Intel's (INTC 0.39%) stock recently jumped after the tech giant announced a $10 billion accelerated buyback plan, ending its five-month suspension of buybacks throughout the COVID-19 pandemic.

Intel previously launched a $20 billion buyback plan last October, and had already repurchased $7.6 billion in shares prior to the suspension. The new accelerated buybacks represent a continuation of that earlier program.

Intel spent $10 billion on Aug. 21 on the buyback, initially receiving 166 million shares with additional shares likely based on the weighted average share price over a set period. Intel's aggressive move suggests its stock is undervalued after its post-earnings plunge in late July, but a big buyback won't solve its biggest problems.

A canvas bag filled with coins.

Image source: Getty Images.

What are Intel's biggest problems?

Intel's problems started in late 2018, when the difficult development of its 10-nanometer (nm) chips caused a shortage of its 14 nm chips at its own foundries. That shortage allowed Advanced Micro Devices (AMD -0.47%), which outsourced the production of its newest CPUs to Taiwan Semiconductor Manufacturing (TSM 2.23%), to gain ground in the PC market.

Meanwhile, TSMC pulled ahead of Intel in the "process race" to create smaller chips, which allowed AMD to launch its first 7 nm CPUs before Intel's first 10 nm CPUs. AMD's 7 nm CPUs are technically comparable to Intel's 10 nm CPUs, since their nodes are measured differently, but AMD's chips offer similar performance to Intel's chips at much lower prices.

AMD's stable supply of cheaper chips caused headaches for Intel. According to PassMark Software, Intel's share of the CPU market plunged from 82.5% to 62.4% between the third quarter of 2016 and the third quarter of 2020, while AMD's share soared from 17.5% to 37.5%.

Intel's investors were likely hoping its foundry could make the leap from 10 nm chips to 7 nm chips without tripping over its own feet again. But last quarter, Intel admitted its 7 nm process was trailing "approximately twelve months" behind its internal target, and that those first chips wouldn't arrive until 2022 or 2023.

By comparison, TSMC started producing 5 nm chips earlier this year, and plans to start testing out 3 nm chips in 2021. That roadmap indicates AMD will widen its lead in the process race against Intel -- regardless of the hotly debated differences between TSMC and Intel's node measurements.

Why a buyback is the wrong (but predictable) move

Stock buybacks make sense when a company doesn't have anywhere to deploy its excess cash. They don't make sense when a company is struggling with execution issues, product shortages, and losing its core market to a much smaller rival.

Chips being manufactured on a wafer.

Image source: Getty Images.

Intel generated $10.6 billion in free cash flow in the first half of 2020, and it expects to generate $17.5 billion in FCF for the full year. It held $25.8 billion in cash, cash equivalents, and investments at the end of the second quarter. Therefore, it can easily afford to spend $10 billion on buybacks.

Intel only expects to spend about $15 billion on capital expenditures this year, compared to $16.2 billion in 2019. That lower capex can be attributed to disruptions from COVID-19, Intel's decision to outsource the production of its Ponte Vecchio GPUs to TSMC, and tough comparisons to its aggressive capacity expansion efforts last year.

Nonetheless, Intel's reduced capex and higher buybacks suggest it has nowhere else to deploy its cash. But even if Intel is satisfied with its current production capabilities, it could spend that $10 billion on investments and acquisitions to expand its automotive, computer vision, Internet of Things, and memory chip businesses. It could even partially fund a bid for SoftBank's (SFTB.Y 2.96%) ARM Holdings -- which is reportedly in buyout talks with NVIDIA (NVDA 5.56%) -- to take out its biggest CPU architecture rival and expand its reach beyond x86 chips.

Unfortunately, Intel's current CEO, Bob Swan, who took over in 2018 after Brian Krzanich's abrupt resignation, is the company's former CFO and has a background in finance instead of engineering. Therefore, it's not surprising to see Swan executing big buybacks instead of aggressively investing in Intel's future.

Intel has already wasted billions on buybacks

Intel's stock was already trading at about 10 times forward earnings prior to the announcement, so the stock looks historically cheap. But it's cheap for obvious reasons -- it lost its process lead to TSMC, AMD is eating its lunch in the PC market, and it delayed its upcoming chips.

Intel bought back 63 million shares in the fourth quarter of 2019 at an average price of $55.32 per share. In the first quarter, it bought back another 71 million shares an average price of $59.15 per share. The stock is currently trading at about $49 as of this writing -- so Intel already blew $7.6 billion on badly timed buybacks.

In short, Intel should stop trying to solve its problems with buybacks. They'll briefly boost its earnings and slightly reduce its valuation, but they won't help the chipmaker regain its edge in the competitive semiconductor market.