Editor’s note: This is general educational information about how to read Federal Government of Nigeria bond auction results. It is not investment advice. It is based on the debt office and exchange documents listed at the end.
The word that attaches itself to a Nigerian bond auction is usually oversubscribed, and it is usually true. At the August 2026 auction the Debt Management Office offered ₦100.00 billion of the 16.2499% FGN APR 2037 and received bids of ₦392.48 billion. Read on its own that looks like an emphatic vote of confidence. Read against the rest of the same table it looks like something narrower and more interesting: a book in which most of the bidders wanted a yield the government was not willing to pay, and were turned away.
What the results table actually publishes
A DMO auction summary carries a fixed set of lines for each instrument. Amount offered. Total bids and successful bids, counted as numbers of bids rather than value. Subscription, meaning the naira value of everything submitted. Non-competitive allotment. Amount allotted to market. Range of bids. Marginal rate.
Only two of those are outcomes. The marginal rate is the yield at which the auction cleared, and the amount allotted is what the government actually raised. Everything else describes the book that was submitted, and a book is a set of offers, not a set of commitments at whatever price the issuer names.
Subscription is not demand
Subscription is the number that generates the headlines and it is the least informative in the table, because it aggregates bids across every yield anyone asked for. The August 2026 auction makes the point three times over.
On the 16.2499% FGN APR 2037 the DMO offered ₦100.00 billion, subscription came to ₦392.48 billion across 171 bids, and 34 bids succeeded. The marginal rate was 17.1900%. On the 22.60% FGN JAN 2035 the office offered ₦250.00 billion, subscription came to ₦513.61 billion across 199 bids, 31 of them succeeded, and only ₦64.13 billion was allotted to the market at a marginal rate of 17.1500%. On the 15.45% FGN JUN 2038, offered at ₦750.00 billion, subscription was ₦821.32 billion, 161 of 225 bids succeeded, and the rate cleared at 17.7900%.
Set that against the previous month. At the July 2026 auction, held on July 20, 2026, the DMO offered ₦400.00 billion of each of the same three bonds. Subscriptions were ₦555.47 billion, ₦665.19 billion and ₦518.00 billion, and the marginal rates were 18.3400%, 18.3500% and 18.4000%. Success rates were far higher: 90 of 184 bids, 126 of 211, and 92 of 161.
The June 2026 auction offered ₦600.00 billion of each of two bonds, drew subscriptions of ₦705.22 billion and ₦708.27 billion, allotted ₦600.90 billion and ₦621.00 billion to the market, and cleared at 18.3400% and 18.3500%, with 135 of 179 and 181 of 215 bids successful.
The pattern across the three months is the opposite of what subscription alone suggests. In June and July, when the marginal rates were above 18%, a majority of bids were accepted. In August, when the clearing rates came in near 17%, subscription cover on the 2037 rose sharply and the proportion of bids accepted fell. A larger book with a smaller share of bids accepted reflects where the allotment was cut, not the size of demand at any single yield.
The line that is easy to miss
The August table contains a figure that reorders the whole reading. Against ₦631.02 billion allotted to the market on the 15.45% FGN JUN 2038, the non-competitive allotment was ₦742.29 billion. On the 22.60% FGN JAN 2035 the non-competitive allotment was ₦10.00 billion. In July the only non-competitive figure was ₦50.00 billion, on the 2038. In June there was none.
A non-competitive allotment is taken at the marginal rate set by the competitive book rather than at a price the taker bid. The bidder does not express a view on yield, and by definition adds nothing to the price discovery that produced the rate. On the 2038 in August the non-competitive amount exceeded the entire competitive allotment. Any statement about demand for that bond that treats the two as the same kind of number is describing something other than what happened.
The instruments are the same each month
The comparison holds up because the auctions are re-openings. The August offer circular confirms that all three instruments were re-openings of previously issued bonds, and that successful bidders in a re-opening pay a price corresponding to the yield-to-maturity bid that clears the volume, plus accrued interest, with the original coupons maintained. The 22.60% FGN JAN 2035 auctioned in August is the identical security to the one auctioned in July and June. So a change in the marginal rate is a change in the market’s required yield on one security, not a comparison of two different bonds.
The instruments also have a secondary market to be priced against. The circular states that FGN bonds are listed on Nigerian Exchange Limited and FMDQ OTC Securities Exchange, and NGX operates a fixed income market making platform through which participants trade listed federal, state and corporate debt on an order book with firm orders, with pre-trade and post-trade reporting. An auction rate that diverges from where the same bond trades on that book is a fact about the auction, not about the bond.
Analysis: what a strong auction establishes
A high subscription figure establishes one thing reliably: that the primary dealer network showed up. The offer circular names the fourteen Primary Dealer Market Makers through whom bids are placed, and a thin book from that group would be a real signal of stress. Nigeria’s auctions have not produced one in this sequence.
What subscription does not establish is the level. The August auction cleared more than a full percentage point below July on the same 2035 bond, at 17.1500% against 18.3400%, while the acceptance rate fell from 90 of 184 bids to 31 of 199. Both of those are consistent with a market that would lend at 18% and an issuer that would only borrow at 17%. They are equally consistent with a genuine repricing of Nigerian duration between the two dates. The auction table cannot distinguish them, and neither can any commentary built solely on the cover ratio.
The three lines that carry the most information are the ones least often quoted. The count of successful bids against total bids shows how much of the book the issuer rejected. The range of bids shows how far apart the participants were, and the August range on the 2038 opened at 16.0000% against a clearing rate of 17.7900%. The non-competitive allotment shows how much of the raise came from participants who took the price rather than set it.
There is a further limit worth stating. The issuance calendar for the third quarter of 2026 published ranges of amount on offer rather than fixed sums, with the lower bound at ₦500 billion in some slots and ₦200 billion in others, with a note that the calendar is provisional and subject to change at short notice. The DMO also reserves in the circular the right to allot at its discretion. An auction where the issuer chose the size, chose the allotment and declined most of the book is not a referendum on the government’s credit. It is a negotiation, and the results table publishes both sides of it for anyone willing to read past the first line.
