Description

Idea

Anritsu is a leading vendor of “Test & Measurement” equipment used predominantly for testing smartphone cell-network connectivity. This is a product upgrade-cycle story that we believe is coming to an end. Anritsu has benefitted from a gold rush in customer demand in the ongoing 5G investment cycle, as participants across the supply chain upgrade old 4G test equipment, which has led to a 5.5x in EPS over the past three years.

We expect Anritsu’s currently elevated earnings to peak in FY 2H21 (ending 3/31/22) alongside patterns established in prior cycles as 1) customers’ equipment upgrade cycle completes and 2) competition intensifies.

Looking out to FY23 (ending 3/31/24), we expect Anritsu to post JPY 75 of EPS, representing ~50% downside to Street. In our base case, we believe Anritsu will trade at 16.0x fwd EPS for a JPY 1,201 stock price at YE FY22 (ending 3/31/23; 33% gross downside). 

Company Overview

Originally dating back to 1895, Anritsu is a leading vendor of telecom-related testing equipment (along with U.S. peer - Keysight Technologies - and German peer - Rohde & Schwarz). It is based in Japan with global operations. Its business is comprised of three segments: Test & Measurement (“T&M”), Product Quality Assurance (PQA), and Other equipment.

Anritsu’s T&M segment is our area of focus and key to the short thesis. Anritsu derived 71% of its revenue and 85% of its EBIT from its Test & Measurement business in FY20, and T&M therefore drives the overall earnings picture. While PQA and Other’s EBIT has been ~stable through the years, T&M’s profitability has been quite volatile through the years:

Test & Measurement Overview

In the core T&M segment, Anritsu mainly develops testing equipment for:

·       Mobile communication (used in R&D and production of smartphones – makes sure smartphones connect properly to wireless networks, e.g., 4G and 5G)

·       Network infrastructure (analyzer for cell-towers/base stations – makes sure telecom radios are performing properly; datacenter connectivity evaluation tools – makes sure optical wires in datacenters are up to spec)

·       Electronics (general purpose measuring equipment for a variety of use-cases)

Within T&M, the mobile business is our focus given it: 1) represents ~59% of T&M sales and a higher fraction of profitability (not disclosed, but we believe 80%+) and 2) is the cyclical subsegment of the group (with network infrastructure and electronics ~ flat over time). 

Thesis

1) Anritsu is enjoying elevated profitability due to a product upgrade cycle that we believe is currently peaking.

2) Anritsu’s share in the current cycle is less than the prior cycle.

3) Testing equipment for R&D purposes are higher margin and we expect will be the first to decline, leading to mix-shift driven margin degradation over the next couple of years.

Thesis Point 1) Anritsu is enjoying elevated profitability due to a product upgrade cycle that we believe is currently peaking

Anritsu’s T&M segment operating margin has gone from 3.4% at the trough of the last T&M cycle in FY17 to 23.7% in FY20. Along with top-line growth, this has led to T&M EBIT 10x’ing over this time-frame. Since FY17, this has led to a 5.5x in overall EPS for the Company and a 2.2x in the stock price. 

We believe this was driven by a product upgrade cycle for its T&M equipment that is now peaking. 

Wireless Test & Measurement Cycles Explained

Each major generation of wireless connectivity standards (e.g. 4G, 5G, etc.) requires upgrades in 1) smartphone hardware and 2) network infrastructure. This new hardware, in turn, requires new Test & Measurement (T&M) equipment for verifying network compatibility at both the R&D and production stages. These sudden capital equipment investment requirements create clear industry boom- busts around major wireless standard changes that last 2-3 years each. 

Below, we show Anritsu’s T&M revenue growth over the past two decades. Growth elevates for ~3 years proceeding the beginning of a new wireless standard upgrade cycle and declines thereafter.

The current 5G cycle is developing similarly. 

Stages of Communication T&M Upgrade Cycle

The wireless-standard upgrade cycle takes place over two main stages of investment: 

R&D 

1) Chipset vendors (e.g. Qualcomm) must first design the new wireless chips (e.g. 4G, 5G “basebands”). 

2) Next, Smartphone OEMs (e.g. Apple) take the next- gen wireless chips, integrate them into prototype smartphones, and test the overall design with their own T&M equipment. 

3) Once the prototype phones are complete, telco carriers (e.g. AT&T) independently test and verify the prototypes with their own T&M equipment. 

Production

4) Once the final smartphone models are developed, manufacturing lines are set up for mass production. Manufacturing companies like Foxconn would need to make new purchases of T&M equipment to support basic wireless testing on the production line. 

When each step is performed for the first time, purchase of T&M equipment supporting the latest wireless standard is required. As T&M equipment have long useful lives (~7 years), these are one-time purchases in nature and create upgrade “cycles” of investment. 

5G Upgrade Cycle Detailed 

For developing smartphones with 5G compatibility, just like prior cycles, three separate R&D steps are taken: A) chipset development, B) smartphone development, and C) carrier acceptance. Finally, test systems are used in the D) production stage. 

We believe 5G chipset R&D peaked in 2020 while smartphone R&D and carrier acceptance are currently peaking. Meanwhile, we believe T&M for production will peak over the next 12 months:

A) We believe Chipset R&D has already peaked

Chipset R&D began in 2H18 and final products shipped to smartphone OEMs for R&D over the ensuing 12-18 months. Therefore, all major chipset makers have already researched and produced a new 5G chip. As such, there should be limited incremental demand for new T&M equipment here. 

We see this playing out in Anritsu’s own commentary: 

·       CY1Q19: “acquired early development demand of the 5G chipsets and devices” 

·       CY3Q19: “Currently, main demand is for the development-related measurement for chipsets and terminals compatible with [5G]” 

·       CY4Q20: “Demand for the development of 5G chipsets ... has been steady” 

We take this to mean chipset demand has peaked. We believe CY20 was the peak for chipset R&D. 

Importantly, the number of 5G chipset manufacturers is limited compared with smartphone manufacturers. Qualcomm, HiSilicon, and Mediatek dominate the market (combined ~3/4ths of the market). 

B) We believe Smartphone R&D is peaking

For smartphone R&D, we surmise that the number of smartphone vendors and models brought to market drives growth for T&M equipment. That is, new 5G vendors provide incremental TAM for Anritsu’s tools. Additionally, existing 5G vendors that increase the number of models brought to market likely also provide incremental TAM (more development required). 

On both accounts, 2021 is shaping up to be a peak year (delta in new 5G models produced in 2021 is lower than 2020; same goes for the delta in new 5G smartphone vendors): 

We believe CY 1H21 was the peak for Smartphone R&D. 

C) We believe Final Carrier Test Is Peaking

We believe final carrier test follows smartphone R&D with only a slight lag (3-6 months). That is, carriers need to be ready to evaluate new 5G smartphone prototypes from the smartphone OEMs. As a flood of new 5G smartphone prototypes hit the market in 2019 and 2020, carriers should have invested in capacity to test final prototypes commensurately. 

As we (a) believe T&M purchases for smartphone R&D peaked in CY 1H21 and (b) expect T&M equipment for production to be peaking in 2H21/1H22, this would argue for a ~6-month lag between smartphone R&D and final carrier-test peak T&M purchases. 

Final Carrier Test related T&M purchases are likely to peak in CY 2H21. 

With chipset and smartphone R&D past their peaks, CY 2H21 is likely to see overall R&D T&M sales step- down (as carrier test is unlikely to offset declining chipset and smartphone R&D related purchases). 

D) We expect 2022 to be the peak for 5G incremental production 

While 5G smartphone units are expected to continue to grow, the rate of change will decline slightly in 2022 and severely in 2023 (151mm units; 42% lower y/y), based on estimates from IDC. This means a smaller amount of new T&M units will be required at manufacturers: 

As a result, CY 2H21/1H22 is likely to be the peak in T&M units delivered to manufacturers. Indeed, as production further grows, more units will be shipped, but the delta is what matters for Anritsu’s revenue. By 2025, new T&M production units required will be 1/4th the peak rate in 2021/2022; Anritsu’s production T&M TAM will similarly be 1/4th the size. While this makes logical sense, it is also rooted in historical cycles. 

Evidence from the 4G Cycle 

We can also look back to the 4G cycle to inform the timing of Anritsu’s peak T&M profitability. Below, we plot total 4G and 5G smartphone shipments (IDC data) vs. Anritsu’s Mobile T&M revenue by calendar year. CY15 marked the peak in year-to-year growth of 4G units produced (425mm more units in CY15 vs. 269mm more units in CY14). As can be seen below, Anritsu’s T&M revenue growth peaked out in CY14 (as customers bought machines in advance of CY15 production): 

CY21 is shaping up similarly as customers pull-in T&M equipment in preparation for the large growth years of CY21/CY22. 

Thesis Point 2) Anritsu’s share in the current cycle is less than the prior cycle

5G T&M Market Share 

There are three main providers of T&M equipment for 5G R&D: Keysight Technologies (U.S.), Anritsu (Japan), and Rohde & Schwarz (“R&S”, Germany). Keysight missed out on the 4G R&D cycle almost completely as it did not have an offering technical enough to fit customers’ needs. Anritsu and R&S, resultantly, enjoyed a ~50/50 split of the 4G R&D market for some time. 

“Anritsu’s R&D tester [market share] remained over 50%” – JPM initiation from May 2012. 

Keysight has since rectified their technical gap, in part through acquisition, and has emerged as the leader in 5G R&D T&M equipment. 

“We have strong competitors in the industry in total, Rohde & Schwarz and Anritsu are 2 companies that come to mind. But we did get out to a very early lead” – Keysight at Wells Fargo Conference 12/19. 

Anritsu is going from effectively a two-player market in the past R&D T&M upgrade cycle to three players in 5G R&D. This should, all else equal, limit Anritsu’s peak earnings power and lead to faster margin degradation. 

Keysight Is Accelerating Its Lead 

Thesis Point 3) Testing equipment for R&D purposes are higher margin and should be the first to decline, leading to mix-shift driven margin degradation over the next couple of years.

T&M Production Margins vs. R&D 

Finally, as the R&D upgrade cycle completes and Anritsu’s business mix-shifts to production T&M equipment, we expect to see margin degradation. 

R&D T&M Equipment 

•  Complex and occasionally custom designs. 

• Carry high ASPs ($500k+) and lucrative gross margins.

• Speed to market is imperative. 

• Three key players: Anritsu, Keysight, R&S. 

Production T&M Equipment 

• Standardized boxes prone to commodification. 

• Carry ~$50k ASPs and lower gross margins. 

• Speed to market less of a concern. 

• Numerous players, of which we believe Anritsu will hold ~20% share. 

While we do not have the exact breakdown in margins between R&D and Production units, past commentary & financial performance informs us that R&D units are meaningfully higher. 

4G Cycle

·       Shared Research (equity research) from 2010: “Early-stage equipment is most expensive with highest profit margin (20%).” 

·       Anritsu CY4Q11 Q&A: “The rise in operating margin reflects better product mix ... demand for high value- added measuring instruments associated with the necessary systems for LTE development.” 

·       Anritsu CY1Q15 Q&A: “Falling prices in the mobile manufacturing market have impeded growth.” 

·       While Anritsu’s T&M revenue continued to grow until CY2Q14, LTM T&M EBIT began to decline significantly in CY2Q13, which we attribute to the mix-shift towards production: 

5G Cycle 

Anritsu CY2Q19 Q&A: Operating margin improved sequentially “due to differences in product mix.” We know that this “mix” aligns with increased sales of 5G chipset and smartphone R&D T&M equipment (and not production, as most 5G production ramped in late 2020). 

Anritsu CY1Q20 Q&A: “main reason for the increase in profit margin in the T&M business was a change in the product mix. Specifically, an increase in the ratio of sales of products for 5G boosted the gross profit margin.” Again, the majority of this is likely R&D related (the first 5G iPhone was not released until 4Q20). 

Putting This All Together 

Putting this all together, we believe Anritsu has just crossed the peak T&M investment phase for customer R&D purposes, while production will peak out over the next 12 months.  Mapping the current cycle vs. 4G shows a very similar story – T&M LTM EBIT declined on a y/y basis after R&D peaked in CY 2Q13, and we think the same happened in CY 2Q21: 

Valuation 

In our base case projection, we have Mobile T&M revenue declining 27% in FY23 vs. FY20 (what we think will be the peak year). This results in a 14% overall decline in T&M sales b/w FY23 and FY20, but 42% lower T&M operating profit given op leverage / mix-shift. 

This compares to a 33% decline in Mobile T&M revenue in the past cycle (FY16 vs. FY13), 22% overall T&M sales decline, and an 84% T&M operating profit decline: 

Payoff Structure

Key drivers to the model are simple: mobile T&M revenue growth and T&M incremental EBIT margin. 

Mobile T&M rev growth: 

• Bear: -13% in FY21, -15% in FY22 & FY23 

• Base: -10% in FY21, FY22, and FY23 

• Bull: 7% growth in FY21 & FY22, 5% in FY23 

Incremental T&M EBIT margin: 

We assume 60% incremental T&M EBIT margin for the rest of FY21, 65% in FY22, and 70% in FY23 across all cases. 

EPS: 

These inputs result in FY23 EPS of JPY 57.9, JPY 75.1, and JPY 152.8 in the bear, base, and bull cases, respectively. 

Multiple: 

Testing equipment peers like Teradyne and Viavi, that we view as less cyclical than Anritsu, trade for ~mid-teens CY22 EPS. 

Putting it all together: 

In our base case, we value Anritsu at 16x fwd FY23 EPS for a JPY 1,201 stock price at YE FY22 (33% downside. Bear case contemplates 14x fwd EPS; on the lower EPS estimate this is -55%. Bull case contemplates 18x fwd EPS (what would be a peak EPS number), for a +54% return. Blended prob weighted return is down 26% through YE FY22.  

Key Risks

• Extended procurement of T&M equipment in the 5G cycle.

• Elevated customer orders for smartphone T&M equipment could sustain due to a tiered introduction of 5G capabilities in smartphones (most carriers only support sub-6 Ghz spectrum in the 5G early years). 

• Accordingly, an upgrade of equipment to mmWave could create additional revenue for Anritsu. 

• Mitigant: We believe that the bulk of 5G R&D equipment installed already has mmWave testing capabilities. Anritsu’s key T&M equipment already supports mmWave (only exception would be modular equipment where the customer specifically chose not to take mmWave equipment). 

• Mitigant: Upgrades to modular equipment would be at a small fraction of the original cost (perhaps at most 25%), therefore providing little uplift. 

• Mitigant: Any follow-on T&M equipment revenue would merely sustain rather than further boost Anritsu’s peak earnings. 

• 5G T&M is used in new end-markets in a major way outside of smartphones. 

• 5G connected autos is seemingly the largest potential here.

• Mitigant:

• Anritsu is not projecting a significant amount of a contribution here. In an old deck, Anritsu included Auto & IoT as a small contribution to its mid-20’s operating target (by eye this looks like sub 10% of sales). 

• The total auto market will be much smaller. 1.5bn smartphones are produced annually vs. ~100mm motor-vehicle units (much of which would not be high-end enough to support 5G).

The author of this post is a private fund manager registered with the Securities and Exchange Commission as an investment adviser. At the time of its publication, no fund or account managed by the author held any position, whether long or short, in Anritsu Corporation.  However, any such fund or account may (or may not), at any time, without making any public or other disclosure or giving notice to any party, except as may be required by applicable law, buy and sell securities of, and hold a long or short position in, Anritsu Corporation (and other companies mentioned in this post). The information set forth in this post does not constitute a recommendation to buy or sell any security, or legal, tax, investment or other advice. This post represents the opinion of the author as of the date of this post. This post contains certain “forward-looking statements,” which may be identified by the use of such words as “believe,” “expect,” “anticipate,” “should,” “planned,” “estimated,” “potential,” “outlook,” “forecast,” “plan” and other similar terms. All are subject to various factors, any or all of which could cause actual events to differ materially from projected events. This post is based upon information reasonably available to the author and obtained from sources the author believes to be reliable; however, such information and sources cannot be guaranteed as to their accuracy or completeness. The author makes no representation as to the accuracy or completeness of the information set forth in this post and undertakes no duty to update its contents.

Catalyst

• cycle turning
• mix shift towards produciton units which hold lower margin