The following information was extracted from the 2010 10k, the 2011 Q2 report, and the most recent proxy filing. I encourage all to check my calculations and methodology, but I fully believe that you'll find I am providing an accurate portrayal of the economics of what has gone on. I believe the MRGE management team is charismatic, friendly, and persuasive. And I also believe it has engaged in a number of transactions with questionable motives.

With majority ownership of MRGE and a large representation on its board of directors, Merrick LLC, a private equity firm, has substantial influence and seems to put its own economic interests above those of the other common shareholders. Merge has seven board members, four of which are either affiliates of Merrick, of Merge or of both companies. Merrick's CEO, Michael W. Ferro Jr., serves as chairman of Merge's board of directors. Justin Dearborn, MRGE's President and CFO is a former affiliate of Merrick. Board member Gregg Hartenmayer has served as an advisor to Merrick, and Merge's CEO, Jeff Surges, is also on the board. With full disclosure and in plain sight, it appears that management (Merrick) has engaged in what, in my opinion, seems to be corporate mismanagement and self-dealing for the benefit of itself at the cost of other shareholders. Here are two transactions that Merrick was party to.

In April of 2010 Merrick issued preferred stock to itself and a few select other qualified investors at an alarmingly steep discount to face value, thereby giving itself and the other original buyers abnormally high returns with the cost borne to other shareholders.

First, the transaction: In exchange for total proceeds of $41,750,000, 41,750 shares of preferred stock and 7,515,000 shares of common stock were issued. The preferred shares accrued an annual 15% dividend, and management said their valuation was performed by an independent consultant.

Let's allocated the $41,750,000 between the two classes of shares as MRGE did, but this time let's use the average market price of MRGE stock in the prior month in our calculations. The average market price of MRGE stock in March of 2010 was $2.18. So the common stock issued had an approximate value of $16,382,700 (7,515,000 shares x $2.18 per share). Subtracting this value from $41,750,000 gives $25,367,300 of value that would then be allocated to the preferred shares. Dividing $25,367,300 by 41,750 shares of preferred gives a purchase price of $607.60 per preferred share. This is equal to a 39% discount from the $1,000 redemption value. (In the 10-k management estimated they were issued at about $643/share. It all depends on what you think the common stock component is worth. I have disclosed my reasoning. Then the residual amount is assigned to the preferred component.) Now, the preferred shares were privately placed, which means that they don't trade freely, so an illiquidity discount to par value could be justified. But 39%? Especially when management, who controls the timing of their redemption, also owns the shares and would benefit in such circumstances?

Continuing, so according to my estimates the preferred shares were issued at approximately 61% (again, $607.60/share) of their $1,000/share redemption value. Then the shares were redeemed just 14 months after they were issued, giving Merrick and any other original preferred shareholder a 93.5% return (($1,000 redemption value per share + $175.5 accrued dividends per share)/ $607.60 purchase price per share ). See the 10Q for Q2 of this year where this is disclosed but not as clearly as I have done above. Merrick owned 10,000 shares, so the round trip transaction netted them $5,679,000 over the course of 14 months. That is, 10,000 shares x ($1,175.5/share total redemption proceeds - $607.6/purchase price). Impressive.

Surely there were cheaper financing alternatives available than issuing the preferreds at such a steep discount. Even a public common equity raise would not have been so costly. Also, why wasn't the entire issue composed of common stock rather than a fraction of it? Were the deep discount preferreds used to sweeten the deal to the other knowledgeable buyers? Lastly, if Merrick was concerned about losing voting control, then a rights offering would have been appropriate. There's more to this transaction though.

The preferred shares were repaid using the proceeds from $52 million in additional debt at 11.5%. Effectively management substituted $52 million in debt for $41.75 million in preferred stock. Even without this added debt burden the company was struggling to cover its existing $200 million debt obligations. Interest coverage as defined by (CFO + Interest expense)/Interest expense for the last four quarters had been 0.2, 2.1, 0.9, and 1.7, respectively. Was this really a time to be taking out more debt? On the Q2 call management said they did this (swapped the preferred for debt) "in order to improve the long-term cash flow of the company." I do not believe this is an accurate statement. I believe management (Merrick) did this to get a 93.5% return - $5,679,000 - over 14 months. While the coupon rate on the debt was lower than the 15% preferred dividend, the difference in financial flexibility is dramatic. Miss a preferred dividend, no big deal. It just cumulates. Miss a coupon payment and you're filing for bankruptcy protection. Leaving the preferred shares outstanding would have been the prudent thing to do, but that would not have provided Merrick with $5,679,000 in such a short time.

Assuming the preferred shares are never redeemed means the minimum annual financial cost is 24.7%: $150 per year (par value of $1,000 x 15%) divided by the proceeds per share of roughly $607.6. If the preferreds are redeemed, then they must be redeemed at $1,000 plus any accrued dividends. If they are redeemed as early as 1 year later (ignoring the early retirement feature that was eventually relaxed), then they cost equityholders an 89.3% annual financing charge (($1000 redemption value + $150 accrued dividend) / $607.6 proceeds from issuance) -1). So depending on the timing of the redemption, the explicit annual cost of the preferreds is somewhere between 24.7% and 89.3%. As I've already said, surely there were cheaper, more shareholder friendly - friendly to all shareholders that is - financing arrangements.

In light of the above transactions, management not being forthright about the composition of Q3 revenue, and the quiet, timely filing of the shelf registration statement only two days after the Q3 call - all things that I have previously published about - I wonder if investors will take notice and start to view management as a little less credible than before. I know I have. And it should come to us as no surprise that the street analysts have been silent on all these issues. Objective analysis is no way to win I-banking business.

Disclosure: I am short [[MRGE]].

OIS makes ophthalmology imaging and informatics systems that, one would imagine, are easily distinguishable from MRGE's other portfolio of offerings. Is it that hard to simply count the number of product sales of the OIS type and any related maintenance/support revenue and add it all up? Or does MRGE not have such internal control procedures?

Only with the help of the $30 million dollar OIS acquisition was MRGE able to "hit" the revenue numbers again. The press has it wrong, and I maintain my stance that this quarter was again a clear and substantial revenue miss. Simple math shows this. See my previous posts.

Would it kill management to provide the requested numbers? No, but it might just kill the stock price.

I believe this management team is keenly aware of how important public perception is, especially when your company is highly levered and has stock that is priced to perfection.

The market has voted to ignore the points I raise about the top line, but the share counts continue to increase. Luckily for MRGE, continuing to have negative income, higher share counts actually bolster its reported GAAP EPS as losses get spread among a larger share base, making it less negative on a per-share basis.

Disclosure: I am short [[MRGE]].

Reported Q4 GAAP revenue = $64.1 million.
Estimated Q4 OIS acquisition contribution = $3.5 million
Estimated Q4 organic GAAP revenue = $60.6 million
Q3 reported revenue (all organic according to managenment) = $60.1 million.
Estimated organic dollar growth from Q3 to Q4 = $0.5 million

Consensus organic revenue expections for Q4 = $64.5 million
Estimated organic reported revenue = $60.6 million
Negative surprise/revenue miss = ($3.9) million.

Q4 GAAP EPS = ($0.01)
Q4 Reported CFO/share = ($0.05)
Q4 Reported FCF/share = ($0.06)

Management continues to emphasize new deals and top line growth, but the company continues to lack the ability to convert these to meaningful GAAP EPS profits and cash flow per share.

Disclosure: I am short [[MRGE]].

On Thursday of this week MRGE will release its full year and fourth quarter financial results for 2011.

The Q4 consensus GAAP organic revenue estimate that excludes any contribution from the OIS acquisition is $64.5 million.

On August 5 of last year MRGE acquired OIS for more than $30 million. Investors will recall that on the Q3 conference call, after having owned OIS for almost two full months - a company that generated revenue of more than $18 million in FY2010 and which is expected to generate $14 million in FY2012 - management said it did not provide a meaningful revenue contribution to the reported GAAP Q3 revenue number. That reported number happened to fall roughly in line with Street expectations for organic growth. (See my previous post entitled "Q3 - OIS Negligible Revenue Contribution?").

As of the close of Q4, having owned OIS for five months, I think the market will naturally be looking for a Q4 reported GAAP revenue number much larger than the original $64.5 million ex. OIS consensus estimate. I think the market will be expecting at least $68 million in GAAP reported revenue and it will not be as forgiving as it was last quarter with respect to the OIS component and the residual, implied, organic number. Here's the simple math:

$64.5 million, consensus GAAP organic revenue estimate for Q4
+$3.5 million, OIS GAAP revenue contribution estimate for Q4 ($14 million/4).
= $68 million expected GAAP Revenue for Q4.

I believe that a reported number less than this will disappoint. And whatever is reported on a GAAP basis, investors will almost surely subtract about $3.5 million (conservative, considering OIS recorded $18 million in 2010) to isolate what amount came from MRGE.

If the numbers are indeed underwhelming, management may attempt to mitigate any sell off by raising FY 2012 guidance and emphasizing the whole Merge Honeycomb project, but I believe this will have a somewhat limited effect because I think management has lost some credibility over the past six months.

Disclosure: I am short [[MRGE]].

In contrast with what the FINRA database lists, primary source SEC filings reveal that the 11.75% bonds were not issued at 97.266 % on 11/2/2010, but rather were issued in April of that year as part of a private placement. The bonds simply began trading in the public markets in early November as they were converted from the original $200 million privately placed issue to a freely traded public issue with the attendant disappearance of an illiquidity discount. Equivalently stated, all else equal, publicly held, freely traded securities are nearly always priced at a premium relative to their privately held counterparts. The private issue's historic coupon rate and yield would reflect this. So, we would expect the bonds to trade up, to a certain degree, upon such a conversion.

I will concede that not all the reduction in yield is due to the disappearance of an illiquidity discount, but I think about 2% is reasonable, especially for junk bonds undergoing the private-to-public transition.

As mentioned in the original article, the bonds are currently yielding 8.42%. Surveying the results from a basic search for callable, non-convertible bonds with the same S&P and Moody's ratings and approximate time to maturity (May 2015 - May 2016) shows that even 8.42% is still well within junk bond territory with the results of this peer group averaging 7.1%. Pricing the bond at a premium to its private market value simply places its yield to maturity on par with other publicly traded junk credits. I don't interpret this as the bond market warming up to the debt.

(Please perform this search as the result link will not transfer.)

http://cxa.marketwatch.com/finra/BondCenter/AdvancedScreener.aspx

Below are some supporting passages and links relevant to my paragraphs above:

The original issuance terms are succinctly stated in the most recent quarterly filing:

In April 2010, we issued $200,000 of Senior Secured Notes (Notes) at 97.266% of the principal amount, which bear interest at 11.75% of principal (payable on May 1st and November 1st of each year) and will mature on May 1, 2015. The Notes were offered in a private placement pursuant to Rule 144A and Regulation S under the Securities Act of 1933, as amended. In connection with the Notes, we incurred issuance costs of $9,015 (which are recorded in other assets on the condensed consolidated balance sheet). - pages 9-10.

http://www.sec.gov/Archives/edgar/data/944765/000114036111051885/form10q.htm

The notes were then converted to a pubic issue in October of 2010 (and apparently began trading freely in the public market in early November). From the first page of the SEC filing:

We are offering to exchange all of our outstanding 11.75% Senior Secured Notes due 2015, which we refer to as the old notes, for new 11.75% Senior Secured Notes due 2015, in an exchange transaction that is being registered hereby. We refer to these new notes as the exchange notes, and together with the old notes, the notes. The terms of the exchange notes are identical to the terms of the old notes except that the transaction in which you may elect to receive the exchange notes has been registered under the Securities Act of 1933, as amended, or the Securities Act, and, therefore, the exchange notes are freely transferable.

http://www.sec.gov/Archives/edgar/data/944765/000095012310091034/c59316b3e424b3.htm

Also, to round out my response I should note that the additional $52 million of debt that was later issued followed the same pattern: a private placement subsequently converted to a public issue.

Disclosure: I am short [[MRGE]].

Here I present a number of reasons why I am short MRGE stock. I’ll post a few more soon.

First, some background on HITECH and Meaningful Use.

The Health Information Technology for Economic and Clinical Health Act (HITECH) – part of the American Recovery and Reinvestment Act (ARRA) passed in 2009 – mandates that all health care providers adopt electronic medical records by 2018 (approximately). Over the next few years, providers must achieve “meaningful use” status by both adopting a government certified electronic medical record (EMR) software product and demonstrating it is being used in a “meaningful” way (= a checklist of functions as defined by the government). The meaningful use (MU) mandate is being rolled out over three phases. We are currently in phase one. Phase two is targeted to begin in 2013 (approx.) Government incentives for early adoption in phase 1 exist, but penalties for non-compliance will occur in later stages. In phase one, individual providers who adopt a certified MU product and demonstrate it is being used properly are reimbursed for $44,000 (or $64,000 in some cases). Hospitals may qualify for much more -“millions”- but the actual amount depends on various factors. If a provider fails to achieve MU in later years, then government reimbursement for procedures will be greatly reduced.

Reason 1: A Crowded Market

Health care IT vendors who supply government certified (meaningful use certified) software will be the direct beneficiaries of the HITECH act. To date, MRGE has only two government certified products, one for Radiology (Merge RIS 7.0) and one for Orthopedics (Merge OrthoEMR version 4.0). Legacy products often require substantial overhauls to become compliant. To date, I do not believe any additional MRGE products are anticipated to become certified.

Here’s a link to the government website that lists products that have been given the government’s stamp of approval: http://onc-chpl.force.com/ehrcert/CHPLHome

Competition is intense and the list seems to be rapidly increasing. Note that across the entire spectrum of specialties and offerings there are already hundreds of vendors with more than a thousand certified products.

As far as Radiology goes, in addition to MRGE, there are already many other vendors with certified radiology offerings. Here are a few that are listed on the site mentioned above.

1.) Advanced Data Systems Corporation’s “Medics DocAssistant”
2.) Carestream Health’s “Vue RIS”
3.) Corepoint Health’s “Corepoint Integration Engine”
4.) GE’s “Centricity RIS-IC”
5.) MagView’s “MagView”
6.) MedInformatix’s “MedInformatix RIS”
7.) RamSoft Inc.’s “RamSoft PowerServer”
8.) StreamlineMD’s “StreamlineMD”
9.) Zotec Partners’ “Electronic Billing Center (EBC) Radiology Information System”

The same can be said about Orthopedics certified offerings. According to the same website, here are a few:

1.) Athenahealth’s “AthenaClinicals”
2.) ChartLogic’s “Charlogic EMR”
3.) Phoenix Ortho’s “Phoenix Ortho EMR”
4.) Neusoft Technologies’s “NueMD EHR”
5.) eClinicalWorks’ “eClinical Works”
6.) Allscripts – numerous certified offerings
7.) ADP’s “ADP AdvancedMD EHR”
8.) Greenway Medical Technologies’s “Primesuite”
9.) AllegianceMD’s “Veracity”

Reason 2: Meaningful Use Adoption Will Likely Be Slower Than Expected. Many will Defer.

To date, there has been a reasonable amount of Meaningful Use (MU) product adoption, but I believe the preliminary numbers are likely to be front-loaded as tech savvy physicians/ practices have been much more quick to adopt. Who remain are the less tech savvy physicians who are still confused about the whole MU process and its requirements. Adding to the confusion is the wide array of certified products available. From a more typical physician’s point of view, getting up to speed on how to actually get MU stage 1 approved, selecting a suitable product, installing that product, training all the docs and employees at the facility and considering all the disruptions in the normal workflow seems like a difficult and painful process all for the $44,000 reimbursement that will be received in a few installments over 4-5 years. I don’t dispute the carrot, but I don’t think there will be a widespread rush to get it, if at all, at this point in time.

In addition, we are only in Stage 1 and the requirements for Stages 2 and 3 have yet to be determined. (Indeed, they are much disputed by policymakers and healthcare professionals and organizations.) It would seem that many physicians will simply wait for Stage 2 to be defined and rolled out and attempt to get compliant then. Sitting still for a year or two seems to be the easiest thing to do. 

As far as MU revenue contribution for 2012 goes, here’s a rough estimate. a.) assume that each software license (with various ancillary implementation costs) runs about $8,000, b.) assume half of the addressable market (loosely defined here as number physicians in each specialty area) adopts a certified EMR product in 2012, and 3.) MRGE captures 5% of the adopters in radiology and Orthopedics. According to the recent MRGE presentation there are about 27,000 radiologists and 20,000 orthopedic specialists. 47,000 docs x 0.5 adoption rate x 0.05 of the market x $8,000 per license = $9.4 million revenue contribution in 2012. 

I will return to this $9.4 million revenue contribution estimate figure later, but for now please keep in mind that the total dollar amount of revenue growth from 2011 to 2012 is expected to be about $59 million. (2012 guidance midpoint of $294 million minus about $235 million for 2011 )

Reason 3: Image Interoperability Not Currently Part of Any MU Stage Requirements: Limited Demand 

The company’s primary focus is on medical image sharing and medical image attachment to a patient’s overall electronic record. Currently, medical images are not part of a patient’s EMR. MRGE’s I-connect product stack is marketed as allowing for universal sharing and incorporation of images regardless of their origin. (A Google search for “Radiology Universal Viewer” produces a few other competitors who also have universal viewer offerings.) Existing traditional image viewers in use are pretty good. They, too, allow for remote access (i.e. off site, at home, etc.) just like I-connect. 

The issue is that image incorporation or similar transfer functions are not currently required to achieve MU status; they are not part of the current government mandate. Without such requirements, I believe future demand for I-connect will be limited to a handful of large, tech savvy, non-financially strapped healthcare providers that have multiple, disparate technologies and or facilities, similar to many of the 30 or so existing I-connect clients. 

Furthermore, the requirements for MU stage 2 and MU stage 3 are not currently defined. Any claims about future functional requirements are merely speculative. Widespread adoption of interoperability technology will likely need to be forced. After all, it does cost money and research - my own channel checks and interviews over the past few months - indicates that hospitals are now doing only the bare minimum to meet MU requirements. Incorporating medical images is not part of these efforts. At this point, there is no government payment incentive to adopt I-connect. MRGE has about 30 I-connect deals so far, I believe, and none of these buyers adopted I-connect for MU stage 1 compliance reasons. 

As a side note, it is worth mentioning that there is also substantial pushback by healthcare organizations such as the American Hospital Association (AHA) regarding the government’s meaningful use timeline and criteria. There is growing concern that the timeline is simply too ambitious and not achievable. While stemming from good intentions of leveraging all the benefits of IT in the healthcare space, essentially MU has been created and crammed down by the government without sufficient input and thought. It is not a coincidence that original timelines have recently been extended: http://www.informationweek.com/news/healthcare/EMR/232200656
Let me be clear. Of course I believe that from a quality of care point of view, interoperability/ image sharing is beneficial. My stance though is that it will not be adopted via I-connect as rapidly as is priced into this stock. I believe the same about adoption rates of MU EMRs. See Reason 2.

On the Q2 recent conference call, Management said that a good revenue number to think about for I-connect (Access + Share) was $500,000 or below per deal. 

Back of the envelope FY 2012 I-connect revenue contribution estimate: aggressively assume that MRGE signs 40 I-connect deals next year or about 10 per quarter. 40 I-connect deals x $0.5 million ASP per deal = $20 million revenue contribution from I-connect. I qualify this as aggressive because I believe the rapid early adopters who already had relationships with MRGE – the “low hanging fruit” if you will – have probably already signed on. 

We will return to this $20 million estimated revenue contribution later in the summary.

Reason 4: The Rollout of ICD-10 Will Likely Constrain MU Spending and Resources

Health care providers and their IT departments are distracted by another, more immediate and financially threating issue than achieving MU1 status: the upgrade and rollout of ICD-10 from ICD-9. ICD-10 is the new coding measure that providers must adopt by October 1, 2013 to ensure proper reimbursement amounts from the government. A lack of full adoption means that reimbursement amounts will decline. See link below for the American Medical Association: 
http://www.ama-assn.org/ama/pub/physician-resources/solutions-managing-your-practice/coding-billing-insurance/hipaahealth-insurance-portability-accountability-act/transaction-code-set-standards/icd10-code-set.page

And for the Department of Health and Human Services: http://www.hrsa.gov/healthit/icd10/#icd9

As it now stands, not being MU compliant for the next few years will not result in payment adjustments (reduced reimbursements) until 2015: https://www.cms.gov/EHRIncentivePrograms/Downloads/EHRIncentProgtimeline508V1.pdf

Not being ICD-10 compliant by October 1, 2013 will result in erosion of a provider’s existing revenue stream via reduced reimbursements commencing on that date. I would think that ICD-10 compliance is more of an immediate concern. 

Furthermore, the new code is very detailed, much more so than ICD-9. IT issues aside, training doctors and administrators to properly use this new coding language in a way that ensures full reimbursement will require substantial resources, leaving much less time and money for MU considerations. See the American Medical Association’s adoption timeline and suggestions for the ICD-10 transition: http://www.ama-assn.org/ama1/pub/upload/mm/399/icd10-timeline-fact-sheet.pdf

All of this creates additional headwinds for discretionary Stage 1 MU and I-connect spending.

Reason 5: Questionable Acquisition Standards

On August 5, 2011, MRGE completed its acquisition of Ophthalmic Imaging Systems (OIS) for about $30 million. 

OIS was publicly traded and had the following reported financial metrics for 2008, 2009, and 2010, respectively: (amounts in $ millions)

Revenue: 12.5, 13.6, and 18.6
EBITDA: (1.4), (6.1), and (2.5)   Yes, negative EBITDA figures.
Net income: (3.4), (7.0), and (2.4)
CFO: (0.7), (0.1), and (3.3)
FCF: (2.6), (0.2), and (3.5)

The OIS acquisition will increase MRGE’s top line by simple addition of the purchased revenue stream, but by historical standards, everything below the revenue line quickly goes negative. OIS hasn’t recently been a profitable business by any measure and it consumes cash. It goes without saying that MRGE will need to achieve substantial synergies and or overhaul the OIS business in order for this $30 million acquisition to not negatively impact shareholder value.

I think MRGE is too fixated on top line growth at the cost of what goes on beneath it. Just because the top line grows does not mean EPS, CFO/share and FCF/share grow at a commensurate rate. Growth needs to be framed in terms of its overall effect on a per share basis. You can always easily buy and grow revenue via acquisition. Increasing EPS and FCF/share, etc. - performing a deal that is accretive to shareholders - is the tricky part. And at some point investors do start to care about traditional measures of value such GAAP reported EPS, CFO/share, FCF/share and actual EBITDA, and the roll-up game ends.

While I’m on the topic of OIS, Management estimates that it will contribute about $14 million to revenue in 2012.  I use this number again in the summary post.