October 9, 2024
About the Company
Nanya is a small Taiwanese DRAM manufacturer. They produce DRAM on trailing-edge process nodes (licensed from Micron) but are developing their own more advanced, technology. Nanya has about a 1.6% share of the DRAM market, a distant fourth, behind Samsung, Hynix, and Micron. The top three companies control a combined 96.4% of the DRAM market.
I don’t own any shares of Nanya and would not buy any, because they lack the scale and the technology to be competitive in DRAM. I keep an eye on their quarterly releases as a data point on the DRAM market.
Market Commentary and Q1 2024 Performance
Seven quarters have passed since Nanya stopped including the time trend of their gross margin with their earnings. Nanya may be following the Micron playbook of holding back inventory to support pricing. Except in Nanya’s case, they don’t have DDR5 and HBM sales to support their financials in the meantime. Revenue declined by 18% quarter-over-quarter, but pricing rose mid-single digits %. Shipments were down low-twenties percent. Gross margin increased slightly, from 2.9% to 3.2%. Because of the large decline in revenue and significant fixed operating expenses, Nanya’s net loss rose to NT$1.49B from a loss of NT$814M last quarter. The company is yet to have an upturn this cycle and are paying a high financial price for their lagging positions in process and product technology. For the last eleven quarters, Nanya’s ASPs have been up mid-single digits %, up low-teens %, up high-single digits %, up low-single digits %, down high-single digits %, down mid-single digits %, down mid-single digits %, down high-single digits %, down mid-20s%, down low-20s%, down mid-single digits %, down mid-single digits %. This is now four quarters of rising DRAM ASPs for Nanya – one full year. Thus, the bottom of the downturn for Nanya was somewhere between May and July of 2024. Over this one year period of rising ASPs for Nanya, their average pricing has risen around 30%. This is paltry compared to the industry leaders. We are now farther off of the prediction I made two quarters ago. Then I forecasted the rate of ASP increases in Q2, Q3, and Q4 to be up mid-teens %, up 20%, and up high-teens %. Instead, we have seen up low-teens% and up mid-single digits %, so reality is well below what I thought would happen. I believed the non-leading edge and non-AI DRAM markets would become more and more undersupplied as 2024 passed. That has not happened. Rather, the lagging edge DRAM markets are oversupplied and the rate of price increases is waning.
Bit Shipments and Capital Expenditures
Because of the softening market, the company has cut back their capital expenditures for the year 2024. Last quarter they raised it from NT$20B to $NT26B. Now they have reversed course and taken the 2024 total spending back down to NT$20B. Total capital expenditures for the first nine months of the year were NT$12.8B, leaving NT$7.2B to be spent in the fourth quarter. That is a similar level of spending to Q3. They have flattened out expenditures. For the whole year 2024, WFE will account for approximately 50% of the total CapEx. Q3’24 bit shipments decreased low-twenties % sequentially. The company is now forecasting bit shipments for all of 2024 to increase by 10% year-over-year. This is down from what they said last quarter, which was an increase of 20% year-over-year. All of this reduction can’t be coming from lower wafer output because cycle time is two months, thus Nanya is planning to hold back more inventory than they previously thought they would.
Q4’24 Market Outlook
Nanya continued their tradition of almost useless market commentary. Demand for AI applications “remains robust.” This is a memory segment for which Nanya has no products, thus they are not benefitting directly from this demand. Demand for non-AI PCs, mobile applications, and consumer products is slower than expected. Major memory suppliers continue to allocate capacity to HBM, DDR5 and advanced nodes. Inventory for DDR4, LPDDR4 and other non-leading-edge memory products are too high.
- · Server Market: DDR5 demand is strong for AI server applications. No comment on non-AI servers signals demand in this area is not noteworthy.
- · Mobile Market: Their commentary is more pessimistic than last quarter, when they said smartphone sales in China are starting to recover. This quarter they only call out that sales of mid-tier and low-tier smartphones has not improved.
- · PC Market: No comment on no-AI PCs, and AI PC demand is in the initial stage of growth.
- · Consumer Market: Demand remains weak. The company said “regional economic improvement is required for demand recovery.” Nanya always cites the economy when the market is weak and they don’t know what is happening with the balance of supply and demand. They are like my uncle who thinks his glass trinkets business is doing poorly for the last twenty-five years because of “the economy” when it is actually the emergence of internet commerce that is killing him.
Analyst Call
Prepared Remarks
- · There was a charge of NT$475M in the quarter because of the earthquake that occurred in Q2. This is about a third of their loss in the quarter.
- · They are “targeting our wafer input exceeding 15% of total capacity by the end of the year” for 10-nanometer class products.
Analyst Q&A
- · Because of their migration to DDR5 and more advanced technology, they believe they are running their fabs at full capacity. Likely this is at lower total wafer output than two years ago when they were running an older technology mix.
- · The company believes improvement in the health of the DDR4 and older DRAM products depends on how fast the big three DRAM makers can sell down their inventory. Nanya is the tail being wagged here.
- · The CEO didn’t say it outright, but he almost said that their ASPs will be down in Q4, because almost all of what they sell is DDR4 or older products.
- · One analyst said they see news saying that Chinese competitors are flooding the market with DDR3, DDR4 and LPDDR4, suppression prices. If this is true, Nanya will feel the bulk of the squeeze from new DRAM bit growth coming out of the PRC.
- · The CEO (Pei-Ing Lee) doesn’t want analysts to name the Chinese competitor (it is CXMT, as one analyst said.) Lee stated the major impact of the “huge capacity” that CXMT has in DRAM is being felt in the mobile segment.
- · Demand from the US and Japan is pretty good. From China and Europe demand is weak.
- · One analyst quoted a price premium of 30% to 40% for DDR5 over DDR4. Lee agreed with that range.
- · The 15% of DDR5 they quoted is percent of wafers. Once those starts are all the way through their fabs, bit output of DDR5 DRAM will be 25% of total bits.
- · Revenue from China for Nanya has been declining since Q1 of 2024. It has come down from 20-something percent then to a little more than 10% now.
Summary
An upturn has not materialized for Nanya. It looks that this memory cycle will pass for the company without seeing them reach profitability. For Micron investors, the takeaway here is that the non-leading edge DRAM market (non-DDR5, non-HBM) has seen pricing and demand peak and is now heading down. Bit shipments for Nanya were down almost 20% and ASPs only rose mid-single-digits percent. The market has weakened enough that they cut back their capital expenditures for the rest of 2024. They had raised it to NT$26B for the full year when they announced Q2 results three months ago. Now they have taken it back down to NT$20B. Nanya is 100% exposed to the PC, mobile, and consumer segments. Demand in all three of these areas has deteriorated in the last ninety days. Nanya is being squeezed by Chinese entrant CXMT on the low end and they don’t have any high end products left. They are hoping that AI demand holds up so more of the bits in the DRAM Big Three get moved out of PC, mobile and consumer to help bring those markets back into undersupply. Their migration to DDR5 products needs to be fast enough to get some of the price premium from there, before that market becomes oversupplied also. This may be the cycle without an upturn for Nanya, a serious risk to their financial viability over the next two to three years. For Micron investors, it is concerning that non-leading edge, non-HBM markets have weakened in the last three months, rather than sliding further into undersupply.
S. Hughes (cyclical long MU)