iptv growth for UT - If UT doubles their iptv subscriber base, they could end 2009 with about 2.5m subscribers. That is a growth of 1.25m (about 1m from China and 250k from outside China). IPTV revenues could be about $156m in STB and $50m in infra/service/maintenance for a total of about $200m.

Uncertainties and potential - Even with the growth in iptv in 2009, the number of iptv users in UT markets will be very small. Based on the continued opex for UT in the second half, management has indicated it will spend to further their markets in iptv, NGN, and broadband. The company had initially projected about $1.1b in revenue for 2008 at the start of the year and now is down to about $900m (numbers adjusted for PCD sale). The company needs to further cut costs while ramping revenues by several hundred million. It seems iptv can add about $100m for 2009. NGN and broadband will show good growth as well but definitely not enough to get to profitability in 2009. On the bright side, the company has built up a nice cash base (ending 2008 with over $300m in net cash), cutting expenses, and potentially selling additional assets (CSBU). More importantly, the iptv revenues (while still lumpy) has significant growth, strategic wins, and worldwide traction behind it.

Quantifying the shortfall

Fellow shareholder Shadow tried to quantify the $100m cash flow shortfall for 2008. Normally, I would just reply on the yahoo board but this is a good discussion so I will post my reply on the blog. Based on $60m in cash usage used in preparation of the phase II BSNL broadband contract, Shadow focused on the $40m cash shortfall as follows:

"I gathered some numbers together from UTSI's 10Q and Tim's blog with the following results. Handset sales (PAS plus China CDMA/GSM) in H1 2008 were around $93M and for year are expected to be $155M indicating H2 sales of only $62M or $31M per quarter. I previously estimated that H2 resulted in an unexpected cash loss of an additional $40M to Analyst Day estimates by management and a loss of $135M in sales (initial sales were to be $1.035B and were revised per Tim's Blog now down to $900M). If you assume H2 handset sales were to match H1 sales then lost sales of $31M are now predicted. At 36% gross margin this would account for gross profit loss of $11.2M in gross profit from this division. If you estimate PAS infrastructure sales loss/delayed recognition of $30M with margin of 45%, you get a loss of another $13.5M. Using the sales constraint above, $135M total sales lost in H2 - $31M (handset division) - $30M (PAS infrastructure) = $74M of lost sales of CDMA handsets due to lost PCD contract. This results in loss of $11.1M in gross profit since gross margins for this product were given to us at 15%. So, total lost gross profit with above assumtions would be $35.8M which is close enough to the $40M in lost cash to be an acceptable solution."

I had commented previously that I was surprised with the $755m in non-PCD revenue expected for 2008 that was in the Analyst Day meeting. Using the mid-range values given late last year, I had come up with $833m. The initial company target for 2008 for the terminals business unit was $210m with a 22% gross margin so the $155m number that Barton mentioned to me a week ago was responsible for the bulk of that. The increase in PCD revenue noted in the Analyst Day meeting made up for the lost TBU revenue but it was also a bad sign already that I should have pursued more. The second item was during the cc to discuss the PCD sale. We learned that the internal handset part of the PCD would have $280m in revenue as noted in the Analyst Day slides. However, Barton was hesitant in backing this number during the CC and only mentioned that it was in the $200m something range. Again, this was something that could have helped in identifying the revenue shortfall.

Anyway, back to Shadow's calculations. I think the PAS handset contribution revenue wise is a good estimate but the gross margin of 36% is on the high side as the company knew even at the beginning of the year that margins would be lower (hence overall 22% GMs). This may lead you to increase the PAS infrastructure contribution of the loss but a large part of the shortfall in cash flow stems from the PCD distribution sale. Also, the $60m+ in TBU sales left might even be less than what you are estimating. The company did not specifically mention PAS infra regarding the cash flow delta in the earning call except in the conversation I had and that was described as "some" so the $30m you estimated is probably on the high side. This is a good discussion because of the importance still of PAS as it still generates about $210m + $155m or $365m in revenue for the company (assuming those numbers are still correct for 2008). I am a little reassured during the conversation that Barton emphasized not to extrapolate the PAS handset declines into the future and that Barton spent time in emphasizing that the PAS infra have much longer time horizons. Blackmore added that the telecom merger impact should help past this short term pain.

MCBU insight - I was going to post on iptv subscribers/revenues and the above discussion is a good backdrop to the MCBUs revenue stream. At the start of the year, the estimate for the MCBU was $315m. Since PAS infra was at $210m, only $115m or so is left for iptv and NGN. This might be more if the PAS infra number that Barton gave me did not include the expected shortfall. I'll continue this on the next post.

In summary, management lost credibility in not identifying the shortfalls in the Analyst Day meeting/slides. Maybe they thought no one followed the company anyway. Maybe they got complacent since they were going to sell the PCD and generate a lot of cash. The 27% one day drop after the earnings call only emphasized that the company has to communicate guidance and manage expectations much better. I had already emphasized my frustration for the discussion I had with Barry Hutton when we were discussing the guidance given in the Q1 earnings call for Q2. I want to point out that our conversation was when the stock was well above $5. While most shareholders will agree with the company's focus on the long term, the company cannot dismiss a drop from almost $6 to $3.5. Can you imagine the uproar of Apple shareholders if the stock drop to $90 in the next two weeks due to some short term issues?