August 3, 2011

Corporate board of directors: how to formulate them for success

These days, more than ever, boards of directors carry the responsibility for the company’s health, fiscal responsibility, and future expansion. This is particularly true for public companies. Over the years, I’ve worked with boards of both for-profit and nonprofit organizations. While some boards work better than others, there are key factors to consider to guide and support the board’s success.

Specialist vs. generalist

Many advisors recommend a board with people with specific areas of expertise, but I couldn’t disagree more. I believe board members should have very broad experience in various types of businesses that are similar to the ones on the boards of which they sit whether in the way they operate, their vertical, or their geographies. When specialists are needed, smaller companies can allocate them to their board of advisors, and larger companies should have the means to hire consultants or outside service providers for their specific needs. If the board members have enough experience, they’ll know just where to find help when they need it.


While size does matter, there’s no universal agreement on the optimal size for the board of directors. According to BoardSource, the average board size is 15, and according to the Corporate Library’s Study, the average board size is just under 10 members (clearly, these two studies were not using the same research base, but I digress…). For smaller companies, any governing body larger than 5-7 members becomes ineffective. Larger/public companies typically have larger boards, however, there are many committees for the board members to participate in (in particular an audit committee and a compensation committee, which are becoming more and more important in the governance of companies), so larger boards for larger companies become necessary. Any board larger than 15 members becomes difficult to manage and typically ineffective. As an example, IBM’s corporategovernance guidelines indicate that “10-14 directors on the board is optimal”. I happen to agree.

Independence from the company

The board of directors is a body of elected or appointed members who jointly oversee the activities of the company, including the CEO. Stacking the board with insiders inherently creates conflicts in oversight of the company and its CEO. Having majority or all outsiders on the board will help the board objectively guide the company in the right direction. An outsider is someone who has never worked at the company, is not related to any of the key employees, and has never worked for a major supplier, customer or service provider.


Many companies don’t call out a particular time-frame for their board’s tenure, however, with the fast changing nature of business, it is best they refresh their talent in the boardroom regularly. While there are benefits to continuity in the boardroom, becoming stale and complacent towards the company’s interests can pose risks to the company’s health and growth. Many boards hold reelections every year. I think having a maximum tenure of 7-10 years will help keep the board’s perspective fresh and objective. This also keeps away career board members who may lose their once-honed operational expertise by supporting themselves solely by sitting on corporate boards (I’d like to have that lifestyle, wouldn’t that be nice!).


Yes they do. Quite nicely, as it turns out. Board compensation can vary greatly based on the company size, location, and industry. Also, the composition of the compensation can vary widely from mostly cash to mostly stocks. According to a study by C-Suite Insights, the level of board compensation increases with the company’s market cap. On average, small-cap board members received $101,750 per annum, their mid-cap peers earned $138,500, and large-cap board members made $190,000. These figures include cash retainer fees, annual equity awards, and total meeting fees paid for service on the board. They exclude any fees paid for committee service. Keep in mind these are averages for public companies. Private companies do not publish their board’s compensation, so it can vary widely. Some Fortune 100 companies are known to pay 7 figures to their board members.

I recently wrapped up a strategy engagement with a long term client, and the board’s composition was the topic of many discussions. Clearly, it’s difficult for the board to self-regulate, and unfortunately, many boards wait for an external event (usually a negative one) to self examine. The best way to ensure a strong and objective board is to regularly examine the board’s composition, and update the bylaws as needed.

May 19, 2011

Kat Shoa moderating panel at Connectivity Week 2011: Integrated Critical Communications Infrastructure

I've been asked to moderate a panel at the Smart Grid Connectivity Week 2011. The topic is of utmost importance to the success of the Smart Grid: how to integrated all the critical communications around the utility infrastructure.

What:  Integrated Critical Communications Infrastructure
When: Wednesday, May 25, 3:30-5pm
Where: Santa Clara Convention Center, Room #204

We have a prestigious panel discussing this topic:

Jon Sessions: Chief Scientist, SAIC
David Witkowski: President, Wireless Communications Alliance
George Flammer: Chief Scientist, Silver Spring Networks
Howard Liu: Network Architect, Southern California Edison

If you haven't registered, you can register on their website and get a 20% discount with the code below.


Hope to see you there! I'll be there on Wednesday only.

March 1, 2011

The direct impact of R&D expenditures on market capitalization: an analysis of large cap companies – Google, Apple, Microsoft, HP, IBM, 3M, Intel, Xerox, Oracle, Cisco, Caterpillar, GE, Johnson & Johnson, Dell

As many of my readers know, I’m a big proponent of R&D expenditures for companies developing products, services, and intellectual properties. And for some time, I’ve been thinking about analyzing the effect of R&D expenditures on two metrics: market capitalization and revenues. Today I’m looking at the impact of R&D on market cap. As important as revenues are, the CEOs of most public companies are forced into keeping market cap (effectively their stock price) front and center in their priority list. I’ll perform similar analysis on the effect of R&D on revenues in the near future.

The analysis is quite simplistic. I plotted the R&D expenditures per number of employees, against the market cap per number of employees. Given that it typically takes 2-3 years for R&D expenditures to pay off, the R&D numbers are taken from FY08 annual reports of these companies, and the market cap numbers are taken from February 28, 2011. The number of employees corresponds with the timing of the metrics so that it evens out any acquisitions or layoffs the companies may have gone through. Clearly, many other variables should be considered in the analysis to be perfect, but for the purposes of this article, I've kept it simple.

Not to my surprise, the analysis shows that there’s a direct correlation between R&D expenditures and market cap. I’ve kept the analysis focused on large-cap products and/or services companies mostly because both their internal operations and their stock prices are typically less volatile than smaller cap companies.

The list of companies and their ticker symbols are as follows:

Google: GOOG
Apple: AAPL
Microsoft: MSFT
Intel: INTC
Xerox: XRX
Oracle: ORCL
Cisco: CSCO
Caterpillar: CAT
Johnson & Johnson: JNJ
Dell: DELL

The following chart shows how these companies compare, and the black line is the trendline (generated by Excel):

Clearly, Apple and Google, have skewed the results here with their huge market cap per employee.  These two Wall Street darlings have clearly impressed the investors resulting in very high stock prices. You can also argue that they run a mean machine since they’re doing so much more “per employee” than the others, especially in Apple's case. And the amount of R&D funds Google spends per employee is mind boggling. It seems their entire operations are focused on R&D.

For the purposes of the analysis though, I removed the two anomalies so we can focus on more typical companies. The following chart compares the list sans Google and Apple: 

Now we get a more clear picture of the trendline. Looking on the lower left side, we can see that GE and DELL spend a pathetically low amount on R&D per employee. For companies with large manufacturing operations, this makes sense since that skews the numbers somewhat. But Dell’s last earnings results show that the company’s sales are still highly dependent on commoditized products, so maybe they need to rev up the R&D expenditures a little bit.

The next group is giant companies HP, IBM, Xerox, 3M, and Caterpillar. Without looking at the operations of each of these companies in detail, it’s really difficult to dissect the results, but I was a bit surprised to see IBM in the same category as Xerox. I admire so much that IBM does and I expected it to show up higher on the scale.

Microsoft sits high up on the top right. Software companies typically have higher R&D costs per employee, and given that Microsoft has had layoffs recently, the market cap per employee has also gone up. By the way, is anyone as surprised as I am about how much Johnson & Johnson spends on R&D? I'm not very familiar with the company's operations, but its position looks pretty respectable on this chart.

Based on this admittedly simplistic analysis, the main takeaway is that the lower right and the top left of the chart are completely empty, i.e., if a company spends a lot of money on R&D, their market cap 2-3 years down the line will not be very low, and vice versa (are you listening Mark Hurd?). Unfortunately, most public companies are forced into showing very short term results, so their CEOs constantly weigh the cost of R&D against their short term bottom line. I really wish I could do this on the entire S&P 500 group to see a better visual, but this took long enough.

Stay tuned for a similar analysis on the effect of R&D expenditures on revenues. I have a feeling the charts will look very similar.

As always, comments are welcome.