Has the Semiconductor Sector bottomed?

The semiconductor industry has historically gone through many boom-and-bust cycles over the past 25 years, whereby supply and demand fall out equilibrium leading to huge swings in the sectors sales and profits. Before the recent pull back of the SOX, (Philadelphia Stock Exchange Semiconductor Index), in January 2022 many investors pointed to the increased usage of chips beyond traditional smart phones and PCs to Electric Vehicles to toasters as a reason for the sector to hold on to all-time highs. The thought was that now that chips have become so main steam, in-bedded in almost all manufactured goods, that severe peaks and troughs where less likely.

Historically, semiconductor demand – such as for personal computers in the 1980s, mobile phones in the 2000s and smartphones in the 2010s – would lead manufacturers to aggressively expand capacity during peaks, while CAPEX would contract during sluggish periods. However, with chips now becoming a key component in industrial applications, vehicles, industrial automation, 5G infrastructure, artificial intelligence, and cloud computing.

And with such applications requiring more chips per unit creating longer cycles than just consumer electronics, this was expected to lessen the industry’s reliance on consumer spending and potentially smoothing revenue. For example, an electric vehicle (EV) uses more than twice the number of semiconductors than its internal combustion engine predecessor, and industry experts expect this gap to widen, basically the EV is the new smart phone.

All-in-all, these arguments made sense until the SOX corrected from January 2022 highs (SOX closed at 4,088) dropping 45% to bottom out in October 2022 (SOX closed at 2,263). Since the sector bottomed from October lows, the SOX has rallied in February 2023 up 32% (SOX close at 2,918). So, the question on all investor’s mind is… “Has the semiconductor sector fully corrected and is poised to rally or are more tough times ahead”? We must remember that the semiconductor sector is the plankton of the Technology and Manufacturing ocean. That is everything requires chips in it to run. Typically, the chip sector will lead the Technology sector out of the valley and will be one of the first sectors to show green sprouts foreshadowing a pending recovery.

At this point there are few signs of a rebound. Expect semiconductor industry revenue to drop in FY2023. Do see sustained long-term strength in cloud/datacenter, telco/enterprise infrastructure, and automotive/industrial segments, offset by continued weakness in consumer markets like PC / smartphones compounded by a downturn in memory fundamentals. In 2023, Semiconductor revenue growth is expected to fall to $566.4 Bn representing a -2.6% drop in the growth rate as capacity builds out to meet current shortages. Following the COVID-19 downturn in 2H2020, the industry has enjoyed eight strong quarters of above-average semiconductor revenue growth.

However, it has been clear that the industry has entered into a cyclical downturn in 3Q22, the first full quarter of negative YoY contraction since emerging out of the COVID-19 downturn. Expect sub-seasonal growth through 2023, which suggests the down-cycle would persist for six quarters (similar to 2018/2019 downcycle), driven by consumer weakness amid a global macro-economic slowdown. If we look at some of the recent results coming out of bell-weather chip makers the results point to flat or downward revisions in sales and profit growth.

Taiwan Semiconductor Manufacturing (TSM), the world's largest contract foundry chipmaker, recently beat expectations for earnings in the fourth quarter but missed on sales. It also guided below estimates for the upcoming 1Q23 quarter as chip demand weakens. TSM noted that it has seen demand soften for data center chips. Meanwhile, demand for chips for smartphones and consumer electronics remains weak.

TSM did comment that it sees the end of the chip shortage that hammered the auto industry. The company expects low single-digit percentage growth in 2023 and implied that the second quarter could mark the revenue bottom for the year. TSM lowered its projected CAPEX budget for new fabs to $36 Bn from a previously projected $40 Bn.

Intel (INTC) has been feeling the chill during the recent broad-based pull back, as chip demand for PCs has dried up after the Covid-19 Pandemic “work from home” spike. Intel recently reported fourth-quarter revenue of $14 Bn, down 32% YoY and below analyst expectations. The U.S. chipmaker also reported a 20% drop in full-year revenue at $63.1 Bn. Both of Intel’s flagship growth units Data Center Group (DCG) and Client Computing (CCG) revenues came in below analyst estimates.

Texas Instruments (TXN) recent 4Q22 results beat sales and EPS estimates. But, for the upcoming 1Q23 the forecast for revenue and profit ranges from the company were slightly muted compared to Wall Street targets, as an economic downturn threatens to shackle demand from the chipmaker's so far resilient markets. TXN saw broad based season decline with the only exception in Automobiles.

Smartphones and personal computing products were the first to feel the pinch of customers cutting back on discretionary spending as interest rates rose, but segments such as industrial have started to come under pressure. Texas said revenue in its industrial business fell 10% in the fourth quarter from the previous period, while it plunged 20% in its communications equipment and enterprise systems businesses.

Dutch-based ASM Lithography (ASML) was one of the chip equipment players to buck the negative trend. ASML recent 4Q22 sales of Euro 6.438 Bn (vs an estimate of Euro 6.38 Bn), +13% YoY. Management’s guidance “that despite uncertainty in the marketplace they are confident that the market will rebound in the 2H23” was very encouraging that a turnaround in the chip equipment sector is in the cards. ASML is considered the “Mercedes Benz” of the capital equipment space as the stock’s trading premium to the peer group (AMAT, KLAC, LCRX) shows. In regards to the lead times, customers like TSM may not immediately need extreme ultraviolet lithography (EUV) tools currently, but will not cancel an order given the possibility of a recovery in the 2H23 and a cancelled order might take 1.5 years to reinstate.

ASML is the only company in the world that makes extreme ultraviolet lithography (EUV) machines – the most sophisticated type of lithography equipment that is required to make every single advanced processor chip used in the world today. Therefore, ASML is not subject the level of cancellations during a downturn as its peers. ASML projected revenue outlook for FY2023 of 25% YoY, along with net bookings that grew 16.8% YoY and a healthy book-to-bill ratio of 1.3x… help drive the stock price higher.

Let’s take a look at the recent boom-to-bust cycles in the chip sector over past 25 years and see if we can determine based on the historical profit valuation metric the price-to-book ratio (P/B) if the sector has truly bottomed out and is poised for a rebound. The P/B ratio is a good metric to judge the valuation of a chip company, as many larger semiconductor companies have built solid balance sheets, with strong cash positions and little or no debt, providing resilience amid an economic slowdown and higher borrowing costs.

The price-to-book value is the ratio of the market value of a company's shares (share price) over its book value of equity. The book value of equity, in turn, is the value of a company's assets expressed on the balance sheet or simply stated the company’s net worth or retained earnings. The value of looking at price-to-book value vs price-to-earnings or price-to-sales ratios is that P/B smooths out some of the extreme highs and lows of chip company’s sales and profits over the course of an up-or-down cycle. No doubt an analyst would not expect a chip stock to trade below book-value-per share or liquidation value as key buy or sell metric. A semiconductor company’s success is driven by its R&D capability to produce new innovative chips that will drive future profit growth, not the liquidation value of machinery or capital equipment.

If we take a look and evaluate the SOX Index’s price-to-book value over the course of the past 25 years we see some interesting relationships (see SOX P/B chart). Right before the “Dotcom Bust” in June 2000 the SOX traded at a P/B ratio of approximately 8.8x. After the “Dot.com Bust” and the Sept. 11 terrorist attack on the World Trade Center, the P/B ratio fell 80% to approximately 1.5x in September 2002. This corresponded with a hard landing or deep recession that severely dented chip companies profits. The SOX slowly improved, but never reached its former grandeur, recovering to about 4.7x P/B ratio in December 2003. After some minor bumps and bruises the SOX dropped again during the “2008 Financial Crisis” to bottom out in November 2008 at 1.25x P/B ratio.

Again, this downturn corresponded with another deep recession that took the S&P 500 Index five-and-half years to fully recover. After that “2008 Financial Crisis” recession the SOX experienced a 13-year secular bull trend with minor blips in December 2018 during the “China/US Trade Sanctions” (P/B fell to 3.5x) and the “Covid-19 Pandemic” on March 2020 (P/B fell to 4x). The P/B ratio for the SOX finally regained its former 2000 high of 8x in April 2021 and then begin to contract, foreshadowing the pull-back in the SOX Index that peaked in January of 2022 and has fallen 30% since.

So, what does this all mean for investors in the chip sector trying to gauge if the group has truly bottomed out? First of all, a clear indicator that the SOX or chip sector is vastly overvalued and poised for a correction is if the P/B ratio trades in the 8x’s range. In both instances that P/B traded in the 8x’s handle, that occurred in 2000 (peak P/B 8.8x) and 2021 (peak P/B 8.2x), the SOX Index eventually dropped precipitously. Secondly, during a mild recession or minor disruption the P/B ratio for the SOX will trade as low as 3.5x to 4x.

This occurred during the minor set-backs like the “China/US Trade Sanctions” (P/B fell to 3.5x) and the “Covid-19 Pandemic” (P/B fell to 4x). Thirdly, during periods of a severe recession (or a hard landing) the P/B ratio for the SOX can trade as low as 1.25x-1.5x. This occurred after the “Dot.com Bust” and the Sept. 11 terrorist attack on the World Trade Center (P/B fell to 1.5x). And again occurred in the aftermath of the “2008 Financial Crisis” (P/B fell to 1.25x).

So, with the current P/B ratio of the SOX Index at 4.5x… what conclusions can we make? If you feel that the economy is poised to undergo a deep recession (or a hard landing) like the ones in 2002/2008 then you would expect the P/B ratio for the SOX to contract further, leading to a more severe correction. If you believe that the current “Fed Tightening” cycle is in the 8th inning and the economy is headed for a minor recession (or a soft landing) then the P/B in 4x handle is not excessive.

Thus, this current level of 4.5x would indicate that the chip sector appears oversold… possibly poised for a turnaround. Each investor must make his/her own determination. My gut feeling is that the current downturn mirrors the “China/US Trade Sanctions” and the “Covid-19 Pandemic” periods, opposed to the Dot.com Bust”/Sept. 11 Terrorist Attack and the “2008 Financial Crisis” eras.

Daniel Morgan, Senior Portfolio Manager