What Norway's sovereign fund exit says about Dhaka's stock market

On 12 August, Norges Bank Investment Management reported the best half-year in its history: a record 1,753 billion kroner earned in six months, a 9.4% return on a portfolio worth $2.3 trillion.

One smaller number moved the other way. The fund's Bangladesh holdings fell 18% to $95.87 million, a six-year low, down from a peak of $248 million in 2020. The world's largest investor grew more confident almost everywhere. Dhaka was the exception.

The fund's weight in Bangladesh is trivial; its behaviour is not. It holds stakes in more than 10,000 companies across 67 countries and thinks in decades. Capital of this kind buys through crashes and wars, as it did during the pandemic and the war in Ukraine. What moves it out of a market altogether is a judgement about structure: whether trades settle reliably and whether the rules that governed the purchase will still govern the sale.

A six-year withdrawal spread across nearly every position therefore carries more information than any quarter of hot-money outflow. Fast money prices earnings. Patient money prices the machinery. When it leaves, it is not simply making a trade; it is delivering a verdict on the machinery.

The domestic record supports that reading. Bangladesh Bank's own survey puts foreign equity holdings at $914.58 million at the end of 2025, roughly 70% below 2020 levels and marking the eighth consecutive year of net outflow.

In June, foreign investors sold a net Tk358 crore of Dhaka-listed shares, the largest monthly sell-off of 2026, against purchases of just Tk6 crore. Yet the index finished the fiscal year up 19%.

Together, the two facts describe a market that has rerated domestically while derating internationally. Local savings returned. International capital, which can hold any market in the world, kept leaving.

The clearest evidence that structure rather than earnings is doing the work comes from the fund's largest local holding.

BRAC Bank produced a record profit in 2025, becoming the first private local bank to cross Tk2,000 crore, followed by a strong first half. Across the same period, foreign ownership fell from 36.7% to below 33%.

Every share represents two claims: one on a company's cash flows, and another on the market's ability to turn that claim back into money. Investors who sell a bank during its best-ever year are not necessarily discounting the company. They are discounting the second claim.

The global case

The counterargument deserves a hearing.

Foreign money touches perhaps 130 of 360 listed companies and is marginal to turnover in a retail-driven market. Frontier-market allocations have shrunk globally. The war pushed capital out of Asian equities. The fund's local manager describes the current conditions as a storm that may pass.

Each point is fair.

But an explanation based on global conditions must account for a retreat that began years before the war, and for a market two seas away that spent those years attracting what Bangladesh was losing. 

Vietnam is the natural experiment. In October 2025, FTSE Russell announced Vietnam's upgrade from frontier to secondary emerging-market status, effective the following month, with passive inflows alone estimated at billions of dollars. The mechanism matters more than the money.

No global fund can conduct its own due diligence on the settlement system of every small market. The industry therefore outsources part of that judgement to index committees. A classification is a coordination device: it tells thousands of allocators simultaneously that a market's machinery has been inspected.

Vietnam reformed to meet the inspection checklist rather than to improve sentiment. It scrapped pre-funding requirements for foreign institutions and established a formal process for failed trades.

Romania ran the same play a decade earlier, and I saw it from inside.

When Romania privatised its gas producer Romgaz in 2013 and electricity distributor Electrica in 2014 — a listing I worked on — the fiscal receipts were only part of the purpose. The listings were capital-markets policy.

The state selected assets large enough to create the index-eligible free float and daily liquidity the market lacked. Liquidity was the final criterion Romania had to overcome. The privatisations helped supply it, and in 2019 FTSE Russell upgraded Romania to emerging-market status.

Patient capital returned to Bucharest because Romania changed what there was to buy. Romania had EU membership behind it, which Bangladesh does not. But the supply-side logic travels.

Necessary reforms, incomplete reform

Dhaka's recent reforms are necessary but incomplete.

Bangladesh Bank eased repatriation for portfolio investors in May, while the securities regulator has proposed direct listings for large companies. The record June sell-off came after both measures, pointing to the deeper problem: investors are pricing the durability of the rules.

A procedure introduced by circular can be withdrawn by circular.

The floor price did lasting damage on this margin. The months of frozen prices may have cost less than the demonstration that prices could be frozen.

MSCI's June review made the consequence explicit: Bangladesh remains a frontier market, with currency convertibility flagged. A return of floor prices could prompt a demotion to standalone status — the category below frontier and outside most investment mandates.

The fund's weight in Bangladesh is trivial; its behaviour is not. It holds stakes in more than 10,000 companies across 67 countries and thinks in decades. Capital of this kind buys through crashes and wars, as it did during the pandemic and the war in Ukraine. What moves it out of a market altogether is a judgement about structure: whether trades settle reliably and whether the rules that governed the purchase will still govern the sale.  

MSCI will resume its Bangladesh reviews in November. November is the deadline. The evidence points to three actions, each with an owner and a test.

First, the BSEC and the exchanges should make Central Counterparty Bangladesh operational before the review. Incorporated seven years ago with Tk300 crore in paid-up capital, it has never cleared a trade. The test is simple: a first cleared trade, on record, before November.

Second, the Financial Institutions Division should nominate one large state-owned enterprise, or agree terms with a willing multinational, and file the market's first large-cap listing under the new public-offer rules, with a free float of at least 10% — the threshold proposed by the regulator itself. The test is a filing on the BSEC's desk this year.

Third, the BSEC and the Financial Institutions Division should issue a joint public commitment that floor prices will not return under any market conditions. A commitment that survives the next downturn is the only kind investors will price. The test is whether the next sell-off passes without a freeze.

None of this requires new money. All of it is visible from Oslo.

Fahim Chowdhury is an investment banker and managing director at RetailBook.

Disclaimer: The views and opinions expressed in this article are those of the author and do not necessarily reflect the opinions and views of The Business Standard.