Check out the latest Broadcom, NXP Semiconductors, and Qualcomm earnings call transcripts.
In a different world, Broadcom (AVGO -4.31%) could have merged with NXP Semiconductors (NXPI -1.93%) already. NXP had a $44 billion buyout offer from fellow embedded microchip maker Qualcomm (QCOM -2.36%), while Qualcomm itself was the target of a hostile takeover attempt by Broadcom. But both deals fell apart, arguably for similarly political reasons, so we can still choose to invest in either one of these companies as a stand-alone stock.
Broadcom may be a decent value investment, but NXP runs circles around that particular investment thesis. Here's how.
What's going on here?
Long story short: NXP's share prices fell 25% in 2018 due to the failed Qualcomm merger. That deal had been simmering for nearly two years, only to fall apart at the finish line, where Chinese regulators simply never approved the massive buyout. So NXP took a $2 billion breakup fee from Qualcomm and the two companies went their separate ways. Meanwhile share prices plunged. Today NXP is trading some 42% below the all-time highs that were set at the height of the Qualcomm mania in early 2018.
The strongly negative market reaction makes it seem NXP can't go on without Qualcomm. That's a terrible mistake.
With or without the merger agreement, NXP remains a market-leading powerhouse in several thriving markets, led by industrial and automotive computing. These two sectors account for roughly 70% of the company's total revenues and represent fantastic growth opportunities for the NXP over the next several years. For instance, NXP's management expects the automotive computing market to grow by 20% a year over the next five years as ordinary cars come with more and more electronics installed -- and self-driving vehicles with even heavier chip loads enter the mainstream.
In a recent industry conference, CFO Peter Kelly shrugged off the failed Qualcomm deal this way:
I think there's a narrative that as soon as we announced the deal, we all went to sleep and didn't do anything for the last two years. And that's clearly not true.Obviously we didn't speak to investors, but we had a strategy that we continue to execute. There was almost no overlap with Qualcomm, so it wasn't like we were saying that we have this business here that we're going to disinvest in. Qualcomm were very supportive of us being aggressive in the areas that we were in. So we continued to drive our investments in automotive and industrial and we grew R&D for both of those by about 15%, compounded annually.
In fact, Kelly and his company really didn't mind losing the Qualcomm merger after two years of fruitless deal-making efforts. At some point it's just better to walk away than to extend the uncertainty of an unfinished contract even further:
In the end, the teams were just relieved to have a way forward. They were saying, in the last few months, 'Just make your minds up -- do the deal or don't do the deal. We don't care, but tell us which one it is.' Once the deal collapsed, we actually had a bit of pickup in morale.
Image source: Getty Images.
"N-X-P": Another way to spell "deep value"
In other words, the NXP we see today is no worse than the NXP that Qualcomm wanted to pay $44 billion for. Getting back to that valuation on the back of ordinary share price moves would roughly double NXP's market cap from today's prices.
Or you can compare and contrast NXP with its sector peers. In a semiconductor industry where many of the largest competitors trade at 20 times trailing earnings or cash flows, you can pick up NXP shares at just 12.5 times earnings and 6.2 times free cash flows. That's a bargain for a healthy growth stock with a firm grip on some of the market's most exciting growth opportunities.
And Broadcom's shares trade at 22 times trailing earnings or 21 times free cash flows. Fair and reasonable? Sure. But Broadcom can't hold a candle to NXP's hugely discounted stock prices. There's nothing terribly wrong with owning Broadcom, but if you want to buy only one chip stock today, NXP should be it.
