The budget’s real question: who lends to the government?

Interest payments of Tk 1,27,500 crore in the budget presented on June 11 will consume nearly one in every five taka of revenue the government hopes to collect. That is the cost of servicing more than Tk 22 lakh crore of public debt from one of the narrowest sovereign investor bases among major economies. The deficit itself, at 3.55 percent of GDP, is moderate. What it leaves open is who lends the money.

The answer, mostly, is the banks. Of the Tk 2,43,000 crore gap, Tk 1,12,000 crore will come from the banking system and just Tk 15,000 crore from savings instruments; external loans and grants will cover the rest. Increasing net foreign borrowing by nearly 90 percent is the budget’s most underrated decision, and the right one. But it treats the symptom. The structural problem is that the state has, in effect, a single domestic lender.

That lender is in no condition to carry the load. Banking-sector capital adequacy turned negative in December 2025, a first in the country’s history; non-performing loans stand at more than 30 percent of all credit; private-sector credit growth has fallen to 4.72 percent, the lowest on record. Every taka the sovereign draws from a bank competes directly with a factory’s working capital.

Countries at Bangladesh’s income level have widened the base instead. The Philippines raised a record P585 billion, or about $10.5 billion, from a single retail Treasury bond in February 2024, sold through mobile apps with a minimum investment of roughly Tk 10,000. In the UK, I helped build the technology that now allows retail investors to buy government bonds and treasury bills directly. The mechanics are ordinary: a published issuance calendar, digital onboarding, and small minimum investments — Bangladesh can do the same.

Bangladesh has better raw material than Manila did: more than 80 million verified mobile-money accounts, remittances exceeding $30 billion a year, diaspora bonds whose ceilings were removed in December 2024 yet remain paper-era products, and a sovereign sukuk that was oversubscribed elevenfold this year. Expanding Sanchaypatra is not the answer. Its administered rates already make it the state’s most expensive source of borrowing, which is why this budget leans away from it. The solution is to put market-priced treasury bills on modern rails.

Converting that appetite into investment requires three steps, none of which requires new legislation. The Finance Division could publish a quarterly retail treasury issuance calendar. Bangladesh Bank could permit the subscription and settlement of treasury bills through mobile wallets, with NID-based onboarding. And the ministry could consider a diaspora bond, priced off the sovereign yield curve and distributed digitally through banking channels in London and the Gulf. India raised $4.2 billion this way in 1998; Kenya’s mobile bond, poorly marketed, raised only a fraction of its target. None of this replaces revenue reform; it gives the state room to manoeuvre while that harder work proceeds.

The budget’s stated theme is economic democratisation. Letting a garment worker in Gazipur or a nurse in Riyadh lend to their own state on market terms would be its most literal expression. And every taka raised that way is a taka the banks can finally lend to someone else.