Why Banks Keep: Understanding Interest Rate Spreads Around the World

The hidden tax on savings

Every time you put money in a bank, you’re making a trade. You give the bank your cash, and in return, they give you interest. The bank then lends your money to someone else — a homebuyer, a small business, a government — and charges them a higher rate. The difference between what they pay you and what they charge borrowers is the interest rate spread, and it’s the banking system’s margin for error, overhead, and profit.

In South Korea, that spread is just 1.2 percentage points. In Zimbabwe, it’s 51 percentage points. Both are real numbers from the same World Bank dataset, and both reveal something fundamental about how those banking systems work — or don’t work.

For a complete ranking of spreads across 105 economies, see our global interest rate map.

What a spread actually measures

The interest rate spread is simple in concept: it’s the weighted average lending rate minus the weighted average deposit rate. But that single number captures a lot of complexity.

When a bank makes a loan, it’s not just passing along your deposit. It’s covering:

Operating costs. Branches, staff, IT systems, compliance, marketing.
Credit risk. Some borrowers won’t pay back. Banks price that expected loss into every loan.
Liquidity risk. Banks need to keep enough cash on hand to meet withdrawals, even if all their loans are long-term.
Regulatory costs. Capital requirements, reserve requirements, deposit insurance premiums.
Profit. Banks are businesses. They need to earn a return for shareholders.

A wide spread doesn’t necessarily mean banks are greedy. It often means the banking environment is expensive to operate in — high default rates, thin competition, heavy regulation, or macroeconomic instability. For context on how credit risk varies across countries, see where bad loans pile up.

The extremes: efficiency vs. dysfunction

At the tight end of the spectrum, spreads under 2 percentage points suggest banking systems that are either highly competitive or tightly regulated. South Korea at 1.2 pp, Switzerland at 2.0 pp, and Qatar at 1.0 pp all fit this pattern. These are economies where banks can operate efficiently, credit risk is manageable, and competition keeps margins thin.

For a look at how Korean households use financial services, see the debit card divide.

At the wide end, spreads above 10 percentage points tell a different story. Zimbabwe at 51 pp, Madagascar at 47 pp, and Brazil at 32.5 pp are places where banking is expensive — for both savers and borrowers. In these economies, banks face high operating costs, significant credit risk, or both, and they pass those costs through in the form of wide margins.

For a deeper look at borrowing costs in these economies, see the most expensive places to borrow and the cost of credit in 2026.

Why spreads matter for financial inclusion

A wide spread is more than a banking metric — it’s a barrier to financial inclusion. When banks keep a large margin between deposit and lending rates, it means:

Savers earn less. Households get poor returns on their savings, which discourages formal saving and pushes people toward informal alternatives — cash under the mattress, livestock, or rotating savings groups.
Borrowers pay more. Small businesses and households face high borrowing costs, which limits investment, consumption, and economic mobility.
Banking stays expensive. Wide spreads often correlate with limited branch networks, high fees, and restrictive account requirements.

The result is a vicious cycle: high spreads discourage both saving and borrowing, which keeps the banking system small and inefficient, which keeps spreads high. Breaking that cycle requires competition, better credit infrastructure, or regulatory intervention — sometimes all three.

For a broader look at how households save across different economies, see how the world saves and where the world parks its money.

The Brazil puzzle

Brazil offers a case study in persistent wide spreads. The country’s 32.5 percentage-point spread in 2024 is the widest among major economies, and it’s been that way for decades. Even as Brazil’s deposit rates fell from hyperinflation-era peaks — from over 9,000% in 1990 to 7.7% in 2024 — the spread remained wide.

Why? Brazil’s banking system is concentrated, with a few large banks dominating the market. Credit risk is high: nonperforming loans have historically been a significant burden. The tax and regulatory environment adds cost. And Brazilian banks have traditionally earned high returns, with little competitive pressure to pass savings to customers.

The result is a banking system that’s profitable for shareholders but expensive for everyone else. For a look at how Brazilian households and firms access finance, see where firms rely on bank loans.

The Korea model

South Korea’s 1.2 percentage-point spread is the tightest among major economies, and it reflects a different set of conditions. Korean banks operate in a competitive market with strong consumer protection. Credit risk is relatively low, thanks to a well-developed credit bureau system and a culture of repayment. Operating costs are manageable, thanks to high digital adoption — for more on how Koreans use digital financial services, see the debit card divide and where salaries are paid digitally.

The tight spread means Korean savers get reasonable returns and borrowers face manageable rates. It’s not a perfect system — household debt is high, and some argue banks take excessive risks — but by the metric of intermediation efficiency, South Korea is a global leader.

What about deposit rates?

The spread tells only half the story. The other half is the deposit rate itself — what banks actually pay savers. A country can have a narrow spread but still pay depositors almost nothing if the overall interest rate environment is low.

Switzerland is a case in point. Its 2.0 pp spread is among the tightest in the world, but its deposit rate in 2024 was just 0.9%. For years, Swiss deposit rates were actually negative — banks charged customers to hold their money. The tight spread meant banks weren’t taking much margin, but the baseline was so low that savers still lost out.

For a complete picture of deposit rates across 113 economies, see our global interest rate map.

China offers a different example. Its deposit rate of 1.5% has been unchanged since 2015, a decade of administered stability. The spread of 2.9 pp is moderate, but the fixed deposit rate means Chinese savers don’t benefit when global rates rise — or suffer when they fall. For a look at how Chinese households save and invest, see where the world parks its money.

The inflation distortion

In high-inflation economies, both deposit rates and spreads can be misleading. Turkiye paid 71% on deposits in 2024, and Argentina paid 54%. Those sound like excellent returns — until you account for inflation that exceeded 60% and 200% respectively. Real returns were deeply negative.

In these environments, the spread can be narrow even when banking is expensive. Argentina had a 7.5 pp spread in 2024, which sounds moderate but is actually quite wide in real terms when inflation is triple-digit. The high nominal rates mask the true cost of intermediation.

For a look at how households in these economies cope, see how the world saves.

What the data doesn’t capture

Interest rate spreads are a useful summary statistic, but they don’t tell the whole story. They don’t capture:

Fees and charges. Account maintenance fees, transaction fees, and penalties can add significantly to the cost of banking.
Credit rationing. In some economies, banks lend at reasonable rates but to a very small pool of borrowers. Most households and firms can’t access credit at any price. For more on this, see where credit runs deepest and where firms rely on bank loans.
Non-price terms. Collateral requirements, loan covenants, and repayment schedules can make a loan more or less accessible, independent of the interest rate.
Informal alternatives. In many economies, households and businesses rely on informal lenders, rotating savings groups, or trade credit rather than formal banks. For a look at how mobile money is changing this in some regions, see credit cards vs. mobile money and the top 10 mobile money countries.

For a broader discussion of how to interpret financial statistics, see reading fintech statistics responsibly.

The policy question

Wide spreads are often a symptom of deeper problems — weak institutions, poor infrastructure, macroeconomic instability. Fixing them requires more than regulatory tinkering. It requires building the foundations of a functioning financial system: property rights, contract enforcement, credit bureaus, competition policy, and macroeconomic stability.

But spreads can also be a policy choice. Some governments keep deposit rates low to make borrowing cheap for favored sectors. Others restrict competition to protect incumbent banks. In these cases, the wide spread is a feature, not a bug — and changing it requires political will, not just technical fixes.

For a look at how different economies approach financial sector policy, see our reports on bank capital adequacy, bank liquidity reserves, and how the world finances business.

Sources & method

All data comes from the World Bank’s World Development Indicators, accessed through the bank lending-deposit rate spread (FR.INR.LNDP) and deposit interest rate (FR.INR.DPST) series. Spreads cover 105 economies; deposit rates cover 113. Data years range from 2020 to 2024 depending on country reporting. For full methodology detail, see the methodology page.