MU - Micron Technology Inc.

[ValuePadawan]

As long as the only companies making DRAM are Samsung, SK Hynix and Micron I have no doubt they will in time behave rationally.

Lol RIP proves me right for having some hubris https://www.telecomlead.com/telecom-chips/korea-approves-sk-hynixs-106-bn-semiconductor-project-99388

Seems like an awful lot of supply coming online. Will be interesting to hear what Micron says about this on their earnings call.

[nwoodman]

Great quarter and the tailwinds seem to be growing.   MS maintains  their target of $105, this is roughly where I have them pegged too.  That GM forecast of 40+% is pretty amazing though.

 

MS take:

We continue to see relative value in the stock: Micron's trough earnings power surprised us, and should bode well for longer term value creation. In the last few quarters, we have been surprised at the resiliency in Micron's gross margins, given the sharp compression in PC and server DRAM. The company's success in graphics DRAM, at improving the mobile mix towards DRAM+NAND modules, and in embedded DRAM, has driven this outcome. As a result, despite numerous substantial challenges, the company was able to earn about $3 in EPS in a difficult FY20 and FY21. As conditions now look to be bottoming out, we would characterize that as trough earnings. While we're not sure we will see EPS back at new highs of $12 plus, if we enter into a phase of improvement, that is where we expect sentiment will quickly go, leading to through cycle earnings potential of $7.40 or so. Given the rerating in the semiconductor space over the last 24 months, we see substantial relative value here. Our price target remains 14.5x that through cycle number (a discount to the group given higher cyclicality and lower free cash flow yield); that leaves our target at $105, and we remain OW the stock.

MICRON_20210401_0000.pdf

[Pelagic]

Isn't this the downside of stock buybacks, especially when it comes to cyclical businesses? When they end up actually buying back stock is when there is +FCF and the stock price is high. Yet in the downturn of the cycle when the stock price is low they hold on to their excess cash and don't buyback stock. I think I'd rather just receive a dividend and spend it as I see fit, then again I hold shares in a tax deferred account.

 

Quote

...so our current obviously approach towards returning cash to shareholders is through the buyback. We mentioned in the second fiscal quarter, we didn't buy back stock nor did we buy back stock in the first quarter and that was really a function of the fact that cash flow was negative and we were protecting our cash -- our net cash position.

But as you say -- suggest and as we talked about we do expect to have good cash flow in the back half of the fiscal year and we do expect this business to generate good free cash flow over time.

So, I think primarily, you can look to us to buy back stock. We're going to return at least 50% of our free cash flow in the form of buybacks and we have in the past returned more than that. So, potentially, we could do that in the future.

 

[mwtorock]

it is true at least to some extent. the timing of buybacks depend a lot on cashflow. they need to spend on CapEx or risk dropping behind in terms of technology. the hope is that management stay cool when share price is high and do not spend all the cash on buybacks. Waiting patiently though is not easy. 

[Tintin]

I've been going deep on MU over the past few weeks, and am now trying to construct a suitable framework to estimate their Owner Earnings - which isn't as easy as it looks.  The way I see things, semiconductor manufacturing is an interesting business in that virtually all capital expenditure has to be classified as 'maintenance' rather than 'growth'.  Although companies are continuously spending huge sums on 'new' product and process technologies in order to lower per gigabit production costs, do they really have a choice in the matter?  I would say that they don't.  If they don't continuously innovate by creating better products at lower costs, they get left behind sitting on a bunch of inventory that nobody wants next year.  So if innovations in 'new' production processes are a fundamental prerequisite to simply maintain their existing sales volumes, how can we view such CapEx as a form of 'growth'?

 

I have therefore come to the conclusion that for a business such as Micron, their free cash flows and owner earnings are essentially the same number.  Is this a fair assessment or have I missed something obvious?

[TheThinkingInvestor]

On 4/2/2021 at 6:20 AM, Tintin said:

I've been going deep on MU over the past few weeks, and am now trying to construct a suitable framework to estimate their Owner Earnings - which isn't as easy as it looks.  The way I see things, semiconductor manufacturing is an interesting business in that virtually all capital expenditure has to be classified as 'maintenance' rather than 'growth'.  Although companies are continuously spending huge sums on 'new' product and process technologies in order to lower per gigabit production costs, do they really have a choice in the matter?  I would say that they don't.  If they don't continuously innovate by creating better products at lower costs, they get left behind sitting on a bunch of inventory that nobody wants next year.  So if innovations in 'new' production processes are a fundamental prerequisite to simply maintain their existing sales volumes, how can we view such CapEx as a form of 'growth'?

 

I have therefore come to the conclusion that for a business such as Micron, their free cash flows and owner earnings are essentially the same number.  Is this a fair assessment or have I missed something obvious?

 I think you’re completey correct. It’s a highly capital Intensive business. So I think it’s fair to assume maintenance CapEX is equal to growth CapEX. As long as they don’t increase CapEX to a high degree that ASP’s come down they can maintain margins with should keep them flush with cash to increase shareholder value.

 

 I’m still a little confused how they plan on returning that cash tho. With EPS and likely the share prices going up, buybacks don’t really seem to be the best use of excess cash. 

[Tintin]

8 hours ago, TheThinkingInvestor said:

 I think you’re completey correct. It’s a highly capital Intensive business. So I think it’s fair to assume maintenance CapEX is equal to growth CapEX. As long as they don’t increase CapEX to a high degree that ASP’s come down they can maintain margins with should keep them flush with cash to increase shareholder value.

 

 I’m still a little confused how they plan on returning that cash tho. With EPS and likely the share prices going up, buybacks don’t really seem to be the best use of excess cash. 

Thanks for the reply!  I know Bruce Greenwald has an interesting take on calculating the amount of 'growth' CapEx within a company, but for a business like semiconductor manufacturing that framework just doesn't appear to work.

 

As for MU returning value to shareholders... given that they won't be paying dividends any time soon, paying off debt is probably their most likely course of action in 2021 as they generate more FCF on the upside of the cycle.

I wouldn't totally write off the prospect of further share buybacks from happening in 2021, but that would be dependent upon a macroeconomic or geopolitical event driving the price down, e.g. a severe escalation in tensions between China and Taiwan might create a buying opportunity.

[Spekulatius]

MU really should pay special dividends from windfall profits at the cyclical top. I believe this will continue to be a cyclical industry , as me work is basically a commodity. The Chinese competitors that are emerging will make sure or that.

 

Buyback at or close to cyclical tops don’t make much sense and at the bottom there are no cash flows to pay for buybacks. That the corundum with cyclical business and I think MU is in the same category.

[ValuePadawan]

https://semiwiki.com/semiconductor-services/297904-spie-2021-applied-materials-dram-scaling/ Good overview of the possibilities and problems with scaling DRAM chips in the next few years.