This article first appeared in Wealth, The Edge Malaysia Weekly on December 26, 2022 - January 1, 2023

In the course of his work, Peter Lim Tze Cheng, a fund manager and chief research officer at Trident Analytics Sdn Bhd, found that the semiconductor industry had been overgeneralised. The former CEO of Inter-Pacific Asset Management and head of research at Equities­Tracker Holdings Bhd says local investors tended to look at these companies as though they were all the same.

This gave him the idea to write a book, What I learnt about semicon & EMS: A sharing of my views on the industry, to help readers and investors break down the complexity of the semiconductor industry (see review). Despite the long title, the book is easily digestible at just 144 pages.

The investment community took notice of the semiconductor industry in early 2020 when the Covid-19 pandemic struck the world. Electronic devices, powered by semiconductor chips and the internet, became the only means for people to work and communicate from home when countries were locked down. That motivated Lim to complete his book.

In an interview with Wealth, he talks about the book and shares his views on the semiconductor industry, including what the government can do to propel the sector forward. The following are excerpts from the interview.

Wealth: What gave you the idea to write the book?

Peter Lim Tze Cheng: What drives me is that the technology sector has been overgeneralised. When I heard people say ‘this and that company is a tech company’, it made me realise that they didn’t really understand the industry. People started comparing different companies as though they were the same. But they are actually quite different in where they sit in the value chain. It is quite a complex industry.

What are the things you want your readers to learn from the book?

I hope the readers can ­really understand the semiconductor industry, what the whole value chain is about and what matters to each part of the value chain. Companies in each part of the value chain play a very different role in the whole industry and their dynamics are very different.

Did you deliberately keep things simple in the book? Did you use only information that is publicly available to write it?

I tried to avoid sharing non-public information. During my research, I discovered more about the companies’ strengths and weaknesses. But this is non-public information. Additionally, if I disclose too much about a certain company, what if their competitors read it? It would be like I was putting some companies at a disadvantage. The real challenge of writing this book was figuring out what not to write. That’s the reason I didn’t go into too many details.

While you are a fund manager who is familiar with semi­conductor companies, you are not an engineer yourself. How did you make sure that your content was correct?

I did a lot of research myself and I did send parts of it to CEOs [for them to read through]. I chose one or two companies from each part of the value chain and let them vet through. This was to ensure there were no incorrect facts.

You touched briefly on the challenges that some semiconductor companies are facing. Can you elaborate on this, and what suggestions do you have for the government?

The labour issue is the biggest challenge they are facing. I think the government has to understand the whole semiconductor value chain when it talks about automation. For instance, the OSAT (outsourced semiconductor assembly and test) players are highly automated. But if you talk about other parts of the industry, like the precision components part of the value chain, you need machine operators, which can be quite labour-intensive. If we get the policy wrong, these companies will face a permanent labour shortage.

I think the government cannot assume that everything in the tech sector can be automated. That’s not the case. There are some parts that you can, yes, but there are other parts that you can’t.

Eventually, we will lose out to countries like Vietnam if we continue to rely on foreign labour. And we will again face the old issue of having too many of them. Is this right?

It is not a question of labour cost, but of labour availability. Even if you’re willing to pay a decent salary for labour, way above the minimum wage, you still can’t get them if the government does not allow you to bring in more workers.

We still have an advantage over countries like Vietnam in terms of management quality, execution, logistics and other things. One of the reasons we are losing out is labour. If we can solve that, we are good.

I am not talking about foreign labour like construction workers. I am referring to skilled workers who operate machines. You might say it is possible to get workers from the local labour supply. But with the gig economy today, if I can ride my motorbike delivering food and parcels and earn RM2,000 a month, why would I want to work at a factory with two shifts a day?

What do you think the government can do to help spur the industry?

The thing we are lacking is incentives. We allow foreign players to set up plants and facilities here. But it should come with some conditions, like requiring them to buy a certain percentage of local products with their capital expenditure. There could also be tax breaks or capital allowances for the purchase of local machinery. On one hand, yes, you lose out on the tax [collection]. But on the other hand, you spur the local economy. Look at how the US is encouraging its companies to expand their wafer facilities in the country: Tax breaks.

Do you think the local semiconductor companies are not moving up the value chain fast enough, as some have claimed?

This is why it is important for people to understand the value chain. Let’s have a very simple perspective by looking at the highest level of the value chain in the industry, the IC (integrated circuit, or chip) design. In this area, all countries including Malaysia, China, the UK and Japan are quite far behind the US.

The second highest level in the value chain is what we call [wafer] fabrication or manufacturing. We always talk about TSMC (Taiwan Semi­conductor Manufacturing Company) being the largest company in this area. But the owners of the machines that fabricate the wafers are companies in the US and Europe, not China or Malaysia.

Then, we come to the machines to produce the chips. You talk about Samsung and TSMC. But who owns the machines that produce the chips? It is also mainly in the US, not even China.

We are far behind [the top levels of the value chain]. We have to be objective about the whole situation. Not only Malaysia is far behind. Everybody is lagging behind the US in this sector. But it doesn’t mean we are bad.

Where is our strongest exposure in Malaysia?

OSAT. We have some of the world’s top companies in this area. Ten years ago, you would have said OSAT players were just manufacturers that bought machines and packaged the chips. They don’t have much to shout about. But now, if you look at the last five to 10 years, you don’t see new OSAT players coming into the market. That is because even the technology to package the chips is very difficult nowadays.

What is your view on the semiconductor industry today?

I think the market is either close to the bottom or at the bottom already. All the bad news is already out. Everybody is talking about the recession now. The risk has already been factored in.

And if you look at inflation, it fell in August and earlier months. The issue is that it has not fallen enough. I’m a strong believer that inflation is on its way down for the rest of the world. It means that in 2023, there will be room for the Fed (US Federal Reserve) to hold off on its rate hikes, or even lower the rate, which would mean a rate cut.

There should be more positive developments than negative ones. Whatever is negative, it is all out in the market.

What is your outlook on the semiconductor industry? What are the names that you like?

I think it will still grow but not at the same rate as in 2020 and 2021. We are now talking about normalisation. But at the same time, valuations have dropped significantly since then. It is now in the range of 20 to 30 times PER (price-earnings ratio), down from 60 to 100 times back then. It is more palatable to me, more reasonable.

I would say it is a good time to start looking at the semiconductor sector. Think about it. The trend will not reverse itself. Do you foresee a future where you won’t need semiconductors anymore? It’s impossible. In Malaysia, which sector is really export-oriented, where the whole world is its market? Semiconductors.

If you look at the whole semiconductor value chain, I like MPI ­(Malaysian Pacific Industries Bhd) and Unisem (M) Bhd in the OSAT space. It has come to a stage where the technology barrier is quite high now. In the production equipment space, I like Elsoft [Research Bhd]. Its strong capabilities are reflected in its margins, which are some of the highest in the industry. As for the EMS (electronics manufacturing services) players, I like P.I.E. [Industrial Bhd] and RGT [Bhd]. They have strong fundamentals and good prospects. Kobay [Technology Bhd] is quite capable in the precision space.

Save by subscribing to us for your print and/or digital copy.

P/S: The Edge is also available on Apple's App Store and Android's Google Play.