LRCX - Lam Research: Going From Bad To Worse

Summary

• We already knew the memory capex turnaround was a mirage.
• Now we know it's even worse, as other segments are following memory.
• And to add insult to injury, current revenues and earnings are being propped up by a one-time effect that's set to disappear soon.

In my previous Lam Research ( LRCX ) article I pointed out that a quick memory capex turnaround wasn't anywhere to be seen , unlike the market was expecting. And in a previous Applied Materials ( AMAT ) article, I explained that weakness would spread from memory to other segments. The AMAT article is also useful to understand why semiconductor equipment companies are so much more cyclical that semiconductor producing companies -- which would argue for them carrying lower valuation multiples in general.

Lam Research recently reported Q2 FY2023 earnings , and the main takeaway from this report was the significant downward guidance for Q3 FY2023.

The new guidance included a 12% cut in revenues versus market consensus, as well as a 16% cut in EPS versus market consensus. These are, by themselves, quite large cuts. Moreover, the guidance was deep enough that it already took consensus to a negative revenue growth rate in Q3 FY2023, and a -25% revenue growth rate in Q4 FY2023.

As deep as the cuts for Q3 FY2023 were, things are still significantly worse than they look. This is so because since Q2 FY2023, has been relying on steep drops in deferred revenue to feed its reported revenue and EPS numbers. Deferred revenues result from sales which have already been invoiced, but not yet booked. For a while LRCX has been suffering from supply chain disruptions which led to delayed and incomplete deliveries. As these normalized, deferred revenues have started coming down to a “normal”, “baseline” level.

When deferred revenues drop, they work like a one-time boost to revenues and earnings. That boost, as per LRCX, will be mostly over in Q3 FY2023 – even with the negative guidance. Come Q4 FY2023, there won’t be any more such boost, and thus revenues and EPS will finally reflect truer ongoing sales.

And what do these ongoing sales imply, according to the current consensus (reached after the recent earnings call)? They imply revenues down as much as 31% year-on-year (Q1 FY2024), and EPS down as much as 44% (also Q1 FY2024). These were not numbers discounted a few months ago – over the last 6 months, the cuts to Q1 FY2024 have amounted to a 27% reduction in revenues, and a 33% drop in earnings.

The markets, 6 months ago, as per my previous AMAT article, weren’t even yet discounting a drop in sales into the memory segment. Here’s the relevant quote from AMAT’s own earnings conference call 6 months ago (bold is mine):

So I think the way we would think about it is we're a few quarters behind what raw demand is, and that's probably what's happening across the industry. When we think about the sustainability of the demand, we study all the end markets, as you've described, all the different ICAPS markets, the edge devices, the leading-edge logic area, et cetera, including the memory components. And we see – we see a consistent growth across the horizon from a demand perspective .

6 months ago, LRCX traded more or less where it trades now. At the time the story was that all segments were growing. Recently (before the latest earnings call), the story was that weakness was mostly concentrated in the memory segment. But, of course and as predicted, during the latest call things were already spreading (bold is mine):

As we look forward into 2023, however, we see a substantially weaker demand environment and the corresponding need to make prudent changes to our near-term operations and priorities. Customers across all segments are exercising caution , especially those in the memory markets. Inventory levels in both NAND and DRAM remain very high, and customers are not only reducing new capacity additions, but also lowering fab utilization levels to bring excess inventory into balance as quickly as possible. In addition, the U.S. government’s new restrictions on sales of equipment, parts and services for specific technologies and customers in China are further impacting equipment demand in a declining market.

So, what do we in fact have here? We have a market that 6 months ago didn’t see weakness in any segment. Then there was massive weakness in memory, but not other segments. And now, in the latest earnings call, weakness has spread to all segments. One would think that the market’s reaction would be to get more negative since the spreading weakness had not been foretold. Was that what happened? Not quite. Instead, what happened was that analysts immediately started looking for a bottom! Take the following question from the earnings call, it reflects this general “looking for a bottom” mentality:

Vivek Arya

Thanks for taking my question. I am trying to gauge what is kind of your trough quarter of this year conceptually, right? I understand that you don’t give exact forecast. But if I go through this deferred revenue math, so let’s say another $500 million comes out suggest that your March shipped revenue conceptually is about $3.3 billion. Does that reflect all the China and memory CapEx cuts so that is sort of pure trough revenue quarter, or do you think there is more to come? So, the trough revenue quarter might be later this year, closer to $3 billion or some other number. I am just trying to conceptually gauge what is the trough quarter for this year, so we can get a sense for what trough earnings power could be.

Doug Bettinger

The best I can do is just say what I have already said. Revenue is somewhat first half weighted, largely because we are pulling the deferred down in March. In March, it’s pulled down to where it’s going to be. And so you got to kind of think about that plus the fact that I told you, we think WFE is fairly balanced first half, second half. And I think if you think that through, you will get it pretty close to where it should be.

This isn't a discounting of the worse, by any measure.

Meanwhile, what is reality?

Reality is that the memory segment weakened first, and the impact we’re seeing now is mostly from the memory segment. But segments other than memory have weakened as well (as per the earnings call), and that weakness will only be truly seen 2 quarters into the future.

Reality is also that impacts from customers' intentions to commit to lower capex now, affect the year in a progressive way looking forward. This is easy to understand – near-term commitments are most often harder to cut than longer-term commitments.

Are There Risks To Being Bearish?

There are always risks to any opinion we might have on any stock. In being bearish, one could at the very least miss upside in this particular stock.

However, we should consider that I'm bearish because this particular industry is facing an extreme slowdown the market didn't foresee (in terms of consensus) 6 months ago. Yet, the stock has barely corrected from what were highs attained during what was (at least for other tech sectors) an obvious bubble. And mind you, most sectors are not as cyclical at manufacturing semiconductor equipment.

The bullish thesis will have you believe that after this cyclical valley, again semiconductor equipment will flourish due to the need for more capacity and new, smaller, processes. And that semiconductor intensity is increasing in many different products (autos are often mentioned).

That's all fine and dandy, but certainly investors should be aware that the valley is getting deeper and deeper, and earnings are going to implode by 50% or more before the rosy days return.

It is my (bearish) view, that investors aren't all that prepared to hold a cyclical stock at 14.4x peak earnings through such a cyclical valley. For sure, in any other cyclical sector, peak multiples are nearly always contracted to 3-4-5-7x P/Es. Just look at housing stocks for an example (and even there, the worst isn't priced in yet).

Conclusion

Over the coming quarters, I believe LRCX will report deeper and deeper revenue and earnings contractions. This is implied both by end-demand weakness and capex intention cuts, as well as the mechanical effect of exhausting deferred revenues.

While now analysts are already looking for a bottom, and semiconductor equipment stocks have been rebounding as result, I expect that this optimism won’t last. Reduced utilization rates mean that even a rebound in semiconductor demand won’t immediately translate into a rebound in semiconductor equipment orders (needed to increase production capacity). At the same time, the weakening of ex-memory segments hasn’t yet run through either the P&L statements or analysts' forward-looking estimates.

All of this means that the bottom for LRCX isn’t here or anywhere close to here. Cyclical stocks like LRCX don’t bottom at 14.5x peak earnings (LRCX FY2023 consensus). They bottom at single-digit peak EPS consensus.

What can you do through this? If you're long, you should be aware of the problem. You could sell or you could sell covered calls, to not realize your profits. If you're neutral, you could use selling short LRCX as a means to reduce net long exposure elsewhere. If you're short already, you just need to be calm, as the present rally is likely unsustainable.

For further details see: