Seeking Alpha
View as an RSS Feed

Valuable Insights  

View Valuable Insights' Comments BY TICKER:
Latest comments  |  Highest rated
  • Clean Diesel Technologies Should Be A $5 Stock On Its Way To $10 [View article]
    Could you walk me through your financial analysis that suggests that $5 is fair value? EV/EBITDA, P/E? The $5 appears to be completely arbitrary with the premise being, it's a low float stock in a space where a lot of stocks have run - let me toss out a big premium (even better, "on it's way to $10") with no basis other than it sounds good.

    I recognize that maybe (then again maybe not) there is a big opportunity ahead, but its important to set price targets for reasons other than it feels/sounds good.

    If I somehow missed the valuation analysis, please point me in the right direction.

    Thanks - VI
    Jan 9, 2014. 06:36 AM | 5 Likes Like |Link to Comment
  • Valuable Insights Positions For 2014: Still An Abundance Of Unique Ideas In Small Caps [View article]
    Hi RP,

    To your first question, yes, I've been wrong many times and hold on to my mistakes and beat myself up about them. I fiercely hate getting it wrong, so I am extremely conscious of where I've been in error, and how best to avoid repeating such errors. I believe much of my investing philosophy reflects what I've learned (sometimes painfully) over the years. I believe my suggestions are useful to all investors, or at a minimum thoughtful and intelligent.

    I don't want to get into a long dialogue on Phil Frost in this forum, perhaps in a future article. There are people emotionally attached to Phil Frost, despite compelling evidence, it's hard to convince them they're wrong - notwithstanding his prior successes.

    There is plenty that's been written on him - much of it available for free on SA and extremely informative. To be blunt, from a value investing (or sanity) perspective, his behavior makes no sense. And since he keeps buying shares (I believe to prop up OPK and other stocks, to make them appear appealing) he kinda, sorta, isn't really using other people's money, is he - although when he eventually cashes out (if he can find away), it's likely he'll have done it at your expense.

    I don't know his motivation. My best guess he's going to donate his inflated stock to charity or put it in some trust to avoid/limit estate taxes. Or he's engaging in some perverse social experiment to see how blindly people will allow themselves to be led to the slaughter by letting others think for them.

    I may be wrong here - pretty sure I'm not - as I mentioned at the outset, I have been before. Still, too many interesting longs with managements that are aligned and at least trying to do the right thing, whether they succeed or not.
    Jan 8, 2014. 01:42 PM | 5 Likes Like |Link to Comment
  • Veeva Systems: SaaS's Biggest Bubble And Tech's Best Short [View article]
    Just to add one additional thought: Having worked in an investment bank we can unequivocally confirm that sellside analysts have a tendency to receive information from various sources (company slides, S-1s, market research orgs, etc.) and then do cut and past jobs (excel copy down and right). They assume growth rates and take management projections at face value. There is a lot of top down analysis and not much bottom up. The rare sell-side analyst does the work.
    This is not to denigrate the sell-side. They have a variety of pressures: meeting with clients, conducting field research with companies they cover, writing research, getting on earnings calls, etc.
    Suhail Capital says they spent a better part of a month putting this together and it still is not 100% perfect. The sell-side doesn't have that luxury (for all the fees they receive, maybe they should hire someone when they have a live IPO to do the research).
    So, all this being said, without a question, I'd rely far more on the data provided in this write-up than anything from the sell-side. They have all the motive in the world to go with management's numbers, and little time to do the necessary work to challenge it.
    VEEV looks like it would be expensive if the market/projections were as big as they claim, but it looks completely absurd given where the market size likely is.
    Disclosure: Short
    Dec 4, 2013. 08:24 AM | 5 Likes Like |Link to Comment
  • Camtek: Printing A 3D Pump And Dump [View article]
    Tom, nice work.

    So, we've personally visited approximately 20 printed circuit board facilities over the past 15 years. The equipment is large, takes up rooms, and utilizes a wide range of materials and chemicals. Beyond the complexity of fabrication, there is a great deal of time, precision and expense involved in testing.

    With Camtek coming from a standing start (i.e. ZERO in 3D printing revenue for PCBs), it is inconceivable that there will be material revenue from this segment in 2014 or likely 2015 - if EVER.

    Go to any website that details PCB fabrication - or call up TTM Technologies (TTMI) or others and request a site visit.

    People just want to believe instead of doing some homework and using common sense.

    Nov 28, 2013. 01:10 PM | 5 Likes Like |Link to Comment
  • The Big Winner From Intel's Analyst Day? ARM Holdings [View article]
    Wow, very meaningful article.

    Ashraf, hold no position in ARM right now, but you know where we've stood on this ( Our point is not at all to gloat. Our point is to congratulate you - NOT for coming around to our view, but for NOT sticking to a view after it's clear that the facts are different than expected. Best possible thing an analyst can do is adjust their view if the facts change or if they discover they're wrong (too many analysts simply change the thesis - or become creative with their interpretation of data points). Better to lose a little face now than to lose longer term credibility. And the fact is, you're obviously highly regarded throughout the SA community.

    Anyhow, never boring in compute land. Thanks for the great work.
    Nov 22, 2013. 02:18 PM | 5 Likes Like |Link to Comment
  • Yes Mrs. Yellen, QE Is Creating An Asset Bubble [View article]
    Come on Akram, let's be cynical and let's be real. She doesn't take the job, she's Janet Yellen. She takes the job and, worst case (and not unlikely case), is the big fail, she gets paid $100k per hour per speech for the rest of her life.
    Today, with the nature of celebrity, there is NO downside EVER to being a high profile political figure. Geez, Palin and her pathetic family will be rich for the rest of their sorry lives despite non-existent IQs.

    Yellen may not (probably will not) succeed as Bernanke, but taking the job is an absolute no brainer win.
    Nov 18, 2013. 10:07 PM | 5 Likes Like |Link to Comment
  • Chipmos: Shares Have Over 100% Upside And Near-Term Catalysts [View article]
    We wanted to address all the comments and private emails in a single response.

    First, thanks for the very favorable feedback. We wish we had ideas like this on a weekly (or monthly) basis, but it's difficult to find names where the facts and set-up are so good. Even following the company's ~30% run since reporting earnings (a miss), shares are still under 2x EV/EBITDA (and there is ZERO net debt).

    Based on several informed comments, it sounds like the listing process is slightly different than we had suggested. Specifically, the company should list on or around April 19th on Taiwan's Emerging Stock Market. It will be a relatively small number of shares. However there will be the potential for additional share sales from underwriters in negotiated transactions. With greater float, there will likely be near-term coverage, even before an uplisting to the Taiex, which would presumably have a very positive impact on valuation. It sounds like an uplisting might not occur for 10-12 months, rather than the 6 we expected, however we view this as a small negative as coverage and liquidity are the key catalysts to achieving appropriate valuation.
    It also sounds like Chipmos may be able to take proceeds from Taiwan share sales to use on additional buybacks of Chipmos Bermuda shares, which would help ensure the closure of any valuation gap. If there is a reader with definitive information on this point, we invite your comment.

    Regarding Keltus' comment on free cash flow versus EPS, we think Hewitt's comment is more than sufficient. ChipMOS is ridiculously conservative. They easily could have reported non-GAAP EPS last year, showing eps excluding FX losses, but they didn't. They could have written off excess equipment, thereby lowering depreciation, but they didn't. Accounting involves many artificial constructs (which we studied ad nauseum during our academic career) and we'd suggest that investment bankers/private equity types rely far more on cash flow and EBITDA metrics than EPS. Of course, perspective like your creates opportunities for investors like us.

    Regarding lower y/y EBITDA, if you break COGS into cash COGS and depreciation related COGS, GM on a like-for-like basis was down about 50bps y/y which is largely attributable to mix, offsetting slightly higher y/y revenues. Operating expenses increased about $2 million y/y. This is essentially the decline.

    While it is almost incomprehensible that any company not on the brink of bankruptcy could trade at 1.5x EV/EBITDA, we posit several reasons.
    First, Chipmos did in fact come from the brink of bankruptcy in 2009. They did pay down $500 million, but the value has no accrued to equity - but given where the company came from it is norton many radar screens.
    Second, Chipmos missed full year revenue guidance of growing 10%. They also disappointed a bit in terms of gross margins expanding as much as hoped for. This strikes us as the ultimate case of investors missing the forest for the trees. If you could buy a million dollar home for a quarter million, and then learned that the door knobs need to be replaced, you'd still buy the home. Investors are laser focused on things that are noise in the scheme of how cheap IMOS is in the big picture. We give Chipmos' 2 largest shareholders, DLS and Chilton a lot of credit for sticking it out through all the noise and understanding what matters.
    Third, there was a big sale of Chipmos shares by SPIL and Thailin in late October - we write about it here This sale was entirely confusing and essentially not explained by management. It was however a necessary step for a Taiwan listing.
    Fourth, management has been excessively conservative. Any company with a 25% free cash flow yield should be buying back share. Chipmos wanted to get to net debt neutral, they are now, and a new buyback kicks in in the next 2-3 weeks. We expect MANY more if the stock remains significantly undervalued - and under $20 is significantly undervalued.
    Fifth, liquidity is difficult for most funds. A half billion dollar hedge fund that wants to build a 1-2% position without making up more than 10% of daily volume would spend 1-2 months. We expect the move from $20 to $30 to be much faster than from $10 to $20. Check out the IMOS chart from September 2003.
    Lastly, typical Wall Street herd mentality. With 1.5x EV/EBITDA something has to be wrong, it has to be a fraud, no one owns it. We'd speculate that at 3x, when the chart starts looking good, and fund managers can name 5 other "smart" funds that own it, it will be a great, cheap idea. Well, we're calling it at 1.5x - we like being early and being right.
    Mar 31, 2013. 09:50 PM | 5 Likes Like |Link to Comment
  • Dear Apple Board, This Is How You Defend The Share Price [View article]
    I advocated not only a massive buyback, but a huge debt issuance (@2-3% interest it's a no-brainer). I suggested this on seeking alpha 11 months ago
    The perception that it will end Apple's innovation is not an excuse for poor capital allocation. It should do it, because it is smart, accretive, and shareholder friendly to do so.
    Steve Jobs was notorious for hating debt. He was also a visionary. Hopefully, he would have had enough vision to recognize the unique time in the credit markets in which high quality companies are being offered near free money.

    Also, you spelled Jeff Gundlach incorrectly in your 2nd sentence.
    Jan 15, 2013. 03:09 PM | 5 Likes Like |Link to Comment
  • ChipMOS Technologies Up 54% In 2 Weeks [View article]
    Bill, I respect that you are making your call based on technicals, but I think it is misleading to write that there is a "paucity of public information."
    At best, this is inaccurate, at worst, this can be construed as very negative and self-serving to your now short position.
    ChipMOS issues monthly sales data, files in the US, has large customers such as Micron and Spansion, and has an updated presentation on its corporate website.
    Investors interested in reading about ChipMOS' fundamentals should read my recent articles.
    Incidentally, I did not write following last Friday's call, but folks should know that the company could have free cash flow of $5 per share in 2012, vs $4.18 in 2011, and be net debt free by the end of the 3rd quarter - meaning shares trade at 3.7x FCF. At a very reasonable 8x FCF we have a $40 stock. You are right, this can be a runaway freight train.
    Again, I respect your investment methodology, I just think there is plenty of info on ChipMOS if one spends very little time and makes a small effort.
    Mar 21, 2012. 07:21 AM | 5 Likes Like |Link to Comment
  • Chipmos: Shares Have Over 100% Upside And Near-Term Catalysts [View article]
    Despite the run in ChipMOS over the past several years, shares remain cheap on both an absolute and relative basis. With an EV/EBITDA of under 5x and a double-digit free cash flow yield (the company is and has been an FCF machine) there is a substantial margin of safety - which explains why despite many iffy quarters, the stock is up almost 5-fold over the past 2 1/2 years.
    The company reports earnings Tuesday after the close. This may be the last chance to get shares on the cheap, although in our view, the play is over the next 6-9 months and not just for the quarter. That said, the stock is cheap, and the catalyst should be the biggest positive earnings revision in a very, very, very long time.
    On Friday ChipMOS posted July revenue of $64.7 million, the best month since October 2007, the second best month in company history, an 8% m/m improvement, and well above our expectation. And despite a rallying market, the stock was down. Frankly, that's one of the things we continue to like about IMOS (despite the occasional frustration) it's under the radar screen and provides an unusual opportunity.
    Assuming typical seasonality (i.e. modest improvement in both August and September) 3Q revenue should be $196-200+ million, well above the consensus of $186mn. Management could set the low end lower, but that would in all likelihood reflect typical conservatism. Given June revenue declined from May, it is probable that July improved over the course of the month, meaning there is a good deal of momentum entering August.
    Typically, a low multiple stock with a big upward guide leaves 1) a lot of room for shares to run, and 2) potential for significant multiple expansion.
    But as we have written about in multiple prior articles, this is not a stock about a given quarter, this is a well-positioned, inexpensive company that should see meaningful growth over multiple quarters. With the advent of 4k2k screens, and growing complexity in cellular handsets, ChipMOS' driver IC packaging/test (duopoly with Chipbond) should enjoy rapid growth. Notably top customers Novatek and Himax guided to double-digit q/q growth for 3Q. And the memory segment, with 18% customer Micron is enjoying well-noted stability in the first time...perhaps forever. In addition, NOR flash customers such as Winbond, Macronix and Spansion are once again growing and outsourcing more test (almost 100% incremental GMs). ChipMOS will likely also discuss new Toshiba (high-margin) test business when they report earnings.

    Beyond being well positioned, ChipMOS is likely to discuss consolidation of their operating entities into a single, simpler corporate structure. If they don't discuss it on their call, then they should sometime after their late August board meeting. This was discussed in a recent article by SA author Jaret Wilson - we believe such an event would result in a reduced sharecount, with the company eliminating a significant number of shares via a recent $300mn+ loan at 2% (of shares yielding FCF >10%). In the likely event the company pursues such a strategy, we'd expect meaningful and pronounced share appreciation.
    Aug 10, 2014. 10:59 PM | 4 Likes Like |Link to Comment
  • Crossroads Systems: Valuable IP And Niche Product Could Lead To Significant Upside [View article]

    Please see our comment from 6/29. Our best guess as to why Eberwein exercised early (in addition to his publicly stated filings that shares are undervalued) is that additional funds via option/warrant exercises from Eberwein and others can be used to pre-pay the Fortress debt. Paying this debt eliminates costly interest and also takes away Fortress' participation in the monetization of the non-972 patent portfolio. If Crossroads genuinely believes that non-972 is worth $10s or $100s of millions or more, then paying off Fortress asap is logical and consistent with CRDS' claims. It wouldn't be surprising if we learned of additional warrant/option exercises by other holders when the co next reports earnings.
    Jul 13, 2014. 10:33 AM | 4 Likes Like |Link to Comment
  • Crossroads Systems: Valuable IP And Niche Product Could Lead To Significant Upside [View article]

    To be clear, it could be far more than $20mn. The point is, that even on $20mn (we'd be disappointed if, over time, it's not a good deal more), EPS is $0.66. We have a view on how the mkt. tends to value 80% GM storage businesses with strong IP and a healthy recurring rev component. And this EPS excludes potential incremental royalty streams from both '972 and non-'972 IP.

    Jun 28, 2014. 11:53 PM | 4 Likes Like |Link to Comment
  • A Huge Gross Margin Impact That Everyone Is Missing! [View article]
    Pinnacle - completely agree. Intel's record in forecasting is a mixed-bag at best. If they were a small company, they would get NO credit for their predictive powers, but because they are the all-mighty Intel, they still have plenty of apologists. Very good chance they will get it wrong.

    Last year prognosticators were calling for 2013 to be the year Intel got going in Mobile. Now they are donating chips to gain share in 2014 (a margin enhancing strategy if I ever saw one), with the real business arriving in 2015 (are they going to get share at Apple or Samsung? Doubt it. So what percent of high-end remains, and what percent will the get of that). Not a lot to get excited about. 'bout that foundry strategy (thesis drift anyone?)

    For anyone excited about Intel and mobile, a brief excerpt from our article from last November 2, 2012

    A quick review of Intel's history in mobile:

    2006: In June Intel sells XScale to Marvell (MRVL) for $600 million in a deal that closes November 2006. Xscale had achieved notable wins in PDAs with Hewlett Packard (HPQ), Dell (DELL), and Palm. Of course, XScale used ARM architecture.

    2007: Presents Menlow and Moorestown at the Intel Developers Forum. We're not aware of any major (or minor) platforms ever brought to market using these chips.

    2008: The atom processor is introduced (Silverthorne), with a focus on mobile devices. It does not gain traction in mobile, but is adopted in the netbook market.

    2010: Intel and Nokia rouse the audience at Mobile World Congress with the amazing MeeGo OS. This foray into developing a mobile operating system is put to sleep..permanently, in 2011.

    2011: Medfield is announced, with samples to come in 2012.

    So of course Intel could get it right, but history is not on their side. And while we recognize the Medfield/Clover Trail SoC could gain traction, this is not just about price/performance. ARM has an established ecosystem with well capitalized licensees including Apple (AAPL), Nvidia and Qualcomm (QCOM). It simply won't be easy to disarm, ARM.
    Dec 23, 2013. 12:32 PM | 4 Likes Like |Link to Comment
  • Micron Roadmap: Evolution Risks, Financials, Valuation And Emerging Trading Dynamics [View article]
    Just one more point - on point 1 (and I will stop with this...and I really could keep going). I can say with complete certainty (and by complete certainty, I mean COMPLETE CERTAINTY) that exactly NONE of the upward pricing the "will happen" in Micron stock over the next 6 to 9 months already took place.
    Dec 16, 2013. 07:02 PM | 4 Likes Like |Link to Comment
  • Taiwan Listing Of ChipMOS In 2014 Set To Unlock Value [View article]
    Nice first article.

    Regarding free cash flow yield, it is likely higher than you state on a normalized basis. Management has commented that next year cap-ex will decline to about $70 million, which should result in FCF of $100+ million or $3.50-$4.00 per share or 17-20% FCF yield. If excess cash were used for repurchases, the yield would be higher (20%+).
    On a normalized basis, if cap-ex is ~15% of revenues, FCF would still be $70-$80mn, which, again assuming share repurchases, would result in FCF yield of 13-15%.
    Any way you cut it, shares are still substantially too cheap.
    Dec 2, 2013. 07:27 AM | 4 Likes Like |Link to Comment