TXN - Avnet May Look Cheap But There Are Reasons

Summary

• Avnet beat sell-side expectations in the fiscal second quarter and guided ahead of Street expectations (and broadly in line with semiconductor bellwether Texas Instruments).
• Past cycles have seen roughly 15% peak-to-trough revenue declines for Avnet; pricing was much stronger in the upcycle, but underlying demand for many key products remains healthy.
• Avnet has long lagged Arrow in operating metrics like operating margin and return on invested capital; management has plans in place to drive better growth and margins.
• If management can deliver better growth (particularly with better margins), there could be large upside, but it's a larger risk to take ahead of a cyclical downturn.

Semiconductor stocks have sold off over the past year on fears of a downturn in demand in 2023, something that has been increasingly confirmed as companies report and issue guidance. But Avnet ( AVT ), one of the largest semiconductor distributors, has beaten the semiconductor index, the S&P 500, and its main rival Arrow ( ARW ) by a wide margin, and it still only trades at the lower end of its historical EBITDA multiple range.

Does that make Avnet a bargain? I’m more than a bit skeptical. For starters, Avnet is only beginning to see a correction in its business, and past cycles have seen peak-to-trough moves of around 15% in revenue. Moreover, Avnet has consistently lagged Arrow in metrics like operating margin and return on invested capital, and while I think Avnet can and will do better, I believe some of that is already reflected in the price and I’m unwilling to give the company too much benefit of the doubt.

The good, and bad, news here is that small changes in basic modeling or valuation assumptions drive big swings in potential value. An extra 0.1% to the long-term average FCF margin drives $2/share in fair value, and a 0.1x change in my forward EBITDA multiple drives a $1.30/share change. With the stock near 5-year highs and ahead of what I do expect to be a meaningful corrective cycle, I think there’s just too much modeling uncertainty to be strongly bullish, though I will absolutely acknowledge meaningful upside if the company can close some of its operating performance gap with Arrow.

Some Sequential Softness Emerges, But Avnet Still Beats

Avnet ’s fiscal second quarter results were better than expected, with meet-or-beat performance on revenue and margin lines. The company also guided higher for the next quarter, which has not been a particularly common event with anything tied to semiconductors lately.

Revenue rose 14% year over year as reported, and 21% in constant currency, but did decline about 0.5% sequentially (0.6% in constant currency), suggesting that the September quarter was quite likely the peak for the cycle – a conclusion that reports from companies like Texas Instruments ( TXN ) would also support – but revenue was still over 3% higher than analysts expected.

By segment, Electronic Components revenue rose 23% yoy in constant currency and fell about 0.3% qoq, while Farnell was flat yoy and down 3.7% qoq, with a shortage of single-board computers impacting results.

Gross margin declined 50bp yoy and improved 30bp qoq to 11.7%, meeting expectations, while adjusted operating income rose almost 40% yoy and 3% qoq. Operating margin improved 80bp yoy and 10bp qoq, beating by 30bp.

By segment, Electronic Components profits rose 57% yoy and 11% (with margin up 120bp yoy and 50bp to 4.7%). Farnell profits fell 39% yoy and 28% qoq, with margin down 470bp yoy and 310bp qoq to 9.0%.

At the midpoint of management’s guidance for the next quarter, Avnet looks set to see a 6% qoq decline in revenue (versus TI’s guide for a 7% decline), though that is still about 4% better than the Street had expected, while the midpoint of EPS guidance ($1.80) was almost 8% higher.

Lead-Times Still High, But Pricing Is A Risk

Lead-times in the semiconductor industry first started contracting in September of 2022 (down 4 days to 26.3 weeks, as per Christopher Rolland at Susquehanna) and the contraction has since accelerated, with December down 8 days (the biggest decline since at least 2017) to 24.0 weeks.

Avnet management confirmed a lot of what has been showing up in this lead-time data – namely that while lead-times for some components are easing (connectors especially), certain products (MCUs, analog, power, IGBTs, and MOSFETs) are still much tighter, and overall lead-times are still 50% above pre-pandemic norms for some products. Even so, Avnet’s book-to-bill fell below 1.0x across all regions for the first time since mid-2020.

With customers largely having their inventories back where they want them (again, excepting certain products like auto MCUs and IGBTs/MOSFETs) and with demand weakening, I expect meaningful price pressure in the coming months for “mainline” semiconductor products. Pricing was a big part of the revenue growth seen in the industry over the past year (as capacity was constrained, volume couldn’t go up that much), and I expect that to correct. Avnet management noted that 25 suppliers raised prices in January, a much smaller figure than in prior quarters.

As volumes ease and prices correct, I expect another 10%-plus peak-to-trough decline for Avnet . On the negative side, the higher-than-normal pricing growth seen during this last upswing could drive a steeper correction. On the positive side, this slowdown looks like it will be relatively brief and mild; auto production (including EVs) will drive higher chip consumption (more content per vehicle, especially for EVs) and I expect a relatively moderate industrial correction.

The Outlook

Avnet has had a lot of operating challenges over the years, including both external challenges (losing Texas Instruments as it shifted to direct distribution and losing business from companies acquired by Analog Devices ( ADI )) and self-inflicted wounds like a poor ERP implementation years ago.

While financials presented in isolation can be misleading, over the last five years Arrow’s average operating margin has been 110bp higher than Avnet’s and its ROIC has been 5 points higher. I’m not going to say that’s the only reason why Arrow has significantly outperformed Avnet over the last five years (as well as the last 10 and 15 years), but I believe it reflects at least part of the issue.

Avnet has been trying to turn it around and drive better results. Simple “blocking and tackling” execution is part of that, but so too is the company’s attempt to continue growing more value-added business lines like its demand creation operations (where the company’s engineers work with customers to provide multiple product design alternatives and where Avnet can earn higher margins on the sales if a given semiconductor supplier’s products are used).

I’m expecting around 4% long-term normalized growth from Avnet, and I’d note that management has targets well above this (5% to 8%) over the next few years. Better execution with demand creation and better leveraging the company’s opportunities in faster-growing auto and industrial end-markets could certainly drive some upside.

On margins, I think the correction in chip prices is likely to make meaningful operating margin progress difficult from FY’23-FY’25, but I do think Avnet can achieve around 30bp of long-term average FCF margin improvement (toward 2%). As I mentioned in the open, exceeding these targets can lead to significant value creation, as the business is highly leveraged to even small margin improvements.

Discounted cash flow modeling suggests limited (around 5%) upside from here on my base-case, but if the company could get FCF margins above 2% on a sustained basis, upside expands to 20% or more pretty easily.

Likewise with EV/EBITDA. While I think an 8x multiple (in the middle of the company’s recent historical range) on trough EBITDA is reasonable given the risks and uncertainties of a near-term downturn, and supports a $51 fair value, every $10M or so more in trough EBITDA would drive about $0.85/share of upside and I may be overestimating the downturn.

The Bottom Line

Simply running the numbers, there’s little doubt in my mind that Avnet shares could do very well if the company can get the business on a sustainably better trajectory. “Will they, though?” is the key question, and I just don’t have the answer to it. I don’t really want to make a big bet on sustained self-improvement going into a downturn, but given the upside potential from even relatively modest improvements, it’s a name worth watching.

For further details see: