The Shilling at a Two-Year Low, and the Anchor Meets Its Test

Brent moved above US$105 after a strike on a Saudi pipeline on 11 September took around four million barrels a day out of the market. Uganda imports essentially all of its fuel, so the dollar bill arrives immediately.

US 10-year yields are near 5%, the highest since 2007, and that pulls capital away from frontier markets such as Uganda. This is the scenario Bank of Uganda listed first among its upside risks in August.

Local businesses are buying more dollars to pay for fuel and other imports, and companies are also stocking up ahead of December. At the same time there are fewer dollars available, which puts more pressure on the shilling.

Gold is Uganda’s biggest export earner, bringing in billions of dollars a year. Gold export earnings fell by about 16% month on month in June, which means fewer dollars coming into the country for the same effort.

At the 9th Annual Bankers Conference on 18 September, Finance Minister Henry Musasizi said lending rates of 18–20% are holding businesses back and called for cheaper credit. He added that private-sector lending must grow from about Ugx 28 trillion today to Ugx 490 trillion by 2040 for the government to reach its target of a US$500 billion economy. Governor Michael Atingi-Ego stressed that banks cannot provide all of that alone: equity investment, diaspora funding and blended finance will be needed too.

There is a bigger issue behind the cost of loans. Banks are putting more of their money into government securities and less into private-sector lending. Government securities grew from 26% to 31% of bank assets between 2021 and 2025, while loans to businesses and individuals fell from 41.2% to 37.3%. In plain terms: when banks lend more to government, there is less money left for everyone else, and borrowing becomes more expensive.

The cash reserve requirement rises from 11% to 13.5%, effective 24 September. Commercial banks must keep a larger share of customer deposits at the central bank instead of lending or investing it. This is the second increase this year, after the move from 9.5% to 11% in May. The Central Bank Rate remains at 9.75%, so the headline interest rate did not change.

The reserve increase matters because it reduces the number of shillings available in the banking system, which can ease pressure on the currency. Despite the move, the shilling has not strengthened yet: it went from Ugx 3,927.74 to Ugx 3,945.96 per dollar by Friday. The requirement only takes effect on 24 September, so it is too early to judge. In plain terms, Bank of Uganda is supporting the shilling by making it less plentiful rather than by selling dollars.

Kenya’s stock market fell sharply last week as a stronger dollar and higher borrowing costs pressed on investors. The All-Share Index fell 4.96% and the NSE 20 fell 6.47%, wiping about Ksh 206 billion off the value of listed companies. Uganda’s market fell only 1.25%, but trading was much lighter. Here, the pressure has been felt through the shilling rather than through share prices.

NIC Holdings’ share price jumped 33.33% to Ugx 16, but only 3,500 shares worth Ugx 51,500 changed hands. That is how thin trading can be on Uganda’s exchange: a very small amount of buying moves a price a long way. A sharp price rise does not mean the company’s underlying value rose by the same amount.

An article in the 18 September edition of New Vision reported that about Ugx 46.6 billion of member contributions remain unallocated to individual accounts, or 0.18% of NSSF’s total member balances. Some of it goes as far back as 2011, held up by incomplete or incorrect member information and by difficulty tracing contributors. The figure has been falling as NSSF reconciles money when members produce evidence of their contributions, and it has published a list of the affected members and employers on its website: about 209,040 individuals and 521 employers. The lesson for a saver is ordinary but expensive to ignore — keep your NSSF records and details accurate and up to date.

Both debut bonds on 9 September cleared at exactly 100.000, at par, because Bank of Uganda picked the coupon and refused anything above it. We expect the same here, and a 15.000% coupon consistent with the 20-year. If it clears above 15.000%, the anchor has broken, and that is the story of the week.

On an existing bond the coupon is already fixed, so investors can only adjust the price they will pay. With money tighter, they are likely to want a higher return, which means the 2-year and 5-year could clear above their August levels. Bank of Uganda may respond by borrowing less rather than accepting much higher rates.

The new cash reserve requirement takes effect on the same day investors have to pay for the bonds. Banks will need to place more money with Bank of Uganda while also finding the cash to settle their purchases. That leaves less to invest, which means fewer bids and potentially higher yields on the re-openings. That is the investor’s side of the argument. The other side is that Bank of Uganda has repeatedly shown it will not borrow expensively, with the longer tenors repricing to around 15.000%. Whose terms are met on Wednesday is the thing to watch.

This auction carries four maturities against three in each of the previous three. With money already tight, that is a bigger amount for the market to absorb, and Bank of Uganda may well end up borrowing less than it has advertised.

A 15.000% return in shillings looks attractive, but if the shilling weakens sharply against the dollar, part of that return disappears when you convert it. The new 20-year is already showing a loss of about 3% in dollar terms eight days after settlement. It can still be a good investment if your goal is to earn in shillings; it looks very different if your goal is to protect or grow money in dollars.

We hold last week’s view. Bonds below ten years keep between 7.767% and 12.470% after tax; ten years and longer keep about 13.669% and above. The important tax break arrives at ten years. Beyond that, you lock your money away far longer without being paid for it — the 25-year keeps 13.823% against the 10-year’s 13.669%, which is not a reward worth fifteen extra years. Until the long bonds pay meaningfully more than the 10-year, there is little reason to take that risk.

The 2-year and 5-year could be more interesting than the new bond. Their coupons are fixed, so only the price moves, and with banks short of cash once the reserve requirement lands, they may come cheaper and yield more. The 2-year is taxed at 20% and the 5-year at 10%. Ten years and longer still win on tax, but a 5-year keeping 12.470% instead of the 11.085% it would keep at 20% withholding is a materially different proposition.