Senior Technology Analyst, Japanese Equities, LightStream Research

Scott Foster

Senior Technology Analyst, Japanese Equities

Long-Short Analysis on Japan, Asia and the World

LightStream Research

LightStream conducts quantitatively rigorous and data-driven analysis with a focus on Japan and Asia.

We consistently cover the Tech, Consumer and Industrial sectors, and provide ECM and Events coverage more broadly.

We supplement traditional qualitative insights from company relationships with a robust quantitative foundation that examines as much historical data as is available.
We further refine the accuracy of our insights and calls through the use of alternative data to flesh out current trends, and through a valuation framework that emphasises non-superficial peer group selection and a proper attempt to understand the metrics and changes that have driven security valuations in the past.

Equity Bottom-Up

<ul>
<li>Sales of semiconductor equipment are running ahead of guidance. Management now sees a considerably stronger 2H and 19% operating profit growth in FY Dec-22 as a whole.</li>
<li>In the longer term, Japan - U.S. semiconductor cooperation should benefit Horiba and other companies in the Japanese semiconductor industry.</li>
<li>Selling at 10.5x new EPS guidance for FY Dec-22. Not meaningfully cheap if a down-cycle is coming, but attractive if it is not imminent and security concerns support demand.</li>
</ul>


Sales of semiconductor equipment are running ahead of guidance. Management now sees a considerably stronger 2H and 19% operating profit growth in FY Dec-22 as a whole.
In the longer term, Japan - U.S. semiconductor cooperation should benefit Horiba and other companies in the Japanese semiconductor industry.
Selling at 10.5x new EPS guidance for FY Dec-22. Not meaningfully cheap if a down-cycle is coming, but attractive if it is not imminent and security concerns support demand.


Horiba (6856 JP): Semiconductor Division Leads Upward Revision for FY Dec-22

<ul>
<li>Measurement and analysis to qualify hydrogen fuel cells, vehicle engines and fueling stations, electrolysis and other hydrogen production processes should be a new growth driver for Horiba.</li>
<li>Semiconductor equipment demand losing momentum, but should hold up in 2022. Weak demand for auto emission measurement systems is a key risk.</li>
<li>Reasonably valued at 15x EPS guidance for FY Dec-21 and 1.45x book value, with a dividend yield of 2.0%.</li>
</ul>


Measurement and analysis to qualify hydrogen fuel cells, vehicle engines and fueling stations, electrolysis and other hydrogen production processes should be a new growth driver for Horiba.
Semiconductor equipment demand losing momentum, but should hold up in 2022. Weak demand for auto emission measurement systems is a key risk.
Reasonably valued at 15x EPS guidance for FY Dec-21 and 1.45x book value, with a dividend yield of 2.0%.


Horiba (6856 JP): New Growth Opportunity in Hydrogen Energy

Pan-Asia Catalysts/Events, Quiddity Advisors

Travis Lundy

Pan-Asia Catalysts/Events

Quiddity Advisors

Quantitative Analysis

Long-time Japan strategist Jonathan Allum, most recently of SMBC Nikko, put a guest post titled <a href="https://www.ft.com/content/a97cf336-518d-45dd-8092-ec0363c18725" target="_blank" rel="noopener">Takeaways from 30 years of Covering the Japanese Market</a> in the Financial Times on 26 November (Thanksgiving) which is well worth a read.He points out that foreign investors in 2020 are likely going to be net dis-investors in Japan for the fourth year in five, fifth year in six, and increasing the net selldown which has occurred since the monster year in 2013. <section class="figure"><section class="figure"><section class="figure"><figure class="editor-figure mceNonEditable"><img src="https://static-prod.smartkarma.com/images/media/000/174/956/original/Screenshot%202020-11-29%20at%203.07.23%20PM.png?1606633655" class="editor-image img-responsive" data-smartkarma-image-id="174956"></figure></section>This data is compiled from the weekly data which shows each investor category. On a net basis across all investor categories, it will be flat because while it shows buybacks on market, crediting the "Business Corporations" category, it does not show the data of investors selling into Tender Offers or the data of investors buying from IPOs.</section>Of course, the data is not that clean because investing doesn't always happen according to calendar year decisions. 
<section class="figure"><figure class="editor-figure mceNonEditable"><img src="https://static-prod.smartkarma.com/images/media/000/174/969/original/Screenshot%202020-11-29%20at%203.40.57%20PM.png?1606635676" class="editor-image img-responsive" data-smartkarma-image-id="174969"></figure></section>It just so happens that the last two really big selloffs in dollar terms (both exceeding ¥6trln on a rolling 52wk basis) did not happen to see start and end points coincide with those of the Gregorian calendar.
There has been an interesting confluence of "lack of participation in the market" this year. After lifting ETF buys in March 2020, the BOJ has slowed down its ETF buying considerably in the past six months (the current 6-month total is near the lower end of the range seen in the past four years), buybacks are down this year as companies slowed their activity in the face of covid-19-induced uncertainty, and foreigners are net sellers. 
Of course Japanese insurers are on the verge of extending what will be a 12-year streak of net sells to a 13th year, and it looks like Japanese banks will be extending their net selling to a 14th year in a row. Trust Banks will be net buyers for the sixth year in seven this year, but for at least the fourth year in a row it will be less than the net buying of ETFs by the BOJ so ex-BOJ trust banks will have been net sellers for several years. 
Business corporations are up ¥1.335trln year-to-date through the 20th of November, and will almost certainly extend their streak to ten years on the buy side. The interesting net flow this year is the "Proprietary" flow at brokers. This is now positive for ten years in a row, and there are particular reasons for this, as discussed below.
Because of the noise around recent articles about the BOJ and foreign activity, I decided to once again dust off the long-term analytics on Japan investor class flows, and present them here. Because flows net off to close to zero (i.e. there are a certain number of shares, and if some investor class is a net seller, other investor classes must be net buyer(s)), deriving a model for behavior is useful.  Long ago, in roles in fixed income and equity derivatives strategy, I built models of net foreign buying, and they uniformly showed that foreign investors chase/push/drive performance. My final model is a multi-input model which provides a simple combination of two factors they chase.
This insight discusses those behavior models, investor class flow dynamics, cycles, and trends and pressures near-term.
The combination of the investor class flows, and the model which drives net foreign activity, is interesting. I think it helps because the top-down aspect matters for understanding allocation flow.
Read on for the nitty gritty (as usual, we start with Conclusions First).</section>

Long-time Japan strategist Jonathan Allum, most recently of SMBC Nikko, put a guest post titled Takeaways from 30 years of Covering the Japanese Market in the Financial Times on 26 November (Thanksgiving) which is well worth a read.He points out that foreign investors in 2020 are likely going to be net dis-investors in Japan for the fourth year in five, fifth year in six, and increasing the net selldown which has occurred since the monster year in 2013. This data is compiled from the weekly data which shows each investor category. On a net basis across all investor categories, it will be flat because while it shows buybacks on market, crediting the "Business Corporations" category, it does not show the data of investors selling into Tender Offers or the data of investors buying from IPOs.Of course, the data is not that clean because investing doesn't always happen according to calendar year decisions. 
It just so happens that the last two really big selloffs in dollar terms (both exceeding ¥6trln on a rolling 52wk basis) did not happen to see start and end points coincide with those of the Gregorian calendar.
There has been an interesting confluence of "lack of participation in the market" this year. After lifting ETF buys in March 2020, the BOJ has slowed down its ETF buying considerably in the past six months (the current 6-month total is near the lower end of the range seen in the past four years), buybacks are down this year as companies slowed their activity in the face of covid-19-induced uncertainty, and foreigners are net sellers. 
Of course Japanese insurers are on the verge of extending what will be a 12-year streak of net sells to a 13th year, and it looks like Japanese banks will be extending their net selling to a 14th year in a row. Trust Banks will be net buyers for the sixth year in seven this year, but for at least the fourth year in a row it will be less than the net buying of ETFs by the BOJ so ex-BOJ trust banks will have been net sellers for several years. 
Business corporations are up ¥1.335trln year-to-date through the 20th of November, and will almost certainly extend their streak to ten years on the buy side. The interesting net flow this year is the "Proprietary" flow at brokers. This is now positive for ten years in a row, and there are particular reasons for this, as discussed below.
Because of the noise around recent articles about the BOJ and foreign activity, I decided to once again dust off the long-term analytics on Japan investor class flows, and present them here. Because flows net off to close to zero (i.e. there are a certain number of shares, and if some investor class is a net seller, other investor classes must be net buyer(s)), deriving a model for behavior is useful.  Long ago, in roles in fixed income and equity derivatives strategy, I built models of net foreign buying, and they uniformly showed that foreign investors chase/push/drive performance. My final model is a multi-input model which provides a simple combination of two factors they chase.
This insight discusses those behavior models, investor class flow dynamics, cycles, and trends and pressures near-term.
The combination of the investor class flows, and the model which drives net foreign activity, is interesting. I think it helps because the top-down aspect matters for understanding allocation flow.
Read on for the nitty gritty (as usual, we start with Conclusions First).

JAPAN FLOW:  Foreign Investor Flows Still Matter (Redux, Reduxed) - Playing Catchup

ja research

ja research is an independent research platform focused exclusively on Japanese equities. Leveraging data and graphic analysis provided by Japan Analytics, we provide insights on small and medium-sized companies that have been ignored by the sell-side as well as challenging conventional wisdom on larger companies. We also offer analyst and corporate access, the latter with translation if required. 

As expected, 1H results fell short of expectations and FY Dec-19 guidance has been cut. But the new full-year guidance, which implies a 10% year-on-year increase in sales and a 13% increase in operating profit in 2H, may still be a stretch.Both of the company's two major profit drivers, Automotive and Semiconductor, reported large year-on-year profit declines in 1H. But despite downward revisions and a weakening economy, management is still guiding for higher sales and profits in 2H. In the case of the Automotive sector, this can be explained by seasonality and project timing. In the case of Semiconductor equipment, management cites an increase in new orders in 2Q while admitting to limited visibility.On the positive side, quarterly data suggests that sales growth and operating profit are bottoming out. Our operating profit forecast for FY Dec-19 is below management's, but not by much, and we expect recovery in 2020. The Automotive business should continue to grow with the introduction of more severe emission regulations in China and India. Demand for semiconductor equipment should slowly improve.At ¥5,410 (Monday, September 2, closing price), the shares are selling at 13.0x our EPS estimate for this year, 12.0x our estimate for next year and 10.6x our estimate for FY Dec-21. These and other projected valuations are close to the bottom of their 5-year ranges, leaving a trade-war induced global recession as the primary risk for investors.

As expected, 1H results fell short of expectations and FY Dec-19 guidance has been cut. But the new full-year guidance, which implies a 10% year-on-year increase in sales and a 13% increase in operating profit in 2H, may still be a stretch.Both of the company's two major profit drivers, Automotive and Semiconductor, reported large year-on-year profit declines in 1H. But despite downward revisions and a weakening economy, management is still guiding for higher sales and profits in 2H. In the case of the Automotive sector, this can be explained by seasonality and project timing. In the case of Semiconductor equipment, management cites an increase in new orders in 2Q while admitting to limited visibility.On the positive side, quarterly data suggests that sales growth and operating profit are bottoming out. Our operating profit forecast for FY Dec-19 is below management's, but not by much, and we expect recovery in 2020. The Automotive business should continue to grow with the introduction of more severe emission regulations in China and India. Demand for semiconductor equipment should slowly improve.At ¥5,410 (Monday, September 2, closing price), the shares are selling at 13.0x our EPS estimate for this year, 12.0x our estimate for next year and 10.6x our estimate for FY Dec-21. These and other projected valuations are close to the bottom of their 5-year ranges, leaving a trade-war induced global recession as the primary risk for investors.

Horiba (6856 JP): Slow Recovery Ahead, Valuations Low

Weak 1Q results, declining demand from the semiconductor industry, renewed pessimism over the U.S. - China trade war and, apparently, profit taking have caused Horiba's share price to drop 27% from its April 19 peak of ¥6,930. At ¥5,040 (Thursday, June 6, closing price), the shares are now close to the bottom of their 5-year valuation ranges, selling at 10.6x management's EPS guidance for FY Dec-19, 11.5x our EPS estimate for this year and 10.4x our estimate for FY Dec-20. Year-on-year comparisons for the three months show total consolidated sales down 14.0%, operating profit down 51.8% and net profit down 58.1%. Automotive sales were down 10.3% and operating profit down 65.3%. Semiconductor equipment sales were down 29.9% and operating profit on those sales down 49.5%. The company has stopped releasing orders data, reducing transparency, but visibility has been poor anyway, with optimism over underlying demand in the semiconductor market giving way to realization that the industry is in the midst of a typical (i.e., severe) downturn. The 1Q decline in Automotive sales and profits, on the other hand, appears to be due primarily to project timing. Management's guidance for the six months to June - sales down 4.0% year-on-year, operating profit down 25.3%, net profit down 33.4% - might be achievable, but it implies higher sales and profits in 2Q, which would be unusual. 2H guidance assumes a 39% sequential increase in semiconductor equipment sales, which seems very unlikely. Management admits that 2H guidance is probably too high, but plans to wait until 1H results are announced (probably during the first week of August) before making any changes to their full-year outlook. Our own estimates for FY Dec-19 are below management's. We expect recovery in 2020.The announcement of 2Q results plus developments on the trade front should set a new baseline for investors. But it appears that a large part of the anticipated downward revision has already been discounted.

Weak 1Q results, declining demand from the semiconductor industry, renewed pessimism over the U.S. - China trade war and, apparently, profit taking have caused Horiba's share price to drop 27% from its April 19 peak of ¥6,930. At ¥5,040 (Thursday, June 6, closing price), the shares are now close to the bottom of their 5-year valuation ranges, selling at 10.6x management's EPS guidance for FY Dec-19, 11.5x our EPS estimate for this year and 10.4x our estimate for FY Dec-20. Year-on-year comparisons for the three months show total consolidated sales down 14.0%, operating profit down 51.8% and net profit down 58.1%. Automotive sales were down 10.3% and operating profit down 65.3%. Semiconductor equipment sales were down 29.9% and operating profit on those sales down 49.5%. The company has stopped releasing orders data, reducing transparency, but visibility has been poor anyway, with optimism over underlying demand in the semiconductor market giving way to realization that the industry is in the midst of a typical (i.e., severe) downturn. The 1Q decline in Automotive sales and profits, on the other hand, appears to be due primarily to project timing. Management's guidance for the six months to June - sales down 4.0% year-on-year, operating profit down 25.3%, net profit down 33.4% - might be achievable, but it implies higher sales and profits in 2Q, which would be unusual. 2H guidance assumes a 39% sequential increase in semiconductor equipment sales, which seems very unlikely. Management admits that 2H guidance is probably too high, but plans to wait until 1H results are announced (probably during the first week of August) before making any changes to their full-year outlook. Our own estimates for FY Dec-19 are below management's. We expect recovery in 2020.The announcement of 2Q results plus developments on the trade front should set a new baseline for investors. But it appears that a large part of the anticipated downward revision has already been discounted.

Horiba (6856 JP): Downward Revision Ahead

Japan

HORIBA, Ltd., manufactures and markets measuring instruments and analyzers.  The Company's main product lines are scientific analyzers, environment monitoring, medical analyzers, engine emission analyzers, and semiconductor test equipment. Horiba operates worldwide.

Price Gauge
52-week Range	0.00 - 0.00
Market Cap (USDm)	0.00
P/E	0.00
Dividend (Yield)	0.00

Horiba Ltd (6856 JP)

Horiba (6856 JP): Semiconductor Division Leads Upward Revision for FY Dec-22

FY Dec-22 guidance raised due to stronger-than-expected semiconductor related demand, especially in 2H. Japan-US semiconductor cooperation a longer...

Scott Foster

Horiba (6856 JP): New Growth Opportunity in Hydrogen Energy

Hydrogen related analytical systems becoming a new growth driver, replacing vehicle emission measurement. Semiconductor related demand slowing, but...

Scott Foster

JAPAN FLOW: Foreign Investor Flows Still Matter (Redux, Reduxed) - Playing Catchup

The Japanese market has unusual flow cyclicality. Foreigners take price. They have been net sellers. They buy when USD-denom TOPIX outperforms....

Travis Lundy

Horiba (6856 JP): Slow Recovery Ahead, Valuations Low

As expected, 1H results fell short of expectations and FY Dec-19 guidance has been cut. But the new full-year guidance, which implies a 10%...

Scott Foster

Horiba (6856 JP): Downward Revision Ahead

Weak 1Q results, declining demand from the semiconductor industry, renewed pessimism over the U.S. - China trade war and, apparently, profit taking...

Scott Foster

Price Gauge

52-week Range

Market Cap (USDm)

P/E

Dividend (Yield)