Innovation, growth and goodwill

Sun's Q1FY09 earnings release 3 weeks ago was a historic moment with the disclosure of revenue per lines of business. This is giving a unique insight into whether Sun's strategy and investments are working out and how. Is Sun's open innovation paying off? Is it creating value for Sun? Interestingly, the Q1FY09 earnings also contained a piece of news about goodwill, a goodwill impairement though. How does all this come together then?

In today's post, I am making an attempt to connect the dots and make sense in all this. Let me start with a disclaimer though. I have no access to sensitive company information, all of the information used in the analysis thereafter is public data, don't look for insider's knowledge here. Also I am not a professional financial analyst, comments below are my own (naive) speculations, don't look for any experienced guidance here.

Known as early as Preinreich (1936), a firm's goodwill can be expressed as follow :

Goodwill (t)P(t)  −   BV(t)  = 

t = 1
E[x(t) − R · BV(t-1)]


P(t) denotes the market value of the firm's equity at year t,
BV() the book value,
x() the net earnings,
E[ ] the expectation operator,
R the discount rate (return on equity).

Note that ax(t) = x(t) − R · BV(t-1) is known as residual income in corporate finance litterature. Similar to the Net Present Value concept, residual income is a measure of how much value a firm creates (or destroys). In a simple perpetual growth model, goodwill becomes :

Goodwill (t) E[ax(t+1)]
R − g

where g denotes the growth rate of income. Goodwill is thus a function increasing in ax(t) --the current profitability-- and in growth g --fueled by the firm's innovation. See here for more details on how the above 2 equations are connected.

From innovation to growth

I am extracting the following revenue data directly from Slide 6 of the Q1FY09 Earnings presentation. FY06 revenue is calculated based on the year-over-year FY07 growth rates provided on the slide. I have grouped OpenStorage with x64 Servers as OpenStorage revenue seems to be primarily attributed to Thumper, which was built off common x64 intellectual property obtained through the Kealia acquisition. Data is ranked based on the Q1FY09 year-over-year growth rates.

FY06 FY07 FY08 CAGR 06-08 Q1FY09 Y/Y
Solaris SPARC CMT Servers 180 609 1119 149% 83%
MySQL / Infrastructure 138 199 208 23% 50%
Solaris, Management and Virtualization 197 222 216 5% 21%
Java 194 219 219 6% 18%
X64 Servers + Open Storage 475 733 798 30% 13%
Disks & Storage Arrays 1398 1412 1527 5% 3%
Tape 1076 989 952 -6% -4%
Other Storage 42 38 59 18% -16%
Solaris SPARC Enterprise Servers 3322 3388 3412 1% -27%
Other Systems Products 2144 1737 1291 -22% -56%
Table 1 - Billings per Business Unit

In the top half of Table 1, we find all of the most recent strategic moves of Sun, into Software and Volume Servers (inc. the hydrid data servers of OpenStorage). I find this very encouraging and an evidence that Sun's open-innovation strategy is working. Sun is able to expand outside of its historical high-end Enterprise IT market and earn marketshare for its differentiated products alongside the commodity platforms (Lintel, Wintel). Because of such good fundamentals, I believe Sun can come out strong of the transition period it is in, of adjusting its cost structure and profitability ratio.

From growth to goodwill

The Q1FY08 earnings release also announced a $1.445B goodwill impairment. Early on, analysts attributed it either to the StorageTek or MySQL acquisitions. The amount of the write-off led to conclude however that MySQL was probably not to blame. If confirmed, this result would be consistent with the growth rates from Table 1; the rate of growth for the Software business is accelerating in Q1FY09 compared to its past trend so, if goodwill should change in any direction for the Software business unit, it should be upwards.

From Sun's SEC filings, I was able to gather the following data, that summarizes Sun's acquisition activity up to the previous major goodwill impairment in 2003, which resetted the goodwill balance close to zero --if we fully attribute the $2B goodwill impairment in 2003 to pre-2003 acquisitions, Sun started the 2003-2008 period with a little $155M net goodwill. I have categorized, to the best of my understanding, the acquired companies using the product categories listed in the Q1FY09 earnings slides (Note: xVM is short for Virtualization and Management; Volume Servers include Sparc CMT and x64 systems).

Acquisition Fiscal Year Total Price ($M) Goodwill ($M) Business Unit Category
Pirus Networks 2003 167 143 Storage Other Storage
Terraspring 2003 30 27 Software xVM
Afara 2003 28 0 Systems Sparc CMT
Pixo 2004 23 17 Software Java
CenterRun 2004 65 46 Software xVM
Waveset 2004 136 77 Software Infrastructure
Nauticus 2004 12 4 Systems Other Systems
Kealia 2004 93 0 Systems x64 Servers
SevenSpace 2005 48 37 Services Services
Procom 2005 52 0 Systems Other Storage
Aduva 2006 13 12 Software xVM
StorageTek 2006 4082 1886 Storage Storage
SeeBeyond 2006 375 252 Software Infrastructure
Tarantella 2006 26 19 Software xVM
Neogent 2007 24 7 Services Services
SavaJe 2007 Software Java
MySQL 2008 904 711 Software MySQL
Innotek 2008 163 103 Software xVM
Cluster FS 2008 Software Solaris
Vauu 2008 Software Infrastructure
Table 2 - Business Combinaisons 2003-2008

Business Unit Goodwill ($M)
Software 1264
Services 44
Storage 2029
Volume Systems 0
Enterprise Systems 4
Total 3341
Table 3 - 2003-2008 Goodwill per Business Unit

Fiscal Year New Goodwill Impairment Balance
2002 - - 2182
2003 171 (2027) 326
2004 144 (64) 406
2005 37 (2) 441
2006 2169
2007 7 (103) 2514
2008 814 (113) 3215
Table 4 - Total Goodwill Balance

This closer look at Sun's annual reports confirms that goodwill is concentrated in the Software and Storage lines of business. It also confirms that only the Storage goodwill is as big as the write-off so the Storage goodwill has necessarily been impaired. If we discard Software altogether based on the growth rate analysis, we also conclude that the bulk of the goodwill impairment can be attributed to the StorageTek acquisition.

Is this write-off bad news? Long-term, I think the rationale for buying StorageTek continues to make sense, i.e. offering the complete range of products, services and solutions to Sun's traditional high-end customers --the needs of mission-critical data and vertically-scalable applications are not today and may not ever fully be addressed by commodity platforms. When these customers (e.g. Wall Street, Manufacturing) come out of recession, Sun will remain a vendor of choice and may in fact gain marketshare. Because Sun uses the same underlying technology innovation in its emerging business lines --OpenStorage is OpenSolaris, Sparc CMT is OpenSparc--, the Enterprise Solaris Sparc platform should gain in credibility and popularity throughout the economic downturn, unlike other Unix/RISC alternatives, I predict.


Great prost, 2 questions:
- What is "other storage"? Honeycomb?
- What is "Other Systems Products" Workstations?
In which unit is the Sun Ray business?


Posted by Dennis on November 19, 2008 at 06:39 AM CET #

Extracted from Slide 6 of the earnings presentation:

\* Other Systems Products includes Workstations, Volume SPARC, Netra SPARC, Integrated Products and Networking products

\* Other Storage includes the NAS products

Hope this helps, Frederic.

Posted by Frederic Pariente on November 19, 2008 at 07:51 AM CET #

Post a Comment:
  • HTML Syntax: NOT allowed

How open innovation and technology adoption translates to business value, with stories from our developer support work at Oracle's ISV Engineering.



« April 2014