Skip to content

Posts from the ‘Radcom (RDCM)’ Category

Week 294: It doesn’t matter how you get there

Portfolio Performance




Top 10 Holdings


See the end of the post for my full portfolio breakdown and the last four weeks of trades

Thoughts and Review

It’s a seminal moment for the blog!  For the first time in what seems like forever my largest position is something other than Radcom.  Thanks to more than doubling in price in the last four months (and even after pulling back from $6 to $5), Identiv now holds that honor.

At the beginning of November I wrote the following about Identiv:

I tweeted a couple of times this morning that I don’t think this stock makes sense at a $20 million market cap… The company has a $55 million trailing twelve month revenue run rate, they are showing growth, they are EBITDA positive now and it’s not an insignificant amount of EBITDA.  That feels like it should warrant at least 1x sales.

We are already at a $55 million market capitalization but with momentum at the company’s back I haven’t sold a share.

A second position, RMG Networks, has also ran up the ladder, and now sits as my fourth largest position at a little less than 5%.

I wrote this about RMG Networks when I first took the position in late June:

With the focus on the new verticals and improve productivity of the sale force new opportunities in pipeline are up over 40%.  And here is where we start to see an inkling that the strategic shift is bearing fruit.  In the sales pipeline, Michelsen said that the number of deals $100,000 or greater has increased by 50% in the last year while the number of $1 million deals have tripled…My hope is that these early signs of sales improvements lead to an uptick in revenues in short order.

We are starting to see that pipeline bear fruit.  The entire move has come in the last two weeks.  The stock has moved from 70 cents to a dollar on news that they had secured contracts in the healthcare vertical and converted one of their previously announced trials into revenue in the supply chain vertical.

Finally, a third company, Combimatrix, which I wrote about earlier this week, is beginning to run and take a more significant position in my portfolio after releasing solid fourth quarter results.

So that’s all great, but the reason I mention these three examples is because they illustrate how bad I am at predicting how things will play out.   In the second half of last year had you asked me what my portfolio would move on I would have replied it will rise and fall on the fortunes of Radcom and Radisys.

Flash forward a few months and my portfolio has moved significantly higher and Radcom and Radisys have done nothing.  Radisys has actually went backwards to the tune of 20%.  Whodathunkit.

This is why I carry so many positions.   A. I’m a terrible timer.  The events that I think are imminent take months or years to play out, while the events that I think are distant have a habit of manifesting much faster.

Second, my favorite ideas are often not my best one’s.  I have no idea why this is.  If I did I would change my favorite ideas.  But it’s uncanny.  I’ll sit on a thesis like Radisys, work it into the ground to understand it in depth, and then along will come a Health Insurance Innovations, which I will buy on a bare thesis (in this case that the Affordable Health Act will be repealed and this is going to be good for HIIQ) and when the dust settles I’ll have more gains from the latter than the former.  Its kinda crazy.

I guess as long as you are moving in the right direction it doesn’t really matter how you get there.

Portfolio Changes – Adding Silicom

I added a couple of new positions this month.  The Rubicon Project and Silicom.

Silicom got hit after releasing what I thought was a pretty good fourth quarter.  The company traded down to $35 from $39 pre-earnings.  I’ll try to get a more detailed write up out on Silicom at some point, but the basic points are:

  • This is a $250 million market capitalization company with $36 million of cash and no debt
  • It’s trading at a little over 2x revenue and just guided 15% growth in the first quarter and double digit growth for the year
  • Their past seven year compounded annual growth rate is 26% and growth was 21% in 2016.

Silicom designs a wide range (over 300 SKUs) of networking, cybersecurity, telecom and storage products. These are generally board level and appliance level hardware solutions.

They expect their security vertical will grow double digits, their cloud vertical will “grow significantly” and that a contribution from SDWAN will kick-in in 2017 and is expected to become a “major growth area”.  They said that over the intermediate term they see a larger opportunity in their pipeline than they have have in the past.

Already the stock has rebounded on news of a significant contract for encryption cards that will ramp in 2017 and reach $8 million in sales in 2018.

I’ll talk more about Rubicon Project in an upcoming post.

Apart from these new positions I did a bit of tweaking of my positions, adding a little to Nuvectra and Combimatrix, reducing my position in Bsquare and selling out of DSP Group.  I also have added to my Vicor position in the last couple of days (subsequent to the update end so not reflected in this update).

Taking advantage of Bovie Medical Weakness

I also added significantly to my position in Bovie Medical.  The stock sold off on news that their pilot project with Hologics for selling the J-Plasma device would not be extended.    As I tweeted at the time, I didn’t think this was as big of news as the market did.

To expand on my reasoning, Hologics has a particular business model they follow for their instrumentation and disposable business, of which J-Plasma would have been a part (from 10-Q):

we provide our instrumentation (for example, the ThinPrep Processor, ThinPrep Imaging System, Panther and Tigris) and certain other hardware to customers without requiring them to purchase the equipment or enter into a lease. Instead, we recover the cost of providing the instrumentation and equipment in the amount we charge for our diagnostic tests and assays and other disposables.

So they go “full razor blade”.  Bovie on the other hand, generates significant sales from generators.   The average selling price (ASP) for a generator is much higher than hand piece so Bovie generates a significant slice of their revenue from it.  From the 2015 fourth quarter conference call :

I guess when you think about it, the generator ASP is north of $20,000, the hand piece ASP is $375

So the models aren’t aligned.

Second, Hologic’s Gyn Surgical business segment (consisting of the NovaSure Endometrial Ablation System and our MyoSure Hysteroscopic Tissue Removal System) is a $400 million business so J-Plasma is microscopic for them.  They may not have been inclined to bend their model for Bovie.

Also worth noting is that Hologics wasn’t even mentioned in the Bovie 10-Q whereas the agreement with Arteriocyte that was mentioned favorably.

Finally the language used on the third quarter conference call around Hologics wasn’t exactly definitive:

Well, as you know, the sales channel partnership with Hologic,right now,is in a pilot phase.  So we wouldn’t be in a position, if we were to disclose the economic relationship, until that’s a permanent agreement.  So the pilot portion of our partnership will go until the end of February.  So you could look at some period after that before we can announce a permanent relationship and we’ll decide at that point in time if we’re going to elaborate on the economics of the relationship.

The agreement with Hologics hadn’t generated material revenue so there is no hit to the bottom line.  And in a separate press release (which oddly was released on the same day as the Hologics information but didn’t get on their website for a couple days after), Bovie reiterated guidance for 2017, including “accelerated growth for J-Plasma”.

I think the stock sold off in the following couple of days because its small, illiquid and under followed, not because this agreement was meaningful to the company.  So I bought.

Portfolio Composition

Click here for the last four weeks of trades.



Week 282: Two Big Events

Portfolio Performance



Top 10 Holdings


See the end of the post for my full portfolio breakdown and the last four weeks of trades

Thoughts and Review

My portfolio bounced back this month.  This was somewhat remarkable given that my two largest positions, Radisys (RSYS) and Radcom (RDCM), continued to perform poorly.  I don’t expect much from either of these stocks until they are able to secure additional contracts with service providers.  With year end coming up, I am hopeful (but not counting on) some news on that front.

The rest of my portfolio did extremely well, benefiting from the rotation to small caps that occurred after the election of Donald Trump.  I didn’t anticipate the market move or the small cap revival.   But I wasn’t the only one, in fact I didn’t hear that prediction from anything I read.  I would be interested if anyone else knows of an expert, newsletter writer or manager that predicted the move?  They would be worth following.

In retrospect it makes sense; expectations of significantly lower taxes and a relaxation of regulations would lead to a market rally with a bias on small caps with domestic exposure and few tax loop holes.  The stocks that have performed the best for me have had that characteristic.

Willdan Group (WLDN) is a text book example.  Willdan has always paid a high tax rate, sometimes over 40%.  If the companies tax is cut in half, which is not impossible under a Republican government, earnings go up by 30%.  They are also essentially an infrastructure play, another positive.  The stock has moved from $16 to $24 in the month since the election.

Adding Healthcare, Infrastructure, Biotech

While I wasn’t positioned for a rally leading into November 8th, I adapted as the market moved higher.  As I’ve written about here, I added Health Insurance Innovations shortly after the election on the expectation that changes to the Affordable Care Act (Obamacare) would open up competition, which would be positive for their business.

I also added an infrastructure play, Smith-Midland (SMID), as it seems that this will be the focus of spending under the Trump administration.  Smith-Midland makes make pre-cast concrete products like barriers, sound walls, small buildings, and manholes.  They have a market capitalization of $25 million and even after having run up to $5 are not expensive.  There is a good article on the company here.  I also added to my existing position in Limbach Holdings, another infrastructure play.

My last move in response to the election result was to add to a few biotech names.  This worked out initially but interest has waned in the last couple of weeks.  I added to my position in Supernus (SUPN), to Bovie Medical (BVX) and added back some TG Therapeutics (TGTX).  I may jettison the latter position soon.

Responding to OPEC

The Trump move was followed by the OPEC move, which I again don’t profess to have predicted.  I was agnostic going into the OPEC meetings; I held my usual weighting of energy positions, but did not pile into them as a bet that a deal would be reached.

Instead, as is my typical strategy, I chased the news, adding to energy names on the heels of the announcement.  By waiting I missed out on the first 10% move, but once the deal was announced it was a far lower-risk entry into stocks on my watchlist.

It can be argued that OPEC’s cut will only lead to high US production, or that it will be diluted by cheating by OPEC members, but nevertheless its difficult to argue that this doesn’t put a floor on prices.  And if there is a floor, stocks that previously had to discount the possibility of another move into the $30’s do not have to anymore.  Therefore stock prices needed to move higher.  I think they still do.

Many of the names I am interested in are small enough that they do not move immediately with the market.  Thus I have been able to add to Journey Energy (JOY) and Zargon Oil and Gas (ZAR) at prices not too different to what they were leading up to the announcement.  There is a good SeekingAlpha article (and comments, in particular note those on the interim CFO hire) on Zargon here.  I haven’t seen any analysis on Journey, and I will try to write up a summary on the stock in the next couple of weeks.

A second energy name that I added to and am in the process of writing up is Swift Energy.  As I tweeted on Friday:

I also added Resolute Energy (REN), a Permian player I have been in and out of over the past 6 months, and added back Granite Oil (GXO).  There was a good comment to my last portfolio update that gave me some perspective on the concerns I had raised about Granite.  I wanted to add to Jones Energy (JONE), but it moved so quickly off of the OPEC news that I didn’t get a chance.

Finally I added to a derivative play, CUI Global.  CUI Global is a bet on Trump as well as OPEC.  The company has said in their presentations that they have struggled gaining traction with their GasPT products in North America because of the dour investment climate for oil and gas infrastructure.  This should change under Trump and with support to oil prices.  Its no guarantee that CUI Global will be the beneficiary, but if their product is as good as they profess it to be, it should be the preferred measurement tool for new projects.

I also added a position in Contura Energy.  It was written up here.  I think this article will move behind the paywall soon, so I would recommend reading it sooner than later.

Where we go from here?

I’ve taken on some risk as the market has moved higher and especially after the OPEC agreement.  But I do not expect this to last long.  I’ll be paring back positions over the next few weeks.

I don’t feel like I know what to expect from this new US regime.  Tweets like the one’s Donald Trump made over the weekend, promising a 35% tax against companies moving production abroad, leave me wondering where we end up?  Are these just empty threats, impossible to implement?  Or is this only going to escalate?

It’s uncharted territory.  If government spending increases significantly, taxes are cut and trade restrictions are imposed, I’m not sure where it leaves us.  Will bond yields rise, setting off a negative market event?  Will investors continue to pile into domestic US equities?  Will stocks based in foreign locales or with manufacturing operations abroad sell-off on concerns over tariffs being implemented.  The answers are just way beyond me.

Lacking confidence in the answers means I have to get smaller.  That’s the only response.  Since July (my year end) I am up nearly 30%.  I feel like I am pushing my luck asking for the same kind of performance in the second half of my fiscal year.

Portfolio Composition

Click here for the last four weeks of trades.


Week 278: Shorter Posts and thoughts on Credentials

Portfolio Performance

Top 10 Holdings

See the end of the post for my full portfolio breakdown and the last four weeks of trades.

Thoughts and Review

I had such a good response from my post on Radisys that I decided to change things up for the blog.  Rather than posting monthly letters summing up all my thoughts, I am going to deliver updates in a more traditional blog format.  I will write as things come up. So this update will be more brief, and will not cover any lengthy company updates.

I had a pullback in the last month.  I guess it shouldn’t be unexpected.  The previous three months were almost parabolic.  Having a portfolio that is weighted only to a few stocks, any kind of lull in the performance of those stocks can cause big fluctuations.  Right now my portfolio is heavily dependent on the performance of Radcom and Radisys. Both stocks had corrections leading into and following their third quarter earnings.

The good news is that nothing has occurred with either to warrant a change in mind.  While I expressed some concerns about Radcom in my earlier post, I felt a lot better about the stock after their Needham conference presentation.  I even bought some back over the last couple days.

I sold out of a number of oil stocks.  I still hold positions in Swift Energy, Journey Energy, Zargon Oil and Gas and a very tiny position in Gastar Exploration.  Other than Swift Energy, none of my positions are very big.  I started by selling Granite Oil after these comments on InvestorsVillage (here and here).  Looks like I was wrong there.  Later, as the price of oil began to break down I sold Jones Energy and Resolute Energy.  Both of these are levered plays and I expect out-sized moves as oil corrects.

I added a couple of new starter positions under the theme of Big Data: Attunity and Hortonworks.  The latter has begun to work out but the former has not at all.  I’m doing some more work to understand if I just made a mistake on Attunity.  With the new blog format I will write-up the positions and my thoughts on the Hadoop market (which led to my investments) in separate posts.

I also added a position in Nimble Storage.  The company has some good technology, can compete with Pure Storage and take market share away from incumbents like NetApp.  Again I’ll give more details in a later post.

Finally I mentioned in my last post that I had taken a very small position in Supernus Pharmaceuticals.  I’ve held that position over the last month and watched the stock correct downwards almost every day.  This is a big biotech sell-off and I don’t think the move has much to do with the company itself.  Supernus is growing very fast, there appears to be plenty of opportunity for further growth, and the pipeline of new drugs seems to be quite robust.  I’m seriously considering adding a big chunk to this one.


One final thought on the topic of credentials.  As I have written in the past, I manage my own family’s money.  Recently I had an opportunity to expand that to a number of friends.  But before going too far down that path I wanted to understand the regulatory requirements.

It turns out that in Canada at least, managing money and taking any sort of payment for it is very regulated.   It requires a number of courses, which is reasonable, but also years of very specific experience under the tutorage of a dealer.

Clearly I am not going to take 3 or 4 years to work as an understudy just so I can start a small part-time business on the side for a few friends.

My frustration is that there is no distinction between someone trying to scale into a large fund, soliciting money from the general investing populace, and someone who wants to do what I was looking into; basically help out some buddies and get paid on performance to do it.  These two activities do not seem equivalent in terms of public risk.  But in the eyes of the regulator there is no distinction.

I’m not a conservative in most ways but this certainly gives me sympathy to the position that abhors regulation.  I’m in a region that is suffering, I have a ready-made opportunity to create a small business, and the government has basically said no you can’t do that, because we know best.  Because as anyone who has read this blog for the past 6 years knows, I am clearly not qualified to pick stocks.

Portfolio Composition

Click here for the last four weeks of trades.


Third Quarter Earnings Updates: RDCM and VICR

I had such a great response from my single story formatted post on Radisys that I decided to change things up a bit for the blog.  Rather than posting monthly consolidated letters where I sum up all my thoughts, I am going to try to deliver updates in a more traditional blog format, writing as things come up.


Its interesting to contrast Vicor with Radisys.  Both companies are in a similar place; they have a large opportunity ahead of them, that opportunity is beginning to realize itself with new orders, the order build should really hit its stride in the second half of 2017, and the next couple of quarters will be relatively weak as we wait for the order book to build up.

The difference between the two companies is the way the market has interpreted the results.  Vicor was up nicely on the day after their earnings release and has held up relatively well as the small cap market has suffered.  Radisys was down and has continued to be down since their release.

Vicor reported  a good quarter.  VI Chip bookings were up 54%.  Picor bookings were up 122%. Only legacy  BBU bookings were down at 14%.

More importantly, on the call the company announced the first of its VR13 orders, “some major million dollar type orders that obviously are well beyond the prototype level”.

For those not following the story, VR13 refers to an Intel Skylake chip based server specification whose ramp has been delayed by continual delays in the Skylake chipset.  Intel announced in their quarterly call that they were now shipping Skylake chips for sampling.  Vicor’s announcement provides follow through to that data point.

Vicor also announced some other interesting development news.   In last 3-4 months, they have been doing work on power on package technology, putting point of load power multiplier on the ASIC to provide 100’s of Amps in 6V to 1V range.  Within several weeks they expect to see the first production of once such solution.

I added a little to my position in Vicor.  I think that the VR13 ramp is upon us and results are more likely to surprise positively than negatively going forward.


There was something a little bit off about the Radcom call.  Maybe it was my expectation, or at least hope, that there would be some new announcements.  There wasn’t.

It didn’t help that a replay of the call wasn’t posted until this morning.  I wonder whether the delayed move in the stock is because of folks that couldn’t listen to the call and didnt’ get their first impression until this morning.

There were some positives.  There was the extension of deals with two CSPs.  There was the use of the word “accelerating” for several of the CSPs that they are engaged with.  And there was color that 2017 has some upside on smaller non-NFV deals with emerging market CSPs.

But there were also a number of things I didn’t like to hear.  I didn’t like the language around the relationship with AT&T being “ok”, which is the term they used a few times.  And I didn’t like that they continued to be vague about the new product that they won’t talk about and haven’t released to the wider market.

Referring back to my post last month on Radcom, I had been hoping that there might be an announcement piggybacking the announcement by Gigamon and their involvement with AWS and AT&T.  The question was actually asked on the call, and while the answer was cryptic enough to leave the question open ended, its hard to verify that a GIMO/AT&T/AWS connection is there.

Also, if I understand the wording around the estimate that Radcom’s service assurance total addressable market (TAM) is $100 million between now and 2020, maybe 2018 given acceleration, it’s not really that big, and it shows they really do need to expand to new products to grow into valuation.

Finally the guidance that “the models” are saying no deals until the second half 2017 is sobering.  They had originally anticipated deals towards the end of this year, beginning of next year.  This isn’t surprising to me, I mean everything that I have read about service providers is that they move at a glacial pace, but its still a slip from the original color.

So I reduced.  I resized Radcom so it is now closer to a regular sized position for me.  No need to try to be a hero.  I wouldn’t sell any more at these levels.  It was a big move down this morning.  And the upside is still significant, so its a stock I want to own.  But I don’t want to have my portfolio heavily dependent on the story if it won’t play out until the second half of 2017.  I have enough “2017 stories” already.  I’ll keep my reduced stake and wait until we get closer and see where the price shakes out.


Week 274: Going with Growth

Portfolio Performance


Top 10 Holdings


See the end of the post for my full portfolio breakdown and the last four weeks of trades

Thoughts and Review

Since March I have been focused on finding companies with growing top line revenue.  I have put less emphasis on EBITDA, earnings and cash flow.  As long as the company is showing growth I’ve been taking a closer look.

This change in strategy has worked.  The market has pushed up the price of stocks with growth prospects, while value stocks, at least the small and microcaps I follow, continue to lag.

But I wouldn’t reclassify myself as a growth or momentum investor just yet.  Once it stops working, I’ll reevaluate, try to figure out what is working next.  I try not to have any allegiance to any particular investing methodology.  I’m just trying to figure out what is working and will take that approach whatever it is.

What remains constant is my process.  Small positions.  Add as they rise and announce positive developments.  Reduce if they fall even if I don’t always understand the reasons.

When things are working at their best this process feels like a manufacturing conveyor of idea generation.  I keep working hard, the ideas keep coming, when an idea doesn’t pan out I throw it away and move on to the next one.

This is the experience of the last few months.  My portfolio gains have been anchored by my two large positions, Radcom and Radisys, but I have been churning out gains from the numerous small positions I have taken on.  As is my method, I have been adding to these positions as they go up.  When a position moves against me, I don’t add and will throw it out if it simply isn’t working.

Last month I wrote:

I have noticed a change in temperament in the subsequent months, and this seemed particularly evident through the second quarter earnings season.   I still had misses.  BSquare had miserable results.  CUI Global was lackluster.  Both stocks fell.  But even then, the moves down didn’t have quite the same conviction as similar such moves in the past year.

Its hard for me to put my finger on exactly what it was that was different, so maybe this is all just sophistry, but it was almost like these stocks were going down grudgingly, because they had to, not because they wanted to.

It still feels this way, though I sense some early signs of headwinds.  But the path of least resistance, at least for most of the stocks I own, seems to be up.  I’ll continue to hold on until there is more evidence that the ride is about to end.

Apart from Radcom, the biotechs and the oils have been working particularly well.  It seems like the biotech stocks have turned the corner.  I took positions in a number of new biotech names: Aequus Pharmaceuticals and Novabay Pharmaceuticals, which I have written about below, as well as Supernus Pharmaceuticals, Verastem and Aeterna Zentaris, which I ran out of time to write about and are all pretty small starter positions anyways.  You will note though that Supernus Pharmaceuticals is a derivative play of Aequus, as the rights of two of the drugs being marketed by Aequus in Canada are being licensed from Supernus, for whom they are growing significantly in the United States.

Oil Gains

I timed the oil stocks well.  As I wrote about last month, I have positions in Jones Energy, Zargon Oil and Gas, Granite Oil, Journey Energy, Gastar Exploration, and Resolute Energy.

However, as I alluded to last month I haven’t had a ton of conviction in the price of oil over the last couple of months.  I wrote:

I don’t have a crystal ball on oil.   I am sympathetic that the builds we have seen (up until this week’s rather massive 14mmbbl draw) are due to the drainage of floating storage.  I’m hopeful that once this runs its course we could see some surprising draws in the shoulder season (this thesis has been expressed by a number of investors on InvestorVillage, not the least being Robry825, who describes his position here and  here).  I’m also cognizant that oil wells do decline and that we simply aren’t drilling like we were a few years ago.  Still, as I’ve said on many occasions before, I play the trend till it ends but when it ends I really don’t know.

Here is where it’s helpful to have knowledgeable folks to follow.  I mentioned Robry and his extremely bullish outlook in my quote above.   There is another poster on InvestorVillage named Sophocles.  I’ve known and followed him for a long time.    He isn’t always right, but he has an intuitive sense for a commodity bottom well before the data has changed or the experts are revising their forecasts upwards.

He has been steadfast in his opinion about the uptrend for oil prices over the last number of months.

Leaning on the calls of Robry, Sophocles and a few others has helped me hold my oil positions through what was a great deal of uncertainty in September.  I doubt I would have do so otherwise.  I would have been shaken out by the negative Goldman note, a couple of the precipitous two or three day drops, or after the price reacted negatively to what should have been extremely bullish storage numbers.

But here we are at $50 and one thing I have learned is to not second guess where you are going simply because the data hasn’t caught up with the price.  In a real bull market price leads data, and you will never know the why until its too late.

With that said, I thought this IV post was a very good summary of the fundamental oil balance heading into next year.

Shorts and Mortgages

The Canadian government laid out the ingredients for a pretty good shorting opportunity this week when they introduced new rules for insured mortgages.  There is no question that these rules will have an impact on the housing market.  There is a lot of uncertainty about the degree.  I suspect it’s going to have a pretty big impact, especially if some of the comments below on declines in transaction volumes come to pass.

There are two major changes.  First there has been a reduction in borrowing limits for anyone applying for an insured mortgage with a 5 year fixed rate.  They will be subject to a much higher interest rate than before:


James Laird from calculated the following impact of the rule change on a typical borrower:


Rob Mclister (from the blog made a similar observation in this clip, saying that an individual with a $50,000 salary, 10% down, prior to the rule changes could take a $300,000 mortgage, now that has been reduced to $240,000.

Mclister went on to point out that about one third of buyers are first time buyers and at least one third of those would have trouble qualifying for a mortgage with the new rules.

The other change, which applies more specifically to the non-bank lenders, is that a number of mortgage products that were previously eligible for bulk insurance are no longer eligible.  Mclister lists refinances, amortization over 25 years, rental properties, and mortgages over $1 million as being subject to the rule change.

Non-bank lenders don’t have a deposit base for funding.   They therefore require securitization to fund new originations.  In order to securitize the mortgages they need to be insured.   So without insurance non-bank lenders basically have to stop originating these products.

Its not wonder that, as Sherry Cooper said on Thursday, “alternative lenders and mortgage brokers…have suspended activities” in some lines of business and that “the part of business that is no longer insurable can no longer be done by these lenders – they get their funds by insuring mortgages and securitizing – they can’t securitize if they can’t insure”.

Mclister said that “non-bank lenders are going to have heck of time trying to dispose of these mortgages, nonbank lenders are going to have to go to banks to buy mortgages, they are going to jack up financing costs unless regulators level the playing field

Dan Eisner, of True North mortgage, was quoted as saying “many of their funding sources have announced substantial changes or are telling them substantial changes are coming.”

Earlier in the week Eisner was on BNN where he made the rather chilling prediction that transactions would decline by 40-50% across the country.

The final clip I will point out is from the Moneytalks radio show this last Saturday (October 8th, 9am about 40 minutes in).  The show has a weekly segment with a real estate expert named Ozzie Jurock.  I have been listening to Jurock for a long time, years, and I have never heard him speak so negatively about the outlook for Canadian housing.  He has typically been very upbeat and optimistic (which, incidentally, has been the correct position for the last number of years).  Worth noting is that Jurock singled out the non-bank lenders as being particularly hard hit by these changes.

While the cause is different, there are two things at play here that are eerily reminiscent to the United States a decade ago.  First, interest rates have, effectively, jumped suddenly for a number of borrowers (like the adjustable rate mortgages did in 2007).  Second, a formerly reliable source of funds (securitization via portfolio insurance) is now no longer available to some lenders.

I’m not an expert on the companies in this area.  I don’t know all the details of the loan books of Home Capital, Equitable Group or the insured portfolio of Genworth, to take a few names.  But from the simple experience of the United States, I know that when transactions decline and funding sources dry up, house prices fall and the tide for the monolines and originators goes out.  How far is the big question.

As I’ve said in the past, I don’t like to talk about my shorts.  There are a few reasons for this. For one, I short to hedge and tend to have weaker theses around the individual names.  For two, its too much of a negative activity.  You run the risk of going offside companies, other investors, who wants the trouble.

But I took a few shorts in the mortgage space this week.

I still own this damn rental property.  So I need a good hedge.

Radcom and Radisys

Lots of interesting puzzle pieces with both of these stocks.

There was an interesting announcement out of AT&T and AWS (Amazon Web Services) this week.  The two companies agreed to forge “a tighter partnership to develop joint cloud, Internet of Things, and security solutions, targeting enterprises and combining AT&T’s network with the AWS cloud.

I think the big move that we saw in Radcom on Thursday is likely due to this announcement, and the potential implications that exist for Radcom.

There are a few puzzle pieces that revolve around this partnership and Radcom.   They don’t all fit together yet, but they are beginning to.

A few of these pieces are centered around Gigamon.  You might remember me lamenting about Gigamon in a past letter.  It’s a stock I looked at in March, thought it was a great opportunity, but waited for a dip into the mid $20’s that never came.

Well Gigamon announced a few months ago that they would be offering the first visibility solution for AWS.  This includes security inspection of traffic and monitoring of resource access and use.

Gigamon is also a supplier of AT&T.  This was mentioned at the Gigamon analyst day but some more detail was given by kerryb in the comment section of this SeekingAlpha post:

I have an investment broker friend who introduced Radcom too contact Alex Henderson of Needham about Gigamon since the stock has done so well and asked the relationship with Radcom, Alex told him that Gigamon is not competition and that Gigamon and Radcom are currently working together for interoperability between their software. Gigamon does the monitoring of the data flows for mainly security and parses out the important information and sends the info to the Radcom MaveriQ software/tool which then takes the information and does the service assurance piece. Gigamon can reduce the processing of the Radcom and other tools by 65-75%.

Gigamon talked about how they were encroaching on the traditional probe vendors at their analyst day.  I listened to that call a couple of months ago, and made note of what they said because it seemed important:

new battlefront in CSP arena is subscriber costs which is the cost to maintain, upsell, attract subs. To do that you have to understand sub experience. Gigamon has capabilities that can reduce this expense significantly, which is exactly why they are seeing traction in CSP arena. In their visibility platform they have ability to look at traffic across all interfaces 2G, 3G, 4G, Wifi, to be able to recognize subscribers intelligently, the devices that subscribers are coming from, the applications that subscribers are using, and to then take high value subscribers and pair them with the tools that are being used to manage subscriber experience. What this allows service providers to do is to match their tooling capacity to the high value subscriber and not waste the tooling capacity on subscribers who really are not subscribing to new services. So as an example in my family, our plan, the people who use it most are the kids, but the person who pays the bills is me, the CSP needs to be able to discern who the high value subscribers are, whether its my traffic or the kids, and then be able to come back and make sure I have a fantastic experience they can bill me correctly for the services I’m using but more importantly can upsell me on new services they want to introduce. They have to have that intelligence in the infrastructure and that is the solution Gigamon is providing. A lot of this functionality used to be part of the traditional probe of the tooling manufacturer but now is part of their visbility platform and is reducing the number of probes that CSPs that to deploy is shrinking or being eliminated because the subscriber visibility in their infrastructure.

The final puzzle piece is Radcom’s announcement on their second quarter call that they were developing a new product for AT&T.  They said this product was “adjacent” to their service assurance implementation and “critical” to making the NFV implementation work.

The scuttle is that the new product provides a security function.  On the second quarter call one of the analysts, George Marima, asked whether the product was the “threat intellect” product or something else and what the licensing model was.  He was referring to AT&T’s announcement in July of “Threat Intellect”, which is a security offering to enterprises “helping to enable businesses to detect, analyze and address security threats faster and more efficiently than ever before.”

Radcom’s response was that it would follow a license model much like the rest of the MaveriQ suite.  But described the model a couple of times to it being an “enterprise” license.

What I wonder is whether Radcom is referring to enterprise in the context of a single license to AT&T or whether this is an enterprise license to the customers of AT&T, specifically those using the threat intellect suite of security solutions that is now likely being integrated as security solution for AWS, see this article for more details on that.  While there are still only a few puzzle pieces here, there is enough to speculate about it, and such a leap would be a huge win for a tiny player like Radcom.

On the Radisys front, in this joint paper (hat tip to kerryb again for digging this up) with Intel Radisys describes their DCEngine in some detail.  On the last page are the following comment about the Verizon deployment of DCEngine this year:

“The success of the initial 20-rack deployment has led the tier 1 CommSP to continue to deploy DCEngine in multiple locations, with hundreds of petabytes of storage and more than 200 racks being installed by the end of 2016.”

On the first quarter conference call CEO Brian Bronson said the orders from Verizon were for “less than 100 racks”.  This was said after the follow-on Verizon order had already been announced so presumably it includes most or all of the orders delivered in the first half of the year.

In the SeekingAlpha comments where this paper was linked to, Mike Arnold added the observation that the average selling price (ASP) of a DCEngine rack is $350,000.  I haven’t been able to confirm that number.  If it’s true, it suggests around 140 racks were sold to Verizon in the first half.

Either way the conclusion is that there has still been a lot of racks sold to Verizon in the second half.  I hoping that this will help lead to an upside surprise when Radisys announces its third quarter results.

Hudson Technologies, R-22, and youtube

I spent one night this month watching youtube videos of air conditioner installers talking about regulatory requirements for refrigerant.

I turned to youtube because I was having a hard time finding information about R-22 conversions.   Turns out that air conditioner techs don’t have time to pen a lot of articles.  They do however like to talk about their trade on camera and take you through the installation process in detail.

Its interesting stuff.  Most interesting is the perception.  The reality is that in the US virgin R-22 is being phased out.  But there is nothing preventing the use of reclaimed R-22.

However the perception for many of the techs I listened to is closer to “the government doesn’t want us using R-22 so we aren’t going to use it”.

It’s worthwhile to weigh these “boots on the ground” opinions before getting too rosy about just how high R-22 prices are going to go.

It’s also becoming clear that there are more legitimate R-22 alternatives than I originally thought.   These are refrigerant blends made by Dupont, Honeywell, Chemours.  This video does a good job of stepping through each of the blend alternatives.

Now again, the reality is that these blends tend to compare poorly to R-22.  They have lower cooling capacity, require more power, and are more likely to damage equipment like compressors on the unit.

Nevertheless, you can drain your unit of R-22, replace a few valves, recharge the unit with the replacement refrigerant, and you are good to go.  Add to that the previously noted perception that the government just doesn’t want you to use R-22, and you probably have more conversions, whether it’s the right thing to do or not.

There is also quite a bit of talk, both in the videos and on some message boards about the price of R-22 rising.  The rising price is a further incentive to switch.  It might not be the right decision, in terms of insuring the best performance and longest life for your unit, but I think its likely getting done more and more as R-22 prices go up.

This video provided some thoughtful  insights about the pros and cons of alternatives.

This has made me a little more cautious about my outlook for Hudson Technologies.  Not enough to sell just yet.  I still don’t think the stock is pricing in the current $15 R-22 price and I also don’t think its reflecting the new contract with the department of defense.  But the upside isn’t quite as high as I originally had anticipated.

New Position: Aequus Pharmaceuticals

This is a really tiny specialty pharma company based in Canada.  Market capitalization is only about $15 million after a recent $2 million capital raise.  The CEO participated in that raise to the tune of 800,000 shares.

The company is focusing on the marketing and sales of drugs in Canada. They also have a pipeline of early stage opportunities that revolve around taking existing drugs and repackaging them into a transdermal delivery mechanism.

The company has the Canadian marketing rights for 4 drugs, Tacrolimus XR has been marketed since December 2015, Vistitan has been marketed since May 2016, while Topiramate XR and Oxcarbazipne XR are still awaiting approval from the Canadian Health board.


Let’s start with Tacrolimus.  The drug is used as an immunosuppressant in patients receiving organ transplants, most commonly kidneys.  Aequus is selling the first generic form of the drug available in Canada.  In other countries, generics make up a significant portion of the market.  In the UK its 53% while in the US its 40%.  Right now Tacrolimus account for $100 million of a total $240 million immunosuppressant market.  The long-term goal here is for Aequus to take a market share that is inline with the US and UK.

Vistitan is a prostaglandin.  Prostaglandins are created naturally within our body and promote or inhibit all kinds of different effects on bodily tissues.  Vistitan contains bimatoprost, which is engineered to increase to fluid flow from the eye, which in turn reduces pressure in the eye.  Therefore Vistitan is used in the treated of elevated interocular pressure, or high eye pressure.  Its most commonly used in glaucoma.  I haven’t been able to get the figure on the Canadian market in 2015, but in 2014 the market for IOP-lowering medication was $140mm and bimatoprost 0.01%, which is a lower dose form of Vistitan that was the only dosage available in Canada, accounted for $42 million of this.

While Topiramate XR and Oxcarbazipne XR are awaiting approval in Canada they have been approved in the United States since 2013 by Supernus.  In the US sales of the two drugs are growing and the combined prescription growth was up 39% year over year in the second quarter while sales were up 47% to $50.3mm.  Assuming similar results in Canada, it doesn’t seem unreasonable to me that sales could exceed $20 million annually.

Its pretty early to be estimating revenue from these products, but in a recent Midas Letter interview, CEO Doug Janzen said:

You have a $20 million business that expects to do $2 to $3 million in sales this year, closer to $15 next year, peaking that at $40 off of existing products

That estimate doesn’t seem unreasonable to me, and may even be low given the corresponding markets in the United States for each drug.

The final puzzle piece here is the pipeline, which consists of 3 drugs in development.  Each of these drugs is already approved and well established.  Aequus is developing a new transdermal delivery method for each.   Transdermal means that the drug will be administered by path.


For the patch they have entered into a multi-product collaboration with a company called Corium International, which specializes in transdermal manufacturing. The two companies split the costs 50/50.

The three products  are still in very early stages.  One advantage though of the new drug application for a transdermal version versus an entirely new drug is that costs and time to deliver are significantly reduced, estimated to be $15 million and 4.5 years for a transdermal development versus $500 million and 10+ years for new drug.

It seems like a lot of irons in the fire, and a lot of potential, for a $15 million market capitalization.  If the stock dips back below 30c I will probably finish off my position with another add.

New Position: NovaBay Pharmaceuticals

I came across Novabay while scouring through presentations and the Rodham and Renshaw conference.  Most of the participants at the conference are early stage drug companies, but Novabay is an exception, having a product in production called Avenova.

Avenova is a really simple product.   It consists of a compound called Hypochlorous acid in a saline solution.   Hypoclorous acid is a natural compound that is given off by while blood cells to fight viruses.  It’s a particularly potent substance and is used in fighting flesh eating disease (for which Novabay provides Avenova for free) and can kill the ebola virus.

A second advantage of Avenova is that it does not create resistance.  Unlike antibiotics, Avenova can be prescribed without worries of creating anti-resistant strains of the virus that become harder to manage.  This is particularly important in chronic applications like bleptharitis, which I’ll expand on in a minute.

Novabay reminds me a lot of Bovie Pharmaceuticals, which I bought and wrote about back in August.  Both are tiny, have a single product driving growth (J-Plasma in Bovie’s case), are seeing really strong growth, and have plenty of room to run.

The two differences are A. Novabay is cheaper than Bovie and B. the potential market for Avenova is bigger than J-Plasma.

Fully diluted including warrants that are close to being in the money, Novabay has 12 million shares outstanding.  That gives it a market capitalization of $43 million at $3.60.  Bovie has a market capitalization of over $120 million.  Novabay is operating at about $10 million of sales run rate.  Bovie is at about $20 million, but keep in mind that only about 20% of this is revenue from J-Plasma, the growth engine here.

Meanwhile, the potential market for Avenova is really big.  Here is a slide from the company’s presentation:


The biggest part of this market is Blepharitis, which is a chronic condition of eyelid inflammation.  You can treat it with antibiotics but because it comes back opthamologists don’t like doing that unless it’s a severe case.  This article describes the benefits of Avenova:

Blepharitis has become more difficult to manage because of the resistance bacteria build against traditional antibiotics used for this condition. Hypochlorous acid utilizes a different mechanism of kill from traditional antibiotics and may therefore be less likely to promote bacterial resistance.

I read through a number of forums, which while anecdotal, generally pointed to positive patient experiences.  The most common complaint I heard was that Avenova is not covered by all insurers.  This is simply a fact of the newness of the product.

There is competition. This article describes Blephadex and Ocusoft which are two of the primary competing products.  Ocusoft published this report comparing their product with Avenova.  I think many of their claims on price comparison have been mitigated by recent insurance approvals.

While there may be some headwinds with competition, its such a big market and the patient experiences appears to be varied enough to support a few different products.  Moreover, the company is showing product and revenue growth. Given what remains a fairly low market capitalization, I’m inclined to think the upside justifies the risks.

Portfolio Composition

Click here for the last four weeks of trades.


Week 270: Change is in the air?

Portfolio Performance


Top 10 Holdings


See the end of the post for my full portfolio breakdown and the last four weeks of trades

Thoughts and Review

The second quarter results are finished and as I look back on August I can’t help wondering if something has changed.

Since February the market has done well but for the first time in over a year I have done better.  I had a miserable first month of the year but after deciding things had gone too far in early February, its really turned around.

Here is the performance of my more significant individual non-oil related positions since that day.


August was a particularly good month.  I was up over 10%. I’ve had a number of big movers over the past few months.

But its more than performance that makes me think that something has changed.

The way that the small and micro-cap stocks I own have been acting has been different than the past year and a half.  For reference, in August of last year I pondered whether you could have a bear market that was never actually defined as a bear market.  At the time I wrote:

When I raise the question of whether we are in a bear market, its simply because even though the US averages hover a couple of percent below recent highs, the movement of individual stocks seems to more closely resemble what I remember from the early stages of 2008 and the summer of 2011.

Through most of 2015 and into the first couple of month of 2016 the movements of small company stocks just didn’t feel right.  Every move up was pressured.  Every move down was too easy.  This of course climaxed with the selling stampede of January and February of this year.

But I have noticed a change in temperament in the subsequent months, and this seemed particularly evident through the second quarter earnings season.   I still had misses.  BSquare had miserable results.  CUI Global was lackluster.  Both stocks fell.  But even then, the moves down didn’t have quite the same conviction as similar such moves in the past year.

Its hard for me to put my finger on exactly what it was that was different, so maybe this is all just sophistry, but it was almost like these stocks were going down grudgingly, because they had to, not because they wanted to.

Even Friday, which was a brutal day for the indexes, didn’t seem as bad as all that through the lens of my positions.  Radcom ended up.  Radisys spent most of the day green before giving up the ghost in the last 30 minutes.  A number of other positions started the day poorly but didn’t really follow through, almost to be saying yeah, we have to be down because the averages are down but there is no reason to panic.  A couple  of the oil stocks I’ve recently added (more on that below), fell at the open only to grind their way back up.

To be sure, part of my perception may be that I’ve picked up my game and picked better stocks.   It took me a while to figure out what my theme was for 2016 but once I settled on growth stories it has served me well.  This replaced sectors like tankers and airlines and REITs where I tried to find gain in 2015 and for the most part failed at it.

So maybe that’s the source of the change.  I’m staying away from value, looking for growth, and that’s what is working.

I’m wary that we have had a big move up, that each of the prior moves up over the last couple years has been followed by a precipitous fall, and that the election in the United States isn’t really setting itself up to be market friendly.

Nevertheless I can’t lose the thought that while we never had an official bear market we did have an extremely ugly period for small and micro-cap stocks and maybe that bear market capitulated in the first couple of months of the year and we are on to something new.  Its worth keeping an eye on.

Oil and my oil stocks

Inline with my tweets over the last few weeks I have been increasing my oil exposure.  I added a number of names and been adding to a couple of others.  Here are the tweets.  I’ll talk about each of these below.

I also took a position in Journey Energy (JOY) late this week but I’m not going to write about that one in this update.

Nevertheless they are all small bets.  Other than Granite Oil, which is about a 3.5% position for me, my oil bets are in the 1% range.

I don’t have a crystal ball on oil.   I am sympathetic that the builds we have seen (up until this week’s rather massive 14mmbbl draw) are due to the drainage of floating storage.  I’m hopeful that once this runs its course we could see some surprising draws in the shoulder season (this thesis has been expressed by a number of investors on InvestorVillage, not the least being Robry825, who describes his position here and  here).  I’m also cognizant that oil wells do decline and that we simply aren’t drilling like we were a few years ago.  Still, as I’ve said on many occasions before, I play the trend till it ends but when it ends I really don’t know.

Its also worth noting that while the US election could turn out to be a gong-show for stocks in general, it sets up as having very little downside and a fair bit of upside for the Canadian oil stocks.  Either Clinton wins and its status quo, which has been already heavily discounted in the disappointment of investors with Canadian oil stocks, or Trump wins and well, regardless of what chaos that entails, he did say he will approve Keystone, which would be a big tailwind for energy producers, particularly given the boondoggle that’s taking place with Energy East.

Nevertheless I don’t like not being in oil.  Part of this is simply hedging.  I am a Canadian investor who primarily invests in US stocks.  Therefore I am naturally short oil via the Canadian dollar.  If oil prices go up, so with the Canadian dollar and I will lose.  Owning some oil stocks helps to balance that out.  If I can hit the odd rocket ship like Clayton Williams or Resolute Energy have been recently, then all the better (note that while I have talked about Clayton Williams in prior posts I haven’t mentioned Resolute because I got the idea from a subscription service run by Keith Schaeffer.  He’s had a few winners for me since I took a subscription and would recommend it).

Empire Industries – China Theme Park Deal

On September 1st Empire Industries announced a “strategic cooperation agreement” with a Chinese company called Altair to build out a new space theme park in the Zhejiang province of China.  The announcement coincided with Justin Trudeau’s visit to China and apparently he was on hand at the signing.

The stock shot up on the announcement and to be honest I got a little too caught up in the hype and added to my position at 45 cents.  While I think those shares will ultimately deliver a positive return, there were some details left out of the news release that warrant caution.

In particular, this article from Business Vancouver fills in some of the details.

On the positive side, Empire will be delivering 6 attractions to the park for a price of $150 million.  This is significant dollar amount that will more than double the backlog.

Also, Empire will have an option to take a 20% stake in the amusement park.  Guy Nelson, president of Dynamic Attractions, the amusement park ride manufacturing subsidiary to which the work is directed, said they would likely exercise this option.  Given that the park has a price tag of $600 million, a 20% stake is a big investment for a company the size of Empire.  It would also makes you wonder about how Empire is going to get paid, as it seems a little too coincidental that their ownership opportunity is about the same dollar value as the cost of the attractions they are to deliver.

The other consideration that contributed to the stock pulling back from its highs was that construction of the park depends on getting the land.  Currently the land belongs to the Chinese government and the agreement that Altair struck was as the reserve bidder, meaning that they get the land if no one else places a bid.

I’m not selling any of the shares I own.  If the deal goes through it should still be a significant positive for the company.  But I should have been more wary of any Chinese dealings and I will wait for more news before acting further.

Granite Oil – Enercom

Granite presented again at the Enercom conference in September. They gave a presentation that an engineer would love and an investor would shrug at.  I don’t believe I heard the acronyms ROR or EBITDAX or even NAV a single time.

Last month I wrote that I had reduced my position in Granite.  After further consideration, a little more research and a more optimistic attitude towards the oil price, I decided to add back what I had sold.  The stock had also sold off into the low $6’s which has been good support in the past.

During the first six months of the year Granite focused on infrastructure.  They drilled one well in the first quarter and three more wells in the second quarter.  They converted a number of older wells into injection wells and added compression to facilitate the gas injection process (as I’ve written before their primary and only asset, Ferguson, is a large oil field where they are implementing an early enhanced oil recovery (EOR) strategy via gas injection.

In the second quarter the company achieved 100% voidage replacement in the heart of the Ferguson field.  The reservoir is now pressurized back to its original pressure.  Granite can move ahead with development drilling and expect the pressure support to limit well declines.

In the second half of the year the company expects to drill 5 wells.  On September 6th they announced results from the first of these wells.  Over a 96 hour test the well flowed at an average 815 boepd and ended the test at 995 boepd.   The company well costs are $1.2 million, down from $1.9 million in the first quarter of the year.

One number Granite did mention at Enercom was that the total contingent original oil in place (OOIP) at Ferguson is 457 million barrels. This is actually from a fairly old estimate by Sproule in 2012.

I contacted the company to clarify how much of their land package the estimate accounted for.  Their response was that the estimate encompassed an area a little larger than the current EOR approved area.  The EOR approved area is about 23 sections, shown below:


The entire land package held by Granite is over 550 sections.  This includes two other discovery wells to the south west:

entire-land-packageIts an extremely large package for a little company with a $200 million market capitalization and very little debt.  The wells aren’t as exciting as Permian wells, but they also come in at about 1/8th the cost.  The company doesn’t get much respect, but I hope that changes if they show some growth in the second half, which is possible as they finish out their drill program and see the full effect of the EOR.

There were a SeekingAlpha article written about Granite here.

Update/Summary of Accretive Health

I’ve owned a small position in Accretive Health since late last year.  And I owned the stock once before that, in 2014.  Yet I’ve never actually written about it.

Part of the reason I have stayed quiet on the position is because it has, until recently, been very small.  But I’ve also kept quiet because the thesis is hard to get behind.  This is the definition of a flyer.

Accretive Health has a few problems:

  1. They have been investigated for abusive billing practices.  There is a VIC article written around the time these issues arose that explains the claims in some detail here
  2. The abusive billing practices were followed up by an SEC investigation into revenue recognition.   The result of the investigation was that the company did not release its financial statements for the period of 2012-2014 until just last year. This article is a decent source for that episode.
  3. The conclusion of the SEC investigation led to a restatement of GAAP results and the subsequent GAAP accounting practices that were enforced make their financial statements pretty much useless.  Accretive can’t book revenue from a client until either the end of the contract term or when the agreement is terminated.  This means they go long periods with very little revenue and if they lose a client (generally a bad thing) they can show a big profit.
  4. Much of their business is dependent on Ascension Health.  Ascension is a large non-profit hospital operator.  They are also a large owner in Accretive.  They have shown their willingness to exploit their leverage as they did when they tried to take Accretive over with a a low-ball offer in mid-2015

Sounds exciting doesn’t it?   Nevertheless the potential upside holds my interest.  Here’s some details about how Accretives business works.

Accretive provides revenue cycle management for hospitals.  This means that they help streamline front- and back-office operations, including patient registration, insurance verification, coding compliance and collections.

Hospitals operate on extremely tight margins.  I have gathered margins can be as low as 1-2% if they are profitable at all.  Accretive’s model is to insert experts into the hospital, integrate their software into the hospital software, and work together with hospital employees to reduce costs, more effectively bill patients (for example finding insurance options, correctly classifying patient visits, etc), and improve revenue collection.

The contract with the hospital is performance based.  Accretive receives a percentage of the savings that they are able to achieve for the hospital, compared against a pre-contract run rate.  They also receive incentive fees that are based off of additional revenue the hospital generates due to their improved collection and billing processes.

This business model means that Accretive doesn’t get anything until later in the contract, after they have implemented their solutions, made changes to the processes, and the hospital has begun to reap the results.

It’s also a tricky business model for revenue recognition; billings are in part based on costs not yet invoiced by the customer and that have to be estimated, and the agreements typically include clawbacks if cost reductions aren’t maintained.  The consequence is that GAAP accounting insists that very little revenue be booked until the end of the contract.

Now lets talk about Ascension Health.  Ascension is Accretive’s biggest customer.  Accretive was actually born out of Ascension in 2004.  In 2015 Ascension accounted for 59% of gross cash generated (Gross cash generated is effectively the company’s internal metric of revenue equivalence it uses to get around GAAP accounting. I’m just going to refer to it as revenue from this point forward.  Likewise, I am going to refer to their net cash generated as EBITDA, which is basically what it is).

The relationship with Ascension is not always friendly.  Just around the time when the black cloud of lawsuits and SEC investigations was beginning to lift Ascension dropped a bomb on Accretive, writing a letter where they discussed taking the company private for about 50% of what was then the market capitalization (around $2.50 per share).  In a not so veiled threat Ascension said that if the takeover didn’t proceed they didn’t plan to enter into a new master services agreement with Accretive once the existing one expired in mid-2017.

In December of last year Accretive came to an agreement with Ascension whereby Ascension would drop the takeover offer, not walk away from their relationship, and enter into a new long term agreement for services.

The agreement is for 10 years.  Right now Accretive is responsible for about $6.5 billion of Ascensions net patient revenue.  The new agreement will add another $8 billion of net patient revenue from new Ascension hospitals and affiliates.

As part of the agreement Ascension gets an even larger piece of Accretive.  Accretive issued them $200 million of preferred stock paying 8% and convertible into shares at $2.50 (so 80 million shares).  Ascension also received 10 year warrants for 60 million shares at a price of $3.50.

Accretive has about 107 million shares outstanding not including any of the above conversion.  The market capitalization is about $200 million, cash is about $200 million and apart from the preferred there is no debt.

To put the size of the deal with Ascension in perspective, Accretive has provided guidance of $200-$220 million of revenue in 2016.  This would be “pre-Ascension new deal” revenue.  On the fourth quarter call management said that for every $1 billion of net patient revenue they on-board, they expect $40 million of revenue (they reiterated this statement on the second quarter call).

So basically Accretive stands to more than double revenue at the end of the on-boarding, which is expected to take 3 years (including this year).  The new business from Ascension will initially be at similar margins to the existing business, and those margins will improve as time passes and Accretive’s own systems and processes are implemented.

The tumult of investigations and Ascension related distractions has caused Accretive to lose some other business over the past year.  Ironically, the company had a blow-out first quarter on a GAAP basis, generating $167 million in net income.  But that’s because of customer attrition, which allowed them to recognize deferred revenue that had been accumulating for those customers.  The last thing you want to see with Accretive is strong GAAP results.  Hospital count dropped from 77 in the first quarter to 72 in the second quarter.

Nevertheless, the amount of new business coming from Ascension is substantial.  Accretive said that they expected EBITDA margins on the combined business  showed be in the mid to high teens.  At $400 million to $500 million of revenue Accretive stands to generate significant cash for a company with a $200 million enterprise value.

Still, there is enough uncertainty to warrant only a small position.  Will Accretive be able to turn around the business outside of Ascension, keep existing customers and win new customers?  Or have the last 4 years been too much.  And what is the state of the company itself?  Presumably after all the lawsuits, restatements, executive shuffles and unrest morale must be low so will they be able to execute?  Finally, the new agreement with Ascension is a departure from the legacy model; Ascension employees will be joining Accretive and Accretive will be taking over most aspects of the revenue cycle, as opposed to just advising and consulting.  Will this new model work and will it ultimately be as profitable as management predicts?

Lots of questions.  Nevertheless, if the questions are answered positively the upside is going to be significant.  While I did add to my position recently, I will keep it relatively small (its a little over 1%) until some of the answers become more clear.

Radcom Moves

Radcom has had no significant news since my last update.   Yet we still only have a single NFV deal.

I’m not too worried about how this plays out in the long term.  What I am bracing myself for is the scenario where Radcom presents third quarter results, gives a positive qualitative update on the pipeline but no quantitative progress.

Radcom has a $220 million market capitalization and a $170 million enterprise value.  It trades at 5.7x EV/sales.  I figure this means they need to grow revenues at 30-40% next year to justify that multiple.  That means another AT&T sized contract.

The market is going to want to see that happening by the third quarter report.  If it doesn’t the stock is likely going to sell back down to a multiple that reflects more skepticism.  Unfortunately these carriers aren’t known for their fast movements.  What’s another 3 months for a CSP looking to overhaul their network?

If such a sell-off occurs, it may be a great buying opportunity, but I’m not thrilled with the idea of getting from here to there.  Nevertheless the opportunity with Radcom seems to still outweigh this risk.  I sold a couple shares at $19 but less than 5% of my position.  So I’m just bracing myself for a potential pullback if carriers live up to their name.

New Position: CUI Global

I bought CUI Global because it has a new and better technology (for measuring gas composition) than the current standard, and they can sell that technology into a large addressable market (gas transmission companies and chemical plants).  But so far its been a struggle to gain a foothold into old-school industries that are not used to change and that are currently dealing with a protracted downturn.

Given this trajectory, it’s most likely that even if the story works out it will A. be delayed longer than anyone would have expected and B. have a number of false-starts and hiccups before finally showing consistent growth.

CUI Global operates two segments, Electronics and Natural Gas Integration.  The electronics business has been treading water and while it may have some medium term potential its not really the focus of my purchase.  There is a good discussion of the electronics division in the Q&A of the last quarter conference call.  For this write-up I am going to focus on Natural Gas Integration.

The natural gas integration segment sells a product for sampling and analyzing gas composition that is quicker and graceful than the existing technology.  Called the GasPT system, it consists of a probe and analyzer, both of which are unique in design and operation:

GasPTThe GasPT system is faster and cheaper than existing solutions.  These legacy techniques have been around for 60 years.  Their process works like this: gas is pulled out of the pipeline using a high pressure probe, transported to a gas conditioning unit which lowers the pressure to sea level, and transported through the low pressure pipeline to a gas chromatogram located in a shed where it takes 20 min to analyze the gas and report to operator the chemical content, from which energy content is inferred.

Installation of the incumbent technology costs a quarter of a million dollars, is a 6 week job, requires that a kiosk is built on site, that concrete, pipe, is laid and the carrier gas has to be replaced monthly.

In comparison, the GasPT system resides right on the pipeline, measures gas directly via the Orbital probe, there is no ancillary gas required, the cost of installation is $55,000 and it takes 90 minutes to install.   There is no operating overhead and the gas is analyzed in seconds.

Management has said on numerous occasions that during the bid process for the recently won Snam Rete business (more on that in a minute), they went up against Emerson, ABB and Elster and the head to head competition demonstrated that there is no comparable technology on the market.  They have also said that they are at least a couple years ahead of the competition in terms of development.

The company sells the GasPT system into two verticals:

  1. Natural Gas Transmission Operators
  2. Process Control for compressors and turbines.

The transmission operators they are trying to sell into are large national pipeline companies that aren’t easily receptive to change.  It took some time but early this year they began to make some inroads.  The company inked a contract with Snam Rete, an Italian gas transmission company, in February 2016.

The initial purchase order was for 400 units.  CUI Global has been delivering units at a rate of 50 per month.  The  total volumes of the contract is for 3,300 units with the opportunity for that to expand to 7,000 units.  They expect that next year volumes will ramp to 100 per month.

The thesis here is that

  1. The Snam Rete will ramp to 100 units next year and that will be enough volume to move the natural gas integration segment to profitability
  2. The Snam Rete contract will open the door to other national transmission operators.

CUI Global is engaged with transmission companies in France, Spain and Germany.  Listening to management I get the impression that they are furthest along in France, where in the second quarter they announced a distribution agreement with Autochim for sales of GasPT units in France and Africa.  They also talked on the second quarter call about their recent engagement with Transcanada Pipelines.

Worth noting is that Snam Rete is a low pressure pipeline delivery company, which is unusual.   Almost all gas transmission at a large scale is done at high pressure.  Because of the unique nature of their operations, Snam Rete did not have to purchase the whole solution from CUI Global, they took the analyzer but not the probe.  The other potential customers will be taking the full GasPT solution.

They are also gaining traction in its second market vertical, gas processing facilities.  On the last couple of calls they have talked about a contract with an ethylene plant operator in Texas.  In the first quarter they said:

Recently Orbital NA announced its receipt of a purchase order from a large scale ethylene plant operator in South Texas to design, build and deliver nine patented, ultra fast and accurate in depth, VE sample probes and sample systems and another purchase order for an additional six VE sample probes. That order totaling almost 1.8 million is a trial project which may result in a similar deployment of our technology across 54 other facilities operated by the same customer worldwide.

In the second quarter they qualified the potential of this operator as being a $100 million opportunity if they can expand the order to all the plants run by this operator.

Extending their reach into North America a bit further, a couple of weeks ago they signed a licensing agreement with Daily Thermetrics for the sale of the VE Technology, which is the probe portion of the GasPT unit.

I’m not sure what to make of management.  They tell a good story.  I came across the idea from a conference they participated in (the rebroadcast has expired but I can provide it if you email me).  What they described was compelling.  But going back through their old calls and financials, the execution hasn’t been great, though this may also be more of a function of the businesses they are trying to tap into.

CUI Global has a market capitalization of $110 million and cash and debt come close to cancelling one another out on the balance sheet.  When I model their business, it looks like they need to double the sales of the energy segment to get meaningful profitability for the company.

That may be a tall order.  But the contracts in the pipeline are big numbers.  The energy segment is operating at a run rate of $7.5 million per quarter.  The contract with Snam Rete for the 3,300 units was for $60 million Euro.  Presumably contracts with other transmission providers would be similar.  Likewise, the potential of this single ethylene plant operator is $100 million.

So the total addressable market (TAM) appears to be significant.  The company addresses its TAM on slide 11 of this recent presentation.

The question, which remains valid, is whether they can really gain traction and become the go-to solution.  The thing is, its binary.  Either they get more contracts or they don’t.  I don’t think you can sit on the sidelines and wait and see how it plays out.  Because the next contract, if it happens, will likely be the big move when it happens, and the stock will gap before you can react.

New Position: Jones Energy

I took a position in Jones Energy because of their recent acquisition of STACK/SCOOP acreage. Jones has long had a large acreage position in Oklahoma, but their target has been northwest and southeast of the STACK/SCOOP prospect, where they have targeted the Cleveland formation.  The new acreage should give better returns and more prolific well results.

The story you want to see play out is one where attention grows for the STACK/SCOOP and the play becomes recognized for its multi-zone potential that is close to, if not on par, with the Permian.   That should push up the price per acre and make the acquisition look even more attractive (it already looks like a good deal to me compared to other recent transactions.  Meanwhile an upcoming drill program should give some prolific IP30 results that will add excitement to the story.

At the moment the STACK/SCOOP does not have quite as good of returns as the Permian but they aren’t far off. Demonstrating the viability of the play even at these prices, Newfield points out (in their June presentation) that rig count is rising at a similar rate as the Permian:  

There are a number of large operators in the STACK. Devon has the biggest position, and Marathon, Cimerex and Newfield are also players. Newfield was the first mover in the play. Gastar (who I will talk about shortly) provides a good map of where the operators are.

Continental is a player in both the STACK and the SCOOP. They have acreage that is in the northwest for STACK (Blaine, Dewey and Custer county) and in Grady county for SCOOP.

Jones acreage (see the map below) is in the southern part of Canadian county, so around the Chaparral acreage, and south of that into Grady county. The northern border of Grady is roughly where the STACK ends and the SCOOP begins.  The STACK and SCOOP regions are considered distinct because the geology changes, as the Meramec formation dips down and is replaced by the upper and lower Sycamore.

Jones bought their acreage at $7,600/acre which is on the low side of other transactions I’ve seen. Marathon bought 61,000 acres at $11,800/acre (here they define their position). Newfield bought 42,000 acres from Chesapeake at $10,000/acre. Back in December Devon bought acreage at $17,000/acre. Continental recently sold SCOOP acreage that is in south Grady and further SW into Garvin county for about $9,000/acre adjusted for production. I don’t see too much evidence that Jones bought an inferior acreage position, other than that the STACK acreage is further to the southwest than the core STACK region at the intersection of Blaine/Kingfisher/Canadian.

As I said, STACK results look comparable but slightly less prolific than Permian. Most of the results are drilling into the Meramec formation, some drilling into Woodford. Well costs seem to be in the range of $4.5 million to $6.5 million for a 10,000ft lateral, which is a big range but I think that is because of changes in depth across the play. Continental is an outlier, they operate much further west than the rest of the operators and their wells cost $9-$10 million. The IP30 for the wells are around 1,500 boepd, with some wells delivering as high as 2,000 boepd.  Continental has seen its more expensive wells hitting 3,000 boepd. EUR per well is around 1 MMboe with again Continental being much higher at 1.7 MMboe. All operators report ROR that is consistently in the range of 70-80% at $50 oil.

The northern part of Jones acreage is prospective for the Meramec but transitions to the Sycamore as you go south (so where the STACK becomes SCOOP).   Though there isn’t a lot of information I’ve found on the Sycamore, what I have gathered appears to validate that well performance is similar to the Meramec.  Jones provides a few well results from Marathon and a company called Citizen Energy (who unfortunately is private and so there is no information on their website) in the map I showed above.

A second prospect in the STACK/SCOOP is the Woodford shale.  Their are a few Newfield well results in the Woodford referred to in the Jones map.   Continental seems have cracked the Woodford code recently.  Continental talks about how they are seeing a 40% increase in EUR on offset wells from recent wells where they have tweaked the completion techniques for $400K incremental cost.


This is pretty interesting, especially because Jones acreage looks like it is in same area as Continental.  Below outlines where Continental has been drilling:

While the STACK/SCOOP lags the Permian in performance, the perception I get from listening to comments from Marathon, Cimerex and Newfield is that the area is earlier in the learning cycle, and we should expect further improvements (like we may have just seen with Continental in their Woodford SCOOP wells).

Cimerex in particular has acreage both in the Permian and in the STACK and they are allocating significant capital to both plays.  The gave a good presentation at Enercom, where they talk about both the Permian and the STACK/SCOOP making it easy to draw some comparisons.

Jones also has legacy acreage that is not too far away from the STACK/SCOOP.   They have 100s or maybe 1,000s of wells in this area targeting the Cleveland formation. In total they have 180,000 acres in Ellis, Roger Mills and Beckham county in OK and in Lipscomb, Ochil tree and Hemphill county in Texas. This acreage is to the northwest of the STACK but its not that far northwest. Much like the STACK/SCOOP, the acreage has multizone potential.  Jones has identified a number of potential zones in addition to the Cleveland. There is the Marmaton and Tonaka for example. Neither has been tested much. This article does a good job of delineating each of the STACK, SCOOP and the Cleveland/Tonkawa.

With respect to the multizone potential of the STACK/SCOOP and their legacy Cleveland land I thought the following comment was interesting from their second quarter call:

John Aschenbeck Got it, very helpful. One more if I could sneak it in on the Osage [ph], I’ve had a few operators say that Meramec EURs could potentially be possibly on the Osage as well leading you to believe Meramec-like returns in the Osage as well. Have any thoughts on that?

Jonny Jones We’ve got a 21-zone stack in the western Anadarko basin, of which these are just a couple of zones. We’ve been believers for a long, long time that there’s many pays out here, some of which are the ones you just mentioned, that really have not been exploited with modern technology. We have a lot of them in our section. People are just now starting to parse all the different trenches of the Mississippi and then that’s all they are. But there is a lot of other things out here besides those zones that are attractive. I think you’re going to see that come to fruition over the next six to nine months as folks actually start trying these different zones. The stack, the scoop, all these different zones right now are not one zone. There’s multiple pays here.

New Position: Gastar Exploration

Gastar is the other Oklahoma operator that I have taken a small position in.  By any traditional metric Gastar is a disaster.  They have about $575 million of debt versus a $125 million market capitalization.   Absent their hedges they aren’t generating enough cash flow to cover their interest payments.

But they own a lot of acreage (84,000 acres) right around the heart of the STACK:

The big question is whether the land in Kingfisher, which is a bit north of where Marathon, Newfield, Devon et al are drilling, is as prospective as the land further south?  I’m not sure, though from a number of presentations I’ve seen delineating the extents of the STACK it is clear the play extends well into Garfield county, so Gastar’s acreage is far from the edge of the formation.

While Gastar fails miserably on traditional valuation techniques, the stock looks pretty attractive if you apply acreage valuations that are comparable to recent land sales prices in the area.  By my calculation the current share price is reflecting about $6,000/acre for the STACK position and $4,000/acre for WEHLU (the WEHLU is on the eastern edge of the STACK and I don’t have as much information on how prospective it is but Gastar has been drilling decent Hunton wells there for a number of years).  The NAV is very levered to appreciation of this acreage.  At $8,000/acre for the STACK acres the NAV rises to $2.22.  If you use $11,000/acre the NAV is over $4/share.

I don’t know if Gastar gets revalued up to reflect the going rate of a STACK acre or whether the company sinks into bankruptcy.  I know they are going to do some drilling to prove out some of their STACK acreage and those results will be the key.  I also know from what happened to Resolute that when a company goes from reflecting bankruptcy to being valued based on current acreage prices the move can be pretty amazing.

New Position: Zargon Exploration

Unlike the first two positions oil and gas positions, Zargon isn’t a play on the STACK/SCOOP.  The company’s operations are primarily in Alberta with a small amount of production in North Dakota that is likely to be sold in the near future.

Up until July Zargon had too much debt and there was some thought that the company would eventually become a victim of bankruptcy.  At the end of the second quarter the company had $122 million of net debt to go against second quarter funds flow of $3.5 million.

But on July 22nd  Zargon announced the sale of is southeastern Sasksatchewan assets for net proceeds of $87.5 million.  The transaction significantly reduces debt and makes it much more likely that the company will make it through to the other side of this oil price downturn.

With the use of proceeds put towards the debt, the company’s net debt position is expected to be around $35 million.  I was buying the stock a bit lower than it is now, but even at 90 cents the market capitalization is only $27 million, meaning that the enterprise as a whole is going for $62 million.

This doesn’t seem like a bad price to me considering what you get.  After the asset sale Zargon is left with about 2,800 boepd of production, 79% of which is liquids.  The production breaks down to 400 boepd from North Dakota, 2,000 boepd from various Alberta land packages, and 530 boepd from their Little Bow Alkali-Surfactant-Ploymer flood project.

All of these properties are low decline.  Little Bow production should actually increase through the end of the quarter before stabilizing at over 600 boepd.  This number could go higher if oil prices recover as the company has suspended alkali and surfactant injection because of the economics of the project at current oil prices.


The other properties are very low decline, 14% according to the September presentation.

The properties as a whole have a proved PV10 of $108 million and and a proved plus probable PV10 of $176 million.  The enterprise value trades at a discount to the proved value of the reserves.  This is at forecasted prices though, and that forecast is assuming a slow oil price recovery through to 2030.

I think the current price is probably reflecting about $45 oil.  What I like about the stock, and why I took a position, is that at $55 oil the stock is probably a double at least.  Its probably not a great long-term hold, as I don’t see anything in particular about the properties that excites me, but as a vehicle for playing a price recovery in oil I think its worth a position.

One Last New Position: Limbach Holdings

I took a position in Limbach Holdings after one  of the funds I follow, Dane Capital, took a position.  Dane Capital is the same fund that I got the idea for RMG Networks from.  They have written up Limbach in two pieces on SeekingAlpha (here) and I’m not going to add more on the name right now because they describe the thesis quite well.

Portfolio Composition

Click here for the last four weeks of trades.

Below you will see that I’ve gone back to my old format for portfolio composition.  Those of you who have followed the blog for a while will know my love/hate relationship with RBC Practice Portfolios.  I used to be able to use the portfolio holdings page provided by RBC, but then they introduced a bug which basically screwed up the gain/loss on a position if you partially sold out or added to an existing position.  So I started tracking my portfolio via spreadsheet as well as with the RBC portfolio.  This is time consuming so when RBC introduced a new portfolio summary that wasn’t great but at least wasn’t totally out to lunch, I began to report it and have used it for the past few updates.  Well this week I realized that this summary has a bug as well, so I am back to reporting via spreadsheet.  I also took the time to add a function to my code that colors the gain and loss columns in green or red.


Week 266: Loving the lack of volatility

Portfolio Performance




Top 10 Holdings


See the end of the post for my full portfolio breakdown and the last four weeks of trades

Thoughts and Review

Its been a good market for my approach to investing.  Volatility is low and the market isn’t really making any big moves.  Individual companies are being judged on the merits of their business and the macro picture is taking a back seat.

I believe I have also benefited from being more selective.  As I have written previously, I’m taking fewer flyers, being more suspicious of value, staying away from dividend payers and focusing on growing businesses and emerging trends.

As a result these have been the best couple months that I have had in a while.

I made a number of changes around the edges of my portfolio, adding starter positions in a number of names, reducing position size in others, but I remained relatively constant in spirit: I have two large positions: Radcom and Radisys, and a number of small 2-4% positions where I am waiting for a reason to make them bigger.

With respect to my two biggest positions, nothing was said or reported over the last month, including their second quarter results, that deters me in my allocation.  I will give a quick synopsis of Radcom’s quarter below.  With Radisys I spent a lot of time on them a couple of months ago so I will wait until the next update to say more. In short though, the results were fine, they are engaged with a lot of carriers, they seem extremely confident that those engagements will lead to more deals and we will wait and see what new business the second half brings.

I did add five new positions to my portfolio in the last month.  I’m not sure what it was about this month that made it such a boon for ideas.  It wasn’t because I worked unusually hard to find them.  I just went about my usual process and for some reason kept coming up with interesting stocks.

I’ll talk about 3 of those positions (Bovie Medical, Hudson Technologies and Sientra) in this post.  In the interest of space I will leave the other two, Mattersight and CUI Global, for next month.  This being an August update I have a lot of earnings news I want to talk about and I don’t want things to go overly long (though I suppose it always goes overly long, so relatively anyways).

With that I will cut-off my general comments and get straight to it, starting with some company updates followed by my new positions.

Radcom – waiting on the next deal

Radcom announced results on August 8th but more importantly was a separate news release giving an official endorsement of Radcom’s technology from AT&T along with the recognition of the role Radcom will be playing in AT&T’s open source NFV backbone ECOMP.

The press release from AT&T validates the work that Radcom has done, and should make the company the go-to player for other service providers following in AT&T footsteps with ECOMP.  I have always felt the big risk with Radcom was the execution of such their large and cutting-edge deployment with AT&T and that something went off the rails.  That seems off the table now.

As an aside, I found this webinar, which includes a description by AT&T of what they are trying to accomplish with ECOMP, to be quite useful in trying to understand the platform.

While the second quarter results were inline, on the conference call Radcom’s CEO Yaron Ravkaie made a number of bullish comments.  He said that Radcom was now in discussions with 9 carriers, up from 5 in the first quarter, and that the earlier discussions had progressed positively, with feedback coming back from carriers that “this is exactly what we need and we want to progress with it”.

Interesting new opportunities are also opening up.  Radcom has soft launched a new adjacent product to MaveriQ that Ravkaie called a “very important component of NFV implementation”.  I am assuming this was a joint effort with AT&T to some degree since the product is already being used by AT&T even though they have only started marketing it to other carriers.

A final interesting comment was made at the end of the call when Ravkaie referred to a Tier 1 carrier that was recently in their office in Tel Aviv, their “head of their NFV program spend hours with them”, he went “deep into their NFV strategy” and was interested in “partnering” and “co-creation of cutting edge stuff”.

The negative with the story is timing; we don’t know when the next deal signs and telecom providers are notoriously slow footed.  I felt like Ravkaie was cautious about timing, making the following comment on the second quarter call:

the next coming quarters is going to change some of these into deals, and again I can’t really comment on when it’s going to happen, and on exactly when is the next order is going to come in, because everything is so new, so it’s hard to predict. And because as of the end of the day, big solutions in the telecom’s environment, so it does take time

Nevertheless it remains a great story.  The stock has moved up strongly since earnings.  I’m not sure where it goes in the short run, it all depends on the “when” with the next deal, but I see no reason to believe they won’t get that next deal done at some point.

Willdan looks good

The single sentence story on Willdan is that the company had very strong second quarter results, beat estimates significantly, guided higher for the year, said on the conference call that they have signed contracts that will be deployed in 2017 that they just can’t announce yet, said that revenue in 2017 will show further sequential growth, and said that their micro-grid strategy is taking hold with “increasing evidence to suggest that overall spending on microgrids will increase significantly in the coming years and we are well positioned to capitalize on this emerging market opportunity.”

Willdan reminds me of Argan, which was a little engineering, procurement and construction company that I found on the Greenblatt Magic Formula list back in 2011 (I just checked and its still there!) when the stock was $8.   I bought the stock but sold it a few months later at $14.  Argan continued to run from single digits to almost $50 now on the back of successful growth of their engineering services.

Argan demonstrates how engineering services can be very profitable if you have a team of the right professionals and you are delivering a service along the right trend.  I have said before that I think that we are on the cusp of changes to the power grid that will mirror what we saw first in computing and storage (ala the advent of the datacenter) and currently in telecom service providers (which is why Radcom and Radisys are my biggest positions).  If I am right about that Willdan is positioning itself to be in the middle of the build-out.

Oclaro was fine but Infinera made it a bit of a debacle

Oclaro had an excellent quarter, beating estimates and putting out strong growth guidance which was good news.  Unfortunately I only held about half the position that I had only a few days before.

I got turned around by a report out of Needham entitled “Optical Super Cycle”.  I don’t get Needham research, but I caught wind of a report they had in mid-July that described an optical upgrade super-cycle and managed to get my hands on it. Its compelling; the explosion of data is causing metro, long-haul and data-center interconnect to require upgrades simultaneously.

Its exactly the reason I am in Oclaro.  But after being convinced by Needham, I felt like I needed more.  So I went with one of their recommendations, a company called Infinera that seemed to be well positioned in long-haul, had the ability to take market share in metro, and wasn’t expensive compared to its peers.  No problem right?

Well Infinera announced results three days later and gave probably the worst guidance I have EVER seen.  Ever.   The company guided third quarter revenues of $180-$190 million versus Capital IQ consensus of $271 million.

This caused a couple of things to happen.  First I lost a large chunk of my investment in Infinera.  Second, it really shook my confidence in Oclaro, which is a much larger investment for me.  After some back and forth I decided to cut the position in Oclaro in half.

My reasoning hear was admittedly a bit suspect.  Infinera didn’t appear to be a big Oclaro customer.  Oclaro has significant business in China, a market where I don’t think Infinera plays at all.  The evidence from the call was that much of Infinera’s problems were internal, in particular that they weren’t gaining share in the metro market like they had anticipated and that a recent acquisition wasn’t bearing expected fruits.  Nevertheless what Infinera’s disaster highlighted to me is something that I have always worried about with Oclaro; that I will be the last to know when the optical cycle turns, and that these cycles turn quickly.

Well Oclaro announced their results and they were stellar.  Both their results and those of some others suggest that the cycle hasn’t turned yet.  Oclaro said they expect at least 30% growth in 2017.  Gross margins are expected to expand into the mid-30s from the high 20’s they were only a quarter ago.  Their 100G transceiver business has doubled in the last year.  The company is well positioned with the two vendors in China (Huawei and ZTE) where their growth depends.   The stock has continued to go up since, nearly hitting $7 on Friday.

But I can’t add.  I’ll stick with my half position but that’s it.   The cyclicality scares me, in particular the fact that they don’t really get to pass through price increases even when the cycle is in their favor, so the benefit is all volume and when the cycle ends or when inventory builds they will be smacked.   Needham could very well be right and maybe that doesn’t happen for 3+ years.  But I haven’t been able to put together enough knowledge to feel really confident about that.   So I think I’ll just leave that one as is.

What is going on with BSquare and DataV is interesting

Here is why I added to BSquare.   They had a crappy quarter. The stock tanked.  The company has a market capitalization of $54 million but subtracting cash the enterprise value is only $27 million.  The market has left them for dead but I’m seeing data pointing to how their new DataV product might on the cusp of taking-off.

Right now DataV has one announced customer, a $4.3 million 3 year contract with an industrial company.  Because this company is so small, and because DataV has 70% margins, it won’t take many contracts to be meaningful.

In my last blog post I highlighted a career opportunity I read about on the BSquare site.    In particular I noted the following language in one of the sales job postings:

Bsquare is investing significantly in marketing demand generation tied to its industry leading DataV IoT platform.  Market response has overwhelmed our current sales capacity, and we are looking for proven inside sales dynamos to join our team

I thought that was pretty positive.  Last week I went and looked at the job postings again.  There were a bunch of new one’s for android developers but also I found they updated the posting I referenced before, (its here) with the following additional language about responsibilities:

Responsible for making 30-70+ outbound calls (including follow up) per day to inbound leads

I don’t know if this is saying what it reads, but an inbound lead should be a company that has first contacted BSquare, so 30-70+ calls to those leads is an awful lot.  Are they exaggerating?  Or is there that much interest in the product?

They also quietly published this interview on their website, describing the integration of DataV with a heavy-duty truck environment.   Whats confusing about this is that there was a question on the first quarter call that suggested to me that the one DataV customer they had was VF Corp, which is not a trucking or industrial company.

The company has like zero following.  There were no questions at all on the last conference call.  The only intelligent questions I’ve heard in the last few calls is some private investor named Chris Cox who I am presently trying to hunt down (feel free to drop me a line if you read this Chris!).  I don’t think anyone cares about this name.  Maybe there is good reason for that.  But maybe not.

RMG Networks – will have to wait another quarter for the inflection

RMGN results were somewhat lackluster, the CEO had said that they would be growing revenues sequentially from here on out on the first quarter call and that didn’t happen. Revenues were flat, the issue was that the international business lagged and some timing of sales issues.  In particular it sounded like the Middle East was down a lot sequentially.

So it wasn’t a great press release but the color on the call was very positive.  In a separate news release they said they had a sales agreement with Manhattan Associates and gave further color on that relationship on the call.  RMG Networks had previously been something called a bronze partner with Manhattan which meant they would be recommended by sales if it came up but there was no compensation to the sales staff for any sales they created.  The new agreement integrates them into the selling package, most importantly now is that sales gets commission on the RMG products that they sell.  Because the products are complimentary this is expected to drive sales.  Manhattan Associates is a supply chain solution company so the partnership is right in line with what RMGN is trying to expand into and is also a $4B company so much bigger than RMGN.

They also said they have similar relationships that have been agreed to in principle but are not press releaseable quite yet with two other partners.

The second positive is that the 3 supply chain trials that they have referred to in the previous quarter has progressed into the purchase phase.  We should start to see revenues from those in the coming quarters.  The CEO gave a lot more color around the sales pipeline and how many leads they are generating and also with respect to the team that they have put together.  He really tells a good story, though that can be taken both ways I guess.  He is bringing people aboard that the market likes, signing agreements and getting their foot in the door where needed so I am inclined to believe this is just a waiting game to see how it translates into revenues in coming quarters.

Medicure has lots of catalysts, have to wait for them to materialize

As for Medicure, results again were probably a bit weak, I would have liked to see revenue at $8 million but $7.7 million is still pretty strong.  The bottom line was hurt by a large stock option expense, and they price their options as awarded so it all hits in a single quarter, you don’t get the spreading out of the option effect that most companies see.

Overall the idea is still there, they are continuing to gain market share with Aggrastat, hospital bags purchased increased 16% sequentially, the company said that June was their highest sales quarter since December and in a response to one of the questions on the Q&A they implied that the disconnect between sales and scripts should resolve itself into high sales, likely in the third quarter.

They are on-track to hear back from the FDA with respect to the bolus vial in the second half of the year, and the Complete Response they got from the FDA back in June for the STEMI indication sounds rectifiable.  They said there were 7 concerns that FDA had identified, 6 have been addressed and agreed to by the FDA and the seventh they have sent in their modification and are awaiting the response.

They also provided some color on Apicore.   It sounds like they expect to plan to purchase the other 95% of the company, said that Apicore  continues to grow over and above the $25 million of revenue they generated last year, and they have a new cardiovascular generic that they are developing along with Apicore that they expect to submit.  They were asked again about the price for their purchase option on Apicore and they again said they wouldn’t disclose, which is unfortunate.  They commented that they have built out their sales and administrative staff in response to the higher Aggrastat demand and it now has the ability to support multiple products.  In particular they can add this new generic with no additional staff increase.  They are also on the hunt for acquisitions of other drugs that are complimentary and low risk.

Vicor was Disappointing

I was disappointed in the second quarter results from Vicor given the expectation they had set the last quarter.  Much of Vicor’s backlog depends on a server standard called VR13.  The new server standard is in turn dependent on a new Skylake chipset being delivered by Intel and the chipset has been delayed (again), this time until the second half of 2017.

The consequence is that rather than orders beginning to ramp beginning in the second half, they likely won’t start receiving order for another 6 months.

I reduced my position on the news.  The company still is a technology leader and they still have the best power conversion solution for the next generation of datacenters, but six months is a long time and I’m betting I can build back the position at lower prices.

As I do continue to hold Vicor, I’ll be sure to follow Intel more closely to see where they are at with the chipset. In the mean time the best Vicor can hope for is to tread water.

New Position: Bovie Medical

I came across Bovie Medical doing a scan of 52-week highs on  This is a scan I like to do as much as possible during earnings season; you can catch stocks that are starting their next leg up because of recently released results.

Unfortunately I was on vacation at the time and so I caught Bovie a little later than I might have otherwise.  I bought the stock at $2.60, which was a couple of days after they had announced earnings. That was up from $2.05, which is what it had opened at the day of their earnings announcement.

Bovie Medical operates in 3 segments:

The “core business” is made up of electrosurgical medical devices (desiccators, generators, electrodes, electrosurgical pencils and lights) and cauteries.  This segment makes up largest percentage of revenues and has flat to low single digit growth historically.   Bovie has said they want to grow the business at mid-single digits which they were able to accomplish in the first half of this year.

The OEM business segment manufactures electro-surgical generators for other medical device companies.  It generated $1.6 million of revenue in the second quarter, up from $941,00 in Q1 and $648,000 in Q2 2015.

The growth from the OEM segment this year was somewhat unexpected.  It was partially due to contract restructuring that staggered contracts – they said that contracts are typically front-end development, back end production so they staggered the contracts to even out the revenue.  The company did say after the second quarter that the expected the growth rate to slow in the second half of the year but still show growth.  From the Q2 conference call comments:

…second quarter performance benefited from purchase orders signed last year and several new contracts were signed in the second quarter that should contribute to revenue growth over the next several quarters.

So the OEM business is posting some interesting numbers but the real reason I bought the stock is a new product called J-Plasma that Bovie recently developed.   J-Plasma is a tool that improves the outcomes of surgeries through an ionized helium stream of plasma. The result is better precision and coagulation without significant heating of the tissue.

The J-Plasma system consists of a helium plasma generator and tool disposable.  Here’s a screen cap of the disposable tool just to get an idea of what it looks like.


The generator (called the ICON GS plasma system) ionizes helium and produces a thin beam of the ionized gas.  You use the beam for cutting, coagulating or ablating the soft tissue.  The disposable is the hand piece for delivering the ionized gas stream.  It sounds like you replace the disposable with nearly every surgery.  There are different disposable hand pieces offered depending on the surgery being performed.  The procedure can be used on delicate tissues like fallopian tubes, ureters, the esophagus, ovaries, bowels and lymph nodes.

The initial generator purchase is in the $20,000 range, and the hand tools average $375.  Bovie said on its second quarter call that they are shifting to a pay per use option with a leasing program to bypass the upfront capital expense of the generator.

The success of J-Plasma didn’t happen immediately.  The company has had the product on the market since January 2015.  For the first year growth was in fits and starts.  In particular, the capital required from the up-front generator was a stumbling block.  The company said the following about the slow progress as recently as the fourth quarter conference call:

The operating metrics and leading indicators for J-Plasma product adoption were strong in the fourth quarter, but sales were below our expectations. We continue to face an exceedingly slow pace of J-Plasma generator sales. While we know that the long sales cycle for capital equipment is an industry wide issue, we also know that our VAC approval track record has been outstanding at over 92%, which makes this situation even more frustrating.

The company is targeting two verticals.   They are currently selling J-Plasma into gynecology (estimated total addressable market, or TAM, of $2 billion) and are moving into the plastic surgery business (estimated TAM of $440 million).

j-plasma-TAMThere are a number of other markets they can expand into including cardiology, urology, oncology and ENT.

There is also the opportunity for the J-Plasma system to be used in robotic surgeries.  Bovie has an expert in robotic surgeries on their medical advisory board that has begun to use J-Plasma in trials.  Results are expected in the first half of next year,  Bovie is also developing an extension to the product, due to be out in 2017, that will integrate into existing robotic surgical systems.  They said on the second quarter call that they are exploring relationships with “existing and emerging surgical robotic systems”.

Sales of J-Plasma increased substantially in the second quarter.  J-Plasma sales were $766,000.  This was up from sales of J-Plasma of $356,000 in the first quarter, which was up from $284,000 in the first quarter of 2015.

The company made a couple of moves in the second quarter that should increase exposure of the product further.  On the second quarter call they announced sales partnerships with two large distributors: Hologix (developer, manufacturer and supplier of premium diagnostic and surgical products) which will add them to their GYN and GYN Surgical line of products ($300mm line), and Arteriocyte, which will start selling J-Plasma to their network of plastic surgeons.

Given that Bovie’s sales force currently consists of a mere 16 employees and 30 independent sales reps, this should increase the reach significantly.  The company said that the two agreements are going to expand their salesforce “by multiples”.

One thing I like about Bovie is that they have a small revenue base to grow from.  In the second quarter revenue was $9.2 million.  This is up from revenue of $7.2 million in the first quarter.

Even though J-Plasma revenue remains in its infancy, incremental growth is still quite accretive to the top line because of the simple math that comes along with the company not being very big.

So we will see how things go in the upcoming quarters.  The one negative of note is when I model the growth out to next year, the company is still only borderline profitable after assuming a similar sales pace to the last couple of quarters.  So we may be a ways away from a real earnings inflection.  Nevertheless, I’m not sure that will matter much if J-Plasma sales can continue at the pace they are at.  If they do, profitability is an eventual inevitability and that is what the market will focus on.

New Position: Hudson Technologies

In an unfortunate turn of events I was listening to the Hudson Technologies presentation at the ROTH conference (the presentation is no longer available but I could send a copy I made if someone wants it) on my bike ride home a few weeks ago.   I was thinking wow, this sounds like a really interesting idea.  So I get home, take a look at the chart and boom!  The stock had jumped from like $3.50 to $5 that very day.


Nevertheless I continued to dig and found that in some ways the stock is actually a better idea now than it was pre-spike.

Hudson is the biggest refrigerant reclaimer in the United States.  The stock jumped on July 18th (the day of my bike ride) because of the announcement of a contract with the department of Defence:

[Hudson] has been awarded, as prime contractor, a five-year contract including a five-year renewal option, by the United States Defense Logistics Agency (“DLA”) with an estimated maximum value over the term of the agreement of $400 million in sales to the Department of Defense.  The fixed price contract is for the management and supply of refrigerants, compressed gases, cylinders and related items to US Military Commands and Installations, Federal civilian agencies and Foreign Militaries.  Primary users include the US Army, Navy, Air Force, Marine Corps and Coast Guard.

So this is a big deal for a little company.  But it wasn’t the primary reason I was looking at the stock.  The story that intrigued me centered around their reclamation of R-22 refrigerant volumes.

R-22 refrigerant, also known as HCFC, is an ozone depleting substance, much like the CFC refrigerant that we all remember from the 1990s.  In fact, R-22 took the place of CFC’s in many applications.  But because R-22 also has an negative environmental impact it was decided by a number of governments t phase the refrigerant out.  In the United States, between now and 2019, production of virgin R-22 will go to zero, as ruled on by the Environmental Protection Agency.

This article did a good job outlining the phase-out cycle by the EPA:

In October 2014, the EPA announced its final phasedown schedule regarding the production and importation of HCFC-22. The order called for an immediate drop from 51 million pounds allowed in 2014 to 22 million pounds in 2015, 18 million pounds in 2016, 13 million pounds in 2017, 9 million pounds in 2018, and 4 million pounds in 2019. No new or imported R-22 will be allowed in the U.S. on or after Jan. 1, 2020.

With R-22 production being phased out, the remaining source of R-22 will be from reclaimed refrigerant.  This is where Hudson comes in as the largest reclaimer in the U.S.  Hudson should benefit from the production restrictions as they gain market share as well as benefit from price.

Its price where things get really interesting.  Over the time period where CFC’s were phased out the price spiked from $1 per lbs to $30 per lbs.

Already we are seeing the price rise.  R-22 ended last year at $10/lbs (up from$7.50), it rose to $12/lbs in the second quarter and prices are currently at $15/lbs.  The company said on their second quarter call saying they are “showing signs of further price improvement.”

Hudson benefits directly from the price rise.  The company has said that they expect that for every $1/lbs rise in the price, 50c should fall as margin.

The opportunity won’t continue forever, but it appears to me that the runway will be measured in years.  While new air conditioning units do not use R-22, the economics of repairs for existing units work in favor of R-22.  At the Roth conference the company said the following:

…lets say you want to repair unit outside of your house, has 10lbs of refrigerant, repairs are $2,500, even if refrigerant is $100/lb its $1,000… [in comparison] new unit is going to cost you $8,000

Hudson has a market capitalization of around $165 million and there is about $30 million of debt.  The company announced 15c EPS in the second quarter.  Keep in mind that this number includes no impact from the department of Defense contract and that R-22 prices have risen another $3 in the third quarter.

I honestly thought the stock was going to take off after the second quarter.  While it flew in after hours (and sucked me in for a few more shares) it gave up those gains the next day and actually traded down 10% in the in suing days.

It turns out that company executives sold a bunch of stock over that time.  The company released a press release on the 10th saying that various executives had sold 1.1 million shares.  That’s a lot of shares for a little company to absorb.

You could of course look at this negatively, but keep in mind that these say executives still hold 20%+ of the company, and so they sold down their positions by about 10%.  They’ve waited a long time for some sort of pay day, its hard for me to put too much stock in them cashing in a bit once it comes.

I really like this idea.

It seems to me that R-22 prices have no where to go but up, that upwards trend has already affirmed itself, and the stock really isn’t reflecting this.  Nor is it reflecting the full impact of the DoD contract, which should be worth around $40 million in annual revenue at similar to higher margins than the existing business right now.  While the stock hasn’t really moved up from my original purchase price so I haven’t been adding much, the size on the high side of what I usually allocate to a new position.

New Position: Sientra

I honestly can’t remember how I came across Sientra.  It was probably some sort of screen, but I have no recollection of what I was screening with.   I’m not getting enough sleep.

At any rate, I came upon the stock the day before it released earnings, which forced me to do some quick work and decide whether I knew enough to take a position or not.

I took a position on that day only an hour before the close.  I tweeted my thought in this tweet (as an aside I’ve decided to try to return to Twitter to comment on my portfolio adds.  I feel like I am missing out on a level of feedback that was often useful).

I always wince when I make a decision like this; I’ve been bitten by acting too fast before.  Yet what drove me to act quickly was because Sientra seemed like a time sensitive situation.

Here’s the scoop with Sientra.  The company sells a breast implant.  Their product was approved by the FDA in March 2012.  By most accounts it is a better product than those that already on the market from Allergan and Johnson & Johnson.  The company was taking market share and sales were increasing.  In the first six months of 2015 sales were $26 million, up from $21 million in the comparable period the year before.  In the third quarter sales were $13 million, up from $10 million in 2014.  Unfortunately the results were overshadowed by manufacturing problems that led to the company taking a $3 million allowance for product returns and suspending further production of the impants.

A few weeks before they announced their 3rd quarterr results the company announced that on October 2nd they had “learned that Brazilian regulatory agencies announced that, as they continue to review the technical compliance related to Good Manufacturing Practices (GMP) of Silimed’s manufacturing facility.”

On October 9th Sientra released a letter that they had sent plastic surgeons regarding the issues with the facility.  Sientra relied on a single source manufacturer.  The compliance issue caused the company to put a voluntary hold on sales.  The stock dropped from over $20 to $10.

Sientra did its best to come clean.  They worked with the FDA and hired a third party firm to verify the safety of the product.  The company was silent for two months and the stock continued to fall, eventually settling at $3 in December.

On January 8th, in a published letter to doctors, the company announced that they had “submitted all of the third testing data of its products to the FDA. He says that, in the company’s opinion, the results show that all are safe and present no significant risk to patients. If the FDA agrees, then their implants will be back on the market.”

The company said that while their investigation found that there were microscopic levels of particulate matter on the products, that data also revealed that even with well controlled manufacturing processes the presence of microscopic particles is unavoidable, and the level of particles from the shedding of a typical laboratory pad would have more particles than their products.

They started selling their implants again in March.  The second quarter was their first full quarter of results. Along with the results the company provided an update on their search for a new manufacturing partner.  As I had hoped, they announced a new partner.

Sientra has entered into a services agreement with Vesta, a Lubrizol LifeSciences company and a leading medical device contract manufacturer of silicone products and other medical devices…Under terms of the agreement, Vesta is establishing manufacturing capacity for Sientra and is working with the Company to finalize a long-term supply arrangement for its PMA-approved breast implants. Sientra anticipates that all project milestones will be achieved for the Company to submit a PMA Supplement to the FDA during the first quarter of 2017.

Sales in the second quarter were $6.2 million, which is only about half of the $14.2 million in sales they had the previous year, but still a very successful initial level considering their limited launch.  In fact the company has to be very careful about how much product they sell before they get manufacturing back up and running.  They do not want to risk another supply disruption.

The response to their return to the market has been positive.  The reason I was so interested in getting into the stock sooner rather than later is because management gave a very positive review of how quickly customers were coming back.  On the first quarter call by CEO Jeffrey Nugent:

We were removed from the market voluntarily and our primary competitors naturally came in and took over those customers that we were no longer able to serve. So what encourages us, and me particularly, is the relative speed and ease of converting those previous customers to come back from those products that they used as a replacement.

So, I could give you a number of other statistics. We have a very high level of analytics inside the Company. We know exactly who is ordering what. We’re following that on a very detailed basis. But as far as pushback, we’re not seeing much. There are virtually no concerns about the safety issues that were raised and we’ve been able to convince those customers that we have the confidence and are giving them the assurance that we are not going to allow them to go back on backorder.

So customers want the product and the market for growth is there.  The breast implant market is large compared to the size of Sientra.  I had to do some searching, and most of the numbers are behind expensively priced reports, but I was able to gather that the US implant market is at least $1 billion.  Sientra had a revenue run rate of around $50 million before it ran into troubles.  It seems that there is plenty of market share left to capture.

The one negative consideration is that this story isn’t going to play itself out in the next couple of months.  The company remains supply constrained and are relying on inventory for sales.  While the details about the manufacturing partner are great, its going to be a while before they are producing new product.  The company said the fourth quarter of 2017, which I believe is probably conservative, but regardless new product is still some time off.

Nevertheless, with a positive response from customers and the path to new product clearing, I suspect we will see the stock move higher as we inch towards that date.  My hope is that we eventually get the stock back to the $20 level, where it was before the roof caved in.

What I sold

I reduced my positions in Air Canada, Granite Oil and Intermap.  Air Canada and Granite continue to be slow to develop.  I did like what I read from Granite’s second quarter earnings release on Friday, and I may add to the position again in the coming days.

Air Canada just continues to lag regardless of what results they post.  I’m keeping what amounts to a start position here, but after seeing the stock struggle for the better part of two years even as the business continues to improve, I just don’t know what it will take for a re-valuation to occur.

Intermap just keeps dragging along with no financing in place.   I sold a little Intermap in the high 30’s but mostly decided to reduce after lackluster news along with the quarterly results released Friday.

In all three cases there is also the consideration that I have come across a number of new ideas as I discussed above and prefer to make room for them.

I also sold Iconix in mid-July but bought back my position before earnings.  However, as I am want to do, I neglected to add back the position in my tracking portfolio and didn’t realize that I had not done so until I reviewed the positions this weekend.  So I will add it back Monday morning.

Portfolio Composition

Click here for the last four weeks of trades.