NVIDIA May Be Worth More Than It Seems

For NVIDIA (Nasdaq: NVDA  ) investors, the sound that most likely comes to mind when thinking about the stock's recent performance is a low, guttural "ugh." It sounds a heck of a lot like the sound that one would make when getting punched in the stomach.

But even though NVIDIA's stock has been clobbered so far this year, investors may be premature in shoveling dirt on the company. The company just went through a very troublesome revamping of its graphics cards, and many investors have been stepping away for fear that the company will lose its competitive edge against tough foes like Advanced Micro Devices (NYSE: AMD  ) .

Some NVIDIA-watchers now think the company's aggressive moves could be set to pay dividends, while the company has been generally seeing much better financial performance after a recession-induced drubbing.

Shares of NVIDIA currently change hands at $9.55 per share. Is that a good deal? Well, first we need to get an idea of what NVIDIA's shares are really worth.

It's a beautiful day in the neighborhood
One way to get an idea of what a stock might be worth is to check out how similar companies are valued. So let's take a look at how NVIDIA stacks up against some semiconductor industry peers.

Company

Total Enterprise Value / Trailing Revenue

Price / Forward Earnings

Price / Book Value

Trailing PEG

NVIDIA

1

16.3

1.9

1.5

AMD

0.9

13.2

6.7

0.8

Analog Devices (NYSE: ADI  )

3

12.7

3.1

1.7

Broadcom (Nasdaq: BRCM  )

2.7

14.3

4.1

1.6

Intel (Nasdaq: INTC  )

2.4

9.8

2.5

1.2

Texas Instruments (NYSE: TXN  )

2.2

9.9

3.1

1.1

Average

2.2

12

3.9

1.3

Source: Capital IQ (a Standard & Poor's company) and Yahoo! Finance.
Average excludes NVIDIA.

Using each of those averages to back into a stock price for NVIDIA, and then taking the average across those results, we can come up with an estimated price per share of right around $13. So far, that looks pretty promising.

Of course looking on this chart, you might notice that on the basis of earnings-related measures -- P/E and PEG -- NVIDIA seems to be trading at a bit of a premium to its peers, but when we compare based on revenue and book value multiples, it looks like a screaming bargain. The reason for this is pretty simple -- NVIDIA's profit margin has recently fallen well below that of its peers and so investors aren't giving as much weight to the value of the company's revenue or equity.

So it would seem that the key for NVIDIA is getting those margins to rebound. By balancing out revenue and book value multiples -- which imply a margin rebound -- with earnings-based multiples -- which don't -- I've taken a bit of a middle ground on the issue.

Collecting the cash flow
An alternate way to value a stock is to do what's known as a discounted cash flow. Basically, this method projects free cash flow over the next 10 years and discounts the tally from each of those years back to what it would be worth today (since a dollar tomorrow is worth less to us than a dollar today).

Because a DCF is based largely on estimates (aka guesses) and it attempts to predict the future, it can be a fickle beast, and so its results are best used as guideposts rather than written-in-stone answers sent down from on high.

For NVIDIA's DCF, I used the following assumptions:

2009 unlevered free cash flow

$340 million

FCF growth 2010-2014

13%

FCF growth 2010-2019

6.5%

Terminal growth

3%

Market equity as a percentage of total capitalization

100%

Cost of equity

12%

Cost of debt

NM

Source: Capital IQ (a Standard & Poor's company) Yahoo! Finance, and author's estimates.

While most of this is pretty standard fare when it comes to DCFs, the academically inclined would probably balk at the way I set the cost of equity. In a "classic" DCF, the cost of equity is set based on an equation that uses beta -- a measure of how volatile a stock is versus the rest of the market -- and a few other numbers that I tend to thumb my nose at.

But when you get right down to it, the cost of equity is the rate of return that investors demand to invest in the equity of that company. So I generally set the cost of equity equal to the rate of return that I'd like to see from that stock.

Based on the assumptions above, a simple DCF model spits out a per-share value just a bit more than $15 for NVIDIA's stock. I should note that a chunk of that value -- more than $3 per share -- comes from NVIDIA's hefty cash hoard. However, also beware that the company's cash flow has been very volatile and therefore tougher to model.

Do we have a winner?
The valuations that we've done here are pretty simple and, particularly when it comes to the DCF, investors would be well advised to play with the numbers further before making a final decision on NVIDIA's stock.

That said, the two valuation methods that we've used here suggest an intrinsic value for NVIDIA between $13 and $15 per share. The midpoint of that range is nearly 50% above NVIDIA's current share price, so it looks as if right now NVIDIA's stock is priced to be bought.

Do you agree that NVIDIA's stock is on sale? Head down to the comments section and share your thoughts.

Want more stock ideas? How about 50 of them? Well, Fool, you better check out the 11 O'Clock Stock!

Intel is a Motley Fool Inside Value pick. NVIDIA is a Motley Fool Stock Advisor recommendation. The Fool owns shares of and has written puts on Intel. Motley Fool Options has recommended buying calls on Intel. Try any of our Foolish newsletters today, free for 30 days. The Motley Fool has a disclosure policy.

Fool contributor Matt Koppenheffer owns shares of Intel but does not own shares of any of the other companies mentioned. You can check out what Matt is keeping an eye on by visiting his CAPS portfolio, or you can follow Matt on Twitter @KoppTheFool or on his RSS feed. The Fool's disclosure policy assures you no Wookiees were harmed in the making of this article.


Read/Post Comments (12) | Recommend This Article (17)

Comments from our Foolish Readers

Help us keep this a respectfully Foolish area! This is a place for our readers to discuss, debate, and learn more about the Foolish investing topic you read about above. Help us keep it clean and safe. If you believe a comment is abusive or otherwise violates our Fool's Rules, please report it via the Report this Comment Report this Comment icon found on every comment.

  • Report this Comment On August 09, 2010, at 11:04 PM, baldheadeddork wrote:

    The same technical reasoning/forecasts built on SWAG was used to support claims that we weren't in a housing bubble three years ago.

    I think your assumptions are based on equally suspect notions. Nvidia is going to average 6.5% annual growth in free cash flow for the next decade? Are you aware that it's en route to being forced out of the motherboard chipset business and that it's low end graphics chip market is imperiled by AMD and Intel moving graphics to the CPU? They are a distant also-ran in the SOC mobile device market, too.

    Put it all together and Nvidia is very likely going to be a one-product (discrete mid and high-end graphics chips) company by 2015. They're really good at this, but the same was once said about countless other tech companies that are now gone. In fact, the company that Nvidia reminds me of most today is Palm circa 2003.

  • Report this Comment On August 10, 2010, at 7:05 AM, TMFKopp wrote:

    @baldheadeddork

    Thanks for the comment.

    Not sure I follow your housing comparison except that bullish predictions on housing three years ago were wrong and you apparently think the NVIDIA valuation is off.

    As for NVIDIA and the numbers here, a few quick thoughts:

    - Growth forecasts are based on analyst expectations, and their numbers should take into account any expected changes in NVIDIA's market share.

    - The model above -- as noted -- is a simple one. It's ideal for investors to play around with the numbers themselves to see how changes like what you reference would impact the company's value.

    - You make the assumption that it will be bullied out of all but one product line... that's as much of an assumption as the expectation that it won't.

    Bottom line is that the issues that you're considering are exactly what investors want to think about when they take a financial model to the next level. Playing with the assumptions, challenging expectations -- that's what will yield the best results. Above is meant to be a starting point.

    I do, however, find myself scratching my head again with the Palm comparison...

    Matt

  • Report this Comment On August 10, 2010, at 8:59 AM, DrBrandonRye wrote:

    I really enjoyed the article but could you give some more details (for us inexperienced investors) on how you calculated the target share price from the P/E, P/B, Trailing PEG and revenue?

  • Report this Comment On August 10, 2010, at 10:28 AM, TEBuddy wrote:

    Your comments make it clear you dont know Nvidia's business very well. Nvidia has money, but isnt going to be making as much any more. They are losing market share in their prime business and as stated are getting left out of areas they previously made considerable margins. Nvidia stopped making chipsets for AMD long ago, and refused to pay a license for Intel's new chipsets. Like they forget to pay on several other licenses too. Nvidia got to where they are by swallowing the competition, but AMD is not going anywhere and has better technology, sell more efficient high end GPUs. AMD and Intel now dominate the integrated graphics market and will only get worse for Nvidia. Thier SOC did not take off, probably because it cost more than other larger companies were willing to build them for.

  • Report this Comment On August 10, 2010, at 5:35 PM, enkidusfriend wrote:

    @TEBuddy

    Agreed and have expressed same sentiments earlier. nV losing their chipset market isn't an "assumption", it is a fact without that QPI licence. Chipset functions are getting swallowed by the CPU anyways, so even without the licence issues, that market is evaporating.

    No one seems interested in actually building marketable products based on the Tegra 2 SOC so far and all those design wins are still so much paper. I sure hope analysts aren't calculating those into their growth forecasts because it seems more and more likely to be a black hole.

    The previous article, reference above about nV trying to retake the mainstream discrete GPU market is also problematic. Their initial pricing was very aggressive on the GTX 460 1GB and 768MB models and prices are already falling on both those mainstream and high-end cards when their main competitor has seen prices on some SKU's persisting at well over the original MSRP for months since launch.

    The GPGPU market looks much more promising for nV and they were prescient in focusing more and more on that as a core market.

    The pain will continue until the rest of their desktop line comes out and perhaps not until their discrete GPU's get a die-shrink.

  • Report this Comment On August 10, 2010, at 6:00 PM, Phoenix002 wrote:

    It's not "they are being forced out of the motherboard chipset market", it's "they have been forced out of the motherboard chipset market". The DMI/QPI spat with Intel is a very costly one for nV, not just financially but also technologically - even if they won the lawsuit along with a DMI/QPI bus license it does not mean nV can come to market with a compelling chipset the next day! The lawsuit also diverted funds from developing a comparable AMD-based solution as well. This was a bad fight to pick, nV!

    As profitable as the HPC market is for nV, my personal middle-term outlook is still a sell until nV can get its core consumer-oriented products in order.

  • Report this Comment On August 10, 2010, at 6:12 PM, TMFKopp wrote:

    @DrBrandonRye

    "could you give some more details (for us inexperienced investors) on how you calculated the target share price from the P/E, P/B, Trailing PEG and revenue?"

    Sure thing.

    As an example, in the above table, the average price-to-forward earnings multiple is 12. NVDA's earnings per share over the next 12 months are expected to be $0.59. So you assume the market puts that multiple of 12 on the forward earnings of $0.59 and you get to an implied price of just over $7.

    It works similarly for revenue, book value, and the others. You take the average multiple, apply it to the results or expected results and you get to the implied value.

    Just remember that any multiple based on a number above the interest line (EBIT, EBITDA, revenue, etc) is going to be enterprise value / number. So in that case your implied value will get you to the enterprise value, then you have to visit the balance sheet to get from that back to the equity value.

    Hope that helps...

    Matt

  • Report this Comment On August 10, 2010, at 6:20 PM, TMFKopp wrote:

    @Phoenix002, enkidusfriend, et al

    Thanks for bringing some great thoughts to the discussion!

    Based on where NVDA is trading, it's obvious that the market/investors don't buy the current growth estimates and it seems like you folks have your finger on the pulse of why.

    Out of curiosity, where would you peg growth for the company over the next few / five years?

    Matt

  • Report this Comment On August 10, 2010, at 6:26 PM, baldheadeddork wrote:

    @ enkidusfriend - Fermi based GPGPU's look great on paper, but Nvidia totally blew it with their thermal output. Heat management is a huge issue for everyone running a server room or data center. Putting three, four or five Fermi GPGPU's in a server doesn't save any money over more servers or new CPU's if you have to spend $$$ on added cooling for your server room.

    No one in my business likes spending $30K on a new quad Xeon server, but when you look at the whole cost of ownership it's a lot less expensive than spending $100K on new air conditioning equipment.

  • Report this Comment On August 10, 2010, at 10:28 PM, investorak wrote:

    Its funny how you have put together the valuation based on all sorts of multiples in the world but left out the most important one for semiconductor companies and thats price/sales. Your analysis is flawed that's the least I can say..

  • Report this Comment On August 11, 2010, at 1:02 AM, TMFKopp wrote:

    @investorak

    Would a rose by any other name smell as sweet? I'm not really sure....

    However, what I am sure of is that enterprise value / trailing revenue -- which is in the table above -- covers the same bases as price / sales.

    Matt

  • Report this Comment On August 14, 2010, at 3:04 AM, Sovestor wrote:

    Have anyone ever thought that NVDA might not even produce positive Free Cash Flow in the next 3-5 years due to increased competition in the sector? Have investors gone blind to project free cash flow based on the past figures? Wake up!!!

Add your comment.

Sponsored Links

Leaked: Apple's Next Smart Device
(Warning, it may shock you)
The secret is out... experts are predicting 458 million of these types of devices will be sold per year. 1 hyper-growth company stands to rake in maximum profit - and it's NOT Apple. Show me Apple's new smart gizmo!

DocumentId: 1264260, ~/Articles/ArticleHandler.aspx, 10/31/2014 6:16:24 PM

Report This Comment

Use this area to report a comment that you believe is in violation of the community guidelines. Our team will review the entry and take any appropriate action.

Sending report...


Advertisement