Why most loan products are designed backwards

By John Pabi Ntifo

The structural misalignment between how credit is packaged and how borrowers actually earn money is not a market failure. It is a design failure, and it is hiding in plain sight.

In Accra’s Makola Market, a fabric trader takes a GH?5,000 loan in January. Her sales are seasonal and her income peaks sharply in April before the Easter period and again in December before Christmas.

Her repayment schedule, however, does not know this. It expects the same fixed monthly payment from February through December, indifferent to the rhythm of her trade. By June, she is borrowing informally to service a formal loan. By August, she is in arrears. The bank records a nonperforming loan. The borrower absorbs the penalty. Both parties lose.

This is not an isolated misfortune. It is the structural condition of consumer and microenterprise lending in Ghana. The system is designed, consciously or otherwise, around the operational requirements of the lender rather than the economic reality of the borrower. Understanding why requires a close examination of how loan products are built, who they are built for, and what assumptions are baked into their architecture from the very beginning.

The Architecture of a Loan Product

When a Ghanaian bank or microfinance institution designs a loan product, the core engineering problem it is solving is risk management. It must assess the probability of default, price that risk into an interest rate, and structure repayment in a way that limits its exposure at every point in the loan’s life. This is rational institution level behaviour. The trouble is that none of these design imperatives require the product to succeed for the borrower. They only require it to succeed for the portfolio.

The standard outcome of this process is a fixed installment loan: a fixed principal disbursed in a lump sum, repaid in equal monthly instalments over a fixed term at a stated interest rate, typically calculated on a reducing or flat balance basis. Ghana’s formal lending market, from the major commercial banks to the proliferating savings and loans companies, overwhelmingly defaults to this structure.

GHANA CREDIT LANDSCAPE KEY INDICATORS

• Non-performing loan ratio, banking sector: 18.9%
• Share of working population in informal / self-employment (est.): 80%
• Typical microfinance loan term: 3 to 12 months
• Borrowers with irregular monthly income (est.): 60%

The fixed instalment structure is not inherently bad. For a borrower with a stable, salaried income paid monthly, it is perfectly adequate. The problem is that it is applied uniformly to a borrower population that is, by any reasonable measure, not uniformly salaried. Ghana’s labour market is dominated by informal workers, petty traders, smallholder farmers, seasonal contractors and microentrepreneurs.

These are people whose income arrives weekly, daily, or in irregular bursts tied to harvest cycles, market days, client payments, or festivals. The Ghana Statistical Service’s Annual Household Income and Expenditure Survey (AHIES) consistently record the informal sector as accounting for approximately 80% of employed Ghanaians, concentrated in agriculture, trade, and services.

Applying a monthly repayment schedule to a weekly income borrower is not a minor inconvenience. It is a fundamental structural mismatch.

The Income Cycle Problem

The Makola trader is not an edge case. Ghana’s informal labour market is characterised by income that is weekly, daily, or seasonal rather than monthly. The Ghana Statistical Service’s GLSS7, conducted across 15,000 nationally representative households, documents that the majority of employed Ghanaians work in self-employment and informal enterprises where earnings arrive daily, weekly, or seasonally rather than in monthly salary payments.

The financial diaries literature, which tracks actual household cash flows fortnightly over a full year, consistently finds that low income households in economies like Ghana’s manage income that is not merely low but fundamentally irregular. It is characterised by volatile weekly inflows, seasonal peaks, and unpredictable gaps. Yet the overwhelming majority of loan products in Ghana’s microfinance portfolio require fixed monthly repayments, creating a systematic structural mismatch for the majority of their borrowers.

This mismatch creates a predictable cascade. The borrower receiving income on Thursday does not have a full month’s repayment sitting idle in an account by the first of each month. They must either hold cash, forgoing reinvestment in their trade, or scramble to assemble the repayment at month’s end from multiple smaller inflows. The cognitive and financial cost of this assembly is real, measurable, and almost entirely invisible in standard credit risk models.

Farmers face an even more acute version of the same problem. In Ghana’s cocoa belt, smallholder farmers receive the bulk of their annual income in a single payment from the Ghana Cocoa Board (COCOBOD) during the main crop season, typically between October and March. A 12 month loan taken in April, requiring equal monthly repayments, asks this farmer to service debt from income that does not yet exist. The structural response of borrowing informally to service formal debt is well documented in the agricultural finance literature and generates a secondary layer of indebtedness entirely invisible to the formal lender.

STRUCTURAL COMPARISON: TWO BORROWER PROFILES

Profile A: Salaried civil servant. Monthly income of GH?3,200 arrives on the 25th of each month. Takes a GH?10,000 loan with monthly repayments of GH?1,100. Cash flow alignment is near perfect. Default probability is low. The product works as designed.

Profile B: Market trader. Weekly income averages GH?800 but varies between GH?400 and GH?1,400. Takes a GH?10,000 loan with identical monthly repayments of GH?1,100. In a weak week, the full monthly repayment exceeds her weekly earnings. In a strong week, it is manageable. The variance, not the average, is the problem. The product is structurally unsuitable but was sold using the same template as Profile A.

The Flat Rate Distortion

Compounding the repayment schedule problem is the widespread use of flat rate interest calculation in Ghana’s microfinance sector. Under a flat rate structure, interest is calculated on the original principal for the entire loan term, regardless of how much has been repaid. A borrower taking a GH?5,000 loan at 5% per month flat for 12 months pays interest on GH?5,000 every month, even in month eleven when only GH?500 of principal remains outstanding.

The effective annual rate of a loan marketed at 5% per month using flat rate calculation is not 60%. It is approximately 110 to 120%, depending on the disbursement and repayment timing. Research has consistently found that a significant portion of microfinance borrowers cannot accurately state the effective interest rate on their current loan. This is not because they are unsophisticated, but because the product is designed in a way that makes true cost opacity a structural feature rather than an accident.

What Designing Backwards Actually Looks Like

The clearest diagnostic for a backwards designed loan product is whether the product structure requires the borrower to adapt their financial behaviour to the product, or whether the product adapts to the borrower’s financial reality. In virtually every conventional Ghanaian lending product, the answer is the former.

PRODUCT FEATURE	LENDER FIRST DESIGN	BORROWER FIRST DESIGN
Repayment schedule	Fixed monthly regardless of income cycle	Aligned to borrower’s earning frequency
Interest calculation	Flat rate on original principal	Reducing balance; transparent effective APR
Grace period	None, or added to principal	Calibrated to income cycle lag (e.g., post harvest)
Loan sizing	Determined by collateral or guarantor	Determined by cash flow capacity
Repayment stress	Addressed through penalties and enforcement	Prevented through design

The irony of lender first design is that it is also, ultimately, bad for lenders. Ghana’s banking sector NPL ratio stood at 20.6% at end December 2023, up from 16.6% in December 2022, according to the Bank of Ghana’s Financial Stability Review. The NPL ratio  stood at 18.9% as at December 2025.

This is one of the highest readings in the sector’s recent history. A substantial portion of these nonperforming loans are not the result of borrowers who could never repay. They are the result of borrowers who could have repaid under a well designed product, but could not repay under a poorly designed one. The cost of design failure is being socialised across the entire portfolio and passed back to borrowers as higher interest rates.

A Path Forward

The solution is neither to extend more credit nor to tighten approval criteria. It is to ask a different first question. Instead of beginning product design with how to manage risk, lenders should begin with how the borrower actually earns money. They should build repayment structures, pricing mechanisms, and loan terms outward from that answer.

This is not a radical proposition. It is standard practice in agricultural finance in Brazil, where BNDES linked rural credit products use harvest aligned balloon structures as a baseline. It is emerging in East Africa, where fintechs like c and Apollo Agriculture have demonstrated that income aligned loan structures can reduce default rates while expanding access. The technology required to implement it in Ghana exists. The market data required to inform it is increasingly available. What has been missing is the design intent.

The second article in this series examines a more fundamental gap: the near total absence of decision support infrastructure for borrowers at the point of taking a loan. A well designed loan product can still be a bad decision for a specific borrower at a specific moment. The Ghanaian lending market currently has no systematic mechanism for catching that distinction before it becomes a default.

NOTES & SOURCES

• Bank of Ghana, Banking Sector Report, Q4 2023. Loan product structure analysis based on publicly available product schedules from the ten largest retail lenders by loan book.
• Ghana Statistical Service, Annual Household Income and Expenditure Survey (AHIES), 2022 Q3. Informal sector employment share consistent across multiple AHIES quarters.
• Mensah, J.V. and Fosu, A.K., ‘Seasonal credit constraints among smallholder cocoa farmers in Ghana,’ Journal of African Economies, 2020.
• Bank of Ghana, Financial Stability Review 2023, published October 2024. NPL ratio December 2023: 20.6%; December 2022: 16.6%

John is a  Product Architect, Credit and Digital Lending

The post  Why most loan products are designed backwards appeared first on The Business & Financial Times.

Read Full Story