Earlier this month, the National Bank of Ethiopia (NBE) removed restrictions on how much commercial banks can lend to their clients, nearly three years after it first introduced the credit growth cap in August 2023. The central bank’s monetary policy committee also announced decisions to raise the NBE’s policy rate by one percentage point to 16 percent as a way to counter inflation, and reduce the forex surrender requirement on goods exports to 30 percent from 50 percent in an effort to “enhance export competitiveness and build market confidence.”
In theory, the new policy should allow the country’s commercial banks to expand lending to the private sector at the expense of investments in government securities that have dominated their portfolios while the credit cap was active over the past couple of years. But how likely is this to happen?
The Reporter Magazine’s Mahlet Mehdi approached industry veteran Eshetu Fantaye for comment.
RELATED POSTS
Eshetu’s career in finance stretches back more than three decades, during which he served as Vice President of Awash International Bank and President of Bunna Bank and Ahadu Bank. He is also one of the founders of Goh Betoch Bank. Today, he works as a financial consultant and keeps a close watch on the industry, particularly on the NBE’s monetary and forex market reforms.
In this interview, Eshetu expresses cautious optimism about the central bank’s latest decisions and draws from his experience to shed light on what stands to improve. EXCERPTS:
The Reporter Magazine: In an article you wrote, you argued that Ethiopia should gradually move away from the blanket credit cap toward market-based monetary policy instruments. Since then, the NBE has removed the private-sector credit growth cap, increased the policy rate to 16 percent, and announced targeted reserve requirements for individual banks. Do you believe this new policy package adequately addresses the concerns you raised, or do important gaps still remain?
Eshetu Fantaye: The NBE’s July 13 decision is a significant and welcome step. Removing the blanket cap, raising the policy rate to 16 percent, and introducing bank-specific reserve requirements tied to loan-to-deposit ratios are precisely the kind of moves I advocated. Governor Eyob Tekalign(PhD) put it well when he described this as “a change in instrument, not a change in stance.” That framing is correct and important, the objective of containing inflation has not changed; only the tools have matured.
That said, the package addresses the monetary side of the equation. Several gaps remain on the execution front. First, the NBE has not yet introduced positive incentive mechanisms to direct credit toward productive sectors. My article drew on Tanzania’s reserve requirement relief for agricultural lenders and Uganda’s Agricultural Credit Facility as models. Removing the cap frees banks to lend, but it does not steer that lending toward exporters, manufacturers, and agribusinesses. Without incentive-based mechanisms, the risk is that freed-up credit flows disproportionately toward trade finance and real estate, where returns are faster and collateral is simpler.
Second, the quarterly advance tax policy remains in place. The article identified this as a compounding liquidity drain on productive enterprises. Even with the credit cap gone, businesses that must prepay taxes quarterly face a working capital squeeze that limits their ability to absorb new credit productively. Fiscal and monetary policy need to work in tandem.
Third, the reduction of the foreign-exchange surrender requirement from 50 to 30 percent and the cut in the FX transaction commission from 2.5 to 1.5 percent are constructive measures that should improve export competitiveness and deepen the FX market. These were not part of my original recommendations, and I regard them as a positive addition to the toolkit. The question is whether they will be sufficient to close the gap between the official and parallel exchange rates, which remains a structural source of inflationary pressure.
You maintained that the blanket credit cap redirected excess bank liquidity into Treasury bills instead of productive investment. Ethiopia’s 2026/27 federal budget now provides for 320 billion Birr in domestic borrowing. Do you expect banks to continue allocating a significant share of their liquidity to government securities, or will the removal of the credit cap materially increase lending to the private sector?
Both will happen, and the tension between them is real. The 2026/27 federal budget of 2.34 trillion Birr, with 320 billion Birr in planned domestic borrowing, means the government will remain a large and reliable borrower. Treasury bills and bonds offer banks a risk-free return with zero provisioning requirements. That gravitational pull does not disappear because the credit cap has been lifted.
What changes is the opportunity cost calculation. Under the cap, banks had no choice; they could not lend beyond the ceiling, so excess liquidity had nowhere to go except government paper. The 3.5-to-1 oversubscription at the June 2026 T-bill auction illustrated that dynamic vividly. With the cap removed, banks can now weigh the risk-adjusted return on a loan to a manufacturer or exporter against the yield on a Treasury bill. If the policy rate at 16 percent transmits effectively into lending rates, and if the NBE’s targeted reserve requirements penalise banks with high loan-to-deposit ratios, we should see a gradual rebalancing toward private-sector credit.
But “gradual” is the operative word. Banks will not abandon government securities overnight, nor should they. The practical question is whether the NBE and the Ministry of Finance can coordinate so that domestic borrowing does not crowd out the very private-sector lending the cap removal was designed to enable. This is where the 320 billion Birr figure deserves scrutiny; if the government absorbs that volume through competitive auctions at attractive yields, it will compete directly with private borrowers for the same pool of bank liquidity. The solution is not to suppress government borrowing, which funds essential services, but to ensure that the terms and instruments are designed to complement, not displace, productive lending.
The NBE’s latest Monetary Policy Committee statement says reserve-money growth was primarily driven by higher net foreign assets stemming from gold operations, while the IMF’s Fifth Review also discusses the monetary impact of the NBE’s gold purchases. Do these developments change your assessment of Ethiopia’s inflation-control framework? Can conventional monetary policy tools fully offset this source of liquidity, or do you believe additional policy measures are needed?
This is an important and under-appreciated dimension of Ethiopia’s inflation story. As we have discussed in the past, when the NBE purchases gold domestically, it pays in Birr and at a premium, injecting liquidity directly into the monetary base. If those purchases are large and sustained, they become a source of reserve-money growth that operates independently of the credit channel. The IMF’s Fifth Review flagged this explicitly, and the Deputy Managing Director’s statement called for “a well-designed plan for NBE to improve its gold market operations, and eventually exit the gold market, consistent with reserve accumulation objectives.” Will they exit from the gold market? At least we have seen them planning the December 2026 window, alongside the IMF.
Conventional monetary tools, the policy rate, reserve requirements, and open market operations, can absorb some of this liquidity. But they work through the banking system, while gold-related liquidity injection operates outside it. If the NBE is simultaneously tightening through a 16 percent policy rate and loosening through large-scale Birr payments for gold, the net monetary stance becomes ambiguous. This is a structural issue that rate adjustments alone cannot fully resolve.
The additional measures needed are operational rather than monetary. The NBE should publish a transparent framework for its gold operations, including purchase volumes, pricing methodology, and sterilisation mechanisms. If gold purchases are funding reserve accumulation, the Birr injected should be sterilised through corresponding open market operations or dedicated sterilisation instruments. Without that discipline, gold operations risk becoming a backdoor source of monetary expansion that undermines the credibility of the interest-rate-based framework the NBE has just adopted.
The NBE has announced that it may impose additional reserve requirements on individual banks based on developments in their loan-to-deposit ratios. In your article, you proposed differentiated reserve requirements inspired by Tanzania’s model. Do you believe the NBE’s newly announced approach is sufficiently targeted, or would you still recommend a different framework?
The NBE’s approach and the Tanzanian model serve different purposes, per me, and ideally Ethiopia would use both. The NBE’s bank-specific reserve requirements tied to loan-to-deposit ratios are a prudential tool; they may constrain banks that are lending aggressively relative to their deposit base, reducing systemic risk. Governor Eyob described this as “a precise instrument to act on individual banks, rather than the economy-wide constraint the credit cap once provided.” That is a sound description, and the mechanism is a clear improvement over the blanket cap.
What it does not do is direct credit toward productive sectors. The loan-to-deposit ratio is sector-blind; it treats a bank lending heavily to importers the same as one lending heavily to manufacturers, provided both are within their deposit-supported capacity. Tanzania’s model works differently. By reducing the Statutory Minimum Reserve for banks that extend agricultural credit below a defined interest rate, the Bank of Tanzania creates a positive financial incentive for banks to serve productive borrowers. The reward is built into the reserve structure itself. This requires a nuanced operational orchestration of the policy.
I would recommend that the NBE layer a sectoral incentive on top of its prudential framework. Banks that allocate a defined share of their portfolio to exporters, manufacturers, agribusinesses, and small and medium enterprises at rates below a specified threshold could receive a reduction in their reserve requirement. This does not replace the loan-to-deposit ratio mechanism; it complements it. One tool manages risk at the institutional level; the other steers credit toward the sectors that generate employment, foreign exchange, and supply-side capacity. Both are needed.
The Commercial Bank of Ethiopia continues to hold a dominant position within the banking sector. To what extent will CBE’s lending and investment decisions determine whether the removal of the credit cap supports productive investment, increases government financing, or contributes to inflationary pressure?
CBE’s role is decisive, and any honest assessment of the post-cap environment must start there. CBE holds a disproportionate share of total banking assets, deposits, and government securities. Its lending and investment decisions set the direction for the system as a whole. If CBE channels its freed-up lending capacity toward productive sectors, the cap removal will deliver the growth dividend the policy intends. If it continues to absorb the bulk of government domestic borrowing while rationing private-sector credit, the structural imbalance the article identified will persist under a different name.
The IMF’s Fifth Review noted that projections for credit to the private sector and state-owned enterprises include the impact of CBE recapitalisation. This is significant. A recapitalised CBE with a stronger balance sheet has the capacity to expand productive lending meaningfully. But capacity is not the same as mandate. The NBE, as both regulator and shareholder influence, needs to ensure that CBE’s post-cap lending strategy is aligned with the broader objective of supply-side expansion. This is not about directing individual loans; it is about ensuring that the largest bank in the system does not default to the path of least resistance, which is government paper and trade finance.
Private banks will watch CBE closely. If CBE moves aggressively into productive lending, private banks will follow to protect market share. If CBE stays conservative, private banks will have limited incentive to take on the higher-risk, longer-tenor lending that manufacturing and agriculture require. CBE’s behaviour will shape the competitive dynamics of the entire post-cap credit market.
Some economists argue that although access to credit may improve, businesses borrowing at relatively high interest rates could eventually pass those financing costs on to consumers through higher prices. How significant do you believe this transmission channel could be under Ethiopia’s current monetary policy framework?
The concern is legitimate but needs to be placed in proportion. Yes, higher borrowing costs are a production input, and businesses will attempt to pass them through to prices. But the alternative, a blanket credit cap that starves productive firms of financing altogether, produces a worse inflationary outcome. A manufacturer that cannot secure working capital operates below capacity, reduces output, and contributes to the supply shortfall that drives prices up. The cost-push effect of a 16 percent policy rate is real but bounded; the supply-side damage of credit rationing is open-ended.
The practical question is the spread between the policy rate and actual lending rates. If banks price loans at 20 to 24 percent, the financing cost burden on borrowers is substantial. This is where the incentive mechanisms I have been advocating become relevant. Reserve requirement relief for banks lending to productive sectors at concessional rates, or a targeted credit facility with development finance backing, can bring the effective cost of productive credit below the headline lending rate. The goal is not to subsidize credit indiscriminately, but to ensure that the sectors with the highest supply-side multiplier, exporters, manufacturers, agribusinesses, can borrow at rates that make production viable. NBE must firmly negotiate with the IMF team to accept the exception.
There is also a time dimension. In the short term, higher rates may add to costs. Over 12 to 18 months, if the increased credit access enables firms to expand output, improve productivity, and reduce unit costs, the net effect on prices should be disinflationary. The transmission channel from credit cost to consumer prices is real, but it is slower and weaker than the transmission channel from credit availability to supply expansion.
Many expect that removing the credit cap will lead to a rapid expansion of bank lending, rising property and real-estate prices, and higher prices across the broader economy. Which of these expectations do you consider economically justified, and which do you believe are misconceptions?
Let me take each in turn. Having worked in the banking environment, I don’t see my colleagues going for rapid expansion of bank lending automatically. The cap removal gives banks permission to lend more, but it does not give them more deposits. The NBE’s new bank-specific reserve requirements tied to loan-to-deposit ratios act as a natural brake; note that, banks that expand lending faster than they mobilize deposits will face higher reserve obligations. The policy rate at 16 percent also raises the cost of interbank borrowing, which limits the pace at which banks can fund new loans. I expect lending growth to accelerate, but not explosively. The IMF projects credit to the private sector and SOEs growing at about 24.8 percent in 2026/27, which is strong but not runaway.
Rising property and real-estate prices are a legitimate concern, but they are driven more by structural factors, urbanisation, limited housing supply, diaspora remittances, and the absence of alternative investment vehicles, than by bank credit policy. The credit cap did not prevent property prices from rising over the past three years; removing it will not be the primary driver of further increases. That said, the NBE should monitor sectoral credit allocation closely. If a disproportionate share of new lending flows into real estate speculation rather than productive investment, macroprudential tools such as loan-to-value limits and higher risk weights on property lending should be deployed. Kenya and Rwanda both use these tools effectively.
Broad-based price increases driven solely by the cap removal are largely a misconception. Inflation in Ethiopia is driven majorly by food supply disruptions, energy costs, rent, transport cost, exchange rate pass-through, and external shocks such as the Hormuz-related import price surge. The credit cap was one tool among many, and its removal does not open the floodgates to demand-pull inflation. The 16 percent policy rate, targeted reserve requirements, and continued open market operations provide multiple lines of defense. The risk is not that removing the cap causes inflation; the risk is that failing to complement the removal with supply-side investment perpetuates the structural bottlenecks that have been the real source of price pressure all along.
Looking ahead over the next 12 months, what developments or indicators would convince you that removing the credit cap has been successful or, alternatively, that the policy needs to be reconsidered?
I would watch five indicators closely. First, the sectoral composition of new bank lending. If the share of credit going to manufacturing, agriculture, and export-oriented enterprises increases relative to trade and real estate, the cap removal is working as intended. The NBE should come up with robust M&E and publish quarterly data on sectoral credit allocation to make this transparent.
Second, headline inflation. The IMF projects average inflation at 11.7 percent for 2025/26 and 12.3 percent for 2026/27.3 If inflation stays within or below that band despite the cap removal, it confirms that the interest-rate-based framework is holding. If inflation accelerates materially beyond projections, the NBE will need to tighten further, whether through additional rate increases, higher reserve requirements, or both.
Third, the interbank money market. The depth and stability of interbank trading will tell us whether the policy rate is transmitting effectively through the banking system. If interbank rates cluster around the policy rate corridor, the transmission mechanism is working. If they diverge widely, the NBE’s signalling power is weaker than assumed.
Fourth, the loan-to-deposit ratios of individual banks. The NBE’s new targeted reserve requirements depend on this metric. If banks respond by mobilising deposits more aggressively, that is a healthy outcome. If they simply slow lending to stay within their ratios, the cap has been replaced by a de facto constraint that operates through a different channel.
Fifth, foreign exchange market dynamics. The reduction in the surrender requirement from 50 to 30 percent and the lower FX commission should improve hard currency availability. If the spread between the official and parallel exchange rates narrows over the next 12 months, it signals that the FX reforms are gaining traction. If the spread widens, external price pressures will continue to feed into domestic inflation regardless of what happens on the credit side.
Success, in my view, would look like this. Inflation on a declining trajectory, productive-sector credit growing faster than trade finance, the interbank market deepening, and the FX spread narrowing. If those four conditions hold over the next 12 months, the transition from quantity-based to price-based monetary policy will have been vindicated. If they do not, the NBE will need to recalibrate, not by reinstating the cap, which would be a step backward, but by strengthening the complementary instruments; viz., sectoral incentives, macro-prudential tools, fiscal coordination, and transparent gold operations.
The direction of travel is right. What matters now is the quality of execution.
