Nvidia’s R&D spending examined

Not what you'd expect

WHAT DO YOU DO when your company does not have a competitive product in the most profitable market segment, your competitors are developing products that will kill your share in the high volume segments and the puppy in the pipeline is not looking that promising?

You cut expenditures and save cash. Lots of cash. And then you invest a lot on R&D or acquisitions to open new market frontiers for you. Nvidia (NYSE: NVDA) got the first part right. It saved cash and reduced expenditures, just like almost every other company has been in this market. The catch is in the second part. It reduced its R&D spending. No, really, even with the puppy late and without a single competitive product on the market, it actually cut down on R&D spending.

If you check Nvidia’s financial statements for FY2009, Q2 and Q3, R&D expenditures were more or less 200 million per quarter. This is a 33 percent decrease from the 300 million spent in the first quarter and a return to the 2008 levels. Poor puppy, it was so badly treated that it had woodscrews and a crudely sawed PCB.

For Nvidia, this low R&D spending means two things:

First, Nvidia is developing Fermi derivatives at a leisurely pace, which means that it doesn’t see big cash generation potential in Fermi derivatives or else it would be rushing them to market.

Second, it gave up on this generation and has nothing new in its pipeline for the short term, so it is looking forward to the next chip generation.

This would seem to mean that Nvidia’s board of directors views this year as practically lost. It is not competitive in the desktop market, its chipset business is doomed, its Quadro Business will be under pressure from AMD, its notebook business is under pressure by Intel’s Arrandale and its plan B, Tegra and Tesla business, stubbornly refuses to take off, design wins or not. For Tesla it even offered a promotion that would make the car dealers green with envy.

Now it’s a matter of generating enough cash to keep the company in the black and try to figure out plan C, which is most likely to sell high volumes of low cost parts to OEMs and gold plated Fermis as Quadro boards. It is betting on a capacity constraint, which will disappear as soon as AMD starts to produce its graphics chips with GlobalFoundries.

By any means, the pain will be there for Nvidia as it does not have the technological leadership in any relevant segment of the market. This site, My Smart Trend quotes a Morgan Stanley report that predicts a better Q4 for Nvidia, but the last bit is very interesting:

“Morgan Stanley sees fiscal 2010 EPS of 22 cents and fiscal 2011 EPS of 48 cents”. 22 cents per share for the entire FY2010. This is merely 10 percent more than what Nvidia reported last quarter.

Boursier, also quoting a Morgan Stanley report, predicts a 40 percent drop in Nvidia’s shares.

For those sites, the trend is clear: Nvidia is pretty much trapped in its current market and will be under pressure in every relevant segment, and this will put pressure on its stock.

Let’s wait for its Q409 statements, they should show us some very interesting trends.S|A

Note: Edited from 50 percent decrease to 33 percent decrease when calculating spending differences. This was a typo and the actual spending figures did not change. You can refer to the links provided for actual numbers to double check our calculations. Ed

Updated February 15th, 2010: 

Some users on our forums have pointed out that Nvidia took a one time charge of 90 million in Q1, which reduces the drop in R&D 10% from 50% as stated above. True. The problem is that if a company has a good product in the pipeline and needs to rush it to market, they would likely increase expenditures to pull in the timeline.  This costs money, the R&D should therefore rise, not stay flat or fall a little.
If you do not rush the product you will concede market share to your competitors. Things are different if you do not have something competitive, a company would likely attempt to conserve cash and resources for the next generation if there isn’t a competitive product in the pipeline.
Fermi is late, it will arrive at least 6 months later than Cypress, there is no official word about the mainstream and budget derivatives, and the mobile derivatives are still in the realm of dreams. Are these spending patterns consistent with a company rushing to put out a top to bottom line of competitive products?  One would expect increased costs for tape-out and debugging of the various derivative parts if there is actually a full line-up and they’re rushing to market as they state publicly.  Their spending on R&D would indicate otherwise.
The following two tabs change content below.


Latest posts by Newsdesk (see all)