Shares of Intel (INTC -0.97%) have fallen 23% since the beginning of the year, due to ongoing concerns about the future of the PC market. After that decline, some investors might consider the stock a bargain.

After all, Intel topped analyst estimates on both the top and bottom lines during its second quarter earnings, the stock trades at just 12 times earnings, and it pays a forward annual dividend yield of 3.4%. But in my opinion, investors should avoid this stock for three simple reasons.

1. The PC market is stuck in the mud
Intel's processors are respectively installed in over 80% and 90% of desktops and notebooks worldwide, according to IDC. But back in March, the company slashed its full-year revenue forecast by nearly $1 billion, due to sluggish demand for PCs worldwide.

Source: Pixabay.

During the second quarter, research firm Gartner reported that worldwide PC sales plunged 9.5% annually in the second quarter, due to longer upgrade cycles, price hikes caused by a strong dollar, a shift toward tablets, and a lack of catalysts ahead of Microsoft's Windows 10 launch. Looking ahead, Gartner expects worldwide PC shipments to rise just 1.5% annually this year.

Intel's results reflected that decline last quarter, with revenue at its Client Computing group, which accounted for 57% of its top line, dropping 13.5% annually. Unless PC makers can get customers excited about desktops and laptops again, that downward trend will likely continue.

2. The performance plateau
For several years, Intel has relied on "tick-tock" launches for its chips. A "tick" launch shrinks the existing microarchitecture with an improved manufacturing process. A "tock" launch introduces a new microarchitecture, but the size remains the same.

With Intel's 32nm chips in 2010 and 2011, the gap between the tick and tock was exactly one year. But that gap expanded to 13 months with its 22nm chips in 2012-2013, and stretched out to 15 months with its 14nm chips in 2014-2015. Intel's latest chip, the 14nm Broadwell, is a "tick" launch. Skylake, the 14nm "tock" launch, is expected to arrive in August -- 11 months after Broadwell's late arrival.

But that doesn't mean that the "tick-tock" schedule is back on track. Cannonlake, the 10nm "tick" launch for Skylake, has been delayed to 2017 due to manufacturing issues. Instead, Intel will launch a second "tock" next year with Kaby Lake, the third 14nm chip after Broadwell and Skylake. This indicates that Intel's ability to keep selling smaller and more powerful chips could soon hit a plateau. If that happens, other competitors in the chipmaking industry -- like AMD and ARM-based rivals like Qualcomm -- might start gaining ground in the PC and Data Center markets.

3. A lack of financial clarity
Intel also reports revenue and operating income in an opaque manner. Starting this fiscal year, it merged its PC and mobile divisions into the Client Computing group. The company claimed that the two units were merged to improve lines of communication between product teams, but I suspect it was done to prevent investors from looking too closely at its mobile losses.

Last year, Intel's mobile division posted a $4.2 billion operating loss, down from a loss of $3.1 billion in 2013. Revenue also plunged from $1.4 billion to $202 million. Those losses were caused by Intel's subsidization of OEM partners with co-marketing agreements, steep discounts on Atom chips, and financial assistance in redesigning logic boards. Now that those losses are hidden away in the Client Computing group, investors won't know exactly how much Intel is spending every quarter on mobile subsidies.

Source: Intel.

During Intel's most recent earnings call, CFO Stacy Smith declared that 70% of its operating income now came from its Internet of Things, NAND, and Data Center groups, suggesting that it had diversified away from its core PC business. Yet Smith didn't mention that 55% of its operating income still came from the Client Computing group. The total exceeds 100% due to an operating loss of $708 million in its "all other" category, which absorbs various expenses and losses.

In my opinion, these red flags indicate that Intel is more interested in mixing up numbers for sound bites than being straightforward with investors.

An unclear future
Intel isn't doomed, but it's getting into the habit of covering up losses in one slipping business unit with gains in a slightly better one. That's why Intel combined its PC and mobile divisions, why it now mentions IoT, NAND, and Data Centers in the same breath, and why it agreed to acquire Altera for nearly $17 billion.

That strategy won't accomplish much in the long run, unless those businesses become large enough to diversify Intel's top and bottom lines away from the PC market. Therefore, Intel stock might look cheap today, but it faces massive headwinds in the near future.