Description
I am recommending a short of Allscripts Healthcare Solutions (MDRX, $13.60). This is the stock I used for my application to VIC, and while I’m a bit less enthusiastic about the entry point now, I still think the risk-reward is good. The numbers in the report are based on a $15.80 stock price.
MDRX has almost tripled in the last two years on the growth in its electronic health record (EHR) product. The stock now trades at 87x untaxed TTM EPS and 34x the consensus untaxed 2006 EPS estimate, which I think the company will miss. Management’s increasingly aggressive accounting in the past several quarters suggests the company is straining to meet analysts’ expectations. Likely disruption from GE’s proposed acquisition of IDX announced Sept 29 and possible slowing growth in MDRX’s market segment are the catalysts for the short. MDRX has derived over 70% of its EHR business from the IDX customer base.
MDRX reports three lines of business:
1) Prepackaged medications: this is a low margin, declining legacy business. 37% of 2005 sales.
2) Software and related services: 100% of the company’s recent growth is from this area. This is where the company reports its EHR software and hardware sales, including TouchWorks. 55% of 2005 sales. Analysts expect growth here to drive over 90% of the company’s growth in 2006.
3) Information services: clinical education programs for physicians and patients, paid for by drug companies. This business has struggled. 8% of 2005 sales.
***Aggressive accounting***
The balance sheet and cash flow statement suggest MDRX has been working hard to make numbers, particularly in 3Q05. Management nonetheless on its 3Q05 conference call maintained its guidance for 2005 bookings growth of 40% and revenues of $120m. Bookings growth YTD in 3Q05 was 35%, so bookings growth needs to accelerate in 4Q.
DSO has been climbing rapidly. In 3Q05 DSO stood at 81, up 7 days y/y. 2Q05 DSO were 74, up 6 days y/y. 1Q DSO were 83, up 14 days y/y.
Management seems to be pulling revenue into current periods at the expense of future ones. Software and services bookings were up 35% y/y for the first nine months of 2005 while revenues grew 57%.
From the 3Q05 10Q regarding software and services revenue: “The increase in both periods [3Q05 and the nine months ended September 2005] is due to an increase in our installed customer base, an increase in revenue from existing customers converting their subscriptions into licenses, and existing customers purchasing additional licenses to resell our software to local and regional small physician groups and to implement our software at additional internal sites.”
1. Management is moving customers from subscriptions to licenses, which is having the effect of pulling revenue into current periods at the expense of future ones.
2. Management does not say if the customers that purchased the licenses for resale have resold or installed those licenses. In light of the rapid rise in DSOs this increase in license sales for resale could indicate the company is stuffing the channel at the end of the quarter.
3Q05 software and services revenue were also lower quality, with a higher proportion of revenues coming from hardware. This poorer mix pressured GM%.
Unbilled receivables are skyrocketing. MDRX uses percentage of completion accounting for a portion of its license revenue and has as far back as I looked (I looked back to 2001; note that the company changed from the output to the input method in January 2003). So, unbilled receivables are not unexpected. Their growth relative to revenue is. Unbilled receivables climbed 135% y/y in 3Q05. The unbilled receivables balance has run at over 20% of software revenues in 2005 (Q1=23%, Q2=23%, Q3=24%) compared to an average of 13% in 2004 (Q1 =14%, Q2=6%, Q3=15%, Q4=16%). It seems likely that MDRX became more aggressive on its revenue recognition assumptions in 2005.
3Q05 packaged medication revenue exceeded most analysts estimates. In the 10Q management notes that there was a higher mix of wholesale revenue during the quarter. Perhaps management was seeking a channel for additional revenues.
Prepaid expenses and other current assets were up 52% y/y in 3Q05 while revenues only grew 19%. Meanwhile accounts payable, accrued liabilities and accrued compensation in aggregate were down 4% y/y. Allscripts is capitalizing assets on its balance sheet instead of expensing them at a rapid clip while simultaneously under accruing for expenses, thereby understating operating expenses on the income statement.
Deferred revenue was up only 10% y/y in 3Q05 while the revenue that the deferred relates to (software and information services) was up 38% y/y. MDRX breaks out deferred revenue for just the software and related services and the story is the same. Deferred revenue in software and related services was up 23% against 57% growth in that category’s revenue. Part of this divergence is likely explained by management’s moving of some customers from subscriptions to licenses. In any case, this shift suggests management is pulling forward revenue at the expense of future periods.
Operating expenses dropped in 3Q05 versus the prior quarter, helped by a reduction of bad debt expense and higher capitalized software according to the 10Q. Well bad debt expense was not only down, but management appeared to have reversed some of the AR allowance into earnings in 3Q05 (bad debt expense was negative $79k on the cash flow statement). If bad debt expense had been at the same average level of the prior four quarters, 3Q05 earnings would have been reduced by $0.01. MDRX coincidentally beat by one penny in 3Q05, reporting 7c of EPS.
The allowance for doubtful accounts dropped monotonically from 12.3% of gross accounts receivable at 4Q04 to 8.9% at 3Q05.
Capitalized software in 3Q05 rose by $300k over the prior four quarter average. Had cap sw stayed at the four quarter average, earnings would have been another penny (really $0.007) less in 3Q05.
All of the above adds up to 3Q05 cash from operations being negative $408k and free cash flow (CFO less capital expenditures less capitalized software) of negative $2M against a reported $2.9m in net income. This is a dramatic divergence versus history. In 2004 MDRX free cash flow averaged 2.3x reported net income. In 1Q05 that ratio dropped to 1.4x, in 2Q05 to 1.0x. Now free cash flow is negative and reported earnings are at their highest level ever.
Management delivered the cash from operations number in an interesting way on the conference call: “…such an increase was reflective of the $2m in cash generated from operations, offset by our $1.4M semi annual interest payment on our convertible debt, and $1m payment related to our semi annual insurance premiums.” Unsurprisingly, at least one analyst wrote in his 3Q05 note that the company generated $2m in cash from ops. Really that figure was negative $408k, but the company made listeners work to derive that figure.
Note that MDRX announced April 2004 that it would change auditors to Grant Thorton from KPMG.
***Market***
The Medical Group Management Association (MGMA) and the University of Minnesota School of Public Health surveyed in early 2005 over 3,300 medical group practices to assess EHR adoption.
ftp://ftp.mgma.com/pub/WEBMISC/Center_for_research/EHR_tables_and_graphs_final.pdf
MDRX derives most of its EHR revenues from physician practices with more than 21 doctors.
The data suggests a slowdown in future implementation growth in MDRX key market. They found that for practices with over 21 physicians, 11.4% had already adopted, 28.5% had an implementation in progress, 15.7% planned an implementation in the next 12 months, 24.2% planned an implementation in the next 13-24 months and 20.2% had no plans to implement in the next 24 months. Note the drop in implementations in progress (29%) to planned implementations in the next 12 months (16%) Also, within a few months 70% (11%+29%+16% / 11%+29%+16%+20%) of the MDRX market that will implement within the next year will have already begun implementation. I don’t normally place a lot of weight on market research, but I find this data interesting in light of the increasingly aggressive accounting that I believe management has used to mask slower than expected growth.
Competition. In practices with greater than 25 physicians, MDRX competes primarily with Epic (private), GE’s Centricity, NextGen (QSII) and Cerner according to KLAS rankings. KLAS is a research firm that focuses on the healthcare IT market. According to KLAS the Allscripts product for practices under 25 physicians does not meet KLAS’ minimum confidence level.
Longer term, healthcare IT probably consolidates around the larger players like GE that can provide an integrated system, similar to what we’ve seen in corporate IT. This could make it difficult for single solution providers like MDRX. Indeed, MDRX has found it difficult to win business outside the IDX customer base, probably because MDRX is the only vendor among top competitors that doesn’t offer an integrated EHR and practice management system. MDRX’s EHR does integrate tightly with IDX’s practice management solution. I discuss this some more in the “IDX Relationship” section below.
Vendors also have to compete with the U.S. government’s free system, OpenVista, which the VA hospitals use. Vista seems to be reasonably well liked by physicians. OpenVista likely will pressure pricing, even if it doesn’t win a large part of the market.
“The Department of Veteran Affairs’ (VA) open source VistA Enterprise Healthcare Information System has been 20+ years in development and backed by a several billion dollar investment by the VA. VistA been successfully implemented at over 1,300 sites of care including 158 medical centers and approximately 850 related clinics, long term care, domiciliary and home healthcare programs. Approximately 60% of all nurses and over 70% of physicians get exposure to the VistA system during their training.” From: http://www.medrecinst.com/conferences/tepr/2005/program/tracksDetailSession.asp?FUNC_ID=222&PARENT_ID=67
Here’s an informative article that discusses some of the EHR alternatives available and a hospital that is implementing OpenVista: http://www.healthcare-informatics.com/issues/2005/05_05/cover.htm
Also note in the article that GE has plans to make commercially available an EHR system that Intermountain Healthcare developed.
***IDX Relationship***
In January 2001 MDRX bought ChannelHealth from IDXC. ChannelHealth products formed the foundation for MDRX’s EHR, TouchWorks. Concurrent with the acquisition MDRX entered into a 10 year strategic alliance with IDX where “Allscripts and IDX agreed to coordinate product development and align [their] respective marketing processes. Under this agreement, IDX granted [MDRX] the exclusive right to market, sell, license and distribute ambulatory point-of-care and clinical application products to IDX customers.”
The MDRX EHR is tightly integrated with IDX’s practice management software. MDRX has derived over 70% of its EHR revenues from the IDXC customer base (80% per recent Lehman initiation report; about 70% of software and services bookings were from the IDX customer base in the nine months ended September 2005). MDRX derived all of its growth from its EHR product in the last year; analysts expect over 90% of 2006 growth to come from its EHR product.
MDRX has been relatively unsuccessful outside of the IDX customer base. The market opportunity is 5x as large outside of the IDX customer base as in it, yet MDRX has derived less then 30% of its EHR business outside of the IDX customer base. This may be because MDRX is not permitted to sell a practice management product under its contract with IDX. Top competitor products NextGen (QSII) and Epic (private company) have an integrated practice management and EHR solution. Clearly the market is more competitive outside the protected IDX customer base.
Note that GE has a competing EHR product called Centricity. GE also partnered with Intermountain to commercialize Intermountain’s EHR product (see article link above). Per the IDX – MDRX contract GE is not permitted to sell an EHR product into the IDX customer base. Considering GE has its own EHR product, GE may be less inclined than IDX was to support MDRX. More on that below.
September 29, 2005 GE announced its intent to acquire IDXC. MDRX management is adamant that its relationship with IDX will remain strong and that GE will honor the terms of the MDRX and IDX contract, one of which is that GE can not sell an EHR into the IDX installed base. Perhaps, but it seems likely that while GE will honor the terms of the contract, it could do so in a way that will be less beneficial to MDRX. Furthermore, it seems likely that at a minimum the announcement of the acquisition will disrupt MDRX’s business with IDX customers.
For example, an excerpt from the contact is “The parties will be required to compensate and incent their sales forces to sell New Allscripts' products to IDX customers and prospects.” It seems likely that GE could curtail incentives for the IDX sales force to sell MDRX products, but still be in compliance with the contract.
Another excerpt: “Either party may terminate the marketing restrictions to which it is subject upon the occurrence of a material adverse change in the business, properties, results of operations or condition (financial or otherwise) of the other party (other than changes that are the result of economic factors affecting the economy as a whole or changes that are the result of factors generally affecting the specific industry or markets in which the party competes). The determination of whether a change is a material adverse change giving rise to a termination right is subject to arbitration.”
GE has the option to terminate the contract if MDRX results of operations deteriorate. GE may have the ability to curtail incentives for the IDX sales force to sell the MDRX product thereby directly influencing MDRX results. Perhaps GE could influence the early termination of the contract.
Another excerpt: “If during the term of the alliance, a change of control occurs with respect to IDX whereby a direct competitor of New Allscripts will control IDX, New Allscripts will, for the remainder of the term of the alliance, receive a revenue share on sales of all IDX "Patient Channel" products equal to the percentage of revenue share to which New Allscripts was entitled at the time of the change of control and, at the end of the term of the alliance, New Allscripts will receive the source code for all IDX "Patient Channel" products.” According to a recent Piper Jaffray note, MDRX hasn’t ever received compensation for the “Patient Channel” product and IDX no longer markets the “Patient Channel” product.
Note that IDX missed 3Q revenue estimates because of GE acquisition disruption. FBR wrote that IDX missed its revenue estimate by 2% and its systems sales estimate by 16%. Several customers postponed signing system contracts in 3Q05. IDX expects similar delays in 4Q05.
***Analyst 2006 estimates***
Analyst estimates are for about 40% growth in 2006 software and services revenue. Analysts’ project the growth in software and services will account for over 90% of the growth in total MDRX revenue in 2006. Analysts seem to be extrapolating management’s 2005 bookings growth guidance of 40%.
Bookings were up 35% nine months YTD, so bookings are back end loaded in 2005 and growth will have to accelerate in 4Q05 for management to hit its target. This may be tough given the IDX acquisition announcement.
Note that projected 2005 revenue growth in software and services is about 50%, 10% points higher than management’s bookings growth target and 15% points higher than Sep 05 YTD bookings growth.
Aggressive revenue recognition in 2005 likely inflated revenues at the cost of 2006 growth. If bookings grow by 40% in 2006 then revenue growth will likely be less than analyst’s estimate of 40% given the pull forward of revenue into 2005 and assuming the same relative backlog growth. Disruption from the GE acquisition of IDX may pressure revenue. Market dynamics detailed in the MGMA survey make 40% growth seem very aggressive as does increasing competition from competitors such as Open Vista.
Software and services growth of 30% next year to me seems like the best case in light of recent trends, competition, etc. Removing 10% points of software and services growth reduces EPS by about $0.05 assuming a 65% GM% and incremental SG&A of 30% of revenue (i.e. a 35% incremental operating margin). MDRX’s 2005 YTD incremental operating margin was 38%.
The 2006 consensus analyst EPS estimate of $0.47 does not include the impact of stock option expense, which will be about $0.10 according to management, if they don’t accelerate option vesting, which they might do.
***IDX overhang***
IDX owns about 7M MDRX shares or about 17% of MDRX’s outstanding shares. IDX has been selling. GE noted that its proposed acquisition price of IDX was $1.2B, net of cash and marketable securities. Most of IDX’s cash and marketable securities are MDRX shares. So the MDRX shares aren’t part of the purchase price. Perhaps GE doesn’t plan on holding MDRX shares for long.
***Shelf filing***
There have been a slew of positive analyst initiations of coverage recently. I was wondering why until the company announced a $200m shelf filing. I’m really not sure what the company will use it for, perhaps an acquisition, which management could use to bury its past accounting gimmickry. Another possibility is that MDRX will use the proceeds to buy its shares from GE. This would remove the share overhang, but probably be quite negative for the stock as it would remove any hope that GE might support MDRX selling into the IDX customer base.
***Valuation and price target ($10 at best)***
MDRX is expensive, especially considering that reported earnings seem overstated and 2006 estimates are probably too high. When I wrote this the stock had closed at 15.80, so ratios reflect this price.
The stock trades at 55x the fully taxed consensus EPS estimate for 2006 compared to 41x for MDRX’s closest comp, QSII. QSII estimates are fully taxed. The consensus untaxed 2006 EPS estimate for MDRX is $0.47. I use a 39% tax rate, the same as QSII, to arrive at a $0.29 fully taxed estimate for MDRX. MDRX could begin to pay taxes in 2008.
MDRX trades at 113x the taxed 2005 consensus EPS estimate versus 51x for QSII. MDRX trades at 69x the consensus 2005 estimate (untaxed).
MDRX generated $5M in TTM free cash flow. MDRX’s market cap is $700m, so it is trading at 140x untaxed TTM free cash flow. QSII trades at 56x TTM free cash flow.
$10 target price. It seems reasonable that if MDRX begins to miss estimates and its growth is less than expected that it will trade at best with a multiple similar to QSII instead of at its current premium. Using QSII’s 40x or so multiple on a fully taxed 2006 estimate for MDRX of about $0.26 (consensus of $0.47 less $0.05 by my estimate and applying a 39% tax rate) yields a $10 price target for MDRX. If MDRX begins to miss earnings I would be surprised if the stock didn’t go even lower than $10.
Implementation of FAS 123R will reduce 2006 EPS by about $0.10 per share, or about $4-5M according to management. Management says it may accelerate the vesting of certain options to reduce compensation expense in 4Q05. This is not included in the earnings multiples above.
***Risks to short***
The EHR sector is strong
Management postpones day of reckoning with a large acquisition
Catalyst
Business disruption from IDX acquisition announcement.
Slowing market growth in MDRX segment (>21 physicians.
Company runs out of room to continue aggressive accounting.