|Shares Out. (in M):||50||P/E||0||0|
|Market Cap (in M):||1,150||P/FCF||0||0|
|Net Debt (in M):||0||EBIT||0||0|
CompuGroup (COP) is a software provider for the health-care industry. We believe COP offers a rare and an extremely compelling investment opportunity to earn more than 3x returns within 3 years while owning a very high quality, sticky and defensive business that trades for a normalized forward P/E of 12x. We expect COP to benefit from material catalysts in the next 24 months and to compound earnings at 40% per year from 2015 to 2020.
COP operates predominantly in Europe and trades on the Deutsche Borse. It was founded in 1979 by its CEO Frank Gotthardt, who controls just over 50% of the equity. Through a combination of organic growth and tuck-in acquisitions of sub-scale competitors, COP became the dominant supplier of software systems for doctors and dentist practices in Europe. The company holds 40% to 85% market share in its core regions of operations. Europe makes up more than 90% of revenues, with Germany making up 50% of revenues, Austria 10% and Sweden 10%. COP is based in Koblenz, Germany.
COP generates about 75% of EBIT from its Ambulatory Information Systems segment (AIS), which sells software suites used to manage the daily operations of small medical offices. The product enables basic functions such as booking appointments, managing patient bills and payments, supporting finance processes and maintaining HR. Typically the system is used by the office receptionist and practice owner. COP augments sales of the basic suite with a wide range of add-on modules that allow the practice to improve efficiency and grow revenue yield. Examples include the ability to submit insurance claims online, manage regulatory filings, archive documents digitally and send prescriptions to pharmacies electronically. It also enables patients to schedule appointments through the web, view their medical records remotely, and setup online consultations with their physician.
In addition to the AIS segment, COP generates another 15% of its EBIT from the Pharmacies Computer Systems (PCS) unit, which the company acquired in 2011. The PCS product is similar to the AIS product but tailored to small pharmacies. COP offers PCS in Germany and recently entered Italy. Combined, the AIS and PCS units run under the Healthcare Providers System I segment (HPS I), which houses 90% of COP’s EBIT. Beyond that, COP runs a business labeled HPS II, which develops IT systems for hospitals and contributes 8% to EBIT. COP gets another 2% of its EBIT from a set of legacy products and services in the HCS segment, which includes an ISP business, a Pharma advertisement tool and a product used by doctors to coordinate research.
The HPS I segment (AIS + PCS units) benefits from 80% recurring revenues from fees paid for maintenance contracts and 20% from sales of initial license products (mostly through a perpetual licensing model but also as SaaS where the market desires this model). An installation of a new system typically includes 1-2 seats and costs the doctor office about E3,500. Maintenance fees are paid monthly and total about E1,100 per year. COP has over 200,000 paying customers in its HPS I segment and churn is 1-1.5% per year. The product allows for strong pricing power and the company has increased prices by over 3% per year in the past, while churn remained stable (there are historical cases when COP raised prices as much as 30% in a given year with no change in churn). The product is crucial to the day to day functioning of the customer’s business, and hence switching costs are very high, while the overall spend on the product by the customer is low.
The stable nature of demand for doctor and dentist services implies that COP benefits from a predictable and stable cash flow base. Consequently, COP’s revenues are very resilient to economic cycles. During 2008, the AIS business grew organically 9%, followed by 14% in 2009. Over the last 7 years, which includes the prolonged EU recession, the business compounded at 9% with the lowest year at over 6%. The entire company’s revenues in 2008 were up 3% as a result of low-single-digit declines in the hospital and HCS segment.
Over time COP leveraged acquisitions of smaller sub-scale competitors to position itself as the dominant player in almost all its markets (aside from the USA). In Germany, which makes up 50% of AIS, the Company holds half the market. The next player in the market is a JV between two small software providers (Doc Expert and MCS). The JV has 20% share and from the looks of the financials of its parents, they are not doing very well and are tiny companies. The balance 35% of the market is held by many local providers who have 1,000-2,000 seats or less. In other regions, COP holds 40% share in Austria, 40% in France, 42% share in Denmark, 44% share in Italy, 60% share in the Czech Republic, 67% in Norway, and 76% in Sweden. Only in the US is COP’s share low, as it holds just 3% of the market.
The competitive outlook for COP is improving as the business of doctors and dentists is growing in complexity - insurance reimbursement is becoming stricter and more tangled to handle, regulatory compliance and reporting is expanding, and doctors face increasing demands to offer online connectivity to various players in the value-chain (patients, payers, suppliers, pharmacies). Expanding complexity of the health practice’s operations translates to more demands out of its IT system. That in turn makes it harder for small software vendors to remain relevant both in terms of breath of offering and depth of solutions. Thus, growing complexity is pushing customers towards COP’s robust and full product offering. COP has been gaining share over all the other players in the market, which has contributed about 50bps per year of additional revenue growth.
In the past, the AIS business grew at an organic compound rate of 9% on a net basis (2007 – 2014). Given a 1.5% churn rate, gross CAGR was around 10.7%. Of this, price added ~3.5%, share gains 0.5% and the number of medical practitioners in the market around 0.25% (aging population requires more doctors). The balance of 6.5% growth came from up-selling ancillary modules into the installed base.
Growth from selling add on modules should continue to support high-single-digit revenue growth going forward. The installed base is far from saturated - in Germany only 30% of the base is online, and yields avg. revenues per customer of E1,100 / year. In Sweden, 100% of the base is online and an average customer generates E5,000 / year. E-Services modules such as booking appointments, remote access to medical files, online consultation and bill presentation are 30% penetrated, compared to only 3% in Germany (where COP started offering in 2012 and reached 1.5% at end of year, doubling to 3% in 2013). Other countries are somewhere in between this range. In addition, no single add-on module has outsized contribution to growth. COP’s catalog includes hundreds of modules, ranging from compliance, digital archiving, online e-Services, functionality for specialized practices (i.e. modules for OBGYN office requirements), remote access through mobile devices, medical management (i.e. Diabetes management), customer loyalty management and customer analytics. A typical module costs E350-500 euro to install and has E10-15/month of recurring maintenance attached. The attractiveness of this model is that instead of customizing the product to specific clients, COP sells a la carte at wonderful margins.
For the sake of conservatism relative to the past, we model about 6% CAGR from 2015 to 2017. Using incremental margins of 45% (in the past they achieved 67% incremental margins), we expect EBITDA in the AIS business to grow at 13.5% per year.
For the PCS segment we also expect around 6% growth as COP pushes price increases in recently acquired business that were undermanaged. We also see a clear path to material increase in these acquired companies’ margins (buying small competitors, raising prices and driving margins to COP’s corporate average has been a successful playbook for COP in the past) and net EBITDA growth of 9% per year. Combined we model EBITDA in the HPS I segment growing at a 13% CAGR from 2015 to 2017.
Beyond the HPS I business we expect the HPS II to remain flat and the HCS segment to decline around 2% (both assumptions are far worse than management expectations) and margins to stay around flat thanks to cost cutting. As a reminder, the HPSII and HCS together make up only 10% of EBITDA. Finally we expect COP to generate modest leverage on its corporate expense at a rate of about 10bps per year.
All together we expect COP to drive 13.5% yoy growth in EBITDA from 2015 to 2017 and reach EBITDA margins of just over 25% (vs. management goal of 30%). From there we see an acceleration of this compounding rate.
COP is primed to benefit from material structural change in its largest end market. The German government is currently running a test project to electronically connect all the constituents in the health-care systems (patients, insurance payers, hospitals, doctor practices, and pharmacies). This is a result of a dire need to reduce health-care cost inflation for a rapidly aging population. A connected health-care system with real-time communications and digital data archiving is crucial in order to reduce friction, improve compliance, drive utilization and enable outcome based performance and error minimization. Connecting the system is no longer a question of if but when.
COP is uniquely positioned to benefit from the transformation of the German market into a digital connected system. COP controls 50% of the market for doctor practices and has a unique lead in selling online connectivity modules to it client base. COP also has scale advantages in the cost of development and can undercut competitors in winning the 50% of the market it does not control as doctors evaluate vendors of choice for migration of their practices into an online world. Looking at other regions in Europe, there is clear evidence that COP’s share is positively correlated to the level of electronic integration of the health-care system where it operates. The Company’s highest share is in Norway and Sweden, where it holds 67% and 76% of the market. Both countries are the most advanced nations in Europe in terms of electronic connectivity (with Sweden at effectively 100%) and regarded as models of success to be followed by other nations.
The project to connect the German healthcare system together is led by Gematik, a government consortium. Gematik picked COP as part of a test project that started this year and will continue into next year. The project was to be a test of the ability to connect all participants in the system in a secure and cost effective manner. However a couple of weeks ago the head of Gematik announced that the test is progressing well and that a full national roll-out is now “beyond the point of no return”. The countrywide roll-out is scheduled to begin in 2016 and will be mandated by law for all healthcare providers (with patients receiving an electronic health-care card as well). The roll-out will be based on a competitive market model open to all IT providers who meet the technological standard of the network.
We believe the market has yet to digest the recent update on the successful progress of the Gematik project and the affirmation of a full national roll-out. Sell-Side analysts also have not incorporated the impact as it is “too far out” relative their typical short-term myopia. However, this is a major development since moving to a connected system should drive the average payment that COP makes from German customers from around E1,100 per year to E5,000 per year on an on-going basis. (and it will generate substantial revenues from installations on a one-time basis). We estimate that during a 5 year penetration period the implementation of a national rollout in Germany will accelerate COP’s growth from around 4% per year to 15% per year and drive over 20% of EBITDA growth from 2018 onwards just from the increase in maintenance fees (with upside to these estimate). As such it will serve as a major catalyst for the shares and will drive multiple re-rating.
Technological Displacement: the trend towards Cloud and SaaS could theoretically form an opportunity for new entrants to showcase their products to COP’s customers and switch them away. However the likelihood of this scenario appears low from several angles. SaaS/Cloud’s main value proposition relative to software on premise is lower costs and better scalability. Both factors lose appeal in the unique case of COP’s installed base: at a yearly cost of E1.1K for an average customer, the savings to be had are immaterial compared to the pain from switching for the doctor (i.e. the loss from three missed appointments in a year during the transition phase to a new product equals a whole year’s worth of spend on COP’s system). Further, given their small size, medical offices have little need to scale up using 3rd party hosted infrastructure. Finally, the Cloud’s main drawback resides around security and risk of breaching confidentiality. Both are enormous pain-points for medical providers. As a benchmark, in 2009 when a well-capitalized new entrant tried to penetrate COP’s market in Germany, the attempt failed completely as doctors didn’t want to take the risk and pain of switching systems (even with this competitor offering the product free of charge for a period of time). Finally we should mention that COP offers Cloud delivery for all its AIS products for customers willing to try this model out.
Management Mistakes in regards to HCS and US business: If management fails to turnaround the US business (which has suffered in the last couple of years due to sub-scale) and invests capital there, returns will suffer. The magnitude of such a move should not be material, though, given the small level of contribution of the business (close to zero EBITDA in 2014). Also, based on discussions with the CFO and comments mgmt made publicly it does not appear that they will be that irrational – they have openly acknowledged the need to diversify away the US if the turnaround plan fails (management is not giving itself years here – our sense is 12-18 months at most). The CEO’s large ownership is a major motivator for rational allocation of capital in this aspect. The case for HCS is similar. The company expects HCS to generate positive EBITDA for years to come (and in fact grow EBITDA), and management is open to divestment need be. Since we ascribe zero EBITDA and value to HCS, all we need is not to lose money or for the biz be sold away.
We see COP compounding earnings at 40% from 2015 to 2020 (with 25% compounding until 2017 followed by over 50% compounding from 2018 into 2020). We expect the market to recognize this outstanding value creation by the end of 2017 and ascribe a multiple of 18x P/E on 2018, yielding a stock price of E66/shr. This implies 3.5x return in just over 3 years or an IRR of around 50%.
|Subject||Typo and Clarification|
|Entry||11/03/2014 05:30 PM|
We expect EBITDA to compound at over 30% from 2018 to 2020 (and not 20% as stated in the post)
|Entry||11/05/2014 09:50 AM|
Thanks for the idea. As well, tip of the cap for your Pulsion Medical idea -- we did quite well.
I have been going through the financials here, and I wanted to make sure that I understand the one complexity -- surrounding the various depreciation/amortization. I am trying to drive at an unlevered FCF number and having one point I need to clarify.
From my reading, there are three different depreciation/amortization going on here:
(A) traditional physical PP&E
(B) self-developed software that is capitalized
(C) goodwill/intangible assets of acquisitions
In 2013 and 2014, what numbers are you using for each?
- I believe that in 2013 there was $41mm total across all 3 types. I further see that there was $29mm of "amortization without amortization of self-developed software." This is confusing because based on its usage in generating cash earnings, I believe this number also excludes traditional PP&E depreciation (the confusion seems to come in that sometimes the translation uses 'amortization' to refer to both what we would call 'depreciation and amortization' and sometimes only 'amortization.').
So, I get $29mm of type (C) and then types (A) + (B) = $12mm. Is this correct?
Further, what capex numbers are you using for traditional PP&E and for self-developed software? This seems a lot more clear and they are investing $6mm-ish on PP&E and $12mm-ish on self-developed software. But I was a bit puzzled at the mis-match between the combined $18mm in capex versus the $12mm of (A)+(B) D&A. It could just be simple explanation: capex is greater than D&A . . . but when I look back historically, it seems like the capex has been consistent enough over time that they should converge more than they do.
Sorry for the long-winded question.
|Entry||11/05/2014 10:11 AM|
I have found note 25 of annual report, which describes D&A in detail. But this strikes me as inconsistent with the cash earnings.
First, there was E41mm of total D&A and $14mm of "Software" (and note that this seems to be self-developed software as there is another line for "Purchased Software Licenses.") This would get us to E27mm of D&A less Amortization of Self-Dev Software.
But this is inconsistent with the E29mm they use.
But even more troubling is why would they add back the full E29mm. This includes at least E6mm of very real depreciation that requires maintenance capex. I understand adding back custy contract amortization and goodwill and brand amort, but I do not understand adding back facility/furniture/office equipment.
If i am reading this correctly, the cash earnings number is a fair amount overstated versus what I would consider a fair FCF-to-equity number.
|Subject||Re: Re: Re: Financials|
|Entry||11/05/2014 03:01 PM|
You somehow did answer my rambling questions.
I am getting about E18mm in normalized capex (E6mm in PPE; E12mm in self-dev software), which is similar to your 3% of revenue (E15mm) and 1:2 ratio. And so I now think their 2013 add-back of E29mm is only slightly aggressive (by about E5mm). I am getting unlevered FCF in 2014 of about E60mm and this is at 18% EBITDA margins.
Thanks for the idea.
|Subject||Re: Re: Re: Re: Financials|
|Entry||11/05/2014 03:54 PM|
Right. Also please notice that the 2014 margin is depressed due to a couple of factors (no margin on the German project, the ERP spend at the OpEx level and the consolidation of the acquired businesses in PCS in Italy prior to pushing through price increases). In 2015 we will see these headwinds abate - the sell-side is modelling 260bps of ebitda margin expansion (we are more modest at 190bps). Combine that with modest revenue growth and reinvestment of FCF at a 7.5% yield and you will get to around 12x fwd P/E (we are at 12.3x as the stock moved up a bit since we posted).
|Subject||Re: Re: Re: Re: Re: Financials|
|Entry||11/05/2014 03:56 PM|
also notice interest expense run-rate is lower in 2015 (CFO sees E10mm run-rate; we assume E12mm)
|Subject||Calculating free cash flow|
|Entry||11/07/2014 12:12 PM|
Thanks for an interesting idea.
Could you explain your basis for capex & intangibles being 3% of sales? I show them spending Eur 104mm in total for the last 5 years on all forms of capex (counting just PPE & capitalized software; sourced from the table they provide each year that breaks down all investment spending, e.g. p. 36 of '13 Annual). That comes to 5.4% of sales over that period. Admittedly some one-time items in there, but every year seems to have some one-time items.
What is your assumed cash tax rate? Their effective tax rate from note 28 in their Annual seems not to dip below 37%, despite the 29.65% German rate. That's possibly distorted by the large goodwill charges. To acccount for that, I just summed EBITDA, then subtracted net interest expense and capital spending. That's my denominator. Then I used "cash income tax paid" from the cash flow statement as my numerator. Over the last 4 years, that averages 32.5%.
If we assume just a 4.5% capex/sales rate on 2015 sales of Eur 550mm, and a 32.5% cash tax rate (applied to 130mm of EBITDA, net of interest and capex), I get a levered fcf of Eur 64mm, about a 7% levered yield, and a 5.6% unlevered yield. That is not too different from yours, but I'd be interested to hear your reaction to it.
Also, I am curious to understand - to get consolidated margin to 25% in 2016, what margin do you assume in the HPS I segment?
|Subject||Re: Calculating free cash flow|
|Entry||11/07/2014 01:49 PM|
I'd like to try and assist by answering on two levels – at the big picture level, which i see as the more important, and on a detailed level in order to address your question directly.
As for the HPS I segment we model an ebitda margin of 33.5% in 2016.
I hope this helps.
|Subject||Re: ARPU accretion|
|Entry||09/21/2015 06:17 AM|
- The ARPU refers to users and COP has about 100k of those in Germany and 250k overall in AIS
- The confidence around the increase in ARPU in founded on several layers: a. speaking with GPs in Germany and hearing that demand is real, doctors are waiting for modules to come out and expect them to help save money, improve efficiency and medical outcomes, b. understanding that GPs compete for customers and that in a connected environment those who wont offer the convenience of real-time, remote and paperless medicine will eventually lose out as customers will expect convenience and quicker service (will take time but direction is clear), C. seeing that the German government is motivated to push more services on the connected system and it is working already i believe on the next phase of the e-Health bill, and D. benchmarks of other countries point towards the direction.
- the competitor was ICW. I see the aethna situation as quite different due to the fragmentation of the US market, the importance of doctor groups in the US, the starting market share structure of the players (and what it means in terms of ability to under price new entrants out of the market quickely) and how SW is distributed (IT distributors in Europe are more important)