Disclaimer: This article is intended for educational and informational purposes only. Nothing in it constitutes — and it should not be construed as — financial, investment, tax, or legal advice, nor as an offer, solicitation, or recommendation to buy, sell, hold, or convert any security, currency, or financial product. This includes any specific companies named above (and their tickers) and any discussion of the rupiah versus the US dollar; these are used purely to illustrate a point, not as endorsements or recommendations.

The information here is drawn from sources we believe to be reliable as of the date of writing, but we make no warranty as to its accuracy or completeness. Figures such as exchange rates, financial ratios, and fiscal numbers move constantly and may be out of date by the time you read this. Any forward-looking statements are opinions, not predictions.

All investing carries risk, including the possible loss of capital, and past performance is never a guarantee of future results. Please do your own research, and consider speaking with a licensed professional before making any decision. Ultimately, your financial decisions are your own.

Reader’s discretion is advised. Hope it is useful in some way :)

I was sitting down in one of my favorite bank branches while waiting my queue number. You know running errands in a bank like a proper adult ;) When it was my turn and I spoke to the teller about my banking needs, I asked them a number of questions regarding the behaviour of other patrons.

In light of the recent weakening Rupiah against the Dollar, I went straight to ask them

“How do you comfort people who are worried about Rupiah depreciation? People usually associate current levels of depreciation like what happened in the past in 1998 specifically.”

The response was

“ummm.. but it is very different now and then. Nowadays the fundamentals is stronger. and there are more regulation that is in place for the banks now.”

I wasn’t very satisfied with the answer. Because I could tell that people are not going to be comforted with that statement. Also, the teller went on telling me that there they have been busy facilitating transactions of people buying the Dollar lately. They could serve around 6 - 7 people in one branch making big dollar buys. So I could tell that such statement does not comfort people.

The fear is real and contagious

Here is the part I have to be upfront about. I did not remember 1998 (I was only 2 years old). So I am not going to present to people who did and create a debate arguing that their fear is irrational. Especially at that time people watched their savings evaporate and their city burn.

You do not argue someone out of a memory like that.

But the elephant on the room is not about the fear per se. It is about the contagiousness of the fear in the age of social media. A comedian Ronny Chieng has a major rant here summarised in a 90 seconds YouTube shorts:

No matter how ridiculous and funny this video might sound, it is painting a real picture of my experience with many boomers. For the first time in human history, we have a class of individuals who are no longer that productive anymore having access to so much information especially when last time (before the internet was around), information was the biggest edge.

Nowadays, too much information slows you down. It causes people to overthink, be indecisive and be worried about making wrong decisions. But we can’t blame the boomers for this because for most of their lives, information is something that allows them to gain a competitive edge. So it is very counterintuitive to unlearn that and it is very hard to have them be aware of their unhealthy levels of information diet.

Which probably makes the very feeling of fear very contagious

Robert Shiller wrote a whole book about exactly this — Narrative Economics — and his core claim is that economic stories spread like diseases. “Economic narratives,” he writes, “follow the same pattern as the spread of disease.” A story does not have to be true to move an economy. It only has to be infectious. And the 1998 story is beautifully infectious: a frightening number, a remembered trauma to pin it to, and a single simple instruction:

Rupiah is going to collapse so buy dollars, now.

But a story cannot spread that easily if there is no element of truth to it. And it is definitely not my style to make a strawman argument and then attempt to beat it down to say that these people are not rational.

So let me steelman the fear properly

If there are further points that you’d like to add, please add them in for my consideration. It might be an open discussion of some sort as well. The idea is not to dismiss it but it is also not to be consumed by it alive. We want to lean into the fear and see it for what it is. I think that is a much productive use of our time, to make household economic decisions on our daily lives.

Start with the budget (APBN)

In 2025, Indonesia ran a deficit of about 2.92% of GDP. This deficit is seen as the widest outside pandemic years and it is very close to the 3% ceiling. For context, the 3% was written into law after the last crisis in 98.

Plus, tax revenue came in roughly 8% under target in 2025.

So it follows that if government revenue does not hit target, we could expect that budget deficit could grow further surpassing the 3% ceiling. In fact, the government has put that 3% law into review. To me, that’s a clear sign that fiscal anchor is starting to drift.

Danantara

Now we have a new player as well which is designed as a sovereign vehicle to drive economic growth but their reporting is still quite unclear to this date.

• What’s their economics like

• What’s their balance sheet like

• Are they profitable, etc

So it is entirely possible that the liabilities could be accumulating off on the side. But not only that the government is running fiscal deficit, we are also seeing contentious policies that are being passed by the government. Such as free nutritious meals (MBG) that is highly controversial, arguing that this would exacerbate the deficit situation considerably.

I mean how it is not a fiscal monster? The budget in 2025 was Rp71tn and in 2026, it will be up to Rp335tn.

This is still being done despite the 3% ceiling limit. And don’t get me started about the unpredictability that the new export policy started. Yes it sounds great on paper, but it is obvious that investors are jittery about the ability for the entity to execute.

IDX

The situation with the stock market isn’t helping much either. Indonesia received a warning from MSCI in January 2026 that it would be downgraded to the frontier market because there was lack of transparency in the stock ownership which causes price engineering to be more rampant.

In response to that, there has been swift transformation in the ruling such as >1% shareholder disclosure rule, 15% free float minimum requirements and so on. But it remains to be seen if such efforts are sufficient to restore investor confidence.

All in all it seems to be like a perfect storm

Put it all together and it does read like the makings of a perfect storm. Every incentive is now lined up for money to leave Indonesia. And a sizeable chunk of the money in our stock market, by value, is in fact, owned by foreigners: locals now own roughly 61% of the equity market, with foreigners near 39%.

That is not nothing and it is a very serious issue of eroding trust. When you put it together, you get a coherent, sober worry: loosening fiscal discipline, money walking out the door, a central bank defending the rupiah, a currency at all-time lows.

So yes, I take all of this seriously in 2026. Now let’s look at how it started last time in 1998.

How it was back then

My parents’ memory of 1998 is the rupiah number. They said that it weakened from around 2,400 to roughly 16,000 against the dollar, with uglier spikes in the worst of the panic. But just like today in 2026, currency was the symptom, not the disease.

What made 1998 catastrophic rather than merely painful was that four crises stacked on top of one another and fed each other:

1. a currency that broke from its zone,

2. a banking system that was rotten with poor oversight,

3. a corporate sector exposed to dollar debt it could not repay,

4. and a government that collapsed politically while trying to manage all three at once.

We can dig deep into all four pillars above but I find it amusing that they are very consistent with the core finding of Carmen Reinhart and Kenneth Rogoff in their book: This Time Is Different: studies eight centuries of financial crises (banking, inflation, currency and sovereign debt).

They mention that banking crises tend to come before sovereign defaults, not after. In their research around their book, they put forth:

“banking crises... often precede or accompany sovereign debt crises. Indeed, we find they help predict them.” The hidden debts that sink a government, they note, often include “private debt that become public... as the crisis unfolds.”

So the order of events matters. The currency didn’t bankrupt the country directly. It blew up the banks, and then the banks blew up the state.

Let me repeat that, because it quietly reframes the whole fear. The leading indicator of a sovereign blowup is likely not the exchange rate. It is the banks. More precisely, it is whether the banks are carrying unhealthy levels of foreign-currency exposure that a falling currency can detonate.

So the honest question is not “what is the rupiah number?” It is “is that dollar bomb sitting on the balance sheets again?”

And the wonderful thing about that question is that, unlike the future of the rupiah, it has an answer you can simply look up in their annual reports.

You nervous? It could be a dollar bomb that explodes you know

Ok let’s take a look that one number which is net open position (NOP).

After 1998, Bank Indonesia introduced a hard cap on the exact thing that killed the banks last time. It is called the Net Open Position (Posisi Devisa Neto). The formula is:

(foreign currency assets - foreign currency liabilities) / bank’s capital

currency assets example:

• banks’ cash position in foreign currency

• banks’ loans given to customers in foreign currency

currency liabilities example:

• banks’ customer deposits in foreign currency

• banks’ debt in foreign currency

In plain language it means how exposed is this bank to a move in the currency.

But why does that matter so much? Think about what happens when a borrower goes bad. If a bank lends to a company and the company defaults, the bank eats the loss — some borrowers always default, and that part is normal.

That’s why we have provisioning so that writing out that debt will not cause liquidity issues. What made 1998 lethal was that the defaults and the currency crash arrived together. Banks had borrowed in dollars and lent in dollars.

When the rupiah collapsed, their borrowers couldn’t pay. Think about it, if you own a business and you take a USD loan, your loan would shoot up overnight because it is in USD but you earn in Rupiah. As a consequence you will default on your loans because your earnings couldn’t keep up with the ballooning debt.

But the banks? They still have to repay the liabilities they take in USD terms too. So they still owe people (depositors) in dollars but they don’t have the money to repay them back. Which ultimately caused banks to be insolvent.

So now we understand why Bank Indonesia caps the Net Open Position. It limits how far a bank's dollar assets and dollar liabilities are allowed to drift apart relative to its capital.

The dangerous side in a crisis is when a bank ends up owing more dollars than it holds: if the rupiah collapses, the value of what it owes balloons faster (USD) than the value of what it holds (Rupiah), and it has to buy dollars at a punishing exchange rate to settle up. This crystallises the loss on its balance sheet. So banks are to keep the gap small, so that loss stays small.

How much does Bank Indonesia allow?

The cap is 20% of capital. The higher the % is, the more vulnerable a bank is towards currency fluctuations. It is now a figure that is heavily monitored. Today it needs to be reported at the end of a working day by law.

Compare that to 1997, when, in the words of a former senior deputy governor of Bank Indonesia, Anwar Nasution, the rules on foreign borrowing “were generally violated” and corporate dollar debt simply “was not recorded.”

I tried digging the NOP numbers back then in 1998 but couldn’t really find a reliable one. Today the figure is as follows:

When we read the data, we see that the most dollar-exposed of the big four, BRI, is running currency risk at about 2.5% of its capital. The rest sits lower — from BCA’s rounding-error 0.1% up to Mandiri’s 1.8%. Industry-wide, bad loans sit a little above 2% and the capital adequacy ratio sits around 26% (roughly three times the global minimum).

If we look at it as it is, these are not 1998’s walking-zombie banks papering over bad loans with creative accounting. They are some of the best-capitalised banks in the region, and they are not holding the bomb.

But none of that makes the rupiah’s weakness fake. A weaker rupiah genuinely hurts importers, dollar-cost borrowers, anyone with bills denominated in foreign currency. That is a real cost, and I won’t wave it away. But it is a completely different from a banking collapse that precedes a sovereign default.

So what do you do with this

I am not going to tell you where the rupiah will go. Nobody can, and a confident “Relax bro, it won’t hit 50,000” is not entirely useful as well. The currency could keep weakening for entirely ordinary reasons — a strong dollar, higher oil prices, money chasing higher yields elsewhere. The fiscal drift I steelmanned earlier is genuinely worth watching.

What I am saying is that “the rupiah is at a record low” and “Indonesia is about to repeat 1998” are two completely different claims. The first is a price. The second is commonly preceded by a banking collapse.

Fortunately, for the second one you can check rather than predict, by reading the balance sheets. As of now, they read clean.

And I guess this is exactly why we do what we do. When your feed is screaming negativity and you are probably tempted to “convert everything to dollars” — at the worst possible exchange rate, paying the spread to the banks and then surrendering your rupiah. Our job is to give you the data and make sure that your decisions are driven by things that you could check and verify.

Because if that fear stops spreading for whatever reason, at the very least you know that you have made your decision without regret.

References

In case you’re interested in reading the materials consulted in writing this article, below are the entry points:

1. Bank Indonesia (2015). Peraturan Bank Indonesia No. 17/5/PBI/2015 — Perubahan Keempat atas PBI No. 5/13/PBI/2003 tentang Posisi Devisa Neto Bank Umum. https://peraturan.bpk.go.id/Details/135521/peraturan-bi-no-175pbi2015-tahun-2015 [The Net Open Position cap of 20% of capital, and the relaxation from 30-minute to end-of-working-day reporting.]

2. Reinhart, C. M. & Rogoff, K. S. (2009). This Time Is Different: Eight Centuries of Financial Folly. Princeton University Press. [The book-length thesis that banking crises and sovereign defaults travel together across history.]

3. Reinhart, C. M. & Rogoff, K. S. (2011). “From Financial Crash to Debt Crisis.” American Economic Review, 101(5), pp. 1676–1706 (NBER Working Paper No. 15795, 2010). https://www.nber.org/system/files/working_papers/w15795/w15795.pdf [Source of the exact quoted lines: “banking crises… often precede or accompany sovereign debt crises. Indeed, we find they help predict them,” and “private debt that become public… as the crisis unfolds.”]

4. Shiller, R. J. (2019). Narrative Economics: How Stories Go Viral and Drive Major Economic Events. Princeton University Press (NBER Working Paper No. 23075). https://www.nber.org/system/files/working_papers/w23075/w23075.pdf [“Economic narratives follow the same pattern as the spread of disease.”]

5. Nasution, A. (2015). “How Indonesia reformed its risky financial sector.” East Asia Forum, 24 April 2015. https://eastasiaforum.org/2015/04/24/how-indonesia-reformed-its-risky-financial-sector/ [Former BI Senior Deputy Governor on 1997–98: rules on foreign borrowing “were generally violated,” corporate dollar debt “was not recorded”; inflation ~58%, rates >80%, GDP −13.1%.]

6. Indonesia Investments. Asian Financial Crisis — Indonesia. https://www.indonesia-investments.com/culture/economy/asian-financial-crisis/item246 [Rupiah ~2,400/USD pre-crisis collapsing to ~16,000; managed band floated from August 1997; 16 banks closed in November 1997 triggering bank runs.]

7. IDNFinancials / Indonesia Stock Exchange (2026). Foreign outflows from Indonesian shares and bonds, 2026. https://www.idnfinancials.com/news/60974/foreign-dump-indonesian-shares-bonds-outflows-at-idr-12-55-trillion [Foreign net selling of Indonesian equities in 2026.]

8. Bloomberg (2026). Indonesia’s fiscal deficit reaches 2.92% of GDP, near legal limit. https://www.bloomberg.com/news/articles/2026-01-08/indonesia-fiscal-deficit-soars-to-2-92-of-gdp-near-legal-limit [2025 budget deficit of 2.92% of GDP, the widest outside the pandemic years.]

9. Bloomberg (2025). Indonesia adds State Finance Law to 2026 legislative priorities. https://www.bloomberg.com/news/articles/2025-09-19/indonesia-adds-state-finance-law-to-2026-legislative-priorities [The 3%-of-GDP deficit cap — set by Law No. 17/2003 after the 1998 crisis — placed under review.]

10. The Diplomat (2026). “Unpacking Indonesia’s 2025 Fiscal Revenue Shortfall.” https://thediplomat.com/2026/01/unpacking-indonesias-2025-fiscal-revenue-shortfall/ [2025 state revenue came in roughly 8% below target.]

11. East Asia Forum (2026). “Indonesia’s fiscal anchor begins to drift.” 15 March 2026. https://eastasiaforum.org/2026/03/15/indonesias-fiscal-anchor-begins-to-drift/ [Danantara’s borrowing sits outside the deficit rule, raising contingent-liability and transparency concerns.]

12. Kompas (2025). “Anggaran MBG Melonjak Jadi Rp 335 Triliun di APBN 2026.” https://money.kompas.com/read/2025/09/24/112056126/anggaran-mbg-melonjak-jadi-rp-335-triliun-di-apbn-2026 [Free Nutritious Meals (MBG) budget rising from Rp71 trillion in 2025 to Rp335 trillion in 2026.]

13. IDNFinancials / Bloomberg (2026). MSCI halts rebalancing; Indonesia risks downgrade to frontier market. https://www.idnfinancials.com/news/60811/msci-halts-rebalancing-indonesia-risks-downgrade-to-frontier-market [January 2026 MSCI warning over share-ownership/free-float transparency; reforms include 1% shareholder disclosure and raising the minimum free float to 15% from 7.5%.]

14. bne IntelliNews (2026). Indonesian rupiah drops to historic low of IDR18,000 per US dollar. https://www.intellinews.com/indonesian-rupiah-drops-to-historic-low-of-idr18-000-per-us-dollar-446678/ [Rupiah at a record low around 18,000/USD, roughly 10% weaker year-on-year.]

15. Annual Reports 2025 — BCA, BNI, Bank Mandiri, BRI. Risk Management / Net Open Position disclosures. [FY2025 Net Open Position ratios: BCA ~0.1%, BNI 1.1%, Bank Mandiri 1.76%, BRI 2.51%; industry CAR ~26%, gross NPL ~2%.]

Disclaimer: story written in this article not a real story. This story is fabricated based on many many calls that I take with potential customers. Also any mention of symbols, ticker or investment instrument is not to be construed as financial advice. This article is for educational purposes only to help readers reflect and see things from a different angle.

In a Gmeet meeting, a potential customer sat on his desk with a blurred background. I could tell from his expression that he was worried. I cordially invite him to share his investment experience. He muttered that he has two main investing portfolio comprising of Indo stocks. One is for long term and the other is for short term.

The idea for the long term portfolio is sound, he would only buy bluechip stocks that are perceived to always go up in price. This makes sense because the companies inside the portfolios are truly well known. Consumer juggernauts all the way to big banks. These aren’t the type of companies that go missing tomorrow.

Then, on the other hand, the short term portfolio is for him to do trades based on market psychology, employing various trading strategies. It tends to contain riskier plays. Often companies that are gossiped to be the next big thing in terms of its price action. The term for this changes depending on which class of investors they are.

Conglo 2025: cacing-cacing naga-naga

Covid 2021: saham pompom

Late 2010s: saham gorengan

1990s-early 2000s: saham bandar

Nowadays in light of the MSCI phenomenon, It is not uncommon at all to see -80%, -60% floating loss of “conglo stocks” that were popular in 2025.

From this, I could tell that most investors already have given up stocks completely. They usually will blame the government and everyone else but themselves about the current predicament. However, there is a small minority who’s brave enough to lean into the fear and take this moment as an opportunity to raise their standards in investing by calling us.

Meanwhile for big banks portfolio is now down 26.2% and the person is usually not too worried because they believe that big banks would grow in the long term. What becomes the point of contention is usually opportunity cost. Is my money working as hard as possible at the current moment? While they wait for the drawdowns to recover.

Now I have seen this a million times

So I thought that this is just the “norm”. Every other professional that I have spoken to over the last two years has the same two buckets (kantong investasi). But to me it is quite confusing.

Why when people have different financial goals, different time horizons, different family context, and so on, somehow converge on the same two buckets, weighted roughly the same way.

And they are smart enough to know that it doesn’t work for most people, perhaps including themselves

For the long term portfolio, they’d know that they are making the dangerous assumption that the price they pay for a share does not matter. The better and more popular the business, the more expensive it gets. And that’s fine by them because they believe that in the long term, the prices of these great companies will always go up. But they also know that when expectations are unmet, the share price could be severely punished. Famous example is Unilever.

Picture this in the mid 2010s, the stock has been rallying for well over a decade (see price chart above). It was championed by brokers, influential people as the stock that is considered evergreen. I still remember how the stock was sold to me personally (I was still in high school back then more than a decade ago).

Broker: Toby, do you know what people would still do during an economic crisis?

Gullible highschooler: What?

Broker: People would still take a shower. So companies selling soap like Unilever will always be relevant. That’s why it is an evergreen stock that will always go up in the long run.

Gullible highschooler: Wow!

Fast forward about 10 years later~

I am not saying that Unilever is a bad company. It is big and massively profitable. It’s just that I bought that company at a huge premium when I started my foray into stock investments. It taught me that you could not blindly buy an asset without caring about its price relative to the value of the company, no matter how good or famous or resilient they are.

For the short term portfolio, the flaw is easier to spot.

It fails the survivorship test

The successful trader on Instagram is a sample of one drawn from a pool of ten thousand. The other 9,999 abandoned their accounts and went back to work. The IG algorithm does not show the ones who went back to work. The trading course teacher does not talk about them. We build our worldview on a curated selection that is, by definition, statistically unrepresentative. And we know this.

The people I’ve met who run these two strategies are intelligent and genuinely knowledgeable. Many of them are doctors, engineers, consultants, founders. If you sat down with them and explained the flaws, they would nod and they will understand the risks of investing in this manner.

Yet they would go back to their portfolio and do the same thing especially at times when the strategy appears to be working.

So why do people want something, but they still do the opposite

Talk to the same people about FIRE. Financial independence, retire early. The movement that explicitly says the goal is freedom from financial obligation. Lower your costs, raise your savings rate, escape the salary trap and live from your investments.

Many of the people in the two-bucket portfolio also love FIRE.

And yet they still want to expose themselves to at least a decade long financial commitment with an uncertain (adjustable) expense that could elevate their future living expenses dramatically.

a.k.a.

They still want to buy a house.

Not just any house.

The house their peers, family, relatives, significant other would approve. These people believe in financial freedom. They want it. But they also sign the largest obligation of their lives, because of a widely held wisdom which is:

properti pasti naik (property will always go up)

Despite the elevated property prices especially high relative to average income earners. Despite the weakening demand. Despite the low yield. Properti pasti naik.

Therefore, they lock themselves up to a huge financial obligation. In their minds, if they need the money suddenly, they could always sell that property. Yet they also know that property is not a liquid asset.

Note: I am not saying that all property investments are bad here. But the kinds that I am hearing certainly is. The people I’ve heard about like to punch way above their pay grade to take up a huge mortgage and commit themselves to a long tenure of financial obligation.

So the puzzle seems to have layers here. Not just in investing in stocks.

I am seeing genuinely smart people making suboptimal decisions when they have rational knowledge that their decisions aren’t optimal. I mean you can agree with me that information is not the bottleneck here.

Even compute and analysis is not. You can ask Claude to make a financial projection for you for your purchase of a recent home. You could estimate your living expense and run simulations having 1, 2 or 3 kids now. If you don’t do it with AI, you could do it with a certified financial planner too. Its no longer expensive nor inaccessible for you to have this knowledge.

So something else is doing the steering of behaviour here.

What an academic says about this

Robert Cialdini is a social psychologist at Arizona State University. He is widely known for the book Influence: The Psychology of Persuasion, has sold millions of copies (hopefully I can sell that many books next time 😝). It is the foundational text on what he calls compliance. On how people behave.

There are many conditions in which people would be “obliged” to behave in a certain way. They are reciprocity, liking, consistency, authority and social proof. His point on social proof finally gave me some sense for what I have been watching all this while.

Cialdini starts with a grim example.

The name was Kitty Genovese in 1964, New York. She was attacked by a man that lasted for more than half an hour, outside her apartment building. The craziest thing is that there were 38 neighbours who saw or heard parts of it from their windows. None of them came down to help Ms Genovese. None of them called the police until it was too late. The journalists who wrote about it framed it as a story about people becoming more individualistic in New York. To be honest, I would probably write the same thing if I were to be the journalist.

But Cialdini does not agree with this. He argues that witnesses were not apathetic. They were reading each others’ action. Each person looked out their window, saw nobody else doing anything, and concluded from that absence: this must not be the kind of emergency that requires my action. The more witnesses there were, the less likely any single person was to act — because each additional silent observer made the social signal of inaction louder.

This case became very famous it earned a name called the bystander effect. When you don’t know that the situation is truly an emergency in public, you’d rather be a bystander and not do anything.

That is what Cialdini means by a behaviour that is driven by social proof. It activates under two specific conditions:

1. uncertainty about what to do,

2. and similarity to the people around you.

The Genovese witnesses had both. They were unsure what they were seeing — was it a domestic dispute, a prank, a real attack? The fact that the event occurred in the evening was a clear smoking gun. They assumed the other watchers were people with judgement roughly like theirs. So they used each other’s inaction as evidence. And Ms Genovese died in the incident.

To me the negative example showcases how powerful this social proof effect is in our behavior.

Another more grim example, we might think that we love our lives. But Cialdini puts forth another example of Jim Jones’s Peoples Temple in Guyana, 1978. Cult members voluntarily commit suicide by consuming poison after they receive the instruction from the leader and after seeing that others oblige. Over 900 people died that day. Many of them watched their neighbours drink first.

If people can kill themselves because of social proof, I’d bet my money that people can buy a house because of social proof 🙂

So here’s what you can observe

Remember that brilliant cousin or university friend you met for dinner the other day? Perhaps they announced that they have just recently purchased a house. An expensive one that is beyond their pay grade at the moment. They say that they take a 20 year mortgage to finance the house and the rates are adjustable after 5 years of instalments.

How do you feel?

Maybe after reading this blogpost you’d feel like:

Naahhh, I don’t want to buy that kinda house. That ain’t me Toby.

But imagine you meeting your parents and they naively ask,

David, when do you plan to buy a house for yourself? You’re not going to rent forever are you?

How would you feel then?

It is a similar innocent question as to when do you plan to have kids? Or when do you plan to get married? They mean well but it is still a loaded question. A question that requires deep thinking and deliberation instead of one that has to be acted upon after seeing your friends’ answers to them.

Stock investments is the same

How would you feel when you know that your smart friends are owning big banks alongside you and DCA-ing S&P500 religiously?

Remember that stock investment has 2 things that social proofs tend to work on really well:

1. Uncertainty

2. Similarity of what people around you do

You very well know that you don’t know what the US president would say or do tomorrow. And today you brush aside his brash actions because your AI stocks are still bringing in acceptable returns.

But what if they stop delivering the returns you’d expect?

How would you feel?

Will you be blaming the US government? Or will you realise that perhaps copying what others are doing at the moment is not the most prudent investment decision?

Most will do the former. A tiny minority will do the latter.

So arming yourself with knowledge is not enough

Social proof is a powerful force and I think that rationality is not a strong enough weapon to overcome it. I believe that combatting conformist thinking requires you to change which signal is closest to you.

Practically, that means three things.

First, get explicit about your own goals on paper, in your own words, before considering anyone else’s opinion. Read them more often than you read your family or friends’ WhatsApp group.

Second, surround yourself with at least one person who is a contrarian, and whose opinion you cannot easily ignore.

The third thing is ask yourself:

“Would I still want to own this thing if nobody knew?”

If the answer to owning something is yes based on your calculations and honest preference, then it is likely a good deal. But if the answer is no; then you probably have to think twice or thrice before pressing that buy button.

It is one thing to buy because you want it for your own interest (be it present or future). And it is a completely different thing for you to buy because you want to fit in.

Knowing the difference helps.

Cheers.

According to her, our experience building Recompound is very relatable to people who are also trying to build their own business.

So here goes. Raw, unfiltered, things presented just as they are.

Hope you enjoy.

Disclaimer: this post is for educational purposes only. It does not constitute investment advice. It is the author’s opinion regarding new export policy that is recently announced and it attempts to consult recent data that is not usually covered otherwise. We hope that our readers could learn a thing or two from the data that we are seeing and we continue to welcome level headed and constructive discussions. Lastly, if you are a client, check your investment thesis in the client dashboard. This write up has been incorporated there also.

This week, per this post is published, President Prabowo announced a PP (Government Regulation) mandating that all exports of palm oil, coal, and ferro alloys must be routed through a designated BUMN (”single-gate” system).

There are very differing opinions regarding this ambitious new export policy amongst pundits on social media. So we’d like to offer our thoughts being investors in the Indonesian capital market ourselves.

First things first

We refrain from making sensationalist comparisons between this current policy vs other failed government policies in the past like BPPC where clove farmers had to sell cloves at a fixed price to a single government entity during the Soeharto era. Just like how we refrain from rushing to extrapolate that a weakening Rupiah necessarily means that a Rupiah is going to fall in the fashion we see in 1998.

We believe that conditions of the present economy, technology, regulation, information and even governments are not the same. Therefore, just because there had been a precedent, it does not mean that it is definitely likely that we will see a repeat of history. For example, it goes without saying that just because we saw COVID in 2020, does not necessarily mean that Hantavirus will cause economies around the globe to see lockdowns again. It might go on a lockdown for real (I am not a public health expert here), but it does not have to be.

And of course this is not an investment recommendation or financial advice. We are first to disclose that we are not a macroeconomics policy expert too.

Why choose to not make sensationalist claims?

Well to be frank because we are not a media company. We don’t get anything from making sensationalist claims. If you like our Substack post or put a comment, we don’t get paid. And getting famous on social media is not our business model. So there’s almost 0 point for us to make such sensationalist alarmist/euphoric claims.

What helps is for us to analyse and reflect on the data that allows us to make an informed decision as a long term investor. Knowing fully our limitations about what the outcome will be in the future.

Don’t get me wrong, it is possible that the policy could be a huge failure. But I stress that it is not wise to dismiss the likelihood of it being successful as well. And we ought to look at both the negatives and the positives as it is so as to stay well informed of what is going on.

Now with that out of the way, let’s get into it.

Key Pointers and Questions

What does the policy actually do?

All exports of three commodities (palm oil US$23B, coal US$26B, ferro alloys US$16B — total US$65B/year, ~23% of national exports) must go through a designated state owned enterprise (BUMN). Implementation is phased: Phase I (June–August 2026) is documentation only — transactions still go directly between company and buyer, while Danantara Sumber Daya Alam (DSI) validates volume, pricing, and delivery. No margin taken. Phase II (September 2026+) is full implementation where DSI takes over all contracts, shipments, and payments. This is the phase the market fears — and the one that may involve a margin.

How massive is the underinvoicing problem?

Based on UN COMTRADE data, cumulative export underinvoicing from Indonesia over the past 34 years (1991–2024) totals US$908 billion (~Rp 15,400 trillion).

For reference, BCA’s profits is about Rp 50 trillion per year. So that 34 years of under invoicing is about 300 years of BCA’s net profits 🤯. Indonesia hasn’t even been around that long. We are in the less than 100 years club (1945 independence).

This is the gap between what Indonesia reported as exports versus what trade partners reported as imports. This isn’t an estimate — its recorded international trade data. From this, I think that the government’s motivation is clear: make more money (increase state revenue) by plugging a leak that is worth hundreds of billions of dollars that has been flowing offshore (primarily to Singapore).

If underinvoicing is successfully curtailed, more USD stays in Indonesia — which is also a positive for the rupiah.

The mechanism is textbook transfer pricing: an Indonesian producer sells to a Singapore affiliate at $100 (below benchmark), the affiliate re-sells to the real buyer at $150 (market price), and the $50 margin stays in Singapore. Finance Minister Purbaya himself stated that he randomly checked 10 CPO companies and selected 3 ships at random — all showed prices to Singapore at half the price to the US. That’s how widespread the practice is.

Who is being targeted?

Not “clean” mining companies — the target is companies that have been invoicing below benchmark. A documented example: Global Witness reported that Adaro used Coaltrade Services International (Singapore) to sell coal below benchmark prices from 2009–2017, avoiding an estimated ~US$125 million in Indonesian taxes. Profits were booked in Singapore (tax rate ~10%) instead of Indonesia (~50%). This policy forces all invoices to be verified against published benchmarks by the BUMN. If a company has always been invoicing at benchmark, the impact is minimal — it’s just a compliance layer.

Who benefits from the practice of under invoicing?

It follows that the ultimate beneficial owners (UBO) who owns the trading companies will benefit. The spread between the sales to the trading company and the sales to the importer gets to stay in the private trading companies (likely owned by majority shareholders).

Are there overseas precedents?

There are many, among others:

• China: Operates a “designated exporter whitelist” for strategic minerals (tungsten, antimony, rare earths). Only government-approved companies can export. Effective for revenue control, but China’s rare earth export quota was struck down by the WTO in 2014 after Japan/US/EU challenged it.

• Malaysia (PETRONAS): Full state control of oil & gas since 1974. Private companies (Shell, Exxon) still operated profitably via Production Sharing Contracts, but long-term investment declined ~40% from peak.

• Chile (CODELCO): Copper was nationalized in 1971, but later opened to a dual-track model — state-owned CODELCO alongside private miners (BHP, Freeport). Private companies remained profitable, just paying higher royalties (up to 46.5%).

• Zambia: The cautionary tale — full nationalization of copper in 1969 ended in disaster. Copper prices fell, the state entity couldn’t respond quickly, mining’s GDP contribution collapsed from 33% to 7.7%. Zambia eventually re-privatized.

Are there more recent local precedents?

PT Timah (TINS) — the most directly relevant domestic precedent: PT Timah formed an internal task force to crack down on illegal tin miners in Bangka Belitung who had been paying zero royalties and zero taxes. The result? Q1 2026 net profit surged

+1,184% YoY

(from Rp 117B to Rp 1.5T), revenue +160%, GPM expanded from 18% to 39%. When the playing field was levelled and illegal competitors were removed, the legitimate company’s profitability transformed overnight. This is the same mechanism as the export rules — when underinvoicing is eliminated, clean companies actually benefit.

There’s another one, Indonesia’s own Berdikari (single-gate importer for SBM / soybean meal): caused short-term price spikes during the transition period, but poultry and feed company GPMs actually improved once stabilized. Although we don’t see this positively (more neutral) because the GPM could also be pushed by increasing poultry prices and volume because of free meal for children policy.

Why is market dominance the key to this policy working?

The logic is straightforward: if buyers (foreigners) have sufficient alternative suppliers, they switch — the added bureaucracy from the BUMN gate becomes a reason to buy from another country. But if the country is dominant and buyers don’t have adequate substitutes, they have to comply.

Indonesia has this leverage in all three targeted commodities: — Thermal coal: #1 exporter globally, 40%+ of global seaborne supply — Palm oil: #1 exporter globally, 55%+ of global supply — Ferronickel: #1 exporter globally

India, China, Bangladesh, and Vietnam cannot replace their entire Indonesian coal supply with Australian or South African alternatives overnight — the volume gap is too large. So as long as the BUMN fee and bureaucratic friction remain reasonable, buyers will continue purchasing from Indonesia. This is fundamentally different from Zambia’s failure — Zambia held only ~7% of global copper supply, so buyers immediately shifted to Chile and Congo when ZCCM’s bureaucracy became burdensome.

Who is the designated BUMN intermediary?

While this can change from time to time, but the one that is identified now is PT Danantara Sumberdaya Indonesia (DSI) — a brand new entity incorporated on 18–19 May 2026 (one day before the announcement), under the Danantara sovereign wealth fund.

Luke Thomas Mahony has been appointed as Direktur Utama (CEO) of DSI. He’s an Australian national based in Jakarta, previously SEVP Business Performance & Optimization at Danantara Indonesia since September 2025. Before that, he was Chief Strategy and Technical Officer at PT Vale Indonesia (INCO) from July 2024 to September 2025, and held senior roles at Vale Base Metals (CTO, then Director-Technical). His earlier career includes stints at BHP Billiton and Xstrata Coal.

The bull case on Mahony: he’s a technical mining person with hands-on operational experience across multiple countries. He understands how multinational mining companies structure their export sales — which is exactly what DSI needs to police. Being a foreigner may also make him more neutral from domestic political pressure.

The skeptic’s case: you’re appointing a former Vale executive to oversee an entity that will audit companies like Vale. Vale Indonesia (INCO) sells ~80% of its output to Vale Canada Limited — this is exactly the kind of related-party export structure that DSI is supposed to scrutinize. There’s probably a potential conflict of interest.

To us, the appointment of a legitimate mining professional rather than a bureaucrat is a net positive signal for execution quality. It partially de-risks the “Zambia scenario” where incompetent state management destroys the industry. But the conflict of interest angle is worth monitoring — especially how DSI handles ferro alloy exports from IMIP/IWIP smelters and nickel shipments.

How will DSI make money?

Danantara CEO Rosan Perkasa Roeslani has confirmed that DSI will operate as a trading company — taking a margin from every export transaction. This is not a flat administrative fee; DSI will function as a commercial intermediary that buys from producers and sells to international buyers, capturing the spread. The stated purpose: to eliminate both

underinvoicing

(selling below benchmark to affiliates offshore) and

overpricing

(inflating import costs to shift profits out).

Where do we see the risk?

How could this hurt miner’s profitability? There are two main risks that we are thinking through:

Risk 1 — DSI’s spread margin is too large. Rosan has confirmed DSI will take a margin — so this is no longer hypothetical. If DSI’s spread exceeds whatever average selling price (ASP) improvement comes from eliminating underinvoicing, the net effect to the exporter is negative. Concrete example: if a miner’s operating profit margin is ~20% and DSI charges a 4% fee on revenue, that’s a ~20% drawdown in operating profit. That’s not trivial.

Risk 2 — DSI becomes a monopsonist and sets buy prices too low. As the sole intermediary, DSI has the power to dictate purchase prices from producers. Exporters lose bargaining power, and profit gets squeezed from the revenue side. This is a bad precedent for foreign investment too — the UNTR’s Martabe gold mine saga shows how this can backfire.

Two scenarios — Scenario 1 (Base case): Nothing changes materially. DSI takes a minimal fee during Phase I, Phase II implementation is gradual, and earnings are broadly unaffected. This is more likely for Phase I (June–August) which is just documentation. — Scenario 2 (Downside): DSI charges a meaningful margin. At a 4% fee on a 20% OPM business, you’re looking at ~20% profit drawdown. This needs to be watched closely once Phase II terms are announced.

The positive case — what if this actually works?

Finance Minister Purbaya’s view during his recent Konferensi Pers is bold: “Profitability must double. It’s time to buy.” Here’s the logic:

1. Win-win if executed well. Underinvoicing is eliminated → ASP rises to true benchmark levels → the government collects more tax and royalties, and minority shareholders get the earnings that were previously leaking to Singapore trading affiliates. More USD stays onshore in Indonesia, supporting the rupiah.

2. Single seller = higher negotiated prices. If DSI successfully consolidates Indonesia’s negotiating position as the world’s #1 exporter of these commodities, it could push global prices higher. Companies with good GCG (corporate governance) benefit directly — they weren’t underinvoicing anyway, so the ASP uplift is pure upside.

3. Non-GCG companies are definitely hurt — and that’s a good thing for the market. Companies that were playing the underinvoicing game lose their edge. The playing field levels. This is structurally positive for investors who pick well-governed companies (which is exactly what we do).

On balance, we believe the positive outcome outweighs the negative

But we are still cautious because there is execution risk. Although Danantara’s recent success in cleaning up PT Timah (TINS) delivered a +1,184% YoY net profit surge is a data point, more successes that lead to better profitability would prove crucial.

Insider signal: does the owner already know before the information was released?

Well we don’t know for sure and I sure wish I could give Mr Anthony Salim a quick call, but a couple months back, our president Prabowo did have a meeting with the conglomerates who own large mining sites. It wouldn’t be far fetched to assume that the bosses have also be briefed regarding the government initiative. Especially the topic of discussion as reported by Kompas: “Indonesia Incorporated”.

And another data point was around this time — when the market was fearful — TP Rachmat increased his personal stake in ADRO from 3.02% to 3.46%. That’s +125.4 million shares added.

Think about that. We acknowledge that this is pure speculation, but the conglomerate owners (TP Rachmat, Boy Thohir, Anthony Salim) likely had early visibility on this policy direction before the public announcement. These are individuals who attend the highest-level government briefings. If TP Rachmat — who would have understood the implications better than anyone — chose to buy more shares rather than sell, that’s not a signal to ignore.

Call it what it is, but to us, insider buying is a bullish signal especially in the middle of widespread market fear. And to be fair, we will check again in the coming months if there are more insider buying or less of insider buying.

But let’s remain level headed and not get carried away. We are investors with limited information anyways. Here are key questions that still need answers:

1. What will DSI’s margin be? Rosan mentioned DSI will take a trading margin, but the rate hasn’t been disclosed. Below 2% = manageable. Above 3% = needs re-evaluation. Worst case: At 4%+ on a 20% OPM business, we’re looking at a ~20% profit drawdown.

2. How will settlement payments work? If DSI adds 30+ days to the payment cycle, working capital can get squeezed.

3. How will DSI handle potential conflicts of interest given its leadership’s prior roles in the mining companies it’s meant to oversee?

Closing Remarks

We will continue monitoring every development that affects the profitability of companies in our portfolio. We think that it is helpful to have every investment decision made based on the best available information at the time — like assembling a puzzle, one piece at a time.

Being right or wrong on any single call is part of the investment business. What matters is that the process is disciplined, the logic is clear, and we keep learning from every outcome. Do our best and never get attached to the outcome. Especially when the outcome is really beyond our control.

I love this question.

Despite:

• transparency over the investment thesis we share

• long term incentive alignment between us and customers through a performance-based fee structure

• and the buying and selling executed independently in our customers’ own brokerage accounts

someone still asks this.

To me, this signals that I look rich (having so much equity that I need exit liquidity) even with my $10 Casio watch haha.

Jokes aside, this signals a deep-seated trust issue that is prevalent in the finance sector. And I get it. Finance has been mired with scandals, shadow banks, frauds and so on.

So instead of being defensive about the question, I want to take it seriously. Let me walk through it the way I’d walk through it in my own head if I were to be a potential customer myself.

First, what is exit liquidity?

You can skip this if you already know what is an exit liquidity.

Tom wants to sell his house in Senopati, South Jakarta. The house is riddled with spooky ghosts.

He bought it five years ago for Rp 17 billion (about USD 1 million). Now he wants out. He finds an agent. “Mr. Tom, I can realistically sell it for USD 800k.” Tom is not satisfied. “I want at least USD 2 million. I’ll double your commission if you pull it off.”

The agent nods and goes to find a buyer.

Not someone who knows Senopati — those people know about the ghosts and won’t pay anywhere near USD 2 million.

So the agent finds John, an American moving to Jakarta to work in SCBD as an expat. The agent pitches the area, the food, the hotels nearby.

John was convinced and John becomes Tom’s exit liquidity. John was ignorant about how bad the place was and he later finds out about the ghosts. He got furious. But the transaction is done.

Tying it back to stock investing, customers as exit liquidity essentially means that they are unsuspecting buyers (retail investors) to purchase our shares of a company at elevated prices.

Back to the question

So, are our customers our exit liquidity?

The answer is: no.

But “no” is not a good enough answer. Because again, it requires trust on your part toward the service provider. And the whole reason you’re asking the question is that you don’t want to extend that trust on faith.

So instead, let’s assume that yes, we want you to be our exit liquidity. That’s why we have a marketing budget. That’s why we work hard to acquire you.

What’s the cost? What’s the reward of doing it?

The reward

Let’s start with the reward and be generous.

Assume that we work with a public company with a market cap of Rp 1 trillion. They want to sell 1% of their shares to the public.

That means they want to sell Rp 10 billion worth of their shares.

Now let’s be generous. Let’s say that our commission is 50% of the sales value. Which means we will get Rp 5 billion. Big, ya?

Note: I want you to understand that if we actually did this, it would be a crazy unethical thing to do because of the conflict of interest. But let’s just play along.

So how many customers do we need to sell Rp 10 billion worth of shares to?

Our minimum committed fund is Rp 25 million. But let’s say that on average our customers have Rp 250 million (10x the minimum).

Also note: If Budi recommends to all-in on one stock, that would be disastrous.

But let’s run the cartoon version of the math first. Rp 10 billion / Rp 250 million per customer = 40 customers.

40 customers and we walk away with Rp 5 billion. That looks like the reward.

Except “all-in on one stock” is not a recommendation any responsible advisor would actually make. Because the customers will just say “no, I don’t want to all in”.

In reality, a position size for a single stock in a concentrated portfolio is more like 10%. At a 10% position size, each customer takes Rp 25 million of this one stock.

Rp 10 billion / Rp 25 million = 400 customers.

So the real number is 400. Four hundred of our customers, all simultaneously convinced that the same one stock deserves a tenth of their portfolio, all on the recommendation, all at the same time. That is what it would take to clear Rp 10 billion of inventory.

Now let me size that against our probable size of a customer base.

Let’s be generous again. Assume that 10% of our Instagram followers (16K) are customers — that is already a wildly optimistic conversion rate. That puts our customer count at around 1,600.

400 out of 1,600 = 25% of our entire customer base.

A whopping quarter of every customer we have ever earned, burned in a single transaction.

But at least we walk away with Rp 5 billion?

But how big is Rp 5 billion, actually?

Let me be honest about the reward side, because Rp 5 billion sounds big in isolation.

If I split it 50-50 with Budi, with no marketing budget set aside, we each walk away with Rp 2.5 billion.

We also have VC investors. I am pretty sure they would be super pissed if they only got a fraction of that Rp 5 billion. So another chunk of it gets eaten there.

And here is the part that actually grounds it for me personally.

If I worked at Goldman at Rp 100 million per month, I would earn Rp 2.5 billion in roughly two years. With none of the conflict-of-interest baggage. With no risk of losing 400 customers after a single transaction.

So the “reward” is not a transformative number. It is roughly two years of a Goldman salary, paid out once, in exchange for torching the business we are trying to build for the next few decades.

When you frame it that way, the question isn’t “is it ethical?” — that one is easy.

The question is, “who would even take that trade?”

The cost

Now the cost side. There are three layers.

Layer 1: the direct churn.

Those 400 customers are gone. Not in the polite, “I need this money” way. Its more of “I will never speak to Recompound folks again” way.

Because in their minds:

“Ni gara2 pake Recompound gua jadi nyangkut di saham X.”

Whatever lifetime value we would have earned from them — gone. With performance-based fees, the customers who stay the longest are the ones we earn the most from, because compounding does its work on both their portfolio and our fees.

400 customers, over a decade of fees, is not a “meaningful chunk of revenue.” For a business at our stage, it is the business.

Layer 2: the word of mouth.

Indonesia’s working professional / entrepreneur community is small. Really small. It runs on dinners. On WhatsApp groups. On who-introduced-whom at last weekend’s BBQ.

400 burned customers don’t leave quietly. They tell people.

Let’s say each one tells 7 people. That’s 2,800 prospects who have been pre-poisoned against us before we ever get a chance to talk to them.

Whats more, the prospects who matter most to us, the analytical, slow-to-trust, asks-hard-questions kind, are exactly the kind who weigh peer testimony heavily. Negative word of mouth doesn’t just lose us 2,800 names on a spreadsheet. It loses us the 2,800 names we most wanted on the spreadsheet.

CAC for that kind of prospect is not cheap. It is some of the hardest customer acquisition work in finance, because the kind of person who can tell the difference between a real fiduciary and a salesman is, by definition, hard to sell to.

Layer 3: the moat.

This is the worst one.

Trust is a moat, especially in finance. Trust is so much of a moat that, arguably, it is the product. The portfolio is the deliverable. The trust is what makes the deliverable worth anything.

Once we burn it, we cannot unburn it.

Every competitor who comes after us will use this story against us forever. “Remember when Recompound did X to their customers?” It becomes the first thing anyone hears about us.

So the full ledger:

• Reward: Rp 5 billion gross — which, after splitting with Budi and accounting for our VC investors, lands closer to two years of a Goldman salary per person, paid out once

• Cost: a decade of LTV from 400 of our hardest-won customers, plus ~2,800 poisoned prospects, plus the permanent destruction of the one moat that actually matters in this business

I suppose it is quite clear that the economics does not make sense 😅

But I insist that this question is a good one

Three things stand out to me when someone asks this question.

First, the fact that you’re even asking is a signal.

It’s a signal that you think analytically and that you don’t take claims at face value. That you want to see the incentives before you trust someone.

These are exactly the kind of people we want as customers. They are also, frankly, the hardest kind to acquire. Burning a customer who asks this question, just to make a one-off windfall, would be a fatal economic decision.

We work real hard to find you.

Second, one-off deals are bad fundamentals.

The kind of business we want to build is one where customers stay for a decade because economically speaking, the longer they stay, the more valuable they become to the business. Where their fund compounds. Where they invite their friends. Where they introduce their kids to us when their kids start earning.

That kind of business does not get built on burning people. It gets built on showing up well, repeatedly, for a very long time.

A one-off Rp 5 billion windfall, paid for by destroying the customer base, is not a business. It’s stupidity at its finest.

Third, again trust is a moat. Why would we destroy our own moat?

In a finance business, trust is the only durable competitive advantage. Anyone can buy the same Bloomberg terminal. Anyone can hire CFAs. Anyone can build a model. What they cannot easily buy is a decade of customers who say,

“These guys did right by me even when no one was looking.”

If we make you our exit liquidity, we are voluntarily destroying the only thing that makes Recompound worth more than the sum of its salaries.

So the answer is no

But more importantly: the answer is no, and you don’t have to trust me on it.

You can do the math and see if it makes sense for yourself 🙂 If you fancy, you can even play with the variables and assumptions on the cost and rewards that are made.

That’s the answer I’d rather put out. Because I believe that trust in finance shouldn’t be a leap of faith. It should be what’s left when you’ve done the analysis and the analysis still says: this is aligned with my goals.

If you’re the kind of person who wants to do that analysis — welcome. You are exactly who we built this for.

Disclaimer: this exposition is an editorial. It is not an academic paper and it should not be seen as facts. It is based on the author’s personal observation and experience. It does not constitute investment advice and is for educational purposes only.

I was playing with Instagram story and created a poll on Recompound. Market was horrendous. And I wanted to check our followers how worried they are about the market. It turns out that a good 30% of the responders were feeling worried. So perhaps this write up could be timely.

We all know where does the worry come from

First things first, we know that worry tends to come from the market itself. The modern stock market is not a place where real ownership exists for many people. It is a place where price movements gets sent to you every moment. Push notification, tickers, news commentary, now we have streams and social media, WhatsApp and Telegram groups among others.

You already know all of this. But it is important to repeat that the market ties you with a rope by design. You get pulled into it the moment you sign up.

Simple observation:

When you see the news saying anything good or bad about the macroeconomics, geopolitics, government policies, etc in whichever outlet, you would be pulled towards checking your investment application. You’d be wondering how does this news or event affect my investments?

This is normal because the market is in the business of transactions, and broadcasting prices loudly is what gets people to act. Feeling something every time prices move is what gets you to trade. Trades generate fees. Fees feed the entire chain.

So knowing where the worry comes from is a good starting point. But it is not good enough cure.

Where does this worry normally lead to?

Worry or anxiety is an unpleasant feeling. So when we are worried, we naturally seek for stimulus or information that counterbalances that worry.

How we do this is usually we search for answers to questions that we have in mind. Among others, these are primary questions that people have in mind:

1. How long is the market going to stay horrendous?

2. Have we made the right investments. I.e. will our holdings recover as the market gets better?

When these two answers are reliably answered by someone we trust, we’d find ourselves comforted. Someone we trust can literally be anyone:

1. It could be your financially savvy brother

2. It could be your capital market friend

3. It could be a YouTuber investment guru

4. It could be an intelligent guy on Substack

5. It could be your super smart colleague in the private equity space

6. You get the idea

While the claims and predictions from people you trust can make you less worried, I argue that the pleasant feeling tends to be fleeting. It is more similar to a pain killer that removes the pain. But likely that the underlying “sickness” is not attended to effectively yet.

Assume that my argument is valid, why do we look for claims and predictions as a natural response to cure our anxiety?

Let’s have a bit of fun first 🙂

Chris Williamson recently interviewed a mentalist named Oz Pearlman. Oz would guess names, birthdays, pin code or numbers they were just thinking about. By the end of the episode, Chris’ face was as red as cherry as he was defeated and felt like he’s a prey in water. Take a look, video is funny.

How do you feel after watching the video? I made Budi and Wilson watch the video at the office. Wilson was entertained. Budi was skeptical. Budi is way too analytical for this stuff. He thought that it must be set up. But I did observe both of them while they were watching. Both of them sat there momentarily in disbelief.

So why does this work? Why are we so intrigued?

Btw, if you say don’t feel anything, please stop trolling me haha.

My guess is because our brain is designed to assume that minds cannot be read. We go through our life thinking that our mind is private and ours alone to observe. What we consciously do not divulge, do not get known by others. So when somebody appears to be able to read minds, the experience feels like breaking law of nature.

Now let’s sit with this for a moment and draw parallel to investments.

Similarly, our brain is also designed to assume that markets cannot be predicted. So when a statement that “looks like” a prediction turns out to be true, the predictor becomes like a wizard. To wit, you could take a look to a conversation that was perceived to be a prediction.

But I promise I didn’t make a prediction. I didn’t say that a certain stock would go to a certain price. I just said that we would be more interested if the price becomes lower.

So here’s what I hope is clearly demonstrated. The feeling when you see a magic trick is very similar, nigh the same as that feeling when you see a market prediction comes true.

That feeling is called wonder.

Wonder is a pleasant feeling when what you thought was impossible, happens. Then you would continue to want to know how on earth does that impossible thing happen.

A mentalist sells wonder through the illusion of accurately reading minds.

An investment person sells wonder through the illusion of accurately predicting the market.

That is the trick.

So what have we learned from watching a mentalist in action

We’ve learned that:

• Worry comes from the market movement

• Our tendency is to search for people we trust who have previously predicted the market because it has generated a sense of wonder.

• When that trusted person makes another prediction that is in favour of your investments, you feel less worried.

So I hope that it is now clear why we naturally look for prediction from people we trust. Because accurate prediction generates wonder → which creates trust. The next time we feel worried, we would go again and seek for the prediction to alleviate our worries.

But here’s what Oz tells us about his tricks

Literally the first question of the podcast:

Chris: “You have said that your career is built on a lie. What’s the lie?”

Oz: “The lie is that I can read people’s minds. I can’t. I wish I could.”

Oz would be the first to tell you that what looks like mind-reading is a very sophisticated set of guesses, observations, and probabilities. The miracle, when you understand it, is just craft of creating an illusion that he’s reading minds.

He has nothing to gain by saying this. He says it anyway, because he is honest about what he does.

And I’d say this to everyone who happens to trust us as well. We believe that the market is too random to make predictions accurately 100% of the time. So we’d rather not make predictions. So when people ask:

“when is the market going to recover?”

“is the currency depreciation going to get better?”

“how is war uncertainty going to impact the market as a whole?”

We will be first to say that we don’t know.

But when we say we don’t know, it does not mean that you must be worried, quite the opposite

There is a neuroscientist named Lisa Feldman Barrett who wrote a book called How Emotions Are Made. She argues something that, if you take seriously, changes how you think about your own worry.

Her argument is that emotions are not universal, hardwired things that just happen to you. They are constructed in your brain, in real time, from concepts. And the most powerful thing is that concepts are not just built from experiences that you observe, but also from words of our language.

Human brains are very unique in the sense that we can construct a reality in our minds using words. The existence of reality depends on a perceiver. Think about it. What is money in a physical sense? Well it is a bunch of bits (or code) that is stored in the database of a bank. If you wire Rp 100.000,- to a monkey, the monkey will not feel a thing because that money to them is not real. But if you send that Rp 100.000,- to Toby. Toby will say thank you and go to Indomaret and buy a box of sparkling water. Indomaret would give Toby that box of sparkling water because they agree with Toby that his Rp 100.000,- is real.

Another example how we construct reality using words.

When you order from Domino’s, you do not say:

“Hi, I’d like a circular flatbread with tomato sauce, melted cheese, and pepperoni baked at 220 degrees. Please.”

You’d say:

“Hi, I’d like to order a pizza. Please”

The word makes the thing real, orderable, recognisable and most importantly gojek-able right to your door step. Without the word, you would have to assemble the concept from scratch every single time, and most of the time you would not bother.

Here is what this means for your worry. How to feel less worried

Barrett’s claim is that emotions as a reality are constructed the same way. The bodily sensations are real — a tightness in the chest, a tug in the stomach. The worrisome thoughts that occur on our mind is also real — I wonder if my investments are going to recover tomorrow or next week.

To feel less worried, you need a competing and name-able state. Btw, naming it is very important as I previously discussed the importance of names.

But what is a good competing state? One that has its own name? Without a competitor, worry has no rival and keeps running. We’d be looking for predictions to make us feel less worried and that’ll feel like a coping mechanism, rather than a long term solution.

Here is the awkward part. Many when asked to name the opposite of worry, would go for calm or peace of mind. But I think that these words are not good enough.

Calm does not specify what changed. If I tell you I am worried, and the next week I tell you I am calm, you do not know why. Did I take a holiday? Did I medicate? Did I just have a good day? Did the market rally? Did I sell everything? Calm describes a state of feeling without describing the underlying condition that produced it.

Peace of mind is something most people associate with a vague spiritual state. It does not feel achievable on a Tuesday morning while you are running a business.

Wonder, in case you are wondering, is not a good opposite too. Because wonder only happens when the predictions have come true. If I hear Warren Buffett says that market would recover, I’d feel wonder only after the market really recovers. Not when it takes another 5% plunge on a Tuesday morning.

So we need a different kind of opposite. Something more specific than calm. More human and grounded than peace. More present than wonder.

Let me propose a word.

Untethered

Most retail investors live with an invisible rope. The rope ties you to the price chart. You do not know the rope is there until it tugs — your stock dips and you feel it in your stomach, your stock jumps and you feel a sudden rush. You check at 9am. You check at lunch. You check at end of day. The rope is taut all day.

Untethered is the day the rope falls off.

It does not feel the same as detachment. Detachment is something you do — a discipline you cultivate. Untethered is something that happens to you once the conditions change. The rope falls off because it is no longer pulling on the other end.

The chart is still there though, just that you are not pulled towards it.

Level of worry depends on who is holding the rope

I want to be honest with you about why feeling untethered is, for me personally, easy.

It is not because I have superior willpower or a better meditation practice. It is because someone else is holding the rope. I use Recompound with whom I can interact with a dedicated advisor. So does my mom. So does my brother. None of us think about the market day in, day out — not because we are unusually disciplined, but because we know there is a dedicated advisor whose actual job is to do the watching. The pull is gone because it has been outsourced to people I trust.

That is the entire mechanism. There is no trick.

Which means the answer to what to do tonight if you feel worried depends on a question only you can answer.

Right now, if I did not look at my portfolio for three months, would I trust that someone is still watching it on my behalf?

If your answer is yes

if you have a wealth steward you genuinely trust, whether that is a finance-savvy sibling, an advisor you have worked with for years, or anyone else competent and aligned with your interests — then here is what to do. Understand them more.

If your financially savvy sibling is taking care of your investments, ask them questions. Understand their philosophy, the assumptions they make, the learnings they have experienced, the mistakes. When you realise that work has already been done, the rope is loose.

If you use Recompound and you feel worried, talk to Budi 🙂

If your answer is no, here is what you can do.

Your worry is not irrational. It is your nervous system accurately reporting that you are alone with the responsibility, in a system designed to keep you tugged. The rope is doing its job. The work is to find someone trustworthy to hold the rope together with you. It can be anyone, but ideally someone honest, competent and aligned with your interest. With whom you can verify their work. Your advisor, your sibling, your uncle, anyone.

P.s. It can be yourself as well. I believe that if you consistently commit to a strategy that is logical and has been proven to work for you, you will likely feel untethered from the volatility of the market overtime too.

Closing remarks

That someone could also be us or it might not. Again, what matters is that the person on the other end of the rope is honest, competent and genuinely on your side. They really don’t have to be a superhuman or a market wizard. But I believe that if those three conditions are met, you can feel untethered. Whoever is holding the rope, make sure they are holding it on a bad week, not just a good one. Because if they don’t hold on to it, you’d end up getting tugged again.

A rope you can name is a rope you can take off. But you take it off by handing the other end to someone competent you trust — not by pulling harder against a system that is built to pull back.

That is what we are doing for the people who work with us. Not predicting the market. Not chasing the trade. But really by helping them hold the rope on their behalf.

So that’s it. While seeking for prediction is an okay solution to feel better for a while, perhaps you could attempt to feel untethered from the market by handing the rope over to someone who’s honest, competent and aligned with your interest. Subsequently, you will feel less worried from market volatility, getting your life back in the process.

Cheers!

References

In case you’re interested in reading or watching the materials consulted in writing this article, below are the entry points:

1. Barrett, L. F. (2017). How Emotions Are Made: The Secret Life of the Brain. Houghton Mifflin Harcourt. [The construction theory of emotion, the role of words and concepts in shaping what we can feel, and the argument that emotions are predicted, not triggered. The pizza-style examples of how language constructs ordinary categories appear throughout.]

2. Williamson, C., interviewer. How to Steal Thoughts Out of Anyone’s Head — Oz Pearlman — #1088. Modern Wisdom Podcast. April 23, 2026. https://chriswillx.com/podcast/ [Source for the “the lie is that I can read people’s minds. I can’t. I wish I could” exchange in the opening minutes, and the “it’s not about me, it’s about them” line later in the episode.]

3. Taleb, N. N. (2001). Fooled by Randomness: The Hidden Role of Chance in Life and in the Markets. Random House. [Background reading on why streaks of correct predictions in finance are usually luck rather than skill, and why we are so easily fooled into mistaking the former for the latter.]

Not only because of the well-known fundamentals — the earnings quality, the governance, the balance sheet — but let’s be honest. It is also because of the social angle.

There’s that subtle validation that comes from owning what every serious/respectable person around you owns. A lot of people I admire ended up with portfolios genuinely concentrated in BBCA, BBRI, BMRI.

To add, you can see that the damn thing is real. You could see bank branches, office towers bearing their names, and ATMs that are virtually everywhere across Indonesia. Even in cities that you have probably never even heard of, like Tambolaka, you see a Bank Rakyat Indonesia.

Mind you the behavior to buy and hold big banks is not easy. You need discipline to set aside your monthly income, month after month, into stocks whose price action is nowhere near as exciting as crypto or the smaller penny stocks that dominate your feed.

You need patience.

You need a certain degree of deafness to shield you from the loud noises of people screaming

“Gua cuan x% beli y”

”I am profitable x% because I bought y”

For years.

You cannot even begin to start counting in your head: how much money making opportunities have I missed because I refuse to listen to the guy next to me talking about these other sexy stocks.

Even worse is the recent drawdown — minus 30 to minus 40 percent from the all-time highs. Portfolios filled almost entirely with big banks. A thing that was assumed to be the prudent action, the prudent decision, turns out to be “blatantly wrong” within the span of less than a year.

All the hard work of slowly saving into a sizeable investment account — gone in a matter of months.

If that’s not painful, I don’t know what is.

Because it must be really difficult to do the prudent thing for so many years and get a less-than-ideal result. It almost feels like a fuck you in the face. From the market, — fuck you, you were supposed to be timing this, trading it, running speculative plays to generate superior returns. Or — fuck you, you should have just stayed in time deposits and risk-free rates.

And if you tried to go one step braver than the risk-free rate — if you bought the bonds of shadow banks offering a couple of extra percent — those shadow banks tend to go under. Another huge risk on your capital you never saw coming.

Painful stuff.

But strangely, watching this has led me to a realization about how I look at failure.

As a kid I was terrified of it. I’ve been getting scholarships, prestigious jobs, venture capital backing, whatever. But underneath all that, I also encountered plenty of failure.

The first time I failed badly was my SAT in high school. I cried for an entire evening because score came back under 2,000. I was so afraid I would end up at a bad university, so afraid I wouldn’t get scholarships. Disappointed in myself.

That night I told myself: I don’t want to fail anymore.

Obviously I didn’t learn anything back then.

A couple of years before we started Recompound, I had a co-founder split. Not catastrophic, but tough. It drained my savings. The opportunity cost was a seat at a prestigious bank. I was questioning everything — what am I doing.

I thought those two events taught me something. They didn’t. Not really. I still carried hesitancy toward failure.

The lesson only arrived when I started feeling it in other people.

Watching many disciplined people fail after doing the right thing. After trying to be a good person. After trying to be a fucking disciplined person who sets aside their monthly wages to plan their future properly, trusting in so-called safe instruments.

And here’s the strange part.

The ones who grow the most from this aren’t the ones who avoid failure — they’re the ones who refuse to resign to it. They suddenly care about why they invested in a stock. They care about the analysis behind it. They care about why a company is actually undervalued — not just whether a more financially savvy person said it was undervalued.

They stop accepting the prescription. They want to verify. They want to understand with their own common sense that a company is genuinely cheap — not take someone’s word for it.

That move — from “this is good because everyone says so” to “this is good because I understand the mechanism” — is a huge, huge shift. It’s the shift that lets people stay invested for the long term.

That’s when I really realise that the biggest growth in a person comes from their biggest failure.

If big banks had never corrected, I imagine these same people would still be doing the same thing, over and over. Maybe the drawdown happened for a reason — to upgrade the critical minds of the general population. To separate the people who want to actively improve from the people who quietly resign to fate.

So wherever you are, if you’re in the middle of this — this phenomenon, this so-called failure — I hope you know which side you want to be on.

I hope you choose growth over stagnation.

Last I checked I don’t think we can time travel and reverse decisions we made in the past. The only thing we can do is improve the next one. And maybe continue to have faith that the next fruit is better because of what this one costs.

If you don’t know who talk to about this, you can always give us a call.

Cheers.

Howdy there. Long time no see.

If you’ve been reading this blog over the past few weeks, you’ll have noticed the byline has been almost entirely Toby’s. That’s because I’ve been head-down with our existing clients — we had to rebalance aggressively to capitalize on the market fire sale triggered by the fear that MSCI was going to downgrade Indonesia from Emerging Markets into Frontier Markets status, compounded by the Iran-war shock and the general panic that came with them.

A lot of names we had been tracking in our watchlist suddenly fell into our buy zone, and when that happens, writing blog posts takes a backseat to actually allocating capital. That work is now largely behind us. I can go back to writing regularly.

And I want to start with something that, if I’m being honest with you, I don’t bring up enough.

I have never worked a single day as a professional equity analyst.

My degree is in Computer Engineering from the Hong Kong University of Science & Technology. My professional career before Recompound was software engineering and data science — most recently, building credit-scoring engines.

Everything I know about investing is otodidak: self-taught through reading, through making real losses in the market (a tuition fee you can only pay in cash), and — when I decided this wasn’t a hobby anymore — through obtaining my Penasihat Investasi (PI) license from OJK.

That’s the pedigree. If you want to verify, my LinkedIn is public.

The paradox

I bring this up because one of the most common questions I get when meeting someone new is: “Wait, you have an engineering and data-science background and you do value investing? Shouldn’t you be doing quant trading?”

It’s a fair question. On paper, my background fits quantitative strategies much more naturally — statistical arbitrage, high-frequency patterns, alphas hidden in the noise, complicated maths all the way down. That was the path of least resistance. I didn’t take it.

The honest reason is that value investing clicked for me in a way that quant never did. And what made it click was not Warren Buffett. It was a concept I had already been using for years in a completely different context, dressed up in different words.

That concept is Margin of Safety.

What Margin of Safety actually is

Most retail investors I meet misunderstand what Margin of Safety actually is. They think it’s a tool for calculating target price — “if fair value is 1,500 and MoS is 30%, I buy at 1,050.” That’s half-true at best. It is also backwards.

Margin of Safety is not a number that tells you how much money you’re going to make.

Margin of Safety is a number that tells you how much you can survive being wrong.

Wrong about what? Wrong about everything. Wrong about the company’s moat. Wrong about the owner’s integrity. Wrong about the commodity cycle. Wrong about the macro regime. Wrong about the discount rate you used because you forgot to check the risk-free rate had moved. Wrong because some black swan none of us priced in showed up uninvited.

The whole point of MoS is to bake humility into the assessment — we don’t know what we don’t know — and let the buffer do the work when we’re inevitably wrong about something.

And that is exactly what I had been doing as an Machine Learning engineer, except we called it something else.

The Machine Learning analog: Regularization

In machine learning, when you train a model on data, your default loss function looks something like:

where w is the vector of model weights. You minimize that loss, and you get the best-fit model for your training data.

Except — there is a problem. The best-fit model for the training data is often a terrible model for the real world. It latches onto noise, memorizes rubbish, and collapses the moment it sees unfamiliar inputs. We call this overfitting. The model is too confident in patterns that aren’t real.

The fix is regularization. You add a penalty term to the loss function:

where R(w) is some measure of model complexity. The two most common forms are L2 (ridge), which penalizes large weights:

and L1 (lasso), which drives many weights to exactly zero:

The coefficient λ controls how much you care about fitting the training data versus keeping the model humble. Crank λ up, and your model will have worse training-set performance but better real-world performance. Because — and this is the sentence I want you to sit with for a second —

The real world is not the training set.

That is literally Margin of Safety, just under a different name.

Why the analogy goes deeper than it looks

A value investor building a thesis is doing what an ML engineer does training a model. You construct a mental representation — this business earns X, grows at Y, the owner behaves like Z, the commodity cycle does W — and you check how well it fits the data you have. If your thesis is sophisticated enough and your data looks clean enough, you can convince yourself very precisely where the stock should trade.

But your data is the training set. The real world is the test set. The real world has Iran-war shocks, RKAB quota surprises, sudden waves of foreign money leaving Indonesia when global index providers change their mind about us, commodity cycles that last one year longer or shorter than the prior ten. You cannot regularize these away. You can only prepare for them.

Howard Marks puts it more memorably than I can:

“Never forget the six-foot-tall man who drowned crossing the stream that was five feet deep on average.”

The stream is the training data. On average it is crossable. The drowning happens at the deep spot that the average conceals. Survival depends on preparing for the worst point on the curve, not the mean of the curve. A strategy that works in normal weather but dies in the tail is not a strategy — it is a delayed obituary.

Regularization — in ML and in investing — is the penalty you willingly pay for that preparation. You give up some upside in exchange for not drowning.

The second analog: diversification as regularization

There is a second place the analogy runs deep, and it is where I want to gently push back on a pattern I see often in Indonesian retail value-investing circles.

Let me be clear about what I am not arguing.

At Recompound we run roughly five to twelve names at any given time.

By most global measures, that is already a concentrated portfolio — a typical US mutual fund holds eighty or more. Five to ten names is a perfectly healthy balance of conviction and diversification. It is what we do, and we are comfortable with it.

What I want to flag is not five or ten. It is one, two, or three.

The standard Buffett-inspired advice is “if you really understand what you own, you only need a handful.” Some retail investors I meet take this much further than Buffett ever intended and end up with one, two, maybe three stocks that represent 80 to 100% of their net worth. Almost always these are people who have read one or two investing books, found their big conviction, and gone all-in because they “know” the thesis.

I understand the impulse — I have felt it myself. I still think it is dangerous.

To see why, two more lessons from machine learning are worth borrowing.

The first is that an ensemble of decent models almost always beats a single brilliant one on out-of-sample data. Random forests beat the best single decision tree. Averaged networks beat the best single network. Boring, but true.

The second is dropout — the practice of randomly disabling neurons during training. Models trained with dropout are more robust than models that aren't, because they are forced to not over-rely on any single path.

The investment analog writes itself. A one-to-three-stock portfolio is a single model, even a brilliant one — and we know what happens to a single brilliant model when the data it didn’t see shows up. A ten-name portfolio with uncorrelated exposures and no two names sharing the same failure mode is an ensemble. And position-sizing discipline that caps every name at a hard max porsi porto is dropout for investors: you force yourself to survive even when one conviction quietly dies.

MoS on each name protects against the specific failure mode of that name. Diversification across names protects against the category of failure you didn’t see coming. Both layers are needed. Neither is sufficient on its own.

Your MoS on that one stock may be excellent. But when a tail-risk event arrives — and it will — a huge MoS on a single name is still a single-name bet. You do not die. You simply stop existing as an investor.

Concentration gives you alpha when you’re right. Diversification gives you survival when you’re wrong. I have been wrong enough in my life — in my ML jobs and in the market — to not trust myself to be right every time on just one or two names.

The cautionary tale: LTCM

The story I keep coming back to is Long-Term Capital Management.

LTCM’s partners included Myron Scholes and Robert Merton — actual Nobel laureates in economics. Their models were more sophisticated than anything I will build in my lifetime. Their training data was cleaner, their maths sharper, their execution more disciplined than 99% of market participants.

They blew up in 1998. A few sigma events that “couldn’t happen” — Russian sovereign default — happened, and the fund had to be rescued by a consortium of banks to prevent systemic contagion.

LTCM did not fail because they were stupid. They failed because they had trained on a world that didn’t include the tail, and then levered up based on the confidence that beautifully-fit model gave them. They had an extraordinarily well-fit model with almost zero regularization.

Kita bukan dewa was a lesson they hadn’t internalized yet.

When I read about LTCM — and I read about it often, the way some people re-read tragic novels — I do not see a story about excessive pride and self confidence.

I see a story about regularization coefficients set too low.

Closing

So here’s the thing. If you have an engineering or a data-science background and you’re wondering whether value investing is for people like us — I promise you it is.

More naturally than quant, actually.

Quant, in its most common form, is optimization without humility (in my humble opinion).

Value investing, done properly, is optimization with a regularizer. You are not trying to build the best model of the market. You are trying to build a model that survives what you did not expect.

Margin of Safety on each position. Diversification across positions. Position sizing that respects your own ignorance. These are not just “be careful with money” platitudes your parents told you.

They are λ’s — little penalty coefficients in the loss function of your life’s capital.

Turn them too low and you will outperform in the training set — and die in production.

Turn them high enough and, like any well-regularized model, you will be slightly less impressive on paper. But you will still be around to keep compounding ten years from now.

And compounding, really, is the only game that matters.

See you on the next one.

I hope you enjoy this conversation, because I did. And this can become a replacement for your Netflix and doom-scroolling nights. Or if you want something to accompany you in (Jakarta’s) traffic.

We have been posting a series of videos on YouTube called I CAN RELATE. It is a series where we talk to our customers, most of whom are entrepreneurs and working professionals, about their lives. They are exceptional but also ordinary human beings, who go through experiences in life that are unique yet also relatable. We talk about their career, life decisions, relationship with people and money, self-improvement, etc etc.

Talking to these people, oftentimes I will think in my head “Omg I can totally relate to that” but have to restrain myself from cutting into the person’s speech midway, so usually I just think that in my head.

This time, I talk to Wilson, who is Recompound’s first ever full-time employee. If you do not know, Recompound is mostly a 2-men business between 2023 and 2024. We did have a couple of part-time soldiers, but in the office it is just me and Budi. As much as I like Budi, I need to see a face that is not just Budi’s. Plus, in 2025, I was hit with a crude reality that I was not a superhuman after all. I realised that I can’t code, research companies, write blog posts, make videos, take client calls, answer client questions, talk to our investors, clean the office (and many more tasks, you get the idea), all on my own. Budi did a lot of the heavy lifting focusing on coding and investments, but he has inflicted a couple of collateral damages with his less-than-ideal diplomacy skills when interacting with clients.

So we took the leap of faith and hired our first ever full-time employee. We also did not randomly hire a random person. I have known Wilson since my university days, where we worked on a social project together. Ever since then I always had a keen memory of him: a can-do guy not deterred by uncharterred territories. So when I think of the person that I want to see in the office daily, and the person that can make me feel secure to hand over the day-to-day running of Recompound while I work on growing the business, I rang him up.

I masked it as a “catch-up” session, with a very clear agenda in my mind. I was secretly happy to hear that he was feeling stuck and wanted a change in his then job. I did try to sound empathetic to his predicament. Then it was his turn to ask me about what was going on in my life. I tried to be low-key but yet did not miss the opportunity to brag about Recompound slightly. I mean, he has to think that we are cool to want to join us, right?

I will leave the rest for you to listen in the conversation. But my long-entrenched faith and admiration for Wilson proved true even until today, almost 2 years into our working together. Some things never change: Wilson’s can-do attitude, as seen in his transformation from someone who does not how to code to calling pull-requests like that’s his daily breakfast.

If you want to decide whether this is something worth listening to, below is a summarised points of the topics we talked about.

• Developing grit and resilience — Wilson studied chemical engineering but pivoted into tech during COVID when oil & gas companies stopped hiring. It took him 11 months to land a job, eventually joining Shopee as a PM with no prior experience.

• Career pivoting and self-discovery — Wilson went from chemical engineer → product manager → data analyst, learning SQL and coding largely through LeetCode, HackerRank, and sheer determination — without the help of AI tools that exist today.

• Positive Mental Attitude (PMA) — Wilson’s mantra on approaching life is by having PMA. When life throws balls at you, get through it with the help of PMA.

• Going deep vs. wide — A key lesson from Shopee: it’s better to master one area deeply than to spread yourself thin across many topics (”jack of all trades, master of none”).

• Finding your strengths — His advice is to try many things cheaply and quickly while you’re young, so you can identify what you’re genuinely good at before life responsibilities pile up.

• Healthy habits and accountability — Wilson is Toby’s gym buddy, and loves the Tobyfit catering that is provided by Bu Tina (Toby’s beloved cook). Maybe Recompound is a cult, but we are a cult that sees core habits (exercise, sleep, eating well, and surrounding yourself with people who have good habits), as key factors directly affecting your clarity of thought and decision-making.

• Humility and continuous learning — He believes humility comes from reading books, meeting inspiring people, and acknowledging that nobody can know everything. The Alchemist by Paulo Coelho his life-changing book.

• Legacy and kindness — He wants to be remembered as someone who brought positive energy and was kind, even to strangers. He encourages starting with small acts of kindness within your own family first.

If you read this far, you are genuinely awesome. I hope you managed to listen to our conversation too.

Do leave us comments if you find this relatable!

A few weeks ago, a friend introduced me to a group of friends at dinner. He said, “This is Toby, he runs a stock picking company.”

I smiled politely. But inside I was like — broo, no ☹️. That is not what we do.

And here’s the weird thing. It wasn’t just that the label was wrong. The label changed how the person across the table treated me. The immediate question directed to me is:

“Bro jadi saham apa ni lagi bagus?” (”What’s a good stock to buy now?”)

asking for a hot tip. Under such a scenario, I smile and wave like the penguins in the Madagascar Movie.

Had my friend said, “This is Toby, he runs a company that acts as a Personal CIO for busy professionals,” the conversation would have gone completely different. The person would have asked me how I think about portfolio construction. How I assess risk. How I stay convicted during a downturn.

Same person. Same dinner. Same Toby. But a different label can produce a different conversation, which could produce a different relationship.

So I’ve been thinking about this quite a bit.

A programmer taught me this first

So I studied Computer Science back in university. One of the first things you learn — before data structures and algorithms, before anything fancy (that I have mostly forgotten) - is how to name things properly.

This sounds trivial. But I have come to realise that learning about naming things properly is not trivial at all.

A code base is like an essay, akin to the one you are reading now, that serves a purpose to instruct computers on what programmers want computers to do.

So even if you don’t code, you could:

• be confused by a codebase OR

• have somewhat of an understanding what the code is trying to do.

depending on how well names are written.

For example, a variable called

data

tells you nothing about the data that it is supposed to store. But a variable called

monthlyRecurringRevenue

tells you a very clear thing. It tells you that it should be storing monthly recurring revenue.

When a developer reads the second name, they would know what the line of code does. They know what the number should look like. The name itself becomes a thinking tool.

Let me say that one more time. The name itself becomes a thinking tool.

You could imagine:

Hmmm, if the company now gets an additional stream of monthly recurring revenue, how should I update the calculation of this monthlyRecurringRevenue ?

vs

Hmmm, if the company now gets an additional stream of monthly recurring revenue, how should I update the calculation of this data ?

Do enough rounds of the 2nd scenario and I probably will start losing hair.

In software, bad naming leads to bugs. Not because the code breaks. But because the developer misunderstands what the code is supposed to do. The logic was fine but label was misleading. And the misleading label led to the wrong assumption which led to incorrect modification or improvement of code.

The same thing happens in investing

When people label what they do as “trading,” they behave like traders. They check prices hourly. They panic when red candles show up. They celebrate green candles. In short, they become a servant to the price chart.

When people label what they do as “owning productive businesses for the long term,” they behave like owners. They study the annual report. They assess the management team. They ask: is this business generating more cash this year than last year? How’s the retained earnings of the business this year like? They hold through the noise because they have a mental model that can absorb the noise.

Warren Buffett is arguably the greatest investor of all time. And it seems to me that one of his biggest edges is not just spreadsheet. But it’s his vocabulary.

He doesn’t call himself a “stock picker.” He calls himself a “business picker.” That single reframe changes everything. A stock picker looks at a ticker and thinks about price movements, entry points, and when to sell. A business picker looks at a company and asks: does this business make high returns without putting up much capital? Will people still want this product in coming decades?

And here’s where it gets crazy. Buffett watched Coca-Cola for 52 years before buying it. Yes, he sold Cokes as a six-year-old from his grandfather’s grocery store. He saw the product dominate the world for half a century. But he didn’t buy a single share.

Why? Because for a lot of his career, he was using Ben Graham’s framework, which labelled a good investment as a “cheap stock.” Coca-Cola was never cheap enough by that definition. It was only when he reframed his approach — from finding “fair businesses at wonderful prices” to finding “wonderful businesses at fair prices” — that he could see what was right in front of him.

Same Buffett. Same Coca-Cola. Different name for what he was looking for. The old label caused him to overlook the opportunity for 52 years. The new label made it obvious. Shocking.

Your emotions work the same way

Here’s where things get even more interesting. I recently read a book about how emotions are constructed by Lisa Feldman Barrett. And her core argument really shook me.

She argues that emotions don’t have universal fingerprints. There’s no single pattern in your brain or body that maps neatly to “anger” or “sadness” or “fear.” Instead, your brain constructs the emotion based on what concepts you have available.

So if you only have the word “stressed,” then every uncomfortable feeling becomes stress. You’re stressed when you’re actually exhausted. You’re stressed when you’re actually lonely. You’re stressed when you’re actually hungry. But if you have richer labels — if you can distinguish between overwhelm, resentment, restlessness, grief — you can respond appropriately. You sleep when you’re exhausted. You call a friend when you’re lonely. You eat when you’re hungry.

Better labels, better responses. Worse labels, worse responses.

This is exactly what happens to investors who lump everything under “the market is scary.” Is the market correcting because of valuation compression? Or because of an earnings miss? Or because of foreigners pulling out? Each of those has a different implication and a different appropriate response. But if your only label is “scary,” your response is confined to panic.

Recompound: Your Personal CIO

We don’t really call ourselves a stock picks service. We don’t call ourselves a robo-advisor. We don’t call ourselves a wealth management platform.

We call ourselves your Personal CIO. Chief Investment Officer. And that is deliberate 🙂

Because a CIO doesn’t give you hot tips. A CIO thinks deeply about capital allocation on your behalf. A CIO has conviction rooted in analysis, not hype. A CIO stays when things get boring — and 99% of investing is boring. And a CIO is here to serve you, the CEO.

The label isn’t really a marketing gimmick. It is our own mental model on how we should behave and do our jobs day to day. And hopefully, the mental model could shape how our clients relate to their own wealth too.

When our clients understand that they have a Personal CIO, they stop asking “what stock should I buy today?” and start asking “how is my portfolio positioned for the next 5 years?” That shift, from stock tips to stewardship, is the name of the game.

So what’s in it for you?

Here’s the practical takeaway. You don’t need to read a neuroscience book or learn to code or become a Recompound customer to apply this.

Just pay attention to the names you give things.

If you call your investment portfolio a “trading account,” you probably will trade. If you call it “my family’s long-term wealth,” you probably will put more thoughts to safeguard it.

If you call your morning routine a “grind”, it can end up feeling like suffering. If you call it “prep”, it will feel like a lifestyle with purpose.

If you call a market downturn a “crash,” you can end up feeling panicky. If you call it “a sale,” you will start asking: “how to look for bargains?”

The names you give things are not just descriptions. They become more of instructions really. They tell your brain how to feel, how to think, and ultimately, how to act.

So choose your names carefully.

Cheers.

Yep, no kidding. I did go for a retreat for one week without looking at my phone in Forest Island, Bali (it was confiscated at the start of the retreat). And all participants were barred from talking to each other. We are not allowed to read books and not allowed to write. Anything that is too stimulating to the mind is not allowed.

To make matters worse, you could take a guess who’s my room mate (all not intended), who I am not allowed to talk to. Go on, take a guess.

Well if you guess Budi Ryan, then you are correct 🙂

Obviously as Recompound Co-Founders, our lives are very plugged in:

• to WhatsApp groups (each of our customer has their own private WhatsApp group with us)

• to the markets (we have to take a look at the market when we want to buy or sell stocks)

• to the internet and AI (need to read a lot, and write a bit of blogpost)

So why are Toby and Budi doing this to themselves?

The short honest answer is because we want to take a holiday (a break) from Recompound. And it is especially timely because during the Idul Fitri period in Indonesia, market is closed and it tends to be more quiet.

Another benefit that we’d like to get from this holiday is a stronger mind.

See, I believe that our mind loves to wonder around thinking about the future, and dwelling in the past. Not fully connected to what’s happening in the present moment. Especially because we are investors. We spend a lot of time thinking about what the companies we invest could become in the future. What’s the price going to be. We also unavoidably spend time dwelling about the past. We should have bought this later, sold this earlier. Hindsight bias tends to be a thing for investors. I don’t think there’s any investor who’s immune to hindsight bias, including ourselves.

So going for this retreat is an attempt to develop a stronger muscle of our mind. Exactly like going to the gym.

So why not just go to the gym to develop a stronger mind? Yes, going to the gym counts as a tool to develop a stronger mind. In the gym we have to focus on progressive overload and focus on feeling the muscle groups that we want to train (back, chest, shoulders, legs). Then a stronger mind that comes out of it is definitely a good bi-product. But I believe there’s an even more specialised tool to develop this stronger mind, which is the main focus of this retreat:

meditation

Yes, so for 7 days 6 nights, Budi and I did nothing but meditation. The retreat fortunately, is designed very nicely for us to learn about meditation step by step. Because I don’t really do meditation at all before the retreat. So it is very noob friendly.

Btw by no means is this an article to ask you to meditate every day. I think going to the gym is great for the body, but I don’t go around and force people to go to the gym. In the same spirit, I now think doing meditation is great for the mind but I will also not go around and force people to meditate ;)

If you are curious about what comes out from the retreat, you could read on. But in this article I will be talking about the core learnings of developing a stronger mind through meditation and how it is extremely related to value investing. It is so friggin related that I wonder if the Founder (Pak Merte Ada) of the meditation retreat secretly read up books on value investing and applied the learnings to the meditation course that he designed some 30 odd years ago.

Hopefully, this article would be helpful to pique your interest to get to know of a tool to actively train your mind to be stronger. Let’s get started.

Four core components of a strong mind

In the retreat, strong mind is communicated as a harmonious mind with four core components. Some people are weaker in one component and stronger in another. So you can choose which core components you would like to train. Really like going to the gym right?

Some people have weaker legs so they can do leg day and do those squats. Others want to train their back a bit more so that they can get that macho V shape look. You get the idea.

Now these 4 components that enable stronger harmonious mind to be developed, I argue, are also core ingredients to be a good long term investor.

First component

First component is focus, or concentration really. During our meditation, when we close our eyes, a key item that we have to do is focus on a single object. That object is usually our breathing. In and out, in and out. If we don’t have a strong focus, we tend to feel other things outside our breathing. For example, the itchiness on the neck. The numbness on the feet. Or the loud sound of birds chirping at the background. When we observe that our focus is scattered, we have to bring it back to our breathing. A single object that we have set our intentions on.

Now isn’t this exactly like value investing?

When we want to analyse a company A for example, we need to direct our attention and focus to one company. There will be temptations or noise from our neighbours about company B, company C and so on. Investment ideas nowadays seem to be quite scattered on social media. Influecer A will say that this stock is good. Influencer B will say another stock is good. As a good long term investor, we’d want to focus our analysis on one particular stock at a time. Because if we hop around (we are definitely guilty of this as we are never perfect investors), the quality of the analysis can get compromised.

So having focus, is ideal for us to understand a company better, filtering out plenty of noise that is circulating out there. With this ingredient developed, we can be a better long term investor.

Second component

Second component is being at the present. In the retreat, this is called being mindful. Because it is no surprise that when we close our eyes and meditate, we would think about what I look forward to eating after I come out from this jail. Or “ah darn it, have I done this or that before coming in for this retreat?”

When we feel that our mind wanders to the past or to the future, we invite it back to be in the present.

Well this is, again, exactly like value investing. We absolutely have no idea what the price of a company we invest is going to be tomorrow. But we tend to think about the future so much.

“Will the stock price plummet because of the war tomorrow?”

“Will IHSG benefit from increase of oil prices?”

“Will government policies like free meals affect foreign investors confidence in the future?”

And don’t get me started talking about thinking excessively about the past.

“Ah, I should have bought more of that stock. It should have 20% allocation instead of 10%!”

“Why did I not sell the stock when the margin of safety is already reduced?”

“I should have never diversified to foreign markets!”

You get the idea 🙂 We tend to think so so much about what will happen in the future and what we should have done in the past. So much so that we can sometimes forget that

“hey we are now invested in a good company, with good management, good product and great competitive edge. We bought it at a good price, but the price has not moved. So we wait.”

Or

“hey the companies that we invested is already priced at a premium by the market. So we can sell now. We don’t have to be greedy about the price going up further in the near future.”

That my reader, is what I learned about developing a more mindful mind which is anchored at present. By anchoring our decisions based on the present moment and the information we have today, we’d develop to be a better long term investor.

Third component

Third component is being gentle. The word that the retreat uses is developing love and kindness. This again makes so much sense because when our minds start to wander, we tend to be hard on ourselves.

“Toby, why are you so easily distracted?”

“Budi, why can’t you stay at present?”

Negative talk is real when we inadvertently fail to focus, and being gentle to ourselves is a potent antidote. Particularly for me, when I notice that my mind starts to wander, I now view it as a training opportunity to bring my mind back to the present and the area of focus.

Again I believe that being gentle to ourselves is vital to becoming a good long term investor. Because investments is never perfect. Things that happens outside of our control can happen both in good and in bad ways. In a good way, the company we invest in can suddenly be acquired by another company and the price could go up like crazy.

It is easy for us to be greedy and forget about our investment thesis when such a positive event happens to the stocks we invest. In a bad way, the company we invest can also underperform relative to our expectations and then the stock price could take a hit. Again, it is easy for us to fearful and forget about our investment thesis.

We can be harsh on ourselves and end up making impulsive decisions. But what I think is more helpful is if we could accept the things happening outside of our control as a given. Then we can treat uncontrollable events as an opportunity for us to respond in a calm and gentle way, instead of treating it as an evidence that we are foolish investors.

Do we buy more? Sell? Or do nothing. With gentleness on our minds, the answers will be a lot clearer than not. With more clarity, I believe this would help us to make better decisions and become better long term investors.

Fourth component

Fourth component is seeing the truth. A man who sees the truth as it is, without adding much salt or sugar on top, is said to be a wise man. In the retreat, there’s one universal truth that we are invited to experience. That truth is the truth of impermanence. Things are constantly changing and in flux. We could observe and experience our breath changes from inhalation to exhalation and then back to inhalation again. We could observe the heat on our chest cool off gently. We could experience the itchiness on our skin that gradually subsides. Everything on our mind and body changes.

When we see that our mind and body is in constant state of flux, we are seeing the truth as it is. We’d grow our hair, we’d grow more muscle, we’d become older and so on. That’s the truth. And when we see this truth as it is, we develop wisdom.

Relating this concept to investing for the long term, the companies we invest are also constantly changing. Their sales would grow or decline. Their assets would get sold. Their debt will get bigger or lower. Their management would change from time to time. Heck their business models can even change. Seeing that a company is like a breathing organism that is never permanent allows an investor to be wiser.

As a result, we don’t have to pretend that we have to be a shareholder of a company forever. If we don’t like the new management or if we are not satisfied with their growth, we can always sell. Or if we think that the company becomes better, gaining more market share, while the price is still well below its intrinsic value, we can always increase our ownership and buy more.

This adaptability and non-attachment, I believe, is developed overtime with a wiser mind. Therefore I believe that seeing the truth (as it is) is also, an essential ingredient to be a good long term investor.

Closing Remarks

Thank you for reading this exposition thus far. I hope that you could see how a long term investor could be better overtime through the lens of a person who practices meditation. It all starts from having a strong harmonious mind.

Are our minds perfect and are we the best long term investors? No and no. Developing our minds to be stronger and our skills to be better long term investors are always a work in progress. There are always others with stronger minds and better investing skills. But we believe that the retreat carries genuine benefit in increasing our resolve to have stronger minds and be better long term investors.

We are definitely not paid to mention the meditation retreat. But if you’d like to challenge yourself to go on for a retreat like what Budi and I did, feel free to check out Bali Usada. If you end up trying them out, I hope that you will find it useful to develop a stronger mind and a healthier body. May all beings be happy!

Cheers.

A sequel to “I had lunch with a cinema owner.”

Since we are talking about the use of AI in my line of work, I would like to preface this write up by clearly communicating what this write up is not.

It is not an expert opinion about how the investment industry is going to take shape in the future

It is not an editorial whether AI will replace more knowledge workers in Jakarta

It is also not about financial advice as there will be a brief discussion about a stock research that was conducted using AI.

You can treat this writeup as a data point for how AI is currently used by Co-Founder of an early stage wealth tech startup in Indonesia.

With that, let’s get cracking.

There is a wise 80 year old man

Who likes to write memos documenting his thoughts embroided with his investing philosophy that he has accumulated for more than 5 decades. He manages close to $200 billion of funds and he generously publishes his thoughts to the public free of charge. One day, a fellow value investor shared with me his most recent memo. He writes a memo about his experience using Claude. I dropped everything and went on to read what he has to say.

In his latest memo, Howard Marks describes asking Claude to write him a tutorial about AI. What came back was 10,000-word essay that referenced his past memos, anticipated counterarguments and Claude even sprinkled humor here and there. I couldn’t imagine being in his shoes, witnessing the amount of technological progress in his life time. I mean he was born in an era where digital photography was not invented yet. Now he is dealing with a machine that writes a note as good as what a competent colleague would do.

But Marks being Marks, did not stop at awe. He pushes further. He questions deeply into what is the likely implication of AI to the field of investing. If AI is so good at gathering and synthesizing data, is there a place for mere mortals like Toby and Budi?

Note: Marks didn’t say Toby and Budi in the memo because we are really just mere mortals

Well the answer comes from a person who studies the spread of diseases, an epidemiologist Marc Lipsitch. He observed that humans make decision using three core components:

1. Facts,

2. Informed extrapolation from analogies to prior experience, and

3. Opinion or speculation

Marks is saying that AI is beyond this world for (1) and (2). Perhaps even better than any human alive. Just like Stockfish (the chess AI) is better than Magnus Carlsen. But he says that AI is genuinely weaker in novel situations. In other words, AI is probably not as good as humans who have lived experience with minimal data in the internet (information used to train AI models).

He further went on to say that humans (at least the best humans) still hold an edge in situations that are genuinely novel, where there is no historical pattern to rely on.

But here is what I kept thinking about after reading the memo.

If AI is this impossible to beat at facts and extrapolation, shouldn’t we trust its opinion too? It does not have anchoring bias. It does not have fear or greed. It does not hold onto a losing position because its ego cannot stomach the loss. It is not swayed by narratives — the very thing Shiller described to us in Part 1.

In the mean time, look at us. Look at the quality of judgment that human beings produce. You can think of the decisions made by leaders around the world. We are emotional, inconsistent, stubborn. We lie to ourselves. We hold onto beliefs long after data and evidence proves them wrong.

So how would humans, a being this flawed, stand a chance in having superior taste, judgement and opinion?

I didn’t fully understand what Marks said.

But above all, I was feeling very competitive. I grew up in a Singaporean kiasu (fear of losing) culture and I simply didn’t want to lose out to this 80 year old wise person. Prior to reading his memo, I never used Claude before.

So like any other kiasu person, I went on to buy the Claude max plan to give it a spin.

I want to have a better grasp of what does Marks mean. So I started where any kiasu person would start.

I wanted to test how far AI can actually go in my work. Not in theory. But in practice, on a real stock that I was actually studying. And I want to be honest — part of me was hoping AI would just be better. That would make my life a lot easier.

The company that I studied was Sarana Menara Nusantara (TOWR)

TOWR is Indonesia’s largest independent owner and operator of telecommunications infrastructure. A Djarum owned company, it owns about 30,000 towers across the country and leases space on them to telco operators like Telkomsel, Indosat, and XL. The tenants install their antennas and equipment on the tower, pay rent, and the contracts are locked in for 10 years. It is a beautifully predictable business.

The current factor that is driving TOWR’s profitability

Aside from average selling price of the lease agreements (which is not the focus in today’s discussion), is also TOWR’s colocation model. Being an independent telco tower operator, TOWR is able to host more than 1 tenants in its infrastructure (example tenants are Telkomsel, XL and Indosat).

Now the capital expenditure for having 1 tenant is large, as you have to either build a new one or acquire towers from existing industry players (TBIG is also in the same industry). But an additional tenant at the same tower involves minimal capital expenditure which has a direct contribution to its profitability. You can imagine getting additional revenue without most of its expense. It would go straight down to profits.

So to repeat, having an additional tenant or a high tenancy ratio (average number of tenants per tower) is highly desirable for the company as it is associated with higher profitability.

But of course, the market has largely been harsh.

You could see below that the price of the share has seen a 60% drop in the last 5 years as this post was written. Markets might be harsh because TOWR might be seen as having an obsolete business especially because telco companies will have lesser need of those towers.

But there’s another phenomenon that might cause immediate headwind to TOWR’s profitability that the market does not like.

Consolidation, consolidation

To give you more context, the telco industry in Indonesia has seen multiple consolidations in the past 5 years. Starting with Indosat and Hutchinson’s merger in 2022. And the most recent merger of XL and FREN back in April 2025.

Why events such as this is pertinent to TOWR’s profitability?

I suspect the answer is because consolidations would reduce tenancy ratio (discussed above). Usually we had 5 telco players prior to the consolidation. Now that we have 3, the merged telco companies no longer need to rent the same tower space.

So of course, tenancy ratio has dropped for the past few years from 1.88 in 2021 to 1.61 in 2025.

So the question: now that there was a recent merger of FREN and XL in 2025, how bad will tenancy ratio actually get, and for how long?

My plan is simple

I want to use AI to help me use Agent Based Modelling and Monte Carlo Simulation technique to model out the probable tenancy ratio outcomes.

For those unfamiliar,

Agent Based Modelling

is the idea that instead of modelling an entire system with a single math equation, you model the individual pieces and let the system emerge from their interactions.

In TOWR’s case, each tower is an “agent” with its own characteristics. Location (urban, suburban, rural), whether it has fibre connectivity, how many tenants it currently has, how much exposure it has to the merging operator.

You give each tower a set of probabilities (chance of gaining a tenant, chance of losing one, chance of getting 5G equipment) and then let all 30,000 of them play out over a period of time. The system-level result — the tenancy ratio — emerges from 30,000 individual stories, not from a top-down assumption.

Monte Carlo Simulation

is the idea that you run the same model hundreds of times, each time with slightly different random outcomes, to see the range of what could happen. Think of it like rolling dice 200 times and tracking the results. Any single roll tells you nothing. Two hundred rolls give you a distribution, what’s likely and what’s extreme.

Combine the two: 30,000 towers, each with their own probabilities, simulated 200 times per scenario. That gives you a probability distribution of where the tenancy ratio might land — not a point estimate, but a range of possible outcomes.

Now here’s where it gets interesting.

Building the model was the easy part. AI handled that quite nicely. The hard part was...

Claude was impressive

It built the entire agent-based model. It ran simulations and produced charts with confidence bands. If you saw the output, you would genuinely think a quant team produced it. Take a look.

So I started looking at the parameters. One of the key inputs was the organic colocation rate — how often a tower gains a new tenant through normal business activity.

Claude estimated this by looking at 2024 data: TOWR added 3,751 tenants across roughly 30,000 towers. That gives you about 2.5% per quarter, which makes sense.

But something about 2024 felt off. I remembered reading TOWR had acquired a competing tower company called IBST in 2024. So I went to NotebookLM — which had the actual company filings loaded — and asked it to separate the organic additions from the acquired ones.

Turns out, a significant chunk of those 3,751 tenants came from the IBST acquisition. They were bought, not won (organic). So Claude had used a contaminated year to set one of the most sensitive parameters in the entire model.

The corrected organic rate was closer to 1.0% per quarter (originally was 2.5%), which is a big shift. That single number meaningfully changed the five-year trajectory of the tenancy ratio.

But the assumptions underneath were wrong. And they were wrong in ways that sounded right.

Here’s another one. The model assumed that when the merged telco operator (XLSMART) decommissions a redundant tower, about 60% of the time they would relocate their equipment to an existing TOWR tower nearby. That sounds reasonable — why build a new tower if there’s already one available?

But it turns out TOWR has a strict no-discount policy. Whether you colocate onto an existing tower or order a brand new one, you pay the same rate. So the operator has zero financial incentive to compromise on location. They always choose the network-optimal spot, which usually means a new build. The actual number was closer to 25% colocating onto existing towers, not 60%.

Nobody would know this from the outside. It is buried in how TOWR structures its commercial relationships. NotebookLM surfaced it only because I asked the right question — and I only knew to ask because the economics felt off. Which I also don’t know how to describe myself.

This happened over and over. Non-MNO tenants that I assumed occupied tower space turned out to be purely fibre clients — the parameter should have been zero, not 0.5%. A bell-curve churn shape that I had assumed turned out to be wrong because tower contracts are irrevocable — churn is driven by contract expiry schedules, not operator discretion.

Each correction required me to ask a question that AI did not think to ask itself. And at the end… I got this results.

The results looked nearly identical to the naked eye, yes?

But the assumptions underneath the second picture have been far more closely scrutinised. I challenged every parameter and in many cases corrected them too. The organic colocation rate was halved. The swap ratio was cut by more than half. Non-MNO tenants were zeroed out. The merger intensity was redesigned entirely.

And yet, even after all that work, the interpretation of these results still differs. An AI would present the base case median as the most likely outcome. A cautious analyst might anchor to the bear case. An investor with experience in Indonesian telcos might look at these charts and see something else entirely based on relationships and context that never made it into the model. Which brings me back to the 80 year old man.

Where the human edge actually lives

Remember Marks and the Lipsitch framework? Facts, extrapolation, opinion. AI dominated the first two. The question was whether humans could still be relevant in the third.

After some time of working with Claude and NotebookLM on this analysis, I think I have a bit more understanding of what Marks was saying. It is not that humans have superior opinion. It is that humans have a bit more situated opinion.

We carry context from our own lives that no model has access to.

Every correction I made came from something I already knew or something that felt off based on lived experience or a prior reading that I came across, and so on. Perhaps from knowing what it feels like to lose money after making a wrong bet, that tends to sharpen our instincts all around and then asking the questions that turn out to be important.

So it is not uncommon to ask “are you being optimistic?” on other sub-areas of research.

It came purely from the gut, not analysis.

I don’t think any of this is genius. It is just being aware of context and applying lived experience to what we are studying.

So what helps in the age of AI

I think being hyper observant helps. We have so much lived experience, that is information we can glean right away from our senses. The internet is a huge source of information. But I don’t think it is right to underestimate the wealth of knowledge we get from walking to Indomaret, or travelling out of town from Jakarta to Manado. These experiences can often carry insights to our line of work in ways that is simply unimaginable to an AI.

I also think that willingness to try things is an important one. While researching and understanding a public company is still difficult work and reading convoluted responses of an AI could be tiring. But hey, I don’t have to write a single line of code in doing monte carlo simulations anymore. That’s bananas. AI is certainly helpful in saving huge precious time and allowing my inquiry to be more robust.

Lastly, intellectual honesty would be the most important ingredient. I don’t think there’s a point of being observant, trying new things and then lying to yourself about deriving a conclusion that I myself want to hear. The study should lead to a better understanding which hopefully would bring us to a better decision all around.

Closing remarks

That’s all folks, thanks for reading this far! If you fancy, give Claude or any other AI tools a spin. I am not just curious on how it is helpful in studying a company. But I am also curious how it applies to other lines of work, like healthcare, technology, consumer, anything. So if you are so kind to share your experience, do share them in the comments!

Cheers.

Those in the investment scene will need no introduction on who Chris is. For the benefits of those not very involved in the investment circles (no shame about this, because you are busy living life in your own niche circle), Chris is the founder of Bibit, which later got acquired by Stockbit. He also found the holding company Clapham, which owns a number of F&B companies.

Beyond the companies that he have founded, though, Chris is just an amazing human being. Speaking to him almost felt like meeting a soul brother. Some of the things he talked about were the things that I privately preached in my tiny circle, but often feel a bit shameful to share publicly. I thought I was just a weird person to have these beliefs and so I kept them to my closest circle. But hearing them coming from Chris just felt like a huge validation boost. There were countless moments when I thought in my head

Omg! I can’t believe you are thinking the same thing!

I genuinely had a great time speaking with Chris. Two hours passed which felt like 30 minutes. I hope this can be a useful companion for you to listen to when you are stuck in traffic, when you are on your long runs, or just want a better evening entertainment (to replace doom-scrolling and Netflix).

We did talk about both investment-related and non-investment related stuff. Here are some breakdowns just to help you decide if this is worth your time.

Investment-related topics:

• Changes in the Indonesian capital market 10 years ago vs now

• Whether value investing is dead

• Whether it is possible to make money in the stock market

• Managing the emotional roller-coaster when investing

• The impact of MSCI downgrading Indonesia

• The future of banking stocks in Indonesia

• How to evaluate banking stocks

• Specific stocks mentioned (not a recommendation): Astra, Gudang Garam, Prodia

Non-investment-related topics:

• Chris’ experience building Bibit

• How work and life would change with AI

• How we should view our relationship with money

• How to make yourself smarter by managing your information diet (hint: Substack)

• What you should know when deciding to invest in global stock market

Thank you , , , , , and many others for tuning into our live video with and ! Join us for my next live video in the app.

Also, drop us a comment on who you want us to talk to next time and the topics/questions you want us to cover.

Until next time, folks!

Get more from Toby Limanto in the Substack app

Available for iOS and Android

Get the app

We were supposed to catch up with him after Budi’s wedding in March 2025. We caught up almost a year later. I am ridiculously horrendous at these things, I know.

It was a weekday afternoon in a dimly lit restaurant. Budi and I sat next to each other across a cinema owner whose family runs theaters across Java. He has an air of calmness that he exudes all around him. So calm that I myself get nervous. But then I told myself “get yourself together, Toby. You got this”.

And I was genuinely eager to learn about his experience in the rapidly changing entertainment industry. I have been seeing more and more people (including myself) consuming AI-produced entertainment, and I figured someone running cinemas across Java — competing against the biggest player in the market — would have a sharp perspective on what’s happening.

Until he said something absolutely shocking.

His eyes lit up, almost like wanting to share a big announcement of the birth of a first son.

Guys! I am using the terminal in my computer for the first time to build an application for my business.

You can imagine this being said in the most calm yet energetic manner.

I know for a fact that this person is not technical. Furthermore, he does not at all have to concern himself with how to use AI to build applications. Yet here he was, showing us a web application he had built, hosted on Github. He built an interface that allows his company to ahem, trade secret ;)

He demo-ed it to us with the kind of pride you see in someone who has discovered something genuinely new about what they’re capable of.

He even recommended me and Budi a podcast: Naval Ravikant’s Motorcycle of the Mind. His message was simple. Use these tools. Use them aggressively. Don’t wait.

So his enthusiasm naturally left me wondering. He was someone with every reason to be defensive about what AI means for his industry. But he’s excited. This cinema owner sees AI as a revolution, while many others see them as the mother of job/industry destroyers.

Why is this? Two intelligent people can see the same tech and arrive at completely opposite emotional responses.

So I reflected deeper and remembered something I had been reading.

The narratives that we’ve been consuming

But before making sense of the optimism with AI, let’s have a sense of the fear first. The book that’s highly useful is Robert Shiller’s Narrative Economics.

Shiller needs no introduction, the Yale man is a Nobel Prize winner in economics. Let me flex a bit because I cannot help it, I also went to Yale. For a semester exchange. But hey stop judging. Pauli Murray College woop woop. Ok I am done.

The fascinating thing about the book is that it doesn’t sound like the economics we used to learn in high school. Scarcity, unlimited wants. Demand supply. GDP, Inflation. It doesn’t even look like the economics we observe on social media lately which is Game Theory.

The book looks like the spread of diseases. Yes, epidemiology.

The central argument that Shiller makes is that

economic narratives - the stories people tell each other about money, jobs, markets, the future - spread exactly like diseases.

They have contagion/recovery rates, they mutate, they go dormant for decades and they flare up again when conditions are ripe for flaring. Most importantly, they influence economic behaviour that household data struggle to explain.

The part that is most striking to me is what Shiller describes as “perennial narratives”. Stories that keep coming back across centuries. Every time they come back, the stories told from the narrative are different. But they are the same narrative. That narrative is: Machines will replace human labor.

Shiller traces it all the way back to the 1800s. The Luddites smashing power looms in 1811, the Swing Riots against threshing machines in 1830s. The depression of the 1870s where labor-saving inventions was the top of mind as a cause of mass unemployment. The 1930s, when “technological unemployment” started becoming a household term. Want to guess what was the terrifying new machine back then? Dial telephone.

What I have noticed is that the narrative has the same structure. Every time.

1. A new technology appears,

2. people fear it will destroy livelihoods on a massive scale.

The story spreads not necessarily because it is true, but because it taps into the very core of our emotions: anxiety. Every time, it would mutate just ever so slightly so that we can say this time is completely brand new. This time is different. This time humans are going extinct for real.

Well I believe AI is the latest mutation.

But what makes this particular mutation unusual, and this I believe will get relevant to how we are all feeling about AI, is something Shiller calls “constellation” effects. Narratives don’t operate alone, instead they cluster. Right now “AI will take your job” narrative is running alongside the AI bubble narrative and the AGI narrative. They stack on top of each other as one narrative makes the others more contagious.

If you want to see what a narrative constellation looks like in practice, look at what happened with Citrini Research a few weeks ago. They published a thought experiment on Substack imagining what the economy might look like in 2028 if AI displacement accelerates. I feel that what made it so contagious was not any single claim.

It was a brilliant piece that wove together:

• the labor replacement narrative,

• the bubble narrative,

• the financial contagion narrative,

• and the knowledge worker being automated narrative

into one vivid, emotionally coherent story.

As a result, markets moved quite wildly. Software stocks had a free fall. Bloomberg even linked the selloffs to the report. Let me repeat this one more time, a thought experiment on Substack shook Wall Street. Not because it contained new data. But because it bundled fears that were already circulating into a single story that hit harder than any of them could have alone.

Well folks, I suppose that is the constellation effect at work.

So if you are feeling anxious about AI without knowing clearly why, this might be it. It is not one story that affects us. It is a cluster of narratives hitting you from every direction. Social media, Dinner table, Cocktail party, Family gathering.

Now what’s the point of all this? Well it does not mean that the fear is wrong. The fear inside us is very real. But rather, the point is seeing the structure of fear more closely. Once we see the structure, we are no longer just inside the narrative, we are observing it.

Like a person watching a movie at the cinema, noticing not just the story in front of them but seeing how beautifully the actors act out a scene. How well setup the lighting is. How crafty is the script writer in playing with our emotions.

Note that I am not at all sharing this to trivialise the gravity of what is currently happening. I believe that disruption is very real and is underway. I see it in my own work every day and I imagine you see it in yours. But there is a difference between experiencing disruption and being swept up in a narrative about disruption.

Shiller helped me see that difference. And I hope you do too. Because it has now been easier for me to ask a more useful question. Instead of asking

“should I be scared or excited”

we can nudge ourselves to ask

“what’s in front of me now, and what can I do with it?”

Just like how the cinema owner has been doing just that, genuinely curious how he can use new technology to his advantage. He probably does not need to read Shiller. He seems to be asking the right questions already.

I have been learning a lot about what happens after you ask that question too. About what it actually looks like to work with AI in my line of work.

But first, that would involve Claude and an 80 year old man (almost).

To be continued

Last Thursday, Budi and I had an incredibly insightful conversation with Chris Angkasa. For those who missed it, I am very sorry. We are trying to push Substack to resolve a technical difficulty and get the live recording (for some reason it does not get saved in my draft). So if that gets recovered, we will publish it. But one key lesson that stood out, amongst many gems that were mentioned, was the emphasis to write down things that we do not know instead of things we know.

Let me repeat. One of the key lessons was to write down the things that we do not know (surrounding our investments) instead of the things we know.

This resonates deeply with Nassim Taleb’s concept of epistemocracy. In his book the Black Swan, Taleb discusses about an epistemocrat, a person who is well aware of the limitations of their knowledge. Instead of being overly confident towards exclaiming their knowledge prowess, they continue to question the truthfulness of their understanding. As a result, they will not be rushed to commit to a brash decision, they will make claims with adequate qualification and they will not be overzealous in their predictions of the uncertain future.

Taleb loves epistemocrats. He even goes to say that a utopia is a world that is ruled by epistemocrats. A society that gives their constant questioning of knowledge legitimacy. He calls that world an epistemocracy.

Well, it needs no convincing that such a world obviously does not exist today. A politician who claims that they don’t have a full picture about how the country would progress amidst global uncertainty will not gain votes. As the overconfident counterpart will make better promises and fare better in elections.

An entity who’s in the investment space (for e.g. Recompound) who does not know the 2nd order impact of a middle eastern conflict to an earnest public company selling ceramics to a domestic market in Indonesia will not gain as many followers as an investment guru who can “reverse engineer” and time when a market crash would occur.

But what’s the value of a person who admits the things they don’t know?

Understandably, you might be thinking:

“what’s so special about being an epistemocrat? All I am seeing is a timid mostly ignorant overthinker who is reluctant in making decisions with respect to their knowledge that they aren’t 100% sure about. How do they gain followers, influence, power and ultimately, money?”

Let me help you clear the confusion.

The value of an epistemocrat is not in the predictions they make. It’s in the predictions they refuse to make. They catalog their ignorance, writing down the things that is uncertain and unknowable. When they are tempted to want to know the unknowable, they avoid doing so.

Let me make this concrete. Picture a pool table.

It is a Friday night and you are playing billiard with your group of friends, a boys night out at a Jazz bar in Canggu. Now it is your turn to hit that 8 ball with the cue ball. Given the initial position of the cue ball, how hard you hit the 8 ball, you could mathematically compute the positions of the two balls after.

Guess what’s the endpoint? Well if you guess that the 8 ball will be in the top right corner as the cue ball remains in the same quadrant, then you are correct.

But, your friend, Bob, are given the position below as the end point after your turn.

Do you think Bob could guess where the 8 ball and the cue ball came from initially? I don’t think so. It is quite impossible because the cue ball and the 8 ball could start initially from anywhere! I hope you’d agree with me on this.

Therefore, it is borderline crazy to say with relative confidence where the ball come from, except Bob starts yapping:

Actually, look at the spin on the 8 ball — see how it’s slightly off-center? That tells you the cue ball struck it at roughly a 35-degree angle with a left English. The only position that produces that angle while leaving the cue ball where it is? Lower-left quadrant, two diamonds from the corner pocket. And your friend is right-handed, tends to favor an elevated bridge, which biases his break angle. Factor that in and honestly, there’s only one plausible starting configuration. Cue ball bottom-left, 8 ball dead center. I’d bet money on it.

Bob is so confident 😂

Well, here are more explanations that you might hear from time to time. They don’t confine themselves to you playing pool with your friends. The overconfidence extends to cover economy, politics, markets and so on.

Either in WhatsApp groups, family conversations or straight up “experts” fire chats. Below quotes might ring a bell. Although I have to give you fair warning that identifying causal errors in the made up sentences can feel unnatural*.*

On COVID-19: “If you were paying attention to the epidemiological data out of Wuhan in late December, the R0 alone told you everything. A novel respiratory virus with asymptomatic transmission in a hyper-connected global economy with daily international flights? There was only one outcome. I flagged this to my team on January 3rd and said — this goes global by March, lockdowns by April. The writing was on the wall. People just chose not to read it.”

On AI replacing jobs: “This was completely predictable. The job was 80% pattern recognition and template-based output — exactly the task profile that transformer models excel at. Once GPT-4 hit the market, anyone doing that kind of work had maybe 18 months. The companies that didn’t restructure were just in denial. I’ve been saying since 2022 — if your job can be described in a two-page SOP, it’s already gone.”

On climate change: “The science has been there since the 1800s — Arrhenius literally calculated CO2 warming in 1896. You had ice core data showing the correlation, satellite measurements confirming it since the 70s, and every single climate model trending in the same direction. When you combine that with post-industrial emission rates doubling every 40 years, there was only one trajectory. I’ve been saying since 2005 — we hit 1.5 degrees by 2030, extreme weather events triple by 2025. This wasn’t a prediction, it was arithmetic.”

You have got to love AI for saving me time digging through my WhatsApp and making me these texts. If needed, you can also use it as your thinking buddy by asking “where’s the causal error in the sentences that I am reading, what other range of possibilities that can occur”.

Value of not knowing things

Great, now that we have established the logical fallacies people are tempted to make in cocktail parties (and their corresponding mispredictions), we can move on to the central argument of this article:

which is being an epistemocrat maximises the chance of long term compounding.

Exponential Growth Bias

Now you have 5 seconds to think and guess. Promise? 5 seconds only. What is the answer to the below multiplication question?

Ok proceed once you have submitted your answer.

This is a math question that was asked in Kahneman and Tversky’s experiment. The median estimate of the group’s response was 512. When the question is flipped, starting from 8 times 7 and so on, the median estimate rises to 2,250. The correct mathematical answer is 40,320. If you guessed incorrectly, how surprised are you? Well certainly my guess was wrong when I first attempted the question.

This experiment tells us that we like to massively underestimate exponentially growing numerical series. We tend to fail to grasp the non-linear acceleration of growth, towards the end.

It is reported that the size of this underestimation is considerable: it is common for two-thirds of subjects to produce estimates that are at or below 10% of the actual (normative) value. For example in another test, many subjects expected a population growth target that would naturally be hit in 1979 to not be reached until the year 2000.

What else does research say about exponential growth bias? A number of things. But one thing that strikes the most to me is

people’s immunity to experience, education, and instruction when it comes to having this bias.

• This bias occurs even if you have a strong mathematical background: A subject’s number of college credits in mathematics had no significant correlation with their ability to accurately predict the growth.

• This bias occurs even if you have an extensive professional experience: An experiment testing professional decision-makers (members of a government Joint Conservation Committee) whose jobs require understanding growth processes showed they performed no better than naive undergraduate subjects. Daily experience with growing processes does not reduce the underestimation.

• This bias even occurs after a direct instruction: When subjects were given a 75-minute lecture on exponential misperception, shown numerical examples, and given immediate feedback before being tested, their core perception of the mathematical exponent only improved slightly. Instead of truly understanding the non-linear curve, these “sophisticated” subjects simply attempted to evade the error by universally inflating all their guesses linearly across the board.

Wow, so compounding, being the 8th wonder of this world also happens to be something that is:

• grossly underestimated

• incomprehensible to our human brain

Now that we know that wrapping our mind into compounding is somewhat of a loss cause, let’s take another look at another property of compounding.

It cannot compound if it gets disrupted

Remember your overconfident friend, Bob, from a moment ago? Yes he’s the one who plays pool but he also likes to say things from time to time about the market.

“Look, it was obvious. The yield curve inverted in August, manufacturing PMI was contracting for three straight months, and the Fed was hiking into a slowing economy. Layer on the fact that insider selling hit a two-year high in Q3 and you had every textbook signal flashing red. Anyone who didn’t see this coming simply wasn’t reading the data. I told my subscribers in September — we’re six weeks from a correction. Sell everything.”

You listened to him as his prediction was correct. So a few months down the line, he muttered the exact same line of argument, this time about the geopolitical tension that is causing investors to be jittery. He decides to again, sell everything and you decide to follow suit.

Only to find out weeks later that geopolitical tension actually causes a steep rally to commodities and commodity producing stocks that you used to own. Because of Bob’s overconfident assertions, you missed 10 best days of your stock portfolio. Well not so much of a big deal right?

Wrong.

According to JP Morgan’s research, $10,000 invested in the S&P 500 from January 2005 to December 2024 would have grown to $71,750 if you stayed fully invested. Miss just the 10 best days — ten days out of roughly 5,000 trading days — and you’re left with $32,871. That’s more than half your wealth. Not because the market crashed. But because you weren’t there when it recovered.

And here’s what makes Bob especially dangerous: those 10 best days occurred within two weeks of the 10 worst days. The biggest rallies can happen right after the sharpest drops — exactly the moments when Bob has already convinced you to sell everything.

But being wrong is not the issue here, it is the act

I make wrong predictions all the time. I predicted that Manchester United would win the Premier League after Sir Alex Ferguson retired under David Moyes (you guys are probably thinking that I was hallucinating, I was not). When Cristiano signed for Man Utd again, I also thought that year was going to be the year. But nope. And you could probably see clearly that my wrong prediction of Man Utd’s glory is not a problem. Why? Because it costs me nothing except some dignity amongst my Liverpool “friends”.

Bob’s prediction cost you $38,879.

Bob’s overconfidence didn’t just make him wrong. It made him act. And the action — not the necessarily wrongness — is what killed the compounding.

So what would an Epistemocrat probably do?

They draw a sharp line between predictions they can enjoy (football, cocktail parties, WhatsApp group debates) and predictions they refuse to act on (portfolio decisions). I suppose like normal people, they still have opinions. But they stop letting opinions that fall on the “unknowable” list touch their money.

Overconfidence attacks compounding from the outside, through unnecessary action. Exponential growth bias attacks it from the inside, by making you feel like nothing is happening so why bother staying invested. The long-term investor is fighting on two fronts at the same time. The epistemocrat’s toolkit — writing down what you don’t know, resisting the urge to act on noise, accepting that years will feel flat — is a defense system against both attacks at once.

That is why being an epistemocrat can maximise your chance of long term compounding. Not because you know more than Bob. But because you know what not to act on.

All of this starts with one shameless/ful admission: we don’t know some things about macro and about our own investments. And yet we invest anyway.

Simple in theory. Counter intuitive to wrap our head around. Very hard to practice over and over again.

I sure hope Substack can find the Live recording with Chris Angkasa last Thursday. There were more gems covered in the talks. Or we can do another one 🙂

References

In case you’re interested in reading the materials consulted in writing this article, below are the entry points:

1. Taleb, N. N. (2007). The Black Swan: The Impact of the Highly Improbable. Random House. Chapter 12: “Epistemocracy, a Dream.”

2. Tversky, A. & Kahneman, D. (1974). “Judgment under Uncertainty: Heuristics and Biases.” Science, Vol. 185, No. 4157, pp. 1124–1131. [The 1 × 2 × 3 × ... × 8 anchoring experiment.]

3. Wagenaar, W. A. & Sagaria, S. D. (1975). “Misperception of Exponential Growth.” Perception & Psychophysics, 18, pp. 416–422. [The foundational paper on exponential growth bias: two-thirds of subjects estimating below 10% of the actual value, immunity to instruction and professional experience.]

4. Stango, V. & Zinman, J. (2009). “Exponential Growth Bias and Household Finance.” The Journal of Finance, 64(6), pp. 2807–2849. [More-biased households borrow more, save less, and favor shorter maturities.]

5. JP Morgan Asset Management (2025). Analysis of S&P 500 Total Return Index, January 2005 – December 2024. Cited in: Megan, L. “Selling Out During the Market’s Worst Days Can Hurt You, Research Shows.” CNBC, April 7, 2025. https://www.cnbc.com/2025/04/07/selling-out-during-the-markets-worst-days-can-hurt-you-research.html

6. Recompound Live with Chris Angkasa. [Recording pending recovery from Substack.]

• Is Indonesia really at risk of being downgraded by MSCI

• What does it mean for your portfolio?

What we discuss include:

• Breaking down the fear vs the facts

• Explain how global index decisions work

• Why short-term headlines shouldn’t drive long-term investment choices

• The importance of entry price, valuation discipline, and staying focused on fundamentals rather than market noise

If you’ve been tracking market swings and wondering how structural decisions like MSCI classifications affect your long-term wealth building, this episode gives a clear, grounded perspective — complete with real-world examples and client questions answered in real time.

Even if you chance upon this long after the MSCI fiasco, the lessons learned are timeless. Adhering to these principles may help you weather through future storms in the equity market, and maybe even in life.

As always, this is not an investment advice on the mentioned ticker. This is for educational purposes only. Always do you own research or consult your investment advisor before making your own investment decisions.

Let’s get into this.

Part of my 2026 new year’s resolution was to get better at running. What better way for me to get better at running by forcing myself to join big running events, so that I could get my coveted 10K or HM Personal Best. One competition for me that stood out was JAKIM 2026. It looked legit as I recalled in the previous years where the competition was organised, there were plenty of road closures in Jakarta area, around the running event.

So when I saw in a running calendar event that JAKIM 2026 was open for registration, I speedily rushed to my phone and attempted to register. The registration process was awful. The instructions was vague and the website load was slow. The worst part? I have to download yet another banking app to register. So I downloaded Bale from BBTN, opened an account and… failed to register because tickets were sold out. I thought to myself, “wow running has been super popular nowadays in Jakarta” but again I thought to myself “wow this Bale app by BBTN must have gotten a lot of users from this running event” so let’s take a look at BBTN.

The app itself must be properly built because it handled large amounts of traffic quite smoothly. As people are forced to make a bank account and settle ticket payments there, I wonder how’s their Current Account / Savings Account ratio (CASA ratio). Banks love low cost funds because they are in the lending business. You’d want the cost of your funds to be low (giving people less interest) so that you can have higher spread when you lend the money out to people, giving you higher net interest margins. So if you hold super established running events and make participants bank with you as a pre-requisite to purchasing those tickets, you’d get access to cheap funds.

Quick note: CASA Ratio measures the proportion of current and savings accounts in a bank’s total deposits. Current accounts are low-cost funds (little to no interest paid), while time deposits are more expensive for banks. The higher the CASA Ratio, the cheaper the bank’s funding base — and generally, the better.

In marketing words, Running is top of funnel (TOFU) and as people get acquainted with the bank’s app and its usability, they get pushed down the funnel as they use the app more extensively for other banking services (like getting a mortgage).

So I went to look at the CASA ratio numbers since 2022 (the first time they held a JAKIM marathon was in 2023).

• Desember 2022: 48.5%

• Desember 2023: 53.7%

• Desember 2024: 54.1%

• December 2025: 48.7%

I’d say the CASA ratio growth is not too shabby, although I have seen better numbers in other banks. But given BBTN’s price to earning’s ratio about ~5 and BBTN’s price to book value which is about ~0.5, it might be interesting for me to go deeper because the bank is showing signs that it wants to stay relevant, but the market is rating it as if it is filled with super bad assets in a sunset industry. So it makes me curious as to what is BBTN’s main business. Why market seems to be quite harsh on BBTN. I don’t think there will be a right or wrong answer to this, but it is fun to explore.

Investors are forward looking

We invest today not to get the returns immediately. There is usually an underlying thesis why the company we invest today will be more valuable tomorrow. We’d be more inclined to invest if the company is poised to grow. So what are the growth drivers for BBTN?

Well, BBTN is a specialised bank that is focused on mortgages. So it really is more of like a property business but with a much faster sales cycle (you don’t have to wait for developers to build and handover to their customer for you to acknowledge sales in your books), the moment a customer takes in a mortgage, the bank gets revenue (remember, you pay the bank an interest if you take mortgages!!). For BBTN’s case, the moment a real estate developer develops their houses, they also get revenue. So they get revenue from both property buyer and real estate developer. But it also has its downside where property developer would get their money as soon as there is a buyer, BBTN has to deal with the uncertainty of home buyers defaulting on their mortgages. Then they are left with illiquid real estate assets.

But what type of property is BBTN focusing on? Affordable housing.

The core portfolio is split between subsidized (FLPP) and non-subsidized mortgages, with the former representing approximately 64% of the total loan book as of 9M25.

Who’s buying the houses? Well we have seen that millennials represent ~79% of the subsidized customer base. This demographic focus typically results in higher Loan-to-Value (LTV) stability and long-term customer retention. But because this is subsidized housing, it follows that not everyone (especially the bourgeoisies) can participate. Your monthly income cannot exceed a certain threshold, otherwise you will have to pay non-subsidised interest on your mortages. But if you do qualify, interest is super attractive and your likelihood to default is also minimised. More encouragingly, the income cap to qualify for FLPP has been raised.

FLPP Income Eligibility Criteria (Updated April 2025)

This tells us that the number of people who can qualify for this loans would also increase. Which translates to better quality loan books for the bank.

Cool, now that the bank potentially will have access to more quality borrowers qualifying for FLPP program, does the bank has enough liquidity to make the loan? Well yes, we can see that the deposit growth is outpacing loan growth. You might think is it because the Finance Minister Purbaya who placed Rp 25 Trillion into the bank? You are correct, but even if we exclude the massive government fund placement, the core deposit growth still stood at a very healthy 9.2% year on year (up to 10.0% in specific institutional/retail metrics).

So it looks like it is going to be very good for shareholders isn’t it? On the cost side we are seeing that macroeconomic factors are giving that extra “ooomph” for the bank to grow. For example, Bank Indonesia lowered its policy rate to 4.75% in Q3 2025.

On the revenue side, we are also seeing the FLPP program growing from the government’s 3 million housing program for millennials who by the way makes up a growing chunk of the population (Indonesia is experiencing a bonus demography).

Since 2015 (back then it was still one million house), BBTN is the dominant contributor to these national figures, consistently contributing ~60% of the national housing realization annually through both demand-side financing (Subsidized Mortgages/KPR) and supply-side financing (Construction Loans for developers)

According to BBTN’s sustainability data, the national achievement of the program per year is as follows:

• 2015: 699,770 units

• 2016: 805,169 units

• 2017: 904,758 units

• 2018: 1,132,621 units

• 2019: 1,257,852 units

• 2020: 965,217 units

• 2021: 632,373 units

• 2022: 875,832 units

• 2023: 878,727 units

Rest assured, BBTN is highly specialised in supporting government programs like this and supporting FLPP would require expertise in cashflow management. Which means it is likely that it will be more and more favoured vehicle to realise the government’s program into the future (2026 and so on). Now we could see that the narrative is there and the case for increase of profitability from higher net interest margin is also there.

But what’s with the market valuing the company as if the company is in a sunset industry? Are investors in IDX crypto bros who believe that blockchain and DeFi will make banks a dinosaur? Jokes aside, let’s try to find out more about potential setbacks that BBTN is facing.

So what are the setbacks

Well COVID. If you noticed during COVID years the number of units that got constructed above took a dip. But more important than that, it is reasonable to believe that COVID would affect the quality of mortgages because many people lost their jobs during that time of lock downs and ceasing of business operations.

During COVID, Bank Indonesia allowed banks to "restructure" loans. This meant a loan could be "non-paying" but still classified as "performing" on the balance sheet. This allows a company to keep booking "accrued interest" as profit even if the customer hasn't paid a single Rupiah in cash.

We see this clearly when we compared accounting profits and actual cash generation. Accounting profits was relatively stable at Rp 1.6 Trillion (ignoring the anomaly in 2019). But Cash receipts dropped from Rp 3.4 Trillion in 2019 to 1.8 Trillion in 2020. So this is already telling me that the bank was hit by a double-whammy: a collapse in new loans (seen in the construction dip) and a deterioration of the existing loan book because of COVID.

If I were in the CEO’s seat, my first priority would be to clean up the balance sheet. Bad assets should be dealt with decisively so they don’t become a structural drag on profitability for years. When Pahala N. Mansury handed over to Nixon L. P. Napitupulu, who assumed the role of Acting CEO in February 2021, the signals suggest that this is precisely what management began to focus on.

One area that reflects this effort is the Loan at Risk (LAR) coverage ratio.

A loan categorized as “at risk” (LAR) has elevated probability of becoming a Non-Performing Loan (NPL), so prudent management would increase provisions to cushion against that possibility.

Historically, throughout the 2010s, LAR coverage hovered around ~15%. Since Mr. Nixon’s appointment, it has been raised to the 19–21% range. Directionally, this is the right move — it signals greater conservatism.

However, as outside investors, we do not know whether this level of coverage is sufficient. Asset quality risk is dynamic, not static. If underlying borrower stress resurfaces, or if post-restructuring performance deteriorates, LAR coverage may need to be increased again. That would come with consequences: higher provisioning, lower near-term earnings, and potentially weaker returns on equity.

In other words, while management appears to be acting prudently, there remains uncertainty around whether current reserves are adequate for the true embedded risk in the loan book. Until that uncertainty clears, the market is unlikely to assign a premium valuation — and as investors, we must acknowledge that limitation.

For now, there seems to be quite a number of puzzle pieces that remain unresolved:

• Is the management going to be able to continue executing the program 3 juta rumah well? Is the management going to steer clear from the non-subsidized mortgages (which is known to be taking a hit nowadays due to recent layoffs)?

• Has the COVID asset quality deterioration been handled well?

• Is the management going to be able to be less reliant on time deposits as source of funds? It is heading towards that direction with the running events, Bale, and corporate initiatives but we need to see a stronger signal while there are other banks are already showing more sexy CASA numbers.

If you are an investor, how would you value this kind of company? Do you think it has enough margin of safety? Multiples wise its cheap but it feels less firm because a number of things could go wrong although the tailwind is right in front of our eyes. Especially given the fact that we are seeing signs that management has been doing the right thing. Do they deserve to be appreciated better? Or no? But regardless, we haven’t made ourselves comfortable in building a position.

Sunday is a family day.

My parents are 60 years old and getting older by the day. My dad suffered a mini illness recently, which has made Sunday an even more sacred day to dedicate time for my family. We’d go “hunting” for new coffeeshops or Kopitiam that just opened in Jakarta and my Mom always has a new place in mind (Instagram victim) that inevitably offers a more attractive place to hangout and drink coffee on a Sunday. And let’s not forget that peanut butter toast with butter (extra emphasis with the butter). Please don’t tell me you have not tried that one yet, although make sure you get your exercise down too.

But obviously with every Chindo (Chinese Indonesian) family, lunch is never just a family lunch for us to chitchat happily about life and spend quality time. It is also a war zone.

The children (in other words me and my siblings) need to fight a psychological battle with my Dad in two crucial battle fields:

• Number one, who made better progress in their business (i.e. making money). How much is your revenue this month? Why does it sound low?

• Number two, who sounds more intelligent in their knowledge about geopolitics or macroeconomics. Hey, I bet that the invasion of country X to country Y is going to cause the oil prices to rally. I managed to sell put options and make a profit from the premiums!

• [Optional - sometimes discussed] Who lives a healthier lifestyle, day to day.

Boy, it is a brutal battle. And usually a losing one.

Especially for me because I run Recompound. You might be thinking, hey isn’t Recompound a good business? Well yes, but not a good one to brag to your Chindo Dad. Let me explain.

The fee structure with customers we handle is on a monthly basis. So if there’s no fee that we charge, because our customers’ portfolio don’t surpass previous all time high, there’s literally nothing to brag about. Not that I brag about our performance to people, which you will find out why later.

Furthermore, we don’t look at macro. If you are a potential customer who are hoping to get macro analysis to predict movements of the market, we will surely disappoint. So don’t come to us if you are looking for macro analysis cause-effect → prediction. Find someone else.

The last banger is that the way we do investing is hyper boring. No promises, no hype, no multibagger rallies. Just investing in earnest businesses with good quality management and healthy cashflow. Helping people hold for the long term. We lost against the index in 2025 as our performance was about 18%.

I mean, who in the right mind would brag about 18% returns to their friends over coffee? 😂 The brags I hear on coffee shops nowadays is 50% profits, minimum.

So during these psychological wars vs my Dad, I’d often stay silent. I’ll be vague about how the business is going. I will also just nod my head about money flows from US to Japan to Gold to Bitcoin and to my pocket. Not because I am an obedient son, but really because I have no idea about the development of geopolitics and international macroeconomics, unless it is a key business variables for the companies we look into. I do not want to pretend to know. Let alone make predictions out of it.

So what’s the emotion that usually come about during these conversation to me?

Well, to be frank, I’d feel that “I am not good enough”.

Now you might be thinking, but you have achieved this and that and yes I am very proud of the achievements that we have managed to earn thus far.

But feelings of being “not good enough” is as real as it gets. Feelings of fear when markets crash after the recent foreign selloffs or after seeing a long post about how the future of the country is bleak is real. Feelings of greed when everything seems up, like there’s no way that your investments are going to suffer from drawdown or there’s no way that target price is not going to get hit, such that you want to topup with more funds — is also very real.

These feelings from my experience almost cannot be controlled or suppressed. They surface whenever they want to. They never ask for permission to appear because we are only human from the way our brain is structured.

But the odd thing is that I keep doing what I do. Even when I feel that “I am not good enough” every weekend during this psychological war. Why?

Because it turns out, by investing in stocks in a conservative way, we get two things in return:

One, our customers’ and my ability to focus on our craft. Our customers are as sharp as a knife when it comes to looking our investment thesis document. They question fair payback period, industry assumptions and diligence that is done on the companies. But, once those are done, they get to focus on their work and career accordingly.

Two, we get to experience lucky breaks, during which we always practise gratitude. From day 1, we’d want our customers to have as low of an expectation as possible. Our team expects very minimal returns from capital appreciation. We carefully assess the quality of company earnings and likelihood of dividend yield. The buying opportunities become a conservative bet to get good earnings via higher dividend yields.

So when companies get appreciated through things that we don’t expect — buybacks, tender offer, or we don’t know, other people noticing how good and cheap the companies really are — we get to practice gratitude.

So yeah, not everything is about returns although returns is a very important component. But the manner in which the returns are produced turns out to be very important for these high achieving individuals.

But I suppose my experience running this thing called Recompound has made me realise that I am most proud of doing what I do because our customers can continue focus on building their career, giving value to other people through their businesses and developing their craft. I realise that feeling of being a “not good enough” son is a potent fuel to push me into action. Even though I may never be good enough of a son, I think my actions and intention allow me to say that “I am good enough” of a person 🙂

Now get your grenades ready.

I used to think Howard Marks was being a bit too simple, maybe even a little naive.

In his famous memos and talks, he often hammers home points like:

• “Macro-forecasting is not critical to good investing”

• “Disavowal of market timing”.

I thought:

Really? In a market as volatile as Indonesia, you’re just going to ignore the macro and not try to time the entry / exits?

Then, the last week of January 2026 happened.

The 2026 MSCI Fiasco: A Reality Check

The recent IHSG crash was a brutal wake-up call for many.

After the election of Thomas Djiwandono as Deputy Governor of BI, the market didn’t just stumble—it went into a full-blown “trading halt” as over-inflated “Konglo” narrative stocks melted down. The fear was real: rumors swirled that Indonesia could be downgraded to a Frontier Market, a move that would have a catastrophic domino effect on our fixed-income & currency markets as well.

Many social media influencers were still shouting “Bull Cycle!” just days before the crash.

Their logic? Indonesia’s weighting in the MSCI is only about 1.25%, so global institutions wouldn’t bother being dumped by us.

They were dead wrong. The institutions did care. The bubble didn’t just leak; it popped. Overnight, reforms were swept in, and leadership at the OJK and BEI were replaced as the government scrambled to fix the coordinated trading activities and “gorengan” actions that led to this mess.

The Report That Saw the “Pucuk”

Just six days before the crash, on January 22, we released our Recompound Monthly Report to our clients:

Excerpt of our January 2026 report to clients

Within the report, we said that we didn’t have a crystal ball.

We didn’t know the MSCI fiasco would happen exactly on January 28.

But we did see the “late-cycle” signals that Howard Marks warns about:

• Retail Participation: Share of Average Daily Trading Value (ADTV) was back at the “mania” levels of 2021.

• Narrative Chasing: Money was crowding into “Konglo” stocks where price action was driven by stories and momentum rather than actual earnings power.

• Public Euphoria: Google search interest for “saham” was printing fresh highs—a classic sign that we were super close to the “Pucuk” (peak).

  

Risk Control, Not Market Timing

When the crash hit, the IHSG dropped over 9% in a matter of days. But our portfolio’s drawdown is minimum:

This wasn’t because we “timed the market” and moved to cash at the perfect second.

It’s because we followed the Oaktree philosophy: The primacy of risk control.

We didn’t just buy “good assets” (read: stocks); we bought them well.

1. Earnings-Backed Stocks: Our holdings are positioned as cash-producing assets with high dividend yields (~9% p.a.). Dividends reduce your reliance on price momentum. When the “narrative” dies, the cash flow keeps you afloat.

2. Margin of Safety: We stayed disciplined. Even when names looked attractive, we stayed on the watchlist if the “margin of safety” was too thin.

3. Avoiding the Crowd: While retail was rushing into the trendy tickers (I won’t name names here, but it should be obvious to you), we were researching companies with stabilizing earnings quality in the real economy.

Buying Well is Your Only Protection

Who could have timed this pop beforehand?

Honestly, no idea. But you don’t need to time the pop if you aren’t holding the balloon.

As we wrote in our report: “Be a student of history, not the victim of history”.

If you buy assets with a wide Margin of Safety and strong cash flows, a broad market correction becomes an opportunity, not a thesis break.

Success in investing isn’t about being the smartest person in the room who predicts the crash. It’s about being the most disciplined person in the room who buys well enough that the crash doesn’t matter.

Note from Toby: this week’s post is a guest post by a hyper intelligent human from University of Cambridge, Shania Mustika. If you find it interesting, do leave a comment!

My friend and I are co-authoring a paper on the spread and death of conventions.

Our question is as such:

What determines the fate of the conventions of a minority group — whether they are abandoned, persist in isolation, or spread to the wider population?

Putting it into concrete life context, here are some examples.

My partner used to like eating out and not like exercising. But because we got into a relationship, he has to spend a lot of time with me. We once went on a vacation with my two other siblings. So there were 3 people from my family — all of whom like to eat home-cooked food and think of going to the gym as part of vacation activity. By the time the vacation ended, the first thing my partner did was to sign up for gym membership and drew up a meal-prep plan.

Here is another example. My older brother is an avid diver. When he started diving 7 years ago, he liked to show off pictures of sea creatures he took, which at the time meant little to me. Who cares about tiny little worms among seafans which you have to squint and spend a good 5 minutes to spot while keeping yourself still and careful not to destroy the corals (the little worms are called nudibranch, if you are curious). But he kept going on and on about his diving trips, kept spamming our family groups with pictures, kept talking about the glorious sharks and mantas and barracudas… Until I felt envious. I decided to give this diving thing a try and now we have an annual diving crew made up of a clan leader (my brother) and 5 followers (me being one of them).

The two examples show how minority culture dies and spreads. The first shows how my partner is a minority among our group and his convention gets abandoned. The second shows how my brother is a minority and his convention gets adopted into the mainstream.

Some math and economics

So my friend and I decided to write this example into mathematical language (maybe because we don’t have anything better to do).

We model this interaction into a game played by individuals. The game is described as follows:

• A person can belong to either a majority or a minority group.

• Each group has their own preferred action. For example, majority prefers action A (going to the gym) and minority prefers action B (not going to the gym).

• Individuals from each group will only be able to interact with each other and get positive payoff if they choose the same action. So if there are two people Alice and Bob, they will only get positive payoff if they both choose A or they both choose B.

• If a majority/minority individual compromises and chooses his non-preferred action, but is able to interact with another individual, he will also get positive payoff, but smaller than what he would get should he choose his preferred action.

• If you have a hard time understanding the set up, here is a concrete example. My partner belongs to the minority group and prefers not going to the gym. I belong to the majority group and want to go to the gym. If we choose our own preferred action, we are both unhappy as we cannot spend time with one another. If my partner compromises and joins us at the gym, I am very happy; he is also happy, but slightly less because he needs to expend energy. Conversely, if I compromise and skip the gym, my partner is happy (because we are likely to go get food at some cafe); I am less happy because I don’t get my brain-derived neurotrophic factor (BDNF, or the feel-good chemical one gets after exercising). Now back to math.

Now the payoff matrices below illustrate this interaction.

So now there is a trade-off. Suppose you are a minority individual. You can stick to doing B (not going to the gym), but that means you cannot interact with the rest of the people. If my partner chooses not to exercise, he might be happy because he doesn’t have to expend energy, but he will be left alone while all of us are at the gym (I promise going to the gym together with your friends has a high payoff. I.e. quite fun).

There are 3 possible outcomes of such interaction.

1. Everyone adopts the same action A (going to the gym). Call this All-A society.

2. Everyone adopts the same action B (not going to the gym). Call this All-B society.

3. Majority keeps its action A while minority keeps its action B, and they don’t interact with each other. Call this a segregation (S) society.

What is the outcome of such game? After some calculations on Nash equilibrium and selecting an equilibrium that is stochastically stable (I will not bore you with this), the result of this model is as such.

In human words: when the share of minority is small, the minority will choose to abandon its convention and adopt the majority convention. When the size of minority is intermediate, both outcomes are possible and we are not sure who will win. When the size of minority is large enough, there is enough people from the minority group to sustain its own conventions.

But minority cultures do spread

Yes, the results above almost mean that minority culture will never spread. But we do observe this in real life.

My brother was the only diver among my family, but managed to get everyone to spend our hard-earned money on diving gears.

You are in a WhatsApp group where 1 person started running and playing padel, but soon that becomes your go-to Sunday activities.

Sometimes bad practises can spread too. For instance you have a colleague who starts coming in slightly later to work. Slowly other people find it okay to come late too and soon people only start arriving at the office at 10/11am.

We think that this can be explained in two ways. First is minority premium. This means that the action of the minority is seen as superior, either because the action itself is superior, or because the minority group is seen as more elite that the rest of the people want to emulate.

The second way this can happen is if there is strong minority attachment. This happens when the minority is more attached to their own action and is more reluctant to change their action.

The result is a payoff matrix that looks slightly different.

Payoff matrices for minority premium case.

Payoff matrices for minority attachment case.

This leads to some changes to the resulting equilibrium. The results become:

When the size of a minority group is small, the minority will choose to abandon its convention and adopt the majority convention. When the size of minority is intermediate, segregation is the outcome (previously it could be A or B). When the size of minority is large enough, the majority will adopt the minority action (previously it was segregation).

The result seems to be in line with common sense. When the minority is very small, they may choose to adopt the majority action so that they can interact with more people. When the minority is mid-sized, each group is happy enough interacting within their own bubble. When the minority is large enough, the majority may want to emulate the minority because it elevates their own status, and this status elevation outweighs their discomfort in changing their behavior (in the case of minority premium). In the case of strong minority attachment, the majority may choose to compromise for the minority, since the latter is not willing to do so.

Below is a pictorial representation of the result.

The takeaway for you

We are always surrounded by people.

Sometimes we unknowingly get influenced by the people around us. If we are influenced positively, then that’s good for us. But if we get influenced negatively, then we want to be cognisant of that and think of how to change our environment to avoid being pulled down.

This also gets me thinking. Maybe the reason why influencers can be so powerful is because they are able to create a minority premium persona. We tend to admire those people because they have the physique that we want, they can go on business class flights, etc etc. We think of these attributes as enviable and therefore we adopt their behavior in hopes of becoming like them.

At the same time, this model may give you strategy to influence the people around you, in positive ways I hope. If you want to get your family to exercise, maybe show them examples of people they admire who also exercise. If you want your kids to sleep early, be headstrong enough to also sleep early (so that they don’t have anything fun to do in the evening) and do fun activities in the morning.

I hope you have fun reading this and do share with me your own experience navigating majority-minority conventions in the comment section!

P.S. If you are interested in reading the paper, with all the mathematical proofs, do leave a comment and I can share it with you when it is ready to be circulated :)