Every federal budget is a stark reminder of how much Pakistan’s federal government is unable to spend within its means. Therefore, the burden to keep fiscal balance somewhat manageable falls on the same few sacrificial lambs, typically the formal sector in terms of collection and development needs for expenditure.
Troublesome as it may be, the country’s gross public debt ratio of 70 per cent is not outrageously high by developing economy standards. However, one big problem is its concentration: over the past decade, commercial banks have held the bulk of the federal government’s debt.
Of Pakistan’s Rs54.5 trillion in domestic debt, the bulk sits in marketable instruments, worth Rs46.6tr; of that, scheduled banks hold Rs36.8tr, or 79pc. Insurers account for under 5pc, mutual and pension funds for about 6pc, and a catch-all bucket of “corporates and others” for the rest. This makes the bank-sovereign nexus extreme by global standards.
A World Bank analysis from the end of 2024 put Pakistani banks’ public-debt holdings at roughly 60pc of total assets, four times the global median and the highest in a sample of over 80 countries. As a result, the effect on credit activity has been highly detrimental, with the industry’s advances-to-deposits ratio hovering below 40pc and the share of small and medium enterprises barely 10pc of private-sector loans.
Shift even a tenth of the Rs54tr domestic stock out of banks and into retail hands at a yield just 150 basis points cheaper, and the annual saving runs into the region of Rs80bn
Since the two balance sheets of banking and sovereign are wound so tightly together, the relationship has curdled into something toxic. The government borrows from banks, taxes the profits from that borrowing, and banks push money away rather than put it to work. Somewhere in this loop, both the depositor and the real economy have been forgotten.
When 79pc of the outstanding paper sits with a small club of institutional buyers, those buyers carry real pricing power into every auction; a market with retail savers, pension funds, insurers and foreign buyers each holding a meaningful slice generates competitive tension that bears down on yields, and a bank-dominated one simply does not.
The institutional money that would normally provide that tension, chiefly the insurers, is too small to matter: at roughly 0.9pc of GDP against about 4pc in India, the entire sector’s asset base is smaller than a single year of government borrowing. That leaves retail, and on paper, the case for it is compelling. There are already millions of Pakistanis lending to the state through the old National Savings Schemes, currently holding Rs3.6tr.
This segment has historically accepted lower yields than banks for the same sovereign credit, so widening the base could also trim the debt-servicing bill. A new policy InsightLab at the Karachi School of Business & Leadership, Karachi, argues that despite new instruments and platforms, the set of creditors holding Pakistan’s debt has barely changed over the past seven years. Banks still hold the vast majority.
The new channels changed how the debt is sold, but not who buys it. Shift even a tenth of the Rs54tr domestic stock, some Rs5.5tr, out of banks and into retail hands at a yield just 50 to 150 basis points cheaper, and the annual saving runs into the region of Rs25bn to 80bn. This would make a noticeable difference to the debt-servicing bill, which has become the single largest line in the budget and only compounds each year.
There is a structural prize too. Banks gravitate toward short-term and floating-rate paper, largely because their liability mix forces them to do so. Pakistani banks hold hardly any fixed deposits, just Rs6.1tr out of Rs37.3tr, so they cannot comfortably warehouse long, fixed exposure. A genuine retail base anchored by long-dated household savings would take on the very tenor the banks shy away from, easing the rollover risk that the current profile does nothing to address.
For a government desperate to rein in its largest expenditure line, retail is the rare lever that lowers both cost and risk at once. But the question is: how does the sovereign reach this segment? Historically, that answer was National Savings, though it is not without shortcomings. Its rates are set by administrative fiat in discrete steps, so they lag the market.
This is attractive to savers when rates fall, but it is a structure that works against the state’s own objectives, is untradable, capped at Rs5 million, and is pitched more as quasi-social security for widows and retirees than as a serious financing tool. The second route runs through the capital markets by issuing Sukuk directly at the Pakistan Stock Exchange. But this has fared no better at changing who holds the paper.
Since December 2023, the government has auctioned Ijarah Sukuk through the exchange to dazzling headline demand, yet the paper is fully Statutory Liquidity Requirement-eligible, individuals cannot bid directly when fewer than 1pc of citizens hold a brokerage account, and banks still end up holding close to 90pc of the stock.
Third is the diaspora channel, the Roshan Digital Account, and truly the one relative win: over 927,000 accounts opened and more than $12.7bn received since 2020, though Naya Pakistan Certificates, the debt instrument inside it, have never crossed 2pc of government external debt.
The newest effort tries to fix the access problem at its root. Investor Portfolio Securities (IPS) accounts have long let individuals hold government paper in principle, but in practice, the channel meant branch visits, manual forms, and bank staff with little incentive to promote it, so few ever used it. The State Bank of Pakistan’s InvestPak portal, launched in November 2025, builds on that plumbing and strips out the friction. It does so by allowing individuals to open an IPS account, bid at auction, and trade securities entirely online.
In theory, it is the most promising of the lot, with one catch: the access still routes through bank-maintained IPS accounts, the very institutions with no commercial reason to usher retail investors toward an asset class they would rather keep for themselves.
India faced the same problem
and took a different route. Its RBI Retail Direct scheme, launched in 2021, lets individuals hold government bonds in an account directly with the central bank, cutting out banks.
If there is a single fix worth making, it is to stop flying blind. Pakistan now runs several parallel retail channels and publishes consolidated data on none of them, so nobody can actually say whether the needle is moving. The holder-wise statistics do not even carry a separate line for individual investors.
The rest follows from there: a genuinely retail-sized product rather than the Rs100,000 minimum tickets that pass for one today, and an honest decision on whether to keep routing retail through the banks or, as India did, around them. None of this pays off in a single budget. But the concentration did not build itself overnight either, and years of inaction cannot be undone in days.
After seven years of new instruments, the holder of last resort is still the bank. It will stay that way until the state builds something savers can actually use and a route that doesn’t run through the institutions it is trying to move beyond.
Mutaher Khan is co-founder of Data Darbar and Head of InsightLab at KSBL. Shahzaib Abbasi is an analysts at InsightLab.
Published in Dawn, The Business and Finance Weekly, June 8th, 2026
