Collector
The fixed-rate trap | Collector
The fixed-rate trap
Business Recorder

The fixed-rate trap

The government’s cash needs never end. External financing is thinner, repayment obligations are still arriving, and a larger share of the fiscal burden is being pushed onto the domestic market. This month alone, Pakistan has had to manage the April 8 Eurobond repayment alongside a wider external-payment squeeze that is already pressuring reserves. At the same time, the sovereign is trying to improve the duration profile of its debt by leaning more on fixed-rate borrowing. That may be sensible in theory. In present conditions, it is colliding with the balance-sheet structure of the banking system. The auction calendar makes the pressure visible. After the March 26 fixed-rate PIB auction with a Rs400 billion target, the government lined up another Rs450 billion for April 20 and Rs500 billion for May 5. The March auction itself drew heavy bidding, but the accepted cut-offs still landed at 12.5 percent on the 2-year, 3-year, and 5-year papers, 12.4 percent on the 15-year, and the 10-year bids were rejected altogether. Including non-competitive bids, total accepted face value crossed Rs466 billion. That is not a sign of effortless demand. It is a sign that duration now must be paid for. The working assumption behind this strategy is that banks can keep absorbing fixed-rate sovereign paper in size. That is where the problem begins. Pakistani banks fund themselves largely through deposits whose pricing resets quickly or whose maturities are too short to comfortably support a large incremental move into long-duration fixed-rate bonds. The asset may be long and fixed. The liability base is not. That mismatch is not an accounting footnote. It is the story. SBP’s December 2025 deposit data show current deposits of about Rs12.9 trillion, saving deposits of about Rs15.0 trillion, and fixed or term deposits of roughly Rs4.9 trillion, against total deposits of around Rs34.5 trillion. More importantly, by arithmetic from the same maturity buckets, well over 90 percent of term deposits mature within two years. So the idea that banks can painlessly fund more long-duration fixed-rate PIBs out of savings and short-tenor term money is fantasy. The only plausible base is current deposits, and even there, “sticky” is a behavioral assumption, not a contractual fact. The capacity argument is also weaker than policymakers seem willing to admit. SBP banking data already show banks sitting on a PIB stock above Rs24 trillion, with roughly Rs9 trillion in fixed-rate PIBs and about Rs15 trillion in floating-rate PIBs. In other words, the sovereign has already used up a large part of the banking system’s tolerance for duration. Asking banks to take materially more fixed-rate risk now does not happen in a vacuum. It happens on top of an already swollen sovereign book. OMO liquidity does not solve that problem. It can help banks fund the government in the short term, but it does not create a stable liability base for warehousing long-duration fixed-rate paper. On April 3, SBP injected about Rs13.3 trillion through 7-day and 14-day OMOs at 10.53 percent and 10.51 percent, respectively, while the policy rate stood at 10.5 percent. That is short-tenor, policy-linked money. Using it to sit on long fixed-rate PIBs only deepens repricing risk and weakens carry. So banks may still buy, but only at yields that compensate them for taking that mismatch risk onto their books. That is why the recent move in cut-offs should not be treated as noise. It is the market telling the government that the balance-sheet elasticity of banks is not infinite. The sovereign wants more fixed-rate funding to reduce its own interest-rate sensitivity. Banks, meanwhile, do not want to inherit that same sensitivity without being paid for it. The result is straightforward: either the government accepts sharply higher fixed-rate borrowing costs, or it backs away from its duration ambitions and remains exposed to floating-rate risk. There is no elegant way out. From here, every financing choice is costly. The only question is which cost the state chooses to recognize upfront.

Go to News Site