Concentration, risks and portfolio sizing

First, a small portfolio update. Until Q2 the portfolio is -9 percent, and YTD it is -9,5 percent, both measured in SEK. Currency movements, a stronger SEK, have hurt the result a lot, but it's at least a stable result. I have sold Uni-Charm Indonesia, since I now consider them to be of too low quality (they aren't profitable enough, Q2 took a turn for the worse). I also decided that I only want to own independent companies. This was not a good investment at all, as you can guess, but I hope I have learned some lessons.

Increases have been made in Argent Industrial, Primeserv and RFM. The money comes primarily from Uni-Charm, but I have reduced my holding in AS a little bit, just for the sake of increasing in those three. They have P/E:s of 6-9, EV/EBITDA of 2,9 to 3,8 and the FCF-yields range between 15-18%. I have also increased my holding in Karooooo a little bit (reinvested dividends). Perhaps too many changes going on this year, but the theme of them is consistent, so lets move on to the subject of today: concentration!

The reasons for concentration
This year, I have become much more focused. I have reduced my portfolio from 14 stocks to 10. There are reasons for that, hopefully good ones! One simple reason is that the really fat pitches, unique opportunities on a global scale, are rare. If you find one of these, you should go for it! Bold words, but when the upside is significant and the downside is low, and you think you understand the business well, you should size big! Time and opportunities are both limited.

The large holdings have significant long-term potential, of course. But, more importantly, there are no major risks. They have net cash or very low debt, business and competition risks are low, and the valuation risks are not high (very important). Hopefully, the worst-case scenario will not be that bad (although unknown risks always exist). To summarize my thinking: Exceptional opportunities are rare, and my high-conviction holdings are low-risk stocks. Taking "high risks" with low-risk stocks is OK!

Three buckets and limits
As I see things now, the number of stocks in my portfolio should be about 8–14, most probably 10-12. The actual number depend on what I find and if I chose to invest in more markets (thinking of Japan). In theory, a normal share is almost 10%, but in practice it's smaller, since I prefer to invest heavily in my high-conviction holdings.

I usually categorize my holdings into three buckets:

• Small: 4–5%
• Core: 7–8%
• High conviction: 12–20% (reduce at 25%)*.
  
  

The bar is set at 4%. If I'm not comfortable owning that percentage in a company, I shouldn't own it at all. I want to be able to keep track of what I own, not have too many stocks to follow and be as updated and sure as I can be about them. Is the thesis intact? Is there any new information? Is the weighting OK? I want to make fewer, better-informed decisions. As for the small or core holdings, some things are holding me back. I like the companies a lot, but they are somewhat riskier, have more uncertainties or just don't offer as good growth or low valuations.

*The limits are for buying. If a small holding decreases to 3%, I should either increase it or sell it. But if a high-conviction holding increases to 21%, I will not automatically reduce it. However, if it increases to 26%, I will automatically reduce it to 25%. I hope this will reduce my risk of confirmation bias, no buying when I already have 20%.

The maximum limit? Why stop so "early"?

I mentioned rare opportunities at the beginning, so why not invest 30% or 40% of your portfolio in one of these? Well:

• Confirmation bias. I have to limit myself. If I allowed myself to invest more, I would be influenced by the time, money, and effort I put into these stocks. I am already affected, but the effects would be more severe. Put simply, I might be more and more wrong without being able to see it.
• The unknown is still there. Even if I quit my daytime job and devoted all my time and effort to following just three companies (the so called best ones), I still wouldn't be safe. Anything can happen in any company. There are things I don't know, I can't know, or anyone can know. The only way to protect yourself against catastrophic events is to diversify a bit more. I want to sleep well.
• The last one isn't about risks, but I look at many markets and ideas. That way, I will certainly come up with several ideas worthy of being high-conviction holdings. If I only invested in one or two markets, I might run out of ideas, but when the pool is 10,000 companies, there will be more than 4 worthy of 10+ percent in your portfolio. But perhaps not if the pool is just 500–1,000 stocks. (I also strongly believe that owning stocks in different stable countries is safer than just one).

These three reasons explain why 25% is my maximum limit, even if the company is the cheapest and most exceptional stock opportunity that I can find in the world. Again, these are big words, and I talk about best in the world according to my strategy (the chance that the stocks really are, is very slim).

What worst case are we talking about? 

A common way of thinking is that a stock might lose 50% in a bad year. This could happen to your largest holding. Let's say it does. If you have 20% of it, you will lose 10% overnight. If you have 40%, you would lose 20%. It's a really bad year, but it's something you can recover from. So why am I afraid? Well, two things.

• A bad year is one thing, but permanent loss of capital keeps me awake at night. What makes me afraid is if the 20% holding goes to zero. Or if the 40% one does. It's very uncommon, especially if we are talking about well researched high quality companies. But it could happen if you're really, really unlucky. The issue is that I invest in small, rare often little-known stocks that on the surface almost seem too good to be true. The research is thorough, and the quality is definitively high but it's still impossible to be 100% certain.
• Losses are harder to recover from. If you lose 20%, you have to regain 25% to break even, but if you lose 40%, you have to regain 67%. Mathematically, it is important to limit your losses.

Opinions?

I think it is logical and it might suit me to be a bit more concentrated. It is also highly individual. However, there could be some problems with my thinking. Having fewer stocks is less fun. You could become narrow-minded. And ten to twelve is a really small number if you invest in many markets. There are often almost equally good opportunities, and choosing one and rejecting the other could be too drastic.

What is your opinion on the portfolio sizing? Is a portfolio of 10-12 stocks a good middle ground, where you bet big on your "best ideas" but still are diversified? Are there effective ways to minimise the risks of confirmation bias and unknown events besides diversification? I guess a clear selling discipline and a critical mindset could be important. Do you have any thoughts on the matter? 

Portfolio changes

I have been more active than usual this year, so it's time to provide an update on my portfolio and share some thoughts on the changes. If you follow me on Twitter/X, this won't be news to you, but it's good to document the changes in a more structured way. This blog functions like a diary, and most importantly, I hope to receive comments, even though I know that blogs are a bit out of fashion nowadays.

The portfolio is shown above, and now some comments on the changes:

The selling

K2 LT

As explained in these tweets, I have sold out. In short, the reasons are increased competition and corporate governance issues. The catalyst was the delayed annual general meeting, which suggests that they may lack active owners. With hindsight, perhaps the company was too small, but the main mistake was overestimating the 'moat'. Nevertheless, I believe that you can hold this company and perform well. It's a nice, inexpensive, non-cyclical company, but I decided to invest in even nicer and/or cheaper companies instead.

Sarantis

I also sold Sarantis, which was a harder decision. The reason was the valuation. I can certainly own slightly pricey stocks, but I decided this was a bit too much. Century Pacific and Karooooo are also a bit pricey, but they are constantly growing their revenue in double digits. Sarantis is very good and stable, and certainly not expensive in a broader context (i.e. Walmart and P&G), but I just think there are better options in the portfolio, for example, Ultrajaya (Consumer) and AS Company (Greece). The bar is high for companies trading at or above a P/E ratio of 20. And it should be.

The buying 

Ultrajaya

I increased my position earlier this spring and again more recently. It's a market leader whose share price has dropped by around 30% this year. Having owned the stock since 2017, I think now could be a good time to increase my holding, but who knows?

They are buying back shares and are trading at a P/E of 12.5 and EV/EBITDA of 7.7 (TTM). For 2027, it's a P/E of 7.5, and EV/EBITDA of 4.3, and a FCF-yield of 13.5% (based on reasonable estimates). This may be as cheap as it gets for a strong consumer company. If Buffett were to buy something in Indonesia, this would be a strong candidate. Milk consumption is low but increasing, and their market share and brand strength is high. They are diversified and, most importantly, vertically integrated. I have thought a lot about consumer companies and I really like that aspect. I considered buying back Kri-Kri, but decided to invest more in Ultrajaya instead.

Arwana

I recently made a small increase. They are also cheap and are buying back shares. It's trading at a P/E of 10 and an EV/EBITDA of 6 (TTM). Arwana and Ultrajaya are the highest quality companies I can find in Indonesia. The reason I don't own more in Arwana is that, unusually for the companies I own, they are quite affected by external factors: Import restrictions and gas price subsidies. Nevertheless, I believe they have a moat and their competitors are certainly more affected.

Uni-Charm Indonesia 

I made a small increase earlier. Extremely cheap, but it could be artificially cheap. However, it's rare to see a P/E of 10, an EV/EBITDA of 1, cash per share of 420, and a stock price of 565. I decided to increase my holding, well aware of the value trap risks. I'm following Indonesian forums closely on this one and the jury is still out. It's notable that some Indonesian local investors have been skeptical from the start (link), and everything is on the table (sell, hold or increase) depending on future reports. As always, but even more so in this case. If I sell, it will be because the quality is too low, that market share or margins deteriorate.

Karooooo

I believe in Karooooo and think they are on the right track. I've been thinking a lot about corporate governance, risk and culture - something I did even before buying the stock in 2021. I did reduce my holdings last summer, but I have increased my holding this spring. It may seem illogical for anyone who calls themselves a value investor to own stocks trading at a P/E of over 30. But my answer is simple: This company is unique. Fast forward five years and you will (probably) see strong, profitable growth, with Asia playing a much bigger role. They differentiate themselves and have clear economies of scale. See the tweets from this spring and the more recent ones for more info about my view. 

AS Company 

It is also a company on the right track. The clarifications they made about their investments in the hospitality sector were important. Otherwise, there was a real risk of diworsification (in the sense of Peter Lynch). Notably, this is the only European stock I own. It's a rare combination of a strong history, good momentum, a very strong balance sheet and a low valuation, not to mention buybacks and high dividends. I think it's quite cheap given its quality, and it's an attractive value stock that's unknown but stable.

Portfolio risks?

Overall, the portfolio has a bias towards emerging markets, so I might perhaps try to find holdings that, over time, reduce that exposure. But I'm not sure. You should allocate most of your portfolio to your best ideas. I certainly find the best ones in Indonesia, the Philippines, South Africa, Singapore and Greece. It's difficult to invest in things you don't believe in just for the sake of diversification and managing portfolio risk. However, I do set 'caps' at country level. And Argent Industrial, a significant holding, is essentially a UK company. And I do look at some companies in Japan right now.

I also think the risks are reduced due to the nature of the businesses. Most are non-cyclical; only Argent, Arwana and Ekadharma have cyclical elements, and a large proportion are consumer staples. All companies have very strong balance sheets; the median company has net cash equivalent to 1.5x EBITDA. If you look at the portfolio this way (the picture below), you could argue that the risks are actually low. Valuations are overall low but Karooooo and Century Pacific "distort" the P/E ratio (the only ones above P/E 13). 

Another observation is that there is a bit more concentration than before. I decided to do it this way since I am getting to know my holdings better, although I may still miss something. And with higher concentration, it's more important than ever to find ways to reduce the risk of confirmation bias. 

What do you think about the portfolio and the changes?

Q1 2025: AI - a gamechanger for exotic investing?

As you can imagine, the quarter was not great. I am down 11 percent in SEK for Q1 2025. The only change in the portfolio is an increase in RFM Corp. There is a lot to be said about presidents, tariffs, currency movements and predictability, but I will not. The only thing I will say is this: If you own companies with low valuation, strong financials, high profitability, operating in somewhat non-cyclical sectors - you will probably do well over time! 

So, let's move on to today's topic: AI and exotic investments.

My new process 

AI is something I have been learning more about over the last few months. I am starting to use AI in my investment process, and it has actually been a game changer. For me, investing in small, obscure companies means no company reports. You are on your own. Well, not anymore. I think AI, and the one I use most is Grok, is excellent at producing "company reports".

My process is now:

• Finding company ideas: FinChat screener, fund holdings, other investors.
• Quick impression of the company: FinChat (numbers), corporateinformation (quality, stability), Financial Times (graphs), Simply Wall street (what stands out). 
• Mini-research: Ask i e Grok about: Market, moat, competition and possible red flags.

Once this quick process is done, the classic research begins. Now we are talking about hours, days, weeks or months. With the help of AI or not. But there is no obvious reason to pass on the company.

The big change?

It all starts with the company, as before. I think I can judge whether it is a good company to invest in (valuation, quality, history, etc.) or not. The big problem for me is judging the market and the "surroundings". This is much easier now. And AI also helps to look deeper into the company.

The effects could be this: 

• For Large cap investors: None! You just read the reports as before. Pick one of the 10-20 reports out there and make up your own mind. 
• For no-coverage investors like me, there is the gamechanger. AI, like Grok, almost give coverage. In practice: I might feel more comfortable, or at least being able to process, much more unknown small caps in regions like Indonesia and Japan. Getting a grip on the surroundings (everything outside the company) really helps. You can also create decent "company reports".

Of course, AI tools cannot be trusted. But they will help, and they are already surprisingly good at digging up facts and painting a picture of a company's place in the broader market. For me, a simple search on moat, market share and uniqueness could replace 25-50 Google searches. In the same time you will "analyze" 5-10x more companies. And most importantly: You can quickly eliminate companies that do not fit your style or criteria.

I mostly use Grok, I think the results are better for analyzing companies than for Chat GPT. But the really interesting question is. Where will these tools be in 3-5 years? Comprehensive reports I guess.

What is your edge?

When almost all information is available regarding all companies. What is your advantage? It is not in deep research if AI can do that (and will evolve). If the majority of companies go from unknown to known (due to AI), is this perhaps a reason to change your investment style? 

Here is my suggestion. Your advantage in a world of information and analysis is to be able to buy companies that funds cannot. To go the extra mile, to look at small companies, to look at markets that others reject, and to use the broker that gives you the most access. There will still be hurdles and obstacles, and perhaps "Home Bias" is the most popular one. 

My edge is being able to buy small obscure companies (courage and mindset). The advantage is thinking outside the investment box.  Another effect could be that knowing something very well will always be an edge, i.e. things that AI fails to observe, gets wrong, and companies that simply screen horribly.

To summarize, my edge is:

• Not having so much money - being able to buy companies that funds cannot touch (a fact).
• Looking at smaller companies that are a bit hard to buy (a mindset).
• Looking at markets that are a bit hard to reach. This is both a mindset, getting away from home bias and a practical thing (finding the brokers with the greatest reach). 
  

For Sweden, the last one is quite simple. I think Swedish brokers have the best reach in Europe, almost. At least if you are comfortable with the old way of placing orders by phone. It's just a shame that more Swedes don't use their advantage. Living in Sweden is a double-edged sword: since investing is popular, it's hard to find hidden gems locally, and almost all companies are already well covered. You can't find both quality and cheapness. But it is easy to escape!

Opinions?

What do you think? Do you use AI in investing? And what is your edge, today and tomorrow?

Q4 2024: Results, three M's and demographics

Results for 2024

The portfolio had a great year. The performance was +39.9%, which is explained by a combination of growth and revaluation and some currency movements. My result is measured in SEK, so a weaker SEK had a positive effect. Overall, the big and simple explanation is that my larger holdings did very well. I don't see this as the new normal and I am cautiously pessimistic about 2025. The top five performers for 2024 were Karooooo (+85%), Argent (+82%), Century Pacific (+35%), Sarantis (+31%) and RFM (+29%) excluding dividends. The only detractors were Ekadharma (-14%), Uni-Charm (-24%) and Delfi (-30%). The portfolio at the time of writing is shown above. 

A few holdings, namely Century Pacific and Karooooo, are now reasonably valued or perhaps a little stretched, at least for a Graham-inspired investor (trading above 20-25x P/E). I still own them because of their excellent quality combined with growth (Century) and strong growth (Karooooo). However, I think a P/E below 15 is to be expected when looking 3 years ahead, at least for Karooooo. As for Century, I may be laughed at by saying this, but it's hard to find non-cyclical companies of the same quality, even if you look at all the listed companies in the world. Nevertheless, I'm thinking of reducing my holding a little, despite some tax implications. 

Most of the portfolio consists of cheap or very cheap non-cyclical companies, despite higher share prices. The median P/E is 14, the median EV/EBITDA is 8 and, importantly, the FCF yield is around 10%. All but one of the companies has net cash. At the portfolio level, the 10-year CAGR is around +8% for sales and +20% for EPS. Future growth is important, but overall, I think I have high quality companies with some growth and reasonable valuations. Now let us move on to today's topic.

Demographics: Always important for consumer companies?

If you like boring defensive companies, as I do, I think demographics are an underrated thing to look at as a long-term investor. For consumer companies and local based companies, demographics could be critical. Ok, I know that "what company at what valuation" is the most important and what you invest in. But in the long term, the market in which the company operates, and structural changes should not be neglected. Focus Compounding used to say something like "the market in which a company operates makes up about 50% of the investment" (if I got that right, I am not sure).

Let us take a quick look at ASEAN (Indonesia and the Philippines), Germany and the USA and try to "predict" the changes that will occur by 2030 and 2040. Let's look at the number of people and GDP, with OECD and Population pyramid as sources.

• Indonesia
• 2024: 283m GDP: 1 400
• 2030: 296m
• 2040: 312m GDP: 4 000
  
  
• Philippines
• 2024: 116m GDP: 471
• 2030: 121m
• 2040: 130m GDP: 1400
  
  
• Germany
• 2024: 85m GDP: 4 700
• 2030: 83m
• 2040: 80m GDP: 5 300
  
  
• USA
• 2024: 345m GDP: 29 100
• 2030: 355m
• 2040: 370m GDP: 32 000
  
  
• Sweden (my home country)
• 2024: 11m GDP: 610
• 2030: 11m
• 2040: 11m GDP: 800

Put simply, a consumer related domestic-focused company in Indonesia or the Philippines should, all things being equal, have a better future than one in Germany, the US, or Sweden. At least if they can capitalize on a country with more and richer people. This is a simplistic and uncertain idea, but still, at least demographics is the most "certain" of the long-term predictions. It's also interesting to note that the US has quite good demographics compared to most of Europe.  

The three M:s, and a D?

I listened to a Swedish podcast some time ago, I can't remember which one, but there was an interesting rule of thumb: you should always be aware of changes in the market, market share and margins when you invest.  

• Market growth: How much will the market grow in 5-10 years? I usually look at some market growth reports. I e telematics (Karooooo) is growing about 15-20% per year and the Indonesian tape consumption not so much. A growing market is good (all other things being equal), but only a start. There are two more M's. The demographic changes if one of many things that affect whether the market is growing or not. 
• Market share: Will the company gain, maintain or lose market share? When a market is growing, the market leaders, which I usually invest in, will sometimes have declining market shares. It's nice with market growh but it also attracts competitors. You could also argue that the simple measure of market share isn't enough, and that relative market share is the most important measure. For example: If the market leaders have 25, 21 and 19% respectively, it's not a big difference. But if the leader has 12% and the next company has 2%, the leader could be in a much stronger position, and there is more of the market to take from weaker competitors.
• Margins: If a company must lower its price to maintain market share, this is very important. Changes in margins should be monitored. If a company can't protect its margins (moat anyone?), it won't benefit from a growing market or larger market share, at least not in a sustainable way.

The best case is a company that benefits from all three M's. They operate in a growing market, they gain market share and can increase their margins over time. 

Now let us go back to the beginning of this article. You might want to start with these three M(usketeers) and add a D('artagnan). As you might have guessed, D stands for demographics. It will in some way affect market growth for any business, at least consumer businesses. But it could go much deeper than that if you think about it. With big demographic and GDP growth, the nation may have more international recognition, its currency may be stronger, more and richer local investors will put more money into the local stock market, and more foreign investors will be attracted. It will become a bigger and more recognized country, which could have a knock-on effect on its stock market. 

As a long-term investor, the three M's are what you should focus on in any investment. But as a bonus, if you are a long-term buy-and-hold investor, you should at least think about demographics.