Thursday, August 12, 2010

Microsoft – an accounting geek’s summarizes his purpose and lessons

This is not a post with investment conclusions.  It is more a little about my own voyage.  If you want the stock posts read the last two posts in order.

My purpose

The purpose of this blog (as stated in my profile) is:

to explore investment ideas. [This blog] differs from most investing blogs as I have no intention of talking my own book. However I will explore what is wrong with my investments and that means I need to talk about positives and negatives. I welcome criticism. I most value criticism which demonstrates that some of my ideas are wrong.

I wrote the post to try and explore what was wrong with Microsoft stock (which – for the record I own in small quantity).  I was slightly annoyed with the title put on the post when it was reprinted by Business Insider because it misrepresented my purpose.  That said – there are real negatives with the Microsoft story – and it is worth exploring them. 

Microsoft is (way) outside my usual beat.  I am an accounting geek.  I like to read accounts to find out where things have been clipped, where the statements are aggressive, where the bodies are buried. 

Microsoft is an unusual company because there are no bodies buried in the accounts.  The business is staggeringly profitable – and those profits are entirely visible in cash (and equivalents).  Usually I add value (if I manage to add it at all) by understanding how the business makes its money and how that is reflected in their accounts.  For me reading Microsoft’s accounts beyond a summary and cash flow statement is pointless.

With a complex financial institution (my usual beat) understanding the accounts can be a difficult job.  Schiff’s Insurance Observer still refers to “the blonde with more curves than Conseco’s financial statements”.  Microsoft’s statements are as curvy as Bill Gates.  This is a fabulous business – arguably the best business in the history of capitalism. 

So the question:

The outcome for Microsoft stock will be driven by whether – in ten years – we wake up and find that Microsoft has retained dominance in the way corporations interact with their data and a decent position in the relationship of retail consumers and digital data. 

If Microsoft is dominant in that they will be making much more money in a decade because there will be much more information – and more computers - and the cost-per-computer is not outrageous even if the total revenue line is. 

If Microsoft loses its dominance then the stock could have another weak decade (notwithstanding the fact that it starts at a low price relative to current cash flow).  The stock price is so low relative to indisputable current cash flows that Microsoft would really have to stuff it up from here to make the stock a worse investment than say the current ten year bond.  But this is technology and it is subject to rapid change.

In my semi-geeky way I tried to run through the threats to understand them.  The threats are obvious: virtualization (and hence the end of the hardware-dependent computer), virtualization (and hence the end of the client-server model Microsoft rode to glory), Linux (and free software generally), the resurgence of the Macintosh platform, and (possibly most importantly and certainly most obviously) the increase in (mostly non-Microsoft) mobile devices (which are increasingly how the young and the affluent consumer relates to music and the internet and to data generally). 

So far – with respect to most of those threats – Microsoft has not been victorious – but has left itself a place at the table.  Even Zune has a place at the table even if it is up the end... 

Microsoft has not had an unambiguous winner in a decade – but then they are so entrenched that even without a winner they remain a strong company.  They produced one (very) weak flagship product (Vista).  But they have a (very) good product now (Windows 7) – it is however sufficiently late to cede a lot of ground to Apple.  Still you could imagine a world with an absolutely brilliant Windows 10 (!) phone which integrates near perfectly with business data (being a thin client for the work computer) but also has a lossless music player, high quality music jacks, integrates with Facebook (or its successor) perfectly – oh – and it is as cool as any future iPhone.  It just has not happened yet – but a super-hot product would add 100 billion to the cumulative cash flow Microsoft might produce and make the stock a fabulous investment.

On virtualization the “clients” in my experience are mostly windows (or often virtualized windows servers sitting on linux boxes).  Microsoft has a position.  But the virtualization platforms are Citrix, VMWare and others.  The underlying operating system is linux.  Indeed Wikipedia gives a huge list of programs and Windows is by no means dominant.  Linux (outside the server market) remains a product for geeks and Baltic grandmothers (but is getting better rapidly).  Apple look like the winner in mobile devices – but it is early in that game still and I think Linux in its Android form is likely to wind up the winner just because Apple extracts too much margin and the Android product is likely to be just as good (except for the killer iPod app).  That said the Windows 7 phone reviews pretty well.  Ten years ago Palm was the likely winner in mobile devices.  These things change!

A comment on the comments

The comments were very high quality. Sure I found someone who believed that Linux must run on top of (pirated) Windows XP.  He simply could not believe that you could run a computer without Windows.  [That level of ignorance is itself interesting because it tells you just how entrenched Microsoft is…]  But I also found down-in-the-trenches people who could not imagine that any business could go all thin client.  [I observed however that quite sophisticated businesses have gone that way already – and it is far superior to the client-server arrangement for security and disaster recovery – but not only is it superior – it is cheaper.]   I found religious open-source zealots but also entirely practical people who thought you could not get fired for buying Microsoft.  (I noted that in the server business for many applications you would be fired for buying Microsoft…) 

I met open-source geeks who thought that Ubuntu – at its current rate of improvement – would be as user friendly as a Mac within 12 months.  That is aggressive. It took me considerable effort to permanently mount the windows-share hard drives on my new computer – enough difficulty that I can be sure Ubuntu is not ready for prime time.  But given the improvement between my first linux machine [a Mandriva 2006 release] and the current Ubuntu release I think Ubuntu will be up to it in a few years.  That said the Mac is more user friendly than Windows and it has not won the war at corporate level.  The better product does not necessarily win because of the advantages of incumbency.   

A comment on the stock selection

This a long way outside my circle of competence.  I reckon I can read a bank’s financial account as well as anybody – but nothing is added by reading Microsoft’s accounts.  So to do this I had to become more geeky than is natural for me.  Alas I discovered there is as much debate amongst geeks as there is debate amongst bank analysts and – like bank analysts – they are probably all wrong about most things. 

I really want to thank everyone who commented.

 

John

12 comments:

  1. If you wish to look at a company where there is a lot more in the accounts than the company would like then have a look at HPQ...very interesting for an accounting geek like yourself. The tax rate in particular bears scrutiny

    ReplyDelete
  2. Hi John, read these comments with interest, you've missed one important thing in the tech sector:

    The way it works is a cash flow generating machine which feeds a R&D machine. *The R&D machine is hit and miss.* It's not the case that Microsoft has better people or engineers than (say) Apple, Oracle, or Sun. It's just that some of these companies get lucky with a product every now and then.

    Don't look at Apple's successes - think about the many many failed Apple products that got released and died to understand why the iPod took off.

    The R&D process is hit and miss. Microsoft have had a few misses - they'll have a hit sooner or later. The question you've got to ask is - "what is the frequency of hits?" - and "how long will MS cashflows stay nicely positive until they have another hit?"

    "Hits" (i.e. original product lines which have massive ROI) seem to come every 3-5 years.

    ReplyDelete
  3. And I should add (I'm the last anonymous) - I'd love to read anything on insurance.

    I hate insurers. I think the only sensible way to invest in them is if you know some industry insiders. Is there another way?

    ReplyDelete
  4. a) One other threat is serious web-based business applications which are neutral as to what platform the client uses.
    b) two other success stories for Microsoft: Office Communications Server as a replacement for an aging PBX, and Sharepoint, which is already becoming entrenched in corporate America. - TZ

    ReplyDelete
  5. "The stock price is so low relative to indisputable current cash flows that Microsoft would really have to stuff it up from here not to make the stock a worse investment than say the current ten year bond."
    That's all you need to know. It's a value stock with a margin of safety. Go ahead and add to your position.

    Absent Ballmer's departure, MS will never be a force in search or mobile OS's. Even if he did leave, I wouldn't bet on it happening.

    ReplyDelete
  6. John, I have a slightly different set of threats to MSFT. I believe they are:

    1) Not virtualization. Virtualization is annoying to MSFT, but every VM still requires a Windows license. So they don't lose much there, just the incremental cost that the virtualization vendors charge for their hypervisors. Hyper-V is "good enough" and will likely steal market share soon, and they will only get better.

    2) Not Linux. Linux is not a big threat anymore, because businesses pay just as much for Red Hat support as they do for MSFT. I think a free operating system was very scary at first, but in the end Linux has it's place in IT, and MSFT has it's place. I think we've reached steady state and both can coexist.

    3) Google. Google has a stream of revenue that is completely orthogonal to MSFT's which makes them very hard to kill. MSFT's usual modus operandi is to dilute their competitors products with just-barely-good-enough-but-free offerings. Because they can fund it with their Windows and Office revenues, they will eventually win due to a war of attrition. The only exception in software, I believe, is Quicken vs MS Money.

    However, Google has the ability to steal revenue from Microsoft, via Google Docs. This is why Bing has been so important, because they have started to make headway against stealing a bit of Google's revenue. This is also why MSFT is paying so much to their salespeople to get customers seated on Online Exchange vs gmail. They are doing whatever they can to cut off the revenue stream to Google so that they can start chipping away at it. Once Google's revenue stream starts getting impacted, they will cut back on perks, the spoiled Silicon Valley engineers will leave because "it sucks here", etc, and things unravel.

    Meanwhile Google is trying the same exact same thing, but it will be harder because MSFT has 4x the revenues and profits. However, as long as they maintain their dominance in the Internet, they will be largely untouchable and can keep jabbing MSFT but keep out of reach. Google Docs is getting better, and if they can make something really "good enough" to Office, then MSFT is in trouble. No one under the age of 18 will buy MS Office (the Student's edition of Office will disappear) and kids will grow up on Google Docs.

    ReplyDelete
  7. John,

    A monopoly is always a good business to be in.

    Take a look at IBM's financial statements pre-1982. High margins with strong top and bottom-line growth, an investor's wet dream.

    Heck, even AT&T pre-1985 looks fantastic. Granted, growth was single-digits vis-a-vis MSFT's current low-teens growth, but AT&T was a much bigger company and look at all that free cashflow generated!

    Anyway, back to MSFT. It's still a buy because things can't get worse from here (Hey MSFT is still a big ship, and if a big ship is slow to turn, that also means it takes a very long time to sink) . And if MSFT turns out to be the next Eastman Kodak, I could always point to this http://brontecapital.blogspot.com/2009/12/kodak-bill-gates-and-efficient-markets.html

    ;)

    ReplyDelete
  8. Virtualization works well when the applications for different OS are compile for the same CPU instruction set(i.e. Intel based cpu). In this case, when you run say a windows app within linux, a large portion of the instruction can be directly executed by the CPU.

    On the other hand, if the app is compiled for one CPU but is ran on a CPU with a different instruction set the app will run much slower (up to an order of magnitude slower). The reason is the cpu running the app will need to emulate another CPU's instruction set one by one.

    This can be done and has been done Take a look at Multiple Arcade Machine Emulator (MAME). Today, I can run any coin-op arcade game from my youth on my PC. However, notice I am using a modern PC that is 100X or more faster than the original hardware of these arcade games.

    If you use one modern machine to emulate another modern machine with a different CPU instruction set, the performance will be very poor. Unless cell phones use CPUs that support intel instruction set, cell phones will probably never satisfactory emulate PC.

    Your theory that virtualization will make the OS less critical is theoretically correct for intel based machines. However, in practice, it is unlikely the average users will put up with the complexity of multiple OS and virtualization.

    To unseat microsoft, it cannot be accomplish by running today's fat client apps (e.g. ms office, adobe creative suite, etc.) using virtualization. The prevalence of this model needs to change.

    What really needs to happen is to move these fat client apps into the cloud. In which case, there is no need for virtualization. All that is needed is a browser. This is already happening with casual office apps (google doc, and many others). However, it is not yet clear cloud computing will work for the demanding users. Although I'd guess in 5 years or so, the odds are good.

    However, even than there are still problems for cloud computing.

    1. privacy - why should I trust my person data to a corporation? There hasn't been a major privacy event to put this issue into people's consciousness. The question of privacy is untested.

    2. cost - I could potentially be held hostage by the corporation that provide the cloud computing service. Why should I have to pay a monthly access fees for data that I've created and own.

    3. network - what if I don't have network access? We can store data and program in the local cache. When computing happens against the data and code in the local cache, than in affect we are still doing local computing. If the PC has to download both data and code locally for later use, why cloud computing again?

    4. there will always be some app, perhaps computing intensive apps, that will not work well in a cloud computing model.

    ReplyDelete
  9. I am not an economist, but I was a strategist at MRM when we had the Microsoft account (I no longer work for MRM, & Microsoft has moved its online marketing to Wunderman). The following are a few reasons I don’t have much hope in the future of Microsoft:

    They are highly introverted and try to enforce their rules on the market. Plugins and Mashups are frowned upon, which has feed the phenomenal growth of Mozilla Firefox. Ironically they sponsor Mashable!

    Barriers to entry into the software market have greatly diminished making it vastly more competitive.

    Interoperatability issues (which once enabled their rapid growth through positive network benefits) virtually no longer exist between windows/mac & linux, and with the growth in SaaS the OS you run becomes even less important.

    Sharepoint is junk (too complex and rigid for fluid collaboration), evidence of this is the growth of companies such as Atlassian and 37Signals. File sharing is becoming a lot easier and accessible, just check out DropBox.

    Lastly, they have taken a defender strategy, ‘Its our market, we have to stop the interlopers’. The mindset of an old king.

    Obviously I don’t have much love for Microsoft today, but at one stage in my life I fought hard for a job in the agency that had their account.

    ReplyDelete
  10. Thanks to people are selling Microsoft to make room for tech stocks Microsoft is now a value stock.

    ReplyDelete
  11. John, you made the comment:"arguably the best business in the history of capitalism". That reminds of something I've read recently by another prominent fund manager (although he is not as successful as you). The statement is probably true, although there are some other contenders too (eg, Exxon). Anyway, the reason for this post is to suggest that you redouble your efforts in terms of avoiding group-think - if you are borrowing an expression from someone else, that might be because you are spending too much time reading other people's publicity and, frankly, your time can be better spent with your nose in annual reports. Remember, you have earned your success by doing your own work, not by following the croud. Cheers.

    ReplyDelete

  12. I wonder how much of Microsoft's revenue comes from their "CAL" model: whenever a corporation running exchange, office, windows hires a new staff member, they have to ship some money off to Microsoft for the network licenses for all of those products.

    And so Microsoft's revenue in a way should somewhat track the (white-collar) labour market as a whole: as people get hired, money is sent to Microsoft. As people get laid off, the already-paid-for CAL's can be recycled toward future hires.

    ReplyDelete