A Stock Market That Does Not Trade: Savings With Nowhere to Go

Executive Summary. Bangladesh's stock market turnover ratio was 14.54% in 2024, the lowest of the five countries reporting, against a peer median of 61.66%: Vietnam 78.39%, India 65.23%, Thailand 58.09%, Pakistan 38.08%. The median peer's market changes hands 4.2 times as fast as Dhaka's. The market is small as well as still. Capitalisation was 5.745% of GDP in 2025, smallest of the six and 8.7% of the peer median of 65.69%, with 360 listed domestic companies against a peer median of 868. Twenty years earlier the ratio was 4.749% of GDP, so two decades produced a rise of 20.97%, a multiple of 1.21. None of this is a shortage of savings. Gross savings ran at 34.81% of GNI, second highest of the five countries reporting and 5.5 points above their median of 29.32%, and broad money reached 48.8% of GDP. Nor does the bank channel take up the slack: private credit was 34.5% of GDP, below the peer median of 40.15%. An economy of 173.6 million people and US$450.1 billion of output has built neither channel to scale.

A ratio that already controls for size

The turnover ratio is the value of shares traded during a year divided by the market's capitalisation. It is a speed, not a size. A country with a small exchange can post a high number, provided the shares that exist actually change hands.

Bangladesh posted 14.54% in 2024. Vietnam posted 78.39%, India 65.23%, Thailand 58.09% and Pakistan 38.08%. Indonesia has no 2024 observation in this series, so this is a comparison of five countries and not six, and Bangladesh is last of the five. The peer median is 61.66%, which puts Dhaka at roughly a quarter of it.

That is not a statement about how small Bangladesh's exchange is, because the exchange's own size is the denominator. It is a statement about whether the shares that have already been issued are traded at all.

Source: World Bank World Development Indicators, stock market turnover ratio (CM.MKT.TRNR), 2024. Indonesia has no observation for that year.

Three mechanisms would produce this number, and the data here separates none of them.

The first is free float. Shares locked in the hands of sponsors, the state or permanent blockholders sit in the capitalisation figure but can never reach the numerator. A market where most of the register never moves prints a low turnover ratio mechanically, whatever the appetite of the investors who remain. WDI carries no float measure, so this essay cannot distinguish a thin float from thin demand.

The second is the absence of institutional money. Where pension books, insurance funds and asset managers hold size, they rebalance, and rebalancing is volume. Nothing in these six indicators measures who owns the shares.

The third is price formation. If a price is administratively prevented from reaching a level where a buyer will bid, the trade does not occur and the volume never appears in the numerator. Rules of that kind are not recorded in any WDI series, so that channel is equally untested here.

Three candidates and no way to test any of them with this data is the boundary of the argument, and where the next piece of evidence has to come from.

Small, and no bigger than in 2005

Market capitalisation of listed domestic companies stood at 5.745% of GDP in 2025, the smallest of the six countries. India recorded 266.9%, Thailand 88.8%, Indonesia 65.69%, Vietnam 61.4% and Pakistan 17.26%. The peer median is 65.69% of GDP, which leaves Bangladesh 59.94 points below it and at 8.7% of the median value.

The listing count tells the same story from the supply side. Bangladesh had 360 listed domestic companies in 2025, the fewest of the six, against Pakistan's 535, Vietnam's 400, Thailand's 868, Indonesia's 956 and India's 8,510. The peer median is 868, so Bangladesh's count is 508 firms short of it, or 41.5% of it.

The time series is the part that should trouble anyone watching the real economy over the same window. Bangladesh's market capitalisation was 4.749% of GDP in 2005 and 5.745% in 2025. That is a rise of about one percentage point of GDP, 20.97% in relative terms, a multiple of 1.21 across twenty years in which the economy itself multiplied several times over. The exchange did not shrink. It simply did not participate.

Source: World Bank WDI, market capitalisation of listed domestic companies (CM.MKT.LCAP.GD.ZS), 2025.

One caution belongs beside those two figures rather than in a footnote. In the underlying series, Bangladesh's 2025 listing count falls by nearly half against its own 2024 reading, and India's 2025 count more than triples in a single year. Breaks of that shape are what a change in what gets counted looks like, not what a year of issuance or a year of delisting looks like. The 2025 cross-section is therefore the weakest evidence in this essay. On the 2024 readings of the same capitalisation series, Bangladesh sat above Pakistan rather than below it. Read "smallest of the six" as a claim about the 2025 vintage, and treat the turnover result, which is 2024 and shows no comparable break in any country's series, as the load-bearing one. The twenty-year comparison is anchored at 2005 and 2025 and inherits the same vintage caution at its endpoint.

The money is not the missing part

The intuitive explanation for a market this quiet is that a poor country has nothing to invest. The savings data does not support it.

Gross savings ran at 34.81% of GNI in 2025, second highest of the five countries reporting, behind Indonesia at 35.04% and ahead of India at 34.7%, Thailand at 23.93% and Pakistan at 15.23%. The peer median is 29.32%, so Bangladesh sits 5.5 points above it. Vietnam has no observation in that year, which is why this is a comparison of five.

Monetary depth points the same way. Broad money was 48.8% of GDP in 2024, second of the four countries reporting, behind Thailand at 141% and ahead of Indonesia at 41.77% and Pakistan at 38.21%. India and Vietnam have no 2024 observation here.

Two honest qualifications. Gross savings as WDI defines it is national income less consumption plus net transfers, so a country receiving large remittance inflows records a higher ratio for reasons that say nothing about whether firms can reach that money. And broad money is a stock of liquid liabilities held at banks. It measures where savings are parked, not what they finance. Neither indicator proves there is a pool of equity capital waiting for a listed instrument to buy. Both make it hard to argue that Bangladeshi households have nothing to allocate.

So the question becomes which channel the savings run through. The bank channel is the obvious candidate, and it is shallower than the peer group too.

Domestic credit to the private sector was 34.5% of GDP in 2025, fourth of the five countries reporting, above only Pakistan at 10.73%. The peer median is 40.15%, leaving Bangladesh 5.65 points below it. India recorded 44.03%, Indonesia 36.28%, and Thailand 143.1%, which is the range the same set of countries spans. Vietnam does not report in this year.

Source: World Bank WDI, domestic credit to private sector (FS.AST.PRVT.GD.ZS), Bangladesh against comparators, 1995-2025.

Note the shape of the finding. Bangladesh is not a bank-based economy that skipped equity markets. It is an economy where the equity channel is the smallest of the six and the bank channel is below the peer median. The savings are near the top of the group and the two formal routes from savings to firms are both narrower than the comparators' routes.

What a market that does not trade costs

Trading is not the point of a stock market. Pricing is, and pricing requires trades.

Where volume is this low, an unlisted Bangladeshi company has no market benchmark for what its own equity is worth. Valuation stops being an observation and becomes a negotiation, which advantages whoever has better information, which is almost never the outside investor. Private capital that does enter has no visible exit, so it demands a return that compensates for illiquidity, which raises the cost of equity for every firm in the economy including the ones that never intend to list.

The consequence runs into firm size. A company that cannot sell equity funds expansion from retained earnings and whatever a bank will lend against collateral. Its scale is then capped by the owner's existing balance sheet rather than by the return on the project in front of it. That is a distributional fact as much as a financial one: it determines who is allowed to build a large firm, and the answer is people who already own assets.

For households, the arithmetic is simpler. With the equity channel effectively shut, savings sit in deposits and in assets these six indicators do not capture at all, land and gold and lending inside a family. The saver's return becomes whatever spread the banking system chooses to pay, in an economy that grew 4.2% in 2024.

None of this is established by the cross-section. These are six countries observed together, and the association between a shallow market and a bank-dependent corporate sector is consistent with causation running either way, or with a third factor, weak contract enforcement being the standard candidate, producing both. Thailand's 143.1% of GDP in private credit and Vietnam's 78.39% turnover ratio sit on top of very different institutional histories. The claim here is that a gap exists and is large, not that the gap explains Bangladesh's growth path.

What would change this conclusion

A checkable test. If the turnover ratio rises above 38.08%, Pakistan's 2024 level and the lowest of the four peers reporting, while the listed company count stays near 360, then the binding constraint was trading friction and not the supply of listable firms, and this essay's weight on issuance is misplaced. If the listed count climbs past 535 while turnover stays near 14.54%, the reverse. Both are observable in the same WDI series within a few releases.

Three moves.

1. Publish a float-adjusted capitalisation series and a definition-stable listing count. A market cannot be managed on a statistic that halves between vintages, and a turnover ratio computed on capitalisation that includes permanently locked shares flatters nothing and diagnoses nothing. Owner: Bangladesh Securities and Exchange Commission, with the Dhaka and Chittagong exchanges. Success signal: a published free-float capitalisation series, and a listing count that no longer breaks between WDI vintages the way the 2025 reading breaks from 2024.
2. Put long-term institutional money on the bid. Provident, pension and insurance pools are the natural marginal buyers of listed equity, and their absence is the most plausible untested explanation for a turnover ratio of 14.54%. Owner: Finance Division, with Bangladesh Bank and the insurance regulator. Success signal: turnover above 38.08% in a WDI release by 2030, which would still leave Bangladesh below the peer median of 61.66%.
3. Make listing worth its cost to mid-sized firms. Three hundred and sixty listed companies in an economy of this size is a supply problem before it is a demand problem, and disclosure cost, tax treatment and the loss of family control are the levers that decide it. Owner: Bangladesh Securities and Exchange Commission, with the National Board of Revenue. Success signal: listed companies above 535 and capitalisation above 17.26% of GDP, both Pakistan's 2025 levels, which would still leave Bangladesh far below the peer medians of 868 firms and 65.69% of GDP but would end a two-decade flatline.

The counterargument

The strongest objection is that an illiquid equity market is not a problem worth solving. Plenty of economies industrialised on bank credit and retained earnings, capital markets came later, and a country at Bangladesh's income level should fix its banks rather than chase a deeper exchange. On this reading, 14.54% turnover is a symptom of a sensible sequencing, not a defect.

Two answers.

First, the sequencing argument presumes the bank channel is doing the work, and the data does not show a bank channel doing the work. Private credit at 34.5% of GDP is below the peer median of 40.15% and fourth of the five countries reporting. A bank-based development model is a real model, but it requires deep banks. Bangladesh has neither deep banks by this comparison nor a market that trades. That is not a sequence, it is a gap in both channels at once.

Second, the objection treats equity as a substitute for credit when the two do different jobs. Equity absorbs loss ahead of deposits, prices risk in public, and gives an owner a way out that does not require selling the firm to whoever is standing nearest. A banking system without a functioning equity market beneath it carries risks on its own balance sheet that shareholders elsewhere carry on theirs. That is a financial stability argument, not a capital markets enthusiasm, and it survives even if every claim about growth in this essay is set aside.

---

Data sources: World Bank World Development Indicators, retrieved from the BDPolicyLab data lake, 2026-08-10. Comparators are India, Pakistan, Indonesia, Thailand and Vietnam; "peer median" is the median of the comparators reporting in the stated year, excluding Bangladesh. Turnover and broad money are 2024, being the latest years with comparable coverage; market capitalisation, listed companies, private credit and gross savings are 2025. Coverage is incomplete in several years: Indonesia does not report turnover for 2024, Vietnam does not report private credit or gross savings for 2025, and India and Vietnam do not report broad money for 2024, so those comparisons are stated as five or four countries rather than six.

Sources

1. World Bank WDI, Stock market turnover ratio (%), CM.MKT.TRNR: https://data.worldbank.org/indicator/CM.MKT.TRNR
2. World Bank WDI, Market capitalization of listed domestic companies (% of GDP), CM.MKT.LCAP.GD.ZS: https://data.worldbank.org/indicator/CM.MKT.LCAP.GD.ZS
3. World Bank WDI, Listed domestic companies, total, CM.MKT.LDOM.NO: https://data.worldbank.org/indicator/CM.MKT.LDOM.NO
4. World Bank WDI, Domestic credit to private sector (% of GDP), FS.AST.PRVT.GD.ZS: https://data.worldbank.org/indicator/FS.AST.PRVT.GD.ZS
5. World Bank WDI, Gross savings (% of GNI), NY.GNS.ICTR.ZS: https://data.worldbank.org/indicator/NY.GNS.ICTR.ZS
6. World Bank WDI, Broad money (% of GDP), FM.LBL.BMNY.GD.ZS: https://data.worldbank.org/indicator/FM.LBL.BMNY.GD.ZS