Omnivision Technology
Introduction, Replies, and Analysis

Format for Printing

Format for printing

Request Reprints


By wishiwaspsychic
March 28, 2005

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

OmniVision Technologies, Inc. (OVTI)

The company profile: OmniVision Technologies, Inc. designs, develops and markets integrated semiconductor image sensor devices. The Company's main product, an image sensing device called the CameraChip, is used to capture an image in a variety of consumer and commercial mass market applications, including digital still cameras, cell phones, security and surveillance cameras and video game consoles. CameraChips are manufactured using the complementary metal-oxide semiconductor (CMOS) fabrication process, a method of producing integrated circuits. The CameraChip integrates a number of distinct functions, including image capture, image processing, color processing, signal conversion and output of a fully processed image or video stream. OmniVision also provides companion chips used to connect its CameraChips to various interfaces, including the universal serial bus. In addition, the Company provides companion chips that perform compression in standardized still photo and digital video formats.

The story:

OVTI is the unquestioned leader in the CMOS market, having garnered approximately 50% market share over the past 18 months. Historically, CMOS sensors have competed with CCD sensors. The former offer more efficiency and better pricing, and have dominated the lower and mid markets. The latter have offered higher quality but with a premium price, and have dominated the high end of the market. Now, this appears to be changing. In late February, the company announced that it had eight design wins in the digital still camera and video camera market.

These design wins represent the first uses of a 5-megapixel CMOS image sensor to process both still and video images in mass market consumer devices. If OVTI and its CMOS sensors can begin taking significant share in the high end of the market, this would give the company a huge boost.

Camera phones remain the company's bread and butter. OVTI has deals with 4 of the 5 major phone producers (the lone holdout being Nokia...and there are rumors that Nokia will come on board). The company reportedly sold some 60-70M units last year (various models), for about a 50% share of the market. This year, the total market for camera phones is expected to increase to 200M units, and in 06, the market is expected to increase to 300M units. If OVTI maintains a 50% share, the future is amazingly bright. But even if its share drops to 40%, in 2006 this would still equate to 120M units shipped, as opposed to 60-70M units for calendar year 04! There will be competition, pushing down ASP a bit, but 100% top line unit growth can withstand some pricing pressure pretty easily. To stay ahead of the field, Omnivision took a calculated risk and retooled its sensor architecture and launched its new line of sensors last July. There have been lingering doubts as to how successful the company would be with this new line, but a release yesterday on a major design win for the company's new line 1.3 MP sensor should help squelch those fears. OmniVision Technologies, Inc. a leading global supplier of CMOS image sensors, announced today that it has secured a major design win to supply its OV9650 1.3 megapixel CMOS image sensors for use in upcoming, high volume camera phones of one of the world's largest handset makers. This follows a series of design wins for this sensor with other handset makers and marks a milestone for this product. "Last year, VGA cameras clearly led the camera phone market, but this year, 1.3 megapixel cameras are quickly gaining popularity," said Jess Lee, OmniVision's Director of Product Marketing. "These successive design wins of our OV9650 underscore our belief that 1.3 megapixel cameras will join VGA in dominating the camera phone market in 2005."

I am not an industry expert, but it seems to me that the transition to the next generation of sensors is going pretty well for OVTI. For more evidence, I look at earnings results versus expectations. If the company was having major issues with the adoption of its new line, I would expect that figures would come in below analyst estimates. But for the last three quarters--from the launch of the new line to now--OVTI has beat earnings expectations by 6.7%, 16.7%, and 22.5%. But the stock is trading at $16.60, far below its 52 week highs, and well off the +20 level reached a few months ago.

If OVTI were only a player in the camera phone market, I would still be mildly bullish based on the growth expectations in that market. But the reason I feel OVTI offers the potential to be a huge gainer (aside from valuation) is that the company has several other markets for its CMOS sensors that could generate revenues equal to, if not greater, than that of the camera phone market. Currently, some 15% of the company's revenues are derived from the security/surveillance market. As the CMOS technology provides higher resolution via the 5MP products, and remains much cheaper than CCD technology, revenues from this area should exhibit solid growth. I also expect that the size of this market will show strong growth, as the nature of today's world and affordability of the products will make security increasingly important.

OVTI is also making a splash in the automobile market. Some may be aware of ads touting new safety features concerning lane change warnings and emergency breaking prior to collisions. Basically, cars are being loaded with tiny cameras (~4-5), and when the car changes lanes, the cameras ensure that there is not a vehicle in the blind spot. Similarly, if for some reason a driver is incapacitated or does not see that he/she is going to run into something, the camera can be used to trigger an emergency brake to stop or limit the severity of a collision. OVTI already has a deal with a premier manufacturer for 06, and this article provides more insight into the opportunity. "OVTI today launched its first image sensor designed specifically for the automotive market. The OV7940 is a new high-performance, highly integrated analog CMOS image sensor based on OmniVision's proprietary OmniPixel(TM) technology. This true single-chip solution can be used in numerous automotive driver assistance applications.

The size of this market looks to be major (especially with higher ASP):

Market analyst firm Techno Systems Research (TSR) projects that application volumes for CMOS image sensors will surpass those for CCD sensors next year, and that by 2008, 20 million CMOS sensors will be in use in the automotive market(a).

OVTI also hopes to enter the medical market with its host of uses for cameras, though I believe this will be a tougher market in which to gain acceptance. The point remains that OVTI has incredible growth opportunities, and though it faces competition and uncertainties, the stock's ridiculously cheap valuation makes it a highly attractive investment.

OVTI valuation:

This is where things get really interesting for OVTI. But before I get into any numbers, it makes sense to explain the major catalyst(s) that have landed OVTI with such a wonderfully cheap valuation. The major catalyst was the company's disclosure that internal and external auditors were looking into the company's accounting practices. This announcement resulted in a 30% haircut, 26 to 17), and a flood of law suits followed, bringing the stock under more pressure. When the company announced August results and lowered guidance because of the transition to its next generation of sensors, the stock made its way into the single digits. I first became aware of the stock after the accounting investigation announcement, and I first became an OVTI purchase after the stock was crushed based on the downward revision in guidance.

In hindsight, the company's accounting problems can be viewed as a non-issue. In the most recent 10Q, the company laid out the problems. First, beginning in the second half of fiscal 2003 and continuing through the first nine months of fiscal 2004, certain distributor sales, for which we recognize revenue on a "sell-through" basis, were not reported to us by one of our distributors in a timely manner. Additionally, in the second and third quarters of fiscal 2004, during the transition of testing operations and certain international sales functions to overseas locations, some shipments made to customers late in the quarter were incorrectly classified as transferring title upon delivery as opposed to upon shipment, and therefore revenue was not recognized when product was shipped. Both of these issues resulted in delayed revenue recognition. Both management and the Audit Committee determined that the errors did not have material effects on quarterly or annual financial statements for fiscal 2003 and, consequently, decided not to restate the quarterly or annual financial results for such periods. What happened for the periods that OVTI did restate earnings? In total, the restatements add $8.9 million in revenue and bump up net income by $2.7 million, or 4 cents a share, over previous totals.

So the restatements actually increased earnings, and during the last CC, the company's CFO commented that the restatements did not change the company's cash flow for the year. The fact that the stock has yet to recover from the initial announcement of the audits should, in my mind, be viewed as an opportunity, as the stock was hammered because accounting fears and competitive fears, and the former has been squelched.

Back to numbers....

-OVTI has a maket cap of $940M

-As of the end of last quarter (Jan.), the company had $315M in cash and equivalents and no debt. That's right, the company is trading for less than 3x its cash position. And with cash on hand, there is simply no possibility that the company is pulling any kind of Enron shenanigans to artificially inflate earnings.

-In the first three quarters of fiscal 05, the company has had cash flows of $100M, so that gives a run rate of ~$133M. On an enterprise value to cash flow basis (not FCF), this gives a ratio of 4.6, which indicates a very, very cheap stock. FCF for the trailing twelve months was $110.8M, giving a EV/FCF ratio of 5.5 (still unbelievably cheap).

-The company's trailing PE is 13, and if the cash and equivalents are backed out, it is 8.7. Analysts estimate a growth rate of 25% for the company for the next five years, giving a PEG ratio of .51 for the company. Compare this with an average PEG ratio of 1.42 for the semiconductor industry (which clearly includes a number of different type of chip/sensor producers, but has few companies that specialize in areas with expected growth rates anywhere near that of the camera phone, security, and automobile sensor markets) and 1.51 for the S&P 500.

-Given the rapidly changing nature of its industry, it is impossible to determine anything close to an exact growth rate for OVTI. Therefore, I have chosen a number of ranges for 5-year earnings growth estimates, and have run some rough discounted cash flow calculations for each. After the 5-year growth period, I have assumed that earnings will level off to 3% growth, and I have used an 11% discount rate. To come up with the total value of the stock, I have added cash on hand back to future earnings (also, note that the growth rates are meant to account for dilution).

5% growth terminal value = $23.32

10% growth terminal value = $27.35

15% growth terminal value = $32.13

20% growth terminal value = $37.76

25% growth terminal value = $44.35

30% growth terminal value = $52.03

35% growth terminal value = $60.92

40% growth terminal value = $71.17

Given that the camera phone market, in which OVTI has commanded a 50% share, is expected to go from some 140M units in 04 to 300M in 06, and the advance of CMOS sensors into the high end of the market, I think that 20-30% 5 year growth estimates are reasonable. This accounts for some decline in ASP and market share erosion, which will be made up for in sheer volume. Also, as I mentioned before, OVTI and its CMOS sensors have wonderful growth prospects in security, automobiles, and the medical field. Smaller markets, like the smart toy market, will be the icing on the cake.

The reason I am so high on OVTI is that it offers the rare combination of a high margin of safety along with the potential to be a true growth play. For those who like to look more at relative valuations, think about this--if OVTI grows earnings at a 20% clip, and trades with a PE of 20 (for a mildly conservative PEG of 1), the company would trade at over $30 based on next year's earnings (ttm earnings of 1.27 x 1.2 x 20 = $30.47). And this ignores the company's substantial cash position!

Omnivision does face challenges in growing on a forward basis, but I believe that at current prices the stock offers considerable upside with very little risk to the downside. If OVTI trades much lower, a larger semiconductor company would likely become very interested in picking up the company.

OVTI issues/concerns:

--The SEC investigation. As is often the case in the post-Enron era, anytime the SEC is mentioned in reference to a stock, it is assumed that the company is headed nowhere quickly. At times, this assumption is correct, but in other cases, it is far from true. For OVTI, I don't believe the SEC's involvement is a major issue. Why? A) The company's restated earnings went up, not down. If management was attempting to dupe investors as to the true operational performance of the company, this would make little sense. B) The company's cash position makes major fraud a virtual impossibility.

--Inventory concerns. In its last earnings report (issued at the end of February), the company noted that inventories were at $67M. This represented a significant increase in inventory, and bears jumped on this increase as evidence that the company was having troubles moving its product. Personally, I view the inventory increase as little more than a minor issue. The inventory is entirely composed of Omnipixel products--i.e. it is all new generation sensors, so obsolescence is not a concern. Also, there are a myriad of external sources that have discussed the growth of the camera phone market. OVTI must have product on hand to meet new orders (the company does not simply accept an order and then go into production), so it makes logical sense that when the company expects demand to grow, it will prepare accordingly. During the CC, Leigh (the CFO), admitted that the company had expected demand to ramp up faster than the reality, but with the announcement of the 1.3 MP design win with one of the world's largest handset makers greatly reduces inventory concerns. It is still unclear as to whether or not this deal might be with Nokia, but regardless of the maker, it is very good news.

On a side note, 2/3's of the inventory was work in progress, or WIP. The concept of inventory divergence argues that an increase in raw materials and WIP should be construed as a positive signal, as it indicates a company stockpiling to meet future demand. On the other hand, a large increase in finished goods is considered to be a negative signal, as it signals that the company is having trouble selling as much product as it expected. Based on this concept, OVTI's inventory build appears more likely to be a positive signal than a negative one.

--Weak forward guidance. OVTI beat estimates in the Jan. quarter by 22%. For the April quarter, OVTI guided for EPS of .29-.34, above analyst expectations of .28. However, the company's revenue range of $93M-$103M did not impress analysts, who had called for revenues of $102M. During the CC, the company explained this "issue," noting that the company unexpectedly booked $3-5M in revenues that had previously been expected to fall in the next quarter. If this ~$4M had fallen in Q4, the revenue range would have been $97M-$107M, and the company would have transferred .03 of earnings surprise to the following quarter. I know I have mentioned OVTI's recent 1.3MP design win repeatedly, but it bears repeating once more. When the company gave guidance, it could not assume this design win. Depending on when shipments for this deal begin, this quarter or next could see solid increases in top and bottom line growth.

--OVTI is in a commodity business, and will not be able to grow earnings because of declining ASP's. OVTI does make sensors. It is in the semiconductor industry. The camera phone market may eventually resemble a commodity business, with gross margins well below OVTI's current 40% levels.

That said, I believe this day is still far away. I like to compare the camera phone market to that of the personal computer market in its younger days. In the early 90's, the differences in computers from year to year was so big that one was compelled to buy a new one regularly. Today, computers can be used for several years, as the basic computer fulfills most uses adequately. For anyone who has seen the difference between a top of the line camera phone or video phone with a bottom of the line, it is clear that they are two entirely different beasts. The rate of innovation is high, and until the products reach a level at which improvements do not make a fundamental difference in the utility garnered by the phone users, this will not be a commodity business. OVTI is more nimble than its biggest competitors (Micron, Agilent), and the company has--to this point--remained a step ahead of the competition. There will always be concerns as to whether the company will maintain its dominant position, but that is the case with just about any business.

So while ASP's will decline, I do not buy the claim that OVTI is playing in a commodity industry. These are not eggs, they are highly complicated sensors, and different companies have wildly divergent success with regard to yield and architecture efficiency. OVTI's management has made it clear that the company has a plan to maintain gross margins even as ASP's decline, by decreasing wafer prices, improving yield, and improving geometry (not too sure as to the technical aspect of the last method).

And lest we forget, if OVTI is successful in the automobile market, this market should be far more "sticky" than the camera/camera phone mWearket. Once manufacturers have adopted a product and integrated into the production schedule, they will be reticent to completely redesign because another company offers a slightly better price.


If OVTI receives some closure from the SEC and reports earnings and guidance above expectations, I believe the stock is poised to make a nice move up. I'd love to hear what others think.



Become a Complete Fool
Join the best community on the web! Becoming a full member of the Fool Community is easy, takes just a minute, and is very inexpensive.