Why Bangladesh Bank should think twice before cutting rates

Highlights:

• Rate cuts risk reigniting inflation via exchange-rate depreciation
• Food inflation drives persistence; largely immune to monetary policy
• Food price shocks fade slowly, keeping headline inflation elevated
• Non-food inflation transmits food and exchange-rate pressures economy-wide
• Monetary policy affects inflation mainly through the exchange rate
• Hold rates; improve food market contestability for durable disinflation

Bangladesh Bank is about to announce its Monetary Policy Statement for the rest of FY26. Once again, a familiar question has surfaced: should it cut the policy rate and begin easing monetary conditions?

The pressure is understandable. Headline inflation has come down from its peak, the real policy rate is positive, and private investment remains weak. Business groups are lobbying hard for cheaper credit. On the surface, a rate cut looks like the obvious—and politically attractive—next step.

But a closer look—based on monthly data from July 2019 to October 2025—at how inflation actually behaves in Bangladesh, how it rises, how it spreads across sectors, and how it eventually cools, suggests that cutting rates now would risk reigniting inflation through the exchange rate, slowing disinflation rather than supporting durable growth.

What makes inflation persistent

Headline CPI inflation in Bangladesh is persistent, but not explosive. When inflation rises, it tends to remain elevated for many months, continuing to erode household purchasing power long after the initial shock. At the same time, inflation does not spiral indefinitely. Over time, it drifts toward a level shaped by underlying price dynamics rather than short-term demand fluctuations.

Those dynamics are dominated by food prices. Food inflation sits upstream in Bangladesh's inflation process and is the primary source of persistence in headline inflation. This is why inflation often remains stubborn even as macroeconomic conditions tighten.

Food inflation is highly inertial. Roughly 85 percent of a food price shock carries over from one month to the next, imparting substantial stickiness to headline inflation. Prices also adjust asymmetrically: food prices rise quickly when shocks hit—whether from supply disruptions, policy frictions, or global price movements—but fall back only slowly when conditions improve.

Crucially, food inflation appears largely insulated from standard macroeconomic levers. It does not respond systematically to domestic credit growth, broad money growth, or even exchange-rate movements. Instead, it is best explained by its own past. Once food inflation rises, it unwinds only slowly—on average, only about 15 percent of the initial shock fades each month, leaving much of it in place for a long time.

This pattern points to structural features of food markets rather than macroeconomic overheating. Dominant intermediaries, markup-setting behavior, and informal extraction weaken competitive discipline. In more contestable markets, exchange rates and liquidity conditions would matter more. In Bangladesh's food markets, they largely do not.

BB's own value-chain evidence reinforces this conclusion: food inflation in Bangladesh reflects structural frictions and weak contestability across key supply chains, not excess demand—explaining both its persistence and its limited sensitivity to monetary tightening.

How shocks spread

If food inflation anchors persistence, non-food inflation is how shocks spread across the economy. Non-food inflation is less persistent than food inflation, but more responsive to macroeconomic conditions—especially movements in the exchange rate.

Persistent food inflation spills into non-food prices through cost-of-living pressures and expectations. When food prices rise, higher rents, service charges, and markups become easier to justify. Roughly half of a sustained increase in food inflation eventually feeds into non-food prices, while the reverse does not occur: non-food inflation does not feed back into food prices.

Non-food inflation is sensitive to exchange-rate movements. A 1 percent depreciation of the taka raises non-food inflation by about 0.15 percentage points within a few months—a meaningful and relatively quick pass-through that builds over time, with roughly one-third of the initial depreciation eventually showing up in prices. This response is strongly asymmetric: depreciation pushes prices up far more forcefully than appreciation pulls them down.

Together, these dynamics explain why inflation in Bangladesh rises quickly and broadly, but falls slowly and unevenly. Even after exchange-rate pressures stabilise and non-food inflation begins to ease, headline inflation remains elevated until food prices finally turn.

Implications for monetary policy

These dynamics imply a different role for monetary policy in Bangladesh than the one embedded in standard demand-based frameworks.

In conventional models, core inflation—typically defined as CPI excluding food and energy—is treated as a proxy for excess demand. Rising core inflation signals overheating; falling core inflation suggests that tighter policy is cooling demand. That interpretation does not fit the Bangladeshi data. Demand-mediated monetary transmission—where tighter policy lowers inflation by cooling domestic spending—appears weak or absent.

This does not make monetary policy irrelevant; it changes the channel through which it matters. In Bangladesh, monetary conditions influence inflation primarily through the external sector. Easier credit and lower interest rates raise demand for imports, increasing pressure on the taka. Exchange-rate depreciation then feeds directly—and asymmetrically—into non-food inflation.

Within this framework, non-food inflation plays a role analogous to "core" inflation, but with a different meaning. It does not reflect excess demand. Instead, it reflects how upstream food and exchange-rate pressures are spreading through the economy, as movements in the taka are rapidly passed through into the prices of imported goods, energy, transport, and other non-food items. Core-like measures are therefore informative as indicators of transmission, but not as signals that demand is overheating or that policy space has opened.

The absence of demand-driven inflation in domestic prices does not imply unconstrained monetary space. While credit expansion may not raise inflation through excess demand, it can still do so indirectly by worsening the external balance, weakening the exchange rate, and triggering cost pass-through into non-food prices. In this system, inflation is not demand-led but exchange-rate–mediated. Because food inflation adjusts downward only slowly, such shocks keep headline inflation elevated even after exchange-rate pressures ease.

Credit expansion can, in principle, raise output through depreciation and higher net exports. In Bangladesh, however, this channel is weak and unstable. Export supply responds slowly, imports are highly input-intensive, and depreciation feeds quickly into domestic costs, eroding real competitiveness. Growth benefits are therefore uncertain and temporary, while the inflation costs are more predictable, asymmetric, and persistent.

The decision at hand

The upcoming Monetary Policy Statement will be judged on whether BB chooses to ease or hold steady. The evidence suggests that cutting rates now would be premature. Easing risks weakening the exchange rate, reigniting inflation, and prolonging the disinflation process.

Monetary policy alone cannot fix food inflation. But BB can still influence food price dynamics by improving contestability in food markets—most directly by reducing non-price frictions in foreign-exchange and trade finance, including clearer, more predictable rules for authorised banks to open import LCs for essential food items and lower discretionary barriers that allow supply to be restricted.

A natural concern is whether easing trade-finance frictions for food imports could weaken the exchange rate and reignite downstream inflation. But improving predictability in foreign-exchange access is not equivalent to monetary easing or import subsidisation. Food import demand responds primarily to supply gaps, while rules-based access reduces hoarding and rent extraction rather than inflating volumes. By easing persistent food price pressures, such measures can support inflation expectations and exchange-rate stability over time.

Staying the course—while supporting currency stability and improving food market contestability—offers a more credible path to sustained disinflation than premature easing in the current inflation regime.

Zahid Hussain is a former lead economist of The World Bank, Dhaka Office