Bank Mandiri’s recent annual report, the 2024 English-language edition, ends its editorial narrative on page 1,227. Pages 1,228 through 1,241 certify the report’s conformity with the ASEAN Corporate Governance Scorecard. Pages 1,242 through 1,263 are an OJK Reference Index. Thirty-six consecutive pages, in other words, whose only job is to show the judges of Indonesia’s Annual Report Award (ARA) that the report contains what they asked for — and all of that before the audited financial statements have so much as begun.

The investor who reads from the front needs no such map. The reader marking the document against the rubric does.

The thirty-six pages tell you, plainly, who the report is now written for.

Malaysia is about to make the same design choices, under a programme called MY Value Up. First, though, how far the Indonesian convention has already spread.

Bank Mandiri is not the outlier. The earliest version of this convention I have been able to identify appears in Bank Permata’s 2008 annual report, which closes with an eight-page Bapepam-LK Cross Reference index at pages 357 to 364. (Bapepam-LK, the capital markets regulator at the time, was absorbed into OJK in 2012.)

From that point the convention spreads. PT Bank Rakyat Indonesia’s 2024 narrative annual report runs to 832 pages excluding the audited financials, and includes its own twenty-two-page reconciliation to OJK rules and the ARA scoring criteria: a concordance, in effect, between the document and the rubric it will be marked against. Every listed bank that competes seriously for the award now produces a document organised in part around indices back to what the rubric asks for.

The report writes the map because the map is what the ARA rubric grades.

How a recognition scheme became a rulebook

This was not where the ARA started. The award was launched in 2002 by seven founding institutions: Bapepam (later Bapepam-LK, now OJK), Bank Indonesia, the Indonesia Stock Exchange, the Tax Directorate, the Ministry of State-Owned Enterprises, the Indonesian Institute of Accountants, and KNKCG (the National Committee on Corporate Governance Policy, today’s KNKG). Its scoring criteria were not clearly delineated, and rested on no underlying regulation.

The initiative was led by KNKCG, then chaired by Herwidayatmo, who at the same time held the chairmanship of Bapepam-LK. A year earlier, in the 2001 Code of Good Corporate Governance, KNKCG had set out the five GCG principles: transparency, accountability, responsibility, independence, fairness, or TARIF. The ARA in 2002 was built to score Indonesian companies against them. Eighty-three entered the first cycle.

The regulatory architecture now sitting beneath the award came later. A 2016 OJK regulation governs the annual reports of issuers; a 2021 circular specifies their form and content. Between them the two run to a hundred and ten pages, and the desk evaluation grades 365 separate disclosure items. All of it was built afterwards, to codify in formal regulation what the ARA had already, in looser terms, begun to ask for.

KNKG’s 2021 revision restated the same principles as PUGKI 2021’s four pillars: ethical behaviour, transparency, accountability, sustainability, or ETAK. The framework moved first, the rubric followed it, and the regulation arrived last to write down what the award had spent years implying.

The relevant fact is the sequence. The rubric did not extend the regulation. The regulation extended the rubric.

And at no public point in twenty-three years did the design process, or any of the revision points, seat foreign institutional investors active in Indonesian equities, or the IR practitioners who actually write these reports, in the room. I have found no public record of investors being asked what the criteria ought to contain, and none of IR practitioners being asked whether what they were optimising against bore any relation to what their investor audiences actually read.

The annual report grows because the rubric grows, and the rubric grows because it is the artefact of a closed conversation among the people who built it.

Five banks, twenty-two years of disclosure growth

Five of the largest listed banks in Indonesia, then and now. Page counts are shown as editorial pages, the narrative and governance sections a human being is actually meant to read, alongside the total bound document where one exists.

BankEarlier reportEditorial / total pp.Recent reportEditorial / total pp.
BCA2002 (bilingual)92 / 2072023 (English; ARA 2024 winner)566 / 743
Bank Mandiri2009 (English)307 / 4962024 (English)1,227 / 1,726
Bank Permata2003 (bilingual)200 / — *2018 (bilingual)670 / 874
BRI2008 (English)150 / — †2024 (English)832 / 1,257
CIMB Niaga2007 (bilingual)262 / 3762021 (English; ARA 2022 winner)697 / 1,044

* Permata’s 2003 edition was issued without the financial statements; no bound total exists. † BRI’s 2008 financial statements were published as a separate 115-page document.

Twenty-two years separate the two columns. Permata’s row is the noisiest, since its ownership and reporting conventions changed mid-period and its 2003 edition skipped the financial statements altogether (one way, admittedly, to keep a page count down). But every row points the same way. Nobody got shorter. Editorial length grew by somewhere between two-and-a-half and sixfold, and the longest reports belong to the most decorated winners — a correlation nobody has tried very hard to deny.

The figure even flatters the reports that switched from bilingual to English-only along the way: drop a language and the page count should fall, not multiply. None of this is the financials’ fault. Audited statements lengthen because accounting standards make them, and they are at least honest about being long. What is growing here is the editorial report, the part meant for a reader, drifting visibly away from one.

Some of that expansion is not the rubric’s doing. IFRS convergence, integrated reporting, and OJK’s own sustainability-reporting mandate under POJK 51/2017 each added pages on their own account, and a fair reading has to grant them. But those mandates added content. The rubric organised the document. What determines where a disclosure sits, what it is captioned, and which index it reconciles back to is the scoring criteria. Which is why the reports terminate not in a conclusion but in a concordance.

Bank Mandiri’s 2009 report, the year before the line bends sharply upward, gives pages 266 to 305 over to branch addresses. Forty consecutive pages of them. No analysis, no commentary, no map: just postal addresses, set in the same handsome serif as the rest of the book, as though a fund manager in London were about to mail a letter to the Medan branch. The ATMs, mercifully, were spared.

Pages 486 to 493, tucked in behind the financial statements, are an index telling you where each OJK-required disclosure can be found. In hindsight the 2009 report is a transitional fossil. The architecture the 2024 report is built around, content shaped to the rubric with indices pointing back to the regulator, is present in 2009 only in trace amounts. By 2024 the trace has become the whole skeleton.

The branch addresses, at least, appear to have gone, someone having worked out that an investor who wants a bank can generally find one.

What the long-only PM in Singapore is trying to read in those 1,227 pages of Mandiri’s most recent narrative is the same thing she was looking for in the 168 pages of the 2002 report: whether the bank is still running at the margin she has modelled, what management will spend capital on over the next two years, and how the book is positioned for the credit cycle she is trying to anticipate.

The content is in there somewhere. It is just not where she would look — the pages are not organised for her. They are organised for the rubric, which is to say for the regulator’s account of what a complete disclosure looks like. It is the same misdirection that, in founder-led companies, leaves the report speaking past the people who price the stock.

This is what regulatory capture of a disclosure regime looks like in practice, though “capture” is almost too dramatic a word for it. Nobody was bought. Nothing was schemed. Every party acted in good faith, and the ARA has, by its own success criteria, plainly worked.

The capture is structural. The regime answers to the people who built it, and the audience it is supposed to serve was simply never in the room when the rules were drawn.

What MY Value Up could do that Indonesia did not

The Securities Commission Malaysia and Bursa Malaysia introduced MY Value Up on 20 April 2026. The programme targets the 88 largest Bursa-listed companies and asks them to articulate mid-to-long-term growth strategies that go beyond compliance-level disclosure. It will be shaped through a series of technical workshops before the MY Value Up Guidebook is published around the middle of this year. SC Chairman Datuk Mohammad Faiz Azmi described the ambition as moving “from a culture of mere compliance to one of active value creation”.

The framing is exactly right, and the architecture is still unsettled. That is a rare and useful place to be.

The single decision that will determine which way MY Value Up goes is the composition of the design conversation between now and the Guidebook. As announced, that conversation has three parties: the regulator, the exchange, and the 88 listed companies, whose feedback in the workshops is the stated mechanism for shaping the programme. It is a good and sensible list.

It is also, almost exactly, the list Indonesia worked from — continuously, for twenty-three years. A rubric built by the regulator, the exchange, and the companies being graded will drift, over time, toward whatever those three can agree is a reasonable expectation. The output is foreseeable, because we have already seen it.

The party not named in that structure is the one the disclosure is actually for. Suppose the conversation also seated a standing panel of foreign institutional investors active in Malaysian equities (long-only funds in Singapore, Hong Kong, London, and the Gulf; sovereign allocators; specialist Asian managers), alongside the IR practitioners who actually write the disclosures. The rubric would come out differently.

Not because investors and IR people are wiser than regulators, but because they ask different questions. A rubric exposed to all three sets of questions while it is being designed is less likely to drift toward something that scores well only against the questions its authors already favoured.

The Qatar Stock Exchange’s IR Excellence Program is the regional precedent worth noting on this one point. Its outcomes are open to argument and the programme is still maturing, so the case for it is not “look at what it produced”.

The case is narrower, and harder to dismiss. When QSE built the framework in 2015, it engaged an independent IR advisory firm to administer it, and the assessment rests substantially on a survey of more than two thousand institutional investors and analysts active in the Qatari market. Whatever the programme produces in the long run, the question it asked at the design stage was a more honest one, because the people whose answers would matter were in the room.

If this dynamic sounds familiar at your company or inside your regulator, that is the conversation we are most useful in. Book a Disclosure Diagnostic.

The narrower regret

Indonesia’s ARA has, by its own metrics, worked. The disclosure quality of the seven hundred-odd Indonesian listed companies that compete for it has, in measurable ways, improved over twenty-three years, and the governance practices captured in the 365 disclosure items are a better set than the 1997-vintage practices the programme was built to replace. None of this is in question.

The regret is narrower than failure. It is that the success criteria were set by a closed conversation, and that no founding agency has, on the public record, asked the people the regime was built to serve whether they would have built it differently. That question is cheap to ask and expensive to skip.

Malaysia can still ask it. The answer will be visible in the Guidebook — in the weeks between now and the middle of the year.

If you sit on a Bursa-listed board, run IR at one of the 88 PLCs, or sit inside SC Malaysia and would like to discuss any of this, write to jonathan@iradvantage.asia or find us on LinkedIn.

IR Advantage advises listed companies across Asia on investor relations strategy and execution. The author held in-house IR roles in Indonesia through the period when annual report length expanded materially. Page counts are taken directly from the issuers’ published reports on their investor relations pages. The account of the ARA’s founding (including Herwidayatmo’s concurrent chairmanship of KNKCG and Bapepam-LK) draws on Mas Achmad Daniri and the ARA organisers’ press materials; the KNKCG-to-KNKG lineage and the post-crisis governance reforms draw on the OJK Indonesia Corporate Governance Roadmap (January 2014). The 365 disclosure items are taken from the Buku Kriteria ARA 2022. The two instruments referenced are OJK Regulation 29/POJK.04/2016 on the annual reports of issuers (18 pages) and OJK Circular Letter 16/SEOJK.04/2021 on their form and content (92 pages).

Advising listed companies representing over $50 billion in aggregate market capitalisation.

We help listed companies disclose to the audience they actually want to attract — not the one the rubric assumes.

Book a Disclosure Diagnostic

30 minutes. No obligation. You’ll know exactly where your disclosure gaps are.

Jonathan Zax Founder & President Director, IR Advantage IRC·ICIR·Wharton MBA·Harvard BA 30 years in investor relations. The author held in-house IR roles in Indonesia through the period when annual report length expanded materially.
← Back to Perspectives