Description

P&I is a leader in European HR/payroll software.  It’s a gem of a business that has remained curiously cheap despite an outlook that went from good to great this past year, and despite a vast improvement in the trading volume of the shares.   I’m guessing that obscurity is to blame, as P&I was quasi-private until very recently and thus off the market’s radar.  (Note: market cap above is in USD)

At 10X FCF for the year ending March ’08, the valuation doesn’t make any sense.  This is a software business with >40% recurring revenues and <3% customer churn, which grows around 10% organically by simply taking share in a fragmented, yet difficult to enter market.  This growth requires minimal capex, so almost all the cash flow can be used for dividends or buybacks.  So 10% growth + 10% yield equals a 20% return, which alone would make this a cheap stock.

But now it could be better than that.  P&I announced deals with ADP and a European company called LogicaCMG this past March.  Long story short, ADP and Logica will be using P&I software for some of their HR outsourcing services, which means that P&I will have two big partners driving lucrative fee revenue with no incremental cost.  The upside here is uncertain, but it wouldn’t take any miracles for FCF to go from 1.75/share today to maybe 4.00/share over the next four years.  Cash would pile up along the way (from the 3.50/share I expect at fiscal year end).

Nice market to be in:

P&I was founded four decades ago in Germany and still gets around 80% of sales from this market, though foreign sales are growing faster.  Their core product does the payroll calculations for about 11 countries in Europe, and a couple of add-on products basically put HR tasks like attendance, scheduling shift workers, and expense approval online and tie it all into a single database.  They serve about 11% of the German market, as measured by the number of paychecks processed by its software.  Its focus is on midsized companies, defined as having 500 to 5,000 employees. 

SAP is the only formidable competitor in this segment.  SAP’s software, however, is often too complex and expensive for midsized companies.  P&I has the “best-of-breed” product, whereas SAP pitches the simplicity of an entire integrated ERP suite (they don’t sell payroll separately, though SAP customers often use HR software from other vendors like P&I).  Also, SAP’s R&D money is better spent on software that can be sold around the world – supply chain management, etc. – whereas payroll requires a huge amount of country-specific development.  Oracle is a similar story, and seems to be targeting only large companies.  ADP, meanwhile, has been burdened with a product that’s increasingly obsolete.

The rest of the middle-market is served by ~15 tiny players which lack the scale to keep on adding more HR functions or geographic coverage, and this disadvantage grows wider over time.  I figure this pool of potential customers is big enough to maintain P&I’s historic growth for a long time – maybe 10 years or so.  The smaller guys don’t always disappear quickly since customers are typically loathe to switch payroll vendors, but that’s fine.  Foreign sales have been rising faster, and outsourcing deals (ADP, etc.) could speed things up too.

You might expect a bit more competition in Europe’s largest economy (28% of the total).  But there’s a barrier to entry here, namely red tape.  Germany’s Kafkaesque labor laws are among the most complex in the world and make payroll a real challenge.  A Spanish company called Meta4 once hired away a P&I manager to develop a solution for the German market, but the effort failed.  PeopleSoft’s first attempt in 1997 was a high-profile disaster.  (They got it right eventually, however, which explains how Oracle got in.)  ADP likewise paid-up to acquire their way in (1997), but the solution they got is fast becoming obsolete.  Basically, there isn’t a big threat of new competition.

Unique product with growing demand:

The other thing they have going for them is country coverage.  P&I handles 11 countries.  SAP probably does more.  I’m pretty sure Oracle doesn’t cover more than 5 or 6.  There are a couple of vendors with 3, and that’s about it.   Having an integrated system saves money for multinational customers since maintaining different systems for each country is expensive, and multiple databases increase the risk of errors and delays in reporting.

Demand from multinational customers is growing and it’s surprising to me that the list of vendors is so small.  Microsoft tried to create a pan-European payroll product in 1994 and eventually gave up.  I’m guessing it’s just not an easy thing to develop.

Why the market doesn’t seem to care:

Size is one reason.  SAP and Oracle together have about a 30% share of the European market and everyone else is tiny in comparison.  So analysts give those two a disproportionate share of the attention.

History is another.  The Carlyle Group bought a 2/3 stake in ’04, back when earnings and market cap were a lot smaller than they are today.  Carlyle’s stake obviously tied up a lot of liquidity, and the remaining float was tightly held by a bunch of European value investors (the ones I know are still invested, btw).  With no real trading volume and no need for outside capital, brokers lost interest and dropped coverage.  P&I dropped off the radar screen.

Liquidity finally returned in August when Carlyle sold its stake at EUR 21/share.  I believe the real trading volume is a multiple of the apparent USD 400K/day average.  Carlyle’s shares were bought by 34 investors in roughly equal-sized lots, so with few retail-sized holders, sellers have turned to the bookrunner on the August sale (Berenberg Bank) to help them find buyers.  So try them if liquidity is a concern.

Now why would Carlyle give us a bargain?  I’m guessing the answer is they had bigger fish to fry – like their new EUR 5.8 billion Europe LBO fund.  In contrast, the P&I stake came to roughly 110 million.  Carlyle had been winding down the fund in which they owned it.  This fund – Carlyle Europe Venture Partners – had already returned somewhere north of a 15% gross IRR since January 2000, and the P&I stake was less than 10% of the total proceeds.  I’m guessing that tying up its European team on this particular sale was less attractive than working on new, much larger investments.

The BPO upside:

Roughly 15% of corporate HR/payroll in Europe is now handled via business process outsourcing, or BPO.  An HR BPO typically involves a client company handing off a whole slew of tasks to a provider and reducing their own staff accordingly.  Companies do this mainly to save costs, but frequently also to spread out costs that might otherwise hit them upfront (like a new IT system) – pricing is pay-as-you-go based on the # of employees covered, for a minimum period of say 4-10 years.

Most BPO activity has been with large companies, but it’s picking up in the middle market.  In Europe, ADP is the only global provider in the middle market.  The others in this space are European IT outsourcers/ systems integrators like Logica and Atos Origin.

ADP now selling their software:

ADP has been selling a German payroll software system called Paisy, which dates back to the 1970’s.  Paisy has become a challenge to maintain and can’t do a lot of the stuff that newer web-based HR applications do - common issues with old software written in COBOL. 

BPO deals are ADP’s fastest growing line of business worldwide and Paisy was holding them back in Germany, so they decided to find a partner for the software.  For midsized companies, BPO deals usually cover a number of HR tasks (payroll, benefits, recruiting, etc.) and for ADP the revenues are about 2.5 to 3X what they charge for payroll processing alone.  Moreover, payroll is usually the first thing to be outsourced and gets your foot in the door to sell other services.  So even though payroll is ADP’s bread-and-butter it made sense for them to partner here.

So the Paisy salesforce is now selling P&I-based solutions and the first installations will go live in January.  P&I gets license and maintenance fees on each deal.  ADP’s salesforce is about the same size as P&I’s (~20).  They do about 4.5M paychecks in Germany.  I figure P&I does 3.5-4M in Germany (out of 4.5M total).  I’d imagine ADP gets a discount off the typical price, but unless it was a really steep discount it seems like this new revenue stream will be significant.  (for what it’s worth, I estimate the average license price has been around EUR 30 per paycheck, plus an 18% maintenance fee)

One more piece of data: ADP offers a multinational payroll BPO service called Globalview which uses SAP software.  It’s not directly comparable since it’s targeted at large companies (~10K employees) and covers more than just Europe, but still the growth is encouraging.  Globalview launched mid-2004 and had 730K paychecks at mid-2007.   The growth seems to have ramped from roughly 25K paychecks per quarter in ’04 to around 75K today.

There’s another part to the ADP deal.  Based on conversations with managers at ADP and SAP, it appears that ADP talked to both SAP and P&I about partnering.  They chose P&I for technical reasons – only their software would make it practical to convert their current BPO customers away from Paisy.  Apparently this isn’t so easy to do, and payroll is something you just can’t screw up. 

There’s no firm commitment yet, but considering ADP’s thought process I’d say the odds are decent at least some customers will be switched over.  For converted customers, ADP would pay based on a “usage” model – a per paycheck per month fee, similar to what its BPO customers pay.  I don’t know what that is, but note that P&I says that in four years it would like to have EUR 15-20M coming from about 1M paychecks, which would imply EUR 1.25-1.66 per check per month.  ADP, ULTI and others are charging around $1.50-3.50 for basic payroll processing in the U.S., so it seems reasonable.  BPO is 2.5 out of their 4.5M paychecks, though ~300K of these are using SAP.  So do the math on 2.2M paychecks a month and this could be a huge piece of business.

LogicaCMG

Logica is European IT outsourcer/ systems integrator, sort of like Accenture.  Payroll is just one of their businesses, but they seem anxious to grow it.   Like ADP, they focus mainly on midsized customers.  Logica ditched its in-house software and is now relying on P&I for their customers in the Netherlands.  I estimate they paid about EUR 6M for these licenses, split between the December and March quarters.  As with ADP, Logica will pay license and maintenance fees on new BPO customers.

I estimate their Netherlands payroll business is doing about EUR 40M annually, but I don’t know the growth rate.  Logica will dedicate about 20 salespeople, which seems like a high number relative to current sales, so I’m wondering if they’ll also be targeting other countries where Logica operates.  So I’ll confess to being a little fuzzy on the upside here.

P&I says it hopes to extend this partnership to cover Logica's UK and Scandinavia customers.  Admittedly, they don’t yet have software for these markets, but it seems a matter of time (the UK is 25% of the European market).  This potential upside could be significant, as Logica does far more revenue in the UK than it does in the Netherlands.

The bottom line:

CEO Vaslos Triadis gave a presentation this past July in which he outlined a goal of 95-100M sales in about four years (FY12), up from about 60M today.  This would come from 30M in maintenance, 30M in consulting, 20M in traditional license sales and 15-20M in usage fees from BPOs.  He didn’t mention margins, but I estimate this implies north of four Euros per share.

Normally I write-off long-term outlooks like this as wishful thinking or BS, but this one actually seems do-able to me.  I can “get there” with the following assumptions:

• Core business keeps historic growth rate and current cost structure
• ADP converts 1M (out of 2.2M) BPO paychecks to P&I over the next four years, at my low-end price estimate of EUR 1.25
• ADP’s BPO business in Germany ramps up to 10% growth in a couple years (vs ~15% market growth)
• Logica’s BPO business in the Netherlands grows at 5%
• ADP and Logica pay license fees equal to half the usual price (this is arbitrary – I used this as the plug to reach the low side of 95-100M)

Final thoughts:

The only thing that bugs me about this idea is that management doesn’t own much stock.  So now that the company is without a controlling shareholder for the first time, I have to wonder if capital allocation could get sloppy.

As far as I can tell, however, the profit growth over the past few years came from more than just Carlyle cracking the whip.  I think Triadis (the CEO) deserves credit too, as does the normal growth of a fundamentally good business.  Triadis continues to work pretty hard from what I see.  Nobody from management is on the Board, whose approval is required for the big decisions.  Plus the investor base is a reasonably close-knit community.  So I expect the company will be run sensibly.

Catalyst