There is a trade that looks obvious right now. Indonesia’s four largest banks — BCA, BRI, Bank Mandiri, BNI — have fallen 23 to 46 percent from their peaks. BBCA trades at a PE of 14x against a ten-year historical average of 24x. BBRI is down 33 percent in fifty-two weeks. BMRI trades at 7.1x earnings with a 19 percent ROE. BBNI is below book value at 0.9x PBV. Every major broker has a Buy rating. Average analyst consensus implies 40 to 53 percent upside.

The retail investor sees this and says: cheap. The semi-sophisticated investor runs a mean reversion model and says: very cheap. The analyst at the brokerage says: accumulate on weakness.

They are not wrong about the number. The number is correct. Indonesian bank stocks are trading at historically low valuations relative to their own past.

But here is what they are missing: cheap relative to the past is only a buy signal if the future looks like the past. And the future of Indonesian banking — for the next two to three years — does not look like the past. It looks like something different. Something that requires a different framework entirely.

This volume is not a bear case on Indonesian banks. It is a precision instrument for understanding exactly what has to happen before they become a buy — and what signals to watch that will tell you it is happening before it ever appears in a financial report.

Because by the time it appears in a financial report, you are already late.

To understand why TGS has been underweight Indonesian banks since 2024, you have to understand one thing clearly: this is not a market accident. It is not a global risk-off story that will reverse when the Fed cuts. It is not a temporary rupiah shock. It is a deliberate, structural policy decision by the Prabowo administration — and it has a name.

We call it the Transition Tax.

When a populist government takes power with a mandate to lift the economic floor — to create jobs, to push money downward, to demonstrate that the state works for the grassroots — it needs a transmission mechanism. In Indonesia, that mechanism is the Himbara banks: BRI, Mandiri, BNI, BTN. These four state-owned institutions are not just commercial banks. They are policy instruments. And in 2025–2026, they have been deployed as such with unusual explicitness.

In September 2025, Finance Minister Purbaya Yudhi Sadewa transferred Rp200 trillion of government funds from Bank Indonesia directly to the Himbara banks. His stated reason, in his own words: “The goal is for banks to have a lot of money and they can’t put it anywhere else except lending it out. We’re forcing the market mechanism to work.”

That sentence is the entire thesis. The government is forcing the banks to lend. At KUR rates of 6 percent. Into segments with higher credit risk. With no meaningful ability to reprice. This is not a market outcome. This is a mandated outcome — and the NIM compression that follows is not a side effect. It is the price the system pays for the policy.

The divergence between BCA and the Himbara banks is not a coincidence. It is a structural signal. BCA is a private bank with the highest Current Account and Savings Account ratio in the system — which means its cost of funds is structurally independent of government rate mandates. When the government pushes cheap lending through Himbara, BCA watches from the side. When Himbara’s NIM compresses, BCA’s holds. This is why BCA outperforms in this specific phase — not because it is a better business, but because it is a different kind of business.

NIM is being squeezed from both sides simultaneously. From the top: the government mandates low lending rates on KUR and priority sectors. From the bottom: depositors demand higher rates because they have alternatives — SBN, government bonds, money market instruments — that are paying competitive yields. The bank is caught in the middle. Revenue ceiling pushed down. Cost floor pushed up. Margin disappears from both directions.

Most investors look at the bottom line. Net profit. That is the number that gets the headline. That is the number that gets compared quarter over quarter. That is the number that drives the consensus Buy or Sell call.

It is also the number that tells you the least about what is actually happening inside a bank’s earnings engine during a transition phase. Here is why — and here is how to read it correctly.

A bank’s profitability has five distinct layers. They do not move together. They move in sequence. And understanding the sequence is the difference between reading a financial report and understanding one.

The critical insight: bank management has discretion over when to recognize provisions, within regulatory limits. In Phase 1 — which is where Indonesia’s Himbara banks are right now — management provisions conservatively, fee income provides buffer, and the headline net profit number remains digestible. In Phase 2, the NPL that has been forming in the real economy for 6 to 18 months finally forces provision recognition at a rate that overwhelms the buffer. Net profit falls sharply. Analysts downgrade. The market finally sees what the data has been signaling for over a year.

The investors who were waiting for the financial report to tell them something was wrong — they are reading the outcome, not the signal.

The Transition Tax is not unique to Indonesia. Every government that has used its state-owned banking system as a policy instrument to drive populist economic transformation has gone through a version of this cycle. The most precise comparative case is India’s public sector bank crisis of 2012 to 2021 — nine years from the beginning of the policy burden to a genuine, sustained recovery.

Nine years. That is the India timeline. Indonesia may be faster — its institutions are different, its debt culture is different, and the global context is different. But the structural logic is identical. A government that uses its banks as a policy instrument creates a transition tax on bank profitability. That tax lasts until the policy transformation either succeeds or is abandoned. And the signal that it is ending does not come from the bank’s financial report. It comes from the economy the policy was trying to build.

This is where most analysis stops. It identifies the problem, maps the historical parallel, and concludes with “wait for conditions to improve.” That is not intelligence. That is description.

What serious capital needs is the specific, observable, pre-financial-statement conditions that signal the transition is working — that the economic transformation the government is attempting is gaining real traction, not just policy paper. Because the reversal in Indonesian banking will not be triggered by a rate cut. It will not be triggered by a government recapitalization. It will be triggered by something more fundamental and more durable: the emergence of genuine grassroots spending power.

Here is the precise thesis: the Transition Tax ends when the policy succeeds. When the population that the populist mandate was designed to serve — the grassroots economy, the micro-borrowers, the MSME owners, the informal sector workers — actually has enough income, enough cash flow, enough economic participation to sustain credit repayment on their own. Not because the government subsidized their borrowing. Because they earned enough to repay.

When that happens, four things change simultaneously: NPL trends downward from genuine repayment rather than restructuring. KUR demand rises from business expansion rather than survival borrowing. The government no longer needs to force lending at 6% because organic demand exists at market rates. And the banks — for the first time in this cycle — can reprice their margins upward without political resistance.

That is the inflection point. And it will appear in the real economy 6 to 12 months before it appears in a bank’s financial report.

Underweighting a sector does not mean sitting in cash. It means recognizing that the same policy environment which is creating a Transition Tax on banking is simultaneously creating tailwinds somewhere else. The Prabowo administration’s populist agenda — cheap credit to the bottom, infrastructure build-out, digital connectivity, food security, energy sovereignty — is a headwind for bank NIM. But it is a direct revenue signal for other sectors.

The intelligent rotation is not away from Indonesia. It is within Indonesia — toward sectors that receive the tailwind from the same policy that creates the banking headwind.

This is not a permanent reallocation. It is a phase-appropriate positioning. When the six leading indicators above begin to align — when grassroots spending power emerges, when KUR NPL shows genuine repayment, when the government is willing to allow rate normalization — the rotation back into banks will be one of the highest-conviction trades available in Indonesian equities. The banks will not just recover. They will re-rate. The question is not whether to own them. It is whether you are using the intervening period to build knowledge or just waiting.

TGS has maintained an underweight position on Indonesian banking — particularly Himbara — since 2024. Not because the banks are badly managed. Not because the Indonesian economy is failing. But because we read the policy direction before it appeared in financial statements, and what we read told us that the next two to three years would impose a structural tax on bank profitability that the market was not pricing.

We were right about the direction. The data has confirmed it quarter by quarter. Himbara aggregate profit growth went from +22.86% in 2023 to -11.26% in Q1 2025. NIM has compressed across every state bank. Provisions are rising. And the financial statements are only now beginning to show what has been building in the real economy for over a year.

We are now in Phase 1 — the slow compression phase, where net profit is declining but still positive, fee income provides buffer, and the headline numbers remain manageable enough that retail investors still see “cheap bank stocks” rather than “early stage earnings deterioration.”

Phase 2 is when the provision cycle accelerates beyond the buffer. When NPL in the micro and MSME segment forces recognition that can no longer be deferred. When quarterly net profit falls sharply enough that the consensus narrative shifts from “accumulate on weakness” to “deteriorating fundamentals.” That is when the uninformed capital exits. And that exit — paradoxically — may create the actual entry point for patient, informed capital.

The dividend yield argument deserves specific rebuttal, because it is the most common reason retail investors give for holding bank stocks through a transition phase. BBRI’s dividend yield at current prices exceeds 12%. That sounds compelling. But a 12% dividend yield paid from earnings that are declining — funded in part by a payout ratio that reached 91.98% in 2025 — is not income. It is capital erosion with the optics of income. If earnings continue to compress in Phase 2, the dividend will be cut. The yield will disappear. And the capital loss will dwarf the dividends received.

This is the final, most important point: holding Indonesian bank stocks today because they pay a high dividend while their earnings are in structural compression is not an income strategy. It is hope dressed as analysis.